Olfaction in vector-host

interactions

edited by:

Willem Takken and Bart G.J. Knols

Ecology and control of vector-borne diseases

Volume 2
Olfaction in vector-host interactions
 Olfaction
in vector-host
interactions

Ecology and control of vector-borne diseases

Volume 2

edited by:

Willem Takken
and
Bart G.J. Knols

Wageningen Academic
P u b l i s h e r s
 This work is subject to copyright. All rights
are reserved, whether the whole or part of
the material is concerned. Nothing from this
publication may be translated, reproduced,
ISBN: 978-90-8686-091-3 stored in a computerised system or published in
e-ISBN: 978-90-8686-698-4 any form or in any manner, including electronic,
DOI: 10.3920/978-90-8686-698-4 mechanical, reprographic or photographic,
without prior written permission from the
publisher:
ISSN: 1875-0699 Wageningen Academic Publishers,
P.O. Box 220, 6700 AE Wageningen,
the Netherlands,
Cover photo: Hans Smid [email protected],
www.bugsinthepicture.com www.WageningenAcademic.com

The individual contributions in this publication

and any liabilities arising from them remain the
First published, 2010 responsibility of the authors.

The publisher is not responsible for possible

© Wageningen Academic Publishers damages, which could be a result of content
The Netherlands, 2010 derived from this publication.
Ecology and control of vector-borne diseases
In the past century, many advances were made in the control of vector-borne diseases. Malaria
disappeared from the northern hemisphere, diseases such as typhus, Bartonella and yellow
fever were seriously reduced in prevalence and in many countries effective methods of disease
control contributed to a greatly reduced incidence of such diseases. Most of these advances were
beneficial to the industrialised world, whereas underdeveloped countries continued to suffer
much as before. Indeed, several diseases such as malaria, Rift Valley fever and African sleeping
sickness are still highly prevalent in specific countries. ‘New’ vector-borne diseases such as dengue,
chikungunya fever and West Nile fever, have emerged and are invading previously disease-free
regions. The discovery of new drugs and vaccines has made great advances and allows for the
effective treatment and control of many diseases. In contrast, vector control has lagged behind in
development, even though it is realised that effective vector control would allow for an immediate
interruption of the transmission of disease, and aid in disease control and eradication. In the last
20 years new initiatives on vector control have been undertaken, leading to a rapid development
of effective and lasting methods of vector control. The development of molecular genetics has
provided new insight in vector biology and behaviour, which is being used for developing new
strategies of vector control. Advances in geographic information systems allow for precision
targeting of interventions. The collective information on new developments in Vector Ecology
and Control of Vector-borne Diseases is scattered over numerous periodicals and electronic
databases. This book series intends to bring together this information in sequential volumes
arranged around selected themes that are currently of interest. Forthcoming themes will include
‘Surveillance methods of vectors of infectious diseases’, ‘Recent advances in biological control
of mosquitoes’, ‘Transgenic tools for vector management’ and ‘Integrated vector management’,
but also fundamental biological topics such as ‘Mating behaviour of disease vectors’, ‘Oviposition
behaviour of disease vectors’ and ‘Parasite-vector interactions’. Other topics will be added as
perceived relevant.

Willem Takken and Bart Knols are well-known experts in the field of Medical and Veterinary
Entomology. Both have extensive experience from field work in the tropics and ecological studies
in the laboratory and field. They also have a wide experience in publishing and other methods
of research dissemination. Willem Takken is Professor in Medical and Veterinary Entomology
at Wageningen University. Bart Knols is visiting scientist at the Division of Infectious Diseases,
Tropical Medicine & AIDS of the Academic Medical Centre in Amsterdam, the Netherlands, as
well as director of K&S Consulting, a company engaged in the global dissemination of medical
information, organisation building, and consultancies in infectious disease control.

Olfaction in vector-host interactions 5

Table of contents
Preface 11

1. Strategic use of chemical ecology for vector-borne disease control 13

Willem Takken and Bart G.J. Knols

2. Molecular regulation of olfaction in mosquitoes 17

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

3. Olfactory physiology of blood-feeding vector mosquitoes 39

Yu Tong Qiu and Joop J.A. van Loon

4. Odour coding and neural connections 63

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

5. The detection of carbon dioxide and its role in the orientation to hosts by
haematophagous insects 91
Alan J. Grant and Robert J. O’Connell

6. Host finding by female mosquitoes: mechanisms of orientation to host odours and

other cues 115
Ring T. Cardé and Gabriella Gibson

7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory 143

Renate C. Smallegange and Willem Takken

8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies 181

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

9. The chemical ecology of sandflies 203

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

10. Understanding and exploiting olfaction for the surveillance and control of Culicoides
biting midges 217
James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

11. Black fly interactions with their hosts 247

James F. Sutcliffe

12. Olfaction in Glossina – host interactions: a tale of two tsetse 265

Steve J. Torr and Philippe Solano

13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding 291
Steve Schofield and Steve J. Torr

14. The role of olfaction in host seeking of triatomine bugs 309

Pablo G. Guerenstein and Claudio R. Lazzari

Olfaction in vector-host interactions 7

15. Chemical ecology of tick-host interactions 327
Sandra A. Allan

16. Parasite-mediated enhancement of transmission by haematophagous insects 349

Hilary Hurd

17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in
Brazil: a case study 365
Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

18. Exploitation of olfactory-mediated behaviour 399

Willem Takken and Bart G.J. Knols

About the editors 409

Contributors 411

Reviewers 415

Keyword index 417

8 Olfaction in vector-host interactions

Preface
The series Ecology and Control of Vector-borne Diseases was launched in 2007 with Volume I
providing an overview on the emergence of pests and vector-borne diseases in Europe. This first
volume has been well received, and was the starting point of an overview of the state of the art of
various topics that are relevant for this field of science. In the past decade the world has witnessed
an increasing number of outbreaks of vector-borne diseases as well as invasions of vectors and/
or pathogens. Whereas in the previous century many of these diseases were eliminated from the
temperate climate zones, some of them are found to expand their distribution, with invasions of
Europe, North America and Australia. Many of these events are associated with environmental
change and find their explanation by the often complex ecological processes that drive the
transmission of the pathogens. The prevention and control of these diseases requires in-depth
knowledge of the arthropods that transmit the pathogens. This book series aims for highlighting
specific aspects of the ecology of these vectors that can be exploited for the prevention and
control of these diseases.

The current volume addresses the topic of how blood-feeding arthropods interact with their
vertebrate hosts. As the transmission of infectious vector-borne pathogens is much dependent
on the contact between vector and host, the efficacy of host location is of profound importance.
Interruption of vector-host contact is considered one of the most effective means of vector-borne
disease control, as is currently witnessed by the successful use of insecticide-treated bed nets for
malaria prevention in sub-Saharan Africa.

As with Volume I, this book is a collaboration of a large number of scientists, each experts in their
specific field, who have written comprehensive overviews of specific aspects of olfaction-mediated
arthropod behaviour. Each chapter was reviewed by independent reviewers, and subsequently
adjusted as needed. We are grateful to the anonymous reviewers, as we realise that this takes time
and effort. We thank Mike Jacobs and Marijn van der Gaag of Wageningen Academic Publishers
for their patience and useful advice. We thank Hans Smid for once more having contributed to the
cover design. Special thanks to Françoise Kaminker, who has language-edited several chapters,
and provided useful advice.

Wageningen, 8 March 2010

Willem Takken and Bart Knols

Olfaction in vector-host interactions 11

1. Strategic use of chemical ecology for vector-borne disease control
Willem Takken and Bart G.J. Knols

Abstract

In the epidemiology and control of vector-borne diseases an accurate estimate of parasite and
pathogen transmission is required. This can be achieved by monitoring the vector population
using traps provided with attractive cues. Volatile chemicals are the principle cues with which the
disease vectors locate their vertebrate hosts. Natural and synthetic blends of such chemicals can
be used to manipulate the vectors, so that transmission risk can be determined and, in some cases,
the vector population can be killed. Novel scientific developments, including molecular genetics,
analytic chemistry and odour-release technologies provide opportunities for the effective use of
chemical ecology in vector-borne disease control.

Keywords: olfaction, insects, ticks, behaviour, disease transmission, nuisance, attractant, repellent

Introduction

In the last decade nuisance insects and disease vectors have received increasing attention
because of their continuing and devastating effects on human and animal health. Examples
are recent outbreaks of chikungunya fever in the Indian Ocean, SE India and Italy (Erin Staples
et al. 2009), the large number of recorded dengue outbreaks in South America (Barclay 2008)
and the arrival of bluetongue virus in NW Europe (Carpenter et al. 2009). In addition, the rapid
growth of the world population coupled with increasing poverty in resource-poor countries is
the cause of much suffering from neglected vector-borne diseases such as malaria, leishmaniasis
and trypanosomiasis (Mathers et al. 2007). Lastly, the current debate on climate change and its
potential effects on human health is causing global concern about emerging diseases such as West
Nile virus, Lyme disease and other tick-borne diseases (Epstein 2001, Gould and Higgs 2009, Jones
et al. 2008). Whereas for several of these diseases effective intervention strategies are available,
and hence their control can be implemented, for many of them the current control strategy is
inadequate. With some exceptions, effective vaccines are lacking, parasites have developed
resistance against drugs, and vector control efforts are hampered by insecticide resistance,
technologically inadequate tools and lack of resources.

It is widely agreed that the first step in monitoring vector-borne disease is accurate risk assessment
(Hay et al. 2004, Mathers et al. 2007). This information is needed to study the epidemiology of
the disease, and to predict its course in time and space. The basic reproductive number (R0)
(Hartemink et al. 2008, Smith et al. 2007) is used to express disease risk, and as the biting rate of
disease vectors is one of the critical parameters in R0, tools to accurately measure this are essential.
Examples in which these data are being widely used are malaria and dengue control programmes:
in endemic situations vector densities provide knowledge on spatial and temporal distribution
of biting intensities, while in epidemic situations, vector populations are measured so that the
course of the outbreak can be monitored. Also, the outcome of vector-targeted interventions,
aimed at the interruption of parasite/pathogen transmission, can be assessed when accurate data
on vector densities are available. Vectors are also collected to determine their infections with
disease agents: the intensity of disease transmission the product of vector density and parasite/
pathogen infection rate. In spite of the importance of accurate sampling of disease vectors, the
methodologies used for insect collection are in many cases based on perception rather than on

Olfaction in vector-host interactions 13

Willem Takken and Bart G.J. Knols

accurate knowledge of vector behaviour. This is because information on vector-host interactions

(at which level the parasites/pathogens are being transferred) is in many cases poor or lacking. It
is therefore of relevance that this knowledge be expanded in order to obtain better, more precise,
data on vector biting density and intensity of parasite transmission.

Since the discovery of the role of arthropods in disease transmission, entomologists have worked
on the development of tools for vector surveillance, and a wide range of traps is available with
which the vector population can be monitored (Qiu et al. 2007, Service 1993). Traps are designed
to attract the target insect(s) using sensory cues such as sound, colour and odour. As chemical
signals are the principal cues with which arthropods identify their blood hosts, many traps employ
odorants to lure the vectors to the trap. The odorants used vary from natural host odours, synthetic
blends of compounds mimicking host odours, or non-host related chemical compounds that
have been found to be attractive to the vectors. Harris (1938) and Van Thiel (1935) were among
the first to report on the use of host odours to trap tsetse flies and mosquitoes, respectively. It
was also established that carbon dioxide (CO2) acts as a universal attractant for many blood-
feeding insects, and many traps are baited with this compound. For example, the Centre for
Disease Control miniature light trap (CDC light trap) (Sudia and Chamberlain 1962) was reported
to produce significantly-enhanced catches of mosquitoes when baited with dry ice (Newhouse et
al. 1966). The CO2-baited CDC trap was the gold standard for establishing mosquito densities for
many years. In the tropics, CO2 was replaced by natural human emanations by operating the CDC
trap next to a bed net under which a human spent the night (Garrett-Jones and Magayuka 1975).
Although these odour-baited traps collected disease vectors, the trap collections represented only
crude data on vector densities, and more accurate measurements were needed.

In the 1970s, Vale (1974), dissatisfied with the variable results of odour-baited traps, began to
experiment with natural host odours to investigate the behaviour of tsetse flies in Zimbabwe.
By using an elegant design of electrified grids and artificial host models, Vale demonstrated that
natural host odours were the principle cues that led tsetse flies to the vicinity of their blood hosts
(Vale 1993). The studies on tsetse olfactory behaviour are still considered to be classic methods
for investigating insect behaviour with respect to their vertebrate hosts. Gillies and Wilkes (1968)
demonstrated that mosquitoes could also be manipulated with host odours, and it was soon found
that synthetic blends of odorants could be used to manipulate mosquito populations (Takken and
Kline 1989). These behavioural studies were gradually augmented with physiological studies to
reveal the sensory processes that made the insects respond to the odorants (Cork 1996, Davis and
Sokolove 1976). The advent of molecular genetics has allowed for detailed insight in the genetics
and molecular regulation of insect olfaction, which provides fundamental knowledge on the
regulation of insect behaviour (Jacquin Joly and Merlin 2004, Lessing and Carlson 1999, Touhara
and Vosshall 2009). This, in turn is being used for the identification of chemical compounds that
affect vector-host interactions, so that the target vectors can be more effectively lured to sampling
devices and/or removed from the environment as a tool for disease control.

In future, odorants will increasingly be used for the monitoring of vector-borne disease
transmission risk as well as for vector control. It is therefore appropriate that the current volume
of the Ecology and Control of Vector-borne Diseases has brought together reviews of the state-
of-the-art of this topic, from the molecular regulation of olfactory senses (Chapter 2) to field
applications of odorants for disease control (Chapter 17). As many disease vectors belong to the
insect order of Diptera, this group of insects is discussed in detail, with mosquitoes being the
example for detailed explanation of olfactory regulation. Chapters on blood-feeding bugs, vectors
of Chagas disease, and blood-feeding ticks, vectors of a variety of parasites and pathogens, are

14  Olfaction in vector-host interactions

 1. Strategic use of chemical ecology for vector-borne disease control

included as well. Finally, as it is increasingly becoming apparent that the olfactory behaviour of
insects and ticks is often mediated by parasites, a chapter on the latest knowledge of this field
of science is also included. Odorants are not only used to attract insects and ticks, but they can
also be used to protect blood hosts from being attacked (Katz et al. 2008). In many cases odorants
act as deterrents or repellents, and where applicable, this aspect of olfaction is discussed as well.

The ecology of vector-borne diseases is much determined by the association between vectors,
parasites and hosts, which is in turn strongly mediated by chemical cues. The chapters in this
volume demonstrate that detailed studies on insect olfaction contribute to our understanding of
the individual components regulating this process, and contribute to finding effective solutions
for disease control.

References

Barclay E (2008) Is climate change affecting dengue in the Americas? The Lancet 371: 973-974.
Carpenter S, Wilson A and Mellor PS (2009) Culicoides and the emergence of bluetongue virus in northern Europe.
Trends Microb 17: 172-178.
Cork A (1996) Olfactory basis of host location by mosquitoes and other haematophagous Diptera. In: Bock GR and
Cardew G (eds) Olfaction in mosquito host interactions. CIBA Foundation Symposium 200. Wiley, Chichester, UK,
pp 71-88.
Davis EE and Sokolove PG (1976) Lactic acid-sensitive receptors on the antennae of the mosquito, Aedes aegypti. J
Comp Physiol 105: 43-54.
Epstein PR (2001) Climate change and emerging infectious diseases: review. Microbes and Infection 3: 747-754.
Erin Staples J, Breiman RF and Powers AM (2009) Chikungunya fever: an epidemiological review of a re-emerging
infectious disease. Clinical Inf Dis 49: 942-948.
Garrett-Jones C and Magayuka SA (1975) Studies on the natural incidence of Plasmodium and Wuchereria infections
on Anopheles in rural East Africa: I-assessment of densities by trapping hungry female Anopheles gambiae Giles,
Species A. WHO/VBC 75.145 Geneva Switzerland.
Gillies MT and Wilkes TJ (1968) A comparison of the range of attraction of animal baits and of carbon dioxide for some
West African mosquitoes. Bull Entomol Res 59: 441-456.
Gould EA and Higgs S (2009) Impact of climate change and other factors on emerging arbovirus diseases. Trans Royal
Soc Trop Med and Hygiene 103: 109-121.
Harris RHTP (1938) The control and possible extermination of the tsetse by trapping. Acta Conventus Tertii de Tropicis
Atque Malariae Morbis 1: 663-677.
Hartemink NA, Randolph SE, Davis SA and Heesterbeek JA (2008) The basic reproduction number for complex disease
systems: defining R(0) for tick-borne infections. American Naturalist 171: 743-754.
Hay SI, Guerra CA, Tatem AJ, Noor AM and Snow RW (2004) The global distribution and population at risk of malaria:
past, present, and future. Lancet Infectious Diseases 4: 327-336.
Jacquin Joly E and Merlin C (2004) Insect olfactory receptors: contributions of molecular biology to chemical ecology.
J Chem Ecology 30: 2359-2397.
Jones KE, Patel NG, Levy MA, Storeygard A, Balk D, Gittleman JL and Daszak P (2008) Global trends in emerging infectious
diseases. Nature 451: 990-993.
Katz TM, Miller JH and Hebert AA (2008) Insect repellents: historical perspectives and new developments. J Am Acad
Derm 58: 865-871.
Lessing D and Carlson JR (1999) Chemosensory behaviour: the path from stimulus to response. Curr Opinion Neurobiol
9: 766-771.
Mathers CD, Ezzati M and Lopez AD (2007) Measuring the burden of neglected tropical diseases: the global burden of
disease framework. PLoS Negl Trop Dis 1: e114.

Olfaction in vector-host interactions 15

Willem Takken and Bart G.J. Knols

Newhouse VF, Chamberlain RW, Johnston JG and Sudia WD (1966) Use of dry ice to increase mosquito catches of the
CDC miniature light trap. Mosquito News 26: 30-35.
Qiu YT, Spitzen J, Smallegange RC and Knols BGJ (2007) Monitoring systems for adult insect pests and disease vectors.
In: Takken W and Knols BGJ (eds) Ecology and control of vector-borne diseases, Vol. 1: emerging pests and vector-
borne diseases in Europe. Wageningen Academic Publishers, Wageningen, the Netherlands, pp 329-353.
Service MW (1993) Mosquito ecology - field sampling methods. second edn. Elsevier Applied Science, London, UK.
Smith DL, McKenzie FE, Snow RW and Hay SI (2007) Revisiting the basic reproductive number for malaria and its
implications for malaria control. Plos Biology 5: e42.
Sudia WD and Chamberlain RW (1962) Battery-operated light trap, an improved model. Mosquito News 22: 126-129.
Takken W and Kline DL (1989) Carbon dioxide and 1-octen-3-ol as mosquito attractants. J Am Mosquito Control Ass
5: 311-316.
Touhara K and Vosshall LB (2009) Sensing odorants and pheromones with chemosensory receptors. Annual Rev Physiol
71: 307-332.
Vale GA (1974) New field methods for studying the responses of tsetse flies (Diptera, Glossinidae) to hosts. Bull Ent Res
64: 199-208.
Vale GA (1993) Development of baits for Tsetse flies (Diptera: Glossinidae) in Zimbabwe. J Med Ent 30: 831-842.
Van Thiel PH (1935) Onderzoekingen omtrent den gedrag van Anopheles ten opzichte van mensch en dier, mede in
verband met rassenstudie bij Anopheles maculipennis (Studies on the behaviour of Anopheles around humans and
animals, associated with a species-specific study of Anopheles maculipennis). Gen Tijdschr Ned Indië 75: 2101-2181.

16  Olfaction in vector-host interactions

2. Molecular regulation of olfaction in mosquitoes
Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

Abstract

Recent advances in our understanding of the molecular mechanisms of mosquito olfaction have
opened opportunities to approach behavioural disruption strategies of disease vectors from a
new vantage point. The genome sequence projects of Anopheles gambiae, Aedes aegypti and
Drosophila melanogaster have allowed the identification of complete gene families that are
involved in odorant sensing and signal regulation. Odorant receptor proteins, in particular, have
been shown to be directly involved in olfactory transduction. The function of these receptors in
various head appendages of the larval and adult mosquitoes has also raised new fundamental
questions pertaining to the nature and mechanisms of olfaction in these insects.

Keywords: gene, g-protein coupled receptor (gpcr), labellum, maxillary palp, odorant-binding
protein (obp), odorant receptor (or), signal transduction

Introduction

Eighty years have elapsed from the initial published (Rudolfs 1922) evidence of olfactory-driven
behaviour of mosquitoes to the most recent discovery and characterisation of several olfactory
gene families in Anopheles gambiae Giles (Fox et al. 2001, Hill et al. 2002), Aedes aegypti (L.) (Bohbot
et al. 2007, Nene et al. 2007) and Culex pipiens quinquefasciatus Say (unpublished at the time of
writing). Emphasis on these three mosquito species derives from their roles as vectors of the
pathogens responsible for several critical human infectious diseases including malaria, West Nile
disease, dengue fever, yellow fever, Rift Valley fever, chikungunya disease, leishmaniasis, filiariasis,
various arboviral encephalitis and others. In these cases the mosquito’s vectorial capacity is not
only regulated by internal factors (physiological state, see Chapter 2 in this volume), but is also
mediated by sensory inputs derived from heat, visual and most importantly, olfactory stimuli
(Figure 1).

Adult anophelines are attracted to odorant stimuli that range from single compounds to complex
odour blends (reviewed in Knols 1996). For example, Ae. aegypti is attracted by a mixture of CO2,
ammonia, L-lactic acid and fatty acids (Bosch et al. 2000, Geier et al. 1999, Kellogg 1970, Steib et
al. 2001) while An. gambiae is attracted by CO2, lactic acid, acetone and 1-octen-3-ol (Dekker et
al. 2002, Healy and Copland 1995, 2000, Smallegange et al. 2005, Takken et al. 1997). In nature,
mosquitoes rely principally on olfaction to locate and choose oviposition (Takken and Knols 1999),
nectar-feeding sites (Foster 1995) and mates (Cabrera and Jaffe 2007, Nijhout and Craig 1971) to
name but a few critical life-cycle behaviours.

The availability of the complete Ae. aegypti (Nene et al. 2007) and An. gambiae (Holt et al. 2002)
genomes has facilitated the identification and annotation of a handful of gene families, which are
potentially involved in the regulation of olfactory behaviours. This effort was further enhanced
by the pioneering studies undertaken to similarly analyse the Drosophila melanogaster Meigen
genome (Adams et al. 2000). Indeed, comparative genomics using these three mosquito species
has provided insights into the evolution of many gene families and the possible conserved
function shared by homologous genes. Functional characterisation of olfactory mechanisms in
mosquitoes is a more recent development and combines information from electrophysiological

Olfaction in vector-host interactions 17

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

CO2 Odorants
Upwind flight orientation Nectar

Mate?

Appetitive/Nest Activation
Attraction
searching Orientation
Nutritional homeostasis Olfaction Olfaction Host
Gonotrophic status Vision Vision
Circadian rythms Gustation
Age, size Mechanoreception
Mating condition Heat sensing

Oviposition

Figure 1. The physiological and sensory basis for mosquito behaviour. Foraging, lekking, preying and
nesting behaviours are modulated by internal and external signals. The reproductive, sexual, nutritional and
developmental states and temporal factors including the circadian clock all influence the mosquito’s responses
to external stimuli, which rely on a combination of sensory modalities.

studies as well as advances in the use of heterologous expression systems including cell-based
assays and the use of Drosophila as a transgenic platform for the study of mosquito genes.

The sensory modality of olfaction is a form of chemoreception specifically dealing with volatile
stimuli and essentially represents the interface between the organism and its immediate
environment. It requires the activity of highly complex peripheral signal transduction cascades
to transform air-borne chemical signals into precise patterns of neuronal activity. At the cellular
level in insects this process is focused on a collection of olfactory receptor neurons (ORNs) that are
encased in small, perforated cuticular structures called sensilla and are distributed on the surface
of various head appendages (Figure 2). The olfactory signal transduction events take place on the
ORN dendrites and their underlying mechanism remains largely uncharacterised and controversial
despite intensive research efforts.

This chapter will review the current knowledge of the molecular components implicated in
insect and especially mosquito olfactory transduction. It will attempt to provide the most recent
theory of olfactory transduction based on current experimental data focusing on two mosquito
species, An. gambiae and Ae. aegypti, and from Drosophila. While Drosophila remains the most
robust genetic model for the study of insect olfaction, significant progress has been made toward
the identification and functional characterisation of mosquito olfactory genes. In the following
paragraphs, we will look at the contribution of genome sequencing projects to the identification
and phylogenetic analyses of olfactory proteins. Rather than merely cataloguing the olfactory

18  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

A. Olfactory tissues of the mosquito B. Olfactory sensillum

Air Sensillum lymph

Cuticle
Lymph
Dendrite
Pore
Pore
Antenna

Accessory
Maxillary palp ORNcells

Labellum Axons
ORN

C. Current model of olfactory transduction

Odorant-gated ion channel model

Odorant

OBP
Ionotropic
ODE pathway
Cations

Extracellular AC

ORx OR7

Intracellular SNMP
Gp

Metabotropic
pathway

cAMP

Figure 2. Olfactory transduction theory in mosquitoes. (A) The antenna, the maxillary palp and the labellum of
mosquitoes are the three sites of olfactory sensing. (B) Sensory organs called sensilla cover these tissues and
house Odorant Receptor Neurons (ORNs), which are dedicated to convert odorant signals into electrical outputs.
Several molecular players have been involved in this transduction pathway. (C) Odorant molecules access the
sensillum lymph through pores in the cuticles and are shuttled to a membrane protein complex constituted of a
Sensory Neuron Membrane Protein (SNMP) and an Odorant Receptor (OR) heteromer (ORx/OR7). In this model,
OR activation triggers two responses. The fast response involves a direct effect on the receptors, which act as an
odorant-gated ion channel. A slower response is mediated via a G protein (Gp) dependent secondary messenger
pathway including Adenyl Cyclase (AC) and cyclic AMP (cAMP) (Sato et al. 2008, Wicher et al. 2008).

Olfaction in vector-host interactions 19

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

gene families and their hypothetical operative machinery, we will attempt to provide a biological
context for these olfactory components as derived from functional data as well as studies that
have elucidated the distribution of olfactory genes in various mosquito olfactory tissues and
developmental stages.

Mosquito olfactory transduction: a road map under construction

Olfaction is a key sensory modality that modulates host-seeking and other behaviours in
mosquitoes (Figure 1). Over two decades of research on invertebrate olfaction have identified
several molecular components that may safely be stated to be involved in the biochemical
processes responsible for the detection and transduction of odorant stimuli. These include
arrestins (Arrs), G-proteins, odorant-binding proteins (OBPs), odorant degrading enzymes (ODEs),
odorant receptors (ORs) and sensory neuron membrane proteins (SNMPs).

While the structures of insect ORs have not been precisely determined, there is little doubt that
they function as a heteromeric complex (Benton et al. 2006) and play a central role in odorant
detection and the activation of downstream transduction machinery. Insect ORNs generally
express a conventional OR conferring its odorant-binding specificity together with a highly
conserved non-conventional OR belonging to the OR83b sub-family of proteins (Jones et al. 2005,
Krieger et al. 2003, Larsson et al. 2004, Pitts et al. 2004). Pioneering experiments in heterologous
expression systems have demonstrated that heterodimeric Drosophila ORs (DORs) elicit cellular
responses following odorant stimulation (Neuhaus et al. 2005, Wetzel et al. 2001), effectively
defining their function as bona fide ORs. This functionality has been repeatedly validated with An.
gambiae (Ditzen et al. 2008, Hallem et al. 2004a, Kwon et al. 2006, Lu et al. 2007, Sato et al. 2008,
Xia et al. 2008), Bombyx mori Linnaeus (Nakagawa et al. 2005, Syed et al. 2006), Heliothis virescens
(Fabricius) (Grosse-Wilde et al. 2007) and Apis mellifera Linnaeus ORs (Wanner et al. 2007). Not
surprisingly, these data indicate that the expression of a given Or gene in insect ORNs determines
their functional specificity (Dobritsa et al. 2003, Hallem et al. 2004a, Hallem et al. 2004b, Jones et
al. 2005).

Whereas the functional role of insect ORs is well established, the molecular events preceding
and following their activation remain decidedly unclear although our understanding of the
processes has dramatically improved of late. Extracellular events associated with odorant
detection involve soluble OBPs and ODEs (Figure 2), and SNMPs are thought to establish a
molecular bridge specifically between ORs and pheromone ligands. While the precise biochemical
processes leading to the activation of downstream molecular targets are not expressly defined,
recent reports suggest that insect ORs exhibit either ionotropic or alternatively both iono- and
metabotropic characteristics. On the one hand, the rapid response kinetics of odorant receptors
observed in heterologous expression systems are consistent with the behaviour of ligand-gated
ion channels that act independently of G-protein signaling (Sato et al. 2008). In contrast, another
study indicates an additional metabotropic response requiring the recruitment of G-protein-
mediated second-messenger pathways that is delayed and slower relative to the initial ionotropic
response (Wicher et al. 2008). Although the question remains open, the incorporation of G-protein
dependent pathways that presumably require the synthesis of second-messenger intermediates is
also consistent with numerous studies linking ORs to GPCR signalling pathways (Kain et al. 2008,
Kalidas and Smith 2002, Woodard et al. 1992).

20  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

Structural and evolutionary aspects of the mosquito olfactory gene families

Odorant-binding and chemosensory proteins

Two classes of soluble proteins, the OBPs and chemosensory proteins (CSPs), have been found in
the lumen of chemosensory sensilla of insects. Odorant-binding proteins were first discovered
in 1981 in the silk moth Antheraea polyphemus (Cramer) (Vogt and Riddiford 1981) and in the
cow (Pelosi et al. 1982). While sharing the same name, mammalian OBPs are characterised as
lipocalin family proteins (Flower et al. 2000), and insect OBPs are not phylogenetically related and,
not surprisingly, have entirely different structures. Classic insect Obp genes encode hexa-helical
proteins (Kruse et al. 2003, Wogulis et al. 2006) whereas mammalian lipocalins exhibit a β-barrel
motif (Bianchet et al. 1996). CSPs were first identified in Drosophila (McKenna et al. 1994). And both
OBPs and CSPs have been identified in a plethora of insects’ genera of both heterometabolous
and holometabolous lineages and are presumed to perform similar functions in odour detection.
Odorant-binding protein and Csp genes encode small, globular hydrophilic proteins (14-20 kDa and
11kDa, respectively) that are excreted in significant amounts out of the cell into the peri-receptor
space. As is the case for so many other components of olfactory signalling pathways, the role of
these proteins remains very much a matter of debate. Nevertheless, the premise they do play
important and perhaps essential roles is supported by numerous studies in all systems studied
to date that demonstrate very high OBP and CSP expression levels in olfactory tissues, as well as
a more limited set of in vitro binding assays (Pelosi et al. 2006, Wanner et al. 2004). Interestingly,
several members of both OBP and CSP families are expressed outside chemosensory tissues and
are therefore likely involved in other processes. Indeed, current evidence suggest that CSPs are
not involved in olfactory processes even though they have been associated with olfactory tissues
(Sabatier et al. 2003). Accordingly, without questioning their relevance, we will not discuss the
potential olfactory involvement of CSPs in this context but rather will focus on OBPs where there
are compelling data insofar as Drosophila pheromone detection is concerned.

Multiple OR functional studies in cell-based assays that are entirely devoid of OBPs have shown
that OBPs are not strictly required for OR activation in these heterologous systems (Neuhaus et al.
2005, Wetzel et al. 2001), although the addition of OBPs to these systems enhances the response
sensitivity to semiochemicals (Matsuo et al. 2007, Syed et al. 2006, Wang et al. 2007). The strongest
evidence for an in vivo olfactory function of OBPs is typified by the studies on the LUSH-OBP/
OBP76a paradigm in Drosophila (Smith 2007). What is the role of LUSH and to what extent its mode
of action represents the OBP family? The LUSH protein is present in the sensillum lymph of the
pheromone sensitive T1 sensilla located on the Drosophila antenna. Elegant physiological (Xu et
al. 2005) and more recent structure-function analyses (Laughlin et al. 2008) have demonstrated
that LUSH itself is the ligand for the pheromone sensitive neurons and that the cVA pheromone,
mediates the conversion of LUSH into its active conformation thereby interacting with SNMP1
and DOR67d. Several amino-acid residues in the LUSH protein have been implicated in this
conformational change and are conserved among other insect LUSH homologs. It is important to
note that other unknown factors are implicated in this process as all three proteins cited above
are not sufficient to mimic the in vivo observations. Importantly, it is likely that this, and perhaps
all pheromone detection pathways represents a distinct paradigm for odorant recognition that
reflects a requirement for extreme sensitivity.

The release of the complete genomes of Drosophila (Adams et al. 2000), An. gambiae (Holt et al.
2002) and Ae. aegypti (Nene et al. 2007) has allowed a full accounting of the Obp gene repertoires
in each of these species. The Drosophila genome carries 49 Obp genes while Ae. aegypti and An.

Olfaction in vector-host interactions 21

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

gambiae possess 57 and 66 putative Obp members, respectively (Figure 3). Among individual
species, OBPs share little overall amino-acid sequence identity, ranging from 13.3%, 16.6% to 21.7%
in Drosophila, An. gambiae and Ae. aegypti, respectively (Zhou et al. 2008). Of the three Dipteran
species, the overall amino-acid identity among the classic 6-cysteine-bearing OBPs is less than 5%.
It has been proposed that this elevated number and sequence diversity of OBPs is an adaptation to
the variety of chemical cues that this system needs to interact with (Mohl et al. 2002).

The Dipteran OBP family consists of several distinct subgroups based on peptide length and more
importantly based on the number of cysteine residues and their positions within the peptide
primary structure. This arbitrary distinction, however, does not hold phylogenetic scrutiny (Pelosi
et al. 2006). Historically, classic OBPs have been characterised by a 6-cysteine motif to which all
other OBPs are gauged. Additional groups displaying a variation on the OBP archetype include
the ‘Plus-C’ (8 cysteine residues), ‘Minus-C’ (4 cysteine residues) and ‘Atypical’ OBPs (elongated
C-terminus). Interestingly, members of the atypical group have so far only been found in
mosquitoes (Xu et al. 2003, Zhou et al. 2008). Their absence in the Drosophila (Xu et al. 2003) and
honeybee (Foret et al. 2007) genomes suggests that this gene subfamily arose sometime after
the fly/mosquito split approximately 250 million years ago (Gaunt and Miles 2002, Zdobnov et
al. 2002). Moreover, at this time there is no hard evidence for the involvement of atypical OBPs in
olfactory processes.

Molecular phylogenetic analyses underscore the high degree of sequence divergence among
these Dipteran species including several instances of species-specific gene expansions (Bohbot
and Vogt 2005, Zhou et al. 2008). By contrast, the genomes of Ae. aegypti, An. gambiae and
Drosophila possess Obp gene members belonging to the OS-E/OS-F, LUSH and OBP19a groups
suggesting that some level of functional conservation applies among these three species (Zhou
et al. 2008). Obp genes tend to occur in clusters within insect genomes (Hekmat-Scafe et al. 2002,
Xu et al. 2003) and within each cluster exhibit similar exon-intron structures (Zhou et al. 2006).
Apart from the intron-less members of the atypical sub-class, Obp genes contain 1 to 4 introns
(Vogt 2002) including one usually found immediately downstream of the sequence encoding the
peptide secretion signal-sequence (Hekmat-Scafe et al. 2002, Xu et al. 2003, Zhou et al. 2008).

OBP OR
Classic +C Atypical Total
Ae. aegypti 34 17 15 66 131
38 Mya
140 Mya Cx. pipiens

Fly/mosquito split An. gambiae 33 19 14 66 79

140-200 Mya

D. melanogaster 37 12 - 49 62

Figure 3. Odorant-Binding Proteins (OBPs) and Odorant Receptors (ORs) in the Dipteran genomes (Hekmat-Scafe
et al. 2002, Wicher et al. 2008, Xu et al. 2003, Zhou et al. 2008).

22  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

The combination of structural and genomic conservation of these genes, particularly between
Ae. aegypti and An. gambiae, and their amino-acid sequence diversity suggest rapid evolutionary
dynamics driven by gene duplication events. While several studies have shown the binding
capability of these proteins in vitro, and despite their presence in the peri-receptor compartment
of chemosensory sensilla, direct functional evidence for a precise role(s) in olfactory signalling
remains elusive.

Sensory neuron membrane proteins

The SNMP was first described as a putative pheromone binding membrane protein in the antennae
of the silk moth Antheraea polyphemus (Rogers et al. 1997, Vogt et al. 1988). Since these initial
observations, highly conserved SNMPs have been identified in several insect families (Nichols
and Vogt 2008). SNMPs are transmembrane receptors, and are directly related to the CD36 family
fatty acid transporters/scavenger receptors which have been shown to bind a number of ligands
in various biochemical pathways including lipid metabolism, homeostasis and others (Rac et al.
2007). Two closely related Snmp genes (Snmp1 and Snmp2) are found in Diptera (Nichols and Vogt
2008) and twelve to fourteen CD36-like genes have been identified in the genome of Drosophila,
Ae. aegypti and An. gambiae.

In Drosophila, SNMP has recently been shown to be involved in pheromone detection

(Benton et al. 2007). In this case, Snmp1 is expressed in trichoid sensilla and is required for the
electrophysiological response of DOr67d-expressing ORNs to the male pheromone cis-vaccenyl
acetate. Furthermore, protein-fragment complementation assays indicate that SNMP1 and
DOR83b are physically in close proximity on the surface of the cell membrane (Benton et al. 2007).
The role of Snmp1 in pheromone sensitivity extends to Lepidoptera where HR13 (Grosse-Wilde et
al. 2007), a pheromone receptor from Heliothis virescens, requires SNMP1 for function, suggesting
that SNMPs may be generally involved in the recognition of the hydrocarbon chain common to
the pheromones of both species (Benton et al. 2007). Lastly, Drosophila Snmp1 is also expressed
in the support cells surrounding pheromone-sensitive ORNs in several chemosensory sensilla
distributed across a variety of appendages where their biological role has yet to be determined
(Benton et al. 2007, Forstner et al. 2008, Nichols and Vogt 2008).

Odorant receptors

From a number of perspectives, ORs remain the central molecular players among olfactory proteins.
Not surprisingly, the first identified insect ORs were found in Drosophila (Clyne et al. 1999, Gao and
Chess 1999, Vosshall et al. 1999). Functional validation of the role of DORs as bona fide ORs came
some time later through extensive use of numerous heterologous expression systems (Dobritsa
et al. 2003, Goldman et al. 2005, Hallem et al. 2004b, Kreher et al. 2005, Wetzel et al. 2001). As was
the case with the majority of gene families, these initial studies in Drosophila opened the door for
similar studies in disease vector insects. Bioinformatics and molecular-based analyses of mosquito
genomes have provided the identification of complete OR repertoire sets in An. gambiae (AgORs)
(Hill et al. 2002) and more recently, in Ae. aegypti (AaORs) (Bohbot et al. 2007). In contrast to the
60 DOr genes, there are 79 (Hill et al. 2002) and 131 (Bohbot et al. 2007) Or genes in An. gambiae
and Ae. aegypti, respectively (Figure 3).

Traditionally classified as G-protein coupled receptors (GPCRs), insect ORs, Aa/AgORs belong
to a novel membrane protein family with no homology with vertebrate ORs. In addition to
their characteristic lack of primary sequence conservation, insect OR proteins are recognised

Olfaction in vector-host interactions 23

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

as functionally obligate heterodimers composed of a conventional and highly divergent OR

and a highly conserved non-conventional co-receptor called OR83b in Drosophila and OR7 in
mosquitoes (~80% amino-acid identity). In contrast, conventional ORs from flies and mosquitoes
hardly share any significant sequence similarity outside relatively conserved ORs such as the
DOR43a/OR2 sub-group (Figure 4). The similarity between the universal co-receptor DOR83b/
OR7 group is a reflection of their conserved function as chaperones involved in protein trafficking
of conventional ORs and their participation in odorant detection (Benton et al. 2006, Jones et
al. 2005, Larsson et al. 2004, Neuhaus et al. 2005). Interestingly, DOR43a and the mosquito OR2
homologs do not recognise the same odorants (Hallem et al. 2004a,b, Storkuhl and Kettler 2001,
Wetzel et al. 2001). Similarly to OBPs (Vogt 2002), Or genes tend to occur in clusters. This feature
is more pronounced in the mosquito genomes (Bohbot et al. 2007, Hill et al. 2002) relative to
that observed in Drosophila (Robertson et al. 2003). The former exhibit gene clusters of 2 to 11
genes while DOrs occur for the most part as scattered singletons and in a few instances in clusters
of up to 3 genes (Robertson et al. 2003). Genes within a given cluster typically display higher
sequence similarity. The evolutionary history leading to these accrued gene duplication events in
mosquitoes is unknown but indicates perhaps that these insects have relied on this mechanism
to drive OR phenotypic diversity thereby increasing their odour space. The comparative molecular
phylogenetic analyses of Aa/AgORs have confirmed the characteristic divergence of these ORs
(generally less than 20% overall amino-acid sequence identity). Within each repertoire are
several OR-specific expansions that likely reflect the evolutionary divergence as well as the
unique chemical ecology of these two mosquito species. It is remarkable that 11 homologous
Aa/AgOR groups can be nevertheless distinguished based on sequence similarity (Figure 4) and
microsyntenic relationships (Bohbot et al. 2007), suggesting that at least in some cases, significant
evolutionary constrains have been imposed upon this gene family.

The olfactory apparatus of vector mosquitoes

The olfactory apparatus of vector mosquitoes is comprised of ORNs and support cells that are
spread over several adult and larval appendages that together provide an array of sensory
capabilities that presumably reflects biological requirements (Figure 2). One method of marking
ORNs is to assess the expression of Or genes. In that context a complete Or expression map of the
larval and adult olfactory appendages is available for Ae. aegypti (Bohbot et al. 2007). The larval
antennae and all three adult olfactory appendages in mosquitoes, the antenna, the proboscis,
and the maxillary palp, are populated by a diversity of porous sensilla containing the sensory
dendrites of ORNs (Lu et al. 2007, Pitts et al. 2004, Pitts and Zwiebel 2006, Xia et al. 2008). In An.
gambiae detailed transcription profiles of the proboscis, the maxillary palp and the larval antenna
have been established (Kwon et al. 2006, Lu et al. 2007, Xia et al. 2008), while the precise expression
profile of olfactory genes in the adult antenna has yet to be fully elucidated.

Olfactory properties of the mosquito antennae

Paradoxically, and in large part as a result of its complexity, the mosquito antennae, which is
the principal adult olfactory appendage, remains the least understood in so far as the molecular
bases of olfactory physiology. That said, a significant body of physiological work has been
completed in Ae. aegypti (Davis 1974) and other vector mosquitoes as reviewed in Davis (1996).
More recently several important studies have begun to unravel the olfactory physiology of An.
gambiae antennae (Meijerink et al. 2001, Meijerink and van Loon 1999, Qiu et al. 2006). At the
same time there is a growing appreciation of the antennal expression patterns of several genes
that encode signal transduction components that have been linked in one degree or another to

24  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

INTERspecific comparison

Specific ORs High identity homologous ORs

DOR83b
AaOR7
AgOR7
AaOR10
Ae. aegypti AgOR10
AaOR9
AaOR2
AgOR2
DOR43a
AaOR8
AgOR8
AgOR11
AaOR11
AaOR12
AaOR6
An. gambiae AgOR6
AaOR23
AgOR23
AaOR40
AgOR40
AaOR31
AgOR31
AaOR22
AgOR22
AgOR67
AgOR66
AaOR66
D. melanogaster AgOR41
AaOR41
AaOR42
AaOR69
AgOR69

Figure 4. Phylogenetic comparison of Dipteran odorant receptor proteins (ORs). ORs are highly divergent at the
protein level, the vast majority being species specific. A small subset of ORs display various levels of sequence
conservation between Aedes and Anopheles mosquitoes. The mosquito OR7 and Drosophila OR83b are true
orthologs and are defined as universal ORs (~80% amino-acid identity). The other similarity between mosquitoes
and flies is restricted to the mosquito OR2 and DOR43a (~36% identity) (Bohbot et al. 2007, Hill et al. 2002).

olfaction. As discussed above, these include antennal arrestins (Merrill et al. 2003), G-proteins
(Rutzler et al. 2006), OBPs (Biessmann et al. 2002, Xu et al. 2003) as well as an ever- increasing
spectrum of ORs. These include representatives from An. gambiae (Fox et al. 2001, Hill et al. 2002),
Ae. aegypti (Bohbot et al. 2007) and Cx. quinquefaciatus (Xia and Zwiebel 2006). Of these, at least
one has been linked to human odorants (Hallem et al. 2004a) while several have been reported
to show sexually biased expression patterns (Fox et al. 2001, Iatrou and Biessmann 2008). Current
efforts in the field are doubtlessly focused on providing a detailed understanding of antennal
odour coding mechanisms.

Olfactory properties of the mosquito labellum

The proboscis is a long slender organ housing the feeding stylets of the adult mosquito, harboring
numerous chemosensory sensilla at its bulbous tip known as the labellum (Figure 5). Early studies
on the mosquito labellum have focused largely on its gustatory function. In Culiseta inornata
(Williston) three morphological and functional types of labellar sensilla have been defined (McIver
1982, Pappas and Larsen 1976); the Type 1 (T1) and the Type 2 (T2) sensilla are distributed on the
external surface of the labellum, while the Type 3 (T3) sensilla are located on the oral surface – all

Olfaction in vector-host interactions 25

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

three types are sensitive to taste stimuli (McIver 1982, Owen et al. 1974, Pappas and Larsen 1976).
Recent electrophysiological studies, however, demonstrated that the labellum of An. gambiae is
also sensitive to odorant stimuli including butylamine and several aliphatic carboxylic acids (Kwon
et al. 2006). Here at least one sensillum that is reminiscent of the Cs. inornata T2 sensilla harbours
at least two ORNs and responds to butylamine, aliphatic acids and several ketone compounds
including acetophenone (Kwon et al. 2006). The presence of this cryptic olfactory capacity in the
primary gustatory organ of An. gambiae is especially interesting inasmuch as butylamine has
been identified in human skin emanations (Ellin et al. 1974). Furthermore, the odorants that elicit
responses from labellar sensilla have relatively low volatility compared to other known mosquito
attractants, raising the suggestion that the labellum of An. gambiae might be involved in the
close-range odour detection at the penultimate stages of host-seeking or oviposition behaviour
(Kwon et al. 2006).

In a decided departure from the Drosophila labellum which is devoid of olfactory function and
putative OR expression, immuno-histochemical studies revealed that the non-conventional and
obligate co-receptor AgOR7 is expressed in about 50 labellar T2 sensilla of An. gambiae (Pitts et

A AL
Butylamine
OL Aliphatic acids
Ketones

A
D
SOG
V
P
B
P
SSR
A

T2 sensillum site

OR6 OR7

Hair

Proboscis

Labellum

Figure 5. The labellum of An. gambiae exhibits olfactory responses. Schematic dorsal view of the mosquito head
and brain, anterior (A), posterior (P), dorsal (D) and ventral (V) axes, as indicated. (A) Neurons of the labellum
project to a posteromedial region of the antennal lobes (AL) in the brain. (B) A subset of T2 sensilla on the labellum
respond to odorant stimuli in single sensillum recording (SSR) experiments. Coincidently, T2 sensilla express OR6
and OR7. SOG: sub-oesophageal ganglion, OL: optic lobe.

26  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

al. 2004). Similarly AaOR7 and CqOR7 staining is localised to the labella of Ae. aegypti and Cx.
quinquefasciatus (Melo et al. 2004, Xia and Zwiebel 2006), respectively. Overall, more than 20
conventional AgOrs are expressed on the labellum of An. gambiae, and in situ hybridisation studies
have thus far demonstrated that at least one of these AgOrs, AgOr6, is co-expressed with AgOR7
in a subset of labellar ORNs (Kwon et al. 2006). It is worthwhile to note that AgOR6, which has
been associated with a T2 sensillum that respond in vivo to butylamine, aliphatic acids and several
ketone compounds (Kwon et al. 2006), has recently been shown to specifically elicit responses
to precisely those compounds when expressed in the Xenopus oocyte heterologous expression
system (Xia et al. 2008).

Surprisingly, only one or two glomeruli in the antennal lobe receive projections from the labellar
ORNs of An. gambiae and Ae. aegypti (Ghaninia et al. 2007, Kwon et al. 2006). These glomeruli are
located in the medial region of the antennal lobe and might be innervated by both labellar and
palpal ORNs (Ghaninia et al. 2007, Kwon et al. 2006). It is possible that the An. gambiae labellum
harbours multiple functional types of ORNs, each expressing a different AgOR in addition to
AgOR7, and this ensemble of ORNs expressing a combinatorial of over 20 AgORs converge on
one or two medial glomeruli in the antennal lobe. This type of organisation represent a significant
departure from established paradigms in the olfactory or gustatory system in insects where ORNs
or gustatory receptor neurons (GRNs) with different receptor profiles will converge onto distinct
regions of the antennal lobe or the suboesophageal ganglion (Rutzler and Zwiebel 2005, Vosshall
and Stocker 2007).

Olfactory properties of the mosquito maxillary palps

In contrast to the antenna, which contains the largest quantity and variety of olfactory sensilla,
the maxillary palp is considerably less complex, harboring a single type of chemosensory
sensillum, the capitate peg (Figure 6). The maxillary palp of mosquitoes consists of five segments,
although female anophelines have much longer palps than those found in culicines, where the 5th
segment is reduced to a knob. Furthermore, whereas in adult female anophelines, capitate pegs
are distributed on palpal segments two, three and four, their distribution is restricted to the 4th
segment in female culicines (McIver 1972).

Extensive single-sensillum electrophysiological recordings (SSRs) have been carried out in

female An. gambiae, Ae. aegypti and Cx. quinquefasciatus (Lu et al. 2007, Syed et al. 2006). These
experiments invariably revealed the existence of a single functional class of capitate pegs across
the maxillary palp (Lu et al. 2007, Syed et al. 2006), although the number of capitate pegs per
palp varies drastically among the three species (McIver 1982). Furthermore, these structures
consistently contain three ORNs that form stereotyped triads within each capitate peg sensillum
(Lu et al. 2007, Syed et al. 2006). The capitate peg (cp) A and B neurons respond with high sensitivity
to CO2 and 1-octen-3-ol, respectively, both of which are established mosquito attractants (Grant
and O’Connell 1996, Lu et al. 2007, Syed et al. 2006). It is possible that An. gambiae and other
mosquitoes use a complex host odour blend as a crucial host-seeking cue, of which CO2 and
1-octen-3-ol are two components. It has been hypothesised (Lu et al. 2007) that by pairing the
CO2-sensitive cpA neuron with the 1-octen-3-ol-sensitive cpB neuron within the capitate peg
sensilla, the mosquito might be able to perceive the ratio between these two compounds within
the host odour blend with high fidelity and spatial resolution. Indeed this organisation is shared
by several cases of insect pheromone-sensing sensilla, in which ORNs responsive to individual
components of the pheromone blend are compartmentalised (Nakagawa et al. 2005, Wojtasek
et al. 1998). Interestingly, the insect repellent DEET (N,N-diethyl-meta-toluamide) inhibits the

Olfaction in vector-host interactions 27

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

AL
A 1
Scale

OL 2
A
D SOG
V
B P P
3

A
Cp
4
cpC
5 Or7-Or28
Acetophenone
2,4,5-trimethylthiazole

Maxillary palp
cpA cpB
Proboscis Gr22-Gr23-Gr24 Or7-Or8
CO2 1-octen-3-ol
DEET

Figure 6. The maxillary palp of An. gambiae exhibits olfactory responses. Schematic view of the mosquito head
and brain, anterior (A), posterior (P), dorsal (D) and ventral (V) axes, as indicated. (A) Palpal ORNs project to a
posteromedial region of the antennal lobes (AL) in the brain. (B) Capitate pegs (Cp) are located on the ventral side
of the 2nd, 3rd and 4th segments of the maxillary palp. Each neuron in the Cp (CpA, CpB, CpC) responds to different
odorant stimuli and expresses a different set of odorant receptors (Ors) or gustatory receptors (Grs). SOG, sub-
oesophageal ganglion, OL, optic lobe.

spontaneous firing (Lu et al. 2007) and 1-octen-3-ol-evoked responses in the An. gambiae cpB
neuron; whereas CO2-evoked responses of the cpA neuron are unaffected (Ditzen et al. 2008). The
response profiles of the cpC neuron seem to differ between anophelines and culicines. Whereas the
An. gambiae cpC neuron responds to a broad panel of odorants including 2,4,5-trimethylthiazole
and acetophenone (Lu et al. 2007), the cpC neurons in Ae. aegypti and Cx. quinquefasciatus seem
to be extremely narrowly-tuned and do not respond to any of the tested stimuli (Syed et al. 2006)
(M. Ghaninia, B.S. Hansson and R. Ignell, personal communication), suggesting that they might be
involved in the detection of mosquito pheromones or kairomones.

Localisation studies using in situ hybridisation indicate that a distinctive set of AgOrs and AgGrs
are expressed in the triad of ORNs in the maxillary palp of An. gambiae (Jones et al. 2007, Lu et al.
2007). The cpA neuron co-expresses three CO2 receptors, AgGr22, AgGr23, and AgGr24; and the
other two ORNs express AgOr7 along with AgOr8 or AgOr28, respectively (Jones et al. 2007, Lu et
al. 2007) (Figure 6). However, additional AgOrs might also be expressed in the maxillary palp of An.
gambiae, as a small fraction of AgOr7-positive ORNs is devoid of AgOr8 or AgOr28 signals in double
in situ hybridisation assays (Lu et al. 2007). Indeed, a recent study showed that in addition to AgOr8
and AgOr28, several other AgOrs could be detected from palpal cDNAs, albeit only after 35 cycles
of PCR amplification (Iatrou and Biessmann 2008). Two AaOrs are expressed in the maxillary palp

28  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

of Ae. aegypti (Bohbot et al. 2007). Of these AaOr8 is a close ortholog of AgOr8 (with 70% identity),
whereas AaOr49 is not closely related to AgOr28 (Bohbot et al. 2007). Three AaGrs, AaGr1, AaGr2,
and AaGr3, are close homologs of AgGr22, AgGr23, and AgGr24 (with 72-89% identity), respectively
(Kent et al. 2007), and are presumably expressed in the Ae. aegypti equivalent of the cpA neuron.
It remains interesting to speculate as to what additional factors are expressed and contribute to
the olfactory signal transduction in the capitate peg sensilla. In addition to these Ors, multiple
Obp genes are expressed in the maxillary palp of the female An. gambiae mosquito although their
role(s) in olfactory signalling remain to be elucidated (Biessmann et al. 2005).

Heterologous expression systems have successfully recapitulated the response properties of

palpal AgORs and AgGRs and assigned them to the An. gambiae cpA, cpB, or cpC neurons (Lu
et al. 2007). Specifically, the response spectra of AgOR8 and AgOR28, when co-expressed with
AgOR7 and a Gα15 subunit in Xenopus oocytes, closely mirror that of the native cpB and cpC
neurons (Lu et al. 2007). Of these, AgOR8 manifests a highly sensitive response to 1-octen-3-ol,
and this is in the absence of the native dendritic structure and environment (Lu et al. 2007). These
data are consistent with the current paradigms that insect ORs provide the primary determinant
of response spectra of ORNs (Hallem et al. 2004b, Nakagawa et al. 2005). AaOR8, the Ae. aegypti
ortholog of AgOR8, also responds to 1-octen-3-ol with high sensitivity when expressed in Xenopus
oocytes (G. Wang and L.J. Zwiebel, unpublished data), and this suggests that AaOR8 is likewise
expressed in the Ae. aegypti cpB neuron. The three An. gambiae CO2 receptors, AgGR22, AgGR23,
and AgGR24, which display high sequence homology to other insect CO2 receptors (Jones et al.
2007, Lu et al. 2007), are sufficient to drive a CO2 response when co-expressed in the Drosophila
antennal basiconic type (ab) 3A ORN (Lu et al. 2007). Importantly, sensitivity to CO2 in this neuron
does not require AgOR7 expression in An. gambiae or DOR83b in Drosophila. Indeed, it would
seem that Drosophila has lost its ortholog of AgGR23 subsequent to the divergence of the two
species (Lu et al. 2007) and ectopic expression of DmGR21a and DmGR63a, the fruit fly ortholog of
AgGR22 and AgGR24, is sufficient to induce CO2 sensitivity in the otherwise CO2-insensitive ab3A
neuron (Jones et al. 2007, Kwon et al. 2007). The function of AgGR23 in the cpA neuron remains
uncharacterised. Whereas one copy of AgGr22 and AgGr24 transgenes is not sufficient to induce
CO2 sensitivity in the ab3A neuron, adding an AgGr23 transgene further to this ORN will make it
CO2-sensitive (Lu et al. 2007). However, compared to the An. gambiae cpA neuron, the response of
three AgGr transgenes to CO2 is weak and its sensitivity is low (Lu et al. 2007), and this limitation
hampers further interpretation of the result.

Anterograde staining of maxillary palp axons revealed that mosquito palpal ORNs innervate a set
of dorso-medial glomeruli in the antennal lobe (Distler and Boeckh 1997, Ghaninia et al. 2007).
Three glomeruli in Ae. aegypti and five to six glomeruli in An. gambiae receive afferent input
from the maxillary palp nerve (Ghaninia et al. 2007, Ignell et al. 2005); and in both species, one
glomerulus is considerably larger than the others and has been suggested to receive projections
from CO2-sensitive ORNs (Anton et al. 2003, Ignell et al. 2005). In the mosquito it would therefore
appear that CO2-sensitive ORNs connect to a dorso-medial glomerulus in the antennal lobe and
mediate attraction to CO2. In contrast, the Drosophila CO2-sensitive ORNs innervate a ventral
glomerulus and mediate avoidance (Suh et al. 2007, Suh et al. 2004), suggesting that different
locations of the CO2-receptive glomerulus in the antennal lobe may reflect divergent neural
circuits that potentially mediate differential behavioural responses. Interestingly, it has recently
been suggested this difference is partially accounted for by miR-279, a regulatory microRNA that
is expressed in the Drosophila maxillary palp to prevent expression of Drosophila CO2 receptors
and drive the projections of CO2-sensitive ORNs away from medial glomeruli (Cayirlioglu et al.
2008). It is possible that Drosophila has gained or started to utilise miR-279 after divergence from

Olfaction in vector-host interactions 29

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

its common ancestor with the mosquito, and that this, along with additional mechanisms, has
shifted expression of the CO2-sensitive ORNs from the maxillary palp to the antenna and modified
their projections in the antennal lobe.

Larval olfaction

The study of larval olfaction in mosquitoes followed seminal work on the larval feeding behaviours
and ecology as important elements in larval control strategies as well as basic studies in Drosophila
that used molecular genetics to dissect sensory and behaviour pathways in the less complex
larval neural system. More recently these approaches have started to merge together to illuminate
several elements of olfactory-driven behaviours in the mosquito.

Contrary to aquatic mosquito larvae that typically develop in nutrient-poor environments,

terrestrial-based Drosophila larvae have access to a virtually unlimited amount of food (Merritt
et al. 1992) and not surprisingly, respond to a wide variety of odorants (Cobb and Domain 2000).
Attractive and repellent responses are mediated by 23-25 ORs (Fishilevich et al. 2005, Kreher et al.
2005) that are distributed in 21 ORNs (Singh and Singh 1984) within the dorsal organ of the larva.
Remarkably, while single ORNs can detect odorants, full behavioural responses seem to require
the combinatorial activation of several different ORs probably enhancing the sensitivity of the
olfactory system.

Until recently, little information was known on the molecular and cellular aspects of the larval
chemosensory system of mosquitoes. The larval antenna as best studied in Culicidae, typically
comprises the sensory cone and the peg organ (Figure 7), the latter housing the two sensilla
considered to be the principal sites for larval chemoreception (Nicastro et al. 1998, Zacharuk et
al. 1971). The sensory cone is a multiporous sensillum while the peg organ is considered a taste
sensillum with a terminal pore on its distal end. A combination of in situ hybridisation and RT-PCR
techniques have shown that 13 AgOrs are expressed in the sensory cone of An. gambiae (Pitts et al.
2004, Xia et al. 2008). Interestingly almost twice as many AaOrs are expressed in the larval antenna
of Ae. aegypti (Bohbot et al. 2007) (Figure 7). Thirty percent of the larval AgOrs are stage specific in
An. gambiae (Xia et al. 2008), while in Ae. aegypti this proportion doubles (Bohbot et al. 2007). The
reasons behind these qualitative and quantitative differences are unknown but certainly reflect
the respective and specific evolutionary history of these insects. It certainly suggests that the
Aedes larval odour space is more complex than its anopheline cousin.

There is behavioural evidence that An. gambiae larvae respond both positively and negatively
to various compounds of different chemical structures and ablation of the An. gambiae larval
antenna suppresses these olfactory-driven behaviours (Xia et al. 2008). Larval attractants include
aromatic compounds such as isomers of methylphenol and indole, which are associated with food
sources and are also known to trigger electrophysiological responses in adult female mosquitoes
(Blackwell and Johnson 2000). 1-Octen-3-ol, in addition to activating the cpB neuron in adult
female An. gambiae (Lu et al. 2007), is a component of mammalian body odours (Ramoni et al.
2001). Two compounds have been shown to induce repellent response in larvae: of particular
significance is the commercial insect repellent DEET, which recently has been postulated to act
by blocking the response activity of a subset of ORs (Ditzen et al. 2008). The fact that DEET is
able to evoke repulsive behaviours in Anopheles larvae and in adult Cx. quinquefasciatus in the
absence of any other sensory stimuli suggests that it is able to act as a direct excito-repellent
(Syed and Leal 2008, Xia et al. 2008). The molecular mechanism responsible for these avoidance

30  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

A Larval head B Larval antenna C Odorant receptors

An. gambiae Ae. aegypti

Sensory
cone 1 L 1
2 2
Peg 6 3
organ 7 7
10 8
28 L 9
34 L 14
L 37 15
L 40 16
48 29
49 L 34
12 ORNs L 52 L 37
L 58 L 40
47
L 48
L 58
L 60
L 61
L 73
L 74
A L 75
D V 78
P L 89
L 90

Figure 7. Larval antenna of mosquitoes exhibits olfactory responses. (A) Schematic dorsal view of a larval head.
(B) Twelve odorant receptor neurons (ORNs) are located in the larval antenna of An. gambiae. Their dendrites
project into the sensory cone. (C) Expression of Or genes in the larval antenna of Ae. aegypti and An. gambiae.
L, larval specific expression.

effects is olfactory mediated as a DEET specific ORN located in the short trichoid sensilla on the
Cx. quinquefasciatus antennae responds to DEET alone (Syed and Leal 2008).

Heterologous expression in Xenopus oocytes coupled with the two-electrode voltage clamp
technique revealed that AgOR1/2/6/10/28/34/48 respond to two or more odorants while the larval
specific AgOR40 is activated by DEET (Xia et al. 2008). Olfactory receptors of An. gambiae (AgORs)
either display broad responses to various odorants or are narrowly tuned to a few compounds,
suggesting redundancy at the level of odour coding at the peripheral level, a feature widely
observed in Drosophila (Kreher et al. 2005). It has been proposed that the partial physiological
convergence between larval and adult olfaction is a reflection of the shared subset of expressed
Ors between the larval and adult stages of mosquitoes (Xia et al. 2008) and flies (Kreher et al. 2005).

Olfactory receptors of larval and adult stages of Drosophila project to discrete, non-overlapping
areas in the antennal lobe. In addition, each fly larval ORN expresses one DOr. This molecular and
cellular feature has been described as the one ORN-one OR-one glomerulus ‘paradigm’. This linear
organisation is also present in the olfactory system of the Anopheles larva, whereby based on the

Olfaction in vector-host interactions 31

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

expression of AgOr7, only twelve ORNs project their dendrites into the sensory cone of Anopheles
(Xia et al. 2008) matching the number of conventional AgOrs identified by RT-PCR. In Aedes, the
situation may be different as the Or/ORN ratio is close to two (Bohbot et al. 2007).

Concluding remarks

Our understanding of the molecular mechanisms underlying olfactory signal transduction in

vector mosquitoes has dramatically improved. The diverse nature of the cellular and molecular
processes involved in olfaction will most likely bring to light several parallel models to better
explain pheromone as well as general odorant perceptions. Over and above these basic studies,
OBP and OR proteins display several features including characteristic and species-specific
molecular diversity, a central role in olfactory transduction and accessibility that make them
potential targets that are amenable to large scale and high-throughput screening with an eye
towards the development of novel tools for vector control through behavioural misdirection.

References

Adams MD, Celniker SE, Holt RA, Evans CA, Gocayne JD, Amanatides PG, Scherer SE, Li PW, Hoskins RA, Galle RF, George
RA, Lewis SE, Richards S, Ashburner M, Henderson SN, Sutton GG, Wortman JR, Yandell MD, Zhang Q, Chen LX,
Brandon RC, Rogers YH, Blazej RG, Champe M, Pfeiffer BD, Wan KH, Doyle C, Baxter EG, Helt G, Nelson CR, Gabor
GL, Abril JF, Agbayani A, An HJ, Andrews-Pfannkoch C, Baldwin D, Ballew RM, Basu A, Baxendale J, Bayraktaroglu L,
Beasley EM, Beeson KY, Benos PV, Berman BP, Bhandari D, Bolshakov S, Borkova D, Botchan MR, Bouck J, Brokstein
P, Brottier P, Burtis KC, Busam DA, Butler H, Cadieu E, Center A, Chandra I, Cherry JM, Cawley S, Dahlke C, Davenport
LB, Davies P, de Pablos B, Delcher A, Deng Z, Mays AD, Dew I, Dietz SM, Dodson K, Doup LE, Downes M, Dugan-
Rocha S, Dunkov BC, Dunn P, Durbin KJ, Evangelista CC, Ferraz C, Ferriera S, Fleischmann W, Fosler C, Gabrielian AE,
Garg NS, Gelbart WM, Glasser K, Glodek A, Gong F, Gorrell JH, Gu Z, Guan P, Harris M, Harris NL, Harvey D, Heiman
TJ, Hernandez JR, Houck J, Hostin D, Houston KA, Howland TJ, Wei MH, Ibegwam C, Jalali M, Kalush F, Karpen GH,
Ke Z, Kennison JA, Ketchum KA, Kimmel BE, Kodira CD, Kraft C, Kravitz S, Kulp D, Lai Z, Lasko P, Lei Y, Levitsky AA, Li
J, Li Z, Liang Y, Lin X, Liu X, Mattei B, McIntosh TC, McLeod MP, McPherson D, Merkulov G, Milshina NV, Mobarry C,
Morris J, Moshrefi A, Mount SM, Moy M, Murphy B, Murphy L, Muzny DM, Nelson DL, Nelson DR, Nelson KA, Nixon
K, Nusskern DR, Pacleb JM, Palazzolo M, Pittman GS, Pan S, Pollard J, Puri V, Reese MG, Reinert K, Remington K,
Saunders RD, Scheeler F, Shen H, Shue BC, Sidén-Kiamos I, Simpson M, Skupski MP, Smith T, Spier E, Spradling AC,
Stapleton M, Strong R, Sun E, Svirskas R, Tector C, Turner R, Venter E, Wang AH, Wang X, Wang ZY, Wassarman DA,
Weinstock GM, Weissenbach J, Williams SM, WoodageT, Worley KC, Wu D, Yang S, Yao QA, Ye J, Yeh RF, Zaveri JS, Zhan
M, Zhang G, Zhao Q, Zheng L, Zheng XH, Zhong FN, Zhong W, Zhou X, Zhu S, Zhu X, Smith HO, Gibbs RA, Myers
EW, Rubin GM and Venter JC (2000) The genome sequence of Drosophila melanogaster. Science 287: 2185-2195.
Anton S, van Loon JJA, Meijerink J, Smid HM, Takken W and Rospars J-P (2003) Central projections of olfactory receptor
neurons from single antennal and palpal sensilla in mosquitoes. Arthropod Structure & Development 32: 319-327.
Benton R, Sachse S, Michnick SW and Vosshall LB (2006) Atypical membrane topology and heteromeric function of
Drosophila odorant receptors in vivo. PLoS Biol 4: e20.
Benton R, Vannice KS and Vosshall LB (2007) An essential role for a CD36-related receptor in pheromone detection in
Drosophila. Nature 450 (7167): 289-293.
Bianchet MA, Bains G, Pelosi P, Pevsner J, Snyder SH, Monaco HL and Amzel LM (1996) The three-dimensional structure of
bovine odorant binding protein and its mechanism of odor recognition [see comments]. Nat Struct Biol 3: 934-939.
Biessmann H, Nguyen QK, Le D and Walter MF (2005) Microarray-based survey of a subset of putative olfactory genes
in the mosquito Anopheles gambiae. Insect Mol Biol 14: 575-589.
Biessmann H, Walter MF, Dimitratos S and Woods D (2002) Isolation of cDNA clones encoding putative odourant binding
proteins from the antennae of the malaria-transmitting mosquito, Anopheles gambiae. Insect Mol Biol 11: 123-132.

32  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

Blackwell A and Johnson SN (2000) Electrophysiological investigation of larval water and potential oviposition chemo-
attractants for Anopheles gambiae s.s. Ann Trop Med Parasitol 94: 389-398.
Bohbot J, Pitts RJ, Kwon HW, Rutzler M, Robertson HM and Zwiebel LJ (2007) Molecular characterization of the Aedes
aegypti odorant receptor gene family. Insect Mol Biol. 16: 525-537.
Bohbot J and Vogt RG (2005) Antennal expressed genes of the yellow fever mosquito (Aedes aegypti L.); characterization
of odorant-binding protein 10 and takeout. Insect Biochem Mol Biol 35: 961-979.
Bosch OJ, Geier M and Boeckh J (2000) Contribution of fatty acids to olfactory host finding of female Aedes aegypti.
Chem Senses 25: 323-330.
Cabrera M and Jaffe K (2007) An aggregation pheromone modulates lekking behavior in the vector mosquito Aedes
aegypti (Diptera: Culicidae). J Am Mosq Control Assoc 23: 1-10.
Cayirlioglu P, Kadow IG, Zhan X, Okamura K, Suh GS, Gunning D, Lai EC and Zipursky SL (2008) Hybrid neurons in a
microRNA mutant are putative evolutionary intermediates in insect CO2 sensory systems. Science 319: 1256-1260.
Clyne PJ, Warr CG, Freeman MR, Lessing D, Kim J and Carlson JR (1999) A novel family of divergent seven-transmembrane
proteins: candidate odorant receptors in Drosophila. Neuron 22: 327-338.
Cobb M and Domain I (2000) Olfactory coding in a simple system: adaptation in Drosophila larvae. Proc Biol Sci 267:
2119-2125.
Davis EE (1974) Identification of antennal chemoreceptors of the mosquito, Aedes aegypti: a correction. Experientia
30: 1282-1283.
Davis EE (1996) Olfactory control of mosquito behaviour. Ciba Found Symp 200: 48-53.
Dekker T, Steib B, Carde RT and Geier M (2002) L-lactic acid: a human-signifying host cue for the anthropophilic mosquito
Anopheles gambiae. Med Vet Entomol 16: 91-98.
Distler P and Boeckh J (1997) Central projections of the maxillary and antennal nerves in the mosquito Aedes aegypti.
J Exp Biol 200: 1873-1879.
Ditzen M, Pellegrino M and Vosshall LB (2008) Insect odorant receptors are molecular targets of the insect repellent
DEET. Science 319: 1838-1842.
Dobritsa AA, van der Goes van Naters W, Warr CG, Steinbrecht RA and Carlson JR (2003) Integrating the molecular and
cellular basis of odor coding in the Drosophila antenna. Neuron 37: 827-841.
Ellin RI, Farrand RL, Oberst FW, Crouse CL, Billups NB, Koon WS, Musselman NP and Sidell FR (1974) An apparatus for the
detection and quantitation of volatile human effluents. J Chromatogr 100: 137-152.
Fishilevich E, Domingos AI, Asahina K, Naef F, Vosshall LB and Louis M (2005) Chemotaxis behavior mediated by single
larval olfactory neurons in Drosophila. Curr Biol 15: 2086-2096.
Flower DR, North AC and Sansom CE (2000) The lipocalin protein family: structural and sequence overview. Biochim
Biophys Acta 1482: 9-24.
Foret S, Wanner KW and Maleszka R (2007) Chemosensory proteins in the honey bee: Insights from the annotated
genome, comparative analyses and expressional profiling. Insect Biochem Mol Biol 37: 19-28.
Forstner M, Gohl T, Gondesen I, Raming K, Breer H and Krieger J (2008) Differential expression of SNMP-1 and SNMP-2
proteins in pheromone-sensitive hairs of moths. Chem Senses 33: 291-299.
Foster WA (1995) Mosquito sugar feeding and reproductive energetics. Annu Rev Entomol 40, 443-474.
Fox AN, Pitts RJ, Robertson HM, Carlson JR and Zwiebel LJ (2001) Candidate odorant receptors from the malaria vector
mosquito Anopheles gambiae and evidence of down-regulation in response to blood feeding. PNAS 98: 14693-14697.
Gao Q and Chess A (1999) Identification of candidate Drosophila olfactory receptors from genomic DNA sequence.
Genomics 60: 31-39.
Gaunt MW and Miles MA (2002) An insect molecular clock dates the origin of the insects and accords with
palaeontological and biogeographic landmarks. Mol Biol Evol 19: 748-761.
Geier M, Bosch OJ and Boeckh J (1999) Ammonia as an attractive component of host odour for the yellow fever
mosquito, Aedes aegypti. Chem Senses 24: 647-653.
Ghaninia M, Hansson BS and Ignell R (2007) The antennal lobe of the African malaria mosquito, Anopheles gambiae –
innervation and three-dimensional reconstruction. Arthropod Struct Dev 36: 23-39.

Olfaction in vector-host interactions 33

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

Goldman AL, Van der Goes van Naters W, Lessing D, Warr CG and Carlson JR (2005) Coexpression of two functional odor
receptors in one neuron. Neuron 45: 661-666.
Grant A and O’Connell R, eds (1996) Electrophysiological responses from receptor neurons in mosquito maxillary palp
sensilla. Wiley, Chichester, UK.
Grosse-Wilde E, Gohl T, Bouche E, Breer H and Krieger J (2007) Candidate pheromone receptors provide the basis for the
response of distinct antennal neurons to pheromonal compounds. Eur J Neurosci 25: 2364-2373.
Hallem E, Fox AN, Zwiebel LJ and Carlson JR (2004a) A mosquito odorant receptor tuned to a component of human
sweat. Nature 427: 212-213.
Hallem E, Ho MG and Carlson JR (2004b) The molecular basis of odor coding in the Drosophila antenna. Cell 117: 965-979.
Healy TP and Copland MJ (1995) Activation of Anopheles gambiae mosquitoes by carbon dioxide and human breath.
Med Vet Entomol 9: 331-336.
Healy TP and Copland MJ (2000) Human sweat and 2-oxopentanoic acid elicit a landing response from Anopheles
gambiae. Med Vet Entomol 14: 195-200.
Hekmat-Scafe DS, Scafe CR, McKinney AJ and Tanouye MA (2002) Genome-wide analysis of the odorant-binding protein
gene family in Drosophila melanogaster. Genome Res 12: 1357-1369.
Hill CA, Fox AN, Pitts RJ, Kent LB, Tan PL, Chrystal MA, Cravchik A, Collins FH, Robertson HM and Zwiebel LJ (2002) G
protein-coupled receptors in Anopheles gambiae. Science 298: 176-178.
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides
R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P,
Atkinson PW, Baden H, de Berardinis V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M,
Cai S, Center A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A,
Dew I, Evans CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun
SL, Hogan JR, Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin
JJ, Lobo NF, Lopez JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O’Brochta DA,
Pfannkoch C, Qi R, Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J,
Thomasova D, Ton LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman
JR, Wu M, Yao A, Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della Torre A, Roth CW,
Louis C, Kalush F, Mural RJ, Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey
PT, Venter JC, Weissenbach J, Kafatos FC, Collins FH and Hoffman SL (2002) The genome sequence of the malaria
mosquito Anopheles gambiae. Science 298: 129-149.
Iatrou K and Biessmann H (2008) Sex-biased expression of odorant receptors in antennae and palps of the African
malaria vector Anopheles gambiae. Insect Biochem Mol Biol 38: 268-274.
Ignell R, Dekker T, Ghaninia M and Hansson, BS (2005) Neuronal architecture of the mosquito deutocerebrum. J Comp
Neurol 493: 207-240.
Jones WD, Cayirlioglu P, Kadow IG and Vosshall LB (2007) Two chemosensory receptors together mediate carbon dioxide
detection in Drosophila. Nature 445: 86-90.
Jones WD, Nguyen TA, Kloss B, Lee KJ and Vosshall LB (2005) Functional conservation of an insect odorant receptor gene
across 250 million years of evolution. Curr Biol 15: R119-121.
Kain P, Chakraborty TS, Sundaram S, Siddiqi O, Rodrigues V and Hasan G (2008) Reduced odor responses from antennal
neurons of G (q)alpha, phospholipase Cbeta, and rdgA mutants in Drosophila support a role for a phospholipid
intermediate in insect olfactory transduction. J Neurosci 28: 4745-4755.
Kalidas S and Smith DP (2002) Novel genomic cDNA hybrids produce effective RNA interference in adult Drosophila.
Neuron 33: 177-184.
Kellogg FE (1970) Water vapour and carbon dioxide receptors in Aedes aegyptii. J Insect Physiology 16: 99-108.
Kent LB, Walden KK and Robertson HM (2007) The Gr family of candidate gustatory and olfactory receptors in the
yellow-fever mosquito Aedes aegypti. Chem Senses 33: 79-93.
Knols BGJ (1996) Odour-mediated host seeking behavior of the Afro-tropical malaria vector Anopheles gambiae Giles.
Thesis, Wageningen Agricultural University, Wageningen, the Netherlands.
Kreher SA, Kwon JY and Carlson JR (2005) The molecular basis of odor coding in the Drosophila larva. Neuron 46: 445-456.

34  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

Krieger J, Klink O, Mohl C, Raming K and Breer H (2003) A candidate olfactory receptor subtype highly conserved across
different insect orders. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 189: 519-526.
Kruse SW, Zhao R, Smith DP and Jones DN (2003) Structure of a specific alcohol-binding site defined by the odorant
binding protein LUSH from Drosophila melanogaster. Nat Struct Biol 10: 694-700.
Kwon HW, Lu T, Rutzler M and Zwiebel LJ (2006) Olfactory responses in a gustatory organ of the malaria vector mosquito
Anopheles gambiae. PNAS 103: 13526-13531.
Kwon JY, Dahanukar A, Weiss LA and Carlson JR (2007) The molecular basis of CO2 reception in Drosophila. PNAS 104:
3574-3578.
Larsson MC, Domingos AI, Jones WD, Chiappe ME, Amrein H and Vosshall LB (2004) Or83b encodes a broadly expressed
odorant receptor essential for Drosophila olfaction. Neuron 43: 703-714.
Laughlin JD, Ha TS, Jones DN and Smith DP (2008) Activation of pheromone-sensitive neurons is mediated by
conformational activation of pheromone-binding protein. Cell 133: 1255-1265.
Lu T, Qiu YT, Wang G, Kwon JY, Rutzler M, Kwon HW, Pitts RJ, van Loon JJ, Takken W, Carlson JR and Zwiebel LJ (2007)
Odor coding in the maxillary palp of the malaria vector mosquito Anopheles gambiae. Curr Biol 17: 1533-1544.
Matsuo T, Sugaya S, Yasukawa J, Aigaki T and Fuyama Y (2007) Odorant-binding proteins OBP57d and OBP57e affect
taste perception and host-plant preference in Drosophila sechellia. PLoS Biology 5: e118 doi: 110.1371/journal.
pbio.0050118.
McIver SB (1972) Fine structure of pegs on the palps of female culicine mosquitoes. Can J Zool 50: 571-576.
McIver SB (1982) Sensilla of mosquitoes (Diptera: Culicidae). J Med Entomol 19: 489-535.
McKenna MP, Hekmat Scafe DS, Gaines P and Carlson JR (1994) Putative Drosophila pheromone-binding proteins
expressed in a subregion of the olfactory system. J Biol Chem 269: 16340-16347.
Meijerink J, Braks MA and Van Loon JJ (2001) Olfactory receptors on the antennae of the malaria mosquito Anopheles
gambiae are sensitive to ammonia and other sweat-borne components. J Insect Physiol 47: 455-464.
Meijerink J and van Loon JJ (1999) Sensitivities of antennal olfactory neurons of the malaria mosquito, Anopheles
gambiae, to carboxylic acids. J Insect Physiol 45: 365-373.
Melo AC, Rutzler M, Pitts RJ and Zwiebel LJ (2004) Identification of a chemosensory receptor from the yellow fever
mosquito, Aedes aegypti, that is highly conserved and expressed in olfactory and gustatory organs. Chem Senses
29: 403-410.
Merrill CE, Pitts RJ and Zwiebel LJ (2003) Molecular characterization of arrestin family members in the malaria vector
mosquito, Anopheles gambiae. Insect Molecular Biology 12: 641-650.
Merritt RW, Dadd RH and Walker ED (1992) Feeding behavior, natural food, and nutritional relationships of larval
mosquitoes. Annu Rev Entomol 37: 349-376.
Mohl C, Breer H and Krieger J (2002) Species-specific pheromonal compounds induce distinct conformational changes
of pheromone binding protein subtypes from Antheraea polyphemus. Invert Neurosci 4: 165-174.
Nakagawa T, Sakurai T, Nishioka T and Touhara K (2005) Insect sex-pheromone signals mediated by specific combinations
of olfactory receptors. Science 307: 1638-1642.
Nene V, Wortman JR, Lawson D, Haas B, Kodira C, Tu ZJ, Loftus B, Xi Z, Megy K, Grabherr M, Ren Q, Zdobnov EM, Lobo
NF, Campbell KS, Brown SE, Bonaldo MF, Zhu J, Sinkins SP, Hogenkamp DG, Amedeo P, Arensburger P, Atkinson
PW, Bidwell S, Biedler J, Birney E, Bruggner RV, Costas J, Coy MR, Crabtree J, Crawford M, Debruyn B, Decaprio D,
Eiglmeier K, Eisenstadt E, El-Dorry H, Gelbart WM, Gomes SL, Hammond M, Hannick LI, Hogan JR, Holmes MH,
Jaffe D, Johnston JS, Kennedy RC, Koo H, Kravitz S, Kriventseva EV, Kulp D, Labutti K, Lee E, Li S, Lovin DD, Mao C,
Mauceli E, Menck CF, Miller JR, Montgomery P, Mori A, Nascimento AL, Naveira HF, Nusbaum C, O’leary S, Orvis
J, Pertea M, Quesneville H, Reidenbach KR, Rogers YH, Roth CW, Schneider JR, Schatz M, Shumway M, Stanke M,
Stinson EO, Tubio JM, Vanzee JP, Verjovski-Almeida S, Werner D, White O, Wyder S, Zeng Q, Zhao Q, Zhao Y, Hill CA,
Raikhel AS, Soares MB, Knudson DL, Lee NH, Galagan J, Salzberg SL, Paulsen IT, Dimopoulos G, Collins FH, Birren B,
Fraser-Liggett CM and Severson DW (2007) Genome sequence of Aedes aegypti, a major arbovirus vector. Science
316: 1718-1723.
Neuhaus EM, Gisselmann G, Zhang W, Dooley R, Stortkuhl K and Hatt H (2005) Odorant receptor heterodimerization in
the olfactory system of Drosophila melanogaster. Nat Neurosci 8: 15-17.

Olfaction in vector-host interactions 35

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

Nicastro D, Melzer R, Hruschka H and Smola U (1998) Evolution of small sense organs: sensilla on the larval antennae
traced back to the origin of the Diptera. Naturwissenschaften 85: 501-505.
Nichols Z and Vogt RG (2008) The SNMP/CD36 gene family in Diptera, Hymenoptera and Coleoptera: Drosophila
melanogaster, D. pseudoobscura, Anopheles gambiae, Aedes aegypti, Apis mellifera, and Tribolium castaneum. Insect
Biochem Mol Biol 38: 398-415.
Nijhout HF and Craig GB (1971) Reproductive isolation in Stegomyia mosquitoes. III evidence for a sexual pheromone.
Entomologia Experimentalis et Applicata 14: 399-412.
Owen WB, Larsen JR and Pappas LG (1974) Functional units in the labellar chemosensory hairs of the mosquito Culiseta
inornata (Williston). J Exp Zool 188: 235-247.
Pappas LG and Larsen JR (1976) Gustatory hairs on the mosquito, Culiseta inornata. Journal of Experimantal Zoology
196: 351-360.
Pelosi P, Baldaccini NE and Pisanelli AM (1982) Identification of a specific olfactory receptor for 2-isobutyl-3-
methoxypyrazine. Biochem J 201: 245-248.
Pelosi P, Zhou JJ, Ban LP and Calvello M (2006) Soluble proteins in insect chemical communication. Cell Mol Life Sci
63: 1658-1676.
Pitts RJ, Fox AN and Zwiebel LJ (2004) A highly conserved candidate chemoreceptor expressed in both olfactory and
gustatory tissues in the malaria vector, Anopheles gambiae. PNAS 101: 5058-5063.
Pitts RJ and Zwiebel LJ (2006) Antennal sensilla of two female anopheline sibling species with differing host ranges.
Malar J 5: 26.
Qiu YT, van Loon JJ, Takken W, Meijerink J and Smid HM (2006) Olfactory Coding in Antennal Neurons of the Malaria
Mosquito, Anopheles gambiae. Chem Senses 31, 845-863.
Rac ME, Safranow K and Poncyljusz W (2007) Molecular basis of human CD36 gene mutations. Mol Med 13: 288-296.
Ramoni R, Vincent F, Grolli S, Conti V, Malosse C, Boyer FD, Nagnan-Le Meillour P, Spinelli S, Cambillau C and Tegoni M
(2001) The insect attractant 1-octen-3-ol is the natural ligand of bovine odorant-binding protein. J Biol Chem 276:
7150-7155.
Robertson HM, Warr CG and Carlson JR (2003) Molecular evolution of the insect chemoreceptor gene superfamily in
Drosophila melanogaster. PNAS 100 Suppl 2: 14537-14542.
Rogers ME, Sun M, Lerner MR and Vogt RG (1997) Snmp-1, a novel membrane protein of olfactory neurons of the silk moth
Antheraea polyphemus with homology to the CD36 family of membrane proteins. J Biol Chem 272: 14792-14799.
Rudolfs W (1922) Chemotropism of mosquitoes. Bulletin New Jersey Agr Exp St 367: 4-23.
Rutzler M, Lu T and Zwiebel LJ (2006) Galpha encoding gene family of the malaria vector mosquito Anopheles gambiae:
expression analysis and immunolocalization of AGalphaq and AGalphao in female antennae. J Comp Neurol 499:
533-545.
Rutzler M and Zwiebel L (2005) Molecular biology of insect olfaction: recent progress and conceptual models. J Comp
Physiol A Neuroethol Sens Neural Behav Physiol: 1-14.
Sabatier L, Jouanguy E, Dostert C, Zachary D, Dimarcq JL, Bulet P and Imler JL (2003) Pherokine-2 and -3. Eur J Biochem
270: 3398-3407.
Sato K, Pellegrino M, Nakagawa T, Nakagawa T, Vosshall LB and Touhara K (2008) Insect olfactory receptors are
heteromeric ligand-gated ion channels. Nature 452 (7190): 1002-1006.
Singh RN and Singh K (1984) Fine structure of the sensory organs of Drosophila melanogaster Meigen larva (Diptera:
Drosophilidae). Int J Insect Morp Embr 13: 255-273.
Smallegange RC, Qiu YT, van Loon JJ and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic acids
as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto (Diptera:
Culicidae). Chem Senses 30: 145-152.
Smith DP (2007) Odor and pheromone detection in Drosophila melanogaster. Pflugers Arch 454: 749-758.
Steib BM, Geier M and Boeckh J (2001) The effect of lactic acid on odour-related host preference of yellow fever
mosquitoes. Chem Senses 26: 523-528.
Storkuhl KF and Kettler R (2001) Functional analysis of an olfactory receptor in Drosophila melanogater. PNAS 98:
9381-9385.

36  Olfaction in vector-host interactions

 2. Molecular regulation of olfaction in mosquitoes?

Suh GS, Ben-Tabou de Leon S, Tanimoto H, Fiala A, Benzer S and Anderson DJ (2007) Light activation of an innate
olfactory avoidance response in Drosophila. Curr Biol 17: 905-908.
Suh GS, Wong AM, Hergarden AC, Wang JW, Simon AF, Benzer S, Axel R and Anderson DJ (2004) A single population of
olfactory sensory neurons mediates an innate avoidance behaviour in Drosophila. Nature 431: 854-859.
Syed Z, Ishida Y, Kimbrell DA and Leal WS (2006) Pheromone reception in fruit flies expressing a moth’s odorant receptor.
PNAS 103: 16538-16543.
Syed Z and Leal WS (2008) Mosquitoes smell and avoid the insect repellent DEET. PNAS 105 (36): 13598-13603.
Takken W, Dekker T and Wijnholds YG (1997) Odor-mediated flight behavior of Anopheles gambiae Giles sensu stricto
and A-stephensi Liston in response to CO2, acetone, and 1-octen-3-ol (Diptera: Culicidae). Journal of Insect Behavior
10: 395-407.
Takken W and Knols BG (1999) Odor-mediated behavior of Afrotropical malaria mosquitoes. Annu Rev Entomol 44:
131-157.
Vogt RG (2002) Odorant binding protein homologues of the malaria mosquito Anopheles gambiae; possible orthologues
of the OS-E and OS-F OBPs OF Drosophila melanogaster. J Chem Ecol 28: 2371-2376.
Vogt RG, Prestwich G. D and Riddiford LM (1988) Sex pheromone receptor proteins. Visualization using a radiolabeled
photoaffinity analog. J Biol Chem 263: 3952-3959.
Vogt RG and Riddiford LM (1981) Pheremone binding and inactivation by moth antennae. Nature 293: 161-163.
Vosshall LB, Amrein H, Morozov PS, Rzhetsky A and Axel R (1999) A spatial map of olfactory receptor expression in the
Drosophila antenna. Cell 96: 725-736.
Vosshall LB and Stocker RF (2007) Molecular architecture of smell and taste in Drosophila. Annu Rev Neurosci 30: 505-533.
Wang P, Lyman RF, Shabalina SA, Mackay TF and Anholt RR (2007) Association of polymorphisms in odorant-binding
protein genes with variation in olfactory response to benzaldehyde in Drosophila. Genetics 177: 1655-1665.
Wanner KW, Nichols AS, Walden KK, Brockmann A, Luetje CW and Robertson HM (2007) A honey bee odorant receptor
for the queen substance 9-oxo-2-decenoic acid. PNAS 104: 14383-14388.
Wanner KW, Willis LG, Theilmann DA, Isman MB, Feng Q and Plettner E (2004) Analysis of the insect os-d-like gene family.
J Chem Ecol 30: 889-911.
Wetzel CH, Behrendt HJ, Gisselmann G, Stortkuhl KF, Hovemann B and Hatt H (2001) Functional expression and
characterization of a Drosophila odorant receptor in a heterologous cell system. PNAS 98: 9377-9380.
Wicher D, Schafer R, Bauernfeind R, Stensmyr MC, Heller R, Heinemann SH and Hansson BS. (2008) Drosophila odorant
receptors are both ligand-gated and cyclic-nucleotide-activated cation channels. Nature 452 (7190): 1007-1011.
Wogulis M, Morgan T, Ishida Y, Leal WS and Wilson DK (2006) The crystal structure of an odorant binding protein from
Anopheles gambiae: evidence for a common ligand release mechanism. Biochem Biophys Res Commun 339: 157-164.
Wojtasek H, Hansson BS and Leal WS (1998) Attracted or repelled? a matter of two neurons, one pheromone binding
protein, and a chiral center. Biochem Biophys Res Commun 250: 217-222.
Woodard C, Alcorta E and Carlson J. (1992) The rdgB gene of Drosophila: a link between vision and olfaction. Journal
of Neurogenetics 8: 17-31.
Xia Y, Wang G, Buscariollo D, Pitts J, R, Wenger H and Zwiebel L (2008) The molecular basis of olfactory-based behavior
in Anopheles gambiae larvae. PNAS 105: 6433-6438.
Xia Y and Zwiebel LJ (2006) Identification and characterization of an odorant receptor from the West Nile Virus mosquito,
Culex quinquefasciatus. Insect Biochemistry and Molecular Biology 36: 169-176.
Xu, P, Atkinson, R, Jones, DN and Smith, DP (2005) Drosophila OBP LUSH is required for activity of pheromone-sensitive
neurons. Neuron 45: 193-200.
Xu, PX, Zwiebel, LJ and Smith, DP (2003) Identification of a distinct family of genes encoding atypical odorant-binding
proteins in the malaria vector mosquito, Anopheles gambiae. Insect Molecular Biology 12: 549-560.
Zacharuk, RY, Yin, LR.-S and Blue, SG (1971) Fine structure of the antenna and its sensory cone in the larvae of Aedes
aegypti (L.). J Morph 135: 273-298.
Zdobnov EM, von Mering C, Letunic I, Torrents D, Suyama M, Copley RR, Christophides GK, Thomasova D, Holt RA,
Subramanian GM, Mueller HM, Dimopoulos G, Law JH, Wells MA, Birney E, Charlab R, Halpern AL, Kokoza E, Kraft CL,
Lai Z, Lewis S, Louis C, Barillas-Mury C, Nusskern D, Rubin GM, Salzberg SL, Sutton GG, Topalis P, Wides R, Wincker

Olfaction in vector-host interactions 37

Jonathan D. Bohbot, Tan Lu and Laurence J. Zwiebel

P, Yandell M, Collins FH, Ribeiro J, Gelbart WM, Kafatos FC and Bork P (2002) Comparative genome and proteome
analysis of Anopheles gambiae and Drosophila melanogaster. Science 298: 149-159.
Zhou, JJ, He, XL, Pickett, JA and Field, LM (2008) Identification of odorant-binding proteins of the yellow fever mosquito
Aedes aegypti: genome annotation and comparative analyses. Insect Mol Biol 17: 147-163.
Zhou, JJ, Kan, Y, Antoniw, J, Pickett, JA and Field, LM (2006) Genome and EST analyses and expression of a gene family
with putative functions in insect chemoreception. Chem Senses 31: 453-465.

38  Olfaction in vector-host interactions

3. Olfactory physiology of blood-feeding vector mosquitoes
Yu Tong Qiu and Joop J.A. van Loon

Abstract

The olfactory organs of blood-feeding mosquitoes are the antennae, maxillary palps and proboscis.
On each of these organs, several morphologically distinct types of multiporous sensilla are present
that typically house two to four olfactory receptor neurons (ORNs). Electrophysiological studies
of ORN responsiveness have allowed the identification of several functional ORN-classes within
each sensillum type. Functional ORN-types are distinguished based on their unique spectrum
of responses to a set of volatile organic compounds (VOCs). Each neuron expresses one or a few
membrane-bound olfactory receptor (OR) proteins that function as the molecular determinant
of the response specificity of the ORN. Genomic analysis has identified 79 and 131 candidate
OR-genes in the malaria mosquito Anopheles gambiae and the yellow fever mosquito Aedes
aegypti, respectively. In the sensillum lymph surrounding the dendrites, water-soluble odorant-
binding proteins (OBPs) fulfil roles in transport and inactivation of VOCs. In Anopheles gambiae 57
OBP-genes have been found and in Ae. aegypti this number is 66. The ORNs have been shown to
encode odour quality, i.e. molecular structure, odour concentration as well as temporal changes in
concentration and spatial distribution. Responses of ORNs belong to two major types: some odour
stimuli elicit excitation whereas other odours cause inhibition relative to the pre-stimulus activity.
Both response modes are controlled by the same OR. Both generalist ORNs and specialist ORNs
are found. The responses of ORNs to odorants are concentration-dependent above a threshold
concentration, and both absolute concentration and the change in concentration can be encoded.
The ensemble of mosquito ORNs responds to odours emitted by hosts, plants and oviposition
sites, which together harbour a substantial diversity of molecular structures. Yet, the number
of VOCs tested on the full array of ORN functional types of any mosquito species is still far from
exhaustive and rarely exceeds 100 VOCs. A substantial concerted effort of molecular biologists,
electrophysiologists and ethologists lies ahead to further unravel the mosquito olfactory system.
The information thus obtained is essential for the development of behavioural disruption methods
that will contribute to the control of mosquito vector populations.

Keywords: sensillum, neuron coding, odour, receptor

Introduction

Blood feeding mosquitoes are disagreeable to humans not only due to their annoying biting habit
but especially because many species transmit fatal human diseases such as malaria, yellow fever,
dengue and West Nile virus. Olfaction is essential in the life of a mosquito to ensure survival and
successful reproduction. Mosquitoes largely rely on the sense of smell to find a mate, nectar, blood
and oviposition sites. Understanding the mechanisms of mosquito olfaction and odour coding
is crucial to develop measures based on behavioural disruption to control mosquitoes. Because
mosquito species that act as vectors of infectious diseases belong predominantly to the genera
Anopheles, Aedes and Culex, we restrict ourselves mainly to these three mosquito genera in this
chapter.

Olfaction in vector-host interactions 39

Yu Tong Qiu and Joop J.A. van Loon

Location and morphology of olfactory organs in adult mosquitoes

The main olfactory organs of insects are the antenna and mouthparts (Figure 1), which carry
specialised cuticular extensions, the so-called sensilla that house olfactory receptor neurons
(ORNs). The majority of olfactory sensilla of adult mosquitoes are situated on the antennae, yet a
smaller number is located on the maxillary palps and the proboscis. Female mosquitoes possess

Figure 1. Scanning electron micrograph of olfactory organs of female Anopheles gambiae. (a) The head
showing An: antenna; MP: maxillary palp; Pr: proboscis. (b) Sensilla on the antennae. SA: sensillum ampullacea;
GP: grooved peg sensillum; TC: sensillum trichodea subtype C. (c) The 13th segment of the antenna. sSC: small
sensillum coeloconicum. (d) One segment of the antenna. lSC: large sensillum coeloconicum; TA, TB, TC, TD or TE:
sensillum trichodea subtype A, B, C, D or E. (e) Capitate peg sensilla (CP) on the maxillary palps. (f) The tip of the
proboscis showing the olfactory sensilla T1 (SEM photos provided by Hans Smid).

40  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

a much larger number of olfactory sensilla than their male counterparts (McIver 1982, Table 1).
Highly specialised in detecting the wing beat sounds from conspecific female mosquitoes (Cator
et al. 2009), the antennae of male mosquitoes are inhabited by fibrillae on most of the antennal
segments, only the proximal and the two most distal segments are populated with olfactory
sensilla (Ismail 1962).

Olfactory sensilla are hair- or peg-shaped and have a multiporous wall structure. A typical insect
olfactory sensillum is composed of two or more bipolar olfactory receptor neurons (ORNs),
although sensilla containing one neuron have also been described (McIver 1982), three auxiliary
cells and the surrounding glia, epidermis and cuticle (Figure 2). The surface of the insect cuticle
is hydrophobic, a property reducing water evaporation and increasing water repellency. Odour
molecules, many of which are non-polar, first have to penetrate through the sensillum wall before
reaching the ORNs. There are two types of sensillum walls: single-walled olfactory sensilla contain
pore channels through which the non-polar components are excreted to the epicuticle during
sensillum formation (Figure 2). Pore channels that remain open are later used for the transportation
of the odour molecules into the sensillum lymph (Steinbrecht 1997). In contrast, a double-walled
olfactory sensillum is composed of hollow cuticular finger-like structures, which are fused to each
other and form at the fusion points, spoke-channels. It is likely that odour molecules get into the
sensillum lumen of double-walled sensilla via these channels (Steinbrecht 1997).

Antennal sensilla

The antennae of adult mosquitoes bear five types of sensilla: chaetica, coeloconica, ampullacea,
trichodea, and grooved pegs (Figure 1). The latter type is also referred to as sensilla basiconica in
some references, which may be confused with the single-walled basiconic sensilla; therefore here
we only use the name grooved peg. Sensilla chaetica are mechanosensilla with a thick wall and
a socket at their base. Anophelines are the only mosquitoes in the family of Culicidae that have a
large type of coeloconic sensilla. Although both large and small sensilla coeloconica have a ‘peg
in pit’ structure, they differ from each other both morphologically and functionally. The small

Table 1. Numbers of olfactory sensilla of mosquito species located on the antennae and maxillary palps (McIver
1982).

Species Palpi Antenna

Trichodea Grooved peg Large coeloconica

♀ ♂ ♀ ♂ ♀ ♂ ♀ ♂

Anopheles gambiae 67 14 629 227 84 20 33 14

Anopheles maculipennis 65 13 618 300 62 27 28 12
Anopheles stephensi 98 9 573 194 107 28 32 8
Aedes aegypti 29 17 532 175 105 36 0 0
Culex pipiens 78 15 901 239 265 95 0 0
Culex territans 39 15 537 216 275 174 0 0
Mansonia africana 68 21 623 147 208 36 0 0
Mansonia uniformis 75 17 571 164 212 39 0 0
Wyomia smithii 34 20 218 189 36 30 0 0

Olfaction in vector-host interactions 41

Yu Tong Qiu and Joop J.A. van Loon

Figure 2. Schematic drawing of the structure of typical olfactory sensilla. AX: axon; CUT: cuticle; D: desmosomes;
DS: dendrite sheath; E: epidermis; GJ: gap junctions; GL: glia; HL: haemolymph; ID: inner dendritic segment; ISL:
inner sensillum lymph space; OD: outer dendritic segment = ciliary dendrite; ORN: olfactory receptor neuron;
OSL: outer sensillum lymph space; PC: pore channel; SJ: septate junction; TH: thecogen cell; TO: tormogen cell; TR:
trichogen cell (from Qiu 2005 after Keil and Steinbrecht 1987).

coeloconica contain a single-walled peg, with possibly one or a few pores at the tip (Clements
1999, McIver 1982). A pair of small coeloconic sensilla is located at the tip of mosquito antennae,
occasionally these sensilla are found on other antennal segments (McIver 1982, Pitts and Zwiebel
2006). Two of the three neurons innervating these small sensilla coeloconica were found to be
sensitive to convection heat, whereas none of the three neurons are olfactory (Davis and Sokolove
1975, Gingl et al. 2005, Wang et al. 2009). Sensilla ampullacea are thought to function as convection
heat detectors because they have a structure similar to that of small sensilla coeloconica (McIver
1982). Only sensilla trichodea, grooved peg sensilla and large coeloconica are innervated by
olfactory neurons.

Sensilla trichodea

The most abundant type of antennal sensillum of mosquitoes are sensilla trichodea (Figure 1b, 1c),
which are single-walled olfactory sensilla normally containing dendrites of two ORNs, occasionally
sensilla trichodea innervated by a single neuron can be found on the most distal antennal flagella
(Boo 1980a, McIver 1982) (Figure 3a, 3c). The average number of sensilla trichodea in a species
ranges from 200 to 1000 (Table 1, based on McIver (1982). Mosquito antennae carry several different
shapes of sensilla trichodea, each of which has distinct functional characteristics. Culicines have

42  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

Figure 3 Schematic diagrams of single- and double-wall structure of olfactory sensilla. (a) Section through a
sensillum trichodeum (single-walled) and pore tubules. HW: hair wall; D: dendrite; P: pores. (b) Coeloconic
sensillum (double-walled), opened to show dendrites therein. The cuticular fingers (F) are hollow and partly fused.
(c) Detail of wall of sensillum trichodeum. PC: pore channel; PT: pore tubules. (d) Detail of the double wall of a
coeloconic sensillum. SC: spoke channels; F: fingers; GR: groove (from Qiu 2005 after Keil 1999).

basically four types of sensilla trichodea: T1, T2, T3 and T4, classified according to their length, wall
thickness and whether the tip is sharp or blunt (reviewed by McIver 1982). This classification does
not fit well for anophelines. Transmission microscope studies by Boo (1980a,b) identified five types
of sensilla trichodea, designated A – E, on the antennae of Anopheles stephensi Liston according to
hair length, shape, diameter, wall thickness, neuron branching and pore channel density.

Van den Broek and Den Otter (1999) compared the sensitivities of ORNs in four anopheline
species with host preferences ranging from zoophilic to anthropophilic and found differences
in the number and sensitivity of ORNs in sensilla trichodea responding to carboxylic acids and
1-octen-3-ol.

Olfaction in vector-host interactions 43

Yu Tong Qiu and Joop J.A. van Loon

Grooved pegs

Grooved peg sensilla are double-walled sensilla containing 2 to 5 neurons one of which is assumed
to be a hygroreceptor (Figure 3b, 3d). In Aedes and Culex species, grooved peg sensilla show great
variation in peg length and it was suggested that two types exist and that the groove length
correlated with the sensitivity of innervating neurons to lactic acid (Bowen 1995). However, large
overlap occurs between length distributions of pegs with different lactic acid sensitivity.

Large coeloconic sensilla

The large coeloconic sensilla are only found on the antennae of anopheline mosquitoes (Table 1,
McIver 1982). Double-walled pegs are contained in sunken pits. The number of large coeloconic
sensilla on each antenna of female mosquitoes range from 28 for An. maculipennis Meigen to 50
for An. ziemanni Grünberg and these are innervated by four or five neurons, the function of which
has not been studied.

Maxillary palps and proboscis

On the maxillary palps the only sensilla innervated by ORNs are the capitate peg sensilla, which in
female anophelines are located on the ventral side of palp segments 2-4 and in male anophelines on
segment 4 (McIver and Siemicki 1975). In both sexes of Culex and Aedes mosquitoes capitate pegs
are located on segment 4 (reviewed by McIver 1982). Three neurons are co-compartmentalised
in each capitate peg sensillum. The dendrite of one neuron forms numerous lamellae (reviewed
by McIver 1982).

At the tip of the labellar lobe there are three types of trichoid sensilla, two of which (T1 and T2) are
located externally and one (T3) internally (McIver 1982, Pappas and Larsen 1976). These sensilla
were thought to contain mechanical and gustatory neurons, however, a recent study revealed that
T2 sensilla contain olfactory receptor proteins and have an olfactory function (Kwon et al. 2006).

Olfactory transduction

Transduction is the process by which quality (molecular structure) and quantity (concentration)
of odorants are converted to the neural code contained in the frequency and temporal patterns
of action potentials (see Chapter 4, this volume). The nature of this process is considered to be
common to all insects, although most information is derived from studies on Lepidoptera and
the fruit fly, Drosophila melanogaster Meigen (Rutzler and Zwiebel 2005). The sensillum lymph
of olfactory sensilla contains water soluble proteins, called odorant binding proteins (OBPs) that
bind odour molecules and transport them to receptor molecules on the dendritic membrane
(Klein 1987, Vogt and Riddiford 1981). The possible functions of OBPs include: selective binding
with odour molecules; binding with irrelevant or harmful compounds to reduce their chance of
coming into contact with the dendritic membrane (Park et al. 2000); selective transport of odour
molecules to specific receptor molecules on the dendritic membrane and selective inactivation
of odour molecules (Steinbrecht 1998). Based on the presence of six conserved cystein residues
and a conserved spacing between the cysteins, 57 putative OBP-genes in An. gambiae Giles and
66 in Aedes aegypti (L.) have been identified (Xu et al. 2003, Zhou et al. 2008).

On the plasma membrane of the olfactory neuron dendrite one to three ligand-binding olfactory
receptor (OR) genes are co-expressed with a conservative co-receptor gene, the Or83b family

44  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

in Drosophila and AgOR7 in An. gambiae. Both OR protein and the co-receptor protein have a
seven-transmembrane domain, but unlike the vertebrate G-protein-coupled receptors (GPCRs)
their amino termini locate intracellularly. From the genome of An. gambiae, 276 G-protein-
coupled receptor homologous genes were identified and 79 of these are considered as putative
OR-genes (Hill et al. 2002). Recently, 131 putative odorant receptor genes have been identified
from the genome sequence of Ae. aegypti (Bohbot et al. 2007; Chapter 2, this Volume). In An.
gambiae conventional AgORs are co-expressed in ORNs with a highly conserved OR gene, AgOR7,
and form a heterodimer (Pitts et al. 2004). Recent studies suggest that insect ORs respond to
odour ligands along two distinct pathways (Sato et al. 2008, Wicher et al. 2008). Firstly, insect
OR heterodimers function as a ligand-gated ion channel. Secondly, the OR heterodimers also
follow a metabotropic pathway by forming a non-selective cation channel activated by cyclic
nucleotides and odorant-sensing units. When the depolarisation of the membrane resulting from
the respective ion flows reaches a threshold, action potentials are generated at the point where
the axon exits the perikaryon and travel along the axon, which projects to the antennal lobe in
the brain (see Chapter 4, this volume).

Peripheral odour coding

Properties of odour coding

By using the single sensillum recording (SSR) method, the action potentials of ORNs can be
recorded in situ; their specificity and sensitivity in response to odour stimuli can be studied by
quantifying the frequency of action potentials over defined post-stimulus time intervals. ORNs
have been shown to encode odour quality, concentration as well as temporal changes in odour
concentration and spatial distribution (De Bruyne et al. 2001, Heinbockel and Kaissling 1996,
Mustaparta 2002).

ORNs respond selectively to odour compounds with certain structural features such as chain
length, electron cloud distribution, the position of double bonds and functional groups of the
odour molecule (Boeckh and Ernst 1983, Liliefors et al. 1985, 1987, Shields and Hildebrand 2001).
The specificity range or odour tuning width of an ORN is largely determined by the olfactory
receptor that is expressed therein. An elegant method to study the function of an single OR-gene
is to delete the endogenous OR gene expressed in a specified neuron in the D. melanogaster
antenna and to replace it by a single exogenous OR gene (Dobritsa et al. 2003, Hallem et al. 2004a).
Not only the differences but also the overlap of the response spectra between different ORNs is
important information for the central nervous system to discriminate the quality of an odour
compound (De Bruyne et al. 2001, Den Otter and Van der Goes van Naters 1993, Sass 1978). These
ORN properties allow several degrees of freedom for odour coding. Based on the response spectra
of ORNs to a panel of odorant stimuli, ORNs and the sensilla containing them can be classified into
different functional groups (De Bruyne et al. 2001, Ghaninia et al. 2007, Qiu et al. 2006).

ORN responses belong to two major types: some odour stimuli elicit excitation whereas other
odours cause inhibition relative to the pre-stimulus activity (De Bruyne et al. 2001, Ghaninia et al.
2007, Qiu et al. 2006, Shields and Hildebrand 2001). This process is controlled by the same OR and
can be explained by a simple model (Hallem et al. 2004b). Without odour stimulus, ORs are present
as two forms in homeostasis: active and inactive. Once an OR binds with a ligand that elicits an
excitation response, the active form is stabilised and the firing frequency is increased; in case an
OR binds with a ligand that elicits an inhibition-type response, the inactive form is stabilised and
the firing frequency is decreased.

Olfaction in vector-host interactions 45

Yu Tong Qiu and Joop J.A. van Loon

The responses of ORNs to odorants are concentration-dependent above a threshold concentration,

and both absolute concentration and the change in concentration can be detected, provided that
the quality of the odorant is coded independently (Todd et al. 1992). It seems that pheromone-
sensitive-ORNs in lepidopterans are only sensitive to the concentration change (Kaissling 1998).
Carbon dioxide-sensitive receptor neurons of some insect species can detect both absolute and
relative concentrations (Bogner 1992, Bogner et al. 1986, Den Otter and Van der Goes van Naters
1992, Grant et al. 1995). More ORs are activated when exposed to higher concentrations of odours,
which might be a mechanism for insects to detect the absolute odour concentrations (Hallem et
al. 2004a).

In most cases two or more ORNs are co-compartmentalised in one sensillum providing the nervous
system with one more dimension for odour coding (Todd and Baker 1999). Such an arrangement
might be particularly important in the detection of odorant mixtures. The fact that sex pheromone-
sensitive receptor neurons in moths, which respond behaviourally to a species-specific ratio of
often two major components, are commonly housed in the same sensillum, implies a function of
co-compartmentalisation in discrimination of ratios between components of odour blends (Baker
et al. 1988, Cossé et al. 1995, Hansson et al. 1987, O’ Connell 1975).

Temporal characteristics of ORN responses to odours are another feature of odour coding, which is
especially important to enable odour discrimination while flying or walking upwind (Baker 1985,
Kennedy et al. 1981, Kramer 1992). A response is called phasic if the frequency of firing action
potentials decreases abruptly shortly after the onset of the excitation response. A tonic response
is characterised by an increase in firing frequency that outlasts the duration of stimulation. The
temporal response characteristics of an ORN to a certain stimulus seem largely independent of
dosage (De Bruyne et al. 2001, Qiu et al. 2006).

Natural odours utilised by mosquitoes

Flower- and honeydew-associated odours

To meet their energetic needs, both female and male mosquitoes of all physiological stages need
carbohydrates, obtained mainly from floral and extra-floral nectar and honeydew (Foster 1995
2008). Flower extracts and synthetic plant odours were shown to attract mosquitoes (Foster
2008, Foster and Hancock 1994, Hancock and Foster 1993, Healy and Jepson 1988, Jepson and
Healy 1988, Mauer et al. 1999). Information on the compounds in floral fragrances attractive to
mosquitoes is scarce (Jhumur et al. 2008). Floral fragrances are composed of a rich diversity of
compounds among which terpenes, phenols, aldehydes, fatty acid derivatives and benzenoids
are most likely attractants of mosquitoes (Foster 1995, Jhumur et al. 2008)

Human odours

Female mosquitoes are guided to their blood hosts by the physical and chemical cues emanating
from these hosts. Heat, moisture and visual cues from blood hosts are perceived by searching
female mosquitoes at a close range. Gillies and Wilkes (1968) reported that carbon dioxide could
attract mosquitoes at distances of 18-36 m and natural odours from a calf at distances of 54-73 m.

Human odour is composed of a complex of volatiles released from human skin and exhaled in
human breath. GC-MS analysis of human skin emanations collected on glass beads revealed 346
compound peaks (Bernier et al. 2000). More than 100 compounds were identified from exhaled

46  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

human breath (Krotoszynski et al. 1977). Ellin et al. (1974) analysed the composition of total human
body effluvia and identified 135 out of more than 300 compounds detected.

Human skin secretes higher amounts of lactic acid compared to other vertebrates, which compound
proved to be a mosquito attractant (Dekker et al. 2002, Smith et al. 1970). Lactic acid is attractive
by itself for Ae. aegypti (Smith et al. 1970) but not for An. gambiae, whereas a synergistic effect
was found between lactic acid, ammonia and a mixture of aliphatic carboxylic acids (Smallegange
et al. 2002). Braks and Takken (1999) reported that incubated human sweat was attractive to An.
gambiae and that most of this effect could be ascribed to the emission of ammonia produced
by microbial activity in sweat. When freshly produced sweat was incubated, the abundance of
other components, such as indole, 1-dodecanol, 6-methyl-5-hepten-2-one and geranyl acetone,
changed. Incubated sweat was found to be highly attractive whereas the attraction to fresh sweat
was weak (Meijerink et al. 2000).

Female mosquitoes were strongly attracted to nylon stockings worn by a human (Pates et al.
2001). GC-MS analysis of the extracts of Limburger cheese, which resembles, to a human nose,
the smell of unwashed human feet, revealed the major components to be carboxylic acids (Knols
et al. 1997). Forty compounds were identified from a diethyl-ether extract of human sweat, the
major components being aliphatic carboxylic acids (Healy and Copland 2000). Oxocarboxylic acids
were found in human blood and urine (Chalmers and Lawson 1982). Six oxocarboxylic acids were
reported to stimulate landing responses of An. gambiae (Healy et al. 2002). Chemical analysis
revealed that the typical human axillary odour is composed of C6 to C11 straight-chain, branched
and unsaturated acids (Zeng et al. 1991, 1996).

Oviposition-site related odours

Mosquitoes rely on olfactory, optical and tactile cues for locating oviposition sites (Beehler et al.
1993, Dhileepan 1997). Substances originating from mosquito larvae, pupae and eggs have been
found attractive to gravid females (reviewed by Bentley and Day 1989, Blackwell and Johnson
2000, Chadee 1993, Zahiri et al. 1997).

Mosquito breeding sites produce semiochemicals, mainly of microbial origin, that stimulate
mosquito oviposition (Beehler et al. 1994a, Hasselschwert and Rockett 1988, Hazard et al. 1967).
Infusions from decaying wood were found attractive to gravid Ae. triseriatus (Say) and an active
component was identified as p-cresol (4-methylphenol) (Bentley et al. 1979, 1981). Grass infusions
were shown to contain oviposition stimuli for Culex mosquitoes, the attractive compounds include,
among others, 3-methylindole, 4-methylphenol and indole (Du and Millar 1999, Millar et al. 1992,
Mboera et al. 2000, Lindh et al. 2008). The behavioural responses of mosquitoes to olfactory cues
is reviewed in detail in the laboratory (Chapter 7), and in the (semi-)field (Chapter 8) elsewhere
in this Volume.

Response spectra of mosquito ORNs

The response characteristics of ORNs are determined by the ORs expressed therein (Hallem et
al. 2004b). Normally ORNs in morphologically different sensilla have different response spectra
indicating that they express different ORs. Moreover, sensilla that appear morphologically similar
may house ORNs with different response spectra.

Olfaction in vector-host interactions 47

Yu Tong Qiu and Joop J.A. van Loon

ORNs can be classified according to their spectrum of response specificity to a panel of odour
stimulants. The selection of the odour panel has a direct bearing on the classification. An ideal
odour panel should contain the best stimulant of each ORN under investigation. Unfortunately,
which stimulants are the best for the ORNs is unknown beforehand and in fact it is the central
research question that first needs to be answered. In most studies, odours that are known to elicit
behavioural activities and odours that typically occur in volatile blends emitted by nectar, hosts
or oviposition sites are included in the test panel. However, these odours are not necessarily the
best stimulants for the ORNs. For example, the most potent stimulant of the C neuron innervating
capitate peg sensilla on the maxillary palps of female An. gambiae is 2,4,5-trimethylthiazole which
is a flavour compound and has not previously been identified as a mosquito infochemical (Lu et
al. 2007).

Anopheline and culicine mosquitoes show similarities as well as differences in the subtypes
of olfactory sensilla and the function of ORNs. We summarised the results of several electro­
physiological studies on different morphological subtypes of sensilla trichodea and grooved peg
sensilla in both mosquito families in an attempt to make a comparison (Table 2). For examples,

Table 2. Function of ORNs in sensilla trichodea and grooved peg sensilla in the mosquito Subfamilies Culicinae
and Anophelinae.

Mosquito subfamily Sensilla Subtype Functional types

Culicinae trichodea short sharp-tipped

sst1

sst2
sst3

sst4

long sharp-tipped
blunt I (long)

sbtI1

sbtI2

blunt II (short)

sbtII1
sbtII2
grooved peg lactic acid excited
lactic acid non-responding

lactic acid inhibited

48  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

ORNs sensitive to ammonia and lactic acid are found in grooved peg sensilla in both families;
indole-sensitive ORNs are found in short sharp-tipped sensilla trichodea; four functional types
have been found in the short sharp-tipped ST in Ae. aegypti, whereas two were found in a similar
morphological type of ST in An. gambiae. This comparison is necessarily limited at this time because
(1) not all sensillar sub-types have been studied; (2) in anophelines three additional morphological
ST-subtypes are present; (3) disparate odour panels have been used in different studies.

Sensilla trichodea (ST)

Among the five subtypes of ST (A-E) of anophelines only the medium-length sharp-tipped
subtype C and short sharp-tipped subtype E have been well studied for their function. By testing
the response of olfactory sensilla to a panel of 44 compounds that are components of human
odour, plant volatile blends or derived from oviposition sites, Qiu et al. (2006) identified four
functional types among ST subtype C and two functional types among subtype E in female An.
gambiae. ‘Generalist’ ORNs that are tuned to a broad range of odours were found in ST subtype
E, whereas ‘moderate specialist’ ORNs that are tuned to a narrow range of odours were found

Compounds eliciting References

Excitation Inhibition

ethyl propanoate, 4-methylcyclohexanol Kuthiala et al. 1992

4-methylcyclohexanol; 2-butoxyethanol; indole; 2-3C acids Ghaninia et al. 2007
ethyl lactate
2-butoxyethanol; indole Ghaninia et al. 2007
4-methylcyclohexanol; 2-butoxyethanol; α-thujone; 3C, 6C acids Ghaninia et al. 2007
ethyl butarate; ethyl propionate
4-methylcyclohexanol; 2-butoxyethanol; ethyl-L- several fatty acids and esters Ghaninia et al. 2007
lactate of fatty acids
2-10C acids volatile oils Lacher 1967
7-10C acids; volatile oils; eugenol; citronellol 2-5C acids; eugenol; Lacher 1967
citronellol
4-methylcyclohexanol; 2-butoxyethanol; α-thujone; Ghaninia et al. 2007
α-pinene; acetic acid; ethyl lactate; methyl-& ethyl
butyrate
ethyl butyrate; ethyl-L-lactate; ethyl propionate; Ghaninia et al. 2007
2-butoxyethanol
esters of 2-4C acids; 1-octen-3-ol; 2-butoxyethanol; 2-3C acids
4methylphenol; α-pinene
2-3C acids; indole Ghaninia et al. 2007
indole propionic acid Ghaninia et al. 2007
lactic acid; 2-3C acids; fluoro-lactate; ammonia DEET; fluoro-lactate Davis and Sokolove 1976
pyruvic acid; ammonia 2-3C acids; pyruvic acid; ethyl Davis and Sokolove 1976
lactate; 2-butoxyethanol
ammonia; water; pyruvic acid; acetic acid lactic acid; DEET; fluro-lactate; Davis and Sokolove 1976
pyruvic acid; 2-3C acids

Olfaction in vector-host interactions 49

Yu Tong Qiu and Joop J.A. van Loon

Table 2. Continued.

Mosquito subfamily Sensilla Subtype Functional types

Anophelinae trichodea C (medium-length sharp- TC1

tipped) TC2
TC3
TC4
E (short sharp-tipped)
TE1

TE2

C or E

grooved peg GP1

GP2
GP3

GP4

GP5

in subtype C, with one ‘specialist’ ORN tuned to only geranyl acetone. Phenols were among the
most effective stimulants for several neuron types belonging to different functional classes in
both subtype C and E. Subtype E, but not subtype C, also houses neurons responding to aliphatic
carboxylic acids, oxocarboxylic acids as well as 1-octen-3-ol and its homologs. ORNs responding
to 4-methylcyclohexanol and 4-methylphenol were found in short sharp-tipped ST of An. stephensi
(Bentley et al. 1982).

The function of ORNs in ST with various shapes was reported for several species of culicines
(Bentley et al. 1979, Bowen 1990, Davis 1976, Klowden and Blackmer 1987, Kuthiala et al. 1992,
Lacher 1967) (Table 2). A recent study by Ghaninia et al. (2007) classified 11 functional groups
within ST using a panel of 16 compounds from six different chemical groups (Table 2). The short
sharp-tipped ST contains at least one neuron that is sensitive to 2-butoxy ethanol, which is a
component of plant volatiles and attracts Culex quinquefasciatus Say (http: //www.pherobase.
net/database/species/species-Culex-.php, accessed 25 August 2009). ORNs in the short sharp-
tipped ST respond strongly to indole, 4-methylcyclohexanol and esters of short-chain carboxylic
acids; these compounds stimulate oviposition by gravid culicine mosquitoes (Allan and Kline
1995, Beehler et al. 1994b, Bentley et al. 1982, Perry and Fay 1967), whereas indole is also a major

50  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

Compounds eliciting References

Excitation Inhibition

ammonia; phenols Qiu et al. 2006

geranyl acetone Qiu et al. 2006
indole; phenols Qiu et al. 2006
ammonia Qiu et al. 2006
4-methylcyclohexanol; 4-methylphenol Bentley et al. 1982
indole; phenols; 1-hexen-3-ol; 5-7C acids; 3-methyl- 3-4C acids Qiu et al. 2006
1-butanol, ammonia
4-methylphenol; 4-ethylphenol; geranyl acetone; Qiu et al. 2006
2-nonanone; 1-hexen-3-ol; 1-hepten-3-ol; 1-octen-
3-ol; 7-octenoic acid; 3-methyl-2-hexenoic acid;
3C, 3M4C,6C, 9C, 10C, 12C acids; oxo-4-5C acids;
ammonia
geranyl acetone, 3-methyl-1-butanol, sulcatone, 2-6C acids Meijerink and van Loon
ammonia, 2-6C acids; 1-octen-3-ol; 3- & 4-methyl 1999, Meijerink et al.
phenol 2001, Van den Broek and
Den Otter 1999
ammonia; 4-5C amines Qiu et al. 2006
ammonia; 4-5C amines; oxo-4-5C acids Qiu et al. 2006
ammonia; 1-butylamine; 2-oxo-butanoic acid; lactic Qiu et al. 2006
acid
ammonia; 1-butylamine; 2-oxo-butanoic acid; lactic 3-6C and 9C acids Qiu et al. 2006
acid
ammonia; 4-5C amines; 4-6C acids Qiu et al. 2006

component of aged human sweat (Meijerink et al. 2000). Short blunt-tipped ST contain neurons
sensitive to esters of short-chain carboxylic acids, 4-methylphenol and alcohols, the former two
types of compounds were reported to attract gravid mosquitoes and considered to be oviposition
stimulants (Bentley et al. 1979, Davis 1976, Klowden and Blackmer 1987). An ORN strongly tuned to
the plant volatile, α-thujone, is also found in the short blunt-tipped ST (Ghaninia et al. 2007). The
long blunt-tipped STs were found responding to C2-C5 carboxylic acids by inhibition, but to C7-
C10 by excitation; they showed excitatory responses to terpineol but were inhibited by eugenol
and citronellol, the latter three being terpenoids of plant origin (Lacher 1967). Except for some
responses to C2-C10 carboxylic acids recorded by Lacher (1967), long sharp-tipped STs were not
responsive to the compounds that have been tested in the study by Ghaninia et al. (2007).

Grooved peg sensilla

Grooved peg sensilla in An. gambiae and An. quadriannulatus Theobald were reported to respond
to polar compounds such as ammonia, lactic acid, acetone, butylamine as well as to complex
natural odour mixtures such as incubated sweat, cow odour and Limburger cheese odour, which

Olfaction in vector-host interactions 51

Yu Tong Qiu and Joop J.A. van Loon

is perceived by the human nose to resemble human foot odour (Meijerink et al. 2001, Van den
Broek and Den Otter 1999).

Qiu et al. (2006) found that all sensilla they recorded from in An. gambiae contained at least one
neuron that responded to ammonia and 1-butylamine. ORNs in GP sensilla were found responding
to oxocarboxylic acids, short-chain carboxylic acids and lactic acid, compounds that are shown to
attract mosquitoes or to enhance landing responses (Braks et al. 2001, Healy and Copland 2000,
Knols et al. 1997, Smallegange et al. 2005).

Similar as for anophelines, grooved peg sensilla in culicine mosquitoes have also been found
responding to short-chain carboxylic acids but hardly to plant and oviposition site-associated
volatiles (Davis 1988, Davis and Sokolove 1976). However, the responses of the GP sensilla in Ae.
aegypti to ammonia and lactic acid are strikingly different from the responses recorded from the
homologous sensilla in An. gambiae. Almost all the GP sensilla in An. gambiae contain ORNs that
were excited by ammonia, whereas only 8% of those in Ae. aegypti responded to ammonia (Davis
and Sokolove 1976, Qiu et al. 2006). In contrast, the LA-excited neuron population in Ae. aegypti
was present in a higher proportion of grooved peg sensilla (58.8%) than that in An. gambiae
(40%) (Qiu et al. 2006). The difference in sensitivity of peripheral neurons to ammonia and lactic
acid reflects the different roles of these two compounds in the behaviour of these two mosquito
species: for An. gambiae ammonia alone is attractive to the females, but lactic acid alone has a
negligible effect; while for Ae. aegypti the relative importance of the two compounds in attraction
appears reversed (Steib et al. 2001, Geier et al. 1999, Smallegange et al. 2005).

Labellar sensilla

Because the characteristic co-receptor AgOR7 was found to be expressed in T2 sensilla on the
proboscis of An. gambiae and Ae. aegypti, it has been hypothesised that the proboscis of these
mosquitoes also functions as an olfactory organ (Melo et al. 2004, Pitts et al. 2004). Subsequent
studies by Kwon et al. (2006) tested the response of ORNs in T2 sensilla on the proboscis of
female An. gambiae to a range of compounds and found that these neurons were responding
to butylamine, acetic acid, oxo-carboxylic acids and ketones with a heterocyclic ring, including
acetothiophene, acetylpyridine, acetylthiazole, and acetylphenone. In situ hybridisation revealed
the co-expression of the conventional AgOR6 localised in sensilla T2.

Capitate peg sensilla

For blood feeding mosquitoes and biting midges (Diptera: Ceratopogonidae), a receptor neuron
sensitive to CO2 was found in sensilla basiconica on the maxillary palps (Grant and Kline 2003, Grant
and O’Connell 1996, Grant et al. 1995, Kellogg 1970). Because almost all the sensilla containing
CO2 receptors are innervated by one neuron with an expanded lamellar structure, it has been
postulated that this is the neuron that is sensitive to CO2. The capitate peg sensilla of mosquitoes
also house a neuron that is highly sensitive to 1-octen-3-ol (Grant and O’Connell 1996).

The function of the three neurons innervating capitate sensilla on the maxillary palps of
An. gambiae was studied by Lu et al. (2007). Activity of three neurons has been identified. A
neuron from which the largest spike amplitude was recorded responded to CO2 over a narrow
concentration range. This neuron showed excitatory responses to several heterocyclic compounds
and inhibitory responses to several compounds including indole. The B-neuron was extremely
sensitive to 1-octen-3-ol, it also responded to several analogues of 1-octen-3-ol, but with lower

52  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

sensitivity, and several ketones including 6-methyl-5-hepten-2-one. The C neuron in the capitate
sensilla was broadly tuned to heterocyclic compounds, ketones and alcohols. In situ hybridisation
indicated the coexpression of three AgGRs (AgGR22, 23 and 24) in neuron A and the co-receptor
AgOR7 was expressed together with AgOR8 in the B neuron and with AgOR28 in the C neuron.
Expression of these AgORs in heterologous systems verified that the response spectra of the
individual receptors matched the in vivo responses of the three neurons.

Effects of physiological state

The behaviour of a mosquito is determined by the prevalent physiological state. For example, a
female mosquito does not take her first blood meal before she reaches a certain age; after a blood
meal has been taken, a female mosquito will not respond to host-derived cues for the next 48 to
72 hours until the eggs have matured, in due time she needs to find a suitable oviposition site
to deposit the eggs. Does the peripheral olfactory system respond to changes in physiological
state? Would mating, age, blood meal, gonotrophic cycle, body size and infection state affect
the response of ORNs to odours? Several studies addressed these questions and provided some
answers, whereas most of the questions remain unanswered.

Age

Davis (Davis 1984a) reported that the sensitivity of grooved peg sensilla in Ae. aegypti increased
with increasing post-emergence time, which was in accordance with increased host-seeking
activity. By comparing the response of the CO2 receptor neuron in capitate peg sensilla of Ae.
aegypti, Grant and O’Connell (2007) found that newly emerged female mosquitoes were much less
sensitive to CO2 than older females, whereas for males no such pronounced difference was found.
The CO2 sensitivity was in accordance with the host-seeking activity.

Blood meal and oviposition

Lactic acid sensitivity was found to be down-regulated after female mosquitoes took a blood
meal, suggesting an involvement of the peripheral olfactory system in the modulation of foraging
behaviour of Ae. aegypti (Davis 1984b). After oviposition, the lactic acid sensitivity returns to the
pre-blood-fed level. Davis and Takahashi (1980) found that ORNs in blunt-tipped type II sensilla
of gravid females were more sensitive to oviposition site-related compounds, namely methyl- and
ethyl-esters of C2-C4 carboxylic acids, than the non-gravid females.

After a blood meal, a new functional type of sensilla trichodea E was found for An. gambiae, which
was highly sensitive to indole and 3-methyl indole, and sensitive to C5-C9 carboxylic acids as
well as 7-octenoic acid, which is a component of human axillary odours (Qiu et al. 2006, Zeng et
al. 1991). Increasing sensitivity to these compounds might facilitate female mosquitoes to locate
oviposition sites and hosts in the next gonotrophic cycle more efficiently. The overall sensitivity
of ORNs in gravid mosquitoes to ammonia and phenols was lower (Qiu et al. 2006), supporting a
role in the observed suppression of host-seeking behaviour. Up to 435 gene products are either
up- or down-regulated after the ingestion of a blood meal by female An. gambiae (Holt et al. 2002,
Ribeiro 2003). One of the putative OR genes (AgOR1) was strongly down regulated 12 h after blood
feeding (Fox et al. 2001). When AgOR1 was expressed in a Drosophila ORN of which the original OR
was deleted (‘empty neuron’), it was found specifically responsive to 4-methylphenol (Hallem et al.
2004a). The overall decrease of sensitivity to phenols found by Qiu et al. (2006) might be caused
by the down-regulation of these and similar genes.

Olfaction in vector-host interactions 53

Yu Tong Qiu and Joop J.A. van Loon

Diapause

The lactic acid-sensitive neurons found in active females do not respond in the diapausing females
of Cx. pipiens Linnaeus, whereas neurons that are sensitive to putative oviposition stimulants are
unaffected during diapause (Bowen et al. 1988). A subsequent study showed that in female Cx.
pipiens that had terminated diapause high sensitivity to lactic acid was restored.

Translation of physiology to behaviour

Input-output relations

As discussed above, mosquito ORNs have varied specificity spectra and sensitivity ranges. Across-
fibre patterning is the most plausible coding principle operating in the mosquito olfactory system
(Ghaninia et al. 2007, Qiu et al. 2006). What conclusions can we draw from the electrophysiological
studies in terms of their behavioural relevance? Can we predict which compound is attractive or
repellent to mosquitoes based on our electrophysiological data? Can we say anything at all about
the behavioural relevance of a compound or compound mixture after the electrophysiological
activity these elicit has been recorded? Does the response intensity of ORNs correlate with
behavioural response? Has one compound a higher behavioural impact when more ORNs are
tuned to it or when it is tuned to only by few specific neurons? Do excitatory and inhibitory
responses have a different behavioural effect? We must establish that we have only preliminary
answers to some of these questions. Correlations between ORN specificity and sensitivity to a
compound and behavioural significance of this compound have been found in several cases, but
a full picture is still out of reach. Several examples are given below.

In all the tested species CO2 receptors are found in the capitate peg sensilla on the maxillary
palps (Grant et al. 1997, Grant and O’ Connell 1996, Lu et al. 2007). As expected, CO2 is indeed
generally attractive to mosquitoes (Dekker et al. 2001, Gillies 1980, Mboera and Takken 1997, Qiu et
al. 2007). It is noteworthy that the CO2 -sensitive neurons respond within a narrow concentration
range and the response intensity is not particularly high. An ORN responding to 1-octen-3-ol is
co-compartmentalised in the same capitate pegs as the CO2 receptor neurons. The sensitivity to
1-octen-3-ol is exceptionally high compared to other ligands, and the response intensity is much
higher than observed for the CO2-sensitive neuron. Yet in behavioural tests, 1-octen-3-ol showed
much more variable effects to various mosquito species and on its own is rarely attractive to
mosquitoes (Kline 1994).

For ammonia, an abundant component of incubated sweat, a better match between

electrophysiology and behaviour has been found. Ammonia elicits excitatory responses in all
functional types of grooved peg sensilla and some types of sensilla trichodea in An. gambiae
(Meijerink et al. 2001, Qiu et al. 2006), which finding nicely correlates with the reliable attraction of
female An. gambiae to sources of this compound (Braks et al. 2001, Smallegange et al. 2005). Lactic
acid has a similar behavioural effect on female Ae. aegypti as ammonia on female An. gambiae,
and neurons that are sensitive to this compound are found in the majority of the grooved peg
sensilla, as mentioned before.

N,N-diethyl-3-methylbenzamide (DEET) is a strong repellent for many species of mosquitoes

and several electrophysiological studies attempted to reveal its mode of action. It was found
in Ae. aegypti that blunt-tipped type I ST contain neurons that can be excited by DEET (Lacher
1967). Other studies suggested that DEET makes female mosquitoes anosmic to lactic acid (Davis

54  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

1985, Davis and Sokolove 1976) and oviposition stimulants (Kuthiala et al. 1992). A recent study
showed that DEET reduced the sensitivity of a neuron expressing AgOR8 responsive to 1-octen-3-
ol (Ditzen et al. 2008). More recently, however, Syed and Leal (Syed and Leal 2008) showed that the
decrease of 1-octen-3-ol activity is due to the adsorbing properties of DEET, as deduced from the
observation that when the two compounds were delivered separately the response to 1-octen-
3-ol was unaffected. These authors also found DEET sensitive neurons on the short sharp-tipped
ST in Cx. quinquefasciatus.

Conclusions and outlook

Understanding the chemical ecology of natural odours mediating mosquito behaviours is

extremely important. Many natural odours from different sources share the same chemical
components. It is the combination of different components and the ratios between the amounts
of these components that determine characteristic profiles of natural odours. For example,
carboxylic acids, indoles and phenols are found among plant volatiles, human emanations and
oviposition site odours alike. From behavioural studies it has become clear that not a single VOC,
but a subset of VOCs (the most effective composition of which is as yet unknown) as occurring in
the complex compound mixtures that natural sources provide, form an ensemble code that allows
the olfactory system to discriminate between a nectar source, a host body or an oviposition site.
Components in a mixture may synergise or antagonise the activity of each other already at the
peripheral level (Ochieng et al. 2002). As an example, it will be of considerable interest to analyse
the possible peripheral interactions occurring between ammonia, lactic acid and carboxylic acids
when these are offered as different blends of these three components that have been shown to
cause behavioural synergism (Smallegange et al. 2005). Electrophysiological studies on volatile
blends are rare, as a strong bias for single compound stimulations has been noted. Subsequent
processing of all olfactory input takes place in the olfactory lobe of the brain to either generate
or suppress a behavioural response (Chapter 4, this volume).

The amount of electrophysiological data on ORN-functional types has grown over the past decade
but is still modest and indeed far from exhaustive. The large number of sensilla (Table 1) and even
more so, ORNs, on mosquito antennae and the absence of regionalised sensillar distributions
make it difficult to carry out an exhaustive inventory. It is to be expected that a larger sample
size of the total ORNs studied enhances the chance of identifying additional functional classes.
Thus, increasing this sample size is still a first and foremost but also tedious task, yet essential to
increase the likelihood that the sample attains a size representative for the total ORN population.
The availability of cloned OR-genes allows for screening of their ligand-specificity spectra gene
by gene using heterologous platforms. This necessitates close collaboration between molecular
biologists and electrophysiologists as for verification of the data thus obtained in the donor
mosquito species itself, the use of SSR-techniques remains a crucial final step.

Looking at mosquito olfaction from the perspective of the complex chemistry of natural VOC-
mixtures released by hosts, oviposition sites and nectar- and honeydew sources, the number
of VOCs tested on the full array of ORN functional types of any mosquito species is still far from
exhaustive and rarely exceeds 100 VOCs. Increasing the panel of VOCs is likely to reveal new
functional ORN-types. Expanding this panel to the 300 or so VOCs that are produced by the human
body is one goal, but including non-natural analogs and indeed, synthetic compounds, is bound
to aid our insight in ORN specificity and olfactory coding in blood feeding vector mosquitoes. The
predictability of olfaction-mediated behavioural responses based on quantified olfactory input is

Olfaction in vector-host interactions 55

Yu Tong Qiu and Joop J.A. van Loon

a real scientific challenge for any sensory physiologist and bears particular societal relevance in
case of mosquito species that are such threatening disease vectors.

References

Allan SA and Kline DL (1995) Evaluation of organic infusions and synthetic compounds mediating oviposition in Aedes
albopictus and Aedes aegypti (Diptera: Culicidae). J Chem Ecol 21: 1847-1860.
Baker TC (1985) Chemical control of behavior. Pergamon Press, Oxford, UK.
Baker TC, Hansson BS, Löfstedt C and Löfqvist J (1988) Adaptation of antennal neurons in moths is associated with
cessation of pheromone-mediated upwind flight. PNAS 85: 9826-9830.
Beehler JW, Millar JG and Mulla MS (1993) Synergism between chemical attractants and visual cues influencing
oviposition of the mosquito, Culex quinquefasciatus (Diptera: Culicidae). J Chem Ecol 19: 635-644.
Beehler JW, Millar JG and Mulla MS (1994a) Protein hydrolysates and associated bacterial contaminants as oviposition
attractants for the mosquito Culex quinquefasciatus. Med Vet Entomol 8: 381-385.
Beehler JW, Millar JG and Mulla MS (1994b) Field evaluation of synthetic compounds mediating oviposition in Culex
mosquitos (Diptera, Culicidae). J Chem Ecol 20: 281-291.
Bentley MD, McDaniel IN, Yatagai M, Lee H-P and Maynard R (1979) p-Cresol: an oviposition attractant of Aedes triseriatus.
Environ Entomol 8: 206-209.
Bentley MD, McDaniel IN, Yatagai M, Lee HP and Maynard R (1981) Oviposition attractants and stimulants of Aedes
triseriatus (Say) (Diptera: Culicidae). Environ Entomol 10: 186-189.
Bentley MD, McDaniel IN and Davis EE (1982) Studies of 4-methylcyclohexanol: an Aedes triseriatus (Say)(Diptera:
Culicidae) oviposition attractant. Environ Entomol 19: 589-592.
Bentley MD and Day JF (1989) Chemical ecology and behavioral aspects of mosquito oviposition. Annu Rev Entomol
34: 401-421.
Bernier UR, Kline DL, Barnard DR, Schreck CE and Yost RA (2000) Analysis of human skin emanations by gas
chromatography/mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for
yellow fever mosquito (Aedes aegypti). Anal Chem 72: 747-756.
Blackwell A and Johnson SN (2000) Electrophysiological investigation of larval water and potential oviposition
chemoattractants for Anopheles gambiae sensu stricto. Ann Trop Med Parasitol 94 389-398.
Boeckh J and Ernst K-D (1983) Olfactory food and mate recognition. Springer, Berlin, Heidelberg, Germany.
Bogner F, Boppré M, Ernst K-D and Boeckh J (1986) CO2 sensitive receptors on labial palps of Rhodogastria moths
(Lepidoptera: Arctiidae): physiology, fine structure and central projection. J Comp Physiol [A] 158: 741-749.
Bogner F (1992) Response properties of CO2-sensitive receptors in tsetse flies (Diptera: Glossina palpalis). Physiol
Entomol 17: 19-24.
Bohbot J, Pitts RJ, Kwon HW, Rutzler M, Robertson HM and Zwiebel LJ (2007) Molecular characterization of the Aedes
aegypti odorant receptor gene family. Insect Mol Biol 16: 525-537.
Boo KS (1980a) Fine structure of the antennal sensory hairs in female Anopheles stephensi. Z Parasitenkd 61: 161-171.
Boo KS (1980b) Antennal sensory receptors of the male mosquito, Anopheles stephensi. Z Parasitenkd 61: 249-264.
Bowen MF, Davis EE and Haggart DA (1988) A behavioral and sensory analysis of host-seeking behavior in the diapausing
mosquito Culex pipiens. J Insect Physiol 34: 805-813.
Bowen MF (1990) Post-diapause sensory responsiveness in Culex pipiens. J Insect Physiol 36: 923-929.
Bowen MF (1995) Sensilla basiconica (grooved pegs) on the antennae of female mosquitos - electrophysiology and
morphology. Entomol Exp Appl 77: 233-238.
Braks MAH and Takken W (1999) Incubated human sweat but not fresh sweat attracts the malaria mosquito Anopheles
gambiae sensu stricto. J Chem Ecol 25: 663-672.
Braks MAH, Meijerink J and Takken W (2001) The response of the malaria mosquito, Anopheles gambiae, to two
components of human sweat, ammonia and L-lactic acid, in an olfactometer. Physiol Entomol 26: 142-148.
Cator LJ, Arthur BJ, Harrington LC and Hoy RR (2009) Harmonic convergence in the love songs of the dengue vector
mosquito. Science 323: 1077-1079.

56  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

Chadee DD (1993) Oviposition response of Aedes aegypti (L.) to the presence of conspecific eggs in the field in Trinidad,
W.I. J Am Mosq Control Assoc 64: 63-66.
Chalmers RA and Lawson AM (1982) Organic acids in man. Chapman & Hall, London, UK.
Clements AN (1999) Adult integumental sensilla: their structure, physiology and connections with the brain. In: Clements
AN (ed) The biology of mosquitoes. Vol. 2: Sensory reception and behaviour. CAB International, Wallingford, UK,
pp 8-54.
Cossé AA, Campbell MG, Glover TJ, Linn CEJ, Todd JL, Baker TC and Roelofs WL (1995) Pheromone behavioral responses
in unusual male European corn borer hybrid progeny not correlated to electrophysiological phenotypes of their
pheromone-specific antennal neurons. Experientia 51: 809-816.
Davis EE and Sokolove PG (1975) Temperature responses of antennal receptors of the mosquito Aedes aegypti. J Comp
Physiol [A] 96: 223-236.
Davis EE (1976) Receptor sensitive to oviposition site attractants on antennae of mosquito, Aedes aegypti. J Insect
Physiol 22: 1371-1376.
Davis EE and Sokolove PG (1976) Lactic acid-sensitive receptors on antennae of mosquito, Aedes aegypti. J Comp Physiol
[A] 105: 43-54.
Davis EE and Takahashi FT (1980) Hormonal modification of chemoreceptor sensitivity in an insect. Am Zool 20: 936-936.
Davis EE (1984a) Development of lactic acid receptor sensitivity and host-seeking behavior in newly emerged female
Aedes aegypti mosquitoes. J Insect Physiol 30: 211-215.
Davis EE (1984b) Regulation of sensitivity in the peripheral chemoreceptor systems for host-seeking behavior by a
hemolymph-borne factor in Aedes aegypti. J Insect Physiol 30: 179-183.
Davis EE (1985) Insect repellents - concepts of their mode of action relative to potential sensory mechanisms in
mosquitos (Diptera, Culicidae). J Med Entomol 22: 237-243.
Davis EE (1988) Structure-response relationship of the lactic acid-excited neurones in the antennal grooved-peg sensilla
of the mosquito Aedes aegypti. J Insect Physiol 34: 443-449.
De Bruyne M, Foster K and Carlson JR (2001) Odor coding in the Drosophila antenna. Neuron 30: 537-552.
Dekker T, Takken W and Cardé RT (2001) Structure of host-odour plumes influences catch of Anopheles gambiae s.s. and
Aedes aegypti in a dual-choice olfactomer. Physiol Entomol 26: 124-134.
Dekker T, Steib B, Cardé RT and Geier M (2002) L-Lactic acid, a human-signifying host cue for the anthropophilic
mosquito Anopheles gambiae sensu stricto. Med Vet Entomol 16: 91-98.
Den Otter CJ and van der Goes van Naters WM (1992) Single cell recordings from tsetse (Glossina m. morsitans) antennae
reveal olfactory, mechano- and cold receptors. Physiol Entomol 17: 33-42.
Den Otter CJ and Van der Goes van Naters WM (1993) Responses of individual antennal olfactory cells of tsetse flies
(Glossina m. morsitans) to phenols from cattle urine. Physiol Entomol 18: 43-49.
Dhileepan K (1997) Physical factors and chemical cues in the oviposition behavior of arboviral vectors Cules annulirostris
and Culex molestus (Diptera: Culicidae). Environ Entomol 26: 318-326.
Ditzen M, Pellegrino M and Vosshall LB (2008) Insect odorant receptors are molecular targets of the insect repellent
DEET. Science 319: 1838-1842.
Dobritsa AA, van der Goes van Naters W, Warr CG, Steinbrecht RA and Carlson JR (2003) Integrating the molecular and
cellular basis of odor coding in the Drosophila antenna. Neuron 37: 827-841.
Du YJ and Millar JG (1999) Electroantennogram and oviposition bioassy responses of Culex quinquefasciatus and Culex
tarsalis to chemicals in odors from bermuda grass infusions. J Med Entomol 36: 158-166.
Ellin RI, Farrand RL, Oberst FW, Crouse CL, Billups NB, Koon WS, Musselman NP and Sidell FR (1974) An apparatus for the
detection and quantitation of volatile human effluents. J Chromatogr 100: 137-152.
Foster WA and Hancock RG (1994) Nectar-related olfactory and visual attractants for mosquitoes. J Am Mosq Control
Assoc 10: 288-296.
Foster WF (1995) Mosquito sugar feeding and reproductive energetics. Annu Rev Entomol 40: 443-447.
Foster WA (2008) Phytochemicals as population sampling lures. J Am Mosq Control Assoc 24: 138-146.
Fox AN, Pitts RJ, Robertson HM, Carlson JR and Zwiebel LJ (2001) Candidate odorant receptors from the malaria vector
mosquito Anopheles gambiae and evidence of downregulation in response to blood feeding. PNAS 98: 14693-14697.

Olfaction in vector-host interactions 57

Yu Tong Qiu and Joop J.A. van Loon

Geier M, Bosch OJ and Boeckh J (1999) Ammonia as an attractive component of host odour for the Yellow fever
mosquito Aedes aegypti. Chem Senses 24: 647-653.
Ghaninia M, Ignell R and Hansson BS (2007) Functional classification and central nervous projections of olfactory
receptor neurons housed in antennal trichoid sensilla of female yellow fever mosquitoes, Aedes aegypti. Eur J
Neurosci 26: 1611-1623.
Gillies MT and Wilkes TJ (1968) A comparison of the range of attraction of animal baits and of carbon dioxide for some
West African mosquitoes. Bull Entomol Res 59: 441-456.
Gillies MT (1980) The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review. Bull Entomol
Res 70: 525-532.
Gingl E, Hinterwirth A and Tichy H (2005) Sensory representation of temperature in mosquito warm and cold cells. J
Neurophysiol 94: 176-185.
Grant AJ, Wigton BE and Aghajanian JG (1995) Electrophysiological responses of receptor neurons in mosquito maxillary
palp sensilla to carbon dioxide. J Comp Physiol [A] 177: 389-396.
Grant AJ and O’ Connell RJ (1996) Electrophysiological responses from receptor neurons in mosquito maxillary palp
sensilla. In: Bock GR and Cardew G (eds), Olfaction in Mosquito Host Interactions Ciba Foundation Symposium 200,
Wiley, New York, USA, pp 233-248.
Grant AJ, Borroni PF and O’ Connell RJ (1997) Pulsed pheromone stimuli affect the temporal response of antennal
receptor neurones of the adult cabbage looper moth. Physiol Entomol 22: 123-130.
Grant AJ and Kline DL (2003) Electrophysiological responses from Culicoides (Diptera: Ceratopogonidae) to stimulation
with carbon dioxide. J Med Entomol 40: 284-293.
Hallem EA, Fox AN, Zwiebel LJ and Carlson JR (2004a) Mosquito receptor for human-sweat odorant. Nature 427: 212-213.
Hallem EA, Ho MG and Carlson JR (2004b) The molecular basis of odor coding in the Drosophila antenna. Cell 117:
965-979.
Hancock RG and Foster WA (1993) Effect of preblood-meal sugar on sugar seeking and upwind flight by gravid and
parous Aedes aegypti (Diptera: Culicidae). J Med Entomol 30: 353-359.
Hansson BS, Lofstedt C and Roelofs WL (1987) Inheritance of olfactory response to sex pheromone components in
Ostrinia nubilalis. Naturwissenschaften 74: 497-499.
Hasselschwert D and Rockett CL (1988) Bacteria as ovipositional attractants for Aedes aegypti (Diptera: Culicidae). The
Great Lakes Entomologist 21: 163-168.
Hazard EI, Mayer MS and Savage KE (1967) Attraction and oviposition stimulation of gravid female mosquitoes by
bacteria isolated from hay infusions. Mosquito News 27: 133-136.
Healy TP and Jepson PC (1988) The location of floral nectar sources by mosquitoes: the long-range responses of
Anopheles arabiensis Patton (Diptera: Culicidae) to Achillea millefolium flowers and isolated floral odour. Bull
Entomol Res 78: 651-657.
Healy TP and Copland MJW (2000) Human sweat and 2-oxopentanoic acid elicit a landing response from Anopheles
gambiae. Med Vet Entomol 14: 195-200.
Healy TP, Copland MJW and Cork A (2002) Landing responses of Anopheles gambiae elicited by oxocarboxylic acids.
Med Vet Entomol 16: 126-132.
Heinbockel T and Kaissling KE (1996) Variability of olfactory receptor neuron responses of female silkmoths (Bombyx
mori L.) to benzoic acid and (±)-linalool. J Insect Physiol 42: 565-578.
Hill CA, Fox AN, Pitts RJ, Kent LB, Tan PL, Chrystal MA, Cravchik A, Collins FH, Robertson HM and Zwiebel LJ, (2002)G
protein-coupled receptors in Anopheles gambiae. Science 298(5591): 176-178.
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides
R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P,
Atkinson PW, Baden H, de B, V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M, Cai S, Center
A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A, Dew I, Evans
CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun SL, Hogan JR,
Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin JJ, Lobo NF, Lopez
JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O’Brochta DA, Pfannkoch C, Qi R,

58  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J, Thomasova D, Ton
LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman JR, Wu M, Yao A,
Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della TA, Roth CW, Louis C, Kalush F, Mural RJ,
Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey PT, Venter JC, Weissenbach
J, Kafatos FC, Collins FH and Hoffman SL, (2002)The genome sequence of the malaria mosquito Anopheles gambiae.
Science 298(5591): 129-149.
Ismail IAH (1962) Sense organs in the antennae of Anopheles maculipennis atroparvus (V.Thiel) and their possible
function in relation to the attraction of female mosquito to man. Acta Trop 19: 1-58.
Jepson PC and Healy TP (1988) The location of floral nectar sources by mosquitoes: an advanced bioassay for volatile
plant odours and initial studies with Aedes aegypti (L.) (Diptera: Culicidae). Bull Entomol Res 78: 641-650.
Jhumur US, Dötterl S and Jürgens A (2008) Floral odors of Silene otites: their variability and attractiveness to mosquitoes.
J Chem Ecol 34: 14-25.
Kaissling KE (1998) Flux detectors versus concentration detectors: two types of chemoreceptors. Chem Senses 23:
99-111.
Keil TA and Steinbrecht RA (1987) Diffusion barriers in silkmoth sensory epithelia: application of lanthanum tracer to
olfactory sensilla of Antheraea polyphemus and Bombyx mori. Tissue Cell 19: 119-134.
Keil TA (1999) Morphology and development of the peripheral olfactory organs. In: Hansson BS (ed) Insect olfaction.
Springer-Verlag, Berlin, Germany, pp 5-47.
Kellogg FE (1970) Water vapour and carbon dioxide receptors in Aedes aegypti. J Insect Physiol 16: 99-108.
Kennedy JS, Ludlow AR and Sanders CJ (1981) Guidance of flying male moths by wind-borne sex pheromone. Physiol
Entomol 6: 395-412.
Klein U (1987) Sensillum-lymph proteins from antennal olfactory hairs of the moth (Saturniidae). Insect Biochemistry
17: 1193-1204.
Kline DL (1994) Olfactory attractants for mosquito surveillance and control: 1-octen-3-ol. J Am Mosq Control Assoc
10: 280-287.
Klowden MJ and Blackmer JL (1987) Humoral control of preoviposition behavior in the mosquito, Aedes aegypti. J Insect
Physiol 33: 689-692.
Knols BGJ, van Loon JJA, Cork A, Robinson RD, Adam W, Meijerink J, de Jong R and Takken W (1997) Behavioural
and electrophysiological responses of the female malaria mosquito Anopheles gambiae (Diptera: Culicidae) to
Limburger cheese volatiles. Bull Entomol Res 87: 151-159.
Kramer E (1992) Attractivity of pheromone surpassed by time-patterned application of two nonpheromone compounds.
J Insect Behav 5: 83-97.
Krotoszynski B, Gabriel G and O’Neill H (1977) Characterization of human expired air: A promising investigative and
diagnostic technique. J Chromatogr Sci 15: 239-244.
Kuthiala A, Gupta RK and Davis EE (1992) Effect of the repellent deet on the antennal chemoreceptors for oviposition
in Aedes aegypti (Diptera: Culicidae). J Med Entomol 29: 639-643.
Kwon HW, Lu T, Rutzler M and Zwiebel LJ (2006) Olfactory responses in a gustatory organ of the malaria vector mosquito
Anopheles gambiae. PNAS 103: 13526-13531.
Lacher V (1967) Elektrophysiologische Untersuchungen an einzelnen Geruchsrezeptoren auf den Antennen weiblicher
Moskitos (Aedes aegypti L.). J Insect Physiol 13: 1461-1470.
Liliefors T, Thelin B and Van der Pers JNC (1985) Chain-elongated analogues of a pheromone component of the turnip
moth, Agrotis segetum. A structure-activity study using molecular mechanics. J Royal Chem Soc, Perkin Transactions
II 1957-1962.
Liliefors T, Bengtsson BO and Hansson BS (1987) Effects of double-bond configuration on interaction between a moth
sex pheromone component and its receptor: a receptor-interaction model based on molecular mechanics. J Chem
Ecol 13: 2023-2040.
Lindh JM, Kannaste A, Knols BGJ, Faye I and Borg-Karlson AK (2008) Oviposition responses of Anopheles gambiae s.s.
(Diptera: Culicidae) and identification of volatiles from bacteria-containing solutions. J Med Entomol 45: 1039-1049.

Olfaction in vector-host interactions 59

Yu Tong Qiu and Joop J.A. van Loon

Lu T, Qiu YT, Wang G, Kwon JY, Rutzler M, Kwon HW, Pitts RJ, van Loon JJA, Takken W, Carlson JR and Zwiebel LJ (2007)
Odor coding in the maxillary palp of the malaria vector mosquito Anopheles gambiae. Curr Biol 17: 1533-1544.
Mboera LEG and Takken W (1997) Carbon dioxide chemotropism in mosquitoes (Diptera: Culicidae) and its potential in
vector surveillance and management programmes. Med Vet Entomol 85: 355-368.
Mboera LEG, Takken W, Mdira KY, Pickett JA (2000) Sampling gravid Culex quinquefasciatus (Diptera: Culicidae) in
Tanzania with traps baited with synthetic oviposition pheromone and grass infusions. J Med Entomol 37: 172-176.
Mauer DJ and Rowley WA (1999) Attraction of Culex pipiens pipiens (Diptera: Culicidae) to flower volatiles. J Med Entomol
36: 503-507.
McIver S and Siemicki R (1975) Palpal sensilla of selected Anopheline mosquitoes. J Parasitol 61: 535-538.
McIver S (1982) Review article: Sensilla of mosquitoes (Diptera: Culicidae). J Med Entomol 19: 489-535.
Meijerink J and van Loon JJA (1999) Sensitivity of antennal olfactory neurons of the malaria mosquito, Anopheles
gambiae, to carboxylic acids. J Insect Physiol 45: 365-373.
Meijerink J, Braks MAH, Brack AA, Adam W, Dekker T, Posthumus MA, van Beek TA and van Loon JJA (2000) Identification
of olfactory stimulants for Anopheles gambiae from human sweat samples. J Chem Ecol 26: 1367-1382.
Meijerink J, Braks MAH and van Loon JJA (2001) Olfactory receptors on the antennae of the malaria mosquito Anopheles
gambiae are sensitive to ammonia and other sweat-borne components. J Insect Physiol 47: 455-464.
Melo ACA, Rutzler M, Pitts RJ and Zwiebel LJ (2004) Identification of a chemosensory receptor from the yellow fever
mosquito, Aedes aegypti, that is highly conserved and expressed in olfactory and gustatory organs. Chem Senses
29: 403-410.
Millar JG, Chaney JD and Mulla MS (1992) Identification of oviposition attractants for Culex quinquefasciatus from
fermented bermuda grass infusions. J Am Mosq Control Assoc 8: 11-17.
Mustaparta H (2002) Encoding of plant odour information in insects: peripheral and central mechanisms. Entomol Exp
Appl 104: 1-13.
O’ Connell RJ (1975) Olfactory receptor responses to sex pheromone components in the red-banded leafroller moth.
J Gen Physiol 65: 179-205.
Ochieng SA, Park KC and Baker TC (2002) Host plant volatiles synergize responses of sex pheromone-specific olfactory
receptor neurons in male Helicoverpa zea. J Comp Physiol [A] 188: 325.
Pappas LG and Larsen JR (1976) Gustatory hairs on mosquito, Culiseta inornata. J Exp Zool 196: 351-360.
Park SK, Shanbhag SR, Wang Q, Hasan G, Steinbrecht RA and Pikielny CW (2000) Expression patterns of two putative
odorant-binding proteins in the olfactory organs of Drosophila melanogaster have different implications for their
functions. Cell Tissue Res 300: 181-192.
Pates HV, Takken W, Stuke K and Curtis CF (2001) Differential behaviour of Anopheles gambiae sensu stricto (Diptera:
Culicidae) to human and cow odours in the laboratory. Bull Entomol Res 91: 289-296.
Perry AS and Fay RW (1967) Correlation of chemical constitution and physical properties of fatty acid esters with
oviposition response of Aedes aegypti. Mosquito News 27: 175-183.
Pitts RJ, Fox AN and Zwiebel LJ (2004) A highly conserved candidate chemoreceptor expressed in both olfactory and
gustatory tissues in the malaria vector Anopheles gambiae. PNAS 101: 5058-5063.
Pitts RJ and Zwiebel LJ (2006) Antennal sensilla of two female anopheline sibling species with differing host ranges.
Malaria Journal 5: 26.
Qiu YT 2005. Sensory and behavioural responses of the malaria mosquito Anopheles gambiae to human odours. PhD
Thesis,Wageningen University, Wageningen, pp 208.
Qiu YT, Van Loon JJA, Takken W, Meijerink J and Smid HM (2006) Olfactory coding in antennal neurons of the malaria
mosquito, Anopheles gambiae. Chem Senses 31: 845-863.
Qiu YT, Smallegange RC, Ter Braak CJF, Spitzen J, van Loon JJA, Jawara M, Milligan P, Galimard AM, Van Beek TA, Knols BGJ
and Takken W (2007) Attractiveness of MM-X traps baited with human or synthetic odor to mosquitoes (Diptera:
Culicidae) in The Gambia. J Med Entomol 44: 970-983.
Ribeiro JMC (2003) A catalogue of Anopheles gambiae transcripts significantly more or less expressed following a blood
meal. Insect Biochem Mol Biol 33: 865-882.

60  Olfaction in vector-host interactions

 3. Olfactory physiology of blood-feeding vector mosquitoes

Rutzler M and Zwiebel LJ (2005) Molecular biology of insect olfaction: recent progress and conceptual models. J Comp
Physiol [A] 191: 777-790.
Sass H (1978) Olfactory receptors on the antenna of Periplaneta: response constellations that encode food odors. J
Comp Physiol [A] 128: 227-233.
Sato K, Pellegrino M, Nakagawa T, Nakagawa T, Vosshall LB and Touhara K (2008) Insect olfactory receptors are
heteromeric ligand-gated ion channels. Nature 452(7190): 1002-1006.
Shields VDC and Hildebrand JG (2001) Responses of a population of antennal olfactory receptor cells in the female
sphinx moth Manduca sexta to plant-associated volatile organic compounds. J Comp Physiol [A] 186: 1135-1151.
Smallegange RC, Geier M and Takken W 2002. Behavioural responses of Anopheles gambiae to ammonia, lactic acid and
a fatty acid in a y-tube olfactometer. In: Proceedings of Experimental & Applied Entomology, N.E.V. Amsterdam,
the Netherlands, pp 147-152.
Smallegange RC, Qiu YT, Van Loon JJA and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic
acids as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto
(Diptera: Culicidae). Chem Senses 30: 145-152.
Smith CN, Smith N, Gouck HK, Weidhaas DE, Gilbert IH, Mayer MS, Smittle BJ and Hofbauer A (1970) L-Lactic acid as
a factor in the attraction of Aedes aegypti (Diptera: Culicidae) to human hosts. Ann Entomol Soc Am 63: 760-770.
Steib BM, Geier M and Boeckh J (2001) The effect of lactic acid on odour-related host preference of yellow fever
mosquitoes. Chem Senses 26: 523-528.
Steinbrecht RA (1997) Pore structures in insect olfactory sensilla: a review of data and concepts. Int J Insect Morphol
Embryol 26: 3-4.
Steinbrecht RA (1998) Odorant-binding proteins: expression and function. Ann N Y Acad Sci 855: 323-332.
Syed Z and Leal WS (2008) Mosquito smell and avoid the insect repellent DEET. PNAS 105: 13598-13603.
Todd JL, Haynes KF and Baker TC (1992) Antennal neurons specific for redundant pheromone components in normal
and mutant Trichoplusia ni males. Physiol Entomol 17: 183-192.
Todd JL and Baker TC (1999) Function of peripheral olfactory organs. Springer-Verlag, Heidelberg, Germany.
Van den Broek IVF and Den Otter CJ (1999) Olfactory sensitivities of mosquitoes with different host preferences
(Anopheles gambiae s.s., An. arabiensis, An. quadriannulatus, An. m. atroparvus) to synthetic host odours. J Insect
Physiol 45: 1001-1010.
Vogt RG and Riddiford LM (1981) Pheromone binding and inactivation by moth antennae. Nature 293: 161-163.
Wang GR, Qiu YT, Lu T, Kwon HW, Pitts RJ, Van Loon JJA, Takken W and Zwiebel LJ (2009) Anopheles gambiae TRPA1 is
a heat-activated channel expressed in thermosensitive sensilla of female antennae. Eur J Neurosci 30: 967-974.
Wicher D, Schäfer R, Bauernfeind R et al. (2008) Drosophila odorant receptors are both ligand-gated and cyclic-
nucleotide-activated cation channels. Nature 24: 1007-1011.
Xu PX, Zwiebel LJ and Smith DP (2003) Identification of a distinct family of genes encoding atypical odorant-binding
proteins in the malaria vector mosquito, Anopheles gambiae. Insect Mol Biol 12: 549-560.
Zahiri N, Rau ME and Lewis DJ (1997) Oviposition responses of Aedes aegypti and Ae. atropalpus (Diptera: Culicidae)
females to waters from conspecific and heterospecific normal larvae and from larvae with Plagiochis elegans
(Trematoda: Plagiorchiidae). J Med Entomol 35: 565-568.
Zeng XN, Leyden JJ, Lawley HJ, Sawano K, Nohara I and Preti G (1991) Analysis of characteristic odors from human male
axillae. J Chem Ecol 17: 1469-1493.
Zeng XN, Leyden JJ, Spielman AI and Preti G (1996) Analysis of characteristic human female axillary odors: qualitative
comparison to males. J Chem Ecol 22: 237-257.
Zhou JJ, He XL, Pickett JA and Field LM (2008) Identification of odorant-binding proteins of the yellow fever mosquito
Aedes aegypti: genome annotation and comparative analyses. Insect Mol Biol 17: 147-163.

Olfaction in vector-host interactions 61

4. Odour coding and neural connections
Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Abstract

In this chapter, we provide a general overview over the structure and function of the peripheral
and central olfactory systems of insects, with special emphasis on the olfactory system of
mosquitoes. We give a short presentation of what currently is known about the functional
genomics of odorant-binding proteins as well as odorant receptors. Emphasis is, however, put on
the structure and functional characteristics of olfactory sensilla to bring the reader up to date of
what has been accomplished within this field over the last couple of years. Furthermore, readers
are given an in depth overview of how the primary olfactory centre, the antennal lobe, of insects
is structured as well as how it deals with input olfactory information. Finally, we give a description
of the structure and function of higher olfactory centres.

Keywords: olfactory signal transduction, olfactory receptor neuron, local interneuron, projection
neuron, coding

Introduction

Disease-vector insects, like most other insects, rely heavily on olfactory cues for survival and
reproduction. Thus, to gain insight into the biology of a particular insect species, it is often of
paramount importance to study its olfactory system, including the structure and function of
both its peripheral and central nervous systems. Our understanding of the neural and peri-neural
aspects of the mosquito olfactory system have significantly expanded in recent years due to the
genomic sequencing of some of the most important disease vectors, the African malaria mosquito,
Anopheles gambiae Giles (Holt et al. 2002), the yellow fever mosquito, Aedes aegypti (L.) (Nene et al.
2007), and the southern house mosquito, Culex quinquefasciatus Say (http: //cpipiens.vectorbase.
org). As a result, we have gained an increased knowledge about the molecular components of
odour detection that ultimately regulate the function of the peripheral olfactory system.

In this chapter, we provide a general description of what is currently known about the molecular
basis of olfactory signal transduction in insects, providing a brief summary on what is known in
mosquitoes [a more thorough presentation is given in Chapter 2 of this volume]. The knowledge
on the initial steps of signal transduction improves our understanding of the functional anatomical
organisation of the olfactory system in insects in general, and in mosquitoes specifically.
Functional anatomical studies in mosquitoes have primarily been conducted on the peripheral
olfactory system. In the next two sections, we provide a general overview of what is known about
the system in insects, and provide some specific data on mosquitoes obtained from our own
ongoing research [a more detailed description on the peripheral system in mosquitoes is given
in Chapter 3 of this volume]. To date, very little is known about central nervous integration of
olfactory information in mosquitoes. However, thorough analyses have been made on the neural
connectivity in the primary olfactory system, the antennal lobe (AL) of both An. gambiae and
Ae. aegypti (Ghaninia et al. 2007a,b, Ignell et al. 2005). These studies emphasise that the central
olfactory system of mosquitoes display great structural resemblance with that of other insects,
suggesting that the coding mechanisms might be similar to that observed in these species.
Furthermore, we provide an overview of what is currently known about odour coding in the AL

Olfaction in vector-host interactions 63

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

and higher olfactory centres. We conclude by highlighting topics that need to be addressed in
future studies to better grasp the neuroethology of these important vector species.

Morphology of the peripheral olfactory system

The olfactory organs of insects include the antennae, and most often a pair of mouthparts acting
as secondary olfactory organs. In arthropods that do not have antennae, e.g. ticks, either a pair
of legs or mouthparts take over this function. Irrespective of the location of the organs, they are
often covered by hair-shaped sensilla, i.e. small cuticular sensory organs. These organs consist of
three structural elements: a cuticular outer sheath structure, sensory cells and enveloping support
cells. Based on the cuticular structure of a sensillum, we are able to discriminate different types
of sensilla through microscopic analysis. According to a system suggested by Altner (1977), these
types are: (a) wall-pore sensilla, (b) tip-pore sensilla, and (c) aporous sensilla; in which wall-pore
sensilla are mainly olfactory and the most abundant type found on the olfactory organs. Wall-pore
sensilla may be further divided, based on cuticular structure, into single-walled (including e.g.
sensilla basiconica and trichodea) and double-walled sensilla (including sensilla coeloconica and
grooved pegs) (Hallberg and Hansson 1999, Keil 1999). The number of bipolar sensory neurons,
in this case olfactory receptor neurons (ORNs) innervating a particular sensillum type varies from
one to approximately fifty (Keil 1999). Surrounding these neurons are often three, but sometimes
four, enveloping support cells (the tormogen, trichogen and thecogen cells), each of which have a
particular role in the development of the sensillum and take an active part in the mature sensillum
(reviewed by Keil 1999).

Olfactory sensilla in mosquitoes

Olfactory sensilla on both antennae and maxillary palps of mosquitoes (Figure 1) have been
described previously for several species (reviewed by Clements 1999, McIver 1982, Sutcliffe 1994).
Also recently, olfactory sensilla were described on the labellar lobes of the proboscis (Figure 1A,
Kwon et al. 2006), previously recognised only as a gustatory organ. Apart from the large coeloconic
sensilla that are absent in culicine mosquitoes, the types of sensilla found on the olfactory organs
of mosquitoes are generally well conserved, also compared with other insects [see Chapter 3, this
volume for more information].

Perireceptor events in insect olfaction

In the last few decades, biochemical, molecular and genomic studies have greatly improved our
understanding of the mechanisms underlying insect olfaction. It is now evident that detection
of chemical cues in insects occurs through a series of events, several of which are mediated by
different protein families in the sensillum lymph (Figure 2). In order to reach the odorant receptors
expressed on the dendrites of the ORNs (see below), odour molecules need to enter through
sensillar pores and pass through an aqueous environment (Figure 2). This transport has been
proposed to be mediated by soluble proteins, referred to as odorant-binding proteins (OBPs), in
the sensillum lymph (Figure 2). Comparison among insect OBPs have revealed common structural
properties such that they are small, water-soluble, extracellular proteins with a signal peptide
sequence cleaved during secretion. Odorant-binding proteins also share the conservation of six
cysteins that is important for the stability of the protein structure. The abundance of OBPs in the
sensillum lymph, their differential distribution in various sensillum types and the specificity of
their ligand-binding properties suggest that they are important players in the first step of odour

64  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Figure 1. (A) Scanning electron micrograph of the head of a female Aedes aegypti shows the olfactory organs, the
antennae (Ant), maxillary palps (Mp) and labellar lobes (LL) of the proboscis. (B) Scanning electron micrograph
of an individual segment of the antenna of the same female shows grooved peg sensilla and four sub-types of
olfactory sensilla trichodea: short sharp tipped (sst), short blunt-tipped I (sbtI), short blunt-tipped II (sbtII), and
long sharp-tipped (lst) (redrawn, with permission, from Ghaninia et al. 2007b).

coding in insects. Here we will focus on the general aspects of insect OBPs and recent findings
that indicate their possible roles in olfactory-driven behaviour of insects.

Odorant-binding proteins

The first insect odorant-binding protein was identified in the male antennae of the moth Antheraea
polyphemus (Cramer) using the radioactively labelled pheromone (E, Z)-6,11-hexadecadienyl
acetate as a probe in ligand-binding experiments. This odorant-binding protein was referred to
as a pheromone-binding protein (PBP) (Vogt and Riddiford 1981). Subsequently, additional OBP
sequences were identified in Lepidoptera (Györgyi et al. 1988, Raming et al. 1989, Vogt et al. 1989)
that showed differential expression pattern and distinct binding properties to different classes
of odorant molecules (Krieger et al. 1996, Pelosi 1998, Vogt and Lerner 1989, Vogt et al. 1991a,b).
Accordingly, insect OBPs are now classified into three groups: PBPs, general odorant-binding
proteins (GOBPs) and antennal binding protein X (ABPX).

Olfaction in vector-host interactions 65

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

OBP
A
Odorants

OR

ORNs

Air Odorants
B
Sensillum
cuticle
P
Sensillum lymph OB

P
OB
E
OB OD
OB P
OD
P E
Ca 2+
K Ca2+
+
Na +
Or83b
ORx

Cyclic AC Gα
nucleotide-gated Ca2+ β γ
K+
ion channel Na+ cAMP ATP Ca2+

Figure 2. Schematic representation of the components of olfactory signal transduction in insects. (A) Odorants
enter the aqueous olfactory sensillum lymph via pores of the cuticular hair wall of the sensillum. In the sensillum
lymph, hydrophilic OBPs are believed to bind and transport the odorants to conventional OR (ORx) located at the
dendritic membrane of the ORNs. (B) Olfactory signal transduction is initiated as the odorant bind to the ORx,
which is associated with the membrane-bound heterotrimeric G-protein. Odorants that stimulate the ORx are
removed from the vicinity of the receptor and degraded by ODEs. Conventional ORx forms a heterodimer with the
Or83b family OR that is required for the proper localisation of the ORx to the dendritic membrane in order to trigger
signal transduction. Upon binding of the odorant molecule to the ORx, a conformational change forms a ORx-
Or83b duplex activating a Gα-protein that, in turn, stimulates AC, resulting in the generation of cAMP. This triggers
cyclic nucleotide-gated ion channel that increases intracellular Ca2+ and Na+ and membrane depolarisation. Two
models for odorant-induced production of cyclic nucleotide signalling have been proposed recently. Wicher et al.
(2008), the model showed here, observed a G-protein coupled cyclic nucleotide-gated ion channel stimulation
and activation of Or83b that allows inward flow of ions. Sato et al. (2008), however, showed the stimulation of
the ORx-Or83b duplex, and influx of cations through ion channels, in the absence of a second messenger system.
These data provide an important evidence for the mechanism of signal transduction in insect olfactory systems;
while, the roles of the second messengers in the signalling pathway are needed to be investigated more in detail.
AC, adenylyl cyclase; ATP, adenosine triphosphate; cAMP, cyclic adenosine monophosphate; Gα, G-protein α
subunit; Gβ, G-protein β subunit; Gγ, G-protein γ subunit; OBP, odorant-binding protein; ODE, odorant-degrading
enzyme; OR, odorant receptor; ORNs, olfactory receptor neurons.

66  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Immunocytochemical localisation of OBPs has been determined in Lepidoptera where

pheromone-sensitive sensilla trichodea express PBPs (Steinbrecht et al. 1992), whereas sensilla
basiconica mainly express GOBPs (Laue et al. 1994). No co-localisation of PBPs and GOBPs in the
same sensillum has been observed (Steinbrecht et al. 1995). Differential expression of OBPs has
also been found in Diptera (Hekmat-Scafe et al. 1997, Park et al. 2000). However, co-localisation
of closely related OBPs (i.e. OS-E and OS-F) in the same sensillum has been shown in Drosophila
melanogaster Meigen (Hekmat-Scafe et al. 1997). This finding shows that more than one OBP can
be expressed in an individual sensillum.

In the post-genomic era, a large number of Obp genes have been identified in Dipteran insects
as well. In D. melanogaster, 51 Obp genes have been identified (Galindo and Smith 2001, Graham
and Davies 2002, Hekmat-Scafe et al. 2002). In An. gambiae and Ae. aegypti, 57 and 66 genes have
been found to encode OBPs, respectively (Biessmann et al. 2005, Xu et al. 2003, Zhou et al. 2008). In
D. melanogaster and An. gambiae, most of these genes have been characterised through reverse-
transcriptase polymerase chain reaction (RT-PCR), in situ hybridisation and GAL4/UAS expression
systems. Comparisons of OBPs among these species have revealed conserved gene homologues
as well as gene expansions (Xu et al. 2003, 2008), which may reflect species-specific functions (see
also Chapter 2, this volume).

Expression of Obp genes is not restricted to olfactory tissues, such as antennae, maxillary palps
and the proboscis. There is a considerable overlapping expression of Obp genes in olfactory and
gustatory tissues of mosquitoes (Biessmann et al. 2005, Li et al. 2005, Sengul and Tu 2008) and
fruit flies (Hekmat-Scafe et al. 2002). It is thus plausible that OBPs are involved in both types of
chemoreception, or alternatively, that both olfactory and gustatory tissues share some common
functions. These findings may contribute significantly to understand the possible roles of OBPs in
olfactory processes (discussed below).

Physiological roles of OBPs in olfaction

Since the discovery of the first insect OBP in moth antennae (Vogt and Riddiford 1981), intensive
efforts have been made to determine the structures of OBPs in various insect species to reveal
the mechanisms behind OBP function. Structural information is so far only available for six insect
OBPs, including the D. melanogaster LUSH protein (Kruse et al. 2003) and OBP1 in An. gambiae
(Wogulis et al. 2006). These studies found some common features, e.g. that OBPs contain α-helices
that form a hydrophobic binding pocket for the ligand and that OBPs undergo a pH-dependent
conformational change that determines ligand binding and release.

Despite a wealth of biochemical and structural information, the exact physiological function
of OBPs still remains elusive. However, the advantage of genetic tools in D. melanogaster has
contributed significantly to the understanding of the function of OBPs in insect olfaction. Xu et
al. (2005) found that a D. melanogaster mutant for an OBP, named LUSH, showed defects in its
perception of the aggregation pheromone, 11-cis vaccenyl acetate (cVA). The expression of the
lush transgene in the mutants restored the LUSH phenotype, indicating that LUSH is required
to bind cVA and, in turn, stimulate the cVA-sensitive neurons to elicit a pheromone-induced
behaviour. More recently, it was shown that LUSH is required for cVA-sensitive receptors (Or67d)
to respond to cVA in vivo (Ha and Smith 2006). Other D. melanogaster OBPs, such as OBP57d
and OBP57e, have also been shown to regulate behavioural differentiation in members of the
D. melanogaster species complex (Matsuo et al. 2007). Flies having the mutant gene Obp57d/e

Olfaction in vector-host interactions 67

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

showed altered responses to the host plant odorants, hexanoic acid and octanoic acid, which
influenced the behaviour related to oviposition site preference.

The availability of the Ae. aegypti and An. gambiae genomes (Holt et al. 2002, Nene et al. 2007) has
allowed for the identification of a substantial number of putative Obp genes and their subsequent
tissue localisation in these species. More interestingly, some of these genes have been shown
to be down-regulated after a blood meal (Biessmann et al. 2005, Justice et al. 2003a, Sengul
and Tu 2008). This may indicate that these Obp genes play an important role in host-seeking
behaviours. Consequently, future research will have to focus on the understanding of how these
genes influence the behaviour of mosquitoes. Experiments to establish transformed mosquitoes
to produce Obp knock-out mutants are highly demanding as it is difficult to assess the phenotype.
However, a recent RNA interference strategy has shown a significant decrease (8-fold and >100-
fold reduction) in the mRNA levels of the two candidate Obp genes in the olfactory tissues of
female Ae. aegypti mosquitoes (Sengul 2008). This finding is very encouraging and shows that
it is possible to establish an efficient method to knockdown Obp genes in olfactory tissues of
mosquitoes. Such analysis opens doors to future systematic analysis of the molecular players
involved in mosquito olfaction.

Transduction mechanisms

When odour molecules have travelled through the sensillum lymph, they reach the surface of the
olfactory receptor neuron dendrite (Figure 2). Here, olfactory receptor (OR) proteins reside inside
the membrane. These proteins are the main key to the specificity of the ORNs. Insect ORs were
first identified in D. melanogaster (Clyne et al. 1999, Gao and Chess 1999, Vosshall et al. 1999).
They were a new family of putative seven-transmembrane G-protein coupled receptors (GPCR).
It is downstream from these receptors that the translation from a chemical signal to an electrical
one, the transduction cascade, takes place (Figure 2B). In vertebrates this cascade has been well
characterised. The binding of a ligand to the OR activates a G-protein coupled to the receptor.
The G-protein in turn activates an adenylyl cyclase, which converts ATP to cAMP. The cAMP is a
well-known signalling molecule in various cell processes, and here acts on the intracellular part
of a cyclic nucleotide-gated ion channel, and changes its conductance. This change in turn opens
chloride channels and thereby changes the charge of the inside of the cell vs. the outside. If this
process proceeds long enough, an action potential is formed. In crustaceans similar processes
have been observed, but here two second messengers, cAMP and IP3, are involved.

It was long considered that insect olfactory transduction processes were very similar to what had
been observed in vertebrates and crustaceans. Recent progress does, however, challenge these
views. Several factors of the insect system seem different. One peculiarity of the insect system was
reported by Benton et al. (2006), who showed that insect ORs are inserted ‘inside-out’ as compared
to the vertebrate system. The insect ORs have an extracellular N-terminus and an intracellular
C-terminus. The functional implications of this architecture are so far more or less unknown.

Vertebrate ORNs always express a single type of OR only. However, when the putative insect ORs
were discovered, it was demonstrated that the ORN-type-specific OR in D. melanogaster was more
or less always co-expressed with a ubiquitous, second OR, OR83b (Figure 2B, Neuhaus et al. 2005).
Later experiments showed that this protein is vital for ORN function and that in its absence the
type-specific OR does not even reach the dendrite during development (Larsson et al. 2004). In
Manduca moths, Nakagawa et al. (2005) showed that expression of a pheromone receptor and the
OR83b homolog was sufficient to obtain channel activation after ligand stimulation. The OR83b

68  Olfaction in vector-host interactions

 4. Odour coding and neural connections

proteins heterodimerise with the ORN-type-specific ORs (Figure 2B). The type-specific ORs thus
form a complex allowing direct interactions between the two proteins. The OR83b-type of protein
has been found in all insects investigated so far, and seems to be a general insect feature in insect
olfactory systems. In mosquitoes, AgOR7, AaOR7, and CqOR7 represent the possible orthologs
of the D. melanogastor OR83b in An. gambiae, Ae. aegypti and Cx. quinquefasciatus, respectively
(Bobhot et al. 1997, Hill et al. 2002, Melo et al. 2004, Xia and Zwiebel 2006).

Recent studies, including the ubiquitous OR83b receptor, have shown that this receptor is in fact
a non-specific cation channel, under both ionotropic and metabotropic control (Figure 2B, Sato
et al. 2008, Wicher et al. 2008). The initial and very fast response is cAMP-independent, while a
second and much more sensitive response is Gs-mediated and cAMP-dependent. More or less
simultaneously Kain et al. (2008) showed very strong evidence for a G q and a phospholipase
C involvement. The transduction system is not clearly understood, and contains a number of
different transduction pathways. The function of each of these components still remains to be
elucidated.

Peripheral olfactory system

The ORNs represent the first neuronal filter in olfactory signalling. Arthropod vectors find
their hosts’ particular odour blend amidst a varied and complex odour environment. It is this
identification of signature volatile blends from the background noise that is decoded by the
vector chemosensory system, and translated into host-seeking behaviours. The first level of
neuronal discrimination occurs in the ORNs in which the electrical signals are coded using odorant
selectivity, volatile sensitivity and timing of the response.

Selectivity

An ORN is the first component in a signal processing channel conveying information concerning
a constrained number of volatile compounds to the AL. This constraint on an individual ORN’s
responses to specific compounds, or families of compounds, is referred to as ‘tuning’ (Hallem et al.
2004, Hamana et al. 2003, Stensmyr et al. 2003). Tuning is largely, possibly exclusively, dependant
on OR-ligand affinity (see previous section on transduction mechanisms) and the perireceptor
events in the aqueous environment surrounding the ORN dendrites (see Figure 2).

The affinity of volatile compounds for their cognate receptors has been determined in many ways:
in vivo functional characterisation of sensilla with single sensillum recording (SSR) describes ORN
tuning, whether it is narrowly tuned, e.g. pheromone-receptive neurons in a multitude of species
(e.g. Bengtsson et al. 1990, Liljefors et al. 1984, 1985, 1987); or broadly tuned, e.g. many kairomone-
receptive antennal trichoid sensilla in mosquitoes (Figure 3, Ghaninia et al. 2007b, Hill et al. 2009,
Qiu et al. 2006a,b). Only a few volatile pheromones have been identified in disease vectors, such
as oviposition pheromones (Davis and Bowen 1994, Du and Millar 1999, Pickett and Woodcock
1996), and in fact for many years, it has been claimed that there was little or no volatile pheromone
communication between arthropod vectors (e.g. Schofield and Moreman 1976). More recently,
however, there have been both identifications and other strong indications of sex/aggregation-
type pheromone communication in a variety of vectors, such as ticks, triatomines, sandflies and
mosquitoes (Brazil and Hamilton 2002, Cabrera and Jaffe 2007, Grenacher et al. 2001, Guerenstein
and Guerin 2004, Vitta et al. 2007)(see also elsewhere in this Volume). The narrow tuning of
pheromone-receptive ORNs has been extensively investigated in many species, and probably most
widely studied in Lepidoptera (e.g. Baker et al. 2004, Domingue et al. 2007). Studies which subtly

Olfaction in vector-host interactions 69

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Nonanal
Linalool
Limonene
Menthone
Eugenol
Citral
α-thujone
3-methylphenol
Indole
3-methylindole
2-undecanone
Benzyl alcohol SBT I 5 SBT I 3
4-methylcyclohexanol (n=4) (n=14)
2-butoxyethanol
1-octen-3-ol
Ethyl butyrate
(-)-ethyl-L-lactate
Propionic acid
Heptanoic acid
Butanoic acid B B
Paraffin oil A A
-50 0 50 100 150 -50 0 50 100 150

Neuronal activity (spikes/s)

Figure 3. An example of odorant receptor neuron (ORN) selectivity (tuning) from the mosquito Culex
quinquefasciatus. The short blunt type I trichoid sensillum 5B (SBT I 5B) is narrowly tuned, responding maximally
to a single compound, 3-methylindole, whereas SBT I 3A responds to linalool, menthone, indole, 3-methylindole,
2-butoxyethanol and 1-octen-3-ol, demonstrating a more broad pattern of tuning. Two spontaneously active
neurons are found in all trichoid types (A and B). The neuronal responses of A and B (See legend) are shown as an
average over (n) replicates (reproduced with permission from Hill et al. 2009).

manipulated the structure of one pheromone component in the moth Agrotis segetum (Denis and
Schiffermüller) revealed that the optimal tuning of a pheromone receptor is dependent on the
molecular shape, chirality and electron distribution of the pheromone component (Bestman 1981,
Bestman and Vostrowsky 1982, Larsson et al. 1999, Priesner 1979, Priesner et al. 1975, Wojtasek
et al. 1998).

The ORN responses of disease vectors to kairomones have come under intense scrutiny in recent
years, due to the colossal potential for kairomones to be used as trap attractants and as repellents
for personal protection (e.g. Ditzen et al. 2008, Justice et al. 2003b, Mason et al. 1993, Pickett and
Woodcock 1996, Schmidt 2005). Small panel SSR screens (<100 compounds) have described
generalised tuning curves for approximately 80% of the ORNs in three mosquito species: An.
gambiae (Kwon et al. 2006, Lu et al. 2007, Qiu et al. 2006b), Ae. aegypti (Ghaninia et al. 2007b, Siju et
al. 2010) and Cx. quinquefasciatus (Hill et al. 2009, Syed and Leal 2007).

70  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Historically, kairomone-responding ORNs have been considered, with a few exceptions, to be

broadly tuned (e.g. Behan and Schoonhoven 1978, Hansson et al. 1989). Many recent studies,
however, have clarified that broad or general tuning, as indicated by previous studies, is a result
of two common experimental practices: the lack of screens identifying the actual key ligands
for these ORNs; and the concentrations at which kairomones were applied were many hundreds
of times higher than environmentally and physiologically relevant (e.g. Anderson et al. 1995,
Dickens 1990, Hansson et al. 1999, Heinbockel and Kaissling 1996, Larsson et al. 2001, Priesner
1979, Shields and Hildebrand 2001, Skiri et al. 2004). Kazama and Wilson (2008), working with D.
melanogaster, have determined that high volatile concentrations, which result in high ORN spike
frequencies, dampen AL input receptivity and thus, high frequency ORN responses are not the
optimal characteristics by which to identify key ligands. In contrast, a better method to gauge ORN
selectivity is to identify those compounds which elicit threshold responses at low concentrations.

In vitro heterologous expression of ORs in mammalian cell-lines and ectopic expression of ORs
in the D. melanogaster empty neuron system have been largely able to reconstruct the in vivo
tuning profiles, with the bonus of identifying the OR responsible for the ligand binding pattern
(e.g. Hallem et al. 2004, Kwon et al. 2007, Rospars et al. 1996, Syed et al. 2007, Carrey et al. 2010).
The high correlation between the in vivo and in vitro response profiles indicates that the binding
between odorant and receptor is chiefly responsible for the selective nature of ORNs. As a result,
the heterologous expression system is well-placed as a technique to identify potential key ligands
through large scale screening methods akin to those used in pharmaceutical-type screens (Hallem
et al. 2004, Malnic 2007, Touhara 2007). Moreover, putative key ligands can be validated in vivo
using SSR.

Sensitivity

For the most part, the olfactory system identifies odours not by the activation of a single processing
channel but through comparisons among all active channels (see below). In this configuration, it
is necessary to know the precise tuning curves of each ORN type before the organism can identify
a complex odour blend (for review see Vosshall and Stocker 2007).

As mentioned above, ORNs are to a greater or lesser degree promiscuous in their responses to
volatiles, e.g. one ORN responds to changes in the flux of more than one volatile compound;
and more than one ORN responds to the same compound, but to varying degrees (Figure 4). It
is the comparison of these differentially tuned curves in the CNS that leads to the identification
of relevant odour plumes (for review see Todd and Baker 1999). Comparisons among the three
mosquito species’ ORN response profiles reveal that these organisms, which have maintained a
similar host choice during evolutionary separation, employ distinct patterns of odour tuning in
the ORNs. Although many similar volatiles are recognised by all three species, there appears to be
little conservation in the individual ORN tuning curves (Ghaninia et al. 2007a, Hill et al. 2009, Qiu
et al. 2006b, Siju et al. 2010).

Olfactory receptor neurons can respond to odorants in extremely low amounts: e.g. sensilla
generally respond to a threshold dose sex-pheromones between 10-7 g and 10-12 g (Angioy et al.
2003, Boeckh and Selsam 1984, Kaissling 1971, Kaissling and Priesner 1970) and to kairomones at
amounts often 10 to 100 fold higher (Hansson et al. 1999, Hill et al. 2009, Jönsson and Anderson
1999, Larsson et al. 2001). It is important to mention that these experimentally derived threshold
values are all relative, and should not be taken as absolute values; since volatility differs greatly
amongst various compounds, and experimental factors such as the speed of airflow and the

Olfaction in vector-host interactions 71

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

A Nonanal
Linalool
Limonene
Menthone
Eugenol
Citral
α-thujone
3-methylphenol
Indole
3-methylindole
2-undecanone
Benzyl alcohol
4-methylcyclohexanol
SBT II 1 SBT II 2 SBT II 3 SBT II 4 SBT II 5
2-butoxyethanol (n=4) (n=5) (n=7) (n=6) (n=5)
1-octen-3-ol
Ethyl butyrate
(-)-ethyl-L-lactate
Propionic acid
Heptanoic acid
Butanoic acid B B B B B
Paraffin oil A A A A A
-50
0
50
100
150
-50
0
50
100
150
-50
0
50
100
150
-50
0
50
100
150
-50
0
50
100
150
Neuronal activity (spikes/s)

B
160
140 sbtII-2A
Response (spikes/sec)

120 n=5
100
Heptanal
80 Octanal
60 Nonanal
Decanal
40
20
0 } other stimuli
-20 0.001 0.01 0.1 1 10
Concentration (%)

Figure 4. An example of odorant receptor neuron (ORN) sensitivity from the mosquitoes (A) Culex quinquefasciatus
in which four of the five short blunt type II trichoid sensilla (SBT II 1-4) contain ORNs responding to varying degrees to
nonanal and another four SBT II (1, 3-5) sensilla respond differentially to the carboxylic acids (butanoic, heptanoic
and propionic acid). Two spontaneously active neurons are found in all trichoid types (A and B). The neuronal
responses of A and B (See legend) are shown as an average over (n) replicates; and (B) Aedes aegypti in which the
sbt II 2A ORN responds with differing sensitivity to four short chain aldehydes (C7-C10) (reproduced with permission
from Hill et al. 2009 and Ghaninia et al. 2008).

72  Olfaction in vector-host interactions

 4. Odour coding and neural connections

duration of stimulus can greatly affect the response of ORNs (Ignell and Hansson 2005). New
techniques such as SSR coupled to a gas chromatograph (GC-SSR) are being used to identify novel
key ligands directly from natural odour collections in physiologically and environmentally relevant
quantities (Ghaninia et al. 2008). Additional analyses such as raster plots, peristimulus histograms
and inter-spike interval superimposition should be employed if more absolute quantification of
ORN threshold response is being sought (Ignell and Hansson 2005).

Temporal coding and stimulus frequency

As mentioned briefly above, the frequency and duration of ORN response is a critical component
of signal coding in the peripheral olfactory nervous system (Hallem et al. 2004, Kazama and Wilson
2008). An intermittent frequency of odour stimulus is often essential in order to elicit a behavioural
response (Baker et al. 1985, Dekker et al. 2005, Geier and Boeckh 1999, Justus et al. 2002, Kennedy
et al. 1981). For example, a homogeneous plume of CO2 is sufficient to activate upwind flight
behaviours in Ae. aegypti, but is insufficient to enable host source-finding; whereas, a filamentous
plume of CO2 with an average pulse frequency of ~6 Hz, which incidentally is much more akin to
a natural host odour plume structures (Mafra-Neto and Cardé 1998, Murlis and Jones 1981), will
substantially enhance source-finding behaviours (Dekker et al. 2005, Geier and Boeckh 1999). At
the level of ORN activity in response to intermittent stimuli, neuronal response in moths has been
shown to mimic the stimulus pulses from 3-12 Hz (Almaas et al. 1991, Baker et al. 1988, Kaissling
1986, Marion-Poll and Tobin 1992) and even up to a staggering frequency of 33 Hz (Bau et al. 2002).

Tracking the temporal characteristics of an odour plume gives the vector the ability to track the
structure of the plume, e.g. size of plume quanta and distance between these quanta; and the
ability to trace the plume back to its source, the ultimate goal of any seeking behaviour (e.g.
Dekker et al. 2005, Lemon and Getz 1998, Moore 1994, Vergassola et al. 2007, Vickers 2006, Willis
and Arbas 1991). The two main temporal coding regimes used in the periphery are phasic and
tonic spiking response patterns (Figure 5). Phasic responses are generated by those neurons
that are able to track stimuli at high pulsed frequencies, thus describing the plume quanta size

40
ORN response (spikes/s)

32

24

16

0
Cell A Cell B

Figure 5. An example of different temporal characteristics of the odorant receptor neurons (ORNs) from the
mosquito Culex quinquefasciatus. The short blunt type I trichoid sensillum 3 (SBT I 3) is depicted responding to a
0.5s stimulus of linalool (10-2 v/v). Each bar represents the response from each of the ORNs contained within the
SBT I 3 sensillum (A and B neuron) during 100ms (reproduced with permission from Hill et al. 2009).

Olfaction in vector-host interactions 73

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

and spacing (Figure 5A). Those ORNs responding with the longer lasting tonic spiking patterns
are believed to make up the short term memory component of the peripheral olfactory system
(Figure 5B), thereby providing the recent stimulus history for direct temporal comparison with
current ORN stimulus response (Almaas and Mustaparta 1991, Den Otter and Van der Goes van
Naters 1993). As many ORNs respond to the same compound with different sensitivities, they also
respond to the same volatile with different temporal patterns (e.g. Ghaninia et al. 2007b, Hill et al.
2009, Qiu et al. 2006b). In Cx. quinquefasciatus, two ORNs within the same sensillum, i.e. SBT I 3,
respond to the green leaf volatile linalool with apparently similar threshold sensitivities, but with
different temporal characteristics (Figure 5). The ‘A neuron’ responds in a fast phasic manner, while
the ‘B neuron’ responds with an initial phasic component followed by a long lasting tonic response
(Figure 5, Hill et al. 2009). This coupling of the long and short term responses to individual stimuli
is not, however, universal. There are many examples of sensilla, both morphologically distinct
and indistinct, which exhibit sensitivity to the same compounds and yet display a continuum of
temporal characteristics (Almaas et al. 1991, Ghaninia et al. 2007b, Heinbockel and Kaissling 1996,
Hill et al. 2009). The diversity with which the arthropods code chemosensory information in the
periphery and transmit these complex signals to the higher olfactory centres through neuronal
selectivity, sensitivity and temporal response patterns, identifies the peripheral olfactory system
as an essential first processing centre in transforming chemical cues into complex and finely tuned
vector behaviours.

Anatomy and function of the primary olfactory centre: the antennal lobe

Antennal lobe glomeruli – the first relay stations in the central nervous system

In insects, ORNs from the antennae (Figure 6A), and in some cases from the mouthparts (Figure
6B), convey chemical information into the primary olfactory centre of the insect brain, the AL
(Figure 6C, Anton and Homberg 1999, Schachtner et al. 2005), producing internal representations
of the odour signals detected at the peripheral level. In ticks, olfactory afferents from Haller’s
organ project into an orthologous region in the central nervous system called the olfactory lobe
(Van Wijk et al. 2006), suggesting that the organisation of the olfactory systems is highly conserved
in the different arthropod lineages. In insects, the formation of these representations has been
shown to involve progressive transformations of the olfactory information between layers of
pre- and postsynaptic neurons in a hierarchical manner (Ache and Young 2005, Hansson and
Christensen 1999, Ignell and Hansson 2005, Urban 2002). The initial step in this transformation
is a reorganisation of the seemingly heterogeneous peripheral olfactory system into a highly
organised chemotopic map in spherical neuropil, glomeruli, of the AL (Hansson and Christensen
1999) that house the synaptic contacts between ORN terminals and AL interneurons. The number,
size and localisation of AL glomeruli is species specific as revealed by mapping and 3-dimensional
reconstructions in various species (reviewed by Schachtner et al. 2005), including the mosquitoes
Ae. aegypti (Figures 6E, 6F and 6G, Ignell et al. 2005) and An. gambiae (Ghaninia et al. 2007a).
Data from the spider mite, Phytoseiulus persimilis Athias-Henriot, also suggest that the olfactory
lobe of mites have a glomerular organisation (Van Wijk et al. 2006). Sexual dimorphism on the
basis of particular specialised glomeruli and on varying glomeruli numbers has been described in
species belonging to Dictyoptera, Hymenoptera, Lepidoptera, and Diptera (Anton and Homberg
1999, Hansson and Anton 2000, Rospars 1988). The functional significance of the observed sexual
dimorphism in mosquitoes, including differences in number, shape and size of particular glomeruli
(Figure 6D, Ghaninia et al. 2007a, Ignell et al. 2005), is still to be revealed. However, examples of
the functional significance of sexual dimorphism abound. For example, in all cockroaches and
moths that use female-produced, long distance sex attractants, a pronounced sexual dimorphism

74  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Figure 6. Confocal images of an AL of a female Aedes aegypti after anterograde staining of the ipsilateral antennal
nerve (A-C). Scale bar represents 25 µm. Individual glomeruli are labeled as in E-G. Note that the perimeter of
individual glomeruli, in most cases, can be clearly seen. (D) Three-dimensional reconstruction of the brain of an
androgynous Ae. aegypti, with surface reconstructions of male (right) and female (left) AL glomeruli as well as
neuropil structures directly or indirectly involved in the integration of olfactory and mechanosensory information.
AL: antennal lobe, AMMC: antennal motor and mechanosensory center, CC: central complex, JOC: Johnston’s
organ center, LAL: lateral accessory lobe, LPC: lateral protocerebrum, MB: mushroom body, OES: esophagus, SOG:
subesophageal ganglion, SPC: superior protocerebrum. (E-G) Three-dimensional model of the AL of a male Ae.
aegypti as seen from a ventral perspective. a: anterior, p: posterior, l: lateral, m: medial, v: ventral, d: dorsal, c:
central. Ventral glomeruli are subsequently removed from the complete model to uncover the dorsal glomeruli. A
total of 49 glomeruli are identified (reproduced with permission from Ignell et al. 2005).

Olfaction in vector-host interactions 75

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

is present in the form of a macroglomerulus or a macroglomerular complex in the males (Hansson

and Christensen 1999).

The reorganisation and convergence of broadly distributed ORNs into AL glomeruli appears to be
a general rule in various insects’ olfactory systems, as revealed through tracing studies of ORNs,
originally in moths (e.g. Christensen et al. 1995, Hansson et al. 1992, Ochieng´ et al. 1995) and now
also reiterated in vector insects like mosquitoes (Figures 6A and 6B, Anton et al. 2003, Ghaninia
et al. 2007a), and through molecular and anatomical mapping of ORNs in the model species D.
melanogaster (Couto et al. 2005, Vosshall et al. 2000). Studies on D. melanogaster have allowed us
to confirm the general validity of the one ORN – one glomerulus hypothesis put forward in the
pre-genomic studies; stating that functionally distinct ORNs converge into distinct glomeruli in
the AL (Couto et al. 2005, Vosshall et al. 2000). The systematic survey of D. melanogaster odorant
receptor (OR) expression at neural resolution have further revealed that although ORNs in distinct
sensillum types project to distinct glomeruli, their neighbour relations are not preserved (Couto
et al. 2005), i.e. they display a non-parallel topography unlike that observed in the visual, auditory
and somatosensory systems. Olfactory receptor neurons that express similar ORs, however, tend
to map to closely appositioned glomeruli in the AL (Couto et al. 2005). This type of organisation
ensures a clustered odour representation of similar odorants at the central nervous level, which is
supported by in vivo imaging studies of ORN ensembles on insects like the honeybee, Apis mellifera
Linnaeus (Galizia and Menzel 2000, Joerges et al. 1997), the moths, Spodoptera littoralis Boisduval
(Carlsson et al. 2002, Carlsson and Hansson 2003) and Manduca sexta Linnaeus (Hansson et al.
2001), and the fly, D. melanogaster (Ng et al. 2002, Wang et al. 2003). This suggests that odours, and
specifically odour quality, ultimately are encoded by spatial patterns of activation in defined sets
of glomeruli before higher order brain centres are able to recognise and discriminate the different
molecular features that are relayed via second-order neurons, the so-called projection neurons
(PNs) (Galizia and Menzel 2000, Hildebrand and Shepherd 1997, Vosshall et al. 2000).

Intra- and interglomerular synaptic interactions in the antennal lobe and quality coding

The ORNs and the different interneuronal elements that arborise within the AL glomeruli are
synaptically interconnected and taken together make up a complex neural network (Figure 6).
The synaptic organisation of the AL has been studied at the ultrastructural level in the cockroach,
Periplaneta americana Linnaeus, and the moth, M. sexta (Boeckh and Tolbert 1993, Distler and
Boeckh 1996, 1997a,b, Malun et al. 1991a,b). The synaptic wiring within single glomeruli provides
anatomical evidence for discrete processing channels within individual AL glomeruli, in which
functionally distinct ORNs converge and make direct synaptic contacts with PNs. Single neuron
tracing of PNs in D. melanogaster along with the stereotypic mapping of ORNs in AL glomeruli also
provide anatomical support for these types of channels (Figure 7B, Couto et al. 2005, Marin et al.
2002). As mentioned above, an odorant stimulus will typically elicit activity in several ORN classes
(Figure 4A), and consequently several processing channels will be activated by any given stimulus
(see above). If no transformation occurs through intercalated local interneurons (LNs, Figure
7A) within or between these channels, a direct feed-forward excitation of the PNs by the ORNs
takes place, i.e. the olfactory information will faithfully be transmitted to higher brain centres, as
suggested by Ng et al. (2003), Root et al. (2007) and Wang et al. (2003). Accumulating evidence
over the last few years from different insects, however, emphasises that LNs that tend to have
broad multiglomerular ramifications (Figure 7A), and thus provide a scaffold for crosstalk between
channels, play a major role in shaping the output response of the processing channels (reviewed by
Ignell and Hansson 2005, see also Bhandawat et al. 2007, Olsen and Wilson 2008, Olsen et al. 2007,
Root et al. 2007, Shang et al. 2007, Wilson and Laurent 2005). In addition, intraglomerular events

76  Olfaction in vector-host interactions

 4. Odour coding and neural connections

appear to have a significant influence on these processes (Bhandawat et al. 2007, Christensen et
al. 2000, Kazama and Wilson 2008, Olsen and Wilson 2008, Wilson et al. 2004).

In recent years, the D. melanogaster AL has been the model of choice for investigating how intra-
and interglomerular events shape the processing channels. Although results emanating from
these studies have expanded our knowledge about the neural mechanisms that shape the PN
odour responses, it is important to acknowledge that these results may not directly transfer to
the olfactory systems of other insects (reviewed by Ignell and Hansson 2005). A growing body of
evidence, however, favours an interaction of interglomerular, or lateral, inhibitory connections
between the processing channels in all insect species studied (Christensen et al. 1998, Olsen and
Wilson 2008, Sachse and Galizia 2002, Waldrop et al. 1987). Lateral inhibition, through GABAergic
LNs acting on both ORNs and PNs, is supported by ultrastructural analysis of their synaptic wiring
(Boeckh and Tolbert 1993, Distler and Boeckh 1996, 1997a,b, Malun et al. 1991a,b). Recent evidence
from D. melanogaster suggests that a significant component of the lateral inhibition is directed
towards ORNs (Olsen and Wilson 2008). The functional significance of this is that lateral inhibition
mediates a gain control in the olfactory system by suppressing strong and redundant responses. As
a consequence, this damping of excessive input decreases cross-correlation between the outputs
of the different processing channels. This in turn is believed to promote a more efficient neural
code for odours that promote quality coding. In A. mellifera and M. sexta, however, physiological
evidence supports a different model in which a substantial component of the lateral inhibition
seems to act to sharpen the tuning curve of PNs rather than the ORN output, creating a more
restricted spatial pattern of output from the odour-activated AL glomeruli (Christensen et al. 1998,
Sachse and Galizia 2002, Waldrop et al. 1987). Evidence for a similar, but minor, component in
D. melanogaster is available, but the functional implication of this arrangement is yet unknown
(Olsen and Wilson 2008).

In addition to the role in regulating patterns of activity in the AL, GABAergic LNs are also involved in
generating temporal patterns of activity within the AL, which has been suggested to be critical for
strengthening the neural representation of odours in other regions of the central nervous system
(reviewed by Christensen and White 2000, Laurent 2002). Disruption of lateral inhibition in the
olfactory network through GABA receptor antagonists results in an overall increase in the output
from the processing channels (Christensen et al. 1998, Sachse and Galizia 2002, Wilson et al. 2004).
As a result, detection and discrimination thresholds are increased as evidenced by psychophysical
data from A. mellifera and M. sexta indicating that the salient odour signal is lost (Hosler et al. 2000,
Mwilaria et al. 2008). In D. melanogaster, the increased output from the processing channels has
been linked to the activity of excitatory cholinergic LN connections between glomeruli (Olsen
et al. 2007, Root et al. 2007, Shang et al. 2007), driven by direct monosynaptic input from ORNs
(Kazama and Wilson 2008). This type of cholinergic connection has, so far, neither been supported
physiologically nor immunohistologically in other insect olfactory systems (Homberg and Müller
1999). Instead, available data support alternative pathways for the observed phenomenon
(Christensen et al. 1998). In D. melanogaster, cholinergic LNs are believed to broaden the tuning
of PNs resulting in an increased separation between odour representations in PN coding space
(Bhandawat et al. 2007, Wilson et al. 2004). As a consequence, PNs are believed to use their dynamic
range more efficiently by protecting the sensory representations from noise contamination added
at later stages in the olfactory processing. Note that this model is in stark contrast to that which
is believed to occur in the olfactory network of insects like moths and honeybees, where lateral
inhibition is acting to decrease the spatial pattern of output. Irrespective of the transformation
process used, the general consensus among olfactory neurophysiologists is that discrimination
of odours, i.e. quality coding, seems to depend on a combinatorial coding process involving the

Olfaction in vector-host interactions 77

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

ensemble activity of a diverse array of PNs in the AL (reviewed by Christensen and White 2000,
Hansson and Christensen 1999).

Additional features of the olfactory signals relayed to higher brain centres

The intra- and interglomerular interactions within the AL provides the olfactory system with a
second level of neuronal signal filtering, with ORNs being the first level, allowing several features of
the olfactory signals – their quality, quantity and temporal characteristics – to filter through and to
be conveyed to higher brain centres (Figure 7C). As mentioned above, a substantial amount of data
is available for quality coding in the AL. In this section, we will present how the olfactory system
relays information about the quantity as well as the temporal characteristics of an odorant signal.

An increase in odour quantity, or concentration, at both the ORN and the PN level, is generally
encoded as an increase in the response frequency of these neurons. Odour-evoked ORN signals
generally seem to be amplified in the postsynaptic PN (Bhandawat et al. 2007, Boeckh and Ernst
1987, Christensen and Hildebrand 1994, Hansson et al. 1991, Hartlieb et al. 1997). This increase
may be partly explained by the spatial summation of convergent ORNs onto PNs (reviewed by
Hansson and Christensen 1999, see also Kazama and Wilson 2008). An additional factor dictating
the sensitivity of PNs appears to be the size of the innervated glomerulus (Kazama and Wilson
2008). Projection neurons with the highest sensitivity often innervate large glomeruli, like the
sexually dimorphic macroglomeruli and macroglomerular complexes observed in male moths
and cockroaches (Hansson and Christensen 1999, Kazama and Wilson 2008). Results from D.
melanogaster indicate that the odour-evoked ORN signals are transformed non-linearly, in which
weak ORN responses are more significantly amplified in PNs compared to strong ORN responses
(Bhandawat et al. 2007). The non-linear transformation seems to be dependent on the strength
of the ORN-PN synapses, in which high concentrations of odour stimuli tend to produce a short-
term depression at the synapse, triggered by vesicular depletion or presynaptic inhibition via
GABAergic LNs (Bhandawat et al. 2007, Kazama and Wilson 2008). So far, presynaptic inhibition
has only been shown in D. melanogaster (Kazama and Wilson 2008). This transience promotes
adaptation but also appears to be a major reason for the more broadly tuned PNs compared to
ORNs at high concentrations (Kazama and Wilson 2008). The non-linear transformation emphasises
that synapses in the olfactory network are optimised for high sensitivity near odour detection
thresholds (Kazama and Wilson 2008). This is supported by behavioural experiments on insects
other than D. melanogaster (Angioy et al. 2003, Kaissling 1971, Kaissling and Priesner 1970).

Although results from D. melanogaster shed novel light on the functional characteristics of
the olfactory system, it is important to acknowledge that other insects seem to have solved
concentration coding differently. In A. mellifera, odour coding seems to be performed by two
parallel neuronal pathways, an intensity-coding channel and a concentration-invariant channel,
that independently extract information about odour quality and concentration (Sachse and Galizia
2003). Psychophysical studies also support this type of coding pathways (Getz and Smith 1991,
Kramer 1976, Sakura et al. 2002). Morphological evidence indicating similar pathways is present
in other insects, including mosquitoes (Anton and Homberg 1999, Ignell et al. 2005).

In addition to extracting information about odour quality and quantity, the olfactory system of an
insect must be able to detect and extract temporal information of a given odour stimulus signal.
This can be explained by the fact that the temporal intermittency of an odour signal is important
for driving the orientation behaviour of an insect (Baker 1990, Baker et al. 1985, Kennedy et al.
1981, Mafra-Neto and Cardé 1994, Vickers and Baker 1994). In all insects studied so far, PNs are able

78  Olfaction in vector-host interactions

 4. Odour coding and neural connections

to follow stimulus pulses at frequencies between 1 and 10 Hz (Christensen and Hildebrand 1988,
Heinbockel et al. 1999, Kazama and Wilson 2008, Lei and Hansson 1999, Lemon and Getz 1998). The
ability of a PN to follow a pulsed stimulus may be linked to a short-term synaptic depression, likely
presynaptic vesicle depletion, at the ORN-PN synapse as revealed in D. melanogaster (Kazama and
Wilson 2008). Alternatively, or perhaps additionally, stimulus tracking of PNs may be regulated by
intercalated inhibitory LNs that trigger characteristic pulse-tracking capabilities of certain PNs, as
observed in moths (Christensen and Hildebrand 1997, Lei and Hansson 1999, Waldrop et al. 1987).

Anatomy and function of higher olfactory centres

The topographic map in the antennal lobe, of at least D. melanogaster, is retained in the higher
olfactory centres, the mushroom body and the lateral horn of the protocerebrum (Jefferies et al.
2007, Lin et al. 2007, Marin et al. 2002, Wong et al. 2002). Projection neurons that innervate a given
glomerulus display stereotypic branching patterns in these centres, whereas projection neurons
innervating different glomeruli exhibit very different arborisation patterns. It is plausible that
olfactory information relayed from the AL of mosquitoes and other haematophagous insects also
is retained in a similar way considering the conserved anatomy of the higher olfactory system in
the insect lineage (Strausfeld et al. 1998).

A central question in olfaction is how the brain discriminates different odours to elicit an
appropriate behavioural response. Stereotypic connectivity maps of odorant-to-OR (Hallem and
Carlson 2006), ORN-to-PN (Couto et al. 2005, Fishilevich and Vosshall 2005), and PN-to-Kenyon cell,
the principal neuron type of the mushroom body (Jefferies et al. 2007, Lin et al. 2007) have allowed
the construction of neural computation of odour discrimination in the D. melanogaster brain.
Stereotypic PN-to-KC connectivity and functional imaging suggest that differential representation
of the odours in the AL is maintained in the MB calyx and possibly further processed in the
different MB neurons (Jefferies et al. 2007, Lin et al. 2007). The distribution of odour responses
across different classes of Kenyon cells and the imposition of odor-sensitive excitatory and
inhibitory responses both appear to enhance distinct neural representations of different odours.
Such complexity of odour representations greatly reduces the possibility of overlap between
spatiotemporal patterns elicited by two different odorants making them easier to discriminate.

Classical lesion experiments (Erber et al. 1980, Heisenberg et al. 1985) and genetic blockade of KC
synaptic output have also shown that mushroom bodies are essential for memory formation and
retrieval in fruit flies (Dubnau et al. 2001, Krashes et al. 2007, McGuire et al. 2001, Schwaerzel et al.
2002). By determining how odours are represented across KCs in comparison to the tuning profile
of ORNs, Turner et al. (2008) found that representations were significantly better separated in the
mushroom body than in the receptor layer. This combination of sparseness and representation
divisibility is desirable of memory systems (Garcia-Sanchez and Huerta 2003, Huerta et al. 2004,
Kanerva 1988, Olshausen and Field 2004, Tsodyks and Feigelman 1988, Willshaw and Lonquet-
Higging 1969) and consistent with the functional role assigned to the mushroom bodies (De Belle
and Heisenberg 1994, Dubnau et al. 2001, Hammer and Menzel 1998, Krashes et al. 2007, McGuire
et al. 2001, Schwaerzel et al. 2002). With the increasing knowledge concerning the anatomy and
physiology of the mushroom body, we are making good progress towards an understanding of
memory formation and retrieval. Recent functional imaging studies of the mushroom body of D.
melanogaster, have, for example, revealed memory traces, manifested as Ca2+ activity, in specific
subtypes of Kenyon cells (Wang et al. 2008, Yu et al. 2006). If the predictions by Turner et al. (2008)
are correct, i.e. that there is a strong selective pressure to generate sparse representations in the

Olfaction in vector-host interactions 79

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

mushroom bodies, it may be proposed that this ‘sparsening’ is an intrinsic requirement of any
memory system.

Anatomical analyses of PN branching patterns in the lateral horn of the protocerebrum (e.g.
Figure 7B) in D. melanogaster reveals a segregation of putative pheromone representing PNs and
most other PNs (Jefferis et al. 2007). Interestingly, these authors found a high degree of sexual
dimorphism in the lateral horn region that emphasise sex-specific integration in the lateral horn
that may underlie sex-specific behaviours. The spatial segregation of pheromone representation
contrasts with the representation of glomeruli that receive input from non-pheromonal PNs.
Many of these PN classes display extensive overlap, which makes the representations of different
odorants overlapping. The findings by Jefferis et al. (2008) thus suggest that olfactory information
is highly intermixed at the LH level compared to the glomerular organisation of the antennal
lobe, but that rather discrete channels are retained for pheromones all the way from the sensory
periphery to the LH. They further propose that the LH is globally organised according to biological
values rather than chemical nature of the odorant information. This finding is reminiscent of the
male silkworm moth, Bombyx mori, where PNs from the macroglomerular complex representing
sex pheromones send axon projections to a discrete area in the lateral protocerebrum (Seki et
al. 2005). Spatial segregation of the pheromone representation in higher olfactory centres may
therefore be a conserved feature in insects.

Future outlook

Investigation of the olfactory systems, including morphological and physiological studies in the
periphery, the AL and the higher brain centres, within a wider variety of vectors is of fundamental
importance in the development of the olfaction-based methods of trapping and repelling vectors,
demonstrated so successfully in other insect species. With the inclusion of descriptions of more
olfactory systems from haematophagous arthropods, there is the increased probability that
common features of olfactory signally processing will emerge, providing further insight into
processes mediating vector-related olfactory behaviours. Such information will ultimately help
identify novel tools to protect human hosts through the target interference with vector-related
behaviours.

With the recent genome sequencing of some of the most important disease vectors, an
accumulation of information about the perireceptor and transduction mechanisms involving
olfactory coding of volatile signals that drive important behaviours like mating, host finding and
oviposition is expected. Access to the genome of, at present, three species of mosquitoes allow
for comparative genome analysis such that the identification of the complete repertoire of OBPs
and ORs provide rich resources for linking molecular, evolutionary and ecological processes.
Systematic physiological analysis of the olfactory system of these species is important to link in
vitro and in vivo data. To better grasp olfactory coding in mosquitoes, central olfactory processes
are needed to be investigated more in detail. For this purpose, previous studies on other insects,
where anterograde staining and optical imaging of ORNs have been efficiently used, can be
applied for a better understanding of odour coding in mosquitoes.

Acknowledgements

We would like to thank the Swedish research council, Formas.

80  Olfaction in vector-host interactions

 4. Odour coding and neural connections

References

Ache BW and Young JM (2005) Olfaction: diverse species, conserved principles. Neuron 48: 417-430.
Almaas TJ and Mustaparta H (1991) Heliothis virescens: response characteristics of receptor neurons in sensilla trichodea
Type 1 and Type 2. J Chem Ecol 17: 953-972.
Almaas TJ, Christensen TA and Mustaparta H (1991) Chemical communication in heliothine moths. I. Antennal receptor
neurons encode several features of intra- and interspecific odorants in the male corn earworm moth Helicoverpa
zea. J Comp Physiol A 169: 249-258.
Altner H (1977) Insect sensillum specificity and structure: an approach to a new typology. Olfaction Taste 6: 295-303.
Anderson P, Hansson BS and Löfkvist J (1995) Plant-odour-specific receptor neurones on the antennae of female and
male Spodoptera littoralis. Physiol Entomol 20: 189-198.
Angioy AM, Desogus A, Barbarossa IT, Anderson P and Hansson BS (2003) Extreme sensitivity in an olfactory system.
Chem Senses 28: 279-284.
Anton S and Homberg U (1999) Antennal lobe structure. In: Hansson BS (ed) Insect olfaction. Springer, Berlin, Germany,
pp 97-124.
Anton S, Van Loon JJ, Meijerink J, Smid HM, Takken W and Rospars JP (2003) Central projections of olfactory receptor
neurons from single antennal and palpal sensilla in mosquitoes. Arthropod Struct Dev 32: 319-327.
Baker TC (1990) Upwind flight and casting flight: complementary phasic and tonic systems used for location of sex
pheromone sources by male moths. In: Doving K (ed) ISOT X, Proc. 10th Int. Symp. Olfaction Taste, GCS/AS, Oslo,
Norway, pp 18-25.
Baker TC, Willis MA, Haynes KF and Phelan PL (1985) A pulsed cloud of sex pheromone elicits upwind flight in male
moths. Physiol Entomol 10: 257-265.
Baker TC, Hansson BS, Löfstedt C and Löfqvist J (1988) Adaptation of antennal neurons in moths is associated with
cessation of pheromone mediated upwind flight. PNAS 85: 9826-9830.
Baker TC, Ochieng’ SA, Cossé AA, Lee SG, Todd JL, Quero C and Vickers NJ (2004) A comparison of responses from
olfactory receptor neurons of Heliothis subflexa and Heliothis virescens to components of their sex pheromone. J
Comp Physiol A 190: 155-165.
Bau J, Justus KA and Cardé RT (2002) Antennal resolution of pulsed pheromone plumes in three moth species. J Insect
Physiol 48: 433-442.
Behan M and Schoonhoven LM (1978) Chemoreception of an oviposition deterrent associated with eggs in Pieris
brassicae. Entomol Exp Appl 24: 163-179.
Bengtsson M, Liljefors T, Hansson BS, Löfstedt C and Copaja SV (1990) Structure-activity relationships for chain-
shortened analogs of (Z)-5-decenyl acetate, a pheromone component of the turnip moth, Agrotis segetum. J Chem
Ecol 16: 667-684.
Benton R, Sachse S, Michnick SW and Vosshall LB (2006) Atypical membrane topology and heteromeric function of
Drosophila odorant receptors in vivo. PlosBiol 4: 240-257.
Bestmann HJ (1981) Pheromon Rezeptor-Wechsilwirkung bei Insekten. Mitt Dt Ges Allg Angew Ent 2: 242-247.
Bestmann HJ and Vostrowsky O (1982) Peripheral aspects of olfacto-endocrine interactions. Structure-activity. In:
Breipohl W (ed) Olfaction and endocrine regulation. IRL Press, London, UK, pp 253-265.
Bhandawat V, Olsen SR, Gouwens NW, Schlief ML and Wilson RI (2007) Sensory processing in the Drosophila antennal
lobe increases reliability and separability of ensemble odor representations. Nat Neurosci 10: 1474-1482.
Biessmann H, Nguyen QK, Le D and Walter MF (2005) Microarray-based survey of a subset of putative olfactory genes
in the mosquito Anopheles gambiae. Insect Mol Biol 14: 575-589.
Bohbot J, Pitts RJ and Kwon HW (2007) Molecular characterization of the Aedes aegypti odorant receptor gene family.
Insect Mol Biol 16: 525-537.
Boeckh J and Selsam P (1984) Quantitative investigation of the odour specificity of central olfactory neurones in the
American cockroach. Chem Senses 9: 369-380.
Boeckh J and Ernst KD (1987) Contribution of single unit analysis in insects to an understanding of olfactory function.
J Comp Physiol A 161: 549-565.

Olfaction in vector-host interactions 81

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Boeckh J and Tolbert LP (1993) Synaptic organization and development of the antennal lobe in insects. Microsc Res
Tech 24: 260-280.
Brazil RP and Hamilton JGC (2002) Isolation and identification of 9-methylgermacrene-B as the putative sex pheromone
of Lutzomyia cruzi (Mangabeira, 1938) (Diptera: Psychodidae). Memorias Do Instituto Oswaldo Cruz 97: 435-436.
Cabrera M and Jaffe K (2007) An aggregation pheromone modulates lekking behaviour in the vector mosquito Aedes
aegypti (Diptera: Culicidae). J Am Mosq Control Assoc 23: 1-10.
Carey AF, Wang G, Su C-Y, Zwiebel LJ and Carlson JR (2010) Odorant reception in the malaira mosquito, Anopheles
gambiae. Nature 464: 66-71.
Carlsson MA and Hansson BS (2003) Dose-response characteristics of glomerular activity in the moth antennal lobe.
Chem Senses 28: 269-278.
Carlsson MA, Galizia G and Hansson BS (2002) Spatial representation of odours in the antennal lobe of the moth
Spodoptera littoralis (Lepidoptera: Noctuidae). Chem Senses 27: 231-244.
Christensen TA and Hildebrand JG (1988) Frequency coding by central olfactory neurons in the sphinx moth Manduca
sexta. Chem. Senses 13: 123-130.
Christensen TA and Hildebrand JG (1994) Neuroethology of sexual attraction and inhibition in heliothine moths. In:
Schildberger K and Elsner N (eds) Neural basis of behavioural adaptations. Progress in Zoology, vol 39. Fischer,
Stuttgart, Germany, pp 37-46.
Christensen TA and White J (2000) Representation of olfactory information in the brain. In: Finger TE, Silver WL and
Restrepo D (eds) The neurobiology of taste and smell. Wiley-Liss Inc, New York, USA, pp 201-232.
Christensen TA, Harrow ID, Cuzzocrea C, Randolph PW and Hildebrand JG (1995) Distinct projections of two populations
of olfactory receptor axons in the antennal lobe of the sphinx moth Manduca sexta. Chem. Senses 20: 313-323.
Christensen TA, Waldrop B and Hildebrand JG (1998) GABAergic mechanisms that shape the temporal response to odors
in moth olfactory projection neurons. Ann NY Acad Sci 855: 475-481.
Christensen TA, Pawlowski VM, Lei H and Hildebrand JG (2000) Multi-unit recordings reveal context dependent
modulation of synchrony in odor-specific neural ensembles. Nat Neurosci 3: 927-991.
Clements AN (1999) The biology of mosquitoes, vol. 2. Sensory reception and behaviour. CABI Publishing, Wallingford, UK.
Clyne PJ, Warr CG, Freeman MR, Lessing D, Kim J and Carlson JR (1999) A novel family of divergent seven-transmembrane
proteins: candidate odorant receptors in Drosophila. Neuron 22: 327-338.
Couto A, Alenius M and Dickson BJ (2005) Molecular, anatomical, and functional organization of the Drosophila
olfactory system. Curr Biol 15: 1535-1547.
Davis EE and Bowen MF (1994) Sensory physiological basis for attraction in mosquitoes. J Am Mosq Control Assoc 10:
316-325.
De Belle JS and Heisenberg M (1994) Associative odor learning in Drosophila abolished by chemical ablation of
mushroom bodies. Science 263: 692–695.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972.
Den Otter CJ and Van der Goes van Naters W (1993) Responses of individual antennal olfactory cells of tsetse flies
(Glossina m. morsitans) to phenols from cattle urine. Physiol Entomol 18: 43-49.
Dickens JC (1990) Specialized receptor neurons for pheromone and host plant odors in the boll weevil, Anthonomus
grandis Boh (Coleoptera: Curculionidae). Chem Senses 15: 311-331.
Distler PG and Boeckh J (1996) Synaptic connection between olfactory receptor cells and uniglomerular projection
neurons in the antennal lobe of the American cockroach, Periplaneta americana. J Comp Neurol 370: 35-46.
Distler PG and Boeckh J (1997a) Synaptic connection between identified neuron types in the antennal lobe glomeruli
of the cockroach, Periplaneta americana: II. Local multiglomerular interneurons. J Comp Neurol 378: 307-319.
Distler PG and Boeckh J (1997b) Synaptic connections between identified neuron types in the antennal lobe glomeruli
of the cockroach, Periplaneta americana: II. Local multiglomerular interneurons. J Comp Neurol 383: 529-540.
Ditzen M, Pellegrino M and Vosshall LB (2008) Insect odorant receptors are molecular targets of the insect repellent
DEET. Science 319: 1838-1842.

82  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Domingue MJ, Musto CJ, Linn CE Jr, Roelofs WL and Baker TC (2007) Altered olfactory receptor neuron responsiveness
in rare Ostrinia nubilalis males attracted to the O-furnacalis pheromone blend. J Insect Physiol 53: 1063-1071.
Du YJ and Millar JG (1999) Electroantennogram and oviposition bioassay responses of Culex quinquefasciatus and Culex
tarsalis (Diptera: Culicidae) to chemicals in odours from Bermuda grass infusions. J Med Entomol 36: 158-166.
Dubnau J, Grady L, Kitamoto T and Tully T (2001) Disruption of neurotransmission in Drosophila mushroom body blocks
retrieval but not acquisition of memory. Nature 411: 476-480.
Erber J, Masuhr T and Menzel R (1980) Localization of short-term memory in the brain of the bee, Apis mellifera. Physiol
Entomol 5: 343-358.
Fishilevich E and Vosshall LB (2005). Genetic and functional subdivision of the Drosophila antennal lobe. Curr Biol 15:
1548-1553.
Galindo K and Smith DP (2001) A large family of divergent Drosophila odorant-binding proteins expressed in gustatory
and olfactory sensilla. Genetics 159: 1059-1072.
Galizia CG and Menzel R (2000) Odour perception in honeybees: coding information in glomerular patterns. Curr Opin
Neurobiol 10: 504-510.
Gao Q and Chess A (1999) Identification of candidate Drosophila olfactory receptors from genomic DNA sequence.
Genomics 60: 31-39.
Garcia-Sanchez M and Huerta R (2003) Design parameters of the fan-out phase of sensory systems. J Comput Neurosci
15: 5–17.
Geier M and Boeckh J (1999) A new Y-tube olfactometer for mosquitoes to measure the attractiveness of host odours.
Entomol Exp Appl 92: 9-19.
Getz WM and Smith KB (1991) Olfactory perception in honeybees: Concatenated and mixed odorant stimuli,
concentration and exposure effects. J Comp Physiol A 169: 215-230.
Ghaninia M, Hansson BS and Ignell R (2007a) The antennal lobe of the African malaria mosquito, Anopheles gambiae -
innervation and three-dimensional reconstruction. Arthropod Struct Dev 36: 23-39.
Ghaninia M, Ignell R and Hansson BS (2007b) Functional classification and central nervous projections of olfactory
receptor neurons housed in antennal trichoid sensilla of female yellow fever mosquitoes, Aedes aegypti. Eur J
Neurosci 26: 1611-1623.
Ghaninia M, Larsson MC, Hansson BS and Ignell R (2008) Natural odor ligands for olfactory receptor neurons of female
mosquitoes, Aedes aegypti, using gas chromatography- single sensillum recordings. J Exp Biol 211: 3020-3027.
Graham LA and Davies PL (2002) The odorant-binding proteins of Drosophila melanogaster: annotation and
characterization of a divergent gene family. Gene 292: 43-55.
Grenacher S, Kröber T, Guerin PM and Vlimant M (2001) Behavioural and chemoreceptor cell responses of the tick, Ixodes
ricinus, to its own faeces and faecal constituents. Exp Appl Acarol 25: 641-660.
Guerenstein PG and Guerin PM (2004) A comparison of volatiles emitted by adults of three triatomine species. Entomol
Exp Applic 111: 151-155.
Györgyi TK, Roby-Shemkovitz AJ and Lerner MR (1988) Characterization and cDNA cloning of the pheromone-binding
protein from the tobacco hornworm, Manduca sexta: a tissue-specific developmentally regulated protein. PNAS
85: 9851-9855.
Ha TS and Smith DP (2006) A pheromone receptor mediates 11-cis-vacenyl acetate-induced responses in Drosophila.
J Neurosci 26: 8727-8733.
Hallberg E and Hansson BS (1999) Arthropod sensilla: morphology and phylogenetic considerations. Microsc Res Tech
47: 428-439.
Hallem EA and Carlson JR (2006). Coding of odors by a receptor repertoire. Cell 125: 143-160.
Hallem EA, Ho MG, Carlson JR (2004) The molecular basis of odour coding in the Drosophila antenna. Cell 117: 965-979.
Hamana H, Hirono J, Kizumi M and Sato T (2003) Sensitivity-dependent hierarchical receptor codes for odours. Chem
Senses 28: 87-104.
Hammer M and Menzel R (1998) Multiple sites of associative odor learning as revealed by local brain microinjections
of octopamine in honeybees. Learn Mem 5: 146-156.

Olfaction in vector-host interactions 83

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Hansson BS and Christensen TA (1999) Functional characteristics of the antennal lobe. In: Hansson BS (ed) Insect
olfaction. Springer, Berlin, Germany, pp 125-161.
Hansson BS and Anton S (2000) Function and morphology of the antennal lobe: new developments. Ann Rev Entomol
45: 203-231.
Hansson BS, Löfstedt C and Foster SP (1989) Z-linked inheritance of male olfactory response to sex pheromone
components in two species of tortricid moths, Ctenopseustis obliquana and Ctenopseustis sp. Entomol Exp Appl
53: 137-145.
Hansson BS, Christensen TA and Hildebrand JG (1991) Functionally distinct subdivisions of the macroglomerular
complex in the antennal lobe of the male sphinx moth Manduca sexta. J Comp Neurol 312: 264-278.
Hansson BS, Ljungberg H, Hallberg E and Löfstedt C (1992) Functional specialization of olfactory glomeruli in a moth.
Science 256: 1313-1315.
Hansson BS, Larsson M and Leal WS (1999) Green leaf volatile-detecting olfactory receptor neurones display very high
sensitivity and specificity in a scarab beetle. Physiol Entomol 24: 121-126.
Hansson BS, Carlsson MA and Kalinovà B (2001) Olfactory activation patterns in the antennal lobe of the sphinx moth,
Manduca sexta. J Comp Physiol A 189: 301-308.
Hartlieb E, Anton S and Hansson BS (1997) Dose-dependent response characteristics of antennal lobe neurons in the
male moth Agrotis segetum (Lepidoptera: Noctuidae). J Comp Physiol A 181: 469-476.
Heinbockel T and Kaissling K-E (1996) Variability of olfactory receptor neuron responses of female silkmoths (Bombyx
mori L.) to benzoic acid and (±)-linalool. J Insect Physiol 42: 565-578.
Heinbockel T, Christensen T and Hildebrand JG (1999) Temporal tuning of odor responses in pheromone-responsive
projection neurons in the brain of the sphinx moth Manduca sexta. J Comp Neurol 409: 1-12.
Heisenberg M, Borst A, Wagner S and Byers D (1985) Drosophila mushroom body mutants are deficient in olfactory
learning. J Neurogenet 2: 1-30.
Hekmat-Scafe DS, Steinbrecht RA and Carlson JR (1997) Coexpression of two odorant-binding protein homologs in
Drosophila: implications for olfactory coding. J Neurosci 17: 1616-1624.
Hekmat-Scafe DS, Scafe CR, McKinney AJ and Tanouye MA (2002) Genome-wide analysis of the odorant-binding protein
gene family in Drosophila melanogaster. Genome Res 12: 1357-1369.
Hildebrand JG and Shepherd GM (1997) Mechanisms of olfactory discrimination: converging evidence for common
principles across phyla. Ann Rev Neurosci 20: 595-631.
Hill CA, Fox AN, Pitts RJ, Kent LB, Tan PL, Chrystal MA, Cravchik A, Collins FH, Robertson HM and Zwiebel LJ (2002) G
protein-coupled receptors in Anopheles gambiae. Science 298: 176-178.
Hill SR, Hansson BS and Ignell R (2009) Characterisation of antennal trichoid sensilla from female southern house
mosquito, Culex quinquefasciatus Say. Chemical Senses 34: 231-252.
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides
R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P,
Atkinson PW, Baden H, de Berardinis V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M,
Cai S, Center A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A,
Dew I, Evans CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun
SL, Hogan JR, Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin
JJ, Lobo NF, Lopez JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O’Brochta DA,
Pfannkoch C, Qi R, Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J,
Thomasova D, Ton LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman
JR, Wu M, Yao A, Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della Torre A, Roth CW,
Louis C, Kalush F, Mural RJ, Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey
PT, Venter JC, Weissenbach J, Kafatos FC, Collins FH and Hoffman SL (2002) The genome sequence of the malaria
mosquito Anopheles gambiae. Science 298: 129-149.
Homberg U and Müller U (1999) Neuroactive substances in the antennal lobe. In: Hansson BS (ed) Insect Olfaction.
Springer, Berlin, Germany, pp 182-204.

84  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Hosler JS, Buxton KL and Smith BH (2000) Impairment of olfactory discrimination by blockade of GABA and nitric oxide
activity in the honey bee antennal lobes. Behav Neurosci 114: 514-525.
Huerta R, Nowotny T, García-Sanchez M, Abarbanel HD and Rabinovich MI (2004) Learning classification in the olfactory
system of insects. Neural Comput 16: 1601-1640.
Ignell R and Hansson BS (2005) Insect olfactory neuroethology – an electrophysiological perspective. In: Christensen
TA (ed) Methods in insect sensory neuroscience. CRC Press, Boca Raton, USA, pp 319-347.
Ignell R, Dekker T, Ghaninia M and Hansson BS (2005) Neuronal architecture of the mosquito deutocerebrum. J Comp
Neurol 493: 207-240.
Jefferis GS, Potter CJ, Chan AM, Marin EC, Rohlfing T, Maurer CR Jr and Luo L (2007) Comprehensive mapsof Drosophila
higher olfactory centers: spatially segregated fruit and pheromone representation. Cell 128: 1187-1203.
Jönsson M and Anderson P (1999) Electrophysiological response to herbivore-induced host plant volatiles in the moth
Spodoptera littoralis. Physiol Entomol 24: 377-385.
Joerges J, Küttner A, Galizia G and Menzel R (1997) Representations of odours and odour mixtures visualized in the
honeybee brain. Nature 387: 285-288.
Justice RW, Dimitratos S, Walter MF, Woods DF and Biessmann H (2003a) Sexual dimorphic expression of putative
antennal carrier protein genes in the malaria vector Anopheles gambiae. Insect Mol Biol 12: 581-594.
Justice RW, Biessmann H, Walter MF, Dimitratos SD and Woods DF (2003b) Genomics spawns novel approaches to
mosquito control. Bioessays 25: 1011-1020.
Justus K, Murlis J, Jones C and Carde R (2002). Measurement of odor-plume structure in a wind tunnel using a
photoionization detector and a tracer gas. Environ Fluid Mech 2: 115-142.
Kain P, Chakraborty TS, Sundaram S, Siddiqi O, Rodrigues V and Hasan G (2008) Reduced odor responses from antennal
neurons of G(q)alpha, phospholipase Cbeta, and rdgA mutants in Drosophila support a role for a phospholipid
intermediate in insect olfactory transduction. J Neurosci 30: 4745-4755.
Kaissling K-E (1971) Insect olfaction. In: Beidler LM (ed) Handbook of sensory physiology, Vol.IV; Chem senses. Springer-
Verlag, Berlin, Germany, pp 351-431.
Kaissling K-E (1986) Chemo-electrical transduction in insect olfactory receptors. Ann Rev Neurosci 9: 121-145.
Kaissling K-E and Priesner E (1970) Die Riechschwelle des Seidenspinners. Naturwissenschaften 57: 23-28.
Kanerva P (1988) Sparse distributed memory. MIT Press, Cambridge, MA, USA.
Kazama H and Wilson RI (2008) Homeostatic matching and nonlinear amplification at identified central synapses.
Neuron 58: 401-413.
Keil TA (1999) Morphology and development of the peripheral olfactory organs. In: Hansson BS (ed) Insect olfaction.
Springer, Berlin, pp 5-47.
Kennedy JS, Ludlow AR and Sanders CJ (1981) Guidance of flying male moths by wind-borne sex pheromone. Physiol
Entomol 6: 395-412.
Kramer E (1976) The orientation of walking honeybees in odour fields with small concentration gradients. Physiol
Entomol 1: 27-37.
Krashes MJ, Keene AC, Leung B, Armstrong JD and Waddell S (2007) Sequential use of mushroom body neuron subsets
during Drosophila odor memory processing. Neuron 53: 103-115.
Krieger J, von Nickisch-Rosenegk E, Mameli M, Pelosi P and Breer H (1996) Binding proteins from the antennae of
Bombyx mori. Insect Biochem Mol Biol 26: 297-307.
Kruse SW, Zhao R, Smith DP and Jones DN (2003) Structure of a specific alcohol-binding site defined by the odorant
binding protein LUSH from Drosophila melanogaster. Nat Struct Biol 10: 694-700.
Kwon HW, Lu T, Rützler M and Zwiebel LJ (2006) Olfactory responses in a gustatory organ of the malaria vector mosquito
Anopheles gambiae. PNAS 103: 13526-13531.
Kwon JY, Dahanukar A, Weiss LA and Carlson JR (2007) The molecular basis of CO2 reception in Drosophila. PNAS 104:
3574-3578.
Larsson MC, Leal WS and Hansson BS (1999) Olfactory receptor neurons specific to chiral sex pheromone components
in male and female Anomala cuprea beatles (Coleoptera: Scarabaeidae). J Comp Physiol A 184: 353-359.

Olfaction in vector-host interactions 85

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Larsson MC, Leal WS and Hansson BS (2001) Olfactory receptor neurons detecting plant odours and male volatiles in
Anomala cuprea beetles (Coleoptera: Scarabaeidae). J Insect Physiol 47: 1065-1076.
Larsson MC, Domingos AI, Jones WD, Chiappe ME, Amrein H and Vosshall LB (2004) Or83b encodes a broadly expressed
odorant receptor essential for Drosophila olfaction. Neuron 43: 703-714.
Laue MR, Steinbrecht RA and Ziegelberger G (1994) Immunocytochemical localization of general odorant-binding
protein in olfactory sensilla of the silkmoth Antheraea polyphemus. Naturwissenschaften 81: 178-180.
Laurent G (2002) Olfactory network dynamics and the coding of multidimensional signals. Nat Rev Neurosci 3: 885-895.
Lei H and Hansson BS (1999) Central processing of pulsed pheromone signals by antennal lobe neurons in the male
moth Agrotis segetum. J Neurophysiol 81: 1113-1122.
Lemon WC and Getz WM (1998) Responses of cockroach antennal lobe projection neurons to pulsatile stimuli. Ann NY
Acad Sci 855: 517-520.
Li ZX, Pickett JA, Field LM and Zhou JJ (2005) Identification and expression of odorant-binding proteins of the malaria-
carrying mosquitoes Anopheles gambiae and Anopheles arabiensis. Arch Insect Biochem Physiol 58: 175-189.
Liljefors T, Thelin B and Van der Pers JNC (1984) Structure-activity relationships between stimulus molecules and
response of a pheromone receptor cell in turnip moth, Agrotis segetum: modifications of the acetate group. J
Chem Ecol 10: 1661-1675.
Liljefors T, Thelin B, Van der Pers, JNC and Lofstedt C (1985) Chain-elongated analogues of a pheromone component of
the turnip moth, Agrotis segetum. A structure-activity study using molecular mechanics. J Chem Soc Perkin Trans
II 1957-1962.
Liljefors T, Bengtsson M and Hansson BS (1987) Effects of double-bond configuration on interaction between a moth
sex pheromone component and its receptor: a receptor-interaction model based on molecular mechanics. J Chem
Ecol 13: 2023-2040.
Lin HH, Lai JS, Chin AL, Chen YC and Chiang AS (2007) A map of olfactory representation in the Drosophila mushroom
body. Cell 128: 1205–1217.
Lu T, Qiu YT, Wang G, Kwon JY, Rutzler M, Kwon HW, Pitts RJ, van Loon JJ, Takken W, Carlson JR and Zwiebel LJ (2007)
Odor coding in the maxillary palp of the malaria vector mosquito Anopheles gambiae. Curr Biol 17: 1533-1544.
Malnic B (2007) Searching for the ligands of odorant receptors. Mol Neurobiol 35: 175-181.
Malun D (1991a) Inventory and distribution of synapses of identified uniglomerular projection neurons in the antennal
lobe of Periplaneta americana. J Comp Neurol 305: 348-360.
Malun D (1991b) Synaptic relationship between GABA-immunoreactive neurons and an identified uniglomerular
projection neuron in the antennal lobe of Periplaneta americana: A double-labeling electron microscopic study.
Histochemistry 96: 197-207.
Mafra-Neto A and Cardé RT (1994) Fine-scale structure of pheromone plumes modulates upwind orientation of flying
moths. Nature 369: 142-144.
Mafra-Neto A and Cardé RT (1998) Rate of realized interception of pheromone pulses in different wind speeds modulates
almond moth orientation. J Comp Physiol A 182: 563-572.
Marin EC, Jefferis GS, Komiyama T, Zhu H and Luo L (2002) Representation of the glomerular olfactory map in the
Drosophila brain. Cell 109: 243-255.
Marion-Poll F and Tobin TR (1992) Temporal coding of pheromone pulses and trains in Manduca sexta. J Comp Physiol
A 171: 505-512.
Mason JR, Bean NJ and Clark L (1993) Development of chemosensory attractants for white-tailed deer (Odocoileus
virginianus). Crop Protection 12: 448-452.
Matsuo T, Sugaya S, Yasukawa J, Aigaki T and Fuyama Y (2007) Odorant-binding proteins OBP57d and OBP57e affect
taste perception and host-plant preference in Drosophila sechellia. PLoS Biol 5: 985-996.
McIver SB (1982) Sensilla of mosquitoes (Diptera: Culicidae). J Med Entomol 19: 489-535.
McGuire SE, Le PT and Davis RL (2001) The role of Drosophila mushroom body signaling in olfactory memory. Science
293: 1330-1333.

86  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Melo AC, Rützler M, Pitts RJ and Zwiebel LJ (2004) Identification of a chemosensory receptor from the yellow fever
mosquito, Aedes aegypti, that is highly conserved and expressed in olfactory and gustatory organs. Chem Senses
29: 403-410.
Moore PA (1994) A model of the role of adaptation and disadaptation in olfactory receptor neurons: implications for
the coding of temporal and intensity patterns in odor signals. Chem Senses 19: 71-86.
Murlis J, Jones CD (1981) Fine-scale structure of odour plumes in relation to distant pheromone and other attractant
sources. Physiol Entomol 6: 71-86.
Mwilaria EK, Ghatak C and Daly KC (2008) Disruption of GABAA in the insect antennal lobe generally increases odor
detection and discrimination thresholds. Chem Senses 33: 267-281.
Nakagawa T, Sakurai T, Nishioka T and Touhara K (2005) Insect sex pheromone signals mediated by specific combinations
of olfactory receptors. Science 307: 1638-1642.
Neuhaus EM, Gisselmann G, Zhang W, Dooley R, Störtkuhl K and Hatt H (2005) Odorant receptor heterodimerization in
the olfactory system of Drosophila melanogaster. Nat Neurosci 8: 15-17.
Nene V, Wortman JR, Lawson D, Haas B, Kodira C, Tu ZJ, Loftus B, Xi Z, Megy K, Grabherr M, Ren Q, Zdobnov EM, Lobo
NF, Campbell KS, Brown SE, Bonaldo MF, Zhu J, Sinkins SP, Hogenkamp DG, Amedeo P, Arensburger P, Atkinson
PW, Bidwell S, Biedler J, Birney E, Bruggner RV, Costas J, Coy MR, Crabtree J, Crawford M, Debruyn B, Decaprio D,
Eiglmeier K, Eisenstadt E, El-Dorry H, Gelbart WM, Gomes SL, Hammond M, Hannick LI, Hogan JR, Holmes MH,
Jaffe D, Johnston JS, Kennedy RC, Koo H, Kravitz S, Kriventseva EV, Kulp D, Labutti K, Lee E, Li S, Lovin DD, Mao C,
Mauceli E, Menck CF, Miller JR, Montgomery P, Mori A, Nascimento AL, Naveira HF, Nusbaum C, O‘leary S, Orvis
J, Pertea M, Quesneville H, Reidenbach KR, Rogers YH, Roth CW, Schneider JR, Schatz M, Shumway M, Stanke M,
Stinson EO, Tubio JM, Vanzee JP, Verjovski-Almeida S, Werner D, White O, Wyder S, Zeng Q, Zhao Q, Zhao Y, Hill CA,
Raikhel AS, Soares MB, Knudson DL, Lee NH, Galagan J, Salzberg SL, Paulsen IT, Dimopoulos G, Collins FH, Birren B,
Fraser-Liggett CM and Severson DW (2007) Genome sequence of Aedes aegypti, a major arbovirus vector. Nature
316: 1718-1723.
Ng M, Roorda RD, Lima SQ, Zemelman BV, Morcillo P and Miesenböck G (2002) Transmission of olfactory information
between three populations of neurons in the antennal lobe of the fly. Neuron 36: 463-474.
Ochieng‘ SA, Anderson P and Hansson BS (1995) Antennal lobe projection patterns of olfactory receptor neurons
involved in sex pheromone detection in Spodoptera littoralis (Lepidoptera: Noctuidae). Tissue Cell 27: 221-232.
Olsen SR and Wilson RI (2008) Lateral presynaptic inhibition mediates gain control in an olfactory circuit. Nature 452:
956-960.
Olsen SR, Bhandawat V and Wilson RI (2007) Excitatory interactions between olfactory processing channels in the
Drosophila antennal lobe. Neuron 54: 89-103.
Olshausen BA and Field DJ (2004) Sparse coding of sensory inputs. Curr Opin Neurobiol 14: 481–487.
Park SK, Shanbhag SR, Wang Q, Hasan G, Steinbrecht RA and Pikielny CW (2000) Expression patterns of two putative
odorant-binding proteins in the olfactory organs of Drosophila melanogaster have different implications for their
functions. Cell Tissue Res 300: 181-192.
Pelosi P (1998) Odorant-binding proteins: Structural aspects. Ann NY Acad Sci 30: 281-293.
Pickett JA and Woodcock CM (1996) The role of mosquito olfaction in oviposition site location and in the avoidance of
unsuitable hosts. Ciba Found Symp 200, pp 109-119; discussion pp 119-123, pp 178-183.
Priesner E (1979) Progress in the analysis of pheromone receptor systems. Ann Zool Ecol Anim 11: 533-546.
Priesner E, Jacobson M and Bestmann HJ (1975) Structure-response relationships in noctuid sex pheromone reception.
Z Naturf 30: 283-293.
Qiu YT, Smallegange RC, Van Loon JJ, Ter Braak CJ and Takken W (2006a) Interindividual variation in the attractiveness
of human odours to the malaria mosquito Anopheles gambiae s. s. Med Vet Entomol 20: 280-287.
Qiu YT, Van Loon JJ, Takken W, Meijerink J and Smid HM (2006b) Olfactory coding in antennal neurons of the malaria
mosquito, Anopheles gambiae. Chem Senses 31: 845-863.
Raming K, Krieger J, Breer H (1989) Molecular cloning of an insect pheromone-binding protein. FEBS Lett 256: 215-218.
Root CM, Semmelhack JL, Wong AM, Flores J, Wang JW (2007) Propagation of olfactory information in Drosophila. PNAS
104: 11826-11831.

Olfaction in vector-host interactions 87

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Rospars JP (1988) Structure and development of the insect antenno deutocerebral system. Int J Insect Morphol Embryol
17: 243-294.
Sachse S and Galizia CG (2002) Role of inhibition for temporal and spatial odor representation in olfactory output
neurons: A calcium imaging study. J Neurophysiol 87: 1106-1117.
Sachse S and Galizia CG (2003) The coding of odour-intensity in the honey bee antennal lobe: local computation
optimizes odour representation. Eur J Neurosci 18: 2119-2132.
Sakura M, Okada R and Mizunami M (2002) Olfactory discrimination of structurally similar alcohols by cockroaches. J
Comp Physiol A 188: 787-797.
Rospars JP, Lánský P, Tuckwell HC and Vermeulen A (1996) Coding of odour intensity in a steady-state deterministic
model of an olfactory receptor neuron. J Comput Neurosci 3: 51-72.
Sato K, Pellegrino M, Nakagawa T, Nakagawa T, Vosshall LB and Touhara K (2008) Insect olfactory receptors are
heteromeric ligand-gated ion channels. Nature 24: 1002-1006.
Schachtner J, Schmidt M and Homberg U (2005) Organization and evolutionary trends of primary olfactory brain
centres in Tetraconata (Crustacea + Hexapoda). Arthropod Struct Dev 34: 257-299.
Schmidt CW (2005) Outsmarting olfaction: the next generation of mosquito repellents. Environ Health Perspect 113:
468-471.
Schofield CJ and Moreman K (1976) Letter: Apparent absence of a sex attractant in adult Triatoma infestans (Klug),
vector of Chagas‘ disease. Trans R Soc Trop Med Hyg 70: 165-166.
Schwaerzel M, Heisenberg M and Zars T (2002) Extinction antagonizes olfactory memory at the subcellular level. Neuron
35: 951-960.
Seki, Y, Aonuma H and Kanzaki R (2005). Pheromone processing center in the protocerebrum of Bombyx mori revealed
by nitric oxide-induced anti-cGMP immunocytochemistry. J Comp Neurol 481: 340-351.
Sengul MS (2008) Two odorant-binding protein genes in mosquitoes: Comparative genomics, expression and function.
Dissertation, Virginia Polytechnic Institute and State University, USA.
Sengul MS and Tu Z (2008) Characterization and expression of the odorant-binding protein 7 gene in Anopheles
stephensi and comparative analysis among five mosquito species. Insect Mol Biol 17: 631-645.
Shang Y, Claridge-Chang A, Sjulson L, Pypaert M and Miesenböck G (2007) Excitatory local circuits and their implications
for olfactory processing in the fly antennal lobe. Cell 128: 601-612.
Shields VDC and Hildebrand JG (2001) Recent advances in insect olfaction, specifically regarding the morphology and
sensory physiology of antennal sensilla of the female sphinx moth Manduca sexta. Micros Res Techniq 55: 307-329.
Siju, KP, Hill SR, Hansson BS and Ignell R (2010) Influence of blood meal on the responsiveness of olfactory receptor
neurons in antennal sensilla trichodea of the yellow fever mosquito, Aedes aegypti. Journal of Insect Physiology
56: 659-665
Skiri HT, Galizia CG and Mustaparta H (2004) Representation of primary plant odorants in the antennal lobe of the moth
Heliothis virescens using calcium imaging. Chem Senses 29: 253-267.
Steinbrecht RA, Ozaki M and Ziegelberger G (1992) Immunocytochemical localization of pheromone-binding protein
in moth antenna. Cell Tissue Res 270: 287-302.
Steinbrecht RA, Laue MR and Ziegelberger G (1995) Immunolocalization of pheromone-binding protein and general
odorant-binding protein in olfactory sensilla of the silkmoths Antheraea and Bombyx. Cell Tissue Res 282: 203-217.
Stensmyr MC, Dekker T and Hansson BS (2003) Evolution of the olfactory code in the Drosophila melanogaster subgroup.
Proc R Soc Lond B Biol Sci 270: 2333-2340.
Strausfeld NJ, Hansen L, Li Y, Gomez RS and Ito K (1998) Evolution, discovery, and interpretations of arthropod mushroom
bodies. Learning Memory 5: 11-37.
Sutcliffe JF (1994) Sensory bases of attractancy: morphology of mosquito olfactory sensilla – a review. J Am Mosq
Control Assoc 10: 309-315.
Syed Z and Leal WS (2007) Maxillary palps are broad spectrum odorant detectors in Culex quinquefasciatus. Chem
Senses 32: 727-738.
Syed Z, Ishida Y, Taylor K, Kimbrell DA and Leal WS (2007) Pheromone reception in fruit flies expressing a moth’s odorant
receptor. PNAS 103: 16538-16543.

88  Olfaction in vector-host interactions

 4. Odour coding and neural connections

Todd JL and Baker TC (1999) Function of peripheral olfactory organs. In: Hansson BS (ed) Insect olfaction. Springer-
Verlag, Berlin, Germany, pp 67-96.
Touhara K (2007) Deorphanizing vertebrate olfactory receptors: recent advances in odorant-response assays.
Neurochem Int 51: 132-139.
Tsodyks MV and Feigelman MV (1988) The enhanced storage capacity in neural networks with low activity level.
Europhys Lett 6: 101-105.
Turner GC, Bazhenov M and Laurent G (2008) Olfactory representations by Drosophila mushroom body neurons. J
Neurophysiol 99: 734-746.
Urban NN (2002) Lateral inhibition in the olfactory bulb and in olfaction. Physiology and Behavior 77: 607-612.
Vergassola ME. Villermaux and Shraiman BI (2007) ‚Infotaxis‘ as a strategy for searching without gradients. Nature 445:
406-409.
Vickers NJ (2006) Winging it: moth flight behavior and responses of olfactory neurons are shaped by pheromone plume
dynamics. Chem Senses 31: 155-166.
Vickers NJ and Baker TC (1994) Reiterative responses to single strands of odor promote sustained upwind flight and
odor source location by moths. PNAS 91: 5756-5760.
Vitta AC, Mota TR, Diotaiuti L and Lorenzo MG (2007) The use of aggregation signals by Triatoma brasiliensis (Heteroptera:
Reduviidae). Acta Tropica 101: 147-152.
Vogt RG and Riddiford LM (1981) Pheromone binding and inactivation by moth antennae. Nature 293: 161-163.
Vogt RG and Lerner MR (1989) Two groups of odorant binding proteins in insects suggest specific and general olfactory
pathways. Neurosci Abstr 15: 1290.
Vogt RG, Köhne AC, Dubnau JT and Prestwich GD (1989) Expression of pheromone binding proteins during antennal
development in the gypsy moth Lymantria dispar. J Neurosci 9: 3332-3346.
Vogt RG, Prestwich GD and Lerner MR (1991a) Odorant-binding-protein subfamilies associate with distinct classes of
olfactory receptor neurons in insects. J Neurobiol 22: 74-84.
Vogt RG, Rybczynski R and Lerner MR (1991b) Molecular cloning and sequencing of general odorant-binding proteins
GOBP1 and GOBP2 from tobacco hawk moth Manduca sexta: comparisons with other insect OBPs and their signal
peptides. J Neurosci 11: 2972-2984.
Vosshall LB and Stocker RF (2007) Molecular architecture of smell and taste in Drosophila. Annu Rev Neurosci 30: 505-533.
Vosshall LB, Amrein H, Morozov PS, Rzhetsky A and Axel R (1999) A spatial map of olfactory receptor expression in the
Drosophila antenna. Cell 96: 725-736.
Vosshall LB, Wong AM and Axel R (2000). An olfactory sensory map in the fly brain. Cell 102: 147-159.
Waldrop B, Christensen TA and Hildebrand JG (1987) GABA-mediated synaptic inhibition of projection neurons in the
antennal lobes of the sphinx moth, Manduca sexta. J Comp Physiol A 161: 23-32.
Wang JW, Wong AM, Flores J, Vosshall LB and Axel R (2003) Two-photon calcium imaging reveals an odor-evoked map
of activity in the fly brain. Cell 112: 271-282.
Wang Y, Mamiya A, Chiang AS and Zhong Y (2008) Imaging of an early memory trace in the Drosophila mushroom body.
J Neurosci 28: 4368–4376.
Wicher D, Schäfer R, Bauernfeind R, Stensmyr MC, Heller R, Heinemann SH and Hansson BS (2008) Drosophila odorant
receptors are both ligand-gated and cyclic-nucleotide-activated cation channels. Nature 24: 1007-1011.
Van Wijk M, Wadman WJ, Sabelis MW (2006) Morphology of the olfactory system in the predatory mite Phytoseiulus
persimilis. Exp Appl Acarol 40: 217-229.
Willis MA and Arbas EA (1991) Odor-modulated upwind flight of the sphinx moth, Manduca sexta L. J Comp Physiol A
169: 427-440.
Willshaw D and Longuet-Higgins HC (1970) Associative memory model. In: Meltzer B, Michie O (eds) Machine
intelligence Vol 5. University of Edinburgh Press, Edinburgh, UK, pp 351-359.
Wilson RI and Laurent G (2005) Role of GABAergic inhibition in shaping odor-evoked spatiotemporal patterns in the
Drosophila antennal lobe. J Neurosci 25: 9069-9079.
Wilson RI, Turner GC and Laurent G (2004) Transformation of olfactory representations in the Drosophila antennal lobe.
Science 303: 366-370.

Olfaction in vector-host interactions 89

Rickard Ignell, Meryem S. Sengul, Sharon R. Hill and Bill S. Hansson

Wogulis M, Morgan T, Ishida Y, Leal WS and Wilson DK (2006) The crystal structure of an odorant binding protein from
Anopheles gambiae: Evidence for a common ligand release mechanism. Biochem Biophysics Res Commun 339:
157-164.
Wojtasek H, Hansson BS and Leal WS (1998) Attracted or repelled? A matter of two neurons, one pheromone binding
protein, and a chiral centre. Biochem Biophys Res Commun 250: 217-222.
Wong AM, Wang JW and Axel R (2002). Spatial representation of the glomerular map in the Drosophila protocerebrum.
Cell 109: 229-241.
Xia Y and Zwiebel LJ (2006) Identification and characterization of an odorant receptor from the West Nile virus mosquito,
Culex quinquefasciatus. Insect Biochem Mol Biol 36: 169-176.
Xu PX, Zwiebel LJ and Smith DP (2003) Identification of a distinct family of genes encoding atypical odorant-binding
proteins in the malaria vector mosquito, Anopheles gambiae. Insect Mol Biol 12: 549-560.
Xu P, Atkinson R, Jones DN and Smith DP (2005) Drosophila OBP LUSH is required for activity of pheromone-sensitive
neurons. Neuron 45: 193-200.
Yu D, Akalal DB and Davis RL (2006) Drosophila alpha/beta mushroom body neurons form a branch-specific, long-term
cellular memory trace after spaced olfactory conditioning. Neuron 52: 845-855.
Zhou JJ, He XL, Pickett JA and Field LM (2008) Identification of odorant-binding proteins of the yellow fever mosquito
Aedes aegypti: genome annotation and comparative analyses. Insect Mol Biol 17: 147-163.

90  Olfaction in vector-host interactions

5. The detection of carbon dioxide and its role in the orientation to
hosts by haematophagous insects

Alan J. Grant and Robert J. O’Connell

Abstract

Host orientation behaviours of mosquitoes and other blood-seeking insects are controlled
to a large degree by signals released by the host, including heat, moisture and sound, as well
as visual and olfactory cues. Clearly olfactory signals play a primary role in mediating such
behaviours. Although other olfactory signals such as lactic acid, 1-octen-3-ol and ammonia have
been implicated in orientation, perhaps the most important of these volatiles is carbon dioxide
(CO2). Carbon dioxide is a by-product of cellular respiration that is released in large amounts
by potential hosts and has been associated with mosquito behaviour since the early 1920s. In
this report we focus on aspects of carbon dioxide detection including the morphological and
physiological characteristics of sensory structures located on the maxillary palps which contain
receptor neurons responsive to CO2. The relevant characteristics of these receptor neurons
include the threshold and slope of the concentration response functions, the temporal pattern
of the discharge, the effects of different background CO2 concentrations and the physiological
condition of the insect. Among these conditions are age, diapause status and whether the
species is autogenous or anautogenous. The physiological characteristics of CO2 detection in
female mosquitoes are contrasted with similar characteristic in male mosquitoes as well as in
other insects such as biting midges (Ceratopogonidae). The recent advances involving molecular
research to further our understanding of the cellular processes of CO2 detection and the role of
deterrents and repellants on such processes are discussed. Finally, the uses of CO2 as a tool in
the design of control strategies including the development of traps for monitoring and potential
control are discussed.

Keywords: carbon dioxide, mosquitoes, biting midges, behaviour, sensory physiology, blood
feeding

Introduction

Each year, there are approximately three hundred million new cases involving the major diseases
transmitted by mosquitoes. One and a half million fatalities can be attributed to these diseases
or their complications each year (Breman 2001, Mayor 2008). For decades, efforts have been
made to better understand the mechanisms that underlie host attraction, feeding and mating of
haematophagous insects (Friend and Smith 1977, Galun 1977). In particular, many have attempted
to describe and synthesise the complex of signals which control host-seeking behaviour in
mosquitoes (Davis and Bowen 1994, Daykin et al. 1965, Takken 1991, Zwiebel and Takken 2004). A
number of factors including: moisture, heat, (Clements 1963, 1999); host volatiles such as: lactic
acid (Acree, Jr. et al. 1968, Smith et al. 1969), carbon dioxide (CO2) (Gillies 1980, Mayer and James
1969) and other compounds (Kline et al. 1990, 1991) appear to be important stimuli for mosquito
orientation.

Many biting insects, including anautogenous mosquitoes, require a blood meal before oogenesis.
For example, a female Aedes aegypti (L.) mates soon after emergence and sometime later she
blood feeds. Within several days of blood feeding, she lays her eggs. The blood serves as a primary

Olfaction in vector-host interactions 91

Alan J. Grant and Robert J. O’Connell

source of energy for egg development. To acquire this blood meal, the female mosquito must
locate a host for feeding. One of the sensory cues involved in this host-seeking behaviour is CO2;
a primary byproduct of cellular respiration, which is released in large quantities by all potential
hosts. Carbon dioxide had been implicated in mosquito orientation as early as the 1920s (Rudolfs
1922, Crumb 1922) and continues to be seen as one of the important chemical signals utilised
by mosquitoes and other haematophagous insects in their host-locating and feeding behaviours
(Bowen 1991, Galun 1977, Gillies 1980). To serve that end, female mosquitoes, as well as many
other biting insects, prove to be equipped with an array of sensors that are highly sensitive
and specifically tuned to respond to behaviourally relevant doses of CO2 (Grant and Kline 2003,
Grant and O’Connell 1996a, Grant et al. 1995, Kellogg 1970). Carbon dioxide-sensitive receptor
neurons have been identified and studied morphologically in mosquitoes (McIver 1972b), and
their physiological properties (Grant et al. 1995, Grant and O’Connell 1996b) have been explored in
several other arthropods, including Lepidoptera (Bogner 1990, Bogner et al. 1986), Hymenoptera
(Lacher 1964, Stange and Diesendorf 1973), other Diptera (Bogner 1992, Grant and Kline 2003)
and Arachnida (Steullet and Guerin 1992). It is clear from all these studies that attraction to hosts
is a complex behaviour that can be modified by a large number of environmental effectors many
of which may be unique to a particular species (Davis 1996).

Since the early extirpation experiments of Roth and others (Jones and Madhukar 1976, Omer
and Gillies 1971, Roth 1951), it has been clear that chemoreceptor neurons in sensilla located on
the maxillary palps together with those on the antenna, play an important role in the detection
and processing of the chemical stimuli implicated in initiating and modulating host location
and feeding behaviours in insects (Bowen 1991, Gillies 1980, Kline et al. 1991, Takken and Kline
1989). Early physiological studies of the peripheral sensory system in mosquitoes had focused
largely on the responses of olfactory receptor neurons in antennal sensilla. The antenna does
contain a highly sensitive L-lactic acid receptor neuron, whose activity can be modulated by the
behavioural repellent N,N-diethyl-m-toluamide (DEET) (Davis 1977, 1985, Davis and Sokolove
1976) The response characteristics of other antennal sensilla will be discussed by other author’s
in this volume and will not be addressed further in this review. A single early report (Kellogg
1970) described electrophysiological responses to CO2 in neurons housed in sensilla basiconica
on the maxillary palp of female mosquitoes. We now know that one of the three sensory neurons
normally found in sensilla basiconica responds to small increments in CO2 concentration and is
sensitive to concentrations that are appropriate for orientation behaviour (Grant et al. 1995). In
addition to the CO2-sensitive neuron, the maxillary palp sensillum is innervated by two other
spontaneously active receptor neurons. The neuron producing the smallest amplitude action
potential is sensitive to stimulation with low concentrations of 1-octen-3-ol (octenol) (Grant and
O’Connell 1996b), another mosquito attractant (Takken and Kline 1989). The neuron producing
the intermediate-sized action potential is sensitive to yet another compound (unpublished data)
and all three of these cells have axons that project directly to the brain. These three cells are
developmentally related to each other so it is possible that they all could share similar functional
purposes guiding host-locating behaviours.

Morphology of the maxillary palp and sensilla

Like many female mosquitoes, Ae. aegypti have specialised sensilla on the ventral lateral surface
of the fourth subsegment from the base of each maxillary palp (McIver 1972a, 1982). This overall
morphology is shared by many species of mosquitoes. For illustrative purposes we show in Figure
1A-B scanning electron micrographs (SEMs) from a female Culex pipiens Linnaeus. Male mosquitoes
also possess morphologically similar sensilla that are distributed along the distal half of the third

92  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

Copyright A.J. Grant C Copyright A.J. Grant D

Figure 1. Scanning electron micrographs (SEMs) of maxillary palps of mosquito species (A-C) and biting midges
(D). (A) Portion of the head of a 21-22 day old female Cx. pipiens showing the maxillary palps (MP), antennae
(ANT), and proboscis (PR). (B) Fourth subsegment of a maxillary palp of a 21-22 day old female Cx. pipiens
showing individual sensilla. Inset: individual sensillum containing a neuron sensitive to CO2. (C) Portion of the
maxillary palp of a male Ae. aegypti depicting the basiconic sensilla containing neurons sensitive to CO2. (D)
Portion of the maxillary palp of a female C. furens (Poey) showing a cluster of CO2-sensitive basiconic sensilla
situated in an enlarged depression on the distal portion of the third subsegment from the base.

subsegment (Figure 1C) (McIver 1972a, 1982). The presence of these sensilla is prevalent among
mosquitoes, but their numbers and distribution varies from species to species (Braverman and
Hulley 1979). Female Ae. aegypti possess approximately 25 sensilla on each palp. The sensilla
basiconica in both Ae. aegypti and Cx. quinquefasciatus Say are restricted to the fourth subsegment
from the head, whereas Anopheles stephensi Liston have sensilla basiconica distributed along the
terminal three subsegments of the palp. It has been suggested that differences in the number

Olfaction in vector-host interactions 93

Alan J. Grant and Robert J. O’Connell

and distribution of particular chemoreceptors on the antennae and maxillary palps of various
species of mosquito are related to differences in the insect’s relative preferences for particular
hosts (Braverman and Hulley 1979) or in their landing and probing responses to them. In addition
to the variation in distribution of sensilla, the overall length of the maxillary palps varies. The palps
of An. stephensi are significantly longer (about 1,450 µm) than the palps of either Ae. aegypti (about
350 µm) or Cx. quinquefasciatus (about 300 µm). Although there is considerable variation in the
external morphology of the palps and the distribution of sensilla along them the physiological
characteristics of the neurons that innervate them show strong similarities (Figure 2); particularly
with regards to the response of the neuron producing the largest amplitude action potential
(up to 300 µV peak to peak), which is reliably sensitive to stimulation with CO2 (Grant et al. 1995,
Kellogg 1970). However, our recent unpublished studies suggest that several Anopheline species
have reduced sensitivity. Lu et al. reported that An. gambiae Giles showed an elevated threshold
and a diminished dose response functions (Lu et al. 2007).

The presence of these CO2-sensitive sensilla is often correlated with blood feeding behaviour.
In other arthropods, sensilla with similar morphological features are found to contain neurons
which also respond to CO2. For example, biting midges possess similar sensilla with neurons that
respond to CO2. Although the Phantom Midge (Family Choaboridae) is taxonomically related to
the mosquito, it had long been thought to be a non-biting insect. Subsequent investigations
revealed the presence of blood in the digestive tract and an anatomical study revealed that these
animals have mouthparts appropriate for biting and that they also possess basiconic sensilla on
the palps which are morphologically similar to those examined here (McKeever and Pound 1979).
To date, this correlation only holds for females of the species.

175 Ae. aegypti

Ae. taeniorhynchus
150 An stephensii
Cx. quinquefasciatus
125 Cs. melanura
Impulses/2 seconds

100

75

50

25

0
0 200 400 600 800 1000
-25
Carbon dioxide (PPM)

Figure 2. Mean±SE number of action potentials during 2 s stimulus pulses of the indicated concentrations of
CO2 from s. basiconica on the maxillary palps of individuals from five different species of female mosquito (Ae.
aegypti, An. stephensi, Cx. quinquefasciatus, Cs. melanura (Coquillett), and Ae. taeniorhynchus (Wiedemann)).
All of these functions were established in a background environment containing 0 ppm CO2.

94  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

Methodology in electrophysiological studies

Standard electrophysiological techniques can be used to record extracellular responses from the
receptor neurons located in the sensilla found on insects. For CO2-sensitive neurons in biting insects
such as mosquitoes and biting midges, details of the recording methodology have been described
fully in previous reports (Grant and O’Connell 1986, O’Connell 1972) and are summarised below.
To insure stable recordings, the insect must be carefully immobilised with the target structure
positioned in a manner that allows unobstructed access of the recording microelectrode to the
selected sensillum. Whole adults are secured on a glass plate with minute strips of sticky tape
or fibre threads to secure the thorax, abdomen, legs and wings. Once the insect is secured, the
preparation is positioned under the objectives of a compound light microscope with an effective
magnification of approximately 750X.

The recording microelectrodes are hand-made from 125-µm diameter straightened tungsten wire.
They are electrolytically sharpened to a tip diameter less than 1 µm in a 10% electrolyte solution
(Galbreath and Galbreath 1977, Hubel 1957). An indifferent electrode of similar design is inserted
into the eye of the insect and the recording electrode is inserted through the cuticle at the base
of an individual sensillum. The microelectrodes are mounted in micromanipulators for positioning.
Following placement of the recording electrode in the fluid space of the sensillum the electrical
signals obtained from the neurons within the sensillum are band-passed filtered, amplified and
sent in parallel to an audio monitor and a computer for subsequent data acquisition, action
potential discrimination, analysis and storage. The basic work station, including the microscope,
manipulators, preparation stage and amplifier are all mounted on a vibration isolation table in a
Faraday cage. The electrical circuits required for this apparatus are isolated and independently
earthed through a dedicated ground.

Data acquisition and analysis are accomplished using a software programme Autospike, developed
by the late Dr. Jan N.C. van der Pers (www.syntech.nl/s). This programme is similar in principle to
one used earlier (O’Connell et al. 1973) and is designed to record and analyse electrophysiological
information from insect chemosensilla. This programme runs on a Windows platform with an
analog-to-digital/digital-to-analog (AD/DA) interface via an IDAC controller. Discrimination and
sorting of action potentials arising from the several neurons usually found in a single sensillum
are easily accomplished with Autospike. Electrophysiological preparations from mosquitoes and
other insects typically last for several hours before there are detectable changes in their response
properties.

Chemical stimulation of olfactory receptor neurons in insects is usually accomplished with two
opposing gas streams directed towards the sensory structure (Borroni and O’Connell 1992). One
of these is normally on and carries the desired background stream (550 mls/min). The second
stream is normally off and carries the desired stimulus (440 mls/min) stream. An independent
controller activates the valves controlling these two streams and also initiates data collection by
the software. To reliably manipulate CO2 stimulus concentration the flow dilution olfactometer
design (O’Connell and Mozell 1969) normally used in olfactory studies should not be used. In
this basic design a stock concentration of CO2 is diluted to the desired stimulus concentration
by adding diluent flow from a pure air stream and adjusting the ratio of flow in the two gas
streams to produce the final desired concentration. Although the bulk concentration of CO2 can
be adjusted in this fashion, its use reveals a considerable variation in neuronal response. This
appears to be due to inconsistent mixing within the stimulus stream. As we shall discuss later, CO2
receptor neurons appear to have an enhanced ability to follow rapid changes in concentration

Olfaction in vector-host interactions 95

Alan J. Grant and Robert J. O’Connell

especially when compared to the more tonic pheromone receptor neurons normally found in
insect antennae (Borroni and O’Connell 1992). Because of this factor, reliable CO2 stimuli should
be delivered without dilution from certified formulated gas cylinders, each containing a calibrated
concentration of CO2. For our studies we chose 0, 150, 300, 350, 600 and 1000 ppm CO2, each
containing 20% purified oxygen, and the remainder made up with purified nitrogen. Between CO2
stimulations, the preparation is bathed in a stream of carbon dioxide-free synthetic air (0 ppm
CO2). For each preparation 2 s pulses of CO2 are sequentially presented at each of the certified
concentrations. The interval between individual stimuli is normally 10-20 s. Following a brief rest
period of approximately one to five minutes the whole protocol is repeated. Reproducibility of
the response to a particular stimulus concentration is very high and individual responses at a
particular concentration rarely differ from each other by more than a few impulses. Three to five
repetitions of this protocol are made for each preparation to form the average. T-tests are then
performed for the grand average response across all preparations at each dose to determine if
there are any statistical differences between treatments.

Temporal pattern of response

The maxillary palp sensilla of both mosquitoes and biting midges (Diptera: Ceratopogonidae)
contain neurons that are responsive to stimulation with low concentrations of CO2. Stimulation
of these neurons with pulses of CO2 elicits typical biphasic action potentials with peak-to peak
amplitudes of several 100 µVs (Figure 3). This neural signal is sent to the insect brain where

Figure 3. Typical electrophysiological response from a female Cx. pipiens in response to a 2 s stimulation of 600
ppm CO2 in a background of 0 ppm CO2.

96  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

environmental CO2 information is processed. Processing in the central nervous system probably is
dependent not only on the presence or absence of impulse activity in particular afferents, but also
on the pattern of action potentials generated in them during the odour stimulus. Generally these
neurons do not produce nerve impulses in environments lacking CO2. To evaluate the temporal
pattern of activity for the CO2-sensitive neurons in the maxillary palp sensilla, we can evaluate
the instantaneous frequency of action potentials generated by these neurons during the stimulus
pulse (Figure 4). Both mosquito and biting midge CO2-sensitive cells respond with a sharp ‘phasic’
burst of activity, lasting 100-200 ms, followed by a lower tonic level of response whose duration
and magnitude is dependent on stimulus concentration and the length of the stimulus pulse. Step
increases in stimulus concentration elicit phasic increases in activity in both mosquito and biting
midge receptor neurons, step decreases in concentration elicit reciprocal phasic reductions in
activity followed by a slower increase in activity as the response returns to a level proportional to
the new background concentration of CO2. This latter pattern can only be observed when steps
in stimulus concentration are given from a background level of CO2. This pattern of discharge is
widely believed to provide a sharpening mechanism (Dionne 1992) that allows the nervous system
to detect the small changes in concentration that likely signals the presence of potential hosts
(Grant and O’Connell 1996a).

The difference in the temporal pattern of discharge may also reflect intrinsic differences in the way
in which the respective chemical signals in the environment are processed. The response properties
of CO2-sensitive receptor neurons to step changes in concentration suggest the presence of at
least two different mechanisms. On one hand, there appears to be a tonic response mechanism
that simply reads out a response rate proportional to the steady state concentration of CO2 with
relatively little desensitisation over time. On the other hand, there is also a rate-sensitive mechanism
whose output rapidly follows both the direction and the rate of change of CO2 concentration. As
is the case with cold receptor neurons in vertebrates (Darian-Smith et al. 1973) and invertebrates

20
Impulses/0.4 seconds

15

10

-5
0 4 8 12
Time (seconds)

Figure 4. The temporal pattern of response from a CO2-sensitive receptor neuron in a female Ae. aegypti. The open
symbols represent the average response to a 4 s stimulus pulse stepping up to 350 ppm CO2 from a background
concentration of 300 ppm. The solid symbols represent the average response to a 4 s stimulus pulse stepping down
to 300 ppm from a level of 350 ppm CO2. In both cases the number of action potentials was averaged in 400 ms
time bins and represents the mean from a single sensillum stimulated 10 times with each pulse protocol.

Olfaction in vector-host interactions 97

Alan J. Grant and Robert J. O’Connell

(Altner and Loftus 1985) these phasic properties provide a sharpening mechanism that greatly
exaggerates the response output of the receptor neuron as stimulus intensity changes rapidly.
For example, the average tonic levels of activity elicited by CO2 backgrounds of 300 and 350 ppm
differ from each other by only about 6 impulses/s (Figure 4). However, a rapid step from 300 to 350
and back to 300 ppm elicits peak phasic responses that differ from each other by approximately
38 impulses/s. Although we do not yet know the ‘following’ capabilities of the CO2 receptor, the
ability of the neuron to produce phasic bursts whose magnitude and sign are a function of the
magnitude and sign of the change in CO2 concentration should have the effect of enhancing
the sensory signal elicited and transmitted to the central nervous system by rapid changes in
concentration. It is likely that this ability of the system to follow rapid changes in concentration
accounts for the difficulty associated with standard dilution techniques of stimulation referenced
earlier.

Depending on the nature of the central processing associated with these temporal inputs, the
ability of the sensory system to follow rapid changes in CO2 concentration might be enhanced by
these characteristics. This pattern of response would also be particularly appropriate if mosquitoes
oriented to hosts by following the rapidly changing increments in CO2 concentration that are likely
to be encountered in nature (Elkinton et al. 1984), as opposed to simply flying upwind toward a
source in a steadily increasing gradient as might be imagined from a point source release of an
odorant. It should be clear then that receptor neurons in insect olfactory sense organs can vary in
both their chemical sensitivity and in their temporal characteristics (see Chapter 4, this volume).

Responses in different background concentrations of CO2

For ease of comparison, most electrophysiological tests reported here were to step increases
in CO2 concentration from a background environment lacking CO2 (0 ppm CO2). The resulting
response magnitude fits a log function dependent simply on CO2 concentration. However, biting
insects in nature never encounter environments without CO2. Therefore, when we tested the same
set of CO2 concentrations applied from different background levels of CO2 (Figure 5), we obtain
some useful additional information about the sensory neuron. Here too, we see the same basic
pattern of response to stimulation in different background concentrations of CO2 in mosquitoes
and biting midges. The average numbers of impulses generated in different background
concentrations during a 2 s stimulus containing a given CO2 concentration are similar or slightly
less as long as the background concentration is lower than the stimulus concentration. However,
if the background concentration is higher than the stimulus concentration, the average number
of impulses generated to the stimulus step is greatly diminished. For example, in backgrounds
containing 0, 150 or 300 ppm CO2 the magnitude of responses elicited by a stimulus containing
600 ppm CO2, are comparable to each other. However, in a background concentration of 1000 ppm
CO2, the magnitude of response to a similar stimulus containing 600 ppm CO2 is greatly reduced.

One implication of this observation is straightforward. One can imagine that the reduced sensory
signal may impede the animal’s ability to find hosts in high backgrounds. Conversely, the animal
is more likely to only find hosts that emit CO2 concentrations higher than the background. From
this we can surmise that continued increases in the environmental concentrations of CO2 might
result in large shifts in host-seeking behaviour. However, without additional information on the
response of hosts to these changes it is impossible to predict how these shifts could alter host
detection and subsequent disease transmission.

98  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

180
160
140
Impulses/2 seconds

120
100
80
60
40
20
0
0
150
300
350
600
1000

0
150
300
350
600
1000

0
150
300
350
600
1000

0
150
300
350
600
1000

0
150
300
350
600
1000

0
150
300
350
600
1000
Background 0 PPM 150 PPM 300 PPM 350 PPM 600 PPM 1000 PPM
concentration
Carbon dioxide (PPM)

Figure 5. Mean + SE responses of CO2-sensitive receptor neurons in female Ae. aegypti. The stimuli consisted of 2
s pulses of different concentrations of CO2 delivered from each of the indicated background levels of CO2.

Detection of CO2 by male mosquitoes

Male mosquitoes possess morphologically similar sensilla to those structures found on the
maxillary palps of females (McIver 1972a, 1980). They prove to contain CO2-sensitive receptor
neurons as well (Grant et al. 1995). The spatial distribution of these sensilla basiconica on the male
palp is not restricted to the fourth sub-segment from the head, as in the female of many species.
Instead, the distribution includes the distal half of the third subsegment from the head (Figure
1C). Neurons that innervate these male sensilla have thresholds and sensitivities to CO2 that are
similar to those in females (Figure 6, Grant and O’Connell 2007).

The apparent ability of males to detect CO2 raises questions about the potential function of
this chemical stimulus in their biology. Male mosquitoes are not known to blood feed, as their
mouthparts are not adapted for piercing (Matheson 1945). But males do feed on plant nectars.
Perhaps their ability to detect CO2 is utilised in their search for these food sources. Depending on
when during the day they are active this input may be useful in detecting either sources or sinks
of CO2 involved with the photosynthetic activity of plants. Alternatively, male Ae. aegypti have
been observed near human hosts where mating occurs as the females seek a blood meal (Hartberg
1971). This suggests that males of some species may use CO2 as a secondary cue to orient to places
where females are likely to be more abundant and available for mating.

Effect of age on CO2 responses

Following emergence, female mosquitoes typically perform a series of behaviours which includes
mating, feeding and oviposition. For an anautogenous mosquito such as Ae. aegypti, a blood
meal is required to obtain the protein needed for egg production. To acquire this blood, the
female engages in periods of host-seeking. Host seeking is governed by sensory cues associated

Olfaction in vector-host interactions 99

Alan J. Grant and Robert J. O’Connell

250

200
Impulses/2 seconds

150

100

50

0
0 200 400 600 800 1000

-50 CO2 (ppm)

Figure 6. Mean±SE concentration-response function from male (open circles; n=20) and female (solid square;
n=30). CO2-sensitive receptor neurons in female Ae. aegypti in response to a 2 s stimulation of the indicated
concentration of CO2. Stimulation with the control concentration of 0 ppm CO2 elicited no response in any
mosquito. Male responses were not statistically different from female responses at any concentration tested.

with host-produced odours (Takken 1991) including CO2. Mating generally occurs within the first
few days following emergence (Christopher 1960, Clements 1999). Blood feeding is not required
for copulation to occur. Numerous behavioural reports suggest that for most anautogenous
mosquitoes relatively little feeding activity occurs during the first few days of adult life and host-
seeking and feeding activities increase dramatically from then to 96 h after emergence (Christopher
1960, Davis 1984a, Seaton and Lumsden 1941). This enhanced host-seeking level is maintained
for several weeks. Since host-seeking behaviour changes as the insect ages, it is interesting to
evaluate the dynamic sensitivity of the CO2 receptor neurons across this timeframe. To that end,
we recorded electrophysiological responses from sensory neurons to stimulation with various
concentrations of CO2 in newly emerged adults (0-2 d old) when host orientation is reduced, and
in older insects (4-7 d old) when host orientation reaches a maximum. Female Ae. aegypti show
pronounced differences in the sensitivity of their CO2 receptor neurons across these young and
older age classes (Figure 7, Grant and O’Connell 2007). Young females have a significantly lower
overall response rate to stimulation with 300, 600 and 1000 ppm CO2 when compared with the
same stimuli in older females. Recording from male Ae. aegypti did not show a comparable age
dependent difference for any of the CO2 concentrations tested (Grant and O’Connell 2007).This
difference in sensitivity as a function of age was also observed for female Cx. pipiens (unpublished
results). Similar differences in the rates of development of sensory systems have been studied
electrophysiologically and reported in other insects as well (Blaney et al. 1986, Crnjar et al. 1990,
Davis 1984b, Schweitzer et al. 1976).

In addition to the variation in overall sensitivity between young and older female mosquitoes, the
temporal pattern of discharge to a continuous stimulus (1000 ppm CO2) also differed between
the two ages (Grant and O’Connell 2007). As we have argued before (Grant and O’Connell 1996b)

100  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

300

250

Impulses/2 seconds 200

150

100

50

0
0 200 400 600 800 1000

-50 CO2 (ppm)

Figure 7. Mean±SE concentration-response functions from young (open square; n=15) and older (solid circles;
n=15) CO2-sensitive receptor neurons in female Ae. aegypti in response to a 2 second stimulation of the indicated
concentration of CO2. Stimulation with the control concentration of 0 ppm CO2 elicited no response in any
mosquito. Responses were statistically different from each other at concentrations of 300, 600 and 1000 ppm. The
solid circles and open squares represent data points and the solid lines represent functions fitted to these points.

the temporal discharge pattern may enhance the neuron’s ability to discriminate rapid changes
in CO2 concentration that are presumed to occur in the downwind stimulus plume emanating
from a potential host. Thus, enhancing the capability of the peripheral sensory system to detect
CO2 changes as the female mosquito matures should enhance host-seeking in older females.
On the other hand, these changes in sensitivity may be accidental and simply reflect the normal
physiological maturation of the sensory system or the changes in the periphery may be timed
by other more central processes in the mosquito which are designed to optimise the sensory
capabilities of the system when it is required to mediate host-seeking behaviour. It is not
possible to state the causal relationships between these host-related sensory capabilities and
the development of host-seeking behaviours. However, one is tempted to speculate that the
maturation we have demonstrated in the sensory system is responsible for the age-related shift
in host-seeking behaviour.

CO2 detection in biting midges

In addition to mosquitoes, other biting insects detect CO2 including but not limited to Tabanidae,
Simuliidae, some Muscidae and Ceratopogonidae. Among the Ceratopogonids or biting midges,
Culicoides is the most important genus with respect to the health and comfort of humans and
animals (Kettle 1965, Linley et al. 1983). Annoyance from adult female biting midges often may
become serious enough to affect outdoor recreational and work-related activities. Their presence
also has an adverse effect on tourism and land development, especially in coastal areas of
the United States (Linley and Davies 1971) and in parts of Europe. Animal and human health

Olfaction in vector-host interactions 101

Alan J. Grant and Robert J. O’Connell

are affected by disease organisms transmitted by adult female biting midges, which include
protozoans, filarial worms and viruses (Blanton and Wirth 1979, Kettle 1965, Linley et al. 1983).

Sensilla basiconica on the maxillary palps of female C. furens (Poey), C. stellifer (Coquillett) and
C. mississippiensis (Hoffman) all contain a receptor neuron which responds to stimulation with
low levels of CO2 (Figure 1D). These neurons produce typical biphasic action potentials with
peak-to-peak amplitudes of up to 350 µV. As in our mosquito work, we recorded responses
to stimulation with 0, 150, 300, 600 and 1000 ppm CO2. The Culicoides CO2-sensitive neurons
have basic physiological characteristics that are similar to those in the mosquito CO2-sensitive
neurons (Grant and O’Connell 1996b, 2007, Grant et al. 1995). Both mosquito and biting midge
receptor neurons respond with a phasic-tonic discharge pattern to increases or decreases in CO2
concentration. The neurons of both are silent in environments without CO2. However, there are
several interesting differences between biting midges and mosquito response properties. First, in
most cases, their response thresholds appear different. Stimulation with 150 ppm CO2 elicited a
response in C. furens, which is statistically larger than the activity seen in response to stimulation
with 0 ppm CO2 (Figure 8). Aedes aegypti, on the other hand, often require higher concentrations
to elicit a significant increase in response (Grant et al. 1995). This indicates that the threshold of
detection is lower for C. furens and as a consequence their neurons are more likely to be active
at ambient levels. Second, the slopes of the concentration response functions are considerably
different. The response function for Ae. aegypti is steeper than for Culicoides. This indicates that a
small change in stimulus concentration will elicit a larger change in impulse activity in Aedes as

160
Ae. aegypti
C. furens
140
C. stellifer
C. mississippiensis
120
Impulses/2 seconds

100

80

60

40

20

0
0 200 400 600 800 1000

Carbon dioxide concentration (ppm)

Figure 8. Concentration-response functions from three species of Culicoides [C. furens (n=15), C. mississippiensis
(n=3) and C. stellifer (Coquillett) (n=6)] in response to graded stimulations of CO2. For comparison, also included
is the concentration response function for Ae. aegypti (n=13) (Grant et al. 1995).

102  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

compared to Culicoides suggesting that the mosquito would be able to detect a smaller change
in CO2 concentration. Although the threshold and slope of the concentration functions were
different, they do overlap with each other near ambient levels (350 ppm CO2).

The significance of a lower threshold of sensitivity to CO2 in Culicoides is unclear. Ambient

atmospheric levels of CO2 are currently about 350 ppm. So, it seems unlikely that having the ability
to detect lower CO2 concentrations than ambient would enhance the insect’s host-seeking ability.
Many nematoceran insects are autogenous and utilise sugar feeding for survival (Stewart and
Kline 1999). The environment contains many sources and sinks for CO2. Actively photosynthesising
plants may produce local reductions in the concentration of CO2. Thus, it is possible that these
insects are using their enhanced CO2 sensitivity to find CO2 levels below ambient where plants
that contain nectar for sugar feeding are located. Culicoides mississippiensis is known to sugar feed
from the plant Yaupon Holly, Ilex vornitoria Aiton, during the day and possibly during the night
(Stewart and Kline 1999). Therefore, it is possible that low concentrations of CO2 in conjunction
with other volatiles may serve as a plant attractant for sugar feeding. In a similar manner, higher
concentrations of CO2 may serve as a host-seeking attractant for blood feeding.

Sensitivity to low concentrations of CO2 also may have no behavioural significance. It has been
suggested (Stange 1997, Stange and Wong 1993) that certain species of moths, which also have
CO2-sensitive neurons, actually are adapted to lower, pre-industrialisation levels of CO2. Although
our methods differed from those of Stange in characterising CO2 sensitivity, perhaps the sensitivity
to lower concentrations of CO2 in Culicoides simply reflects an earlier sensitivity, which has not yet
been genetically adapted to fit the dramatic increase caused by human-based emissions of CO2.

Diapause and CO2 detection

Culex pipiens is a vector of several medically important viruses in the United States including
West Nile. Under environmental conditions of short day length and low ambient temperatures
during late larval and early pupal stages, the mosquitoes are induced to enter diapause (Spielman
and Wong 1973). Diapausing adults demonstrate a slower rate of development of primary
ovarian follicles. In addition, host-seeking and feeding behaviours are reduced. Because of these
differences we sought to determine if the CO2-sensitivity of palp receptor neurons was affected by
diapausing status. To test this, we recorded responses from the CO2-sensitive peripheral sensory
neurons in the maxillary palps of adult female mosquitoes to graded stimulation with CO2 in
diapausing and non-diapausing populations. Non-diapausing Cx. pipiens were reared from pools
of egg rafts to adults in environmental chambers at 18 °C, 15:9 h light-dark cycle, and 75% RH.
To induce diapause, alternate groups selected from the same raft pools were reared with short
days 7:17 h light-dark cycle. To confirm the effectiveness of our rearing protocols, primary ovarian
follicle and germarium lengths were measured in a portion of each cohort to determine diapause
status, and the stage of ovarian development was determined according to Christopher (1911).

Our electrophysiological data (Figure 9) indicate that the diapause state does not appear to affect
the sensitivity of these sensory neurons. Diapausing and non-diapausing female Cx. pipiens have
similar electrophysiological responses to stimulation with 0, 150, 300, 600, and 1000 ppm CO2.
These finding suggest that the reduction in host-seeking behaviour observed in diapausing Cx.
pipiens is not determined by changes in the CO2 sensitivity of peripheral sensory neurons. We
note that these data suggest that the changes in behaviour that occur with diapause must arise
because of factors that affect some other stage of central nervous system processing (see Chapter
4, this volume).

Olfaction in vector-host interactions 103

Alan J. Grant and Robert J. O’Connell

250

Impulses/2 seconds 200

150

100

50

0
0 150 300 600 1000
CO2 (ppm)

Figure 9. Mean + SE concentration-response from CO2-sensitive neurons on female Cx. pipiens that were reared
under temperature and light conditions to produce non-diapausing adults (light grey; n=35) and diapausing
adults (dark grey; n=26). Those adults reared under diapausing condition were not statistically different from
those reared under non-diapausing conditions at any concentrations tested.

CO2 responses from receptor neurons in autogenous mosquitoes

Culex pipiens forms a complex of mosquitoes whose taxonomic status is still unclear. Culex pipiens
was first described by Linneaus in 1758 and Cx. molestus by Forskal in 1775 (Vinogradova 2000).
Morphologically these two mosquitoes are very similar. Physical differences such as comparative
measures of maxillary palp to proboscis length in males have been suggested as a potential
diagnostic trait. However, such lengths appear quite variable from population to population,
greatly reducing their predictive value. In our hands, this metric discriminated potential Cx.
molestus Forskal from Cx. pipiens with approximately 85% accuracy. Although considered by some
a distinct species or subspecies, many are inclined to consider Cx. molestus as an ‘intraspecific
category, i.e. a variety, eco-physiological race or ecotype’ (Vinogradova 2000). Whatever the
taxonomic classification, Cx. molestus are autogenous and are able to produce eggs without
taking a blood meal, whereas Cx. pipiens are anautogenous and require a blood meal for egg
production. Both Cx. pipiens and Cx. molestus produce an egg raft, however, the size of the rafts
differ considerably. Culex molestus (sugar-fed) rafts contained 78.9±7.0 (SE, n=13) eggs, whereas
an average blood-fed Cx. pipiens raft contained 190.7±7.7 (SE, n=3) eggs.

Since the life cycle of the autogenous Cx. molestus does not require blood feeding, we were
interested in characterising the CO2-sensitive receptor neurons in their palps and comparing their
sensitivity with those found in Cx. pipiens. The females of both groups of mosquitoes possess
maxillary palps with similar basiconic sensilla. Although no detailed study of fine structure of
female maxillary palps has been done, Cx. molestus and Cx. pipiens palps and sensilla appear

104  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

very similar. However, recordings from the CO2-sensitive neurons revealed some interesting
differences in their overall sensitivity to CO2 stimulation. Culex molestus are significantly less
sensitive as compared to Cx. pipiens (Figure 10). Extrapolating from the dose-response curves;
both Cx. molestus and Cx. pipiens have similar thresholds of slightly over 100 ppm CO2; however
the slopes of the concentration response functions are different. These results are interesting
because as was the case with the difference between young and older Ae. aegypti and Cx. pipiens,
the group that shows reduced or no host-seeking behaviours possessed sensory neurons less
sensitive to CO2. Culex molestus can be induced to blood feed and we are currently conducting
experiments to determine if the nutrient source (sugar or blood) affects the receptor neurons
of the resulting offspring. Preliminary data from Cx. molestus suggest that aspects of the CO2
sensitivity is dependent on diet.

250

200
Impulses/2 seconds

150

100

50

0
0 150 300 600 1000
CO2 (ppm)

Figure 10. Mean + SE concentration-response relationships for female Cx. p. pipiens (light grey; n=15) and Cx.
p. molestus (dark grey; n=20). All recording were from insects 5-8 days post emergence. Values were statistically
different at 150, 300, 600 and 1000 ppm CO2 (t-test, two tailed; P<0.01).

Potential effects of changes in atmospheric levels of CO2

We are becoming increasingly convinced that our environment is changing (Gore 2006). Pressures
associated with post-industrial development and methods of environmental management have
led to large changes in the levels of chemicals in our environment. One of these chemicals is CO2,
the ambient level of which has been increasing for several centuries. Pre-industrial revolution
environmental levels of CO2 were estimated at or below 200 ppm. Current ambient atmospheric
levels range between 330 and 350 ppm. Predictions of future levels have been as high as 500
ppm. Increasing levels of environmental CO2 as well as other greenhouse gasses are believed
by some to contribute to an elevation in global temperatures. There are many consequences of
global warming including sea level elevation and wide-spread regional climate change. Some
have speculated that such environmental changes in temperature could lead to a shifting of

Olfaction in vector-host interactions 105

Alan J. Grant and Robert J. O’Connell

habitat range for many species including mosquitoes (Eisen et al. 2008). Such a change in range
could potentially affect the transmission of vector-borne diseases. Diseases that are prevalent in
tropical climates might become more prevalent in temperate areas with significantly more public
health impact.

Given that CO2 is one of the primary activators-attractants for biting insects and that many biting
insects possess highly specialised sensory neurons to detect CO2, we suggest that alterations in
CO2 levels alone could also have effects on host-seeking behaviour. We know that the threshold
of response of these sensory neurons lies around 100-120 ppm CO2. Based on the response
characteristics of these neurons, we question how potentially increasing environmental levels
of CO2 could affect the ability of the sensory neurons to process CO2 changes and consequently
how they will modulate orientation behaviour. We know that when these neurons are challenged
with graded stimulation with increasing doses of CO2, the background CO2 concentration is
important. In other words, a stimulus pulse of 1000 ppm CO2 in a background of 150 ppm CO2
elicits a significantly different response compared to a similar stimulus pulse of 1000 ppm CO2
from a background of 600 ppm CO2 (Figure 5). The signal received by the CNS is different. At
present we do not know how this change in signal will be interpreted by the mosquito CNS and
what behavioural consequences may result. In addition, little is understood about the effects of
changes in long-term stimulation of these sensory neurons. The CO2-sensitive system seems to
be rather plastic and it is likely that the overall consequences for behaviour will not be obvious.
However, it is clear now that a more thorough understanding of the system would be useful.

Molecular studies of CO2 detection

With the advent of molecular methods of study there have been many exciting advances in our
understanding of the cellular mechanism underlying olfactory detection in insects. Early and
continuing studies have been conducted with the model systems Drosophila (Dobritsa et al.
2003, Hallem et al. 2006, Jones et al. 2007, Rutzler and Zwiebel 2005). With the sequencing of the
Anopheles genome in 2002 (Holt et al. 2002) much work has focused on An. gambiae (Goldman
et al. 2005, Hallem et al. 2004, Hill et al. 2002, Kwon et al. 2007, Lu et al. 2007). Recent published
work has begun to unravel and identify the molecular nature of the receptor proteins involved
in CO2 detection. Vosshall and her colleagues (Jones et al. 2007) report that receptor neurons in
Drosophila co-express two receptor proteins necessary for CO2 detection. These proteins Gr21and
Gr63a have analogues in mosquitoes, GPRGR22 and GPRGR24. Both of these are expressed in
olfactory neurons found in maxillary palp sensilla of An. gambiae (Jones et al. 2007). This work
also indicates that the expression of these two proteins in CO2-insensitive cells confers sensitivity
and the removal of Gr63a causes mutant flies to lose this sensitivity both at a behavioural and
sensory level. Work from other labs has implicated a third receptor (AgGR23) whose addition
clearly augments CO2 responsiveness (Lu et al. 2007). The roles of each of these proteins remains
to be determined in the presumed cascade of events that leads to the detection and processing
of this important environmental cue. It is tempting to speculate that these proteins will have
differential effects on the sensitivity and overall rate of response to CO2.

Work by Ditzen et al. (2008) has utilised the molecular understanding of receptor protein systems
to understand the mode of action of the repellent DEET (N,N-diethyl-meta-toluamide). They report
that DEET strongly inhibits the 1-octen-3-ol response of An. gambiae and Drosophilia melanogaster
Meigen by mediating a response in the OR83b receptor protein and the mosquito analogue
GPROR7 receptor complex. However, using electrophysiological techniques the detection and
sensitivity of the CO2 receptor neuron appears to be unaffected. These responses confirm that the

106  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

addition of DEET strongly inhibited the octenol response but did not affect the CO2 sensitivity. In
a similar fashion we also observed that DEET did not modulate the CO2 response in An. gambiae
Giles (Figure 11). More recent work in Cx. quinquefasciatus has suggested an alternative mechanism
for DEET action on the octenol response. Syed and Leal (2008) have proposed that DEET acts
through another channel by directly stimulating receptor neurons found in a specific antennal
trichoid sensillum (Pickett et al. 2008, Syed and Leal 2008). Molecular understanding of receptor
mechanisms will lead to further elucidation of the mode of action of attractants and repellents.
Such understanding should enhance the development of semiochemicals and ultimately lead to
the development of novel control strategies.

A B 0 ppm

50,0

40,0
600 ppm

30,0
Impulses/ 2 Seconds

20,0

10,0
600 ppm
+100 μg DEET
0,0
0 PPM CO2

600 PPM CO2

600 PPM CO2 +

100ug DEET

Figure 11. Electrophysiological responses from basiconic sensilla on the maxillary palps of female An. gambiae.
(A) Mean + SE responses (n=3) to 2 s stimulations of 0 ppm CO2, 600 ppm CO2 and 600 ppm CO2+100 µg of
DEET. Preparations were held in a background environment of 0 ppm CO2 (these data were presented at the 1998
Association for Chemoreception Sciences meetings).

The use of CO2 in insect trapping

Carbon dioxide has long been used as an attractant for female mosquitoes and other biting
insects in trapping technology. For most, but not all species of biting insects, CO2 serves to
selectively attract female biting insects in a concentration dependent manner (Figure 12A-B). For
several species, a significant but smaller number of males are also attracted. Reports in the 1960s

Olfaction in vector-host interactions 107

Alan J. Grant and Robert J. O’Connell

A 2,200

Mean number of culicidae/night

1,800

1,400

1000

600

200

100 250 500 1000

-200
CO2 (ppm)
B 300
Mean culicoides (mg)/night

250

200

150

100

50

0
100 250 500 1000
CO2 (ppm)

Figure 12. Mean + SE concentration-response relationship of trap catch of (A) mosquitoes and (B) biting midges as
a function of CO2 release in field experiments conducted in Cedar Key, FL, USA. The trap was a passive counter-flow
trap whose CO2 source was a compressed tank. Release concentration amounts were determined by restrictive line
orifices and regulated pressure. Each point represents eight nights of trapping.

demonstrated the utility of adding CO2 to surveillance traps (Carestia and Savage 1967, Newhouse
et al. 1966, Vickery, Jr. et al. 1966, Wright 1967). Often CO2 was added in the form of dry ice that
sublimates to gas. In CDC style traps, blocks of dry ice are added above the traps in an insulated
chamber with a port releasing CO2 into the trap stream. In some instances, a bag of dry ice could
be simply hung in the vicinity of the trap attracting insects to the general area. Several pounds
of dry ice are often sufficient to generate CO2 for 12-14 hours. Alternately, regulated cylinders

108  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

containing compressed CO2 could be used to release the gas at a standard rate (often 500 mls/
minute). This rate could be adjusted to allow a standard 20 pound tank to last for days to weeks.
This works well for surveillance system where trapping periods are relatively short. To further
enhance catch, other attractants such as 1-octen-3-ol, lactic acid or ammonia are added. These
supplemental attractants are often used to ‘tailor’ catches to specific problem species, such as
anthropophilic day-biters.

In the 1990s, newer technology for commercial traps appeared on the market. They were designed
to operate continuously for longer periods of time. These traps generated CO2 by catalytically
combusting propane on a bed of coated beads. Heat generated from this burn process is converted
to electricity, which is used to power fans that dispersed the CO2 attractant plume and vacuumed
approaching insects into nets. An added advantage to this technology was the co-generation of
heat and moisture, two addition attractants for many biting insects (Kline 2002). The advantages
of such traps are several-fold. First, the trap is able to generate attractants in the form of CO2, heat
and moisture and collect insects continuously for over one month on a standard 20-lb ‘barbeque
grill’ sized propane tank. Second, by generating its own power, the trap was not reliant on battery
or line voltage. Consequently this allowed traps to be placed at some distance from living areas.
The rational is that proper placement of the trap at some distance from commonly used areas can
result in biting insects being drawn away from areas where they would most likely encounter a
host for a blood meal. Third, these traps tended to be very specific usually attracting few non-
target insects.

The release of CO2 from traps can be modulated in several ways to tune attractiveness. First, the
overall concentration released can be modified. Generally larger amounts of CO2 will result in
larger catches. Second, the strategy of release can affect catch. For example, counter flow release
systems, where the plume dispensing the CO2 stream is countercurrent to the aspiration stream,
enhances species diversity and numbers in the catch. Third, the temporal nature of the CO2 release
can affect catch rate. Pulsing the attractant can be used to release discrete amounts of CO2 at
specific intervals or times of day. This aspect of release is complicated by the fact that even a
continuous release of CO2 is easily and quickly broken into pockets by a turbulent stream of wind.
The importance of such discontinuous signals has been studied and discussed for field (Murlis
and Jones 1981); bioassay (Dekker et al. 2001) and sensory (Grant et al. 1997) systems. In pulsing
systems, several variables can be modulated including: concentration of CO2 release, duration of
release and interval between pulses. All these must be evaluated to determine optimal release
conditions. Clearly traps can serve as an important tool to monitor population fluctuations and to
assay for potential health risks by collecting and testing a variety of hematophagous insects for
potentially dangerous viruses and other pathogens.

Conclusion

In this brief review we summarised our current understanding of the properties and role of CO2
sensitive chemosensory systems in modulating host-seeking behaviours in mosquitoes. We see
that these orientation behaviours are complicated and potentially plastic. They represent a dynamic
system whose sensitivity may be modulated by environmental factors. Under what conditions
does the system’s sensitivity change? We have shown here that the age of the insect as well as
the blood feeding tendencies affect CO2 sensitivity, but there are many other developmental and
environmental conditions that may also modulate the response. For example, does the infected
state of an individual mosquito change peripheral sensitivity?

Olfaction in vector-host interactions 109

Alan J. Grant and Robert J. O’Connell

There are many steps in the chain of events that lead from the detection of an environmental
cue to a change in animal behaviour. Each provides a potential point of attack where treatment
may alter the behavioural outcome. A complete understanding of this chain of events in disease
vector organisms can provide new approaches to mechanisms that may be effective in decreasing
disease transmission. We suggest that a failure to appreciate this fact supports another cautionary
tale with widespread impact.

References

Acree F, Jr., Turner RB, Gouck HK, Beroza M and Smith N (1968) L-lactic acid: a mosquito attractant isolated from humans.
Science 161: 1346-1347.
Altner H and Loftus R (1985) Ultrastructure and function of insect thermo- and hygroreceptors. Annu Rev Entomol 30:
273-295.
Blaney WM, Schoonhoven LM and Simmonds MSJ (1986) Sensitivity variations in insect chemoreceptors; a review.
Experientia 42: 13-19.
Blanton FS and Wirth WW (1979) The sand flies (Culicoides) of Florida (Diptera: Ceratopogonidae): Florida Department
of Agriculture and Consumer Services, USA.
Bogner F (1990) Sensory physiological investigation of carbon dioxide receptors in Lepidoptera. J Insect Physiol 36:
951-957.
Bogner F (1992) Response properties of CO2-sensitive receptors in tsetse flies (Diptera: Glossina palpalis. Physiol
Entomol 17: 19-24.
Bogner F, Boppré M, Ernst K-D and Boeckh J (1986) CO2 sensitive receptors on labial palps of Rhodogastria moths
(Lepidoptera: Arctiidae): physiology, fine structure and central projections. J Comp Physiol [A] 158: 741-749.
Borroni PF and O’Connell RJ (1992) Temporal analysis of adaptation in moth (Trichoplusia ni) pheromone receptor
neurons. J Comp Physiol [A] 170: 691-700.
Bowen MF (1991) The sensory physiology of host-seeking behavior in mosquitoes. Annu Rev Entomol 36: 139-158.
Braverman Y and Hulley PE (1979) The relationship between the numbers and distribution of some antennal and palpal
sense organs and host preference in some Culicoides (Diptera: Ceratopogonidae) from southern Africa. J Med
Entomol 15: 419-424.
Breman JG (2001) The ears of the hippopotamus: manifestations, determinants, and estimates of the malaria burden.
Am J Trop Med Hyg 64 (1-2 Suppl): 1-11.
Carestia RR and Savage LB (1967) Effectiveness of carbon dioxide as a mosquito attractant in the CDC minature light
trap. Mosquito News 27: 90-92.
Christopher SR (1911) The development of the egg follicle in anophelines. Paludism 2: 73-88.
Christopher SSR (1960) Aedes aegypti (L), the yellow fever mosquito: its life history, bionomics and structue. Cambridge
University Press, Cambridge, UK.
Clements AN (1963) The physiology of mosquitoes. Pergamon Press, Oxford, UK.
Clements AN (1999) The biology of mosquitoes. Volume 2 Sensory reception and behaviour. CABI Publishing,
Wallingford, UK.
Crnjar R, Yin C-M, Stoffolano jr. JG, Barbarossa IT, Liscia A and Angioy AM (1990) Influence of age on the electroantennogram
response of the female blowfly (Phormia regina) (Diptera: Calliphoridae). J Insect Physiol 36: 917-921.
Crumb SE (1922) A mosquito attractant. Science 55(1426): 446-447.
Darian-Smith I, Johnson KO and Dykes R (1973) ‘Cold’ fiber population innervating palmar and digital skin of the
monkey: responses to cooling pulses. J Neurophysiol 36: 325-346.
Davis EE (1977) Response of the antennal receptors of the male Aedes aegypti mosquito. J Insect Physiol 23 613-617.
Davis EE (1984a) Development of Lactic acid-receptor sensitivity and host-seeking behavior in newly emerged female
Aedes aegypti mosquitoes. J Insect Physiol 30: 211-215.
Davis EE (1984b) Regulation of sensitivity in the peripheral chemoreceptor systems for host-seeking behaviour by a
haemolymph-borne factor in Aedes aegypti. J Insect Physiol 30: 179-183.

110  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

Davis EE (1985) Insect repellents: concepts of their mode of action relative to potential sensory mechanisms in
mosquitoes (Diptera: Culicidae). J Med Entomol 22: 237-243.
Davis EE (1996) Olfactory control of mosquito behaviour. In: Bock GR and Cardew G (eds) Olfaction in mosquito-host
interactions. John Wiley & Sons, Chichester, UK, pp 48-53.
Davis EE and Bowen MF (1994) Sensory physiological basis for attraction in mosquitoes. J Mosquito Control Assoc 10:
316-325.
Davis EE and Sokolove PG (1976) Lactic acid-sensitive receptors on the antennae of the mosquito, Aedes aegypti. J
Comp Physiol [A] 105: 43-54.
Daykin PN, Kellogg FE and Wright RH (1965) Host-finding and repulsion of Aedes aegypti. Can Ento 97: 239-263.
Dekker T, Takken W and Carde RT (2001) Structure of host-odour plumes influences catch of Anopheles gambiae s.s. and
Aedes aegypti in a dual-choice olfactometer. Physio Entomology 26: 124-134.
Dionne VE (1992) Chemosensory responses in isolated olfactory receptor neurons from Necturus maculosus. J Gen.
Physiol 99: 415-433.
Ditzen M, Pellegrino M and Vosshall LB (2008) Insect odorant receptors are molecular targets of the insect repellent
DEET. Science 319(5871): 1838-1842.
Dobritsa AA, Van der Goes van Naters, Warr CG, Steinbrecht RA and Carlson JR (2003) Integrating the molecular and
cellular basis of odor coding in the Drosophila antenna. Neuron 37: 827-841.
Eisen L, Bolling BG, Blair CD, Beaty BJ and Moore CG (2008) Mosquito species richness, composition, and abundance
along habitat-climate-elevation gradients in the northern Colorado Front Range. J Med Entomol 45: 800-811.
Elkinton JS, Cardé RT and Mason CJ (1984) Evaluation of time-average dispersion models for estimating pheromone
concentration in a deciduous forest. J Chem Ecol 10: 1081-1108.
Friend WG and Smith JJB (1977) Factors affecting feeding by bloodsucking insects. Ann Rev Ento 22: 309-331.
Galbreath RA and Galbreath NH (1977) A simple method for making fine dissecting needles. Ent News 88: 143-1447.
Galun R (1977) Responses of blood-sucking arthropods to vertebrate hosts. In: Shorey HH and McKelvey JrJJ (eds)
Chemical control of insect behavior theory and application. John Wiley & Sons, New York, USA, pp 103-115.
Gillies MT (1980) The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review. Bull Ent Res
70: 525-532.
Goldman AL, Van der Goes van Naters, Lessing D, Warr CG and Carlson JR (2005) Coexpression of two functional odor
receptors in one neuron. Neuron 45: 661-666.
Gore A (2006) An inconvenient truth: the planetary emergency of global warming and what we can do about it. Rodale
Books, Emmaus, PA, USA. 328 p.
Grant AJ and O’Connell RJ (1996a) Electrophysiological responses from receptor neurons in mosquito maxillary palp
sensilla. In: Bock GR and Cardew G (eds) Olfaction in mosquito-host interactions. John Wiley & Sons, Chichester,
UK, pp 233-248.
Grant AJ and O’Connell RJ (2007) Age-related changes in female mosquito carbon dioxide detection. J Med Entomol
44: 617-623.
Grant AJ, Borroni PF and O’Connell RJ (1997) Pused pheromone stimuli affect the temporal response of antennal
receptor neurones of the adult cabbage looper moth. Physiol Entomol 22: 123-130.
Grant AJ and Kline DL (2003) Electrophysiological responses from Culicoides (Diptera: Ceratopogonidae) to stimulation
with carbon dioxide. J Med Entomol 40: 284-293.
Grant AJ and O’Connell RJ (1986) Neurophysiological and morphological investigations of pheromone-sensitive sensilla
on the antenna of male Trichoplusia ni. J Insect Physiol 32: 503-515.
Grant AJ and O’Connell RJ (1996b) Electrophysiological responses from receptor neurons in mosquito maxillary palp
sensilla. Wiley, Chichester, UK.
Grant AJ, Wigton BE, Aghajanian JG and O’Connell RJ (1995) Electrophysiological responses of receptor neurons in
mosquito maxillary palp sensilla to carbon dioxide. J Comp Physiol [A] 177: 389-396.
Hallem EA, Dahanukar A and Carlson JR (2006) Insect odor and taste receptors. Annu Rev Entomol 51: 113-135.
Hallem EA, Nicole FA, Zwiebel LJ and Carlson JR (2004) Olfaction: mosquito receptor for human-sweat odorant. Nature
427(6971): 212-213.

Olfaction in vector-host interactions 111

Alan J. Grant and Robert J. O’Connell

Hartberg WK (1971) Observations on the mating behaviour of Aedes aegypti in nature. Bull WHO 45: 847-850.
Hill CA, Fox AN, Pitts RJ, Kent LB, Tan PL, Chrystal MA, Cravchik A, Collins FH, Robertson HM and Zwiebel LJ (2002) G
protein-coupled receptors in Anopheles gambiae. Science 298(5591): 176-178.
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides
R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P,
Atkinson PW, Baden H, de B, V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M, Cai S, Center
A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A, Dew I, Evans
CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun SL, Hogan JR,
Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin JJ, Lobo NF, Lopez
JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O’Brochta DA, Pfannkoch C, Qi R,
Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J, Thomasova D, Ton
LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman JR, Wu M, Yao A,
Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della TA, Roth CW, Louis C, Kalush F, Mural RJ,
Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey PT, Venter JC, Weissenbach
J, Kafatos FC, Collins FH and Hoffman SL (2002) The genome sequence of the malaria mosquito Anopheles gambiae.
Science 298(5591): 129-149.
Hubel DH (1957) Tungsten microelectrodes for recording from single units. Science 125(549): 550.
Jones JC and Madhukar BV (1976) Effects of antennectomy on blood-feeding behavior of Aedes aegypti mosquitoes.
Ent Exp & Appl 19: 19-22.
Jones WD, Cayirlioglu P, Kadow IG and Vosshall LB (2007) Two chemosensory receptors together mediate carbon dioxide
detection in Drosophila. Nature 445(7123): 86-90.
Kellogg FE (1970) Water vapour and carbon dioxide receptors in Aedes aegypti. J Insect Physiol 16: 99-108.
Kettle DS (1965) Biting ceratopogonids as vectors of human and animal diseases. Acta Tropica 22: 356-362.
Kline DL (2002) Evaluation of various models of propane-powered mosquito traps. J Vector Ecology 27: 1-7.
Kline DL, Dame DA and Meisch MV (1991) Evaluation of 1-octen-3-ol and carbon dioxide as attractants for mosquitoes
associated with irrigation rice fields in Arkansas. J Am Mosquito Control Ass 7: 165-169.
Kline DL, Takken W, Wood JR and Carlson DA (1990) Field studies on the potential of butanone, carbon dioxide, honey
extract, 1-octen-3-ol, L-Lactic acid and phenols as attractant for mosquitoes. Med Vet Ent 4: 383-391.
Kwon JY, Dahanukar A, Weiss LA and Carlson JR (2007) The molecular basis of CO2 reception in Drosophila. PNAS 104:
3574-3578.
Lacher V (1964) Elektrophysiologische untersuchungen an einzelnen rezeptoren für geruch, kohlendioxyd,
luftfeuchtigkeit und temperatur auf den antennn der arbeitsbiene und der drohne (Apis mellifera L.). Zeitschrift
für vergleichende Physiologie 48(587): 623.
Linley JR, Hoch AL and Pinheiro FP (1983) Biting midges (Diptera: Ceratopogonidae) and human health. J Med Entomol
20: 347-364.
Linley JR and Davies JB (1971) Sandflies and tourism in Florida and the Bahamas and Caribbean Area. J Econ Ento 6410:
264-278.
Lu T, Qiu YT, Wang G, Kwon JY, Rutzler M, Kwon HW, Pitts RJ, van Loon JJ, Takken W, Carlson JR and Zwiebel LJ (2007)
Odor coding in the maxillary palp of the malaria vector mosquito Anopheles gambiae. Curr Biol 17: 1533-1544.
Matheson R (1945) Guide to the insects of connecticut part VI. the Diptera or true flies of connecticut. State of
Connecticut State Geological and Natural History Survey, USA.
Mayer MS and James JD (1969) Attraction of Aedes aegypti (L.): Responses to human arms, carbon dioxide, and air
currents in a new type of olfactometer. Bull Ent Res 58: 629-642.
Mayor S (2008) WHO report shows progress in efforts to reduce malaria incidence. BMJ 337: a1678.
McIver SB (1972a) Fine structure of pegs on the palps of female culicine mosquitoes. Can J Zoo 50: 571-576.
McIver SB (1972b) Fine structure of the sensilla chaetica on the antennae of Aedes aegypti (Diptera: Culicidae). Ann
Entomol Soc Amer 65: 1390-1397.
McIver SB (1980) Sensory aspects of mate-finding behavior in male mosquitoes (Diptera: Culicidae). J Med Entomol
17: 54-57.

112  Olfaction in vector-host interactions

 5. The detection of carbon dioxide and its role in the orientation to hosts by haematophagous insects

McIver SB (1982) Sensilla of mosquitoes (Diptera: Culicidae). J Med Entomol 19: 489-535.
McKeever S and Pound JM (1979) The mouthparts of female Corethrella brakeleyi and C. wirthi (Diptera: Chaoboridae).
J.Morph. 161: 157-168.
Murlis J and Jones CD (1981) Fine-scale structure of odour plumes in relation to insect orientation to distant pheromone
and other attractant sources. Physiol Entomol 6: 71-86.
Newhouse VF, Chamberlain RW, Johnston JG and Sudia WD (1966) Use of dry ice to increase mosquito catches of the
CDC miniature light trap. Mosquito News 26: 30-35.
O’Connell RJ and Mozell MM (1969) Quantitative stimulation of frog olfactory receptors. J Neurophysiol 32 51-63.
O’Connell RJ (1972) Responses of olfactory receptors to the sex attractant, its synergist and inhibitor in the red-banded
leaf roller, Argyrotaenia velutinana. In: Schneider D (ed) Olfaction and taste IV. Wissenschaftliche Verlagsgesellschaft
MBH, Stuttgart, Germany, pp 180-186.
O’Connell RJ, Kocsis WA and Schoenfeld RL (1973) Minicomputer identification and timing of nerve impulses mixed in
a single recording channel. Proc IEEE 61: 1615-1621.
Omer SM and Gillies MT (1971) Loss of response to carbon dioxide in palpectomized female mosquitoes. Ent Exp &
Appl 14: 251-252.
Pickett JA, Birkett MA and Logan JG (2008) DEET repels ORNery mosquitoes. PNAS 105: 13195-13196.
Roth LM (1951) Loci of sensory end-organs used by mosquitoes (Aedes aegypti (L.) and Anopheles quadrimaculatus Say)
in receiving host stimuli. Ann Entomol Soc Amer 44: 59-74.
Rudolfs W (1922) Chemotropism of mosquitoes. New Jersey Agricultural Experimienrt Stations Bulletin 367: 5-23.
Rutzler M and Zwiebel LJ (2005) Molecular biology of insect olfaction: recent progress and conceptual models. J Comp
Physiol [A] 191: 777-790.
Schweitzer ES, Sanes JR and Hildebrand JG (1976) Ontogeny of electroantennogram responses in the moth, Manduca
sexta. J Insect Physiol 22: 955-960.
Seaton DR and Lumsden WHR (1941) Observations on the effects of age, fertilization and light on biting by Aëdes
aegypti (L.) in a controlled microclimate. AnnTrop Med Parasit 35: 23-36.
Smith K, Thompson G and Koster HD (1969) Sweat in schizophrenic patients: identification of the odorous substance.
Science 166(3903): 398-399.
Spielman A and Wong J (1973) Environmental control of ovarian diapause in Culex pipiens. Ann Entomol Soc Amer 66:
905-907.
Stange G (1997) Effects of changes in atmospheric carbon dioxide on the location of hosts by the moth Cactoblastis
cactorum. Oecologia 110: 539-545.
Stange G and Diesendorf M (1973) The response of the honeybee antennal CO2-receptors to N20 and Xe. J Comp Physiol
86: 139-158.
Stange G and Wong C (1993) Moth response to climate. Nature 365 699.
Steullet P and Guerin PM (1992) Perception of breath components by the tropical bont tick, Amblyomma variegatum
Fabricus (Ixodidae) I. CO2-excited and CO2-inhibited receptors. J Comp Physiol [A] 170: 665-676.
Stewart RG and Kline DL (1999) Sugar feeding by Culicoides mississippiensis (Diptera: Ceratopogonidae) on the Yaupon
Holly, Ilex vomitoria. J Med Entomol 36: 268-271.
Syed Z and Leal WS (2008) Mosquitoes smell and avoid the insect repellent DEET. PNAS 105: 13598-13603.
Takken W and Kline DL (1989) Carbon dioxide and 1-octen-3-ol as mosquito attractants. J Am Mosquito Control Assoc
5: 311-316.
Takken W (1991) The role of olfaction in host-seeking of mosquitoes: a review. Insect Sci Applic 12: 287-295.
Vickery CA, Jr., Meadows KE and Baughman IE (1966) Synergism of carbon dioxide and chick as bait for Culex nigripalpus.
Mosquito News 26: 507-508.
Vinogradova EB (2000) Culex pipiens pipiens mosquitoes: taxonomy, distribution, ecology, hysiology, genetics, applied
importance and control. Pensoft Publishers, Sofia, Bulgaria.
Wright RH (1967) A mosquito trap. J Econ Ento 60: 876-877.
Zwiebel LJ and Takken W (2004) Olfactory regulation of mosquito-host interactions. Insect Biochem Mol Biol 34: 645-
652.

Olfaction in vector-host interactions 113

6. Host finding by female mosquitoes: mechanisms of orientation to
host odours and other cues1

Ring T. Cardé and Gabriella Gibson

Abstract

The mechanisms by which mosquitoes locate their blood hosts are difficult to observe and analyse
under natural conditions, given the small size of these insects, the crepuscular/nocturnal activity
rhythms of most medically important species, and the distances of less than ten to hundreds of
metres travelled to obtain a blood meal. Aspects of these mechanisms can be characterised from
observations of the responses of mosquitoes in the laboratory and field to host cues at each stage
in the process of host location. The main stages are typically described as activation, orientation
to wind direction and odour-plume following, and arrestment near the source. Volatile chemicals
emitted by the host extend the range over which it can be detected much further than any other
cue, such as visual features or body temperature. Once a mosquito detects airborne odour cues,
the evidence indicates that it relies primarily on an interaction between olfactory and visual inputs
to move upwind toward the odour source by optomotor-guided anemotaxis, utilising contact with
the host odour plume that is carried by the airstream to update its heading with respect to wind
direction by visual cues, until it arrives near the odour source, when upwind progress is arrested
and its responses to other host stimuli, such as heat, moisture and visual features of the host are
thought to bring the mosquito into contact with the host. Knowledge of the sensory systems and
physiological factors that control behaviour provide clues as to how mosquitoes locate hosts. For
example, (1) the daily timing of quiescence and activity is mainly controlled by an endogenous
circadian clock that determines the times of day, and hence the environmental conditions, such as
light intensities, at which mosquitoes are most active, (2) mosquito eyes are highly specialised to
detect high contrast features at low light levels at the expense of resolution, and, therefore, they
can probably see hosts only <5-10 metres away, whereas they can be attracted by host odours
originating tens of metres upwind, (3) the concentration and spatial structure of the plume of
semiochemicals affect the type of response elicited in mosquitoes, and (4) responsiveness to
particular odours is influenced by various hormones released on adult emergence, as a result of
mating and throughout the gonotrophic cycle.

Keywords: carbon dioxide, odour plume, kairomone, upwind anemotaxis, klinokinesis, visual
cues, odour salience

Introduction

A female mosquito’s successful location of a host for feeding is under considerable selective
pressure for reliability, energetic efficiency, and stealth to avoid predation, as only through
securing a blood meal can the females of most species reproduce. When the host is beyond visual
range, orientation manoeuvres modulating host finding are mainly related to contacting a plume
of host-emitted odour and following it upwind, although other kinetic strategies are possible,
such as ‘aim and shoot,’ whereby mosquitoes at rest on a surface detect the direction of air currents
by mechanoreception and take-off upwind when stimulated by host odours. Plume following
strategies ought to be sufficiently flexible to enable a mosquito to reach a prospective host under

1 Both authors contributed equally to this review.

Olfaction in vector-host interactions 115

Ring T. Cardé and Gabriella Gibson

Box 1. Glossary of terms used in this chapter, emphasising their use in

orientation to odours in wind.

Anautogenous Female mosquitoes that require a blood meal to produce eggs.

Anemotaxis Maintaining a set angle to wind flow; positive anemotaxis
(usually what is meant by ‘anemotaxis’) is orientation upwind.
Autogenous Female mosquitoes that produce their first ovarian cycle without
a blood meal.
Crepuscular Activity at twilight, either dawn or dusk.
Klinokinesis Orientation using successively sensed changes in stimulus
concentration.
Longitudinal klinokinesis Orientation using successively sensed changes in stimulus
concentration in the direction toward the origin of the stimulus,
e.g. changes in odour concentration sensed within the plume as
the organism moves towards its origin.
Mechanoreception Sensing displacement of air movement by mechanical deflection
of hair or body extensions, e.g. Johnston’s organ at the base of
the antennae.
Optomotor Use of visual feedback to regulate manoeuvres, e.g. sensing and
directing displacement by visual monitoring of changes of the
organism’s position in space.
Spontaneous activity When kept in constant environmental conditions (sound-proof
and light-proof chambers), mosquitoes fly spontaneously at
certain phases of the circadian day, even though there are
no external stimuli to trigger such a response. This pattern of
spontaneous activity is thought to reveal the underlying pattern
of daily cycles of various behaviours under natural conditions.
Transverse klinokinesis Orientation using successively sensed changes in stimulus
concentration lateral to the direction towards the origin of the
stimulus, e.g. orienting by using differences in concentration
of the odour plume sensed during lateral (across plume)
movement, which may include detection of the plume’s edge.

the naturally varying conditions of the field, where turbulent wind flow creates a spatially and
temporally intermittent odour plume of fluctuating directionality. The sensory inputs available to
modulate this process vary enormously with mosquito species, because of wide variance in the
breadth of acceptable hosts and the odour cues they present, in the structure of habitat and its
influence on plume dispersion, and environmental conditions, particularly those of wind speed,
wind turbulence and light levels (Gibson and Torr 1999).

This review focuses mainly on those orientation manoeuvres that are presumed to operate at
long range – that is, when the female mosquito is many metres or more away from a prospective
host and often beyond visual range. However, as mosquitoes reach the host’s immediate vicinity
other cues, such as visual aspects of the host, heat and humidity may add to the suite of sensory
inputs, and together these can influence continued orientation towards the host, hovering near
the host, landing on the host, exploration of the host’s surface, and biting. In some mosquitoes,
there appear to be two orientation steps: flight to the host habitat, followed by pre-attack ‘resting’,

116  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

and then flight to the host. For those species most closely associated with human habitats, this
resting phase has been observed most often to occur just before or after house entry (Clements
1999), a stage in host location that we know even less about than odour-plume following.

As mosquitoes approach a host, the overall plume structure, concentration of odours and nature
of visual stimuli generally change, possibly providing mosquitoes with some information as to
the proximity of the host. In the case of house-entering, apart from the effects of an aperture
(i.e. open windows or eaves) on air flow, plume structure, and visual stimuli, the olfactory inputs
available to a mosquito are likely to change as it approaches a prospective host. Indoors the
concentration of odours generally increases, and some ‘minor’ odour components may play an
increasingly greater role as these reach threshold concentrations. At close range, a mosquito
may lose contact with the plume of carbon dioxide when its flight path allows exploration of
body areas removed from exhalations. Then odours from specific regions of the host’s body
may become more important for close-range orientation and landing. Little is known, however,
about the spatial and temporal pattern of host odours in the relatively stagnant conditions of
habitations, nor about the neurophysiological effects of relatively continuous exposure to host
emanations on mosquito olfactory systems (e.g. sensory adaptation or habituation?). In a stagnant
environment it is possible that there are remnants of exhalations and odours from surfaces with
prior host contact everywhere (e.g. worn clothing, bedding material). The profile of odours
(the behaviourally relevant constituents, as well as their ratios and concentrations) emitted by
individuals of a given host species also can differ among individuals, with distinctive chemical
signatures postulated to influence individual variability in host attractiveness.

Much of what has been learned about the mechanisms of insect flight along odour plumes is based
not only on work with mosquitoes, but with other insects, particularly moths, tsetse and Drosophila.
Together these model systems have been used to establish the kinds of manoeuvres that flying
insects use to orient along plumes in wind, and, in several moth species, how orientation toward
an odour source may occur in still air (reviewed by Cardé and Willis 2008, Hardie et al. 2001). How
mosquitoes orient to a host under windless conditions remains a largely unexplored question.
Many of the laboratory and field assays that have contributed to our understanding of mosquito
orientation have been designed, not to define orientation mechanisms, but either to improve
the efficacy of odour-baited mosquito traps or to aid in the identification of the semiochemicals
mediating orientation. Notwithstanding, some of these studies have added directly to our
understanding of orientation, in part by helping define the probable role of different odours
in long-distance and close-range orientation. Some of the field-based methods used to explore
long-range responses of tsetse to host odours been applied to mosquitoes, with promising results
(Knols et al. 1998, Torr et al. 2008).

This review will emphasise what is known about plume-orientated behaviours of female mosquitoes
in host finding, identify gaps in our understanding, and suggest areas for future exploration.
We will give particular attention to how changes in plume structure may influence orientation
strategies. We also will consider how internal cues (e.g. circadian rhythms and physiological state)
and environmental cues (e.g. visual cues related to wind direction and speed and thermal cues
related to convection currents) interact with odour in influencing host finding. Important reviews
of this topic include those by Clements (1999), Gibson and Torr (1999), Sutcliffe (1987), Takken
(1991) and Takken and Knols (1999).

Olfaction in vector-host interactions 117

Ring T. Cardé and Gabriella Gibson

Activation

The circadian rhythms of spontaneous activity for several species of medically important species
have been established (Clements 1999), demonstrating that the daily timing of active and
quiescent phases in mosquitoes is controlled mainly endogenously, thereby restricting most
mosquito flight to species-specific times of day (e.g. nocturnal, diurnal, crepuscular). Accordingly,
species-specific host-location strategies have evolved in the context of a specific range of general
environmental conditions, e.g. a range of light intensities, patterns of wind dynamics, atmospheric
carbon dioxide concentrations and activity rhythms of potential hosts and predators (Gibson and
Torr 1999).

The pattern of spontaneous activity obtained under constant environmental conditions is perhaps
best thought of as an indication of the fluctuating threshold of sensitivity to external and internal
cues (Brady 1975). Both the daily timing of maximum sensitivity to stimuli and the type of stimuli
that give rise to a response change throughout a mosquito’s life, particularly for females. These
behavioural changes are generally triggered by changes in hormone levels and physiological
states brought on by significant events, such as mating, the intake of either blood or plant juices,
and oviposition, and more gradual processes such as changes in nutritional status (Clements 1999).
For example, it has been shown for several species that the patterns of circadian activity of newly
emerged uninseminated females are similar to those of conspecific males (Anopheles gambiae
Giles s.s., Jones and Gubbins 1978; An. stephensi Liston, Rowland 1989; Culex quinquefasciatus Say,
Jones and Gubbins 1979, Clements 1999), and coincides with the timing of swarming and mating
behaviour in the field. Uninseminated females and males aggregate at swarms sites, and hence
they must respond similarly to swarm-formation stimuli since they arrive at specific swarming sites
at about the same time. There is evidence, however, that males join swarms slightly in advance of
females; the onset of spontaneous activity in uninseminated female Cx. quinquefasciatus is ~10
min earlier than in males of the same age (Gibson 1983). Once mated, however, the pattern of
female circadian activity changes to coincide with the timing of blood feeding observed in the
field (Gillies 1988, Clements 1999) and females become refractory to mating (Jones and Gubbins
1979, Shutt et al. 2010). These changes in pattern and type of activity have been shown for several
Culex species to be induced by male accessory gland (MAG) proteins, which are normally passed
from male to female during mating (Chiba et al. 1992, Clements 1999, Jones and Gubbins 1979).

In contrast, Aedes aegypti (L.) females are active for the first few and last few hours of daylight,
irrespective of their insemination status, and respond to host odours even before they mate. Aedes
aegypti males are also attracted to host animals (see also Chapter 5, this volume), intercepting
females as they approach the host, hence mating and blood feeding occur during the same
circadian phase and without any obvious post-mating change in the stimuli to which females
respond (Corbet and Smith 1974), although MAG proteins cause them to become refractory to
reinsemination once mated (Craig 1967, Shutt et al. 2010).

Once females are in the active phase of their circadian rhythm, laboratory studies have shown that
they ‘take-off’ in response to activating cues, such as carbon dioxide (Eiras and Jepson 1991, Gillies
1980, Healy and Copeland 1995), or spontaneously, without any particular environmental stimulus
required. Little is known about what stimulates mosquitoes to take flight from their ‘resting sites’
in the field. In one of the few experimental observations of activation in the field, Gillies (1988)
reported that 10,000 An. gambiae s.l. released inside an experimental hut earlier in the day, left the
hut in a mass exodus within 10-20 min of sunset. In a study of tsetse behaviour in the field, Torr (1988)
concluded that that the initial activation of resting flies is primarily mediated through endogenous,

118  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

rather than host stimuli: no more than 35% of resting flies responded to a range of visual and
olfactory cues known to be highly attractive to flying tsetse. The rest gradually left the resting-site
refuge in a stochastic pattern over the course of their ‘active phase.’ Thus, both ‘spontaneous activity’
and responsiveness to particular environmental cues are likely to activate mosquitoes, causing them
to leave sheltered resting sites which brings them into contact with a wider range of more accessible
environmental stimuli, which evoke subsequent behaviour, depending on their physiological state
(i.e. stage in the gonotrophic cycle of blood feeding and egg laying) and the nature of the stimulus
(Clements 1999, Jones and Gubbins 1978, Rowland 1989).

Strategies for locating a plume of host odour

A mosquito’s first contact with host odour may arise through activation, followed by ‘ranging
flight’ – to put it teleologically – flight that occurs ‘in search of’ a host. Little is known about the
orientation of ranging flights, but there remains some debate (reviewed by Cardé and Willis 2008)
about the optimal strategy for discovering an odour plume: should a mosquito head crosswind,
upwind or downwind? Alternatively, is a random heading with respect to the current wind flow
optimal? Two key issues governing the likely heading of a particular direction of the flight path
with respect to wind flow are: the balance between energy expenditure that wind direction and the
mosquito’s airspeed dictate and the rapidity and probability of plume contact. Ranging strategies
in wind have been debated from theoretical perspectives, with the predictions on optimality
generally contingent on how frequently the wind shifts direction and an insect’s ability, if any, to
keep track of its position relative to recent changes in wind direction (Figure 1).

Insects that are strong fliers, and/or fly significantly faster than the wind speed can fly at an angle
to the wind, and still make progress against it. Tsetse typically fly at 6-8 m/s in winds of ~0.5 m/s. A
field study on tsetse in Zimbabwe showed that the orientation to wind direction when no obvious
visual or olfactory cues were present was predominantly downwind, and significantly crosswind
when they left a visual target they had been circling, again with no odour present (Gibson et al.
1991).

Smaller, weaker fliers, such as mosquitoes, that fly at ~20 cm/s in still air (Gibson 1995) are more
at the mercy of the wind; they can be blown off-course easily by a wind of 0.5 m/s. Therefore, we
might expect them to adopt an orientation to the wind that is most stable, i.e. to fly as directly
upwind as possible, correcting for drift across their visual flow-field to maintain an even flow of
images from front to back in all areas of the eye. This may be why some species of mosquitoes
fly upwind in wind tunnels, even in the evident absence of olfactory cues (Costantini et al. 2001,
Takken et al. 1997). At wind speeds of less than their preferred flight speed, however, they would
be more able to maintain a course at a fixed angle to the wind. Therefore, within the range of wind
speeds wherein flight is feasible, the ‘search’ strategy may, to some extent at least, be wind-speed
dependent.

In the case of mosquitoes, we have little field evidence to bear on this question. Gillies and
Wilkes (1974) proposed that mosquitoes tended to head downwind (in accord with an optimal
expenditure of energy model). They compared rates of mosquito capture with and without a
large (57 m long, 2.9 m high) semi-circular screen, permeable to air flow but not mosquitoes,
set 18 m away from human bait and oriented on nights of relatively uniform wind direction to
form a barrier to upwind movement. As rates of capture were comparable in the two situations,
mosquitoes were assumed to generally head downwind before plume contact. Gillies and Wilkes
(1978) later retracted this interpretation, because they determined that screen fences had an

Olfaction in vector-host interactions 119

Ring T. Cardé and Gabriella Gibson

wind

Figure 1. Possible consequences of a shifting wind direction on optimal strategies for foraging for an odour plume
(see Sabelis and Schippers 1984). When the wind shifts over 60° (α), it would appear more advantageous to head
downwind rather than crosswind, because the plume’s width (a) will exceed its downwind length (b), and then it is
more efficient energetically to head with the mean wind flow (shown by the arrow). Most important, the probability
of plume contact would be increased (Sabelis and Schippers 1984). Depending on the insect’s flight capabilities and
the wind speed, heading upwind might also be more efficient than a crosswind preference for rapidly contacting
the plume. However, there are several implicit assumptions in this model (see Cardé and Willis 2008 for a full
discussion). If the insect is to maintain a course that is on average headed downwind (or upwind) along the mean
wind direction over the time interval of a sweep exceeding 60° of wind direction, then it would need to keep track
of its heading in the environment (its cardinal direction) over the time interval during which the wind direction
shifted over 60°. Moreover, the insect would need to somehow ‘calculate’ the mean wind direction over this time
interval, which of course is not a constant. These requirements would appear beyond an insect’s capabilities. Likely,
if upwind, downwind, or crosswind ranging strategies are adopted (as opposed to a random heading with respect
to wind flow), the insect would make such a navigational decision based on the then current wind direction. This
model also does not consider that because of the forces of turbulent diffusion, the instantaneous distribution of
odour within the sweep would be patchy; instead the model assumes an above threshold concentration throughout
the sweep (Modified from Sabelis and Schippers (1984) and used with permission of Springer).

insignificant effect on the mosquitoes’ eventual direction of movement. Mosquitoes approaching

such nets generally ascended the net and then continued onwards with the same heading, so that
all captured mosquitoes could have simply headed upwind to reach the human bait.

Wind tunnel studies have produced conflicting evidence. Anopheles gambiae s.s. mosquitoes
in evident absence of host odours (other than ambient carbon dioxide) generally head upwind
(Costantini et al. 2001, Takken et al. 1997). Aedes aegypti mosquitoes, on the other hand, generally
take flight and head upwind only after contacting a plume of either carbon dioxide or human skin
odour (Dekker et al. 2005), possibly because they are diurnal, and have more visual information
about wind direction, they are more able to maintain diverse orientations to wind. Wind-tunnel
trials may be able to provide some information useful in establishing strategies of ranging flight
with respect to wind flow, although natural mosquito behaviour may be affected by spatial
constraints of the wind tunnel structure. In this artificial environment, mosquitoes can readily
fly upwind or downwind and the duration of such flights can be extended for many minutes

120  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

by moving a visual pattern below on the tunnel’s floor in the direction appropriate to simulate
the visual effect of the mosquito being displaced along its intended course (e.g. Kennedy 1940).
Crosswind flight in such tunnels, however, rapidly carries the insect to the wind tunnel’s wall and
so this assay method can only document short bouts of crosswind ranging. Specialised wind
tunnels that permit wind direction to shift or that mimic wind shifts by manipulation of the visual
surround may be useful in establishing instantaneous strategies for dealing with fluctuations in
wind direction (Drosophila, Zanen et al. 1994; tsetse, Colvin et al. 1989), but again they do not
allow simulation of a directional preference over even moderate distances (less than a meter) or
sustained time intervals. Therefore such assays are of limited value in establishing possible
ranging strategies in the field.

Rather than engage in ranging flight, some mosquitoes may perch on foliage, waiting for a host to
pass nearby. If the host is mobile, a sit-and-wait strategy may be useful, especially for day-active
species, as then the visual cues will be most prominent (Gibson and Torr 1999). Visual cues from
the host may serve to initiate flight (activation), and collimate flight toward the host, even after
encounter of the odour plume. Although visual cues would be expected to be useful for day-active
mosquitoes, they may also aid nocturnal orientation. However, it is important to bear in mind
that even diurnal species, such as Ae. aegypti, have eyes that are designed to maximise their light
gathering potential at the expense of resolution. In this respect their eye design is more similar
to nocturnal species, such as An. gambiae s.s., than to true diurnal species, such as Toxorhynchites
brevipalpis Theobald or Sabethes cayaneus (Fabricius) (Land et al. 1999). In practical terms, this
means that the light sensitivity of An. gambiae s.s. enables orientation to high-contrast images at
light intensities equivalent to only 1 log unit of starlight (Gibson 1995), i.e. typical of a moonless
starlit night, but their resolution is so poor (~8° angle of resolution) they would have difficulties
resolving an image the width of a person from a distance of 5 m, even in bright light, as compared
to ~50 m for a tsetse. If a host is moving, the range of detection would increase, although for
nocturnal species that feed on quiescent hosts, movement would be of relatively little importance.
The visual resolution of Ae. aegypti, which is generally referred to as diurnal even though it is most
active in the daylight hours nearest dawn and dusk, is only ~6°, as compared to 3° for a fully diurnal
mosquito such as Tx. brevipalpis, and 1-4° for higher flies, such as tabanids and tsetse (Gibson
and Young 1991). Nonetheless, by being active during daylight hours, the visual capabilities of
Ae. aegypti are greater than for strictly nocturnal species; the extra light enables it to detect and
respond to moving objects that it could probably not resolve at night (Muir et al. 1992).

Field studies at night are consistent with what we might expect from these physiological
measurements. Bidlingmayer and Hem (1979, 1980) used arrays of suction traps with and without
an associated dark visual cue approximately the height of a human, finding that several mosquito
species were capable of orienting to such a silhouette over a distance of at least 15 m in the
absence of any host-odour cues.

Bidlingmayer (1971) and Bidlingmayer and Hem (1981) also showed that mosquito species could
be characterised as ‘open grassland’ species or ‘woodland’ species, by their responsiveness to
visual features in the environment; grassland species tended to move out of woodland at dusk,
remaining close to the woodland edge at night, i.e. possibly following the sharp visual ‘edge’ of the
dark treeline against the lighter sky, whereas woodland species were only caught in the open field
at night if they were trapped in the vicinity of tall shrubs, which suggests they may fly from tree to
tree by attraction to the visual stimulus of the tree against the background of lighter vegetation.

Olfaction in vector-host interactions 121

Ring T. Cardé and Gabriella Gibson

Correlated with their response to visual features, grassland and woodland species may also differ
in their response to the types of odour plumes that occur in these two habitats. The behaviour of
wind, and hence odour plumes, is greatly affected by obstacles such as trees (see above). Studies
of the plume structure of pheromones released by moths (Willis et al. 1994) and bark beetles
(Wyatt et al. 1993) have shown that trees create turbulence in the pheromone-bearing odour
plumes, which causes odours to move vertically up and down the leeward side of the tree. The
flight paths of male moths and beetles in these plumes are quite different to plumes from point-
sources generated in open air. Similarly, the odour plumes of hosts in trees, such as nesting birds
and small mammals, are likely to be quite different to odour plumes emitted by animals in the
open. Accordingly, the flight pattern of woodland mosquito species may have evolved in response
to both visual features of the woodland and the plume-structure of woodland-dwelling hosts.

Pre-attack resting

Another useful strategy may be to find a host’s habitat and remain there either until a host appears
or until the time for biting. It has been proposed that ornithophagous Culex mosquitoes use
odours from bird droppings to find their hosts’ nocturnal roosting sites (Cooperband et al. 2008).
Highly anthropophilic species such as An. gambiae s.s. and Ae. aegypti enter human dwellings
where they may sit on interior surfaces, until they respond to sedentary hosts or to mobile hosts
that enter their dwellings during the mosquito’s normal blood-feeding time (Clements 1999). In
eastern Kenya, the propensity of Ae. aegypti to enter houses appears to be a heritable trait, based
on crosses of so-called domestic and feral strains. In mark-release-recapture trials with human
bait, the domestic strain had good fidelity for house-entering, the feral strain rarely entered
houses, and hybrids showed intermediate levels of house-entering (Trpis and Hausermann 1978).

Clements (1999) cites several intriguing examples of Anopheles spp. moving to human dwellings
or thorn bush cattle enclosures early in the evening and then host feeding at a later time. In
Anopheles gambiae s. l. (likely An. arabiensis Patton), there was a 4-hr interval between moving to
pre-attack resting sites near cattle hosts and biting (Smith 1958). Unresolved is whether the same
suite of host odours governs orientation to pre-attack resting sites near hosts and subsequent
orientation to the host. The act of ‘resting’ and its duration may be influenced by the same odour
cues that induced movement to the pre-attack site. Of interest is the evident differential daily
timing of these two orientation events and their different endpoints; our traditional laboratory
bioassay methods may be inadequate for capturing these differences. Indeed, many laboratory-
based bioassay studies report about a 50% response rate over several minutes (usually ~15 min),
which may be due to a natural stochastic rate at which individual mosquitoes become responsive
during the course of their ‘active’ phase. The duration of an assay also can colour its outcome
(Kennedy 1977); some host odours evoke a relatively rapid attraction, whereas others that induce
little response within the first few minutes become attractive, given sufficient assay time (e.g. 15
min, Dekker et al. 2001).

Following an odour plume in wind

Kennedy (1940) was the first to establish how an airborne insect orients upwind along an odour
plume, and, satisfyingly, his experimental subject was a mosquito! Using a small wind tunnel
(38 cm long, 10 cm wide and 7.5 cm high), Kennedy induced upwind flight in Ae. aegypti by
introducing his breath (thereby increasing the concentration of carbon dioxide, and other host
volatiles) into the airflow. By moving a pattern that was projected onto the tunnel’s floor, it was
clear that the mosquito adjusted its trajectory and ground speed by detecting the flow of its visual

122  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

surround. The optomotor response induces the subject to turn so as to maintain an even flow of
images across the visual field from front to back, which leads a flying insect to correct for wind
drift and maintain a stable orientation to the wind under field conditions (Gilbert and Kuenen
2008). This mechanism is termed odour-mediated optomotor-guided, positive anemotaxis and
its role in plume following behaviour has been verified in many other insect groups (moth spp.,
Kennedy and Marsh 1974, Willis and Cardé 2008; tsetse, Colvin et al. 1989). Indeed, it is the only
verified mechanism by which an airborne organism can gauge the direction of wind flow and then
set and maintain a heading relative to wind direction.

David et al. (1982) demonstrated that under general conditions of fluctuating wind direction
with a relatively constant velocity over an open space, at any point in an odour plume where
odours would be detected, the instantaneous wind direction in the immediate area points directly
towards the odour source (Figure 2B), even if the wind direction at the source has changed since
the odour packet left the source. In this general case, flying insects should set a steady course
directly upwind whenever they detect a host odour. This strategy works well when the distance
between the responding insect and the odour source is small (a few tens of metres or less) and the
wind direction does not vary much. When wind speed varies considerably (see below) and in more
complex landscapes, however, simply heading straight upwind after having detected odour, is a
less reliable strategy for locating an odour source. As a plume of odour is transported downwind,

A B C D

Figure 2. Schematic of the effects of a shifting wind direction on the path of meandering odour plumes, in this
case of sex pheromones, as viewed from above. Pheromone is shown as a series of puffs (circles) that expand
by turbulent diffusion over time. The paths of the puffs are depicted by the lines leading from the source along
the centre of each puff; the arrow tips designate the instantaneous wind direction within each puff. The dashed
lines and arrow tips represent the mean direction of wind flow at the source over the time course of each plume’s
meandering. (A) The instantaneous wind directions within the plume assumed by behavioural biologists before
David et al. (1982). The plume’s centreline and the instantaneous wind direction at any point are assumed to be
aligned and so simply would keep the responding insect within the plume’s boundaries and this course would
lead directly to the odour’s source. (B) Conforms with the experiments of David et al. (1982) in an open field with a
varying wind direction but a relatively constant wind speed. Heading upwind in the downwind sector of the plume
often leads the responder out of the plume. (C) Matches observations of Elkinton et al. (1987) beneath a forest
canopy and Brady et al. (1989) in an open savannah; this effect is caused by changes in wind direction and an
increase in velocity. Again, heading upwind can lead out of the plume. (D) Shows a steady wind direction – during
such ‘favourable’ and typically brief alignments of wind direction and the plume’s centreline, insects can make
rapid progress toward the source (reprinted with permission of Wiley-Blackwell from Elkinton et al. 1987, p. 400).

Olfaction in vector-host interactions 123

Ring T. Cardé and Gabriella Gibson

turbulent forces, especially those created by obstacles such as vegetation or environmental

features on a landscape scale, fragment the plume into filaments and create odour-free (or below
odour threshold of detection) gaps which can be large enough to impair orientation. Even in a
turbulent flow, however, filaments of relatively undiluted odour can persist over many metres of
transport, even as the average concentration of odours within the plume’s boundaries decreases
(Murlis and Jones 1981, Murlis et al. 1992). The presence of relatively undiluted odour filaments
permits detection of the odour plume farther downwind of the odour source than would be the
case if the plume was thoroughly mixed. In this case, flying insects may need to alter their heading
and ‘cast’ across the odour plume (as is the case in moths, Kuenen and Cardé 1994) or even back
track to some degree (as for tsetse; Torr and Mangwiro 1996) to re-establish contact with the
odour plume.

A second challenge for insects flying toward an odour source is that within an odour plume the
instantaneous direction upwind may lead, not along the plume’s long axis, but instead to the
plume’s boundaries and beyond. Shifts in wind direction realign the plume’s long axis away from
due upwind, even within a few metres of the odour source (Elkinton et al. 1987, Brady et al. 1989).
Moths following a pheromone plume upwind and then entering an odour-free area (caused either
by turbulent diffusion or by wind shifts) have a strategy for re-contacting the plume: they ’cast’
back and forth across the wind without making progress upwind. If this manoeuvre enables them
to recontact the plume, upwind flight along the plume resumes; if not, after several seconds
of casting, they shift to ranging flight. This strategy is most well-understood for moths, which
have a relatively low flight speed in relation to wind speed, so that by casting they more-or-less
hold station with respect to the ground, and hence with respect to the odour source, while they
continue to fly crosswind to increase their chances of re-encountering the odour stimulus (Kuenen
and Cardé 1994, Cardé and Willis 2008). Tsetse, however, which generally fly many times faster
than the mean wind speed, respond quite differently to losing odour by inadvertently leaving
an odour plume. They slow down and make a smooth turn of ~120° upwind, which takes them
roughly back to where they had last detected host odour, but further upwind and hence closer
to the odour source (Gibson and Brady 1985, 1988). In the case of tsetse and other blood feeding
insects, the odour source may not be stationary, so it may pay to be less precise in flying directly
upwind to where a host might have been (Williams 1994) than it is for moths to calling females.

The strategies used by plume-following mosquitoes to recontact the lost scent of a host are
unknown. It is clear, however, that a mosquito’s detection of a plume of host odour does not
ensure that she will be able to follow the odour plume to its source. Mosquitoes fly at about the
same ground speed in relation to wind speed as moths, and we might expect to find a strategy
of casting more likely than the smooth turns associated with tsetse. In wind-tunnel trials, Dekker
(2002) found that Ae. aegypti surged upwind upon contact with a carbon dioxide plume and
headed crosswind upon loss of contact – suggestive of moth casting upon loss of contact with of
pheromone plume.

Given that there is a tendency for wind speeds to cycle daily, with the highest speeds tending
to occur midday and the lowest speeds in the middle of night, wind direction should be more
readily detectable during the day when its speed is on average highest. Of course, there can
be lulls in wind flow at anytime, and even during the day the wind speed may fall below an
insect’s threshold of detection by the optomotor response. Detection of wind flow at night
generally is more difficult, however, because wind speed is on average reduced and low light
levels impair the visual detection of drift. Low wind speeds can be an advantage for weak fliers,
such as mosquitoes, sandflies and biting midges, however, which are easily blown off course at

124  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

higher wind speeds. Although mosquito eyes are certainly well-adapted for detecting strongly
contrasting visual features at low light intensities (Land et al. 1999), trap catches and landing
catches can be significantly higher on moonlit nights than on moonless overcast nights (Clements
1999), indicating that mosquitoes may well be operating near the limit of their visual abilities
under natural lighting.

It is difficult to generalise about the comparative challenges for mosquitoes navigating along
odour plumes in daytime, twilight, or night time conditions. Higher wind speeds typical of
daytime are correlated with less variability in the magnitude and degree of shifts in wind direction,
although turbulence levels also increase, causing odour gaps of larger size. Thus, at night plume
following should, in principle, be more challenging than in daytime, because at night there is
not only a reduction in wind velocity, but the low light level poses an increased difficulty in
detection of wind-induced drift, a prerequisite to determination of the direction of airflow.
However, the navigational capabilities of each species may be attuned to the characteristics of its
habitat and the meteorological conditions typical of the daily time it forages for hosts. In daytime,
the structures of odour plumes in open forests and in grassy fields differ markedly (Elkinton et
al. 1987) and in daytime the wind speed may be too high to permit plume following. However,
the wind flow at night in certain structured environments, such as valleys, can aid host-seeking
behaviour; catabatic winds produced by cool air moving down the valley floor under nocturnal
atmospheric inversions can give rise to a non-turbulent air current that flows gently downhill,
thereby producing a relatively steady stream of olfactory cues.

There are two other proposed strategies for flying upwind along an odour plume, neither of which
involves optomotor feedback. As wind velocity increases with distance from the ground, Gillett
(1979) suggested that mosquitoes dipping toward ground level might sense differences in their
acceleration within the wind shear via changes in pressure on mechanoreceptors and use this
feedback to gauge the direction upwind. If the mosquito was headed upwind, for example, a dip
would produce an increased acceleration because the mosquito would encounter a decrease in
airflow. Successive comparisons of accelerations versus decelerations might provide the needed
directional information. Mechanoreceptive feedback could be provided by sensory hairs, the
halteres or by the Johnston’s organ at the base of the antennae. This mechanism has not been
tested experimentally, and it assumes in part that mosquitoes continue to orient upwind in light
levels that are too dim to permit conventional optomotor response.

The second alternative mechanism is ‘aim-then-shoot’, wherein orientation over a distance

consists of a sequence of short flights. A stationary organism detects odour by olfactory sensory
organs and wind flow direction by mechanoreceptors while at rest on a substrate and then flies
upwind in a straight line, guided by visual cues that are set upon takeoff. Landing occurs if contact
with odour is broken and flight is reinitiated if odour is again sensed. Such a series of steps may
eventually lead to the odour source. This is a mechanoreceptive anemotaxis (Kennedy 1986) and
there is some evidence that it occurs in cabbage root flies and tsetse (see Cardé and Willis 2008),
although optomotor anemotaxis is also employed by tsetse (Gibson et al. 1991). The occurrence
of aim then shoot in mosquito host seeking remains speculative.

Orientation to odour in still air

Mechanisms for finding an odour source under either windless conditions or when airflow is
undetectable by visual means has been studied in depth only in moths. In wind-tunnel assays,
moths orientating along a pheromone plume formed in wind usually continue along the plume in

Olfaction in vector-host interactions 125

Ring T. Cardé and Gabriella Gibson

a zigzag path when the wind flow is quickly stopped (e.g. Kuenen and Baker 1983, Willis and Cardé
1990). These observations established that moths are capable of using a form of klinokinesis,
that is, orientation using successively sensed changes in pheromone concentration. In these
wind-tunnel manipulations, moths travelled about a metre along the plume in still air, apparently
having already set a course while in the wind using visual cues, which appears to have been
sufficient to enable them to continue in the same direction, albeit without wind. Changes in odour
concentration near the boundaries of a wind-formed odour plume could guide a flying insect
toward the source by ‘transverse klinokinesis’. If this mechanism is employed by mosquitoes, it
likely occurs when wind forms an odour plume, and then ceases or falls below a speed that can
be detected visually.

Mosquitoes could use longitudinal klinokinesis to self-steer a course toward an odour source.
The collimating cue would be increases in odour concentration detected during an approach
to the odour source. In still air the odour gradient would be caused by molecular diffusion and,
if this mechanism applies to mosquito orientation, it most likely occurs only very near (within
decimetres or less) of a host, particularly indoors on a still night when the inhabitants of the
dwelling are asleep. Such a gradient also can exist within a wind-formed odour plume, but
changes in concentration are steep enough only within decimetres of the source to be of potential
value in orientation (Murlis et al. 1992), and this mechanism would be redundant to simply using
anemotaxis.

Another orientation strategy useful close to a host would be to follow plumes formed by heat-
induced convention currents from a warm-blooded host. Such a strategy would enable orientation
to the body surface in still air. Odour plumes that are formed in this way would rise and orientation
downward toward the host could involve a range of cues, including; an optomotor response,
gravitational cues, sensing of a vertical heat gradient, sensing of a vertical odour gradient, plume
edges, or some combination of these cues. It is even conceivable that entirely ‘non-directional’
orientation, for example, by changes in concentration governing the rate of turning (klinokinesis)
or speed of flight (orthokinesis), could bring the mosquito close to the host (Williams 1994).
Convection currents appear to guide the selection of biting sites in members of the An. gambiae
complex (De Jong and Knols 1996, Dekker et al. 1998). Vertical flight has been shown in Ae. aegypti
to be influenced by optomotor cues (Daykin 1967). In a wind tunnel with vertical airflow, moving
stripes were used to simulate the visual effect of displacement due to an airstream; mosquitoes
tended to fly with respect to the stripes as though they were flying toward the odour source.
Once again, the optomotor response could account for downward descent toward a host along
a convective airstream.

Mosquitoes also may approach a host by following an increasing temperature gradient formed by
convection currents by klinokinesis, in a manner similar to following increasing chemical gradients
in still air. Their antennae are known to be particularly sensitive to small changes in temperature.
In Ae. aegypti, Davis and Sokolove (1975) found one set of thermoreceptors to be sensitive to
increases in temperature and a second set responded to sudden decreases in temperature.
Together, these receptors were capable of sensing changes in temperature as minute as ± 0.2 °C
(Davis and Sokolove 1975).

Because convection currents are created by a differential between body temperature and ambient
temperature, with the warmer air rising, the speed of the air current generated by convection
currents will depend on the magnitude of the difference between these temperatures. In the
tropics at night, there is often less than a 10 °C difference and therefore convection currents may

126  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

not move quickly enough or extend far enough from the source before the air reaches ambient
temperature. Nonetheless, with a sensitivity to temperature differences of as little as ±0.2 °C,
mosquitoes may be capable of following convection currents by following a temperature gradient
that varies by at least this much. Alternatively, they could close in on a host via klinokinesis and/
or orthokinesis, as described earlier in the case of movement along a vertical odour gradient
(Warnes 1995).

Eiras and Jepson (1994) found that some Ae. aegypti mosquitoes flew downward along heat
convention currents in the absence of any host cues in a small bioassay chamber; orientation was
enhanced by addition of water vapour and even more so by extracted human sweat. A human
hand, however, produced the highest proportion of source location – likely because of the suite
of natural host odours emitted. In such assays when the stimulus is a human hand it is difficult to
disentangle the effects of a temperature gradient from the potential effects of changes in odour
concentration. Aedes aegypti mosquitoes descended along a heat gradient presented alone, but
human skin emanations alone induce upwind flight (e.g. Dekker et al. 2005) and, in the absence of
wind or other cues such as heat, skin odours may induce downward flight. These kinds of assays
establish the cues contributing to downward movement in a natural convention current that is
laden with human skin odour. Just how the mosquito determines a downward direction is not
firmly established.

Which odours induce orientation?

In addition to being an ‘activator’, carbon dioxide seems to be a universal ‘attractant’ for female
mosquitoes (Gillies 1980, Reeves 1953, Rudolfs 1922, see also Chapter 4, this volume) and generally
all haematophagous arthropods, and it is widely employed as ‘bait’ in traps used for mosquito
surveillance. Carbon dioxide is of course ubiquitous as a natural constituent of the atmosphere (at
~0.04%), and it comprises approximately 4% of vertebrate exhalations. A host-generated plume
of carbon dioxide may be detectable over background for a considerable distance downwind.
Zöllner et al. (2004) released carbon dioxide at a rate equivalent to two cattle (20 l/min) in a
riparian habitat in daytime and detected the plume with a carbon dioxide sensor 64 m downwind
(the maximum distance sampled) as fluctuations of 300 ppm over background, albeit with
frequency of contact inversely proportional to distance from the source. Of course, the rate of
release of carbon dioxide from smaller vertebrate hosts would be less – chickens produce about
30-40 ml/min (Barott and Pringle 1946, Clements 1999). The distances over which carbon dioxide
alone is effective as an attractant have been measured with ramp traps arrayed in a radial pattern
from a central release of carbon dioxide at a rate equivalent to two calves (Gillies and Wilkes
1969, 1970, 1972). In these experiments the distances of ‘attraction’ of Aedes, Anopheles, Culex
and Mansonia species ranged from about 8 to 18 m. It is worth bearing in mind that traps are not
perfectly efficient at capturing all mosquitoes in the area, so the numbers caught may fall below
detectability when few mosquitoes approach it, and therefore this estimate of range of attraction
is likely to be an underestimate.

A wind-borne plume of carbon dioxide fluctuates in intensity against a background concentration

and so to detect a plume of carbon dioxide mosquitoes could rely on sensing its intermittent
structure, an idea dating to Wright and Kellogg (1962). Mosquitoes detect carbon dioxide by
receptors on the maxillary palps and an increase or decrease in concentration of as little as 50 ppm
from ambient clearly modifies receptor firing rates in Aedes, Anopheles and Culex mosquitoes (Grant
and O’Connell 1996). In assays with odour plumes filling an 8 cm-diameter Y-tube olfactometer,
a fluctuating intensity of carbon dioxide induced a higher proportion of net upwind flight than

Olfaction in vector-host interactions 127

Ring T. Cardé and Gabriella Gibson

a homogeneous cloud of carbon dioxide in Ae. aegypti (Geier et al. 1999a). In a wind tunnel, Ae.
aegypti turned upwind and accelerated its flight speed after a brief encounter with an individual
filament of carbon dioxide (Dekker et al. 2005). These two observations suggest that in Ae. aegypti
plume intermittency is associated with sustained upwind movement, stimulated by repetitive
encounters with carbon dioxide filaments, similar to the mechanism proposed for upwind
orientation of male moths along a turbulent pheromone plume (Mafra-Neto and Cardé 1994,
Vickers and Baker 1994). These findings with Ae. aegypti suggest that it is the fluctuating intensity
of carbon dioxide over background that induces sustained upwind flight. The applicability of this
principle to the upwind orientation other mosquitoes remains to be investigated.

Odour salience

In a wind tunnel assay with Ae. aegypti, a single, transient (<100 ms) exposure to a filament of
carbon dioxide lowered the threshold of response to human skin odour by a factor of more than
ten (Dekker et al. 2005). This finding could mean that for Ae. aegypti the ’long-distance’ attractant
is carbon dioxide and not human skin odour, and/or that carbon dioxide lowers the threshold
for attraction to a human skin odour, acting at some distance from the host. Alternatively, as this
species is highly anthropophilic and would be expected to be in or near human dwellings before
the final stages of host orientation, the presence of a fluctuating intensity of carbon dioxide could
signal that a potential host is nearby and initiate further orientation mediated by both carbon
dioxide and skin odours.

Odours other than carbon dioxide influence orientation and it has been proposed that for some
mosquito species with narrow host ranges, such as some highly anthropophilic species, carbon
dioxide alone may not induce orientation. Instead, at long-range, mosquitoes would sense
specific host odours in addition to the carbon dioxide plume and carbon dioxide alone would
be of diminished importance to the induction of plume following (e.g. Mboera and Takken 1997,
Mboera et al. 1998, Takken and Knols 1999, Pates et al. 2001). Whether host odours other than
carbon dioxide induce plume following at long range in such species is an issue that remains
unresolved and which will be difficult to verify experimentally.

Traditionally, mosquitoes have been characterised by their ‘host preferences’, which is mainly
based on the outcome of their responsiveness to host cues: the origin of the blood imbibed by
individual mosquitoes. Even after taking into account biases due to relative host availability (e.g.
as measured by various indices, including relative number or biomass of hosts present, but not
protected from mosquitoes), it is clear that even amongst species that transmit human pathogens,
the profiles of ‘host preference’ are highly variable between species, and even within species over
their geographic range of distribution (Clements 1999). At first sight, one might expect differences
in host preference to be due to differences in the profiles of host odours to which each species
is attracted. There may be host-specific odours, or ratios of blends to which particular species
are most responsive, but so far, no combination of odours has been found to be as attractive as
a live host with carbon dioxide at biologically relevant doses), or to mimic a specific type of host
(Chapter 7, this volume). However, we do know that the pattern of responsiveness to whole body
odours and carbon dioxide vary considerably among species and even within a species. ‘Host
preference’ may reflect differences between species in the response to particular chemicals. In
addition, differences between host preferences as striking as that noted between members of the
An. gambiae species complex (An. gambiae s.s. is one of the most anthropophilic species in the
world, An. arabiensis is also highly anthropophilic, but rather more opportunistic, often feeding
on cattle and An. quadriannulatus (Theobald) feeds almost entirely on bovids) may be due to

128  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

behaviours associated with host location, such as house entry, that are confounded with plume
following. Current methods of assaying the response to odours in the field, such as odour-baited
traps of all types, cannot distinguish between these two kinds of orientation.

To investigate further the relative responsiveness of mosquitoes to carbon dioxide and other host
volatiles, Costantini et al. (1996) undertook a series of dose-response assays in the field using
odour-baited entry traps (OBETs) to test the relative importance of carbon dioxide to a range
of mosquito species, and found that for each species the catch increased with dose of carbon
dioxide in a similar log-linear fashion. If the dose-response data are considered in relation to a
standard human-landing catch, however, the behaviour of each species was quite different. At
one extreme, even the highest dose of carbon dioxide did not catch more An. gambiae s.l. (highly
anthropophilic) than a single human landing catch, whereas for Mansonia uniformis (Theobald)
(zoophilic, with generalist host preference) the three highest doses of carbon dioxide caught
significantly more mosquitoes than a human bait, and An. funestus Giles (also anthropophilic)
did not respond to carbon dioxide alone in sufficient numbers to assess a dose response. These
findings demonstrate that the response to carbon dioxide and other human volatiles is strikingly
species specific, and, therefore, we should expect to discover a variety of mechanisms by which
species effectively ‘choose’ a host. With respect to An. gambiae s.s., these results suggest that
although increasing doses of carbon dioxide increased the catch, the responsiveness to host-
specific cues may take over in importance after an initial attraction to carbon dioxide; in the
absence of human-specific cues, it appears that An. gambiae s.s. was either less attracted to enter
the OBET (proxy for house entry) and/or did not stay in the vicinity of the trap long enough to be
caught in high numbers as the relative abundance of human odours declined. For Ma. uniformis,
carbon dioxide alone appears to be sufficient to arrest females near the source and elicit ‘entry’
behaviour even in the absence of other host cues.

Still-air bioassays

The precise role assigned to carbon dioxide in mediating orientation may be contingent on
the type of assay used. Because An. gambiae s.s. fly upwind in unscented air in a wind tunnel
(Costantini et al. 2001, Takken et al. 1997), this assay method is difficult to use in orientation
studies with this species, although upwind flight centred within a discrete odour plume may
signify attraction. Much of what we know about its odour-mediated behaviour and the behaviour
of other An. gambiae s.l. complex species stems from a series of two-port, still air olfactometer
studies comparing the highly anthropophilic An. gambiae s.s. with zoophilic species such as
An. quadriannulatus. A large, still-air olfactometer determines ‘attraction’ by capture in odour-
emitting ports; it does not allow upwind flight along a wind-borne plume, except very close to the
port entrance. This bioassay method was not designed to explore how odour plumes modulate
orientation, however, but to identify potential host-emitted attractants.

In one study, port entrance of An. gambiae s.s. was induced by odour sources previously in contact
with human skin, whereas carbon dioxide was ineffective in inducing port entrance (Pates et al.
2001), leading to the conclusion that carbon dioxide alone played a minor role in orientation.
Other studies, however, found that An. gambiae s.s. were intercepted by an electrocution net
placed 1 cm in front of carbon dioxide sources (either homogeneous or turbulent presentations),
even though these odour sources of carbon dioxide were not effective for port entry (Dekker et
al. 2001). When the carbon dioxide source was positioned 5 cm downwind of the port entrance,
the capture rate for human skin emanations was elevated (Spitzen et al. 2008), further confirming
that carbon dioxide could play a role in guiding An. gambiae s.s. to a host.

Olfaction in vector-host interactions 129

Ring T. Cardé and Gabriella Gibson

Anopheles quadriannulatus, however, was not captured by carbon dioxide alone in this assay, but
cow odour plus carbon dioxide was efficacious, and, unexpectedly, human odour alone was even
more effective (Pates et al. 2005). The port-entry, still-air bioassay has produced a wealth of basic
information on mosquito odour preference (and repellency) on species in the An. gambiae s.l.
complex, but this method does not provide substantive insight into orientation per se. Rather this
bioassay measures the culmination of what is presumed to be a complex repertoire of behaviours,
and this method may not be helpful in determining specific orientation manoeuvres that occur
along odour plumes meters or more downwind of the source.

In the case of Ae. aegypti, human skin odour alone also induces upwind flight and it seems to induce
the highest proportion of rapid source location when it is presented in a relatively homogeneous
rather than an intermittent plume, the opposite effect of plume structure with carbon dioxide
(Dekker et al. 2005, Geier et al. 1999a). A relatively homogeneous plume of skin odour may occur
very close to the host. Numerous human odours potentially involved in attraction have been
identified from skin head space (e.g. Bernier et al. 2000) and sweat (Braks et al. 2001, Cork and Park
1996), but relatively few compounds have proved ‘attractive’, that is, induce upwind movement
over short distances in laboratory assays. L-lactic acid, short-chain fatty acids and ammonia are
attractive for Aedes and Anopheles (e.g. Acree et al. 1968, Bosch et al. 2000, Braks et al. 2001, Dekker
et al. 2002, Geier et al. 1999b). Carboxylic acids, alcohols and aldehydes are attractive for Culex
(Puri et al. 2006). Generally these assays measure either displacement in an upwind direction,
movement upwind over relatively short distances or into an entrance port of a still-air bioassay
chamber, often over many minutes. It is unclear if these odour-induced movements correlate with
host-finding manoeuvres in the field or if the airborne concentrations of odours are presented at
levels comparable to those that would occur naturally (see Chapter 7 and Chapter 8, this volume).

None of these compounds has been tested in an assay that would verify that these compounds (or
some combination) lure mosquitoes upwind over long (multi-metre) distances. These compounds
may combine with the carbon dioxide plume to act additively or synergistically to induce
attraction from a distance. Alternatively, these odours may modulate only late-stage orientation
manoeuvres such as hovering near or landing on the host (e.g. Healy and Copeland 2000, Healy et
al. 2002), contributing little to orientation at the 10 to 20 metre range documented as the range
of attraction of carbon dioxide (Gillies and Wilkes 1969, 1970, 1972).

Hovering and landing entail very different visual feedback than generated by upwind flight.
Recent studies of the heading of Drosophila tethered in flight simulators (Fry et al. 2009) have
provided confirmation of the role that visual feedback plays in regulating upwind flight to an
odour source. In this setup, the flow of the visual surround from computer-controlled banks of
LEDs was set to pass from ahead to around each side of the fly. This simulated what the fly would
experience during forward flight in wind and indeed flies directed their heading towards the
visually simulated upwind direction. Forward flight also requires input from mechanosensory
cues, provided at least in part by the Johnston’s organ (Budick et al. 2007). Drosophila also orient
towards vertical silhouettes, which in the flight simulator is represented by a vertical band of
LEDs that are turned off; this visual stimulus presumably mimics a suitable landing or feeding
site (Maimon et al. 2008). In nature, such a visual stimulus directly ahead would appear to expand
during approach, causing the fly to slow its progress before landing. When upwind flight is
stimulated, however, flies generally avoid objects ahead, turning either right or left. (A fly in free
flight heading upwind would crash into a rapidly looming visual stimulus if it did not treat the
object as an aversive visual cue.)

130  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

As a mosquito approaches a host, perhaps first hovering nearby before landing, it also must alter
the pattern of visual feedback from a flow of the visual surround signifying upwind movement
to frontal object expansion. It is tempting to speculate that there are specific stimuli, such as
particular host odours perceived at close range, that trigger this change, as suggested by the work
of Healy and Copeland (2000) and Healy et al. (2002).

Field traps baited with carbon dioxide and other host odours

Carbon dioxide has long been used as attractive bait in traps used in surveillance programmes
(e.g. CDC-type traps, see Qiu et al. 2007a). The most commonly used trap configurations capture
mosquitoes by drawing them upwind along a carbon dioxide plume leading to the vicinity of
suction near the trap entrance. Three-dimensional video observations of Cx. quinquefasciatus
and Culex tarsalis Coquillett mosquitoes orienting to such traps in a large field wind tunnel
(Cooperband and Cardé 2006a,b) have shown that of the mosquitoes reaching the origin of the
carbon dioxide plume, relatively few are actually captured (Figure 3). The reason for such a low
proportion of capture is not certain, but it appears that reaching the origin of the plume does
not routinely place the mosquito within an area of suction and that the most mosquitoes do not
engage in localised ‘search’ near the origin of the carbon dioxide plume long enough to enter an
area of suction. Most mosquitoes simply depart after a few to tens of seconds in the vicinity of the

Figure 3. The track of a female Culex tarsalis flying in the vicinity of a suction trap baited with carbon dioxide (dry
ice) shown from three perspectives. This mosquito was not captured and eventually departed the trap’s vicinity.
This track was video recorded in 3-D in a field wind tunnel. XZ is a side view with the air flowing from right to
left, XY is an overhead perspective, and YZ is a view from upwind of the trap (reprinted with permission of Wiley-
Blackwell from Cooperband and Cardé 2006b, p. 24).

Olfaction in vector-host interactions 131

Ring T. Cardé and Gabriella Gibson

trap. It seems likely that capture rates are contingent in part on how long a given mosquito spends
in the vicinity of the trap’s effective area of suction: the longer it lingers, the greater the chance
its meandering flight will take it into the region of ‘no return’. Similar capture inefficiencies apply
to other mosquito species with different types of traps, as shown by Torr et al. (2008), described
in the next section, although much remains to be determined.

Adding additional odours such as 1-octen-3-ol to carbon dioxide can improve trap catch for some
mosquitoes (e.g. Takken and Kline 1989), but the reasons for this are speculative. Such compounds
may lower the threshold for attraction, thereby extending the downwind reach (active space)
of the plume. These compounds also may act additively or synergistically with carbon dioxide,
evoking a response in a higher proportion of mosquitoes. Both of these effects would act at a
distance and increase the number of mosquitoes lured to the trap. Alternatively, these compounds
may add to time spent near the trap, thereby increasing the chance that a circuitous flight path
will intersect the area of suction, and so elevate the probability of capture. At our present state
of understanding it is difficult to know which of these alternatives may apply to compounds that
can augment trap catch over carbon dioxide alone. It is also noteworthy that, with few exceptions,
these chemicals typically provide only a modest increase in catch when added to carbon dioxide:
1-octen-3-ol, for example, doubled or tripled catches of Aedes species, for example, whereas for
some Culex species it can depress the catch (Kemme et al. 1993, Van Essen et al. 1994). Curiously,
unbaited suction traps captured no more Culex species than traps baited with 1-octen-3-ol
(Mboera et al. 2000, Qiu et al. 2007b).

Field trapping with electric nets (E-nets) and odour-baited entry traps (OBETs)

One approach to establishing which odours mediate attraction at long range and close to the
odour source is to use electric nets set a distances of interest, as has been used by Vale and
colleagues with great success in dissecting the orientation responses of tsetse to odours and
visual cues (e.g. Vale 1982). E-nets are designed to intercept tsetse passively, and can be used to
establish details of the approach to an odour source and the relative numbers of flies arriving at
different olfactory and/or visual baits without the need for any additional behavioural responses
that are normally required to catch tsetse as they arrive at an odour source, such as flying into a
trap, or landing on a target. E-nets are proving to be useful for similar purposes with mosquitoes
(Knols et al. 1998, Torr 1994, Torr et al. 2008).

Another device that is useful for assaying the responses of anthropophilic mosquitoes to host
cues is the odour-baited entry trap (OBET), which is intended to simulate the odour plume
emanating from a human dwelling and permits mosquitoes to enter the trap via a window-sized
port (Costantini et al. 1993). Entering an OBET is assumed to be equivalent to house-entering
behaviour and it serves as a proxy for an open window. OBET entry differs, however, in one
respect from house entry: the odour plume from an OBET would be relatively homogeneous
in structure, because it is driven by a fan, exiting at a velocity of about 0.5 m/s from the port.
Artificial ‘smoke’ was used to establish the patterns of odour dispersing from experimental ‘huts’
that were designed to typify entry points of human dwellings in West Africa. The issuing of this
visual surrogate for odour suggested a more intermittent plume structure (Snow 1987) than the
fan-generated, homogeneous outflow from OBETs.

Torr et al. (2008) have combined the E-net and OBET techniques to compare the orientation of
An. arabiensis and An. quadriannulatus in the field. Such a comparative approach measures two
successive anthropophilic behaviours involved in orientation: attraction and house-entering. It

132  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

can be used to compare odour sources such as carbon dioxide with or without added odours and
also natural baits such as cattle or humans. In these comparisons it was clear that the propensity
to be attracted to the vicinity of an odour source did not always correlate with house entering.
For example, An. arabiensis was equally attracted to the vicinity of human and cattle baits, but was
far more apt to enter an OBET baited with human odour. This finding indicates that there may be
an interaction between the response to a particular host odour and the house-entry response.
Although this observation has not been fully explored, it goes some way to explaining why this
species is paradoxically the second most important vector of malaria in sub-Sahara Africa and
yet is considered to be significantly less anthropophilic than its sibling species, An. gambiae s.s.;
the human blood index (proportion of blood meals of human origin) is virtually 100% for An.
gambiae s.s in most of areas, but varies for An. arabiensis across its distribution, from 80-100% in
western Africa, to 7-60% in eastern Africa and <10% in Madagascar (Garrett-Jones et al. 1980). In
field studies in Ethiopia, Tirados et al. (2006) found the An. arabiensis population inherently to be
anthropophagic, but this was counterbalanced by its tendency to blood feed outdoors and remain
outdoors after feeding, where it mainly came into contact with cattle at night. The disparity is
presumably indicative of differential odour suites mediating attraction from a distance and house-
entering. Of course, this also reveals that we do not have the observational tools necessary to fully
understand what mediates the full sequence of host-finding behaviours.

How far do mosquitoes travel to locate hosts?

‘Dispersal studies’, generally conducted by mark-release-recapture experiments, provide some

insights into how far mosquitoes travel during the course of their ‘search’ for resources such
as blood-meal hosts and oviposition sites. Obviously, mark-release-recapture studies can only
assess the summation of all distances travelled from the time of release to capture, and provide
no information about the route taken by individuals that are captured or the cues that guided
this movement. ‘Dispersion’ in the sense used here refers to flight that carries mosquitoes away
from a release point, either guided by olfactory cues, wind direction and environmental features
or non-directed flight (i.e. random with respect to environmental stimuli). Across the range of
species and habitats that have been investigated, species fall into several categories with respect
to measured patterns of dispersal. At one extreme, salt marsh breeding species are known to fly
kilometres away from breeding sites. There is evidence that this may be true migratory flight;
they appear to be unresponsive to host stimuli, flying too high to detect host-odour plumes
and they are generally carried by the wind (Silver 2008). Anopheles gambiae has been shown to
disperse from breeding sites for hundreds of metres to kilometres, depending on environmental
conditions. Where breeding sites are closely associated with human dwellings, they disperse less
far than in areas where they breed in large irrigated fields at a distance from villages (Gillies 1988,
Takken et al. 1998). It is not known to what extent these flights are migratory vs. directed flight in
response to host stimuli. Finally, the maximum distance that Ae. aegypti appears to disperse is on
the order of hundreds of metres, probably because they tend to breed in relatively high density
urban habitats, where the main resources required for blood feeding and breeding occur in high
density (Maciel-de-Freitas 2010, Reiter et al. 1995, Scott et al. 2000) Rather than providing much
insight into the mechanism by which mosquitoes locate hosts, dispersal studies may be best used
to establish the density of species-specific resources in localised areas, and to indicate something
about how far and in what direction with respect to environmental features mosquitoes need to
fly before they encounter a resource-rich area that arrests further flight.

Olfaction in vector-host interactions 133

Ring T. Cardé and Gabriella Gibson

Discussion

The mechanism of host location in mosquitoes over distances in the range of tens of metres
(beyond visual range) is thought to be confined to tracking a plume of wind-borne host odour
upwind toward the odour source by optomotor anemotaxis. At close range in still or nearly still
air (as would occur in a human dwelling) a mosquito might orient by klinokinesis, using changes
in odour concentration, or following convection current temperature gradients to steer toward
the host, but these mechanisms are so far unverified. Gross visual cues from the host also could
serve as collimating cues at ranges of ten or fewer metres and if the mosquito enters the host
odour plume, both cues may be used in orientation. Near the host, the final stage of orientation
and then landing may be governed by visual and odour cues combined with host-produced
heat and humidity. In still air, orientation may involve downward flight along heat-convected
odour plumes. The probable and presumed most common navigational mechanisms for host
orientation, categorised by distance from the host and the presumed sensory inputs governing
these manoeuvres, are summarised in Table 1.

There is enormous variation in conditions under which orientation occurs, contingent on habitat,
how wind disperses the plume, time of day or night and associated light levels, availability of
visual cues from the host, and host movement. The rate of carbon dioxide emission varies with
host size and the odour profiles of acceptable hosts vary among individuals of the same species.
In humans there is considerable individual variation in odour-based attractiveness (e.g. Lindsay
et al. 1993, Logan et al. 2008), but how such variability affects mosquito orientation is not well
understood.

Table 1. A simplified overview of the orientation mechanisms and their probable sensory inputs used by female
mosquitoes for location of a human-sized host. At distances of ≈1 to10 m from the host, orientation to hovering
range can be mediated by solely by visual cues (a gradual expansion of the host fixed in the frontal visual field) and
the mosquito need not be in contact with the plume of host odour. Hovering near the host may include orientation
to specific body regions that are particularly favoured by given species. These simplifications do not account for
other sequences, such as orientation to host odour in still air or orientation to the host’s habitat (such as a human
dwelling) with a delay in biting. Mechanosensory input is also important to maintenance of upwind movement.

Distance

>10 m ≈10 to1 m ≈1 m host contact

Odour cues fluctuating CO2 fluctuating CO2 + fluctuating CO2? + fluctuating CO2?
other kairomones other kairomones kairomones inducing
heat? humidity? landing?
heat? humidity?
Behavioural output upwind anemotaxis upwind anemotaxis hovering near host landing on host
Visual input front-to-rear flow front-to-rear flow steady size of large rapid expansion of
of downward visual of downward visual frontal visual field frontal visual field
field field
Visual cues visual surround host body host body host body

134  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

Much of what is known about the plume-following manoeuvres of female mosquitoes has been
inferred from assays aimed, not at defining orientation mechanisms, but rather designed to aid in
the identification of the host-emitted volatiles mediating attraction. Do these bioassays monitor
responses involved in orientation? If not, are behaviourally active compounds being overlooked?
As responses may be contingent on aspects of the physiological state of the mosquitoes used in
the assay which have not been fully explored, such as distance flown and duration of exposure
to host cues, and characteristics of stimuli presentation, such as concentration, odour plume
structure, presence of carbon dioxide, particular odour mixtures and their ratios, devising
diagnostic bioassays is likely to remain a difficult task.

In other odour-based orientation systems in insects, some of these questions are answered with
comparative ease. The pheromone blend emitted by female moths is, for example, relatively
simple, comprising one to four constituents in nearly all species. All stages of response, from
the initial detection of the pheromone plume at a distance to contact with the female appear
governed by the entire pheromone blend, which is assumed to vary little because of stabilising
selection. In mosquitoes, the initial detection of the host odour plume may involve only carbon
dioxide or it may involve carbon dioxide plus additional host-emitted odours. Close to the host,
these added odours may be more significant in evoking specific behaviours, such as habitat
finding (pre-attack resting sites), house-entering, and landing, as was found to be the case with
the variability in response of tsetse to different ratios and concentrations of known attractants
in field studies (Brady and Griffiths 1993). In contrast to moths where the pheromone signal is
relatively simple and invariant, the odour profile of mosquito hosts is spectacularly variable and
involves hundreds of potential components, adding to the difficulty of establishing which odour
constituents contribute to orientation.

Future studies

Much of what we know about mosquito orientation to hosts and the odours they emit is based on
studies of a handful of model species, often from colonies reared in the laboratory for decades,
and mainly of species that vector human pathogens. Future studies should validate behavioural
findings with wild mosquitoes to be sure we are not misguided by behavioural artefacts of
colonisation. A wide range of approaches to exploring host location can be undertaken in the
field and in the laboratory. First, in wind-tunnel and still-air bioassays of flight along odour plumes,
it is possible to control features of plume structure and air movement while monitoring mosquito
movement in 3-D with video. With such bioassays it is also possible to introduce additional host-
related cues thought to contribute to orientation: heated surfaces, humidity gradients, and visual
objects. Although the characteristics of plume structure can be measured at fixed points in a
wind tunnel (e.g. Justus et al. 2002, Dekker et al. 2005), because odour intensity fluctuates at
any given spatial position, it is not possible to specify how instantaneous odour input matches a
particular orientation manoeuvre. Nonetheless, it is possible to experimentally control the overall
structure of the odour plume and then determine when the mosquito is clearly within or outside
it, although the plume’s ragged boundary will remain an area of uncertainty. When a mosquito
enters an odour plume, its change in orientation may be attributed to odour contact.

Identifying the odour stimuli evoking close approach to the host and alighting may require new,
specialised bioassays (e.g. Healy and Copeland 2000, Healy et al. 2002). Different mosquito species
appear to select favoured body regions of humans for biting (De Jong and Knols 1996), possibly
cueing in on region-specific odours or their interaction with the carbon dioxide plume and other
breath constituents. Many kinds of sensory input are available in close proximity to the host –

Olfaction in vector-host interactions 135

Ring T. Cardé and Gabriella Gibson

increased humidity, heat, and visual cues, adding to the challenge of bioassay design. All of these
factors appear to be enough to guide mosquitoes to a host in virtually still air, apart from the air
currents generated by convection, which also appears to be important to selection of a biting site
(Dekker et al. 1998). Further studies are warranted on the behaviour of mosquitoes in still air in
the presence of this suite of stimuli to understand more fully the mechanisms of host orientation
within enclosed spaces.

A second approach is to document manoeuvres in the field. This has the obvious advantage
of observing mosquitoes behaving under natural circumstances. This technique is technically
challenging, however, because mosquitoes are small and it is difficult to accurately monitor their
manoeuvres, in part because in the field the background is visually heterogeneous. If more than
one species is present, it may be impossible to know which species is being observed. Large, field
wind tunnels (Cooperband et al. 2006a,b) offer a bridge to observing freely flying mosquitoes in
a relatively natural environment, but the relatively laminar flow of wind and the plume structure
that is generated may not be fully reflective of the patterns that are encountered under natural
conditions, and of course, the distance to be travelled is limited by the size of the tunnel. Moreover,
whereas it is possible in a wind tunnel setting to specify where the odour plume’s boundaries
are relative to a manoeuvring mosquito, in the field, it is difficult to establish precisely when a
mosquito is in contact with the odour plume. Nonetheless, plume structure is clearly a relevant
cue and one that future work should emphasise.

Another approach is to monitor the orientation manoeuvres in field by E-nets and OBETs (Torr et
al. 2008). These techniques allow documentation of orientated responses (‘attraction’) to a live
host, to odour sources of carbon dioxide, or to synthetic kairomones. Attraction can be compared
to entry into dwellings baited with odours or hosts, as measured with OBETs. These responses can
differ dependent on how the odour source is presented. In other words, for a given species, what
appears to ‘attractive’ with host as the odour source may prove less so when measured with an
OBET. One unresolved concern with OBETs is that the odour plumes they generate may not mimic
the structure of naturally fenestrated odour plumes.

Identification of the suite of natural host odours inducing attraction and other host-directed
behaviours would greatly facilitate our ability to carry out studies of the mechanisms of orientation.
For most of their lives female mosquitoes cycle between responding to host cues, resting-site cues
while they digest blood and eggs develop, and oviposition cues. Now that the genomes of An.
gambiae s.s, Ae. aegypti and Culex pipiens L. have been sequenced, it will be possible to investigate
correlations between gene expression and behaviour throughout the gonotrophic cycle, enabling
the identification of the proteins involved with odour detection, which may in turn help to identify
particular odorants to which mosquito sensory systems are primed to be most responsive to
during each physiological stage (see Chapter 2, this volume).

Acknowledgments

We are very grateful to Steve Schofield for his comments on this review.

References

Acree, F, Turner R, Grouck HK, Beroza M and Smith N (1968) L-lactic acid: a mosquito attractant isolated from humans.
Science 161: 1346-1347.

136  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

Barott HG and Pringle EM (1946) Energy and gaseous metabolism of the chicken from hatch to maturity as affected by
temperature. J Nutr 31: 35-50.
Bernier UR, Kline DL, Barnard DR, Schreck CE and Yost RA (2000) Analysis of human skin emanations by gas
chromatography/mass spetrometry. 2. Identification of volatile compounds that are candidate attractants for the
yellow fever mosquito (Aedes aegypti). Anal Chem 72: 747-756.
Bidlingmayer WL (1971) Mosquito flight paths in relation to the environment. 1. Illumination levels, orientation, and
resting areas. Ann Entomol Soc Am 64: 1121-1131.
Bidlingmayer WL and Hem DG (1979) Mosquito (Diptera: Culicidae) flight behaviour near conspicuous objects. Bull
Entomol Res 69: 691-700.
Bidlingmayer WL and Hem DG (1980) The range of visual attraction and the effect of competitive visual attractants upon
mosquito (Diptera: Culicidae) flight. Bull Entomol Res 70: 321-342.
Bidlingmayer WL and Hem DG (1981) Mosquito flight paths in relation to the environment. Effect of the forest edge
upon trap catches in the field. Mosq News 41: 55-59.
Bosch OJ, Geier M and Boeckh J (2000) Contribution of fatty acids to olfactory host finding of female Aedes aegypti.
Chem Senses 25: 323-330.
Brady JN (1975) Circadian changes in central excitability – The origin of behavioural rhythms in tsetse flies and other
animals? J Entomol (A) 50: 79-95.
Brady J and Griffiths N (1993) Upwind flight responses of tsetse flies (Glossina spp.) (Diptera: Culicidae) to acetone,
octenol and phenols in nature: a video study. Bull Entomol Res 83: 329-333.
Brady J, Gibson G and Packer MJ (1989) Odour movement, wind direction and the problem of host-finding by tsetse
flies. Physiol Entomol 14: 369-380.
Braks MAH, Meijerink J and Takken W (2001) The response of the malaria mosquito, Anopheles gambiae, to two
components of human sweat, ammonia and L-lactic acid, in an olfactometer. Physiol Entomol 26: 142-148.
Budick SA, Reiser MB and Dickinson MH (2007) The role of visual and mechanosensory cues in structuring forward flight
in Drosophila melanogaster. J Exp Bio 210: 4092-4103.
Cardé RT and Willis MA (2008) Navigational strategies used by insects to find distant, wind-borne sources of odor. J
Chem Ecol 34: 854-866.
Chiba Y, Shinkawa Y, Yoshii M, Matsumoto A, Tomioka K and Takahashi SY (1992) A comparative study on insemination
dependency of circadian activity pattern in mosquitoes. Physiol Entomol 17: 213-218.
Clements AN (1999) The biology of mosquitoes, vol 2. CAB International, Wallingford, UK.
Colvin J, Brady J and Gibson G (1989) Visually guided, upwind turning behaviour of free-flying tsetse flies in odour-laden
wind: a wind-tunnel study. Physiol Entomol 14,: 31-39.
Cooperband MF and Cardé RT (2006a) Comparison of plume structures of carbon dioxide emitted from different
mosquito traps. Med Vet Entomol 26: 1-10.
Cooperband MF and Cardé RT (2006b) Orientation of Culex mosquitoes to carbon dioxide-baited traps: flight
manoeuvres and trapping efficiency. Med Vet Entomol 26: 11-26.
Cooperband MF, McElfresh JS, Millar JG and Cardé RT (2008) Attraction of Culex quinquefasciatus Say (Diptera: Culicidae)
to odors from chicken feces. J Insect Physiol 54: 1182-1194.
Corbet PS and Smith SM (1974) Diel periodicities of landing of nulliparous and parous Aedes aegypti (L.) at Dar es
Salaam, Tanzania (Diptera: Culicidae) Bull Entomol Res 64: 111-121.
Cork A and Park KC (1996). Identification of electrophysiologically-active compounds for the malaria mosquito,
Anopheles gambiae, in human sweat extracts. Med Vet Entomol 10: 269-276.
Costantini C, Gibson G, Brady J, Merzagora L and Coluzzi M (1993) A new odour-baited trap to collect host-seeking
mosquitoes. Parassitologia 35: 5-9.
Costantini C, Gibson G, Sagnon N’F, della Torre A, Brady J and Coluzzi M. (1996) Mosquito responses to carbon dioxide
in a West African Sudan savanna village. Med Vet Entomol 10: 220-227.
Costantini C, Birkett MA, Gibson G, Ziesmann J, Sagnon N’F, Mohammed HA, Coluzzi M and Pickett JA (2001)
Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae to human-specific sweat
components. Med Vet Entomol. 15: 259-266.

Olfaction in vector-host interactions 137

Ring T. Cardé and Gabriella Gibson

Craig GB (1967) Mosquitoes: female monogamy induced by male accessory gland substances. Science 156: 1499-1501.
David CT, Kennedy JS, Ludlow AR, Perry JN and Wall C (1982) A reappraisal of insect flight toward distant point source
of wind-borne odor. J Chem Ecol 8: 1207-1215.
Davis EE and Sokolove PG (1975) Temperature responses of antennal receptors of the mosquito, Aedes aegypti. J Comp
Physiol 96: 223-236.
Daykin PN (1967) Orientation of Aedes aegypti in vertical air currents. Can Entomol 99: 303-308.
De Jong R and Knols BGJ (1996) Selection of biting sites by mosquitoes. In: Block GR and Cardew G (eds) Olfaction in
mosquito-host interactions. Ciba Foundation Symposium 200, John Wiley and Sons, Chichester, UK, pp 89-103.
Dekker T (2002) Influence of plume structure, composition and concentration on orientation behavior of mosquitoes
to host odors. PhD thesis, University of California, Riverside, USA.
Dekker T, Takken W, Knols BGJ, Bouman E, van de Laak S, de Bever A and Huisman PWT (1998) Selection of biting sites
on a human by Anopheles gambiae s.s., An. arabiensis and An. quadriannulatus. Entomol Exp Appl 87: 295-300.
Dekker T, Takken W and Cardé RT (2001) Structure of host-odour plumes influences catch of Anopheles gambiae s.s. and
Aedes aegypti in a dual-choice olfactometer. Physiol Entomol 26: 124-134.
Dekker T, Steib B, Cardé RT and Geier M (2002). L-Lactic acid: a human-signifying host cue for the anthropophilic
mosquito Anopheles gambiae. Med Vet Entomol 16: 91-98.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972.
Eiras AE and Jepson PC (1991) Host location by Aedes aegypti (Diptera: Culicidae): a wind tunnel study of chemical cues.
Bull Entomol Res 81: 151-160.
Eiras AE and Jepson PC (1994) Responses of Aedes aegypti (Diptera: Culicidae) to host odours and convection currents
using an olfactometer bioassay. Bull Entomol Res 84: 207-211.
Elkinton JS, Schal C, Ono T and Cardé RT (1987) Pheromone puff trajectory and upwind flight of male gypsy moths in
a forest. Physiol Entomol 12: 399-406.
Fry SN, Rohrseitz N, Straw AD and Dickinson MH (2009) Visual control of flight speed in Drosophila melanogaster. J Exp
Biol 212: 1120-1130.
Garrett-Jones C, Boreham PFL and Pant CP (1980) Feeding habits of anophelines (Diptera: Culicidae) in 1971-78, with
reference to the human blood index: a review. Bull Entomol Res 70: 165-185.
Geier M, Bosch OJ and Boeckh J (1999a) Influence of odour plume structure on upwind flight of mosquitoes towards
hosts. J Exp Biol 202: 1639-1648.
Geier M, Bosch OJ and Boeckh J (1999b) Ammonia as an attractive component of host odour for the yellow fever
mosquito, Aedes aegypti. Chem Senses 24: 1-7.
Gibson, G. (1983) The swarming behaviour of a mosquito. PhD thesis, University of Sussex, UK.
Gibson G (1995) A behavioural test of the sensitivity of a nocturnal mosquito, Anopheles gambiae, to dim white, red and
infra-red light. Physiol Entomol 20: 224-228.
Gibson G and Brady J (1985) ‘Anemotactic’ flight paths of tsetse flies in relation to host odour: a preliminary video study
in nature of the response to loss of odour. Physiol Entomol 10: 395-406.
Gibson G and Brady J (1988) Flight behaviour of tsetse flies in host odour plumes: the initial response to leaving or
entering odour. Physiol Entomol 13: 29-42.
Gibson G and Torr SJ (1999) Visual and olfactory responses of haematophagous Diptera to host stimuli. Med Vet Entomol
13: 2-23.
Gibson G and Young S (1991) The optics of tsetse fly eyes in relation to their behaviour and ecology. Physiol Entomol
16: 273-282.
Gibson G, Packer M, Steullet P and Brady J (1991) Orientation of tsetse flies to wind within and outside host odour
plumes in the field. Physiol Entomol 16: 47-56.
Gilbert C and Kuenen LPS (2008) Multimodal integration: visual cues help odor-seeking flies. Curr Biol 18: R295-R297.
Gillett JD (1979) Out for blood: flight orientation up-wind in absence of visual clues. Mosq News 39: 221-229.
Gillies MT (1980) The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review. Bull Entomol
Res 80: 525-532.

138  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

Gillies MT (1988) Anopheline mosquitoes: vector behaviour and bionomics. In: Wernsdorfer WH and Gregor IM (eds)
Malaria: principles and practice of malariology. Churchill and Livingstone, Edinburgh, UK, pp 453-485.
Gillies MT and Wilkes TJ (1969) A comparison of the range of attraction of animal baits and of carbon dioxide for some
West African mosquitoes. Bull Entomol Res 59: 441-456.
Gillies MT and Wilkes TJ (1970) The range of attraction of single baits for some West African mosquitoes. Bull Entomol
Res 60: 225-235.
Gillies MT and Wilkes TJ (1972) The range of attraction of animal baits and carbon dioxide for mosquitoes. Bull Entomol
Res 61: 389-404.
Gillies MT and Wilkes TJ (1974) Evidence for downwind flight by host-seeking mosquitoes. Nature 252: 388-389.
Gillies MT and Wilkes TJ (1978) The effect of high fences on the dispersal of some West African mosquitoes (Diptera:
Culicidae). Bull Entomol Res 68: 401-408.
Grant AJ and O’Connell RJ (1996) Electrophysiological responses from receptor neurons in mosquito maxillary palp
sensilla. In: Block GR and Cardew G (eds) Olfaction in mosquito-host interactions. Ciba Foundation Symposium
200, John Wiley and Sons, Chichester, UK, pp 233-248.
Hardie J, Gibson G and Wyatt TD (2001) Insect behaviours associated with resource finding. In: Woiwod IP, Reynolds
DR and Tomas CD (eds) Insect movement: mechanisms and consequences. CAB International, Wallingford, UK, pp
87-109.
Healy TP and Copeland MJW (1995) Activation of Anopheles gambiae mosquitoes by carbon dioxide and human breath.
Med Vet Entomol 9: 331-336.
Healy TP and Copeland MJW (2000) Human sweat and 2-oxopentanoic acid elict a landing response from Anopheles
gambiae. Med Vet Entomol 14: 195-200.
Healy TP, Copeland MJW, Cork A, Przyborowska A and Halket JM (2002) Landing responses of Anopheles gambiae elicited
by oxocarboxylic acids. Med Vet Entomol 16: 126-132.
Jones MDR and Gubbins SJ (1978) Changes in the circadian flight activity of the mosquito Anopheles gambiae in relation
to insemination, feeding and oviposition. Physiol Entomol 3: 213-220.
Jones MDR and Gubbins SJ (1979) Modification of female flight activity by a male accessory gland pheromone in the
mosquito Culex pipiens quinquefasciatus. Physiol Entomol 4, 345-351.
Justus KA, Murlis J, Jones C and Cardé RT (2002) Measurement of odor-plume structure in a wind tunnel using a
photoionization detector and a tracer gas. Environ Fluid Mech 2: 115-142.
Kemme JA, Van Essen PHA, Ritchie SA and Kay BH (1993) Response of mosquitoes to carbon dioxide and 1-octen-3-ol
in Southeast Queensland, Australia. J Am Mosq Control Assoc 9: 431-435.
Kennedy JS (1940) The visual responses of flying mosquitoes. Proc Zool Soc Lond Ser A 109, 221–242.
Kennedy JS (1977). Behaviorally discriminating assays of attractants and repellents. In: Shorey HH and McKelvey JJ, Jr
(eds) Chemical control of insect behavior. Wiley Interscience, New York, USA, pp 215-229.
Kennedy JS (1986) Some current issues in orientation to odour sources. In: Payne TL, Birch MC, and Kennedy CEJ (eds)
Mechanisms in insect olfaction. Clarendon Press, Oxford, UK, pp 11-25.
Kennedy JS and Marsh D (1974) Pheromone-regulated anemotaxis in flying moths. Science 184: 999-1001.
Knols, BGJ, Mboera LEG and Takken W (1998) Electric nets for studying odour-mediated host-seeking behaviour of
mosquitoes. Med Vet Entomol 12: 116-120.
Kuenen LPS and Baker TC (1983) A non-anemotactic mechanism used in pheromone source location by flying moths.
Physiol Entomol 8: 277-289.
Kuenen LPS and Cardé RT (1994) Strategies for contacting a lost pheromone plume: casting and upwind flight in the
male gypsy moth. Physiol Entomol 19: 15-29.
Land MF, Gibson G, Horwood J and Zeil J (1999) Fundamental differences in the optical structure of the eyes of nocturnal
and diurnal mosquitoes. J Comp Physiol [A] 185: 91-103.
Lindsay SW, Adiamah H, Miller JE, Pleass RJ and Armstrong JRM (1993) Variation in attractiveness of human-subjects to
malaria mosquitoes (Diptera: Culicidae) in The Gambia. J Med Entomol 30: 368-373.

Olfaction in vector-host interactions 139

Ring T. Cardé and Gabriella Gibson

Logan JG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue Luntz AJ and Pickett JA (2008) Identification
of human-derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. J Chem Ecol 34:
308-322.
Maciel-de-Freitas R, Souza-Santos R, Codeco, CT and Lourenco-de-Oliveira R (2010) Influence of the spatial distribution
of human hosts and large size containers on the dispersal of the mosquito Aedes aegypti within the first gonotrophic
cycle. Med Vet Entomol 24: 24:74-82.
Mafra-Neto A and Cardé RT (1994) Fine-scale structure of pheromone plumes modulates upwind orientation of flying
moths. Nature 369: 142-144.
Maimon G, Straw AD and Dickinson MH (2008) A simple vision-based algorithm for decision making in flying Drosophila.
Curr Biol 18: 463-470.
Mboera LEG and Takken W (1997) Carbon dioxide chemotrophism in mosquitoes (Diptera: Culicidae) and its potential
in vector surveillance and management programmes. Rev Med Vet Entomol 85: 355-368.
Mboera LEG, Knols BGJ, Takken W and Huisman PWT (1998) Olfactory responses of female Culex quinquefasciatus Say
(Diptera: Culicidae) in a dual-choice olfactometer. J. Vector Ecol 23: 107-113.
Mboera LEG, Takken W and Sambu EZ (2000) The response of Culex quinquefasciatus (Diptera: Culicidae) to traps baited
with carbon dioxide, 1-octen-3-ol, acetone, butyric acid and human foot odour in Tanzania. Bull Entomol Res 90:
155-159.
Muir LE, Kay BH and Thorne MJ (1992) Aedes aegypti (Diptera: Culicidae) vision: response to stimuli from the optical
environment. J Med Entomol 29: 455-450.
Murlis J and Jones CD (1981) Fine-scale structure of odour plumes in relation to distant pheromone and other attractant
sources. Physiol Entomol 6: 71-86.
Murlis J, Elkinton JS and Cardé RT (1992) Odor plumes and how insects use them. Annu Rev Entomol 37: 505-532.
Pates HV, Takken W, Stuke K and Curtis CF (2001) Differential behaviour of Anopheles gambiae sensu stricto (Diptera
Culicidae) to human and cow odours in the laboratory. Bull Entomol Res 91: 289-296.
Pates HV, Takken W and Curtis CF (2005) Laboratory studies on the olfactory behaviour of Anopheles quadriannulatus.
Entomol Exp Appl 114: 153-159.
Puri SN, Mendki MJ, Sukumaran D, Ganesan K, Prakash S and Sekhar K (2006) Electroantennogram and behavioral
responses of Culex quinquefasciatus (Diptera: Culicidae) females to chemicals found in human skin emanations. J
Med Entomol 43: 207-213.
Qiu YT, Spitzen J, Smallegange RC and Knols BGJ (2007a) Monitoring systems for adult insect pests and disease vectors.
In: Takken W and Knols BGJ (eds) Emerging pests and vector-borne diseases in Europe. Ecology and control of
vector-borne diseases, vol 1. Wageningen Academic Publishers, Wageningen, the Netherlands, pp 329-353.
Qiu YT, Smallegange RC, Ter BC, Spitzen J, Van Loon JJ, Jawara M, Milligan P, Galimard AM, Van Beek TA, Knols BG and
Takken W (2007b) Attractiveness of MM-X traps baited with human or synthetic odor to mosquitoes.(Diptera:
Culicidae) in The Gambia. J Med Entomol 44: 970-983.
Reeves WC (1953) Quantitative field studies on a carbon dioxide chemotrophism of mosquitoes. Am J Trop Med Hyg
2: 325-331.
Reiter P, Amador MA, Anderson RA and Clark GG (1995) Dispersal of Aedes aegypti in an urban area after blood-feeding
as demonstrated by rubidium-marked eggs. Am J Trop Med Hyg 52: 177-179.
Rowland M (1989) Changes in the circadian flight activity of the mosquito Anopheles stephensi associated with
insemination, blood-feeding, oviposition and nocturnal light intensity. Physiol Entomol 14: 77-84.
Rudolfs W (1922) Chemotrophism of mosquitoes. Bull NJ Agric Exp Stn 367: 4-23.
Sabelis MW and Schippers P (1984) Variable wind directions and anemotactic strategies of searching for an odour
plume. Oecologica 63: 225-228.
Scott TW, Morrison AC. Lorenz LH, Clark GG, Strickman D, Kittayapong P, Zhou H and Edman JD (2000) Longitudinal
studies of Aedes aegypti (Diptera: Culicidae) in Thailand and Puerto Rico: population dynamics. J. Med. Entomol
37: 77-88.
Shutt B, Stables L, Aboagye-Antwi F, Moran J and Tripet F (2010) Male accessory gland proteins induce female
monogamy in anopheline mosquitoes. Med Vet Entomol 24: 91-94.

140  Olfaction in vector-host interactions

 6. Host finding by female mosquitoes: mechanisms of orientation to host odours and other cues

Silver JB (2008) Mosquito Ecology, 3rd edn. Springer Science+Business Media B.V., Dordrecht, The Netherlands.
Smith A (1958) Outdoor cattle feeding and resting of A. gambiae Giles and A. pharoensis Theo. in the Pare-Taveta area
of East Africa. East African Med J 35: 559-567.
Snow WF (1987) Studies of house-entering habits of mosquitoes in The Gambia, West Africa: experiments with
prefabricated huts with varied wall apertures. Med Vet Entomol 1: 9-21.
Spitzen J, Smallegange RC and Takken W (2008) Effect of human odours and positioning of CO2 release point on trap
catches of the malaria mosquito Anopheles gambiae sensu stricto in an olfactometer. Physiol Entomol 33: 116-122.
Sutcliffe JF (1987) Distance orientation of biting flies to their hosts. Insect Sci Applic 8: 611-616
Takken W (1991) The role of olfaction in host-seeking of mosquitoes: a review. Insect Sci Applic 12: 287-295.
Takken W and Kline DL (1989) Carbon dioxide and 1-octen-3-ol as mosquito attractants. J Am Mosq Contr Assoc 5:
311-316.
Takken W, Charlwood DJ, Billingsley PF and Gort G (1998) Dispersal and survival of Anopheles funestus and A.gambiae
s.l. (Diptera: Culicidae) during the rainy season in southeast Tanzania. Bull Entomol Res 88: 561-566.
Takken W and Knols BGJ (1999) Odor-mediated behavior of Afrotropical malaria mosquitoes. Annu Rev Entomol 44:
131-157.
Takken W, Dekker T and Wijnholds YG (1997) Odor-mediated flight behavior of Anopheles gambiae Giles sensu stricto
and An. stephensi Liston in response to CO2, acetone, and 1-octen-3-ol. J Insect Behav 10: 395-407.
Tirados I, Costantini C, Gibson G and Torr SJ (2006) Blood-feeding behaviour of the malarial mosquito Anopheles
arabiensis: implications for vector control. Med Vet Entomol 20: 425-437.
Torr SJ (1988) The activation of resting tsetse flies (Glossina) in response to visual and olfactory stimuli in the field.
Physiol Entomol 13: 315-325.
Torr SJ (1994) The tsetse (Diptera: Glossinidae) story: implications for mosquitoes. J Am Mosq Contr Assoc 10: 258-265.
Torr SJ and Mangwiro TNC (1996) Upwind flight of tsetse (Glossina spp) in response to natural and synthetic host odour
in the field. Physiol Entomol 21: 143-150.
Torr SJ, Della Torre A, Calzetta M, Costantini C and Vale GA (2008) Towards a fuller understanding of mosquito
behaviour: use of electrocuting grids to compare the odour-orientated responses of Anopheles arabiensis and An,
quadriannulatus in the field. Med Vet Entomol 22: 93-108.
Trpis M and Hausermann W (1978) Genetics of house-entering behaviour in East African populations of Aedes aegypti
(L.) (Diptera: Culicidae) and its relevance to speciation. Bull Entomol Res 68: 521-532.
Vale GA (1982) The trap orientated behaviour of tsetse flies (Glossinidae) and other Diptera. Bull Entomol Res 72: 71-93.
Van Essen PHA, Kemme JA, Ritchie SA and Kay BH (1994) Differential responses of Aedes and Culex mosquitoes to
octenol or light in combination with carbon dioxide in Queensland, Australia. Med Vet Entomol 8: 63-67.
Vickers NJ and Baker TC (1994) Reiterative responses to single strands of odor promote sustained upwind flight and
odor source location by moths. PNAS 91: 5756-5760.
Warnes ML (1995) Field studies on the effects of cattle skin secretion on the behaviour of tsetse. Med Vet Entomol 9:
284-288.
Williams B (1994) Models of trap seeking by tsetse flies: anemotaxis, klinotaxis and edge detection. J Theor Biol 168:
105-115.
Willis MA and Cardé RT (1990) Pheromone-modulated optomotor response in male gypsy moths, Lymantria dispar L.:
upwind flight in a pheromone plume in different wind speeds. J Comp Physiol A 167: 699-706.
Willis MA, David CT, Murlis J and Cardé RT (1994) Effects of pheromone plume structure and visual stimuli on the
pheromone-modulated upwind flight of male gypsy moths (Lymantria dispar L.) in a forest. J Insect Behav 7: 385-409.
Wright RH and Kellogg FE (1962) Responses of Aedes aegypti to moist convection currents. Nature 194: 402-403.
Wyatt TD, Phillips ADG and Grégoire J-C (1993) Turbulence, trees and semiochemicals: wind-tunnel orientation of the
predator, Rhizophagus grandis to its barkbeetle prey, Dendroctonus micans. Physiol Entomol 18: 204-210.
Zanen PO, Sabelis MW, Buonaccorsi JP and Cardé RT (1994) Search strategies of fruit flies in steady and shifting winds
in the absence of food odours. Physiol Entomol 19: 335-341.
Zöllner GE, Torr SJ, Ammann C and Meixner FX (2004) Dispersion of carbon dioxide plumes in African woodland:
implication for host-finding by tsetse flies. Physiol Entomol 29: 381-394.

Olfaction in vector-host interactions 141

7. Host-seeking behaviour of mosquitoes: responses to olfactory
stimuli in the laboratory

Renate C. Smallegange and Willem Takken

Abstract

This chapter reviews the current understanding of odour-mediated host-seeking behaviour in

mosquitoes based on laboratory (indoor) studies. Most recent studies have focused on Aedes
aegypti L., Anopheles gambiae sensu stricto Giles and Culex quinquefasciatus Say, as these species
are all strongly associated with human biting and disease transmission. Where relevant, reference
to other mosquito species is provided. Host seeking in mosquitoes is mediated primarily by
chemical cues, and therefore most studies focus on the response to host odours and identification
of specific compounds or odour blends. Laboratory tools for the study of odour-mediated
behaviour include Y-tube and dual-choice (or dual-port) olfactometers, wind tunnels and room-
size arenas in which mosquitoes can fly freely. Mosquitoes are observed individually or in groups,
usually at a specified age or range of ages, at a pre-determined time of the day when they are
considered to be naturally responsive to host cues. Aedes aegypti and An. gambiae respond
strongly to natural human skin emanations, whereas Cx. quinquefasciatus shows variation in this
behaviour dependent on its geographic origin. Carbon dioxide acts as a principal stimulus for
each of these mosquito species although with a variable species-specific effect. Several host-
derived compounds have been identified that cause behavioural responses in these mosquito
species, including ammonia, L(+)-lactic acid and aliphatic carboxylic acids. In addition, acetone
and dimethyl disulfide enhance the attraction of blends of these compounds to Ae. aegypti. Some
other ketones, such as 6-methyl-5-hepten-2-one and geranyl acetone disrupt the host-seeking
process of Ae. aegypti and An. gambiae. Several aldehydes, such as octanal and nonanal, are
attractive to Cx. quinquefasciatus, and reduce upwind flight and the total number of landings in
Ae. aegypti. Based on these data odour blends have been identified that are attractive to these
mosquito species, albeit less so than natural human odour, indicating that additional, hitherto
unidentified, semiochemicals are involved in host-seeking behaviour that are likely to improve the
efficacy of blends. Behavioural research combined with studies on molecular olfaction is ongoing
to discover compounds that affect the host-seeking process, and further research under semi-field
and field conditions is required to explore the effectiveness of putative attractants and repellents
in natural circumstances.

Keywords: bioassay, host-odours, kairomone, odour baits, olfaction, repellent, synthetic blend,
vector-borne disease

Introduction

Some species of mosquitoes are vectors of pathogens that give rise to several important infectious
diseases of humans and animals. Several of these diseases thrive in tropical climates, where the
abundance of vectors, aided by favourable environmental conditions, allows for high levels of
transmission. Approximately half of the world’s population, especially in less developed countries,
are at risk (Jones et al. 2008, WHO 2008, 2009). Although new progress has recently been made
in the prevention and control of vector-borne diseases, current methods appear insufficient
to prevent many cases of illnesses and deaths due to various factors (e.g. societal, economical,
political, mosquito and parasite resistance) (Greenwood et al. 2008, Morrison et al. 2008, Takken

Olfaction in vector-host interactions 143

Renate C. Smallegange and Willem Takken

and Knols 2009). In addition, the recent expansion of vector-borne diseases into more temperate,
industrialised countries, such as the occurrence of West Nile virus and Chikungunya in the USA
and Europe, respectively (Gould 2003, Gould and Fikrig 2004, Takken and Knols 2007, Vazeille et
al. 2008), demonstrate the need for further development and implementation of measures to
protect people at risk (e.g. with bednets, vaccines) and to control mosquito vectors (e.g. with
insecticides, biological control agents, treated surface and materials, the sterile insect technique,
monitoring tools).

In tropical areas, An. gambiae sensu stricto (hereafter termed An. gambiae), Ae. aegypti and Cx.
quinquefasciatus are medically important mosquito vectors of infectious pathogens such as malaria,
dengue and bancroftian filariasis, respectively (Cook 1996). The females of these mosquito species
are considered to be anthropophilic (Takken and Knols 1999) and often live in close association
with humans. Since host seeking by mosquitoes is mainly mediated by olfactory stimuli (Takken
1991, Takken and Knols 1999), it is likely that the odour complex of the blood-host contains host-
specific and non-host-specific chemicals to which these mosquito species respond (Gillies 1988).
It is important to note that the semiochemicals involved in the attraction of mosquitoes to hosts
may include both repellents and attractants, and the effects of some chemicals are likely to be
dose-dependent. Once these semiochemicals have been identified, it is expected that they can be
incorporated into technologies that can be used to trap, kill or repel mosquitoes (Cook et al. 2007,
Day and Sjogren 1994, Hassanali et al. 2008, Kline 2006, 2007, Logan and Birkett 2007, Takken and
Knols 2009) as has been successfully demonstrated for tsetse flies (see Chapter 12, this volume;
also see Chapter 17, this volume). Ultimately, some of these products may improve or even replace
current control technologies or sampling tools in epidemiological studies of mosquito-borne
diseases. Improvement of present methods could be achieved by adding an odour bait to an
existing mosquito trap, by replacing a bait with a better performing one, or by combining a newly
developed odour trap with current mosquito control tools.

Crumb (1922) was the first to describe the role of host-derived semiochemicals in mosquito
behaviour. In addition, Rudolfs (1922) and Van Thiel (1937) discovered the unique and important
role of carbon dioxide (CO2) (see review of Mboera and Takken 1997). Since then many mosquito
attractants have been discovered, but there is little evidence to explain how these compounds act
as olfactory mediators of behaviour and none are as attractive as natural odour blends (Bernier et
al. 2007, Bosch et al. 2000, Bowen 1991, Okumu et al. 2010, Spitzen et al. 2008, Smallegange et al.
in press, Takken and Knols 1999). Only a few odorants are used as baits for mosquito surveillance
or control, in contrast to the extensive use of host-derived baits for other insects of medical or
veterinary importance (e.g. tsetse flies, screwworm, biting midges, triatomine bugs) (Cork and Hall
2007, Cruz-Lopez et al. 2001, Logan and Birkett 2007, Qiu et al. 2007b, Vale 1993, see also Chapters
10 and 12 of this volume).

Since the review of Takken and Knols (1999), which summarised our knowledge on the role of
semiochemicals in the life history of African malaria mosquitoes at the close of the last century,
new techniques and relevant information have become available. In this chapter, we describe
recent results of laboratory studies on the identification of semiochemicals that play a role in
the host-seeking behaviour of mosquitoes, with emphasis on An. gambiae, Ae. aegypti and Cx.
quinquefasciatus, the species that have been studied in most detail over the last decade. It is not
the purpose of this chapter to go into details of each step involved in the host-seeking process
(Takken 1996, see also Chapter 1 of this volume) or of the strategies employed by mosquitoes
to find distant odour sources (the latter has recently been reviewed by Cardé and Willis (2008)
for insects in general). Instead, we will provide an overview of the individual chemicals and

144  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

synthetic odour blends that have been tested for their possible effect on the behaviour of
host-seeking mosquitoes. We also briefly describe methods and bioassays deployed to identify
candidate semiochemicals. Subsequently, we will discuss problems and knowledge gaps that are
encountered in this research area. Studies on anopheline host-seeking behaviour performed in
semi-field and field situations are discussed in Chapter 8 (this volume).

Identification of candidate semiochemicals

Anopheles gambiae and Ae. aegypti are considered to be highly anthropophilic mosquito species,
i.e. the females express a preference for biting humans to get their blood meal (Gillies 1964,
Harrington 2001, Takken and Knols 1999, White 1974). Several hundred volatile compounds
emitted by humans have been identified, especially in the last twenty years (Bernier et al. 1999,
2000, 2002, Cork and Park 1996, Curran et al. 2005, 2007, Deng et al. 2004, Ellin et al. 1974,
Gallagher et al. 2008, Hasegawa et al. 2004, Haze et al. 2001, Healy and Copland 2000, Healy et al.
2002, Krotoszynski et al. 1997, Meijerink et al. 2000, Natsch et al. 2006, Penn et al. 2007, Perry et
al. 1970, Philips 1997, Zeng et al. 1991, 1996). It is assumed that the host-seeking behaviour of
mosquitoes is mediated by a blend of specific compounds, and not by the complete assemblage
of compounds present in host odour (Zwiebel and Takken 2004). Theoretically, knowledge about
the compounds present in the human odour profile can be used to test which compounds play
a role in the host-seeking behaviour of anthropophilic mosquito species. However, testing the
behavioural effect of each compound in different concentrations and in all possible combinations
is not practical because of the large number of compounds present in human emanations (Bernier
et al. 2000, Curran et al. 2005, Penn et al. 2007, Philips 1997). Here we outline some techniques
that can be used to narrow down the selection of candidate compounds (Logan and Birkett 2007,
Smallegange et al. 2003, Van der Goes van Naters and Carlson 2006).

It has been shown that some persons are bitten by mosquitoes more often than others (Curtis
1986, Lindsay et al. 1993, Scott et al. 2006). This may be the result of a differential attractiveness of
the body odours of human individuals to mosquitoes (Bernier et al. 2002, Knols et al. 1995, Logan
et al. 2008, Lindsay et al. 1993, Mukabana et al. 2002, Qiu et al. 2004a, 2006a, Schreck et al. 1990).
Comparing the odour profiles of highly and poorly attractive individuals can reveal compounds
that play a role in host selection. Gas-chromatography (GC) and mass-spectrometry (MS) are
used to study the differences in the odour profiles of individuals and to identify the compounds
involved (Bernier et al. 1999, 2000, 2002, Logan et al. 2008). These differences may be qualitative
and/or quantitative, which makes it difficult to identify the compounds involved in host selection.
The discovery that human skin emanations collected on glass beads lose their attractiveness
after a certain period of time suggests that volatile compounds decreasing in relative abundance
during this period play a role in the attractiveness of the emanations (Bernier et al. 2000, 2002, Qiu
et al. 2004a, Schreck et al. 1981, Smallegange et al. 2003). Comparing odour profiles of differential
attractive individuals resulted in weak attractants and putative repellents for Ae. aegypti (Bernier
et al. 2002, Logan et al. 2008).

Skin microbiota plays a role in the production of human body odour (Leyden et al. 1981, Shelley et
al. 1953). Blood agar incubated with microbiota from human feet or Staphylococcus epidermidis, a
bacteria species commonly found on human skin, is attractive to An. gambiae. Identifying volatile
compounds produced by microorganisms present on human skin revealed 10 putative attractants
for this mosquito species (Verhulst et al. 2009).

Olfaction in vector-host interactions 145

Renate C. Smallegange and Willem Takken

Although not a new technology we should add here that, in addition to identifying volatile
organic compounds emanating from the human skin and comparing odour profiles of individuals,
electrophysiological methods, such as electroantennography (EAG) and single sensillum recording
(SSR) techniques, either without or in combination with GC and MS, can be applied to identify
which compounds in the human odour profile stimulate the olfactory sensilla on mosquito
antennae or maxillary palps (see Chapter 3 in this volume for more details, as well as Logan et al.
2008, Lu et al. 2007, Qiu et al. 2004b). Ultimately, the effects of odours on mosquitoes must be
investigated directly, with behavioural assays. The sensory systems of a mosquito may be able to
detect a volatile compound, but this does not necessarily mean that a motor response is triggered,
or that the response is necessarily related to the behaviour under investigation. Behavioural
observations are required to establish the mechanisms by which odorants guide mosquitoes to
a host, e.g. they may act as attractants, repellents, activators or arrestants (Costantini et al. 2001,
Logan et al. 2008, Mukabana et al. 2004).

The publication of the full gene sequence of An. gambiae (Holt et al. 2002) and Ae. aegypti
(Nene et al. 2007) has enabled rapid progress in molecular genetics. The sequencing of the Cx.
quinquefasciatus genome has almost been completed (Pelletier and Leal 2009, Waterhouse et al.
2008). A large family of olfactory receptor genes (ORs) and gustatory receptor genes (GRs) has
been identified (Van der Goes van Naters and Carlson 2006). These genes play a crucial role in the
binding of olfactory molecules to neural membranes (see Chapter 2, this volume). The molecular
regulation of olfaction appears highly conserved in the animal kingdom, although the OR and GR
families are highly specific to species (Robertson and Wanner 2006). This would explain why some
insects express a highly selective host preference, as found in An. gambiae and Ae. aegypti. Work
on the molecular regulation of olfaction in Drosophila and Xenopus has elucidated the processes
by which small molecules control physiological activity in insects (Hallem et al. 2006, Rutzler and
Zwiebel 2005). This knowledge is currently being used for the identification of semiochemicals
to which mosquitoes respond (Carey et al. 2010, Hallem et al. 2004, Justice et al. 2003, Leal et al.
2008, Wang et al. in press and Chapter 2, this volume).

Combined with powerful electrophysiological techniques (Chapter 3, this volume), molecular

genetics can provide new insights into the olfactory behaviour of mosquitoes and can be used
as a tool for the rapid identification of candidate semiochemicals. Behavioural studies, however,
remain essential for the final assessment of the role of these chemicals in the insect’s life-history.
The olfactory receptor neurons may be able to detect certain chemical stimuli, the brain will
associate them with a specific behavioural response (Klowden 1995, see also De Bruyne and Baker
2008).

Similar strategies are used to discover olfactory cues that affect mosquitoes with more variable
host preferences, such as Cx. quinquefasciatus. The host-preference of Cx. quinquefasciatus appears
to be region-dependent, as the following examples illustrate. Field experiments in Tanzania
showed that more Cx. quinquefasciatus were caught in human-odour baited traps than in traps
baited with goat or calf odour (Mboera and Takken 1999). By contrast, blood meal analyses have
shown that in Kenya this species has a preference for livestock over human hosts (Muturi et al.
2008). In a study in Tucson, USA, blood meal analysis revealed that 50% of Cx. quinquefasciatus
females fed on humans and 32% on birds, respectively (and therefore this species is considered
to be a potential West Nile virus vector; Zinser et al. 2004), whereas in Tenessee, USA, and Mexico
they were found to feed less on humans than on birds (Eilizondo-Quiroga et al. 2006, Savage et al.
2007). In addition, a greater percentage of Cx. quinquefasciatus females, from a colony originating
in Florida, USA, responded to chicken than to human odour in an olfactometer study (Allan et al.

146  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

2006b). These latter results indicate that in the USA Cx. quinquefasciatus is less anthropophilic and
more ornithophilic than in Tanzania. Similarly, in Senegal and Madagascar An. gambiae populations
were found to feed predominantly on cattle (Diatta et al. 1998, Duchemin et al. 2001). In addition,
Ae. aegypti responds not only to human odour, but also to mouse and chicken odour (Allan et al.
2006b, McCall et al. 1996, McIver 1968), suggesting that this species is actually an opportunistic
feeder. The reliability of results obtained in host preference studies, however, depends on the
method used (field and laboratory), the geographic area covered (field) or geographic origin
of the mosquitoes (laboratory) and the availability of host species (field and laboratory), which
demands for caution in using labels such as ‘anthropophilic’ and to confine research to analysing
only human odours (Kilpatrick et al. 2006, Lefèvre et al. 2009, Torr et al. 2008).

A wider host-range broadens the search for volatile organic compounds mediating host-seeking
behaviour. Overlap in odour profiles of different blood-host species of a mosquito species, such
as human and chicken, may reveal attractive compounds such as nonanal (Syed and Leal 2009).
However, differences in these odour profiles may explain differences in relative attraction to these
hosts due to allomones or additional kairomones (Dicke and Sabelis 1988) emitted by the host.

Compounds present in bovine emanations have been thoroughly examined in relation to the
development of surveillance and control tools for tsetse, cattle and stable flies using the techniques
described above and have been published (e.g. Birkett et al. 2004, Bursell et al. 1988, Schofield
and Brady 1997, Schofield et al. 1995, 1997, Spinhirne et al. 2003, 2004, Torr et al. 1995, Vale 1980,
Vale and Hall 1985a,b, Vale et al. 1988 and Chapter 12, this volume). Also, several compounds
emitted by chickens have been identified, for example by solvent extraction and volatile trapping
of faeces, feathers, feet or skin, and include alcohols, aldehydes, carboxylic acids, diones and
ketones (Bernier et al. 2008, Cooperband et al. 2008, Haathi and Fales 1967, Williams et al. 2003).

Although in nature mosquitoes do not feed on blood directly, females of both Cx. quinquefasciatus
and Ae. aegypti were found to be attracted to the odour of bovine and avian blood. Culex females
landed more on membranes filled with avian than with bovine blood, whereas Aedes mosquitoes
showed no landing preference (Allan et al. 2006a). The volatile chemicals emanating from blood
have been identified (Ashley et al. 1992, Issachar et al. 1982, Sastry et al. 1980) but few of these
have been tested for their effects on mosquitoes in the laboratory, although components found in
exhaled breath and skin emanations may be similar to those present in blood (Sastry et al. 1980).

An initial screening of compounds for their effect on the host-seeking behaviour of mosquitoes in
a laboratory before testing them under field conditions has the advantage that the environmental
circumstances during the bioassays and physiological state (e.g. age, time since sugar or bloodmeal,
stage in gonotrophic cycle) of the mosquitoes can be controlled during the experiments. Before
the results of a bioassay are put into practical use, they should first be validated in semi-field
and/or field experiments (Knols et al. 2002, see also Chapter 8 in this volume). (Semi-)field studies
provide information about the effects of natural temperature, humidity, wind, the presence of
competitive odours, etc. on the responses to odours, and can be used to test the effectiveness of
an odour bait over larger distances than can be tested in the laboratory (Njiru et al. 2006, Okumu
et al. 2010, Williams et al. 2006a). Field evaluations can also reveal simultaneously whether other
vectors are attracted or repelled by the odour compounds tested (Jawara et al. 2009, Qiu et al.
2007a).

Olfaction in vector-host interactions 147

Renate C. Smallegange and Willem Takken

Bioassay techniques in the laboratory

In the laboratory, host-seeking behaviour in response to odour can be evaluated by observing

individual insects. This can be done by direct observation (human landing catches) in a cage (De
Jong and Knols 1995, Dekker et al. 1998, Knols et al. 1994a) or wind tunnel (Takken et al. 1997) or
by video recording (Beeuwkes et al. 2008, Cooperband and Cardé 2006, Dekker et al. 2005, Healy
and Copland 1995, Young et al. 1993).

Y-tube and dual-port (or dual-choice) olfactometers are used to assess the response of mosquitoes
to a choice between an odour and clean air or two different odours (Geier and Boeckh 1999,
Knols et al. 1994b, Posey et al. 1998). Y-tube olfactometers (Figure 1A) are conveniently small
devices for rapid screening of odours. Aedes aegypti in particular responds well in Y-tubes (Geier
and Boeckh 1999, Logan et al. 2008). Dual-port olfactometers (Figure 1B) are more suitable for
mosquito species that perform better in larger arenas, such as An. gambiae (Knols et al. 1994b).

Mosquito responses to odours can also be studied in large indoor cages, for instance inside a
bednet or a closed room. These tools have been used for studies on the selection of biting sites
on volunteers (De Jong and Knols 1995, Dekker et al. 1998, Knols et al. 1994a) and for studies with
insect repellents (Curtis and Hill 1988). By placing mosquito traps in indoor cages (Figure 1C), the
effect of different odorants on trapping efficacy can be investigated (Silva et al. 2005, Smallegange
et al. 2009).

Several methods have been developed for testing putative repellents. The most commonly used
methods to test putative repellents for employment directly on the human skin have recently
been reviewed by Barnard et al. (2007). Various types of olfactometers have been used to test
compounds for their efficacy as spatial repellents (Bernier et al. 2005, Dogan and Rossignol 1999,
Grieco et al. 2005, Kline et al. 2003, Logan et al. 2008, Waka et al. 2006).

Laboratory-based studies have been done with individual mosquitoes or with groups of up to
50 mosquitoes per test. In wind tunnels, experiments are usually done with one insect at a time
(Healy and Copland 1995, Takken et al. 1997), whereas in olfactometers most studies are based on
the results of groups of 15 to 50 insects. Because of the possibility of learning (Alonso and Schuck-
Paim 2006, McCall and Eaton 2001, Tomberlin et al. 2006), individual mosquitoes should not be
tested more than once, and each test should be performed with naïve insects.

Effects of physiology on behaviour

It should be taken into account that the host-seeking process of mosquitoes includes different
phases. After an inactive period during which reproductive maturation (Lehane 1991) and
development of the antennal sensitivity takes place (Davis 1984), female activity increases
(influenced by circadian rhythm; see Klowden 1996). This increase makes it more likely for the
females to encounter host-derived stimuli. Activation takes place as soon as these stimuli have
been detected. This phase is followed by the orientation process that will bring the mosquitoes
in the immediate vicinity of the host. As soon as the host has been located, the female will land
and start blood feeding (Takken 1996). The mosquitoes will respond to different stimuli, among
which olfactory, within each phase. Therefore, the choice for a certain bioassay may influence
the discovery of semiochemicals that play a role in a specific phase of the mosquito host-seeking
process.

148  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

B Pressurised air

Active
charcoal
filter
Release
Flight chamber cage
60 cm
Warm
water
Trapping devices

60 cm
160 cm
Zips
C
233 cm

Release cage 250 cm

330 cm
Figure 1. Schematic drawings of (A) an Y-tube olfactometer (from Geier et al. 1999a), (B) a dual-port olfactometer
(modified after Spitzen et al. 2008) and (C) two MM-X traps in an indoor netted cage.

Olfaction in vector-host interactions 149

Renate C. Smallegange and Willem Takken

Olfactory cues are important external factors affecting host-seeking behaviour from a distance,
whereas visual and physical cues, such as heat and moisture, play a role in close vicinity of
the blood-host (Takken and Knols 1999). However, even when host-stimuli are present, host-
seeking behaviour may not be expressed as a result of endogenous, physiological factors. These
physiological factors include the circadian rhythm, age, nutritional state, mating condition, and
presence of eggs, and should be considered in any study on insect behaviour (Klowden 1995,
1996).

Aedes aegypti is a diurnal mosquito and its host-seeking behaviour should therefore be studied
during daytime under illuminated conditions (Geier and Boeckh 1999, Klowden 1995). Both An.
gambiae and Cx. quinquefasciatus are nocturnal. Biting activity of the latter species was observed
throughout the night (Chandra 1995), allowing olfactometer experiments during the scotophase
(Mboera et al. 1998). Puri et al. (2006), however, observed that the mosquitoes were most active
during a period of one hour after dusk, and therefore performed experiments in dim light during
this short period of time only. Experiments examining the host-seeking behaviour of An. gambiae
are usually performed, in the dark or under moonlight conditions (Knols et al. 1994b), during the
last 4 hours of the scotophase, when this species has frequently been reported to be naturally
active (Haddow and Ssenkubuge 1973, Killeen et al. 2006, Maxwell et al. 1998). Shifts in timing
of host-seeking activity, however, have occasionally been observed, chiefly because of vector-
control interventions (Braimah et al. 2005).

Host-seeking behaviour of Ae. aegypti increases from no response to host cues directly after
adult emergence to more than 90% of the females responding to host-odours 102 hours post
emergence (Davis 1984). Therefore, the best results can be obtained with females of 4-days-old
or older. Generally, adults are tested within a specified range of ages, which varies overall from
3-10 days, to control for the effects of aging.

A blood meal changes the mosquito’s physiology and behaviour, by affecting the sensitivity of
the olfactory receptor organs (Davis 1984, Qiu et al. 2006b, see also Chapter 3 in this volume).
After a full blood meal, An. gambiae is no longer attracted to host stimuli for 24 hours, with host-
seeking behaviour returning to pre-blood meal levels 72 hours post feeding (Takken et al. 2001).
In mated Ae. aegypti, the inhibition begins 30 hours after ingestion of the blood and reaches its
maximum between 36 and 72 hours post feeding. The inhibition is less pronounced in unmated
Ae. aegypti females (Klowden and Lea 1979). Nutritional status also affects the intensity of the
inhibition (Klowden 1986). Test insects may be starved for up to 18 hours prior to testing. This can
be done by replacing the carbohydrate food source with water before the experiment. Infection
with pathogens/parasites may alter the host-seeking behaviour as well (Hurd 2003, Schaub 2006).

Bioassay results with semiochemicals in laboratory studies

In this section an overview is given of semiochemicals identified to play a role in the host-seeking
process of Ae. aegypti, An. gambiae, Cx. quinquefasciatus, and some other mosquito species.
The compounds, the bioassay in which they were tested and the effect that has been found are
summarised in Tables 1-3. As some bioassays cannot differentiate between an inhibitory and a
repellent effect, both terms are used interchangeably. Although, as Kennedy (1978) pointed out,
this term does not reveal the many processes and underlying mechanisms involved, we use the
term ‘attractant’ to refer to a chemical or blend that has been found to induce mosquitoes to
orientate towards the odour source.

150  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Aedes aegypti

A human hand as the odour source stimulates Ae. aegypti females to fly upwind in a Y-tube
olfactometer. A similar result was achieved by using high doses of heated or unheated ethanol
skin extracts (Geier and Boeckh 1999, Geier et al. 1996). Liquid chromatography of the skin extract
revealed active fractions of which one contained lactic acid and another probably ammonia (Geier
et al. 1996, 1999a).

In humans, L-(+)-lactic acid (hereafter lactic acid) is present in breath and on the skin. Human skin
rubbing extracts contain lactic acid in significantly higher quantities than skin extracts of 12 other
mammals and chickens (Dekker et al. 2002) and have, presumably for that reason, received specific
attention as a kairomone for anthropophilic mosquitoes. More interestingly, lactic acid was found
in higher amounts on glass beads that were handled by individuals that were more attractive to
Ae. aegypti than individuals with lower amounts of the compound in their skin emanations. From
numerous studies it was shown that lactic acid plays a major role in the host-seeking behaviour of
Ae. aegypti (Table 1) (Acree et al. 1968, Bernier et al. 2002, Eiras and Jepson 1991, Geier et al. 1996,
Smith et al. 1970). Lactic acid is a weak attractant on its own but acts synergistically with carbon
dioxide (CO2), a major component of exhaled breath, and other volatiles emanating from human
skin (Acree et al. 1968, Allan et al. 2006a, Bernier et al. 2002, Bosch et al. 2000, Eiras and Jepson
1991, Geier and Boeckh 1999, Geier et al. 1996, 1999a,b, Smith et al. 1970, Williams et al. 2006a). An
ethanol extract of human skin was no longer attractive to Ae. aegypti females, presumably because
the lactic acid had been removed from the extract (Geier et al. 1996).

Carbon dioxide on its own is an activator and weak attractant for Ae. aegypti, increasing take-off,
flight velocity and flight duration, although homogeneous CO2 plumes decrease upwind flight
and trap catches (Allan et al. 2006b, Bernier et al. 2007, Bosch et al. 2000, Dekker et al. 2001b, 2005,
Eiras and Jepson 1991, Geier and Boeckh 1999, Geier et al. 1999a,b, Kline et al. 2006). An encounter
with a single CO2 filament increases the responsiveness to human skin odours (Dekker et al. 2005).

The presence of carboxylic acids on the skin is unique to humans compared to other mammals,
birds and rodents (Nicolaides et al. 1968). Saturated aliphatic carboxylic acids with chain length
C1-C3, C5-C8, and C13-C18 were found to increase the attractiveness of lactic acid at certain
doses to Ae. aegypti females. The attractiveness increased even further when a second carboxylic
acid was added, as long as one is a short chain (C1-3) and the other a medium length (C5-8)
carboxylic acid. Nonanoic and undecanoic acid reduced the effect of lactic acid, whereas butanoic,
decanoic and dodecanoic acid had no effect (Bosch et al. 2000). Acetic acid on its own was found
to be attractive to host-seeking Ae. aegypti females and to elicit a landing response. Several other
carboxylic acids, including two aromatics, induced landing responses (Table 1) (Allan et al. 2006a).
In contrast, hexanoic acid reduced the number of landings (Douglas et al. 2005), probably due
to the high doses used. Propanoic, butanoic, 3-methylbutanoic, pentanoic, heptanoic, decanoic,
dodecanoic, tetradecanoic, pentadecanoic, hexadecanoic, heptadecanoic, and octadecanoic acid
did not attract Ae. aegypti females, even though some of these acids were present at higher levels
in the skin emanations of highly attractive persons (Allan et al. 2006a, Bernier et al. 2002, Bosch
et al. 2000). Propanoic and pentanoic acid did not influence the attractiveness of CO2 (Bosch et
al. 2000).

Ammonia was also found to enhance the attractiveness of lactic acid, although it is not attractive
on its own or in combination with CO2 (Geier et al. 1999a). Propanoic and pentanoic acid enhanced
the effect of a lactic acid and ammonia blend. A blend of ammonia, lactic acid, and two carboxylic

Olfaction in vector-host interactions 151

Renate C. Smallegange and Willem Takken

Table 1. Overview of chemical compounds that have been tested for their impact on the host-seeking behaviour
of Aedes aegypti using behavioural assays in the laboratory. (Details about concentrations of compounds are
omitted).

Compound Response Behavioural assay Reference

carbon dioxide (CO2) (Weakly) attractive on its own; Wind tunnel Eiras and Jepson (1991)
flight activating effect; synergism Wind tunnel tube Geier et al. (1999a)
with lactic acid but not ammonia Y-tube Geier et al. (1999b)
Y-tube Geier and Boeckh (1999)
Y-tube Bosch et al. (2000)
Dual-port olfactometer Allan et al. (2006b)
Dual-port olfactometer Kline et al. (2006)
Dual-port olfactometer Bernier et al. (2007)
Homogeneous CO2 reduced trap Dual-port olfactometer Dekker et al. (2001b)
catch
Sensitizes to skin odours Wind tunnel (3D) Dekker et al. (2005)
ammonia (NH3) Not attractive on its own Y-tube Geier et al. (1999b)
Enhanced the effect of lactic acid Y-tube Williams et al. (2006a)
NH3+hexanoic acid improved
blend of lactic acid + acetone +
dimethyl disulfide
Alcohols
4-hexen-1-ol Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
1-hepten-3-ol Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
glycerol Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
Aldehydes
hexanal Reduced the number of landings Landing assay Douglas et al. (2005)
heptanal Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
octanal Reduced the number of landings Landing assay Douglas et al. (2005)
Reduced upwind flight Y-tube Logan et al. (2008)
nonanal Not attractive Dual-port olfactometer Bernier et al. (2002)
Reduction of upwind flight Y-tube Logan et al. (2008)
decanal Reduced upwind flight Y-tube Logan et al. (2008)
z-4-decanal Reduced the number of landings Landing assay Douglas et al. (2005)
Aliphatic carboxylic acids
lactic acid (LA) Weakly attractive on its own; Dual-port olfactometer Acree et al. (1968)
Synergism with CO2 and other skin Dual-port olfactometer Smith et al. (1970)
components Wind tunnel Eiras and Jepson (1991)
Y-tube Geier et al. (1996)
Y-tube Geier and Boeck (1999)
Wind tunnel tube Geier et al. (1999a)
Y-tube Geier et al. (1999b)
Y-tube Bosch et al. (2000)
Dual-port olfactometer Bernier et al. (2002)
Y-tube Williams et al. (2006a)
Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
methanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)

152  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Table 1. Continued

Compound Response Behavioural assay Reference

acetic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)

Attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
propanoic acid Not attractive on its own Y-tube Bosch et al. (2000)
Dual-port olfactometer Allan et al. (2006a)
Enhanced the effect of LA Y-tube Bosch et al. (2000)
Enhanced the effect of LA + NH3 Y-tube Bosch et al. (2000)
No landing response Landing assay Allan et al. (2006a)
butanoic acid No effect on attractiveness of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
3-methyl butanoic acid Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
pentanoic acid Not attractive on its own Y-tube Bosch et al. (2000)
Enhanced the effect of LA
Enhanced the effect of LA + NH3
hexanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Attractive when combined with Y-tube Williams et al. (2006a)
LA + NH3
Hexanoic acid + NH3 improved Y-tube Williams et al. (2006a)
blend of lactic acid + acetone +
dimethyl disulfide
Reduced number of landings Landing assay Douglas et al. (2005)
heptanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
octanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
nonanoic acid Reduced the attractive effect of LA Y-tube Bosch et al. (2000)
decanoic acid No effect on the attractiveness of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
undecanoic acid Reduced the attractive effect of LA Y-tube Bosch et al. (2000)
dodecanoic acid No effect on the attractiveness of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
tridecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
tetradecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
pentadecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
hexadecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)

Olfaction in vector-host interactions 153

Renate C. Smallegange and Willem Takken

Table 1. Continued

Compound Response Behavioural assay Reference

heptadecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)

Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
octadecanoic acid Enhanced the effect of LA Y-tube Bosch et al. (2000)
Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
Blend of lactic acid, Attractive Y-tube Bosch et al. (2000)
ammonia and two
carboxylic acids (C1-
C3 and C5-C8)
Blend of octanal, Reduced number of landings Landing assay Douglas et al. (2005)
hexanal, Z-4 decenal,
decanal, hexanoic
and octanoic acid
Aromatic carboxylic acids
benzoic acid Induced landing response Landing assay Allan et al. (2006a)
2-hydroxybenzoic acid Induced landing response Landing assay Allan et al. (2006a)
Ketones
acetone Attractive on its own; Enhanced Dual-port olfactometer Bernier et al. (2003)
the effect of LA Dual-port olfactometer Bernier et al. (2007)
Improved blend of LA + NH3 + Y-tube Williams et al. (2006a)
hexanoic acid
butanone Weakly attractive Dual-port olfactometer Bernier et al. (2002)
2-pentanone Weakly attractive Dual-port olfactometer Bernier et al. (2002)
3-pentanone Weakly attractive Dual-port olfactometer Bernier et al. (2002)
6-methyl-5-hepten- Weakly attractive Dual-port olfactometer Bernier et al. (2002)
2-one Reduced upwind flight Y-tube olfactometer Logan et al. (2008)
2-decanone Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
geranyl acetone Reduced upwind flight Y-tube olfactometer Logan et al. (2008)
Sulfides
methyl sulfide Attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
carbon disulfide Attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
dimethyl disulfide Attractive on its own Dual-port olfactometer Bernier et al. (2003, 2007)
Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
Enhanced the effect of LA Dual-port olfactometer Bernier et al. (2003, 2007)
Enhanced the effect of lactic acid Lab. trapping exp’s Silva et al. (2005)
+ acetone Y-tube Williams et al. (2006a)
Dual-port olfactometer Bernier et al. (2007)
ethyl disulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)

154  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Table 1. Continued

Compound Response Behavioural assay Reference

methyl propyl disulfide Not attractive Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
dimethyl trisulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
Miscellaneous
dichloromethane Attractive on its own Dual-port olfactometer Bernier et al. (2003)
Enhanced the effect of LA
pyridine Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
4-pyridinamine Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
benzene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
toluene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
heptane Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
1H-indole Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
2-octene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
styrene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
2-nonene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
1,3-butanediamine Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
pentacosane Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
cholesterol Not attractive on its own Dual-port olfactometer Bernier et al. (2002)
squalene Not attractive on its own Dual-port olfactometer Bernier et al. (2002)

acids from the C1-C3 and C5-C8 groups was almost as attractive as an extract of human skin
residues (attracting 68.1% and 77.9% of the tested mosquitoes, respectively). The difference in
attractiveness may have been caused by ‘missing’ compounds in the synthetic blend or because
the concentrations and proportions of the compounds in the blend were not optimal (Bosch et
al. 2000).

Another promising attractive blend is the combination of lactic acid, acetone, and dimethyl
disulfide (Bernier et al. 2007, Silva et al. 2005, Williams et al. 2006a). The latter two compounds
have been detected in human breath and in human skin emanations (Bernier et al. 2000, Philips
1997). MM-X traps (Woodstream, Lititz, PA, USA; obtained originally from American Biophysics,
East Greenwich, RI, USA) baited with this synthetic blend were attractive to Ae. aegypti females
under controlled laboratory conditions (Silva et al. 2005). However, natural human odour was
significantly more attractive than the triple blend in two-choice olfactometer experiments
(Bernier et al. 2007). The addition of ammonia and hexanoic acid improved the attractiveness
of the blend, and the addition of acetone to a blend of lactic acid, ammonia and hexanoic acid
increased attraction to the latter blend (Williams et al. 2006a). BG-Sentinel traps (BioGents AG,
Germany) baited with the four blends mentioned above (i.e. combinations of ammonia, hexanoic
acid, lactic acid, acetone and dimethyl disulfide) appeared to be highly effective and Ae. aegypti-
specific in field experiments in Australia (Williams et al. 2006a,b). However, this seemed to be due
to visual more than olfactory cues (Williams et al. 2006a).

Olfaction in vector-host interactions 155

Renate C. Smallegange and Willem Takken

Acetone, dimethyl disulfide and dichloromethane, attractants on their own, were found to act
synergistically when combined in a binary blend with lactic acid (Allan et al. 2006a, Bernier et
al. 2003, 2007). The binary blends were less attractive than the triple blend mentioned above
(Bernier et al. 2007). Aedes aegypti females also responded to other sulfides that are found on
human skin, in human breath or may be present in human blood (Ashley et al. 1992, Bernier et al.
2000, Krotoszynski et al. 1977, Philips 1997); carbon disulfide, methyl propyl disulfide and methyl
sulfide, either in a dual-choice olfactometer and/or in landing assays were found to be attractive,
but dimethyl trisulfide and ethyl disulfide were not (Allan et al. 2006a) (Table 1).

Several ketones, which were found to be present in higher amounts on glass beads handled by
a person that was more attractive to Ae. aegypti, proved to be weak attractants in a dual-port
olfactometer (Table 1). Two of these compounds, butanone and 2-pentanone, decreased after
the skin sample had aged for 8 hours. In contrast, one of the ketones that was found to be a weak
attractant, 6-methyl-5-heptene-2-one, was also found in lower quantities on a day when a person
was more attractive (Bernier et al. 2002). Similarly, Logan et al. (2008) showed that 6-methyl-5-
heptene-2-one and especially geranyl acetone may be responsible for the decreased attraction
of certain individuals. The former inhibited flight activation and probing activity but not relative
attraction, whereas the latter reduced flight activity over all doses tested as well as relative
attraction and probing activity.

The aldehydes octanal, nonanal and decanal, which are present at high levels in the skin emanations
of less attractive individuals, caused a significant reduction in upwind flight of Ae. aegypti females
and reduced attraction to a human hand at certain concentrations (Bernier et al. 2002, Logan et
al. 2008). This indicates that compounds that are present on human skin may act as repellents or
‘mask’ the attractive effects of other skin compounds. The abundance of these compounds may
explain the relative attractiveness of different individuals to this mosquito species.

Hexanal, octanal and z-4-decanal, all aldehydes produced by the seabird crested auklet (Aethia
cristatella Pallas), were shown to decrease the number of landings on a human hand by Ae. aegypti
mosquitoes at doses similar to concentrations produced by this bird species (Douglas et al. 2001,
2005). Octanal (2.5%) was as repellent as a mixture of four aldehydes, including octanal and two
carboxylic acids mimicking the auklet odorant in its natural composition. Both 2% hexanal and 2%
hexanoic acid reduced the number of mosquito landings to the same extent (Douglas et al. 2005).

Several other compounds, which were interesting because they were more abundant in the skin
emanations of a less attractive individual or were found in higher or lesser amounts when the
skin emanations of a certain person were more attractive on some days than on other days, have
been screened (Bernier et al. 2002). These compounds, which include two alcohols, 4-hexen-1-ol
and 1-hepten-3-ol, with structural similarity to 1-octen-3-ol, an attractant for some mosquito
and haematophagous fly species (e.g. Takken and Kline 1989, Schofield et al. 1997, Vale and Hall
1985a,b), were not attractive when tested individually (Table 1).

Aedes albopictus (Skuse), a competent vector for many arboviruses, which exhibits an opportunistic
host-feeding pattern and is extending its range around the world (Gratz 2004, Richards et al. 2006,
Turell et al. 2005) has been shown to be activated (in a Y-tube olfactometer) to a greater degree by
hexanoic acid, ethyl butyrate and dimethyl disulfide than by lactic acid. These compounds were
also shown to be attractive to this mosquito species (Wang et al. 2006). In contrast, Shirai et al.
(2001) found a repellent, or inhibitory, effect of lactic acid for this mosquito species when applied
on normally attractive human or mouse skin.

156  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Anopheles gambiae s.s.

Experiments in an African village conducted by Haddow (1942) indicated that the host-seeking
behaviour of An. gambiae females is mediated primarily by human odour. This has been proven
by many studies since (reviewed by Takken and Knols 1999). In the laboratory, An. gambiae is
strongly attracted to human emanations, in particular skin emanations. A single finger (Dekker
et al. 2001b, Smallegange et al. 2002), ethanol washings of human hands and feet (Braks 1999),
human sweat (Braks and Takken 1999, Meijerink et al. 2000, Smallegange et al. in press), and skin
emanations of human hands or feet transferred to fabric (Dekker et al. 2001a, Pates et al. 2001b,
Qiu et al. 2004b, Smallegange et al. in press, Spitzen et al. 2008) or glass beads (Qiu et al. 2004a,
2006a) are highly attractive when tested in wind tunnels, Y-tube or dual-choice olfactometers or
indoor experiments with mosquito traps.

Several compounds present in human sweat or on human skin have been tested individually or
in blends under laboratory conditions to identify their effect on An. gambiae females (Table 2).
Ammonia is an important kairomone for this mosquito species; it is attractive over a wide range
of concentrations and suppresses the inhibitory or repellent effect of a carboxylic acid mixture
(Braks et al. 2001, Smallegange et al. 2005). However, at high doses ammonia has a repellent, or
inhibitory, effect (Smallegange et al. 2005).

The role of lactic acid in the host-orientation phase seems to be different in An. gambiae compared
to Ae. aegypti; on its own it is either not an attractant, or only a weak attractant (Braks et al.
2001, Dekker et al. 2002, Healy and Copland 2000, Smallegange et al. 2002, 2005). However, it
has been found to augment the attractiveness of natural human skin odour, carbon dioxide, and
synthetic odour blends, indicating that lactic acid is an essential compound in the orientation of
An. gambiae to human hosts (Dekker et al. 2002, Qiu 2005, Smallegange et al. 2005, 2009).

Variable results have been obtained in field studies with carbon dioxide (Costantini et al. 1996,
Knols et al. 1998, Mboera and Takken 1997, Qiu et al. 2007a). For this species the compound is
considered to indicate the presence of a potential host while host-selection is accomplished
by host-specific cues (Takken and Knols 1999). Dual-port olfactometer studies have shown that
CO2 is a poor kairomone for An. gambiae that can even have a repellent, or inhibitory, effect
depending on the structure of the plume and the positioning of the release point (Dekker et
al. 2001b, Spitzen et al. 2008). The compound, however, activates An. gambiae, guiding them
towards the release point of CO2 after which human skin emanations attract the mosquitoes
into a trapping device (Dekker et al. 2001b, Healy and Copland 1995, Knols et al. 1994b, Spitzen
et al. 2008). This was demonstrated by the fact that CO2 released in a turbulent plume in front of
the trap entrance enhanced the attractiveness of skin emanations. This combination was more
attractive than the synthetic blend of ammonia, lactic acid and CO2 (although the former two
compounds increased the effect of CO2), indicating that additional compounds are involved in the
host-seeking behaviour of this mosquito species (Spitzen et al. 2008). Recently, (semi-)field studies
confirmed that the addition of CO2 to human emanations increases the number of An. gambiae
caught in odour baited mosquito traps (Jawara et al. 2009, Njiru et al. 2006, Schmied et al. 2008).

Human sweat has been shown to stimulate An. gambiae females to land at the odour source,
whereas a mixture of 22 aliphatic carboxylic acids, which have been identified from human sweat,
did not elicit a landing response (Healy and Copland 2000). However, a blend of 12 aliphatic
carboxylic acids was found to be attractive when highly diluted (Knols et al. 1997). A similar blend,
although repellent or inhibitory on its own, was attractive when combined with ammonia and

Olfaction in vector-host interactions 157

Renate C. Smallegange and Willem Takken

Table 2. Overview of chemical compounds that have been tested for their impact on the host-seeking behaviour of
Anopheles gambiae s.s. using behavioural assays in the laboratory. (Details about concentrations of compounds
are omitted).

Compound Response Behavioural assay Reference

carbon dioxide (CO2) Activation and upwind flight Wind tunnel Healy and Copland (1995)
(Weakly) attractive on its own Dual-port olfactometer Knols et al. (1994b)
Homogeneous CO2 reduced trap Dual-port olfactometer Dekker et al. (2001b)
catch; Turbulent CO2 caused Dual-port olfactometer Spitzen et al. (2008)
activation and attractive synergism
with skin emanations
ammonia (NH3) Attractive or repellent / inhibitory Dual-port olfactometer Braks et al. (2001)
effect depending on doses Dual-port olfactometer Smallegange et al. (2005)
Y-tube Smallegange et al. (2002)
Alcohols
3-methyl-1-butanol Repellent / inhibitory effect when Dual-port olfactometer Qiu (2005)
combined with NH3 + LA
4-ethylphenol Repellent / inhibitory effect when Dual-port olfactometer Qiu (2005)
combined with NH3 + LA
1-dodecanol No effect when combined with Dual-port olfactometer Qiu (2005)
NH3 + LA
Aliphatic carboxylic acids
lactic acid (LA) No landing response Wind tunnel Healy and Copland (2000)
Not or weakly attractive on its own Dual-port olfactometer Braks et al. (2001)
Y-tube Smallegange et al. (2002)
Augmented attractiveness of CO2 Y-tube Dekker et al. (2002)
and human odour
Synergism when combined with Dual-port olfactometer Smallegange et al. (2005)
NH3 and carboxylic acids Dual-port olfactometer Smallegange et al. (2009)
acetic acid No effect in combination with NH3 Dual-port olfactometer Smallegange et al. (2009)
+ LA
propanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
2-methylpropanoic No effect in combination with NH3 Dual-port olfactometer Smallegange et al. (2009)
acid + LA
butanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
3-methyl butanoic Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
acid NH3 + LA
pentanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
hexanoic acid No effect on its own Y-tube Smallegange et al. (2002)
Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA or repellent / inhibitory
effect depending on doses

158  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Table 2. Continued.

Compound Response Behavioural assay Reference

heptanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA or repellent / inhibitory
effect depending on doses
octanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
nonanoic acid No effect when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
decanoic acid No effect when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 and LA
dodecanoic acid No effect when combined with Dual-port olfactometer Smallegange et al. (2009)
+ LA
tridecanoic acid No effect when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
tetradecanoic acid Synergism when combined with Dual-port olfactometer Smallegange et al. (2009)
NH3 + LA
Attractive when combined with Indoor trapping Smallegange et al. (2009)
NH3 + LA experiment

hexadecanoic acid No effect in combination with NH3 Dual-port olfactometer Smallegange et al. (2009)
+ LA
blend of 12 Attractive Dual-port olfactometer Knols et al. (1997)
carboxylic acids Repellent / inhibitory effect Dual-port olfactometer Smallegange et al. (2005)
Synergism when combined with Dual-port olfactometer Smallegange et al. (2005)
NH3 + LA
blend of 22 No landing response Wind tunnel Healy and Copland
carboxylic acids (2000)
blend of ammonia, Attractive Dual-port olfactometer Smallegange et al. (2009)
lactic acid and 7 Indoor trapping Smallegange et al. (2009)
carboxylic acids experiment
Unsaturated carboxylic acids
7-octenoic acid Attractive when combined with Dual-port olfactometer Qiu (2005)
NH3 + LA
combination of (E/Z)- Reduced response to CO2 Dual-port olfactometer Costantini et al. (2001)
3-methyl-2-hexenoic
acid isomer mixture
and 7-octenoic acid
Oxocarboxylic acids
2-oxopropanoic acid No landing response Wind tunnel Healy et al. (2002)
2-oxobutanoic acid Induced landing response Wind tunnel Healy et al. (2002)
2-oxo-3- Induced landing response Wind tunnel Healy et al. (2002)
methylbutanoic acid
2-oxopentanoic acid Induced landing response Wind tunnel Healy and Copland
(2000)
Wind tunnel Healy et al. (2002)

Olfaction in vector-host interactions 159

Renate C. Smallegange and Willem Takken

Table 2. Continued.

Compound Response Behavioural assay Reference

2-oxo-3- Induced landing response Wind tunnel Healy et al. (2002)

methylpentanoic
acid
2-oxo-4- Induced landing response Wind tunnel Healy et al. (2002)
methylpentanoic
acid
2-oxohexanoic acid Induced landing response Wind tunnel Healy et al. (2002)
2-oxooctanoic acid No landing response Wind tunnel Healy et al. (2002)
2-hydroxypentanoic No landing response Wind tunnel Healy et al. (2002)
acid
Ketones
acetone Activating effect when combined Wind tunnel Takken et al. (1997)
with CO2
Enhanced the effect of LA, Dual-port olfactometer Qiu (2005)
decreased effect of NH3
No aditional attractiviness when Dual-port olfactometer Qiu (2005)
combined with NH3 + LA
6-methyl-5-hepten- Repellent / inhibitory effect when Dual-port olfactometer Qiu (2005)
2-one combined with NH3 + LA
geranyl acetone Repellent / inhibitory effect when Dual-port olfactometer Qiu (2005)
combined with NH3 + LA
Miscellaneous
benzothiazole Repellent / inhibitory effect Dual-port olfactometer Qiu et al. (2004a)
indole Repellent / inhibitory effect when Dual-port olfactometer Qiu (2005)
tested alone and when combined
with NH3 + LA
blend of 1-butanol, Attractive Indoor trapping Verhulst et al. (2009)
2,3-butanedione, experiment
2-methyl-1-butanol,
2-methylbutanal,
2-methylbutanoic acid,
3-hydroxy-2-butanone,
3-methyl-1-butanol,
3-methylbutanal,
3-methylbutanoic acid,
benzeneethanol

lactic acid (Smallegange et al. 2005). In dual-choice olfactometer experiments, this blend was less
attractive than a worn sock, suggesting that more volatiles play a role in the host-orientation of
An. gambiae (Smallegange et al., in press).

In addition, seven individual saturated aliphatic carboxylic acids (propanoic, butanoic, 3-methyl-
butanoic, pentanoic, heptanoic, octanoic and tetradecanoic acid) and an unsaturated carboxylic

160  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

acid (7-octenoic acid) increased the attractiveness of ammonia when combined in a tripartite blend
together with lactic acid (Qiu 2005, Smallegange et al. 2009). A blend consisting of ammonia, lactic
acid and the seven saturated carboxylic acids was attractive both in a dual-choice olfactometer
and in indoor trapping experiments in a netted cage (Smallegange et al. 2009). Moreover, this
synthetic blend, combined with CO2, attracted more mosquitoes than humans when present in
different experimental huts, whereas it was equally or less attractive than humans when present
within the same hut (Okumu et al. 2010).

Hexanoic and heptanoic acid were found to have inhibitory/repellent effects at certain doses
(Smallegange et al. 2002, 2009). A range of other carboxylic acids (acetic, 2-methylpropanoic,
nonanoic, decanoic, dodecanoic, tridecanoic and hexadecanoic acid) had no effect on the
attractiveness of ammonia combined with lactic acid (Smallegange et al. 2009). The combination
of 7-octenoic acid and the isomer mixture of (E/Z)-3-methyl-2-hexenoic acid reduced the response
to CO2 in a dual-choice olfactometer (Costantini et al. 2001). The latter is a major component in
human axillary odour, the former a minor component (Zeng et al. 1991, 1996).

Six oxocarboxylic acids (2-oxobutanoic, 2-oxo-3-methylbutanoic, 2-oxopentanoic, 2-oxo-3-

methylpentanoic, 2-oxo-4-methylpentanoic and 2-oxohexanoic acid), three of which have been
identified in human sweat, caused a landing response in An. gambiae females (Table 2). More
mosquitoes landed in response to 2-oxopentanoic acid than to the other oxocarboxylic acids.
The landing response to this compound was found to be temperature dependent and decreased
to control levels within 10 minutes after application (Healy and Copland 2000, Healy et al. 2002).

Three alcohols (1-dodecanol, 4-ethyl phenol, 3-methyl-butanol) individually added to ammonia

and lactic acid had no effect, inhibited or repelled An. gambiae females (Qiu 2005), although two
of them (1-dodecanol, 3-methyl-butanol) were found at high levels in incubated sweat, which
was found to be more attractive than fresh sweat (Braks and Takken 1999, Meijerink et al. 2000).

Mosquitoes preferred ammonia over a blend of ammonia, lactic acid and either geranyl acetone or
6-methyl-5-hepten-2-one, depending on the doses of the ketones (Qiu 2005). These two ketones
were found in equal quantities in fresh and incubated sweat (Meijerink et al. 2000).

Acetone is well known as an attractive kairomone for tsetse flies. This compound has been found
in blood, milk, urine and breath of cattle, which is the host of these vectors of trypanosomiasis
(Jordan 1986, Torr et al. 1995, Vale 1980, Vale and Hall 1985a,b). It is also relatively abundant in
human breath (Philips 1997) and was found in fresh but not in incubated sweat (Meijerink et al.
2000). A human-equivalent physiological concentration of acetone had an activating effect on An.
gambiae when added to CO2 (Takken et al. 1997). It was also found to increase the attractiveness
of lactic acid, but decreased the attractiveness of ammonia and was not attractive on its own.
Although the triple blend of ammonia, lactic acid and acetone was attractive, it was as attractive
as ammonia alone (Qiu 2005).

When tested alone, indole, which is abundantly present in incubated sweat (Meijerink et al. 2000),
caused either no effect, inhibition or repelled, depending on the concentration tested. A blend of
ammonia, lactic acid and indole was significantly less attractive than ammonia alone (Qiu 2005).
A compound found in the headspace of worn nylon stockings but suspected to be not specifically
human, benzothiazole, repelled An. gambiae females at close range in a dual-port olfactometer
(Qiu et al. 2004b).

Olfaction in vector-host interactions 161

Renate C. Smallegange and Willem Takken

An. gambiae is attracted to volatiles produced by human skin microbiota both in dual-choice
olfactometer and in indoor trapping experiments. Moreover, MM-X traps baited with a blend of
10 compounds present in the headspace of human feet microbiota (1-butanol, 2,3-butanedione,
2-methyl-1-butanol, 2-methylbutanal, 2-methylbutanoic acid, 3-hydroxy-2-butanone, 3-methyl-1-
butanol, 3-methylbutanal, 3-methylbutanoic acid, benzeneethanol) caught significantly more An.
gambiae during indoor trapping experiments than unbaited traps (Verhulst et al. 2009).

Another member of the An. gambiae complex, An. quadriannulatus (Theobald), has long been
considered to be zoophilic and therefore of no medical importance (Takken et al. 1999). Hence its
host-seeking behaviour has rarely been studied. Pates et al. (2001a), however, showed that this species
has an equal preference for humans and calves. In addition, An. quadriannulatus preferred human
odour over a synergistic combination of cow odour with CO2 in a dual-port olfactometer. Carbon
dioxide alone was not attractive, whereas CO2 with 1-octen-3-ol had a inhibitory or repellent effect.
The effect of acetone was not clear (Pates et al. 2005). A recent field study with An. quadriannulatus
confirmed the finding that this species is attracted to human and cattle odours (Torr et al. 2008).

Culex quinquefasciatus

Almost 60% of the Cx. quinquefasciatus females from a colony that originated from Sri Lanka that
were released individually in a dual-port olfactometer were attracted within three minutes to
polyamide stockings that had been worn by a human volunteer for the preceding 48-60 hours
(Mboera et al. 1998), whereas, in a separate study, less than 25% of the mosquitoes from a colony
in the USA released in groups of 50-70 females were attracted to a human hand (Allan et al.
2006b). Skin rubbings and ethanol washings of human hands and feet or the back were attractive
to groups or individual female mosquitoes (Braks et al. 1999). These results demonstrate that Cx.
quinquefasciatus from different parts of the world are attracted to sources of volatiles that are
present on human skin.

In laboratory studies carbon dioxide has little or no attractive effect on this species. A high
concentration was found to be repellent. Contrary to what has been found for other mosquito
species, no synergistic effect of CO2 in combination with skin emanations has been observed
(Allan et al. 2006b, Braks et al. 1999, Mboera et al. 1998).

Culex quinquefasciatus was attracted to lactic acid in dual-port olfactometer studies (Allan et
al. 2006a, Braks et al. 1999). Ethanol washings from human hands and feet were significantly
more attractive than lactic acid alone, demonstrating that lactic acid is not the only attractive
component on human skin (Braks et al. 1999). Eighteen other compounds present in human skin
emanations were found to be attractive at certain doses to this species when tested individually
in a Y-tube olfactometer (Table 3). Some of these attracted over 75% of the mosquitoes released:
heptanal, nonanal, ethylene glycol, benzyl alcohol, and eleven carboxylic acids (Puri et al. 2006).
Several of these compounds, however, were not found to have an effect on mosquitoes in a dual-
port olfactometer study, possibly due to the concentrations at which these compounds were
tested (Table 3) (Allan et al. 2006a,b). Other alcohols and carboxylic acids had either no effect or
exhibited decreased behavioural responses compared to the control, depending on the doses
applied in a Y-tube olfactometer (Table 3) (Puri et al. 2006).

Recent field experiments showed that not only CO2 enhances trap catches, but also nonanal. The
latter compound is dominantly present in the odour profile of pigeons, chickens and humans from

162  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Table 3. Overview of chemical compounds that have been tested for their impact on the host-seeking behaviour of
Culex quinquefasciatus using behavioural assays in the laboratory. (Details about concentrations of compounds
are omitted).

Compound Response Behavioural assay Reference

carbon dioxide (CO2) Not or weakly attractive on its Dual-port olfactometer Mboera et al. (1998)
own; repellent / inhibitory effect at Dual-port olfactometer Braks et al. (1999)
high concentration; no synergism Dual-port olfactometer Allan et al. (2006b)
with host emanations
Alcohols
ethylene glycol Attractive Y-tube Puri et al. (2006)
glycerol Repellent / inhibitory effect Y-tube Puri et al. (2006)
phenol No response Y-tube Puri et al. (2006)
benzyl alcohol Attractive Y-tube Puri et al. (2006)
decanol Attractive or repellent / inhibitory Y-tube Puri et al. (2006)
effect depending on doses
cholesterol Attractive Y-tube Puri et al. (2006)
Aldehydes
propanal Attractive Y-tube Puri et al. (2006)
hexanal No response Dual-port olfactometer Allan et al. (2006b)
heptanal Attractive Y-tube Puri et al. (2006)
benzaldehyde No response Y-tube Puri et al. (2006)
Dual-port olfactometer Allan et al. (2006b)
nonanal Attractive Y-tube Puri et al. (2006)
No response Dual-port olfactometer Allan et al. (2006b)
Aliphatic carboxylic acids
lactic acid (LA) Attractive Dual-port olfactometer Allan et al. (2006a)
Dual-port olfactometer Braks et al. (1999)
No landing response Landing assay Allan et al. (2006a)
acetic acid Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
propanoic acid Attractive Y-tube Puri et al. (2006)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
butanoic acid Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
3-methyl butanoic Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
acid No landing response Landing assay Allan et al. (2006a)
hexanoic acid Attractive Y-tube Puri et al. (2006)
heptanoic acid Attractive Y-tube Puri et al. (2006)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
octanoic acid Attractive Y-tube Puri et al. (2006)
nonanoic acid Attractive Y-tube Puri et al. (2006)
decanoic acid Attractive Y-tube Puri et al. (2006)
undecanoic acid Attractive Y-tube Puri et al. (2006)
dodecanoic acid Attractive Y-tube Puri et al. (2006)
tridecanoic acid Attractive Y-tube Puri et al. (2006)

Olfaction in vector-host interactions 163

Renate C. Smallegange and Willem Takken

Table 3. Continued.

Compound Response Behavioural assay Reference

tetradecanoic acid Attractive Y-tube Puri et al. (2006)

Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
pentadecanoic acid Repellent / inhibitory effect Y-tube Puri et al. (2006)
hexadecanoic acid Repellent / inhibitory effect Y-tube Puri et al. (2006)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
heptadecanoic acid Attractive Y-tube Puri et al. (2006)
octadecanoic acid Repellent / inhibitory effect Y-tube Puri et al. (2006)
Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
Aromatic carboxylic acids
benzoic acid No landing response Landing assay Allan et al. (2006a)
2-hydroxybenzoic No landing response Landing assay Allan et al. (2006a)
acid
Sulfides
methyl sulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
carbon disulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
dimethyl disulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
ethyl disulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
No landing response Landing assay Allan et al. (2006a)
methyl propyl Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
disulfide Induced landing response Landing assay Allan et al. (2006a)
dimethyl trisulfide Not attractive on its own Dual-port olfactometer Allan et al. (2006a)
Induced landing response Landing assay Allan et al. (2006a)
Miscellaneous
meso-2,3-butanediol Not attractive on its own Dual-port olfactometer Allan et al. (2006b)
2,3-butanediol Not attractive on its own Dual-port olfactometer Allan et al. (2006b)
α-pinene Not attractive on its own Dual-port olfactometer Allan et al. (2006b)
β-myrcene Not attractive on its own Dual-port olfactometer Allan et al. (2006b)
2,3-docosanediol Not attractive on its own Dual-port olfactometer Allan et al. (2006b)
blend of hexanal, Not attractive Dual-port olfactometer Allan et al. (2006b)
α -pinene,
benzaldehyde,
β-myrcene, and
nonanal

164  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

various ethnic backgrounds. Traps baited with both nonanal and CO2 caught the highest number
of Cx. quinquefasciatus (Syed and Leal 2009).

Culex quinquefasciatus females assayed in a dual-port olfactometer responded to a different extent

to odour sources of animal origin (bovine blood, avian blood, chickens, chicken feathers, chicken
feathers combined with CO2, cotton balls rubbed over chicken feathers and hexane extracts
of chicken feathers) (Allan et al. 2006a,b). However, there was no response to five compounds
identified from feathers (Williams et al. 2003) individually or as a blend (hexanal, benzaldehyde,
nonanal, α-pinene, and β-myrcene), or to three individual compounds (meso-2,3-butanediol,
2,3-butanediol and 2,3-docosanediol) that have been identified in the preen gland of hens (Allan
et al. 2006b, Haathi and Fales 1967; Table 3).

In a Y-tube olfactometer, fresh chicken faeces was attractive to Cx. quinquefasciatus. Eight aldehydes
((E)-2-decanal, undecanal, dodecanal, tetradecanal, pentadecanal, hexadecanal, heptadecanal and
octadecanal) identified in the headspace of the faeces elicited electroantennogram responses, as
well as eight unknown compounds (Cooperband et al. 2008), and should be tested for their effect
on the host-seeking behaviour of Cx. quinquefasciatus females.

Volatile compounds associated with blood (aliphatic and aromatic carboxylic acids and sulphides)
showed no attractiveness in olfactometer assays, with the exception of lactic acid (Table 3).
However, in cage assays, significantly more landings were observed on collagen sausage casings
treated with acetic, hexadecanoic, and octadecanoic acid, dimethyl trisulfide or methyl propyl
disulfide than on casings treated with water (Allan et al. 2006a).

Female Cx. nigripalpus Theobald, another mosquito species associated with West Nile virus
transmission in the USA (Godsey et al. 2005, Sardelis et al. 2001), were attracted to tetradecanoic
acid, dimethyl disulfide and methyl propyl disulfide and landed on casings treated with lactic
acid, octadecanoic acid, dimethyl trisulfide, ethyl disulfide or methyl propyl disulfide, although
responses to bovine blood were low (Allan et al. 2006a). Culex nigripalpus and Cx. tarsalis Coquillett
did not respond to human odour, whereas they were attracted to chicken odour in a dual-port
olfactometer. Culex tarsalis was significantly more responsive to CO2 than were Ae. aegypti, Cx.
quinquefasciatus and Cx. nigripalpus (Allan et al. 2006b).

Discussion

In the last decade much progress has been made in the identification of kairomones and
allomones of Ae. aegypti, An. gambiae and Cx. quinquefasciatus. However, in the laboratory assays
these compounds and blends are less attractive than natural odours and work should continue
to identify the ‘missing’ compounds. Recent advances in molecular olfaction and neurophysiology
are expected to contribute to a rapid identification of essential odorants that are lacking from the
current blends.

Robust behavioural assays are available for screening olfactory cues to which mosquitoes respond,
although some devices are not suitable for all species. For example, An. gambiae behaviour is
erratic in Y-tube assays (Smallegange et al. 2002), whereas Ae. aegypti and Cx. quinquefasciatus
behaviour is normal and consistent in this device (Geier and Boeckh 1999, Puri et al. 2006).

The main bottleneck of high-throughput testing of candidate semiochemicals is the relatively

time-consuming operation of the behavioural bioassays, in contrast to neurophysiological studies.

Olfaction in vector-host interactions 165

Renate C. Smallegange and Willem Takken

Three-layer olfactometers, in which three sets of mosquitoes can be tested simultaneously, as

developed by the USDA research laboratory in Gainesville, Florida, USA (Posey et al. 1998), are
a valuable means of increasing the rate at which mosquitoes can be tested. Nevertheless, each
batch of novel odour blends needs to be tested in sufficient replicates, which slows down the
bioassay evaluation. The development of tools that might accelerate the behavioural assays
may be very useful so that new producers of semiochemicals will not be disappointed. These
behavioural assays require large numbers of female mosquitoes, which can be a burden on the
research budget as this requires the availability of climate-controlled insectaries and trained staff.

Progress in the development of digital recording techniques allows for the study of insect
behaviour using 3D image analysers (see publications for details of products available). Patterns
of flight behaviour in response to specific host-odours and blends of odours have been studied,
revealing interesting behavioural changes during odour-mediated upwind anemotaxis (Beeuwkes
et al. 2008, Braks et al. 2005, Cooperband and Cardé 2006, Chapter 6 in this volume). Thus, the
specific role of a semiochemical in the mosquito behavioural repertoire can be established. This
will facilitate the discovery of more effective odorant blends with which to modify or control
mosquito behaviour.

Current techniques in semiochemistry allow for the rapid analysis of volatiles of human and
animal origin (Bernier et al. 1999, Curran et al. 2007, Penn et al. 2007). This poses a potentially
huge problem for the behavioural ecologist, as it is unlikely that all compounds present in host
emanations are involved in host-seeking behaviour (Zwiebel and Takken 2004), and it is impractical
to examine each compound in a behavioural assay. High-throughput testing of semiochemicals is
possible at the neurophysiological level (Chapter 2 and 3, this volume), but encounters logistical
obstructions at the behavioural level. Thus, chemical ecologists will probably rely on testing
selected compounds and blends, based on results from neurophysiology and historical evidence.
The exploration of chemical libraries for small molecules that might affect mosquito behaviour,
as is currently in progress at different research facilities, will add a potentially large number of
compounds that require behavioural testing (Justice et al. 2003, Zwiebel and Takken 2004).

At present, a novel and promising technique is under development by several research groups, in
which differences in attractiveness to mosquitoes between human individuals is being exploited
to study chemical profiles with the aim of identifying critical chemical groups or compounds that
cause these differences (Bernier et al. 2002, Logan et al. 2008, Smallegange et al. 2003, Qiu et al.
2006a). Such compounds may be either kairomones or allomones (Dicke and Sabelis 1988). If the
compounds that are significantly more abundant in highly attractive persons can be identified, it is
likely to provide a rapid method for the development of attractive odour blends for mosquitoes, as
it is expected that these compounds play a role in the attraction of these individuals. Alternatively,
poorly attractive individuals may produce compounds that quench or even suppress the activity
of other chemicals, so that the mosquito can no longer identify the source or, perhaps, is even
repelled from it (Logan et al. 2008).

The relative concentrations of semiochemicals in odour blends is a fundamental factor in

determining their attractiveness. For example, Cx. quinquefasciatus was not attracted to certain
synthetic volatiles of chicken origin (Allan et al. 2006b), whereas it did respond to natural
bird odours. This may indicate that either critical volatiles were missing or that the relative
concentrations of components in the blend were not appropriate. This example illustrates some
of the difficulties inherent in the development of effective synthetic odour baits. Knowledge of
the naturally occurring ratios of compounds in host volatiles is not necessarily useful; artificial

166  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

baits for trapping and killing tsetse represent the most successful application of semiochemicals
to control vectors in the world, and yet the blend of chemicals in the bait is only loosely related to
the naturally occurring ratios and concentrations of compounds emitted by live hosts (Vale and
Hall 1985a,b). It is also worth noting that compounds and blends of compounds that are attractive
at a particular concentration, may be repellent or cause inhibition at higher concentrations (e.g.
Knols et al. 1997, Smallegange et al. 2005, Vale and Hall 1985a). Therefore, single compounds and
blends should be tested across a range of concentrations in a methodical way, similar to range
testing doses of drugs or toxins. Differences in the concentrations of compounds tested may
explain differences in the results of assays based on the same compounds (Allan et al. 2006a,b, Puri
et al. 2006). Similarly, the type of assay used (Y-tube olfactometer, dual-port olfactometer, landing
assay) may affect the results obtained.

A more directed approach may help to identify the most important compounds controlling
host-seeking behaviour of particular species. For example, having found that hexane extracts
of chicken feathers were attractive to Cx. quinquefasciatus while ether extracts were not (Allan
et al. 2006b), Bernier et al. (2008) examined these extracts and found that the ether extracts
contained aldehydes, which were also present in the hexane extracts, whereas the hexane
extracts also contained alcohols, ketones and diones. Subsequent behavioural assays can focus
now on compounds of the latter three groups or can be preceded by either on-line (coupled
with GC) or off-line EAG recordings to determine whether components present in the hexane
extract are detected by the mosquito olfactory structures as Cooperband et al. (2008) did with
acidified chicken faeces. A similar approach could be used to identify more active compounds, in
addition to lactic acid and ammonia, which are present in the fractions of skin extracts attracting
Ae. aegypti (Geier et al. 1996, 1999a). Unidentified compounds may yet be found that increase the
attractiveness of the best blends available so far (Williams et al. 2006a).

It has been shown repeatedly that a compound may be more attractive in mixtures than when
applied singly (Geier et al. 1999a, Smallegange et al. 2005). Aedes aegypti females are more
attracted to triple blends of lactic acid, acetone, dimethyl disulfide or dichloromethane than
to binary blends, whereas binary blends are more attractive than single compounds (Bernier et
al. 2007). A similar observation was found for tsetse flies: a mixture of three chemicals is more
attractive than individual chemicals and more tsetse flies were trapped when the mixture was
applied at higher doses for all components, except octenol, which reduces trap catches at higher
doses, showing again the importance of controlling concentrations (Vale and Hall 1985b).

Compounds that have no clear effect on behaviour when used individually may, however, play an
important role in a mixture, as was shown for lactic acid; although it is not or only weakly attractive
to An. gambiae when applied alone, it has a synergistic effect when presented with ammonia and
carboxylic acids (Qiu 2005, Smallegange et al. 2005, 2009). Similarly, Ae. aegypti is not attracted
to ammonia alone, whereas this compound enhances the effect of lactic acid (Geier et al. 1999a).
This phenomenon makes it even harder to ‘predict’ which mixture of compounds will produce an
effective synthetic blend that is highly attractive to mosquitoes.

For Ae. aegypti and An. gambiae particular blends of kairomones attract a reasonably large number
of mosquitoes in bioassays when tested on their own. However, the natural odour complex of
the human host was significantly more attractive than any synthetic blend when compared
directly against each other (Bernier et al. 2007, Spitzen et al. 2008, Smallegange et al. in press).
It is therefore likely that essential components are lacking in these, otherwise, attractive blends.

Olfaction in vector-host interactions 167

Renate C. Smallegange and Willem Takken

Another complication in the identification of kairomones is that a chemical may be attractive

when applied alone, but shows an inhibitory or a (spatial) repellent effect in combination with
human odours. Such an effect has been found for the widely used mosquito repellent DEET (N,N-
diethyl-3-methylbenzamide), which was found to be (weakly) attractive to Ae. aegypti when used
on its own (Bernier et al. 2005, Dogan et al. 1999, Kline et al. 2003, Mehr et al. 1990), but caused
(spatial) repellence or inhibited attraction when combined with natural human odours (Bernier et
al. 2005, Curtis 1992, Dogan et al. 1999, Kline et al. 2003). Recently, however, Syed and Leal (2008)
demonstrated true behavioural repellency of Cx. quinquefasciatus by DEET in the absence of other
chemostimuli.

In addition, the method of testing putative repellents can conceal its possible effect. Based on the
definition of a repellent as ‘a chemical which causes insects to make oriented movements away from
its source’ (Dethier et al. 1960), it remains unclear from most Y-tube and dual-choice olfactometer
assays whether compounds that attract significantly fewer mosquitoes than the control are truly
repellents (Dogan and Rossignol 1999), because in these devices the insects cannot move away
from the source. However, when the term repellent is used to designate products that intent to
reduce the biting-rate of haematophagous insects (see White 2007), this term can be used for
chemicals that have been shown to reduce the number of mosquitoes attracted to a natural or
synthetic odour source when tested in a Y-tube or dual-choice olfactometer. Especially area or
spatial repellents, which prevent mosquitoes from reaching their blood-host and are thus effective
at a distance from the source of application (see Strickman 2007), may be discovered with these
kind of bioassays. Several methods have been developed to test chemicals for their ability to
be applied as mosquito repellents, either as a spatial or topical repellent. Each method has its
advantages and disadvantages (e.g. Barnard et al. 2007, Chareonviriyaphap et al. 2002, Dogan and
Rossignol 1999, Grieco et al. 2005, Hao et al. 2008, Kline et al. 2003, Klun et al. 2005, Rutledge and
Gupta 2004). Which bioassay should be used may depend on the research question and possible
future application method and may depend on the mode of action of the repellent. For these
reasons we suggest to name chemicals that show a inhibitory or repellent effect in laboratory
bioassays ‘putative repellents’ until their effectiveness has been proven in the field.

The previous paragraphs show that the bioassay used to test semiochemicals as behaviour-
mediating compounds for mosquitoes may provide an indication that a certain compound plays
a role in the host-seeking process. However, the result may not be conclusive of the exact role
of this compound. For example, most Y-tube and dual-choice olfactometers cannot discriminate
between an inhibitory and a repellent effect (Dogan and Rossignol 1999); additional, appropriate
bioassays should be performed. Since the host-seeking process exists of different phases (Takken
1996), the choice for a certain type of bioassay determines which phase is examined. For example,
the activating effect of a chemical may be missed in a landing assay or an olfactometer when it is
not combined with a chemical that induces a landing response or that is an attractant, whereas it
may be noticed when video recording in combination with a 3D image analyser is used. Therefore,
when no response has been found in a specific bioassay care should be taken to conclude that
a certain compound has no effect on the host-seeking behaviour of a certain mosquito species;
the compound may play a role in another host-seeking phase. In an olfactometer, attraction may
not be visible, even though the mosquitoes make oriented movements towards the source of a
chemical (Dethier et al. 1960), when the insects do not fly into the trapping system with which
the olfactometer is applied.

Although similarities have been found in the response to certain olfactory cues (e.g. ammonia,
lactic acid, aliphatic carboxylic acids, some ketones and aldehydes in Ae. aegypti, An. gambiae

168  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

and Cx. quinquefasciatus), it is clear that different mosquito species use different odour cues to
find their blood-host, and the role of the individual compounds varies between species, even for
species that have the same host preference. Finally, much research is required to establish the
range of variation in behaviour within wild populations of a given mosquito species across its
geographical distribution, for variability within populations, and for the effects of colonisation on
natural host-seeking behaviour, as indicated by some disparities between the results of studies
based on different populations of all three species reviewed here (Lefèvre et al. 2009; Williams et
al. 2003, 2006c). It is likely that different attractive blends, odour delivery systems and trap types
have to be used to manage different mosquito populations (Kline 2007).

Conclusions

Several compounds found in human emanations have been shown to cause a behavioural response
in the laboratory in at least one of the three mosquito species, Ae. aegypti, An. gambiae and Cx.
quinquefasciatus. Carbon dioxide is a universal kairomone, but with distinctly different effects on
the three species. In the laboratory, Aedes aegypti is strongly activated by this compound, whereas
for An. gambiae it is also a synergist and effective when released at a distance from the other
kairomones. The behavioural response of Cx. quinquefasciatus to CO2 is at best weak.

Composed odour blends are attractive to mosquitoes in the laboratory. These include a
combination of ammonia, lactic acid, hexanoic acid, acetone and dimethyl disulfide for Ae. aegypti,
and of ammonia, lactic acid and aliphatic carboxylic acids for An. gambiae. However, there is not
yet a blend that is equally attractive or more so than natural human odours. Studies comparing
(1) odours from poorly and highly attractive persons, (2) odours from a preferred host and a non-
host and (3) blood-host preferences of mosquitoes from different geographic regions are likely
to reveal critical compounds responsible for the strong attractiveness of humans to the three
mosquito species reviewed here.

Six-methyl-5-hepten-2-one and geranyl acetone are candidate repellents, as these cause inhibition
or repellence in Ae. aegypti (Logan et al. 2008) and An. gambiae (Qiu 2005). Interestingly, the
release of these compounds from a human arm has been found to decrease when treated with
DEET (Syed and Leal 2008).

Synthetic blends can be exploited to lure mosquitoes into odour-baited traps, manipulate them
away from human houses or cause general disruption to normal host-seeking behaviour by mass
release of human odorants (Day and Sjogren 1994, Kline 2007). Chapter 8 of this volume discusses
how laboratory-based results may be transferred to the (semi-)field for further testing prior to
their application.

It is expected that the studies in progress by several research consortia will lead to novel classes
of odorant blends with which mosquito disease vectors can be monitored or controlled. Recent
results from trapping with odour-baited MM-X or BG-Sentinel traps (Kline 2007, Kröckel et al. 2006,
Mboera et al. 2000, Njiru et al. 2006, Qiu et al. 2007a, Schmied et al. 2008, Williams et al. 2006b)
suggest that the application of such traps, provided the odorant stimulus is highly competitive
with natural host odour, are likely to become reality in the near future. Novel synthetic odorant
blends that disrupt the natural behaviour of mosquito vectors may be used in addition to current
control methods (Greenwood 2008, Takken and Knols 2009) to achieve a greater effect than is
presently possible.

Olfaction in vector-host interactions 169

Renate C. Smallegange and Willem Takken

Acknowledgements

We thank Prof Marc J. Klowden en Dr Gabriella Gibson for constructive comments and suggestions
on an earlier version of this manuscript.

References

Acree F, Jr, Turner RB, Gouck HK, Beroza M and Smith N (1968) L-lactic acid: A mosquito attractant isolated from humans.
Science 161: 1346-1347.
Allan SA, Bernier UR and Kline DL (2006a) Attraction of mosquitoes to volatiles associated with blood. J Vector Ecol
31: 71-78.
Allan SA, Bernier UR and Kline DL (2006b) Laboratory evaluation of avian odors for mosquito (Diptera: Culicidae)
attraction. J Med Entomol 43: 225-231.
Alonso WJ and Schuck-Paim C (2006) The ‘ghosts’ that pester studies on learning in mosquitoes: guidelines to chase
them off. Med Vet Entomol 20: 157-165.
Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Holler JS, Needham LL and Patterson DG, Jr (1992) Determining volatile
organic compounds in human blood from a large sample population by using purge and trap gas chromatography/
mass spectrometry. Analytical Chemistry 64: 1021-1029.
Barnard DR, Bernier UR, Xue R-D and Debboun M (2007) Standard methods for testing mosquito repellents. In: Debboun
M, Frances SP and Strickman D (eds) Insect repellents. Principles, methods, and uses. CRC Press, Boca Raton, FL,
USA, pp 103-110.
Beeuwkes J, Spitzen J, Spoor CW, Van Leeuwen JL and Takken W (2008) 3-D flight behaviour of the malaria mosquito
Anopheles gambiae s.s. inside an odour plume. Proceedings of Experimental and Applied Entomology, N.E.V.,
Amsterdam 19: 137-146.
Bernier UR, Allan SA, Quinn BP, Kline DL, Barnard DR and Clark GG (2008) Volatile compounds from the integument of
White Leghorn chickens (Gallus gallus domesticus L.): Candidate attractants of ornithophilic mosquito species. J
Sep Sci 31: 1092-1099.
Bernier UR, Booth MM and Yost RA (1999) Analysis of human skin emanations by gas chromatography/mass
spectrometry. 1.Thermal desorption of attractants for the yellow fever mosquito (Aedes aegypti) from handled
glass beads. Anal Chem 71: 1-7.
Bernier UR, Furman KD, Kline DL, Allan SA and Barnard DR (2005) Comparison of contact and spatial repellency of catnip
oil and N,N-diethyl-3-methylbenzamide (Deet) against mosquitoes. J Med Entomol 42:306-311.
Bernier UR, Kline DL, Allan SA and Barnard DR (2007) Laboratory comparison of Aedes aegypti attraction to human odors
and to synthetic human odor compounds and blends. J Am Mosq Control Assoc 23: 288-293.
Bernier UR, Kline DL, Barnard DR, Schreck CE and Yost RA (2000) Analysis of human skin emanations by gas
chromatography/mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for
yellow fever mosquito (Aedes aegypti). Anal Chem 72: 747-756.
Bernier UR, Kline DL, Posey KH, Booth MM, Yost RA and Barnard DR (2003) Synergistic attraction of Aedes aegypti (L.)
to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl disulfide. J Med Entomol 40: 653-656.
Bernier UR, Kline DL, Schreck CE, Yost RA and Barnard DR (2002) Chemical analysis of human skin emanations:
comparison of volatiles from humans that differ in attraction of Aedes aegypti (Diptera: Culicidae). J Am Mosq
Control Assoc 18: 186-195.
Birkett MA, Agelopoulos N, Jensen K-MV, Jespersen JB, Pickett JA, Prijs HJ, Thomas G, Trapman JJ, Wadhams LJ and
Woodcock CM (2004) The role of volatile semiochemicals in mediating host location and selection by nuisance and
disease-transmitting cattle flies. Med Vet Entomol 18: 313-322.
Bosch OJ, Geier M and Boeckh J (2000) Contribution of fatty acids to olfactory host finding of female Aedes aegypti.
Chem Senses 25: 323-330.
Bowen MF (1991) The sensory physiology of host-seeking behavior in mosquitoes. Annu Rev Entomol 36: 139-158.

170  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Braimah N, Drakeley C, Kweka E, Mosha F, Helinski M, Pates H, Maxwell C, Massawe T, Kenward MG, Curtis C (2005) Tests
of bednet traps (Mbita traps) for monitoring mosquito populations and time of biting in Tanzania and possible
impact of prolonged insecticide treated net use. Int J Trop Insect Sci 25: 208-213.
Braks MAH (1999) Human skin emanations in the host-seeking behaviour of the malaria mosquito Anopheles gambiae.
PhD thesis Wageningen University, Wageningen, the Netherlands. 123 pp.
Braks MAH, Cooperband MF and Cardé RT (2005) 3-D flight track analyses of Culex mosquito orientation to oviposition-
related odors. In: Noldus LP (ed) Proceedings Measuring Behavior 2005, Wageningen, the Netherlands, pp 209-210.
Braks MAH, Koenraadt CJM, Briet OJT and Takken W (1999) Behavioural responses of Culex quinquefasciatus (Diptera:
Culicidae) to human odour, carbon dioxide and lactic acid. Proceedings of Experimental and Applied Entomology,
N.E.V., Amsterdam 10: 91-97.
Braks MAH, Meijerink J and Takken W (2001) The response of the malaria mosquito, Anopheles gambiae, to two
components of human sweat, ammonia and L-lactic acid, in an olfactometer. Physiol Entomol 26: 142-148.
Braks MAH and Takken W (1999) Incubated human sweat but not fresh sweat attracts the malaria mosquito Anopheles
gambiae sensu stricto. Journal of Chemical Ecology 25: 663-672.
Bursell E, Gough AJE, Beevor PS, Cork A, Hall DR and Vale GA (1988) Identification of components of cattle urine
attractive to tsetse flies, Glossina spp. (Diptera: Glossinidae). Bull Entomol Res 78: 281-291.
Cardé RT and Willis MA (2008) Navigational strategies used by insects to find distant, wind-borne sources of odor. J
Chem Ecol 34: 854-866.
Carey AF, Wang G, Su C-Y, Zwiebel LJ and Carlson JR (2010) Odorant reception in the malaria mosquito Anopheles
gambiae. Nature (in press) doi:10.1038/nature08834.
Chandra G (1995) Short report: peak period of filarial transmission. Am J Trop Med Hyg 53: 378-379.
Chareonviriyaphap T, Prabaripai A and Sungvornyothrin S (2002) An improved excito-repellency test chamber for
mosquito behavioral tests. J Vector Ecol 27: 250-252.
Cook GC (1996) Manson’s tropical diseases. 20th edn. WB Saunders Company Ltd, London, UK.
Cook SM, Khan ZR and Pickett JA (2007) The use of push-pull strategies in integrated pest management. Annu Rev
Entomol 52: 375-400.
Cooperband MF and Cardé RT (2006) Orientation of Culex mosquitoes to carbon dioxide-baited traps: flight manoeuvres
and trapping efficiency. Med Vet Entomol 20: 11-26.
Cooperband MF, McElfresh JS, Millar JG, Cardé RT (2008) Attraction of female Culex quinquefasciatus Say (Diptera:
Culicidae) to odors from chicken feces. J Insect Physiol 54: 1184–1192.
Cork A and Hall MJR (2007) Development of an odour-baited target for female New World screwworm, Cochliomyia
hominivorax: studies with host baits and synthetic wound fluids. Med Vet Entomol 21: 85-92.
Cork A and Park KC (1996) Identification of electrophysiologically-active compounds for the malaria mosquito,
Anopheles gambiae, in human sweat extracts. Med Vet Entomol 10: 269-276.
Costantini C, Birkett MA, Gibson G, Ziesmann J, Sagnon NF, Mohammed HA, Coluzzi M and Pickett JA (2001)
Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae to human-specific sweat
components. Med Vet Entomol 15, 259-266.
Costantini C, Gibson G, Sagnon N, Della Torre A, Brady J and Coluzzi M (1996) Mosquito responses to carbon dioxide in
a West African Sudan savanna village. Med Vet Entomol 10: 220-227.
Crumb SE (1922) A mosquito attractant. Science 55: 446-447.
Cruz-Lopez L, Malo EA, Rojas JC and Morgan ED (2001) Chemical ecology of triatomine bugs: vectors of Chagas disease.
Med Vet Entomol 15: 351-357.
Curran AM, Rabin SI, Prada PA and Furton KG (2005) Comparison of the volatile organic compounds present in human
odor using SPME-GC/MS. J Chem Ecol 31: 1607-1619.
Curran AM, Ramirez CF, Schoon AA and Furton KG (2007) The frequency of occurrence and discriminatory power of
compounds found in human scent across a population determined by SPME-GC/MS. J Chromatogr B 846: 86-97.
Curtis CF (1986) Fact and fiction in mosquito attraction and repulsion. Parasitology Today 11: 316-318.
Curtis CF (1992) Personal protection methods against vectors of disease. Rev Med Vet Entomol 80: 543-553.
Curtis CF and Hill N (1988) Comparison of methods of repelling mosquitoes. Entomol Exp Appl 49: 175-179.

Olfaction in vector-host interactions 171

Renate C. Smallegange and Willem Takken

Davis EE (1984) Development of lactic acid-receptor sensitivity and host-seeking behaviour in newly emerged female
Aedes aegypti mosquitoes. J lnsect Physiol 30: 211-215.
Day JF and Sjogren RD (1994) Vectro control by removal trapping. Am J Trop Med Hyg 50: 126-133.
De Bruyne M and Baker TC (2008) Odor detection in insects: volatile codes. J Chem Ecol 34: 882-897.
De Jong R and Knols BGJ (1995) Selection of biting sites on man by two malaria mosquito species. Cellular and Molecular
Life Sciences (formerly Experientia) 51: 80-84.
De Jong R and Knols BGJ (1996) Selection of biting sites by mosquitoes: In: Bock GR and Cardew G (eds) Olfaction in
mosquito host interactions. John Wiley and Sons Ltd, Chichester, UK, pp 89-103.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972.
Dekker T, Steib B, Cardé RT and Geier M (2002) L-lactic acid, a human-signifying host cue for the anthropophilic mosquito
Anopheles gambiae. Med Vet Entomol 16: 91-98.
Dekker T, Takken W and Braks MAH (2001a) Innate preference for host-odor blends modulates degree of anthropophagy
of Anopheles gambiae sensu lato (Diptera: Culicidae). J Med Entomol 38: 868-871.
Dekker T, Takken W and Cardé RT (2001b) Structure of host-odour plumes influences catch of Anopheles gambiae s.s.
and Aedes aegypti in a dual-choice olfactomer. Physiol Entomol 26: 124-134.
Dekker T, Takken W, Knols BGJ, Bouman E, Van de Laak S, De Bever A and Huisman PWT (1998) Selection of biting sites
on a human host by Anopheles gambiae s.s., An. arabiensis and An. quadriannulatus. Entomol Exp Appl 87: 295-300.
Deng C, Zhang X and Li N (2004) Investigation of volatile biomarkers in lung cancer blood using solid-phase
microextraction and capillary gas chromatography–mass spectrometry. Journal of Chromatography B 808: 269-
277.
Dethier VG, Browne LB and Smith CN (1960) The designation of chemicals in terms of the responses they elicit from
insects. J Econ Entomol 53: 134-136.
Diatta M, Spiegel A, Lochouarn L, and Fontenille D (1998) Similar feeding preferences of Anopheles gambiae and A.
arabiensis in Senegal. Trans R Soc Trop Med Hyg 92: 270-272.
Dicke M and Sabelis MW (1988) Infochemical terminology: based on cost-benefit analysis rather than origin of
compounds? Funct Ecol 2. 131-139
Dogan EB, Ayres JW and Rossignol PA (1999) Behavioural mode of action of deet: inhibition of lactic acid attraction.
Med Vet Entomol 13: 97-100.
Dogan EB and Rossignol PA (1999) An olfactometer for discriminating between attraction, inhibition, and repellency in
mosquitoes (Diptera: Culicidae). J. Med. Entomol. 36: 788-793.
Douglas HD, Co JE, Jones TH and Conner WE (2001) Heteropteran chemical repellents identified in the citrus odor of
a seabird (crested auklet: Aethia cristatella): evolutionary convergence in chemical ecology. Naturwissenschaften
88: 330-332.
Douglas HD, Co JE, Jones TH, Conner WE and Day JF (2005) Chemical odorant of colonial seabird repels mosquitoes. J
Med Entomol 42: 647-651.
Duchemin J. B., Leong Pock Tsy J. M., Rabarison P, Roux J, Coluzzi M, and Costantini C (2001) Zoophily of Anopheles
arabiensis and An. gambiae in Madagascar demonstrated by odour-baited entry traps. Med Vet Entomol 15: 50-57.
Elizondo-Quiroga A, Flores-Suarez A, Elizondo-Quiroga D, Ponce-Garcia G, Blitvich BJ, Contreras-Cordero JF,
Gonzalez-Rojas JI, Mercado-Hernandez R, Beaty BJ and Fernandez-Salas I (2006) Host-feeding preference of Culex
quinquefasciatus in Monterrey, Northeastern Mexico. J Am Mosq Control Assoc 22: 654-661.
Eiras AE and Jepson PC (1991) Host location by Aedes aegypti (Diptera Culicidae): a wind tunnel study of chemical cues.
Bull Entomol Res 81: 151-160.
Ellin RI, Farrand RL, Oberst FW, Crouse CL, Billups NB, Koon WS, Musselman NP and Sidell FR (1974) An apparatus for the
detection and quantitation of volatile human effluents. J Chromatogr 100: 137-152.
Gallagher M, Wysocki CJ, Leyden JJ, Spielman AI, Sun X and Preti G (2008) Analyses of volatile organic compounds from
human skin. Br J Dermatol 159:780-791.
Geier M and Boeckh J (1999) A new Y-tube olfactometer for mosquitoes to measure the attractiveness of host odours.
Entomol Exp Appl 92: 9-19.

172  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Geier M, Bosch OJ and Boeckh J (1999a) Ammonia as an attractive component of host odour for the Yellow fever
mosquito Aedes aegypti. Chem. Senses 24: 647-653.
Geier M, Bosch OJ and Boeckh J (1999b) Influence of odour plume structure on upwind flight of mosquitoes towards
hosts. J Exp Biol 202: 1639-1648.
Geier M, Sass H and Boeckh J (1996) A search for components in human body odour that attract female Aedes aegypti.
In: Bock GR and Cardew G (eds) Olfaction in mosquito host interactions. John Wiley and Sons Ltd, Chichester, UK,
pp 132-148.
Gillies MT (1964) Selection for host preference in Anopheles gambiae. Nature 203: 852-854.
Gillies MT (1988) Anopheline mosquitos: vector behaviour and bionomics. In: Wernsdorfer WH and McGregor I (eds)
Malaria - principles and practice of malariology. Churchill Livingstone, Edinburgh, UK, pp 453-485.
Godsey MS, Jr., Blackmore MS, Panella NA, Burkhalter K, Gottfried K, Halsey LA, Rutledge R, Langevin SA, Gates R,
Lamonte KM, Lambert A, Lanciotti RS, Blackmore CG, Loyless T, Stark L, Oliveri R, Conti L and Komar N (2005) West
Nile virus epizootiology in the southeastern United States, 2001. Vector Borne Zoonotic Dis 5: 82-89.
Gould EA (2003) Implications for Northern Europe of the emergence of West Nile virus in the USA. Epidemiology and
Infection 131: 583-589.
Gould LH and Fikrig E (2004) West Nile virus: a growing concern? J Clin Invest 113:1102-1107.
Gratz NG (2004) Critical review of the vector status of Aedes albopictus. Med Vet Entomol 18: 215-227.
Greenwood BM (2008) Control to elimination: implications for malaria research. Trends in Parasitology 24: 449-454.
Greenwood BM, Fidock DA, Kyle DE, Kappe SHI, Alonso PL, Collins FH and Duffy PE (2008) Malaria: progress, perils, and
prospects for eradication. J Clin Invest 118: 1266-1276.
Grieco JP, Achee NL, Sardelis MR, Chauhan KR and Roberts DR (2005) A novel high-throughput screening system to
evaluate the behavioral response of adult mosquitoes to chemicals. J Am Mosq Control Assoc 21: 404-411.
Haathi EOA and Fales HM (1967) The uropygiols: identification of the unsaponifiable constituent of a diester wax from
chicken preen glands. J Lipid Research 8: 131-137.
Haddow AJ (1942) The mosquito fauna and climate of native huts at Kisumu, Kenya. Bull Entomol Res 33: 91-142.
Haddow AJ and Ssenkubuge Y (1973) The mosquito of bwamba county, Uganda. IX. Further studies on the biting
behaviour of an outdoor population of the Anopheles gambiae Giles complex. Bull Entomol Res 62:407-414.
Hallem EA, Dahanukar A and Carlson JR (2006) Insect odor and taste receptors. Annu Rev Entomol 51: 113-135.
Hallem EA, Fox AN, Zwiebel LJ and Carlson JR (2004) Olfaction - Mosquito receptor for human-sweat odorant. Nature
427: 212-213.
Hao H, Wei J, Dai J and Du J (2008) Host-seeking and blood-feeding behavior of Aedes albopictus (Diptera: Culicidae)
exposed to vapors of geraniol, citral, citronellal, eugenol, or anisaldehyde. J Med Entomol 45: 533-539.
Harrington LC (2001) Why do female Aedes aegypti (Diptera: Culicidae) feed preferentially and frequently on human
blood? J Med Entomol 38: 411-422.
Hasegawa Y, Yabuki M and Matsukane M (2004) Identification of new odoriferous compounds in human axillary sweat.
Chem Biodivers 1: 2042-2050.
Hassanali A, Herren H, Khan ZR, Pickett JA and Woodcock CM (2008) Integrated pest management: the push–pull
approach for controlling insect pests and weeds of cereals, and its potential for other agricultural systems including
animal husbandry. Phil Trans R Soc B 363: 611-621.
Haze S, Gozu Y, Nakamura S, Khono Y, Sawano K, Ohta H and Yamzaki K (2001) 2-Nonenal newly found in human body
odor tends to increase with aging. J Invest Dermatol 116: 520-524.
Healy TP, Copland MJ, Cork A, Przyborowska A and Halket JM (2002) Landing responses of Anopheles gambiae elicited
by oxocarboxylic acids. Med Vet Entomol 16: 126-132.
Healy TP and Copland MJW (1995) Activation of Anopheles gambiae mosquitoes by carbon dioxide and human breath.
Med Vet Entomol 9: 33 1-336.
Healy TP and Copland MJW (2000) Human sweat and 2-oxopentanoic acid elicid a landing response from Anopheles
gambiae. Med Vet Entomol 14: 195-200.
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides
R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P,

Olfaction in vector-host interactions 173

Renate C. Smallegange and Willem Takken

Atkinson PW, Baden H, de Berardinis V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M,
Cai S, Center A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A,
Dew I, Evans CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun
SL, Hogan JR, Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin
JJ, Lobo NF, Lopez JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O’Brochta DA,
Pfannkoch C, Qi R, Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J,
Thomasova D, Ton LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman
JR, Wu M, Yao A, Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della Torre A, Roth CW,
Louis C, Kalush F, Mural RJ, Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey
PT, Venter JC, Weissenbach J, Kafatos FC, Collins FH and Hoffman SL (2002) The genome sequence of the malaria
mosquito Anopheles gambiae. Science 298: 129-149.
Hurd H (2003) Manipulation of medically important insect vectors by their parasites. Annu Rev Entomol 48: 141-146.
Issachar D, Holland JF and Sweeley CC (1982) Metabolic profiles of organic acids from human plasma. Analytical
Chemistry 54: 29-32.
Jawara M, Smallegange RC, Jeffries D, Nwakanma DC, Awolola TS, Knols BGJ, Takken W and Conway DJ (2009).
Optimising odour baited trap methods for collecting mosquitoes during the malaria season in The Gambia. PLoS
ONE 12: e8167.
Jones KE, Patel NG, Levy MA, Storeygard A, Balk D, Gittleman JL and Daszak P (2008) Global trends in emerging infectious
diseases. Nature 451: 990-993.
Jordan AM (1986) Trypanosomiasis control and African rural development. Longman Group Ltd., New York, USA.
Justice RW, Biessmann H, Walter MF, Dimitratos SD and Woods DF (2003) Genomics spawns novel approaches to
mosquito control; Bioassays 25: 1011-1020.
Kennedy JS (1978) The concepts of olfactory ‘arrestment’ and ‘attraction’. Physiol. Entomol 3: 91-98.
Killeen G, Kihonda J, Lyimo E, Oketich F, Kotas M, Mathenge E, Schellenberg J, Lengeler C, Smith T and Drakeley C(2006)
Quantifying behavioural interactions between humans and mosquitoes: evaluating the protective efficacy of
insecticidal nets against malaria transmission in rural Tanzania. BMC Infect. Dis. 6:161.
Kilpatrick AM, Kramer LD, Jones MJ, Marra PP and Daszak P (2006) West Nile virus epidemics in North America are driven
by shifts in mosquito feeding behavior. PLoS Biol 4: e82.
Kline DL, Allan SA, Bernier UR and Posey KH (2006) Olfactometer and large cage evaluation of a solid phase technology
for the controlled production of CO2. J Am Mosq Control Assoc 22: 378-381.
Kline DL (2006) Traps and trapping techniques for adult mosquito control. J Am Mosq Control Assoc 22: 490-496.
Kline DL (2007) Semiochemicals, traps targets and mass trapping technology for mosquito management. J Am Mosq
Control Assoc 23: 241-251.
Kline DL, Bernier UR, Posey KH, and Barnard DR (2003) Olfactometric evaluation of spatial repellents for Aedes aegypti.
J Med Entomol 40: 463-467.
Klowden MJ (1986) Effects of sugar deprivation on the host-seeking behaviour of gravid Aedes aegypti mosquitoes. J
Insect Physiol 32: 479-483.
Klowden MJ (1995) Blood, sex and mosquito. The mechanism that control mosquito blood feeding behavior. Bioscience
45: 326-331.
Klowden MJ (1996) Endogenous factors regulating mosquito host-seeking behaviour. In: Bock GR and Cardew G (eds)
Olfaction in mosquito interactions. Wiley, Chichester, UK, pp 212-225.
Klowden and Lea (1979) Humoral inhibition of host-seeking in Aedes aegypti during oocyte maturation. J Insect Physiol
25: 231–235.
Klun JA, Kramer M, Debboun M: (2005) A new in vitro bioassay system for discovery of novel human-use mosquito
repellents. J Am Mosq Control Assoc 21:64-70.
Knols BG, Njiru BN, Mathenge EM, Mukabana WR, Beier JC and Killeen GF (2002) MalariaSphere: A greenhouse-enclosed
simulation of a natural Anopheles gambiae (Diptera: Culicidae) ecosystem in western Kenya. Malar J 1: 19.
Knols BG, Takken W and De Jong R (1994a) Influence of human breath on selection of biting sites by Anopheles
albimanus. J Am Mosq Control Assoc 10: 423-426.

174  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Knols BGJ, De Jong R and Takken W (1994b) Trapping system for testing olfactory responses of the malaria mosquito
Anopheles-gambiae in a wind-tunnel. Med Vet Entomol 8: 386-388.
Knols BGJ, Mboera LEG and Takken W (1998) Electric nets for studying odour-mediated host-seeking behaviour of
mosquitoes. Med Vet Entomol 12: 116-120.
Knols BGJ, Takken W, Charlwood D and De Jong R (1995) Species-specific attraction of Anopheles mosquitoes (Diptera:
Culicidae) to different humans in South Tanzania. Proceedings of Experimental and Applied Entomology, N.E.V.,
Amsterdam 6: 201-206.
Knols BGJ, Van Loon JJA, Cork A, Robinson RD, Adam W, Meijerink J, De Jong R and Takken W (1997) Behavioural
and electrophysiological responses of the female malaria mosquito Anopheles gambiae (Diptera: Culicidae) to
Limburger cheese volatiles. Bull Entomol Res 87: 151-159.
Kröckel U, Rose A, Eiras AE and Geier M (2006) New tools for surveillance of adult yellow fever mosquitoes: comparison
of trap catches with human landing rates in an urban environment. J Am Mosq Control Assoc 22: 229-238.
Krotoszynski B, Gabriel G and O’Neill H (1977) Characterization of human expired air: A promising investigative and
diagnostic technique. J Chromatogr Sci 15: 239-244.
Leal WS, Barbosa RMR, Xu W, Ishida Y, Syed Z, Latte N, Chen AM, Morgan TI, Cornel AJ and Furtado A (2008) Reverse and
conventional chemical ecology approaches for the development of oviposition attractants for Culex mosquitoes.
PLos ONE 3: e3045.
Lefèvre T, Gouagna L-C, Dabire KR, Elguero E, Fontenille D, Costantini C and Thomas F (2009) Evolutionary lability of
odour-mediated host preference by the malaria vector Anopheles gambiae. Trop Med Int Health 14:228-236.
Lehane MJ (1991) Biology of blood-sucking insects. Harper Collins Academic, London, UK. 288 pp.
Leyden JJ, McGinley KJ, Holzle E, Labows JN and Kligman AM (1981) The microbiology of the human axilla and its
relationship to axillary odor. J Investig Dermatol 77: 413-416.
Lindsay SW, Adiamah JH, Miller JE, Pleass RJ and Armstrong JR (1993) Variation in attractiveness of human subjects to
malaria mosquitoes (Diptera: Culicidae) in The Gambia. J Med Entomol 30: 368-373.
Logan JG and Birkett MA (2007) Semiochemicals for biting fly control: their identification and exploitation. Pest Manage
Sci 63: 647-657.
Logan JG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue AJ and Pickett JA (2008) Identification of human-
derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. J Chem Ecol 34: 308-322.
Lu T, Qiu YT, Wang G, Kwon JY, Rutzler M, Kwon HW, Pitts RJ, Van Loon JJA, Takken W, Carlson JR and Zwiebel LJ (2007)
Odor coding in the maxillary palp of the malaria vector mosquito Anopheles gambiae. Current Biology 17: 1533-
1544.
Maxwell CA, Wakibara J, Tho S and Curtis CF (1998) Malaria-infective biting at different hours of the night. Med Vet
Entomol 12:325-327.
Mboera LEG, Knols BGJ, Braks MAH and Takken W (2000) Comparison of carbon dioxide-baited trapping systems for
sampling outdoor mosquito populations in Tanzania. Med Vet Entomol 14: 257-263.
Mboera LEG, Knols BGJ, Takken W and Huisman PWT (1998) Olfactory responses of female Culex quinquefasciatus Say
(Diptera: Culicidae) in a dual-choice olfactometer. J Vector Ecol 23: 107-113.
Mboera LEG and Takken W (1997) Carbon dioxide chemotropism in mosquitoes (Diptera: Culicidae) and its potential in
vector surveillance and management programmes. Rev Med Vet Entomol 85: 355-368.
Mboera LEG and Takken W (1999) Odour-mediated host preference of Culex quinquefasciatus in Tanzania. Entomol Exp
Appl 92: 83-88.
McCall PJ, Harding JR, Roberts J and Auty B (1996) Attraction and trapping of Aedes aegypti (Diptera: Culicidae) with
host odors in the laboratory. J Med Entomol 33:177-179.
McCall PJ and Eaton G (2001) Olfactory memory in the mosquito Culex quinquefasciatus. Med Vet Entomol 15: 197-203.
McIver SB: Host preferences and discrimination by the mosquitoes Aedes aegypti and Culex tarsalis (Diptera: culicidae).
J Med Entomol 1968, 5:422-428.
Mehr ZA, Rutledge LC, Buescher MD, Gupta RK and Zakaria MM (1990) Attraction of mosquitoes to diethyl
methylbenzamide and ethyl hexanediol. J Am Mosq Control Assoc 6: 469-476.

Olfaction in vector-host interactions 175

Renate C. Smallegange and Willem Takken

Meijerink J, Braks MAH, Brack AA, Adam W, Dekker T, Posthumus MA, Van Beek TA and Van Loon JJA (2000) Identification
of olfactory stimulants for Anopheles gambiae from human sweat samples. J Chem Ecol 26: 1367-1382.
Morrison AC, Zielinski-Gutierrez E, Scott TW and Rosenberg R (2008) Defining challenges and proposing solutions for
control of the virus vector Aedes aegypti. Plos Med 5: e68.
Mukabana WR, Takken W, Coe R and Knols BG (2002) Host-specific cues cause differential attractiveness of Kenyan men
to the African malaria vector Anopheles gambiae. Malar J 1: 17.
Mukabana WR, Takken W, Killeen GF and Knols BGJ (2004) Allomonal effect of breath contributes to differential
attractiveness of humans to the African malaria vector Anopheles gambiae. Malar J 3: 8.
Muturi EJ, Muriu S, Shililu J, Mwangangi JM, Jacob BG, Mbogo C, Githure J and Novak RJ (2008) Blood-feeding patterns
of Culex quinquefasciatus and other culicines and implications for disease transmission in Mwea rice scheme, Kenya.
Parasitol Res 102: 1329-1335.
Natsch A, Derrer S, Flachsmann F and Schmid J (2006) A broad diversity of volatile carboxylic acids, released by a
bacterial aminoacylase from axilla secretions, as candidate molecules for the determination of human-body odor
type. Chem Biodivers 3: 1-20.
Nene V, Wortman JR, Lawson D, Haas B, Kodira C, Tu ZJ, Loftus B, Xi Z, Megy K, Grabherr M, Ren Q, Zdobnov EM, Lobo
NF, Campbell KS, Brown SE, Bonaldo MF, Zhu J, Sinkins SP, Hogenkamp DG, Amedeo P, Arensburger P, Atkinson
PW, Bidwell S, Biedler J, Birney E, Bruggner RV, Costas J, Coy MR, Crabtree J, Crawford M, Debruyn B, Decaprio D,
Eiglmeier K, Eisenstadt E, El-Dorry H, Gelbart WM, Gomes SL, Hammond M, Hannick LI, Hogan JR, Holmes MH,
Jaffe D, Johnston JS, Kennedy RC, Koo H, Kravitz S, Kriventseva EV, Kulp D, Labutti K, Lee E, Li S, Lovin DD, Mao C,
Mauceli E, Menck CF, Miller JR, Montgomery P, Mori A, Nascimento AL, Naveira HF, Nusbaum C, O’Leary S, Orvis
J, Pertea M, Quesneville H, Reidenbach KR, Rogers YH, Roth CW, Schneider JR, Schatz M, Shumway M, Stanke M,
Stinson EO, Tubio JM, Vanzee JP, Verjovski-Almeida S, Werner D, White O, Wyder S, Zeng Q, Zhao Q, Zhao Y, Hill CA,
Raikhel AS, Soares MB, Knudson DL, Lee NH, Galagan J, Salzberg SL, Paulsen IT, Dimopoulos G, Collins FH, Birren B,
Fraser-Liggett CM and Severson DW (2007) Genome sequence of Aedes aegypti, a major arbovirus vector. Science
316: 1718-1723.
Nicolaides N, C. FH and Rice GR (1968) The skin surface lipids of man compared with those of eighteen species of
animals. J. Investigative Dermatology 51: 83-89.
Njiru BN, Mukabana WR, Takken W and Knols BG (2006) Trapping of the malaria vector Anopheles gambiae with odour-
baited MM-X traps in semi-field conditions in western Kenya. Malar J 5: 39.
Okumu FO, Killeen GF, Ogoma S, Biswaro L, Smallegange RC, Mbeyela E, Titus E, Munk C, Ngonyani H, Takken W, Mshinda
H, Mukabana WR and Moore SJ (2010) Development and field evaluation of a synthetic mosquito lure that is more
attractive than humans. PLoS ONE 5: e8951.
Pates HV, Takken W and Curtis CF (2005) Laboratory studies on the olfactory behaviour of Anopheles quadriannulatus.
Entomol Exp Appl 114: 153-159.
Pates HV, Takken W, Curtis CF, Huisman PW, Akinpelu O and Gill GS (2001a) Unexpected anthropophagic behaviour in
Anopheles quadriannulatus. Med Vet Entomol 15: 293-298.
Pates HV, Takken W, Stuke K and Curtis CF (2001b) Differential behaviour of Anopheles gambiae sensu stricto (Diptera:
Culicidae) to human and cow odours in the laboratory. Bull Entomol Res 91: 289-296.
Pelletier J and Leal WS (2009) Genome analysis and expression patterns of odorant-binding proteins from the southern
house mosquito Culex pipiens quinquefasciatus. PLoS ONE 4: e6237.
Penn DJ, Oberzaucher E, Grammer K, Fischer G, Soini HA, Wiesler D, Novotny MV, Dixon SJ, Xu Y and Brereton RG (2007)
Individual and gender fingerprints in human body odour. J R Soc Interface 4: 331-340.
Perry TL, Hansen S, Diamond S, Bullis B, Mok C and Melançon SB (1970) Volatile fatty acids in normal human physiological
fluids. Clin Chim Acta 29: 369-374.
Philips M (1997) Method for the collection and assay of volatile organic compounds in breath. Anal Biochem 247:
272-278.
Posey KH, Barbard DR and Schreck CE (1998) Triple cage olfactometer for evaluating mosquito (Diptera: Culicidae)
attraction responses. J Med Entomol 35: 330-334.

176  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Puri SN, Mendki MJ, Sukumaran D, Ganesan K, Prakash S and Sekhar K (2006) Electroantennogram and behavioral
responses of Culex quinquefasciatus (Diptera: Culicidae) females to chemicals found in human skin emanations. J
Med Entomol 43: 207-213.
Qiu YT (2005) Sensory and behavioural responses of the malaria mosquito Anopheles gambiae to human odours. PhD
thesis, Wageningen University, Wageningen, the Netherlands, 208 pp.
Qiu YT, Smallegange RC, Hoppe S, Van Loon JJA, Bakker EJ and Takken W (2004a) Behavioural and electrophysiological
responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera: Culicidae) to human skin
emanations. Med Vet Entomol 18: 429-438.
Qiu YT, Smallegange RC, Smid H, Van Loon JJA, Galimard AMS, Posthumus MA, Beek TAv and Takken W (2004b) GC-EAG
analysis of human odours that attract the malaria mosquito Anopheles gambiae sensu strict. Proc Exp Appl Ent,
N.E.V., Amsterdam 15: 59-64.
Qiu YT, Smallegange RC, Ter Braak CJF, Spitzen J, Van Loon JJA, Jawara M, Milligan P, Galimard AM, Van Beek TA, Knols BGJ
and Takken W (2007a) Attractiveness of MM-X traps baited with human or synthetic odor to mosquitoes (Diptera:
Culicidae) in The Gambia. J Med Entomol 44: 970-983.
Qiu YT, Spitzen J, Smallegange RC and Knols BGJ. (2007b). Monitoring systems for adult insect pests and disease vectors.
In: Takken W and Knols BGJ (eds) Emerging pests and vector-borne diseases in Europe. Ecology and control of
vector-botne diseases, vol. 1. Wageningen Academic Publishers, Wageningen, the Netherlands, pp 329-354.
Qiu YT, Smallegange RC, Van Loon JJA, Ter Braak CJF and Takken W (2006a) Interindividual variation in the attractiveness
of human odours to the malaria mosquito Anopheles gambiae s. s. Med Vet Entomol 20: 280-287.
Qiu YT, van Loon JJA, Takken W, Meijerink J and Smid HM. 2006b. Olfactory coding in antennal neurons of the malaria
mosquito, Anopheles gambiae. Chem Senses 31:845-863.
Richards SL, Ponnusamy L, Unnasch TR, Hassan HK and Apperson CS (2006) Host-feeding patterns of Aedes albopictus
(Diptera: Culicidae) in relation to availability of human and domestic animals in suburban landscapes of central
North Carolina. J Med Entomol 43: 543-551.
Robertson HM and Wanner KW (2006) The chemoreceptor superfamily in the honey bee, Apis mellifera: expansion of
the odorant, but not gustatory, receptor family. Genome Res 16: 1395-1403.
Rudolfs W (1922) Chemotropism of mosquitoes. Bull NJ Agric Exp Stn 367: 4-23.
Rutledge LC and Gupta RK (2004) Evaluation of an in vitro bloodfeeding system for testing mosquito repellents. J Am
Mosq Control Assoc 20: 150-154.
Rutzler M and Zwiebel LJ (2005) Molecular biology of insect olfaction: recent progress and conceptual models. J Comp
Physiol A Neuroethol Sens Neural Behav Physiol 191: 777-790.
Sardelis MR, Turell MJ, Dohm DJ and O’Guinn ML (2001) Vector competence of selected North American Culex and
Coquillettidia mosquitoes for West Nile virus. Emerging Infectious Diseases 7: 1018-1022.
Sastry SD, Buck KT, Janak J, Dressler M and Preti G (1980) Volatiles emitted by humans. In: Waller GR and Dermer OC (eds)
Biochemical applications of mass spectrometry. John Wiley and Sons, New York, USA, pp 1085-1127.
Savage HM, Aggarwal D, Apperson CS, Katholi CR, Gordon E, Hassan HK, Anderson M, Charnetzky D, McMillen L, Unnasch
EA and Unnasch TR (2007) Host choice and West Nile virus infection rates in blood-fed mosquitoes, including
members of the Culex pipiens complex, from Memphis and Shelby County, Tennessee, 2002-2003. Vector Borne
Zoonotic Dis 7: 365-386.
Schaub GA (2006) Parasitogenic alterations of vector behaviour. Int J Med Microbiol 296: 37-40.
Schofield S and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol on the flight responses of the stable
fly, Stomoxys calcitrans, in a windtunnel. Physiol Entomol 22: 380-386.
Schofield S, Witty C and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol on the activity of the stable
fly, Stomoxys calcitrans. Physiol Entomol 22: 256-260.
Schofield S, Cork A and Brady J (1995) Electroantennogram responses of the stable fly, Stomoxys calcitrans, to
components of host odor. Physiol Entomol 20: 273-280.
Schmied WH, Takken W, Killeen GF, Knols BGJ and Smallegange RC (2008) Evaluation of two counterflow traps for testing
behaviour-mediating compounds for the malaria vector Anopheles gambiae s.s. under semi-field conditions in
Tanzania. Malar J 7:230.

Olfaction in vector-host interactions 177

Renate C. Smallegange and Willem Takken

Schreck CE, Kline DL and Carlson DA (1990) Mosquito attraction to substances from the skin of different humans. J Am
Mosq Control Assoc 6: 406-410.
Schreck CE, Smith N, Carlson DA, Price GD, Haile D and Godwin DR (1981) A material isolated from human hands that
attracts female mosquitoes. J Chem Ecol 2: 429-438.
Scott TW, Githeko AK, Fleisher A, Harrington LC and Yan GY (2006) DNA profiling of human blood in anophelines from
lowland and highland sites in western Kenya. Am J Trop Med Hyg 75: 231-237.
Shelley WWB, Hurley HHJ Jr., Nichols AAC (1953) Axillary odor; experimental study of the role of bacteria, apocrine
sweat, and deodorants. AMA Arch Derm Syphilol 68: 430-446.
Shirai Y, Kamimura K, Seki T and Morohashi M (2001) L-lactic acid as a mosquito (Diptera: Culicidae) repellent on human
and mouse skin. J Med Entomol 38: 51-54.
Silva IM, Eiras AE, Kline DL and Bernier UR (2005) Laboratory evaluation of mosquito traps baited with a synthetic human
odor blend to capture Aedes aegypti. J Am Mosq Control Assoc 21: 229-233.
Smallegange RC, Geier M and Takken W (2002) Behavioural responses of Anopheles gambiae to ammonia, lactic acid
and a fatty acid in a y-tube olfactometer. Proc Exp Appl Ent, N.E.V., Amsterdam 13: 147-152.
Smallegange RC, Knols BGJ and Takken W (2010) Effectiveness of synthetic versus natural human volatiles as attractants
for Anopheles gambiae (Diptera: Culicidae) sensu stricto. J Med Entomol 47: 338-344.
Smallegange RC, Qiu YT, Bukovinszkiné-Kiss G, Van Loon JJA and Takken W (2009) The effect of aliphatic carboxylic acids
on olfaction-based host-seeking of the malaria mosquito Anopheles gambiae sensu stricto. J Chem Ecol 35: 933-943.
Smallegange RC, Qiu YT, Galimard AMS, Posthumus MA, Van Beek TA, Van Loon JJA and Takken W (2003) Why humans
are attractive to malaria mosquitoes. Entomol Berichten 63: 50-53.
Smallegange RC, Qiu YT, Van Loon JJA and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic
acids as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto
(Diptera: Culicidae). Chem Senses 30: 145-152.
Smith CN, Smith N, Gouck HK, Weidhaas DE, Gilbert IH, Mayer MS, Smittle BJ and Hofbauer A (1970) L-Lactic acid as
a factor in the attraction of Aedes aegypti (Diptera: Culicidae) to human hosts. Ann Entomol Soc Am 63: 760-770.
Spinhirne JP, Koziel JA and Chirase NK (2003) A Device for Non-invasive On-site Sampling of Cattle Breath with Solid-
Phase Microextraction. Biosystems Engineering 84: 239–246.
Spinhirne JP, Koziel JA and Chirase NK (2004) Sampling and analysis of volatile organic compounds in bovine breath by
solid-phase microextraction and gas chromatography-mass spectrometry. J Chromatogr A 1025: 63-69.
Spitzen J, Smallegange RC and Takken W (2008) Effect of human odours and positioning of CO2 release point on trap
catches of the malaria mosquito Anopheles gambiae sensu stricto in an olfactometer. Physiol Entomol 33: 116-122.
Strickman D (2007) Area repellent products. In: Debboun M, Frances SP and Strickman D (eds) Insect repellents.
Principles, methods, and uses. CRC Press, Boca Raton, FL, USA, pp 385-395.
Syed Z and Leal WS (2008) Mosquitoes smell and avoid the insect repellent DEET. PNAS 105: 13598-13603.
Syed Z and Leal WS (2009) Acute olfactory response of Culex mosquitoes to a human- and bird-derived attractant.
PNAS 106: 18809-18814
Takken W (1991) The role of olfaction in host-seeking of mosquitoes: a review. Insect Sci Appl 12: 287-295.
Takken W (1996) Synthesis and future challenges: the response of mosquitoes to host odours. In: Bock GR and Cardew
G (eds) Olfaction in mosquito host interactions. John Wiley and Sons Ltd, Chichester, UK, pp 302-320.
Takken W, Dekker T and Wijnholds YG (1997) Odor-mediated flight behavior of Anopheles gambiae Giles sensu stricto and
An. stephensi Liston in response to CO2, acetone, and 1-Octen-3-ol (Diptera: Culicidae). J Insect Behav 10: 395-407.
Takken W, Eling W, Hooghof JDT, Hunt R and Coetzee M (1999) Susceptibility of Anopheles quadriannulatus Theobald
(Diptera: Culicidae) to Plasmodium falciparum. Trans R Soc Trop Med Hyg 93: 578-580.
Takken W and Kline DL (1989) Carbon dioxide and 1-octen-3-ol as mosquito attractants. J Am Mosq Control Assoc 5:
311-316.
Takken W and Knols BGJ (1999) Odor-mediated behavior of afrotropical malaria mosquitoes. Ann Rev Entomol 44:
131-157.
Takken W and Knols BGJ (2007) Emerging pests and vector-borne diseases in Europe. Ecology and control of vector-
borne diseases Vol. 1. Wageningen Academic Publishers, Wageningen, the Netherlands, 499 pp.

178  Olfaction in vector-host interactions

 7. Host-seeking behaviour of mosquitoes: responses to olfactory stimuli in the laboratory

Takken W and Knols BGJ (2009) Malaria vector control: Current and future strategies. Trends Parasitol 25: 101-104.
Takken W, Van Loon JJA and Adam W (2001) Inhibition of host-seeking response and olfactory responsiveness in
Anopheles gambiae following blood feeding. J Insect Physiol 47: 303-310.
Tomberlin J, Rains G, Allan S, Sanford M and Lewis W (2006) Associative learning of odor with food- or blood-meal by
Culex quinquefasciatus Say (Diptera: Culicidae). Naturwissenschaften 93:551-556.
Torr S, Della Torre A, Calzetta M, Costantini C and Vale GA (2008) Towards a fuller understanding of mosquito
behaviour: use of electrocuting grids to compare the odour-orientated responses of Anopheles arabiensis and An.
quadriannulatus in the field. Med Vet Entomol 22: 93-108.
Torr SJ, Hall DR and Smith JL (1995) Responses of tsetse flies (Diptera: Glossinidae) to natural and synthetic ox odours.
Bull Entomol Res 85: 157-166.
Turell MJ, Dohm DJ, Sardelis MR, guinn ML, Andreadis TG and Blow JA (2005) An update on the potential of North
American mosquitoes (Diptera: Culicidae) to transmit West Nile virus. J Med Entomol 42: 57-62.
Vale GA (1980) Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to carbon dioxide, acetone
and other chemicals. Bull Entomol Res 70: 563-570.
Vale GA (1993) Development of baits for tsetse flies (Diptera: Glossinidae) in Zimbabwe. J Med Entomol 30: 831-842.
Vale GA and Hall DR (1985b) The use of 1-octen-3-ol, acetone and carbon dioxide to improve baits for tsetse flies,
Glossina spp. (Diptera: Glossinidae). Bull Entomol Res 75: 219-231.
Vale GA and Hall DR 1985a The role of 1-octen-3-ol, acetone and carbon dioxide in the attraction of tsetse flies, Glossina
spp. (Diptera: Glossinidae), to ox odour. Bull Entomol Res 75: 209-217.
Vale GA, Hall DR and Gough AJE (1988) The olfactory responses of tsetse flies, Glossina spp. (Diptera: Glossinidae), to
phenols and urine in the field. Bull Entomol Res 78: 293-300.
Van der Goes van Naters W and Carlson JR (2006) Insects as chemosensors of humans and crops. Nature 444: 302-307.
Van Thiel P (1937) Quelles sont les excitations incitant l’Anopheles maculipennis atroparvus a visiter et a piquer l’homme
ou le batail? Bulletin Soc Path Exot 30: 193-203.
Vazeille M, Jeannin C, Martin E, Schaffner F and Failloux AB (2008) Chikungunya: a risk for Mediterranean countries?
Acta Trop 105: 200-202.
Verhulst NO, Beijleveld H, Knols BGJ, Takken W, Schraa G, Bouwmeester HJ and Smallegange RC (2009). Cultured skin
microbiota attracts malaria mosquitoes. Malar J 8: 302.
Waka M, Hopkins RJ, Glinwood R and Curtis C (2006) The effect of repellents Ocimum forskolei and deet on the response
of Anopheles stephensi to host odours. Med Vet Entomol 20: 373-376.
Wang G, Carey AF, Carlson JR and Zwiebel LJ (in press) The molecular basis of odor coding in the malaria vector mosquito
Anopheles gambiae. PNAS (in press).
Wang ZY, Mo JC and Zhang SM (2006) Laboratory and field evaluations of potential human host odors for Aedes
albopictus Skuse (Diptera: Culicidae). J Agric Urban Entomol 23: 57-64.
Waterhouse RM, Wyder S and Zdobnov EM (2008) The Aedes aegypti genome: a comparative perspective. Insect Mol
Biol 17: 1-8.
White GB (1974) Anopheles gambiae complex and disease transmission in Africa. Trans R Soc Trop Med Hyg 68: 278-301.
White GB (2007) Terminology of insect repellents. In: Debboun M, Frances SP and Strickman D (eds) Insect repellents.
Principles, methods, and uses. CRC Press, Boca Raton, FL, USA, pp 31-46.
WHO (2009) Malaria. WHO factsheet No 94. World Heath Organisation, Geneva, Switzerland.
WHO (2008) Dengue and dengue haemorrhagic fever. WHO factsheet No 117. World Heath Organisation, Geneva,
Switzerland.
Williams CR, Bergbauer R, Geier M, Kline DL, Bernier UR, Russell RC and Ritchie SA (2006a) Laboratory and field
assessment of some kairomone blends for host-seeking Aedes aegypti. J Am Mosq Control Assoc 22: 641-647.
Williams CR, Kokkinn MJ and Smith BP (2003) Intraspecific Variation in odor-mediated host preference of the mosquito
Culex annulirostris. J Chem Ecol 29: 1889-1903.
Williams CR, Long SA, Russell RC and Ritchie SA (2006b) Field efficacy of the BG-Sentinel compared with CDC Backpack
Aspirators and CO2-baited EVS traps for collection of adult Aedes aegypti in Cairns, Queensland, Australia. J Am
Mosq Control Assoc 22: 296-300.

Olfaction in vector-host interactions 179

Renate C. Smallegange and Willem Takken

Williams CR, Ritchie SA, Russell RC, Eiras AE, Kline DL and Geier M (2006c) Geographic variation in attraction to human
odor compounds by Aedes aegypti mosquitoes (Diptera: Culicidae): a laboratory study. J Chem Ecol 32: 1625-1634.
Young S, Hardie J and Gibson G (1993) Flying insects in the laboratory. In: Wratten SD (ed) Video techniques in animal
ecology and behaviour. Chapman and Hall, London, UK, pp 17-32.
Zeng C, Leyden JJ, Spielman AI, and Preti G (1996) Analysis of characteristic human female axillary odors: qualitative
comparison to males. J Chem Ecol 22: 237-257.
Zeng XN, Leyden JJ, Lawley HJ, Sawano K, Nohara I and Preti G (1991) Analysis of characteristic odors from human male
axillae. J Chem Ecol 17: 1469-1493.
Zinser M, Ramberg F and Willott E (2004) Culex quinquefasciatus (Diptera: Culicidae) as a potential West Nile virus vector
in Tucson, Arizona: blood meal analysis indicates feeding on both humans and birds. J Insect Sci 4: 20.
Zwiebel LJ and Takken W (2004) Olfactory regulation of mosquito-host interactions. Insect Biochem Mol Biol 34: 645-
652.

180  Olfaction in vector-host interactions

8. H
 ost-seeking behaviour of Afrotropical anophelines: field and
semi-field studies

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Abstract

The search for a synthetic mosquito attractant based on one or more human-derived kairomones
has been the goal of many laboratory studies. Besides alleviating the occupational risk to which
volunteers participating in vector surveillance are subjected whilst performing landing catches,
discovery of potent attractants also underpins the development and deployment of mass
trapping devices for controlling transmission of mosquito-borne diseases. Whereas a few potential
synthetic attractants have recently been developed and tested, not much has been done under
open field conditions. Odour delivery methodologies are still needed and the potency of available
attractants is too small at present to warrant full-scale application of this technology for malaria
vector surveillance and control. Cheap traps and trapping devices that operate energy-free or
utilise an affordable or renewable source of energy to power them, a cheap source of carbon
dioxide, a cheap and easy method for attractant delivery, and discovery of generic attractants
that can trap multiple vectors need to be developed. Only then shall the potential impact of this
technology on incidence and prevalence of various vector-borne diseases come to fruition.

Keywords: Anopheles, malaria, kairomones, host-seeking behaviour, Africa

Introduction

About one million deaths and close to five hundred million clinical episodes of malaria occur
throughout the world each year (Breman and Holloway 2007). Ninety percent of malaria-related
deaths occur in sub-Saharan Africa (Boutin et al. 2005), mainly among pregnant women and in
children below five years of age (Philips 2001). Malaria hampers socioeconomic development by
interfering with savings and investment, lowering workforce productivity, promoting absenteeism,
causing premature mortality and increasing medical costs (Sachs and Malaney 2002). This makes
malaria-endemic countries not only poor but also associates them with lower and slower rates
of economic growth. The global distribution of per-capita gross domestic product (GDP) shows a
distinct correlation between malaria and poverty. The reverse is true for countries where malaria
has been eradicated (Gallup and Sachs 2001).

Human malarias are transmitted by Anopheles mosquitoes and are caused by single or multiple
infections of four species of Plasmodium parasites. These include P. falciparum, P. malariae, P. ovale
and P. vivax. Recently, P. knowlesi has been reported to cause malaria in Southeast Asia (Cox-
Singh et al. 2007, Kim-Sung et al. 2009). While P. vivax is most widely distributed, P. falciparum,
which is the most widespread Plasmodium species in sub-Saharan Africa, causes the most severe
complications. The main vectors of malaria in Africa are mosquitoes belonging to the Anopheles
gambiae and An. funestus complexes.

The An. gambiae complex consists of seven morphologically indistinguishable sibling species
namely An. gambiae s.s., An. arabiensis, An. bwambae, An. merus, An. melas, An. quadriannulatus
species A (Coetzee et al. 2000, White 1974) and An. quadriannulatus species B (Hunt et al. 1998).
Anopheles gambiae, the nominal taxon, and An. arabiensis are the most important vector species

Olfaction in vector-host interactions 181

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

in the complex (Coetzee et al. 2000, Gillies and De Meillon 1968, Gillies and Coetzee 1987, White
1974). The occurrence of 80% of the world’s malaria in tropical Africa (WHO 1993) is due to the
strong human biting habits (anthropophily) and relatively long lifespan of these two species
(Collins and Besansky1994), besides An. funestus which has similar biological characteristics.
Anopheles gambiae is a more efficient vector because of its endophilic and anthropophilic
characteristics (Costantini et al. 1999). The vectorial capacity of An. arabiensis is slightly lower
than that of Anopheles gambiae because of its ability to feed on other animals when human
hosts are not available (Gillies and Coetzee 1987). Distinct chromosomal ‘forms’ of An. gambiae,
which are strongly associated with specific habitats, exist in West Africa. Three of these so-called
ecophenotypes (i.e. BAMAKO, MOPTI, and SAVANNAH) have been found to occur in sympatry
at numerous sites (Coluzzi et al. 1979). Two other members of the complex, namely An. melas
in West Africa and An. merus in East Africa, are localised vectors depending on their levels of
contact with people. Of the other sibling species, An. bwambae is responsible for localised malaria
transmission among the Bambute pygmies of Bwamba in Uganda whereas An. quadriannulatus
is not considered a vector, in spite of its competence to transmit P. falciparum (Takken and Knols
1999). The Ethiopian population of An. quadriannulatus was recently recognised as being distinct
(from its South African counterpart) and is designated An. quadriannulatus species B (Hunt et al.
1998).

The An. funestus complex consists of a group of nine species including An. funestus, An. rivulorum,
An. parensis, An. vaneedeni, An. leesoni, An. fuscivenosus, An. aruni, An. brucei and An. confusus
(Cohuet et al. 2004). Anopheles funestus is the main species within the An. funestus group that
transmits malaria (Cohuet et al. 2004, Gillies and De Meillon 1968, Wilkes et al. 1996). It is endophilic
and anthropophilic and is considered as one of the major vectors of malaria in Africa (Cohuet et al.
2004). Whereas An. funestus, An. rivulorum, and An. leesoni are widely distributed throughout sub-
Saharan Africa, the other members of the group are more locally distributed: An. parensis and An.
confusus are found in Eastern Africa, An. vaneedeni in the northern areas of South Africa, An. aruni
in Zanzibar, An. fuscivenosus in Zimbabwe, and An. brucei in Nigeria. The distribution, ecological
and behavioural aspects of the main vectors of malaria in Africa are presented in Table 1.

African malaria mosquitoes locate their blood meal hosts largely based on olfactory cues (Takken
and Knols 1999). Physical cues, encompassing heat and moisture, also play a role that is hitherto
not well understood (e.g. Mukabana et al. 2004, Olanga et al. 2010, Takken et al. 1997). Dissecting
and analysing the broad spectrum of human emanations (Costantini et al. 1993) can provide
an important basis for developing synthetic compounds or blends with desirable attractant
(Logan and Birkett 2007), repellent or attractant ‘masking’ properties (Logan et al. 2008). Several
compounds identified from human emanations have been demonstrated to exhibit attractant
properties under varying experimental conditions. In this chapter we explore the extent to
which odour baits have been employed to assess the host-seeking behaviour of African malaria
mosquitoes under semi-field and field conditions. Emphasis is placed on studies conducted in the
last decade, as earlier work is reviewed by Takken and Knols (1999).

Odour-mediated host-seeking behaviour of African malaria vectors in the field

Haddow (1942) was among the first to recognise the importance of host odours in the behaviour
of African malaria vectors. He demonstrated that human body odour attracted anophelines to
a house, and that the attraction was proportional to body mass. Since then numerous studies
demonstrated the role of human odour in the attraction of malaria vectors (reviewed in Takken
and Knols 1999). Few studies attempted to unravel the role of individual chemicals constituting

182  Olfaction in vector-host interactions

 
Table 1. Distribution, ecological and behavioural characteristics of the main vectors of malaria in Africa.

Species and distribution Ecology and behaviour Comments References

An. arabiensis and An. gambiae
Angola, Benin, Botswana, Burkina Adults: An. arabiensis is anthropophilic but rather The stamp of human Gillies and DeMeillon 1968,
Faso, Burundi, Cameroon, Cape opportunistic, sometimes preferring to feed on domestic activity, direct or Gillies and Coetzee 1987,
Verde, Central African Republic, animals rather than man; rests outdoors frequently, indirect, is implicit in Costantini et al. 1998, 1999, Van
Chad, Congo, Cote d’Ivoire, indoors more frequently; feeds outdoors and indoors most breeding sites. den Broek and Den Otter 1999,
Democratic Republic of Congo, frequently. An. gambiae is highly anthropophilic; mainly An. gambiae is common Coetzee et al. 2000, Duchemin
Djibouti, Eritrea, Guinea, Equatorial feeding and resting indoors. in the wet season et al. 2001, Gimnig et al. 2001,
Guinea, Eritrea, Ethiopia, Gabon, Larvae: breed in temporary pools of water that are and in high rainfall Dia et al. 2003, Pock Tsy et al.

Olfaction in vector-host interactions

Gambia, Ghana, Guinea Bissau, typically small, shallow, open and sunlit; the water is areas. In Madagascar 2003, Moffett et al. 2007
Kenya, Liberia, Madagascar, normally warm, could be turbid, and could contain both species are
Malawi, Mali, Mauritania, Mauritius, algae but no surface film and/or aquatic vegetation. The predominantly zoophilic.
Mozambique, Namibia, Niger, diversity of habitats could include burrow pits, drains,
Nigeria, Rwanda, Senegal, Sierra brick-pits, ruts, car tracks, foot and hoof prints, water
Leone, Somalia, South Africa, Sudan, holes, river overflows, pools left by receding rivers,
Tanzania, Togo, Uganda, Zambia, backwaters, rice fields, etc.
Zimbabwe
An. melas
Angola, Congo, Cote d’Ivoire, Adults: feed readily indoors and outdoors; highly High organic pollution Gillies and DeMeillon 1968,
Democratic Republic of Congo anthropophilic but with considerable levels of zoophily; is typical of the larval White 1974, Bryan et al. 1987,
Equatorial Guinea, Gambia, Guinea, endophily and exophily almost balance; outdoor resting habitats. The species Gillies and Coetzee 1987,
Gabon, Gambia, Ghana, Liberia, sites are mainly natural – shaded earth banks, tree bases is confined to the west Coetzee et al. 2000, Bogh et al.
Nigeria, Senegal, Sierra Leone / buttresses, termite mounds, sides of crab holes, and coast of Africa. 2001, Moffett et al. 2007
underneath vegetation.
Larvae: typically breed in salt- or brackish water habitats –
pools, ponds, lagoons, marshes and swamps.

183
8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies
 Table 1. Continued.

184 
Species and distribution Ecology and behaviour Comments References
An. merus
Madagascar, Kenya, Mauritius, Adults: can be caught resting indoors by day but are Water in pools Gillies and DeMeillon 1968,
Mozambique, Somalia, Seychelles, mainly exophilic; in the absence of domestic animals favoured by An. merus Mosha and Subra 1982, Mosha
South Africa, Swaziland, Tanzania they bite man readily both indoors and outdoors, but is sometimes black, and Petrarca 1983, Paskewitz
have very high zoophilic tendencies. showing a high degree et al. 1983, Sharp et al. 1984,
Larvae: most commonly found in brackish lagoons, of organic (vegetable) Mnzava and Kilama 1986,
ponds, swamps, pools and puddles. pollution. An. merus is Gillies and Coetzee 1987,
confined to the east Coetzee 1989, La Grange 1995,
coast of Africa. Van Rensburg et al. 1996,
Coetzee et al. 2000, Pock Tsy et
al. 2003, Moffett et al. 2007
An. quadriannulatus
Ethiopia, Malawi, Mozambique, Adults: are characteristically zoophagic and exophilic. Incriminated as a Mpofu 1985, Gillies and
South Africa, Tanzania (Zanzibar Larvae: same as for An. arabiensis and An. gambiae. competent vector in the Coetzee 1987, Coetzee 1989,
only), Zimbabwe laboratory but not (yet) Coetzee et al. 1993, La Grange
in the field. 1995, Van Rensburg et al. 1996,
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Hunt et al. 1998, Takken et al.

1999, Fettene et al. 2002, 2004,
Pates et al. 2006, Mzilahowa
et al. 2008, Moffett et al. 2007,
Sharp et al. 2007
An. coustani
Common in most parts of Sub- Adults: readily attack humans outdoors penetrating - Gillies and DeMeillon 1968,
Saharan Africa down to the Cape indoors only to a limited extent; conversely they readily Gillies and Coetzee 1987,
Province in South Africa. Also enter stables and animal-baited traps; females rest Logan et al. 1991, Adugna and
present in Madagascar and Mauritius primarily in tall vegetation; biting activity is fairly evenly Petros 1996, Antonio-Nkondjio
spread out over the whole night. et al. 2002, 2005, Dia 2005,
Larvae: breeds in natural collections of clear water with Muturi et al. 2006, Moffett et
aquatic and semi-aquatic vegetation such as swamps, al. 2007
ponds, backwaters of streams, springs, ditches and rice
fields.

Olfaction in vector-host interactions

 
Table 1. Continued.

Species and distribution Ecology and behaviour Comments References

An. funestus
Angola, Botswana, Burkina Faso, Adults: highly prefer to feed and rest indoors; very highly Larval habitats typically Watson 1953, Bruce-Chwatt
Burundi, Cameroon, Central African anthropophilic. contain emergent and Gockel 1960, Gillies and
Republic, Chad, Congo, Cote Larvae: breed typically in large, less turbid, semi- vegetation (short grass, DeMeillon 1968, Gillies and
d’Ivoire, Democratic Republic of permanent water bodies [swamps (near edges if deep), tall reeds, trailing plants, Coetzee 1987, Gimnig et al.
Congo, Guinea, Eritrea, Ethiopia, weedy banks of streams, rivers, furrows or ditches, etc.). Larvae difficult 2001, Moffett et al. 2007,
Gabon, Gambia, Ghana, Kenya, protected portions of lake shores, ponds, etc.] or in to find: they dive and Jawara et al. 2009
Liberia, Madagascar, Malawi, Mali, water such as seepages which are fed from underground remain under water for

Olfaction in vector-host interactions

Mauritania, Nigeria, Rwanda, permanent sources. Shade from trees or bush is not prolonged periods.
Senegal, Sierra Leone, Somalia, critically important.
South Africa, Sudan, Tanzania,
Togo, Uganda, Zambia, Zimbabwe,
Equatorial Guinea
An. moucheti
Cameroon, Democratic Republic of Adults: highly anthropophilic/domestic; feed on man An. moucheti is Gillies and Coetzee 1987,
Congo, Ecuatorial Guinea, Gabon, indoors; rest to variable extents indoors. essentially a forest Adugna and Petros 1996, Elissa
Kenya, Nigeria, Uganda Larvae: are found along the borders of slow moving species and an et al. 1999, Sylla et al. 2000,
streams and large rivers, or in pools and ponds where important vector of Antonio-Nkondjio et al. 2002,
the water movement may be imperceptible. Although malaria in almost all 2005, Cano et al. 2006, Muturi
restricted to forest areas the larvae are always found areas where it occurs. et al. 2006, Moffett et al. 2007
where the canopy is broken with some penetration of
direct sunlight.
An. nili
Benin, Burkina Faso, Botswana, Adults: are anthropophilic (readily bite man indoors and - Elissa et al. 1999, Antonio-
Cameroon, Chad, Côte d’Ivore, outdoors), frequently resting indoors by day in some Nkondjio et al. 2002, 2005,
Ethiopia, Gabon, Ghana, Gambia, areas; have little contact with man and are rarely found Kenge et al. 2003, Awone-
Kenya, Liberia, Mali, Nigeria, Senegal, indoors in other areas. Ambene et al. 2004, Dia et al.
Sierra Leone, South Africa, Sudan, Larvae: breed within vegetation or dense shade along 2005, Moffett et al. 2007
Tanzania, Uganda the edges of streams and large rivers.

185
8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies
 Table 1. Continued.

186 
Species and distribution Ecology and behaviour Comments References
An. pharoensis
Angola, Burkina Faso, Burundi, Adults: endophagic and exophagic; highly exophilic; Is a secondary vector of Bruce-Chwatt and Gockel
Cameroon, Chad, Cote d’Ivoire, feed on humans but are more zoophilic; mainly rest on malaria. 1960, Gillies and DeMeillon
Democratic Republic of Congo, vegetation; females often found between stems of rice 1968, Gillies and Coetzee 1987,
Egypt, Ethiopia, Gabon, Gambia, plants and on stems of reeds. Wernsdorfer and Wernsdorfer
Ghana, Israel, Kenya, Malawi, Larvae: normally confined to fresh water and breed 2003, Moffett et al. 2007,
Mauritania, Niger, Nigeria, Rwanda, primarily in large vegetated swamps. Other habitats: Jawara et al. 2009
Senegal, Sierra Leone, Somalia, grass shores of lakes among floating plants, rice fields,
South Africa, Sudan, Syria, Tanzania, stagnant desert water, edges of streams, ditches,
Togo, Uganda, Zambia, Zimbabwe overgrown wells and temporarily flooded areas.
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Olfaction in vector-host interactions

 8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

attractive odours. Carbon dioxide (CO2) was recognised as a universal mosquito attractant,
including African anophelines (Mboera and Takken 1997). Apart from the studies that focused
on CO2 only, the number of semi-field and field studies that have been undertaken between
1993 and 2010 to further our understanding of odour-mediated host-seeking behaviour of
African malaria vectors are limited (Table 2). These largely centered on understanding mosquito
behavioural responses to crude, unrefined odour samples, mostly whole human odour (Table 2).
Other unrefined odour bait sources that have been investigated within this period include human
foot odour (Jawara et al. 2009, Murphy et al. 2001, Njiru et al. 2006, Olanga et al. 2010, Okumu et
al. 2010a, Schmied et al. 2008), human breath (Knols et al. 1998, Mukabana et al. 2004) and sheep,
goat, pig (Mahande et al. 2007), cattle (Costantini et al. 1993, 1998, Dekker and Takken 1998,
Duchemin et al. 2001, Kweka and Mahande 2009, Kweka et al. 2009, Mahande et al. 2007, Tirados
et al. 2006) and monkey odours (Costantini and Diallo 2001). A handful of these studies have been
carried out in West Africa where malaria vector species, characterised by certain peculiarities
(Costantini et al. 1999), differ in some ways from those found in eastern Africa, including the
islands of the Indian Ocean (Duchemin et al. 2001).

Specific objectives of the studies conducted are diverse but can be grouped into four. These
included: (1) development of mosquito sampling and surveillance tools (Costantini et al. 1993, Dia
et al. 2005, Govella et al. 2009, Jawara et al. 2009, Knols et al. 1998, Laganier et al. 2003, Mathenge
et al. 2002, 2004, 2005), (2) investigation of the basis of differences of attractiveness of humans
to mosquitoes (Brady et al. 1997, Knols et al. 1995, Lacroix et al. 2005, Lindsay et al. 1993, Mboera
et al. 1997, Mukabana et al. 2002, 2004); (3) determination of the interspecific host preferences of
different mosquitoes species (Costantini and Diallo 2001, Costantini et al. 1998, Dekker and Takken
1998, Duchemin et al. 2001, Kweka and Mahande 2009, Kweka et al. 2009, Mahande et al. 2007,
Tirados et al. 2006, Torr et al. 2008) and (4) identification of chemical attractants for Afrotropical
anophelines (Beavers et al. 1998, Costantini et al. 1996, 2001, Gibson et al. 1997, Murphy et al.
2001, Njiru et al. 2006, Okumu et al. 2010a, Qiu et al. 2007). To date, nothing has been reported on
the actual deployment of chemical attractants for operational mosquito control or surveillance
in Africa.

The species of African Anopheles mosquitoes mentioned in the published studies are diverse and
include Anopheles gambiae s.s., An. arabiensis, An. coustani, An. funestus, An. nili, An. pharoensis,
An. quadriannulatus, An. sergenti, An. squamosus and An. ziemanni. In terms of host preference An.
gambiae s.s. (Costantini and Diallo 2001, Costantini et al. 1993, 1998), An. funestus (Duchemin et
al. 2001) and An. pharoensis (Costantini et al. 1998) preferred to enter human-baited over animal-
odour baited traps in dual-choice assays. This was also the case for An. arabiensis (Kweka et al.
2009, Torr et al. 2008). However, in some cases for An. gambiae, An. arabiensis (Duchemin et al.
2001, Kweka and Mahande 2009) and An. funestus (Costantini et al. 1998) this was the reverse.
Carbon dioxide was identified as a cause for differences in human attractiveness to mosquitoes
(Brady et al. 1997) and the chemical compounds 1-octen-3-ol + CO2 (Beavers et al. 1998), L-lactic
acid + CO2 (Murphy et al. 2001), 7-octenoic acid (Costantini et al. 2001), and a blend containing
ammonia + L-lactic acid + CO2 + 3-methylbutanoic acid (Qiu et al. 1997) attracted various African
mosquito species in the field.

Olfaction in vector-host interactions 187

 Table 2. Published field and semi-field studies on odour-mediated host-seeking behaviour of African Anopheles mosquitoes undertaken between 1993 and 2010.

188 
F= Field study, S= Semi-field study.

Odour baits, F/S Objective(s) Study site and procedures Results

Reference
Carbon dioxide F Develop an odour-baited Study site: Burkina Faso. An Odour Baited Anopheles gambiae s.l., An. funestus and Mansonia spp. were the
(CO2), human trap for field studies of Entry Trap (OBET) with live bait as sources most abundant species caught when traps were baited with a
odour, calf odour; mosquito host-seeking of host stimuli was developed. Volatiles or human, calf or CO2. Species composition varied with different baits.
Costantini et al. behaviour. gases emitted by live baits concealed in tents Caught mosquitoes were mainly non-fed, with 100% survival rate.
1993 were sucked and channelled into a mosquito
trapping device.
CO2, body odour F Determine the basis Study site: Burkina Faso. Attractiveness of In the HBC, the ‘most attractive’ catcher in each team collected twice
(including breath); of differences in humans to mosquitoes was investigated as many mosquitoes as the ‘least attractive’ catcher. The differences
Brady et al. 1997 attractiveness of humans based on did change and were independent of catchers’ skills. OBET assays
to mosquitoes. Biting Catch (HBC) and OBETs baited with found equal significant rankings for An. gambiae s.l. but ranking was
odour of individuals performing the HBC. lost when the men’s CO2 outputs were artificially standardised.
Catches using OBETs were repeated with the
individuals’ CO2 outputs equalised.
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

CO2, body odour F Investigate the influence Study site: Noungou village, Burkina Faso. More An. gambiae s.l. chose the human OBET when traps lay side
(including breath); of CO2 relative to other Direct choice tests were done using two by side. When placed 20 m apart the CO2 baited traps caught 50,
Costantini et al. host odours in the host OBETs placed side by side or 20 m apart. 40, 65 and 200% of An. gambiae s.l., An. funestus, Ma. uniformis
1996, Gibson et al. seeking behaviour of An. One OBET contained CO2 alone; the other and An. pharoensis of the human odour. All species gave similar
1997 gambiae, An. arabiensis. emitted the same concentration of CO2 plus dose-response curves when amounts of CO2 were varied but
human odour. behaviours differed when catches were compared with human bait:
An. arabiensis chose CO2 OBETs with a higher probability than An.
gambiae s.s.
CO2, human body F Determine (1) the relative Study site: Kikulukutu village, Kilombero Human odour attracted significantly more An. gambiae and An.
odour; Mboera et attractiveness of human district, Tanzania. Odours from a male funestus than ‘odour’ from an empty pit. Physical presence of a
al. 1997 odour to mosquitoes in volunteer sitting in an underground pit were volunteer did not increase his attractiveness over odour, from
the presence or absence pumped into an untreated bed net present a competing human subject, pumped from a pit. Tents baited
of a human; (2) mosquito inside a PVC tent. The relative attractiveness with humans attracted a larger proportion of both species than
responses to different of the volunteer’s odour against an empty tents baited with CO2 (300 ml/min), which in turn caught more
levels of CO2. tent, a tent occupied by a second volunteer mosquitoes than unbaited tents. This result was unchanged for An.
or a tent with CO2 (released at 300 or 1,500 gambiae when the amount of CO2 was increased (1,500 ml/min).

Olfaction in vector-host interactions

ml/ min) was then evaluated. Releasing CO2 at 1,500 ml/min equalled catches of An. funestus to
those by man.
 
Table 2. Continued.

Odour baits, F/S Objective(s) Study site and procedures Results

Reference
CO2, 1-octen-3-ol; F Evaluate CO2 and Study site: Qara and Matruh Governorate, Larger numbers of Ae. caspius and An. sergenti were collected in
Beavers et al. 1998 1-octen-3-ol as mosquito Egypt CO2 and CO2 + octenol traps than octenol alone. Traps baited with
attractants octenol + CO2 resulted in greater capture rates but results were
not significant when compared with CO2 alone. At the release rate
tested (0.5 mg/h at 21 °C), octenol was not an effective attractant.
Human, calf odour; F Investigate the role of Study site: Noungou village, Burkina Faso. An. gambiae s.l. and An. pharoensis predominantly chose the
Costantini et al. whole host odours in Choice tests were performed using two human-baited trap; the calf-baited trap had high numbers of Cx.

Olfaction in vector-host interactions

1998 the determination of OBETs placed side by side. Odours of a antennatus, An. rufipes, Cx. duttoni and Cx. nebulosus. An. funestus,
mosquito preferences for human and a calf (of similar mass) were Ma. africana, Ae. dalzieli and Ae. hirsutus were caught in low
a calf and a human. drawn from two concealed tents and blown numbers but with a bias for the human-baited trap. Ma. uniformis
into the OBETs. was distributed randomly between the two traps.
Whole human F Investigate the response Study site: South Africa. CO2 was delivered In the choice test between man and CO2 (human equivalent of 250
odour, calf, CO2; of two sibling species as dry ice with average release rate of 800- ml/min), 81% of An. quadriannulatus and 20% of An. arabiensis were
Dekker and Takken An. arabiensis and An. 900 ml/min. Lower rates of 250 and 180 ml/ attracted to CO2. High CO2 release rates (cow equivalent of 800 ml/
1998 quadriannulatus to CO2, a min were obtained from a CO2 gas cylinder. min) attracted significantly more An. quadriannulatus than CO2
man or a calf. Combinations of baits were compared in released at 250 ml/min. Up to 33% of the attraction of An. arabiensis
dual-choice tests using two mosquito nets to human emanations was attributed to CO2. An. quadriannulatus
placed 2.5 m apart and 10 cm off the ground. was equally attracted to a calf and CO2 (calf equivalent of 180 ml/
Mosquitoes entered the nets from below and min).
were collected by a suction tube.
Human breath, F Test the feasibility of Study site: Namawala village, Tanzania. It was possible to sample An. squamosus, An. coustani and Mansonia
CO2; Knols et al. electric nets for sampling Mosquitoes were sampled using electric africana using human breath-baited and CO2 baited electric nets.
1998 mosquitoes. nets.
3-methyl-2- F Synthesise and Study site: Burkina Faso. The acid(s) were Only 7-octenoic acid significantly increased An. gambiae s.l. trap
hexenoic acid, investigate role of the released with either ‘pure CO2’ (flow-rate 200 catches of CO2. The treatment trap caught significantly fewer An.
7-octenoic acid, listed odourants as ml/min), or ‘whole human odour’ provided gambiae s.l. than the control trap, for both ‘pure CO2’ and ‘whole
CO2, whole human components of human- by an adult male sleeping inside a tent. human odour’ treatments, when 7-octenoic acid was offered in
odour; Costantini specific kairomones. These treatments were compared to ‘pure combination with 3-methyl-2-hexenoic acid.
et al. 2001 CO2’ or ‘whole human odour’, which served
as controls.

189
8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies
 Table 2. Continued.

190 
Odour baits, F/S Objective(s) Study site and procedures Results
Reference
Human and F Assess the tendency of Study site: Senegal. Two adult Cercopithecus An. gambiae s.l, An. funestus, and An. nili clearly expressed a
monkey odours; An. gambiae and other aethiops monkeys and a child of similar preference for human odour, with >90% of captured mosquitoes
Costantini and malaria vectors to prefer mass slept inside separate tents and their caught in the human baited trap. These results support the
Diallo 2001 human over monkey odours were drawn to paired OBETs so that hypothesis that the strongly anthropophilic feeding preferences of
odour. approaching mosquitoes could experience An. gambiae did not evolve from an ancestral association with non-
both odour-laden streams before entering human primates.
one of the two traps.
Cow, human F Assess host preferences Study site: Madagascar. Odours from a An. funestus ‘preferred’ human odour whereas An. gambiae and An.
odour; Duchemin of An. gambiae, An. man and a calf of similar mass, concealed arabiensis preferred calf odour.
et al. 2001 arabiensis and An. in different tents were drawn by fans to
funestus. separate OBETs.
Limburger cheese F Test if attractants derived Study site: Kenya. Hexanoic acid, L-lactic CDC miniature light traps baited with L-lactic acid plus CO2
volatiles, L-lactic from or related to acid and cheese volatiles were offered in attracted significantly more mosquitoes than those baited only with
acid, CO2, hexanoic human emanations elicit combination with CO2 (dry ice). Emanations CO2 or the other attractants.
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

acid, foot odour, behavioural responses from a human or his worn socks were not
human odour; to An. gambiae and An. combined with CO2. Attractants were offered
Murphy et al. 2001 funestus. on 100% cotton 3.8 cm long dental wicks.
Whole human F Compare relative Study site: Barkédji and Ngari villages, HBC was more effective indoors for surveying the anopheline
odour; Dia et al. performance of OBETs Senegal. fauna. Both methods were effective in sampling An. gambiae, An.
2005 and HBC. arabiensis, An. funestus and An. nili, and mosquito age structure and
infectivity rates did not differ between the methods.
Human foot odour, S Assess reponses of An. Study site: Kenya. Lab-reared mosquitoes Increases in catches were observed as follows: 1-octen-3-ol >
CO2, ammonia, gambiae to these odours were released in a 7×11 m screenhouse ammonia > foot odour > foot odour + ammonia > CO2 > foot odour
1-octen-3-ol; Njiru alone or in combination compartment with and given a choice of + ammonia + CO2
et al. 2006 baited MM-X traps.

Olfaction in vector-host interactions

 
Table 2. Continued.

Odour baits, F/S Objective(s) Study site and procedures Results

Reference
Cow odour; whole F Monitor the blood Study site: Ethiopia. Female An. arabiensis Where the ratio of cattle to humans was 0.6:1, 51% of outdoor
human odour; feeding behaviour of An. mosquitoes were collected by host-baited resting mosquitoes and 66% of those collected indoors had fed on
Tirados et al. 2006 arabiensis. methods (light traps, human landing catches, humans; human baits outdoors caught >2.5 times more mosquitoes
cattle-baited traps) and from resting sites than those indoors. Where humans slept outdoors close to their
(huts and pit shelters). Anthropophily cattle, ~46% of resting mosquitoes collected outdoors had fed
was gauged by comparing numbers of on humans despite the high cattle to human ratio of 17:1. In both
mosquitoes caught in OBETs baited with sites, high proportions of blood meals were mixed cow + human:

Olfaction in vector-host interactions

either human or cattle odour. 22-25% at Fuchucha and 37% in the cattle camps. In the OBETs, the
human-baited trap caught five times as many mosquitoes as the
cattle-baited one.
Human, cattle, F Assess feeding behaviour Study site: Mabogini village, Moshi, Tanzania. Cattle, human, sheep, goat and pig odours attracted 90.3%, 9.7%,
sheep, goat of An. arabiensis. Feeding behaviour of An. arabiensis was 9.7%, 7.2%, and 7.3% of An. arabiensis mosquitoes, respectively (n
and pig odours; assessed using OBETs. = 506).
Mahande et al.
2007
CO2, L-lactic F Investigate effects of Study site: Walikunda village, The Gambia. Addition of CO2 to synthetic odours increased the catch of all
acid, ammonia, synthetic odour blends Attractants were placed in vials and placed mosquito species in the MM-X traps. The blend of ammonia
carboxylic fatty on mosquito catches in MM-X traps with or without CO2. Human + L-lactic acid + CO2 + 3-methylbutanoic acid attracted most
acids, ketones, with or without CO2 as an odour was piped to an MM-X trap by mosquito species.
alcohols, whole additional stimulus. drawing air from a human-occupied nylon
human odour; Qiu tent through lay-flat tubing. Sixteen odour
et al. 2007 blends with CO2 and fourteen odour blends
without CO2 were tested. The control odour
consisted of ammonia + L-lactic acid + CO2.
Foot odour, foot S Compare various odours Study site: Ifakara, Tanzania. Female An. The BGS trap caught about three times as many mosquitoes as
odour + CO2, against An. gambiae using gambiae s.s. were released in an experimental the MM-X trap (P=0.002). Adding CO2 (500 ml/min) to foot odour
or foot odour two different trap designs flight arena that was placed inside a semi- increased the number of mosquitoes caught by 268% for the MM-X
+ repellent (BGS or MMX-trap). field system. (P<0.001) and 34% (P=0.051) for the BGS trap, compared to foot
(lemongrass); odour alone. When lemongrass leaves were added to foot odour,
Schmied et al. 2008 mosquito catches were reduced by 39% (BGS, P<0.001) and 38%

191
8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

(MM- X, P=0.353), respectively.

 Table 2. Continued.

192 
Odour baits, F/S Objective(s) Study site and procedures Results
Reference
Cattle odour, F Compare odour- Study site: Rekomitjie, Mana pools game Electric nets baited with odour from a single ox or a single man
human odour, oriented responses of reserve, Zimbabwe. Number of mosquitoes caught similar numbers of An. arabiensis; increasing the dose of
CO2, acetone, An. arabiensis and An. attracted and mosquito entry responses human odour from one to three men increased the catch 4-fold.
1-octen-3-ol, quadriannulatus in the were estimated using electrocuting nets An. quadriannulatus catches from E-nets increased up to 6-fold in
4-methylphenol, field. (E-nets) and OBETs, respectively. Mosquito the progression: man, 3 men, ox and man plus ox. Entry responses
3-n-propylphenol; landing responses were estimated by of An. arabiensis were stronger with human odour (62%) than with
Torr et al. 2008 comparing catches from E-nets and cloth ox odour (6%) or both (15%). An. arabiensis did not exhibit a strong
targets covered with an electrocuting grid. entry response to carbon dioxide (0.2-2 l/min). Entry responses of
An. quadriannulatus were low (<2%) with both cattle and human
odour. Catches from an electrocuting target baited with either
CO2 or a blend of acetone, 1-octen-3-ol, 4-methylphenol and
3-n-propylphenol – components of natural ox odour – showed that
virtually all mosquitoes arriving there alighted on it.
Cow, human F Compare human landing Study site: Mabogini village, Lower Moshi, Both An. gambiae s.l. and Cx. quinquefasciatus mosquitoes were
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

subject; Kweka catches, pit shelters, Tanzania. The number of mosquitoes collected. In general human landing catches recovered significantly
and Mahande indoor resting collections recovered indoors, from a pit shelter, from a fewer mosquitoes than the pit shelter and the cow-baited entry
2009 and man/cow-baited cow-baited entry trap, from a human-baited trap. The human biting catch collections did not differ significantly
entry traps for sampling entry trap and human landing catches were from the indoor resting collections. Though otherwise stated
mosquitoes at low compared during a season of low mosquito human landing catches were apparently not different from
density. density and malaria transmission. collections using the human-baited entry trap.
Cow body odour, F Evaluate human landing Study site: Mabogini village, Lower Moshi, Significantly more An. arabiensis mosquitoes were collected from
cow urine, human catch collections and Tanzania. Human landing catch collections urine baited resting boxes than by the human landing catches, an
subject; Kweka et cow odour-baited resting and an unbaited resting box, a cow odour- unbaited resting box and a resting box baited with cow odour only.
al. 2009 boxes as sampling tools baited resting box and a cow urine-baited
for An. arabiensis. resting box were evaluated as sampling tools
for An. arabiensis mosquitoes.

Olfaction in vector-host interactions

 
Table 2. Continued.

Odour baits, F/S Objective(s) Study site and procedures Results

Reference
CO2, foot odour, F Determine if traps areStudy site: Walikunda village, The Gambia. The mosquito species collected included An. gambiae s.l. (9.1%),
human subjects; best placed indoors MM-X traps were either placed indoors An. pharoensis (2.3%), An. ziemanni (1.4%), other anophelines
Jawara et al. 2009 or outdoors, the best or outdoors in experimental huts, each (0.3%), Culex species (5.3%), Aedes (0.4%), and Mansonia species
height and distance away
occupied by one human subject. The traps (81.2%). The effect of trap position (indoors or outdoors) and the
from a hut for collecting
were baited with CO2 and human foot presence or absence of a sleeper did not have any significant
mosquitoes and the odour (socks worn for 12 hours). Traps were effect on the number of An. gambiae s.l. collected. The optimal
possibility of competition
operated from 09:00 pm to 07:00 am the placement of human odour-baited MM-X traps is at a height of 15

Olfaction in vector-host interactions

among outdoor traps next morning. Hut windows were slightly cm immediately outside huts. This maximizes capture of wild An.
and their influence onopened at 6:00 pm to leave a gap of about gambiae s.l. mosquitoes. The number of An. gambiae s.l. collected
numbers of mosquitoes 10 cm, simulating a poorly fitting window as within the huts at the end of each night was not significantly
entering huts. observed in village huts, and they remained affected by the presence of single or multiple traps immediately
in this position until 6:30 am, while the doors outside, compared to huts that had no trap outside.
remained closed.
CO2, ammonia, 7 F, S Develop and evaluate Study site: Tanzania. Binary choice assays The blend consisting of carbon dioxide, ammonia, L-lactic acid
carboxylic acids, under field conditions a were conducted inside a semi-field enclosure and seven aliphatic carboxylic acids attracted three to five times
foot odour, human super-attractive synthetic using baited traps placed 20 m apart to find more An. gambiae s.l. (mostly An. arabiensis) and An. funestus than
subjects; Okumu et odor blend consisting optimum concentrations at which individual humans when the two baits were in different experimental huts (10-
al. 2010 of known mosquito candidate attractants need to be added 100 m apart), but was equally or less attractive than humans when
attractants. to constitute a blend maximally attractive compared side by side within the same hut.
to lab-reared An. gambiae mosquitoes.
Field experiments were conducted inside
experimental huts, where the blend was
compared with adult male humans, directly
and indirectly.

193
8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies
Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Synthetic odour baits attractive to African anophelines under semi-field and field
conditions

The discovery of synthetic odour baits that are capable of attracting mosquitoes much the same
as a human being (Brady et al. 1997) can enhance the development of powerful tools for vector
surveillance (Takken and Knols 1999) and control (Day and Sjogren 1994, Kline 2007, Logan and
Birkett 2007, Takken and Knols 2009). Individual chemical compounds and blends thereof have
been shown to attract African Anopheles mosquitoes under semi-field and field environments.
The chemicals, which comprise of commonly known kairomones like CO2, carboxylic acids,
ketones, phenols, L-lactic acid, and ammonia, are described in subsequent paragraphs. This
chapter is restricted to semi-field and field studies on African anophelines where CO2 was tested
in combination with crude odourants or at least one synthetic attractant (Table 2). The selection
of candidate attractants for these studies was in many cases informed by laboratory studies,
described in Chapter 7.

Carbon dioxide, a major constituent of human exhaled air (300-500 ml/min.), has been identified
as an attractant for many mosquito species including the main vectors of malaria in Africa (Gillies
1980, Mboera and Takken 1997, Mboera et al. 1997, 2000, Takken 1991, Takken and Knols 1999).
Gillies (1980) suggested that this compound acts as an activator, initiating flight responses, as well
as being an attractant. There is strong evidence that CO2 acts synergistically with other chemical
compounds to attract host-seeking mosquitoes (Dekker et al. 2002, Kline et al. 1990, Murphy et
al. 2001, Njiru et al. 2006, Takken and Kline 1989). Indeed, addition of CO2 to traps baited with
various synthetic compounds significantly increased catches of African anophelines including
An. gambiae s.s. (Costantini et al. 2001, Murphy et al. 2001), An. arabiensis (Costantini et al. 2001,
Murphy et al. 2001, Torr et al. 2008), An. quadriannulatus (Torr et al. 2008), An. funestus (Murphy et
al. 2001), An. sergenti (Beavers et al. 1998), An. pharoensis (Qiu et al. 2007) and An. ziemanni (Qiu
et al. 2007). In a field study in Burkina Faso, CO2 added to 7-octenoic acid significantly increased
the number of An. gambiae s.s. and An. arabiensis attracted to odour-baited entry traps. In The
Gambia, MM-X traps baited with a synthetic odour blend consisting of ammonia, lactic acid,
CO2 and 3-methylbutanoic acid attracted large numbers of mosquitoes belonging to the genera
Mansonia, Anopheles and Culex (Qiu et al. 2007). Field studies in Egypt demonstrated that CO2 plus
1-octen-3-ol attracted similar numbers of An. sergenti as CO2 alone (Beavers et al. 1998).

Limburger cheese, the smell of which is reminiscent of human foot odour, is known to attract
African anopheline mosquitoes under laboratory conditions (De Jong and Knols 1995, Knols
and De Jong 1996). One field study carried out in western Kenya, using Limburger cheese in its
original form, also demonstrated this (Owino 2006) while another, carried out in the same region
using a synthetic analogue of the cheese, did not (Murphy et al. 2001). In these studies Limburger
cheese odours were delivered using either MM-X counter flow geometry traps (Owino 2006) or
CDC miniature light traps (Murphy et al. 2001). Species of African malaria mosquitoes that were
successfully trapped using Limburger cheese in its original form included An. gambiae s.s., An.
arabiensis and An. funestus (Owino 2006). The difference in results may be attributed to usage of
synthetic versus authentic forms of the cheese and differences in odour delivery methodologies.
In the case of Murphy et al. (2001), a combination of lactic acid and CO2 was the only treatment
that significantly attracted more An. gambiae s.l. and An. funestus mosquitoes, although numbers
caught were very low. This may be attributed to the difficulty of baiting CDC light traps, the fan
of which disperses odours widely.

194  Olfaction in vector-host interactions

 8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

Synthetic blends consisting of acetone, 1-octen-3-ol, 4-methylphenol and 3-n-propylphenol,

which are components of cattle odour, have recently been used to bait mosquito electrocuting
nets (Torr et al. 2008). Anopheles arabiensis did not exhibit a strong entry response to CO2 (0.2-2 l/
min) baited entry traps. Catches from an electrocuting target baited with either CO2 or a blend of
acetone, 1-octen-3-ol, 4-methylphenol and 3-n-propylphenol induced overwhelmingly high An.
gambiae s.l. landing rates (Torr et al. 2008). In a recent study, an odour blend consisting of known
mosquito attractants namely CO2, ammonia and carboxylic acids was evaluated (Okumu et al.
2010a; see Table 2 for compound names). Field experiments conducted inside experimental huts
found the blend to attract three to five times more An. arabiensis than humans when the two baits
were in different experimental huts (10-100 meters apart). However, the blend was equally or less
attractive than humans when compared side by side within the same huts (Okumu et al. 2010a).

From the foregoing it is clear that one or more synthetic analogues of human sweat combined
with CO2 (Costantini et al. 2001, Qiu et al. 2007) or CO2 combined with various aliphatic carboxylic
acids (Okumu et al. 2010a) are currently the best attractants for host-seeking Afrotropical malaria
mosquitoes under field conditions. Further research on the development of potent attractants
needs to be carried out with the aim of (1) identifying additional candidate human specific
odours (2) determine their optimum concentrations to attract host-seeking mosquitoes and (3)
optimising existing synthetic odour blends.

Semi-field research on odour bait development

Historically, semi-field systems, which in essence are large outdoor screened cages, have been
used for mosquito research in several countries and for several mosquito species (reviewed by
Ferguson et al. 2008). Only since the late 1990s have such systems become a research tool to
identify important kairomones for African anophelines. A major advantage of such systems is the
fact that a fixed and controllable number of mosquitoes can be introduced in them, which delivers
constant and comparable results. Moreover, by placing such systems directly in areas where
malaria vectors occur in nature it is possible to rear offspring of field-collected gravid females,
thereby ensuring that the genetic background of test mosquitoes is similar to that of the field
population. Availability of efficient traps that can easily be baited with CO2 and other attractants,
such as the MM-X trap (see Njiru et al. 2006) and the BGS trap (see Schmied et al. 2008) has further
strengthened semi-field studies on anopheline host-seeking behaviour. When using two such
traps in simple cross-over designs to account for positional effects, rapid progress can be made
to improve baits. For instance, the recent step-wise and incremental improvement of a powerful
blend to attract An. gambiae s.s. by Okumu et al. (2010a) started with a basic blend of CO2 and
ammonia, and was then augmented with various concentrations of L-lactic acid, until an optimal
concentration was found. Thereafter, each of seven aliphatic carboxylic acids was added to this
base blend, at varying concentrations, until again the optimal increase in catch was reached. This
approach culminated in the development of a blend that attracted 3-5 times as many mosquitoes
than a human volunteer, when tested in experimental huts in a field setting. These developments
clearly demonstrate the power of semi-field research, and we advocate the use of such systems
not only for further attractant development, but also for the development of low-cost trapping
devices, an aspect of bait-trap technology that has not received enough attention to date.

Olfaction in vector-host interactions 195

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

The promise of using odour baits for malaria vector surveillance and control in
Africa

Although the goals of developing potent insect attractants are diverse, the central goal lies in
vector control and surveillance (Kline 2007, Logan and Birkett 2007). With respect to mosquitoes,
impact on target populations can be achieved through mass trapping (Anonymous 2005, Kline
and Lemire 1998), lure and kill technology (Day and Sjogren 1994, Kline 2007) or lure and
contaminate technology (e.g. by using biopesticides like entomopathogenic fungi). The efficacy
of mass trapping, despite its underlying conceptual, technical, logistical and financial limitations,
has been successfully demonstrated for mosquito population reduction in the USA (Anonymous
2005, Kline 2007, Kline and Lemire 1998). However, this strategy is still being developed, with
current efforts being centered on searching for new attractants and attractant formulations
(Healy and Copland 2000, Healy et al. 2002, Okumu et al. 2010a, Qui et al. 2007), improving on
existing ones (Smallegange et al. 2005), and developing trapping devices (Kline 2006, Kröckel et al.
2006). Efficacy trials of candidate synthetic attractants under field (Qiu et al. 2007) and semi-field
conditions (Njiru et al. 2006, Okumu et al. 2010a, Olanga et al. 2010) are also underway.

In terms of surveillance the need of replacing the risk prone human landing catch for measuring the
degree of contact between humans and mosquitoes (Service 1993) is urgent. The key challenge is
to find synthetic attractants, which by acting as human surrogates can be used in sampling devices
for malaria vectors (Brady et al. 1997). Whereas identification of a potent synthetic attractant will
go a long way in helping to develop malaria vector surveillance tools and mass trapping devices,
other challenges relate to suitable ways and means of delivering attractant odours. In the studies
carried out thus far delivery methods of synthetic attractants included use of wicks (Murphy et al.
2001), glass vials (Costantini et al. 2001, Qiu et al. 2007), sealed polythene sachets (Torr et al. 2008),
and nylon strips (Okumu et al. 2010b). The use of wicks, glass vials and nylon strips has the major
disadvantage that whereas release rates of synthetic odours can be measured by weighing before
and after placement in the field, measurement may not be as precise, especially for compounds
which tend to absorb water such as octenol. This is in contrast to sealed polythene sachets,
which guarantee measurable constant release rates of synthetic attractants. However, the use of
pressurised cylinders (Costantini et al. 2001, Qiu et al. 2007, Torr et al. 2008) or dry ice (Murphy et
al. 2001) as delivery methods of CO2 is disadvantageous as the methods are not logistically sound.
They are expensive and laborious for use in the field. Handy sources and more effective methods
of delivering CO2 and other gaseous kairomones should be sought.

Besides the disadvantages associated with odour delivery technologies, the physical gadgets useful
for handling and dispensing attractant-impregnated materials are limiting. Field experiments
have so far relied on using odour-baited entry traps (Brady et al. 1997, Costantini at Diallo 2001,
Costantini et al. 1993, 1996, 1998, Dia et al. 2005, Gibson et al. 1997, Kweka and Mahande 2009,
Kweka et al. 2009, Mahande et al. 2007, Tirados et al. 2006, Torr et al. 2008); mosquito magnet
model X (MM-X) traps (Jawara et al. 2009, Okumu et al. 2010a, Qiu et al. 2007), CDC miniature light
traps (Murphy et al. 2001), BGS traps (Schmied et al. 2008), electric nets (Knols et al. 1998, Torr et al.
2008) and resting boxes (Kweka et al. 2009). Apart from the MM-X and BGS trap none of the other
gadgets is well suited at present for delivering attractants in a desirable manner.

In general the bottlenecks that might impede the use of synthetic attractants for malaria vector
surveillance and control in Africa include (1) lack of cheap traps and trapping devices, (2) lack
of affordable sources of energy to power traps, (3) lack of cheaper sources of CO2, this being
a key kairomone for many malaria vectors and may be a chemical that will remain an integral

196  Olfaction in vector-host interactions

 8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

component of attractive blends, (4) lack of cheap and easy methods for delivering mosquito
synthetic attractants, (5) low diversity of synthetic attractants, and (6) lack of a generic attractant
that can trap multiple vectors.

Concluding remarks

This chapter has focused on odour baits that have been employed to assess the host-seeking
behaviour of African Anopheles mosquitoes under field and semi-field conditions. The reported
findings on mosquito responses to synthetic attractants provide optimism that the search for a
surrogate human is possible. A potent synthetic human odour will enhance the development
of powerful mosquito trapping devices that can be exploited to increase the success of control
programmes through forecasting epidemics accurately and for formulating, planning and rolling
out control activities.

References

Adugna N and Petros B (1996) Determination of the human blood index of some anopheline mosquitoes by using
ELISA. Ethiop Med J 34: 1-10.
Anonymous, 2005. Collier Mosquito Control District - completed projects. Stevens’ Landing. Available at: http: //www.
collier-mosquito.org/stevens_landing.php (accessed November 30, 2009).
Antonio-Nkondjio C, Awono-Ambene P, Toto JC, Meunier JY, Zebaze-Kemleu S, Nyambam R, Wondji CS, Tchuinkam T and
Fontenille D (2002) High malaria transmission intensity in a village close to Yaounde the capital city of Cameroon.
J Med Entomol 39: 350-355.
Antonio-Nkondjio C, Simard F, Awono-Ambene P, Ngassam P, Toto J, Tchuinkam T and Fontenille D (2005) Malaria
vectors and urbanization in the equatorial forest region of south Cameroon. Trans R Soc Trop Med Hyg 99: 347-354.
Awone-Ambene HP, Kengne P, Simard F, Antonio-Nkondjio C and Fontenille D (2004) Description and bionomics of
Anopheles (Cellia) ovengensis (Diptera: Culicidae) a new malaria vector species of the Anopheles nili group from
South Cameroon. J Med Entomology 41: 561-568.
Beavers GM, Hanifa HA and Tetreault GE (1998) Response of mosquitoes (Diptera: Culicidae) to carbon dioxide and
octenol in Egypt. J Egypt Soc Parasitol 28(2): 303-312.
Bøgh C, Clarke SE, Pinder M, Sanyang F and Lindsay SW (2001) Effect of passive zooprophylaxis on malaria transmission
in The Gambia. J Med Entomol 38: 822-828.
Boutin JP, Pradines B, Pages F, Legros F, Rogier C and Migliani R (2005) Epidemiology of Malaria. Rev Prat 55(8): 833-840.
Brady J, Costantini C, Sagnon N, Gibson G and Coluzzi M (1997) The role of body odours in the relative attractiveness of
different men to malarial vectors in Burkina Faso. Annal of Trop Med Parasitol 91: 121-122.
Breman JG and Holloway CN (2007) Malaria Surveillance Counts. Am J Trop Med Hyg, 77 (Suppl 6): 36-47.
Bruce-Chwatt LJ and Göckel CW (1960) A study of the blood-feeding patterns of Anopheles mosquitoes through
precipitin tests. Bull WHO 22: 685-720.
Bryan, JH, Petrarca V, Di Deco MA and Coluzzi M (1987) Adult behaviour of members of the Anopheles gambiae complex
in the Gambia with special reference to An. melas and its chromosomal variants. Parasitologia 29: 221-249.
Cano J, Angel Desalzo M, Moreno M, Chen Z, Nzambo S, Bobuakasi L, Buatiche JN, Ondo M, Micha F and Benito A (2006)
Spatial variability in the density distribution and vectorial capacity of anopheline species in a high transmission
village (Equatorial Guinea). Malar J 5: 21.
Coetzee M (1989) Comparative morphology and multivariate analysis for the discrimination of four members of the
Anopheles gambiae group in Southern Africa. Mosq Sys 21: 100-116.
Coetzee M (2004) Distribution of the African malaria vectors of the Anopheles gambiae complex. Am J Trop Med Hyg
70: 103-104.
Coetzee M, Craig M and Le Sueur D (2000) Distribution of African malaria mosquitoes belonging to the Anopheles
gambiae complex. Parasitology Today 16: 74-77.

Olfaction in vector-host interactions 197

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Coetzee M, Hunt RH, Baack L and Davidson G (1993) Distribution of mosquitoes belonging to the Anopheles gambiae
complex, including malaria vectors, south of latitude 15uS. S Afr J Sci 89: 227-231.
Cohuet A, Simard F, Wondji CS, Antonio-Nkondjio C, Awono-Ambene P and Fontenille D (2004) High malaria transmission
intensity due to Anopheles funestus (Diptera: culicidae) in a village of Savannah-forest transition area in Cameroon.
J Med entomol 41: 901-905.
Collins FH and Besansky NJ (1994) Vector biology and the control of malaria in Africa. Science 264: 1874-1875.
Coluzzi M, Sabatini A, Petrarca V and di Deco MA (1979) Chromosomal differentiation and adaptation to human
environments in the Anopheles gambiae complex. Trans R Soc Trop Med Hyg, 73: 483-497.
Costantini C and Diallo M (2001) Preliminary lack of evidence for simian odour preferences of savanna populations of
Anopheles gambiae and other malaria vectors. Parassitologia 43: 179-182.
Costantini C, Birkett MA, Gibson G, Ziesmann J, Sagnon NF, Mohammed HA, Coluzzi M and Pickett JA (2001)
Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae to human-specific sweat
components. Med Vet Entomol 15: 259-266.
Costantini C, Gibson G, Brady J, Merzagora L and Coluzzi M (1993) A new odour-baited trap to collect host-seeking
mosquitoes. Parassitologia 35: 5-9.
Costantini C, Gibson G, Sagnon N, della Torre A, Brady J and Coluzzi M (1996) Mosquito responses to carbon dioxide in
a West African Sudan savanna village. Med Vet Entomol 10: 220-227.
Costantini C, Sagnon N, Della TA, Diallo M, Brady J, Gibson G and Coluzzi M (1998) Odour-mediated host preferences of
West African mosquitoes, with particular reference to malaria vectors. Am J Trop Med Hyg 58: 56-63.
Costantini C, Sagnon NF, Torre AD and Coluzzi M (1999) Mosquito behavioural aspects of vector-human interactions in
the Anopheles gambiae complex. Parassitologia 41: 209-217.
Cox-Singh J, Davis TME, Lee K, Shamsul SG, Matusop A, Ratnam S, Rahman HA, Conway DJ and Singh B (2007)
Plasmodium knowlesi malaria in humans is widely distributed and potentially life threatening. Clinical infectious
Diseases 46: 16-171.
Day JF and Sjogren RD (1994) Vector control by removal trapping. Am J Trop Med Hyg 50: 126-133.
De Jong R and Knols BGJ (1995) Olfactory responses of host-seeking Anopheles gambiae s.s. Giles (Diptera: Culicidae).
Acta Tropica, 59: 333-335.
Dekker T and Takken W (1998) Differential responses of mosquito sibling species Anopheles arabiensis and An.
quadriannulatus to carbon dioxide, a man or a calf. Med Vet Entomol 12: 136-140.
Dekker T, Steib B, Cardé RT and Geier M (2002) L-lactic acid: a human-signifying host cue for the anthropophilic mosquito
Anopheles gambiae. Med Vet Entomol 16: 91-98.
Dia I, Diallo D, Duchemin J, Ba Y, Konate L, Costantini C and Diallo M (2005) Comparisons of human-landing catches and
odour-baited entry traps for sampling malaria vectors in Senegal. J Med Entomol 42(2): 104-109.
Duchemin JB, Pocktsy JM, Rabarison P, Roux J, Coluzzi M and Costantini C (2001) Zoophily of Anopheles arabiensis and
An. gambiae in Madagascar demonstrated by odour-baited entry traps. Med Vet Entomol 15: 50-57.
Elissa N, Karch S, Bureau P, Ollomo B, Lawoko M, Yangari P, Ebang B and Georges AJ (1999) Malaria transmission in a
region of savanna-forest mosaic Haut-Ogooue Gabon. J Am Mosq Control Assoc 15: 15-23.
Fettene M, Hunt RH, Coetzee M and Tessema F (2004) Behaviour of Anopheles arabiensis and Anopheles quadriannulatus
sp. B mosquitoes and malaria transmission in Southwestern Ethiopia. Afr Entomol 12: 83-87.
Fettene M, Koekemoer LL, Hunt RH and Coetzee M (2002) PCR assay for identification of Anopheles quadriannulatus
species B from Ethiopia and other sibling species of the Anopheles gambiae complex. Med Vet Entomol 16: 214-217.
Gallup JL and Sachs JD (2001) The economic burden of malaria. Am J Trop Med Hyg 64: 85-96.
Gibson G, Constantini C, Sagnon F, Torre A and Coluzzi M (1997) The responses of Anopheles gambiae and other
mosquitoes in Burkina Faso to CO2-the start of a search for synthetic human odor. Annal Trop Med Parasitol 91:
123-124.
Gillies MT (1980) The role of carbon dioxide in host finding by mosquitoes (Diptera: Culicidae): a review. Bull Entomol
Res 70: 525-532.
Gillies MT and Coetzee M (1987) A supplement to the Anophelinae of African South of the Sahara. South African Institute
of Medical Research, Johannesburg, South Africa.

198  Olfaction in vector-host interactions

 8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

Gillies MT and De Meillon B (1968) The Anophelinae of Africa South of the Sahara. South African Institute of Medical
Research, Johannesburg, South Africa.
Gimnig JE, Ombok M, Kamau L and Hawley WA (2001) Characteristics of larval anopheline (Diptera: Culicidae) habitats
in western Kenya. J Med Entomol 38: 282-288.
Govella NJ, Chaki PP, Geissbuehler Y, Kannady K, Okumu F, Charlwood JD, Anderson RA and Killeen GF (2009) A new
tent trap for sampling exophagic and endophagic members of the Anopheles gambiae complex. Malar J 8: 157.
Haddow AJ (1942) The mosquito fauna and climate of native huts at Kisumu, Kenya. Bull Entomol Res 33: 91-142.
Healy TP and Copland MJW (2000) Human sweat and 2-oxopentanoic acid elicit a landing response from Anopheles
gambiae. Med Vet Entomol 14: 195-200.
Healy TP, Copland MJW, Cork A, Przyborowska A and Halket JM (2002) Landing responses of Anopheles gambiae elicited
by oxocarboxylic acids. Med Vet Entomol 16: 126-132.
Hunt RH, Coetzee M and Fettene M (1998) The Anopheles gambiae complex: a new species from Ethiopia. Trans R Soc
Trop Med Hyg 92: 231-235.
Jawara M, Smallegange RC, Jeffries D, Nwakanma DC, Awolola TS, Knols BGJ, Takken W and Conway DJ (2009) Optimizing
odor-baited trap methods for collecting mosquitoes during the malaria season in The Gambia. PLoS One 4: e8167.
Kenge P, Awono-Ambene P, Antonio-Nkondjio C, Simard F and Fontenille D (2003) Molecular identification of the
Anopheles nili group of African malaria vectors. Med Vet Entomol 17: 67-74.
Kim-Sung L, Cox-Singh J, Brooke G, Matusop A and Singh B (2009) Plasmodium knowlesi from archival blood films:
further evidence that human infections are widely distributed and not newly emergent in Malaysian Borneo. Int J
Parasitol. 2009 August; 39(10): 1125-1128.
Kline L (2006) Traps and trapping techniques for adult mosquito control. J Am Mosq Control Assoc 22(3): 490-496.
Kline DL (2007) Semiochemicals, traps/targets and mass trapping technology for mosquito management. J Am Mosq
Control Assoc 23: 241-251.
Kline DL and Lemire GF (1998) Evaluation of attractant baited traps/targets for mosquito management on Key Island,
Florida, USA. J Vect Ecol 23: 171-185.
Kline DL, Takken W, Wood JR and Carlson D (1990) Field studies on the potential of butanone, carbon dioxide, honey
extract, l-octen-3-ol, lactic acid, and phenols as extracts for mosquitoes. Med Vet Entomol 4: 383-391.
Knols BGJ and De Jong R (1996) Limburger cheese as an attractant for the malaria mosquito Anopheles gambiae s.s.
Parasitol Today, 12: 159-160.
Knols BGJ, De Jong R, Takken W (1995) Differential attractiveness of isolated humans to mosquitoes in Tanzania. Trans
Roy Soc Trop Med Hyg 89: 604-606.
Knols BGJ, Mboera LEG and Takken W (1998) Electric nets for studying odour-mediated host seeking behavior of
mosquitoes. Med Vet Entomol 12: 116-120.
Kröckel U, Rose A, Eiras A and Geier M (2006) New tools for surveillance of adult Aedes aegypti: comparison of trap
catches with human landing collections in an urban environment. J Am Mosq Control Assoc 22: 229-238.
Kweka EJ and Mahande AM (2009) Comparative evaluation of four mosquitoes sampling methods in rice irrigation
schemes of lower Moshi, northern Tanzania. Malar J 8: 149
Kweka EJ, Mwang’onde BJ, Kimaro E, Msangi S, Massenga CP and Mahande AM (2009) A resting box for outdoor
sampling of adult Anopheles arabiensis in rice irrigation schemes of lower Moshi, northern Tanzania. Malar J 8: 82
La Grange JJP (1995) Survey of anopheline mosquitoes (Diptera: Culicidae) in a malarious area of Swaziland. Afr Entomol
3: 217-219.
Laganier R, Randimby FM, Rajaonarivelo V and Robert V (2003) Is the Mbita trap a reliable tool for evaluating the density
of anopheline vectors in the highlands of Madagascar? Malar J 2: 42.
Lacroix R, Mukabana WR, Gouagna LC and Koella JC (2005) Malaria infection increases attractiveness of humans to
mosquitoes. PLoS Biol 3: 1590-1593.
Lindsay SW, Adiamah JH, Miller JE, Pleass RJ and Armstrong JRM (1993) Variation in attractiveness of human subjects
malaria mosquitoes (Diptera: Culicidae) in The Gambia. J Med Entomol 30: 368-373.
Logan JG and Birkett MA (2007) Semiochemicals for biting fly control: their identification and exploitation. Pest Manag
Sci, 63: 647-657.

Olfaction in vector-host interactions 199

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Logan JG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue AJ and Pickett JA (2008) Identification of human-
derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. J Chem Ecol 34: 308-322.
Logan TM, Linthicum KJ, Thande PC, Wagateh JN and Roberts CR (1991) Mosquito species collected from a marsh in
western Kenya during the long rains. J Am Mosq Control Assoc 7: 395–399
Mahande A, Mosha F, Mahande J and Kweka E (2007) Feeding and resting behaviour of malaria vector, Anopheles
arabiensis with reference to zooprophylaxis. Malaria J 6: 100.
Mathenge EM, Killeen GF, Oulo DO, Irungu LW, Ndegwa PN and Knols BGJ (2002) Development of an exposure-free
bednet trap for sampling Afrotropical malaria vectors. Med Vet Entomol 16: 67-74.
Mathenge EM, Misiani GO, Oulo DO, Irungu LW, Ndegwa PN, Smith TA, Killeen GF and Knols BG (2005) Comparative
performance of the Mbita trap, CDC light trap and the human landing catch in the sampling of Anopheles
arabiensis, An. funestus and culicine species in a rice irrigation in western Kenya. Malar J 4: 7.
Mathenge EM, Omweri GO, Irungu LW, Ndegwa PN, Walczak E, Smith TA, Killeen GF and Knols BG (2004) Comparative
field evaluation of the Mbita trap, the Centers for Disease Control light trap, and the human landing catch for
sampling of malaria vectors in western Kenya. Am J Trop Med Hyg 70: 33-37.
Mboera LEG and Takken W (1997) Carbon dioxide chemotropism in mosquitoes (Diptera: Culicidae) and its potential in
vector surveillance and management programmes. Annu Rev Entomol 85: 355-368.
Mboera LEG, Knols BGJ, Braks MAH and Takken W (2000) Comparison of carbon dioxide-baited trapping systems for
sampling outdoor mosquito populations in Tanzania. Med Vet Entomol 14: 257-263.
Mboera LEG, Knols BGJ, Takken W and Della Torre A (1997) The response of Anopheles gambiae s.l. and An. funestus
(Diptera: Culicidae) to tents baited with human odour or carbon dioxide in Tanzania. Bull Entomol Res 87: 173-178.
Mnzava AE and Kilama WL (1986) Observations on the distribution of the Anopheles gambiae complex in Tanzania.
Acta Trop 43: 277-282.
Moffett A, Shackelford N and Sarkar S (2007) Malaria in Africa: vector species’ niche models and relative risk maps. PLoS
One 2: e824.
Mosha FW and Petrarca V (1983) Ecological studies on Anopheles gambiae complex sibling species on the Kenya coast.
Trans R Soc Trop Med Hyg 77: 344-345.
Mosha FW and Subra R (1982) Ecological studies on Anopheles gambiae complex sibling species in Kenya 1 preliminary
observations on their geographical distribution and chromosomal polymorphic inversions. Document WHO/
VBC/82867. World Health Organization, Geneva, Switserland.
Mpofu SM (1985) Seasonal vector density and disease incidence patterns of malaria in an area of Zimbabwe. Trans R
Soc Trop Med Hyg 79: 169-175.
Mukabana WR, Takken W, Coe R and Knols BGJ (2002) Host-specific cues cause differential attractiveness of Kenyan men
to the malaria mosquito Anopheles gambiae. Malar J 1: 17.
Mukabana WR, Takken W, Killeen GF and Knols BGJ (2004) Allomonal effect of breath contributes to differential
attractiveness of humans to the African malaria vector Anopheles gambiae. Malar J 3: 1.
Murphy MW, Dunton RF, Perich MJ and Rowley WA (2001) Attraction of Anopheles (Diptera: Culicidae) to volatile
chemicals in Western Kenya. J Med Entomol 38: 242-244.
Muturi E, Shililu J, Jacob B, Gu W, Githure J and Novak R (2006) Mosquito species diversity and abundance in relation to
land use in a riceland agroecosystem in Mwea, Kenya. J Vector Ecol 31: 129-137.
Mzilahowa T, Ball AJ, Bass C, Morgan JC, Nyoni B, Steen K, Donnelly MJ and Wilding CS (2008) Reduced susceptibility
to DDT in field populations of Anopheles quadriannulatus and Anopheles arabiensis in Malawi: evidence for larval
selection. Med Vet Entomol 22: 258-263
Njiru BN, Mukabana WR, Takken W and Knols BGJ (2006) Trapping of the malaria vector Anopheles gambiae with odour-
baited MM-X traps in semi-field conditions in western Kenya. Malaria J 5: 39.
Okumu FO, Killeen GF, Ogoma S, Biswaro L, Smallegange RC, Mbeyela E, Titus E, Munk C, Ngonyani H, Takken W, Mshinda
H, Mukabana WR and Moore SJ (2010a) Development and field evaluation of a synthetic mosquito lure that is more
attractive than humans. PLoS One 5: e8951.
Okumu FO, Biswaro L, Mbeyela E, Killeen GF, Mukabana WR and Moore SJ (2010b) Using nylon strips to dispense
mosquito attractants for sampling the malaria vector Anopheles gambiae s.s. J Med Entomol (in press).

200  Olfaction in vector-host interactions

 8. Host-seeking behaviour of Afrotropical anophelines: field and semi-field studies

Olanga EA, Okal M, Mbadi PA, Kokwaro E and Mukabana WR (2010) Attraction of An. gambiae to odour baits augmented
with heat and moisture. Malar J 9: 6.
Owino EA (2006) Field evaluation of Limburger cheese as an odour bait source for sampling afrotropical malaria vectors.
MSc Thesis, University of Nairobi, Nairobi, Kenya.
Paskewitz SM, Ng K, Coetzee M and Hunt RH (1993) Evaluation of the polymerase chain reaction method for identifying
members of the Anopheles gambiae (Diptera: Culicidae) complex in southern Africa. J Med Entomol 30: 953-957.
Pates HV, Takken W, Curtis CF and Jamet H (2006) Zoophilic Anopheles quadriannulatus species B found in a human
habitation in Ethiopia. Ann Trop Med Parasitol 100(2): 177-179.
Phillips RS (2001) Current status of malaria and potential for control. Clin Microbiol Rev 14: 208-226.
Pock Tsy JM, Duchemin JB, Marrama L, Rabarison P, Le Goff G, Rajaonarivelo V and Robert V (2003) Distribution of the
species of the Anopheles gambiae complex and first evidence of Anopheles merus as a malaria vector in Madagascar.
Malar J 8: 33.
Qiu YT, Smallegange RC, Ter BC, Spitzen J, Van Loon JJ, Jawara M, Milligan P, Galimard AM, Van Beek TA, Knols BG
and Takken W (2007) Attractiveness of MM-X traps baited with human or synthetic odor to mosquitoes (Diptera:
Culicidae) in The Gambia. J Med Entomol 44: 970-983.
Sachs J and Malaney P (2002) The economic and social burden of malaria. Nature 11: 681–685.
Schmied WH, Takken W, Killeen GF, Knols BG and Smallegange RC (2008) Evaluation of two counterflow traps for testing
behaviour-mediating compounds for the malaria vector Anopheles gambiae s.s. under semi-field conditions in
Tanzania. Malar J 7: 230.
Service MW (1993). Mosquito ecology, field sampling methods, 2nd ed. Elsevier Applied Science, London, UK.
Sharp BL, Kleinschmidt I, Streat E, Maharaj R, Barnes KI, Durrheim DN, Ridl FC, Morris N, Seocharan I, Kunene S, La Grange
JJP, Mthembu JD, Maartens F, Martin CL and Barreto A (2007) Seven years of regional malaria control collaboration
– Mozambique, South Africa, and Swaziland. Am. J. Trop. Med. Hyg. 76: 42-47.
Smallegange RC, Qiu YT, van Loon JJ and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic acids
as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto (Diptera:
Culicidae). Chem Senses 30: 145-152.
Sylla EHK, Kun JFJ and Kremsner PG (2000) Mosquito distribution and entomological inoculation rates in three malaria-
endemic areas in Gabon. Trans R Soc Trop Med Hyg 94: 652-656.
Takken W (1991) The role of olfaction in host-seeking of mosquitoes: a review. Insect Sci Appl 12: 287-295.
Takken W and Kline DL (1989) Carbon dioxide and l-octen-3-ol as mosquito attractants. J Am Mosq Control Assoc 5:
311-316.
Takken W and Knols BGJ (1999) Odor-mediated behaviour of Afrotropical malaria mosquitoes. Annu Rev Entomol 44:
131-157.
Takken W and Knols BGJ (2009) Malaria vector control: current and future strategies. Trends Parasitol 25: 101-104.
Takken W, Knols BGJ and Otten H (1997) Interactions between physical and olfactory cues in the host-seeking behaviour
of mosquitoes: the role of relative humidity. Ann Trop Med Parasitol 91: 119-120.
Tirados I, Costantini C, Gibson G and Torr SJ (2006) Blood-feeding behaviour of the malarial mosquito Anopheles
arabiensis: implications for vector control. Med Vet Entomol 20: 425-437.
Torr S, Della Torre A, Calzetta M, Costantini C and Vale GA (2008) Towards a fuller understanding of mosquito
behaviour: use of electrocuting grids to compare the odour-orientated responses of Anopheles arabiensis and An.
quadriannulatus in the field. Med Vet Entomol 22: 93-108.
Van den Broek IVF and Den Otter CJ (1999) Olfactory sensitivities of mosquitoes with different host preferences
(Anopheles gambiae s.s., An. arabiensis, An. quadriannulatus, An. m. atroparvus) to synthetic host odours. J Insect
Physiol 45: 1001-1010.
Van Rensburg AJ, Hunt RH, Koekemoer LL, Coetzee M, Shiff CJ and Minjas J (1996) The polymerase chain reaction
method as a tool for identifying members of the Anopheles gambiae complex (Diptera: Culicidae) in northeastern
Tanzania. J Am Mosq Control Assoc 12: 271-274.
Watson M (1953) African highway. John Murray, London, UK.

Olfaction in vector-host interactions 201

Wolfgang R. Mukabana, Evelyn A. Olanga and Bart G.J. Knols

Wernsdorfer G and Wernsdorfer WH (2003) Malaria at the turn from the 2nd to the 3rd millenium. Wien klin Wochenschr
115 Suppl 3: 2-9.
Wilkes TJ, Matola YG and Charlwood JD (1996) Anopheles rivulorum, a vector of human malaria in Africa. Med Vet
Entomol 10: 108-110.
White GB (1974) Anopheles gambiae complex and disease transmission in Africa. Trans R Soc Trop Med Hyg 68: 278-301.
World Health Organization (1993) World malaria situation in 1991. Part I. Weekly Epidemiol Rec 34: 246.

202  Olfaction in vector-host interactions

9. The chemical ecology of sandflies
Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

Abstract

In the 25 years in which chemical ecology of sandflies (Diptera: Psychodidae) has been investigated,
most work has focussed on Lutzomyia longipalpis, the primary vector of Leishmania chagasi in
South and Central America. Studies of sex pheromones produced by male L. longipalpis have
aided in understanding the population structure of this species complex, found to consist of a
number of reproductively isolated sibling members, which may differ in their vectorial capacity.
Furthermore, identification and synthesis of these pheromones is leading to the development
of new monitoring and control devices, which aim to supplement and improve existing vector
control strategies. Complementary behavioural studies have also explored the role of sex
pheromones in mate choice, and the complex interactions that occur between males and females
during courtship. Studies of oviposition pheromones, kairomones and the effects of Leishmania
infection on host attractiveness have also contributed to our understanding of the intricacy of
chemical ecology in L. longipalpis, with broad implications relevant to the development of control
strategies in this and other insect vectors. In comparison, the chemical ecology of other sandfly
species, many of which are also important disease vectors, has largely been ignored, in part
because of the cost and difficulty of maintaining colonies for laboratory studies. Although much
work has been done, many challenges remain, and considerable research effort will be required to
understand the chemical ecology of sandflies in general, and the molecular basis of their intricate
communication systems.

Keywords: 3-methyl-α-himachalene, 9-methylgermacrene, 9-methylgermacrene-B, attractant,

cembrene, Lutzomyia longipalpis, oviposition pheromone, sex pheromone

Introduction

The economic and public health significance of the Diptera to man and his domestic animals
cannot be overstated. Their ability to transmit viral and parasitic diseases results in a very
significant human and animal disease burden and consequent economic cost. The phlebotomine
sandflies are important vectors of single cell protozoan parasites of the genus Leishmania and
the World Health Organization estimates that the disease burden in disability-adjusted life years
(DALYs) is second only to that caused by malaria and that burden is carried disproportionately by
the World’s poor and disadvantaged peoples.

Chemical ecology in the Diptera involves a wide range of inter and intraspecies interactions and
is employed in conjunction with tactile, visual and auditory signals. In particular, chemical signals
have a key role in both courtship and sexual behaviour in general. Indications of pheromonal
activity in the family Psychodidae was initially reported by Fuerborn (1922) who first described
an eversible mesothoracic ‘scent gland’ in the moth fly Ulomyia fuliginosa (Meigen). In the
haematophagous sub-family Phlebotominae, Lane and Ward (1984) published the first description
of a pheromone releasing papule in Lutzomyia longipalpis s.l. (Lutz and Neiva), although the
underlying secretory cell was drawn by Barth (1961) nearly 30 years earlier and described as an
‘odiferous gland’.

Olfaction in vector-host interactions 203

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

It is eight years since we last reviewed the progress that had been made in our understanding
of the chemical ecology of sandflies (Ward and Hamilton 2002). Since then, significant advances
have been made in our appreciation of their chemical communication systems, but much further
research is still needed to understand basic biological aspects of chemical signalling, and to
exploit our developing knowledge in creating new approaches to vector control. For example,
having established the structures of the male sex pheromone in two populations of Lu. longipalpis
s.l. (Hamilton et al. 1996a,b, 1999a,b), we have developed a cheap and efficient synthesis of the
sex pheromone S-9-methylgermacrene-B and its analogue, 9-methylgermacrene (Hamilton and
Krishnakumari 2004, Krishnakumari et al. 2004) to enable the initiation of field feasibility trials in
Brazil.

Apart from the genus Lutzomyia there has recently been some progress in our investigations into
the chemical ecology of two other phlebotomine genera; Sergentomyia and Phlebotomus. Results
from laboratory-bioassays indicate that the members of the genus Phlebotomus use chemical
communication between the sexes, the precise nature of which is under investigation (I Chelbi
and JGC Hamilton, unpublished observations). Some Sergentomyia species have been found to
produce compounds closely related in structure to the sex pheromones of the Lu. longipalpis
species complex, but their role has not been confirmed (JGC Hamilton unpublished data).

Progress in other areas of sandfly chemical ecology has included studies of attractants and
pheromones relating to oviposition behaviour, and limited investigations into defensive responses
of immatures. Host odour work has continued and some new insights gained into the role of
carbon dioxide obtained and other attractants in host-seeking behaviour. We also have evidence
that indicates that animals infected with Leishmania infantum Nicolle, the causative agent of
visceral leishmaniasis in the New World and southern Europe are more attractive to blood seeking
female sandflies than uninfected animals (Nevatte 2006, O’Shea et al. 2002).

There have also been some significant advances in our understanding of the Lu. longipalpis s.l.
species complex and we are now in a position to attempt to determine its significance in relation
to disease transmission (Bauzer et al. 2007, Maingon et al. 2008). Finally, new insights have been
gained through detailed analysis of courtship interactions, the results of which indicate that
contact pheromones, along with audio and long range chemical signals, may also have a role to
play in sexual behaviour (Bray and Hamilton 2007a).

Oviposition in response to chemical cues

Our laboratory has been active in examining the factors driving oviposition, concentrating on
the New World vector of visceral leishmaniasis Lu. longipalpis s.l. We showed that eggs were
attractive to conspecific ovipositing females (Dougherty et al. 1992, El Naiem and Ward 1991,
1992). Although the age of the eggs appeared not to be an important factor, the numbers of eggs
involved appeared to be critical to produce significant attraction. The origin of the pheromone
was traced to the accessory glands of the post bloodfed female and the substance was eventually
identified as n-dodecanoic acid, a C12 fatty acid (Dougherty and Hamilton 1997). Other work
showed that the green leaf volatiles hexanal and 2-methyl-2-butanol synergise the attraction
of the oviposition pheromone (Dougherty et al. 1995) and a combination of synthetic green
leaf volatiles and the oviposition pheromone mimicked the natural attraction of sandfly eggs
(Dougherty and Hamilton 1997).

204  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

Recently we have returned to this topic and in preliminary work have shown that the Old World
vector of Leishmania donovani Laveran and Mesnil in the Indian sub-continent Phlebotomus
argentipes Annandale and Brunetti also lays eggs that are attractive to conspecific gravid females
and stimulate oviposition (SF Kinsmore, RD Ward, PJ Taylor, JGC Hamilton 2008, unpublished
observations). Work to precisely describe the oviposition behaviour and determine the chemical
nature of the attraction is underway.

Sandfly responses to host odour

Host odours act as important signals to many species of blood-feeding Diptera, facilitating both
host location and discrimination (Colvin and Gibson 1992, Takken 1991). Understanding how
sandflies choose and locate their hosts could aid in developing odour-baited traps for monitoring
and control (Kelly and Dye 1997, Pinto et al. 2001) and help explain variation between individuals
in attractiveness and subsequent biting risk (Hamilton and Ramsoondar 1994, Rebollar-Tellez et
al. 1999). Despite considerable progress in recent years, there is still less known about sandfly
responses to host odours than those of many other biting insects, including mosquitoes.
Nevertheless, a number of studies have highlighted the importance of kairomones in sandfly
host-seeking, both by demonstrated attraction to odours produced by animal baits in the field
(Campbell-Lendrum et al. 1999, Christensen and Herrer 1973, Montoya-Lerma and Lane 1996, Pinto
et al. 2001, Quinnell et al. 1992), and to whole host odour and its constituents in the laboratory
(Bray and Hamilton 2007b, Morton and Ward 1989, Oshaghi et al. 1994, Quinnell et al. 1992).

Many sandflies (including the New World vectors of zoonotic leishmaniasises) are opportunistic
feeders, and may be attracted by ubiquitous cues that indicate the presence of a living vertebrate,
but provide little information as to the species of a potential host (Pinto et al. 2001, Quinnell et
al. 1992). Carbon dioxide is a major constituent of host breath, and is a near universal activator
and attractant of mosquitoes (Gillies 1980, Takken 1991). In the field, attraction of sandflies to
CO2 released from traps has been recorded in Malaysia (Knudson et al. 1979) and more rigorously
demonstrated in both Panama (Chaniotis 1983) and Brazil (Pinto et al. 2001). In the latter study, an
approximate linear relationship was found between concentration of carbon dioxide released and
numbers of sandflies (Lu. intermedia (Lutz and Neiva) and Lu. whitmani (Antunes and Coutinho))
captured. Increased attraction towards sources of greater CO2 release has been proposed as an
explanation for the observed preference of several sandfly species for hosts of a larger size, or
greater host density (Lu. longipalpis, Quinnell et al. 1992; Lu. evansi (Nuñez-Tovar) Montoya-Lerma
and Lane 1996; Lu. whitmani Campbell-Lendrum et al. 1999). However, although many species of
sandfly will respond to carbon dioxide to some extent, it is unlikely to be the sole mediator of
attraction. In field experiments in Brazil, carbon dioxide attracted less than half the number of
sandflies caught using whole human odour (Pinto et al. 2001). In addition, results of laboratory
experiments have been inconclusive regarding the response of Lu. longipalpis to CO2: while
host breath can activate flies to begin probing (Ready 1978) removal of carbon dioxide from
preparations of host odour and pheromone did not decrease their attractiveness in bioassays
(Nigam and Ward 1991).

Volatile molecules, emanating from breath and skin, may provide more specific information
regarding a potential host than carbon dioxide alone. In particular, chemicals produced as a
consequence of bacterial metabolism on the surface of the skin are likely to vary both between
and within host species (Takken 1991). As a consequence, theses chemicals may be important
both in attracting sandflies, and facilitating discrimination between potential sources of a blood
meal. That sandflies respond to host volatiles was first demonstrated with laboratory hamsters,

Olfaction in vector-host interactions 205

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

with males and female Lu. longipalpis attracted to volatile extracts presented in the absence of
any other host cue, including CO2 (Oshaghi et al. 1994).

Similarly, both sexes of Lu. longipalpis are attracted to Petri dishes recently held in human hands
(Hamilton and Ramsoondar 1994). That chemicals emanating from skin cause the observed landing
response was confirmed by transferring the attractive odours from handled to un-handled dishes
in non-polar solvents (Rebollar-Tellez et al. 1999). In both studies using humans, emanations taken
from different test subjects differed significantly in their attractiveness, suggesting variation in
the extracted odours does contribute to differences in individual attractiveness. In addition, flies
from different Lu. longipalpis sibling-species (colonies originating from Jacobina and Lapinha in
Brazil) were found to respond differently in trials, with flies from Jacobina more strongly attracted
to human odour (Hamilton and Ramsoondar 1994). Differences between sibling species in their
attraction to animals and human hosts may contribute to the uneven distribution of leishmaniasis
in Brazil (Ward et al. 1983). Determining which chemicals are responsible for this attraction, and
how sandflies of different species (or sibling-species) respond to them, would be a major step
forward in understanding the ecology of the disease.

While relatively simple bioassays can aid in measuring attraction of sandflies to host-produced
odours, more sophisticated techniques may be required to identify the individual components
responsible. Host odour is comprised of a large number of different chemicals, only a few of
which may elicit any biological response in host-seeking insects. Dougherty et al. (1999) used
gas-chromatography-linked single sensillum recording (GC-SSR) to identify some of the
components of fox odour (Vulpes vulpes Linnaeus) that can be perceived by Lu. longipalpis. Using
electrophysiological readings taken from the ascoid sensillum on the flies’ antennae, females
were found to possess specific neurones perceptive to 16 chemicals in fox gland extract, which
were identified by mass spectrometry. While none of the individual compounds were found to
possess the same attractive qualities as whole fox odour, a synthetic blend of four or more of
these chemicals did attract flies upwind. It appears female Lu. longipalpis may therefore use a
number of volatile kairomones to locate a source of a blood meal, the ability to perceive a range of
chemicals perhaps facilitating their wide host range (Dougherty et al. 1999). Variation in response
to kairomones produced by different animals may be one of the mechanisms through which
Lu. longipalpis sibling-species develop preferences for particular hosts. Electrophysiology may
therefore be one method by which anthropophagic populations, and those likely to be important
leishmaniasis vectors, can be identified.

Although only female sandflies bloodfeed, males of some species are also attracted to hosts as
sites of mating aggregations (Lane et al. 1990, Miles et al. 1976, Quinnell and Dye 1994). Both field
and laboratory studies indicate there are differences in how the sexes respond to components of
host odour (Lutzomyia spp. (Chaniotis 1983); Lu. whitmani (Campbell-Lendrum et al. 1999, Pinto
et al. 2001); Lu. intermedia (Pinto et al. 2001), Lu. longipalpis, (Bray and Hamilton 2007b, Hamilton
and Ramsoondar 1994, Oshaghi et al. 1994), which presumably originate from differences in
both the reason for seeking a host and reproductive strategy (Kelly and Dye 1997, Pinto et al.
2001). In the case of Lu. longipalpis, males arrive earlier and stay longer at the host, presumably
to maximise reproductive success through number of matings, while females are only present for
the time required to mate once and blood feed (Kelly and Dye 1997). In laboratory experiments,
however, males have been found to be less attracted to both human volatiles (Hamilton and
Ramsoondar 1994) and whole hamster odour (Bray and Hamilton 2007b), and more likely to land
on control traps in trials with hamster volatile lures (Oshaghi et al. 1994). Although pheromones
play an important part in the mating system of Lu. longipalpis, initial attraction to lekking sites is

206  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

presumably mediated by host odour. It is possible therefore that males are attracted to cues not
present in laboratory experiments, or that male behaviour is not adequately described by simple
measures of attraction, such as number of flies captured, or number of landings made. Although
males do not transmit leishmaniasis, understanding their responses to host odour in more detail
could aid in both predicting where mating aggregations are likely to occur, and controlling
populations through disruption of normal reproductive behaviour (Bray and Hamilton 2007b).

Identifying components of host odour which are attractive in the laboratory is only the first step
in developing an effective lure for use in the field. To date, the only kairomone found to catch
sandflies in the wild is carbon dioxide. Traps which release CO2 in combination with 1-octen-3-ol
have been produced commercially for controlling other biting insects (Mands et al. 2004), but the
cost of buying and maintaining traps that require gas and power supplies may render this option
uneconomical for use against sandflies, particularly in poor areas where leishmaniasis is prevalent.
Use of volatile blends, which could be manufactured artificially and more easily transported and
sold as formulated lures, may have more potential. However, to date Lu. longipalpis is the only
species for which attraction to volatile chemicals alone has been demonstrated: 1-octen-3-ol (Hall
et al. 1984), one of the components of cattle odour known to attract tsetse flies, failed to increase
attraction of P. argentipes to traps in Sri Lanka (Cameron et al. 1991). More work is needed to
identify those chemicals present in host odour that can attract other important vector species, and
to test these potential lures under field conditions for their efficacy in capturing both males and
females. For those insects known to produce pheromones, combinations of the mating attractant
and kairomones may be the only effective solution.

Sandfly sex pheromones

The sex pheromones of male Lu. longipalpis s.l. are produced by glandular tissue underlying
cuticular papules on tergite 4 or tergites 3 and 4. The underlying glandular tissue and papules
give the appearance of spots of lighter coloured cuticle, these were first observed by Mangabeira
(1969). Because of these males have been classified as one spot (1S) or two spot (2S). Live males,
extracts of males and extracts made in organic solvent from the dissected tergites were found to be
attractive on their own and in combination with host odour to unfed virgin females (Lane and Ward
1984, Lane et al. 1985, Morton and Ward 1989). Chemical fractionation and behavioural analysis
subsequently showed that the most abundant chemical within the extracts was responsible for
the observed attraction (Hamilton et al. 1994).

Coupled gas chromatography mass spectrometry has shown that the active compounds are
terpenes, a class of natural products constructed from three or four five-carbon (C5) isoprene
units. These are a common class of compounds in nature where they are used mainly as defensive
chemicals in plants and termites. However, the Lu. longipalpis compounds are unusual in that the
compounds found in two of the members of the species complex have an additional methyl group
probably added in the early biosynthesis of the molecule (Hamilton et al. 1999c) i.e. they have 16
carbons (C16, mw 218). Thus the sex pheromone of one member of the complex has been shown
to be the novel C16 (3×C5 + CH3) monocyclic terpene (S)-9-methylgermacrene-B (Hamilton et al.
1996a, 1999a) and another, the novel C16 bicyclic terpene, 3-methyl-α-himachalene (Hamilton
et al. 1996b, 1999b). The sex pheromones of another two members of the species complex have
been shown to be different cembrene isomers i.e. they are C20 (mw 272) monocyclic diterpenes
(4 x C5). Work to fully characterise these compounds is continuing.

Olfaction in vector-host interactions 207

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

Studies to determine if terpenes are present in other New World sandfly species have been
undertaken and show that although present in other Lutzomyia species they are not ubiquitous.
It appears that Lutzomyia spp. can be subdivided into three groups; those whose males produce
terpenes and have cuticular papules e.g. Lu. longipalpis, Lu. lenti (Mangabeira), and Lu. carmelinoi
Ryan; those that do not produce terpenes but still have the associated papules e.g. Lu. evandroi
Costa Lima and Antunes, Lu. tejadai Galati and Caceres and Lu. migonei (França) and those that
have neither terpenes nor papules e.g. Lu. whitmani and Lu. neivai (Pinto) (Hamilton et al. 2002). In
some cases the compounds are oxygenated methylsesquiterpenes but more commonly they are
diterpenes (Hamilton and Ward 1994, Hamilton et al. 1999c).

An investigation revealed that two Old World species Sergentomyia minuta Rondani and S.
fallax Parrot also produce a range of terpenes. Both species have cuticular papules similar to
those found on Lu. longipalpis. Interestingly both species produced a much greater range of
sesquiterpenes and methyl sesquiterpenes than Lutzomyia spp. Sergentomyia fallax was found to
produce eight known sesquiterpenes (C15, mw 204) the most abundant being allo-aroma-dendr-
9-ene (53%) and β-barbatene (23%). Sergentomyia minuta produces a novel unsaturated tricyclic
homosesquiterpene (C16, mw 216) as the major terpene (68%) with (E)-β-caryophyllene (C15, mw
204) as the next most abundant terpene (15%). The remaining five compounds that were present
in amounts greater than 1% of the total terpene content are novel predominantly unsaturated
homosesquiterpene hydrocarbons. Although the biological function of terpenes in species other
than Lu. longipalpis has not been confirmed it is likely that they are sex pheromones. It may also
be possible to infer that the evolution of terpenes in the Phlebotominae is from highly complex
tricyclic C16 compounds to the monocyclic C20 compounds found in Lutzomyia spp. (J.G.C.H.
2008, unpublished observations).

The presence of sex pheromones in Old World Phlebotomus spp. is an open question. Terpenes
such as we have found in some Lutzomyia and Sergentomyia spp. appear to be absent e.g. P.
argentipes, P. papatasi Scopoli, P. orientalis Parrot and P. duboscqi Leveu-Lemaire. However we have
recently carried out extensive behavioural studies with P. papatasi from Tunisia and the results
indicate that females are attracted in an olfactometer to live male headspace volatiles (I Chelbi
and JGC Hamilton 2008 unpublished observations). This would suggest that sex pheromones may
be more widely distributed in sandflies than previously recognised.

Exploitation of sandfly pheromones

Oviposition pheromones and stimulants may offer potential for exploitation through monitoring or
control traps. However, to date little interest has been shown in their development, with most effort
focussing on the potential application of male pheromones in vector management. Interestingly,
because male pheromones attract females a significantly greater impact could be made on the
fecundity of the population than if a female pheromone, attractive to males, was used. We have
synthesised the sex pheromone 9-methylgermacrene-B, and an analogue 9-methylgermacrene
(Krishnakumari et al. 2004). These compounds can be synthesised from a readily available plant
intermediate, the analogue is more stable than the natural sex pheromone and has similar activity
in laboratory bioassays (JGC Hamilton and K Krishnakumari, unpublished results). We are currently
formulating these chemicals and carrying out laboratory and field experiments to determine
their attractiveness. Our preliminary results indicate that the synthetic compounds are attractive
to both male and female Lu. longipalpis (DP Bray, RP Brazil, K Krishnakumari and JGC Hamilton,
unpublished results).

208  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

We have also found in laboratory studies that although sex pheromone is attractive to female
sandflies, sex pheromone presented in combination with host odour is significantly more
attractive. This strong synergistic interaction between host odour and male pheromone suggests
that synthetic pheromone lures should be positioned near host animals until the attractive
elements of host odour have been identified and incorporated into the sex pheromone lure.

Chemical communication and mating behaviour in sandflies

Identifying semiochemicals that are attractive to sandflies, and which could therefore function
as lures, represents an important stage in developing novel methods of control effective
against vectors of leishmaniasis. In agriculture, integrated pest management strategies, which
use insecticides in conjunction with pheromone lures and mating disruption techniques, have
proven effective against a number of agricultural pests (Shani 2000). A similar approach applied
to sandflies could mean cheaper, more efficient control, less reliant on spraying insecticides
(Alexander et al. 2003, Kelly and Dye 1997) However, such strategies have at their foundation
a thorough knowledge of the mating system of the target insect (Girling and Cardé 2006). For
many species of sandfly, our current data may be insufficient: unless we understand how different
chemical signals are used, and their function in mating and courtship behaviour, developing
effective methods of pheromone-based control is likely to be difficult.

Few studies have explored the mating system of sandflies in any detail, with Lu. longipalpis the
most well studied. Most attention in this species had focussed on the ability of male-produced
pheromones to attract female flies from a distance. As these chemical signals only attract potential
mates from the same pheromone-producing sibling species, they are likely to be important in
maintaining reproductive isolation between different members of the Lu. longipalpis complex
(Ward and Morton 1991). While these chemicals may function over considerable distances, their
role in the small-scale interactions which occur between courting pairs remains unclear.

Laboratory studies of mating behaviour in Lu. longipalpis have shown that males establish
territories on a host animal, which they defend through wing-flapping displays that escalate into
occasional fights (Jarvis and Rutledge 1992). When a visiting female is detected, males approach
them while wing-flapping, and continue to flap during copulation (Jarvis and Rutledge 1992,
Jones and Hamilton 1998, Ward et al. 1988). The importance of male wing-flapping in courtship
has been demonstrated through mating trials: the occurrence of male wing-flapping directed at
females is a predictor of eventual copulation (Bray and Hamilton 2007a) and males which display
for longer are more likely to be chosen by females (Jones and Hamilton 1998). However, the
methods through which these behaviours signal to females are unknown. These displays may have
a chemical component, and it has been suggested that the movement of the wings could help
to disperse sex pheromones towards both males and females (Jones and Hamilton 1998, Lane et
al. 1985, Ward et al. 1989), though this remains to be confirmed experimentally. In mate choice
experiments, females were found to be more likely to choose males that had more pheromone
remaining in their glands after the experiment, but there was no correlation between time spent
wing flapping and amount of pheromone stored (Jones and Hamilton 1998).

Young male Lu. longipalpis can achieve mating prior to onset of pheromone production, and are
occasionally chosen by females over older, pheromone producing males (Boufana et al. 1986,
Jones 1997, Jones and Hamilton 1998). This suggests that the currently identified male pheromone
is not the only signal involved in Lu. longipalpis mating behaviour. Sound recordings of courtship
in this species have demonstrated that wing-flapping also produces audio signals that resemble

Olfaction in vector-host interactions 209

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

the love songs integral to mating behaviour in Drosophila (Souza et al. 2002, 2004, Talyn and
Dowse 2004, Ward et al. 1988). Furthermore, the audio patterns produced during copulation differ
between sibling species of Lu. longipalpis, suggesting they may play a role in reproductive isolation
(Ward et al. 1988). Females also perform short wing-flapping displays, which appears to signal
receptivity to a potential mate (Jones and Hamilton 1998), and the occurrence of this behaviour
during courtship is a strong predictor of eventual copulation (Bray and Hamilton 2007a). As no
female pheromone has been identified, it is assumed that these displays produce audio signals
only. Playback experiments, similar to those conducted with Drosophila, are needed to isolate the
audio and chemical components of wing-flapping displays, in order to determine their relative
importance in reproductive isolation and female mate choice Bray and Hamilton 2007a, Souza et
al. 2004).

Recently, Bray and Hamilton (2007a) conducted more detailed studies of the small-scale
interactions that occur between male and female Lu. longipalpis prior to copulation. They found
that touching was a common occurrence during courtship, initiated by both males and females,
with physical contact most often taking place between the tips of the legs and the antennae. This
behaviour may represent the transfer and detection of cuticular hydrocarbons, important signals
in courtship of many insect species (Howard and Blomquist 2005), but previously overlooked
both in Lu. longipalpis and sandflies in general. In Drosophila, these signals can provide a
range of information to a female about a potential mate prior to copulation, including age and
sexual experience (Ferveur 2005). Chemical analyses are required to determine whether similar
compounds are present in both Lu. longipalpis and other species of sandfly. If so, behavioural
studies will be needed to determine their role in courtship, and their potential for exploitation in
control programmes.

Interest in sandfly chemical ecology extends beyond developing novel methods of vector
management, to the study of broader questions regarding the role of chemical communication
in mate choice and reproductive isolation. Within a Lu. longipalpis lek, a small number of males
will receive the majority of matings, a product more of female choice than male competition
(Jones et al. 1998). How females discriminate between males, and the advantage of doing so are
unclear, though there may be benefits in terms of the viability and attractiveness of offspring
produced. In the laboratory, females prefer middle-aged males (4 to 6 d old), producing a greater
proportion of hatching eggs when mated with males less than 8 d old, and attractive males tend
to have attractive sons, demonstrating that there is a heritable component to male reproductive
fitness (Jones et al. 1998). There is an intriguing possibility that differences in female preferences
and associated male traits may have led to the reproductive isolation now observed between
sibling species. Clearly, this proposal is worthy of attention for those interested in speciation, and
the current population structure of Lu. longipalpis.

Descriptions of sandfly courtship behaviour in species other than Lu. longipalpis have been limited
to qualitative descriptions of behaviour carried out in the laboratory, or notes based on field
observations. However, the information that has been collected suggests that there are aspects
of courtship that are common across species. Wing-flapping, similar to that described in Lu.
longipalpis, has been observed to occur in Lu. lichyi Floch et Abonnenc (Hamilton et al. 1999c), Lu.
vexator Shannon (Chaniotis 1967), P. argentipes (Lane et al. 1990), P. orientalis (Ashford 1974), P.
duboscqi (Valenta et al. 2000), P. longipes Parrot and Martin (Gemetchu 1976) and P. martini Parrot
(Beach et al. 1983). Males of several species also sway their abdomen from side to side when in
proximity with a female (Lu. vexator (Chaniotis 1967), P. longipes (Gemetchu 1976), P. martini (Beach
et al. 1983)). Both wing-flapping and abdomen swaying could act to disperse hitherto unidentified

210  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

pheromones. If so, there may be considerable potential in developing semiochemical-based

control measures effective against a range of sandfly species. It is hoped the progress being made
with Lu. longipalpis will encourage further research into the mating systems of other important
leishmaniasis vectors, in both the New and Old World.

Future directions

An enormous amount of basic investigation into the chemical ecology of sandflies remains to be
carried out. Although the majority of our knowledge so far is based upon the ‘Drosophila’ of the
sandfly world, Lu. longipalpis, there are at least 1000 species of sandfly where we know virtually
nothing of their chemically moderated inter and intraspecific interactions. From a phylogenetic
basis alone there is fundamental and useful information to be gained from identifying pheromones,
to say nothing of the potential for their application in vector control.

Furthermore, studies of basic mating behaviour are required to determine how sandflies employ
chemical signals (critical to developing semiochemicals as tools for vector management), and to
understand the ecological forces driving evolution of pheromones in the Phlebotominae.

In the longer term, there is also the potential to develop new, innovative molecular approaches to
controlling haematophagous insects through identifying those genes responsible for pheromone
synthesis. For example, Chertemps et al. (2006) recently demonstrated that RNAi-knock-down of
the desaturase gene desatF in Drosophila melanogaster Meigen led females to produce more male
contact pheromone and less female pheromone than normal, resulting in reducing courtship
and mating behaviour. Using newly available gene sequence data in conjunction with modern
molecular techniques, it may similarly be possible to examine the relative importance of different
chemical signals in the mating behaviour of the Psychodidae, through targeting of homologous
genes with a role in pheromone production and to manipulate them for vector control.

While research continues into identifying semiochemicals attractive to sandflies, there remains
little information on the receptors responsible for their detection. So far studies have concentrated
only on the ascoid sensillae, which contain several sex pheromone receptors (Dougherty et al.
1999). However, at least eight other sensillae types are present on the antennal segments of blood
feeding sandflies (Boufana 1990) whose function is unknown. Similarly, we need to understand
how haematophagous insects perceive semiochemicals at the molecular level, presumably
through use of odour-binding proteins (see Chapter 2, this volume). The genes for such proteins
and the cDNAs which encode them in the mosquitoes Culex quinquefasciatus Say and Anopheles
gambiae Giles sensu stricto have now been cloned and recombinant proteins generated (Logan
and Birkett 2007). Sandflies are yet to become the focus of a similar endeavour.

The questions raised above are clearly important and from our own experience can be laborious
and time consuming to answer. In the twenty plus years we have studied sandfly chemical ecology
there have been some significant advances in our knowledge. However, it must be emphasised
that there is a real need for wider participation and perspectives in the topic if the potential value
of such work is to be fully and quickly realised in the future.

References

Alexander B and Maroli M (2003) Control of phlebotomine sandflies. Med Vet Ent 17: 1-18.

Olfaction in vector-host interactions 211

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

Ashford RW (1974) Sandflies (Diptera: Phlebotomidae) from Ethiopia: taxonomic and biological notes. J Med Ent 11:
605-616.
Barth R (1961) Sobre o aparelho genital interno do macho de Phelbotomus longipalpis Lutz & Neiva, 1912 (Diptera:
Psychodidae). Mem Inst Oswaldo Cruz 59: 23-36.
Bauzer LGSR, Souza NA, Maingon RD and Peixoto AA (2007) Lutzomyia longipalpis in Brazil: a complex or a single species?
A mini-review. Mem Inst Oswaldo Cruz 102: 1-12.
Beach R, Young DG and Mutinga MJ (1983) New phlebotomine sandfly colonies: rearing Phlebotomus martini,
Sergentomyia schwertzi and Sergentomyia africana (Diptera: Psychodidae). J Med Ent 20: 579-584.
Boufana B (1990) The tergal gland and antennal sensilla of the sandfly Lutzomyia longipalpis (Diptera: Psychodidae).
PhD Dissertation, University of Liverpool, Liverpool, UK.
Boufana B, Ward RD and Phillips A (1986) Development of the tergal pheromone gland in male Lutzomyia longipalpis
(Diptera: Psychodidae). Trans Roy Soc Trop Med Hyg 80: 33.
Bray DP and Hamilton JGC (2007a) Courtship behaviour in the sandfly Lutzomyia longipalpis, the New World vector of
visceral leishmaniasis. Med Vet Ent 21: 332-338.
Bray DP and Hamilton JGC (2007b) Host odor synergizes attraction of virgin female Lutzomyia longipalpis (Diptera:
Psychodidae). J Med Ent 44: 779-787.
Cameron MM, Amerisinghe FP and Lane RP (1991) The field response of Sri Lanka sandflies and mosquitoes to synthetic
cattle derived attractants. Parassitologia 33: 119-126.
Campbell-Lendrum DH, Pinto MC, Brandão-Filho SP, de Souza AA, Ready PD and Davies CR (1999) Experimental
comparison of anthropophily between geographically dispersed populations of Lutzomyia whitmani (Diptera:
Psychodidae). Med Vet Ent 13: 299-309.
Chaniotis BN (1967) The biology of California Phlemotomus (Diptera: Psychodidae) under laboratory conditions. J Med
Ent 4: 221-233.
Chaniotis BN (1983) Improved trapping of phlebotomine sand flies (Diptera: Psychodidae) in light traps supplemented
with dry ice in a neotropical rain forest. J Med Ent 20: 222-223.
Chertemps T, Duportets L, Labeur C, Ueyama M and Wicker-Thomas C (2006) A female-specific desaturase gene
responsible for diene hydrocarbon biosynthesis and courtship behaviour in Drosophla melanogaster. Ins. Mol.
Biol. 15: 465-473.
Christensen HA and Herrer A (1973) Attractiveness of sentinel animals to vectors of leishmaniasis in Panama. Am J Trop
Med Hyg 22: 578-584.
Colvin J and Gibson G (1992) Host-seeking behaviour and management of tsetse. Ann Rev Ent 37: 21-40.
Dougherty MJ, Guerin P and Ward RD (1995) Identification of oviposition attractants for the sandfly Lutzomyia longipalpis
(Diptera: Psychodidae) present in volatiles of faeces from vertebrates. Phys Ent 20: 23-32.
Dougherty MJ, Guerin PM, Ward RD and Hamilton JGC (1999) Behavioural and electrophysiological responses of the
Phlebotomine sandfly Lutzomyia longipalpis (Diptera: Psychodidae) when exposed to canid host odour kairomones.
Phys Ent 24: 251-262.
Dougherty MJ and Hamilton JGC (1997) Dodecanoic acid is the oviposition pheromone of Lutzomyia longipalpis. J
Chem Ecol 23: 2657-2671.
Dougherty MJ, Ward RD and Hamilton JGC (1992) Evidence for the accessory glands as the site of production of the
oviposition pheromone of Lutzomyia longipalpis (Diptera: Psychodidae). J Chem Ecol 18: 1165-1175.
ElNaiem D and Ward RD (1991) Response of the sandfly Lutzomyia longipalpis to an oviposition pheromone associated
with conspecific eggs. Med Vet Ent 5: 87-91.
ElNaiem D and Ward RD (1992) Oviposition attractants and stimulants for the sandfly Lutzomyia longipalpis (Diptera:
Psychodidae). J Med Ent 29: 5-12.
Ferveur JF (2005) Cuticular hydrocarbons: their evolution and roles in Drosophila pheromonal courtship. Behav Genet
35: 279-293.
Fuerborn HJ (1922) Der sexuelle Reizapparat (Shmuck-, Duft- und Beruhrungsorgane) der Psychodiden nach
biologischen und physiologischen Gesichtspunkten untersucht. Arch für Natur A 88: 1-137.

212  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

Gemetchu T (1976) The biology of a laboratory colony of Phlebotomus longipes Parrot and Martin (Diptera:
Phlebotomidae). J Med Ent 12: 661-671.
Gillies MT (1980) The role of carbon dioxide in host finding by mosquitoes (Diptera: Culicidae): a review. Bull Ent Res
70: 525-532.
Girling RD and Cardé RT (2006) Analysis of the courtship behavior of the navel orangeworm, Amyelois transitella (Walker)
(Lepidoptera: Pyralidae), with a commentary on methods for the analysis of sequences of behavioral transitions.
J Insect Behav 19: 497-520.
Hall DR, Beevor PS, Cork A and Vale GA (1984) 1-octen-3-ol, a potent olfactory stimulant and attractant for tsetse
isolated from cattle odours. Insect Sci Appl 5: 335-339.
Hamilton JGC, Brazil RP, Campbell-Lendrum D et al. (2002) Distribution of putative male sex pheromones among
Lutzomyia sandflies (Diptera: Psychodidae). Ann Trop Med Parasit 96: 83-92.
Hamilton JGC, Brazil RP, Morgan ED and Alexander B (1999c) Chemical analysis of oxygenated homosesquiterpenes: a
putative sex pheromone from Lutzomyia lichi (Diptera: Psychodidae). Bull Ent Res 89: 139-145.
Hamilton JGC, Dawson GW and Pickett JA (1996a) 9-Methylgermacrene-B, a novel homosesquiterpene from sex
pheromone glands of Lutzomyia longipalpis (Diptera: Psychodidae) from Lapinha, Brazil. J Chem Ecol 22: 1477-1491.
Hamilton JGC, Dawson GW and Pickett JA (1996b) 3-Methyl-α-himachalene; sex pheromone of Lutzomyia longipalpis
(Diptera: Psychodidae) from Jacobina, Brazil. J Chem Ecol 22: 331-2340.
Hamilton JGC, Dougherty MJ and Ward RD (1994) Sex pheromone activity in a single component of tergal gland extract
of Lutzomyia longipalpis (Diptera: Psychodidae) from Jacobina, Northeastern Brazil. J Chem Ecol 20: 141-151.
Hamilton JGC, Hooper AM, Mori K, Pickett JA and Sano S (1999b) 3-Methyl-α-himachalene confirmed, and the relative
stereochemistry defined, by synthesis as the sex pheromone of the sandfly Lutzomyia longipalpis from Jacobina,
Brazil. Chem Comms 355-356.
Hamilton JGC, Ibbotson HC, Hooper AM, Mori K, Pickett JA and Sano S (1999a) 9-Methylgermacrene-B confirmed
by synthesis as the sex pheromone of the sandfly Lutzomyia longipalpis from Lapinha, Brazil, and the absolute
stereochemistry defined as 9S. Chem Comms 2335-2336.
Hamilton JGC and Krishnakumari K (2004) Sex pheromone analogue for controlling Lutzomyia longipalpis. BR Patent
I0500799, 20 Sept 2006.
Hamilton JGC and Ramsoondar TMC (1994) Attraction of Lutzomyia longipalpis to human skin odours. Med Vet Ent 8:
375-380.
Hamilton JGC and Ward RD (1994) Chemical analysis of a putative sex pheromone from Lutzomyia pessoai (Diptera:
Psychodidae). Ann Trop Med Parasit 88: 405-412.
Howard RW and Blomquist GJ (2005) Ecological, behavioral, and biochemical aspects of insect hydrocarbons. Ann Rev
Ent 50: 371-393.
Jarvis EK and Rutledge LC (1992) Laboratory observations on mating and leklike aggregations in Lutzomyia longipalpis
(Diptera, Psychodidae). J Med Ent 29: 171-177.
Jones TM (1997) Sexual selection in the sandfly, Lutzomyia longipalpis. PhD Thesis, University of London.
Jones TM and Hamilton JGC (1998a) A role for pheromones in mate choice in a lekking sandfly. Anim Behav 56: 891-898.
Jones TM, Quinnell RJ and Balmford A (1998b) Fisherian flies: benefits of female choice in a lekking sandfly. Proc R Soc
Lond B Biol Sci 265: 1651-1657.
Kelly DW and Dye C (1997) Pheromones, kairomones and the aggregation dynamics of the sandfly Lutzomyia longipalpis.
Anim Behav 53: 721-731.
Knudsen AB, Lewis DJ, Tesh RB, Rudnick A, Jeffery J and Singh I (1979) Phlebotomine sand flies (Diptera: Psychodidae)
from a primary hill forest in western Malaysia. J Med Ent 15: 286-291.
Krishnakumari B, Sarita Raj K and Hamilton JGC (2004) Synthesis of 9-methylgermacrene from germacrone, an active
analogue of (S)-9-methylgermacrene-B, sex pheromone of the Phlebotomine sandfly, Lutzomyia longipalpis,
from Lapinha Brazil. Paper presented at the IUPAC international conference on biodiversity and natural products:
chemistry and medical applications, New Delhi, India, 26-31 January 2004.

Olfaction in vector-host interactions 213

Daniel P. Bray, Richard D. Ward and James G.C. Hamilton

Lane R and Ward RD (1984) The morphology and possible function of abdominal patches in males of two forms of
Lutzomyia longipalpis (Diptera: Phlebotominae). Cahiers ďOffice de la Recherche Scientifique et Technique Outre-
Mer. Ent Méd Parasit 22: 245-249.
Lane R, Phillips A, Molyneux DH, Procter G and Ward RD (1985) Chemical-analysis of the abdominal glands of two forms
of Lutzomyia longipalpis - site of a possible sex-pheromone. Ann Trop Med Parasit 79: 225-229.
Lane RP, Pile MM and Amerasinghe P (1990) Anthropophagy and aggregation behaviour of the sandfly Phlebotomus
argentipes in Sri Lanka. Med Vet Ent 15: 132-139.
Logan JG and Birkett MA (2007) Semiochemicals for biting fly control: their identification and exploitation. Pest Man
Sci 63: 647-657.
Maingon RD, Ward RD, Hamilton JG, Bauzer LG and Peixoto AA (2008) The Lutzomyia longipalpis species complex: does
population sub-structure matter to Leishmania transmission? Trends in Parasit 24: 12-17.
Mands V, Kline DL and Blackwell A (2004) Culicoides midge trap enhancement with animal odour baits in Scotland.
Med Vet Ent 18: 336-342.
Mangabeira OF (1969) Sóbre a sistemática e biologia dos Phlebotomus do Ceará. Rev Bras Malariol Doencias Trop 12:
3-26.
Miles CT, Foster WA and Christensen HA (1976) Mating aggregations of male Lutzomyia sandflies at human hosts in
Panama. Trans R Soc Trop Med Hyg 70: 532.
Montoya-Lerma J and Lane RP (1996) Factors affecting host preference of Lutzomyia evansi (Diptera: Psychodidae), a
vector of visceral leishmaniasis in Columbia. Bull Ent Res 86: 43-50.
Morton IE and Ward RD (1989) Laboratory response of female Lutzomyia longipalpis sandflies to a host and male
pheromone source over distance. Med Vet Ent 3: 219-223.
Nevatte T (2006) Changes in golden hamster behaviour and attractiveness to the sand fly Lutzomyia longipalpis as a
result of Leishmania infection. PhD thesis, University of Keele, Keele, UK.
Nigam Y and Ward RD (1991) The effects of male sandfly pheromone and host factors as attractants for female Lutzomyia
longipalpis. Physiol Ent 16: 305-312.
O’Shea B, Rebollar-Tellez E, Ward RD, Hamilton JG, el Naiem D and Polwart A (2002) Enhanced sandfly attraction to
Leishmania infected hosts. Trans Roy Soc Trop Med Hyg 96: 117-118.
Oshaghi MA, McCall PJ and Ward RD (1994) Response of adult sandflies, Lutzomyia longipalpis (Diptera: Psychodidae),
to sticky traps baited with host odour and tested in the laboratory. Ann Trop Med Parasitol 88: 439-444.
Pinto MC, Campbell-Lendrum DH, Lozovei AL, Teodoro U and Davies CR (2001) Phlebotomine sandfly responses to
carbon dioxide and human odour in the field. Med Vet Ent 15: 132-139.
Quinnell RJ and Dye C (1994) Correlates of the peridomestic abundance of Lutzomyia longipalpis (Diptera, Psychodidae)
in Amazonian Brazil. Med Vet Ent 8: 219-224.
Quinnell RJ, Dye C and Shaw JJ (1992) Host preferences of the phlebotomine sandfly Lutzomyia longipalpis in Amazonian
Brazil. Med Vet Ent 6: 195-200.
Ready PD (1978) The feeding habits of laboratory-bred Lutzomyia longipalpis (Diptera: Psychodidae). J Med Ent 14:
545-552.
Rebollar-Tellez EA, Hamilton JGC and Ward RD (1999) Response of female Lutzomyia longipalpis to host odour
kariomones from human skin. Phys Ent 24: 220-226.
Shani A (2000) Chemical communication agents (pheromones) in integrated pest management. Drug Dev Res 50:
400-405.
Souza NA, Vigoder FM, Araki AS, Ward RD, Kyriacou CP and Peixoto AA (2004) Analysis of the copulatory courtship songs
of Lutzomyia longipalpis in six populations from Brazil. J Med Ent 41: 906-913.
Souza NA, Ward RD, Hamilton JG, Kyriacou CP and Peixoto AA (2002) Copulation songs in three siblings of Lutzomyia
longipalpis (Diptera: Psychodidae). Trans Roy Soc Trop Med Hyg 96: 102-103.
Takken W (1991) The role of olfaction in host-seeking of mosquitos - a review. Insect Sci Appl 12: 287-295.
Talyn BC and Dowse HB (2004) The role of courtship song in sexual selection and species recognition by female
Drosophila melanogaster. Anim Behav 68: 1165-1180.

214  Olfaction in vector-host interactions

 9. The chemical ecology of sandflies

Valenta DT, Killick-Kendrick R and Killick-Kendrick M (2000) Courtship and mating by the sandfly Phlebotomus duboscqi,
a vector of zoonotic cutaneous leishmaniasis in the Afrotropical region. Med Vet Ent 14: 202-212.
Ward RD and Hamilton JGC (2002) Chemical and auditory signals in Phlebotomine sandfly behaviour. In: Killick-Kendrick
R (ed) Canine leishmaniasis: moving towards a solution. Proceedings 2nd International Canine Leishmaniasis Forum,
Seville, Spain.
Ward RD and Morton IE (1991) Pheromones in mate choice and sexual isolation between siblings of Lutzomyia
longipalpis (Diptera: Psychodidae). Parassitologia 33: 527-533.
Ward RD, Morton IE, Lancaster V, Smith PA and Swift A (1989) Bioassays as an indicator of pheromone communication in
Lutzomyia longipalpis (Diptera: Psychodidae). In: Hart DT (ed) Leishmaniasis, the current status and new strategies
for control. NATO ASI Series A: Life Sciences, vol 163. Springer, New York, USA, pp 239-247.
Ward RD, Phillips A, Burnet B and Marucondes CB (1988) The Lutzomyia longipalpis complex: reproduction and
distribution. In: Service MW (ed) Biosystematics of haematophagous insects. Oxford University Press, Oxford, UK,
pp 257-269.
Ward RD, Ribeiro AL, Ready PD and Murtagh A (1983) Reproductive isolation between different forms of Lutzomyia
longipalpis (Lutz and Neiva), (Diptera: Psychodidae), the vector of Leishmania dovani chagasi Cunha & Chagas and
its significance to kala-azar distribution in South America. Mem Inst Oswaldo Cruz 78: 269-280.

Olfaction in vector-host interactions 215

10. Understanding and exploiting olfaction for the surveillance and
control of Culicoides biting midges

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

Abstract

Culicoides midges (Diptera: Ceratopogonidae) are found worldwide with the exception of only
a few countries including New Zealand, Patagonia, the Hawaiian Isles and Antarctica. They are a
nuisance pest to human beings, but transmit a number of diseases that mainly affect livestock.
Like many haematophagous insects, midges have evolved highly developed and sensitive
olfactory mechanisms which allow them to locate a potential mate or a suitable vertebrate host.
This chapter provides a brief overview of the Culicoides midge lifecycle, its status as a vector of
disease and offers an extensive review of Culicoides chemical ecology research, which has led
to the identification of semiochemicals that could be exploited to control them. Host location
processes can be influenced by many factors including midge physiology, host kairomones, the
complex interactions of other volatile chemicals that confer differential attractiveness of hosts
and also physical cues such as light levels and climatic conditions. These factors are considered in
detail and future areas of work are discussed.

Keywords: biting midges, Ceratopogonidae, behaviour, olfaction

Introduction

Of the five major sensory systems (olfaction, taste, touch, vision and sound) by which organisms
perceive their environment olfaction can be said to be the most important for biting flies such as
midges. This sensory modality provides adult female Culicoides (Diptera: Ceratopogonidae) with
essential cues for the location and identification of hosts that are required for a blood meal in order
to grow their eggs. The detection of small, volatile compounds borne on the wind provides the
main cues for activation and upwind flight. This, coupled with other cues such as heat, moisture
and visual cues at closer range and taste on landing drive the behavioural sequence known as
‘host location’. Although some of these stimuli may be involved in other aspects of midge ecology,
their role in the location of a suitable host has been mostly studied and has been exploited to
manipulate the behaviour of host-seeking adult females. For example, odour-baited traps can be
used to lure, capture and kill adults. Although, this method has been used with some success with
other biting Diptera (e.g. Glossina spp. tsetse flies), research into the olfactory processes involved
in midge-host interactions is not fully understood but certainly has the potential to provide a
potential means of Culicoides control.

Few insects that attack man are more annoying during their peak of activity than Culicoides midges
but previous control efforts have been met with limited success, with repellents not providing full
protection and with insecticides failing to provide adequate control of populations. To this end,
a thorough understanding of the host-location process of each Culicoides species of medical and
veterinary importance is still needed. This chapter will review current knowledge of olfaction in
midge-host interactions including knowledge of experimental and commercially available traps.
Protective measures are essential not only for man but also for domestic animals against midges
that vector diseases such as Bluetongue of sheep and cattle and African Horse sickness. Effective

Olfaction in vector-host interactions 217

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

control of targeted species is urgently required, particularly in light of the recent Bluetongue
outbreak in Europe.

Culicoides midge biology and ecology

Lifecycle

Culicoides midges are members of the order Diptera, one of the largest orders within the class
Insecta. They are one of four haematophagous genera within the Ceratopogonidae family and
are considered to have the most medical and veterinary importance due to their biting action
and ability to vector pathogens that cause disease in livestock (Kettle 1965, Linley et al. 1983).
There are approximately 1,400 species of biting midge and they have a wide range of larval and
adult biologies.

Midges have mostly subaquatic or subterrestrial immature stages, developing in/nearby pools,
streams or bogs. However, some can also be found in a variety of terrestrial or semiterrestrial
locations such as animal droppings and vegetation. The northern Palaearctic species complex C.
obsoletus, implicated in Bluetongue virus (BTV) transmission in northern Europe, is fairly catholic
in its breeding habitats occurring not only in dung but also a variety of habitats from woodland to
open country and bogland to marshland (Hill 1947, Kettle and Lawson 1952). Culicoides obsoletus
Meigen s.l. is strongly associated with shaded habitats and a higher green leaf density than the
Old World BTV vector C. imicola Kieffer that is found in much drier, warmer habitats (e.g. straw/
dung mix at dripping water pipes) in full sun (Conte et al. 2007). Many other species troublesome
to humans breed in brackish, sandy habitats along coastlines.

Culicoides are holometabolous, having four larval instars followed by a pupal stage where the adult
form develops before emergence, although some species can over-winter as a diapausing fourth
instar (Figure 1). Climate change, and more specifically alterations in temperature, may affect
larval stages of some Culicoides species. Allingham (1991) and Bishop et al. (1996) demonstrated
that Culicoides brevitarsis Kieffer emerge between 17-36 °C, with most emergence occurring in the
25-36 °C range. They also showed that average temperature in the field prior to their experiment
starting, i.e. when C. brevitarsis was allowed to oviposit, played a role in determining development
time. This species can survive for periods outside the optimum temperature range, but only
emerge if conditions become favourable again within 50 days. Additionally, a return to favourable
temperatures is not necessarily problem free. When temperatures in one experiment were raised
to 25 °C after an extended period at 12 °C the adults that emerged appeared too small to produce
viable offspring. The effect of climate change on the interaction between adult Culicoides midges
and their host is an important and topical issue which is discussed in more detail later in this
chapter.

The adult midge has a body length of about 1-2 mm, a weight of approximately 0.5 μg and has two
compound eyes. Most adult females require blood to develop eggs, although some species, e.g. C.
impunctatus Goetghebuer, are autogenous and therefore do not require a blood meal to develop
their first batch of eggs (Blackwell et al. 1992, Boorman and Goddard 1970). All female midges
must locate hosts to obtain a blood meal and this complex process can be heavily influenced and,
to some extent, controlled by various environmental factors such as solar radiation, wind velocity,
temperature and humidity, all of which are thought to limit flight activity (Bhasin 1996, Blackwell
et al. 1997). This aspect of their ecology is discussed in more detail later in this chapter.

218  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Human host
Human/animal host

ADULT

EGGS
soil surface
PUPA

1 st
instar

4 th LARVAE
instar 2 nd
3 rd instar
instar

Figure 1. Illustrative representation of a female Culicoides midge lifecycle (Lynda Castle, Rothamsted Research,
Harpenden, UK).

Blood meals are obtained preferentially from a variety of vertebrate hosts. For example, cattle,
sheep and deer are among the predominant hosts of C. impunctatus although they will also readily
feed on human beings (Blackwell et al. 1994b, Mands et al. 2004). ELISA tests on blood-fed C.
imicola collected at farms in Ondersterpoort, RSA revealed that most midges had fed more than
once on horse, cattle and sheep with horse being the most frequent (AWA Brown, AJ Mordue
(Luntz) and GJ Venter, unpublished observation). The coat of cattle offers midges a greater surface
area of accessible skin from which to feed than is available on sheep whose fleece they cannot
penetrate so easily (Muller and Murray 1977). After blood feeding, as egg batches mature, females
are thought to rest before seeking a suitable oviposition site (Boorman and Goddard 1970).

Vector status/ nuisance value and geographical distribution

Within the Ceratopogonidae, Culicoides is the most important genus regarding human and livestock
comfort/welfare and health (Kettle 1965, Linley et al. 1983). Midges are found worldwide with
the exception of only a few countries including New Zealand, Patagonia, the Hawaiian Isles and
Antarctica. They have evolved to inhabit most temperate and tropical regions and therefore, in this
chapter, we focus on those that are considered to be of high medical and veterinary importance.

Olfaction in vector-host interactions 219

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

With emerging vector-borne diseases in certain countries it is vital that we understand which
species spread pathogens, the hosts that are most heavily afflicted and the midge geographical
distribution and habitat.

Diseases transmitted by midges mainly affect livestock. For example, Culicoides are well documented
as vectors of livestock arboviruses (Kettle 1984, Purse et al. 2005), including bluetongue virus of
sheep and cattle (BTV), African horse sickness virus (AHSV) and epizootic haemorrhagic disease
(EHD) of deer (Mellor 1990, 1993, 1994, Sellers and Maarouf 1991). Some species also vector the
filarial nematode Onchocerca cervicalis Railliet and Henry, which affects horses (Mellor 1972), or
cause ‘sweet itch’, which is not a disease, but a distressing skin condition of horses caused by
midge saliva (Mellor and McCaig 1974, Townley et al. 1984).

Bluetongue virus infects domestic and wild ruminants and can cause more than 70% mortality in
sheep (Jennings and Mellor 1988). It is prevalent in Central, West and South America, Africa, Asia,
the Middle East and Australia (Holbrook and Tabachnick 1995, Mellor 1990). More recently it has
spread throughout Europe. Culicoides imicola is believed to be one of the main vectors of BTV.
Following the introduction of C. imicola to Italy and the subsequent establishment of BTV (Baylis et
al. 2001) several studies have been conducted to predict the likely spread of BTV (Tatem et al. 2003,
Wittmann et al. 2001). In 2005-2006, Takken et al. (2008) investigated the presence and distribution
of Culicoides spp. at different habitats in the Netherlands to identify where potential BTV outbreaks
may occur. This information was coupled with climate data to develop a BTV risk model (De Koeijer
and Elbers 2007). A mixture of habitats comprising wetland areas, peat land areas, floodplains and
livestock farms was examined. In total 15 species were collected and identified with variations in
abundance for each habitat type. Culicoides impunctatus was most prevalent in wetland and peat
bogs/moors. Nine other species were found in the wetland areas while the diversity was much
lower in the peat land environments, with only C. obsoletus and C. pulicaris species complexes
being identified. Culicoides obsoletus were more prevalent in floodplains and livestock farms than
C. impunctatus. Culicoides scoticus Downes and Kettle, C. dewulfi Goetghebuer and C. chiopterus
Meigen were also present at the livestock farms.

In August 2006, BTV spread throughout northern Belgium, France, Germany, Luxembourg and the
Netherlands. This extremely worrying outbreak occurred further north than the occurrence of C.
imicola, the main vector of the southern Europe outbreaks. However, Culicoides dewulfi, which
breeds in the dung of cattle and horses, was found RT-PCR positive to the BTV-8 serotype that
had caused the northern European outbreak (Meiswinkel et al. 2007). This may present further
problems as it may be possible for this species to also transmit AHSV as the virus is closely related
to BTV (Meiswinkel et al. 2008). Similarly, C. obsoletus sensu stricto and C. scoticus (which comprise
two of the four species in the obsoletus complex), were found to be RT-PCR positive to BTV-8.
Culicoides pulicaris Linnaeus s.s. has previously been incriminated in the field transmission of BTV
(Caracappa et al. 2003), and C. impunctatus (another member of the Pulicaris complex) has been
implicated in the laboratory (Carpenter et al. 2006, Jennings and Mellor 1988). Despite this, the
information available suggests that C. obsoletus played a significant role in the European outbreak
(Meiswinkel et al. 2008). In June 2007 it was reported that an animal from a holding in Germany
had displayed evidence of BTV-8 infection in April. This represented the first indication that the
virus had successfully overwintered in the region (IAH 2008a). In September 2007 the first case
of BTV was identified in the UK (IAH 2008b). There are currently 136 premises affected by BTV
(DEFRA 2008). At the time of the European outbreak, temperatures soared to a record high and
this appeared to promote the replication and transmission of BTV by a number of Culicoides spp.
(Meiswinkel et al. 2008). It is also possible that other viruses such as AHSV, Akabane virus (AKAV),

220  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

epizootic haemorrhagic disease virus (EHDV) and equine enchephalosis virus (EEV) may also be
circulated in northern Europe under favourable climatic conditions (Meiswinkel et al. 2008). AHSV
is currently prevalent in sub-Saharan Africa and causes mortality rates that exceed 90% in horses
(Mellor 1994).

One of the best studied midge species is C. impunctatus, which is found throughout northern
Europe, Russia and northern China but is particularly prevalent on the west coast of Scotland
(Campbell and Pellham-Clinton 1960, Hill 1947). Although at least seventeen other species exist
in this area, C. impunctatus outnumbers the others considerably (Blackwell et al. 1992). Culicoides
impunctatus is among the most numerous of haematophagous flies in the UK where it can reach
huge population densities (Hendry and Godwin 1988). Although C. impunctatus does not vector
diseases in the UK presently, it causes significant losses to the economy, especially in terms of
reduced tourism and outdoor activities in Scotland, where it is responsible for more than 90% of the
biting attacks on humans (Blackwell et al. 1997). This is mainly due to the painful reactions caused
by their bites, which discourage tourists and disrupts outdoor industries including agriculture and
forestry. An estimated 20% loss in working hours has been reported in the forestry industry each
year due to midge bites alone (Hendry and Godwin 1988). Their intense numbers during such
attacks was recently highlighted by Mordue (Luntz) and Mordue (2003) who reported a rate of
around 40,000 landings per hour on the forearms of volunteers. Similarly high numbers are found
attacking people in coastal areas of Australia and the United States. In these countries several
species of biting midges breed in brackish salt marshes and mangrove swamps. The encroachment
of residential developments and recreational facilities into areas close to these breeding locations
and the manmade extensions of breeding habitats has resulted in biting midges becoming an
increasing problem.

Human pestilence by Culicoides biting midges is an acute problem along Australia’s tropical and
temperate shores and significant attack in the sublittoral and hinterland environs is confined to
the southern parts of the country. Of the three littoral species groups involved the C. immaculatus
group is of lesser importance than the other two. Both the C. ornatus group and the C. molestus
group include the most seriously offending species. The former group is usually associated
with mangrove habitats and the latter with the sandier reaches of open estuaries subject to
tidal influence. The C. ornatus group is very large. It includes 50 species overall. A number of
undescribed species in this group may ultimately be shown to contribute significantly towards the
pestilence problem of the tropical coastline. Culicoides ornatus Taylor, C. longior Hagan and Reye
and C. marmoratus (Skuse) are well known as serious pest species. Biting midges (e.g. Culicoides
ornatus) are also serious pests around the coast of the Northern Territory. Culicoides ornatus is
the dominant species in the Darwin area and numbers collected at certain locations reflect a
relatively intense pest problem (Reye and Lee 1963). Prior to 1963 little was known of the biting
midges on the Gold Coast of Australia as their larval habitat was isolated and they were of little
consequence to man. However, this changed with the development of residential areas in close
proximity to creek estuaries. Culicoides subimmaculatus Lee and Reye is the dominant pest species
in these areas. With the construction of residential canal estates on the Gold Coast a new breeding
habitat was created for the species C. molestus (Skuse). Here man interfered with the environment
and created a vast new breeding ground for what was once a minor species. Culicoides molestus
became the dominant midge species in the Gold Coast and interfered with the idyllic outdoor
lifestyle for which the Gold Coast is famous. As residential development spread further to the
north of the Gold Coast C. marmratus and C. henryi became the most prevalent midge species.
On the Gold Coast biting midges rapidly replaced mosquitoes as the most prevalent nuisance
or public health pest insect. The C. molestus group is smaller, with 11 recognised species overall.

Olfaction in vector-host interactions 221

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

Perhaps the supreme pest species of the Queensland coastline is included in this group. Ironically
it remains undescribed and has been regularly misidentified as C. subimmaculatus Lee and Reye
from which it differs in morphological detail and most likely in its biology.

The C. immaculatus group contains a single significant human pest species in Australia, namely C.
immaculatus Lee and Reye. Its preferred habitat is along open shores and its larval habitat remains
unknown.

In the United States the tremendous numbers of C. furens (Poey), C. barbosai Wirth and Blanton,
C. hollensis (Melander and Brues), C. melleus (Coquillett) and C. mississippiensis Hoffman make
living in coastal areas almost unbearable at certain times of the year. Abundance of these species
varies geographically along the extensive USA coastline, and spatially and temporally at different
geographic locations. Tourist business is greatly affected by biting midge abundance at these
locations. Kettle (1962) succinctly puts it, ‘One midge is an entomological curiosity, a thousand
can be hell’. However, the population level or midge density at which female midges must reach
before affecting people is dependent upon personal tolerance. Established residents of a midge-
infested area will tolerate high midge densities whilst new residents a low density is too much
to tolerate. The tourist business is greatly affected by midge (also called sand fly in this area)
problems along many Florida beaches where there are significant populations of C. furens, C.
barbosai, C. hollensis (Melander and Brues), C. melleus and C. mississippiensis. In some inland
localities, particularly in the ‘Highland’ and ‘Crystal Springs’ counties, C. floridensis Beck and C.
tissoti Wirth and Blanton may become a nuisance. Linley and Davies (1971) presented an excellent
review of the problem of midges versus tourism, a summary of the biology of the most important
species, and recommendations for control.

Culicoides furens makes life miserable in the Caribbean and on the eastern seaboard of America
from New York in the north to Sao Paulo in Brazil in the south. With the aid of the wind this species
may disperse 3 or 4 miles (5 to 7 km) (Linley and Davies 1971). Another species, C. barbosai,
had previously been confused with C. furens (Wirth and Blanton 1956). This was an important
discovery because, as the local Research Unit showed later, the breeding sites of C. furens and C.
barbosai were quite different. The existing measures which were directed against C. furens were
less effective against C. barbosai. Culicoides furens bred in the open and in areas shaded by the
white and black mangroves (Languncularia racemosa L. and Avicennia nitida Jacq., respectively)
while C. barbosai bred in wetter areas in the shade of the red mangrove (Rhizophora mangle L.).
Removal of the mangrove cover had no effect on the breeding of C. furens but a year later C.
barbosai had almost disappeared from the cleared area (Davies 1969).

Nuisance biting and the ability to transmit diseases, as described above, along with the recent
spread of Culicoides populations mean that the need for effective control is extremely important.
However, the olfactory processes and behaviours associated with host location and host preference
is still poorly understood and requires much research.

Midge/host interactions and influencual factors

Host location is essential for Culicoides midges to find a suitable blood meal. Midges are known
to detect visual, olfactory and gustatory stimuli, which aid in their location of a host. External
stimuli provided by vertebrate hosts induce a series of behaviours that ultimately lead to the
female finding a host and obtaining a blood meal. Host-derived chemical odours are considered
to be important as they can be detected by the insects at considerable distances and act as flight

222  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

activators or attractants (kairomones). As midges approach their host convective heat (Brown et
al. 1951), humidity, and visual cues such as shape, size, colour and contrast may also play a role
(Bernier et al. 1999, Bidlingmayer 1994, Khan 1977, Muir et al. 1992, Takken 1991, Chapter 6 in this
Volume). Bishop et al. (2008) used two-dimensional plastic cattle shapes and simulated stimuli to
investigate host location for C. brevitarsis. They subsequently proposed that initial attraction was
to visual stimuli, whilst host odours such as CO2 and 1-octen-3-ol were secondary and probably
occurred closer to the host. Landing is suggested to be a visual response to the interface between
the backline of the host and the background. Additionally, this work also revealed information on
the preferential biting sites of this vector. Sticky traps placed over the replica cattle shape showed
that preference is for the ridge line of the back, extending from the neck to the tail. Interestingly,
attraction of the mosquito Culex annulirostris Skuse was also investigated, and the visual element
was found to play no role at all. For C. impunctatus, landing responses on targets painted with black
and white vertical or horizontal stripes showed that females exhibited a clear preference for black
stripes irrespective of the presence of host odour or whether the target was moving (rotating) or
stationary. The target was shown to be unattractive to females at a distance. Striped targets were
however important at inducing a landing response with horizontal black stripes, perhaps related to
edge perception, being an effective landing stimulus (A Bhasin, unpublished observation).

Most adult Culicoides are \crepuscular, with peak activity around dawn and dusk, and to a lesser
extent through the night, although a few species bite during the day (Blackwell 1997, Mellor et
al. 2000). Culicoides impunctatus follows this crepuscular pattern (Blackwell et al. 1997) with the
endogenous rhythm of diel flight periodicity greatly enhanced in the presence of host odour
(CO2) (A Bhasin, unpublished observation). Although for Culicoides species the onset of activity
is usually triggered by falling light intensity (Mellor et al. 2000), meteorological variables other
than light intensity appear to trigger activity in C. impunctatus (Blackwell 1997). Light levels are
thought to accentuate an underlying rhythmically determined periodicity (Blackwell et al. 1997).
This theory corresponds with that of Marsh (1986) who predicted a bimodal circadian rhythm for
C. impunctatus.

Like other arthropods, Culicoides females employ semiochemicals for host-location (Mordue
(Luntz) 2003). Host-related kairomones for C. impunctatus include CO2, acetone, butanone and
L-(+)-lactic acid (Bhasin et al. 2000a,b, 2001). However, the full set of olfactory stimuli has not yet
been identified. To achieve this, we must examine the response of the insects to host odours and
then identify the components that are physiologically relevant within those stimuli. Recently,
Mands et al. (2004) investigated the response of C. impunctatus to animal extracts in a Y-tube
olfactometer. They showed that water buffalo extract was the most active (threshold 0.22 g/ml),
similar to deer extract, whereas other host extracts including pony, calf and sheep were more
than ten times less active. The effects of CO2 and 1-octen-3-ol on three species of Culicoides were
investigated in salt marshes, Georgia, USA (Kline et al. 1994). It was found that CO2 was an effective
attractant for all three species: C. furens, C. hollensis and C. melleus. However, each species did
possess a different response pattern. Contrastingly, 1-octen-3-ol was only an effective attractant
for C. furens, and was tested both singly and in combination with CO2, where it was synergistic.
Repellent effects were reported for C. hollensis and C. melleus. 1-Octen-3-ol has also been shown
to be attractive to C. impunctatus (Blackwell et al. 1996, Mands et al. 2004) and C. vexans, C. delta
C. pulicaris, C. lupicaris and C. albicans (Mands et al. 2004) in the field.

Bhasin et al. (2000b) investigated the behavioural responses of C. impunctatus to nine synthetic
compounds that had been identified as components of mammalian host odour: 1-octen-3-ol,
acetone, butanone, L-(+)-lactic acid, phenol, 3-n-propylphenol, 3-methylphenol, 4-methylphenol

Olfaction in vector-host interactions 223

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

and 4-ethylphenol. The non-phenolic compounds all induced an attractive dose-dependent

response over the range tested. As doses of butanone increased to supra-optimal levels arrestment
of responses was noted. Furthermore, for phenol and lactic acid, supra-optimal concentrations
actually appeared repellent. 1-Octen-3-ol has also been shown to arrest responses or even repel at
high concentrations (Blackwell et al. 1996). Culicoides nubeculosus (Meigen) behavioural responses
were tested for 1-octen-3-ol and acetone, and were shown to perform in a similar fashion to C.
impunctatus but with lower thresholds of response (Bhasin et al. 2000b). Trapping experiments in
southeastern Queensland, Australia, showed that the addition of CO2 to traps increased catches of
C. histrio and C. subimmaculatus significantly. Addition of 1-octen-3-ol and CO2 increased catches
of C. molestus only (Ritchie et al. 1994).

Pheromones

An interesting aspect of the host-location process is that it may also include the detection of
pheromones that act synergistically with host odours. Mating in C. nubeculosus occurs either when
a female flies into a swarm of males, with copulation taking place during descent to the ground,
or when a female is already feeding upon a host (Pomerantzev 1932, Downes 1950, Downes
1955). In the latter case, volatiles from the blood of a vertebrate host (produced either at the
wound itself, or from the excretion products resulting from concentration of the blood meal)
could stimulate sex pheromone production and grooming behaviour, contributing to dispersal
of the pheromone and attracting males whilst on the host. Virgin female C. nubeculosus perform a
greater amount of grooming behaviour in response to blood volatiles, than to controls (Figure 2).
It is hypothesised that this grooming behaviour is directly associated with the release of the sex
pheromone, n-heptadecane, and is initiated by the presence of blood volatiles (Figure 3; Mordue
(Luntz) and Mordue 2003). The behaviour could be a way of transferring the pheromone from
the abdomen to the wings, with wing flutter aiding pheromone dispersal. An effective trapping
system for this species, therefore, may involve the combination of host-derived olfactory stimuli
as well as the sex pheromone. Similarly, during feeding, C. impunctatus females are thought to
produce an aggregation or ‘recruitment’ pheromone (Blackwell et al. 1994a). This is believed to
work synergistically with host-derived kairomones such as 1-octen-3-ol and cause aggregation
of female midges on a vertebrate host. This is suggested to be a means of protection from host
defence by using the ‘safety in numbers’ strategy or a strategy that enables a midge to find a
rare but non-resource-limiting host (Blackwell et al. 1994a, 1996). A contact sex pheromone,
which stimulates copulation, has been identified for C. melleus and comprises 2-methyldocosane,
8-methyldocosane and 9-methyltricosane (Linley and Carlson 1978).

Indoors/outdoors

Culicoides midges were believed to be purely exophagic (i.e. feed on animals outdoors) and
exophilic (i.e. they will not enter inside buildings). Indeed, anecdotal evidence by Muller (1991)
showed that covered areas (yards, stables and sheds) significantly reduced numbers of C. brevitarsis
reaching host cattle but had little effect on mosquitoes. However, Bishop (2002) proposed the lack
of response by the midge species to be due to blocked vision, rather than an aversion to enclosed
spaces. Baldet et al. (2008) recently found Culicoides indoors albeit with a lower species diversity
(six species) than outdoors (nine species). Similar results have been obtained with C. impunctatus
that will readily enter through a window or door of a tent or a chalet containing a human volunteer
or with a light left on (JG Logan, unpublished observation).

224  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

A 1.2

Mean proportion of insects responding ± s.e.

Grooming behaviour
1.0 ***

0.8

0.6

0.4

0.2

0.0
Control 1 Control 2 Control 3 Sheep blood

Odour stimulus

B
Mean proportion of insects responding ± s.e.

0.16
Copulation attempts ***
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
Control 1 Control 2 Control 3 Sheep blood

Odour stimulus
Figure 2. Behavioural responses of 24-48 h old adult Culicoides nubeculosus females to sheep blood volatiles in
Alsevers solution. (A) Grooming behaviour, (B) Copulation attempts.
Control 1: empty flask; Control 2: distilled water; Control 3: Alsevers solution. Mean proportions ±s.e. are shown.
Asterisks indicate statistically significant differences from the controls (ANOVA *** P<0.001; n=100 replicates and
6 insects (3 males and 3 female) per replicate).

In northern France, adult parous female C. obsoletus/C. scoticus activity is greater indoors than
outdoors with a high proportion being freshly blood fed, but only during the autumn months.
Similarly, C. bolitinos Meiswinkel, C. dewulfi and C. chiopterus are thought to be strongly endophagic
whereas C. pulicaris and C. punctatus Meigen are more exophagic. However, it could be that the
housing of all farm animals may cause exophagic/exophilic species to enter buildings to feed
(Meiswinkel et al. 2000). The olfactory processes and other behaviours associated with Culicoides

Olfaction in vector-host interactions 225

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

80 *

70

60
% response (±s.e.)

50

40

30
n-Heptadecane
20 Sex pheromone extract
Control
10

0
Control 1×10e-9 1×10e-8 1×10e-7 5×10e-6 1×10e-6 1×10e-5

Stimulus concentration (g)

Figure 3. Behavioural responses of Culicoides nubeculosus females to a pheromone air entrainment extract and
six doses of n-heptadecane in hexane, in a Y-tube olfactometer.
Mean percentages of midges in the treated arm of the Y-tube ± s.e. are shown. Asterisks indicate statistically
significant differences from the control (ANOVA, * P<0.05; n=17, except pheromone extract where n=6).

midges entering buildings are not fully understood and may be quite different to those associated
with finding a host in an open environment and requires further investigation. This is particularly
important since some authorities recommend that livestock should be housed at night to reduce
Culicoides attack rates.

Climatic conditions

The interaction between Culicoides midges and their host can be heavily influenced by climatic
conditions. Midges display a preference for feeding in still, warm and humid conditions. Wind
speeds of 1-2 m/s are enough to inhibit flight and subsequent host-location activities (Lehane
2005). Mellor et al. (2000) reviewed information on the suppression of flight for various Culicoides
species, ranging from 3 m/s for C. imicola in Kenya (Walker 1977) to 2.2 m/s for C. brevitarsis in
Australia (Murray 1987). It has also been noted that limits are lower for mating swarm formation
by certain species. In addition to the negative effects of wind on activity, it can also affect the
abundance of Culicoides through dispersal of the adults (Baylis and Rawlings 1998). Similarly,
lower temperature limits for flight range from 10 °C for C. variipennis (Coquillett) (Nelson and
Bellamy 1971) to 18 °C for C. brevitarsis (Murray 1987).

In our own landing catch experiments on the west coast of Scotland, the effect of weather on
host-seeking C. impunctatus females can clearly be seen (Figure 4). Although this species is active
in light rain and heavy cloud cover, it is less so during heavy rain. Also, as one might expect,
fewer midges land on the arms of human volunteers with increasing wind speed. In contrast to

226  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

A 2.0 B 2.0
Rain Wind
1.8 1.8
Mean number of midges

Mean number of midges

1.6 1.6
1.4 1.4
1.2 1.2
1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
Dry Light rain Heavy rain Still Slight Moderate Strong
breeze wind wind
C 3.5 D 2.0
Temperature Cloud
1.8
3.0
Mean number of midges

Mean number of midges

1.6
2.5 1.4
2.0 1.2
1.0
1.5 0.8
1.0 0.6
0.4
0.5
0.2
0.0 0.0
9 10 11 12 13 14 15 16 17 18 19 20 Clear Light Cloudy Overcast
skies cloud
Temperature (˚C)
Figure 4. Mean number of Culicoides impunctatus midges landing on the forearms of volunteers under different
weather conditions on the west coast of Scotland during a 3 min repellency trial.

other species, their host-seeking behaviour is high above 9 °C and optimum at around 11 °C, but
decreases with increasing temperature and with very little activity above temperatures of 17 °C.
(JG Logan, unpublished observation)

Transmission of BTV and AHSV by C. variipennis sonorensis Wirth and Jones is favoured by high
temperatures (27-30 °C) as the reduction in longevity is compensated for by the shorter viral
extrinsic incubation period (EIP). Contrastingly, at lower temperatures where adult longevity was
extended, the EIP was disproportionately prolonged, meaning few midges could survive long
enough to transmit the virus (Wittmann 2000).

Intra and interspecific differential responses to host

Midges are differentially attracted to their vertebrate hosts. For example, odours collected from
different host species such as waterbuck, pony, red deer, cattle and sheep result in differential trap
catches of C. impunctatus females (Mands et al. 2004). Interestingly, there is anecdotal evidence
that individual human hosts also show variation in their attractiveness, and it is hypothesised that
more unattractive individuals produce ‘masking compounds’ that might indicate host unsuitability

Olfaction in vector-host interactions 227

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

to midges (Mordue (Luntz) and Mordue 2003). Much evidence exists that individual human beings
are differentially attractive to mosquitoes (Brady et al. 1997, Brouwer 1960, Burkot 1988, Khan
et al. 1966, Lindsay et al. 1993, Mayer and James 1969, Qiu et al. 2004, Schreck et al. 1990) and
there is much anecdotal evidence for biting midges. Recently, C. impunctatus has been shown to
respond differentially to odours collected from individual human volunteers (Logan et al. 2009).
Certain chemicals that were found in greater amounts in extracts from volunteers that caused low
attractiveness, elicit a repellent effect in laboratory assays and repellency trials in the field. It is
suggested that differences in the production of these natural human-derived compounds could
help to explain differential ‘attractiveness’ between different human hosts to C. impunctatus. A
mixture of two compounds in particular, 6-methyl-5-hepten-2-one and geranylacetone ((E)-6,10-
dimethylundeca-5,9-dien-2-one), showed significant repellency in the field and have the potential
to be developed as novel repellents. These types of semiochemicals may interfere with olfactory
mechanisms from a distance, preventing the insects from being attracted to their hosts, as has
also been demonstrated for Ae. aegypti L. (Logan et al. 2008). Similar studies have shown that
some individual cattle produce masking compounds such as the isoprenoid derivative, 6-methyl-
5-hepten-2-one, that make them less attractive to the fly, Hydrotaea irritans (Fallen), than others
within a herd (Birkett et al. 2004).

Most chemicals that are described as causing positive behavioural responses in Culicoides and
other haematophagous insects, such as CO2, 1-octen-3-ol, lactic acid, ammonia, acetone and fatty
acids are found ubiquitously in all humans and in many other vertebrates. Logan and Birkett
(2007), in their review of semiochemicals for biting fly control, proposed such chemicals may
comprise a basic ‘core’ suite of olfactory signals that, when present, convey information to an
insect that a vertebrate is nearby. Alteration of this ‘core’ by addition, or increase, of certain other
chemicals that can ‘mask’ activity of these attractants, or act as repellents (Pickett et al. 1998),
could be a way by which inappropriate or unsuitable vertebrate hosts are avoided by host-seeking
insects.

There is some evidence that the production of repellents in some vertebrates is genetically
determined (Thomas et al. 1987). If this is true for Culicoides hosts, it could be exploited to develop
breeding programmes to select for unattractive individuals. However, our knowledge of Culicoides
host preference to date is very poor for many Culicoides species and particularly for the vectors of
BTV and therefore, this requires further investigation.

Investigating host-derived semiochemicals for the control of midges

The exploitation of host-derived semiochemicals in control technologies first requires an in-

depth knowledge and understanding of ecologically relevant compounds that are likely to be
of practical use. It is therefore vital that we appreciate how target insects interact with such
compounds at the behavioural and physiological level by investigating the ecology of the insect-
host interaction. Various chemical ecology-based techniques can facilitate this. Semiochemicals
can be isolated from relevant odour sources by air entrainment, solvent washings, solid phase
micro extraction (SPME) and vacuum distillation. Behavioural and electrophysiological assays can
then be performed to locate and identify biologically and ecologically relevant semiochemicals.
Subsequent analytical chemistry techniques, such as gas chromatography (GC) and GC-mass
spectrometry (GC-MS) can identify the chemicals accurately. Fractionation has been used in the
past although this can be a laborious task with compounds being missed. A more efficient method
is to use gas chromatography combined with electroantennography (GC-EAG) or single-sensillum
recordings (GC-SSR), which is a unique and effective way of locating relevant components

228  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

within a complex array of chemicals in natural extracts. Such techniques are used to identify
semiochemicals that can be integrated into methods for the management and control of insect
pests as well as enhancing our knowledge and understanding of the behaviour and evolution of
mosquitoes and midges.

Olfaction and electrophysiology

Midges, like most haematophagous insects, have a well developed olfactory system and use their
antenna and maxillary palps to detect semiochemicals from vertebrates, giving them information
about the location and suitability of a host. A full understanding of how Culicoides detect olfactory
signals from their host to date is a poorly studied area, but it is vital to the development of
strategies aimed at controlling these insects (Grant and Kline 2003). Such an understanding of
the olfactory process can provide important information about the way Culicoides midges detect
and respond to host odour stimuli.

Insects detect odour molecules (kairomones) using olfactory receptors known as sensilla. The
olfactory sensilla of different arthropods show a high degree of similarity (Hallberg and Hannson
1999). This suggests that there may consequently be a degree of similarity in odour responses of
different arthropod species. Although this is the case for ubiquitous compounds such as CO2, it is
often not the case for other olfactory stimuli. The sensilla are porous hairs located on the antennae
and maxillary palps of the insect. Branched dendrites are found extending up into the hair and
these dendrites join to a sheathed neurone, which has axons leading to the central nervous system.
The sensilla are well designed for odour capture with the whole of the surface area involved in
the process. An odour molecule impacting upon a sensillum diffuses laterally through the surface
layers. These include an outer layer of polymerised lipid with a thin outer membrane composed
of electron transport material, and an electron dense layer. After diffusing through these layers,
the odour molecule reaches the receptor lymph where it is thought to be bound to an odourant
binding protein (OBP) resulting in the formation of a complex. This complex is then bound by a
receptor binding protein present in the membrane of the dendrite, which subsequently causes
a change in membrane potential. The altered membrane potential allows an ion flux across it
creating a receptor potential. The receptor potential may be large or small depending on the
magnitude of the original stimulus. From the dendrite, the receptor potential spreads to the cell
body and action potentials are produced. These action potentials are very rapid, lasting only a
few milliseconds, and fire at different frequencies proportional to the magnitude of the receptor
potential. Therefore, it can be seen how an impacting odour molecule is converted into electrical
signals that can be processed by the central nervous system of the insect (Zwiebel and Takken
2004, see also Chapters 2 and 4 in this Volume). By connecting microelectrodes to an antennal
preparation of an insect, the detection of a compound can be recorded as an electroantennogram
(EAG), which is the electrical potential from the insect’s antennae.

Culicoides antennae and maxillary palps have several olfactory sensilla types including trichodea
which have multiporous walls and a number of neurones which are typical of olfactory receptors.
They also possess basiconic and coeloconic sensilla which have double walls and unbranched
dendrites which could be olfactory, thermo-, or hygroreceptors (Blackwell et al. 1992, Chu-Wang
et al. 1975). Morphological characteristics of the olfactory apparatus, such as the size and shape of
the subapical labral sensilla (also known to contain olfactory receptors), the size and depth of the
palpal sensory pit, and the number and shape of heads of the palpal sensilla, can be used to classify
and identify Culicoides species (Blackwell 2004, Braverman and Hulley 1979, Jamnback 1965, Lane
1984). In some cases, it is believed that differences in the number of sensilla coeloconica and

Olfaction in vector-host interactions 229

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

palpal sensilla can even determine host preferences. For example, antennae having few antennal
flagellomeres with pits and less coleloconic sensilla might be associated with the need to detect
large amounts of odours and thus could be characteristic of mammalophilic species (Blackwell
2004, Jamnback 1965). Blackwell (2004) noted that some species found in the West Indies, such as
C. darlingtonae Wirth and Blantone, C. glabellus Wirth and Blantone, C. insinuates Ortiz, C. paraensis
(Goeldi), and C. pseudodiabolicus Fox are morphologically characteristic of mammalian feeders
when compared to other species such as C. heliconiae Fox and Hoffman and C. flavivenula Lima
whose host preferences were unknown. Although such morphological studies can be useful to
identify certain Culicoides species, they may also provide some information about host preference
and may even indicate the types of compounds they can detect. However, to fully understand the
olfactory process, a physiological investigation is usually required.

To date, electrophysiological studies on Culicoides midges are scarce. Most work has been
performed on the European species, C. nubeculosus, to look at the response of host odour extracts
and individual compounds. Whilst laboratory models are useful, where possible, working with
the insect of interest is preferable. It is also preferable to work with field caught insects as it
has been shown for some insects, such as the tsetse fly, that responses can diminish within
a few generations. The better studied (at the behavioural level) C. impunctatus has also been
investigated using electrophysiology, but with more logistical difficulties due to the fact that this
species is only available from field populations. Culicoides impunctatus females have a range of
olfactory chemosensilla on their antennae that respond to host-derived chemicals (Bhasin et al.
2000b, Blackwell et al. 1996, Pappenberger 1996, Sutcliffe 1994) and on the maxillary palps that
contain the receptors for CO2 (Bhasin et al. 2001). Culicoides impunctatus has antennal receptors
that respond to a variety of other host derived kairomones including 1-octen-3-ol and other plant-
derived compounds including methyl salicylate (Blackwell et al. 1996, 1997). Blackwell et al. (1995,
1996) were the first authors to describe the olfactory basis of host seeking in C. impunctatus and
since then, other authors have demonstrated EAG activity for other host-derived kairomones,
including CO2, 1-octen-3-ol, phenols and acetone (Bhasin et al. 2000a, Blackwell et al. 1997,
Grant and Kline 2003). Most recently, fifteen EAG-active compounds were found in human odour
extracts and some new compounds for Culicoides midges were identified including benzaldehyde,
6-methyl-5-hepten-2-one, octanal, 2-ethylhexanol, limonene, dihydromyrcenol, nonanal, linalool,
(E)-2-nonenal, menthol, naphthalene, decanal, geranylacetone and alpha-isomethylionone (Logan
et al. 2009).

It is also thought that measuring the electrophysiological response of the receptors on the
antennae can determine host preferences. For example host-odour extracts were assayed on
parous female C. impunctatus response in a Y-tube olfactometer and by EAG on C. nubeculosus
laboratory-reared parous females. Positive behavioural responses to host odours were dose-
dependent with a water buffalo extract being most active (threshold 0.22 g/ml), similar to deer,
whereas other host extracts were less active. Correspondingly, the EAG threshold was lowest for
water buffalo, 10-fold greater for deer, calf and pony, but not detected for sheep (Mands et al.
2004).

Recordings can also be made from single olfactory receptor neurones (ORNs) by inserting a
microelectrode into the sensilla to detect which receptor is responding to specific compounds. In
Culicoides, this has not been done for receptors on the antenna, but has for receptors on the maxillary
palps. The sensilla basiconica, which are located on the maxillary palps of Culicoides, contain an
ORN that responds to CO2 (Grant and Kline 2003) and shares physiological characteristics with
mosquitoes (Grant et al. 1995). The CO2 receptor in Culicoides midges is thought to be extremely

230  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

sensitive with the threshold of detection being significantly lower than that of mosquitoes. For
example, stimulation with 150 ppm CO2 elicits a response in C. furens, whereas in Ae. aegypti,
higher concentrations are required to elicit a response (Grant et al. 1995, Grant and O’Connell, this
Volume). This suggests that Culicoides ORNs to CO2 are likely to be more active at ambient levels.

A study involving EAG is useful to screen whole extracts or single compounds that are likely to have
ecological relevance to Culicoides midges. However, to fully understand how they find their hosts
by way of olfaction and to identify novel semiochemicals, natural host odour, which comprises a
complex mixture of relevant and irrelevant compounds, must be separated and tested. Screening
all host-derived compounds (often in excess of 400 volatile chemicals) by EAG and in behavioural
assays would be a time-consuming process and is highly impractical. A more efficient solution is to
analyse human odour extracts by coupled gas chromatography-electroantennography (GC-EAG).
This technique allows the location of EAG-active chemicals within complex extracts by taking
advantage of the high-resolution of the GC while simultaneously recording the response of an
antennal preparation of a female midge. This technique was first reported in 1969 (Moorhouse et al.
1969) and has been used for many insects (Logan et al. 2008, Pickett and Woodcock 1996). However,
due to the difficulties with rearing Culicoides midges in the laboratory, electrophysiological
techniques are complicated. Although portable EAG kits are available (see Syntech, http://www.
syntech.nl/), they can be difficult to use. When searching for novel host-derived semiochemicals,
it is essential to combine the electrophysiology techniques with that of gas chromatography. This
poses a massive logistical problem with such high-tech equipment. However, in our laboratory,
we are able to send C. impunctatus adult females (in a moist box), previously caught in Argyllshire
in Scotland using a sweep net to our laboratories at Rothamsted in Hertfordshire (a distance of
~400 miles and taking around 24-48 hours). The coupled GC-EAG has been advanced to possess
the capabilities of also detecting responses of single olfactory cells (GC-SCR) (Wadhams 1982), but
this has never been performed on Culicoides midges.

Generally, insect olfactory systems are not only sensitive to specific molecular structures, but
also to ubiquitous compounds. In the later case, sophisticated mechanisms such as coincidence
detection, whereby blends of volatiles in specific ratios from a host plant are detected by insects
within a complex background of volatiles from non-host plants, may be used. This might be
facilitated by paired or clustered ORNs that allow fine scale resolution of such complex signals.
Odours from a host are likely to occur in ‘pockets’ emanating from a host and therefore receptor
cells that respond simultaneously may indicate the presence of a host, whereas the same receptor
cells being stimulated with a delay would not (Bruce et al. 2005). Although this hypothesis has only
been described for phytophagous (plant-feeding) insects thus far, coincidence detection might
also occur in biting flies including Culicoides midges considering the complexity of odours within
their environments (Chapter 6, this Volume).

Laboratory-based experimentation

To understand how an insect responds to natural semiochemicals, laboratory bioassays are

commonly used as a first step. These behavioural experiments can be useful to determine the
effect of semiochemicals on midge behaviour whilst minimising other environmental variables
and provide a simple and rapid means of quantifying behavioural responses to many types of
stimuli. Some of the earliest behavioural studies with haematophagous insects were conducted
using host-seeking mosquitoes and used simple bioassays such as enclosed chambers or arenas
to examine responses to temperature and moisture stimuli (Bar Zeeve 1960, Howlett 1910,
Parker 1948, Thomson 1938). Such bioassays were only able to measure short range responses

Olfaction in vector-host interactions 231

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

(Bowen 1992). More recently olfactometers that allow and encourage flight responses to odour
stimuli have been developed, particularly for mosquitoes, and provide crucial information about
responses of insects to olfactory stimuli in many investigations. Dual choice olfactometers, such
as Y-tubes, have also been used with midges and mosquitoes and allow insects to respond to two
air streams containing different odour stimuli, often comprising one test air stream containing a
stimulus and one control (blank) air stream (Bhasin et al. 2000b, Blackwell et al. 1996, Brown et al.
1951, Dekker et al. 2001, Geier et al. 1999, Logan et al. 2008, 2009, Mands et al. 2004, Price et al.
1979, Roessler and Brown 1964, Smart and Brown 1957, Qiu et al. 2005).

For Culicoides midges, most laboratory-based olfactometer studies have been conducted with C.
impunctatus. Due to the problems associated with rearing this and many other Culicoides species in
the lab, these assays can be difficult. However, where possible, taking portable equipment to a field
laboratory can overcome this. For example, on the West coast of Scotland, a portable Perspex wind
tunnel and small glass Y-tube olfactometers (Blackwell et al. 1996, Bhasin et al. 2000b) have been
used to examine the response of wild populations of Culicoides to odour stimuli. The end of the
wind tunnel is placed outside a window of the field lab and the number of midges that fly upwind
from outside can be counted by the number of flashes on a killing grid. For other experiments,
insects can be caught by a sweep net and used in the olfactometer inside field laboratory (Logan
et al. 2009). In these smaller Y-tubes, the insects walk towards a light and enter one of two arms
which can contain an odour stimulus. Logan et al. (2009) showed that Culicoides midges respond
to certain chemicals in the Y-tube olfactometer including benzaldehyde, 6-methyl-5-hepten-2one,
octanal, 2-ethylhexanol, limonene, dihydromyrcenol, nonanal, linalool, (E)-2-nonenal, menthol,
naphthalene, decanal, geranylacetone and alpha-isomethylionone.

In this Y-tube assay light is used to stimulate upwind movement towards the arms of the Y-tube,
thus ‘forcing’ the midges to make a preferential movement to either arm. The Y-tube allows
for walking and some attempts at flight but not sustained flight and as such is not entirely
representative of a natural situation where C. impunctatus females would normally fly toward
their host by way of positive odour mediated anemotaxis. Insects may not respond the same to
odour stimuli when walking as opposed to flying (Kennedy 1977) and it might allow chemotactic
responses to odour stimuli to occur. Additionally, there can be steep odour gradients at the
interface between airstreams and this may allow chemotactic and chemokinetic responses
to occur which might result in insects remaining within one air stream (Kennedy 1977). Most
olfactometers may allow several behaviours to occur simultaneously such as orthokinesis,
klinokinesis, chemotaxis and anemotaxis and do not allow the discrimination between close and
long range responses. Therefore, caution must be taken when interpreting behavioural results.
Despite such complications, olfactometric experiments are a useful tool designed to give an
indication of the response of biting insects to odour stimuli under controlled conditions and are
thus useful to test host odours without the host suffering the discomfort of being bitten. This is
particularly important for Culicoides research where midge populations can be so immense that
field trials become ethically unacceptable. Encouragingly, previous studies have demonstrated
corresponding results between such laboratory trials and the field for biting midges. For example
C. impunctatus show positive responses to 1-octen-3-ol in a Y-tube olfactometer in the laboratory
and trap catches are enhanced by the addition of this compound to traps in the field (Bhasin et al.
2000b, Blackwell et al. 1996, Mands et al. 2004). With this in mind, it is important that studies are
also conducted in the field to test semiochemicals in a more realistic and natural setting.

232  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Control and field based experimentation

Insecticide applications against adult biting midges have had a limited effectiveness especially
when long-term relief from pestiferous populations is desired (Linley and Davies 1971).
Furthermore, reduction of midge populations via larval control by the application of insecticides
to the soil, or physical modification of wetland developmental sites, is not an option in many
locations such as coastal wetlands and acidic bogs because immature developmental sites are
extremely vast. Insecticide use in these areas is often forbidden due to environmental regulatory
issues. Repellents only provide minimal and temporary relief and are not considered a long-term
solution (Carpenter et al. 2005, Schreck and Kline 1981, Trigg 1996). Apart from altering one’s
life-style when midges are presented by staying indoors, repellents are presently the only means
of self-protection. These provide an inexpensive reasonably efficient protection method against
biting midges. DEET (N,N-Diethyl-m-toluamide), available since 1957, is still the most widely used
repellent against insects and arthropods of medical importance such as mosquitoes and ticks and
can show some efficacy against midges.

Ultimately, more effective control of Culicoides will need to rely on the development of integrated
pest management programs using multiple control strategies to relieve the biting pressure
of these pestiferous biting midges. New techniques are being investigated due in part to
environmental concerns and desires to reduce the use of chemical insecticides. One technique
that has received increased attention is the use of traps for control (Day and Sjogren 1994). The
feasibility of reducing local biting midge populations through the use of removal trapping has
recently been investigated by Kline and Lemire (1998; DL Kline unpublished data) in Florida, USA.
They demonstrated that a single line barrier of either attractant-baited traps (CDC-type, Model
512, John Hock Company, Gainesville FL, USA) or cloth targets impregnated with an insecticide
(lambdacyhalothrin) reduced mosquito and biting midge populations on a barrier island resort in
Southwest Florida. Traps and targets were baited with CO2 and 1-octen-3-ol. Subsequently, Day et
al. (2001) reported the reduction of C. furens populations in two delimited areas along the Atlantic
coast in South Florida and on an island in the Bahamas, using insecticide-treated fabric targets
baited with carbon dioxide and 1-octen-3-ol. This mass trapping programme against Culicoides
resulted in a significant reduction in the number of Culicoides throughout the island. Cilek et al.
(2003) and Cilek and Hallmon (2005) attempted to reduce Culicoides populations around individual
homes in north-western Florida using either one Mosquito Magnet® Pro (American Biophysics
Corporation, North Kingston RI, USA1) or one ABC PRO suction trap (Clarke Mosquito Products Inc.,
Roselle IL, USA) per backyard. The ABC PRO traps were baited with CO2 (500 ml/min) and a 4:1:8
1-octen-3-ol/phenol mixture of 1 octen-3-ol:3-n-propylphenol:4-methylphenol, released from 16
ml screw capped glass vials via a single wick (Dills TM pipe cleaner, United States Tobacco Sales
and Marketing, Greenwich, CT, USA). Overall release rates during the study averaged 5.39±0.54
mg/h. Population reduction, however, was very inconsistent (ranged from 2.3% to 70.6%) over
time and these researchers concluded that 1 trap per backyard was insufficient for consistent
Culicoides population reduction. They suggested that using more than one trap per backyard, or
a perimeter of traps, might be the key to achieving consistent population reduction. At certain
times during the study it was obvious that the attractant-baited traps were ‘overwhelmed’ by the
sheer volume of adult blood seeking midges moving into the targeted control zone. Another
possibility to achieve consistent population reduction was tested by Lloyd et al. (2008) by placing
traps in all the adjacent neighbouring yards, which in effect, created a barrier along the perimeter
of their backyards. Each trap was placed in the centre of a 0.30-0.40-ha area, which resulted in full

1 American Biophysics Corp went out of business in 2006

Olfaction in vector-host interactions 233

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

coverage of this residential area. In this study two commercially available attractant-baited traps
were compared (two models of Mosquito Magnet™ [MM], American Biophysics Corporation, North
Kingston RI, USA) on the island of Rye Key in Cedar Key, FL, USA to determine which model is more
successful at capturing Culicoides spp. Functionally, the two trap models are similar with some
difference in their bait-plume temperature ranges and CO2 output. The plume temperature for the
MM-Freedom® (Freedom) ranges from 33.34 °C to 36.67 °C compared to 33.34 °C to 40.56 °C for
the MM-Liberty Plus® (Liberty Plus) (K McKenzie, personal communication). Both traps utilised the
counterflow geometry design (Kline 2002) and their suction fans were powered thermoelectrically
by the combustion of propane, which also resulted in the production of the attractants CO2,
moisture and heat. The Freedom produces ca. 420 ml/min CO2, compared to ca. 550 ml/min CO2,
for the Liberty Plus. Both traps caught Culicoides spp. very well. Operationally, the Liberty Plus
was the more reliable trap. There was an apparent species preference for the traps. The overall
mean daily collection of C. furens in Freedom traps was significantly higher than the number
collected in the Liberty Plus traps. In contrast C. mississippiensis selected the MM-Liberty Plus
over the MM-Freedom. This could be due to temperature and CO2 differences between the traps.
Culicoides furens is attracted to heat (Kline and Lemire 1995), CO2 (Kline et al. 1990) and 1-octen-
3-ol (Takken and Kline 1989). It is possible that the Freedom and Liberty Plus are presenting the
attractants differently (Cooperband and Cardé 2006), which could affect a trap’s ability to attract
and capture host-seeking insects. Kline et al. (1994) reported that C. furens was the only species
of three major coastal Culicoides species that were attracted to a combination of 1-octen-3-ol
bait and CO2 in Sea Island, GA, USA. They reported that all Culicoides spp. were attracted to CO2.
The combination of heat, 1-octen-3-ol and CO2 may be why C. furens was captured in such high
numbers when compared with C. mississippiensis in this study. It is not certain what kairomones
are used by C. mississippiensis during host location. Minimal work has been done with this species.
What this means is that, for each target species, the trap type and attractant combination needs to
be determined. This will be key to adult biting midge reduction using removal trap technology. For
many Culicoides spp, especially those associated with livestock, this will require additional basic
studies (such as those discussed earlier) to address this complex and challenging problem. The
problem of protecting livestock on farms where a bite from a single viraemic midge is sufficient for
transmission of BTV or AHSV remains a challenge (Mordue (Luntz) and Mordue 2006). Trap number
and placement are also important points to consider. Just as important, adequate funding needs
to be made available for such evaluations in order that innovative alternative control technologies
can be transferred to the public in a timely manner.

Push-pull control strategy

Several ongoing research projects have evolved from these removal trapping studies to evaluating
the feasibility of developing and utilising push-pull systems (DL Kline, unpublished data). Push-
pull strategies involve the behavioural manipulation of pest species via the integration of stimuli
that act to make the protected resource hard to locate, unattractive or unsuitable to the pests
(push) while luring them toward an attractive source, such as an attractant-baited trap (pull) from
where the pests are subsequently removed. Push-pull strategies maximise efficacy of behaviour-
manipulating stimuli through the additive and synergistic effects of integrating their use (Cook
et al. 2007). Development and use of push-pull strategies requires a thorough knowledge of a
combination of behaviour-modifying stimuli that can be used to manipulate the distribution and
abundance of pest insects for pest management. It requires a clear scientific understanding of
the pest’s biology and the behavioural/chemical ecology of the interactions with its hosts and
its environment. The specific combinations of components differ in each strategy according to
the pest species to be controlled (its specificity, sensory abilities, and mobility) and the resource

234  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

targeted for protection, for example people or farm animals. The pests are repelled or deterred
away from this resource (push) by using stimuli that mask host apparency or are repellents or
deterrents. These stimuli may act over the long or short range and ultimately lead to the pest being
repelled or deterred from the resource or not even approaching it. Long range stimuli represent
the first line of defence, preventing or reducing infestation in the first place. Stimuli that act over
the short range, however, can be powerful tools in preventing specific pestiferous behaviours
(Cook et al. 2007). In pull components of push-pull strategies, attractive stimuli are used to divert
pests from the protected resource to a trap or target. The stimuli used to achieve this act mostly
over a long distance. However, short-range stimuli can be useful additions to arrest and retain the
pests in a predetermined place to facilitate the concentration of their populations and to prevent
them from returning to the protected resource. The stimuli can be delivered in a variety of ways.
Strategies primarily include visual and chemical cues or signals. Manipulation of colour, shape, or
size has been studied in a limited way in Australia. Chemical stimuli, in particular semiochemicals,
have the most versatility and potential for use in pest management. The principles of the push-pull
strategy are to maximise control efficacy, efficiency, sustainability, and output, while minimising
negative environmental effects.

Development of effective ‘push-pull’ systems for biting midge population management has
been hindered by a lack of resources to develop the necessary components. Limited research
at several locations worldwide has been dedicated to studies necessary to reveal the details of
biting midge biology, life cycles and chemical ecology that could be applied to their control by
behavioural means. Studies at these locations have mainly focused on nuisance pests of people.
Very little work has been conducted on the ’push’ component. Some studies are underway in
the U.S.A. to determine the ‘masking’ effect of human volatiles and compounds such as linalool
and dehydrolinalool (DL Kline, unpublished data) on host attraction of Culicoides to humans and
horses, and by exploiting compounds discovered during investigations on the natural differential
attractiveness within a host species (U. Bernier, personal communication). In the UK, repellents
such as 6-methyl-5-hepten-2-one and geranylacetone that were identified from odours of
‘unattractive’ volunteers are now under development as commercially available topical repellents.
Other studies have focused on the effectiveness of synthetic repellent products with either DEET
or Picaridin as the active ingredient. Several chemicals related to DEET have been tested against
midges and shown to be similar or more effective than DEET with respect to length of and degree
of protection. One such compound, Bayrepel®, tested by Carpenter et al. (2005) on the Scottish
biting midge proved to be effective for up to 8 hours post application. Plant essential oils such
as citronella and eucalyptus, PMD (p-menthane-3,8-diol), isolated from lemon eucalyptus oil of
Eucalyptus citriodora Hook. have been found to be effective against C. impunctatus (Trigg 1996).
Eucalyptus and other plant-derived repellents lose effectiveness more quickly with time than
synthetic compounds such as DEET. Blackwell et al. (2004) demonstrated the effectiveness of
neem against C. impunctatus and C. nubeculosus. Recently, spatial repellents, which repel from a
distance under field conditions, have been identified for C. impunctatus (JG Logan, unpublished
observation).

Most of the research efforts have been invested in determining what kinds of semiochemicals
and traps/targets can be used in the ‘pull’ component. A limited number of the well known host
kairomones that mediate behavioural responses in other haematophagous insects (Bursell et al.
1988, Hall et al. 1984, Hassanali et al. 1986) have been shown to be active in the field against
biting midges. Kairomones such as those found in breath, skin emanations and urine are the main
sensory cues used by haematophagous insects to find their hosts. The emphasis has been on
kairomones associated with humans and livestock including mammalian-associated volatiles such

Olfaction in vector-host interactions 235

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

as CO2, 1-octen-3-ol and acetone from the breath, and a mix of body odours (Bhasin et al. 2001,
Gibson and Torr 1999, Logan and Birkett 2007). Carbon dioxide (CO2) was first shown by Nelson
(1965) to be an attractant for members of the C. variipennis complex and, more recently, 1-octen-
3-ol has been used to trap C. furens (Kline et al. 1990, Takken and Kline 1989). Carbon dioxide has
a particularly important role in the attraction of haematophagous Diptera to their hosts (Gillies
1980). Stimulation by above-threshold levels of the gas, downwind of the host, activates them
to respond to other kairomones and elicits upwind orientation (Bursell 1984, Healy and Copland
1995, Schofield and Brady 1997). It is released periodically from a mammalian host at an expired
concentration of 4-5% above background (0.03%), and forms a filamentous plume, the exact
structure of which is determined by factors such as source size and wind speed (Brady et al. 1989,
Hargrove et al. 1995, Chapter 5 in this Volume). Response thresholds for activation by CO2 are
typically low: 0.006-0.01% for Stomoxys calcitrans (L.) (Schofield et al. 1997, Warnes and Finlayson
1985) and 0.01% for tsetse (Bursell 1984). Combination baits used have typically utilised CO2 and
1-octen-3-ol (Ritchie et al. 1994, Takken and Kline 1989). Field studies have shown that CO2 and
1-octen-3-ol act together to exert a synergistic effect on catches of C. furens (Kline et al. 1994).
Maximal upwind responses occur to CO2 plus 1-octen-3-ol or CO2 plus acetone but only at host
emission concentrations with higher concentrations causing arrestment (Bhasin et al. 2001). A
combination of 1-octen-3-ol and a pheromone produced by a host-seeking female C. impunctatus
has shown activity, both in the field and in the laboratory (Blackwell et al. 1996). For the Scottish
biting midge traps containing host kairomones such as CO2 with 1-octen-3-ol, acetone, a mixture
of alkyl phenols, or human sweat components have been tested in the field (Bhasin et al. 2001,
W Takken and AJ Mordue (Luntz), unpublished data) and promising combinations of attractant
cues are being defined. Our work and that of others is beginning to reveal that certain human
sweat components such as 3-methyl-2-hexanoic acid or 7-octenoic acid are either attractive to,
or ignored by, midges in their repertoire of host attractants (AJ Mordue (Luntz) and J Pickett,
unpublished data). Using knowledge of host specificity and preferences, the attractiveness of
synthetic host odour blends can be maximised.

One difficulty with the current emphasis on host-seeking kairomones is that some species like
C. impunctatus (Blackwell et al. 1992) and C. mississippiensis (Davis 1981) are autogenous species
(Blackwell et al. 2004), which produce a significantly larger egg-batch in their first, host-free, period
of oogenesis (Blackwell et al. 1992, Boorman and Goddard 1970). This, together with the paucity of
blood meals in most areas of the highlands, would suggest that strategies acting upon the parous,
biting proportion of the C. impunctatus population are unlikely to lead to more than short-term
changes in the biting rates experienced, without constant, high intensity trapping. Thus research
efforts should also be made to discover potential nectar sources. Both sexes utilise nectar and
sugar feeding for various life cycle activities. Nectar sources are of particular importance because
nectar provides energy for sustained flight in mating swarms (Downes 1969), increases adult
longevity (Linley 1966a), and may play a role in egg maturation (Linley 1966b). The females of
some species are not able to survive to the time of the first oviposition without nectar (Downes
1958). Volatiles from flowers may be utilised by adult biting midges to locate a nectar source. If
these volatiles can be isolated and identified they can be used as attractants with traps or targets.
Possible nectar sources have been determined for adult C. mississippiensis near Yankeetown, Levy
County, Florida. Large numbers of adults have been found in early spring on flowering shrubs such
as Ilex vomitoria Aiton. Other habitat related cues (apneumones) may be used by females to locate
suitable oviposition or resting sites. Carpenter et al. (2001) identified the odour of Juncus spp.
infusions together with the visual cue of upper layer photosynthetic sphagnum to be important
in the induction of oviposition of C. impunctatus.

236  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Numerous types of traps are available for use in the ‘pull’ component. Only the Midgeater
and Midg-IT were developed specifically for biting midges. Their design is very similar to the
Mosquito Magnet™ series of traps. A plethora of traps with various killing mechanisms including
electrocution, drowning, sticky boards, live catching and desiccation, originally developed for
mosquito control have been evaluated for their efficacy in capturing biting midges (DL Kline,
unpublished data). The most widely used trap is the MM-Liberty Plus. New advances in trap
technology continually need to be identified and subsequently evaluated. Various commercially
available lures (several manufacturers of 1-octen-3-ol lures, Lurex [lactic acid], Lurex3 [lactic acid
+ ammonium bicarbonate] and BG Mesh lure [lactic acid + ammonia + propionic acid] as well
as proprietary experimental lures, are currently being evaluated at Cedar Key, FL, USA. Visual
aspects of trap design on biting midge attraction have not been evaluated very much, but need
to be. Colour combinations, size and shape may be very important. Visual factors may enhance
the effectiveness of the olfactory stimuli. For example, blue and black traps, approximating
the size of a mammalian host, are used to control cattle tsetse fly (Glossina spp.). Crucial to the
development of efficient traps was the finding that black stimulates landing (Gibson and Torr
1999). Traps need to be designed with visual (A Bhasin, unpublished observation) and olfactory
‘pull’ stimuli that help them compete effectively with the hosts that they are trying to protect
and with other surrounding environmental stimuli. Trap design and positioning are important
and can be maximised by adopting a systematic approach in which the behaviour of the insect
is closely observed.

Culcioides and climate change

Changes in distribution and abundance of insects are likely to be amongst the most important
and immediate effects of climate change (Wittmann and Baylis 2000). In addition to increases in
temperature, changes in precipitation, wind patterns and climate variability are also predicted
to occur (Houghton 1997). The global atmospheric concentration of CO2 is rising markedly as
a result of human activity. The average level has increased from about 280 ppm immediately
before the industrial revolution to a daily average of 380 ppm in 2005 and is increasing at a rate
of 2 ppm per year (Guerenstein and Hildebrand 2008). Such an increase in CO2 is expected to
affect the biology of a number of living organisms including insects and therefore is stimulating
research on what those effects could be. Mondor et al. (2004) quoted an Intergovernmental Panel
on Climate Change (IPCC) report from 2001 which states CO2 and ozone (O3) have risen 31% and
35% respectively since the mid-1800s. As well as contributing to effects of temperature, and the
aforementioned effects on insects, these increasing levels of CO2 and O3 affect plants considerably.
Elevated concentrations of these gases alter the nutritional and defensive characteristics of plants,
and subsequently these effects can cascade through ecosystems and in turn impact upon higher
trophic levels, such as insect herbivores and their natural enemies (Percy et al. 2002). All these
changes may help create novel vector species by removing some of the barriers that render
many Culicoides species refractory to infection. It is worth noting that it is not just the adult
insects that may be affected. Borkent (2005) states that a variety of environmental factors have
been studied for a few pest species, and the results indicate that temperature, moisture levels,
chemical composition of the substrate, and population density all affect the rate of development
of immatures.

Climate and weather have dramatic effects on Culicoides populations and, consequently, the
epidemiologies of midge-borne viral diseases are similarly affected (Mellor et al. 2000). Due to
their small size, adult Culicoides are particularly susceptible to desiccation and even brief periods
at low humidity can reduce longevity (Murray 1991). However, changing meteorological factors

Olfaction in vector-host interactions 237

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

are not necessarily detrimental to the insects. While there are limits to vector competence, for
example BTV and AHSV are unable to develop in C. variipennis sonorensis at temperatures below
14-15 °C (Mullens 1995, Wellby et al. 1996, Wittmann 2000), as stated previously, suitable changes
in temperature may allow an increase in viable vectors.

Arndt (1995) investigated whether air pollutants react with pheromones, which would have
potential ecological and economic consequences. Specifically, he studied the effects of ozone
(O3) on the pheromone-producing insect, Drosophila melanogaster Meigen. It was found that
total pheromone containing extracts as well as commercially available pheromones lost their
biological activity after short-term ozone fumigation at environmentally realistic levels. At the
time the authors could find only one other reference relating to the subject. These findings, if true
for other insects, suggest that climate change may have great impact on intraspecific and feasibly
interspecific communication, particularly with the involvement of pheromones such as those for
C. impunctatus and C. nubeculosus.

More recently Mondor et al. (2004) showed the existence of divergent pheromone-mediated
behaviours in insects under conditions of global atmospheric change. Elevating CO2 and O3
levels, they investigated effects on a common aphid, Chaitophorus stevensis Sanborn. They showed
that the aphids principle means of defence, dispersal response to alarm pheromone, decreased
under elevated CO2, however, increased under elevated O3. Although the mechanism behind this
differential dispersal response remains unknown, it is clear that elevated greenhouse gases may
impact greatly upon intraspecific olfactory communication.

A further impact of climate change may be upon the voltinism of a species. Culicoides impunctatus,
for example, is bivoltine, with two distinct emergence periods (Blackwell 1997) and has recently
been shown to be orally susceptible to BTV in a laboratory study (Carpenter et al. 2006). However,
alterations to temperature may create favourable conditions that allow additional generations
within a given year. Tobin et al. (2008) investigated such effects using a phenology model of the
grape berry moth, Paralobesia viteana (Clemens). It was designed to incorporate temperature-
dependent development and diapause termination, as well as photoperiod-dependent diapause
induction. They found that increases in mean surface temperature of greater than 2 °C can cause
shifts in the ovipositional period resulting in dramatic effects on insect voltinism. Interestingly,
they mention anecdotal evidence from grape growers that already suggests a shift, with increases
in late summer populations. This will affect hosts and susceptibility and could even alter host
preferences with changing farm animals and farming practices.

Such effects may have substantial implications not just from an ecological perspective, but also
for how insect-derived compounds are used in pest management programmes. Different trapping
methods can suggest different activity patterns and some meteorological conditions can affect
the efficiency of trapping as well as the activity of the midge (Mellor et al. 2000).

Grant and Kline (2003) suggest that that a larger change in stimulus concentration is required to
elicit a larger change in impulse activity than Aedes spp. mosquitoes indicating that Culicoides
spp. perceive larger changes in carbon dioxide concentration than Aedes spp. It is possible that
CO2 at low concentrations is similar to that produced by photosynthesising plants and therefore,
in conjunction with other plant-derived volatiles, may provide information enabling location
of nectar for sugar feeding. Additionally, higher concentrations may serve as a host-seeking
attractant for blood feeding. Ambient levels of CO2 are higher now than ever and therefore the
CO2-sensitive neurons are possibly adapted to lower, pre-industrialisation levels of carbon dioxide

238  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

(see also Chapter 5, this Volume). This could simply reflect an earlier sensitivity, which has not
genetically adapted to the dramatic CO2 increase caused by accelerated human-based emissions
of carbon dioxide (Grant and Kline 2003). With a higher level of ambient CO2 this could be altering
the ability of Culicoides spp. midges to respond to CO2. Because ambient concentrations are much
higher, more subtle (or smaller) changes in CO2 need to be detected when host-seeking. This could
hamper the ability of Culicoides spp. midges to host-seek and mosquitoes may have adapted to be
able to do this. This demonstrates that specific information about Culicoides spp. midges is vital for
the development of odour baited traps using CO2 and other semiochemicals as the responses at
the physiological and behavioural level differs to that of other haematophagous insects. However,
this sensitivity to low concentrations of CO2 may not have any effect on behaviour.

Conclusions and future prospects

For many Culicoides species, most aspects of their ecology and behaviour, and in particular the
olfactory processes, remain undefined. This is in part due to the fact that Culicoides research is
hindered by the difficulties associated with rearing these insects. Although C. sonorensis and C.
nubeculosus have been successfully colonised, many species, including some other vectors of
BTV, have not. This means that most studies are carried out with the ‘laboratory models’ or have
to rely on field populations. For C. impunctatus, work has been carried out leading towards their
successful colonisation. Suitable conditions for adult blood feeding, oogenesis, oviposition and
adult survival (Carpenter et al. 2001, 2006), in the laboratory have been established but culture
of the larvae remains a challenge for this species. Additional colonies of Culicoides species would
allow much needed basic studies under controlled laboratory conditions to be carried out.

There are many other aspects of midge ecology that need to be elucidated. For example, climatic
conditions heavily influence host location behaviour. Although there is some knowledge for
species like C. impunctatus, other species are unstudied. This lack of knowledge, and particularly
in light of climate change, is likely to affect and potentially hamper future control efforts.
Additionally, oviposition site cues, those associated with the location of nectar, and pheromones
involved in mate recognition and/or aggregation behaviour could provide new semiochemical
tools. Investigating host preferences (at the intra and interspecific level) has already revealed
potential new semiochemicals that could be exploited to develop better attractants for traps and
new repellents for protection. Taking this further and understanding how unattractive odours are
produced by vertebrates at the molecular/genetic level could lead to new breeding programmes
whereby the trait for being unattractive is selected for. All of these areas of research warrant
further investigation.

The combination of attractants and repellents could provide a push-pull control strategy, but
further research is required on how this could be achieved for the many different Culicoides
species. Additionally, formulation and delivery systems studies are needed to provide optimum
technologies for specific species in specific geographical regions. Synthetic production of
semiochemicals, and their formulation as sprays or in slow-release dispensers, will ensure
standardisation and will contribute to the robustness of such strategies. GIS technologies could
also be incorporated into biting midge control programs. An improved understanding of the
spatial-scale effects of pest population dynamics and potential host interactions, coupled with
increased capability of spatially explicit computer models, will enable us to deploy more accurate
components of the various management strategies in terms of the quantities needed and their
spatial distribution. Ultimately, a better understanding of the behaviour of Culicoides species
midges, enabled by advances in analytical techniques, synthesis procedures, and formulation

Olfaction in vector-host interactions 239

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

science, may provide us with a larger and more effective armoury of semiochemicals and other
stimuli for future use.

References

Allingham PG (1991) Effect of temperature on late immature stages of Culicoides brevitarsis (Diptera: Ceratopogonidae).
J Med Entom 28: 876-881.
Arndt U (1995) Air pollutants and pheromones – a problem? Chemosphere 30(6): 1023-1031.
Baldet T, Delécolle JC, Cêtre-Sossah C, Mathieu B, Meiswinkel R and Gerbier G (2008) Indoor activity of Culicoides
associated with livestock in the bluetongue virus (BTV) affected region of northern France during autumn 2006.
Prev Vet Med 87(1-2): 84-97.
Bar-Zeeve (1960) The reaction of mosquitoes to moisture and high humidity. Entomologica Experimentalis et Applicata
3: 198-211.
Baylis M and Rawlings P (1998) Modelling the distribution and abundance of Culicoides imicola in Morocco and Iberia
using climatic data and satellite imagery. Arch Virol Suppl 14: 137-153.
Baylis M, Mellor, PS, Wittmann EJ and Rogers DJ (2001) Prediction of areas around the Mediterranean at risk of
bluetongue by modelling the distribution of its vector using satellite imaging. Vet Record 149: 639-643.
Bernier UR, Booth MM and Yost RA (1999) Analysis of human skin emanations by gas chromatography mass
spectrometry. 1. Thermal desorption of attractants for the yellow fever mosquito (Aedes aegypti) from handled
glass beads. Analytical Chemistry 71: 1-7.
Bhasin, A (1996) Host location cues of Culicoides spp. (Diptera: Ceratopogonidae). PhD thesis, Department of Zoology.
University of Aberdeen, Aberdeen, UK. 260 pp.
Bhasin A, Mordue AJ and Mordue W (2000a) Responses of the biting midge Culicoides impunctatus to acetone, CO2 and
1-octen-3-ol in a wind tunnel. Med Vet Ent 14: 300-307.
Bhasin A, Mordue AJ and Mordue W (2000b) Electrophysiological and behavioural identification of host kairomones as
olfactory cues for Culicoides impunctatus and C. nubeculosus. Phys Ent 25: 6-16.
Bhasin A, Mordue AJ and Mordue W (2001) Field studies on efficacy of host odour baits for the biting midge Culicoides
impunctatus in Scotland. Med Vet Ent 15: 147-156.
Bidlingmayer WL (1994) How mosquitos see traps - role of visual responses. J Am Mosq Contr Ass 10: 272-279.
Birkett MA, Agelopoulos N, Jensen V, Jespersen MB, Pickett JJA, Prijs J, Trapman HTGJ, Wadhams JLJ and Woodcock
CM (2004) The role of volatile semiochemicals in mediating host location and selection by nuisance and disease-
transmitting cattle flies. Med Vet Ent 18: 313-322.
Bishop AL (2002) The response of Culicoides brevitarsis to livestock undercover. Report to Biosecurity Australia, Canberra,
ACT, Australia.
Bishop AL, McKenzie HJ and Spohr LJ (2008) Attraction of Culicoides brevitarsis Kieffer (Diptera: Ceratopogonidae) and
Culex annulirostris Skuse (Diptera: Culicidae) to simulated visual and chemical stimuli from cattle. Aus J Entom 47:
121-127.
Bishop AL, McKenzie HJ, Barchia IM and Harris AM (1996) Effect of temperature regimes on the development, survival
and emergence of Culicoides brevitarsis Kieffer (Diptera: Ceratopogonidae) in bovine dung. Aus J Entom 35: 361-
368.
Blackwell A (1997) Diel flight periodicity of the biting midge Culicoides impunctatus and the effects of meteorological
conditions. Med Vet Ent 11: 361-367.
Blackwell A. (2004) A morphological investigation of Culicoides spp. biting midges (Diptera: Ceratopogonidae) from
the Caribbean. J Vector Ecol 29: 51-61.
Blackwell A, Brown M and Mordue W (1995) The use of an enhanced Elisa method for the identification of Culicoides
bloodmeals in host-preference studies. Med Vet Ent 9: 214-218.
Blackwell A, Dyer C, Mordue AJ, Wadhams LJ and Mordue W (1994a) Field and laboratory evidence for a volatile
pheromone produced by parous females of the Scottish biting midge, Culicoides impunctatus. Phys Entom 19:
251-257.

240  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Blackwell A, Dyer C, Mordue (Luntz) AJ, Wadhams LJ and Mordue W (1996) The role of 1-octen-3-ol as a host-odour
attractant for the biting midge, Culicoides impuctatus Goetghebuer and the interactions of 1-octen-3-ol with a
volatile pheremone produced by females. Phys Entom 21: 15-19.
Blackwell A, Evans KA, Strang RHC and Cole M (2004) Toward development of neem-based repellents against the
Scottish Highland biting midge Culicoides impunctatus. Med Vet Ent 18: 449-452.
Blackwell A, Mordue AJ, Young MR and Mordue W (1992) Bivoltinism, survival rates and reproductive characteristics of
the Scottish biting midge, Culicoides impunctatus (Diptera, Ceratopogonidae) in Scotland. Bull Ent Res 82: 299-306.
Blackwell A, Mordue (Luntz) AJ and Mordue W (1994b) Identification of bloodmeals of the Scottish biting midge,
Culicoides impunctatus, by indirect enzyme-linked-immunosorbent-assay (Elisa). Med Vet Ent 8: 20-24.
Blackwell A, Wadhams LJ, and Mordue W (1997) Electrophysiological and behavioural studies of the biting midge,
Culicoides impunctatus Goetghebuer (Diptera: Ceratopogonidae): Interactions between some plant-derived
repellent compounds and a host-odour attractant, 1-octen-3-ol. Phys Entom 22: 102-108.
Boorman J and Goddard PG (1970) Observations on the biology of Culicoides impunctatus (Diptera: Ceratopogonidae)
in southern England. Bull Ent Res 60: 189-198.
Borkent A (2004) The biting midges, the Ceratopogonodae (Diptera). In: Marquardt WC (ed) Biology of disease vectors.
2nd Edition. Elsevier Academic Press, Burlington, USA, pp 113-126.
Bowen MF (1992) Terpene-sensitive receptors in female Culex pipiens mosquitoes: electrophysiology and behavious. J
Insect Phys 38: 759-764.
Brady, J., Costantini, C., Sagnon, N., Gibson & Coluzzi, M. 1997. The role of body odours in the relative attractiveness of
different men to malarial vectors in Burkina Faso. Ann Trop Med Parasitol, 91, S121-S122.
Brady J, Gibson G and Packer MJ (1989) Odour movement, wind direction and the problem of host-finding by tsetse
flies. Phys Entom 8: 329-333.
Braverman Y and Hulley PE (1979)The relationship between the numbers and distribution of some antennal and palpal
sense organs and host preference in some Culicoides (Diptera: Ceratopogonidae) from southern Africa. J Med
Entomol 15: 419-424.
Brouwer R (1960) Variations in human body odour as a cause of individual differences of attraction for malaria
mosquitoes. Trop Geogr Med 12: 186-192.
Brown AWA, Sarkaria DS and Thompson RP (1951) Studies on the responses of the female Aedes mosquito. 1. The search
for attractant vapours. Bull Ent Res 42: 105-106.
Bruce, TJA, Wadhams, LJ and Woodcock, CM (2005) Insect host location: a volatile situation. Trends Plant Sci 10: 269-274.
Burkot TR (1988) Non-random host selection by anopheline mosquitoes. Parasitol Today 4: 156-162.
Bursell E (1984) Observations on the orientation of tsetse flies (Glossina pallidipes) to wind-borne odours. Phys Entom
9: 133-137.
Bursell E, Gough AJE, Beevor PS, Cork A, Hall DR and Vale GA (1988) Identification of components of cattle urine
attractive to tsetse flies, Glossina spp. (Diptera: Glossinidae). Bull Ent Res 78: 281-291.
Campbell AE and Pellham-Clinton EC (1960) A taxonomic review of the British species of Culicoides latrielle (Diptera:
Ceratopogonidae). Proc R Soc Lond [B]67: 181-302.
Caracappa S, Torina A, Guercio A, Vitale F, Calabro A, Purpari G, Ferrantelli V, Vitale M and Mellor PS (2003) Identification
of a novel bluetongue virus vector species of Culicoides in Sicily. Vet Rec 153: 71-74.
Carpenter S, Eyres K, McEndrick I, Smith L, Turner J, Mordue W and Mordue (Luntz) AJ (2005) Repellent efficiency of
BayRepel against Culicoides impunctatus (Diptera: Ceratopogonidae). Parasitol Res 95: 427-429.
Carpenter S, Lunt HL, Arav D, Venter GJ and Mellor PS (2006) Oral susceptibility to Bluetongue virus of Culicoides
(Diptera: Ceratopogonidae) from the United Kingdom. J Med Entomol 43: 73-78.
Carpenter S, Mordue (Luntz) AJ, and Mordue W (2001) Oviposition in Culicoides impunctatus under laboratory conditions.
Entomologia Experimentalis et Applicata 101: 123-129.
Chu-Wang I-W, Axtell RC and Kline DL (1995) Antennal and palpal sensilla of the sand fly Culicoides furens (Poey)
(Diptera: Ceratopogonidae). Int J Insect Morphol Embryol 4: 131-149.
Cilek JE and Hallmon CF (2005) The effectiveness of the Mosquito Magnet® trap for reducing Culicoides (Diptera:
Ceratopogonidae) populations in coastal residential backyards. J Am Mosq Contr Ass 21: 218-221.

Olfaction in vector-host interactions 241

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

Cilek JE, Kline DL and Hallmon CF (2003) Evaluation of a novel removal trap system to reduce Culicoides (Diptera:
Ceratopogonidae) populations in Florida backyards. J Vector Ecol 28: 23-30.
Conte A, Goffredo M, Ippoliti C, Meiswinkel R (2007) Influence of biotic and abitotic factors on the distribution and
abundance of Culicoides imicola and the obsoletus Complex in Italy. Vet Parasitol 150: 333-344.
Cook SM, Khan ZR and Pickett JA (2007) The use of push-pull strategies in integrated pest management. Ann Rev
Entomol 52: 375-400.
Cooperband MF and Carde RT (2006) Comparison of plume structures of carbon dioxide emitted from different
mosquito traps. Med Vet Ent 20: 1-10.
Davies JB (1969) Effect of felling mangroves on emergence of Culicoides spp. Jamaica Mosquito News 29: 566-571
Davis EL (1981) Laboratory studies on life cycle development and adult blood-feeding of Culicoides mississippiensis
Hoffman (Diptera: Ceratopogonidae), MSc Thsesis, University of Florida, Gainesville, FL, USA.
Day JF and Sjogren RD (1994) Vector control by removal trapping. Am J Trop Med Hyg50: 126-133.
Day JF, Duxbury CV, Glasscock S and Paganessi JE (2001) Removal trapping for the control of coastal Culicoides
populations. Techn Bull Florida Mosquito Contr Ass 3: 15-16.
DEFRA (2008) Bluetongue: latest situation. Department for Environment, Food and Rural Affairs (DEFRA), UK. Available
at: http://www.defra.gov.uk/foodfarm/farmanimal/diseases/atoz/bluetongue/latest/index.htm. Accessed July
2008.
Dekker T, Takken W and Cardé RT (2001) Structure of host-odour plumes influences catch of Anopheles gambiae s.s. and
Aedes aegypti in a dual-choice olfactometer. Phys Entom 26: 124-134.
De Koeijer AA and Elbers ARW (2007) Modelling of vector-borne diseases and transmission of bluetongue virus in
Northwest Europe. In: Takken W. and Knols BGJ (eds) Emerging pests and vector-borne diseases in Europe. Ecology
and control of vector-borne diseases, vol. 1. Wageningen Academic Publishers, Wageningen, the Netherlands, pp.
99-112.
Downes JA (1950) Habits and life cycle of Culicoides nubeculosus Mg. Nature 166: 510-511.
Downes JA (1955) Observations on the swarming flight and mating of Culicoides (Diptera: Ceratopogonidae). Transcr
Royal Entomol Soc London 106: 213-236.
Downes JA (1958) The feeding habits of biting flies and their significance in classification. Ann Rev Entomol 3: 249-266.
Downes JA (1969) The swarming and mating flight of Diptera. Ann Rev Entomol 14: 271-98.
Geier M and Boeckh J (1999) A new Y-tube olfactometer for mosquitoes to measure the attractiveness of host odours.
Entomologia Experimentalis et Applicata 92: 9-19.
Geier M, Bosch OJ and Boeckh J (1999) Ammonia as an attractive component of host odour for the yellow fever
mosquito, Aedes aegypti. Chem Senses 24: 647-653.
Gibson G and Torr SJ (1999) Visual and olfactory responses of haematophagous Diptera to host stimuli. Med Vet Ent.
13: 2-23.
Gillies MT (1980) The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review. Bull Ent Res
70: 525-532.
Grant AJ, Wigton BE, Aghajanian JG and O’Connell RJ (1995) Electrophysiological responses of receptor neurons in
mosquito maxillary palp sensilla to carbon dioxide. J Comp Physiol [A] 177: 389-396.
Grant AJ and Kline DL (2003) Electrophysiological responses from Culicoides (Diptera: Ceratopogonidae) to stimulation
with carbon dioxide. J Med Entomol 40: 284-292.
Guerenstein PG and Hildebrand JG (2008) Roles and effects of environmental carbon dioxide in insect life. Ann Rev
Entomol 53: 161-178.
Hall DR, Beevor PS, Cork A, Nesbitt BF and Vale GA (1984) 1-octen-3-ol, a potent olfactory stimulant and attractant for
tsetse isolated from cattle odours. Insect Sci Appl 5: 335-339.
Hallberg E and Hannson BS (1999) Arthropod sensilla: morphology and phylogenetic considerations. Microsc Res
Tech47: 429-440.
Hargrove JW, Holloway MTP, Vale GA, Gough AJE and Hall DR (1995) Catches of tsetse flies (Glossina spp, Diptera:
Glossinidae) from traps and target baited with large doses of natural and synthetic host odor. Bull Ent Res 85:
215-227.

242  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Hassanali A, McDowell PG, Owaga MLA and Saini PK (1986) Identification of tsetse attractants from excretory products
of a wild host animal, Syncerus caffer. Insect Sci Appl 7: 5-9.
Healy TP and Copland MJW (1995) Activation of Anopheles gambiae mosquitoes by carbon dioxide and human breath.
Med Vet Ent 9: 331-336.
Hendry G and Godwin G (1988) Biting midges in Scottish forestry: a costly irritation or a trivial nuisance? Scottish
Forestry 42: 113-119.
Hill MA (1947) The life-cycle and habits of Culicoides impunctatus Goetghebuer and Culicoides obsoletus Meigen,
together with some observations on the life-cycle of Culicoides odibilis Austen, Culicoides pallidicornis Kieffer,
Culicoides cubitalis Edwards and Culicoides chiopterus Meigen. Ann Trop Med Parasitol 41: 55-115.
Holbrook FR and Tabachnick WJ (1995) Culicoides variipennis (Diptera: Ceratopogonidae) complex in California. J Med
Entomol 32: 413-419.
Houghton J (1997) Global warming: the complete briefing. 2nd edition. Cambridge University Press, Camebridge, UK.
Jamnback H (1965) The Culicoides of New York state (Diptera: Ceratopogonidae). New York state Museum and Science
Service, Albany, USA. Bulletin 399, 154 pp.
Howlett FM (1910) The influence of temperature upon the biting mosquitos. Parasitology 3: 479-484.
IAH (2008a) Bluetongue virus-infected adult midges surviving winter may have been responsible for the re-emergence
of bluetongue this year in northern Europe. Institute of Animal Health, UK. Available at: http: //www.iah.ac.uk/
press_release/BT_UK_2007/BT_Statement2.html. Accessed July 2008.
IAH (2008b) Institute for Animal Health diagnoses bluetongue in Suffolk. Institute of Animal Health, UK. Available at:
http: //www.iah.ac.uk/press_release/BT_UK_2007/BT_Statement1.html. Accessed July 2008.
Jennings DM and Mellor PS (1988) The vector potential of british Culicoides species for bluetongue virus. Vet Microbiol
17: 1-10.
Kennedy JS (1977) Behaviourally discriminating assays of attractants and repellents. In: Shorey HH and McKelvey JJ
[eds], Chemical control of insect behaviour. John Wiley & Sons, New york, USA, pp 91-98.
Kettle DS (1965) Biting Ceratopogonids as vectors of human and animals diseases. Acta Tropica, Basel 22(4): 356-362.
Kettle DS (1984) Medical veterinary entomology. Croom Helm, London, UK.
Kettle DS and Lawson JWH (1952) The early stages of British biting midges Culicoides Latreille (Diptera: Ceratopogonidae)
and allied genera. Bull Ent Res 43: 421-467.
Khan AA (1977) Mosquito attractants and repellents. In: Shorey HH and McKelvey JJ (eds) Chemical control of insect
behaviour. John Wiley & Sons, New York. pp 305-325.
Khan AA, Maibach HI, Strauss WG and Fenley WR (1966) Quantitation of effect of several stimuli on the approach of
Aedes aegypti. J Econ Entomol 59: 690-694.
Kline DL (1994) Olfactory attractants for mosquito surveillance and control - 1-octen-3-ol. J Am Mosq Contr Ass 10:
280-287.
Kline DL (2002) Evaluation of various models of propane-powered mosquito traps. J Vector Ecol 27: 1-7.
Kline DL and Lemire GF (1995) Field evaluation of heat as an added attractant to traps baited with carbon dioxide and
octenol for Aedes taeniorhynchus. J Am Mosq Contr Ass 11: 454-456.
Kline DL and Lemire GF (1998) Evaluation of attractant-baited traps/targets for mosquito management on Key Island,
Florida, USA. J Vector Ecol 23: 171-185.
Kline DL, Hagan DV and Wood JR (1994) Culicoides responses to 1-octen-3-ol and carbon dioxide in salt marshes near
Sea Island, Georgia, U.S.A. Med Vet Ent 8: 25-30.
Kline DL, Takken W, Wood JR and Carlson DA (1990) Field studies on the potential of butanone, carbon dioxide, honey
extract, 1-octen-3-ol, L-lactic acid and phenols as attractants for mosquitoes. Med Vet Ent 4: 383-391.
Lane RP (1984) Morphometric symmetry in antennae of Culicoides (Diptera: Ceratopogonidae) J Nat History 18: 651-656
Lehane M (2005) The biology of blood-sucking in insects. 2nd Edition. Cambridge University Press, New York, USA. 336
pp.
Lindsay SW, Adiamah JH, Miller JE, Pleass RJ and Armstrong JRM (1993) Variation in attractiveness of human subjects
to malaria mosquitoes (Diptera: Culicidae) in The Gambiae. J Med Entomol 30: 308-373.

Olfaction in vector-host interactions 243

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

Linley JR (1966a) The ovarian cycle in Culicoides barbosai Wirth and Blanton and C. furens (Poey) (Diptera:
Ceratopogonidae). Bull Ent Res 57: 1-17.
Linley JR (1966b) Effects of supplementary carbohydrate feeding on fecundity and life-length in Leptoconops becquaerti
(Kieff ). Bull Ent Res 57: 19-22.
Linley JR and Carlson DA (1978) A contact mating pheromone in the biting midge, Culicoides melleus. J Insect Phys 24:
423-427.
Linley JR and Davies JB (1971) Sandflies and tourism in Florida and the Bahamas and Caribbean Area. J Econ Entomol
64: 264-278.
Linley JR, Hoch AL and Pinheiro FP (1983) Biting midges (Diptera: Ceratopogonidae) and human health. J Med Entomol
20: 347-364.
Lloyd AM, Kline DL, Hogsette JA, Kaufman PE and Allan SA (2008) Evaluation of two commercial traps for the collection
of Culicoides (Diptera: Ceratopogonidae). J Am Mosq Contr Ass 24(3): 253-262.
Logan JG and Birkett MA (2007) Semiochemicals for biting fly control: their identification and exploitation. Pest Manag
Sci 63: 647-657.
Logan JG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue (Luntz) AJ, Pickett JA (2008) Identification of
Human-Derived Volatile Chemicals that Interfere with Attraction of Aedes aegypti Mosquitoes. J Chem Ecol 34(3):
308-22.
Logan JG, Seal NJ, Cook JI, Stanczyk NM, Birkett MA, Clark SJ, Gezan SA, Wadhams LJ, Pickett JA and Mordue (Luntz) AJ
(2009) Identification of human-derived volatile chemicals that interfere with attraction of the Scottish biting midge
and their potential use as repellents. J Med Entomol 46: 208-219.
Mands V, Kline DL and Blackwell A (2004) Culicoides midge trap enhancement with animal odour baits in Scotland.
Med Vet Ent 18: 336-342.
Marsh PM (1986) Ecological studies on Culicoides impunctatus (Diptera: Ceratopogonidae) with reference to its control
in the highlands of Scotland. PhD thesis, University of Edinburgh, Edinburgh, UK.
Mayer MS and James JD (1969) Attraction of Aedes aegypti (L.): responses to human arms, carbon dioxide, and air
currents in a new type of olfactometer. Bull. Entomol Res 58: 629-643.
Meiswinkel R, Baldet T, De Deken R, Takken W, Delecolle J-C and Mellor PS (2008) The 2006 outbreak of bluetongue in
northern Europe – The entomological perspective. Prev Vet Med 87(1-2): 55-63.
Meiswinkel R, Baylis M and Labuschagne K (2000) Stabling and the protection of horses from Culicoides bolitinos
(Diptera: Ceratopogonidae), a recently identified vector of African horse sickness. Bull. Entomol Res 90: 509-515.
Meiswinkel R, Van Rijn P, Leijs P and Goffredo M (2007) Potential new Culicoides vector of bluetongue virus in northern
Europe. Vet Rec 161: 564-565.
Mellor, P.S. 1972. Studies on Onchocerca cervicalis (Railliet & Henry) and its development in Culicoides (Latreille). PhD
thesis, University of London, London, UK.
Mellor PS (1990) The replication of Bluetongue virus in Culicoides vectors. Curr Top Microbiol Immunol 162: 143-161.
Mellor PS (1993) African horse sickness: transmission and epidemiology. Vet Res 24: 199-212.
Mellor PS (1994) Epizootology and vectors of African horse sickness virus. Comparative Imm Microbiol Inf Dis 17: 287-
296.
Mellor PS and McCaig J (1974) Probable cause of sweet-itch in England. Vet Rec 95: 411-415.
Mellor PS, Boorman J and Baylis M (2000) Culicoides biting midges: Their role as arbovirus vectors. Ann Rev Entomol
45: 307-340.
Mondor EB, Trembaly ME, Awmack CS and Lindroth RL (2004) Divergent pheromone-mediated insect behaviour under
global atmospheric change. Global Change Biol 10: 1820-1824.
Mordue (Luntz) AJ (2003) Arthropod semiochemicals: mosquitoes midges and sealice. Biochem Soc Trans 31: 128-133.
Mordue (Luntz) AJ and Mordue W (2003) Biting midge chemical ecology. Biologist 50: 159-162.
Moorhouse JE, Yeadon R, Beevor PS and Nesbitt BF (1969) Methods for use in studies of insect communication. Nature
232: 1174-1175.
Muir LE, Thorne MJ and Kay BH (1992) Aedes aegypti (Diptera: Culicidae) vision - spectral sensitivity and other perceptual
parameters of the female eye. J Med Entomol 29: 278-281.

244  Olfaction in vector-host interactions

 10. Understanding and exploiting olfaction for the surveillance and control of Culicoides biting midges

Mullens BA (1995) Flight activity and response to carbon dioxide of Culicoides variipennis sonorensis (Diptera:
Ceratopogonidae) in Southern California. J Med Entomol 32(3): 310-315.
Muller M (1991) Report of the Bluetongue Research Management Committee, 11-13 Septmeber, Brisbane, Australia.
Muller MJ and Murray MD (1977) Blood-sucking flies feeding upon sheep in eastern Australia. Aus J Zool 25: 75-85.
Murray MD (1987) Akabane epizootics in New South Wales: evidence for long-distance dispersal of the biting midge
Culicoides brevitarsis. Autralian Veterinary Journal 64: 305-308.
Murray MD (1991) The seasonal abundance of female biting-midges, Culicoides brevitarsis (Diptera: Ceratopogonidae),
in coastal New South Wales. Aus J Zool 39: 333-342.
Nelson RL (1965) Carbon dioxide as an attractant for Culicoides. J Med Entomol 2: 56-57.
Nelson RL, Bellamy RE (1971) Patterns of flight activity of Culicoides variipennis (Coquillett) (Diptera: Ceratopogonidae).
J Med Entomol 8: 283-291.
Pappenberger B, Geier M, and Boeckh J (1996) Responses of antennal olfactory receptors in the yellow fever mosquito
Aedes aegypti to human body odours. In: Cardew G and Goode J (eds) Mosquito olfaction and olfactory mediated
mosquito-host interactions. Ciba Foundation Symposium 200, pp 254-266.
Parker AH (1948) Stimuli involved in the attraction of Aedes aegypti, L., to man. Bull Ent Res 39: 387-397.
Percy KE, Awmack CS, Lindroth RL, Kubiske ME, Kopper BJ, Isebrands JG, Pregitzer KS, Hendrey GR, Dickson RE, Zak DR,
Oksanen E, Sober J, Harrington R and Karnosky DF (2002) Altered performance of forest pests under atmospheres
enriched by CO2 and O3. Nature 420: 403-407.
Pickett JA and Woodcock CM (1996) The role of mosquito olfaction in oviposition site location and in the avoidance
of unsuitable hosts. In: Cardew G and Goode J (eds) Mosquito olfaction and olfactory mediated mosquito-host
interactions. Ciba Foundation Symposium 200, pp 109-123.
Pickett J, Wadhams LJ and Woodcock CM (1998) Mate and host location by insect model systems for exploiting olfactory
interactions. The Biochemist 8: 13.
Pomerantzev, BI (1932) Beitrage zur Morphologie und Anatomie der Genitalien von Culicoides (Diptera: Nematocera).
Mag Parasitol 3: 183-214.
Price GD, Smith N and Carlson DA (1979) The attraction of female mosquitoes (Anopheles quadrimaculatus Say) to stored
human emmanations in conjunction with adjusted levels of relative humidity, temperature and carbon dioxide. J
Chem Ecol 5: 383-395.
Purse BV, Mellor PS, Rogers DJ, Samuel AR, Mertens PPC and Baylis M (2005) Climate change and the recent emergence
of bluetongue in Europe. Nat Rev Microbiol 3: 171-181.
Qiu YT, Smallegange RC, Hoppe S, Van Loon JJ, Bakker EJ and Takken W (2004) Behavioural and electrophysiological
responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera: Culicidae) to human skin
emanations. J Med Vet Entomol 18: 429-438.
Reye EJ and Lee DJ (1963) The influence of the tide cycle on certain species of Culicoides (Diptera, Ceratopogonidae).
Proce Linnean Soc New South Wales 87: 377-387.
Ritchie SA, Van Essen PA, Kemme JA, Kay BH and Allaway D (1994) Response of biting midges (Diptera: Ceratopogonidae)
to carbon dioxide, 1-octen-3-ol and light in southeastern Queensland. Aus J Med Entomol 31: 645-648.
Roessler P and Brown AWA (1964) Studies on the responses of the female Aedes mosquito. X. - Comparison of oestrogens
and amino acids as attractants. Bull Ent Res 55: 395-403.
Schofield S and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol on the flight responses of the stable
fly, Stomoxys calcitrans, in a wind tunnel. Phys Entom 22: 380-386.
Schofield S, Witty C and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol on the activity of the stable
fly, Stomoxys calcitrans. Phys Entom 22(3): 256-260.
Schreck CE and Kline DL (1981) Repellency determination of four commercial products against six species of
Ceratopogonid Culicoides. Mosquito News 41: 7-10.
Schreck, C.E., Kline, D.L. & Carlson, D.A. 1990. Mosquito attraction to substances from the skin of different humans. J
Am Mosq Contr Ass 6: 406-410.

Olfaction in vector-host interactions 245

James G. Logan, James I. Cook, A. Jennifer Mordue (Luntz) and Dan L. Kline

Sellers RF and Maarouf AR (1991) Possible introduction of epizootic hemorrhagic-disease of deer virus (Serotype-2)
and bluetongue virus (Serotype-11) into British-Columbia in 1987 and 1988 by infected Culicoides carried on the
wind. Can J Vet Res 55: 367-370.
Smart MR and Brown AWA (1957) Studies on the responses of the female Aedes mosquito. Part VII. The effect of skin
temperature, hue and moisture on the attractiveness of the human hand. Bull Ent Res 47: 89-101.
Sutcliffe JF (1994) Sensory bases of attractancy - morphology of mosquito olfactory sensilla - a review. J Am Mosq
Contr Ass 10: 309-315.
Takken W (1991) The role of olfaction in host-seeking of mosquitos - a review. Insect Sci Appl 12: 287-295.
Takken W and Kline DL (1989) Carbon dioxide and 1-octen-ol as mosquito attractants. J Am Mosq Contr Ass 5: 311-316.
Takken W, Verhulst N, Schotte E, Jacobs F, Jongema Y and Van Lammeran R (2008) The phenology and population
dynamics of Culicoides spp. in different ecosystems in the Netherlands. Prev Vet Med 87: 41-54.
Tatem AJ, Baylis M, Mellor PS, Purse BV, Capela R, Pena I and Rogers DJ (2003) Prediction of bluetongue vector
distribution in Europe and north Africa using satellite imagery. Vet Microbiol 97: 13-29.
Thomson RCM (1938) The reactions of mosquitoes to temperature and humidity. Bull Ent Res 29: 125-140.
Thomas G, Prijs HJ and Trapman JJ (1987) Factors contributing to differential risk between heifers in contracting summer
mastitis. In: Thomas G, Overvecht U and Nansen P. Summer Mastitis. Martinus Nijhof, Dordrecht, the Netehrlenads,
pp 30-35.
Tobin PC, Nagarkatti S, Loeb G and Saunders MC (2008) Historical and projected interactions between climate change
and insect voltinism in a multivoltine species. Global Change Biol 14: 951-957.
Townley P, Baker KP and Quinn PJ (1984) Preferential landing and engorging sites of the Culicoides species landing on
a horse in Ireland. Equine Vet J 16: 117-120.
Trigg JK (1996) Evaluation of a eucalyptus-based repellent against Culicoides impunctatus (Diptera: Ceratopogonidae)
in Scotland. J Am Mosq Contr Ass 12: 329-330.
Wadhams LJ (1982) Coupled gas-chromatography - single cell recording - a new technique for use in the analysis of
insect pheromones. Zeitschrift Für Naturforschung 37c: 947-952.
Walker AR (1977) Seaeonal fluctuations of Culicoides species (Diptera: Ceratopogonidae) in Kenya. Bull Ent Res 67:
217-233.
Warnes ML and Finlayson LH (1985) Responses of the stable fly, Stomoxys calcitrans (L.) (Diptera: Muscidae), to carbon
dioxide and host odours. II. Orientation. Bull Ent Res 75: 717-727.
Wellby MP, Baylis M, Rawlings P and Mellor PS (1996) Effect of temperature on virogenesis of African horse sickness virus
in Culicoides variipennis sonorensis (Diptera: Ceratopogonidae) and its significance in relation to the epidemiology
of the disease. Bull Ent Res 86: 715-720.
Wirth WW and Blanton FS (1956) A new species of salt marsh sandfly from Florida, the Bahamas, Panama and Ecuador,
its distribution and taxonomic differentiation from Culicoides furens (Poey) (Diptera, Heleidae). Florida Entomol
39: 157-162.
Wittmann EJ (2000) Temperature and the transmission of arboviruses by Culicoides biting midges. PhD Thesis. University
of Bristol, Bristol, UK.
Wittmann EJ and Baylis M (2000) Climate change: effects on Culicoides-transmitted viruses and implications for the UK.
Vet J 160: 107-117.
Wittmann EJ, Mellor PS and Baylis M (2001) Using climate data to map the potential distribution of Culicoides imicola
(Diptera: Ceratopogonidae) in Europe. OIE Sci Tech Rev 20: 731-740.
Zwiebel LJ and Takken W (2004) Olfactory regulation of mosquito-host interactions. Insect Biochem Mol Biol 34: 645-
652.

246  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

11. Black fly interactions with their hosts

James F. Sutcliffe

Abstract

Haematophagous black flies (Diptera: Simuliidae) seek mammal and bird hosts as blood sources
and are responsible for significant human and animal hardship and economic loss due to
annoyance, effects of biting and disease transmission. Simuliids locate and choose their hosts
by orienting to host-originating chemical, visual and temperature stimuli. Little is known about
the nature of the stimuli involved as chemical mediators though carbon dioxide is influential and
certain other breath and body odours also clearly play roles. Effects of visual stimuli are limited to
the effective visual range of the host to the fly which is determined by host size and environment.
Vision appears to take precedence over olfaction when the host-seeking fly makes visual contact
with the host. Landing and biting site choices appear to depend on a combination of visual and
skin-associated chemical and thermal stimuli. Black fly responses to potential host stimuli seem
dependent on the context in which the stimuli are received. Some stimuli (e.g. warmth) induce
landing or biting for flies in the host seeking mode (or context) but induce escape reactions
for non-host-seeking flies. Host choice (or rejection) may be based on the ability of these host
stimuli, encountered in an appropriate sequence, to sustain the host seeking mode up to its
consummation (blood feeding). Research on black fly host seeking has been slowed by the relative
difficulty of doing meaningful behavioural experiments in the lab with this group, by the dearth of
lab colonies and the difficulties and costs of maintaining them and by the lack of adequate field
tools and experimental approaches.

Keywords: attractants, blood feeding, host-seeking behaviour, sensory ecology, simuliids, visual
cues

Introduction

General

Black flies (Diptera: Simuliidae) are a geographically widespread insect family found in almost
all areas, including many oceanic islands, where there is running water. Crosskey (1990) includes
1,554 species1 in his world list and notes that many more species will eventually be added as
species complexes are resolved taxonomically.

Immature stages of simuliids occur in running water ranging, depending on species, from small
streams such as those found in the highlands of Central America or on the Precambrian shield of
Canada and Europe, to large rivers such as the Niger in West Africa or the North Saskatchewan
that flows through the parklands of Alberta and Saskatchewan, Canada.

For the most part, black flies are pestiferous in numbers in rural and wilderness settings. Various
factors, many of them not well-understood, appear to exclude simuliids from urban locations.
In eastern Canada, black flies are a pest of humans in the spring and early summer months (late
May, June, early July) while in western Canada the large river species (mainly Simulium arcticum

1 According to Adler et al. (2003), Crosskey recognizes 1,774 valid black fly species in his 2002 update published by the

British Museum of Natural History.

Olfaction in vector-host interactions 247

James F. Sutcliffe

s.l. and S. luggeri Nicholson and Mickel) have multiple generations and are potential pests all
summer long.

In their comprehensive treatment of the systematics of black flies, Adler et al. (2004) provide
brief sketches of several aspects of general black fly biology (including behaviour, ecology,
physiology) for North American species. Crosskey (1990) covers similar topics more extensively
with a worldwide perspective.

Adler et al. (2003) provide an exhaustive revision of the taxonomic status of all North American
black fly species which, because many are boreal, includes many Palearctic species as well. For
the purposes of this review I use species names as reported in the original publications; however,
the reader can determine the probable valid names for the species discussed by referring to the
‘Taxonomic Accounts’ section Adler et al.’s invaluable book.

Females of most simuliid species are blood feeders on mammals or birds though females of several
species are autogenous in their first gonotrophic cycle and a few do not blood feed at any time
having adapted to certain extreme conditions by adopting wholly autogenous reproductive
strategies (Anderson 1986). While a few historical references have been made to black flies feeding
on other insects, fish, frogs, etc., these are considered either mistaken or spurious (Crosskey 1990).
The most credible report of black flies interacting with a non-mammal or non-bird vertebrate
is by Smith (1969) who observed S. venustum Say swarming around a snapping turtle (Chelydra
serpentina (Linnaeus)) in Algonquin Park, Ontario although none appeared to feed on it.

Black flies host seek only during the daylight hours (Crosskey 1990) suggesting that vision may be
an important sensory modality in their host-seeking behaviour. While black flies are potentially
active throughout the daylight hours, many species exhibit various peaks of activity in the early
to mid-morning and during the late afternoon and evening. This is presumably to take advantage
of the normally calmer conditions at these times and to avoid the desiccating heat of the mid-day
and early afternoon hours.

Medical and veterinary significance of black fly blood feeding

Swarms of females can sometimes be massive around hosts and extremely annoying and
stressful. Black fly bites can be painful and have, depending in the number of bites received
and the sensitivity of the individual, consequences ranging from mild itching and irritation to
severe illness and anaphylactic shock. Livestock in the parklands of western Canada have been
killed by attacks of large numbers of Simulium arcticum (s.l.). Claims that this has been due to
exsanguination are doubtless incorrect. Some combination airway blockage by fluid build-up
around the tracheae in response to salivary antigens, allergic reactions to innumerable bites, and
stress due to harassment and heat, especially in animals not previously exposed to black flies, is
more credible as the cause of death (Beck 1980). Numerous examples of the pest significance of
black flies are documented in many parts of the world (see Crosskey 1990 for several accounts).

Their blood feeding habit also makes black flies of significance as vectors of several important
parasites and parasitic diseases of humans and animals. The most important parasite transmitted
by black flies is the filarial nematode, Onchocerca volvulus (Leuckart), responsible for the condition
called (among many other names) River Blindness. River Blindness is transmitted by various
simuliid species in vast tracts of West and Central Africa and more focally in parts of Central and

248  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

South America. Thylefors and Alleman (2006) provide a brief history and up-to-date summary of
onchocerciasis control efforts in Africa and in the New World.

Black fly species in many parts of the world are also vectors of various other microfilarial parasites
of animals including a number of Onchocerca species of livestock (Lok et al. 2000) and Dirofilaria
ursi Yamaguti (Addison 1980) of black bears (Ursus americanus Pallas) in North America.

A simuliid-vectored disease of great significance in birds is an ‘avian malaria’ caused by

geographically widespread members of the sporozoan genus Leucocytozoon. Leucocytozoon
simondi Mathis and Leger is a common cause of death especially for young birds on the nesting
grounds in spring and L. smithi Laveran and Lucet affects geese, turkeys and domestic fowl (Cupp
1986).

The demonstration that Vesicular Stomatitis Virus (VSV) occurs at high levels in members of some
black fly species (Francy et al. 1988) and that S. vittatum Zetterstedt is capable of transmitting
VSV biologically in the lab (Cupp et al. 1992) has further established the vector significance of
this family. VSV does not establish high viremias in its vertebrate hosts (e.g. cattle) making its
transmission by black flies (and other blood feeders) something of a mystery that appears to have
been solved by Mead et al. (2000) who showed that VSV transmission can occur between black
flies as they feed on the same host.

Overview of black fly host preferences

Black flies are conventionally divided into ornithophilic and mammalophilic groups in terms of
broad host preferences, though the gap between these two types of hosts is spanned by some
species. Black flies are generally considered to be fairly host specific within these host ranges. At
one extreme, S. euryadminiculum Davies apparently restricts itself to feeding from the common
loon (Adler et al. 2004) and is the poster species for the supposed high host-specificity of black
flies. Mammalophilic species also appear in some cases to draw some clear distinctions between
potential hosts. For instance, S. arcticum Malloch in Alberta forms massive swarms around cattle
through which humans can walk and within which humans can work and rarely, if ever, get bitten
(Shemanchuk 1986). Many other mammalophilic species appear to have broader host preferences
(see Crosskey 1990 for a more detailed review of known black fly host preferences and Adler et
al. 2004 for more recent information for North American species). It would be of great interest to
know what role black fly responses to host-originating stimuli play in the evolution and expression
of these host preferences.

Aims of this review

Various authors have reviewed aspects of host orientation and responses to blood hosts by biting
flies. These reviews have different scopes ranging from those that look at the roles of specific types
of stimuli (e.g. olfaction in mosquitoes – Takken 1991; visual stimuli in biting flies – Allan et al.
1987) to those that focus on particular parts of the process (e.g. ‘distance orientation’ in biting flies
– Sutcliffe 1987) to others that summarise the broad scope of hosting seeking behavioural control
for a number of biting fly taxa (e.g. Gibson and Torr 1999). In the first part of this review, I will
deal mainly with the part of the host-black fly interaction that corresponds to active orientation
of the insect to the host up to and including landing on the host and seeking out a biting site.
This represents a part of the overall host location-blood feeding process in black flies that was
interpreted by Sutcliffe (1986) as beginning with activation, when the blood feeder first detects

Olfaction in vector-host interactions 249

James F. Sutcliffe

the presence of the host, and as culminating in gorging followed by the departure of the blood
fed insect out of the reach of the host’s defensive reactions.

In the second part of this review I will discuss related topics including what the details of black
fly responses to host stimuli may say about the evolution of host preferences, the nature of host
seeking-blood feeding as a behavioural package and, finally, some of the challenges inherent in
research on host seeking in simuliids.

Black fly – host interactions (outside of active orientation phase)

In the following paragraphs I provide a brief overview of the components of the black fly – host
interaction outside of active orientation for purposes of setting active orientation (the main topic
of this review) in its larger behaviourally meaningful context.

Activation

This larger context includes the activation step which precedes host seeking proper. In activation,
the insect is stimulated out of some ground or resting state into an actively host-seeking mode by
contact with host-originating stimuli (see section on ‘Host seeking-blood feeding as a behavioural
mode’ for further discussion of this concept). In other biting flies, CO2 and other host odours are
widely considered to be activating (see reviews by Gibson and Torr 1999, Takken 1991). This is
likely the case for most simuliids too since many species are caught in larger numbers at traps
and targets baited with CO2 only than at traps and targets without it (Sutcliffe 1986, 1987). This
makes a logical case for CO2 as an activating stimulus in these instances since for flies to be
caught at a trap or target they first need to have been activated. Nonetheless, there are clear
exceptions to the CO2 as activator rule (if it is a rule) since certain black fly species have been
caught in situations where only non-CO2 host stimuli were present. For example, in Algonquin
Park, Ontario S. euryadminiculum swarmed near sources of loon uropygial odour (e.g. filter papers
bearing residues from ether extracts of uropygial gland – Lowther and Wood 1964) whether CO2
was co-released or not and the inclusion of CO2 did not greatly increase numbers swarming (Fallis
and Smith 1964; JF Sutcliffe personal observations). In this case, it must be concluded that it
was some of the odours from uropygial gland secretions that were the activating principal(s).
Examples where CO2 appears not to be necessary to catch black flies (though it may augment
catches achieved with other odours) at traps or targets are also known. In field experiments in
Cameroon, Thompson (1976b) found that traps baited with odours from human sweat would
catch large numbers of S. damnosum Theobald s.l. even in the absence of CO2. By the same logic,
these substances must to possess odour components that have activating properties for this
species. Uropygial secretions of ducks also appear capable of activating some duck-biting species
such as S. rugglesi Nicholson and Mickel and S. anatinum Wood (Fallis and Smith 1964). While
odours are often cited as activating stimuli for host seeking in black flies and other biting flies,
species in open savanna or parkland habitats may also rely on visual stimuli because of the longer
site lines available to them. Thompson (1976a) presents evidence that suggests the savanna form
of S. damnosum may require no more than the sight of the human form to be activated while the
same appears not to be the case for the purportedly less anthropophilic (Crosskey 1990) ‘forest
form’ of S. damnosum.

250  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

Biting site selection

Post active orientation (i.e. post landing) components of the host-black fly interaction include
several complex activities such as biting site selection, biting, gorging and disengagement from
the host followed by escape.

Biting site selection in many black flies is characterised by rapid movement over the skin, fur or
feathers accompanied by a patting or scanning action of the prothoracic legs. The prothoracic tarsi
bear larger numbers of gustatory sensilla than the mesothoracic or metathoracic tarsi (Sutcliffe
and McIver 1976, 1987). This suggests that chemical cues are involved in mediating the biting
response though we still know very little about the nature of these cues or whether others (e.g.
thermal cues) also play a role in biting site selection. See the discussion under ‘Heat and moisture
cues’ and ‘Non-breath odours (including natural repellents)’ below for further consideration of
this topic.

Biting proper is accomplished by a complex interaction of the mouthparts with the skin. This
process is quite well understood from a mechanical – anatomical perspective. Various authors
have examined simuliid mouthparts in detail (Sutcliffe 1985, Sutcliffe and McIver 1982, Wenk
1962) and Sutcliffe and McIver (1984) provided a detailed analysis of the mouthpart sensilla of S.
venustum s.l. and their various possible roles during biting. The make-up and properties of black
fly saliva has been investigated as well. Like other blood feeders’ saliva, black fly saliva contains
a range of substances many of which are suited to interrupting the haemostatic process (Cupp
and Cupp 1997) thus allowing feeding to proceed as quickly and efficiently as possible. Studies
by (Stallings et al. 2002) on S. vittatum further reveal that black fly saliva has properties that may
help microfilariae in nearby sub-cutaneous tissues to move toward the black fly’s mouthparts
during biting.

Active feeding

The onset of active feeding (gorging) is known to be stimulated by certain substances (gorging
stimulants, phagostimulants) in the blood. Cessation of feeding is probably due to inputs from gut
stretch receptors though no specific information about this exists for black flies. Phagostimulants
for S. venustum s.l. were studied using artificial feeding methods (Smith and Friend 1982, Sutcliffe
and McIver 1979) and include various nucleotides (in particular ADP) which can stimulate full
gorging at concentrations as low as 10-5 M in saline.

Of particular interest in further understanding how biting and feeding are controlled in black
flies are electrophysiological studies of the tarsal, antennal, and mouthpart and food canal
sensilla. Putative functions for these receptors are reviewed by McIver and Sutcliffe (1986) based
on sensillar morphology and placement but until tested electrophysiologically, these proposed
functions must be considered unconfirmed.

Active orientation to the host

Carbon dioxide

Carbon dioxide is a powerful mediator of host seeking for many biting flies including many black
fly species. For this reason, CO2 from a block of dry ice or a tank has been a standard method
used to increase the black fly catch at many types of traps and targets. In such applications, CO2

Olfaction in vector-host interactions 251

James F. Sutcliffe

is often referred to as an ‘attractant’. This is a term that has been used loosely by many authors
(including this one, e.g. Sutcliffe 1986) over the years. For a chemical mediator of host seeking to
qualify as a true attractant it must induce upwind anemotaxis. It is not possible to know simply
by observing the accumulated swarm of flies around the trap, or by the increased catch of a sticky
target or entry trap, whether the greater numbers are due to an attractant effect of CO2 or whether
they are, instead, due to elevated CO2 levels simply stimulating increasing unoriented black fly
flight activity that, in turn, causes more flies to encounter the arresting stimulus of the visually
conspicuous trap or get caught on the sticky target.

To address this and similar questions, a method of intercepting black flies in their host-seeking
flight is needed. If interception traps can be deployed some distance away around a CO2 source,
it should be possible to differentiate between attraction and increased unoriented flight activity
since the former will be characterised by high catches on the downwind sides of traps while
the latter will be characterised by more or less equal catches on downwind and upwind sides.
Working in Alberta on populations made up mostly of S. arcticum (IIS-10.11), Sutcliffe et al. (1995)
deployed six Tangletrap-coated clear Lexan panels (upright sticky traps – USTs) as interception
traps (see section entitled ‘Challenges in the study of black fly host seeking behaviour’ for further
discussion of these traps) in a 12 m diameter ring around a CO2-baited cow silhouette trap (CST).
Approximately 82% of the black flies caught were on the downwind sides of the panels supporting
the interpretation that they had flown upwind in the CO2 plume. These results are similar to those
of Golini and Davies (1971) for S. venustum Say in Algonquin Park, Ontario. Such results justify
the use of the term ‘attractant’ with respect to CO2 for these particular black fly species and, they
may also be taken as some indication of CO2 acting as an attractant for other black fly species.
Nonetheless, this is a term that should always be used with due consideration.

In general in simuliids there appears to be good evidence for an increased response (in the form
of increased trap catches) with increased release of CO2, i.e. a dose-effect. However, the shape of
the dose-response curve, whether it is different for different species and what affects the curve’s
shape is poorly known. Part of this is due to the fact that much of the field work that reports on
the effects of CO2 dose on numbers of black flies caught has been based on very rough methods
of controlling and measuring CO2 release rate. To be most worthwhile, such work should employ
precision release valves and accurate rotameter type flow gauges to regulate and measure release
rates from CO2 tanks fitted with two stage regulators.

Fallis et al. (1967) measured the CO2 dose-response for black flies in Algonquin Park, Ontario,
Canada and found that catch rate at CO2-baited fan traps increased in decreasing proportion to
CO2 release rates from 0 to 800 ml/min. Sutcliffe and Schofield (unpublished results) have also
found a clear dose-response of CO2 for UST catches for several black fly species in Alberta and
Ontario. As in Fallis et al. (1967), these dose-effects ‘top out’ at higher, though still physiologically
relevant for cattle, CO2 release rates. It is not clear whether this is due to the plume simply
recruiting more and more black flies in its active space, until there are no more flies left to recruit,
or if it is due to an ever-increasing recruitment by the plume coupled with a fall off in numbers
that succeed in navigating the greater distances to the USTs. Torr (1990) found that CO2 plumes
were not readily navigable for Glossina pallidipes Austen but became much more so when acetone
was co-released. It would be interesting to see the effects of other chemical mediators of black fly
host-seeking on the shape of the CO2 dose-response curve.

Livestock under black fly attack can often be observed to bunch up apparently for the same
reasons other herd animals bunch up when under threat of attack from a predator – there is

252  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

safety in numbers. What this means in the context of biting fly attack is not clear. The ‘topping
out’ of the CO2 dose-response in black flies may provide the explanation for this behaviour in
animals under black fly attack since it suggests that though a large group of animals will produce
a great deal of CO2, the per unit return in terms of black flies attacking will not be linear, i.e. the
ratio of flies to individual cattle should drop and so, therefore, should individual host biting rate.
Ratti et al. (2006) present quantitative data showing a dilution effect on black fly biting on black
grouse proportional to group size. They do not discuss why this occurs but the shape of CO2 dose-
response curve may be part of the explanation.

Schofield and Sutcliffe (1996) show that the amount of CO2 produced by human individuals also
appears to account for individual host ‘attractiveness’ for human biting black flies in Algonquin
Park. Black fly catch on sticky Lexan panels mounted above the heads of seated, similarly dressed
human subjects proved to be significantly and consistently different over two years of experiments.
Removal of breath from the vicinity of the subjects (by having them breathe out through long
tubes with their openings many meters away at right angles to the wind) eliminated individual
differences and reduced catches on the panels by about 85%. This was shown to have been due to
the removal of the CO2 component, and not the non-CO2 breath odours, by the fact that release
of tanked CO2 from a tube at the subjects’ head levels at subject-specific rates restored individual
panel catches to about 90% of their previous levels and completely restored individual differences.

Non-CO2 breath odours

To date there is very little known about the chemical identities of non-CO2 breath odours that
mediate black fly host seeking though there is a variety of evidence that such substances exist for
at least some species (Sutcliffe 1986, 1987). It may be significant, for instance, that Schofield and
Sutcliffe (1996) were only able to restore about 90% of the whole breath catch on sticky panels
mounted above the heads human subjects in field experiments in Algonquin Park (see experiment
description above) when subject-specific amounts of tanked CO2 were released. This could mean
that non-CO2 components of breath are responsible for larger numbers of black flies responding
to the plume or that non-CO2 breath components play a role in ‘focussing’ the responders so
that more of them were intercepted by the sticky panels. The latter explanation is supported by
results from field work in Algonquin Park by Dean (2004). In these experiments, a yellow square
(46×46 cm) or cylinder (20 cm high×15 cm diameter) target was baited either with nothing, CO2
at 300 ml/min or with whole breath (adjusted to contain 300 ml/min of CO2) of a single adult male
subject. At the same time, a black cloth target or cylinder of the same configuration as its yellow
counterpart, was placed across the wind from the yellow target at separations of 0 m, 1 m or 3
m. Total number of flies caught on sticky interception screens placed immediately in front of the
yellow and black targets were not significantly different for CO2 and breath-baited trials but, for
any given separation, yellow targets baited with whole breath captured a higher proportion of the
total than did CO2 baited yellow targets. This suggests that whole breath contained substances
that ‘focussed’ the responding flies more tightly in the odour plume (perhaps by making the
plume more navigable) but that these substances did not recruit more flies to host seeking.
Similar properties have been found in tsetse flies for acetone and octenol which Torr (1990) found
improve the navigability of a CO2 plume for Glossina pallidipes.

Sutcliffe et al. (1994) tested several substances found in cattle breath for their abilities to enhance
catches of black flies in Athabasca, Alberta. These included acetone and 1-octen-3-ol (hereafter
termed octenol), 1- and 2-octanol and octanoic (caprylic) acid each at ‘high’ and ‘low’ release rates
and each accompanied by CO2 released from a pressurised tank. Only acetone at the high rate

Olfaction in vector-host interactions 253

James F. Sutcliffe

produced significantly higher catches than the CO2-only control. Unfortunately, this work was
plagued by inadequate CO2 flow control problems and the release rates of the chemicals was not
quantified. This, and the fact that the octenol effect in those host seekers it is known to occur in
is dose dependent (Takken and Knols 1999, Vale et al. 1985) means the lack of an octenol effect in
these experiments should not be taken for the lack of an octenol effect. Subsequent work done in
Alberta incorporating more precise CO2 flow control and using UST catching methods (JF Sutcliffe
unpublished results) failed to reproduce an acetone effect and further failed to show any effect
of octenol.

Atwood and Meisch (1993) also investigated the effects of octenol odours (presented with and
without CO2) on black fly catches in tent traps in Arkansas. Results were mixed though octenol
(released from pipe cleaner wicks inserted into vials) plus CO2 resulted in significantly larger
catches of Cnephia pecuarum Riley than CO2 alone in one trial of three. In the single trial in which
S. meridionale Riley catches are reported, octenol plus CO2 produced much smaller catches than
CO2 alone.

Results from experiments such as those reported in this section reveal some of the problems
that must be addressed for field experiments on odour mediation of black fly host seeking to
be maximally valuable. For instance, neither study could report precise or even estimated dose
rates for chemicals tested in the field because release from wicks (Atwood and Meisch 1993) and
from vials with different sized holes in the caps (Sutcliffe et al. 1994) would have varied through
the day depending on environmental conditions. The importance of such information in the
interpretation of results like these is difficult to over-state. Other features of these studies, such as
inadequate control of CO2 release rate and collecting devices (tent trap and CST, respectively) that
require landing and or trap entry for the fly to become part of the sample are also problematic.
The former is easily done with high precision valves and flow meters attached to a CO2 tank with
a two-stage regulator. The latter can best be addressed by using simple interception traps such as
USTs (Sutcliffe et al. 1995, see ‘Challenges in the study of black fly host seeking behaviour’ section
for further discussion of this).

Non-breath odours

One of the most dramatic examples of non-breath chemical mediation of black fly host seeking is
the effect of secretions of the uropygial glands of certain waterfowl on some ornithophilic species.
Lowther and Wood (1964) first reported that S. euryadminiculum is strongly and apparently
exclusively ‘attracted’ to loon uropygial gland extracts with or without a visual stimulus and with
or without CO2 (Fallis and Smith 1964). In this same study, catches of other waterfowl biters (e.g.
S. rugglesi, S. anatinum) in sweep net collections over filter papers impregnated with various
extracts of duck uropygial glands were also greater than control sweeps though, in this case, the
addition of CO2 significantly increased catch of these species. The active specifics in loon and duck
uropygial secretions have not been isolated and identified though we have unpublished data (SW
Schofield and JF Sutcliffe) that indicates that long chain alcohols in loon uropygial gland extracts
may play a role.

Sutcliffe et al. (1994) examined the ability of crude and fractionated ‘whole cow odour’ (WCO) and
cow urine to enhance CST catches of black flies (largely S. arcticum IIS-10.11) in Alberta. While
crude WCO plus CO2 resulted in catches significantly higher than the control (CO2 only), none of
the fractions did so suggesting a possible synergism between two or more components of the
WCO. Cow urine in combination with CO2, while it resulted in higher-than-control catches in CSTs

254  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

than CO2 alone, did not have a significant effect though it was close to significance (P<0.12). In
work done subsequently in Alberta (JF Sutcliffe, unpublished results) with better CO2 flow control
and using USTs instead of CSTs, cow urine (stored for no more than 3 weeks in a 4 °C refrigerator)
plus CO2 was found to result in larger catches than CO2 alone.

Information on human odours other than breath as mediators of black fly host orientation is
scant. The first evidence that such substances influence human biters was provided by Thompson
(1976a,b) who found that clothing that had been worn for some time enhanced trap catches of
the forest form of S. damnosum s.l. in Cameroon even in the absence of CO2. Thompson (1976b)
makes a circumstantial case for the source of this substance being apocrine sweat though this has
not been investigated further.

Schofield and Sutcliffe (1997) provide the only other evidence to date for an effect of human skin
secretions on biting responses of black flies. In field-lab experiments with human biting species
(S. venustum s.l.) in Algonquin Park, they first established that human test subjects had different
‘bitabilities’ (defined as each subject’s likelihood of being bitten by a given black fly) by placing
black flies caught individually in vials off a CO2-baited cloth target, directly onto a subject’s skin
and observing for a biting response for three minutes. Different subjects had significantly different
bitabilities that were maintained over several weeks. To isolate a possible skin substance effect
from other subject-specific characteristics that may have been influencing biting, similarly caught
black flies were given the opportunity to bite artificial latex rubber membranes that had been
treated with subject skin secretions and then stretched over a warmed (approx 32-37 °C) aluminum
block. The hierarchy of individual bitability and subject-membrane bitability corresponded closely
although in an unexpected manner. Treated membranes were bitten less frequently than control
membranes and membranes of more bitable individuals were so because they were less ‘repellent’
than membranes corresponding to less bitable individuals. Nothing is known of the chemical
properties of these apparent biting inhibitors though the observation that flies in vials on treated
membranes stayed on the sides and top of the vials could suggest that the effect was, at least in
part, due to volatiles. This work raises the possibility of what Schofield and Sutcliffe (1997) refer to
as ‘antibitants’ in human skin. Many a black fly field worker has experienced severe black fly attack
immediately after bathing and hair washing. The temporary removal of antibitants from the skin
and hair brought about by bathing and shampooing may help explain this phenomenon. While
the notion of antibitants or natural ‘repellents’ has not been pursued further for simuliids, there
is mounting evidence for host-derived compounds that variously interfere with aspects of host
seeking and blood feeding in other haematophagous arthropods (e.g. Logan et al. 2008).

While on the face of it the treatment effect described above looks like repellence, interpretations
of such experiments must be done carefully. Schofield and Sutcliffe (1997) recognise this and
propose alternatively that what is an apparent repellence could actually be the effect of skin
substances that stimulate walking activity that normally occurs when the insect lands on the skin
(see description of ‘patting’ behaviour in ‘Biting site selection’ section). Confinement of the black
fly in the vial might make this look like an attempt to escape a repellent substance.

A final, indirect, mode of chemical interaction between the host and the blood feeding black
fly may be what McCall and Lemoh (1997) refer to as the ‘invitation effect’ observed in certain
members of the S. damnosum complex. This is suggested to be a pheromone effect between black
flies in the act of biting the host and black flies that have not yet selected a biting site. The effect
is to increase the number of flies biting on a given area of the host’s skin. The sensory basis of
this effect has not been elucidated nor has it been determined who (the sender or the receiver)

Olfaction in vector-host interactions 255

James F. Sutcliffe

benefits from the message though the term ‘invitation effect’ implies the sender. It is arguable,
however, that by biting where saliva from previous bites has already increased blood flow to the
area, the receiving individual may benefit by being able to feed rapidly at that site. If this is the
case, it might be better to re-name this the ‘party crasher effect’.

Visual interactions with hosts

In Sutcliffe’s (1986) framework for host seeking in simuliids, chemically-mediated orientation to

the host eventually brings the fly into visual contact with the host. This contact would first be with
the host animal as a whole but with closer approach, the fly would be able to see and respond to
particular physical and textural features of the host.

All other things being equal, the range at which the host first becomes visible for different species
would be determined by the physical setting. Visual obstructions and the complexity in woodland
surroundings may make it more difficult for the host seeker to see a host animal of a given size,
shape and colouration than in relatively less complex, more open settings such as savanna and
parkland. However, the relative ambiguity of odour information about the location of an odour
source (such as a host animal – Brady et al. 1990) should mean that the host-seeking fly will
give priority to much less ambiguous visual information when it becomes available. This, in turn,
suggests that woodland species may rely on olfactory cues for longer than open landscape species
which may rely more on visual cues. Relatively little information on this question is available
though some support for this in the savanna form versus the forest form of S. damnosum is
presented in the section titled ‘Activation’.

Based on field experiments, Sutcliffe et al. (1995) estimated that host seeking S. arcticum (IIS-10.11)
responded visually to a host-mimicking CST (110 cm long by 46 cm wide) from about 8 m. At this
distance, they calculated that the image of the CST would subtend an angle of about 8° (in the
horizontal plane) and would cover all or parts of up 15 ommatidia. If other species of black flies
have a similar visual resolution capability for their hosts, it would mean that something the size
of a human would be resolved at a distance of approximately 20 m while a host the size of a loon
would not become apparent until it was within about 100 cm. Black fly species that bite smaller
birds in leafed out, relatively dimly lit wooded areas might not make effective visual contact until
within a few centimetres.

When visual and chemical stimuli are both ‘in play’, which take precedence? The reasoning
presented at the beginning of this section indicates that the less ambiguous visual signal should.
In field experiments in Algonquin Park, Ontario, Canada, Dean (2004) undertook to examine this
question for host-seeking mammalophilic black flies. In these experiments a yellow square (46×46
cm) or cylindrical (20 cm high × 15 cm diameter) target was baited either with nothing, CO2 at
300 ml/min or with whole breath of a single adult male adjusted to 300 ml per minute CO2. At the
same time, a black cloth target or cylinder of the same configuration as its yellow counterpart, was
placed either beside the yellow target (0 m separation) or at separations of 1 or 3 m. When at 0 m
separation, the sticky screens in front of black target always caught significantly more than half of
the flies irrespective of the bait. However, when the targets were separated by 1 or 3 m, the sticky
screens in front of the baited (yellow) targets collected relatively more flies than those in front of
the dark targets. Clearly, our understanding of how responses to different modalities of stimulus
are ordered during host seeking in simuliids requires further examination and probably needs to
integrate elements such host/target movement.

256  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

Crosskey (1990) provides a detailed discussion of the preferred feeding sites for different groups
of mammal-biting black flies. While details vary from host to host, he notes that most species
biting large mammals do so on either on the muzzle and ears or on the belly, chest and front
legs. The adaptive reasons for black flies having preferred biting areas are poorly researched but
presumably relate to some degree to avoidance of host defensive behaviours. Other reasons for
certain areas of the body being preferred for biting may also relate to avoiding competition with
other black fly or biting fly species. In ornithophilic species such as S. euryadminiculum and S.
rugglesi, preferred biting sites are on the head and neck Bennett et al. (1972) again presumably
because the animal is less able to defend itself from flies in such places.

Visual cues play an important role in determining where on the host the black fly will land, i.e.
in leading the fly to its preferred biting area on the host. Extensive study of colour and shape
preferences of a number of black fly species (reviewed by Sutcliffe 1986, 1987) have established
that black flies generally prefer to land on darker colours and matte surfaces. While black flies
do not seem to assess overall shape prior to landing, some species do exhibit a preference for
landing near edges and on prominences. Choice of landing site appears to be largely independent
of odour stimuli. Yee and Anderson (1995) showed that females of several black fly species they
studied in California would land on the head ends of deer-mimicking three dimensional models
irrespective of whether CO2 used to bait the models was released from the head end of the model
or from beneath it. Similarly, the black fly Wilhelmia equina L. will orient to the head end of a host
model even of the odour bait is at the tail end (Wenk and Schlorer 1963). When not given a head
or ears to indicate which end of a CST is which, S. arcticum landed equally on both ends (Sutcliffe
et al. 1995). These and similar findings suggest that though there is a range over which the host
seeking black fly is guided by host odours, at very close range, visual stimuli take over and guide
landing/biting site selection.

Heat and moisture cues

The importance of heat/thermal stimuli for host-seeking black flies is implied by the fact that
simuliids bite only warm-blooded hosts (birds and mammals). The role of thermal stimuli in
mediating landing and biting in human biting black flies in Algonquin Park Canada was examined
by Sutcliffe and McIver (1979), who showed that female flies collected individually in small vials
after landing on a human host and subsequently transferred onto a warmed latex membrane,
would bite the membrane more readily when the temperature differential between the membrane
and the air just above it was greater. They argued that this means that skin (or membrane)
temperature alone is not the key stimulus and suggested that the thermal differential may have
served as a short range landing cue. Schofield and Sutcliffe (1997) in their experiments on the
bitability of latex membranes treated with human skin secretions also measured the effect of the
temperature differential between the air and the artificial biting surface. Again, black flies that
experienced a greater temperature differential (14-17 °C) between the air and the membrane were
much more likely to bite the membrane than flies that experienced a lesser differential (11-14 °C).
The question of whether the temperature differential is a short range orientation/landing stimulus
(Sutcliffe and McIver 1979) or a biting stimulus (Smith and Friend 1982) remains an open one.

Virtually no information exists for the effects of water vapour either in breath or from the skin on
black fly host seeking.

Olfaction in vector-host interactions 257

James F. Sutcliffe

Discussion

Host seeking-blood feeding as a behavioural mode

The concept of ‘activation’ is used extensively in the discussion of host seeking in biting Diptera
though what is meant by the term is not always clear. Sutcliffe (1986, 1987) suggested that
activation is the event that causes a ‘switching over’ from some ground state of behaviour into
a host seeking ‘behavioural package’ or mode. While in the host seeking mode the host seeker
would be biased to respond to host originating stimuli in a manner appropriate to the host
seeking circumstances but, under different circumstances (e.g. in an oviposition behavioural
mode), the insect might respond to the same stimulus differently or not at all. This suggests that
the activation and continuation of black fly host seeking may fit the ‘rolling fulcrum’ model (Miller
and Strickler 1984) developed as a theoretical framework to explain host-seeking behaviour in
plant feeding insects.

During artificial feeding experiments in Algonquin Park, Sutcliffe and McIver (1979) observed
that black flies captured individually in vials after landing on a host would only bite the skin-
simulating artificially-warmed membrane if given the opportunity to do so within a few minutes
of collection. Flies held for too long would often not bite when placed on the membrane and
many appeared agitated and to be trying to get away from the warmth of the membrane. They
interpreted this as the result of the flies having switched out of the host-seeking mode during the
holding period. In the framework of the rolling fulcrum model, the host-related stimuli that served
to establish (activate) and sustain host seeking-appropriate responses had been interrupted long
enough for some of the flies (those that would not bite) to revert to some more basic behavioural
mode in which sources of heat might be interpreted as representing and increased desiccation
risk. When black flies were held in vials for different lengths of time after being collected from a
host, we found that there was a gradual decrease in the proportion of the sample that would bite
the artificial membrane such that the ‘half life’ of biting persistence was only a few minutes (JF
Sutcliffe, unpublished results). Dean (2004) investigated this phenomenon in human biting black
fly species in Algonquin Park and tried to determine factors that account for some flies switching
out of the host seeking mode before others and to see if biting persistence could be extended in
captive female black flies by providing flies with continued contact with simulated host stimuli (in
this case, slightly elevated levels of CO2 in an air stream) during the holding period. The treatment
had no effect and has not been followed up though it would be interesting to do so with more
complex and complete mixtures of host-odour stimuli.

The existence of a behavioural mode that appears to establish a behavioural context for the host
seeking stimulus-responses has implications for the interpretation of results from experiments
designed to study the host seeker-host interaction and for the circumstances under which such
work can be done meaningfully. In my experience, it is simply not possible to get female black flies
that have been held after being collected in the field to respond ‘appropriately’ to host stimuli in
the lab. This makes behavioural work on these problems ‘out of season’ and anywhere but in the
field problematic. Thus, lab-based olfactometer and wind tunnel experiments, odour choice tests,
artificial feeding experiments using previously collected flies or using flies reared from pupae and
so-on are exceedingly difficult to perform for this group. Even if it were possible to get black flies
to ‘behave’ in the lab, the fidelity with which lab responses reflect what would happen in a natural
setting would be open to question. Indeed, research on host-seeking responses in biting flies in
general should be carefully interpreted since the concept of host seeking as a context-sensitive
behavioural mode is not unique to black flies.

258  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

Host preferences

While the range of potential black fly host preferences is determined at one level by ecological
and temporal factors (i.e. the fly and the prospective host species must occur in the same habitat
at the same time), whether a given black fly species actually prefers a particular host species is
presumably determined by subtleties of sensory cues given off by hosts and by the responses of
the host seeker to those stimuli.

The precise dimensions of the host preferences for most black fly species are not well known
partly because many black fly species are very difficult to tell apart in the field and can often only
be differentiated at the species level with cytogenetic or molecular methods. Another reason
for our poor understanding of black fly host preferences is that is very difficult to field-collect
black flies with appreciable amounts leftover blood meals in the field. Thus, it has been difficult
to apply ELISA and other immunological methods to the analysis of black fly blood meal sources.
However, the current widespread availability of very sensitive molecular methods may make it
more feasible than ever to determine the origins of blood meal remnants in flies even several days
after feeding. Though this opportunity has not been exploited widely yet, some researchers have
taken advantage (e.g. Malmqvist et al. 2004) and the approach has already yielded information
that has implications for black fly interactions with their hosts. Hellgren et al. (2008) found that
three of four specimens of S. annulus (Lundstrom) collected with a car top trap at points 200
km apart in northern Sweden had all fed from Common Cranes (Grus grus (Linnaeus)) despite
the fact this bird is not numerous in the study area. This suggests that S. annulus may orient
preferentially to G. grus because of specific odour or visual stimuli the bird possesses. Results like
these when investigated further may lead to many more examples of specific, host stimulus-based
preferences in black flies like those that exist between, for instance, S. euryadminiculum and loons
and between S. rugglesi and waterfowl.

A better understanding of the stimuli that the host seeking black fly encounters and how the
fly reacts to these stimuli may explain much about the sensory and behavioural mechanisms
that result in host preferences. For instance, where in the host-seeking process does host the
preference exert itself (i.e. where does host discrimination occur)? Is it from the start of active
orientation or does the insect begin host seeking based on initial minimal information about
the prospective host and make decisions about whether to continue further host seeking when
subsequent cues fail to, or succeed in, satisfying certain ‘expected’ criteria (e.g. host-specific
body or breath odours, visual host qualities that the insect will respond to in the host seeking
behavioural context) that confirm that the prospective host is indeed suitable. That the insect
could have enough information about the prospective host from initial contact seems highly
unlikely unless the stimulus is highly specialised such as in the case of loon uropygial odours for
S. euryadminiculum. For many more generalist species, first contact may be with more generalised
stimuli such as elevated levels of CO2. In the latter situation, host discrimination must be ongoing
as the insect continues to move closer to the host coming into contact with more and more host-
related stimuli that ultimately allow the insect to determine whether it should continue to invest
effort and invoke risk both of which host seeking and the act of blood feeding entail.

The range of host preferences exhibited in black flies, and in biting flies in general, must reflect
evolutionary ‘choices’ made by different species or lineages. Different host-seeking strategies
would come with their own sets of advantages and disadvantages. A high degree of host species
specialisation could be combined with a narrow range of specific host-related stimuli that the
insect would require to initiate and engage in sustained host seeking. This strategy would narrow

Olfaction in vector-host interactions 259

James F. Sutcliffe

the number of host animals available for the specialist to feed on but would help ensure that its
investment of energy in host seeking would be most likely to result in a blood meal with minimum
risk. Alternatively, for biting flies with broader host ranges, a better strategy might be to respond
initially to more general host-related stimuli (e.g. CO2) and discriminate as more information, in
the form of host-related stimuli, becomes available. This greatly enlarges the number of potential
host animals available but it also increases the probability of investing energy and undertaking
additional risk for no return.

Members of the Anopheles gambiae Giles mosquito species complex are host specialists and
appear to conform to this pattern. In behavioural experiments in Burkina Faso, Costantini et al.
(2001) showed that An. gambiae s.l. females relied less on CO2 cues and more on human-specific
odours to mediate host seeking. Perhaps An. gambiae made the evolutionary ‘choice’ to specialise
on humans long ago. This is plausible for an African species since it could have co-evolved with
humans over several hundred thousand years. The human biting black fly counterparts to An.
gambiae could be the savannah and forest forms of African species S. damnosum s.l. Thompson
(1976a,b) working in Cameroon showed that substances in human skin and sweat (with or without
CO2) result in large swarms of the forest form around traps while the human profile without any
skin or breath odours, is swarmed heavily by savannah form. Certain black flies that bite non-
humans are also highly host specific and also fit the pattern. Simulium euryadminiculum appears
to specialise on loons and swarms en masse to odours of extracts of loon uropygial gland. The
addition of CO2 to uropygial gland odour appears not to increase numbers. Species such as S.
rugglesi and S. anatinum with reportedly somewhat broader host preferences (various ducks and
to a limited extent other birds) only respond to duck uropygial gland secretions if accompanied
by CO2 and to a lesser extent to CO2 alone (Fallis and Smith 1964). Other ornithophilic species
such as S. aureum s.l. and Prosimulium decemarticulatum Twinn which have even broader host
preferences among waterfowl and passerines, respond in large numbers to CO2 alone but not to
uropygial gland secretions.

Given the broad occurrence of CO2 as a sensory cue for insects of many types including for many
non-biting Diptera (Guerenstein and Hildebrand 2008), perhaps CO2 represents the ancestral host
odour for black flies and sensitivity to other host odours evolved subsequently. This process could
have resulted in the various degrees of host specificity that are represented in the family today
and to a gradual de-emphasis of the CO2 cue for those with the greatest degree of host-specificity.

Challenges in the study of black fly host-seeking behaviour

Notwithstanding the concerns about the interpretation of lab-evoked ‘host seeking’ responses
(above), work with many mosquito species and with certain other biting flies (e.g. stable flies,
tsetse) from lab colonies has helped elucidate various features of host seeking in these groups.
However, to date, virtually all work that sheds light on aspects of host seeking in black flies has
been field-based because black fly colonies are few and far between because they are difficult to
establish and labour-intensive and costly to maintain. Edman and Simmons (1986) summarise the
challenges, which include the infrastructure requirements associated with maintaining an insect
that oviposits and whose larval stages occur in flowing water as well as the particular solutions
that must be found to inducing mating, blood feeding, and oviposition by adults. According to
Adler et al. (2004), only four North American black fly species have been successfully colonised on
a continuing basis and only a handful of colonies currently exist.

260  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

The usefulness of colony reared black flies for host seeking work, if they were available, would
probably also be limited for reasons already discussed. Indeed, many attempts to colonise black
flies have involved autogenous species since they do not have to blood feed (or to be induced to
perform some lab equivalent of host seeking) to produce eggs. Though inducing meaningful host-
seeking responses with colonised black flies is problematic, there are other related constructive
uses that black flies from a lab colony could be put to. For instance, colonised black flies could
be used for electroantennogram-screening of candidate host-mediating chemicals thus helping
focus efforts in the field.

Tsetse fly trap and target responses are relatively well worked out (Gibson and Torr 1999; Chapter
12, this volume) making it possible to design field experiments appropriately for the purposes
intended. Black fly host-seeking research over the years has relied on a variety of host mimicking
traps and targets of various shapes and sizes but, unlike in the tsetse world, how black fly traps
and targets work is not well understood nor do we know in detail the responses they evoke from
the host-seeking black fly. Electric nets in various configurations and applications have been one
of the key tools in the analysis of tsetse host seeking behaviour because they can be used to
intercept host orienting flies in a non-interactive way. Black fly work has not had an equivalent of
the electric net. In an attempt to address this, we have used both sticky Lexan squares and sticky
neutral coloured fibreglass window screening in such applications (e.g. Schofield and Sutcliffe
1996, Sutcliffe et al. 1996). In experiments where the desire is to assess the effects of candidate
attractants without the introduction of other variables such colour preferences, trap entry
behavior, and so-on, these methods are superior to more complex systems such as CSTs. However,
sticky Lexan sheets and screens still have not been assessed for their ability to truly intercept
host seeking black flies. This would be a worthwhile undertaking as would an assessment of the
applicability of electric net technology to black fly host seeking research.

Conclusions

As we attempt to come to grips with the most basic of questions relating to host-seeking
behaviour in black flies and biting flies in general, the landscape itself is changing. The most
recent ingredient added to the host-insect interaction picture is the growing appreciation that
the parasites that blood feeding Diptera transmit manipulate many aspects of their vertebrate and
insect hosts’ physiologies including those aspects that mediate host-seeking. The ability of certain
malaria parasites to manipulate their mosquito vectors during blood feeding has been known for
many years (e.g. Rossignol et al. 1984, 1986) but evidence is now mounting that parasite-infected
host seekers may respond differently than non-infected host seekers during host seeking. There
is also considerable evidence that parasite-infected vertebrate hosts may be more attractive for
vector species (see reviews by Hurd 2003, Lefevre and Thomas 2008; Chapter 16, this volume).
Kruppa and Burchard (1999) compared the ability of Onchocerca volvulus-infected individuals
to ‘attract’ filarial-carrying and non-carrying black flies but found no differences. Nonetheless,
there is no doubt that there are parasite manipulations of hosts that influence the host-biting
fly interaction (indeed, that make this a three-way interaction) and that they will also have to be
accounted for before a full understanding of biting fly host seeking behaviours can be achieved.

References

Addison EM (1980) Transmission of Dirofilaria ursi Yamaguti, 1941 (Nematoda: Onchocercidae) of black bears (Ursus
americanus) by blackflies (Simuliidae). Can J Zool 58: 1913-1922.

Olfaction in vector-host interactions 261

James F. Sutcliffe

Adler PH, Currie DC and Wood DM (2003) The black flies (Simuliidae) of North America. Comstock Publishing, Cornell,
USA.
Allan SA, Day JF and Edman JD (1987) Visual ecology of biting flies. Annu Rev Entomol 32: 297-316.
Anderson, JR (1986) Reproductive strategies and gonotrophic cycles of black flies. In: Kim KC and Merritt RW (eds) Black
flies. Ecology, population management and annotated world list. Pennsylvania State University, University Park,
PA, USA, pp 63-76.
Atwood DW and Meisch MV (1993) Evaluation of 1-octen-3-ol and carbon dioxide as black fly (Diptera, Simuliidae)
attractants in Arkansas. J Amer Mosq Contr Assoc 9: 143-146.
Beck BE (1980) Clinical studies of the effect of the black fly Simulium arcticum on cattle. In: Haufe WO and Croome GCR
(eds) Control of black flies in the Athabasca River. Alberta Environment, Lethbridge, Alberta, Canada, pp 233-237.
Bennett GF, Fallis AM and Campbell AG (1972) The response of Simulium (Eusimulium) euryadminiculum Davies (Diptera:
Simuliidae) to some olfactory and visual stimuli. Can J Zool 50: 793-800.
Brady J, Packer MJ and Gibson G (1990) Odor plume shape and host finding by tsetse. Ins Sci Applic 11: 377-384.
Costantini C, Birkett MA, Gibson G, Ziesmann J, Sagnon N’F, Mohammed HA, Coluzzi M and Pickett JA (2001)
Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae to human-specific sweat
components. Med Vet Entomol 15: 259-266.
Crosskey RW (1990) The natural history of blackflies. Wiley, Chichester, UK.
Cupp EW (1986) The epizootiology of livestock and poultry diseases associated with black flies. In: Kim KC and Merritt
RW (eds) Black flies. Ecology, population management and annotated world list. Pennsylvania State University,
University Park, PA, USA, pp 387-395.
Cupp EW and Cupp MS (1997) Black fly (Diptera: Simuliidae) salivary secretions: Importance in vector competence and
disease. J Med Entomol 34: 87-94.
Cupp EW, Mare CJ, Cupp MS and Rowley FB (1992) Biological transmission of Vesicular Stomatitis Virus (New Jersey) by
Simulium vittatum (Diptera, Simuliidae). J Med Entomol 29: 137-140.
Dean BL (2004) Factors affecting host location behaviour in a field population of the black fly Simulium venustum (sensu
lato) (Diptera: Simuliidae). MSc Dissertation, Trent University, Peterborough, Canada.
Edman JD and Simmons KR (1986) Maintaining black flies in the laboratory. In: Kim KC, Merritt RW (eds) Black flies.
Ecology, population management and annotated world list. Pennsylvania State University, University Park, PA,
USA, pp 305-312.
Fallis AM and Smith SM (1964) Ether extracts from birds and CO2 as attractants for some ornithophilic simuliids. Can J
Zool 42: 723-730.
Fallis AM, Bennett GF, Griggs G and Allen T (1967) Collecting Simulium venustum females in fan traps and on silhouettes
with the aid of carbon dioxide. Can J Zool 45: 1011-1017.
Francy DB, Moore CG, Smith GC, Jacob WL, Taylor SH and Calisher CH (1988) Epizootic vesicular stomatitis in Colorado,
1982 – isolation of virus from insects collected along the northern Colorado Rocky Mountain front range. J Med
Entomol 25: 343-347.
Gibson G and Torr SJ (1999) Visual and olfactory responses of haematophagous Diptera to host stimuli. Med Vet Entomol
13: 2-23.
Golini VI and Davies DM (1971) Upwind orientation of female Simulium venustum Say (Diptera) in Algonquin Park,
Ontario. Proc Entomol Soc Ont 101: 49-54.
Guerenstein PG and Hildebrand JG (2008) Roles and effects of environmental carbon dioxide on insect life. Annu Rev
Entomol 53: 161-178.
Hellgren O, Bensch S and Malmqvist B (2008) Bird hosts, blood parasites and their vectors – association uncovered by
molecular analyses of blackfly blood meals. Molecular Ecol 17: 1605-1613.
Hurd H (2003) Manipulation of medically important insect vectors by their parasites. Annu Rev Entomol 48: 141-61.
Kruppa TF and Burchard GD (1999) Similar attraction by onchocerciasis patients and individuals putatively immune to
Onchocerca volvulus. Trans R Soc Trop Med Hyg 93: 365-367.
Lefèvre T and Thomas F (2008) Behind the scene, something else is pulling the strings: Emphasizing parasitic
manipulation in vector-borne diseases. Infection, Genetics and Evolution 8: 504-519.

262  Olfaction in vector-host interactions

 11. Black fly interactions with their hosts

Logan LG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue Luntz AJ and Pickett JA (2008) Identification
of human-derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. J Chem Ecol 34:
308-322.
Lok JB, Walker ED and Scoles GA (2000) Filariasis. In: Eldridge BF, Edman JD (eds) Medical entomology. A textbook on
public health and veterinary problems caused by arthropods. Kluwer, Dordrecht, the Netherlands, pp 299-375.
Lowther JK and Wood. DM (1964) Specificity of a black fly, Simulium euryadminiculum Davies, towards its host, the
common loon. Can Entomol 96: 911-913.
Malmqvist B, Strasevicius D, Hellgren O, Adler PH, Bensch S (2004) Vertebrate host specificity of wild-caught blackflies
revealed by mitochondrial DNA in blood. Proc R Soc Lond B 271: S152-S155.
McCall PJ and Lemoh PA (1997) Evidence for the ‘‘invitation effect’’ during bloodfeeding by blackflies of the Simulium
damnosum complex (Diptera: Simuliidae). J Insect Behav 10: 299-303.
McIver SB and Sutcliffe JF (1986) Sensory basis of behavior and structural adaptations for feeding in black flies. In: Kim
KC, Merritt RW (eds) Black flies. Ecology, population management and annotated world list. Pennsylvania State
University, University Park, PA, USA.
Mead DG, Ramberg, FB, Besselsen DG and Mare CJ (2000) Transmission of vesicular stomatitis virus from infected to
noninfected black flies co-feeding on nonviremic deer mice. Science 287: 485-487.
Miller JR and Strickler KS (1984) Finding and accepting host plants. In: Bell WJ, Cardé TR (eds) Chemical ecology of
insects. Chapman Hall, London, UK, pp 127-157.
Ratti O, Ojanen U and Helle P (2006) Increasing group size dilutes black fly attack rate in Black Grouse. Ornis Fennica
83: 86-90.
Rossignol PA, Ribeiro JMC and Spielman A (1984) Increased intradermal probing time in sporozoite-infected mosquitoes.
Am J Trop Med Hyg 33: 17-20.
Rossignol PA, Ribeiro JMC and Spielman A (1986) Increased biting rate and reduced fertility in sporozoite-infected
mosquitoes. Am J Trop Med Hyg 35: 277-279.
Schofield SW and Sutcliffe JF (1996) Human individuals vary in attractiveness for host-seeking black flies (Diptera:
Simuliidae) based on exhaled carbon dioxide. J Med Entomol 33: 102-108.
Schofield SW and Sutcliffe JF (1997) Humans vary in their ability to elicit biting responses from Simulium venustum
(Diptera: Simuliidae). J Med Entomol 34: 64-67.
Shemanchuk JA (1986) Host-seeking behavior and host preference of Simulium arcticum. In: Kim KC and Merritt RW (eds)
Black flies. Ecology, Population Management and Annotated World List. Pennsylvania State University, University
Park, PA, USA, pp 250-260.
Smith SM (1969) The black fly, Simulium venustum, attracted to the turtle, Chelydra serpentina. Entomol News 80: 107-
108.
Smith JJB and Friend WG (1982) Feeding behaviour in response to blood fractions and chemical phagostimulants in
the black-fly, Simulium venustum. Physiol Entomol 7: 219-226.
Stallings T, Cupp MS and Cupp EW (2002) Orientation of Onchocerca linealis Stiles (Filaroidea: Onchocercidae)
microfilariae to black fly saliva. J Med Entomol 39: 908-914.
Sutcliffe JF (1985) Anatomy of membranous mouthpart cuticles and their roles in feeding in black flies. J Morphol 186:
53-68.
Sutcliffe JF (1986) Black fly host location: a review. Can J Zool 64: 1041-1053.
Sutcliffe JF (1987). Distance orientation of biting flies to their hosts. Insect Sci Appl 8: 611-616.
Sutcliffe JF and McIver SB (1976) External morphology of the sensilla on the legs of selected black fly species (Diptera:
Simuliidae). Can J Zool 54: 1779-1787.
Sutcliffe JF and McIver SB (1979) Experiments on biting and gorging behaviour in the black fly, Simulium venustum.
Physiol Entomol 4: 393-400.
Sutcliffe JF and McIver SB (1982) Innervation and structure of mouthpart sensilla in females of the black fly, Simulium
venustum (Diptera: Simuliidae). J Morphol 171: 245‑258.
Sutcliffe JF and McIver SB (1984) Mechanics of blood feeding in black flies (Diptera: Simuliidae). J Morphol 180: 124-144.

Olfaction in vector-host interactions 263

James F. Sutcliffe

Sutcliffe JF and McIver SB (1987) Fine structure of tarsal sensilla of male and female Simulium vittatum (Diptera:
Simuliidae). J Morphol 192: 13-26.
Sutcliffe JF, Shemanchuk JA and McKeown DB (1994) Preliminary survey of odours that attract the black fly, Simulium
arcticum (IIS-10.11)(Diptera: Simuliidae) to its cattle hosts in the Athabasca region of Alberta. Insect Sci and Its
Application 15: 487-494.
Sutcliffe JF, Steer DJ and Beardsall D (1995) Studies of host location behaviour in the black fly, Simulium arcticum Malloch
(IIS-10.11)(Diptera: Simuliidae): Aspects of close-range trap orientation. Bull Entomol Res 85: 415-424.
Takken W (1991) The role of olfaction in host-seeking of mosquitoes: a review. Insect Sci Applic 12: 297-295.
Takken W and Knols BG (1999) Odor-mediated behavior of Afrotropical malaria mosquitoes. Annu Rev Entomol 44:
131-157.
Thompson BH (1976a) Studies on the attraction of Simulium damnosum s.l. (Diptera: Simuliidae) to its hosts. I. The
relative importance of sight, exhaled breath, and smell. Tropenmed. Parasitol 27: 355-473.
Thompson BH (1976b) Studies on the attraction of Simulium damnosum s.l. (Diptera: Simuliidae) to its hosts. II. The
nature of substances of the human skin responsible for attractant olfactory stimuli. Tropenmed Parasitol 27: 83-90.
Thylefors B and Alleman M (2006) Towards the elimination of onchocerciasis. Ann Trop Med Parasitol 100: 733-746.
Torr SJ (1990) Dose response of tsetse flies (Glossina) to carbon dioxide, acetone and octenol in the field. Physiol
Entomol 15: 93-103.
Vale GA and Hall DR (1985) The role of 1-octen-3-ol, acetone and carbon dioxide in the attraction of tsetse flies, Glossina
spp. Bull Entomol Res 75: 209-217.
Wenk P (1962) Anatomie des Kopfes von Wilhelmia equina L. Females (Simuliidae syn. Melusinidae, Diptera). Zool Jahrb
Abt Anat Ontog Tiere 80: 81-134.
Wenk P and Schlorer G (1963) Wirstorientierung und Kopulation bei blutsaugenden Simuliiden (Diptera). Z. Tropenmed
Parasitol 14: 177-191.
Yee WL and Anderson JR (1995) Trapping black flies (Diptera-Simuliidae) in northern California. 2. Testing visual cues
used in attraction to CO2-baited animal head models. J Vector Ecol 20: 26-39.

264  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

12. Olfaction in Glossina – host interactions: a tale of two tsetse

Steve J. Torr and Philippe Solano

Abstract

The role of olfaction in the host-oriented behaviour of tsetse is reviewed. For all tsetse, odours
play a role in host location but the relative importance of olfactory stimuli varies between species.
For Morsitans-group tsetse such as Glossina pallidipes and G. m. morsitans, host odours elicit long-
range (~100 m) orientation whereas Palpalis group species such as G. palpalis spp. and G. fuscipes
spp., while showing significant olfactory responses, are much less responsive. Hence baiting traps
with natural host odours can increase the catches of some Morsitans tsetse ten-fold compared
to only a doubling with Palpalis tsetse. Species that are less sensitive to odours seem to be more
responsive to visual stimuli. The semiochemicals present in host odour include carbon dioxide,
1-octen-3-ol, acetone, butanone, 4-methylphenol and 3-n-propylphenol. For Morsitans tsetse,
odours from hosts are not, in general, significantly different from those of non-hosts and thus
olfaction does not seem to play an important role in host selection. The notable exception to this is
human odour which is repellent to some species of Morsitans tsetse. Traps and insecticide-treated
targets baited with blends of 1-octen-3-ol, ketones and phenols are, arguably, the most successful
application of artificial baits to vector control worldwide. The current use and future prospects of
using natural and artificial baits to control tsetse are discussed.

Keywords: Glossina, biting flies, host stimuli, vector behaviour, semiochemicals, allomones,
kairomones, attractants, repellents

Introduction

Tsetse (Glossina spp.) are unlike most other haematophagous Diptera in that both sexes of tsetse
rely exclusively on blood for their development and maintenance and hence, every ~3 days they
visit their hosts to feed. Tsetse also have an unusual form of reproduction: adenotrophic viviparity.
From the age of ~6 days, females produce a single egg which develops within the uterus over a
period of 7-12 days, dependent on temperature. The mature larvae, consisting of equal numbers
of males and females, are then deposited on the soil where they burrow to pupariate, emerging
~30 days later as adults. If there were no deaths of pupae or adults, and larvae were produced
with perfect efficiency, then the minimum period that a female must live to ensure population
replacement is ~25 days, but in reality the average female must live nearly twice as long. This
combination of regular blood feeding and longevity makes tsetse efficient cyclical vectors of
Trypanosoma spp., the causative agents of trypanosomiasis in humans (sleeping sickness) and
livestock (nagana); the trypanosomes have sufficient time to develop in the fly and a high
probability of subsequent transmission to other hosts.

The unusual bionomics of tsetse have several important impacts on efforts to control tsetse-
borne diseases. First, classical interventions directed against the immature stages are impossible
– the larva is contained within the mother and the pupa is buried in the ground. Second, the
slow reproductive rate means that applying a relatively low, but persistent, mortality against a
tsetse population can drive it to extinction; a control measure that kills ~3% of the adult females/
day will eliminate a population of tsetse (Hargrove 1988). Third, the absolute reliance of tsetse
on feeding regularly from their hosts makes them vulnerable to interventions that disrupt – or
exploit – this behaviour. Indeed, this vulnerability was apparent from virtually the moment that

Olfaction in vector-host interactions 265

Steve J. Torr and Philippe Solano

the role of tsetse in the transmission of trypanosomes was recognised. In the 1890s, an epidemic
of rinderpest devastated ungulate populations across East and southern Africa and thus had a
huge impact on the distribution and abundance of tsetse hosts. Consequently, large areas of
southern Africa became tsetse free. The realisation that tsetse could be controlled by eliminating
their hosts led to the widespread use of game elimination to control tsetse-borne diseases.

Some of the earliest interventions against tsetse attempted to kill flies as they approached
natural or artificial hosts. On the island of Principe, sticky panels were attached to the backs of
humans working in tsetse-infested area; the density of tsetse was reduced greatly although the
relative contribution of the sticky-backed humans to this success is unclear. In South Africa, Harris
developed traps that simulated some of the visual properties of natural hosts and used these in an
attempt to control tsetse in present-day KwaZulu-Natal. Huge numbers of tsetse were captured,
and the population numbers declined, but local elimination of the tsetse population was never
achieved. In Zululand too, however, the contribution of traps is also in doubt, since hunting
operations to eliminate the hosts of tsetse were performed at the same time (Buxton 1955).

The early interest in these approaches to tsetse control did not persist, however: concerns about
the gross environmental impact of game elimination, the failure of traps as a control tool, and
the development of organochlorine insecticides led to a decline in host-based interventions
and, in their place, there emerged control methods which relied on the widespread application
of insecticides to the resting sites of tsetse. In Francophone West Africa, the wide-spread use
of drugs rather than vector control formed the mainstay of efforts against Gambian sleeping
sickness. In present-day Cameroon and Burkina Faso, Eugene Jamot developed the use of highly
mobile medical teams to detect and treat cases of sleeping sickness and between 1925 and 1935
hundreds of thousands of people were treated (see reviews by Courtin et al. 2008 and Maudlin
2006). This emphasis on case detection and treatment to combat Gambian sleeping sickness has
persisted to the present day as evidenced by the strategy currently promoted by WHO (Simarro
et al. 2008). It was not until the 1970s that interest in exploiting the host-orientated behaviour
of tsetse was revived. The research that underpinned this revival, and its subsequent impact on
tsetse control, is the subject of this review.

The geopolitics of tsetse and humans

Tsetse infest some 10 million km2 of sub-Saharan Africa but within this area the various groups
of tsetse are associated with particular ecological zones (Table 1). The Fusca group of species is
generally associated with densely-forested regions of Central Africa; the Palpalis group occurs
mainly in riverine areas of Central and West Africa, and the Morsitans group is associated with
savannah regions across the continent. These differences in the geographical distribution of tsetse
species have important epidemiological implications. The savannah areas are the main areas for
cattle production and hence species from the Morsitans group are important in the epidemiology
of animal trypanosomiasis. On the other hand, the Palpalis group is associated with mangrove
swamps and densely-settled riverine systems of West Africa where T. brucei gambiense occurs
(Solano et al. 2008); this species of Trypanosoma accounts for >90% of sleeping sickness cases and
hence the Palpalis group is particularly important in the transmission of this disease. The forest
regions infested by Fusca group are sparsely populated and hence species from this group play
little part in the transmission of trypanosomiasis in most places.

Not surprisingly, most research has been focussed on species of epidemiological importance. As
with many aspects of tsetse biology and control, an account of the olfactory interactions of tsetse

266  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

Table 1. Species and subspecies of tsetse.

Subgenus Nemorhina (Palpalis group)

1a. G. palpalis palpalis Robineau-Desvoidy, 1830
1b. G. palpalis gambiensis Vanderplank, 1949
2. G. tachinoides Westwood, 1850
3a. G. pallicera pallicera Bigot, 1891
3b. G. pallicera newsteadi Austen, 1929
4a. G. fuscipes fuscipes Newstead, 1910
4b. G. fuscipes martinii Zumpt, 1935
4c. G. fuscipes quanzensis Pires, 1948
5. G. caliginea Austen, 1911
Subgenus Glossina s. str. (Morsitans group)
1. G. longipalpis Wiedemann, 1830
2a. G. morsitans morsitans Westwood, 1850
2b. G. morsitans centralis Machado, 1970
2c. G. morsitans submorsitans Newstead, 1910
3. G. pallidipes Austen, 1903
4. G. austeni Newstead, 1912
5. G. swynnertoni Austen, 1923
Subgenus Austenina (Fusca group)
1a. G. fusca fusca Walker, 1849
1b. G. fusca congolensis Newstead and Evans, 1921
2a. G. nigrofusca nigrofusca Newstead, 1910
2b. G. nigrofusca hopkinsi Van Emden, 1944
3. G. fuscipleuris Austen, 1911
4. G. severini Newstead, 1913
5. G. vanhoofi Henrard, 1952
6. G. nashi Potts, 1955
7. G. tabaniformis Westwood, 1850
8. G. longipennis Corti, 1895
9. G. brevipalpis Newstead, 1910
10. G. medicorum Austen, 1911
11. G. schewtzi Newstead and Evans, 1921
12. G. haningtoni Newstead and Evans, 1922
13. G. frezili Gouteux, 1987

and their hosts has two parts: one concerned with the savannah species of East and southern
Africa and another concerned with the riverine flies of West Africa.

Host-orientated behaviour of savannah flies

Species within the Morsitans-group have associations with particular types of savannah woodland.
For instance, G. m. morsitans Westwood and G. m. centralis Machado in East and southern Africa
are associated with ‘miombo’ woodland, dominated by species of Isoberlinia, Brachystegia
and Julbernadia, and the West African G. m. submorsitans Newstead is found in similar ‘doka’
woodland dominated by Isoberlinia doka. Glossina Pallidipes Austen is regarded generally as

Olfaction in vector-host interactions 267

Steve J. Torr and Philippe Solano

being predominant in dense thicket. However, G. pallidipes and G. m. morsitans may also be found
together in, for example, the extensive mopane (Colophospermum mopane) woodlands of Zambia
and Zimbabwe. Several species have relatively confined geographical distributions: G. swynnertoni
Austen is found only in the savannah woodlands of the Maasai steppe and G. austeni Newstead is
associated with evergreen thicket lying within 200 km of the east coast of Africa, but extending in
a strip from Somalia to South Africa. The Morsitans fly can also display distributions akin to riverine
flies, especially in the dry season when suitable resting sites are confined to riparian woodland
(Hargrove and Vale 1980, Torr and Hargrove 1999; Vale 1971). The diet of Morsitans group tsetse is
less variable than that of Palpalis group species. The diet of all Morsitans-group species comprises
blood from suid and bovid hosts (Clausen et al. 1998). The particular host species with these
families can vary according to availability; meals from cattle, for instance, may be more important
in settled areas whereas buffalo and kudu may be important in game reserves. Nonetheless this
variability is less than that of Palpalis group flies which can persist feeding on largely on reptiles
or mammals. Hence, the removal of just six key host species can eliminate Morsitans-group tsetse
(Vale and Cumming 1976) but not Palpalis group flies.

Are host odours important?

Until the 1970s, it was generally assumed that savannah species of tsetse located their hosts
largely by vision, an opinion based primarily on the observation that tsetse were attracted to
moving objects such as cars and walking men. It was not until the development of techniques
for separating the responses to olfactory and visual stimuli that this assumption could be tested
objectively. Two technical developments were particularly important for this. First, electric nets
(Vale 1974a) provided an objective method for assessing the numbers of tsetse in the vicinity of
moving or stationary hosts. Second, the use of ventilated pits (Vale 1974b) provided a means of
producing natural host odour with or without any associated visual stimulus. Hence, Vale (1974b)
was able to compare the numbers of tsetse in the vicinity of a moving or stationary warthog, or the
numbers approaching a visual model of a warthog with or without natural host odour. In a series of
ground-breaking experiments, Vale (1974b) demonstrated unequivocally that tsetse were indeed
attracted to mobile hosts largely in response to the animal’s movement rather than its odours. On
the other hand, most (~90%) tsetse arriving at a stationary ox or warthog did so in response to the
host’s odours. Increasing the dose of odour, by placing more hosts in the ventilated pits, increased
the numbers of tsetse attracted; each ten-fold increase in dose producing a ~2.5-fold increase in
the catch (Hargrove and Vale 1978, Hargrove et al. 1995).

Following earlier reviewers (e.g. Gibson and Torr 1999, Sutcliffe 1987), it is useful to consider three
main phases in the odour-orientated responses of tsetse: (1) activation, which marks the initiation
of host-orientated responses; (2) long-range responses, which bring the activated fly to the vicinity
of the host and (3) short-range responses, which culminate in it landing and feeding on a host. The
host-finding sequence ceases with the insect alighting on the host, and consequently we do not
consider the probing and feeding responses.

Activation

Tsetse are faced with the problem of locating highly mobile hosts that are frequently hidden in
dense vegetation. To overcome this challenge, there are two possible search strategies: either
‘sit-and-wait’ for an animal to pass by or ‘range’ so as to increase the probability of encountering
a host. Tsetse are diurnally active (Brady and Crump 1978, et ante) but their total daily flight time
is limited to <30 minutes (Bursell and Taylor 1980). For most of the day, they rest on branches

268  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

and boles (Hadaway 1977) or, when temperatures are >32 °C, in ‘refuges’ such as holes in trees.
Torr (1988a) studied the responses of wild tsetse resting in refuges to natural ox odour or a visual
stimulus similar in its size, colour and speed of movement to a passing warthog. These stimuli
were presented at 15 min intervals over a 2 hour period in the late afternoon, the period when
savannah flies are most active (Hargrove and Brady 1992). Only ~15% of tsetse were activated by
natural ox odour compared to ~35% in response to the visual stimulus or 5% in the absence of
any stimuli. Thus most flies left the refuge in the apparent absence of any host stimuli, presumably
in response to their endogenous rhythm of spontaneous activity modulated by nutritional
condition, environmental temperature and falling light intensity (Brady 1972, Brady and Crump
1978, Torr and Hargrove 1999). Vale (1980b) also used electric nets to estimate, amongst other
things, whether tsetse arriving at a stationary host had been resting or ranging when they first
encountered host odour. His results also indicated that most (>80%) tsetse were ranging when
they first responded to host stimuli.

Video studies of tsetse flying in the apparent absence of any host stimuli, and thus presumed to be
‘ranging’ for a host, showed that they flew with a downwind bias (Gibson et al. 1991). This accords
with theoretical considerations of strategies for maximising the probability of encountering host
odours. In typical tsetse habitat, variations in wind direction (Griffiths and Brady 1995, Zollner et al.
2004) are likely to create wide swathes of odour which are more likely to be intercepted by flying
up- or downwind, with the latter being more energetically efficient (Sabelis and Schippers 1984).

Long-range responses

Vale (1977b) showed that tsetse respond to the odours produced by a single ox at ~90 m downwind
and hence we regard ‘long-range responses’ as being those occurring at >10 m downwind of a
host.

Field studies, using video and various arrangements of electric nets, have shown that in response
to host odour, tsetse fly upwind (Gibson and Brady 1988, Torr 1988b, Vale 1974b) at ~6 m/s (Gibson
and Brady 1985), <0.5 m above ground (Torr 1988b). On losing contact with odour, they execute a
reverse turn to bring them back into the odour plume where they turn upwind (Gibson and Brady
1988, Torr 1988b). Wind tunnel studies (Colvin et al. 1989) showed that tsetse use optomotor
anemotaxis (Kennedy 1977) to orientate upwind, and video observation of tsetse in odour plumes
also indicated that they detected wind direction in flight (Gibson and Brady 1988) and altered
their course accordingly.

Griffiths et al. (1995) measured the speed with which marked flies navigated up an odour plume.
Half of those that navigated successfully up the plume did so with a mean speed of 4 m/s, which
accords with the notion that they flew up the entire plume in virtually straight and continuous
flight. The remainder however arrived at the odour source with constant probability for at least
20 minutes after take-off. The proline-based metabolism of tsetse prevents them from flying for
extended periods (Hargrove 1975) and it seems likely that these late arrivals must have landed
at some point en route to the odour source. Direct observation of tsetse in the field showed that,
after losing contact with odour, tsetse land, wait for variations in wind direction to bring the plume
back to them and then take-off upwind when contact is re-established (Bursell 1984a). Bursell
(1984a,b) suggested that tsetse assessed wind direction while on the ground and moved up the
plume in a series of straight flights upwind.

Olfaction in vector-host interactions 269

Steve J. Torr and Philippe Solano

Which way is up?

In the absence of vegetation, packets of odour-laden air are carried downwind in virtually straight
lines (David et al. 1982). Hence, a strategy of flying directly upwind whenever odour is detected
should lead a fly to its host. However, a series of studies in tsetse-infested woodland in Zimbabwe
suggested that this simple view of plume dispersal is not applicable. To visualise the dispersal
of an odour plume, Brady et al. (1989) observed the movement of parcels of smoke. They found
that at the low (<1 m/s) wind speeds typically prevailing when tsetse are active, the parcels did
not travel in straight lines through vegetation. Instead, they changed direction such that even 5
m from the source, 25% of flies travelling upwind would be heading >90° away from host. This
error was even greater at lower wind speeds, greater distances from the source and in areas with
dense vegetation.

However such navigational errors may not be as serious as Brady et al. (1990) suggested. First,
theoretical analysis of host-location strategies suggests that for tsetse starting 100 m downwind
of a host, even a 20-40% upwind bias in flight direction would still enable virtually all flies to arrive
at the source within 300 s (Williams 1994). Second, in nature, the vegetation and local topography
constrains the direction of flight and hence flight directly towards the source may be frequently
impossible. Indeed, there is evidence from Kenya and Zimbabwe that tsetse fly along game paths
and through gaps in bushes (Paynter and Brady 1993, Vale 1998). In such situations, tsetse may
need only to gauge whether to fly up or down a path rather than the precise direction of the host.

The general impression of the long-range orientation of tsetse is therefore not of a precisely
orientated navigation up an odour plume but, rather, a ‘quick-and-dirty’ strategy of fast, mainly
upwind, flight that rarely leads directly to the host (Griffiths et al. 1995). For fast-flying insects such
as tsetse, this so-called ‘biased random walk’ strategy for locating odour sources may be optimal
(Brady et al. 1990, Griffiths et al. 1995).

Short-range responses

Odour-orientated responses bring the fly to the vicinity of the host, but the final location is largely
a response to visual cues. Indeed, tsetse are unable to locate an odour source precisely without
a visual target (Vale 1974b) and flies approaching an odour source can be diverted towards an
odourless visual target (Torr 1989). The colour, shape and size of the target govern the strength
of the response. For example, blue or black objects are more attractive than green-yellow; larger
objects attract more flies than small ones, and horizontal oblongs or squares are more attractive
than vertical oblongs. In general, these same cues also increase landing responses, with the
exception that black elicits a stronger landing response than blue. Host odours appear to have
no effect on close-range orientation but they do increase landing responses (Hargrove 1980, Vale
1974b, Warnes 1995).

Components of host odour

A combination of random screening and chemical analysis of host odours using gas chromatography
linked with electroantennography (GC-EAG) (e.g. Bursell et al. 1988, Hall et al. 1984) has identified
ten chemicals as being important host kairomones in ox odour. The most active of these are:
carbon dioxide, acetone (Vale 1980a), 1-octen-3-ol (henceforth termed octenol; Vale and Hall
1985a), 4-methylphenol and 3-n-propylphenol (Bursell et al. 1988, Hassanali et al. 1986) and
butanone (Torr et al. 1995). When dispensed at the known natural doses, a synthetic blend of

270  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

these is about half as effective as natural ox odour, indicating that other kairomones must also
exist (Hargrove et al. 1995, Torr et al. 1995, 2006).

Behavioural roles of host kairomones

Various field and laboratory studies have assessed whether the components of host odour have
specific behavioural effects.

Activation

Laboratory and field studies both suggest that carbon dioxide has an important activating effect
whereas other components such as acetone or octenol do not (Bursell 1984a, Torr 1988a).

Long-range responses

Wind-tunnel studies have shown that tsetse exhibit optomotor anemotaxis (Kennedy 1977) in
response to carbon dioxide (Colvin et al. 1989), acetone and octenol (Bursell 1984a; Paynter and
Brady 1993). The field evidence for this is however equivocal: video studies suggest that upwind
anemotaxis occurs in response to all the known attractants (Brady and Griffiths 1993, Paynter
and Brady 1993) whereas studies using electric nets (Torr 1988c, 1990, Torr and Mangwiro 1996)
or visual observations (Bursell 1987, Torr 1988c) suggest that upwind flight occurs only in the
presence of carbon dioxide. Field studies have also shown strong synergistic effects between
carbon dioxide and both octenol and acetone (Torr 1990, Vale and Hall 1985a). For instance, Torr
(1990) showed that carbon dioxide and acetone dispensed individually at doses of 1,200 l/h and
50 mg/h, respectively, doubled the catch compared to a 16-fold increase when both odours were
dispensed together. He suggested that the synergy arises because strong upwind flight is elicited
only when acetone and carbon dioxide are both present. Various laboratory studies have indicated
that carbon dioxide, acetone and octenol also elicit orthokinetic and klinokinetic responses
(Paynter and Brady 1993, Warnes 1990) but field evidence for this is poor (Gibson and Brady 1988).

Short-range responses

Torr (1989) showed that the close-range orientation of tsetse towards a host was unaffected by
odour composition, but the final alighting response is affected by the presence of carbon dioxide
(Vale and Hall 1985b) and unidentified chemicals present on the skin (Warnes 1995). However,
visual factors such as colour type and pattern (Brady and Shereni 1988, Doku and Brady 1989,
Gibson 1992, Green 1986), shape (Vale 1974b, Torr 1989) and size (Hargrove 1980) appear to be
more important.

All the above studies of the responses of tsetse to specific host kairomones highlight the
importance of carbon dioxide. Earlier studies (e.g. Brady and Griffiths 1993, Torr 1988b) suggested
that at any great distance downwind of a host, the background noise of carbon dioxide in the
atmosphere would obscure that produced by the host. Thus, it was argued that host carbon
dioxide could not play a role in long range responses and activation – behavioural responses
which are likely to occur up to 90 m downwind of the host. However, detection of carbon dioxide
is not necessarily limited by its presence in the atmosphere but rather, its variability (Zollner et
al. 2004). Using high performance (10 Hz to an accuracy of ±0.1 ppm) infra-red carbon dioxide
analysers, Zollner et al. (2004) showed that carbon dioxide released 4-20 l/min can be detected

Olfaction in vector-host interactions 271

Steve J. Torr and Philippe Solano

up to 64 m downwind. The resolution and sensitivity of the instruments are comparable to those
of an insect, suggesting that, in principle, carbon dioxide is detectable by insects at long range.

The role of olfaction in host selection

Savannah species of tsetse feed on a wide range of animals but some species are fed on more than
others. For instance, meals from pigs (warthog and bushpig) and ungulates (e.g., buffalo, kudu)
are repeatedly identified as being important hosts whereas blood meals from primates, including
humans, are relatively rare (Clausen et al. 1998). Moreover, while ruminants consistently form a
large part of the diet, some relatively common species such as impala are notable by their virtual
absence in the diet.

Various studies have assessed whether host selection is governed by olfactory responses. Vale
(1974b) showed that the numbers of tsetse attracted to different hosts was roughly correlated
with their respective masses. For example, odour from an ox (~450 kg) attracted 5× more than
a goat (~32 kg) whereas the number attracted to an impala and a bushpig (both ~73 kg) were
roughly equal. The only exception to the general correlation between mass and attractiveness
was for humans: the numbers of tsetse attracted to odour from a single human (74 kg) was a
fifth of that attracted to goat odour, and adding human odour to ox odour virtually annulled the
attractiveness of the latter. Human odour depressed not only the numbers of tsetse attracted to
an odour source but also the alighting (Hargrove 1976) and feeding (Vale 1977a) responses. In
further studies, Vale (1979) showed that the repellent effects of humans were due to unidentified
chemicals present in body odour rather than the breath.

The preponderance of Suidae in the diet of tsetse has prompted several detailed studies of the
responses of tsetse to odours from bushpig and warthog. Vale (1974b) reported that many tsetse
landed near the eyes of warthogs and he suggested that this response may be related to secretions
from the pre-orbital gland. Torr (1994) confirmed Vale’s observations but suggested that the
landing response was a visual response to the dark patch produced by eye secretions rather than
an olfactory response. Moreover, adding natural warthog odour to a blend of artificial ox odour (i.e.
carbon dioxide, acetone, octenol and phenols) did not increase the catch significantly, suggesting
that there are no kairomones unique to warthogs (Torr 1994). Studies of the responses of tsetse to
odours from bushpig on the other hand do suggest that this host does may produce unidentified
attractants; placing hessian sacks soiled with residues from a live bushpig increased the catch of
tsetse (Vale et al. 1986a). Vale et al. (1986a) suggested that the unidentified kairomones present
in bushpig residues were not identical to those in ox urine (i.e. various phenols) but no direct
comparisons of bushpig residues and phenols have been undertaken to test this suggestion.

Several papers by Gikonyo and colleagues have assessed the responses of tsetse to waterbuck
(Kobus ellipsiprymnus (Ogilby)), an ungulate which, they argue, is fed on less than might be
expected from its relative abundance. Gikonyo et al. (2000) compared the responses of tsetse to
waterbuck and oxen. Single flies were placed in a cage, which was then applied to a live host, or a
feeding membrane treated with secretions from the host’s coat. There was no significant difference
in initial landing rates – but landing duration, probability of probing and feeding rates were all
lower with waterbuck. They suggested that these differences were due to allomones unique to
waterbuck. Further investigations showed that the chemical profiles of waterbuck and buffalo
differ (Gikonyo et al. 2002), that tsetse show electrophysiological responses to these chemicals
(Gikonyo et al. 2003) and that some of these chemicals elicit upwind flight in olfactometers
(Gikonyo et al. 2003, Mwangi et al. 2008). Curiously, the bioassays of putative allomones did

272  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

not repeat Gikonyo’s (2003) initial bioassay based on landing responses, and hence there is no
evidence that the putative allomones are indeed those responsible for differences in landing
responses. Moreover, the behavioural implications of results from olfactometers with small (<2
m long × 0.24 m wide, Gikonyo et al. 2003) flight chambers are unclear. In the only field study of
waterbuck odours, Madubunyi et al. (1996) found that traps baited with urines from buffalo, cattle
and waterbuck caught similar numbers of tsetse, and chemical analyses showed that these urines
contained 4-methylphenol and 3-n-propylphenol in similar ratios. Thus there are no field data to
confirm the hypothesis that waterbuck odours are repellent to tsetse.

On the other hand, there are examples of host choice being influenced by responses to cues
other than host odours. For example, the low feeding rates of tsetse on impala and goat seems
to be explained by their inherently high rate of grooming, which deters tsetse from landing and
engorging; an impala attracted fewer flies than an ox, consistent with its smaller size, but no flies
fed on it whereas 35% of tsetse approaching an ox fed to completion (Vale 1977a). Visual stimuli
may also play a role in host selection: the low number of feeds from zebra seems to be due, in part
at least, to the cryptic effect of their stripes (Gibson 1992).

The advent of DNA-based tools for identifying the individual specific sources of blood meals
allows us to determinate whether some individuals within a herd are bitten more than others (Torr
et al. 2001). The results show that for tsetse feeding on cattle there is a marked preponderance
of meals from larger/older individuals within a herd (Torr et al. 2007b). This bias seems to be
due to younger cattle having higher rates of defensive behaviour (Torr and Mangwiro 2000) and
producing low doses of kairomones, particularly carbon dioxide and phenols (Torr et al. 2006).

Repellents

The search for attractants and analyses of odours from hosts has led, almost incidentally, to the
identification of a number of chemicals naturally present in host odours that reduce the catch of
tsetse from traps (Vale 1979, 1980a, Vale et al. 1988a). These findings might suggest that these
chemicals may play a role in host selection (Torr et al. 1996). However, results with traps do not
necessarily indicate that these chemicals reduce the numbers of tsetse attracted to and/or feeding
on hosts in nature. Thus, while 2-methoxyphenol reduces the numbers of G. pallidipes caught by
odour-baited traps by ~85%, it has no significant effect on the numbers attracted to and feeding
on an ox (Torr et al. 1996). More recently, Saini and Hassanali (2007) showed that 4-methyl-2-
methoxyphenol reduced the numbers of tsetse feeding on cattle by 95%. However, this chemical
has not been detected in natural host odours.

Intraspecific variation in olfactory responses

Much of the foregoing description of olfactory responses relates primarily to G. pallidipes and G. m.
morsitans populations in Zimbabwe and Kenya. Other species and populations of Morsitans-group
flies have been less intensively studied; nonetheless, there is evidence that there are similarities
and differences between species and populations. G. longipalpis Wiedemann seems to be the
species most similar to G. pallidipes being responsive to phenols, acetone and octenol (Späth
1995). The G. morsitans subspecies and G. swynnertoni are responsive to acetone and octenol
but show little if any response to the phenols (G. m. morsitans: Vale 1980a, Vale et al. 1988a; G.
m. submorsitans: Mérot et al. 1988b, Politzar and Merot 1984; G. m. centralis Machado: Modo
1999; G. swynnertoni: Brightwell and Dransfield 1997, Ndegwa and Mihok 1999). Glossina austeni
seems to be the least responsive species, with only carbon dioxide being an effective kairomone

Olfaction in vector-host interactions 273

Steve J. Torr and Philippe Solano

(Kappmeier-Green 2001). There is evidence of intra-specific variations in the olfactory responses

of Morsitans-group species. Studies of G. pallidipes in Somalia showed that acetone or octenol
were only effective in the presence of 4-methylphenol and 3-n-propylphenol, and baiting a target
with the odour from an ox only doubled the catch. By contrast, acetone, octenol and phenols are
effective on their own for G. pallidipes in Zimbabwe (Vale 1980a, Vale and Hall 1985a) and Kenya
(Baylis and Nambiro 1993), and odour from a single ox increases the catch from a target ~10×
(Vale 1974b).

Fusca group species

Fusca group species play a less important role in the transmission of trypanosomes and hence
their olfactory responses have been studied less than those of the Morsitans group. Nonetheless a
few species are locally important as vectors of disease and/or are sympatric with Morsitans group
species and hence we do have some knowledge of their responses. The impression from these
few studies is that the group displays somewhat similar responses to Morsitans group species.
For instance, baiting traps with acetone, octenol or 4-methylphenol all increase the catches of G.
longipennis Corti (Baylis and Nambiro 1993, Kyorku et al. 1990), G. brevipalpis Newstead (Kappmeier
and Nevill 1999). Studies of G. medicorum Austen in west Africa suggest that 3-methylphenol,
4-methylphenol and octenol are effective, although acetone is not (Späth 1995).

West side story: olfactory responses of riverine flies

Ecology of riverine flies

Riverine species of tsetse show marked associations with particular ecotypes, which, in turn, have
important implications for their interactions with hosts. It is therefore pertinent to summarise
briefly the main features of the important Palpalis group species (reviewed in Solano et al. 2009).

Distribution

Glossina tachinoides Westwood

This species generally occurs in narrow thickets along river courses in thickets consisting of
Mimosa pigra and Morelia senegalensis (Sudanese ecotype), swampy forests of Mitragyna inermis
and in the ‘open’ gallery forests consisting predominantly of Syzygium guineense, Cola laurifolia
and Pterocarpus santalinoides (Sudano-Guinean ecotype). It is often found with G. p. gambiensis
Vanderplank in riverine habitats, and occasionally with G. p. palpalis Robineau-Desvoidy in forested
areas. The distribution of G. tachinoides is largely between 12° - 8°N and 2°W - 21°E, but there are
also a few isolated populations found as far east as Ethiopia.

Glossina palpalis

The distribution of this species extends along the Atlantic coast from Senegal to Angola. Glossina
p. palpalis is found closer to the coast within degraded forest areas and wet savannahs, whereas
G. p. gambiensis is found in the riparian networks within drier woodland savannahs and ‘niayes –
lowland areas associated with oil palm plantations in north-west Senegal. Thus G. p. gambiensis
is associated with the Guinean and Sudano-Guinean ecotypes extending from Senegal to Benin,
while G. p. palpalis occurs from Liberia to Angola. Both subspecies can be found in mangrove
areas along the coast, and in areas ranging from peridomestic areas, such as ‘sacred woodlands’

274  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

or mango plantations in savannah areas and coffee/cocoa plantations in forested areas, to sites
within major urban centres such as Abidjan and Dakar. This close association with humans makes
this species a particularly important vector of human and animal trypanosomiases.

Glossina fuscipes Newstead

This species is associated with the riverine and lacustrine systems of the Congo basin and Lakes
Victoria and Tanganyika, particularly vegetated areas associated with rivers, lakes and the forest-
savannah mosaics. G. fuscipes is an important vector of human trypanosomiasis in Central Africa
(DRC, Congo Brazzaville) and some countries of East Africa (Uganda, Kenya), as well as animal
trypanosomiasis in, for example, the Central African Republic.

Seasonal effects on local distribution

The local distribution of Palpalis group species varies markedly between seasons, with flies being
closer to water bodies during the drier months. Within the riverine system, flies disperse linearly
along the river, oscillating up- and down stream over relatively short distances (200-300 m). In the
dry season, displacements are largely within the river whereas in the wet season they are more
likely to disperse away from the river, crossing adjacent savannah woodlands and moving from
one hydrographic network to another. In the hot season, some individuals can fly long distances
(17 km in three days for G. tachinoides; 22 km in five days for G. p. gambiensis) along rivers.

Host range

The diet of Palpalis group species seems more variable than that of Morsitans group flies, being
considered more ‘opportunistic’ in their diet (Sané et al. 2000; Simo et al. 2008 ). For instance,
in Côte d’Ivoire, G. palpalis feeds mainly on humans in settlements in the absence of pigs, but
takes 75% of its blood meals on pigs when they are abundant and in plantations, it feeds at
equal rates on bushbuck and humans. In the hot season, when humans and small ungulates are
concentrated around water bodies, the fly takes 40% of its blood meals from bushbuck and 35-
55% from humans. In the cold season on the other hand, it feeds mainly on reptiles (54-67%).

Tsetse fly populations (G. palpalis, G. tachinoides) that inhabit atypical habitats around villages of
wet areas can feed exclusively on pigs bred by villagers. There is some evidence that individual G.
p. gambiensis may learn to feed on a particular species in some circumstance (Bouyer et al. 2007a).

Olfactory responses of Palpalis group flies

In general, Palpalis group species show markedly weaker responses to host odours. This has meant
that researchers have seldom published their findings and, hence, previous reviewers (e.g. Gibson
and Torr 1999, Green 1994) have had to rely on unpublished data and personal communications.
At the time of writing, published data on the responses of Palpalis group flies is still confined
largely to just two species: G. tachinoides and G. fuscipes.

Glossina tachinoides

Merot et al. (1986), in Burkina Faso, compared the numbers attracted to sources of natural odour
from a man (60 kg), a pig (60 kg) and a cow (150 kg) following Vale’s (1974b) protocol of placing the
hosts in a ventilated pit. Tsetse were captured using an electrocuting target placed at the odour

Olfaction in vector-host interactions 275

Steve J. Torr and Philippe Solano

source. The results showed that these odours increased the catch only 1.2×. Even placing four
cattle in the pit to increase the dose of odour produced only a 1.8× increase in catch. In accordance
with these relatively modest increases in catch, carbon dioxide dispensed at 3 l/min increased the
catch only 1.2×. In a subsequent study, Galey et al. (1986) showed that baiting traps with carbon
dioxide (0.5-20 l/min) increased the catch 1.2-3.3×. Filledier and Mérot (1989b) showed that urine
from a Baoule taurine tripled the catch of G. tachinoides.

Merot et al. (1988a) undertook a series of experiments to determine whether the kairomones
known to attract G. pallidipes were also effective for G. tachinoides. Their results (Mérot et al. 1988a)
showed that a blend of carbon dioxide (1.5 l/min), acetone, octenol and phenols doubled the
catch compared to a 1.3× increase produced by the odour from three cattle. Studies with traps
suggested that (1) a combination of phenol(s) and octenol doubled the catch, but (2) these odours
were ineffective on their own, and (3) their effect was antagonised by acetone. Filledier and Mérot
(1989a) showed that baiting traps with a blend of 3-methylphenol and octenol doubled (1.3-
2.5×) the catch of tsetse. Späth (1995) suggested a blend of 3-methylphenol, 4-methylphenol and
acetone were more effective but his results do not show that the addition of 4-methylphenol was
significantly more effective than the combination of octenol and 3-methylphenol alone.

Blood meal analyses of G. tachinoides show a preponderance of meals from hippopotamus,

ruminants (mainly bushbuck) and monitor lizards (Clausen et al. 1998). Accordingly, Späth (1997)
assessed the effect of baiting traps with odours from these hosts. Placing a caged monitor lizard
adjacent to a trap had no effect but hexane skin washings from the lizard increased the catch
1.3×. Skin washings from a warthog and the phenolic fraction of bushbuck urine also increased
the catch significantly 1.4× and 1.8×, respectively. No studies have been undertaken to confirm
these findings or to test whether these products contain novel attractants.

G. fuscipes fuscipes Newstead

Like other Palpalis group species, G. f. fuscipes is associated with riverine and lacustrine habitats.
Studies of their diet consistently show a high proportion of meals from reptiles (Clausen et al.
1998), particularly monitor lizards (Varanus niloticus niloticus Linnaeus), and this has prompted
several studies of the responses of G. f. fuscipes to lizards.

In the Central African Republic, Gouteux et al. (1995) showed that placing a caged monitor lizard
adjacent to a trap increased the catch of G. f. fuscipes significantly. The cages were hidden by
covering them with leaves and hence Gouteux et al. argued that the catch increases were due to
olfactory rather than visual stimuli. In Kenya, Mohamed-Ahmed (1998), used various arrangements
of traps and electric nets to study the responses of G. f. fuscipes to monitor lizards. Placing a
cage containing three lizards adjacent to an electric net doubled the catch of tsetse, although
the increase was significant only for females. The lizards within the cage were visible and hence
the increase could be due to either visual or olfactory responses. Accordingly, Mohamed-Ahmed
compared the numbers of tsetse landing on a black cylinder (1.5×0.1 m) with or without a single
lizard (~5 kg) inside. The cylinder was perforated with small (13 mm dia) holes along its length to
allow odours from the lizard to escape. An electric grid that covered the surface of the cylinder
caught flies as they landed. The results showed that when the lizard was inside, twice as many
tsetse landed, but the increases were not significant for either sex. Baiting cylinders with lizard
urine increased the catch of tsetse attracted to the vicinity of the cylinder but the increase (2×)
was only significant for females. By contrast, baiting traps with lizard urine increased significantly
the catch of males (~1.4×) but not females.

276  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

Glossina f. fuscipes is an important vector of Trypanosoma spp. (Waiswa et al. 2006, Welburn et al.
2006) and hence, not surprisingly, meals from primates – including presumably humans – and
ungulates are also relatively common (Clausen et al. 1998). However, despite feeding on cattle, G. f.
fuscipes does not appear responsive to the tsetse attractants present naturally in ox odour; baiting
traps with acetone, octenol or phenols, dispensed either on their own or in blends known to be
effective for other species of tsetse, did not have any significant effects (Mwangelwa et al. 1995).
Mohamed-Ahmed and Mihok (1999) studied the numbers of G. f. fuscipes attracted to traps baited
with carbon dioxide. In an experiment conducted in a ‘linear forest’, baiting a trap with carbon
dioxide (5 l/min) had no significant effect on the catch of tsetse. However, when the experiment
was repeated in a ‘dense forest’, carbon dioxide (2.5 l/min) doubled the catch of females but had
no significant effect for males. In a third experiment, conducted in ‘dense forest’, they used electric
nets to gauge the numbers of tsetse in the vicinity of a trap with or without carbon dioxide (2.5 l/
min). Their results show that carbon dioxide doubled the total (i.e. catch from trap+electric nets)
catch of females but had, yet again, no significant effect for males. The similar effects obtained
with either (1) traps alone or (2) traps+electric nets suggest that the main effect of carbon dioxide
was to increase the numbers of tsetse attracted to the odour source rather than increasing trap
efficiency (Vale and Hargrove 1979). Mohamed-Ahmed and Mihok (1999) argued that the absence
of any effect for carbon dioxide in the linear forest was due to local topography and vegetation;
for example, if tsetse were confined to the linear forest, which was only 5-10 m wide, then the area
from which they could be recruited by host odours would be necessarily restricted. Nonetheless,
even in habitats where carbon dioxide or lizard odours were effective, it is striking that the overall
effects are relatively small (~2×) and confined to females only.

Other Palpalis group species

Published data on other Palpalis group species are extremely scant. In Liberia, Cheke and Garms
(1988) showed that acetone and octenol doubled the catch of G. p. palpalis from traps but these
odours had no significant effect when dispensed in combination with a blend of phenols. Finally,
in Congo-Brazzaville, Frezil and Carnevale (1976) showed that baiting a trap with carbon dioxide
increased the catch of G. f. quanzensis Pires ~40×.

Current research on Palpalis group species

Currently, a major research programme supported by the Bill and Melinda Gates Foundation and
the EC, amongst others, is assessing the olfactory responses of five Palpalis group species: G. p.
gambiensis, G. p. palpalis, G. f. fuscipes, G. f. quanzensis and G. tachinoides. Preliminary findings
(Omolo et al. 2009, Rayaisse et al. 2010) suggest that these species show moderate but significant
responses to host odours. For instance, studies in Burkina Faso have shown that there is a
significant ~1.5× increase in the numbers of G. p. gambiensis attracted to a source of ox odour,
and that a blend of acetone, 4-methylphenol, 3-n-propylphenol and octenol can double the catch
from traps (IAEA 2003, Rayaisse et al. 2010). Studies of G. fuscipes species (Omolo et al. 2009)
confirm that G. f. fuscipes responds to odours from lizard but not to those from cattle, humans or
pigs whereas G. f. quanzensis appears to show significant responses to pig odours. The absence
of any significant response by G. f. fuscipes to cattle odour is particularly intriguing since some
experiments showed that even the odour from four cattle, which would have contained ~8 l/min
of carbon dioxide, was ineffective. Previous studies (Mohamed-Ahmed and Mihok 1999) have
also reported a variable response to carbon dioxide (for details see above). For the other Palpalis
group species, carbon dioxide does appear to be effective but not all of the olfactory response is
accounted for by this kairomone. Overall, Palpalis group flies show a significant response to host

Olfaction in vector-host interactions 277

Steve J. Torr and Philippe Solano

odours, but the effect is small in comparison to, say, G. pallidipes. Research is currently in progress
to identify the kairomones present in lizard, pig and cattle odours.

Attacking tsetse with attractants

Not surprisingly, the practical use of attractants for controlling and monitoring tsetse has been
largely confined to operations directed against the Morsitans group. Ironically, however the
revival of bait methods to control tsetse was led by Claude Laveissière and co-workers in West
Africa who used unbaited traps to control Palpalis group species in sleeping sickness foci of Ivory
Coast (see Table 2 in Green 1994). These operations were initiated in the late 1970s, some four
years before the first large-scale use of odour-baited traps and targets (see Table 1 in Green 1994).
Nonetheless, for the purposes of this review we focus mainly on the use of odour baits against
Morsitans group species.

Use of odour baits to control Morsitans group species

The first successful application of odour-baited traps and targets to control tsetse was undertaken
in Zimbabwe where traps and insecticide-treated targets baited with synthetic host odours were
used to eliminate G. m. morsitans and G. pallidipes from an island (5 km2) in Lake Kariba (Vale et
al. 1986b). Following on from this success, large-scale use of the technology began in 1984, with
the deployment of insecticide-treated targets over an area of ~1000 km2 in the Zambezi valley
of Zimbabwe. At the commencement of the operation in 1984, these targets were baited with
acetone and octenol and these devices, deployed at a density of 4/km2 eliminated G. m. morsitans
and reduced the population of G. pallidipes by >99.9% (Vale et al. 1988b). These early successes led
to a rapid adoption of bait methods for tsetse control in Zimbabwe (Figure 1) and elsewhere (e.g.
Zambia: Willemse 1991; Kenya: Dransfield et al. 1990). By the late 1980s, phenols were identified
as attractants for G. pallidipes and since then a blend of acetone (~100 mg/h), octenol (~0.5 mg/h),
4-methylphenol (~1 mg/h) and 3-n-propylphenol (~0.1 mg/h) has become the standard blend
for traps and targets used to monitor and/or control G. pallidipes (Torr et al. 1997), while blends
of acetone and octenol have been used in operations against G. morsitans spp. (Kgori et al. 2006,
Willemse 1991). For further details of control operations based on the use of odour-baited traps
and targets, see reviews by Green (1994), Hargrove (2003) and Torr et al. (2005).

The relatively low cost of odour-baited traps and targets, and the possibility that local people
could make and deploy them, made the use of such devices appealing to local communities
and, more especially, the government institutions, donor agencies and NGOs that advised them.
Accordingly, the 1990s saw a proliferation of community-based interventions against tsetse. Over
the last decade, artificial baits have been somewhat eclipsed by the increasing use of insecticide-
treated cattle, as illustrated by the case of Zimbabwe (Figure 1). Insecticide-treated cattle are
identical to insecticide-treated targets in their effect on a tsetse population but they have several
practical advantages including being: cheaper, less susceptible to theft and vandalism and
effective against tick-borne diseases (Eisler et al. 2003). A further advantage of insecticide-treated
cattle over artificial baits is that livestock keepers are more likely to treat their cattle than use traps
and targets which are regarded as ‘public goods’ (Kamuanga et al. 2001).

Recent studies of the behaviour of tsetse feeding on cattle (Bouyer et al. 2007b; Torr et al. 2007a,b)
showed that by applying insecticide to the legs and belly of the larger cattle within a herd, the
cost of trypanosomiasis control can be reduced still further to ~$2 per head per year (Shaw et al.

278  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

50,000 12,000
Ground spraying
Sequential Aerosol Technique (SAT)
Targets 10,000
40,000 Insecticide-Treated Cattle (ITC)

8,000
Square kilometres

30,000

Cases/year
6,000

20,000
4,000

10,000
2,000

0 0
1982 1984 1986 1988 1990 1992 1994 1996 1998
Year

Figure 1. Areas of Zimbabwe treated annually by various methods of tsetse control (vertical bars) and annual
number of trypanosomiasis cases (solid line) (redrawn from data presented in Vale and Torr 2004 and Torr et al.
2005).

2006). Hence in areas where cattle are already present it seems that, in principle, the use of natural
rather than artificial baits will be the method of choice.

Discussion
Variation in the olfactory responses of tsetse

East vs. West

Intraspecific comparisons of the olfactory responses of different species in different

zoogeographical zones are hampered by the range of experimental methods used by various
researchers. To overcome this, let us consider only experiments where an electrocuting target was
baited with odour from a single adult ox or cow, thereby ignoring the potentially confounding
effects of using different trap designs, and the effects of host odour on trap efficiency. The results
(Figure 2) show that the large responses to host odours are neither exclusive to Morsitans group
species (cf. G. longipennis, a Fusca group species) nor do all Morsitans group species show strong
responses to host odour (cf. G. austeni, G. m. submorsitans, G. pallidipes in Somalia). On the other
hand, published data, and studies currently being undertaken by us in West and Central Africa,
do indicate that no Palpalis group species show the strong olfactory responses exhibited by, say,
G. pallidipes in Zimbabwe.

Olfaction in vector-host interactions 279

Steve J. Torr and Philippe Solano

G. pallidipes (ZW)
G. pallidipes (SOM)
G. m. morsitans (ZW) Morsitans
G. m. submorsitans (BF)
G. austeni (RSA)
Species (country)

G. brevipalpis (RSA)
G. longipennis (K) Fusca

G. tachinoides (BF)
G. f. fuscipes (K)
G. f. quanzensis (DRC) Palpalis
G. p. gambiensis (BF)
G. p. palpalis (CI)
0 2 4 6 8 10 12 14

Catch index

Figure 2. Catch of tsetse from an electrocuting target baited with odour from 1-2 cattle. Catches are expressed
as a proportion (‘Catch index’) of that from an unbaited target. An index of 1 (dotted vertical line) indicates that
the odour has no significant effect; solid bars indicate indices that are significantly (P<0.05) different from unity.
Indices for each species are derived from: Vale 1974b (G. m. morsitans & G. pallidipes in Zimbabwe (ZW)); Mérot
et al. 1986 (G. m. submorsitans & G. tachinoides in Burkina Faso (BF)); Torr et al. 1989 (G. pallidipes in Somalia
(SOM)); Baylis and Nambiro 1993 (G. longipennis in Kenya (K); Kappmeier-Green 2001 (G. brevipalpis and G.
austeni in South Africa (RSA)); Omolo et al. 2009 (G. fuscipes spp and G. f. quanzensis in Kenya (K) and the
Democratic Republic of Congo (DRC)); Rayaisse et al. 2009 (G. p. gambiensis and G. tachinoides in Burkina Faso
(BF)); G. p. palpalis in Côte d’Ivoire (CI)).

The current model of the odour-orientated behaviour of tsetse, based largely on the responses
of G. pallidipes in Zimbabwe and Kenya, may not be as widely applicable as is assumed. We
suggest that there is an urgent need to analyse the basic host-orientated responses of these
‘less responsive’ species. For instance, the current paradigm for the long-range orientation of
tsetse is based on upwind anemotactic responses to host odours. However, might klinokinetic
or orthokinetic responses (Warnes 1990) rather than anemotactic responses be more important
for Palpalis group species? Might host kairomones modulate the visual responses of tsetse (Torr
1989)? Having established what species do not do, we now need to find out what they do do.
A better understanding of the behaviour of these species could provide improved means of
controlling tsetse in much the same way that Vale’s discovery of the olfactory responses of tsetse
in Zimbabwe led to a revolution in tsetse control methods there (Vale 1993).

Intraspecific genetic variation

The differences in the olfactory responses of G. pallidipes from Zimbabwe and Somalia may be due
to genetic differences between these populations. Virtually no information is currently available
on relationships between genetic structure and the olfactory responses of tsetse populations.
The few studies dealing with tsetse population genetics in East Africa have focused on genetic
differentiation measures between geographically distant populations (see Gooding and Krafsur

280  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

2005 for review). In West Africa, several studies have focused on within-population structure (e.g.
Ravel et al. 2007, Solano et al. 2000). These studies suggest that, on a microgeographic scale, G.
palpalis populations are composed of several ‘clusters’ or subgroups, genetically differentiated
from each other with separate geographical and host ranges (as illustrated in Figure 3). Such a
population structure might explain, in part, why Palpalis group flies have variable responses to
host odours. For example, a trap baited with reptile kairomones might only be effective for the
sub-population that feeds on lizards and not those that feed on, say, humans or pigs. Further
research is needed to assess this possibility.

Tsetse infected with

T. brucei s.l., T. congo
- High and significant heterozygote deficiency
Cl1 (WAHLUND EFFECT)
- Infection by different trypanosome groups
- Different bloodmeal sources

Village/
encampment
Cl2

High genetic
differentiation

Tsetse infected with

T. brucei gambiense
Cl3

Tsetse infected with Tsetse trap

Reptile trypanosomes Tsetse dispersal
Figure 3. Representation of the Wahlund effect with a population of G. palpalis, as suggested by genetic results
obtained using microsatellite DNA markers in West Africa (Solano et al. 2000, Ravel et al. 2007). Tsetse caught in
different traps in the environs of a single ‘village’ comprise flies from different clusters (Cl 1 to Cl 3) that are spatially
and genetically distinct. In this example, Cl 1 is associated with pigs infected by T. brucei and T. congolense, Cl
2 is the vector of pathogenic trypanosomes causing sleeping sickness and Cl 3 feeds mostly on monitor lizards
infected with non-pathogenic trypanosomes.

Future directions

Attractants

The past decade has seen a decline in research aimed at developing improved baits for tsetse,
perhaps because there was a general perception that all the important attractants had been
identified for savannah species and there was none to find for riverine flies. However, it is clear that
neither view is correct: unidentified attractants for savannah species remain to be identified (Torr
et al. 1995, 2006) and, more excitingly, Palpalis group flies do respond to host kairomones other
than carbon dioxide (Filledier and Mérot 1989a; Mohamed-Ahmed 1998; Rayaisse et al. 2010).

Olfaction in vector-host interactions 281

Steve J. Torr and Philippe Solano

Would the identification of novel attractants have an impact on tsetse control? The basic studies
of the responses of some species to host odours might suggest that host odours will be of little
practical importance. For instance, the modest (~1.3×) increases in catch for G. pallidipes in Somalia
or both G. m. submorsitans and G. tachinoides in Burkina Faso produced by baiting a target with ox
odour (Figure 2) might have suggested that the effect is due largely to carbon dioxide only and
hence there are no useful attractants to identify. In practice however, traps baited with blends of
acetone, octenol and phenols have been shown to produce 2-4× increases in catch (Filledier and
Mérot 1989a, Politzar and Merot 1984, Torr et al. 1989). We should be wary therefore of assuming
that the modest catch increases obtained with natural odours means that we cannot do better
with synthetic blends.

The identification of odours that improve the performance of traps and targets, say, four-fold could
have important implications for the control of tsetse, especially those riverine species, which play
an important role in the transmission of T. brucei spp.. For the control of Palpalis group species,
baits must currently be deployed at densities of 30 targets/km2 or more, compared to 4 targets/
km2 for savannah species. These higher densities make the method prohibitively expensive for
community-based schemes; Shaw et al. (2006) estimated that the field costs of using artifical
baits deployed at a conservative density of just 10 targets/km2 to control G. f. fuscipes in Uganda
was $706/km2 compared to the $283/km2 needed for 4 targets/km2 to control G. pallidipes. The
identification of cheap attractants that improved the performance of baits 2-4× might reasonably
be expected to make the costs of controlling riverine and savannah species comparable.

Moreover, while targets must be deployed over large (>500 km2) areas to control savannah species,
experience with the use of baits against riverine species shows that relatively small- scale (<100
km2) interventions can have a major impact on tsetse and sleeping sickness (see examples in
Green 1994, and Laveissière and Grébaut 2003). In general, their more restricted distribution and
lower densities make populations of Palpalis group species more amenable to control. In addition,
the lower rates of infection with T. brucei s.l., compared to T. vivax or T. congolense, mean that a
given level of tsetse control has a greater impact on human than animal trypanosomiasis (Rogers
1988). Thus if better baits for riverine flies can be developed, the prospects for their wider use,
especially to control sleeping sickness, seem good.

While the prospects of halving the costs of baits to control riverine flies seem reasonable, the cost
(~$300/km-2) will still be much greater than that ($12/km-2) of using insecticide-treated cattle
(Shaw et al. 2006). There are no examples of large-scale uptake of traps and targets by communities
to control tsetse, whereas livestock keepers are prepared to pay the cost (~2 cents/animal/month)
of treating their cattle with insecticide to control ticks and tsetse, as evidenced by the sale of
~37,000 doses to Ugandan farmers (I. Maudlin, personal communication) over two months in
2008. Given the decline in government and donor support for tsetse control, and the greater role
assumed by livestock keepers and communities living in tsetse-infested areas, it seems likely that
the insecticide-treated cattle rather than artificial baits will be the mainstay of tsetse control.
However, while insecticide-treated cattle offer a particularly cheap and sustainable method of
tsetse, we suggest that odour-baited targets and traps will continue to play an important role. First,
cattle are not evenly distributed and thus to achieve success it is frequently necessary to deploy
artificial baits in areas where cattle are absent or where other hosts are abundant (Hargrove et al.
2003, Warnes et al. 1999). Second, while insecticide-treated cattle may be effective for controlling
vectors of T. b. rhodesiense, this approach may be less applicable for vectors of T. b. gambiense,
where cattle play a less important role in both disease epidemiology and tsetse diet. Third, all
tsetse control operations are bedevilled by tsetse re-invading from either neighbouring areas of

282  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

infestation, or pockets of survivors, and the only effective method of preventing reinvasion from
these is to deploy a barrier of insecticide-treated targets (Cuisance and Politzar 1983, Hargrove
1993, Kgori et al. 2006, Muzari and Hargrove 1996). Finally, all control operations rely on surveys
to assess the extent of the tsetse infestation and the impact of control operations; odour-baited
traps play a key role in the success of these surveys, including current control operations in West
Africa (I. Sidibe, personal communication).

Repellents

In some respects it is surprising that more research has not been conducted on identifying
repellents for tsetse. We know that standard repellents such as DEET are not effective against
savannah species of tsetse (Vale 1979; Torr et al. 1996) but potent, unidentified, repellents are
present in human odours (Vale 1974b). Presumably, the methods that were used to identify
attractants for tsetse could also be used for repellents.

Torr et al. (1996) suggested that repellents would be most effective in areas where there are low
densities of tsetse and low infection rates in the vector populations, conditions that are typical
of sleeping sickness foci in West Africa. It would therefore be pertinent to assess the responses
of Palpalis group species to chemicals (e.g. 4-methyl-2-methoxyphenol) and natural odours (e.g.
human body odour) which appear to be repellent for savannah flies.

The application of science

The last 40 years has seen a tumultuous body of research on the olfactory responses of tsetse
to their hosts, resulting in new methods of vector control and surveillance. However, measured
in terms of impact on vector populations and disease incidence, the impact of this research has
been disappointing: tsetse are, roughly, as widespread as ever, sleeping sickness still kills ~40,000
people a year and the alleviation of animal trypanosomiasis is more dependent on drugs than
vector control.

Baits – artificial and natural – offer the prospect of playing a greater role in the future, but it is
clear that we need to understand and promote the correct use of these technologies; too often
they have been used inappropriately and hence had little or no sustained impact (Torr et al. 2005).
The development of web-based systems (e.g. www.tsetse.org; www.sleeping-sickness.idr.fr) to
promote the correct use of baits goes some way to addressing this need.

More significantly, the formation of the Pan-African Tsetse and Trypanosomiasis Eradication
Campaign (PATTEC) signals a new determination by African governments and leaders to support
greater efforts against tsetse. The emergence of cheap and effective methods of vector control
and surveillance, based largely on host-orientated behaviour, seem set to play an important role
in this initiative.

Acknowledgements

We thank Professors Glyn Vale and John Hargrove and Dr Gay Gibson for helpful comments on
earlier drafts of this chapter. We also thank our colleagues Drs. Johan Esterhuizen and Maurice
Omolo and messrs. Jean-Baptiste Rayaisse and Inaki Tirados for helpful discussions.

Olfaction in vector-host interactions 283

Steve J. Torr and Philippe Solano

References

Baylis M and Nambiro CO (1993) The responses of Glossina pallidipes and G. longipennis (Diptera: Glossinidae) to odour-
baited traps and targets at Galana Ranch, south-eastern Kenya. Bull Entomol Res 83: 145-151.
Bouyer J, Pruvot M, Bengaly Z, Guerin PM and Lancelot R (2007a) Learning influences host choice in tsetse. Biol Letters
3: 113-U111.
Bouyer J, Stachurski F, Kabore I, Bauer B and Lancelot R (2007b) Tsetse control in cattle from pyrethroid footbaths. Prev
Vet Med 78: 223-238.
Brady J (1972) Spontaneous, circadian components of tsetse fly activity. J Insect Physiol 18: 471-484.
Brady J and Crump AJ (1978) Control of circadian activity rhythems in tsetse flies - environment of physiological clock?
Physiol Entomol 3: 177-190.
Brady J and Griffiths N (1993) Upwind flight responses of tsetse flies (Glossina spp) (Diptera: Glossinidae) to acetone,
octenol and phenols in nature - a video study. Bull Entomol Res 83: 329-333.
Brady J and Shereni W (1988) Landing responses of the tsetse fly Glossina morsitans morsitans Westwood and the stable
fly Stomoxys calcitrans (L) (Diptera, Glossinidae and Muscidae) to black-and-white patterns - a laboratory study.
Bull Entomol Res 78: 301-311.
Brady J, Gibson G and Packer MJ (1989) Odour movement, wind direction, and the problem of host-finding by tsetse
flies. Physiol Entomol 14: 369-380.
Brady J, Packer MJ and Gibson B (1990) Odour plume shape and host finding by tsetse. Insect Sci Appl 11: 377-384.
Brightwell R and Dransfield R (1997) Odour attractants for tsetse: Glossina austeni, G. brevipalpis and G. swynnertoni.
Med Vet Entomol 11: 297-299.
Bursell E (1987) The effect of wind-borne odours on the direction of flight in tsetse flies. Physiol Entomol 12: 149-156.
Bursell E (1984a) Effects of host odour on the behaviour of tsetse. Insect Sci Appl 5: 345-349.
Bursell E (1984b) Observations on the orientation of tsetse flies (Glossina pallidipes) to wind-borne odours. Physiol
Entomol 9: 133-137.
Bursell E and Taylor P (1980) An energy budget for Glossina (Diptera: Glossinidae). Bull Entomol Res 70: 187-196.
Bursell E, Gough AJE, Beevor PS, Cork A, Hall DR and Vale GA (1988) Identification of components of cattle urine
attractive to tsetse flies, Glossina spp (Diptera: Glossinidae) Bull Entomol Res 78: 281-291.
Buxton PA (1955) The natural history of tsetse flies - an account of the biology of the genus Glossina (Diptera). H.K.
Lewis, London, UK. 816 pp.
Cheke RA and Garms R (1988) Trials of compounds to enhance trap catches of Glossina palpalis palpalis in Liberia. Med
Vet Entomol 2: 199-200.
Clausen PH, Adeyemi I, Bauer B, Breloeer M, Salchow F and Staak C (1998) Host preferences of tsetse (Diptera:
Glossinidae) based on bloodmeal identifications. Med Vet Entomol 12: 169-180.
Colvin J, Brady J and Gibson G (1989) Visually-guided, upwind turning behaviour of free-flying tsetse flies in odour-
laden wind: a wind-tunnel study. Physiol Entomol 14: 31-39.
Courtin F, Jamonneau V, Duvallet G, Garcia A, Coulibaly B, Doumenge JP, Cuny G and Solano P (2008) Sleeping sickness
in West Africa (1906-2006): changes in spatial repartition and lessons from the past. Trop Med Int Health 13: 334-
344.
Cuisance D and Politzar H (1983) Etude sur l’efficacité contre Glossina palpalis gambiensis et G. tachinoides de barrières
constituées d’écrans ou de pièges biconiques imprégnés de DDT, de deltaméthrine, ou de dieldrine. Rev Elev Med
Vet Pays Trop 36: 159-168.
David CT, Kennedy JS, Ludlow AR, Perry J.N. and Wall C (1982) A re-appraisal of insect flight towards a point source of
wind-borne odour. J Chem Ecol 8: 1207-1215.
Doku C and Brady J (1989) Landing site preferences of Glossina morsitans morsitans Westwood (Diptera: Glossinidae) in
the laboratory: avoidance of horizontal features. Bull Entomol Res 79: 521-528.
Dransfield RD, Brightwell R, Kyorku C and Williams B (1990) Control of tsetse fly (Diptera: Glossinidae) populations using
traps at Nguruman, south-west Kenya. Bull Entomol Res 80: 265-276.

284  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

Eisler MC, Torr SJ, Coleman PG, Machila N and Morton JF (2003) Integrated control of vector-borne diseases of livestock
- pyrethroids: panacea or poison? Trends Parasitol 19: 341-345.
Filledier J and Mérot P (1989a) Pouvoir attractif de l’association m-crésol, 1-octen-3-ol dans un type de diffuseur
pratique pour Glossina tachinoides au Burkina Faso. Rev Elev Med Vet Pays Trop 42: 541-544.
Filledier J and Mérot PR (1989b) Etude de l’attractivité de solutions isolées par fractionnement de l’urine de bovin
Baoulé pour Glossina tachinoides Westwood, 1850 au Burkina Faso. Rev Elev Med Vet Pays Trop 42 453-455.
Frezil J-L and Carnevale P (1976) Utilisation de la caroglace pour la capture de Glossin fuscipes quanzensis Pires, 1948,
avec le piège Challier-Laveissière. Consequences épidémiologiques. Cahiers ORSTOM séries entomologie médicale
et parasitologie 14: 225-233.
Galey JB, Mérot P, Mitteault A, Filledier J and Politzar H (1986) Efficacité du dioxyde de carbone comme attractif pour
Glossina tachinoides en savane humide d’Afrique de l’Ouest. Rev Elev Med Vet Pays Trop 39: 351-354.
Gibson G (1992) Do tsetse ‘see’ zebras? A field study of the visual response of tsetse to striped targets. Physiol Entomol
17: 141-147.
Gibson G and Brady J (1985) Anemotactic’ flight paths of tsetse flies in relation to host odour: a preliminary video study
in nature of the response to loss of odour. Physiol Entomol 10: 375-406.
Gibson G and Brady J (1988) Flight behaviour of tsetse flies in host odour plumes - the initial response to leaving or
entering odour. Physiol Entomol 13: 29-42.
Gibson G, Packer MJ, Steullet P and Brady J (1991) Orientation of tsetse flies to wind, within and outside host odour
plumes in the field. Physiol Entomol 16: 47-56.
Gibson G and Torr SJ (1999) Visual and olfactory responses of haematophagous Diptera to host stimuli. Med Vet Entomol
13: 2-23.
Gikonyo NK, Hassanali A, Njagi PGN, Gitu PM and Midiwo JO (2002) Odour composition of preferred (buffalo and ox)
and nonpreferred (waterbuck) hosts of some savannah tsetse flies. J Chem Ecol 28: 969-981.
Gikonyo NK, Hassanali A, Njagi PGN and Saini RK (2000) Behaviour of Glossina morsitans morsitans Westwood (Diptera:
Glossinidae) on waterbuck Kobus defassa Ruppel and feeding membranes smeared with waterbuck sebum
indicates the presence of allomones. Acta Tropica 77: 295-303.
Gikonyo NK, Hassanali A, Njagi PGN and Saini RK (2003) Responses of Glossina morsitans morsitans to blends of
electroantennographically active compounds in the odors of its preferred (buffalo and ox) and nonpreferred
(waterbuck) hosts. J Chem Ecol 29: 2331-2345.
Gooding RH and Krafsur ES (2005) Tsetse genetics: contributions to biology, systematics, and control of tsetse flies. Ann
Rev Entomol 50: 101-123.
Gouteux JP, Blank F, Cuisance D, D’Amico F and Kota Guinza A (1995) Trials of olfactory attractants to enhance trap
catches of Glossina fuscipes fuscipes (Diptera: Glossinidae) in the Central African Republic. Vet Res26: 33-340.
Green CH (1986) Effects of colours and synthetic odours on the attraction of Glossina pallidipes and Glossina morsitans
morsitans to traps and screens. Physiol Entomol 11: 411-421.
Green CH (1994) Bait methods for tsetse fly control. In: Baker JR and Muller R (eds) Advances in Parasitology, Vol 34,
Academic Publishers, pp. 229-291.
Griffiths N and Brady J (1995) Wind structure in relation to odour plumes in tsetse fly habitats. Physiol Entomol 20:
286-292.
Griffiths N, Paynter Q. and Brady J (1995) Rates of progress up odour plumes by tsetse flies: a mark-release video study
of the timing of odour source location by Glossina pallidipes. Physiol Entomol 20: 100-108.
Hadaway AB (1977) Resting behaviour of tsetse flies, and its relevance to their control with residual insecticides.
Miscellaneous Report No. 36. Centre for Overseas Pest Research, London, UK, 11 pp.
Hall DR, Beevor PS, Cork A, Nesbitt BF and Vale GA (1984) 1-octen-3-ol - potent olfactory stimulant and attractants for
tsetse isolated from cattle odours. Insect Sci Appl 5: 335-339.
Hargrove JW (1975) Some changes in the flight apparatus of tsetse flies, Glossina morsitans and G. pallidipes during
maturation. J Insect Physiol 21: 1485-1489.
Hargrove JW (1976) Effect of human presence on behaviour of tsetse (Glossina spp) (Diptera: Glossinidae) near a
stationary ox. Bull Entomol Res 66: 173-178.

Olfaction in vector-host interactions 285

Steve J. Torr and Philippe Solano

Hargrove JW (1980) The effect of model size and ox odour on the alighting response of Glossina morsitans Westwood
and Glossina pallidipes Austen (Diptera: Glossinidae). Bull Entomol Res 70: 229-234.
Hargrove JW (1988) Tsetse - the limits to population growth. Med Vet Entomol 2: 203-217.
Hargrove JW (1993) Target barriers for tsetse flies (Glossina spp) (Diptera: Glossinidae): quick estimates of target
densities and barrier widths. Bull Entomol Res 83: 197-200.
Hargrove JW (2003) Tsetse eradication: sufficiency, necessity and desirability. DFID Animal Health Programme, Centre
for Tropical Veterinary Medicine, University of Edinburgh, Edinburgh, UK. 134 pp.
Hargrove JW and Brady J (1992) Activity rhythms of tsetse flies (Glossina spp) (Diptera: Glossinidae) at low and high
temperatures in nature. Bull Entomol Res 82: 321-326.
Hargrove JW, Holloway MTP, Vale GA, Gough AJE and Hall DR (1995) Catches of tsetse (Glossina spp.) (Diptera: Glossinidae)
from traps and targets baited with large doses of natural and synthetic host odour. Bull Entomol Res 85: 215-227.
Hargrove JW, Torr SJ and Kindness HM (2003) Insecticide-treated cattle against tsetse (Diptera: Glossinidae): what
governs success? Bull Entomol Res 93: 203-217.
Hargrove JW and Vale GA (1978) Effect of host odour concentration on catches of tsetse flies (Glossinidae) and other
Diptera in the field. Bull Entomol Res 68: 607-612.
Hargrove JW and Vale GA (1980) Catches of Glossina morsitans morsitans Westwood and Glossina pallidipes Austen
(Diptera: Glossinidae) in odour-baited traps in riverine and deciduous woodlands in the Zambezi valley of
Zimbabwe. Bull Entomol Res 70: 571-578.
Hassanali A, McDowell PG, Owaga MLA and Saini RK (1986) Identification of tsetse attractants from excretory products
of a wild host animal, Syncerus caffer. Insect Sci Appl 7: 5-9.
IAEA (2003) Improved attractants for enhancing tsetse fly suppression. Final report of a co-ordinated research project
1006-2002, International Atomic Energy Agency, Vienna, Austria.
Kamuanga M, Swallow BM, Sigue´ H and Bauer B (2001) Evaluating contingent and actual contributions to a local public
good: tsetse control in the Yale agro-pastoral zone, Burkina Faso. Ecol Econ 39: 115-130.
Kappmeier-Green K (2001) Strategy for monitoring and sustainable integrated control or eradication of Glossina
brevipalpis and G. austeni (Diptera: Glossinidae) in South Africa. PhD thesis, University of Pretoria: 245.
Kappmeier K and Nevill EM (1999) Evaluation of conventional odour attractants for Glossina brevipalpis and Glossina
austeni (Diptera: Glossinidae) in South Africa. Ond J Vet Res 66: 307-316.
Kennedy J S (1977) Olfactory responses to distant plants and odor sources. In: Shorey HH and McKelvey JJ (eds) Chemical
control of insect behavior: theory and application. John Wiley & Sons, New York, USA, pp 67-91.
Kgori PM, Modo S and Torr SJ (2006) The use of aerial spraying to eliminate tsetse from the Okavango Delta of Botswana.
Acta Tropica 99: 184-199.
Kyorku C, Brightwell R and Dransfield RD (1990) Traps and odour baits for the tsetse fly, Glossina longipennis (Diptera:
Glossinidae). Bull Entomol Res 80: 405-415.
Laveissière C and Grébaut P (2003) The risk of sleeping sickness transmission: simplifying the calculation of the index.
Insect Sci Appl 23: 99-102.
Madubunyi LC, Hassanali A, Ouma W, Nyarango D and Kabii J (1996) Chemoecological role of mammalian urine in host
location by tsetse, Glossina spp (Diptera: Glossinidae). J Chem Ecol 22: 1187-1199.
Maudlin I (2006) African trypanosomiasis. Ann Trop Med Parasitol 100: 679-701.
Mérot P, Filledier J and Mulato C (1988) Pouvoir attractif, pour Glossina tachinoides, de produits chimiques isolés des
odeurs animales. Rev Elev Med Vet Pays Trop 41: 79-85.
Mérot P, Galey JB, Politzar H, Filledier J and Mitteault A (1986) Pouvoir attractif de l’odeur des hôtes nourriciers pour
Glossina tachinoides en zone soudano-guinéenne (Burkina Faso). Rev Elev Med Vet Pays Trop 39: 345-350.
Modo S (1999) Odour attractants for the tsetse fly Glossina morsitans centralis Machado in Botswana. MSc thesis,
University of Zimbabwe, Harare, Zimbabwe.
Mohamed-Ahmed MM (1998) Olfactory responses of Glossina fuscipes fuscipes (Diptera: Glossinidae) to the monitor
lizard Varanus niloticus niloticus. Bull Entomol Res 88: 311-317.
Mohamed-Ahmed MM and Mihok S (1999) Responses of Glossina fuscipes fuscipes (Diptera: Glossinidae) and other
Diptera to carbon dioxide in linear and dense forests. Bull Entomol Res 89: 177-184.

286  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

Muzari MO and Hargrove JW (1996) The design of target barriers for tsetse flies, Glossina spp (Diptera: Glossinidae).
Bull Entomol Res 86: 579-583.
Mwangelwa MI, Dransfield RD, Otieno LH and Mbata KJ (1995) The responses of Glossina fuscipes fuscipes Newstead to
odour attractants and traps. J Afr Zool 109: 23-30.
Mwangi MT, Gikonyo NK and Ndiege IO (2008) Repellent properties of delta-octalactone against the tsetse fly, Glossina
morsitans morsitans. J Insect Sci 8: paper 43.
Ndegwa PN and Mihok S (1999) Development of odour-baited traps for Glossina swynnertoni (Diptera: Glossinidae).
Bull Entomol Res 89: 255-261.
Omolo MO, Hassanali A, Mpiana S, Esterhuizen J, Lehane MJ, Solano P, Rayaisse J-B, Vale GA, Torr SJ and Tirados I (2009)
Prospects for developing odour baits to control G. fuscipes spp., the major vector of human African trypanosomiasis.
PloS Negl Trop Dis 3(5): e435. doi:10.1371/journal.pntd.0000435.
Paynter Q and Brady J (1993) Flight responses of tsetse flies (Glossina) to octenol and acetone vapour in a wind tunnel.
Physiol Entomol 18: 102-108.
Politzar H and Merot P (1984) Attraction of the tsetse fly Glossina morsitans submorsitans to acetone, 1-octen-3-ol, and
the combination of these compounds in West Africa. Rev Elev Med Vet Pays Trop 37: 468-473.
Ravel S, de Meeus T, Dujardin JP, Zeze DG, Gooding RH, Dusfour I, Sane B, Cuny G and Solano P (2007) The tsetse
fly Glossina palpalis palpalis is composed of several genetically differentiated small populations in the sleeping
sickness focus of Bonon, Cote d’Ivoire. Infect Genet Evol 7: 116-125.
Rayaisse J-B, Solano P, Dramane K, Torr SJ, Tirados I, Lehane MJ and Esterhuizen J (2010) Responses of G. tachinoides and
G. palpalis spp. to natural host odours. PLoS Negl Trop Dis 4(3): e632. doi:10.1371/journal.pntd.0000632.
Rogers DJ (1988) A general model for the African trypanosomiases. Parasitology 97: 193-212.
Sabelis MW and Schippers P (1984) Variable wind directions and anemotactic strategies of searching for an odour
plume. Oecologia 63: 225-228.
Saini RK and Hassanali A (2007) A 4-alkyl-substituted analogue of guaiacol shows greater repellency to Savannah tsetse
(Glossina spp.). J Chem Ecol 33: 985-995.
Sané B, Laveissière C and Meda HA (2000) Diversité du régime alimentaire de Glossina palpalis palpalis en zone forestière
de Côte d’Ivoire: relation avec la prévalence de la trypanosomiase humaine africaine. Trop Med Int Health 5: 73-78.
Shaw A, Torr S, Waiswa C and Robinson T (2006) Comparable costings of alternatives for dealing with tsetse: estimates for
Uganda. Proceedings 11th International Congress of Parasitology ICOPA XI, Medimond International Proceedings,
Bologna, Italy, Glasgow, UK.
Simarro PP, Jannin J and Cattand P (2008) Eliminating human African trypanosomiasis: Where do we stand and what
comes next? Plos Med 5: 174-180.
Simo G, Njiokou F, Mbida Mbida JA, Njitchouang GR, Herder S, Asonganyi T and Cuny G (2008) Tsetse fly host preference
from sleeping sickness foci in Cameroon: epidemiological implications. Infect Genet Evol 8: 34-39.
Solano P, de La Rocque S, de Meeus T, Cuny G, Duvallet G and Cuisance D (2000) Microsatellite DNA markers reveal
genetic differentiation among populations of Glossina palpalis gambiensis collected in the agro-pastoral zone of
Sideradougou, Burkina Faso. Insect Mol Biol 9: 433-439.
Späth J (1995) Olfactory attractants for West African tsetse flies, Glossina spp (Diptera: Glossinidae). Trop Med Parasitol
46: 253-257.
Späth J (1997) Natural host odours as possible attractants for Glossina tachinoides and G. longipalpis (Diptera:
Glossinidae). Acta Tropica 68: 149-158.
Sutcliffe JF (1987) Distance orientation of biting flies to their hosts. Insect Sci Appl 8: 611-616.
Torr SJ (1988a) The activation of resting tsetse flies (Glossina) in response to visual and olfactory stimuli in the field.
Physiol Entomol 13: 315-325.
Torr SJ (1988b) The flight and landing of tsetse (Glossina) in response to components of host odour in the field. Physiol
Entomol 13: 453-465.
Torr SJ (1988c) Behaviour of tsetse flies (Glossina) in host odour plumes in the field. Physiol Entomol 13: 467-478.
Torr SJ (1989) The host-orientated behaviour of tsetse flies (Glossina): the interaction of visual and olfactory stimuli.
Physiol Entomol 14: 325-340.

Olfaction in vector-host interactions 287

Steve J. Torr and Philippe Solano

Torr SJ (1990) Dose responses of tsetse flies (Glossina) to carbon dioxide, acetone and octenol in the field. Physiol
Entomol 15: 93-103.
Torr SJ (1994) Responses of tsetse flies (Diptera: Glossinidae) to warthog (Phacochoerus aethiopicus Pallas). Bull Entomol
Res 84: 411-419.
Torr SJ, Hall DR, Phelps RJ and Vale GA (1997) Methods for dispensing odour attractants for tsetse flies (Diptera:Glossinidae).
Bull Entomol Res 87: 299-311.
Torr SJ, Hall DR and Smith JL (1995) Responses of tsetse flies (Diptera: Glossinidae) to natural and synthetic ox odours.
Bull Entomol Res 85: 157-166.
Torr SJ and Hargrove JW (1999) Behaviour of tsetse (Diptera: Glossinidae) during the hot season in Zimbabwe: the
interaction of micro-climate and reproductive status. Bull Entomol Res 89: 365-379.
Torr SJ, Hargrove JW and Vale GA (2005) Towards a rational policy for dealing with tsetse. Trends Parasitol 21: 537-541.
Torr SJ and Mangwiro TNC (1996) Upwind flight of tsetse (Glossina spp) in response to natural and synthetic host odour
in the field. Physiol Entomol 21: 143-150.
Torr SJ and Mangwiro TNC (2000) Interactions between cattle and biting flies: effects on the feeding rate of tsetse. Med
Vet Entomol 14: 400-409.
Torr SJ, Mangwiro TNC and Hall DR (1996) Responses of Glossina pallidipes (Diptera: Glossinidae) to synthetic repellents
in the field. Bull Entomol Res 86: 609-616.
Torr SJ, Mangwiro TNC and Hall DR (2006) The effects of host physiology on the attraction of tsetse (Diptera: Glossinidae)
and Stomoxys (Diptera: Muscidae) to cattle. Bull Entomol Res 96: 71-84.
Torr SJ, Maudlin I and Vale GA (2007a) Less is more: restricted application of insecticide to cattle to improve the cost
and efficacy of tsetse control. Med Vet Entomol 21: 53-64.
Torr SJ, Parker AG and Leigh-Browne G (1989) The responses of Glossina pallidipes Austen (Diptera: Glossinidae) to
odour-baited traps and targets in Somalia. Bull Entomol Res 79: 99-108.
Torr SJ, Prior A, Wilson PJ and Schofield S (2007b) Is there safety in numbers? The effect of cattle herding on biting risk
from tsetse flies. Med Vet Entomol 21: 301-311.
Torr SJ, Wilson PJ, Schofield S, Mangwiro TNC, Akber S and White BN (2001) Application of DNA markers to identify the
individual-specific hosts of tsetse feeding on cattle. Med Vet Entomol 15: 78-86.
Vale GA (1971) Artificial refuges for tsetse flies (Glossina spp). Bull Entomol Res 61: 331-350.
Vale GA (1974a) New field methods for studying responses of tsetse flies (Diptera; Glossinidae) to hosts. Bull Entomol
Res 64: 199-208.
Vale GA (1974b) Responses of tsetse flies (Diptera, Glossinidae) to mobile and stationary baits. Bull Entomol Res 64:
545-588.
Vale GA (1977a) Feeding responses of tsetse flies (Diptera-Glossinidae) to stationary hosts. Bull Entomol Res 67: 635-649.
Vale GA (1977b) Flight of tsetse flies (Diptera: Glossinidae) to and from a stationary ox. Bull Entomol Res 67: 297-303.
Vale GA (1979) Field responses of tsetse flies (Diptera: Glossinidae) to odours of men, lactic acid and carbon dioxide.
Bull Entomol Res 69: 459-467.
Vale GA (1980a) Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to carbon dioxide, acetone
and other chemicals. Bull Entomol Res 70: 563-570.
Vale GA (1980b) Flight as a factor in the host-finding behaviour of tsetse flies (Diptera: Glossinidae). Bull Entomol Res
70: 299-307.
Vale GA (1993) Development of baits for tsetse flies (Diptera: Glossinidae) in Zimbabwe. J Med Entomol 30: 831-842.
Vale GA (1998) Responses of tsetse flies (Diptera: Glossinidae) to vegetation in Zimbabwe: implications for population
distribution and bait siting. Bul Entomol Res 80 (Suppl 1): 1-59.
Vale GA and Cumming DHM (1976) Effects of selective elimination of hosts on a population of tsetse flies (Glossina
morsitans morsitans (Diptera: Glossinidae)). Bull Entomol Res 66: 713-729.
Vale GA, Flint S and Hall DR (1986a) The field responses of tsetse flies, Glossina spp. (Diptera: Glossinidae), to odours of
host residues. Bull Entomol Res 76: 685-693.
Vale GA and Hall DR (1985a) The role of 1-octen-3-ol, acetone and carbon dioxide in the attraction of tsetse flies,
Glossina spp. (Diptera: Glossinidae), to ox odour. Bull Entomol Res 75: 209-217.

288  Olfaction in vector-host interactions

 12. Olfaction in Glossina – host interactions: a tale of two tsetse

Vale GA and Hall DR (1985b) The use of 1-octen-3-ol, acetone and carbon dioxide to improve baits for tsetse flies,
Glossina spp. (Diptera: Glossinidae). Bull Entomol Res 75: 219-231.
Vale GA and Torr SJ (2004) Development of bait technology to control tsetse. In: Maudlin I, Holmes PH and Miles MA
(eds) The trypanosomiases. CABI, Wallingford, UK, pp 509-523.
Vale GA, Hall DR and Gough AJE (1988a) The olfactory responses of tsetse flies, Glossina spp (Diptera: Glossinidae) to
phenols and urine in the field. Bull Entomol Res 78: 293-300.
Vale GA and Hargrove JW (1979) Methods of studying the efficiency of traps for tsetse flies (Diptera: Glossinidae) and
other insects. Bull Entomol Res 69: 183-193.
Vale GA, Hargrove JW, Cockbill GF and Phelps RJ (1986b) Field trials of baits to control populations of Glossina morsitans
morsitans Westwood and Glossina pallidipes Austen (Diptera: Glossinidae). Bull Entomol Res 76: 179-193.
Vale GA, Lovemore DF, Flint S and Cockbill GF (1988b) Odour-baited targets to control tsetse flies, Glossina spp. (Diptera,
Glossinidae), in Zimbabwe. Bull Entomol Res 78: 31-49.
Waiswa C, Picozzi K, Katunguka-Rwakishaya E, Olaho-Mukani W, Musoke RA and Welburn SC (2006) Glossina fuscipes
fuscipes in the trypanosomiasis endemic areas of south eastern Uganda: Apparent density, trypanosome infection
rates and host feeding preferences. Acta Tropica 99: 23-29.
Warnes ML (1990) The effect of host odour and carbon dioxide on the flight of tsetse flies (Glossina spp) in the laboratory.
Journal of Insect Physiology 36: 607-611.
Warnes ML (1995) Field studies on the effect of cattle skin secretion on the behaviour of tsetse. Med Vet Entomol 9:
284-288.
Warnes ML, van den Bossche P, Chihiya J, Mudenge D, Robinson TP, Shereni W and Chadenga V (1999) Evaluation of
insecticide-treated cattle as a barrier to re-invasion of tsetse to cleared areas in northeastern Zimbabwe. Med Vet
Entomol 13: 177-184.
Welburn SC, Coleman PG, Maudlin I, Fevre EM, Odiit M and Eisler MC (2006) Crisis, what crisis? Control of Rhodesian
sleeping sickness. Trends Parasitol 22: 123-128.
Willemse L (1991) A trial of odour baited targets to control the tsetse fly, Glossina morsitans centralis (Diptera: Glossinidae)
in west Zambia. Bull Entomol Res 81: 351-357.
Williams BG (1994) Models of trap seeking by tsetse flies: anemotaxis, kinesis and edge detection. J Theor Biol 168:
105-115.
Zollner GE, Torr SJ, Ammann C and Meixner FX (2004) Dispersion of carbon dioxide plumes in African woodland:
implications for host-finding by tsetse flies. Physiol Entomol 29: 381-394.

Olfaction in vector-host interactions 289

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

13. Behavioural modalities of ‘non-vector’ biting Diptera: from

olfaction to feeding

Steve Schofield and Steve J. Torr

Abstract

Though they are not considered to be important vectors, stable flies and related species can have
direct economic impacts, and as well might influence pathogen transmission by affecting the
feeding behaviour of flies such as tsetse. We discuss here the responses of Stomoxyinae to host-
based kairomones. Further, we consider the interplay between these Diptera, other vectors and
their common hosts as it relates to disease epidemiology and control strategies. In general, we find
that the odour-evoked responses of stable flies in the laboratory are similar to other groups of fast
moving Diptera that blood feed (e.g. tsetse), i.e. activity increases and flights are steered upwind.
In field studies, several compounds have been associated with increased catch of Stomoxyinae
from traps, but only CO2 consistently achieved this effect. On the whole, results suggest that non-
CO2 kairomones are not that important to host finding by Stomoxys and, further, that synthetic
host odours in general might not have a place in sampling or control schemes for this pest. We
also describe the interplay between Stomoxyinae and their hosts, highlighting the role that their
feeding behaviour might have as regards the epidemiology of animal trypanosomiasis. Finally,
we discuss briefly some of the potential practical applications that might arise from a better
understanding of the interrelationship between Stomoxys species and their hosts.

Keywords: behaviour, host odour, olfaction, Stomoxys, Glossina, stable fly, tsetse fly

Introduction

The blood feeding habit has arisen independently in several orders of insects (Lehane 2005) and,
with this, has come a means of spreading a variety of human and veterinary pathogens. These
range from those with a modest or focal impact on human population health (e.g. Murray Valley
encephalitis, La Crosse encephalitis), to others that play a significant role more-or-less globally
(e.g. malaria, dengue). It is our desire to manage the threats presented by vector-borne pathogens
that, to a large extent, has driven research to understand the behavioural ecology underpinning
haematophagy. Such efforts have aimed to provide novel means of controlling disease by, for
example, luring vectors to lethal baits or developing means of preventing them from finding
and biting a host. Unfortunately, and outside some notable exceptions, efforts in this area often
have not contributed meaningfully to control efforts. There are many reasons for this, but it no
doubt at least partly reflects a systematic failure on the part of investigators to appreciate and/or
evaluate behaviour sensu stricto. Instead, the tendency has been to measure the final outcome of
a complex repertoire of behaviours. To paraphrase, we have spent entirely too much time catching
biting insects in traps (or huts, or on hosts, or in olfactometer ports) and not nearly enough time
analysing, quantitatively, the behavioural processes that got them there in the first place. Why
this should be the case is difficult to fathom, especially given that some of the earliest research
(Kennedy 1939) in the field seemed to avoid this pitfall. Moreover, early workers did warn of the
potential problems with taking a ‘black-box’ approach to behavioural research. One of the most
vocal in this regard was John Kennedy, who often discussed the potential for bias in behavioural
work, especially where experimental endpoints are framed through an anthropomorphic lens
(Kennedy 1992). In any event, one result of not heeding these cautions is that our understanding

Olfaction in vector-host interactions 291

Steve Schofield and Steve J. Torr

of the behavioural ecology of vectors is poorer than it ought to be. Whether or not this failing has
impacted on the effectiveness of interventions that target vectors remains unclear.

On the upside, in recent years the need to analyse vector behaviour objectively and quantitatively
seems to have been re-emphasised, and there has been technological improvement and new
funding opportunities to support this research. The result has been a greater volume and quality
of work on vector behaviour. Examples include other chapters in this book, as well as recent
studies in the literature (Cooperband and Cardé 2006, Dekker et al. 2005, Torr et al. 2008).

Chapter scope

This book focuses on disease vectors, whether veterinary or medical. So the inclusion of a
chapter on ‘non-vector’ biting flies seems odd. However, it is probably more precise to say that
the main subjects of this chapter are not especially good or important vectors. By no means are
we suggesting that they do not have direct or indirect impacts on health or productivity, because
they might and do (Hansens 1951, Mullens et al. 2006, Pickard 1968, Schofield and Torr 2002,
Schwinghammer et al. 1986, 1987, Zumpt 1973). We also feel that the information consolidated
herein might have some relevance to vectors more generally, for the emphasis of this book is on
behaviour rather than diseases or pathogens. In particular, it can be revealing to contrast the
behaviours of non-vector biting flies with better-studied vectors such as the Glossinidae, not so
much to make note of similarities, which abound, but rather to highlight intriguing differences.

For simplicity, and to allow a more in-depth treatment of the subject material, we decided to limit
our coverage to taxa of non-vector biting flies that occur globally, and for which a reasonable body
of laboratory and field-based behavioural data exists. The resulting list is very short indeed, i.e. it
is arguably comprised of species that fall within a single group, i.e. the Stomoxyinae. Within this
subfamily, our emphasis will be on the synanthropic and cosmopolitan species Stomoxys calcitrans
(L.) (the stable fly).

In terms of chapter structure, our thoughts are presented along a continuum that starts with the
insect as a sensory receptor, and then moves through events that might culminate in feeding
on a vertebrate host. Hence, though much of the discussion focuses on the role of olfaction in
distance orientation, visual cues are touched upon and near- or on-host feeding behaviours are
discussed. We also consider laboratory data related to distance orientation in tsetse in this chapter.
Finally, while we acknowledge that interesting olfactory work has been done in several areas of
stable fly biology (e.g. see Carlson et al. 2000, Romero et al. 2006), we focus on behaviour related
to haematophagy.

A framework for host orientation

Despite our above caution regarding the need to be aware of bias and anthropomorphisms,
we begin our discussion with a model for host-seeking behaviour that is somewhat dated,
oversimplified, perhaps biased and almost certainly anthropomorphic. Nevertheless, it provides
a useful framework around which thoughts – and manuscripts – can be organised.

Many species of haematophagous Diptera, including stable flies, must find their hosts from a
distance. Elements of this process can be divided along spatial and temporal lines and, though
it is impossible to attribute the logical framework to a single author, it is well summarised by
Sutcliffe (1986, 1987). In this model, the receptivity of a given insect to host-based stimuli is

292  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

modulated by factors such as hunger (Klowden and Lea 1979, Klowden et al. 1987), competing
behavioural priorities (e.g. oviposition) and endogenous activity cycles (Jones and Gubbins 1978,
Nayar and Sauerman 1971). Presuming an insect that is receptive, host orientation sensu stricto is
thought to be ‘activated’ by some type(s) of host cue. What follows are behavioural manoeuvres,
often mediated by odour, that bias movement towards a host. For flying insects, environmental
landmarks are critically important during this period, e.g. to steer flights. As the orienting insect
approaches the blood source, host-based visual and/or other cues are thought to become
increasingly relevant as behavioural modulators. Ultimately the searching insect might find and
land upon the host and then feed to repletion.

Blood lust: the role of blood feeding in the biology of Stomoxys

Much, though not all (Kunz and Monty 1976, Mihok et al. 1995, 1996, 2007, Zumpt 1973) of our
knowledge concerning Stomoxys biology comes from work done on stable flies. Like tsetse, stable
flies are day-active and both sexes blood feed, which provides the materials necessary for males
to develop reproductive competence (Anderson 1978, Venkatesh and Morrison 1980, 1982),
and females to develop ovaries (Chia et al. 1982, Kuzina 1942, Spates et al. 1988, Venkatesh and
Morrison 1980). In contrast to tsetse, stable flies also feed on plant-related sugar sources (Jones
et al. 1992, Parr 1962) and blood feeding is often daily (Venkatesh and Morrison 1980) or even
more frequent (S Schofield, unpublished data). In terms of endogenous activity, laboratory studies
suggest a single daily peak, with a free running period of approximately 26 hours (Schofield and
Brady 1996). Diel field activity is often bimodal (Beresford and Sutcliffe 2006, Charlwood and
Lopes 1980, Kunz and Monty 1976, Schofield 1998), for example with morning and evening peaks,
though unimodal patterns have also been described (Berry and Campbell 1985, Harley 1964).
Regardless of the specific feeding periodicity or activity cycle, the reality for stable flies, especially
females, is that they feed often. From a host location perspective, this presumably means that
unless they have recently fed, they are likely to be receptive to host-based stimuli.

Stable flies as olfactory receptors

The molecular and physiological bases for insect olfaction are well covered elsewhere, including
Chapters 2, 3 and 4 in this book. We therefore limit our discussion to a brief overview of electro-
physiological studies that have evaluated stable fly responses to odour.

In this fly, olfactory receptors are concentrated on the third antennal segment (Lewis 1971, 1972),
where single cell recordings by Lewis (1972) and later Bidgood (1980) demonstrated responses to
the ubiquitous kairomone CO2. Warnes and Finlayson (1986) similarly demonstrated responses to
CO2, but in this case it was through use of electroantennogram (EAG). With the same technique, they
also recorded responses to acetone, 1-octen-3-ol, cattle odour, cattle odour filtered through soda-
lime, human breath and the odour of cattle faeces. Schofield et al. (1995) were the first to screen
a modest number of compounds using stable flies as electrophysiological receptors. Like Warnes
and Finalyson, they used EAG and found that 1-octen-3-ol evoked substantial activity, whereas
acetone did not. For saturated primary aliphatic alcohols, responses increased with carbon length
to octan-1-ol; and for 1-octen-3-ol with dose over several orders of magnitude. Responses to a
C8 aldehyde, carboxylic acid and ketone were less that to C8 alcohols. Several recent studies have
expanded upon the number of compounds screened for electrophysiological activity. Birkett et al.
(2004) worked with a number of biting and nuisance flies and discerned stable fly EAG responses
to several types of compound including phenols (m- and p-cresol), alcohols (including 1-octen-
3-ol), ketones, linalool and citronellol. These authors also tested the horn fly, Haematobia irritans

Olfaction in vector-host interactions 293

Steve Schofield and Steve J. Torr

(L.), with responses being similar to those reported for stable flies, though with a few exceptions,
notably, for example, an absence of response to phenols. Jeanbourquin and Guerin (2007a,b)
focused on rumen and dung odours, and reported that 1-octen-3-ol and dimethyl trisulphide
evoked the strongest EAG responses, but carboxylic acids, short chain alcohols and aldehydes,
ketones, indoles, phenols and terpenes also elicited significant EAG responses. So what do the
electrophysiological results tell us? For one, they indicate that some compounds consistently
evoke responses (e.g. 1-octen-3-ol), whereas others do not (e.g. ketones, carboxylic acids). There
are many possible explanations for such differences, some of which are explored by Schofield
et al. (1995). Second, that stables flies act as receptors for a variety of host-derived compounds,
and that these same compounds would be reasonable targets for further investigation. What
electrophysiological outcomes do not address is whether or not active compounds have any
behavioural (or biological) relevance – it is this question that comprises the subject matter of the
next several chapter sections.

Starting the search engine: initiation of host-oriented behaviour

Activation is thought to represent a transition in behavioural state whereby the insect becomes
(more) receptive to host-based cues. Hence, measurement of activation itself is not possible.
Rather, early and observable events in the host-seeking process are used as indicators that
activation has occurred. While one can debate as to whether activation is a useful concept, i.e. it
is not testable, the behaviours that have been used as surrogate markers do provide insight into
host-seeking processes. In this context, measurement of activation has traditionally employed
frequency-based measures (e.g. flight periodicity) as behavioural indicators. As well, at least for
stable flies, studies have been informed by results for other groups of blood feeding Diptera, in
particular as regards the positive association between exposure to CO2 and mosquito (Daykin et
al. 1965) or tsetse (Bursell 1984, Turner 1971) activity.

To our knowledge, the earliest work on behaviour initiation by stable flies was done by Gatehouse
and Lewis (1973), who recorded the number of stable fly landings on targets situated within
a wind tunnel. They observed more landings in the presence of CO2, and suggested that this
kairomone evoked flights and/or increased the probability that a flight would terminate on a
target. Warnes and Finlayson (1985a) followed this up with a study using stable flies that were
placed singly into a test chamber and exposed to various concentrations of CO2, or to human
breath. They found that flight activity generally increased with CO2 concentration, was positively
associated with hunger status, and that breath was associated with higher levels of activity than
could be accounted for by CO2 alone. Alzogaray and Carlson (2000) also suggested an activation-
associated effect for CO2, in this case based on net movement of stable flies away from one end
of an olfactometer. In a more comprehensive study, Schofield et al. (1997) evaluated responses
of groups of four stable flies that were placed into a test chamber situated in a wind tunnel and
that were then exposed to CO2. In line with the results of earlier studies, CO2 was associated with
increased flight frequency, decreased latency to take-off, and increased likelihood that flights
would terminate on a visual target placed in the test chamber. Moreover, the authors pointed
out that the response threshold to CO2 was similar in their and Warnes and Finalyson’s work,
e.g. 0.006% compared to 0.01%. However, they did caution against reading too much into this
similarity. First, because inter-study differences in plume structure could affect responses. Second,
because these thresholds, if expressed as odour flux, were about an order of magnitude different.
Finally, because the activity increase was manifest as a step function, whereas in Warnes and
Finlayson’s work there was a dose-response relationship.

294  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

Schofield et al. (1997) also evaluated responses to acetone and 1-octen-3-ol. For acetone, outcomes
were more muted but followed a similar pattern to those for CO2, i.e. flight activity increased with
a response threshold of about 0.01 ug/l, and the proportion of flights terminating on a target
increased in the presence of this compound. By contrast, the principle impact of 1-octen-3-ol was
a decrease in flight activity at the highest concentrations. This latter finding matches well with
at least some work on tsetse where high release rates of 1-octen-3-ol have been associated with
reduced: trap catch (Vale and Hall 1985a), upwind turning response (Paynter and Brady 1993), and
activity (Bursell 1984).

More recently, Jeanbourquin and Guerin (2007a,b) examined the behavioural responses of stable
flies to rumen and dung-based volatiles. In one experiment, groups of five flies were placed into a
release chamber situated in a wind tunnel, with ‘activation’ defined by a fly leaving the chamber.
Increased rates of leaving were evident for raw rumen volatiles, as well as several individual
chemicals (dimethyl trisulphide, p-cresol and butanoic acid). As with previous studies, 1-octen-
3-ol was not associated with an increase in activity. Nor, as was documented in the Schofield et
al. (1997) study, was it associated with a decrease in activity. In another experiment, single flies
placed into a wind tunnel were exposed to raw dung volatiles, with activation being defined as
three or more take-offs within the flight area. Both cow and horse dung were associated with
increased ‘activation’ compared against the control group, e.g. about 70% of flies were activated
compared against 30%. Unfortunately, the authors did not characterise responses over a range
of compound exposures, nor did they use survivorship analyses to explore the data. As a result,
potentially informative comparisons that might have been made with earlier studies regarding,
for example, latency of response or nature of the dose-response curve, are not possible.

In-flight responses to odour: wind tunnel results

If the intent of behavioural work is to inform vector control, then research should ultimately be
done in, or at least translate to, the field. As pointed out earlier in this chapter, much of the
existing field research on haematophagous Diptera has not looked at behaviour per se, but rather
has recorded behavioural endpoints, e.g. trap catch. Clearly, this partly reflects the difficulty of
executing and interpreting behavioural assays in the field. It is for this reason that laboratory
assays of behaviour can be especially useful, i.e. they can document behaviour under carefully
controlled and relatively easy to set up conditions. Despite this, most such work has, just as for
field studies, measured behavioural endpoints, e.g. position in an olfactometer, net displacement
in a wind tunnel. For example, using the set-up described earlier, Gatehouse and Lewis (1973)
suggested that while CO2 increased flight activity, it was not associated with a landing bias
on a baited versus unbaited target, and hence they concluded that it did not elicit directional
responses. On the other hand, they found that emanations from a hand resulted in relatively
more landings on a baited target and therefore inferred direction orientation. Of course, in the
absence of actual measurements of behaviour per se, it is impossible to validate these inferences.
Indeed, subsequent work on stable flies, as we shall see, contradicts their conclusion concerning
CO2. Warnes and Finlayson (1985b) evaluated responses of stable flies to CO2 and acetone in a
low speed wind tunnel, using net movement over a three-minute period to infer (or not) directed
orientation. They found that various concentrations of CO2 resulted in net upwind movement, as
did acetone (0.8 g/min), expired breath and odour from guinea pigs. Certainly, a parsimonious
explanation for these results is upwind anemotaxis, but by no means was this behavioural strategy
specifically demonstrated. Indeed, it has been argued that the results for acetone might also
reflect an arrestant artefact because cessation of upwind flight responses to acetone have been
documented at a concentration/flux some 20 times less than was tested in this study (Schofield

Olfaction in vector-host interactions 295

Steve Schofield and Steve J. Torr

and Brady 1997). In the last couple of years, several additional studies have evaluated orientation
behaviours of stable flies in the laboratory. Jeanbourquin and Guerin (2007a,b) were interested in
stable fly responses to rumen and dung volatiles. ‘Attraction’ in these experiments was determined
on the basis of whether flies showed a minimum upwind displacement in a wind tunnel. As above,
while these positive outcomes are suggestive of directional upwind movement, they do not
establish causation. In any event, these authors reported that a number of compounds, including
horse and cow dung, raw rumen volatiles, dimethyl trisulphide, p-cresol and butanoic acid were
associated with attraction in a wind tunnel. Though flights were also classified as being oriented
(or not) on the basis of ‘sinuous flights in the direction of the odour source’, the authors did not
provide the detail necessary to validate this classification scheme, nor the related interpretations.
Most recently, Beresford and Sutcliffe (2008), looked at the influence of male age on the
responsiveness of stable flies to CO2 in a wind tunnel. Interestingly, net upwind displacement at
the end of a two-hour observation period was inversely correlated with age, which the authors
suggest might reflect an age-based bias in the propensity of males to seek blood meals versus
female conspecifics. However, inferring orientation on the basis of fly position after an extended
period cannot be considered a reliable indicator of directional orientation, especially because a
negative control was not employed, e.g. these results also could reflect age-based variability in
activity levels or dispersal tendencies.

To our knowledge, the only laboratory experiment that has recorded and critically evaluated in-
flight manoeuvres of stable flies was done by Schofield and Brady (1997). This work has much
in common previous work on tsetse flies where upwind anemotaxis (Colvin et al. 1989, Paynter
1991, Paynter and Brady 1993, 1996) and ortho- and klinokinetic responses (Warnes 1990,
Paynter and Brady 1993, 1996) to host-based odour were observed. All of these studies analysed,
quantitatively, the actual flight tracks of test insects. In the interest of efficiency, we will discuss
the tsetse work here (instead of in Chapter 12, this volume), with particular attention being paid
to studies (Colvin et al. 1989, Paynter and Brady 1993, 1996) that used a set-up very similar to that
employed for the stable fly experiments. Bursell (1984) was one of the first to evaluate carefully
tsetse behaviour in the laboratory, and suggested that resting flies took off and flew in an upwind
direction when exposed to CO2. Following this, Colvin et al. (1989) released tsetse flies (Glossina
morsitans morsitans Westwood) from a crosswind position into a wind tunnel and analysed their
responses to: (1) the olfactory stimulus of a CO2 odour plume; (2) the visual stimulus of a moving
floor with a strong visual pattern and (3) the physical stimulus generated by wind moving at
0.25 m/s. By using clever combinations of these three stimuli, the authors demonstrated that
flights were indeed steered into the wind in the presence of this compound, and that this was
guided by visual flow, i.e. that upwind anemotaxis was occurring. This reinforced results from field
work where upwind turning had been observed (see Chapter 13), but more importantly provided
strong evidence for the mechanism of in-flight steering. Paynter and Brady (1996) expanded upon
this work by examining tsetse responses to CO2 at various wind speeds. In addition to upwind
anemotaxis, they described kinetic responses to CO2, specifically that tsetse slowed down and
turned more in its presence. However, these effects and upwind anemotaxis were not evident at
a low wind speed (e.g. 0.1m/s). The authors suggested that this might reflect a limit on the part of
tsetse to detect wind-induced drift. Not surprisingly, stable flies tested using a similar apparatus
(Schofield et al. 1997) also moved upwind when exposed to CO2. In this case, responses were dose-
dependent with flies showing a substantial propensity to move and exit and turn upwind, and to
execute a major turn over three orders of magnitude of CO2 concentration (Table 1). Moreover,
flight kinetics differed between the negative control and the odour treatments, and between
different concentrations of CO2 (Table 2). In other words, and as with tsetse, stable flies slowed
down, turned more, and headed upwind in the presence of CO2. However, they differed from

296  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

Table 1. Percentage of flights exiting upwind, exhibiting major turns and turning upwind in CO2, acetone, and
1-octen-3-ol. Controls consisted of clean air. CO2 concentration is given as percent above ambient (from Schofield
and Brady 1997).

Concentration Exit up Major turn Turn up

CO2 (%)
Control 25 38 54
0.0012 47 42 63
0.006 58 65 74
0.012 69 68 73
0.06 70 80 84
0.12 84 80 88
Acetone (μg/l)
Control 30 33 43
0.001 34 32 50
0.01 59 61 79
0.1 50 57 73
1.0 43 79 40
10.0 20 72 34
1-octen-3-ol (μg/l)
Control 30 39 48
0.0002 58 61 64
0.002 58 60 62
0.02 57 60 76
0.2 26 66 36
2.0 19 61 21

n for each row: CO2 = 73-92, acetone = 43-74, 1-octen-3-ol = 57-79.

tsetse by displaying substantially slower flight speeds and higher turning rates in clean air, and
in plumes containing CO2.

In-flight responses of tsetse and stable flies to acetone and 1-octen-3-ol have also been evaluated
in the laboratory. For tsetse and over a range of doses, both of these compounds were associated
with an increased probability of upwind flight (Paynter and Brady 1993). As regards kineses, tsetse
exposed to 1-octen-3-ol slowed down and turned more at the highest doses tested, whereas
acetone did not produce consistent concentration-dependent kinetic effects. For stable flies,
exposure to acetone or 1-octen-3-ol yielded modest upwind responses with more flies flying
upwind and/or executing a major turn over several concentration increments. However, the
highest 1-octen-3-ol exposures were associated with a reduction in upwind flight compared
against the negative control (Table 1).

Overall, the results of laboratory studies implicate a number of kairomones as being of potential
importance to host-orientation for stable flies (and tsetse flies). These include CO2, acetone and
1-octen-3-ol. Perhaps more importantly, they also highlight specific behavioural strategies (taxes
and kineses) that occur in the presence of odour that reasonably can be expected to increase the
probability of host finding. Indeed, it has been postulated that taxes and kinesis, depending on

Olfaction in vector-host interactions 297

Steve Schofield and Steve J. Torr

Table 2. Mean (±SE) values for angular velocity, sinuosity and velocity of video-recorded stable fly flights in CO2,
acetone, and 1-octen-3-ol. Controls consisted of clean air (from Schofield and Brady 1997).

Concentration Angular velocity (°/s) Sinuosity (°/cm) Velocity (cm/s)

CO2 (%)
Control 114.6±10.8 0.7±0.11 170.9±5.3
0.0012 129.1±10.5 0.8±0.09 188.4±5.1
0.006 162.8±10.1 1.1±0.09 172.2±4.9
0.012 192.2±10.9 1.5±0.12 151.3±5.3
0.06 191.5±9.4 1.5±0.08 139.4±4.6
0.12 212.9±9.9 1.9±0.09 133.3±4.8
Acetone (μg/l)
Control 102.2±12.4 0.6±0.11 169.8±5.6
0.001 124.7±11.5 0.8±0.10 160.3±5.2
0.01 122.9±9.7 0.9±0.08 149.8±4.4
0.1 151.9±9.5 1.1±0.08 157.1±4.3
1.0 167.0±9.4 1.3±0.12 145.5±4.3
10.0 156.3±8.4 1.2±0.07 140.3±3.8
1-octen-3-ol (μg/l)
Control 114.2±9.3 0.7±0.08 170.2±3.7
0.0002 151.4±10.9 1.1±0.10 142.0±4.3
0.002 148.4±11.2 1.3±0.10 136.9±4.4
0.02 152.7±10.4 1.1±0.08 148.6±4.1
0.2 151.5±9.4 1.2±0.08 139.5±3.7
2.0 137.7±11.4 1.2±0.10 123.5±4.8

n for each row: CO2 = 73-92, acetone = 43-74, 1-octen-3-ol = 57-79.

the environment, might play a more or less important role in the orientation response. This idea
was explored by Williams (1994), who suggested that the consistency of wind direction could be
an important determinant of the specific strategy deployed by foraging tsetse. In this model, if
wind-generated directional information was reliable at the time of odour detection, then upwind
anemotaxis with various odour sampling frequencies should optimise the probability of host
encounter. By contrast, if directional information was very poor then kinetic responses to odour
detection should be preferred, i.e. the fly should slow down and turn more. Supporting this model,
there is field evidence to suggest that tsetse can move upwind very quickly to an odour source
(Griffiths et al. 1995). While the model was developed for tsetse, the general approach and its
conclusions are applicable to any fast-flying insect responding to an odour source.

Of course, as with all laboratory studies, the important uncertainty is their relevance to the field
situation. Aside from the usual litany of issues such as the use of laboratory-reared insects and
artificial test conditions, we make note here of several additional points for consideration. First,
stable flies and tsetse move quickly relative to some other groups of blood feeding Diptera (e.g.
mosquitoes), or other insects for which orientation behaviours have been well documented
(e.g. moths). The upshot is that laboratory observations of Stomoxys and tsetse are, by default,
usually limited to relatively few behavioural events, at least compared to the types of records
that can be obtained for moths and mosquitoes. The consequence is that we have not been

298  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

able to observe directly how sequences of behaviours are put together for these groups, the
comparator being the elegant work that has been done on moths as they fly into and out of
odour plumes (Murlis et al. 1992, Chapter 6 in this volume). This limitation does come with some
analytic benefit. In particular, ‘less’ behaviour allows for easier analyses, and the rapid horizontal
movement displayed by these groups makes two-dimensional analyses more tractable. Second,
analyses have focused on concentration as a measure of exposure, with limited consideration
of odour flux (i.e. amount of material moving through area per unit time, often expressed as:
moles/s/cm2). It remains uncertain which, if either, of these measures is more appropriate. Is this
a critical issue? Perhaps not, but certainly it can affect inter-study validation, e.g. as pointed out
above in our discussion of the threshold at which CO2 evokes stable fly flight activity. However,
because flux loses its distinction from concentration where flights are substantially faster (e.g.
stable flies, tsetse) or slower than wind speed (Paynter and Brady 1996, Schofield 1996), and
because it is more challenging computationally, it seems prudent to express exposure intensity
based on concentration. However, this comes with the caveat that impacts of test conditions on
odour presentation should be considered when interpreting results.

Field work on the responses of Stomoxyinae to host odour

Before discussing the specifics of field studies evaluating responses of Stomoxyinae to host
kairomones, we wish to present a conclusion from our review. As far as we can tell, there is little
evidence to indicate that trap-out strategies have broad-based utility as a control intervention
for stable flies, and even less data to suggest that host-odour might contribute significantly to
any such approach.

While we cannot say whether this conclusion will hold into the future, the current lack of evidence
does beg a question – what is the practical aim of carrying out such work? For one, it might allow
for heightened sensitivity of surveillance systems. Of course this needs to be weighed against the
need for a surveillance system itself (i.e. for a non-vector), and the net benefit accrued by adding
host odour into the surveillance equation. Regarding the former, the nuisance and consequent
economic costs associated with Stomoxyinae seem to validate the need for at least targeted
surveillance, though the need to employ host odour as part of a surveillance system has not
been convincingly demonstrated. Another area where this work has proved to be useful, albeit
tangentially, is the investigation of host-stable fly interactions. Work in this area, which partly
originated due to interest in responses of flies to host odour, has yielded useful insights into intra-
and interspecific host-parasite relationships (see below)

In any event, visual cues, as well as traps and targets based on various combinations of these
have long been used as surveillance tools for stable flies (Schofield 1996). It has been around
these technologies that most odour-based research has been carried out. With the exception of
a few studies where trap efficiency was also considered (Mihok et al. 2007, Schofield 1998, Vale
and Hall 1985b), the bulk of the field work on Stomoxyinae has used a behavioural endpoint, i.e.
trap catch, as the outcome measure. As regards specific kairomones, the only compound that
has yielded consistent and positive results from the field is CO2. Hoy (1970) reported that Malaise
traps releasing CO2 at 3 l/min caught about three times more stable flies than unbaited traps, and
Roberts (1972) reported that catches from a Malaise trap baited with 3.5 l/min CO2 were similar
to those from a trap baited with odour from a single steer. More recently, Mihok et al. (1996)
found that Vavoua traps baited with CO2 at 2 l/min caught about twice as many Stomoxyinae
flies as unbaited traps. Some of the early work on tsetse responses to odour also yielded useful
information for Stomoxyinae. Vale (1980), while investigating the responses of tsetse to host

Olfaction in vector-host interactions 299

Steve Schofield and Steve J. Torr

odour, also collected Stomoxyinae; later work (Schofield 1998) suggests that the Stomoxys spp.
collected by Vale comprised a mixture of three species: S. niger niger Macquart, S. sitiens Rhodani
and S. calcitrans. Flies were intercepted using electric nets that were positioned downwind of a
stationary target. When CO2 was released at 2.5 or 15 l/min from near to the target, catch was
increased by more than an order of magnitude. For the variety of other compounds tested in this
study, which included ketones, aldehydes, alcohols and acids, the only one that was observed to
affect the behaviour of Stomoxyinae was acetic acid (at 0.7 g/hr), which reduced the catch. Vale
and Hall (1985a) found that catches of Stomoxyinae at beta traps or electrified targets increased
with CO2 release rate and was some 6 or 16 times higher, respectively, at the maximum tested (20
l/min) when compared against unbaited controls. The potency of CO2 was again demonstrated
by Vale and Hall (1985b) with catches at baited targets (2 l/min) being >10× greater than at an
unbaited trap. Several de facto dose responses to CO2 also have been obtained from Zimbabwe. In
these experiments, Stomoxyinae were collected from electric targets baited with the odour from
various sized groups of cattle (Hargrove et al. 1995, Torr et al. 2007). For both studies, catch rose
with bait mass in a logarithmic fashion (Figure 1). Schofield (1998) also deployed electric targets
in Zimbabwe to evaluate the influence of CO2 (2 l/min CO2) and found that it more than doubled
catches when compared against an unbaited target. Target efficiency, estimated by positioning an
electric screen next to the target, was > 50% in this study and was not affected by CO2.

Can non-CO2 odours derived from hosts increase the catch of stable flies in the field? Some of
the early studies in Zimbabwe suggested not (Vale 1980, Vale and Hall 1985a,b). However, since
this work, evidence has emerged to indicate that other kairomones, especially 1-octen-3-ol, can
sometimes affect trap catch. Holloway and Phelps (1991) found that F3 traps baited with 1-octen-
3-ol (0.7 mg/hr) caught about twice as many Stomoxys spp. as unbaited traps, though subsequent
work carried out in the same area (Schofield 1996) with the same type of trap was not able to

3
Catch index

0
1000 2,000 3,000 4,000 5,000
Herd mass (kg)
Figure 1. Catch indices of tsetse (solid circles) and Stomoxys (open circles) attracted to herds of cattle of various
size. Indices are calculated as the detransformed mean catch from a herd divided by that from a single ox weighing
400 kg. Lines fitted by regression of log10 (herd mass) against log10 (catch index); regression co-efficients for tsetse
and Stomoxys were 0.41 (±0.070, SE) and 0.25 (±0.093), respectively (adapted from Torr et al. 2007).

300  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

duplicate this result, and earlier studies, also in this area (Vale and Hall 1985a,b), were only able
to show a ‘repellent’ effect for 1-octen-3-ol (5-500 mg/hr) when it was released with ox odour. Of
importance, most of the above studies only classified catch to the genus level, and hence inter-
species variability might have affected outcomes.

In Kenya, Mihok et al. (1995) evaluated the influence of acetone, 1-octen-3-ol and various urines
and dungs on the catch of Stomoxys spp. and other Stomoxyinae (e.g. Haematobosca, Prostomoxys,
Rhinomusca, Stygeromyia) using Vavoua traps. Most of the test compounds did not affect trap
catch, though 1-octen-3-ol yielded mixed results and sometimes was associated with increased
(about 3-fold) catch, especially at high doses (2 mg/hr). In a similar study, Mihok et al. (1996)
did not find a consistent effect for 1-octen-3-ol. In two experiments it did not enhance catch of
Stomoxys compared to CO2 alone, and when released by itself caught less than half the number
of flies as a CO2-baited trap, which is similar to the outcome (Mihok et al. 1995) for a CO2-baited
Vavoua trap against an unbaited trap. However, catch of Haematobosca spp. were about 2-3 times
higher for one of the experiments in which CO2 was compared against CO2 and 1-octen-3-ol. In
work carried out in North America (Cilek 1999), release of 1-octen-3-ol from alsynite cylinders was
not associated with increased stable fly catch, though acetone (50 mg/hr) or a mixture of phenols
were. Finally, a series of studies carried out in North America and Africa (Mihok et al. 2007) using
Nzi traps yielded similarly mixed results, i.e. 1-octenol sometimes was associated with modest
increases in catch, and sometimes not. These same authors also reported that traps baited with
acetone (890 mg/hr) in Ethiopia caught about twice as many flies as unbaited traps, but increases
were not evident when acetone was combined with 1-octen-3-ol and/or cattle urine.

Mihok et al. (2007) put forward an interesting hypothesis to explain the variable results for 1-octen-
3-ol. Specifically that it is not efficacious when presented in combination baits or in areas where
livestock are near. We refer the reader to the paper for a fuller explanation, but do point out that
in addition to the exception noted by the authors (Cilek 1999), Vale and Hall (1985b) and Schofield
(1996) tested 1-octen-3-ol without cattle in the immediate vicinity, and did not observe increased
trap catch. Albeit, substantial numbers of wild ungulates were present around the test sites.
Moreover, even though they might not have been immediately adjacent to the test area, there was
a large (>100) herd of cattle at the research station in Zimbabwe, as well as wild ungulates, when
Holloway and Phelps (1991) carried out their studies. Several alternative, though not necessarily
mutually exclusive, explanations for this inconsistent response are possible. For one, even where
1-octen-3-ol has had demonstrable effects, catch increases have been rather modest (e.g. about
2-3 fold). Assuming a substantial random error term and a small expected difference between
treatment and control, experiments might have been underpowered for detecting differences.
Second, 1-octen-3-ol has been shown to have both positive (e.g. upwind anemotaxis) and negative
(activation, reduction in upwind movement, reduced trap catches) effects in the laboratory
and field, and exposure to high levels of 1-octen-3-ol have been associated with physiological
refractoriness in stable flies (Schofield et al. 1995). Given this variability, it would seem that there
is a fairly narrow window to capture ‘positive’ responses to 1-octen-3-ol, which might be manifest
as highly variable results for field work. Finally, we suggest that non-CO2 kairomones simply might
not be that important to host orientation in stable flies. Yes, they are active in the laboratory
and sometimes in the field, but their relative importance seems small compared to CO2. This last
perspective is supported by the work of Torr et al. (2006) who found that differences between
cattle in terms of ‘attractiveness’ to Stomoxys were reduced by standardising CO2 release.

Hence, the story for Stomoxys, though heavily influenced by tsetse work, turns out to be somewhat
different as regards the diversity of host-based kairomones that consistently evoke field responses.

Olfaction in vector-host interactions 301

Steve Schofield and Steve J. Torr

For tsetse, there are several kairomones (Chapter 12, this volume), for Stomoxys there is just one.
Of course, additional compounds might be identified in the future, and novel uses for these
compounds could be developed. However, as it stands, and likely for at least the near future,
there seems to be limited operational advantage to deploying non-CO2 odours, or arguably even
CO2, for the purpose of stable fly surveillance or control.

Feeding by Stomoxyinae: strategies and host selection

The work of Torr et al. (2006) described above highlights that conspecific hosts can vary in their
ability to draw Stomoxys to their vicinity, something that also has been demonstrated for other
blood feeding insects (Mukubana et al. 2002, Schofield and Sutcliffe 1996). Of course, once an
insect gets near to a host a variety of cues, perhaps including host odours, influence whether or
not feeding occurs. As it turns out, several groups of Stomoxyinae have received attention in this
regard. For example, the load of Haematobia irritans (L.) on cattle has been shown to be affected
by a variety of factors (e.g. breed, hair density, sebum, host size, host colour) (Jensen et al. 2004).
Similarly and as pointed out by Mullens et al. (2006), it has long been known that cattle show
substantial variation in the stable fly load that they support (see Cheng 1958, Mullens and Meyer
1987). Most of the work that critically examines this relationship has been published in the last
decade or so. The exception is Warnes and Finlayson (1987). These authors carried out experiments
in a laboratory arena and found that calves that displayed higher rates of defensive behaviour
had lower fly roads. Subsequent to this, and often in concert with studies examining tsetse,
several authors have considered the interrelationship between stable flies, their hosts and other
insects. Torr and Mangwiro (2000) evaluated interactions between tsetse feeding success, host,
and populations of other flies by observing individual cattle placed within an incomplete ring of
electric nets. Though not intended to focus on Stomoxys behaviour, the study was relevant because
in addition to documenting that tsetse feeding ‘success’ varied between hosts, related largely to
the host’s age, it linked this effect to the intensity of host defensive behaviour. Further, rather than
defensive behaviour primarily being regulated by tsetse load, the strongest association was with
the numbers of Stomoxys attacking the host. Hence, their suggestion that interspecific interaction
(driven by Stomoxys populations) might have important effects on feeding outcomes for tsetse,
and by extension could affect trypanosomiasis epidemiology. Of note, while this study considered
stable fly populations, it did not collect detailed information on their feeding behaviour. The
interplay suggested by Torr and Mangwiro (2002) found substantial support in a follow-up study
(Torr et al. 2007). In this case, per capita load of Stomoxys was associated with a significant and
negative reduction in tsetse feeding success, from about 80% at a per capita load of 20 flies to about
40% at a per capita load of 1000 flies (Figure 2). Though not assessed for Stomoxys specifically,
results from this study are also intriguing because they demonstrate the substantial heterogeneity
that is possible with respect to host selection, something that can have interesting and potentially
important implications for disease epidemiology. A more specific evaluation of Stomoxys feeding
behaviour was made by Schofield and Torr (2002) who carried out two experiments in which host-
defensive behaviour and feeding success was directly observed. Interestingly, most Stomoxys were
disturbed while feeding (73%), whether by other flies (65%) or by host-defensive behaviour (35%).
Moreover, feeding success varied with host (Figure 3) and, somewhat paradoxically, there was a
positive (though not significant) trend between Stomoxys density and increased feeding success.
This last observation runs contrary to the conventional wisdom that this interaction, if anything,
should be in the negative direction. The plausible explanation for this, which is somewhat unique
as regards the extant literature, is that the majority of Stomoxys were dislodged by non-biting
Diptera versus host-based defensive behaviour. Hence, increasing the number of Stomoxys on the
host would presumably reduce their per capita exposure to non-biting Diptera, and in turn might

302  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

A B
1.0 1.0

0.8 0.8

Feeding probability
Feeding probability

0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
100 1000 10 100 1000
Total daily catch Per capita daily catch
Figure 2. Feeding probability of Glossina plotted against (A) total daily catch of Stomoxys or (B) per capita catch
of Stomoxys for a single ox (solid dots) or a herd of 12 cattle (open circles). Line fitted by logistic regression
(adapted from Torr et al. 2007).

0.4

0.3
Feeding success

0.2

0.1

0.0
537 386 oxen calf cow
Stomoxys spp. Glossina spp.

Figure 3. Feeding success (±SE) of Stomoxys on two different adult oxen, designated by ear tags 537 and 386,
and feeding success of tsetse on different types of cattle in Zimbabwe (adapted from Schofield and Torr 2002).

increase the probability of a complete feed. In terms of the comparison of Stomoxys and tsetse,
the principal differences observed were that the former showed a higher rate of feeding success,
a lower relative rate of host-induced disturbance, a longer average period on the host (whether
or not disturbed) and a non-constant rate of leaving the host. The authors suggest that these
differences are compatible with the life attributes of these groups. On the one hand, Stomoxys are
rather short lived (daily survival has been estimated to be about 0.75-0.9 (Berry et al. 1981, Scholl
1986)), and feed daily or even more often. Hence, expectation is for a more persistent, ‘riskier’

Olfaction in vector-host interactions 303

Steve Schofield and Steve J. Torr

feeding strategy. On the other hand, tsetse are longer lived (daily survival >0.95; Hargrove 1988)
and likely feed only every ~3 days. Thus, at least compared to stable flies, they should be ‘choosier’
feeders. So, just as for kairomones, while some of the gross patterns of behaviour for these groups
are similar (e.g. host-selection heterogeneities), they show marked differences at a finer scale.

One of the most ambitious studies of interactions between stable flies and their hosts was
undertaken in California by Mullens et al. (2006). In this work, individuals in four herds of 25 cows
were observed 8-10 times per week for 12 weeks. For each observation period, the numbers of
flies and defensive behaviours were recorded. In addition to a very strong correlation (r2 >0.8)
between the weekly mean estimates for defensive behaviour and fly populations, the authors
were able to demonstrate that ranked fly load per animal was consistent over the experiment, as
was ranked intensity of defensive behaviour. Consistent with work for tsetse, older cows tended
to maintain higher fly loads and demonstrated a lower rate of effective defensive manoeuvres
(i.e. leg stamps) than younger animals. Finally, the per capita rate of defensive behaviour against
stable flies decreased with time, suggesting that animals became habituated to stable fly attack.
Though the authors were unable to characterise economically relevant impacts of fly attack (e.g.
decreased milk yield), they make the valid point that more work is required to relate economic
impact – to host behaviour – to biting fly pressure. Finally, they suggest that monitoring defensive
behaviour (instead of fly population) might provide a simpler, more efficient and sensitive indictor
for decision-making around pest management.

In contrast to the situation for distance-orientation kairomones, results from studies evaluating
close-range or on-host behaviours/interactions would seem to have direct operational relevance
to stable fly management. This might simply be in terms of developing efficient surveillance
indicators (Mullens et al. 2006), or more ambitiously could be articulated in the context of
developing more sustainable and/or cost-effective control strategies (see Mullens et al. 2006, Prior
2003, Torr et al. 2001, 2007).

Conclusion

While odour-based orientation to host animals is necessary to stable flies, it might not be readily
exploited from a control perspective. This is substantially different to the situation for at least
some species of tsetse, where research on and eventual use of host odour has powered successful
control programs (see Chapter 12 in this volume). The example of tsetse research also shows how
a better appreciation of the host-orientation behaviour of flies can enhance our understanding
of host-parasite interactions, which in turn might lead to improved methods of control (e.g.
insecticide-treated cattle to control tsetse, see Torr et al. 2007). Whether or not this also will be
the case for stable flies remains to be seen.

Aside from these practical considerations, the odour-oriented behaviour of Stomoxys and tsetse
provide many examples of where host selection is governed more by its defensive behaviour
than its kairomone profile. By contrast, much research on identifying ‘attractants’ and ‘repellents’
for mosquitoes and midges has focussed on comparing the odour profiles of individuals that are
bitten more or less (see other chapters in this book). The relative importance of odours in host
selection by different species of biting fly may be related, in part, to their respective activity
periods: day-active species can make better use of visual cues such as host -colour, -movement and
-shape whereas odour is, perhaps, the only cue that can be relied on to distinguish one sleeping
host from another. However, we also suggest that this might reflect an underlying assumption
that odours alone are responsible for inter-host variability in attractiveness. The cases of Stomoxys

304  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

and tsetse provide numerous examples where this is not the case. Perhaps we should be checking
that the same does not apply to other vectors?

References

Alzogaray RA and Carlson DA (2000) Evaluation of Stomoxys calcitrans (Diptera: Muscidae) behavioral response to
human and related odors in a triple cage olfactometer with insect traps. J Med Entomol 37: 308-315.
Anderson JR (1978) Effects of a blood meal on the mating drive of Stomoxys calcitrans males and its necessity as a
prerequisite for proper insemination of females. J Econ Entom 71: 379-386.
Beresford DV and Sutcliffe JF (2006) Studies on the effectiveness of coroplast sticky traps for sampling stable flies
(Diptera: Muscidae), including a comparison to Alsynite. J Econ Entomol 99: 1025-1035.
Beresford DV and Sutcliffe JF (2008) Male stable fly (Stomoxys calcitrans) response to CO2 changes with age: evidence
from wind tunnel experiments and field collections. J Vector Ecol 33: 247-254.
Berry IL and Campbell JB (1985) Time and weather effects on daily feeding patterns of stable flies (Diptera: Muscidae).
Environ Entomol 14: 336-342.
Berry IL, Scholl PJ and Shugart JI (1981) A mark and recapture procedure for estimating population sizes of adult stable
flies. Environ Entomol 10: 88-93.
Bidgood HM (1980) Host location in Stomoxys calcitrans (L.). Dissertation, University of Birmingham, Birmingham, UK.
Birkett MA, Agelopoulos N, Jensen KM, Jespersen JB, Pickett JA, Prijs HJ, Thomas G, Trapman JJ, Wadhams LJ and
Woodcock CM (2004) The role of volatile semiochemicals in mediating host location and selection by nuisance
and disease-transmitting cattle flies. Med Vet Entomol 18: 313-322.
Bursell E (1984) Effects of host odour on the behaviour of tsetse. Insect Sci Appl 5: 343-349.
Carlson DA, Alzogaray RA and Hogsette JA (2000) Behavioral response of Stomoxys calcitrans (Diptera: Muscidae) to
conspecific feces and feces extracts. J Med Entomol 37: 957-961.
Charlwood JD and Lopes, J. (1980) The age structure and biting behavior of Stomoxys calcitrans (Diptera: Muscidae)
from Manuas, Brazil. Bull Ent Res 70: 549-556.
Cheng T (1958) The effect of biting fly control on weight gain in beef cattle. J Econ Entomol 51: 275-278.
Chia LS, Baxter JA and Morrison PE (1982) Quantitative relationship between ingested blood and follicular growth in
the stable fly, Stomoxys calcitrans. Can J Zool 60: 1917-1921.
Cilek JE (1999) Evaluation of various substances to increase adult Stomoxys calcitrans (Diptera: Muscidae) collections
on alsynite cylinder traps in north Florida. J Med Entomol 36: 605-609.
Colvin J, Brady J and Gibson G. (1989) Visually-guided, upwind turning behaviour of free-flying tsetse flies in odour-
laden wind: a video study. Physiol Entomol 14: 31-39.
Cooperband MF and Cardé RT (2006) Orientation of Culex mosquitoes to carbon dioxide-baited traps: flight manoeuvres
and trapping efficiency. Med Vet Entomol 20: 11-26.
Daykin PN, Kellogg FE and Wright RH (1965) Host-finding and repulsion of Aedes aegypti. Can Entomol 97: 239-263.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972.
Gatehouse AG and Lewis CT (1973) Host location behaviour of Stomoxys calcitrans. Entomol Exp Appl 16: 275-290.
Griffiths N, Paynter Q and Brady J (1995) Rates of progress up odour plumes by tsetse flies: a mark-release video study
of the timing of odour source location by Glossina pallidipes. Physiol Entomol 20: 100-108.
Hansens EJ (1951) The stable fly and its effect on seashore recreational areas in New Jersey. J Econ Entomol 44: 482-487.
Hargrove JW (1988) Tsetse - the limits to population growth. Med Vet Ent 2: 203-217.
Hargrove JW, Holloway MTP, Vale GA, Gough AJE and Hall DJ (1995) Catches of tsetse flies (Glossina spp.) (Diptera:
Glossinidae) from traps baited with large doses of natural and synthetic host odour. Bull Ent Res 85: 215-227.
Harley JMB (1965) Seasonal abundance and diurnal variations in activity of Stomoxys and Tabanidae in Uganda. Bull
Ent Res 56: 319-332.
Holloway MTP and Phelps RJ (1991) The responses of Stomoxys spp. (Diptera: Muscidae) to traps and artificial host
odours in the field. Bull of Ent Res 81: 51-55.

Olfaction in vector-host interactions 305

Steve Schofield and Steve J. Torr

Hoy JB (1970) A recent test of Gambusia for mosquito control in a rice field. Proc Pap Annu Conf Calif Mosq Control
Assoc.38: 73-74.
Jeanbourquin P and Guerin PM (2007a) Sensory and behavioural responses of the stable fly Stomoxys calcitrans to
rumen volatiles. Med Vet Entomol 21: 217-224.
Jeanbourquin P and Guerin PM (2007b) Chemostimuli implicated in selection of oviposition substrates by the stable
fly Stomoxys calcitrans. Med Vet Entomol 21: 209-216.
Jensen KM, Jespersen JB, Birkett MA, Pickett JA, Thomas G, Wadhams LJ and Woodcock CM (2004) Variation in the load
of the horn fly, Haematobia irritans, in cattle herds is determined by the presence or absence of individual heifers.
Med Vet Entomol 18: 275-280.
Jones MDR and Gubbins SJ (1978) Changes in the circadian flight activity of the mosquito Anopheles gambiae in relation
to insemination, feeding and oviposition. Physiol Entomol 3: 213-220.
Jones CJ, Milne DE, Patterson RS, Schreiber ET and Milio JA (1992) Nectar feeding by Stomoxys calcitrans (Diptera:
Muscidae): effects on reproduction and survival. Environ Entomol 21: 141-147.
Kennedy JS (1939) The visual responses of flying mosquitoes. Proc Zool Soc Lond 109: 221a-242a.
Kennedy JS (1992) The new anthropomorphism. Cambridge University Press, Cambridge, UK.
Klowden MJ and Lea AO (1979) Effect of defensive host behaviour on the bloodmeal size and feeding success of natural
populations of mosquitoes (Diptera: Culicidae). J Med Entomol 15: 514-517.
Klowden MJ, Davis EE and Bowen MF (1987) Role of the fat body in the regulation of host-seeking behaviour in the
mosquito Aedes aegypti. J Insect Physiol 33: 643-646.
Kunz SE and Monty J (1976) Biology and ecology of Stomoxys nigra Marquart and Stomoxys calcitrans (L.) (Diptera:
Muscidae) in Mauritius. Bull Ent Res 66: 745-755.
Kuzina OS (1942) Gonotrophic relationships in Stomoxys calcitrans L. and Haematobia stimulans L. Med Parisitol 11:
70-78.
Lehane MF (2005) The biology of blood sucking insects. Cambridge University Press, New York, USA.
Lewis CT (1971) Superficial sense organs on the antennae of the fly, Stomoxys calcitrans. J Insect Physiol 17: 449-461.
Lewis CT (1972) Chemoreceptors in haematophagous insects. Behavioural aspects of parasite transmission. Zoot. J.
Linn. Soc. 51 Suppl. I: 201-213.
Mihok S, Carlson DA and Ndegwa PN (2007) Tsetse and other biting fly responses to Nzi traps baited with octenol,
phenols and acetone. Med Vet Entomol 21: 70-84.
Mihok S, Kang’ethe EK and Githaiga, KK (1995) Trials of traps and attractants for Stomoxys spp. (Diptera: Muscidae). J
Med Entomol 32: 283-289.
Mihok S, Maramba O, Munyoki E and Saleh K (1996) Phenology of Stomoxyinae in a Kenyan forest. Med Vet Entomol
10: 305-316.
Mukabana WR, Takken W, Coe R and Knols BG (2002) Host-specific cues cause differential attractiveness of Kenyan men
to the African malaria vector Anopheles gambiae. Malar J 1: 17.
Mullens BA and Meyer JA (1987) Seasonal abundance of stable flies (Diptera: Muscidae) on California dairies. J Econ
Entomol 5: 1039-1043.
Mullens BA, Lii KS, Mao Y, Meyer JA, Peterson NG and Szijj CE (2006) Behavioural responses of dairy cattle to the stable
fly, Stomoxys calcitrans, in an open field environment. Med Vet Entomol 20: 122-137.
Murlis J, Elkinton JS and Cardé RT (1992) Odour plumes and how insects use them. Ann Rev Entomol 37: 505-532.
Nayar JK and Sauerman DM (1971) The effect of light regimes on the circadian rhythm of flight activity in the mosquito
Aedes taeniorhynchus. J Exp Biol 54: 745-756.
Parr HCM (1962) Studies on Stomoxys calcitrans L. in Uganda, East Africa. II Notes on life history and behaviour. Bull Ent
Res 53: 437-443.
Paynter QE (1991) The effects of host odours on the flight behaviour of tsetse flies. Dissertation, Imperial College,
University of London, London, UK.
Paynter Q and Brady J (1993). Flight responses of tsetse flies (Glossina) to octenol and acetone vapour in a wind-tunnel.
Physiol Entomol 18: 102-108.

306  Olfaction in vector-host interactions

 13. Behavioural modalities of ‘non-vector’ biting Diptera: from olfaction to feeding

Paynter Q and Brady J (1996) The effect of wind speed on the flight responses of tsetse flies to CO2: a wind-tunnel study.
Physiol Entomol 21: 309-312.
Pickard E (1968) Stomoxys calcitrans (L.) breeding along TVA reservoir shorelines. Mosquito News 28: 644-646.
Prior, A.R. (2003) Written in blood: the use of microsatellite markers to study the feeding behaviour of haematophagous
Diptera. PhD thesis, University of Greenwich, Greenwich, UK. 240 p.
Roberts RH (1972) The effectiveness of several types of malaise traps for the collection of Tabanidae and Culicidae.
Mosquito News 32: 542-547.
Romero A, Broce A and Zurek L (2006). Role of bacteria in the oviposition behaviour and larval development of stable
flies. Med Vet Entomol 20: 115-121.
Schofield S (1996) Visual and olfactory responses of the stable fly. Dissertation, Imperial College, University of London,
London, UK.
Schofield S (1998) Field activity and seasonal changes in Stomoxys populations in Zimbabwe. Bull Ent Res 88: 627-632.
Schofield S and Brady J (1996) Circadian activity pattern in the stable fly, Stomoxys calcitrans. Physiol Entomol 21: 159-
163.
Schofield S and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol on the flight responses of the stable
fly, Stomoxys calcitrans, in a wind tunnel. Physiol Entomol 22: 380-386.
Schofield S and Torr SJ (2002) A comparison of the feeding behaviour of tsetse and stable flies. Med Vet Entomol 16:
177-185.
Schofield S, Cork A and Brady J (1995). Electroantennogram responses of the stable fly, Stomoxys calcitrans, to
components of host odour. Physiol Entomol 20: 273-280.
Schofield SW and Sutcliffe JF (1996) Human individuals vary in attractiveness for host-seeking black flies (Diptera:
Simuliidae) based on exhaled carbon dioxide. J Med Entomol 33: 102-108.
Schofield S, Witty C and Brady J (1997) Effects of carbon dioxide, acetone and 1-octen-3-ol flux on the landing responses
of the stable fly, Stomoxys calcitrans. Physiol Entomol 22: 256-260.
Scholl PJ (1986) Field population studies of Stomoxys calcitrans (L.) in Eastern Nebraska. Southwest Entomol 11: 155-160.
Schwinghammer KA, Knapp FW and Boling JA (1987) Physiological and nutritional response of beef steers to combined
infestations of horn fly and stable fly (Diptera: Muscidae). J Econ Entomol 80: 120-125.
Schwinghammer KA, Knapp FW, Boling JA and Schillo KK (1986) Physiological and nutritional response of beef steers
to infestations of the horn fly (Diptera: Muscidae). J Econ Entomol 79: 1010-1015.
Spates GE, DeLoach JR and Chen AC (1988) Ingestion, utilization and excretion of blood meal sterols by the stable fly,
Stomoxys calcitrans. J Insect Physiol 34: 1055-1061.
Sutcliffe JF (1986) Black fly host location: a review. Can J Zool 64: 1041-1053.
Sutcliffe JF (1987) Distance orientation of biting flies to their hosts. Insect Sci Appl 8: 611-616.
Torr SJ and Mangwiro TN (2000) Interactions between cattle and biting flies: effects on the feeding rate of tsetse. Med
Vet Entomol 14: 400-409.
Torr SJ, Della Torre A, Calzetta M, Costantini C and Vale GA (2008) Towards a fuller understanding of mosquito
behaviour: use of electrocuting grids to compare the odour-orientated responses of Anopheles arabiensis and An.
quadriannulatus in the field. Med Vet Entomol 22: 93-108.
Torr SJ, Mangwiro TN and Hall DR (2006) The effects of host physiology on the attraction of tsetse (Diptera: Glossinidae)
and Stomoxys (Diptera: Muscidae) to cattle. Bull Ent Res 96: 71-84.
Torr SJ, Prior A, Wilson PJ and Schofield S (2007) Is there safety in numbers? The effect of cattle herding on biting risk
from tsetse flies. Med Vet Entomol 21: 301-311.
Torr SJ, Wilson PJ, Schofield S, Mangwiro TN, Akber S and White BN (2001) Application of DNA markers to identify the
individual-specific hosts of tsetse feeding on cattle. Med Vet Entomol 15: 78-86.
Turner DA (1971) Olfactory perception of live hosts and carbon dioxide by the tsetse fly Glossina morsitans orientalis
Vanderplank. Bull Ent Res 6: 75-96.
Vale GA (1980) Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to carbon dioxide, acetone
and other chemicals. Bull Ent Res 70: 563-570.

Olfaction in vector-host interactions 307

Steve Schofield and Steve J. Torr

Vale G and Hall D (1985a). The role of 1-octen-3-ol, acetone and carbon dioxide in the attraction of tsetse flies, Glossina
spp. (Diptera: Glossinidae), to ox odour. Bull Ent Res 75: 209-217.
Vale G and Hall D (1985b). The use of 1-octen-3-ol, acetone and carbon dioxide to improve baits for tsetse flies, Glossina
spp. (Diptera: Glossinidae). Bull Ent Res 75: 219-231.
Venkatesh K and Morrison PE (1980) Some aspects of oogenesis in the stable fly, Stomoxys calcitrans (Diptera: Muscidae).
J Insect Physiol 26: 711-715.
Venkatesh K and Morrison PE (1982) Bloodmeal as a regulator of triacylglyceral synthesis in the haematophagous stable
fly, Stomoxys calcitrans. J Comp Physiol 147: 49-52.
Warnes ML and Finlayson LH (1985a) Responses of the stable fly, Stomoxys calcitrans (L.) (Diptera: Muscidae), to carbon
dioxide and host odours. I. Activation. Bull Ent Res 75: 519-527.
Warnes ML and Finlayson LH (1985b) Responses of the stable fly, Stomoxys calcitrans (L.)(Diptera: Muscidae), to carbon
dioxide and host odours. II. Orientation. Bull Ent Res 75: 717-727.
Warnes ML and Finlayson LH (1986) Electroantennogram responses of the stable fly, Stomoxys calcitrans, to carbon
dioxide and other odours. Physiol Entomol 11: 469-473.
Warnes ML and Finlayson LH (1987) Effect of host behaviour on host preference in Stomoxys calcitrans. Med Vet Entomol
1: 53-57.
Warnes ML (1990) The effects of host odour and carbon dioxide on the flight of tsetse flies (Glossina spp.) in the
laboratory. J Insect Physiol 36: 607-611.
Williams B (1994) Models of trap seeking by tsetse flies: anemotaxis, klinotaxis and edge detection. J Theor Biol 168:
105-115.
Zumpt F (1973) The stomoxyine biting flies of the world. Gustav Fischer Verlag, Stuttgart, Germany.

308  Olfaction in vector-host interactions

14. The role of olfaction in host seeking of triatomine bugs
Pablo G. Guerenstein and Claudio R. Lazzari

Abstract

Triatomine bugs (Hemiptera: Reduviidae) are hemimetabolous insects that depend on a blood
meal along their entire life cycle. They live in close association with their vertebrate hosts,
including humans. Triatomines are the vectors of Chagas disease in the Americas. They transmit
the protozoan parasite Trypanosoma cruzi during or immediately after feeding by depositing
their faeces on a host. Olfactory cues play a major role in host seeking in triatomines. Some of
the odorants to which the odour receptor cells respond to are already known. Most of those
odorants (known to also be detected by other haematophagous insects) are of vertebrate origin,
and activate odour receptor cells in antennal grooved pegs and basiconic sensilla. Behavioural
responses (activation and/or attraction) to single odorants often occur mainly at concentrations
close to those that represent the response threshold for the related olfactory receptor cell.
Information about relevant single odorants (e.g. CO2) is used to build synthetic odour mixtures
capable of attracting the bugs efficiently. Thus, a few odour mixtures that evoke higher attraction
than that evoked by its odorant constituents when presented singly (i.e. showing additive or
synergistic effects) have been developed. L (+) lactic acid plays a fundamental role in enhancing
the attraction to those mixtures. Triatomine bugs are nocturnal, and the behavioural response to
host odours is temporally modulated. Thus, behavioural responsiveness to CO2 is controlled by a
circadian clock, and is limited to the early night, the moment when the bugs leave their refuges to
search a host. A prototype of odour baited trap that exploits a particular behaviour of triatomines
has been developed. However, more research is needed in order to achieve a practical, cheap and
efficient odour lure for a trap able to accomplish the necessary task of surveillance of the bugs
in the field.

Keywords: attraction, chemoreception, modulation, odour bait, olfactory sensilla, orientation,

Rhodnius, Triatoma

Introduction
Triatomine bugs, also named kissing bugs, cone-noses or assassin bugs, are Heteropterans
belonging to the family Reduviidae, subfamily Triatominae. They are hemimetabolous insects with
a life cycle that lasts from a few months to a year depending on the species, and that includes
five wingless nymphal (larval) stages followed by a winged adult stage. Triatomines are obligate
haematophagous along their entire life cycle, as nymphs, males and females feed on the blood
of vertebrates. Most of their hosts are warm-blooded vertebrates (birds and mammals), but some
species live in association and feed on ectothermic animals or even are able to suck haemolymph
from other insects (e.g. Almeida et al. 2000).

Around 120 species of triatomines have been described in the New World, half of them capable
of transmitting the flagellate protozoan Trypanosoma cruzi Chagas, the causative agent of Chagas
disease, which menace about 28 million people living in Latin America and accounts for 670,000
disability-adjusted life years per annum (DALY; Tarleton et al. 2007). The epidemiologically most
important species are Triatoma infestans Klug (Figure 1), T. dimidiata Latreille, and Rhodnius prolixus
Stål. About 13 species are known in the Old World although they are not vectors of Trypanosoma

Olfaction in vector-host interactions 309

Pablo G. Guerenstein and Claudio R. Lazzari

Figure 1. An adult female of Triatoma infestans, an epidemiologically important species.

cruzi. They are endemic in southern China, in Malaysia, Indonesia, New Guinea, northeastern
Australia and India (Lent and Wygodzinsky 1979).

Taking a blood meal could take up to 30 min, according to the species, and such a meal most
often represents several times their initial body weight. The bite is often painless and, if the host
is infected with Trypanosoma cruzi, the bug may acquire the infection with its meal. The parasite
then colonises the hindgut of the insects. The bugs can transmit the parasite to a vertebrate
host when infected excrements (faeces) are left over the host’s skin. The parasite can then enter
the host’s bloodstream through the skin or mucous membrane, such as that of the mouth if the
insect or its dejections are ingested. An important factor affecting the vectorial capacity of a given
triatomine species is the time elapsed between feeding and defecation. Those species releasing
excrements during or just after feeding increase the probability of parasite transmission.

The importance of triatomines as disease vectors was first established by the Brazilian scientist
Carlos Chagas in 1909, who discovered the first human cases of a hitherto unknown trypanosomiasis.
Chagas proved that the vector of the causative agent of the disease, Trypanosoma cruzi, was an
insect of the order Hemiptera, Panstrongylus megistus Burmeister (Lent and Wygodzinky 1979).
Further research on the vectors revealed that the parasite is actually present in several species
of the subfamily Triatominae found in and around human dwellings as well as in those living in
sylvatic habitats. The finding that armadillos, opossums and other mammals are naturally infected
with the parasite made clear that Chagas’ disease is a zoonosis, and that the secondary adaptation
of some vector species to human habitations resulted in the domestic cycle of the parasite (Lent
and Wydgodzinsky 1979). No vaccine against the parasite responsible for Chagas’ disease is at

310  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

present available. Control of this disease depends on the surveillance and chemical elimination
of the triatomine vectors (by applying insecticides), screening of blood banks and detection of
parasites in pregnant women, followed by treatment of the newborns from infected mothers.
Being the presence of the vectors strongly associated to the precarious structure of houses in less
favoured rural areas of endemic countries, house improvement to reduce the number of potential
refuges (e.g. wall crevices) for bugs constitutes is another way to reduce transmission.

Regional coordinate actions against Chagas disease started in 1992 by the Southern Cone
Initiative, signed by Brazil, Argentina, Bolivia, Chile, Paraguay, Uruguay, and Peru. It was followed
by two similar multinational initiatives against Chagas disease in Central America and the Andean
Pact region, both launched in 1997, and a large-scale Chagas disease surveillance initiative for the
countries of the Amazon basin launched in 2002 (Schofield and Kabayo 2008).

The surveillance of the arrival to houses and the detection of bugs present in low-density are
important components of control strategies. So, the possibility to exploit olfactory responses
to actively attract bugs into traps or detection devices becomes relevant (e.g. Guerenstein et al.
1995), as well as the characterisation of their chemosensory system and olfactory responses.

Most triatomine bugs live in close association with their vertebrate hosts, usually inside human
housings, in the case of the most antropophilic species, or in the nests or burrows of marsupials,
edentates, rodents, carnivores, bats, and birds. Triatomines are also found among rocks and
in stone walls, where they feed on small rodents and eventually on iguanas and other lizards
that share their habitat. In addition, the insects are found under fallen logs and in hollow trees,
among exposed roots and under loose bark, in palm fronds and in epiphytic bromeliads (Lent and
Wygodzinsky 1979). Many triatomines have become peridomestic, occurring in stables and corrals,
in rabbit and guinea pig houses, and in chicken and pigeon coops. From an epidemiological point
of view, the most important habitat of triatomine bugs is human dwellings, in particular in rural
areas of Latin America. Thus, triatomines can be classified into sylvatic and domestic species, with
an intermediate category of peridomestic species. Inside houses, during daytime, triatomines hide
in dark and sheltered places such as wall cracks, furniture, trunks, suitcases, and boxes filled with
old clothes or papers. They are found among curtains and clothes hanging from the walls, behind
pictures or calendars, in various types of roofing materials, such as thatch, palm fronds, or even
loose shingles, among mattresses and in cots and beds of all kinds, in hammocks and cradles, and
in woodpiles (Lent and Wygodzinsky 1979).

Bugs spend daytime in akinesis, inside their ‘refuges’ in the neighbourhood of their hosts,
presumably to avoid detection by predators, including their active hosts. The akinesis behaviour
is controlled and maintained by external cues, such as aggregation arresting pheromones and
tactile stimulation by physical contact with the substrate and with conspecifics (thigmotaxis),
in combination with an internal motivational state of inactivity, controlled by a circadian clock.
During the early night, the internal circadian clock determines that one to two hours after dusk,
when their hosts are resting, the bugs become active (Lazzari 1992), and leave their refuges
searching for cues revealing the presence of a potential host to feed on.

Once a blood meal has been obtained, bugs walk away from the host, but do not return to
their refuge immediately (Lorenzo and Lazzari 1998). The time spent outside just after feeding
would allow bugs to complete most of their copious diuresis that follows a blood meal. At dawn,
they return to their shelters guided by chemical cues present in their faeces, which are actively
deposited around the refuge accesses (Lorenzo and Lazzari 1996). Chemicals present in the

Olfaction in vector-host interactions 311

Pablo G. Guerenstein and Claudio R. Lazzari

faeces constitute one of the aggregation pheromones of triatomines (Lorenzo Figueiras et al.
1994, Schofield and Patterson 1977). A second one, associated with the cuticle, is left behind on
surfaces over which the bugs walked (Lorenzo Figueiras and Lazzari 1998). The two pheromones,
together with thigmotaxis, cause the insects to remain inactive inside their refuge during daytime.

Olfactory physiology: detection of host odours

The antenna of triatomines consists on four segments, a proximal scapus, a pedicelus and two
flagellar articles. All segments bear sensilla, hair-like cuticular structures that house sensory
cells of different modalities, including chemoreception, mechanoreception, thermoreception
and hygroreception (Barth 1952, 1953, Catalá 1994, Catalá and Schofield 1994, Diehl et al.
2003, Guerenstein and Guerin 2001, Mayer 1968, Bernard 1974, Insausti et al. 1999, Justo and
Tramezzani 1977, McIver and Siemicki 1984, 1985, Lazzari and Wicklein 1994, Taneja and Guerin
1997, Wigglesworth and Gillet 1934).

Triatomine chemoreceptor cells specialised in odour detection (odour receptor cells, ORCs), and
contained in olfactory sensilla, are the best studied. Thus, the odour ‘tuning’ of many of them is
known. To date, the structure of three morphologically distinguishable types of olfactory sensilla
has been described on the antennae of triatomines. Those sensilla are the grooved pegs (GP; Diehl
et al. 2003, Guerenstein and Guerin 2001, Taneja and Guerin 1997, named basiconica in Catalá
1997), the basiconic sensilla (BS, Guerenstein and Guerin 2001, named trichoid thin-walled in
Catalá 1997) and the trichoid sensilla (Guerenstein 1999, named trichoid thick-walled in Catalá
1997). The latter are 35 µm long single (thick)-walled wall-pore sensilla (Bernard 1974). They are
the most numerous olfactory sensilla on the adult antenna and yet the odour tuning of its ORCs
(1-2 in T. infestans nymphs, Bernard 1974; 5-6 in R. prolixus adults, Wigglesworth and Gillet 1934) is
unknown because no olfactory stimulus tested up to date evoked a response from them (Bernard
1974, PG Guerenstein and P Guerin, unpublished results). An olfactory function is attributed to
this sensillum type because of its morphological characteristics. On the contrary, the odour tuning
and physiology of some ORCs in the other two types of olfactory sensilla have been studied in
detail (see below).

To study the odour tuning of identified ORCs, single sensillum recordings (SSRs) are performed
on different, individual sensilla. A number of monomolecular odours (odorants), individually
placed in odour cartridges, can then be screened and the responses to them recorded for
analysis. Alternatively, a headspace extract of a natural odour source (e.g. a host odour) can be
prepared. This extract can then be injected into a gas chromatograph (GC) that is coupled to the
antennal preparation. As the constituents of the extract elute from the GC column, the bioactive
odorants are detected by the insect ORCs as recorded by means of a SSR, a technique named
gas chromatography coupled to single sensillum recording (GC-SSR; e.g. Guerenstein and Guerin
2001, see also Chapter 3 in this volume).

The morphology and ultrastructure of the GPs of 5th instar nymphs of T. infestans have been
studied in detail (Diehl et al. 2003). They are non-articulated, double-walled wall-pore sensilla, 8-18
µm long, with a diameter of about 2.5 µm at their base. The wall pores in the 13-18 longitudinal
grooves communicate to the inner sensillum lymph cavity through radial spoke channels. These
sensilla house four (ca. 54% of GPs) or five receptor cells with an unbranched outer dendritic
segment that can reach the tip of the sensillum. Areas of direct contact between cell membranes of
the soma of the receptor cells are found, and may represent gap junctions. This suggests possible
interaction between the responses of the sensory cells. Some pegs have a terminal pore whereas

312  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

others do not. To date, the proportion of GPs with a terminal pore has not been established and,
moreover, it is unknown if this pore has any functional role. Fifth instar nymphs posses 80-110 of
such sensilla per antenna (Diehl et al. 2003).

As in other insects (e.g. Altner et al. 1977), the ORCs of this type of sensillum respond to polar
odorants. According to the odour tuning of their ORCs, different functional types of GPs are found
(Table 1). In 5th instar nymphs GP1 represents ca. 36% of the total GP population, whereas GP2
and GP3 represent 10 and 43% respectively. The receptor cells of 11% of the GPs did not respond
to any of the olfactory stimuli tested so far (Diehl et al. 2003). At least one of the unresponsive
receptor cell types is a hygroreceptor cell (Bernard 1974).

All odorants that evoke responses in ORCs of GPs are known to be emitted by vertebrate hosts.
Thus, short-chain carboxylic acids are constituents of vertebrate odour (see Guerenstein and
Guerin 2001), including human sweat (e.g. Cork and Park 1996). GC-SSR experiments using the
antenna of T. infestans nymphs allowed the identification of one of them, isobutyric acid, as an
active constituent of the headspace extract of rabbit odour (Guerenstein and Guerin 2001). In
addition, ammonia and short-chain aliphatic amines have been identified in vertebrate effluvia,
including sweat and urine odour (see Diehl et al. 2003, Taneja and Guerin 1997).

Our knowledge about the morphology and physiology of BS is more limited. As for GPs, data were
collected mainly from 5th instar nymphs of T. infestans. BS are non-articulated, single (thin)-walled
wall-pore sensilla, about 26 µm long, with a diameter of about 2 µm at their base (Guerenstein and
Guerin 2001). The numerous wall-pores are linked to pore tubules (Bernard 1974). These sensilla
house a highly variable number of receptor cells, from 21 to 41, with extensive dendritic branching
in the sensillar lumen (Guerenstein and Guerin 2001).

Table1. Odorants that evoke a response in grooved peg sensilla of 5th instar T. infestans. Best stimuli are indicated
whereas less effective stimuli are in parentheses. Responses consist on an increase in spike rate unless otherwise
indicated. Data based on Bernard (1974), Diehl et al. (2003), Guerenstein and Guerin (2001) and Taneja and
Guerin (1997).

GP1 GP2 GP3

ORC1 ammonia (methyl-, isobutyric acid (butyric, C4-5 aliphatic aminesb

(high amplitude spike) dimethyl-, ethyl-amine) isovaleric, 2-methylbutyric
acids)
ORC2 -a ammonia (methyl-, -a
(intermediate amplitude spike) dimethyl-, Ethyl-amine)
ORC3 isobutylamine isobutylamine isobutylamine
(low amplitude spike) (isopentylamine) (2-butylamine, butylamine, (2-butylamine,
3-methyl 2-butylamine, butylamine, 3-methyl
isopentylamine) 2-butylamine,
isopentylamine)

a No odorant tested up to date evoked a response.

b Response consists on a decrease in spike frequency.

Olfaction in vector-host interactions 313

Pablo G. Guerenstein and Claudio R. Lazzari

Experiments with GC-SSR using sheep odour indicated that BS posses an ORC highly sensitive to
aliphatic aldehydes, responding maximally to heptanal, octanal and nonanal (Guerenstein and
Guerin 2001). Such an ORC is also present in BS of 1st and 5th instar nymphs and adults of T.
infestans and in 5th instar nymphs and adults of R. prolixus and Dipetalogaster maxima (Uhler)
(Guerenstein 1999). In addition, screenings of odorants in BS ORCs of 1st and 5th instar nymphs
of T. infestans indicated cells selectively sensitive to terpenes (Guerenstein 1999). One of these
ORCs responds maximally to α(+) pinene whereas another responds to both α(+) pinene and (-)
limonene. Finally, an ORC that responds to terpinen-4-ol and α(+) terpineol is also present. Sensilla
basiconica of adult T. infestans possess an ORC sensitive to the heterocyclic-aromatics pyridine
and furan (Mayer 1968).

As in the case of GPs, most odorants that evoke responses in ORCs of BS are of vertebrate origin.
Thus, C7-C9 saturated aliphatic aldehydes occur in different vertebrate odours including chicken
feather odour (Guerenstein and Guerin 2001), and human effluvia (e.g. Preti et al. 1977). In addition,
α(+) pinene, (-) limonene, and α(+) terpineol have been identified, for example, in chicken feather
and human urine odour (see Guerenstein 1999). On the other hand, terpinen-4-ol appears not
to be emitted by vertebrates nor by triatomines, and is known as a constituent of the male sex
pheromone of some heteropterans (Aldrich 1995).

It should be noted that in adults, the spatial pattern of chemosensory sensilla in the antennal
pedicelus shows differences between species or even populations of the same species. Thus, for
example, in the tribe Triatomini, the density and number of types of chemosensory sensilla in this
antennal segment is positively correlated with the habitat range of the species (Carbajal de la
Fuente and Catalá 2002, Catalá 1997).

Olfactory behaviour: host-seeking

Olfactory cues play a major role in host seeking in triatomines although they are not the only
relevant sensory cues. The heat emitted by the host body constitutes a main orienting cue for
triatomines, whose thermal sense and their use of thermal cues have been the object of diverse
studies (e.g. Ferreira et al. 2007, Flores and Lazzari 1996, Guerenstein and Lazzari 2009, Lazzari
and Núñez 1989a,b, Wigglesworth and Gillett 1934). The behavioural sensitivity of triatomines to
heat is extremely high and they remain the only group of blood-sucking insects where the ability
to perceive the infrared radiation emitted by the host body has been demonstrated (Lazzari and
Núñez 1989a, Schmitz et al. 2000).

As mentioned in the introduction, triatomines are nocturnal and stay inactive, hidden in their
refuges, during daytime. Shortly after dusk the bugs become active. Around that time their hosts
are asleep, emitting volatile odorant blends released by their skin and breath that are dispersed
by air currents. Some of these chemicals activate nearby bugs, which increase their locomotion
activity (e.g. Núñez 1982), thus, increasing the probability of encountering other host-associated
cues after leaving their refuge. The bimodal stimulation of antennal receptors by mechano-
(air movement) and chemical (odours) stimuli triggers long-range, positive odour-modulated
anemotaxis, which here we also refer to as upwind orientation or attraction (e.g. Barrozo and Lazzari
2004a, Taneja and Guerin 1995, see also Chapter 6 in this volume). Provided that airstreams pass
the host’s body before reaching the bugs, they carry the host odours. Then, just walking upwind
(positive anemotaxis) represents a simple solution to reach the host. As the bug approaches the
host, cues of additional modalities (heat, moisture) are detected and integrated at the CNS level,
assigning an appetitive value to the complex, multimodal signal.

314  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

The behavioural response of triatomines to a number of odorants that are constituents of natural
odour blends, has been recorded using different experimental devices including an olfactometer
(e.g. Núñez 1982) and servospheres (e.g. Barrozo and Lazzari 2004a, Taneja and Guerin 1995). A
servosphere (also called locomotion compensator) is an experimental device in which a walking
insect is constantly kept at the apex of a sphere. The insect’s displacements are indirectly tracked
by sensors that detect the displacements of the sphere, which moves as a result of the walking
activity of the insect. Thus, parameters related to activation, such as walking speed and distance
walked, can be measured. In addition, attraction can be assessed by recording the walking angles
with respect to an airstream carrying the odour stimulus, and the percent displacement in the
upwind direction. In a servosphere the insect cannot change its position with respect to the
stimulus (i.e. it is an ‘open loop’ device). Two models of servosphere were used in triatomine
research, the ‘Kramer sphere’ (Kramer 1976), and the ‘Dahmen sphere’ (Dahmen 1980, modified
by Lazzari). The two models differ technically but measure virtually the same parameters. In the
case of triatomines, two different odour delivery systems were used in servosphere experiments.
Thus, one (e.g. Taneja and Guerin 1995) or two (e.g. Barrozo and Lazzari 2004a) airstreams reached
the walking insect. In the former case (used with the Kramer sphere), the control consisted of the
response of the bugs towards clean air delivered before and after odour delivery. In the latter
(used with the Dahmen sphere), one of the streams carried the test stimulus whereas the other
(at 180° from the first) represented an odourless, simultaneous control.

Different odorants evoke activation and/or attraction on triatomines even when presented singly.
One such odorant is carbon dioxide (CO2) (Guerenstein and Hildebrand 2008). It is not completely
understood if CO2 activates triatomines per se because in different studies, using the different
experimental set-ups mentioned above, CO2 stimuli evoked little (Guerenstein and Guerin 2001,
Taneja and Guerin 1995) or no (Barrozo and Lazzari 2004a, Núñez 1982) activation of the bugs. In
any case, a strong activating effect has never been reported. On the contrary, a clear-cut oriented
response of the bugs (nymphs and adults) towards a CO2 source has been shown (Barrozo and
Lazzari 2004a, Guerenstein and Guerin 2001, Núñez 1982, Taneja and Guerin 1995, see also
Trapping of triatomines using host-odour cues). Thus, the mean walking angle was statistically
not different from the direction of the CO2 stimulus. On the Dahmen servosphere the response
threshold falls between 300 and 400 ppm above ambient CO2 background (see also below for
more information on threshold), whereas the response increases with concentration up to 2,300
ppm, the highest level tested on that device. For these tests the CO2 level in the control airstream
was ambient CO2 (Barrozo and Lazzari 2004a).

Nymphs of T. infestans are able to orientate (walk) towards a CO2 source under continuous
stimulation (Barrozo and Lazzari 2006). This is in contrast to CO2-tracking in other, flying,
haematophagous insects like mosquitoes, which are able to follow a CO2 plume only if this
odorant is presented in an intermittent (pulsed) manner (e.g. Geier et al. 1999). Thus, mosquitoes
make use of the high-frequency intermittent pattern of CO2 presentation within a natural host
odour plume to orientate towards a CO2 source. In triatomines, however, pulsed stimulation of CO2
evokes repellency when delivered at a frequency of 1 Hz (Barrozo and Lazzari 2006). The reason
why T. infestans bugs evolved to orientate within a continuous plume may be related to the fact
that in their usually domestic habitat, CO2 may disperse (and be encountered by walking bugs) in
a more homogeneous way (Lehane 2005). To attempt to clarify this, it would be interesting to test
the orientation towards CO2 in the sylvatic triatomine species.

Olfaction in vector-host interactions 315

Pablo G. Guerenstein and Claudio R. Lazzari

CO2 is sufficient but not necessary (Barrozo and Lazzari 2004a,b, Guerenstein and Guerin 2001,
Núñez 1982, Taneja and Guerin 1997) to evoke an attractive behavioural response in triatomines,
similarly to some mosquitoes (e.g. Bernier et al. 2003, Bosch et al. 2000, Gibson and Torr 1999).

Another host odorant that evokes a behavioural response per se is nonanal (see Olfactory
physiology: detection of host odours). Nonanal, at 8 ppb in ambient air, activates (increases
walking speed) but does not attract triatomine nymphs on the servosphere (Guerenstein and
Guerin 2001). Lower and higher doses do not evoke any response. The activating concentration
of 8 ppb is just a few ppb under the estimated response threshold of the aldehyde ORC in BS
(Guerenstein and Guerin 2001).

In addition, ammonia (see Olfactory physiology: detection of host odours) both activates (increases
walking speed) and attracts triatomines at doses of 3-17 ppb but not at a lower dose (Taneja and
Guerin 1997). The behavioural threshold of 3 ppb is just one ppb above the estimated response
threshold of the ammonia ORC in GP1 (Taneja and Guerin 1997), which is similar to that of the
ammonia ORC in GP2 (Diehl et al. 2003).

Other odorants, isobutyric acid and 1-octen-3-ol, both constituents of human sweat odour
(Cork and Park 1996), also evoke a response, attraction, when presented singly to nymphs on a
servosphere (Guerenstein and Guerin 2001, Barrozo and Lazzari 2004a). In the case of 1-octen-3-ol,
five concentrations in log-steps were tested and yet attraction is observed only at an intermediate
dose (Barrozo and Lazzari 2004a). Attraction to isobutyric acid is obtained at doses (0.02-10
ppb in air) that evoke low to moderate (<40 Hz) increases in the firing rate of the ORC excited
by carboxylic acids in GP2 sensilla, whereas at a dose (100 ppb) that evokes higher firing rates
(>40 Hz) the bugs are not attracted (Guerenstein and Guerin 2001). As in the case of nonanal,
the behavioural threshold of 0.02 ppb for isobutyric acid is just under the estimated response
threshold of the carboxylic acid ORC in GP2 (Guerenstein and Guerin 2001). It should be mentioned
that isobutyric acid has also been identified as a main constituent of the alarm pheromone of
triatomines (e.g. Cruz López et al. 1995, Guerenstein and Guerin 2004, Manrique et al. 2006, Rojas
et al. 2002, Schofield 1979a,b). Thus, a parsimonious use of this odorant by these insects has been
proposed (Guerenstein and Guerin 2001, 2004, Guerenstein and Lazzari 2009) because it is used by
triatomines as, at least, host attractant and alarm signal depending on its concentration, sensory
(in particular odour blend) context and the physiological state of the perceiving bug. It should be
noted that odorants tested singly are often effective only within a narrow range of concentrations
(Barrozo and Lazzari 2004a, Guerenstein and Guerin 2001).

Testing odorants singly is important to assess the significance of a blend constituent per se, and
as a control for any enhanced (additive, positive-synergistic) or diminished response to any odour
mixture that includes that constituent. However, in nature, the insects are confronted to blends
of odours consisting of several constituents presented simultaneously at particular proportions.
Identifying those blends correctly should help increase the certainty that the odour source is the
expected host (Lehane 2005). In fact, natural odour blends and even some synthetic mixtures
developed attract triatomines more efficiently than single odorants (e.g. Barrozo and Lazzari
2004b, Taneja and Guerin 1995). A list of synthetic mixtures that evoke higher attraction than
that evoked by its odorant constituents per se is provided in Table 2.

A number of other mixtures tested elicited no attraction, evoked attraction at levels not higher
than those evoked by the odorants when presented singly or lacked the controls of its odorant
constituents tested singly to assess if the response to the mixture was different to that to the

316  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

Table 2. Synthetic odour mixtures that evoke higher attraction than that evoked by its odorant constituents when
presented singly (i.e. additive or synergistic effects observed). Tests carried out with T. infestans nymphs (unless
otherwise indicated).

Mixture Mixture constituentsa Increased attraction References

1 L-LA + CO2 yes c Núñez 1987b, Barrozo and Lazzari 2004a

2 L-LA + C3 yes c
L-LA + C4 Barrozo and Lazzari 2004b
L-LA + C5
3 L-LA + C3+ C4+ C5 yes c Barrozo and Lazzari 2004b
4 L-LA + C3+ C4+ C5+CO2 yes c,d Barrozo and Lazzari 2004b

a L-LA: L (+) lactic acid; C

3-6: n-aliphatic carboxylic acids.
b Tested on nymphs of R. prolixus.
c Synergistic effect.
d Attraction similar to that evoked by mouse odour.

constituents (Barrozo and Lazzari 2004a,b, Guerenstein and Guerin 2001, Otálora-Luna et al. 2004).
The mixtures that do not evoke attraction higher than that elicited by their odorant constituents
when presented singly include L (+) lactic acid (L-LA) + C6 n-aliphatic carboxylic acid, C3+ C4+
C5 n-aliphatic carboxylic acids, C3+ C4+ C5 n-aliphatic carboxylic acids + CO2, isobutyric acid +
nonanal, isobutyric acid + ammonia, isobutyric acid + CO2, and 1-octen-3-ol + CO2. At present,
results from mixtures that evoked no additive or synergistic response are of limited value because
only one or a limited number of proportions and concentrations have been tested.

The synergistic effect between L-LA (a human skin odorant; e.g. Cork and Park 1996) and CO2 in
mixture 1 of Table 2 is evinced when measuring the threshold of the oriented response towards
CO2. Thus, this threshold diminishes from 300-400 ppm when CO2 is tested singly (see above) to
75-150 ppm when presented together with L-LA, which does not evoke a response per se (Barrozo
and Lazzari 2004a). On the other hand, when testing different doses of L-LA with a fixed level
of CO2 (300 ppm), only the intermediate dose out of five tested evokes attraction (Barrozo and
Lazzari 2004a). In mosquitoes, a mixture of L-LA and CO2 also acts in a synergistic way (e.g. Geier
and Boeckh 1999).

Mixtures of L-LA plus C3, C4 or C5 n-aliphatic carboxylic acids (mixtures 2 in Table 2) also evince a
synergistic effect as none of those odorants evoke attraction when presented singly, even when
those carboxylic acids (present in human sweat and skin odour; Bernier et al. 2000, Cork and Park
1996) have been tested at five different doses, including the same and higher concentrations than
that in the mixture (Barrozo and Lazzari 2004b). In addition, a mixture of L-LA plus C3, C4 and C5
n-aliphatic carboxylic acids (mixture 3 in Table 2), at a particular odorant ratio and concentration,
is attractive even when that C3-5 mixture is not, again showing synergistic attraction (Barrozo and
Lazzari 2004b). Moreover, when the latter mixture is combined with CO2 (mixture 4 in Table 2), an
additional synergistic effect is revealed, making that mixture as attractive as a mouse, when tested
on the servosphere (Barrozo and Lazzari 2004b). This is remarkable considering that the positive
control (a mouse) is a source of a plethora of sensory cues (including heat, water vapour and a
complete, natural, odour blend) and triatomines are able to detect many of them (Guerenstein
and Lazzari 2009).

Olfaction in vector-host interactions 317

Pablo G. Guerenstein and Claudio R. Lazzari

It should be noted that all mixtures in Table 2 include L-LA but not all of them have CO2 as
constituent. This emphasizes the role of L-LA in synergistic interactions between odorant
constituents (Barrozo and Lazzari 2004a,b). Moreover, as L-LA is not attractive by itself, its role in
triatomine host seeking seems to be limited to enhance the response to odour mixtures, a role
similar to that found in Anopheles gambiae Giles mosquitoes (Smallegange et al. 2005).

Table 3 includes the full list of host odorants known to be detected by triatomines to date,
according to neurophysiological and behavioural data. Most of those odorants are known to be
also detected by other haematophagous insects (Guerenstein and Lazzari 2009).

Trypanosoma cruzi, the parasite that triatomines transmit, is considered subpathogenic for the
vectors. Thus, the parasite affects the capacity to resist starvation but neither the developmental
nor the mortality rate (Vallejo et al. 2009). However, changes in the behaviour of infected bugs
also occur. Thus, it has been suggested that bugs infected with T. cruzi detect a host more rapidly
than uninfected insects (Botto-Mahan et al. 2006). To date, it is unclear if the olfactory system is
involved in this behavioural change. Other changes reported as induced by parasite infection
include a higher biting rate (Botto-Mahan et al. 2006).

Table 3. Host odorants known to be detected by triatomines. Both behavioural and neurophysiological data were
considered to develop this list.

Odorant Reference

CO2 Guerenstein and Hildebrand 2008, Guerenstein and

Lazzari 2009 (reviews)
C7-9 n-aliphatic aldehydes Guerenstein and Guerin 2001
1-octen-3-ol Barrozo and Lazzari 2004a
L(+) lactic acid Núñez 1987, Barrozo and Lazzari 2004a
isobutyric acid and C3-5 n-carboxylic acids (and C6 Guerenstein and Guerin 2001, Diehl et al. 2003, Barrozo
n-carboxylic acid)1 2003, Barrozo and Lazzari 2004b
ammonia, (ethylamine, dimethylamine)1 Taneja and Guerin 1997, Diehl et al. 2003
C4-5 aliphatic amines as isobutylamine1 Diehl et al. 2003
terpenes as α(+) pinene and terpinen-4-ol1 Guerenstein 1999
pyridine, furan1 Mayer 1968
amyl acetate1 Bernard 1974

1 No behavioural response known (only neurophysiological evidence).

The modulation of olfactory responses

The behaviour of an animal, for instance its response to environmental cues, does not only depend
on exogenous factors, such as the presence of a stimulus, but it is also affected by endogenous
physiological factors. Haematophagous insects are not an exception, and such dependence on the
physiological state has been analysed in detail in mosquitoes, where feeding and reproduction
alternate, giving place to a gonotrophic cycle (Klowden 1981, 1990, 1995, Klowden and Briegel
1994, Takken et al. 2001).

318  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

In the case of triatomines, the physiological state-dependency of their response to host-associated

cues has been analysed recently (Bodin et al. 2008, 2009a,b, Barrozo and Lazzari 2004a). The
responses to chemical and thermal cues of insects of different physiological state have been
characterised as a function of the time of the day, the moulting cycle, and the nutritional and
reproductive states (Bodin et al. 2008, 2009a,b).

Triatomine insects exhibit a marked circadian organisation of many activities (e.g. locomotor
activity, egg hatching, ecdysis, oviposition, thermopreference, visual sensitivity (Barrozo et al.
2004b)). These different physiological and behavioural processes take place at either one or two
temporal windows, one at the beginning and the other at the end of the scotophase. Particularly,
they exhibit a maximal motivation to feed at dusk (Lorenzo and Lazzari 1998), and during this time
they leave their shelters and seek for a blood meal guided by CO2 and other olfactory cues relevant
for host searching (see Olfactory behaviour: host-seeking). At dawn, the bugs return to their
shelters using the aggregation pheromone present in their excrements as chemical landmarks
to guide them back to refuges (Lorenzo Figueiras et al. 1994). For this, they continuously deposit
faecal drops at the entrance of their shelters (Lorenzo and Lazzari 1996). As a consequence, they
must discriminate and react to distinct olfactory stimuli always present in their habitat, at the
right moment.

It has been shown that some insect species, such as fruit flies and cockroaches, vary their olfactory
sensitivity along the day, under the influence of a circadian clock, which modifies the sensitivity
of the antenna to chemical stimuli. In those species, the sensitivity of all ORCs is modulated
simultaneously, making the insects more or less sensitive to all odours at the same time (Krishnan
et al. 1999, Page and Koelling 2003). This is not the case of modulation of olfactory sensitivity in
triatomines. Recent work in two species of triatomines, T. infestans and R. prolixus, has revealed
that the responsiveness to CO2 is limited to a particular temporal window corresponding to the
early night, the moment when they leave their refuges to search for a host (Barrozo and Lazzari
2004a, Barrozo et al. 2004a, Bodin et al. 2008). Furthermore, their responsiveness to aggregation
pheromones is also limited to a particular moment of the day, but in this case, the end of the
night, when they use them as chemical landmarks for returning to their refuges (Bodin et al. 2008).
Interestingly enough, the mechanism controlling both responses (to CO2 and to aggregation
pheromones) are not the same. Whereas the behavioural responsiveness to CO2 is submitted to
the control of a circadian clock, responsiveness to aggregation pheromone is not, and it follows
exogenous cues, such as the light-dark cycle (Bodin et al. 2008).

The particularities of the circadian organisation of activities in triatomines can be summarised

as follows. First, the temporal modulation of their chemical sensitivity is not general, but their
responsiveness to each odour is specifically modulated. Second, this control is executed by a
combination of endogenous and exogenous mechanisms acting together. Finally, the sensitivity
to each odour is maximal at the moment when that odour becomes biologically relevant and
thus, when the insect displays a particular related behaviour. Thus, the sensitivity is specifically
modulated according to the behavioural (temporal) context, each odour is associated with.

Concerning the moulting cycle, and the nutritional and reproductive states, triatomines offer the
possibility of studying their effect independently. Because of their haemimetabolous development
and obligate haematophagy of nymphs and adults of both sexes, those factors can be analysed
avoiding any interference from the others. This is not the case of mosquitoes, were only adult
females feed on blood and nutrition and reproduction are closely related.

Olfaction in vector-host interactions 319

Pablo G. Guerenstein and Claudio R. Lazzari

First studies on the effect of the nutritional state suggest no effect of the starvation time on the
responsiveness of T. infestans nymphs to host-related cues (Barrozo and Lazzari 2004a), but this
was recently shown not to be always the case. Bodin et al. (2009a,b) characterised the influence
of all three factors, nutrition, reproduction and moulting on the response of R. prolixus to host-
associated odours and heat, as well as the motivation to feed on blood, in nymphs, males and
females. Thus, insects recently fed are not attracted by host odours, and the proximity of the
ecdysis, both before and after, reduces the responsiveness of the insects to host signals, as well as
their motivation to feed. The same depression in feeding behaviour was observed in females a few
days after feeding, when they are ready for egg-laying (Bodin et al. 2009b). These results emphasise
the necessity of precisely controlling the physiological state and the temporal synchronisation of
experimental animals in the study of the behavioural responses of triatomine bugs, as well as
to consider the temporal context (daytime) associated to the response and cues under analysis
(Lazzari et al. 2004, Takken 2005).

Trapping of triatomines using host odour cues

When seeking a host, many triatomines let themselves fall from ceilings upon detection of host
odours from below. In fact, one of the most popular common names for triatomines in Spanish,
‘vinchuca’, is a word derived from Quechua, and meaning ‘bug that falls’. Taking advantage of this
behaviour, an olfactometer was designed to study the attractiveness of different odour stimuli
and, furthermore, to test the ability of those odours to capture bugs (Guerenstein et al. 1995).

This dual-choice ‘trapping olfactometer’ consists on a closed rectangular arena with three plastic
tubes connected to it from below. In one side of the arena, one of the tubes (the ‘release tube’)
contains the bugs at the beginning of the experiment. Bugs could reach the arena surface by
climbing onto a piece of cardboard inside this tube. On the opposite side of the arena, two
‘capture-tubes’ (control and test) are placed. The insects can drop into the capture-tubes but
cannot escape from them. All tubes have cloth mesh at the bottom whereas below the capture-
tubes an odour source and its control are placed. Air is pulled out from the release tube and,
therefore, an airstream from the odour source (and its control) constantly reaches the release tube
via the capture-tubes and the arena (Guerenstein et al. 1995). This device has already been useful
not only to provide proof of concept for the experimental design of the olfactometers but also
to test the odour emission from a culture of yeast (Saccharomyces cerevisiae Meynen ex Hansen,
cultured with water and sugar) as activator and attractant for triatomines (Guerenstein et al. 1995).

Interestingly, triatomines are significantly activated by yeast odours, that is, more bugs fall into
the capture tubes when yeast odour is present below the test tube (a test experiment) than in
control experiments in which both capture tubes are ‘controls’ (water with sugar). Moreover, the
odour from the yeast culture also attracts triatomines in test experiments, as most of the bugs
captured fell in the test (yeast) tube while very few were found in the control tube. Under these
experimental conditions, the levels of activation and attraction evoked by yeast do not differ
significantly from those evoked by a living mouse (Guerenstein et al. 1995). Carbon dioxide is likely
the main responsible for the responses to yeast because its removal using potassium hydroxide
resulted in a loss of response by the bugs (Guerenstein et al. 1995). The high levels of activated
(68-85%) and attracted (76-87%) bugs throughout the experiments indicate that the rationale
behind the design of the trapping olfactometer is compelling.

Inspired by the trapping olfactometer, a baited trap for triatomines has been designed (Guerenstein
et al. 1995; Figure 2). The trap consists of a circular ramp that surrounds a plastic container. At the

320  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

volatiles
p
am trapping
sr
es area odour
acc source

2 cm
Figure 2. Trap for triatomines. A source of volatile odorants (e.g. a culture of baker yeast, Guerenstein et al. 1995)
is placed inside the trap. When bugs climb the circular ramp, they perceive the odours from below. This induces
the bugs to jump inside the trapping area, from where they cannot escape.

centre of the container, a vial with an odour source is placed. The odour is forced to descend to the
floor of the trap where it reaches a series of exit slits in the rim of an inverted cup. The odour then
diffuses up to the dark lid of the trap device. When host seeking bugs reach the upper edge of the
ramp they detect the odour as coming from below and drop into the plastic container from which
they cannot escape. For laboratory assays, a test and a control trap are placed in opposite sides
of an experimental arena; bugs are released in the evening and collected the following morning.
Experiments with such a trap confirmed that a yeast culture activates (52-66%) and attracts (86-
89%) of the bugs. Moreover, the results showed that a yeast culture constitutes an suitable odour
bait, and that the trap is efficient in laboratory tests (Guerenstein et al. 1995).

The trap developed is effective for different, relevant, species of triatomines. Thus, in the laboratory,
the yeast-baited trap can be used to catch T. infestans (e.g. Guerenstein et al. 1995), T. dimidiata,
T. pallidipennis (Stål) (Pimenta et al. 2007), T. sordida (Stål), Panstrongylus megistus (Burmeister)
(Pires et al. 2000) and R. prolixus (Lorenzo et al. 1999). However, some species of triatomines (T.
brasiliensis (Neiva 1911) and T. pseudomaculata (Correa and Espínola 1964)) do not fall into the
traps (Pires et al. 2000), possibly because of particularities of their host-seeking behaviour.

The triatomine trap is also capable to catch bugs under semi-natural and field conditions. Thus,
the yeast-baited trap is able to catch significant numbers of bugs in chicken-coops located in
open fields exposed to natural climatic conditions at an endemic area for Chagas Disease (Lorenzo
et al. 1998). It should be mentioned, however, that no host was present during those tests and,
therefore, the traps did not have to compete with the multimodal cues from such a host, including
its natural odour blend. In the field, a trap model using the same principle as the one described
above, caught wild, sylvatic, triatomines when a source of pure CO2 (dry ice) was used as bait
(Botto-Mahan et al. 2002).

In large-scale field tests, a trap that uses a live bait (a mouse) successfully caught triatomines (e.g.
Noireau et al. 2002). However, to date, no trap baited with synthetic odorants has been tested in
such field tests. Further understanding of the olfactory behaviour of triatomines should lead to
the development of practical, cheap and efficient baited trap devices, able to monitor the bugs in
the field and thus help implement a rational control strategy that includes selective intervention
procedures (Schofield and Kabayo 2008, Schofield et al. 2006, Tarleton et al. 2007).

Olfaction in vector-host interactions 321

Pablo G. Guerenstein and Claudio R. Lazzari

Acknowledgments

The National Research Council (CONICET) and National University of Entre Rios, Argentina, and
ANR, CNRS and University of Tours, France, are acknowledged for contributing financial support.

References:

Aldrich J (1995) Chemical communication in the true bugs and parasitoid exploitation. In: Cardé RT and Bell WJ (eds)
Chemical ecology of insects 2. Chapman and Hall, New York, USA. pp 318-363.
Almeida C, Costa Vinhaes M, Silveira A, Silveira AC and Costa J (2000) Monitoring the domiciliary and peridomiciliary
invasion process of Triatoma rubrovaria in the state of Rio Grande do Sul, Brazil. Mem Inst O Cruz 95: 761-768.
Altner H, Sass H and Altner I (1977) Relationship between structure and function of antennal chemo-, hygro-, and
thermoreceptive sensilla in Periplaneta americana. Cell Tissue Res 176: 389-405.
Barrozo RB (2003) Orientación al hospedador en la vinchuca Triatoma infestans (Heteroptera: Reduviidae): Claves
sensoriales responsables. PhD Thesis, University of Buenos Aires, Argentina.
Barrozo RB and Lazzari C (2004a) The response of the blood-sucking bug Triatoma infestans to carbon dioxide and other
host odours. Chem Senses 29: 319-329.
Barrozo RB and Lazzari CR (2004b) Orientation behaviour of the blood-sucking bug Triatoma infestans to short-chain
fatty acids: synergistic effect of L-lactic acid and carbon dioxide. Chem Senses 29: 833-841.
Barrozo RB and Lazzari CR (2006) Orientation response of haematophagous bugs to CO2: the effect of the temporal
structure of the stimulus. J Comp Physiol A 192: 827-831.
Barrozo RB, Minoli SA and Lazzari CR (2004a) Circadian rhythm of behavioural responsiveness to carbon dioxide in the
blood-sucking bug Triatoma infestans (Heteroptera: Reduviidae). J Insect Physiol 50: 249-254.
Barrozo RB, Schilman PE, Minoli SA and Lazzari CR (2004b) Daily rhythms in disease-vector insects. Biol Rhythm Res
35: 79-92.
Barth R (1952) Estudos anatômicos e histológicos sôbre a subfamília Triatominae (Hemiptera-Reduviidae). II parte: Um
novo orgao sensíbel das Triatominae. Bol Inst Oswaldo Cruz 5: 1-4.
Barth R (1953) Estudos anatômicos e histolôgicos sôbre a subfamília Triatominae (Hemiptera-Reduviidae). III parte:
Pesquisas sôbre o mecanismo de picada das Triatominae. Mem Inst O Cruz 51: 11-33.
Bernard J (1974) Etude électrophysiologique de récepteurs impliqués dans l’orientation vers l’hôte et dans l’acte
hématophage chez un Hémiptère: Triatoma infestans. PhD Thesis, University of Rennes, Rennes, France.
Bernier UR, Kline DL, Barnard DR, Schreck CE and Yost RA (2000) Analysis of human skin emanations by gas
chromatography/mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for the
yellow fever mosquito (Aedes aegypti). Anal Chem 72: 747-756.
Bernier UR, Kline DL, Posey KH, Booth MM, Yost RA and Barnard DR (2003) Synergistic attraction of Aedes aegypti (L.)
to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl disulfide. J Med Entomol 40: 653-656.
Bodin A, Barrozo RB, Couton L and Lazzari CR (2008) Temporal modulation and adaptive control of the behavioural
response to odours in Rhodnius prolixus. J Insect Physiol 54: 1343-1348.
Bodin A, Vinauger C and Lazzari CR (2009a) State-dependency of host seeking in Rhodnius prolixus: The post-ecdysis
time. J Insect Physiol 55: 574-579.
Bodin A, Vinauger C and Lazzari CR (2009b) State-dependency of host seeking in blood-sucking insects: Behaviour and
physiology. J Exp Biol 212: 2386-2393.
Bosch O, Geier M and Boeckh J (2000) Contribution of fatty acids to olfactory host finding of female Aedes aegypti.
Chem Senses 25: 323-330.
Botto-Mahan C, Cattan P and Canals M (2002) Field tests of carbon dioxide and conspecifics as baits for Mepraia spinolai,
wild vector of Chagas disease. Acta Trop 82: 377-380.
Botto-Mahan C, Cattan P and Medel R (2006) Chagas disease parasite induces behavioural changes in the kissing bug
Mepraia spinolai. Acta Trop 98: 219-223.

322  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

Carbajal de la Fuente A and Catalá S (2002) Relationship between antennal sensilla pattern and habitat in six species
of Triatominae. Mem Inst O Cruz 97: 1121-1125.
Catalá S (1994) The cave organ of Triatominae (Hemiptera, Reduviidae) under scanning electron-microscopy. Mem Inst
O Cruz 89: 275-277.
Catalá S (1997) Antennal sensilla of Triatominae (Hemiptera, Reduviidae): A comparative study of five genera. Int J Insect
Morphol Embryol 26: 67-73.
Catalá S and Schofield C (1994) Antennal sensilla of Rhodnius. J Morphol 219: 193-203.
Cork A and Park K (1996) Identification of electrophysiologically-active compounds for the malaria mosquito, Anopheles
gambiae, in human sweat extracts. Med Vet Entomol 10: 269-276.
Cruz López L, Morgan E and Ondarza R (1995) Brindley’s gland exocrine products of Triatoma infestans. Med Vet Entomol
9: 403-406.
Dahmen H (1980) A simple apparatus to investigate the orientation of walking insects. Experientia 36: 685-687.
Diehl PA, Vlimant M, Guerenstein P and Guerin PM (2003) Ultrastructure and receptor cell responses of the antennal
grooved peg sensilla of Triatoma infestans (Hemiptera: Reduviidae). Arthropod Struct Dev 31: 271-285.
Ferreira RA, Lazzari CR, Lorenzo MG and Pereira MH (2007) Do haematophagous bugs assess skin surface temperature
to detect blood vessels? PLoS One 2: e932. doi: 10.1371/ journal.pone.0000932.
Flores GB and Lazzari CR (1996) The role of the antennae in Triatoma infestans: orientation towards thermal sources. J
Insect Physiol 42: 433-440.
Geier M and Boeckh J (1999) A new Y-tube olfactometer for mosquitoes to measure the attractiveness of host odours.
Entomol Exp Appl 92: 9-19.
Geier M, Bosch O and Boeckh J (1999) Influence of odour plume structure on upwind flight of mosquitoes towards
hosts. J Exp Biol 202: 1639-1648.
Gibson G and Torr S (1999) Visual and olfactory responses of haematophagous Diptera to host stimuli. Med Vet Entomol
13: 2-23.
Guerenstein PG (1999) Sensory and behavioural responses of Triatoma infestans to host and conspecific odours. PhD
Thesis, University of Neuchâtel, Neuchâtel, Switzerland.
Guerenstein PG and Guerin P (2001) Olfactory and behavioural responses of the blood-sucking bug Triatoma infestans
to odours of vertebrate hosts. J Exp Biol 204: 585-597.
Guerenstein PG and Guerin P (2004) A comparison of volatiles emitted by adults of three triatomine species. Entomol
Exp Appl 111: 151-155.
Guerenstein PG and Hildebrand J (2008) Roles and effects of environmental carbon dioxide in insect life. Ann Rev
Entomol 53: 161-178.
Guerenstein PG and Lazzari CR (2009) Host-seeking: how triatomines acquire and make use of information to find
blood. Acta Trop 110: 148-158.
Guerenstein PG, Lorenzo MG, Núñez JA and Lazzari CR (1995) Bakers-yeast, an attractant for baiting traps for Chagas-
disease vectors. Experientia 51: 834-837.
Insausti TC, Lazzari CR and Campanucci VA (1999) Neurobiology of behaviour. A: Morphology of the nervous system
and sense organs. In: Carcavallo RU, Girón Galíndez I, Jurberg J and Lent H (eds) Atlas of Chagas’ disease vectors in
America, Vol 3, Editora Fiocruz, Rio de Janeiro, Brazil, pp. 1017-1051.
Justo S and Tramezzani J (1977) Estructuras cuticulares de la cabeza del Triatoma infestans (Klug, 1834). Physis Sección
C 37: 343-367.
Klowden M (1981) Initiation and termination of host-seeking inhibition in Aedes aegypti during oocyte maturation. J
Insect Physiol 27: 799-803.
Klowden M (1990) The endogenous regulation of mosquito reproductive behavior. Experientia 46: 660-670.
Klowden M (1995) Blood, sex, and the mosquito, the mechanisms that control mosquito blood-feeding behavior.
Bioscience 45: 326-331.
Klowden M and Briegel H (1994) Mosquito gonotrophic cycle and multiple feeding potential: contrasts between
Anopheles and Aedes (Diptera: Culicidea). J Med Entomol 31: 618-622.

Olfaction in vector-host interactions 323

Pablo G. Guerenstein and Claudio R. Lazzari

Kramer E (1976) The orientation of walking honeybees in odour fields with small concentration gradients. Physiol
Entomol 1: 27-37.
Krishnan B, Dryer S and Hardin P (1999) Circadian rhythms in olfactory responses of Drosophila melanogaster. Nature
400: 375-378.
Lazzari CR (1992) Circadian organization of locomotion activity in the haematophagous bug Triatoma infestans. J Insect
Physiol 38: 895-903.
Lazzari CR and Núñez JA (1989a) The response to radiant heat and the estimation of the temperature of distant sources
in Triatoma infestans. J Insect Physiol 35: 525-529.
Lazzari CR and Núñez JA (1989b) Blood temperature and feeding behavior in Triatoma infestans (Heteroptera:
Reduviidae). Entomol Gener 14: 183-188.
Lazzari CR and Wicklein M (1994) The cave-like sense organ in the antennae of blood-sucking bugs. Mem Inst O Cruz
89: 643-648.
Lazzari CR, Minoli SA and Barrozo RB (2004) Chemical ecology of insect vectors: The neglected temporal dimension.
Trends Parasitol 20: 506-507.
Lehane M (2005) The biology of blood-sucking in insects. Cambridge University Press, New York, USA.
Lent H and Wygodzinsky P (1979) Revision of the Triatominae (Hemiptera, Reduviidae), and their significance as vectors
of Chagas disease. Bull Am Mus Nat His 163: 123-520.
Lorenzo MG and Lazzari CR (1996) The spatial pattern of defecation in Triatoma infestans and the role of faeces as a
chemical mark of the refuge. J Insect Physiol 42: 903-907.
Lorenzo MG and Lazzari CR (1998) Activity pattern in relation to refuge exploitation and feeding in Triatoma infestans
(Hemiptera: Reduviidae). Acta Trop 70: 163-170.
Lorenzo MG, Reisenman CE and Lazzari CR (1998) Triatoma infestans can be captured under natural climatic conditions
using yeast-baited traps. Acta Trop. 70: 277-284.
Lorenzo MG, Manrique G, Pires HH, de Brito Sánchez MG, Diotaiuti L and Lazzari CR (1999) Yeast culture volatiles as
attractants for Rhodnius prolixus: electroantennogram responses and captures in yeast-baited traps. Acta Trop 72:
119-124.
Lorenzo Figueiras AN and Lazzari CR (1998) Aggregation in the haematophagous bug Triatoma infestans: a novel
assembling factor. Physiol Entomol 23: 33-37.
Lorenzo Figueras AN, Kenigsten A and Lazzari CR (1994) Aggregation in the haematophagous bug Triatoma infestans:
Chemical signals and temporal pattern. J Insect Physiol 40: 311-316.
Manrique G, Vitta AC, Ferreira RA, Zani CL, Unelius CR, Lazzari CR, Diotaiuti L and Lorenzo MG (2006) Chemical
communication in Chagas disease vectors. Source, identity, and potential function of volatiles released by the
metasternal and Brindley’s glands of Triatoma infestans adults. J Chem Ecol 32: 2035-2052.
Mayer M (1968) Response of single olfactory cell of Triatoma infestans to human breath. Nature 220: 924-925.
McIver S and Siemicki R (1984) Fine structure of antennal mechanosensilla of adult Rhodnius prolixus Stäl (Hemiptera:
Reduviidae). J Morphol 180: 19-28.
McIver S and Siemicki R (1985) Fine structure of antennal putative thermo-/hygrosensilla of adult Rhodnius prolixus Stal
(Hemiptera: Reduviidae). J Morphol 183: 15-23.
Noireau F, Abad-Franch F, Valente SAS, Dias-Lima A, Lopes CM, Cunha V, Valente VC, Palomeque FS, Carvalho-Pinto CJ,
Sherlock I, Aguilar M, Steindel M, Grisard EC and Jurberg J (2002) Trapping Triatominae in sylvatic habitats. Mem
Inst O Cruz 97: 61-63.
Núñez JA (1982) Food source orientation and activity in Rhodnius prolixus Stal (Hemiptera, Reduviidae). Bull Entomol
Res 72: 253-262.
Núñez JA (1987) Behaviour of Triatominae bugs. In: Brenner R and Stoka A (eds) Chagas’ disease vectors, Vol II, CRC
Press, Boca Raton, USA, pp 1-28.
Otálora-Luna F, Perret J and Guerin P (2004) Appetence behaviours of the triatomine bug Rhodnius prolixus on a
servosphere in response to the host metabolites carbon dioxide and ammonia. J Comp Physiol A 190: 847-854.
Page T and Koelling E (2003) Circadian rhythm in olfactory response in the antennae controlled by the optic lobe in the
cockroach. J Insect Physiol 49: 697-707.

324  Olfaction in vector-host interactions

 14. The role of olfaction in host seeking of triatomine bugs

Pimenta F, Diotaiuti L, Lustosa Lima A and Lorenzo MG (2007) Evaluation of cultures of Saccharomyces cerevisae as baits
for Triatoma dimidiata and Triatoma pallidipennis. Mem Inst O Cruz 102: 229-231.
Pires HH, Lazzari CR, Diotaiuti L and Lorenzo MG (2000) Performance of yeast-baited traps with Triatoma sordida,
Triatoma brasiliensis, Triatoma pseudomaculata and Panstrongylus megistus in laboratory assays. Pan Am J Public
Health 7: 384-388.
Preti G, Smith A and Beauchamp G (1977) Chemical and behavioral complexity in mammalian chemical communication
systems: guinea pigs (Cavia porcellus), marmosets (Saguinus fuscicollis) and humans (Homo sapiens). In: Muller-
Schwarze D and Mozell M (eds) Chemical signals in vertebrates. Plenum Press, New York, USA, pp 95-114.
Rojas JC, Rios-Candelaria E, Cruz-López L, Santiesteban A, Bond-Compean JG, Brindis Y and Malo EA (2002) A
reinvestigation of Brindley’s gland exocrine compounds of Rhodnius prolixus (Hemiptera: Reduviidae). J Med
Entomol 39: 256-265.
Schmitz H, Trenner S, Hofmann MH and Bleckmann H (2000) The ability of Rhodnius prolixus (Hemiptera; Reduviidae) to
approach a thermal source solely by its infrared radiation. J Insect Physiol 46: 745-751.
Schofield C (1979a). Behavior of Triatominae (Hemiptera, Reduviidae) – review. Bull Ent Res 69: 363-379.
Schofield C (1979b) Demonstration of isobutyric acid in some Triatomine bugs. Acta Trop 36: 103-105.
Schofield C and Patterson J (1977) Assembly pheromone of Triatoma infestans and Rhodnius prolixus nymphs (Hemiptera:
Reduviidae). J Med Entomol 13: 727-734.
Schofield C and Kabayo J (2008) Trypanosomiasis vector control in Africa and Latin America. Paras. Vectors 1: 24. doi:
10.1186/1756-3305-1-24.
Schofield C, Jannin J and Salvatella R (2006) The future of Chagas disease control. Trends Parasitol 22: 583-588.
Smallegange RC, Qiu YT, Van Loon JJ and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic acids
as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto (Diptera:
Culicidae). Chem Senses 30: 145-152.
Takken W (2005) Chemical ecology of insect vectors: temporal, environmental and physiological aspects. Trends
Parasitol 21: 57.
Takken W, Van Loon J and Adam W (2001) Inhibition of host-seeking response and olfactory responsiveness in Anopheles
gambiae following blood feeding. J Insect Physiol 47: 303-310.
Taneja J and Guerin P (1995) Oriented responses of the triatomine bugs Rhodnius prolixus and Triatoma infestans to
vertebrate odors on a servosphere. J Comp Physiol A 176: 455-464.
Taneja J and Guerin P (1997) Ammonia attracts the haematophagous bug Triatoma infestans: Behavioural and
neurophysiological data on nymphs. J Comp Physiol A 181: 21-34.
Tarleton RL, Reithinger R, Urbina JA, Kitron U and Gürtler RE (2007) The challenges of Chagas disease – Grim outlook or
glimmer of hope? PLoS Med 4: e332. doi: 10.1371/journal.pmed.0040332.
Vallejo G, Guhl F and Schaub G (2009) Triatominae–Trypanosoma cruzi/T. rangeli: Vector-parasite interactions. Acta Trop.
110: 137-147.
Wigglesworth V and Gillet (1934) The function of the antennae in Rhodnius prolixus (Hemiptera) and the mechanism of
orientation to the host. J Exp Biol 11: 120-138.

Olfaction in vector-host interactions 325

15. Chemical ecology of tick-host interactions
Sandra A. Allan

Abstract

Ticks impact the health of livestock and humans world-wide through their roles as pests and
vectors of diseases such as heartwater, theileriosis, anaplasmosis, babesiosis, Lyme disease,
erhlichiosis, tick-borne encephalitis and Crimean Congo Haemorrhagic fever. Intrinsic to their
capacity to serve in these roles is the ability to locate vertebrate hosts. Olfactory cues are the
primary method used for location of potential hosts and these include both emanations from
hosts and pheromones produced by ticks feeding on the host. Through behavioural, chemical
and electrophysiological methods, volatile compounds from host breath, ruminant digestion, skin
and glandular substances, and other blood-feeding ticks have been identified and their role in the
host-finding process elucidated. These volatile compounds can provide the basis for development
of surveillance methods or be used in the development of push-pull strategies.

Keywords: host, tick, vector, control, surveillance, CO2, pheromone, ruminant, Amblyomma, Ixodes

Introduction

Ticks are highly specialised ectoparasites challenged by the requirement of finding suitable hosts
for blood feeding several times through their life cycle. While strategies vary greatly between
species with some species feeding on a single host during a life cycle, and others requiring
multiple hosts to complete the life cycle, the location of potential hosts is a problem faced by all
individuals. Their environment is complex and filled with a multitude of chemical and physical
cues most of which are irrelevant to host-finding. Olfactory information relating to the location
of potential hosts identified with specialised receptors guides their host-seeking behaviours that
enhance the likelihood of host contact. In addition to host emanations, tick-produced odours also
enhance contact between ticks and a potential host, and these are reviewed in this chapter. For
species such as Amblyomma variegatum (Fabricius), the roles of host-produced and tick-produced
odours in host location are heavily intertwined. Essential to a discussion of tick-vector interactions
are definitions of several behavioural terms (Box 1).

Strategies for host-finding

Strategies for host-finding by ticks can be described as either nidicolous or non-nidicolous

(Sonenshine 1991). Nidicolous ticks generally live in nests, burrows or roosts in close association
with hosts and include most of the soft ticks (Argasidae) and some hard ticks (Ixodidae)(e.g. Ixodes
angustus Neumann). While these ticks may live many years, most of that time is spent off the host,
with feeding occurring for short periods of time (≤1 hr) and nymphs and adults feeding repeatedly
(Klompen et al. 1996). Nidicolous ticks respond to host cues (i.e. carbon dioxide (CO2)), heat and
moisture with these stimuli generally detected over relatively short distances (Sonenshine 1993).
Host-finding by these ticks occurs in a relatively protected environment and in close proximity to
potential hosts. Observations of host-finding by four Australian ticks feeding on reptiles revealed
that while ticks responded to host odour by adapting a questing position, the responding ticks
were always close to the host (<10 cm) (Belan and Bull 1991).

Olfaction in vector-host interactions 327

Sandra A. Allan

Box 1. Definitions of behavioural terms used in tick-vector interactions.

Activation: Initiation of movement from a resting behaviour to locomotory behaviour

similar to the activity caused by stimulants described by (Dethier et al. 1960).
Attraction: Oriented movement towards a stimulus source (Dethier et al. 1960).
Allomone: A chemical that favours the emitter but is detrimental to the perceiving
individual.
Kairomone: A chemical released from one organism that induces a favourable response
by the receiving individual (Shorey 1977). These are compounds produced
by prospective hosts (e.g. from breath, glandular secretions, urine) that elicit
responses by host-seeking ticks are kairomones.
Pheromone: A chemical that is released from one organism and induces a response by
another individual of the same species (Shorey 1977).
Questing: An off-host behaviour of ticks consisting of a resting position with the
forelegs held rigidly in front (Lees 1948). After contact with host stimuli,
forelegs are waved in the air presumably for detection of volatile compounds
by Haller’s organs on the tarsi.
Repellent: Initially described as a chemical that causes an insect to make oriented
movements away from the source (Dethier et al. 1960), this term is used
loosely here to include actions such as attraction-agonists, spatial repellents
and attraction inhibitors (Bernier et al. 2007).

In contrast, non-nidicolous ticks occur in open and exposed habitats such as forests, savannahs,
scrub, moorlands, pastures, meadows, brush and deserts, although some may have an association
with nests and burrows (Sonenshine 1993). Most of the hard ticks (Ixodidae) are non-nidicolous
and for these ticks, blood-feeding is slow (i.e. 3-14 days), occurs only once per life stage and 94-
97% of the life is spent off of the host. Hosts of different stages may be the same individual but
usually are different individuals and often different species. In many cases, hosts for immatures
are smaller vertebrates (e.g. birds, rodents) and hosts for adults are larger (e.g. cattle, deer, dogs).
Hosts may be present over a large area in the environment and host stimuli maybe effective over
longer distances than for nidicolous ticks.

Strategies used by hard ticks for location of vertebrate hosts are often classified into two groups
(Waladde and Rice 1982). The first group utilises a passive ambush tactic. Ambush strategists climb
vegetation and rest on the tops of grass blades or twigs with legs folded. Once in contact with
stimuli such as host-associated odours, heat, vibrations or shadows, the forelegs are extended
and waved presumably to enhance contact of odours with sensory structures on the tarsus of the
foreleg. If a host is contacted as it brushes past, the ticks cling to the host, locate a feeding site and
attach. Species that are examples of this type of strategy include Ixodes ricinus (Linnaeus) (Lees
1948) and Rhipicephalus (Boophilus) microplus (Canestrini) (Wilkinson 1953).

Other ticks such as Hyalomma dromedarii Koch, A. variegatum and A. hebraeum Koch use an active
host-seeking or hunter strategy and actively crawl or run towards the host. Increased activity
occurs in the response to a stimulus such as CO2 or attraction-aggregation-attachment (AAA)
pheromones in the case of large mammal-feeding Amblyomma. Species may vary between
strategies between life stage or may exhibit both types of strategies (i.e. A. americanum (L.))
(Sonenshine 1993).

328  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Host specificity varies considerably between ticks and the several species considered to be one-
host ticks are restricted almost exclusively to a single host (e.g. R. microplus on cattle, I. texanus
Banks on raccoons). Larvae of R. microplus are the only life stage that seeks hosts as all of the
remaining active life stages occur on the bovine hosts. While there are some 2-hosts ticks, most
species are 3-host ticks and may feed on different host species during each life stage. With a broad
host range, a large number of host species may be used by a species (i.e. I. persulcatus Schulze has
200 species of recorded hosts) (Anderson and Magnarelli 2008).

Clearly there are different challenges in location of potential hosts according to the type of
habitats that the ticks are in and the type of strategy used for host location. For all of these species,
olfactory cues contribute to the determination of a potential host.

Odour detection

The primary chemosensory structures used by ticks to detect volatile host chemosensory stimuli
are located on the dorsal surface of the tarsi of the first pair of legs (forelegs) in a unique organ.
This structure, Haller’s organ, consists of a posterior capsule (Pc) (Figure 1), which is covered by a

Figure 1. Scanning electron micrograph of the tarsus of the foreleg of Dermacentor variabilis showing Haller’s
organ. Arrows show the location of olfactory sensilla. Ap = Anterior pit; Ap1 = multiporose olfactory sensilla; Pc =
Posterior capsule. Bar = 50 µm (from Sonenshine 2004).

Olfaction in vector-host interactions 329

Sandra A. Allan

membrane with a slit in most species, and an anterior pit (Ap) that contains several small sensilla
(Figure 2). Amblyomma variegatum, which is one of the best studied ticks, has 19 tarsal olfactory
sensilla, three of which are in the anterior pit and seven are in the capsule of Haller’s organ. These
sensilla consist of 68-94 olfactory receptors per tarsus for adults while larvae with 14 sensilla have
57-77 olfactory receptors (Hess and Vlimant 1986). The number of sensilla and receptors in each
location varies slightly between species.

Receptors present in the capsule and anterior pit of Haller’s organ that have been characterised
by single sensillum electrophysiology consist of receptors for CO2 (Steullet and Guerin 1992a),
hydrogen sulphide (H2S) (Steullet and Guerin 1992b), aromatic aldehydes, lactones and phenols
as well as short-chain fatty acids and aldehydes (Steullet and Guerin 1994a,b) (Table 1 and 2).

The presence of two CO2 receptors in A. variegatum that differ in sensitivity optimises the ability
to detect very small changes in host emission of this kairomone (Steullet and Guerin 1992a).
The CO2-inhibited receptor is sensitive to very small concentration changes just above ambient
CO2 levels. The CO2-excited receptor monitors changes in concentration above 0.1% which
would include concentrations encountered near a potential host (Steullet and Guerin 1992a).
Previously, Holscher et al. (1980) recorded responses to CO2 from single sensillum recordings from
A. americanum, A. maculatum Koch and Dermacentor variabilis (Say).

Figure 2. Scanning electron micrograph illustrating the large olfactory sensillum, Ap1 in the anterior pit of Haller’s
organ surrounded by other sensillae. Bar = 5 µm. Ap = anterior pit; Ap1 = olfactory sensillum (from Sonenshine
2004).

330  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Table 1. Olfactory receptor types related to vertebrate host location in the capsule of Haller’s organ on the forelegs
of ticks as characterised by single sensillum electrophysiology (adapted from Steullett and Guerin 1994a).

Receptor types Tick species Characteristics Reference

Aldehyde – type 1 A. variegatum Saturated and unsaturated C6 aliphatic Steullet and Guerin 1994a
aldehydes; most sensitive to hexanal
Aldehyde – type 2 A. variegatum Saturated aldehydes such as heptanal Steullet and Guerin 1994a
and hexanal
Aldehyde – type 3 A. variegatum Unsaturated aliphatic aldehydes such as Steullet and Guerin 1994a
E-2-heptenal
Benzaldehyde A. variegatum Most sensitive to benzaldehyde, also to Steullet and Guerin 1994a
furfural
CO2-excited A. variegatum Sensitive to concentrations above 0.1% Steullet and Guerin 1992a
CO2-inhibited A. variegatum Sensitive to small concentration changes Steullet and Guerin 1992a
between 0 and 0.2%; Increase of 0.001-
0.002% inhibits
H2S – type 1 A. variegatum More firing with H2S; no response to Steullet and Guerin 1992b
dimethyl sulfide
H2S – type 2 A. variegatum Ethyl mercaptan elicits stronger response; Steullet and Guerin 1992b
dimethyl sulfide responses
2-hydroxybenzaldehyde A. variegatum Most sensitive to 3-hydroxybenzaldehyde Steullet and Guerin 1994a
Lactone A. variegatum Most sensitive to γ-valerolactone, also Steullet and Guerin 1994a
8-butyrolactone, 6-caprolactone
Methyl salicylate A. variegatum AAA pheromone component Hess and Vlimant 1986
I. ricinus AAA pheromone component Steullet 1993
R. microplus AAA pheromone component Steullet 1993

Two types of hydrogen sulfide (H2S) receptors have been identified in separate wall-pore sensilla
in the capsule of the Haller’s organ of A. variegatum (Steullet and Guerin 1992b) (Table 1). Both
receptor types were highly sensitive to H2S with a response threshold of 0.1 ppb and sensitive up
to 10 ppm; the responses to sulfides and mercaptans were dissimilar. The type 1 receptor fired
more with stimulation from H2S, while the type 2 receptor responded more to eythlmercaptan
whereas dimethyl sulfide only stimulated the type 2 receptor. In a behavioural assay H2S at 0.02
and 1 ppm aroused 60% of resting ticks with two-thirds questing and the remainder searching
actively. In mixtures with CO2, H2S appeared to act antagonistically with H2S at 1 ppm mixed with
CO2 decreased questing behaviour. It is likely that these sulfide receptors activate with other
sulfide compounds and that their presence in other body odours such as those from axillary
secretions may play a role in attraction to potential hosts or predilection sites (Steullet and Guerin
1992b). Response to these sulfides may explain the reported attraction of D. variabilis to carrion
odour (Kneidel 1984, McNemee et al. 2003).

Volatiles collected from human breath, axillary secretions and rabbit odour collected on
adsorbent material (Porapak) did not stimulate any receptors on the surface of the foreleg tarsus
of A. variegatum (Steullet and Guerin 1994b). Bovine odour, however, did stimulate receptors in
the DII.1 sensilla, and 2-nitrophenol and 4-methyl-2-nitrophenol also stimulated these sensilla.
Using other known vertebrate-associated volatiles, numerous receptor types were characterised

Olfaction in vector-host interactions 331

Sandra A. Allan

Table 2. Olfactory receptor types related to vertebrate host location in the anterior pit near Haller’s organ on the
forelegs of ticks as characterised by single sensillum electrophysiology (adapted from Steullet and Guerin 1994b).

Receptor type Tick species Sensilla in Characteristics Reference

anterior
pit

Ammonia – type 1 A. variegatum DII.6 Responsive to NH3 Steullet and Guerin 1994b
Ammonia – type 2 A. variegatum DII.6 Responsive to NH3 with different Steullet and Guerin 1994b
dose-response curve
Ammonia – type 3 A. variegatum DIV Weakly stimulated by high Steullet and Guerin 1994b
concentrations
R. sanguineus Haggart and Davis 1980
Fatty acid – type 1 A. variegatum DII.1 Most sensitive to pentanoic acid, Steullet and Guerin 1994b
also to 3-methylbutanoic acid and
butanoic acid
Fatty acid – type 2 A. variegatum DII.5 C4 and C5 fatty acids, weak Steullet and Guerin 1994b
response to hexanoic, heptanoic
and nonanoic acids
Fatty acid – type 3 A. variegatum DII.5 Most sensitive to Steullet and Guerin 1994b
2-methylpropanoic acid
Lactone A. variegatum DI.1 Most sensitive to γ-valerolactone, Steullet and Guerin 1994b
also 6-caprolactone
A. hebraeum Steullet 1993
R. microplus Steullet 1993
I. ricinus Leonovich 2004
Nitrophenol A. variegatum DII.1 Sensitive to 2-nitrophenol and Steullet and Guerin 1994b
(phenol) 4-methyl-nitrophenol
A. variegatum Schöni et al. 1984
R. microplus De Bruyne and Guerin 1994
I. ricinus Leonovich 2004
3-pentanone A. variegatum DIV Responded to 3-pentanone Steullet and Guerin 1994b

(Steullet and Guerin 1994a). These included a benzaldehyde receptor, a hydroxybenzaldehyde

receptor, three types of receptors responding differently to aliphatic aldehydes, three receptors
responding to ammonia and a lactone receptor (Table 2). Ammonia receptors were previously
reported by Haggart and Davis (1980). Leonovich (2004) also reported that a distal knoll sensillum
of Haller’s organ in I. ricinus was sensitive to steer hair extract as well as o-cresol (2-methyl phenol)
and γ-valerolactone.

Gustatory receptors detect components of the arrestment pheromones. Grenacher et al. (2001)
demonstrated that the arrestment behaviour in the sheep tick, I. ricinus to its own faeces was
mediated by gustatory receptor cells housed in the terminal wall-pore sensilla on the foreleg
tarsus. Waladde (1987) speculated that olfactory receptors on the palpal organ of ticks may serve
to perceive olfactory stimuli at very close range.

332  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Behavioural responses to host-produced compounds

Breath

Breath from potential vertebrate hosts strongly stimulates ticks with an increase in host-seeking
behaviour, generally including questing behaviour or movement toward the source (Guerin et al.
2000). Breath contains various cues such as moisture and heat in addition to volatile compounds
such as CO2, acetone, ammonia, and H2S and behavioural responses to these compounds are
presented in Table 3. In ruminants, numerous compounds that are by-products of rumen digestion,
such as 1-octen-3-ol, also induce behavioural responses (Donzé et al. 2004).

Table 3. Compounds other than CO2 eliciting responses associated with host seeking in ticks.

Chemical Tick species Component of Behaviour Reference

Acetone A. variegatum Breath (bovine) Attraction (lab) McMahon and Guerin 2002
Acetic acid A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004
increased speed
O. erraticus Breath Walking in a dish El-Ziady 1958
Ammonia O. erraticus Breath Walking in dish El-Ziady 1958
R. sanguineus Breath Questing Haggart and Davis 1980
Benzaldehyde A. hebraeum AAA pheromone Attraction (lab) Apps et al. 1988
A. hebraeum, A. AAA pheromone Attraction (lab) Yunker et al. 1992
variegatum
Benzoic acid R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Butanoic acid A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004,
increased speed
R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
2-ethylhexanoic R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
acid
Guanine A. walkerae Arrestment pheromone Akinesis Neitz and Gothe 1984
A. cohaerens, R. Arrestment pheromone Akinesis Otieno et al. 1985
appendiculatus,
A. persicus
I. scapularis Arrestment pheromone Akinesis Allan and Sonenshine 2002
A. americanum Arrestment pheromone Akinesis Yoder et al. 2008
Hematin I. scapularis Arrestment pheromone Akinesis Allan and Sonenshine 2002
Heptadecane A. hebraeum, A. AAA pheromone Attraction Yunker et al. 1992
variegatum
Heptanoic acid R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Hexanal R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Hexanoic acid R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Hydrogen sulfide A. variegatum Breath Questing Steullet and Guerin 1992b
Hypoxanthine A. persicus Arrestment pheromone Akinesis Otieno et al. 1985
I. scapularis Arrestment pheromone Akinesis Allan and Sonenshine 2002
Inosine I. scapularis Arrestment pheromone Akinesis Allan and Sonenshine 2002
Isobutanoic acid A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004
increased speed

Olfaction in vector-host interactions 333

Sandra A. Allan

Table 3. Continued.

Chemical Tick species Component of Behaviour Reference

Isobutyric acid A. variegatum AAA pheromone Attraction Apps et al. 1988

3-methylindole A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004
increased speed
4-methylphenol A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004
increased speed
2-methyl A. hebraeum, A. AAA pheromone Attraction (lab) Yunker et al. 1992
propanoic acid variegatum
R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Methyl salicylate A. variegatum AAA pheromone Questing, increased McMahon and Guerin 2000
angular velocity
and walking
speed
A. hebraeum, A. AAA pheromone Attraction (lab) Yunker et al. 1992
variegatum
Nonanoic acid A. hebraeum, A. AAA pheromone Attraction (lab) Yunker et al. 1992
variegatum
Nitric oxide A. variegatum Breath (bovine) Attraction (lab) McMahon and Guerin 2002
2-nitrophenol R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
A. variegatum AAA pheromone Questing, increased McMahon and Guerin 2000
angular velocity
and walking
speed
A. variegatum AAA pheromone Long-range Norval et al. 1991, 1993
attraction
A. variegatum AAA pheromone Aggregation, Schöni et al. 1984
attachment
A. hebraeum AAA pheromone Attraction Norval et al. 1991
A. hebraeum AAA pheromone Attachment Yunker et al. 1992
1-octen-3-ol A. variegatum Breath (bovine), Attraction (low McMahon et al. 2001
disturbed tick levels)
R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Pentanoic acid R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Propanoic acid A. variegatum Breath/rumen (bovine) Upwind movement, Donzé et al. 2004
increased speed
Pyruvate R. microplus Skin/hair (bovine) Questing Osterkamp et al. 1999
Salicylaldehyde A. hebraeum, A. AAA pheromone Attraction (lab) Yunker et al. 1992
variegatum
Squalene A. americanum, Skin Attraction Yoder et al. 1998
D. variabilis
Uric acid A. persicus Arrestment pheromone Akinesis Otieno et al. 1985
I. scapularis Arrestment pheromone Akinesis Allan et al. 2002
I. uriae Arrestment pheromone Akinesis Benoit et al. 2008
A. americanum Arrestment pheromone Akinesis Yoder et al. 2008
Xanthine A. persicus Arrestment pheromone Akinesis Otieno et al. 1985
I. scapularis Arrestment pheromone Akinesis Allan and Sonenshine 2002

334  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

CO2 is a strong activating stimulus for many tick species with ticks moving from a quiescent state
with legs curled under the body to questing behaviour or walking. In addition to being an activator,
CO2 has also been reported as an attractant source in both laboratory (Anderson et al. 1998, Sauer
et al. 1974) and field studies (Barré et al. 1997, Gray 1985, Wilson et al. 1972). In a wind tunnel,
CO2 acted as both a locomotor stimulant and an attractant for A. variegatum (Steullet and Guerin
1992a). The presence of CO2 often enhances or synergises responses to host odours and tick
pheromones (Barré et al. 1997, Beelitz and Gothe 1991, Maranga et al. 2003, Norval et al. 1989a).
In a laboratory study with the soft tick, Argas walkerae Kaiser and Hoogstraal, over 90% of unfed
females became active within 17 seconds and ran quickly to the drum containing the chicken
when exposed to a gradient of CO2 that increased close to the bird (Beelitz and Gothe 1991). When
the same treatment was tested, but with CO2 removed with an adsorbent (sodium hydroxide),
only 67.5% of ticks migrated to the chicken. If CO2 was present in concentrations different than
the gradient towards a chicken, tick movement was altered. If only a heat source was presented in
the absence of chicken odour or CO2, 55% of the ticks migrated to the source. Beelitz and Gothe
(1991) concluded that odours from the chicken represented an essential stimulus modality with
heat and CO2 as secondary, non-specific stimulants. In an olfactometer with pulsed CO2 delivery,
Perritt et al. (1993) reported that adult A. americanum and D. variabilis perceived and responded to
a CO2 level of 9 ppm above the mean ambient level. While, attraction to a CO2 source in the field
has been reported as far as 21 m for A. americanum (Wilson et al. 1972), it is generally effective over
a shorter distance. Ixodes spp. appear to respond less to CO2 than do species of other tick genera
(Anderson et al. 1998, Crooks and Randolph 2006, Fourie et al. 1993, Ginsberg and Ewing 1989,
Schulze et al. 1997) and it has been postulated that CO2 might be more effective as an excitant or
attractant for ‘hunter’ ticks then ‘ambush’ ticks.

Human breath, which contains H2S, stimulates the sulfide-sensitive cells in Haller’s organ of A.
variegatum. Ruminants produce large amounts of H2S during eructation and can likely be detected
by sulfide receptors at considerable distances from the host (Steullet and Guerin 1992b). Ticks,
even in the absence of CO2, may be stimulated into activation or questing by the presence of
sulfides (Steullet and Guerin 1992b).

In a comparison of walking responses of A. variegatum on a servosphere to five components of

breath, none of the compounds were as attractive as natural breath. Acetone, however, was the
most attractive compound after natural breath and induced the greatest reduction in speed and
highest angular velocity (McMahon and Guerin 2002). However, field evaluation of acetone has
not been sucessful and Norval et al. (1987) reported no attraction of A. hebraeum. Nitric oxide and
isoprene were also breath constituents evaluated by McMahon and Guerin (2002) with no activity
for isoprene and weak attraction reported for nitric oxide.

Ammonia is commonly associated with vertebrates and present in breath as well as skin odours and
urine (Taneja and Guerin 1997). Ammonia has been reported to elicit leg waving in R. sanguineus
(Latreille) in laboratory assays (Haggart and Davis 1980). The soft tick Ornithodoros erraticus (Lucas)
walked more on the sides of petri dishes containing 0.5% ammonia than sides without odour (El-
Ziady 1958). McMahon and Guerin (2002) did not detect any effects of ammonia on the walking
behaviour of A. variegatum in servosphere assays and speculated that this compound may need
to be presented in combination with other compounds to elicit a behavioural response.

The physiological state of ticks impacts host-seeking behaviour. Questing, movement and latency
(time after exposure to questing) to CO2 were evaluated for different stages of A. hebraeum in a
laboratory study (Anderson et al. 1998). Responsiveness to CO2 as indicated by behaviour was

Olfaction in vector-host interactions 335

Sandra A. Allan

greatest for ticks six weeks after molting compared to newly molted ticks, and by partially-fed
females compared to replete females. Males removed from a host at any stage of feeding remained
highly responsive to CO2.

Ruminant by-products

Wild and domestic ruminants are critical in the maintenance of populations of many of the hard
tick species associated with debilitating human and livestock diseases such as Lyme disease,
erhlichiosis, babesiosis, heartwater, and theileriosis (Sonenshine 1993). Ruminants regularly
eruct gases from the foregut to relieve pressure and Donzé et al. (2004) reported attraction in the
laboratory of adult A. variegatum, A. hebraeum, I. scapularis Say, I. persulcatus and I. ricinus to the
odour of rumen fluid from freshly slaughtered cattle. Three species (A. variegatum, I. persulcatus
and I. ricinus) significantly increased speed, and removal of the odour from the air stream resulted
in intense local search behaviour by A. variegatum. Using gas chromatography (GC) coupled with
electrophysiological recordings from the wall-pore olfactory sensilla on the first leg tarsus of A.
variegatum and I. ricinus, butanoic, isopentanoic, pentanoic and hexanoic acids, 2-nitrophenol,
4-methyl-2-nitrophenol, 4-methylphenol, indole and 3-methylphenol were determined to be
biologically active. Receptor cells from both tick species also responded to acetic acid, propionic
acid, phenol and 2-methylphenol. Attraction of A. variegatum was observed with butanoic and
isobutanoic acid, 4-methylphenol or 3-methylindole individually. The dose and proportions
of these compounds in a mixture were critical in eliciting oriented walking responses from A.
variegatum and I. ricinus. Of the components identified from cattle, 1-octen-3-ol is an effective
attractant for tsetse (Hall et al. 1984) and other biting flies but it did not attract A. hebraeum in the
field (Norval et al. 1987). Location of large hosts such as ruminants appears to be facilitated by the
attraction of ticks to volatile rumen end-products of digestion.

Urine

Phenolic components of urine are known to attract haematophagous insects (Bursell et al. 1988,
Hassanali et al. 1986). There is, however, little evidence to support the role of host urine in tick
attraction. Carroll et al. (1995) observed arrestant responses by I. scapularis females to urine from
a white-tailed deer (Odocoileus virginianus (Zimmerman)) doe in a laboratory assay. Urine from
bucks resulted in avoidance responses on similar assays (Carroll 1999) and responses to urine
appeared to be enhanced at 50% RH compared to 95% RH (Carroll 2000).

Skin and feathers

Questing ticks contact the skin and associated hair or feathers of a host and presumably, based
on the suitability of the stimuli perceived, continue contact on the host until attachment occurs
or detach and continue searching until a suitable host is found. Host-seeking D. variabilis are
often present along trails (Carroll et al. 1991, Newhouse 1983) and this may reflect the presence of
compounds from hosts (e.g. dogs) present along the trails. Lees (1969) reported large aggregations
of unfed I. canisuga Johnson on walls of dog kennels at dog height and speculated that the ticks
were attracted by odours left by dogs as they rubbed against the wall. In laboratory assays,
rubbings from dog ears resulted in arrestment responses by female I. scapularis, A. americanum
and both sexes of D. variabilis (Carroll 2002) and walking responses by I. ricinus (Crooks and
Randolph 2006). Extracts of canine skin and hair were attractive to Ixodes spp. (Dobrotvorsky et
al. 2000) and D. variabilis, A. americanum and R. sanguineus nymphs (Dukes and Rodriguez 1976)
in laboratory assays.

336  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

The questing behaviour of R. microplus (one-host tick), and I. ricinus (with a broad host range) larvae
were examined to compounds extracted from either the hair or skin surface from bovines and
several other animals was examined and divided into lipid and volatile components (Osterkamp et
al. 1999). Bovine skin lipid extracts and volatile extracts elicited the strongest questing responses
from R. microplus larvae. Larvae of I. ricinus responded strongly to cattle, pig and human lipid
extracts and equally to volatile extracts of cattle, humans, mice, and deer. Based on GC analysis of
extracts, a blend of 37 compounds was formulated that elicited high questing responses. Seven
key components of the blend essential for questing responses were identified and consisted of
benzoic acid, 2-ethylhexanoic acid, hexanoic acid, 2-nitrophenol, 1-octen-3-ol, pentanoic acid and
pyruvate. Comparisons of questing responses of the two species to bovine odour and different
blends of compounds indicated inter-specific species differences in responses to compound
combinations. For R. microplus, 2-nitrophenol and 1-octen-3-ol accounted for 80% of the response
to the entire blend of seven compounds. Osterkamp et al. (1999) concluded that responses to
individual compounds were obtained at high concentrations. However, when tested mixtures,
lower doses were effective in eliciting responses thereby indicating additive or synergistic effects
in the mixtures. Squalene, a component of mammalian skin, and also a glandular secretion of D.
variabilis, was found to be attractive to A. americanum and D. variabilis adults (Yoder et al. 1998)
and may play a role in host selection.

Compounds present on skin or feathers also may serve as natural repellents against ectoparasites.
The crested auklet (Aethia cristatella (Pallas)) emits a seasonally elevated citrus scent and GC analysis
of whole-bird headspace collections and of neck feather extracts revealed several aldehydes
(Douglas et al. 2004). Of these, octenal and hexanal as well as a synthetic blend representing the
crested auklets natural odorants were repellent against nymphs of A. americanum and I. uriae
(White).

Glandular substances

Glandular secretions from integumentary glands are an important form of chemical communication
for many vertebrates. These secretions from deer and antelope have been reported as kairomones
for host-seeking Ixodes ticks. Substances from the ante-orbital glands of the klipspringer antelope
(Oreotragus oreotragus (Zimmerman)) rubbed on twigs elicited attraction and aggregation of
adults of I. neitzi (Clifford, Walker and Keirans) (Rechav et al. 1978). These aggregations increase
the likelihood of future encounters with potential hosts as ticks often quest from these vantage
points. Carroll et al. (1996) reported arrestment responses of male and female I. scapularis to
rubbings from tarsal, interdigital and preorbital glands of white-tailed deer (Odocoileus virginianus
(Zimmerman)). Only weak arrestment responses were reported from rubbings of forehead, back
and non-glandular areas of the feet. Tarsal gland substances from female deer (does) but not male
deer (bucks) elicited arrestment responses of male and female I. scapularis (Carroll et al. 1995).
In the same study, interdigital gland substances influenced climbing behaviour associated with
arrestment. These arrestment responses were verified in the field using field-collected ticks and
wooden skewers rubbed with gland substances (Carroll et al. 1996). Both foreleg and hind leg
interdigital glands produce substances that elicit arrestment responses by female I. scapularis
(Carroll 2001). Doe urine, under certain circumstances, elicited arrestment responses from
female I. scapularis. Avoidance, however, appeared to occur in response to urine from dominant
reproductive bucks (Carroll 2000). White-tailed deer urinate on tarsal glands as part of their scent-
marking behaviours and it is difficult to differentiate possible attractant responses to urine or
tarsal gland secretions as in general urine was repellent (Carroll 2000).

Olfaction in vector-host interactions 337

Sandra A. Allan

Glandular secretions of the tick, Haemaphysalis leachi (Audouin), a tick commonly found on dogs,
contain a dog-repelling allomone that inhibits removal of ticks by grooming (Burger et al. 2006).
Based on responses in bioassays, the allomones may consist of six aliphatic aldehydes from
hexanal to undecanal.

Feeding-site preference

Some species of tick exhibit a strong predilection for feeding on certain sites on the hosts body.
Two such species are R. appendiculatus Neumann and R. evertsi Neumann, which exhibit strong
preferences to feed inside the ears and around the anal regions of catlle, respectively. In a study
by Wanzala et al. (2004), ticks of each species were released on six different body locations
on cattle with most ticks of each species migrating over the body and reaching the preferred
feeding site. Given the apparent avoidance of the preferred feeding site by the other species, the
authors proposed that a push-pull effect, with some compounds attracting and some compounds
repelling, influenced the orientation behaviour. Using dichloromethane extracts of volatiles from
ears or the anal region adsorbed onto charcoal or silica, tick responses were assayed in a tick
climbing bioassay apparatus. Odours trapped from cattle ears attracted R. appendiculatus but
repelled R. evertsii. Odour from the anal region had the opposite effect. It seems likely that other
species with strong predilection sites for feeding are similarly guided by chemical cues.

One impact of feeding aggregations of ticks, particularly those containing immatures, is the
potential for pathogen transmission through nonsystemic means by co-feeding of infected and
uninfected ticks (Randolph et al. 1996). This is best supported from field data with Lyme disease
spirochetes and Ixodes ticks (Kimura et al. 1995, Stafford et al. 1995).

Behavioural response to tick pheromones that enhance host contact

For some tick species, vector-host interactions are facilitated by the presence of tick pheromones.
The pheromones that facilitate host contact fall into two categories: attraction-aggregation-
attachment (AAA) pheromones or arrestment (assembly) pheromones.

Attraction-aggregation-attachment pheromones

In many of the Amblyomma species that feed on large hosts such as cattle, the location of
suitable hosts and the development of feeding aggregations of ticks on hosts are facilitated
by the emission of volatile compounds by blood-feeding conspecific males on the hosts. These
attraction-aggregation-attachment (AAA) pheromones are a multi-component blend that can
elicit a series of behavioural steps that ultimately result in the feeding aggregations observed
on hosts.

This strategy has been well documented for the bont ticks, A. variegatum and A. hebraeum (Apps et
al. 1988, Diehl et al. 1991, Norval and Rechav 1979). Pioneer male ticks locate a suitable host and
migrate on the host to attach at a preferred feeding site. After blood-feeding for five or more days,
males emit a mixture of volatile compounds that elicit attraction of unfed conspecific nymphs
and adults (secondary colonisers) to the infested host individual (Hess and deCastro 1986; Norval
et al. 1989a 1989b; Steullet 1993). Once on the host, ticks migrate to form aggregations in the
preferred attachment sites around the feeding males emitting the volatiles. At this point, they will
attach to the host and feed. While males may attach to potential hosts in the absence of females,
females will only attach and feed near a source of pheromone (blood-feeding males) (Norval and

338  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Rechav 1979). The principal components of the AAA pheromone for A. variegatum and A. hebraeum
include 2-nitrophenol (o-nitrophenol), methyl salicylate,

, 2,6-dichlorophenol, with some differences in presence and abundance between species (Diehl
et al.1991, Lusby et al. 1991, Schöni et al. 1984). In general, 2-nitrophenol and methyl salicylate
stimulate long range, interspecific attraction and aggregation; whereas, methyl salicylate and
nonanoic acid induce attachment (Lusby et al. 1991).

The behavioural effects of these pheromones are enhanced in combination with CO2 (Hess and
deCastro 1986, Norval et al. 1989a) and host odours (which are not attractive on their own)
(McMahon and Guerin 2000) with attraction in the field from over 10 m distance. Hosts with
attached feeding males attract more ticks than uninfected hosts (Norval et al. 1989b). A similar
pheromone is produced by blood-feeding males of A. maculatum. A hexane extract of blood-fed
males that was placed on one ear (preferred feeding site) of cattle resulted in 4-fold increase of ticks
attached to the treated ear over the control ear (Gladney et al. 1974). A similar pheromone appears
to be present in A. cajennense (F.) but may only induce attachment, not aggregation (Rechav et al.
1997). Maranga et al. (2003) examined responses of released A. variegatum to different doses of the
synthetic pheromone blend (o-nitrophenol, methyl salicylate and nonanoic acid in paraffin oil) in
the presence or absence of CO2. Up to 90% of the released ticks were attracted to the pheromone
source in the presence of CO2 within three hours. CO2 alone was unattractive and the pheromone
blend alone was only moderately attractive.

An additional tick-produced compound, 1-octen-3-ol, was identified from various tick materials
(disturbed adults, exuviae, faeces, carcasses) of A. variegatum (McMahon et al. 2001). In servosphere
trials, 1-octen-3-ol elicited attraction in unfed A. variegatum, however, it is unclear how this
compound mediates tick responses and if it would play any role in host attraction. Previous
studies investigated 1-octen-3-ol as a possible attractant for A. variegatum as it is a component of
ruminant digestion, however, no significant attraction was detected (Norval et al. 1987).

Arrestment pheromones

Arrestment or assembly pheromones cause locomotion to stop and result in clusters of individuals
in the environment. Upon contacting these pheromones, ticks cease crawling, curl their legs
under the body and become akinetic (Sonenshine 2004, 2006). These pheromones act within and
between sexes, may affect immatures and tend to be present in caves, under ledges, rookeries, in
cracks or crevices or other sheltered locations (Sonenshine 1991). They are particularly widespread
in nidicolous soft ticks (at least 14 species in the Argasidae family) but are also reported from at
least eight species of hard ticks (e.g. I. scapularis, I. ricinus, R. appendiculatus (Sonenshine 1991)).
These assemblies may enhance contact with potential hosts, enhance survival by keeping ticks in
favourable microhabitats and ensuring mating for those species that mate off hosts. Guanine and
other related purines such as uric acid, xanthine, hypoxanthine and inosine which are constituents
of tick faeces and present in exuvia comprise the active components of this pheromone (Allan
and Sonenshine 2002, Dusbábek et al.1991, Grenacher et al. 2001, Hassanali et al. 1989, Otieno
et al. 1985, Sonenshine et al. 2003). Haematin and 8-azaguanine also contributed to assembly
responses (Grenacher et al. 2001, Sonenshine et al. 2003). Excretion of nitrogenous waste products
by ticks generally occurs after blood-feeding and immediately after moulting. Locations with
accumulations of these materials are likely indicative of an environment favourable for survival
and the historic presence of a host (Otieno et al. 1985). Benoit et al. (2008) reported that guanine
and other components of tick excreta function as an assembly pheromone for the seabird tick, I.

Olfaction in vector-host interactions 339

Sandra A. Allan

uriae, in promoting the formation of off-host aggregations, while uric acid and other components
of penguin guano functioned as kairomones used by the tick to locate its penguin host. Similarly,
unfed A. americanum responded to an assembly pheromone, which contains uric acid and
suggests a kairomonal role of uric acid from bird and reptile faeces. The latter animals are the
preferred hosts for immatures of this species (Yoder et al. 2008). Such clustering of unfed ticks in
the environment greatly enhances successful host contact if clusters are located optimally.

Factors that enhance chemical attractants

In nature, chemical attractants are rarely present without other factors that often enhance or
synergise responses of tick responses. Although eyes and light receptors are simple in ticks, visual
contrast against a background or detection of movement are known to enhance responses (Philis
and Cromroy 1977, Waladde and Rice 1982). Radiant heat has been reported to be synergistic when
combined with host odours (Lees 1948, Sonenshine 1993) and is detected by thermoreceptors on
the tarsi of the first pair of legs, which are waved in the air during active searching.

Practical applications

Surveillance

Host cues have been used successfully for the development of effective surveillance tools for
arthropods. As CO2 is known to activate and possibly attract ticks at a considerable distance from
the host, it has been examined for tick surveillance. A trap using CO2 as an attractant used for
collection of I. ricinus (Gray 1985) attracted ticks from distances of up to 3.5 m and subsequently
attraction for numerous species of ticks has been reported. Attraction of A. americanum to CO2
was reported to be from several meters (Sonenshine 1993) and Wilson et al. (1972) collected A.
americanum from as far away as 21 m. CO2-baited traps have been used to sample a wide range
of species including O. turicata Duges (Adeyeye and Butler 1991). Solberg et al. (1992) reported
that dry ice-baited traps collected larger number of I. scapularis and A. americanum than did drag
or human-host collections.

Host odours or tick-produced odours have also been combined with CO2 for surveillance.
Attraction of A. hebraeum to traps baited with CO2 in conjunction with an AAA pheromone was
effective from distances as far as 11 m in 90 minutes (Norval et al. 1989a). Subsequently these traps
were used for surveillance of Ehrlichia (Cowdria) ruminantium in field populations (Norval et al.
1990). Use of synthetic AAA pheromone (methyl salicylate, o-nitrophenol, nonanoic acid) greatly
increased the efficacy of CO2 traps for A. variegatum in Guadeloupe (Barré et al. 1997). Attraction
of A. variegatum to different doses of AAA pheromone and CO2 (500 ml/min) resulted in collection
of up to 90% of the released ticks (Maranga et al. 2003) and these traps were used to evaluate
efficacy of a subsequent control study (Maranga et al. 2006). Because attractants are considered
particularly useful for ticks using the hunter strategy for host-finding (Sonenshine 1993), they
have promise for development of effective surveillance strategies for these ticks.

Another standard approach for tick surveillance is the use of the tick drag or flag (Sonenshine
1993). The addition of host odour (from sheep wool) onto the cloth increased efficacy from 2.4-2.8
fold (for flagging and dragging, respectively) for I. rubicundus Neumann. Additionally, duration of
tick contact on the cloth increased from 9.1 seconds to 32.9 seconds (Fourie et al. 1995).

340  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Targeted on-host control

Chemical attractants have been evaluated for targeted on-host control of ticks by combining
attractant compounds with acaricides. In an initial study by Gladney et al. (1974), extracts of
blood-fed males of A. maculatum, which contained an attractant pheromone, combined with
insecticide were used to kill female ticks that attached to the treated site. A tick decoy concept was
developed using the AAA pheromones for targeted control of the bont ticks, A. variegatum and A.
hebraeum, on cattle (Norval et al. 1996). Components of the AAA pheromone (methyl salicylate,
o-nitrophenol) combined with a tick sex pheromone (2,6-dichlorophenol) and another attractant
(phenylacetaldehyde), along with insecticides, were attached to tails of cattle. The pheromones
served as attractants for unfed nymphs, males and females to the site on the body where pesticide
was present. Once at the treated site, prolonged contact during the aggregation, attachment and
feeding behaviour resulted in mortality. When these tags were evaluated in Zimbabwe against A.
hebraeum, tick numbers decreased on cattle over a 12 week period (Norval et al. 1996). In the two
trials, efficacy ranged from 87.5 to 99.3% for cyfluthrin and flumethrin-impregnated tags. In an
additional study in Guadeloupe, acaricide and AAA-impregnated tags were attached to the necks
and tails of cattle for control of A. variegatum (Allan et al. 1998). In this trial, tick numbers were
reduced similarly between pheromone and acaricide tags and acaricide tags only. Cattle supplied
with pheromone tags had higher proportions of ticks on the neck (18.2%) and hind (81%) regions
of the body than did cattle with no pheromone tags (necks 8.2%, hind 62.5%), indicating that
pheromone delivery did affect tick aggregations. While this approach appears promising, efficacy
is difficult to evaluate as dead ticks may not remain on the host.

There is potential for a push-pull approach for tick control to be developed with the discovery of
semiochemicals that mediate attraction to different attachment sites by the vectors of East Coast
fever, R. appendiculatus and R. evertsi on cattle. Such an approach could combine the use of a
synthetic or botanical tick repellent at the ear and an attractant-baited trap treated with a fungal
pathogen or acaricide located on the back of the animal (Hassanali et al. 2008).

Targeted off-host control

Several approaches have been taken for targeted off-host control of ticks that incorporate olfactory
cues that facilitate host location. The combination of CO2 (500 ml/min) with AAA pheromones
as attractants to self-treatment traps baited with spores of fungal conidia (Beauveria bassiana
(Bals.) and Metarhizium anisopliae (Sorok.)) appeared effective in treating ticks released onto a
field plot (Maranga et al. 2006). Three weeks after the treatment only 34% of the released ticks
were recovered on plots with fungi whereas 76 to 85% of ticks were recovered on control plots.

The combination of arrestment pheromone components with acaricides was effective in laboratory
studies with Argas persicus (Oken) (Dusbábek et al. 1997). For I. scapularis, the combination of
pheromone components into a slow-release matrix along with permethrin provided promising
results when evaluated in a petri dish assay (Sonenshine et al. 2003). Application of this material
on vegetation in a field trial resulted in significant reduction of I. scapularis nymphs near Armonk,
New York (T. Daniels and R. Falco, personal communication).

Repellents

One of the most effective methods for prevention of tick-borne diseases of humans is the use
of repellents. Towards this end, considerable effort has been made to evaluate compounds that

Olfaction in vector-host interactions 341

Sandra A. Allan

interfere with host-finding by ticks or deter ticks from treated sites. Different systems for testing
tick repellents have been reviewed by Dautel (2004) who differentiated between assays that
evaluate repellency in the absence or presence of host stimuli. In the absence of host stimuli, such
in petri dishes or tick climbing assays, repellency is measured by the absence of ticks in a location
containing the potential repellent. These assays do not correspond directly to sensory perception
of potential hosts but may impact behavioural steps that contribute to successful host-finding
(i.e. climbing vegetation, arrestment in locations suitable for the host). In assays that include host
stimuli, repellency is measured by avoidance in the presence of host stimuli. Studies by Carroll
et al. (2005) indicated that I. scapularis and A. americanum responded to DEET (N,N-diethyl-3-
methylbenzamide) and SS220 (1S,2S’-2-methylpiperidinyl-3-cyclohexene-1-carboxamide) in the
vapour phase to avoid host contact. Similarly Dautel et al. (1999) reported that I. ricinus nymphs
would not contact DEET-treated filter paper. McMahon et al. (2003) reported that indalone in
an airstream caused A. variegatum to walk downwind, however airborne DEET did not repel
ticks attracted to the AAA pheromone. Permethrin, while a toxicant, also repels a range of tick
species and is used for treatment of military uniforms (Mehlhorn et al. 2003, McCain and Leach
2007, Schreck et al. 1995). In contrast, a less common effect has been reported in which fabric
impregnated with permethrin as a repellent enhanced attachment behaviour of H. dromedarii
with more rapid attachment with greater numbers attached than the untreated controls (Fryauff
et al.1994). These results were presumably related to the increased neurosecretory production
which activated salivary secretion in H. dromedarii ticks exposed to permethrin (Mohamed et al.
2000). Questions remain as to the mode of action of many repellents (i.e. deterrency, inhibition of
attraction, masking of detection)(Bernier et al. 2007) and as to whether repellent compounds act
by olfaction, contact with gustatory receptors or both.

Host-derived volatiles that interfere with attraction are potential sources for new repellent
compounds for several insect species (Bernier et al. 2007, Logan et al. 2009, Weldon and Carroll 2007)
and may also hold potential for ticks. Research on volatile compounds emitted from unsuitable
predilection sites on hosts such as cattle (Wanzala et al. 2004) may provide new approaches for
management of livestock from tick infestation and tick-borne diseases using a push-pull approach
(Hassanali et al. 2008). The basis for naturally ‘tick-resistant’ breeds or species of hosts appears
to be related to host immune response (Martinez et al. 2006), however, differences in volatile
emissions (Weldon and Carroll 2007), if present, provide potential compounds useful for livestock
protection. Plant-based compounds evaluated against ticks may provide new active ingredients
for protection of human and even livestock (Bissinger et al. 2009, Carroll et al. 2007, Dietrich et al.
2006, Feaster et al. 2009, Palsson et al. 2008).

The use of host odours for development of in vitro feeding systems for hard ticks can address ethical
concerns, high costs of maintaining suitable hosts and provides opportunities for development of
products against ticks (Kröber and Guerin 2007). Feeding on membranes has been facilitated by
the addition of the AAA pheromones, host hair, tick faeces and bovine pelage (Kröber and Guerin
2007, Kuhnert et al. 1995). Development of these feeding systems could accelerate development
of methods for high throughput screening of candidate repellents.

Conclusions

The challenges for ticks to find potential hosts in a complex environment are met, at least partially,
by an array of olfactory cues. These cues are primarily host-produced but for some species, tick-
produced odours also contribute to host location. One of the most well-studied tick species is
A. variegatum, which feeds primarily on large ungulate mammals. This tick, highly responsive

342  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

to rumen odours and AAA pheromones, may only provide some insight into the olfactory cues
used by other tick species. Interactions between ticks and potential hosts are complex and our
understanding of them has been advanced significantly through the use of electrophysiological,
sophisticated behavioural and highly sensitive analytical chemical methods. Certainly, additional
research into the chemical ecology of additional tick-host interactions will provide compounds
that can be used to develop new surveillance methods or tools for development of push-pull
strategies against ticks that are both pests of vectors of human and livestock diseases.

The field of insect olfactory ecology is advancing quickly with the convergence of genomic,
physiological, and ecological data and detailed examination of the processes between odour
molecules and odour receptor proteins (De Bruyne and Baker 2008). Relatively little, however, is
known for ticks about the three major groups of proteins (odorant receptors, odorant-binding
proteins and odour-degrading enzymes) that are of critical importance to insect olfactory systems
for detection of many of the same chemicals. These may provide feasible molecular targets for the
development of attractants and novel repellent strategies. For ticks, progress in our understanding
of the molecular basis for olfaction has been slower and important differences between insect and
acarines may be discovered. Robertson and Kent (2009) examined the evolution of three protein
genes related to the carbon dioxide receptor in three species of mosquitoes, the silk moth and red
flour beetles. That gene lineage is absent from the I. scapularis genome indicating that ticks utilise
another pathway to detect carbon dioxide. Unique gene evolutionary processes in ticks have also
been reported by Wang et al. (2007), and with completion of sequencing of different tick genome,
a better understanding of these processes will be obtained. The degree of conservation within
insect odorant receptor genes and their relation to acarine genes is yet to be determined and this
information will provide insight into approaches for rational design of novel control strategies
targeted against gene orthologs.

References

Adeyeye OA and Butler JF (1991) Field evaluation of carbon dioxide baits for sampling Ornithodoros turicata (Acari:
Argasidae) in gopher tortoise burrows. J Med Entomol 28: 45-48.
Allan SA, Barré N, Sonenshine DE and Burridge MJ (1998) Efficacy of tags impregnated with pheromone and acaricide
for control of Amblyomma variegatum. Med Vet Entomol 12: 141-150.
Allan SA and Sonenshine DE (2002) Evidence of an assembly pheromone in the black-legged tick, Ixodes scapularis. J
Chem Ecol 28: 15-27.
Anderson JF and Magnarelli LA (2008) Biology of ticks. In: Edlow JA (ed) Infectious disease clinics of North America: tick
borne diseases, Lyme disease. WB Saunders, Philadelphia, USA, pp. 195-215.
Anderson RB, Scrimgeour GJ and Kaufman WR (1998) Responses of the tick, Amblyomma hebraeum (Acari: Ixodidae), to
carbon dioxide. Exp Appl Acarol 22: 667-681.
Apps PJ, Viljoen HW and Pretorius V (1988) Aggregation pheromone of the bont tick Amblyomma hebraeum:
identification of candidates for bioassay. Onderstepoort J Vet Res 55: 135-137.
Barré N, Garris GI and Lorvelec O (1997) Field sampling of the tick Amblyomma variegatum (Acari: Ixodidae) on pastures
in Guadeloupe; attraction of CO2 and/or tick pheromones and conditions of use. Exp Appl Acarol 21: 95-108.
Belan I and Bull CM (1991) Host detection by four Australian tick species. J Parasitol 77: 337-340.
Beelitz P and Gothe R (1991) Investigations on the host-seeking and finding of Argas (Persicargas) walkerae (Ixodoidea:
Argasidae). Parasitol Res 77: 622-628.
Benoit JB, Lopez-Martinez G, Phillips SA, Elnitsky MA, Yoder JA, Lee RE Jr and Denlinger DL (2008) The seabird tick, Ixodes
uriae, uses uric acid in penguin guano as a kairomone and guanine in tick feces as an assembly pheromone on the
Antarctic Peninsula. Polar Biol 31: 1445-1451.

Olfaction in vector-host interactions 343

Sandra A. Allan

Bernier UB, Kline DL and Posey KH (2007) Human emanations and related natural compounds that inhibit mosquito
host-finding abilities. In: Debboun M, Frances SP and Strickman D (eds) Insect repellents principles, methods, and
uses. CRC Press, New York, USA, pp 77-100.
Bissinger BW, Apperson CS, Sonenshine DE, Watson DW and Roe RM 2009. Efficacy of the new repellent BioUD® against
three species of ixodid ticks. Exp Appl Acarol 48: 239-250.
Burger BV, Marx A, Roux ML and Oelofsen BW (2006) Characterization of dog repellent factor from cuticular secretion
of female yellow dog tick, Haemaphysalis leachi. J Chem Ecol 32: 125-136.
Bursell E, Gough AJE, Beevor PS, Cork A, Hall DR and Vale GA (1988) Identification of components of cattle urine
attractive to tsetse flies, Glossina spp. (Diptera: Glossinidae). Bull Entomol Res 78: 281-291.
Carroll JF (1999) Responses of three species of adult ticks (Acari: Ixodidae) to chemicals in the coats of principal and
minor hosts. J Med Entomol 36: 238-242.
Carroll JF (2000) Responses of adult Ixodes scapularis (Acari: Ixodidae) to urine produced by white-tailed deer of various
reproductive conditions. J Med Entomol 37: 472-475.
Carroll JF (2001) Interdigital gland substances of white-tailed deer and the response of host-seeking ticks (Acari:
Ixodidae). J Med Entomol 38: 114-117.
Carroll JF (2002) How specific are host-produced kairomones to host-seeking ixodid ticks? Exp Appl Acarol 28: 155-161.
Carroll, JF, Russek-Cohen E, Nichols JD and Hines JE (1991) Populations dynamics of American dog ticks (Acari: Ixodidae)
along park trails. Environ Entomol 20: 922-929.
Carroll JF, Klun JA and Schmidtmann ET (1995) Evidence for kairomonal influence on selection of host-ambushing sites
by adult Ixodes scapularis (Acari: Ixodidae). J Med Entomol 32: 119-125.
Carroll JF, Mills GD and Schmidtmann ET (1996) Field and laboratory responses of adult Ixodes scapularis (Acari: Ixodidae)
to kairomones produced by white-tailed deer. J Med Entomol 33: 640-644.
Carroll JF, Klun JA and Debboun M (2005) Repellency of Deet and SS220 applied to skin involves olfactory sensing by
two species of ticks. Med Vet Entomol 19: 101-106.
Carroll, JF, Cantrell CL, Klun JA and Kramer M (2007) Repellency of two terpenoid compounds isolated from Callicarpa
americana (Lamiaceae) against Ixodes scapularis and Amblyomma americanum ticks. Exp Appl Acarol 41: 215-224.
Crooks E and Randolf SE (2006) Walking by Ixodes ricinus ticks: intrinsic and extrinsic factors determine the attraction
of moisture or host odour. J Exp Biol 209: 2183-2142.
Dautel H (2004) Test systems for tick repellents. Int J Med Microbiol 293 Suppl 37: 182-188.
Dautel H, Kahl O, Siema K, Oppenrieder M, Muller-Kuhrt L and Hilker M (1999) A novel test system for detection of tick
repellents. Ent Exp Appl 91: 431-441.
De Bruyne M and Baker TC (2008) Odor detection in insects: volatile codes. J Chem Ecol 34: 882-897.
De Bruyne M and Guerin PM (1994) Isolation of 2,6-dichlorophenol from the cattle tick Boophilus microplus: receptor
cell responses but no evidence for a behavioural response. J Insect Physiol 40: 143-154.
Dethier VG, Baron Browne L and Smith CN (1960) The designation of chemicals in terms of the responses they elicit
from insects. J Econ Entomol 53: 134-136.
Diehl, PA, Guerin PM, Vlimant M and Steullet P (1991) Biosynthesis, production site, and emission rates of aggregation-
attachment pheromone in male of two Amblyomma ticks. J Chem Ecol 17: 833-847.
Dietrich G, Dolan MC, Peralta-Cruz J, Schmidt J, Piesman J, Eisen RJ and Karchesy JJ (2006) Repellent activity of
fractioned compounds from Chamaecyparis nootkatensis essential oil against nymphal Ixodes scapularis (Acari:
Ixodidae). J Med Entomol 43: 957-961.
Dobrotvorsky AK, Tkachev AV, Maumov RL and Carroll JF (2000) Study of host odour determinants of questing behavior
in Ixodes ticks. In: Proc 3rd International conference on Tick and Tick-borne Pathogens, High Tatra Mountains,
Slovakia, pp 235-240.
Donzé G, McMahon and Guerin PM (2004) Rumen metabolites serve ticks to exploit large mammals. J Exp Biology 207:
4283-4289.
Dukes JC and Rodriguez JC (1976) A bioassay for host-seeking response of tick nymphs (Ixodidae). J Kansas Entomol
Soc 49: 562-566.

344  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Dusbábek F, Simek P, Jegorov A and Troska J (1991) Identification of xanthine and hypoxanthine as components of
assembly pheromone in excreta of argasid ticks. Exp Appl Acarol 11: 307-316.
Dusbábek F, Rupeš V, Šimek P and Zahradínčková H (1997) Enhancement of permethrin efficacy in acaricide-attractant
mixtures for control of the fowl tick Argas persicus (Acari: Argasicae). Exp Appl Acarology 21: 293-305.
Douglas HD III, Co JE, Jones TH and Conner WE (2004) Interspecific differences in Anthia spp. auklet odorants and
evidence for chemical defense against ectoparasites. J Chem Ecol 30: 1921-1934.
El-Ziady S (1958) The behavior of Ornightodoros erraticus (Lucas, 1849), small form (Ixodoidea, Argasidae), towards
certain environmental factors. Ann Entomol Soc Amer 51: 317-336.
Feaster JE, Scialdone MA, Todd RG, Gonzalez YI, Foster JP and Hallahan DL (2009) Dihydronepetalactones deter feeding
activity by mosquitoes, stable flies, and deer ticks. J Med Entomol 46: 832-840.
Fryauff DJ, Shoukry MA and Schreck CE (1994) Stimulation of attachment in a camel tick, Hyalomma dromedarii (Acari:
Ixodidae): the unintended result of sublethal exposure to permethrin-impregnated fabric. J Med Entomol 31: 23-29.
Fourie LJ, Snyman A, Kok DJ, Horak IG and van Zyl JM (1993) The appetence behavior of two south African paralysis-
inducing ixodid ticks. Exp Appl Acarol 17: 921-930.
Fourie LJ, Van der Lingen F and Kok DJ (1995) Improvement of field sampling methods for adult Karoo paralysis ticks,
Ixodes rubicundus (Acari: Ixodidae), through addition of host odour. Exp Appl Acarol 19: 93-101.
Ginsberg HS and Ewing CP (1989) Comparison of flagging, walking, trapping and collecting from hosts as sampling
methods for northern deer ticks, Ixodes dammini,and lone star ticks, Amblyomma americanum (Acari: Ixodidae).
Exp Appl Acarol 7: 313-322.
Gladney WJ, Grabbe RR, Ernst SE and Oehler DD (1974) The Gulf Coast tick: evidence of a pheromone produced by
males. J Med Entomol 11: 303-306.
Gray JS (1985) A carbon dioxide trap for prolonged sampling of Ixodes ricinus L. populations. Exp Appl Acarol 1: 35-44.
Grenacher S, Kröber T, Guerin PM and Vlimant M. (2001) Behavioral and chemoreceptor cell responses of the tick, Ixodes
ricinus, to its own faeces and faecal consitituents. Exp Appl Acarol 25: 641-660.
Guerin PM, Kröber T, McMahon C, Guerenstein P, Genacher S, Vlimant M, Diehl P, Steullet P and Syed Z (2000)
Chemosensory and behavioural adaptations of ectoparasitic arthropods. Noza Acta Leopoldina NF 83 nr 316: 213-
229.
Haggart DA and Davis EE (1980) Ammonia-sensitive neurons on the first tarsi of the tick, Rhipicephalus sanguineus. J
Insect Physiol 26: 517-523.
Hall DR, Beevor PS, Cork A, Nesbitt BF and Vale GA (1984) 1-octen-3-ol. A potent olfactory stimulant and attractant for
tsetse isolated from cattle odours. Insect Sci Appl 5: 335-339.
Hassanali A, McDowell PC, Owaga MLA and Saini RK (1986) Identification of tsetse attractants from excretory products
of a wild host animal, Syncerus caffer. Insect Sc Appl 7: 5-9.
Hassanali A, Nyandat E, Obenchain FD, Otieno DA and Galun R (1989) Humidity effect on response of Argas persicus
(Oken) to guanine, an assembly pheromone of ticks. J. Chem Ecol 15: 791-798.
Hassanali A, Herren H, Khan ZR, Pickett JA and Woodcock CM (2008) Integrated pest management: the push-pull
approach for controlling insect pests and weeds of cereals, and its potential for other agricultural systems including
animal husbandry. Phil Trans R Soc B 363: 611-621.
Hess E and De Castro JJ (1986) Field tests of the response of female Amblyomma variegatum (Acari: Ixodidae) to the
synthetic aggregation-attachment pheromone and its components. Exp Appl Acarol 2: 249-255.
Hess E and Vlimant M (1986) Leg sense organs of ticks. In: Sauer JR and Hair JA (eds) Morphology, physiology and
behavioural biology of ticks. Ellis Horwood, Chicester, UK, pp 361-390.
Holscher KH, Gearhart HL and Barker RW (1980) Electrophysiological response of three tick species to carbon dioxide
in the laboratory and field. Ann Entomol Soc Am 73: 288-292.
Kimura K, Isogai E, Isogai H, Kamewaka Y, Nishikawa T, Ishii N and Fujii N (1995) Detection of Lyme disease spirochetes
in the skin of naturally infected wild Sika deer (Cervus nippon yesoensis) by PCR. Appl Environ Micro 61: 1641-1642.
Klompen JSH, Black WC IV, Keirans JE and Oliver JH Jr (1996) Evolution of ticks. Ann Rev Entomol 41: 141-161.
Kneidel KA (1984) Carrion as an attractant to the American dog tick, Dermacentor variabilis (Say). J NY Entomol Soc 92:
405-406.

Olfaction in vector-host interactions 345

Sandra A. Allan

Kröber T and Guerin PM (2007) An in vitro feeding assay to test acaricides for control of hard ticks. Pest Manag Sci 62:
17-22.
Kuhnert F, Diehl PA and Guerin PM (1995) The life-cycle of the bont tick Amblyomma hebraeum in vitro. Int J Parasitol
25: 887-896.
Lees AD (1948) The sensory physiology of the sheep tick, Ixodes ricinus (L.) J Exp Biol 25: 145-207.
Lees AD (1969) The behaviour and physiology of ticks. Acarologia 11: 397-406.
Leonovich SA (2004) Phenol and lactone receptors in the distal sensilla of the Haller’s organ in Ixodes ricinus ticks and
their possible role in host perception. Exp Appl Acarol 32: 89-102.
Logan, JG, Birkett MA, Clark SJ, Powers S, Seal NJ, Wadhams LJ, Mordue (Luntz) JA and Pickett JA (2009) Identification
of human-derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes J Chem Ecol 34:
308-322.
Lusby WR, Sonenshine DE, Yunker CE, Norval RA and Burridge MJ (1991) Comparison of known and suspected
pheromonal constituents in males of the African ticks, Amblyomma hebraeum Koch and Amblyomma variegatum
(Fabricius). Exp Appl Acarol 13: 143-152.
Maranga RO, Hassanali A, Kaaya GP and Mueke JM (2003) Attraction of Amblyomma variegatum (ticks) to the attraction-
aggregation-attachment pheromone with or without carbon dioxide. Exp Appl Acarol 29: 121-130.
Maranga RO, Hassanali A, Kaaya GP and Mueke JM (2006) Performance of a prototype baited-trap in attracting and
infecting the tick Amblyomma variegatum (Acari: Ixodidae) in field experiments. Exp Appl Acarol 38: 211-218.
Martinez ML, Machado MA, Nascimento CS, Silva MVGB, Teodoro RL, Furlong J, Prata MCA, Campos AL, Guimaraes MFM,
Azevedo ALS, Pires MFA and Verneque RS (2006) Association of BoLA-DRB3.2 alleles with tick (Boophilus microplus)
resistance in cattle. Genetics Mol Res 5: 513-524.
McCain WC and Leach GJ (2007) Repellents used in fabric: The experience of the U.S. military. In: Debboun M, Frances
SP and Strickman D (eds) Insect repellents principles, methods, and uses. CRC Press, New York, USA, pp. 261-274.
McNemee RB Jr, James WJ IV and Maloney FA Jr (2003) Occurrence of Dermacentor variabilis (Acari: Ixodidae) around a
porcupine (Rodentia: Erthethizontidae) carcass at Camp Ripley, Minnesota. J Med Entomol 40: 108-111.
McMahon C and Guerin PM (2000) Responses of the tropical bont tick, Amblyomma variegatum (Fabricius), to its
aggregation-attachment pheromone presented in an air stream on a servosphere. J Comp Physiol A 186: 95-103.
McMahon C and Guerin PM (2002) Attraction of the tropical bont tick, Amblyomma variegatum, to human breath and
to the breath components acetone, NO and CO2. Naturwissen 89: 311-315.
McMahon C, Guerin PM and Syed Z (2001) 1-octen-3-ol isolated from bont ticks attracts Amblyomma variegatum. J
Chem Ecol 27: 471-486.
McMahon C, Kröber K and Guerin PM (2003) In vitro assays for repellents and deterrents for ticks: differing effects of
products when tested with attractant or arrestment stimuli. Med Vet Entomol 17: 370-378.
Mehlhorn H, Schmahl G, Mencke N and Bach T (2003) In vitro and in vivo studies on the effect of a combination
containing 10% imidacloprid and 50% permethrin against Ixodes ricinus ticks. Parasitol Res 89: 323-325.
Mohamed FSA, Abbassy MM, Darwish ZEA, Tetreault, Matzouk AS, Shoukry MA, Fryauff DJ and Beavers GM (2000)
Effects of permethrin on the salivary glands and neuroendocrine organs of unfed female Hyalomma (Hyalomma)
dromedarii (Ixodoidea: Ixodidae). J Med Entomol 37: 393-400.
Neitz AWH and Gothe R (1984) Investigations into the volatility of female pheromones and the aggregation-inducing
property of guanine in Argas (Persicargas) walkerae. Onderstepoort J Vet Res 51: 197-201.
Newhouse V (1983) Variations in population density, movement and rickettsial infection rates in a local population of
Dermacentor variabilis (Acarina: Ixodidae) ticks in the Piedmont of Georgia. Environ Entomol 2: 1737-1746.
Norval RAI and Rechav YH (1979) An assembly pheromone and its perception in the tick Amblyomma variegatum
(Acarina, Ixodidae). J Med Entomol 16: 507-511.
Norval RAI, Yunker CE and Butler JF (1987) Field sampling of unfed adults of Amblyomma hebraeum Koch. Exp Appl
Acarol 3: 213-217.
Norval RAI, Butler JF and Yunker CE. (1989a) Use of carbon dioxide and natural or synthetic aggregation-attachment
pheromone of the bont tick, Amblyomma hebraeum, to attract and trap unfed adults in the field. Exp Appl Acarol
7: 171-180.

346  Olfaction in vector-host interactions

 15. Chemical ecology of tick-host interactions

Norval, RAI, Andrew HR and Yunker CE (1989b) Pheromone-mediation of host-selection in bont ticks (Amblyomma
hebraeum Koch). Science 243: 364-365.
Norval, RAI, Andrew HR and Yunker CE (1990) Infection rates with Cowdria ruminantium of nymphs and adults of the
bont tick (Amblyomma hebraeum) collected in the field in Zimbabwe. Vet Parasitol 36: 277-283.
Norval RAI, Peter T, Yunker CE, Sonenshine DE and Burridge MJ. (1991) Responses to the ticks Amblyomma hebraeum
and A. variegatum to known or potential components of the aggregation-attachment pheromone. II. Attachment
stimulation. Exp Appl Acarol 13: 19-26.
Norval RAI, Peter T, Yunker CE, Sonenshine DE and Burridge MJ (1993) Response of the ticks Amblyomma hebraeum
and A. variegatum to known or potential components of the aggregation-attachment pheromone. III. Short-range
attraction and aggregation. Exp Appl Acarol 16: 237-245.
Norval RAI, Sonenshine DE, Allan SA and Burridge MJ (1996) Efficacy of pheromone-acaricide-impregnated tail-tag
decoys for controlling the bont tick, Amblyomma hebraeum (Acari: Ixodidae) on cattle in Zimbabwe. Exp Appl
Acarol 20: 31-46.
Osterkamp J, Wahl U, Schmulfuss G and Haas W (1999) Host-odour recognition in two tick species is coded in a blend
of vertebrate volatiles J Comp Physiol A 185: 59-67.
Otieno DA, Hassanali A, Obenchain FD, Sternberg A and Galun R (1985) Identification of guanine as an assembly
pheromone of ticks. Insect Sci Appl 6: 667-670.
Palsson K, Jaenson TGT, Baeckstrom and Borg-Karlson A (2008) Tick repellent substances in the essential oil of Tanacetum
vulgare. J Med Entomol 45: 88-93.
Perritt, DW, Couger G and Barker RW (1993) Computer–controlled olfactometer system for studying responses of ticks
to carbon dioxide. J Med Entomol 30: 571-578.
Philis WA and Cromroy HL (1977) The microanatomy of the eye of Amblyomma americanum (Acari: Ixodidae) and
resultant implications of its structure. J Med Entomol 13: 685-698.
Randolph SE, Gern L and Nuttall PA (1996) Co-feeding ticks: Epidemiological significance for tick-borne pathogen
transmission. Parasitol Today 12: 472-479.
Rechav Y, Goldberg M and Fielden L (1997) Evidence for attachment pheromones in the Cayenne tick (Acari: Ixodidae).
J Med Entomol 34: 234-237.
Rechav Y, Norval RAI, Tannock J and Colborne J (1978) Attraction of the tick Ixodes neitzi to twigs marked by the
klipspringer antelope. Nature 275: 310-311.
Robertson HM and Kent LB (2009) Evolution of the gene lineage encoding the carbon dioxide receptor in insects. J
Insect Sci 9: 19, available online: insectscience.org/9.19. Accessed 5 Oct 2009.
Sauer JR, Hair JA and Houts MS (1974) Chemo-attraction in the lone star tick (Acarina: Ixodidae). 2. Responses to various
concentrations of CO2. Ann Entomol Soc Amer 67: 150-152.
Schöni, R, Hess E, Blum W and Ramstein K. (1984) The aggregation-attachment pheromone of the tropical bont tick
Amblyomma variegatum, Fabricius (Acari: Ixodoidea): isolation, identification and action of its components. J Insect
Physiol 30: 613-618.
Schreck CE, Fish D and McGovern TP (1995) Activity of repellents applied to skin for protection against Amblyomma
americanum and Ixodes scapularis ticks (Acari: Ixodidae). J Amer Mosq Control Assoc 11: 136-140.
Schulze TL, Jordan RA and Hung RW (1997) Biases associated with several sampling methods used to estimated
abundance of Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae). J Med Entomol 34: 615-623.
Shorey HH (1977) Interaction of insects with their chemical environment. In: Shorey HH and McKelvey JJ Jr (eds)
Chemical control of insect behavior theory and application. John Wiley & Sons. New York, USA, pp. 1-6.
Solberg VB, Neidhardt K, Sardelis MR, Hildbrandt C, Hoffmann FJ and Boobar LR (1992) Quantitative evaluation of
sampling methods for Ixodes dammini and Amblyomma americanum (Acari: Ixodidae). J Med Entomol 29: 451-456.
Sonenshine DE (1991) Biology of ticks, Volume 1. Oxford University Press, New York, USA.
Sonenshine, DE (1993) Biology of ticks, Volume 2. Oxford University Press, New York, USA.
Sonenshine DE (2004) Pheromones and other semiochemicals of ticks and their use in tick control. Pararsitol 129:
S405-S425.
Sonenshine DE (2006) Tick pheromones and their use in tick control. Ann Rev Entomol 51: 557-580.

Olfaction in vector-host interactions 347

Sandra A. Allan

Sonenshine DE, Adams T, Allan SA, McLaughlin JR and Webster FX (2003) Chemical composition of some components of
the arrestment pheromone of the black-legged tick, Ixodes scapularis (Acari: Ixodidae), and their use in tick control.
J Med Entomol 40: 849-859.
Stafford KC III, Bladen VC and Magnarelli LA (1995) Ticks (Acari: Ixodidae) infesting wild birds (Aves) and white-footed
mice in Lyme, CT. J Med Entomol 32: 453-466.
Steullet P (1993) Perception of vertebrate volatiles in the tropical bont tick, Amblyomma variegatum Fabricius.
Dissertation, Université de Neuchâtel, Neuchâtel, Switzerland.
Steullet P and Guerin PM (1992a) Perception of breath components by the topical bont tick, Amblyomma variegatum
Fabricius (Ixodidae) I. CO2-excited and CO2-inhibited receptors. J Comp Physiol A 170: 665-676.
Steullet P and Guerin PM (1992b) Perception of breath components by the tropical bont tick Amblyomma variegatum
Fabricius (Ixodidae) II. Sulfide receptors. J Comp Physiol A 170: 677-685.
Steullet P and Guerin PM (1994a) Identification of vertebrate volatiles stimulating olfactory receptors on tarsus 1 of the
tick Amblyomma variegatum Fabricius (Ixodidae) I. Receptors within the Haller’s organ capsule. J. Comp Physiol A
174: 27-38.
Steullet P and Guerin PM (1994b) Identification of vertebrate volatiles stimulating olfactory receptors on tarsus 1 of the
tick Amblyomma variegatum Fabricius (Ixodidae) II. Receptors outside the Haller’s organ capsule. J. Comp Physiol
A 174: 39-47.
Taneja J and Guerin PM (1997) Ammonia attracts the hemaematophagous bug Triatoma infestans: behavioural and
neurophysiological data on nymphs. J Comp Physio A 181: 21-34.
Waladde SM (1987) Receptors involved in host location and feeding in ticks. Insect Sci Appl 8: 643-647.
Waladde SM and Rice MJ (1982) The sensory basis of tick feeding behaviour. In: Obenchain FD and Galun R (eds) The
physiology of ticks. Pergamon Press, Oxford, UK, pp 71-118.
Wanzala W, Sika NFK, Gule S and Hassanali A (2004) Attractive and repellent host odours guide ticks to their respective
feeding sites. Chemoecology 14: 229-232.
Wang M, Guerrero FD, Pertea G and Nene VM (2007) Global comparative analysis of ESTs from the southern cattle tick,
Rhipicephalus (Boophilus) microplus. BMC Genomics 8: 368.
Weldon PJ and Carroll JF (2007) Vertebrate chemical defense: secreted and topically acquired deterrents of arthropods.
In: Debboun M, Frances SP and Strickman D (eds) Insect epellents principles, methods, and uses. CRC Press, New
York, USA, pp. 47-75.
Wilkinson PR (1953) Observations on the sensory physiology and behaviour of larvae of the cattle tick, Boophilus
microplus (Can.)(Ixodidae). Austr J Zool 1: 345-516.
Wilson JG, Kinzer DR, Sauer JR and Hair JA (1972) Chemo-attraction in the lone star tick (Acarina: Ixodidae). I. Response
of different developmental stages to carbon dioxide administered via traps. J Med Entomol 9: 245-252.
Yoder JA, Atwood AD and Stevens BW (1998) Attraction to squalene by ticks (Acari: Ixodidae): first demonstration of a
host-derived attractant. Internat J Acarol 24: 143-147.
Yoder JA, Ark JT and Farrell AC (2008) Failure by engorged stages of the lone star tick, Amblyomma americanum, to react
to assembly pheromone, guanine and uric acid. Med Vet Entomol 22: 135-139.
Yunker CD, Peter T, Norval RAI, Sonenshine DE, Buridge MJ and Butler JF (1992) Olfactory responses of adult Amblyomma
hebraeum and A. variegatum (Acari: Ixodidae) to attractant chemicals in laboratory tests. Exp. Appl. Acarol 13: 295-
301.

348  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

16. Parasite-mediated enhancement of transmission by

haematophagous insects

Hilary Hurd

Abstract

Blood-sucking arthropods are vectors for a variety of parasites and pathogens, including many
that infect man and his domestic animals. Infective stages are imbibed during blood feeding
and their life cycle is completed when they are transferred back to their vertebrate host during
another blood-feeding episode. Transmission of vector-borne parasites is therefore dependent
upon contact between hosts and haematophagous arthropods. In order for blood feeding to
occur, a vector performs a series of behaviour patterns that are initiated in response to both
endogenous and exogenous triggers; the latter including odour signals from the host. If the host
is infected these signals may change in ways that makes them more attractive to the vector of the
parasite. In addition, infection-induced lethargy in the host may make blood feeding less risky and
blood that has fewer erythrocytes is easier to imbibe. When vectors are infected several aspects of
blood-feeding behaviour may be changed in ways that increase host contact or facilitate feeding.
Mosquitoes infected with rodent malarias have been shown to exhibit increased response to host
odours, to have a lower blood volume threshold at which they cease feeding and to show increased
feeding persistence in the face of host-defensive behaviour. Increased feeding persistence is also
observed in Leishmania-infected sand flies, as is increased biting behaviour. The latter is due to
blockage of the gut with promastigote secretory gel. Vector infections also alter mechanisms that
combat host haemostasis, such as the mosquito anti-platelet aggregation factor, apyrase. This
causes additional probing attempts when sporozoite infections are present. The evidence that
some vector-transmitted parasites and pathogens increase transmission prospects by altering the
blood-feeding process suggests that they are manipulating the host or the vector. Host contact
determines the spread of infection thus information concerning parasite-induced changes in
haematophagy needs to be incorporated into epidemiological models.

Keywords: vectors, parasites, transmission, haematophagy, blood feeding, behaviour manipulation

Introduction

The particular interest in the olfactory response of haematophagous insects witnessed in this
volume is undoubtedly prompted by their capacity to transmit the causative agents of infectious
diseases to humans and their domestic animals. The majority of these vector-borne infectious
agents have complex lifecycles that require two or more hosts and one of the major problems
associated with these parasitic lifestyles is the safe transmission from host to host. Parasites have
evolved different strategies to solve the various problems associated with transmission, one of
which is a reliance on insect vectors to ferry them between hosts. Insects that are used as vectors
invariably visit vertebrates to feed upon their blood or other body fluids, thus a prerequisite
for this form of transmission is that the parasite is in easy reach of an insect vector when it is
feeding. The transmission stages of these parasites thus inhabit the blood or superficial skin layers
(reviewed in Lehane 2005) and their vectors are usually females requiring a protein rich meal for
egg production (but see Otranto et al. 2008). Not only does vertebrate blood provide a nutritious
food source for haematophagous insects, it is also an ideal environment for the development of a

Olfaction in vector-host interactions 349

Hilary Hurd

multitude of parasitic organisms ranging in size from arboviruses to helminths (for lists see Lehane
2005 and Otranto et al. 2008).

Vector insects were once regarded as flying syringes, doing little more than sucking up parasites in
one host and injecting them into another. It is now recognised that complex molecular, biochemical
and behavioural interactions occur between parasites and vectors. These interactions affect the
fitness of the insect and the outcome of parasite transmission (Hurd 2003). The exception to
this is when mechanical transmission occurs, during which pathogens are rapidly moved from
insect to insect on contaminated mouth parts. One example is Stomoxys nigra Macquart, which
transmit Trypanosoma evansi Steel in this way (Mihok et al. 1995, Sumba et al. 1998). The more
general pattern is biological transmission, where the parasite may multiply in the insect vector,
change into another developmental stage or both change and multiply (Hamilton and Hurd 2002,
Lehane 2005). All of the developmental stages may occur within the midgut lumen, as is the case
for Leishmania spp. infecting sand flies or Trypanosoma cruzi Chagas that multiply in the crop and
midgut of the triatomine bug, Triatoma infectans Klug, and are deposited on the skin when the bug
defecates during blood feeding. In many cases, however, the gut wall is traversed and parasites
invade tissues such as the thoracic muscle in the case of the filarial worms Wuchereria bancrofti
(Cobbold) and Onchocerca volvulus Bickel or the Malpighian tubules, as does Dirofilaria immitis
(Leidy). Once these worms have moulted into L3 larvae they make their way to the mouthparts,
ready for transmission back to a mammalian host during the vector’s next blood meal. Although
the malaria parasite, Plasmodium, passes through the midgut cells, it remains on the gut wall
beneath the basal lamina undergoing sporogony to produce the infective sporozoites which
migrate to the salivary gland and are transferred into the vertebrate host in the saliva.

Pathogens transmitted by haematophagous insects interact with their vertebrate hosts at

many different levels, often causing severe morbidity or mortality. Many of the diseases that
disproportionately affect the poor and marginalised populations of the world are caused by
insect-transmitted parasites. These diseases include malaria, caused by protozoan parasites of the
genus Plasmodium; sleeping sickness, caused by the flagellates Trypansoma brucei rhodesiense and
T. b. gambiense; Chagas’ disease, caused by Trypanosoma cruzi; leishmaniasis cause by Leishmania
spp.; lymphatic filariasis caused by Wuchereria bancrofti or Brugia malayi Brug; river blindness
caused by Onchocerca volvulus and dengue fever, caused by DEN viruses.

Once established in haematophagous insects, many vector transmitted parasites interact with
their hosts in ways that result in a decrease in vector longevity and /or reproductive fitness (see
reviews by Ferguson and Read 2002, Hurd 1993, Hurd 2003 and Moore 1993). In keeping with
the theme of this volume, this chapter will focus on the point of contact between vector and
vertebrate host and how this contact is affected by pathogens and parasites that may be present
in host or vector.

For transmission to occur, a minimum of two contacts must be made, one to transmit the infectious
organism to the vector and one to return it to a vertebrate host. Thus I will consider transmission
from an infected host to an uninfected insect and from an infected insect back to a vertebrate;
thereby considering transmission at both stages of the parasite’s life cycle. The blood feeding
that causes both of these contacts to occur can be broken down into a series of vector behaviour
patterns namely; appetitive search, attraction to the host, landing, probing and feeding (Hamilton
and Hurd 2002). There is sparse, but growing, evidence that several aspects of this sequence of
events can be altered by infectious agents and that, although the mechanism underlying these
changes may differ, the resulting consequences are similar. This chapter will present an overview

350  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

of parasite-induced changes in blood-feeding behaviour that affect each point of contact between
vector and host. As much of this was reviewed in Hamilton and Hurd (2002) and Hurd (2003), and
as space is limiting, the focus of this article will be upon detailing more recent studies.

The attractiveness of the odour of infected-hosts

Parasitic infection is known to modify the odour profile of host urine, exhaled breath or skin
microbial flora (Braks et al. 1999, Hamilton and Hurd 2002, Kavaliers and Colwell 1995, Penn and
Potts 1998). These odours can be detected by conspecifics and may act as chemical signals during
social interactions (Kavaliers and Colwell 1995). In mice they play a role in avoidance behaviours
and mate choice (Barthelemy et al. 2005, Kavaliers et al. 1998). It is conceivable that over
evolutionary time selection pressures will have favoured parasites that exploit or manipulate host
odour changes that influence the host-seeking behaviour of the respective vector, and thereby
enhance their transmission.

The odour profile of individual hosts differ considerably (Qiu et al. 2006, Rebollar-Tellez et al.
1999) and haematophagous insects clearly distinguish between them, as reviewed by Clements
(1999). Surprisingly, there are very few published studies that compare the response of vectors
to parasite-infected hosts and most of these have examined the response of mosquitoes to
malaria or arboviral-infected hosts (reviewed by Clements (1999)). Sand flies have been shown
to be more attracted to Leishmania-infected dogs (Coleman et al. 1988) and, in a bioassay of
the entrained odours of hamsters infected with L. infantum Nicolle, they were shown to be
more attractive to the sand fly, Lutzomyia longipalpis (Lutz and Neiva), than uninfected hamster
odour (O’Shea et al. 2002). Likewise, pilot studies with rodent malaria-infected mice showed a
similar preferential attraction of uninfected anopheline mosquitoes to the odours of infected
hosts, attractiveness increasing with parasitaemia. (reviewed by Hamilton and Hurd (2002)). The
latter studies corroborated the earlier findings of Day and colleagues (Day et al. 1983) studying
mice infected with rodent malarias. Increased attractiveness of a host will only benefit parasite
transmission if stages infective to the vector are present. In the case of malaria parasites these are
the gametocytes. Rodent malarias have been shown to cause the mouse host to be most attractive
when gametocytes are circulating, suggesting that these hosts may receive more bites and thus
infect more mosquitoes (Day and Edman 1983). A positive association was found between both
asexual parasitaemia and gametocyte density and the proportion of mosquitoes feeding on P.
chabaudi-infected mice at 14 days post-infection (when gametocytes are present in this species)
(Ferguson et al. 2003). Mosquitoes four days post-infection with P. chabaudi I. Landau & A. Chabaud
were also found to be more attracted to infected mice than uninfected ones when offered a
second blood meal (Ferguson and Read 2004). However, chickens infected with P. gallinaceum
Brumpt were found to be less attractive to Aedes aegypti (L.) than uninfected ones (Freier and
Friedman 1976). Although this latter study did not state whether gametocytes were present, other
experiments presented in the paper suggest their chickens would have had a parasitaemia of at
least 28%. Plasmodium gallinaceum gametocytes are present in chickens with rising parasitaemias
of 10+% (Warburg et al. 1992), thus in this species gametocyte presence does not seem to make
the host more attractive and the authors ascribed the decrease in attractiveness of infected hosts
to infection associated metabolic changes that lead to odour changes (Day and Edman 1983).

The majority of studies of rodent malarias are made in the laboratory, using host / parasite / vector
models that do not occur in nature and would not have co-evolved. If increased host attractiveness
at a time of peak infectivity also occurs in malaria-infected humans then the increase in biting rate
and hence increased parasite transmission that this may cause would have important implications

Olfaction in vector-host interactions 351

Hilary Hurd

for disease epidemiology. This is predicted for example in the simulations made by Kingsolver
(1987) using increasing-preference and switching behaviour models that took account of non-
random feeding behaviours. For example, increasing consistent host preference for infected hosts
was predicted to make it easier to maintain a stable infection, relative to the random choice
model (Kingsolver 1987). A very elegantly designed study undertaken in the field in western Kenya
was able to confirm that non-random feeding on malaria-infected humans does occur (Lacroix
et al. 2005). Twelve groups of three children between 3 and 5 years old were initially exposed to
Anopheles gambiae Giles s.s. mosquitoes. One child in each group was uninfected, one was naturally
infected with the asexual stages of P. falciparum Welch (that are not transmissible to mosquitoes)
and the third child harboured gametocytes. None of these children had symptomatic malaria. The
same children were then given anti-malaria treatment and again exposed to mosquitoes when
parasites had been cleared from the blood. In the first assay, gametocytaemic children were twice
as attractive as other children, but this increased attractiveness disappeared once the infection
had been cleared. Thus increased attractiveness was a function of the presence of transmission
stages of the parasite alone, and not due to an intrinsic attractiveness of this group of children.
In contrast, a field study undertaken in Papua New Guinea failed to detect any increase in the
attractiveness of Anopheles punctulatus Dönitz to malaria or filarial-infected individuals (Burkot
et al. 1989).

It is likely that other vector-transmitted parasites also increase host attractiveness as has been
shown in a field studiy that detected greater attraction of Glossina pallidipes Austen towards oxen
infected with T. congolense Broden than uninfected or T. vivax Grassi and Feletti-infected oxen
(Baylis and Mbwabi 1995b). However there are surprisingly few studies that have examined this
issue, even though it is of such importance for disease epidemiology. Other practical aspects
that may follow from studies of host odour changes are the potential for diagnosis of infectious
diseases using odour profiles and the value of using infection-specific odours as baits to lure
haematophagous insects away from their targets.

Host temperature and attractiveness

Alteration in long range attraction to infected hosts is likely to be the result of changes in odour
profile, however, at close range host temperature may also come into play. Are vectors attracted
to hosts with elevated temperatures? Increased body temperature is associated with many
parasitic infections making this an attractive proposition, but this hypothesis is not supported by
empirical evidence, in fact quite the contrary. Studies of arbovirus-infected chickens, Rift Valley
Fever virus-infected lambs and rodents infected with malaria all failed to make an association
between increased feeding and elevated host-temperatures. Indeed, mice infected with rodent
malaria experience depressed rather than elevated temperatures, yet are more attractive to Ae.
aegypti and Culex quinquefasciatus Say (reviewed by Clements 1999). Although these studies did
not demonstrate a role for infection-induced temperature elevation in altering the attractiveness
of infected hosts, the relationship between host temperature and parasite transmission may not
be obvious. For instance, the increased attractiveness of pregnant women to mosquitoes has
been ascribed to elevation of body temperature and an increased release of volatile substances
(Lindsay et al. 2000). This may, in part, explain the increased vulnerability to malaria experienced
by pregnant women (Ansell et al. 2002).

352  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

Defensive behaviour of infected hosts

Animals react to insect blood feeding with defensive behavioural responses that are likely to
interrupt the vector feeding process and lead to the taking of multiple meals (Klowden and Lea
1979). This increases the chance that a vector will feed on several different hosts, including some
infected ones. However, the pathology associated with some vector transmitted diseases induces
host lethargy thus decreasing defensive behaviour, making it easier for the vector to fully engorge.
For example, Plasmodium infection causes lethargy in infected mice, which limits their defensive
behaviour; the timing of the onset of lethargy varying depending upon the species of rodent
malaria. The vulnerability of infected mice to mosquito biting was shown to increase in synchrony
with increased lethargy (Day et al. 1983). The advantage of feeding on a defenceless host seems
obvious, yet few examples of increased blood feeding on lethargic hosts exist, as discussed by
Clements (1999) and Moore (1993). This could be a consequence of the lack of investigations
that would demonstrate it. Alternatively, it may not be as advantageous for the vector as it first
seems. If a parasite decreases the fitness of its vector then an easy to come by, but infective, meal
will have its own costs. Selection pressured that mitigate against favouring infected hosts when
foraging may be operating. Additionally, in terms of transmission success, the times at which
transmission stages of the parasite are present may not coincide with disease symptoms that
induce pathology, as seen for instance in P. falciparum infection, where gametocytes are largely
present in asymptomatic people.

Even when defensive behaviour fails, vertebrates are not the passive victims of the blood feeding
attempts of other organisms. They have a battery of internal defence mechanisms to combat
this theft of blood. These are reviewed by Lehane (2005) and include haemostasis mechanism
such as coagulation and platelet aggregation that control bleeding at the wound site. As is to be
expected, an arms race has ensued to produce counter measures to these defences. Thus the saliva
of haematophagous arthropods in a complex mix of biologically active proteins that assist the
feeding mechanism by resisting host haemostasis, including anticoagulants, immunomodulators,
vasodilators and anti-platelet compounds. It is becoming evident that many of these molecules
have been usurped by vector transmitted parasites to enhance their transmission prospects.

One such example is seen in ticks. Unlike the brief encounter that blood-sucking insects have
with their hosts, ticks engorge on vertebrates for several days, engendering the recruitment of
inflammatory cells such as neutrophils to the bite site in an attempt to disrupt tick feeding. These
efforts are countered by an array of salivary gland proteins (Nuttall and Labuda 2003). One such is
the antioxidant Salp25D, expressed in the midgut and salivary glands of the tick Ixodes scapularis
Say (Das et al. 2001). This tick is a vector for several pathogens including the spirochete Borrelia
burgdorferi (Johnson), causative agent of Lyme disease. By silencing salivary gland salp25D using
dsRNA injection, Narasimhan and colleagues (Narasimhan et al. 2007) showed that, although
engorgement was not affected, acquisition of a spirochete burden in the midgut was significantly
reduced by the first 24 h of feeding, compared to control ticks. In contrast, transmission of Borrelia
from infected ticks to mice was not impaired. This study further demonstrated that Salp25D is
likely to protect the incoming spirochete from the damaging hydroxylradicles generated at the
bite site. Thus transfer of this pathogen from the vertebrate host to the tick is enhanced by the
activity of a salivary gland protein used by the vector to produce an anti-inflammatory response.
The spirochete is, in effect, piggy backing on a mechanism involved in normal feeding behaviour.

Olfaction in vector-host interactions 353

Hilary Hurd

Blood feeding on an infected host

As we have seen, blood feeding is a dangerous strategy for insects to pursue, as the act of piercing
or biting the host engenders defensive actions on the part of the host that are likely to be lethal
to the vector. The majority of haematophagous arthropods thus visit their feeding site only briefly
and the morphology and mechanics of their highly specialised feeding apparatuses have evolved
to efficiently penetrate the host tissue and obtain a meal in as short a time as possible. Blood is
a viscous fluid and flows up the food canal as a result of the negative pressure applied by the
pharangeal pump and/or the cibarial pump, both located in the head capsule. Models used to
describe the flow of blood during vector feeding suggested that the radius of the food canal has
the largest influence over feeding time (Daniel and Kingsolver 1983), however, the Daniel and
Kingsolver model has been challenged, as discussed by Clements (1999). Terminal diameters of
feeding canals are not much greater than the diameter of single red blood cells and it would appear
that the size may be limited by the need to pierce the skin efficiently and without detection. This
is discussed in detail by Lehane (2005).

Host blood quality

In addition to the mechanics of the insect’s feeding apparatus, the nature of the blood will
influence feeding success. The proportion of erythrocytes present (haematocrit) will affect the
viscosity of the blood and hence its rate of flow, its nutritious value (as much of blood protein
is contained in the red blood cells) and, in the case of parasites that infect erythrocyte, the level
of parasitaemia. Chronic parasitic infections such as malaria can induce long term reductions
in haematocrit which may affect vector feeding success. This was demonstrated in a study of
mice infected with the rodent malaria, P. yoelii nigeriensis (Killick-Kendrick). Mice exhibited a drop
in haematocrit (measured as packed cell volume (PCV)) as days-post-infection and parasitaemia
rose. Early in infection, when PCV has only dropped 1-3%, mosquitoes allowed to feed on
anaesthetised infected mice were able to imbibe significantly larger amounts of haemoglobin
(a proxy measurement for red blood cells) than mosquitoes feeding on control mice. However,
as parasitaemia increased and PCV fell below 6% of the control mice the haemoglobin taken by
mosquitoes feeding on infected mice was significantly reduced (Taylor and Hurd 2001). Thus,
initially, the predicted decrease in viscosity as PCV falls appears to enhance erythrocyte uptake.
The infective, gametocyte stages of P. y. nigeriensis are present early in infection when haematocrit
has fallen only slightly. It is thus likely that their transmission to mosquitoes will be enhanced by
a small drop in PCV.

In addition to increasing erythrocyte uptake, the speed at which blood was initially imbibed was
also demonstrated to increase when the yellow fever mosquito, Ae. aegypti, fed on P. chabaudi
infected mice or hamsters infected with the Rift Valley Fever virus compared with their uninfected
counterparts (Shieh and Rossignol 1992). This was due to a reduction in the time taken to locate
blood during probing which, the authors proposed, may be due to an infection-induced reduction
in the number of host platelets.

Both the reduction in probing time and the increased erythrocyte uptake seen when mosquitoes
feed on slightly anaemic blood may be advantageous. It will enhance the chances that mosquitoes
take enough protein for egg production before host-defensive behaviour is initiated. However,
mosquitoes feeding on very anaemic blood will need to seek further feeding opportunities,
thereby increase risky host contact. If the advantage of obtaining a quick meal outweighs the
disadvantage of incurring an infection, then mosquitoes may favour feeding on slightly anaemic

354  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

hosts. Nevertheless, it must be noted that associations used in these study do not occur in the
field and it would be necessary to repeat these findings in a naturally occurring host/parasite/
vector system before the value of enhanced feeding on an anaemic host could be hypothesised to
enhance transmission. Whether these trade-offs operate in favour of human parasite transmission
in a malaria endemic area, where slight anaemia is likely to be commonplace, has yet to be tested.

The affect of host haematocrit on blood feeding has also been investigated in tsetse flies. The
consumption of ox blood by G. pallidipes decreases as PCV exceeded 30% (Baylis and Mbwabi
1995a). However, although tsetse fly feeding was greater on cattle infected with T. congolense
and T. vivax, this was not attributed to host anaemia but to vasodilation caused by T. congolense
attaching to the wall of the host microvasculature (Moloo et al. 2000).

The importance of correct parasite location in the host

Many vectors exhibit peak biting activity at specific times of the day. For example, An. gambiae
are nocturnal feeders with host landing building up from 22:00 h to 04:00 h and continuing until
just before day break. In contrast, An. funestus Giles show a peak feeding activity around 6:00 h
(reviewed by Clements (1999). Some vector- transmitted parasites have a circadian rhythm that
matches that of their vector. They are present in the peripheral blood vessels during the period
that their vector is most likely to bite. This phenomenon is particularly evident in species of filarial
nematodes. It is known as ‘microfilarial periodicity’ and has been discussed by Lehane (2005). Of
particular interest is the difference in periodicity shown by two strains of W. bancrofti. In one strain
microfilariae congregate in the lungs during the day and are only found in the peripheral blood
vessels at night; coinciding with the biting pattern of their vector Cx. quinquefasciatus. In contrast,
the strain transmitted by the day-biting vector, Ae. polynesiensis (Marks), are subperiodic and
can also be found circulating during the day (Hawking 1962). Parasite periodicity, known as the
‘Hawking phenomenon’, has also been reported to occur in several species of malaria (Gautret and
Motard 1999), but could not be demonstrated following a field study of biting behaviour of An.
gambiae feeding on P. falciparum-infected children in Kenya (Githeko et al. 1993). The significance
of parasite periodicity as a means of enhancing transmission has not been fully determined but,
especially in situations where parasitaemia is low, it could enhance the chances that transmission
stages will be imbibed by a suitable vector and, in the case of Plasmodium, in numbers that
maximise the chances of fertilisation.

Parasites are also known to congregate in locations that vectors have access to. Lutzomyia
longipalpis (Lutz & Neiva) preferentially probes on skin lesions rich in Leishmania mexicana
(Biagi) amastigotes, the infective stage to the vector (Coleman and Edman 1988). Developing
gametocytes of P. falciparum are sequestered in the microcirculature whilst developing, thus
they are only exposed to potential transmission when they are mature. There is also evidence
that suggests that circulating gametocytes are aggregated, rather than having a homeogeneous
distribution. This aggregation is reflected in infection patterns in mosquitoes and may help to
facilitate the chance of macrogametocyte fertilisation in the midgut (Pichon et al. 2000).

The blood-feeding behaviour of infected vectors

Arthropod haematophagy consists of a series of physiological and behavioural steps beginning

with an appetitive search, and proceeding through activation and orientation, attraction,
landing, probing or biting and finally imbibing blood (Clements 1999, Hamilton and Hurd 2002,
Lehane 2005). The initiation and completion of all of these stages is influenced by exogenous

Olfaction in vector-host interactions 355

Hilary Hurd

and endogenous factors including, in some cases, the presence of parasites. Although parasite
transmission will only occur during the latter two behavioural phases, all of the preliminary events
are crucial. We can surmise that changes to any of them that result in more host contact will
increase the prospects of parasite transmission, once the parasite has developed into the life
cycle stage that is transmitted back to the vertebrate host (Hurd 2003, Lefevre et al. 2006). Various
vector-transmitted parasite, including bacteria, protozoans and nematodes do indeed alter one
or more stages of blood-feeding behaviour and, in a few cases, this has been shown to increase
transmission chances. A few such examples have been well documented in the reviews quoted in
this chapter and here the focus will be upon vector/parasite associations that have recently been
explored in greater depth.

Appetitive search and attraction

Once parasites have reached the infective stage of their development, any host contact on the part
of their vector will provide an opportunity for transmission. We can therefore argue that changes
in vector behaviour that enhance appetitive searches and make the vector more responsive to
hosts odours will favour parasite transmission back to the vertebrate, thereby completing its life
cycle. Unfortunately, there are few studies that specifically measure these response in vectors
infected with transmissible stages of parasites. Comparisons need to be made between vectors
with immature and mature parasite stages. Increased attraction at an inappropriate time would
not benefit the parasites and may just be a response, for instance, to food robbery caused by the
parasite increasing a hunger response rather than manipulated behaviour (as seen in T. cruzi-
infected triatomines (Schaub 2006)).

Those studies that have specifically investigated the effect of infection on host attractiveness
have focused on mosquitoes. For example, an increase in the olfactory response of P. gallinaceum
sporozoites-infected Ae. aegypti to host odours has been reported (Rossignol et al. 1986).
Additionally, the proportion of An. stephensi infected with P. chabaudi that took a second blood
meal from an anaesthetised mouse rose from 0.39 to 0.56 if they were infected with oocysts and
this change in behaviour was not influenced by the burden of infection (Ferguson and Read 2004).
The significance of the latter finding is, however, unclear as increased host contact at this stage
of infection would not benefit parasite transmission and the experimental design did not make
it possible to clarify whether the behaviour change was actually due to a change in response to
host odours.

Feeding persistence and the role of host-defensive reactions

As has already been discussed, in a natural situation haematophagy is often interrupted by host-
defensive responses that can result in the death of the insect. Feeding persistence is the repeated
attempt to feed, despite these interruptions.

In the face of host-defensive responses, mosquitoes infected with sporozoites of the rodent
malaria P. y. nigeriensis are significantly more persistent than uninfected mosquitoes in their
feeding attempts when interrupted before any blood is imbibed. This is a complete reversal of
the reduction in feeding persistence that occurs when the parasites is developing on the midgut
wall, before it reaches the infective stage (Anderson et al. 1999). Stage specific alteration of biting
behaviour was also seen by Koella and colleagues, investigating the size of blood meal that would
inhibit further blood feeding following interruption. They showed that Ae. aegypti infected with
oocysts of P. gallinaceum had a lower threshold value for blood meal size and were thus less likely

356  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

to re-feed than uninfected mosquitoes. In contrast, the presence of sporozoites increased the
threshold volume required to inhibit host-seeking behaviour success (Koella et al. 2002).

In addition to these laboratory investigations, several studies have examined the feeding
behaviour of malaria-infected mosquitoes in field situations. Not only are peak feeding times
of anopheline mosquitoes changed (reviewed in Hamilton and Hurd (2002) and Hurd (2003)),
infected mosquitoes have been shown to engorge more fully (Koella and Packer 1996, Koella et
al. 1998, Maxwell et al. 1998) suggesting that more tenacious feeding behaviour also occurs in
naturally infected mosquitoes.

A similar increase in blood-feeding persistence has also been observed to occur in sand flies
infected with the metacyclic promastigote forms (infective to mammalian hosts) of L. infantum or
L. mexicana. Sand flies were allowed to feed on an anaesthetised mouse but feeding was artificially
interrupted by brushing the antennae of the flies. Rogers and Bates found a positive correlation
between feeding persistence and the number of metacyclic promastigotes per fly (Rogers and
Bates 2007). In both cases only the infective-sand fly stage caused this behavioural change.

Changes in biting and probing behaviour

In addition to an increase in feeding persistence, vector biting or probing behaviour has been
shown to be affected by several vector/parasite associations including malaria/mosquitoes,
Leishmania/sand flies, Trypanosoma/tsetse flies and Yersinia pestis/fleas. Of these, recent studies
of the changes in biting behaviour induced by Leishmania infections are beginning to elucidate
mechanisms underlying behavioural changes that have been recognised for over a century and
have been reviewed by Hurd (2003), Ready (2008) and Rogers and Bates (2007).

Sand flies become infected with amastigote forms of Leishmania following an infected blood
meal and the parasites complete their entire developmental process within the sand fly gut.
The final phase takes place in the stomodeal valve where they transform into motile metacyclic
promastigote forms that are infective to mammals. Transmission occurs when these promastigotes
are regurgitated within a gel-like material that fills the cardia and stomodeal valve, causing a
blockage. This promastigote secretory gel (PSG) is formed by the parasites and has been shown
to enhance cutaneous infections when deposited onto the skin with sand fly saliva (Rogers et al.
2004, Titus 1998). A major component of PGS is a filamentous glycoprotein, proteophosphoglycan
(fPPG) (Ilg et al. 1996), which forms the 3D matrix of the plug that blocks the gut (Stierhof et
al. 1999) and keeps the valve open. The production of PSG has been shown to occur is several
Leishmania-sand fly combinations (Stierhof et al. 1999). It has been suggested that the plug
facilitates regurgitation and, by restricting blood flow into the fly, prolongs feeding time and
causes flies to bite more, a concept known as ‘the blocked fly hypothesis’.

Sand flies harbouring infective stages of Leishmania take longer to feed on an anaesthetised
mouse than uninfected flies, but do not probe more often (Rogers and Bates 2007). These authors
concluded that the interruption of feeding as a result of host-defensive behaviour (which is
not occurring when the host is anaesthetised) and the subsequent persistence in attempts to
feed is crucial to engendering multiple feeding attempts, and may be more important than the
‘blocked fly hypothesis’. They suggest that the presence of fPPG may inhibit the functioning of
the mechanoreceptors in the foregut that detect blood flow, thereby altering the hunger state or
increasing the threshold blood volume at which feeding is deceased.

Olfaction in vector-host interactions 357

Hilary Hurd

If this behavioural change is to be advantageous to the parasite, it must increase the basic
reproductive number (R0) of the infection by increasing the chance that an infectious vector will
feed on more than one host. Rogers and Bates (2007) showed that infected sand flies were more
likely to feed on multiple hosts in a laboratory setting.

Promastigote secretory gel is one of the few examples of a product produced by parasites that,
via its action in the vector, appears to manipulate blood-feeding behaviour in a way that increases
parasite fitness by enhancing transmission. Rogers and Bates (2007) argue that this is an example
of true manipulation as it fulfils the criteria set out by (Poulin 1995) and discussed further by
Thomas and colleagues (Thomas et al. 2005). Briefly, it is an example of a complex change in
behaviour that achieves the purpose of increasing transmission chances, it fits its purpose as
increased feeding behaviour only occurs once the infective stages are present and it occurs in
several sand fly/Leishmania combinations. PGS could thus be regarded as a manipulator molecule.

The saliva of haematophagous insects contains a cornucopia of molecules that, in one way or
another, assist the blood-feeding process including anaesthetics, anti-coagulation factors, anti-
platelet factors and vasodilators than combat the haemostatic defences of the host. Our current
knowledge of the nature and wide range of these molecules is reviewed in detail by Lehane (2005).
Many species contain an enzyme, apyrase, which inhibits the action of the platelet-aggregation
factor ADP by converting it to AMP and orthophosphate. The presence of apyrase at the site of a
blood vessel wound made by a probing insect will thus inhibit the sealing of a lanced blood vessel.
This would allow a haematoma to form and the insect to locate blood more quickly and feed more
efficiently, and for longer. This has been aptly demonstrated in mosquitoes where feeding time
has been related to the quantity of apyrase in the glands of different anopheline mosquitoes
(Ribeiro et al. 1985). More recently, Boisson and colleagues (Boisson et al. 2006) confirmed the role
of apyrase in efficient mosquito feeding using gene silencing to reduced the expression of the
apyrase gene AgApy. Saliva from mosquitoes injected with ds-AgAyp exhibited reduced enzyme
activity and was less efficient at inhibiting platelet activity. ds-AgAyp-injected females took twice
as long probing before obtaining a blood meal. The authors suggested that their results indicate
that there may be redundant apyrase activity and also that products of the AgApyLike1 and Ag9
genes may also be involved in inhibiting platelet aggregation.

Malaria sporozoites are transmitted to the vertebrate host during the probing phase of feeding
(Rodriguez and Hernandez-Hernandez 2004). Salivary gland infections are associated with an
increase in the time vectors spend probing. Median blood-location time for Ae. aegypti infected with
the avian malaria, P. gallinaceum, thus increases three-fold compared to uninfected mosquitoes.
This is probably as a result of a four fold decrease in the activity of salivary apyrase, even though
the volume of saliva produced is constant (Rossignol et al. 1984). A similar increase in probing
behaviour has been detected in malaria-infected anopheline mosquitoes, as reviewed recently
(Hamilton and Hurd 2002; Hurd 2003). One study provided evidence that sporozoite infections
cause multiple feeding episodes that lead to enhanced transmission. Using microsatellite markers
to identify the source of blood meals in the midguts of mosquitoes feeding in a field situation,
Koella and colleagues (Koella et al. 1998) were able to demonstrate that 22% of P. falciparum-
infected An. gambiae had fed on more than one person during the night, compared with only 10%
of uninfected females. Multiple feeding on different hosts had previously been detected in Ae.
aegypti infected with P. gallinaceum (Kelly and Edman 1992). However, an alternative interpretation
of these observations is that mosquitoes that become infected may have a propensity to multiple
feeding.

358  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

Malaria sporozoites inhabit the distal lobes of mosquito salivary glands, the same location in
which apyrase is produced (Sterling et al. 1973). Surprisingly we still do not know the mechanism
causing the parasite-associated reduction in apyrase. Is it caused by damage incurred as the
sporozoites migrate through the glands or do the sporozoites secrete an inhibitor that suppresses
transcription or inhibits translation? The former would indicate that transmission is enhanced
by an accidental by-product of pathology the latter that this may be another example of vector
manipulation.

A final example comes not from mosquitoes but from fleas infected with plague bacilli. In nature,
Y. pestis, the causative agent of plague, circulates amongst rodent species generally causing little
overt disease symptoms. Violent outbreaks resulting in high mortality periodically occur amongst
susceptible rodent populations and can increase the risk of human infection. Plague transmission
occurs via flea bites. Within the flea gut, ingested bacteria are found in a mass of coagulated blood
that blocks the proventriculus. The blockage results in the regurgitation of plague bacilli into the
mammalian host during feeding. A coagulase- and fibrinolysin associated (pla) gene encoded
on the bacilli plasmid, pKYP1, has been associated with coagulase activity that facilitates the
formation of the proventricular mass (Cavanaugh 1971). Mcdonough and co-workers (Mcdonough
et al. 1993) have shown that fleas infected with pla+ bacteria had higher mortality than that caused
by pla- bacteria and that mortality was not linked to bacterial load. They suggest that the pla gene
product may disrupt the flea’s digestive process leading to blockage, starvation, and early death.
Increased feeding attempts early in this process will enhance plague transmission. Although
further studies are needed to substantiate these suggestions, the pla gene can be regarded as a
promising manipulator molecule candidate with which Y. pestis may modify its vector’s feeding
behaviour to suit its own ends.

Are parasites manipulating vectors?

Much has been written about host manipulation by parasites. There are many incredible examples
that are favourites amongst teachers and students alike (Moore 2002). In comparison with the
fate of crickets infected with hair worms, or snails with trematode sporocysts in their tentacles,
changes in vector feeding persistence or probing behaviour are far from dramatic. Despite this,
the co-evolutionary implications of host and vector behavioural changes are similar. Many of the
changes we see can be expected, and in a few cases have been shown, to lead to an increase in
vector transmission that may be accompanied by a decrease in vector fitness due to the greater
danger associated with increased host contact. Do these unequal consequences for the symbiotic
partners imply that the parasite is manipulating its host? Alternatively, is the behavioural change
a by-product of pathology caused by the infection or are we seeing attempts by the vector to
compensate for the presence of the parasite? Here is not the place to detail the pros and cons of
these theories, which have been much discussed (most recently by Lefevre et al. (2006) and Lefevre
and Thomas (2008), and until we fully understand the mechanisms that govern them, it is difficult
to do much more than speculate.

Parasite fitness can be expressed as R0 which, for microparasites, can be defined as the number of
secondary infections that arise from a primary infection. Following the seminal work of Ross and
Macdonald (Macdonald 1957), models developed to take into account factors that affect R0 were
developed by Anderson and May (1979, 1992) and have since been modified and refined by many
authors interested in the epidemiology of infectious disease.

Olfaction in vector-host interactions 359

Hilary Hurd

Host contact is probably the most important factor limiting the spread of vector transmitted
diseases and this concept, known as vectorial capacity, is incorporated into many of these models.
Saul (2003) developed two deterministic models, one for stable transmission and one for epidemic
situations. They were based on the cyclic feeding model he and his colleagues had previously
developed (Saul et al. 1990) rather than the a continuous feeding model refined from Ross’ model
by Macdonald (1957). Saul’s model assumed that mosquitoes will not feed for a few days after
becoming engorged; whilst the respective gonotrophic cycle is completed (cyclical feeding).
His model showed that search-related mortality of mosquitoes will have an important impact
on malaria transmission. The sensitivity of malaria transmission to mosquito mortality has been
also recognised by others including Smith and McKenzie (2004). One parameter of Saul’s model
(Saul 2003) is the ‘constant attraction rate constant’. Saul recognised that this will, in fact, be
dependent upon factors such as host emission of attractants and host availability to vectors (e.g.
are they sleeping under bednets?). Saul’s model was used to explore the effectiveness of using
‘bait’ animals to lure vectors away from biting humans and predicted that insecticide treatment
of livestock will have an appreciable effect on the human inoculation rate (Saul 2003).

The classical models did not recognise that the parasite itself may have a role to play in affecting
host contact. We have seen that there is evidence that parasitic infections of both mammalian hosts
and arthropod vectors alter feeding behaviour in ways that increase host contact; that infected
hosts are more attractive than uninfected ones and that mosquito biting is heterogeneous.
The assumptions that mosquitoes, for instance, genuinely feed at random and that infected
mosquitoes make the same host contacts as uninfected ones is probably erroneous. Smith et al.
(2007) discuss classical and neoclassical models that estimate R0 for malaria and reconsider R0 in
finite population with heterogeneous biting. They recognise the importance of R0 as an important
metric with which to plan malaria control programmes and that, amongst other measures, it may
be more productive to target interventions towards those individuals who are bitten most, than
those who are bitten least. Clearly it is germane to incorporate many more specific features of the
parasite’s life cycle and its effect on vector feeding behaviour into mathematic models that are to
be of use for planning control strategies.

Conclusions
Parasite-induced changes in vector haematophagy have been recognised for over a century.
However the overriding impression that I have gained whilst preparing this chapter is that, relative
to the number of parasites transmitted by vectors, the documented cases of parasites changing
vector feeding behaviour is very small and, in particular, the epidemiological impacts of these
changes have been neglected. Investigations have mostly focused on malaria and Leishmania
infections and we are certainly in no position yet to assume that this is a common strategy that
parasites employ to enhance transmission. Hopefully, the burgeoning of studies of olfaction in
vector-host interactions will provide further opportunities to incorporate more parasites into
these investigations.

Although observations of biting behaviour in a laboratory setting are useful, they are severely
limited as to the information they can give us about the consequences for parasite transmission in
the field. Studies performed in field settings with natural hosts that can display normal defensive
behaviours must be performed to evaluate laboratory findings, although ethical issues associated
with investigating human infections must be considered. The few investigations of this kind that
have been performed to date are exemplars to be followed.

360  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

Parasitic infection can alter several host behaviour patterns simultaneously and these changes
can themselves alter as parasites mature. Experiments need to be planned to isolate individual
aspects of feeding behaviour and determine which are affected, and when. Only then can we can
decide whether, for instance, changes in feeding persistence have more influence on parasite
transmission than changes in biting or probing.

In addition, to gaining an insight into the degree to which vector feeding behaviour can affect
disease transmission, a knowledge of the mechanisms underlying these changes will help us
to evaluate the evolutionary implications of behavioural change (Lefevre et al. 2006). As the
genome sequences of more and more vectors and parasites are published and we move further
into the post genomic era, investigation of protein profiles and gene transcription of nervous and
other tissues that may govern changes in odour perception, flight behaviour, landing, probing
and imbibing blood is becoming feasible. Some promising headway has recently been made by
Lefevre and colleagues who have examined changes in the head proteome of tsetse flies infected
with trypanosomes and of malaria-infected mosquitoes (Lefevre et al. 2007a,b).

The recent expansion of studies on olfactory behaviour into a wide range of disease vectors and
the advanced technologies for more precise understanding of these behaviours, as described
in the preceding chapters, is likely to stimulate incorporation of pathogens and parasites in
these studies. If this happens we will be able to use knowledge such as this to determine who
is manipulating whom, to understand the implication for parasite transmission and to apply our
understanding to the planning of novel disease control strategies such as the deployment of
transgenic technology to alter feeding behaviour and divert vectors away from anthropophagy.

Acknowledgements

The helpful comments of an anonymous referee helped to improve this chapter and they are
gratefully acknowledged.

References

Anderson RA, Koella JC and Hurd H (1999) The effect of Plasmodium yoelii nigeriensis infection on the feeding persistence
of Anopheles stephensi Liston throughout the sporogonic cycle. Proc R Soc Lond B Biol Sci 266: 1729-1733.
Anderson RM and May RM (1979) Population biology of infectious diseases:Part 1. Nature 280: 361-367.
Anderson RM and May RM (1992) Infectious diseases of humans. Oxford University Press, Oxford, UK.
Ansell J, Hamilton KA, Pinder M, Walraven GE and Lindsay SW (2002) Short-range attractiveness of pregnant women to
Anopheles gambiae mosquitoes. Trans R Soc Trop Med Hyg 96: 113-116.
Barthelemy M, Gabrion C and Petit G (2005) Does chronic malaria modify the odours of its male mouse host? Can J
Zool 83: 1079-1086.
Baylis M and Mbwabi AL (1995a) Effect of host packed cell volume on the bloodmeal size of male tsetse flies, Glossina
pallidipes. Med Vet Entomol 9: 399-402.
Baylis M and Mbwabi AL (1995b) Feeding behaviour of tsetse flies (Glossina pallidipes Austen) on Trypanosoma-infected
oxen in Kenya. Parasitology 110 (Pt 3): 297-305.
Boisson B, Jacques JC, Choumet V, Martin E, Xu JN, Vernick K and Bourgouin C (2006) Gene silencing in mosquito salivary
glands by RNAi. Febs Letters 580: 1988-1992.
Braks MAH, Anderson RA and Knols BGJ (1999) Infochemicals in mosquito host selection: Human skin microflora and
Plasmodium parasites. Parasitology Today 15: 409-413.
Burkot T, Paru A, Graves P and Garner P (1989) Human host selection by Anophelines: no evidence for preferential
selection of malaria or microfilariae-infected individuals in a hyperendemic area. Parasitology 98: 337-342.

Olfaction in vector-host interactions 361

Hilary Hurd

Cavanaugh DC (1971) Specific effect of temperature upon transmission of the plague bacillus by the oriental rat flea,
Xenopsylla cheopis. Am J Trop Med Hyg 20: 264-273.
Clements AN (1999) The biology of mosquitoes, volume 2 sensory reception and behaviour. CABI Publishing, Oxford,
UK, 740 pp.
Coleman RE and Edman JD (1988) Feeding-site selection of Lutzomyia longipalpis (Diptera: Psychodidae) on mice
infected with Leishmania mexicana amazonensis. J Med Entomol 25: 229-233.
Coleman RE, Edman JD and Semprevivo LH (1988) Interactions between malaria (Plasmodium yoelii) and leishmaniasis
(Leishmania mexicana amazonensis): effect of concomitant infection on host activity, host body temperature, and
vector engorgement success. J Med Entomol 25: 467-471.
Daniel TL and Kingsolver JG (1983) Feeding strategy and the mechanics of blood sucking in insects. J Theor Biol 105:
661-677.
Das S, Banerjee G, DePonte K, Marcantonio N, Kantor FS and Fikrig E (2001) Salp25D, an Ixodes scapularis antioxidant, is
1 of 14 immunodominant antigens in engorged tick salivary glands. J Infect Dis 184: 1056-1064.
Day JF, Ebert KM and Edman JD (1983) Feeding patterns of mosquitoes (Diptera: Culicidae) simultaneously exposed to
malarious and healthy mice, including a method for separating blood meals from conspecific hosts. J Med Entomol
20: 120-127.
Day JF and Edman JD (1983) Malaria renders mice susceptible to mosquito feeding when gametocytes are most
infective. J Parasitol 69: 163-170.
Ferguson HM and Read AF (2002) Why is the effect of malaria parasites on mosquito survival still unresolved? Trends
Parasitol 18: 256-261.
Ferguson HM and Read AF (2004) Mosquito appetite for blood is stimulated by Plasmodium chabaudi infections in
themselves and their vertebrate hosts. Malar J 3: 12.
Ferguson HM, Rivero A and Read AF (2003) The influence of malaria parasite genetic diversity and anaemia on mosquito
feeding and fecundity. Parasitology 127: 9-19.
Freier JE and Friedman S (1976) Effect of host infection with Plasmodium gallinaceum on the reproductive capacity of
Aedes aegypti. J Invert Pathol 28: 161-166.
Gautret P and Motard A (1999) Periodic infectivity of Plasmodium gametocytes to the vector. A review. Parasite 6: 103-
111.
Githeko AK, Brandlingbennett AD, Beier M, Mbogo CM, Atieli FK, Owaga ML, Juma F and Collins FH (1993) Confirmation
that Plasmodium falciparum has aperiodic infectivity to Anopheles gambiae. Med Vet Ent 7: 373-376.
Hamilton JCG and Hurd H (2002) Parasite manipulation of vector behaviour. In: Lewis EE, Cambell JF and Sukhdeo MVK
(eds) The behavioural ecology of parasites. CABI, Wallingford, UK, pp 259-281.
Hawking F (1962) Microfilaria infection as an instance of periodic phenomena seen in host-parasite relationships. Ann
N Y Acad Sci 98: 940-953.
Hurd H (1993) Reproductive disturbances induced by parasites and pathogens of insects. In: Beckage NE, Thompson SN
and Federici BA (eds) Parasites and pathogens of insects. Academic Press Inc, San Diego, USA, pp 87-105.
Hurd H (2003) Manipulation of medically important insect vectors by their parasites. Annu Rev Entomol 48: 141-161.
Ilg T, Stierhof YD, Craik D, Simpson R, Handman E and Bacic A (1996) Purification and structural characterization of a
filamentous, mucin-like proteophosphoglycan secreted by Leishmania parasites. J Biol Chem 271: 21583-21596.
Kavaliers M and Colwell DD (1995) Discrimination by female mice between the odours of parasitized and non-parasitized
males. Proc R Soc Lond B Biol Sci 261: 31-35.
Kavaliers M, Colwell DD and Choleris E (1998) Parasitized female mice display reduced aversive responses to the odours
of infected males. Proc R Soc Lond B Biol Sci 265: 1111-1118.
Kelly R and Edman JD (1992) Multiple transmission of Plasmodium gallinaceum (Eucoccida, Plasmodiidae) during serial
probing by Aedes aegypti (Diptera, Culicidae) on several hosts. J Med Ent 29: 329-331.
Kingsolver JG (1987) Mosquito host choice and the epidemiology of malaria. Am Nat 130: 811-827.
Klowden MJ and Lea AO (1979) Effect of defensive host behavior on the blood meal size and feeding success of natural
populations of mosquitoes (Diptera: Culicidae). J Med Entomol 15: 514-517.

362  Olfaction in vector-host interactions

 16. Parasite-mediated enhancement of transmission by haematophagous insects

Koella JC and Packer MJ (1996) Malaria parasites enhance blood-feeding of their naturally infected vector Anopheles
punctulatus. Parasitology 113 (Pt 2): 105-109.
Koella JC, Rieu L and Paul REL (2002) Stage-specific manipulation of a mosquito’s host-seeking behavior by the malaria
parasite Plasmodium gallinaceum. Behav Ecol 13: 816-820.
Koella JC, Sorensen FL and Anderson RA (1998) The malaria parasite, Plasmodium falciparum, increases the frequency
of multiple feeding of its mosquito vector, Anopheles gambiae. Proc R Soc Lond B Biol Sci 265: 763-768.
Lacroix R, Mukabana WR, Gouagna LC and Koella JC (2005) Malaria infection increases attractiveness of humans to
mosquitoes. Plos Biol 3: 1590-1593.
Lefevre T, Koella JC, Renaud F, Hurd H, Biron DG and Thomas F (2006) New prospects for research on manipulation of
insect vectors by pathogens. Plos Pathogens 2: 633-635.
Lefevre T and Thomas F (2008) Behind the scene, something else is pulling the strings: Emphasizing parasitic
manipulation in vector-borne diseases. Infect Genet Evol 8: 504-519.
Lefevre T, Thomas F, Ravel S, Patrel D, Renault L, Le Bourligu L, Cuny G and Biron DG (2007a) Trypanosoma brucei brucei
induces alteration in the head proteome of the tsetse fly vector Glossina palpalis gambiensis. Insect Mol Biol 16:
651-660.
Lefevre T, Thomas F, Schwartz A, Levashina E, Blandin S, Brizard JP, Le Bourligu L, Demettre E, Renaud F and Biron DG
(2007b) Malaria Plasmodium agent induces alteration in the head proteome of their Anopheles mosquito host.
Proteomics 7: 1908-1915.
Lehane MJ (2005) The biology of blood-sucking insects. Cambridge University Press, Cambridge, UK. 321 pp.
Lindsay S, Ansell J, Selman C, Cox V, Hamilton K and Walraven G (2000) Effect of pregnancy on exposure to malaria
mosquitoes. Lancet 355: 1972-1975.
Macdonald G (1957) The epidemiology and control of malaria. Oxford University Press, London, UK.
Maxwell CA, Wakibara J, Tho S and Curtis CF (1998) Malaria-infective biting at different hours of the night. Med Vet
Entomol 12: 325-327.
Mcdonough KA, Barnes AM, Quan TJ, Montenieri J and Falkow S (1993) Mutation in the Pla gene of Yersinia pestis alters
the course of the plague bacillus-flea (Siphonaptera, Ceratophyllidae) interaction. J Med Ent 30: 772-780.
Mihok S, Maramba O, Munyoki E and Kagoiya J (1995) Mechanical transmission of Trypanosoma spp by African
Stomoxyinae (Diptera, Muscidae). Trop Med Parasitol 46: 103-105.
Moloo SK, Sabwa CL and Baylis M (2000) Feeding behaviour of Glossina pallidipes and G. morsitans centralis on Boran
cattle infected with Trypanosoma congolense or T. vivax under laboratory conditions. Medical and Veterinary
Entomology 14: 290-299.
Moore J (1993) Parasites and the behavior of biting flies. J Parasitol 79: 1-16.
Moore J (2002) Parasites and the behaviour of animals. Oxford University Press Inc, New York, USA.
Narasimhan S, Sukumaran B, Bozdogan U, Thomas V, Liang X, DePonte K, Marcantonio N, Koski RA, Anderson JF, Kantor F
and Fikrig E (2007) A tick antioxidant facilitates the Lyme disease agent’s successful migration from the mammalian
host to the arthropod vector. Cell Host Microbe 2: 7-18.
Nuttall PA and Labuda M (2003) Dynamics of infection in tick vectors and at the tick-host interface. Adv Virus Res 60:
233-272.
O’Shea B, Rebollar-Tellez E, Ward RD, Hamilton JGC, El Naiem D and Polwart A (2002) Enhanced sandfly attraction to
Leishmania-infected hosts. T Roy Soc Trop Med H 96: 117-118.
Otranto D, Stevens JR, Cantacessi C and Gasser RB (2008) Parasite transmission by insects: a female affair? Trends
Parasitol 24: 116-120.
Penn D and Potts WK (1998) Chemical signals and parasite-mediated sexual selection. Trends Ecol Evol 13: 391-396.
Pichon G, Awono-Ambene HP and Robert V (2000) High heterogeneity in the number of Plasmodium falciparum
gametocytes in the bloodmeal of mosquitoes fed on the same host. Parasitology 121: 115-120.
Poulin R (1995) ‘Adaptive’ changes in the behaviour of parasitized animals: A critical review. Int J Parasitol 25: 1371-1383.
Qiu YT, Smallegange RC, Van Loon JJA, Ter Braak CJF and Takken W (2006) Interindividual variation in the attractiveness
of human odours to the malaria mosquito Anopheles gambiae s. s. Med Vet Entomol 20: 280-287.
Ready PD (2008) Leishmania manipulates sandfly feeding to enhance its transmission. Trends Parasitol 24: 151-153.

Olfaction in vector-host interactions 363

Hilary Hurd

Rebollar-Tellez EA, Hamilton JGC and Ward RD (1999) Response of female Lutzomyia longipalpis to host odour
kairomones from human skin. Physiol Entomol 24: 220-226.
Ribeiro JM, Rossignol PA and Speilman A (1985) Salivary gland apyrase determines probing time in anopheline
mosquitoes. J Insect Physiol 31: 689-692.
Rodriguez MH and Hernandez-Hernandez FD (2004) Insect-malaria parasites interactions: the salivary gland. Insect
Biochem Molec 34: 615-624.
Rogers ME and Bates PA (2007) Leishmania manipulation of sand fly feeding behavior results in enhanced transmission.
Plos Pathogens 3: 818-825.
Rogers ME, Ilg T, Nikolaev AV, Ferguson MAJ and Bates PA (2004) Transmission of cutaneous leishmaniasis by sand flies
is enhanced by regurgitation of fPPG. Nature 430: 463-467.
Rossignol PA, Ribeiro JM and Spielman A (1984) Increased intradermal probing time in sporozoite-infected mosquitoes.
Am J Trop Med Hyg 33: 17-20.
Rossignol PA, Ribeiro JM and Spielman A (1986) Increased biting rate and reduced fertility in sporozoite-infected
mosquitoes. Am J Trop Med Hyg 35: 277-279.
Saul A (2003) Zooprophylaxis or zoopotentiation: the outcome of introducing animals on vector transmission is highly
dependent on the mosquito mortality while searching. Malaria J 2: 32.
Saul AJ, Graves PM and Kay BH (1990) A cyclical feeding model for pathogen transmission and its application to
determine vectorial capacity from vector infection-rates. J Appl Ecol 27: 123-133.
Schaub GA (2006) Parasitogenic alterations of vector behaviour. Int J Med Microbiol 296 37-40.
Shieh JN and Rossignol PA (1992) Opposite influences of host anemia on blood feeding rate and fecundity of mosquitoes.
Parasitology 105: 159-163.
Smith DL and McKenzie FE (2004) Statics and dynamics of malaria infection in Anopheles mosquitoes. Malar J 3: 13.
Smith DL, McKenzie FE, Snow RW and Hay SI (2007) Revisiting the basic reproductive number for malaria and its
implications for malaria control. PLoS Biol 5: e42.
Sterling CR, Aikawa M and Vanderberg JP (1973) The passage of Plasmodium berghei sporozoites through the salivary
glands of Anopheles stephensi: an electron microscope study. J Parasitol 59: 593-605.
Stierhof YD, Bates PA, Jacobson RL, Rogers ME, Schlein Y, Handman E and Ilg T (1999) Filamentous proteophosphoglycan
secreted by Leishmania promastigotes forms gel-like three-dimensional networks that obstruct the digestive tract
of infected sandfly vectors. Eur J Cell Biol 78: 675-689.
Sumba AL, Mihok S and Oyieke FA (1998) Mechanical transmission of Trypanosoma evansi and T-congolense by Stomoxys
niger and S-taeniatus in a laboratory mouse model. Med Vet Entomol 12: 417-422.
Taylor PJ and Hurd H (2001) The influence of host haematocrit on the blood feeding success of Anopheles stephensi:
implications for enhanced malaria transmission. Parasitology 122: 491-496.
Thomas F, Adamo S and Moore J (2005) Parasitic manipulation: where are we and where should we go? Behav Processes
68: 185-199.
Titus RG (1998) Salivary gland lysate from the sand fly Lutzomyia longipalpis suppresses the immune response of mice
to sheep red blood cells in vivo and concanavalin A in vitro. Exp Parasitol 89: 133-136.
Warburg A, Touray M, Krettli AU and Miller LH (1992) Plasmodium gallinaceum: antibodies to circumsporozoite proetein
prevent sporozoites from invading the salivary glands of Aedes aegypti. Expt Parasitol 75: 303-307.

364  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

17. Practical application of olfactory cues for monitoring and control

of Aedes aegypti in Brazil: a case study

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Abstract

The increasing importance of managing the mosquito vectors of dengue, mainly Aedes aegypti, is
recognised world-wide as the disease becomes more prevalent in many countries. The complexity
of the urban environment where the vectors breed makes it especially difficult to organise
control efforts to reduce their numbers and their impact on disease transmission. Existing control
programmes depend on larval surveys for population assessment but these are often unreliable
and become available too late for the application of control measures to have a significant impact
on disease transmission. The need for better and more timely information about the mosquito
vectors of dengue has been recognised in Brazil and several new technologies for more reliable
information on vector distribution and population densities have emerged that are making
control efforts more effective. Olfactory cues from oviposition sites are now being used to lure
gravid Ae. aegypti females into sticky traps for population assessments and indices of dengue
risk. The development of the so-called MosquiTRAP which incorporates an oviposition attractant
is described and its efficiency relative to more traditional monitoring methods is discussed.
Volatile chemicals that attract host-seeking female Ae. aegypti have also been identified. These
have been formulated for controlled release (‘BG-Lure’) and incorporated into novel trap designs
(‘BG-Sentinel’) with much improved capture of Ae. aegypti compared to the standard traps that
have been used previously. Both trapping systems are now being used to monitor and survey
these important disease vectors. The MosquiTRAP is also an important cornerstone of an on-
line GIS-based surveillance system called MI Dengue (Dengue Intelligent Monitoring System)
which is being implemented in Brazil. Results to date have demonstrated that the accuracy of
its predictions relating to dengue risk is much greater than that obtained using more traditional
methods. Real-time decision-making related to vector management is now possible using the new
surveillance systems, and this should greatly impact the management of the disease.

Keywords: Aedes aegypti, oviposition attractant, host odours, MosquiTRAP, BG-Sentinel, MI-
Dengue, disease risk assessment, Brazil

Introduction

Dengue is ranked as the most important mosquito-borne viral disease in the world, where in
the last 50 years incidence has increased 30-fold. The World Health Organization (WHO) has
estimated that 2.5 billion people live in over 100 countries of Africa, South-East Asia, the eastern
Mediterranean, the western Pacific, and the Americas where dengue viruses can be transmitted.
Up to 50 million infections occur annually, with 500,000 cases of dengue haemorrhagic fever and
22,000 deaths, mainly among children (WHO 2009).

The virus responsible for dengue fever (DF) and dengue haemorrhagic fever (DHF) belongs to the
family Flaviviridae (the same family as the yellow fever virus), and four different viral serotypes
exist: DEN-1, DEN-2, DEN-3, and DEN-4 (Gubler 2004). Infection with one dengue serotype provides
immunity for years, but it does not protect against infection with the other serotypes. Dengue
haemorrhagic fever is more common in children and it can lead to shock from blood loss, and even

Olfaction in vector-host interactions 365

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

death (Gubler 1998). Dengue virus is transmitted to humans during blood feeding by infected
mosquitoes, which have usually acquired the virus by ingesting blood of infected and viremic
humans and possibly also by transovarial transmission. The mosquito Aedes aegypti (L.) (Diptera:
Culicidae) is considered the main vector, although other Aedes species, including Aedes albopictus
(Skuse), have been implicated in rural epidemics as well as some urban ones (Effler at al. 2005). The
mosquito Ae. aegypti lives in close association with humans in urban and suburban environments.
The mosquito feeds predominantly on human blood and breeds in artificial containers such as
drums, buckets, tyres, flowerpots, and vases (Forattini et al. 1995). Therefore, the epidemiology of
dengue is highly related to the biology of the mosquito vector and human behaviour, as well as
to the environment and the virus itself.

As there is no effective vaccine for dengue, vector control is the main approach for control and
prevention. Although insecticide spraying has been used extensively, larval source reduction
(eliminating or cleaning water-filled containers that can harbour Ae. aegypti larvae) is considered
the most effective way of reducing and controlling the mosquito populations (Gubler 1998).

Scientific interest in research on dengue has grown in many countries, and opportunities for
investigators to obtain funds for that research have considerably improved. Funding agencies, such
as governments and international organisations, have increased the amount of money available
for research in response to dengue’s growing global incidence and impact on public health. In
Brazil, special programme funds have been allocated by the Federal and State governments
directed towards the achievement of new knowledge, control tools and new technologies for
dengue and other neglected diseases (CNPq 2009).

In subtropical countries such as Brazil, mosquito population density, dengue transmission, and the
number of DF cases start to increase at the beginning of the rainy season (October), with dengue
cases peaking more than three months later (Coutinho et al. 2006, Forattini et al. 1995). In Brazil,
dengue fever re-emerged as a major urban epidemic in 1986 (Marques et al. 1993) and currently is
considered the most important arthropod-borne viral disease (MS 2007a,b). In 2000, all 26 states
had reported DF cases and during 1986-2007 a total of 4,559,818 cases were officially reported
(MS 2007a) with 493 deaths (MS 2007b).

Transmission of dengue depends on a variety of variables in the relation between virus, mosquito
vector and human host (Focks et al. 1995, Kuno 1997). In the assessment of dengue risk, important
factors are the level of herd immunity in the human population and the ratio of mosquito to human
density, which influences the probability of vector-host contacts. To measure mosquito density,
Ae. aegypti control programmes usually still mainly rely on the surveillance of immature mosquito
stages (eggs, larvae and pupae) to determine the dengue risk. These surveillance methods rely
on tedious and repetitive house-to-house surveys. Data from these surveys are used to calculate
indices like the Premise Index (PI) or the Breteau Index (BI). The Premise Index was described in
the early 1920s for yellow fever outbreaks and is based on the percentage of premises infected
with Aedes larvae or pupae (Connor and Monroe 1923), whereas the BI indicates the number of
containers with larvae and/or pupae of Ae. aegypti (Breteau 1954). Both indices are used by WHO
as indicators of dengue transmission risk (Focks 2003).

In general, these traditional indices are poorly related to the risk of dengue transmission (Coelho
et al. 2008, Focks 2003, Méndez et al. 2006, Reiter et al. 1991, Scott and Morrison 2002). More
advanced assessments of immature stages therefore focus only on pupae, as their abundance
is more closely linked to adult population size (Focks 2003). A direct surveillance of the adult

366  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

mosquito population was previously possible only with much time and effort (using aspirators
or human landing collections) or were inefficient due to low catching rates of nonspecific adult
traps (Schoeler et al. 2004). However, studies in Brazil, combining newly-developed specific traps
for adult Ae. aegypti with geographic information systems, are now showing that an efficient and
real-time surveillance of adult vectors is possible, resulting in a quick determination of dengue
risk and effective vector control.

Overview of standard collections methods for dengue vectors

The collection of host-seeking dengue mosquitoes directly from human volunteers (human-
landing collection or catch) can provide a real-time surveillance of the mosquito population that
is actually involved in producing a dengue risk. During a dengue outbreak, however, when it is
important to assess the efficacy of the control measures, human-landing collections expose the
field workers to an intolerable risk of infection. In addition, the human-landing collection is time
consuming, labour intensive and yields variable results due to differences in human attractiveness
and in the skill of the field workers. Therefore other methods to collect adult mosquitoes are
usually preferred for the collection of dengue vectors. These methods focus either on the detection
of dengue mosquito eggs in oviposition traps (ovitraps), on finding their larvae or pupae, or on
collecting adult mosquitoes (Service 1993, Silver 2008).

Resting adults can be collected using aspirators, which is labour intensive and costly, but collects
large numbers of mosquitoes. Conventional adult traps use only visual cues to attract mosquitoes
(Fay-Prince trap, Wilton trap). Catch rates are usually too low when set in relation to their price
and the fact that these traps need electricity adds to their inconvenience. The addition of carbon
dioxide raises the attractiveness of visual traps, but its use on a larger scale will also be too costly
and complicated in most urban dengue-risk areas.

Recent research has led to the development of new traps and attractants that are highly specific
for dengue mosquitoes. Sticky traps with oviposition attractants capture mosquitoes that have
already ingested a quantity of blood sufficient to produce eggs (e.g. MosquiTRAP™). A new
ventilator trap (BG-Sentinel™) used with a combination of odorant substances found on human
skin focuses on host-seeking dengue mosquitoes. The BG-Sentinel trap has been shown to have
much better catching rates than conventional traps (Kröckel et al. 2006, Meeraus et al. 2008). Both
trap types allow for more specific and refined vector control campaigns in situations where the
traditional tools are often marginally effective.

Dengue mosquito monitoring through the detection of eggs: ovitraps

Eggs are collected with an oviposition trap (ovitrap), which is commonly used to detect the
presence of Aedes mosquitoes in low density areas. The ovitrap was first described by Fay and
Eliason (1966) and consists of a black container filled partially with water and a wooden paddle
placed vertically as an oviposition substrate. This trap requires laboratory infra-structure and
human resources for counting the eggs deposited on the oviposition substrate and identifying
the larval species of Aedes.

Historically, ovitraps have provided useful information in the early detection of new infestations,
spatial (presence or absence) and temporal (seasonal) distribution and in monitoring the impact
of control measures on Ae. aegypti, including those using pesticides (Focks 2003, Reiter and
Nathan 2001, Reiter et al. 1991). However, many comparative studies between the larval survey

Olfaction in vector-host interactions 367

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

and ovitrap use have shown that the ovitrap, besides being more sensitive in the detection of
Ae. aegypti, also has a lower cost but demonstrates limited operational viability in entomological
surveillance (Braga et al. 2000, Rawlins et al. 1998). When ovitraps are used in urban areas where
several species of the Stegomyia subgenus occur sympatrically, the eggs are morphologically
indistinguishable and, thus there is a need for identification of larvae.

Further studies are necessary in order to use ovitraps as a methodological approach to indicate risk
of dengue fever and yellow fever, because of the difficulties involved in calculating the population
density of adult vectors (Focks 2003).

Dengue mosquito monitoring through the detection of larvae and/or pupae

Active search for immature stages

The search of premises for containers that harbour larvae or pupae is still a widespread surveillance
method for dengue vectors. The results are usually used to calculate surveillance indices such
as the House Index (percentage of houses with larvae and/or pupae, also: Premise Index), the
Container Index (percentage of water-containers with larvae and/or pupae) or the Breteau Index
(number of positive containers per 100 houses) (Silver 2008). However, these indices do not
provide for a significant assessment of the number of adults that are actually produced and the
associated dengue transmission risk (Coelho et al. 2008, Focks 2003, Méndez et al. 2006, Reiter et
al. 1991, Scott and Morrison 2002).

Because the time from pupa to adult is short and pupal mortality is low, the number of pupae
found at a given premise provides the best estimation of adult density (Focks 2003). Linking the
number of pupae to human density can thus give an improved estimate of an important factor in
dengue risk assessment, the relation of mosquito density to human density. Focks (2003) therefore
promotes a ‘pupal/ demographic survey method’. This method combines interviews to determine
the number of people living or sleeping at the examined premises with the examination of the
water-holding containers for pupae. Pupae are collected and, in the case of the presence of various
container-breeding species, kept in the laboratory until adult emergence for identification. Using
standardised classification methods to identify and focus on the most productive containers
should both make surveillance and the control through source reduction more efficient (Barrera
et al. 2006, Chadee et al. 2007, Focks and Alexander 2006, Koenraadt et al. 2007). However, pupal
habitats may vary greatly in different settings, making a comparison and standardisation difficult.

Larval traps

Larval traps of various designs made of automobile tyres (tyre-section traps) have been used
for monitoring Aedes oviposition activity (WHO 1999). A standard larval trap (‘larvitrap’) consists
of a water-filled section of a tyre that facilitates visual inspection of the water in situ and allows
for the ready transfer of larvae to another container for examination. However, such a trap has
many disadvantages such as low sensitivity and the eggs laid in the trap have to be in contact
with water for hatching. In Brazil, the tyre-section larval trap was evaluated in the field and the
results showed that ovitraps are much more sensitive than larval traps (Marques et al. 1993). In
spite of the difference in sensitivity, larval traps are still widely used in Ae. aegypti surveillance
programmes and studies in Brazil (Dibo et al. 2005) and other countries, such as Argentina (Micieli
and Campos 2003), Peru (Morrison et al. 2004), Nicaragua (Lugo et al. 2005), Thailand (Tsuda et al.

368  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

2006), Columbia (Cuéllar-Jiménez et al. 2007, Maestre-Serrano et al. 2008) or Cuba (Suárez Ramírez
and Colás Bonne 2008).

Floating larval trap

An alternative larval trap, the ‘floating larval trap’, was developed and evaluated for eliminating
biases commonly associated with human collection (larval survey) (Harrison et al. 1982). The
trap collects diving mosquito larvae and pupae when they float back to the water surface. Both
field and laboratory data revealed that the trap was highly efficient and sensitive for detecting
immature stages of Ae. aegypti and Culex quinquefasciatus Say. The trap was claimed to function
as a random sampling device, providing an equal chance for the capture of immatures. However,
fewer 1st and 2nd instars larvae are collected because they often remain closer to the water surface
and do not dive as deeply when disturbed. Specimens captured in the trap remained alive for at
least 48 h. Although the size of this instrument (>13 cm diam.) could probably be reduced, its use
excludes important Ae. aegypti larval habitats such as vases, tyres, small plastic containers and
most footbaths.

Detection of adult dengue mosquitoes

Human landing collection

Although laborious, collecting mosquitoes directly from a human host obviously provides the best
indication of the number of mosquitoes that are currently involved in the transmission cycle and
can therefore give key information in the assessment of the dengue risk. The method has been
used to determine daily biting periodicity in different areas of the world (Chadee and Martinez
2000). However, catch rates depend on the time of day and the skill, motivation and attractiveness
of the individual field worker. Results are therefore variable and may lack the accuracy needed for
a reproducible and meaningful comparison of data, especially in the study of seasonal variations.
Because the technique exposes field workers to infected mosquitoes, it is currently undergoing
ethical restrictions (Focks 2003, PNCD 2002). The human landing collection is therefore not used
as a standard surveillance method for dengue mosquitoes, but remains an important reference
method in the assessment and comparison of other collection methods (Jones et al. 2003, Kröckel
et al. 2006, Schoeler et al. 2004).

Sampling resting adults with aspirators

Aspirators are used to capture resting adult mosquitoes, especially indoors (Nasci 1981) in dark
places and this method provides an estimate of adult density in the sampled houses (Morrison et
al. 2004; Schoeler et al. 2004). Aspirators are either carried on the back of the technician (e.g. CDC
backpack aspirator, John W. Hock Co., Gainesville, FL, USA) or hand held as in the Nasci aspirator. As
the field worker needs to enter private homes, this method is invasive and requires high tolerance
of the inhabitants. It is also labour intensive and results may be affected by the efficiency of field
technicians (Clark et al. 1994, Favaro et al. 2006). Using backpack aspirators is therefore usually
considered more for research than for routine activities of dengue control programmes (Favaro
et al. 2008).

Olfaction in vector-host interactions 369

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Traps for gravid Aedes aegypti

Artificial oviposition sites can be used to attract and catch gravid mosquito females. Their
attractiveness for dengue mosquitoes can be augmented by the addition of decomposing organic
matter in fermenting infusions made from grass, hay, leaves or other plant material (Chadee 1993,
Du and Millar 1999, Reiter et al. 1991, Sant’Ana et al. 2006, Takken and Mboera 2000). The CDC
gravid trap (John W. Hock Co., Gainesville, FL, USA) was originally designed by Reiter (1983). A
battery-powered fan draws mosquitoes attracted by grass or hay infusions into a catch bag. In
contrast, sticky oviposition traps use glue boards to capture landing mosquitoes and do not need
electricity. Different models have been developed to capture gravid Ae. aegypti (Fávaro et al.
2006, Gama et al. 2007, Kay et al. 2000, Muir and Kay 1998, Ordóñez-Gonzalez et al. 2001, Ritchie
et al. 2003, 2004) and Ae. albopictus (Facchinelli et al. 2007). Like the CDC gravid trap, they use
decomposing organic material in water to attract mosquito females searching for an oviposition
site. Although these infusions are relatively easy to produce, they need approximately one to three
weeks time to ferment. Their composition also changes with age and time in the traps, potentially
altering their attractiveness and making a comparison of trapping results more difficult.

Traps using visual cues

The use of light traps constitutes a method for capturing a large variety of adult insects and
is often used to collect nocturnal mosquito species. However, light traps are not effective at
capturing the diurnal Ae. aegypti (Silver 2008). The Fay-Prince trap (John W. Hock Co., Gainesville,
FL, USA) is a daytime trap with black and white contrasts and uses a fan to draw approaching
mosquitoes into a catch bag (Fay and Prince 1970). Another trap designed to capture day-active
mosquitoes such as Ae. aegypti is the Wilton trap (Wilton 1985) (John W. Hock Co., Gainesville, FL,
USA). The Wilton trap consists of a glossy black cylinder with a fan and collection cup inside. Both
traps catch mosquitoes of both sexes and all physiological stages. The catch rates for visual traps
are, however, only a fraction of the human landing collection (Jones et al. 2003, Kröckel et al. 2006,
Schoeler et al. 2004).

Traps using carbon dioxide

Carbon dioxide is a general attractant to a large majority of mosquito species. Liberated either
from dry ice or pressurised gas bottles, CO2 enhances the catch rates of the Fay-Prince and the
Wilton traps, but results for Ae. aegypti are poor compared to human landing collections (Canyon
and Hii 1997, Schoeler et al. 2004). Some trap types that are commercially available for the end
consumer1 produce CO2 from the combustion of propane gas. The reaction is usually supported
by a catalyst and yields warm, humid CO2 of a concentration attractive to mosquitoes (Kline 2002).
The attractiveness of these traps for dengue mosquitoes has not been investigated. As these traps
are relatively expensive, they are not practical for use in routine monitoring programmes. Other
traps with carbon dioxide that are routinely used to monitor mosquitoes, such as the EVS-trap or
the CDC-light-traps with CO2, also catch Ae. aegypti and Ae. albopictus (Russel 2004), but the costs,
efforts and logistics that come along with the use of CO2 from any of these sources currently do
not make theirr application feasible in most field situations.

1 E.g.
Woodstream Mosquito Magnet, Blue Rhino Skeeter Vac, Flowtron Mosquito Power Trap, Lentek Mosquito Trap
Guardian (list does not claim to be exhaustive).

370  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

The need for simple, efficient, and specific collection tools

A suitable tool for collecting adult dengue vectors in routine monitoring operations should be
sensitive enough to detect adults also when they occur in low densities. The tool should be robust
and workable for routine uses under various conditions and the caught mosquitoes should be
easily removable from the collection device to be identified and/or processed for further virus
detection. The collection rate should correlate very closely with human landing collections in
order to establish meaningful risk indicators.

MosquiTRAP for collection of gravid Aedes aegypti

Olfactory cues

The deposition of eggs by mosquitoes is mediated by olfactory cues, which originate from the
breeding site as a result of microbial activity (Bentley and Day 1989; Clements 1999). Apparently,
Metha (1934) was the first to report the existence of olfactory cues in recognition and selection
of oviposition sites by mosquitoes. Water containing decomposing organic matter was found
attractive for gravid mosquitoes and has been widely used for mosquito surveillance and possibly,
mosquito control (Chadee et al. 1993, Eiras 2001, Millar et al. 1992, Reiter et al. 1991, Takken and
Mboera 2000). Hay and grass infusions have been reported since the 1980’s to be attractive
for ovipositing females of a wide range of mosquito species, including Ae. albopictus (Holck et
al. 1988), Ae. aegypti (Chadee et al. 1993), Ae. triseriatus (Say) (Holck et al. 1988), Cx. nigripalpus
Theobald (Ritchie 1984), Cx. pipiens Linnaeus (Reiter 1986), Cx. quinquefasciatus (Mboera et al.
2000). The infusions are usually made of many different organic materials, such as leaves, grass
and sod (Allan and Kline 1995, Chadee et al. 1993, Rawlins et al. 1998, Reiter et al. 1991).

There are differences in the selection of breeding sites between Ae. albopictus and Ae. aegypti. The
former species breeds in natural and manmade containers and is often most abundant in rural and
sylvan areas, but is able to rapidly colonise suburban an urban areas with sylvan characteristics
(Hornby et al. 1994). The second species is a container-inhabiting mosquito that occurs in urban
areas and oviposits in many breeding sites such as tyres, plant vases, and other water-holding
containers associated with human activities. Many reports have shown that gravid Ae. aegypti and
Ae. albopictus lay more eggs in ovitraps when baited with grass infusions compared to clean water
(Chadee et al. 1993, Reiter et al. 1991, Sant’Ana et al. 2006).

Many organic materials have been used for increase the efficacy of ovitraps for sampling Aedes
eggs in the field, such as oak leaves (Quercus spp.) (Allan and Kline 1995), hay infusions of Axonopus
compressus (Sw.) Beauv. (Rawlins et al. 1998), Bermuda hay infusions (Reiter et al. 1991), infusions
of dried Sclerica bractaeda (L.) (Chadee et al. 1993), palletised plant-based animal feeds (Ritchie et
al. 2003) and leaves of Panicum maximum Jacq. (Sant’Ana et al. 2006). Sant’Ana et al. (2006) have
shown that the oviposition responses of Aedes (Stegomyia) mosquitoes changed with respect to
grass species, stage of plant development and type of fermentation, thus suggesting that grass
leaves have different chemical profiles and bacteria and other microorganisms present in the
infusions and that they may produce several chemical lures for gravid mosquitoes. Neglecting
to take into account the age of infusions for monitoring seasonal abundance of mosquitoes by
ovitraps baited with grass infusions may constitute a significant source of variability in the results
of these studies. The period of fermentation of grass infusions has a significant effect on the
response of gravid mosquitoes and infusions of known and standardised fermentation periods
should be used only. For instance, studying various ages of fermentations (e.g. 1, 3, 5, 7, 15, 20 and

Olfaction in vector-host interactions 371

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

30 days), Sant’Ana et al. (2006) demonstrated that the best fermentation ages for grass infusions of
P. maximum were 15 and 20 days. Most likely, the period of fermentation may change the chemical
profile of volatiles produced by grass infusions and, consequently, the oviposition response of Ae.
aegypti and Ae. albopictus may be affected.

Identification of oviposition attractants for Aedes aegypti from Panicum maximum grass infusion

The use of specific synthetic oviposition attractant chemicals in monitoring programmes for Aedes
(Stegomyia) mosquitoes would eliminate the need to create infusions and would allow for the
preparation of precise formulations of an oviposition medium. Therefore, synthetic oviposition
attractants can eliminate variation in infusion quality and attractiveness.

One important aspect of the use of oviposition attractants in trapping devices is to understand the
role of the individual active chemicals. Several studies used dual-choice bioassays at close-range,
where the oviposition behaviour comes at the end of a behavioural sequence and thus, they did
not discriminate between long-range and short-range responses. As it is likely that oviposition
stimulants act mainly at close-range, such studies may underestimate the role of these stimulants
at long range.

There are several publications on the evaluation of synthetic oviposition attractants for mosquitoes
in the field (Allan and Kline 1995, Beehler et al. 1994, Mboera et al. 2000). However, only few
chemical identifications of oviposition attractants have been conducted for Culex (Du and Millar
1999, Hwang et al. 1980, Leal et al. 2008) and Aedes mosquitoes (Bentley et al. 1979, Ponnusamy
et al. 2008, Sant’Ana 2003). Some compounds such as nonanal, p-cresol, indole and 3-methyl-
indole have been identified as oviposition attractants for Cx. quinquefasciatus and Cx. tarsalis
mosquitoes (Du and Millar 1999, Leal et al. 2008) and Aedes mosquitoes (Bentley et al. 1979). Some
specific bacteria associated with carboxylic acids and methyl esters have been described in binary
choice assays as potent oviposition stimulants for gravid Ae. aegypti, but the elucidation of these
compounds has not yet been completed (Ponnusamy et al. 2008)

Oviposition attractants for gravid Ae. aegypti were isolated and identified from P. maximum grass
infusions (Sant’Ana et al. 2006) and these are considered an important tool for increasing the catch
of gravid Ae. aegypti in the sticky trap MosquiTRAP (Eiras and Resende 2009, Eiras and Sant’Ana
2002, Favaro et al. 2006, 2008, Lourenço-de-Oliveira et al. 2008). The chemical identification of
volatiles produced by grass infusion of P. maximum that enhanced oviposition responses of gravid
female Ae. aegypti was possible only by means of a combination of studies using electrophysiology
(electroantennography), gas-chromatography coupled with mass-spectrometry (GC-MS), gas-
chromatography coupled with electroantennography (GC-EAD) and behaviour using ovitraps
(Eiras et al. 2001, Eiras and Sant’Ana 2002; Sant’Ana 2003; Eiras et al. 2004). Firstly, the grass infusions
of different ages (Sant’Ana et al. 2006) were extracted individually by solvent and by solid phase
micro extraction (SPME) and submitted individually to GC-MS, GC-EAD and EAG and compared
with results from oviposition responses. It was interesting that the responses of gravid female Ae.
aegypti antennae to EAG extract of grass infusions of different ages followed a similar pattern to
the oviposition responses to grass infusion extracts in the field (Figure 1), showing that either the
quantity of active compounds increased with ageing of the grass infusion or that new compounds
were produced as the grass infusions fermented. Later, the active compounds present in infusion
extracts of high EAG and oviposition responses were submitted to GC-MS for identification. Seven
compounds were identified as candidates for an oviposition stimulant or attractant of gravid Ae.
aegypti: nonanal, benzothiazol, decanal, p-cresol, indole, 3-methyl indole and limonene) (Eiras and

372  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

1.5 50
EAG Egg collection
40

Means egg collected

1.0
EAG (mV)

30

20
0.5

10

0.0 0
Control 1 3 5 7 10 15 20 30
Age of grass infusions
Figure 1. Behavioural (ovitrap) and electrophysiological (EAG) responses of gravid Aedes aegypti to Panicum
maximum grass infusions of different ages (Sant’Ana 2004).

Sant’Ana 2002 - patent pending). The EAG studies of the seven individually identified compounds
showed that the aldehydes (nonanal) elicited the highest antennal responses of gravid female Ae.
aegypti. The GC analysis showed that the grass infusion produced a large number of volatiles. The
compound 3-methylindole has a very unpleasant and strong smell even in small concentrations,
and has been shown to modify the response of gravid female Ae. aegypti (Allan and Kline 1995,
Sant’Ana 2003). However, 3-methylindole was revealed to have a repellent effect on Ae. albopictus
in laboratory assays and in the field using ovitraps (Trexler et al. 2003). By contrast, 3-methylindole
is an oviposition stimulant for Cx. quinquefasciatus at very low concentrations (Mboera et al. 2000).

Based on the fact that many substances are produced by grass infusions and that only a few key
chemical compounds are responsible for attracting gravid mosquitoes to breeding sites, many of
the volatiles present in infusions of organic material are likely to be produced by other sources,
such as plants or decaying materials. The responses of gravid Ae. aegypti mosquitoes to a single
compound could result in attraction to an inadequate place for oviposition, but the complexity
of chemical signals may be crucial as an indicator of a potential breeding site.

Formulation of lures for Aedes aegypti from synthetic oviposition attractants

The majority of semiochemicals that have been described as oviposition attractants for Aedes
species would volatilise very quickly if simply added to the water surface of an oviposition trap.
Some form of controlled-release technology is required to release the active substances at a rate
that is optimal for continued attraction of the target species to the oviposition trap. In the case
of Aedes species, two types of controlled-release devices have been used. One was a solid matrix
polymer, within which the active ingredient nonanal was evenly distributed while the other was
released from a reservoir system that took the form of sealed tube. The reservoir system tend
to release more chemicals at the beginning of their field life in contrast with that released from
solid matrix polymer, which was more towards the end of their field life (Figure 2). There is a high

Olfaction in vector-host interactions 373

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

50

40
Release rate (mg)

30

20

Solid matrix polymer

10 Sealed tube

0
0 10 20 30 40 50 60

Days
Figure 2. Release rate of nonanal (AtrAedes) at solid matrix tube and sealed tube dispensers at 25.1±1.87 °C (Data
not published).

evaporation rate during the first 10 days before both devices get a steady release rate. This is often
referred to as a ‘first order’ release profile, while the preferred profile would be one where the
amount of active substance released remains constant during the whole field life of the dispenser
and referred to as a ‘zero order’ release profile. Dispensers with a reservoir system tend to show a
zero order release profile and therefore are preferred as dispensers, as the attraction rate remains
more consistent during the whole length of the life of the dispenser in the field.

Laboratory bioassays

Behavioural studies of gravid female Ae. aegypti exposed to synthetic lures were carried out
in a dual-port olfactometer (Geier and Boeckh 1999) in response to log-doses of nonanal. It is
interesting that the response of host-seeking Ae. aegypti to a human hand in the olfactometer
bioassay is usually about 85-100%, whereas the response of gravid females to grass infusions
is about 40-50% (Mota 2003). The response of gravid Ae. aegypti to oviposition attractants was
consistently within the range of 31 to 45% responding whereas the response of host-seeking
mosquitoes to human odour was about 80% (Table 1). The synthetic oviposition attractant
(nonanal) and the AtrAedes lure elicited similar responses in gravid Ae. aegypti.

Semi field test

Semi-field systems have been used for a variety of studies on mosquitoes, especially on the
behavioural ecology of malaria vectors (Knols et al. 2002, Schmied et al. 2008) and for evaluating
trapping devices for Ae. aegypti (Kline 2002). Although there are some limitations associated
with these studies, for instance that long-range dispersal cannot be studied. Semi-field cages are
considered a controlled environment that offers intermediate conditions between the laboratory
and the field for behavioural studies, where both laboratory and wild-caught insects can be tested.

374  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

Table 1. Behavioural responses of host seeking and gravid female Aedes aegypti to different stimuli in a dual port
olfactometer (Data not published).

Stimulus N Control Test

Host-seeking female Human hand 30 81.9%±10.8 1.1%±2.6*

Gravid female Grass infusion 33 35.0%±4.0 21.1%±3.8ns
Nonanal (10 µg) 8 31.8%±13.1 22.6%±6.4ns
AtrAedes™ Lure 8 43.4%±13.6 25.2%±6.0ns

* Means P<0.05 and ns = means P>0.05.

In Brazil, a semi-field experimental area with eight test cages was developed for studying
behavioural responses of gravid Ae. aegypti mosquito to evaluate prototype traps, controlled
release devices and the development of new formulations of synthetic oviposition attractants
(Roque and Eiras 2008). As laboratory-reared mosquitoes are used in these field cages, the age
and physiological status of test mosquitoes, which are free of viruses and bacterial contamination,
are well established. Field cage results showed that gravid Ae. aegypti mosquitoes had similar
responses in a field cage as in field tests in urban areas, confirming that the semi-field set up is
reliable, safe and gives reproducible results.

Behavioural studies in semi-field cages (Roque and Eiras 2008) showed that the MosquiTRAP baited
with a formulation of synthetic oviposition cues caught similar proportions of gravid Ae. aegypti to
traps baited with infusions of 10-45 days of fermentation (Figure 3). However, the oviposition cues
caught a significantly higher proportion of mosquitoes than infusions aged 5 or 60 days (P<0.01)
(Roque 2007). Therefore, the synthetic oviposition cue can replace the natural grass infusions.
The behavioural response of gravid Ae. aegypti is likely to be governed by a complex mixture of

75
Infusions AtrAedes *
*
60
Response (mean±s.e.)

45

30

15

0
5 10 15 20 30 45 60

Age of grass infusions (days)

Figure 3. Behavioural responses of gravid females Aedes aegypti to the sticky trap (MosquiTRAP) within field
cages to grass infusion (Panicum maximum) and synthetic oviposition attractants (AtrAedes) (Roque 2007).
* Means P<0.05.

Olfaction in vector-host interactions 375

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

several compounds having specific release rates. A single compound at a very precise release rate
can replace the grass infusion. Again, oviposition behaviour is probably mediated by a complex
combination of many compounds rather than by one single compound (Eiras 2001). Moreover, the
concentration of the compound in the odour plume may play an import role as chemical signal
to gravid mosquitoes.

Field evaluation of oviposition lures for Aedes aegypti

Field studies showed that the AtrAedes lures are still attractive after 30 days of field use (Eiras
et al. 2004) and their performance as bait for the Brazilian sticky trap (MosquiTRAP) has been
evaluated with adult collection methods, using an aspirator. In Brazil, the sensitivity of the sticky
trap was compared with manual aspirations (Nasci 1981) to detect the presence of A. aegypti
females during the rainy season for 23 weeks and it was concluded that the sensitivities of both
methods were similar (Favaro et al. 2008). The authors pointed out that the relative efficiency of
use of the MosquiTRAP was 30 traps per person per day whereas that of manual aspirators was
five collections per person per day. Collection with aspirators requires the collector to enter indoor
sites, thus causing potential problems with the residents. In a dengue endemic area in Rio de
Janeiro city during the rainy season, the sticky trap baited with AtrAedes lures was also compared
with CDC backpack aspirators in a mark-release-recapture experiment. From a total of 1,240 gravid
dust-marked Ae. aegypti released in the field, 127 (10.2%) were recaptured by baited MosquiTRAPs
whereas only 47 (3.8%) were recaptured with the backpack aspirator at 10, respectively 8, days
after release(Maciel de Freitas et al. 2008).

Traps for collection of gravid adult Aedes aegypti

Only a few traps are specifically designed for catching previously blood fed, i.e. gravid female Ae.
aegypti. The CDC gravid trap (John W. Hock Co., Gainesville, FL, USA) was originally designed by
Reiter (Reiter 1993) to attract and catch gravid mosquitoes. It has a battery-powered fan mounted
inside that draws mosquitoes attracted by grass or hay infusions into a catch bag. Later, the CDC-
gravid trap was modified by Dennet et al. (2007) and visual aspects were incorporated to increase
the catch of gravid Ae. albopictus. The modified gravid trap did not turn out to be effective at
catching gravid Ae. albopictus, probably because the standard oviposition attractant used had
been developed for Culex spp. and may not have been attractive to Ae. albopictus.

Several sticky trap models have been developed to capture gravid Ae. aegypti (Favaro et al. 2006,
Gama et al. 2007, Kay et al. 2000, Muir and Kay 1998; Ordonez et al. 1997, Ordonez-Gonzalez et al.
2001, Ritchie et al. 2003, 2004) and Ae. albopictus (Facchinelli et al. 2007). The efficacy of a sticky
ovitrap lined with an adhesive paper strip was tested in a dengue area in Guadelupe, Mexico
using mark-release-recapture (MRR) methods (Ordonez-Gonzalez et al. 2001). A similar type of
sticky ovitrap with natural oviposition attractants was developed and evaluated in an Ae. aegypti
infested area in Cairns, Australia, where an association between the number of adult females
captured and the risk of dengue transmission was found (Ritchie et al. 2004). The Australian
sticky ovitraps have also been used for surveillance and behavioural investigations of Ae. aegypti
and Ae. polynesiensis (Marks) in Moorea, French Polynesia (Russell and Ritchie 2004) and in the
Torres Straits, Australia (Ritchie et al. 2006). In Italy, Facchichelli et al. (2007) also developed a new
sticky trap design, which was shown to be specific for Ae. albopictus in field trials in Rome. The
MosquiTRAP has been used for studies of mark-release-recapture in a suburban residential area
of Rio de Janeiro (Brazil) (Maciel de Freitas and Lourenço de Oliveira 2009, Maciel de Freitas et al.
2008) and in Cairns, Australia (Russell et al. 2005).

376  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

In Brazil the MosquitoTRAP is used for monitoring populations of Ae. aegypti in urban areas
(Eiras and Resende 2009). Such sticky traps are a low-cost device without the need for electricity,
laboratory facilities or human resources, providing faster results than the ovitrap, larval survey and
other adult traps for dengue control. Mosquito behaviour in sticky traps appears to be similar in
different areas (Favaro et al. 2006, Gama et al. 2007, Ritchie et al. 2004); once stuck, the mosquito
remains in resting position. Those that escape, usually loose one or more legs that remain adhered
to the sticky card. Identification of collected mosquitoes is still possible with their thoraces, which
usually remain somewhat visible (Gama et al. 2007, Ritchie et al. 2003).

The MosquiTRAP Version 1.0 (patent pending) was developed in 2001 at the Federal University of
Minas Gerais and consisted of a 1 litre black plastic cylinder filled with 300 ml of grass infusion (as a
natural oviposition attractant) and a removable sticky card on which the mosquitoes are captured
(Eiras 2002). During the dry season of 2003, field studies showed that the MosquiTRAP Version
1.0 was more sensitive to detect the presence of Ae. aegypti in urban areas than the traditional
larval survey used in Brazil (Gama et al. 2007), especially during the dry season when the larval
method cannot detect this mosquito. Therefore, the Brazilian National Programme of Dengue
Control (NPDC) stimulated and provided funds for further studies. In 2004-2005, the Brazilian
NPDC sponsored field evaluations of the MosquiTRAP version 2.0 and the MI-Dengue technology
in 10 Brazilian cities, aiming to supplement new tools and develop new entomological indices for
dengue vector control (Eiras et al. 2005, 2007). For such studies, it was necessary to develop a new
sticky trap prototype that could be produced on a large scale. The MosquiTRAP Version 2.0 was
developed at the University of Minas Gerais and prototyped by professional designers, looking
for easy and rapid functionality for field workers. This trap version was funded by several Brazilian
public grants that encouraged the creation of a ‘spin-off’ company, namely Ecovec Ltda., in order
to commercialise it.

The MosquiTRAP Version 2.0 was evaluated and compared with larval surveys and ovitraps in 10
Brazilian cities. The results demonstrated that field workers were able to identify 95 to 100% of Ae.
aegypti captured in MosquiTRAPs and that the larval survey was less sensitive than the sticky trap,
whereas the ovitrap was more sensitive than both methods (Eiras et al. 2005).

The position in which the sticky trap is placed in houses is very important in order to avoid rejection
by the residents. Studies in Brazil demonstrated that the MosquiTRAPs placed outdoors captured
five times more females than traps placed indoors (Favaro et al. 2006). This is probably because Ae.
aegypti feeds on human blood indoors and lays eggs outdoors, where more breeding places are
available. Outdoor traps have also a great advantage of being a non-invasive method of mosquito
sampling because trap inspection does not require house entry for mosquito inspections, which
is obviously inconvenient for homeowners as well as for vector control workers. Beside the sticky
trap being easy to manipulate, it was readily accepted by the residents and local community.
A study in Brazil has shown that the MosquiTRAP had a high level of acceptance by the local
community (95.6 to 99.3%), whereas the larval survey had a low acceptance (60-90%). The time
spent for sticky trap inspection is about 3-5 min whereas the larval survey is about 12-25 min (Eiras
et al. 2005). The selectivity of the MosquiTRAP in catching gravid Ae. aegypti was shown by Favaro
et al. (2006): of 488 dissected females, 426 were gravid (87.3%).

Several studies have been conducted to compare the sensitivity of the sticky trap with that of
the ovitrap (Facchinelli et al. 2007, Fávaro et al. 2006, 2008, Gama et al. 2007, Honorio et al. 2009,
Lourenço-de-Oliveira et al. 2008, Ritchie et al. 2003, 2004) and the aspirator (Favaro et al. 2008,
Maciel de Freitas et al. 2008). Generally, gravid Ae. aegypti lay eggs in both natural and artificial

Olfaction in vector-host interactions 377

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

containers, indoors or outdoors and exhibit an oviposition pattern known as ‘skip-oviposition’,

which means that a single gravid female lays her eggs in several containers (Cobert and Chadee
1993). Consequently, one recipient, such as an ovitrap, may contain eggs from several Ae. aegypti
females (Reiter et al. 1991), which precludes the possibility of knowing the number of females that
laid eggs in the ovitrap. The sensitivity of the ovitrap (presence of at least one egg) and the sticky
trap (presence of at least one adult female Ae. aegypti) was measured in field trials. Gama et al.
(2007) reported that during the dry season, when the population of Ae. aegypti is low, the ovitrap is
more sensitive (16-67%) than the sticky trap (0-32%) whereas Favaro et al. (2006) showed a similar
sensitivity of the MosquiTRAP (82.1%) and the ovitrap (89.7%) during the rainy season. In Australia,
Ritchie et al. (2003) reported that the sensitivity of sticky ovitraps (67.5%) was slightly higher
than that of oviposition traps (64%). The sensitivity of a trap to detect the presence of disease
vectors and to measure population fluctuations with time is a key aspect in effective monitoring
of dengue risk and dengue control programmes.

A new MosquiTRAP design (Version 3.0) (Figure 4) consists of a polyethylene mat black container
(33 cm high by 15 cm wide), divided into two parts: a base, filled with approximately 300 ml of
tap water, and an upper part with a funnel-shaped opening, facilitating the mosquito’s entry and
hindering its exit (Eiras and Resende 2009). The sticky card is attached from the water line in the
base to the upper part of the trap and the synthetic oviposition attractant (AtrAedes) is attached to
the sticky card. This new MosquiTRAP model #3.0 was compared with the previous model #2.0 in
semi-field and urban areas of Brazil. In the semi-field, the new version caught a significantly higher
number of gravid Ae. aegypti (total of 244; mean 5.5±0.46) than the previous version (total of 91;
mean 2.1±0.29) (Eiras et al. 2005). In urban areas (Boa Vista, Roraima State, Brazil) 40 sticky traps of
each version were set in 40 blocks and the evaluation ran for 12 weeks. The previous MosquiTRAP
version 2.0 caught 47 gravid female Ae. aegypti whereas the new version 3.0 caught 127.

This new MosquiTRAP version has been evaluated for dispersion of Ae. aegypti (Maciel de Freitas
and Lourenço-de-Oliveira 2009, Maciel de Freitas et al. 2007, 2008) and temporal patterns of Ae.
aegypti population dynamics (Honorio et al. 2009) and in a novel field data acquisition of adult
female Ae. aegypti and processing-generating GIS real-time web-site surveillance system (Eiras
and Resende 2009).

Figure 4. The Brazilian sticky trap MosquiTRAP™: (A) Sticky trap assembled and (B) its inner parts: (1) upper
opening; (2) funnel shape entering; (3) sticky card set on rings and (4) lower pot with water (Eiras et al. 2007).

378  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

Traps for collection of host-seeking adult Aedes aegypti

Host odours attractive for Aedes aegypti

Composition of host odours attractive for Aedes aegypti

In the collection of mosquitoes in general, the most effective attractant is carbon dioxide (Chapter
5, this volume). For the highly anthropophilic Ae. aegypti, carbon dioxide is, however, not very
effective if used as the sole attractant for gravid females. As the gas is either liberated from dry ice,
from pressurised bottles or from the combustion of propane gas, its use is costly, complicated and
not feasible in typical dengue endemic urban environments. The search for Ae. aegypti attractants
other than carbon dioxide first focused on lactic acid. This substance, present on human skin as
well as in breath, was repeatedly shown to attract Ae. aegypti, but until the beginning of the 1990’s
only in combination with carbon dioxide (Acree et al. 1968, Eiras and Jepson 1991, 1994, Smith
et al. 1970).

Attraction to lactic acid without the addition of carbon dioxide was first shown in a Y-tube
olfactometer that allowed for an especially fine distinction of the effect of olfactory attractants
(Geier and Boeckh 1999). The apparatus also demonstrated the attraction of female Ae. aegypti to
a human skin extract. This extract had been produced by rubbing the skin of a human test subject
with ethanol-soaked cotton pads and a subsequent extraction of these pads. Although lactic acid
was also a component of the skin extract, the extract was more attractive than lactic acid alone
and was the same as that of a human finger introduced into the olfactometer.

Ammonia has also been found to be an attractive component. It had already been shown to be
attractive for other haematopageous insects (Hribar et al. 1992, Taneja and Guerin 1997), but
earlier authors could not demonstrate the attraction of Ae. aegypti to ammonia (Brown et al.
1951, Müller 1968, Rössler 1961;). It had never before been tested in combination with lactic acid.
Reinvestigation of this combination showed a synergistic effect of ammonia with lactic acid (Geier
et al. 1999a). However, the combination of lactic acid and ammonia was again not as attractive as
the complete skin extract, indicating that yet other components play an additional role.

Because fatty acids are widely present on human skin, their possible role as attractants for
mosquitoes had already been proposed by earlier authors (Carlson et al. 1973, Knols et al. 1997,
Rössler 1961). This was confirmed for Ae. aegypti by Bosch et al. (2000), whose experiments
suggested two groups of carboxylic acids with different chain lengths: C1 to C3 and C5 to C8 (C9
and C11 carboxylic acids reduced the attractiveness of lactic acid in the olfactometer).

Later investigations showed that hexanoic acid (caproic acid) was one substance with which the
attractiveness of the lactic acid and ammonia combination for Ae. aegypti could be augmented
further, but only in a closely defined proportion to each other. A dispenser liberating the different
substances in the right quantities for a prolonged period of time would have to account for the
different physical and chemical properties of the individual components. Due to their relatively
large volume and vapour pressure, reservoir devices have been used mainly for lactic acid
controlled release. Tube structures such as those described earlier for oviposition attractants
have been used with the factor determining the release rate being the material from which the
tube is made. For ammonia and hexanoic acid, solid polymer release matrices have been used.
In this way, each substance in the three-component lure has its own release mechanism so that
the ratios between the three components remain the same during the whole of the field life of

Olfaction in vector-host interactions 379

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

the lure. The same principle of separated release devices for each compound in one dispenser
unit is also used in an advanced version of a commercially-available dispenser, the Sweetscent™
dispenser, in which thin polymer coatings determine the release rate of each compound out of
its reservoir pouch. Figure 5 shows the BG-Lure™ and the Sweetscent™, which are made up of the
three controlled release formulations of the active substances, all contained in a mesh bag for ease
of manipulation and placement in the BG Sentinel trap.

Figure 5. (A) Bg-Lure™ and (B) the Sweetscent™ dispensers containing lactic acid, ammonia and hexanoic acid
(Source: Biogents AG).

Plume structure of host odours attractive for Aedes aegypti

Not only the composition, but also the structure of attractant odour plumes carries important
information for host-seeking Ae. aegypti. This was indicated for the first time by a closer analysis
of its flight behaviour in a Y-tube olfactometer (Geier et al. 1999b). The attractiveness of carbon
dioxide, of human skin odours rubbed on a glass support and of L-(+)-lactic acid were tested in
homogeneous, turbulent and filamentous odour plumes. A marked difference between carbon

380  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

dioxide on the one side and the skin odours on the other side was observed. A turbulent and
filamentous carbon dioxide plume with large fluctuations in the concentration lured more
mosquitoes upwind than a homogeneous CO2 plume. The opposite was found with plumes of
human skin odour: the highest number of mosquitoes flew upwind in the homogeneous plume,
whereas in filamentous plumes, their numbers were significantly lower. 3-D video analyses of
the flight behaviour of individual Ae. aegypti females in a larger wind tunnel confirmed these
findings (Dekker et al. 2005). Heterogeneous odour plumes, both from skin odour and from carbon
dioxide, elicited a meandering flight pattern with many turns and transversal parts. Homogeneous
skin odour resulted in a quick and direct upwind flight, while mosquitoes presented with a
homogeneous carbon dioxide plume flew mostly sidewards, much as in the control experiments
(see also Chapter 6, this volume).

The Biogents-Sentinel trap design

The patented design of the BG-Sentinel incorporates attractive host-odour plumes for Ae.
aegyti (Figure 6) (Geier et al. 2004 – patent pending). A faint upward air stream, charged with
artificial skin odours, is produced through a gauze making up the top of the trap. The individual
components are lactic acid, ammonia and caproic acid, all liberated in a fixed ratio from a specially
designed dispenser, the BG-Lure, which is placed inside the trap. The artificial host-odour plume
mimics the upward convection currents produced by a human body. Approaching mosquitoes are
furthermore attracted over short range by the black and white contrast between the top of the
trap and the centre tube, where the mosquitoes are drawn into a catch bag. One fan produces the
attractive plume and sucks the mosquitoes into a catch bag.

A B

Figure 6. (A) BG-Sentinel trap™ assembled and (B) its functional diagram: longitudinal section of the trap. CB:
catch bag; T: black tube; F: fan; arrows indicate the direction of the airflow.

Olfaction in vector-host interactions 381

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

A number of field studies have demonstrated that the BG-Sentinel trap is an excellent tool for
capturing Ae. aegypti (Feltner and Ferrao 2008, Krueger and Hagen 2007, Maciel de Freitas et al.
2006, 2007, Meeraus et al. 2007, 2008, Morrison et al. 2008, Williams et al. 2007, Solberg et al. 2007).
Kröckel et al. (2006) compared the BG-Sentinel to the Fay-Prince trap and a trap that produces
carbon dioxide through the combustion of propane gas (the Mosquito Magnet Liberty™), and
human landing collections in Belo Horizonte, Brazil. They reported significantly better catching
rates for female Ae. aegypti with the BG-Sentinel (between 3 and 15 times better catching rates
than the MML trap and between 8 and 15 times better rates than the Fay-Prince traps (Figure 7).
Compared to human landing collections, the BG-Sentinel trap (without the addition of carbon
dioxide) catches about 40% fewer female Ae. aegypti than a human collector. The trap also captures
male Ae. aegypti.

In comparison to a backpack aspirator (used in a standard protocol of 20 minutes per premise) in

Rio de Janeiro, Brazil, the BG-Sentinel (placed at each premise for 24 hours) captured significantly
more female and male Ae. aegypti (Maciel de Freitas et al. 2006). A similar field study in Cairns,
Australia, confirmed significantly higher catching rates of the BG-Sentinel than a back pack
aspirator and EVS-trap baited with carbon dioxide (Williams et al. 2006). Additional experiments
demonstrated the BG-Sentinel’s potential as an Ae. aegypti-population sampling tool both in
dengue risk assessment as well as the assessment of the success of mosquito control programmes
(Williams et al. 2007). As with a conventional CDC light trap, the BG-Sentinel requires electrical
power to operate the ventilator. Although the ventilator requires only about three Watts, it has
some disadvantages in routine monitoring operations. Passive trapping systems that use sticky
surfaces to catch mosquitoes do not require electricity and are low cost tools, but the catching
rates of these systems are much lower (between 10 to 50 times, Kröckel et al. 2006). For a sensitive
and high resolution spatial analysis of dengue vectors both trapping systems are useful, the
Mean of collected female Aedes aegypti / 3 h

3.0

2.5

2.0

1.5

1.0

0.5

0.0
BG-Sentinel without CO2 trap (catalytic Trap with visual Human landing
CO2 propane cues only collection
combustion)
Figure 7. Mean values (± standard error) for caught Aedes aegypti (syn. Stegomyia aegypti) from three-hour
catching periods over an eight week period. BG-Sentinel trap with lure dispenser BG Lure (n=32); CO2 trap:
Mosquito Magnet Liberty generates moist, warm CO2 (n=32); Trap with visual cues only: Fay Prince trap with
black and white contrast (n=30); Human landing collection: Voluntary test person as a mosquito catcher (n=32)
(Kröckel et al. 2006).

382  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

sticky traps to provide a high spatial resolution and the BG-Sentinels to provide a fast and highly
sensitive measure of the density of adult vectors.

The high collection rates of the BG-Sentinel allow an accurate determination of vector population
densities and provide sufficient sample material to establish virus infection rates by pooling
mosquitoes. In the transmission of vector-borne diseases, two factors play a role: first, the density
of vectors and, second, the incidence of disease agents in their population (infection rate). With
efficient sampling tools for the vectors involved and with quick tests for the presence of viruses
in these vectors, it is thus possible to assess transmission risk by calculating the Vector Index.
This gives the most meaningful risk index which can be used in the efficient planning and rapid
implementation of the control of vector-borne diseases (J Arias, personal communication).

Although originally developed for catching host-finding females, it has been demonstrated that
the BG-Sentinel trap also catches gravid females, parous and nulliparous females, males, and even
blood-fed females in larger numbers than expected (Maciel de Freitas et al. 2006, Williams et al.
2006). Catching a broad spectrum of mosquitoes in different physiological stages allows a more
precise analysis of the mosquito population in a certain area.

A semi-field study showed that an established population of Ae. aegyti in a greenhouse could be
completely eradicated by the placement of one BG-Sentinel trap. Already two days after the trap’s
placement, the human biting rate in the greenhouse was reduced from 15 bites per minute to less
than one bite per minute (Obermayr 2006). Within six weeks of continuous operation of the trap,
the mosquito population was completely extinguished with no eggs, larvae or adult mosquitoes
being found. A large scale field study with about 600 households is currently ongoing in Manaus,
Brazil to demonstrate that the traps can be also used for control of local mosquito populations
and for lowering the risk of dengue transmission in private homes.

Traps for sampling host-seeking Aedes albopictus

The BG-Sentinel with the BG-Lure has also proven to be a useful trap for another species of the
subgenus Stegomyia, Aedes albopictus. This was reported for the first time in 2005, when Ae.
albopictus were detected in two locations in Italy where ovitraps failed to collect this species
(Bitzhenner et al. 2005). In the same year, Ae. albopictus was detected in BG-Sentinel traps placed
on islands in the Torres strait near Australia (Ritchie et al. 2006). In 2006, Ae. albopictus was found
in BG-Sentinels placed in Libreville, Gabon (Krüger and Hagen 2007). A study comparing the CDC
Light trap baited with carbon dioxide to the BG-Sentinel and another new trap, the CMT-20TM
was published by Meeraus et al. (2007). The BG-Sentinel baited only with the BG-Lure captured six
times more Ae. albopictus that the CDC trap with carbon dioxide. Adding CO2 to the BG-Sentinel
and the BG-Lure improved the catching rate to 33 times that of the carbon dioxide-baited CDC
trap. The CMT-20 (with SkinLureTM, an attractant containing lactic acid) collected significantly
more Ae. albopictus when used with CO2 than without CO2, but did not collect significantly more
Ae. albopictus than the CO2-baited CDC light trap.

The high catching rates of the BG-Sentinel traps for Ae. albopictus have led to the idea of using the
traps for mosquito control. A long-term field study in Cesena, Italy, was conducted from May to
October 2008 in order to investigate the effect of the traps on local mosquito populations. Human
biting rates as well as egg collections with ovitraps were significantly reduced in areas with trap
placement compared to areas without traps (Englbrecht et al. 2009).

Olfaction in vector-host interactions 383

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Practical application of population data of Aedes aegypti

The role of GIS in public health surveillance

Geographic Information Systems (GIS) have been widely used by many countries for health
surveillance. In Canada, the Public Health Agency of Canada developed a Disease Surveillance On-
Line web site that provides mapping and other services for cancer, cardiovascular diseases, major
chronic diseases, notifiable diseases, and injury surveillance (PHAC 2009). Mosquito surveillance
systems are also currently using GIS as an important tool to monitor the distribution and spread
of West Nile virus (BCCDC 2009), dengue (Sithprasasna et al. 1997) and malaria (WHO 2009).

Geographic information systems consist of an automated computer-based system that has the
ability to capture, retrieve, manage, display and analyse large quantities of spatial and temporal
data in a geographical context. The system comprises hardware (computer and printer), software
(GIS software), digitised base maps, information and a whole set of procedures such as data
collection, management and updating (WHO 2009). Vector-borne diseases such as that carried
by Ae. aegypti can be mapped temporally and spatially. Such information, when mapped together
creates a powerful tool for monitoring and controlling the dengue vector. Operations for dengue
control based on GIS were established in Singapore in 1998. Streets, residential buildings and
other relevant databases were obtained and mapped to form the base map layer, such as Aedes
breeding sites, dengue case incidences, complainants’ addresses, sensitive areas, weather data
(rainfall, temperature and relative humidity) were mapped into the GIS (Ai-leen and Song 2000).

In Recife, northeastern Brazil, a new approach to dengue vector surveillance based on the use
of spatial analysis techniques, such as the Kernel density estimator, for the identification of
hotspots of vector populations using data from oviptraps, can be very useful for guiding vector
control operations. The long-term use of modified ovitraps provided with Bacillus thuringiensis
var. israelensis (Bti) as a larvicide was evaluated in various intra-urban landscapes. Results from
the first year of egg collections indicated that this could be a promising strategy for detecting
Ae. aegypti population outbreaks (Regis et al. 2008). It is also possible to integrate remote sensing
satellite imagery, trap placement, meteorological data and census data. Similarly, environmental
data such as temperature, relative humidity, and rainfall can be recorded daily at meteorological
point-stations and also integrated into the system. The main disadvantage of the spatial analysis
technique is that it still relies on the use of oviposition traps, requiring much personnel and
laboratory facilities. Although their use is recommended for the detection of low populations of Ae.
aegypti, their value in the assessment of vector abundance, also between blocks or neighborhoods
is questionable (Focks 2003).

Intelligent Dengue Monitoring System

Because of the difficulties of field data collection of larval surveys by the Brazilian National
Programme for Dengue Control, a new field data acquisition of adult female Ae. aegypti and
processing-generating GIS real-time web-site surveillance system was developed and evaluated
in Brazil. The technology is known as ‘Intelligent Dengue Monitoring system’ (MI-Dengue) and
was developed by a ‘spin-off’ company of the Federal University of Minas Gerais, Belo Horizonte,
Brazil. The MI-Dengue consists of (a) a MosquiTRAP baited with synthetic oviposition attractant
that captures adult female Ae. aegypti, which allows for the identification of the adult vectors
during trap inspection; (b) recording and sending entomological data on electronic spreadsheets
or by mobile phone during trap inspection and (c) an internet site that is an integrated real-

384  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

time adult mosquito surveillance system providing entomological indices and GIS technology
(Eiras and Resende 2009). The MI-Dengue permits researchers to send and make data available
on the internet for municipalities’ health managers to access information on the density of gravid
female Ae. aegypti on georeferenced maps and in analytical tables of the sites monitored with
MosquiTRAP.

The MI-Dengue web-data system consists of three software programmes developed to simplify
information: (a) ‘geo-dengue collection’ installed in portable devices (e.g. palmtops or cell phones)
to record Ae. aegypti field capture data; (b) ‘geo-dengue monitoring’ to process the field data
and produce tables with entomological indices and graphs with trends for analysis; and (c) ‘geo-
dengue’, which produces georeferenced maps of mosquitoes captured with MosquiTRAP and
makes them available to users on the Internet on a weekly basis.

The MI-Dengue technology provides a wide range of GIS tools for public health surveillance and
can be adapted for surveillance of other diseases, such as AIDS, tuberculosis, malaria, leishmaniosis,
etc.

Data acquisition

Field data collected by ovitrap, larval surveys, adult traps and aspirator are traditionally recorded
on printed spreadsheets, attached to a clipboard by field workers. Later, the spreadsheets are sent
to be typed and consolidated on a personal computer. Not only is this method prone to typing
errors, it is also time consuming and labour intensive if a large amount of field data is collected.

An alternative and faster method of acquisition and transport of entomological data is by means
of an electronic spreadsheet installed in palmtop computers or mobile phone (Eiras and Resende
2009). These data are transferred automatically to the municipality’s database, and the internet
site automatically generates the GIS maps and entomological tables for the city council’s dengue
control service. The electronic spreadsheet allows for recording household information (e.g.
resident’s name, address, postal code, and emplacement of the trap within the residence) as well
as number of adult mosquitoes caught in each trap/premise.

There are several advantages of using electronic spreadsheet or mobile phone over conventional
field data acquisition systems. The field data can be accessed immediately (premises visited and
scheduled for visits, trap locations, residents’ names, and so on.) and the entomological indices
can be produced automatically. The operation cost is low because these applications and systems
are maintained and operated automatically and there is no delay between data reported to the
database and data available for web mapping and public health access.

GIS for identifying infestations of female adult Aedes aegypti

Infestations of gravid female Ae. aegypti are measured by the MI-Dengue system in five different
ways for focused vector control: (1) weekly monitoring of Ae. aegypti infestation in the urban
blocks; (2) monitoring re-infestation of blocks; (3) hot spot areas; (4) entomological index and (5)
monitoring infection of adult mosquitoes with dengue virus (Eiras and Resende 2009).

All blocks of urban areas are weekly monitored for Ae. aegypti infestation and the number of female
Ae. aegypti captured with the sticky trap is used to establish colour categories for classifying blocks
as green (zero captures), yellow (one female Ae. aegypti/MosquiTRAP/week = low infestation),

Olfaction in vector-host interactions 385

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

orange (two females/trap/week = medium infestation), and red (> three females/trap/week =
high infestation) (Eiras and Resende 2009). These parameters were based on studies by Ritchie
et al. (2004) showing that in Australia, more than two gravid females A. aegypti/trap/week was
associated with a risk of dengue cases.

These weekly data on infested sites and vector infestation levels provided important information
to the municipal dengue control workers who assisted municipal health managers in targeting and
optimising their Ae. aegypti control activities. The georeferenced maps produced by MI-Dengue
technology and provided weekly on the internet, allowed municipal health managers to identify
urban blocks by colour according to the number of female Ae. aegypti specimens captured. Thus,
dengue control activities could be focused on infested blocks within a 200-meter radius. This
focused vector control strategy, supported by a weekly infestation monitoring system, allowed
spatial localisation of vector infestation and the evaluation of control measures within the radius
represented by the trap.

Entomological index

Two indices are used to calculate the EI: (1) the Mean Female Aedes Index (MFAI), which is the
mean number of female Ae. aegypti specimens caught by traps installed per week and is used for
monitoring vector infestations in neighborhoods and municipalities (Figure 8) (Eiras and Resende
2009); (2) the Temporal Mean Female Aedes Index (MFAIt), which is the average of MFAI values for
the three previous, consecutive weeks and is calculated on a monthly basis for Ae. aegypti. The
MFAIt in neighbourhoods are colour-coded green (MFAIt<0.2), yellow (0.2<MFAIt<0.4), and red
(MFAIt>0.4). The MFAIt was developed based on correlations between the number of dengue cases
and the MFAI entomological index during an outbreak of dengue in Vitoria city, Espírito Santo State,
Brazil, where clusters of dengue cases and mosquitoes were observed. The entomological index
t is classified as ‘risk of dengue’ if MFAIt≥0.4, while ‘dengue alerts’ are issued when MFAIt ranges
between 0.20 and 0.39. The situation is considered ‘risk-free’ in terms of dengue when MFAIt≤0.2.
Using this trategy, the municipality of Presidente Epitacio, Sao Paulo, Brazil, experienced a shift
from ‘dengue alert’ to ‘risk free’ was observed). A reduction in entomological indices after several
weeks of monitoring indicated a strong change in the Ae. aegypti population density that was
influenced either by environmental conditions or by vector control in the targeted and priority
areas, or both (Table 2) (Eiras and Resende 2009).

Monitoring re-infestation of blocks

Three variables are used as follows: (a) recurrence of positive traps (at least one female Ae. aegypti
caught); (b) time (chronological order of the week in which the capture occurred), and (c) number
of female Ae. aegypti in the traps during the previous four weeks in the monitored blocks (Eiras
and Resende 2009).

Monitoring block infestations with adult female Aedes aegypti using GIS

The city council’s mosquito control unit can access GIS-maps on the website for the weekly
infestations of the blocks by female Ae. aegypti, and can thus obtain a spatial and temporal
understanding of the infestations (Eiras and Resende 2009). The black dots on the maps represent
the exact trap location by GPS. The users (city council) can have access to the georeferenced
information (latitude and longitude), household address, block and neighbourhood numbers,
and number of mosquitoes captured during each epidemiological week by clicking on these

386  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

1.0
Pres. Epitácio (SP) 2007 2008

0.8

0.6
MFAI

0.4

0.2

0.0
11 14 17 20 23 26 29 32 35 38 41 44 47 50 1 4 7 10 13 16 19

1.0
Três Lagoas (MS) 2007 2008

0.8

0.6
MFAI

0.4

0.2

0.0
11 14 17 20 23 26 29 32 35 38 41 44 47 50 1 4 7 10 13 16 19

1.0
Bastos (SP) 2007 2008

0.8

0.6
MFAI

0.4

0.2

0.0
11 14 17 20 23 26 29 32 35 38 41 44 47 50 1 4 7 10 13 16 19

Weeks
Figure 8. Weekly monitoring of Mean Female Aedes Index (MFAI) furnished by MI-Dengue in three municipalities
in 2007 and 2008 (Eiras and Resende 2009).

Olfaction in vector-host interactions 387

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Table 2. Example of evolution in entomological indices in neighbourhoods in the Municipality of Presidente

Epitacio (São Paulo State, Brazil), classified according to temporal Mean Female Aedes Index (MFAIt) in the
previous four epidemiological weeks (Eiras and Resende 2009).

Entomological weeks % neighbourhood classified

Non-critical (MFAIt<0.2) Dengue alert Critical (MFAIt>0.4)

(0.2<MFAIt<0.4)

7-10 0.0 44.4 55.6

8-11 22.2 33.3 44.4
9-12 22.2 44.4 33.3
10-13 22.2 66.7 11.1
11-14 33.3 55.6 11.1
12-15 66.7 33.3 0.0
13-16 88.9 0.0 11.1
14-17 88.9 11.1 0.0

dots (traps). The weekly infestation levels are provided by colour-coded information on the
blocks, based on the number of female Ae. aegypti captured in each block. The dengue vector
control measures can be based on the weekly observations of infestation levels, where effort is
concentrated in given areas or neighbourhoods of red blocks, thus facilitating implementation of
targeted control measures.

Monthly monitoring of Ae. aegypti reinfestation in the neighbourhoods

Eiras and Resende (2009) showed in preliminary studies that using the MI-Dengue technology in
three cities in Brazil reduced entomological indices after seven weeks of monitoring adult female
Ae. aegypti followed by directional control measures. The MI-Dengue technology was used in
three districts (30,000-75,000 inhabitants) during one year and vector control was based on the
website information of ‘dengue alert’ or ‘hot spots’ of adult female Ae. aegypti. Results showed that
there was a trend of shifting from categories ‘dengue alert’ and ‘critical’ to ‘risk free’, indicating a
strong seasonal variation in the Ae. aegypti population density that was probably influenced by
the climatic conditions and/or targeted control measures. The incidence of dengue cases was also
significantly reduced when compared with the neighbourhood districts (Figure 9). Although, this
was a pilot study, further data are currently being collected in 45 Brazilian cities during 2009 with
the populations of the cities ranging from 30,000 to 700,000 inhabitants.

Hot spot areas

Cluster analysis of the capture of female mosquitoes by MosquiTRAP is based on the weekly mean
number of female Aedes index per block. Space-time Permutation Scan Statistical Model (STPSSM)
(Kulldorff et al. 2005) and Poisson Scan Statistic Model (PSSM) (Kulldorff and Nagarwalla 1995,
Kulldorff et al. 1998) were used to identify whether Ae. aegypti clusters in space are randomly
distributed or not (Figure 10).

388  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

A Andradina-SP Presidente Venceslau-SP Rancharia-SP

10,000
Incidence per 100,000 habitants

1000
-94.6
-69.9
100 -84.5

10

1
2006 2007 2008

B Divinópolis-MG Pará de Minas-MG Santa Luzia-MG

1000
Incidence per 100,000 habitants

100 -32.0
+76.9
+109.4

10

1
2006 2007 2008

C Três Lagoas-MS Presidente Epitácio-SP Bastos-MG

10,000
Incidence per 100,000 habitants

1000

-82.5
100
-99.0

10 -99.7

1
2006 2007 2008
Figure 9. Total autochthonous dengue cases and incidence in municipalities that (A) did not adopt MI-Dengue, (B)
partially MI-Dengue and (C) fully adopted MI-Dengue in early 2008. Numbers in parenthesis means percentage of (-)
reduction or (+) increasing in incidence of autochthonous dengue in 2008 (adapted from Eiras and Resende 2009).

Olfaction in vector-host interactions 389

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Figure 10. (A) Presence of hot spot areas of gravid female Aedes aegypti in Vitoria (Espirito Santo State, Brazil)
detected by MI-Dengue weekly monitoring at week 12 (2009). The window shown on the screen allowed to follow
weekly maps. (B) GIS of infected mosquitoes with dengue virus (small circles are dengue cases and bigger circles
are infected Aedes aegypti mosquitoes with dengue virus).

390  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

Measurement of dengue- infected Aedes aegypti mosquitoes using GIS

In order to control and prevent dengue transmission, it is also important to rapidly detect and type
the virus in adult mosquitoes. For instance, in Brazil, the control of adult Ae. aegypti is trigged only
when a person gets dengue fever (PNCD 2002). Thus, information indicating the number of Ae.
aegypti infected by dengue virus may replace the current indicator for adult Ae. aegypti control.

Recently in Brazil, a rapid and stable method for detection and typing of dengue virus (DEN) in Ae.
aegypti was established and associated with the MI-Dengue technology. Vitoria city, Espirito Santo
State, Brazil, has been using the MI-Dengue technology since January 2007 and approximately
1,450 sticky traps were placed all over the urban area. The entomological index and GIS maps
are produced on a weekly basis. The adult female Ae. aegypti mosquitoes that are trapped in the
MosquiTRAP, are pooled in groups of about 20 mosquitoes and submitted for virus detection and
type identification by reverse transcriptase RT-PCR. Data on infected mosquitoes is plotted in GIS
web-online for the city council. At the moment of writing, it takes 10 days to analyse mosquitoes
for virus infection because samples need to be transported and the analysis is done manually. In
future, it is expected that the detection of virus in mosquitoes can be automated. It was found
that Ae. aegypti was infected with virus DENV-1, DENV-2 and DENV-3 in urban areas during many
months (Figure 10B). The hot spot areas with infected mosquitoes were considered priorities for
vector control. In a preliminary study of an outbreak of dengue in Vitoria city, it was observed
that 85% of dengue cases occurred within a 200 m radius of sticky traps having caught infected
female Ae. aegypti (AE Eiras, unpublished data) This method is very helpful and it has been used
as baseline to develop a new early surveillance system to provide early warning for dengue fever
epidemics, in furnishing information for epidemiologic studies, and for effective vector control
measures.

Monitoring gravid female Ae. aegypti with MosquiTRAP and MI-Dengue allows the Municipal
Health Secretariats to conduct weekly follow-up of infestation trends in neighbourhoods and
municipalities. The practicality of MosquiTRAP and MI-Dengue to monitor female adult Ae. aegypti
in real time is substantial when compared with other georeferenced studies that use sampling
of immature forms of the mosquito (eggs and larvae) (Ai-leen and Song 2000; Chansang and
Kittayapong 2007), which require time and infrastructure for identification, quantification, and
data processing. Our experience has shown that these innovative technological tools are not
difficult to learn nor are they difficult to integrate into existing dengue control programmes.

Potential use of traps for dengue suppression

Efficient tools for the monitoring of adult Ae. aegypti, such as the MosquiTRAP and the BG-Sentinel
trap, have a wide range of applications in both dengue control operations and research. Both trap
types allow the detection of virus-positive female Ae. aegypti during standard weekly monitoring
(even if the incidence of these females is very low) and at the peak of acute epidemics.

A novel strategy has been proposed to effectively suppress dengue transmission in urbanised
areas by the use of innovative mosquito traps. This has not been tried before, probably because
of the low performance, high costs, and complexity of existing trapping technologies. The BG-
Sentinel and MosquiTRAP have the potential to be produced economically at a large scale due
to their simple technology. Although the efficacy of the trap has been demonstrated already in
scientific studies and monitoring programmes, the trap has not been used before as a control
tool to suppress dengue transmission. In 2009, two pilot projects were designed to prove that

Olfaction in vector-host interactions 391

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

a lure and kill control strategy encompassing environmentally-friendly traps for adult vectors
can suppress dengue transmission. Both projects are currently being implemented in Manaus
city, Amazon State, Brazil, and the Salvador, Bahia State, Brazil). Using an integrated community-
based approach for proper education and training of the trap users, the traps are installed in
the household or premises, public buildings, and working places. They can be placed indoors
as well as outdoors, depending on the situation and prevalence of the vectors. The traps should
run continuously throughout the whole season or year. The theoretical benefits of both traps are
increased vector mortality, reduction of human biting rates, early indication of breeding sites
through monitoring, enhancement of individual responsibility in community based programmes,
environmentally friendly (no toxic substances are used) and sustainable. These are important
preconditions for a successful implementation of the traps and the acceptance of the users.
Education and training of the trap user is very important to achieve all the benefits of the traps.
Therefore the implementation of the proposed trapping strategy has to be integrated in already
existing or newly implemented community-based dengue control programs. The use of the traps
is expected to have a positive effect on the continued motivation of individuals to take action
against the vectors, by source reduction, using additional traps, or applying insecticide treatments.

Acknowledgments

Alvaro Eiras thanks to CNPq, FAPEMIG, SEBRAE, FINEP and the Programa Nacional de Controle
da Dengue (PNCD), the Brazilian Health Ministry and the ‘spinoff’ Ecovec for supporting the
development and evaluation of the MI-Dengue technology.

References

Acree F Jr., Turner, RB, Gouck HK, Beroza M and Smith N (1968) L-Lactic acid: a mosquito attractant isolated from humans.
Science 161: 1346-1347.
Ai-leen GT and Song RJ (2000) The use of GIS in ovitrap monitoringfor dengue control in Singapore Dengue Bull 24:
110-116.
Allan SA and Kline DL (1995) Evaluation of organic infusions and synthetic compounds mediating oviposition in Aedes
albopictus and Aedes aegypti (Diptera: Culicidae). J Chem Ecol 21: 1847-1860.
Barrera R, Amador M and Clark GG (2006) Use of the pupal survey technique for measuring Aedes aegypti (Diptera:
Culicidae) productivity in Puerto Rico. Am J Trop Med Hyg 74: 290-302.
Beehler JW, Millar JG and Mulla MS (1994) Field evaluation of synthetic compounds mediating oviposition in Culex
mosquitoes (Diptera: Culicidae). J Chem Ecol 20: 281-291.
Bentley MD and Day JF (1989) Chemical ecology and behavior aspects of mosquito oviposition. Annu Rev Entomol
34: 401-421.
Bentley MD, McDaniel IN, Yatagai M, Lee H-P and Maynard R (1979) p-Cresol: an oviposition attractant of Aedes triseriatus.
Environ Entomol 8: 206-209.
Bitzhenner M, Guaraglia Ch, Geier M, Rose A and Talbalaghi A (2005) Evaluation of the BG-Sentinel, a new monitoring
trap for mosquitoes, in northern Italy. Poster presentation at the 4th International Congress of Vector Ecology, Reno,
Nevada, USA, 2-7 October 2005.
Bosch JO, Geier M and Boeckh J (2000) Contribution of fatty acids to olfactory host finding of female Aedes aegypti.
Chem Senses 25: 323-330.
Braga IA, Gomes AC, Nelson M, Mello RCG, Bergamaschi DP and Souza JMP (2000) Comparação entre pesquisa larvária
e armadilha de oviposição, para detecção de Aedes aegypti. Rev Bras Med Trop 33: 347-353.
Breteau H (1954) La fiève jaune en Afrique occidentale française. Un aspect de la médecine préventive massive. Bulletin
WHO 11: 453-481.

392  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

British Columbia Centre for Disease Control (BCCDC) (2009) Interactive GIS mapping for West Nile virus. Available at
http: //maps.bccdc.org.
Brown AWA, Sarkaria DS and Thompson RP (1951) Studies on the responses of the female Aedes mosquito. Part I. - The
search for attractant vapours. Bull Ent Res 42: 105-114.
Canyon DV and Hii JLK (1997) Efficacy of carbon dioxide, 1-octen-3-ol, and lactic acid in modified Fay-Prince traps as
compared to man-landing catch of Aedes aegypti. J Am Mosq Control Assoc 13: 66-70.
Carlson DA, Smith N, Gouck HK and Goodwin DR (1973) Yellow fever mosquitoes: compounds related to lactic acid that
attract females. J Econ Entomol 66: 329-331.
Chadee DD (1993) Oviposition response of Aedes aegypti (L.) to the presence of conspecific eggs in the field in Trinidad,
WI. J Florida Mosq Contr Ass 64: 63-66.
Chadee DD, Lakkan A, Ramdath WR and Persad RC (1993) Oviposition response of Aedes aegypti mosquitoes to different
concentrations of hay infusion in Trinidad, West Indies. J Am Mosq Control Assoc 9: 346-348.
Chadee DD and Martinez R (2000) Landing periodicity of Aedes aegypti with implications for dengue transmission in
Trinidad, West Indies. J Vector Ecol 25(2): 158-163.
Chadee DD, Doon R and Severson DW (2007) Surveillance of dengue fever cases using a novel Aedes aegypti population
sampling method in Trinidad, West Indies: the cardinal points approach. Acta Trop 104: 1-7
Chansang C and Kittayapong P (2007) Application of mosquito sampling count and geospatial methods to improve
dengue vector surveillance. Am J. Trop Med Hyg 77: 897-902.
Clark GG, Seda H and Gubler DJ (1994) Use of the ‘CDC backpack aspirator’ for surveillance of Aedes aegypti in San Juan,
Puerto Rico. J Am Mosq Control Assoc 10: 119-124.
Clements AN (1999) The biology mosquitoes. Volume 2: sensory reception and behaviour. CABI, Wallingford, UK.
Cobert PS and Chadee DD (1993) An improved method for detecting substrate preferences shown by mosquitoes that
exhibit skip oviposition. Physiol Entomol 18: 114-118.
Coelho GE, Burattini MN, Teixeira MG, Coutinho FAB and Massad E (2008) Dynamics of the 2006/2007 dengue outbreak
in Brazil. Mem Inst Oswaldo Cruz 103: 535-539.
Connor ME and Monroe WM (1923) Stegomyia índices and their value in the yellow fever control. Am J Trop Med 3: 9-19.
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (2009) Edital doenças neglegenciadas - MCT/
CNPq/CT-Saúde/MS/SCTIE/DECIT No. 034/2008. Available at: http: //www.cnpq.br/editais/ct/encerrados.htm.
Coutinho FAB, Burattini MN, Lopez LF and Massad E (2006) Threshold conditions for a non-autonomous epidemic
system describing the population dynamics of dengue. Bull Math Biol 68: 2263-2282.
Cuéllar-Jiménez ME, Velásquez-Escobar OL, González-Obando R and Morales-Reichmann CA (2007) Detección de Aedes
albopictus (Skuse) (Diptera: Culicidae) en la ciudad de Cali, Valle del Cauca, Colombia. Biomédica 27: 273-279.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972
Dennett JA, Wuithiranyagool T, Reyna-Nava M, Bala A, Tesh RB, Parsons RE and Bueno RJr (2007) Description and use
of the Harris County Gravid trap for West Nile virus surveillance 2003–06. J Am Mosq Control Assoc. 23: 359-362.
Dibo MR, Chiaravalloti-Neto F, Battigaglia M, Mondini A, Favaro EA, Barbosa AC and Glasser CM (2005) Identification of
the best ovitrap installation sites for gravid Aedes (Stegomyia) aegypti in residences in Mirassol, state of São Paulo,
Brazil. Mem Inst Oswaldo Cruz,100: 339-343
Du YJ and Millar JG (1999) Electroantennogram and oviposition bioassay responses of Culex quinquefasciatus and Culex
tarsalis (Diptera: Culicidae) to chemicals in odors from bermuda grass infusions. J Med Entomol 36: 158-166.
Effler PV, Pang L, Kitsutani Pl, Vorndam V, Nakata M, Ayers T, Elm J, Tom T, Reiter P, Rigau-Perez JG, Hayes JM, Mills K,
Napier M, Clark GG and Gubler (2005) Dengue fever, Hawaii, 2001–2002. Emerg Infect Dis 11: 742-749.
Eiras AE (2001) Mediadores químicos entre hospedeiros e insetos vetores de doenças médico-veterinárias. In: Vilela
EF and Lúcia MTD (eds) Feromônios de insetos: biologia, química e emprego no manejo de pragas. Editora Holos,
Ribeirão Preto, SP, Brasil.
Eiras AE (2002) Armadilha para captura de mosquitos. Brazilian Patent PI0203907, 9 Sept 2002.
Eiras AE and Jepson PC (1991) Host location by Aedes aegypti (Diptera: Culicidae): A wind tunnel study of chemical cues.
Bull Entomol Res 81: 151-160.

Olfaction in vector-host interactions 393

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Eiras AE and Jepson PC (1994) Responses of female Aedes aegypti (Diptera: Culicidae) to host odours and convection
currents using an olfactometer bioassay. Bull Entomol Res 84: 207-211.
Eiras AE and Resende MC (2009) Preliminary evaluation of the ‘Dengue-MI’ technology for Aedes aegypti monitoring
and control. Cad Saude Publica 25 Suppl 1: S45-58.
Eiras AE and Sant’Ana AL (2002) Atraentes de oviposição para mosquitos. Brazilian Patent PI 0106701-0, Dec 2001.
Eiras AE, Resende MC and Silva IM (2007) Proposta de validação da MosquiTRAP e do sistema de monitoramento
(MI-Dengue): uma nova tecnologia para a geração de novos índices de vigilância entomológica para o Programa
Nacional de Controle da Dengue (PNCD). Research Report, Fundação Nacional de Saúde (FUNASA), Brasilia, Brasil.
Eiras AE, Sant’Ana AL and Stein K (2001) Identification of volatiles from grass infusions that attract gravid Aedes aegypti
mosquito. Presentation at the 3rd International Congress of Vetor Ecology, Barcelona, Spain, 16-21 September 2001.
Eiras AE, Silva IM and Resende MC (2005) Proposta de novo método de monitoramento e de novos índices de vigilância
entomológica usando MosquiTRAP, uma nova armadilha para a captura de adultos do mosquito Aedes aegypti.
Research Report, Fundação Nacional de Saúde (FUNASA), Brasilia, Brasil.
Eiras AE, Silva IM, Roque RA, Matosinhos IM and Geier M (2004) Behavioural responses of gravid Aedes aegypti (Diptera:
Culicidae) to synthetic oviposition attractants identified from grass infusions volatiles. Presentation at the XXII
International Congress of Entomology, Brisbane, Queensland, Australia, 15-21 August 2004.
Englbrecht Ch, Venturelli C, Rose A and Geier M (2009) Continuous trapping of adult Asian tiger mosquitoes (Aedes
albopictus) with BG-Sentinel traps reduced the human landing rate and density indices in an urban environment
in Cesena, Italy. Oral presentation at the 5th European Mosquito Control Association Workshop, Turin, Italy, 9-13
March 2009.
Facchinelli L, Valerio L, Pombi M, Reiter P, Constantini C and Della Torre A (2007) Development of a novel sticky trap for
container-breeding mosquitoes and evaluation of its sampling properties to monitor urban populations of Aedes
albopictus. Med Vet Entomol 21: 183-195.
Favaro AE, Dibo MR, Mondini A, Ferreira AC, Barbosa AAC, Eiras AE, Barata EAMF and Chiaravalloti-Neto F (2006)
Physiological state of Aedes (Stegomyia) aegypti mosquitoes captured with MosquiTRAPs in Mirassol, São Paulo,
Brazil. J Vector Ecol 31: 285-291.
Favaro EA, Mondini A, Dibo MR, Barbosa AAC, Eiras AE, Barata EAMF and Chiaravalloti-Neto F (2008) Assessment of
entomological indicators of Aedes aegypti (L.) from adult and egg collections in São Paulo, Brazil. J Vector Ecol 33:
8-16.
Fay RW and Eliason DA (1966) A preferred oviposition site as a surveillance method for Aedes aegypti. Mosq News 26:
531-535.
Fay RW and Prince WH (1970) A modified visual trap for Aedes aegypti. Mosq News 30: 20-23.
Feltner H and Ferrao P (2008) Evaluating efficacy of the BG-Lure attractant using three mosquito trap designs in the city
of Alexandria, Virginia. Presenation at the 33rd Annual Conference of the Mid-Atlantic Mosquito Control Association,
Baltimore, Maryland, USA, 27-29 Feb 2008. Available at: http://www.bg-sentinel.com/bilder/Feltner_Ferrao-2008-
Evaluating_efficacy_of_BG-Lure.pdf.
Focks DA (2003) A review of entomological sampling methods and indicators for dengue vectors. TDR/ IDE/Den/03.1
WHO, Geneva, Switzerland.
Focks DA and Alexander N (2006) Multicountry study of Aedes aegypti pupal productivity survey methodology: findings
and recommendations. TDR/IRM/Den /06.1 WHO, Geneva.
Focks DA, Daniels E, Haile DG and Keesling JE (1995) A simulation model of the epidemiology of urban dengue fever:
literature analysis, model development, preliminary validation, and samples of simulation results. Am J Trop Med
Hyg 53: 489-506.
Forattini OP, Kakitani I, Massad E and Marucci D (1995) Studies mosquitoes (Diptera: Culicidae) and anthropic
environment. 9. Synanthropic and epidemiological vector role of Aedes scapularis in South-Eastern Brazil. Rev
Saúde Pública 29: 199-207.
Gama RA, Silva EM, Silva IM, Resende MC and Eiras AE (2007) Evaluation of the sticky MosquiTRAP™ for detecting Aedes
(Stegomyia) aegypti (L.) (Diptera: Culicidae) during the dry season in Belo Horizonte, Minas Gerais, Brazil. Neotrop
Entomol 36: 294-302.

394  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

Geier M and Boeckh JA (1999) A new Y-tube olfactometer for mosquitoes to measure the attractiveness of host odours.
Entomol Exp Appl 92: 9-19.
Geier M, Bosch JO and Boeckh J (1999) Ammonia as an attractive component of host odour for the yellow fever
mosquito, Aedes aegypti. Chem Senses 24: 647-653.
Geier M, Bosch JO and Boeckh J (1999) Influence of odour plume structure on upwind flight of mosquitoes towards
hosts. J Exp Biology 202: 1639-1648.
Geier M, Rose A and Eiras AE (2004) Insect trap. International Patent WO2004/054358A2, 1 Jul 2004
Gubler DJ (1998) Dengue and dengue hemorrhagic fever. Clin Microbiol Rev 11: 480-496.
Gubler DJ (2004) The changing epidemiology of yellow fever and dengue, 1900 to 2003: full circle? Comp Immun
Microbiol Infect Dis 27: 319-330.
Harrison BA, Callahan MC, Watts DM and Panthusiri L (1982) An efficient floating larval trap for sampling Aedes aegypti
populations (Diptera: Culicidae). J Med Entomol 19: 722-727.
Holck AR, Meek CL and Holck JC (1988) Attractant enhanced ovitraps for the surveillance of container breeding
mosquitoes. J Am Mosq Control Assoc 4: 97-98.
Honorio NA, Codeço CT, Alves FC, Magalhães MAFM and Lourenço-de-Oliveira R (2009) Temporal distribution of Aedes
aegypti in different districts of Rio de Janeiro, Brazil, measured by two types of traps. J Med Entomol 46: 1001-1014
Hornby JA, Moore DE and Miller TW (1994) Aedes albopictus, distribution, abundance, and colonization in Lee County,
Florida, and its effect on Aedes aegypti. J Am Mosq Control Assoc 10: 397-402.
Hribar LJ, Leprince DJ and Foil LD (1992). Ammonia as an attractant for adult Hybomitra lasiophthalma (Diptera:
Tabanidae). J Med Entomol 29: 346-348.
Hwang YS, Kramer WL and Mulla M (1980) Isolation and identification of oviposition repellents for Culex mosquitoes.
J Chem Ecol 6: 71-80.
Jones JW, Sithiprasasna R, Schleich S and Coleman RE (2003) Evaluation of selected traps as tools for conducting
surveillance for adult Aedes aegypti in Thailand. J Am Mosq Control Assoc 19: 148-150.
Kay BH, Sutton KA and Russell BM (2000) A sticky entry-exit trap for sampling mosquitoes in subterranean habitats. J
Am Mosq Contr Assoc 16: 262-265.
Kline DL (2002) Evaluation of various models of propane-powered mosquito traps. J Vector Ecol 27: 1-7.
Knols BGJ, Van Loon JJA, Cork A, Robinson RD, Adam W, Meijerink J, De Jong R, Takken W (1997) Behavioural and
electrophysiological responses of the female malaria mosquito Anopheles gambiae s.s. Giles (Diptera: Culicidae)
towards Limburger cheese volatiles. Bull Entomol Res. 87: 151-159.
Knols BGJ, Njiru BNN, Mathenge EM, Mukabana WR, Beier JC and Killeen GF (2002) MalariaSphere: A greenhouse-
enclosed simulation of a natural Anopheles gambiae (Diptera: Culicidae) ecosystem in western Kenya. Malaria J
1: 19.
Koenraadt CJM, Jones JW, Sithiprasasna R and Scott TW (2007) Standardizing container classification for immature
Aedes aegypti surveillance in Kamphaeng Phet, Thailand. J Med Entomol 44: 938-944.
Kröckel U, Rose A, Eiras AE and Geier M (2006) New tools for surveillance of adult yellow fever mosquitoes: Comparison
of trap catches with human landing rates in an urban environment. J Am Mosq Control Assoc 22: 229-238.
Krueger A and Hagen RM (2007) Short communication: first record of Aedes albopictus in Gabon, Central Africa. Trop
Med Int Health 12: 1105-1107.
Kulldorff M, Athas WF, Feurer EJ, Miller BA and Key CR (1998) Evaluating cluster alarms: a space-time scan statistic and
brain cancer in Los Alamos, New Mexico. Am J Pub Health 88: 1377-1380.
Kulldorff M, Heffernan R, Hartman J, Assunção R and Mostashari F (2005) A space-time permutation scan statistic for
disease outbreak detection. PLoS Med 2: e59.
Kuno G (1997) Factors influencing the transmission of dengue viruses. In: Gubler DJ and Kuno G (eds) Dengue and
dengue hemorrhagic fever. CABI, Wallingford, UK, pp 61-88.
Leal WS, Barbosa RMR, Xu W, Ishida Y and Syed Z (2008) Reverse and conventional chemical ecology approaches for the
development of oviposition attractants for Culex mosquitoes Plos One 3: e3045.
Lourenço-de-Oliveira R, Lima JB, Peres R, Alves FC, Eiras AE and Codeço CT (2008) Comparison of different uses of adult
traps and ovitraps for assessing dengue vector infestation in endemic areas. J Am Mosq Control Assoc 24: 387-392.

Olfaction in vector-host interactions 395

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Lugo E del C, Moreno G, Zachariah MA, López MM, López JD, Delgado MA, Valle SI, Espinoza PM, Salgado MJ, Pérez R,
Hammond SN and Harris E (2005) Identification of Aedes albopictus in urban Nicaragua. J Am Mosq Control Assoc
21: 325-327.
Maciel de Freitas R, Codeco CT and Lourenço de Oliveira R (2007) Daily survival rates and dispersal of Aedes aegypti
females in Rio de Janeiro, Brazil. Am J Trop Med Hyg 76: 659-665.
Maciel de Freitas R, Eiras AE and Lourenço de Oliveira R (2006) Field evaluation of effectiveness of the BG-Sentinel, a new
trap for capturing adult Aedes aegypti (Diptera: Culicidae). Mem Inst Oswaldo Cruz 101: 321-325.
Maciel de Freitas R, Eiras AE and Lourenco de Oliveira R (2008) Calculating the survival rate and estimated population
density of gravid Aedes aegypti (Diptera, Culicidae) in Rio de Janeiro, Brazil. Cad Saúde Pública 24: 2747-2754.
Maciel de Freitas R and Lourenco de Oliveira R (2009) Presumed unconstrained dispersal of Aedes aegypti in the city of
Rio de Janeiro, Brazil. Rev Saúde Pública 43: 8-12.
Maestre-Serrano R, Vergara-Sanchez C, Berrueco-Rodriguez G, Bello-Novoa B and Brochero H (2008) Presencia
de Haemagogus equinus Theobald, 1903 (Diptera: Culicidae) en los municipios de Soledad y Malambo en el
departamento del Atlántico, Colombia, 1998-2005. Biomédica 28: 99-107.
Marques CCA, Marques GRAM, Brito M, Santos Neto LG, Ishibashi VC and Gomes FA (1993) Estudo comparativo de
eficácia de larvitrampas e ovitrampas para vigilância de vetores de dengue e febre amarela. Rev Saúde Pública
27: 237-241.
Mboera LEG, Takken W, Mdira KY and Pickett JA (2000) Sampling gravid Culex quinquefasciatus (Diptera: Culicidae) in
Tanzania with traps baited with synthetic oviposition pheromone and grass infusions. J Med Entomol 37: 172-176.
Meeraus W, Johnson J and Arias JR (2007) Field comparison of novel and industrial standard traps for collecting Aedes
albopictus in northern Virginia. Poster presentation at the 73rd Annual Meeting of American Mosquito Control
Association, Orlando, Florida, USA, 1-5 April 2007.
Meeraus WH, Armistead JS and Arias JR (2008) Field comparison of novel and gold standard traps for collecting Aedes
albopictus in northern Virginia. J Am Mosq Control Assoc 24: 244-248.
Méndez F, Barreto M, Arias JF, Rengifo G, Muñoz J, Burbano ME and Parra B. (2006) Human and mosquito infections by
dengue viruses during and after epidemics in a dengue-endemic region of Colombia. Am J Trop Med Hyg. 74(4):
678-83.
Metha DR (1934) Effect of saline and free ammonia on the oviposition of Anopheles culicifacies and Anopheles subpiscus
(Rossi). Rev Mal Surv India 4: 411-420.
Micieli MV and Campos RE (2003) Oviposition activity and seasonal pattern of a population of Aedes (Stegomyia) aegypti
(L.) (Diptera: Culicidae) in subtropical Argentina. Mem Inst Oswaldo Cruz 98: 659-663.
Millar JG, Chaney JD and Mulla MS (1992) Identification of oviposition attractants for Culex quinquefasciatus from
fermented Bermuda grass infusions. J Am Mosq Control Assoc 8: 11-17.
Morrison AC, Gray K, Getis A, Astete A, Sihuincha M, Focks D, Watts D, Stancil JD, Olson JG, Blair P and Scott TW (2004)
Temporal and geographic patterns of Aedes aegypti (Diptera: Culicidae) production in Iquitos, Peru. J Med Entomol
41: 1123-1142.
Morrison AC, Zielinski-Gutierrez E, Scott TW and Rosenberg R (2008) Defining challenges and proposing solutions for
control of the virus vector Aedes aegypti. PLoS Med 5: 0362-0366.
Mota RN (2003) Construção de um olfatômetro de dupla escolha para estudos de orientação de fêmeas grávidas de
Aedes aegypti (Linnaeus, 1792) (Diptera: Culicidae) à atraentes de oviposição. MSc Dissertation, Universidade
Federal de Minas Gerais, Brazil.
Ministério da Saúde (MS) (2007a) Graphic showing the historical series of reported dengue cases between 1980 and
2005. Available at: http://portal.saude.gov.br/portal/arquivos/ pdf/Graficos_DNC_dengue_verde.pdf.
Ministério da Saúde (MS) (2007b) Graphic showing the historical series of mortality due to dengue cases between
1980 and 2005. Available at: http: //portal.saude.gov.br/portal/arquivos/pdf/Graficos_DNC_dengue_vermelho.pdf.
Muir LE and Kay BH (1998) Aedes aegypti survival and dispersal estimated by mark-release-recapture in Northern
Australia. Am J Trop Med Hyg 58: 277-282.
Müller W.(1968). Die Distanz- und Kontaktorientierung der Stechmücken Aedes aegypti (Wirtsfindung, Stechverhalten
und Blutmahlzeit). Z vergl Physiol 58: 241-303.

396  Olfaction in vector-host interactions

 17. Practical application of olfactory cues for monitoring and control of Aedes aegypti in Brazil: a case study

Nasci RS (1981) A light weight battery-powered aspirator for collecting mosquitoes in the field. Mosq News 41: 808-811.
Obermayr R (2006) Are new trapping technologies useful for mosquito control interventions? Vector Ecol Newsl 37
(3): 11-12.
Ordonez JG, Fernandez Salas I and Flores-Leal A (1997) Monitoring dispersal of marked Aedes aegypti females under
field conditions using sticky ovitraps in Monterrey, northeastern Mexico. J Am Mosq Assoc 13: 121.
Ordóñez-Gonzalez JG, Mercado-Hernandez R, Flores-Suarez AE and Fernández-Salas I (2001) The use of sticky ovitraps
to estimate dispersal of Aedes aegypti in Northeastern Mexico. J Am Mosq Contr Assoc 17: 93-97.
PNCD (2002) Programa Nacional de Controle da Dengue. Ministério da Saúde Fundação Nacional de Saúde, Brasília,
Brasil. Available at: http://bvsms.saude.gov.br/bvs/publicacoes/pncd_2002.pdf.
Public Health Agency of Canada (PHAC) (2009) Notifiable diseases on-line. Available at: http://dsol-smed.phac-aspc.
gc.ca/dsol-smed/ndis/index-eng.php.
Ponnusamy L, Xu N, Nojima S, Wesson DM, Schal C and Apperson CA (2008) Identification of bacteria and bacteria-
associated chemical cues that mediate oviposition site preferences by Aedes aegypti. PNAS 105: 9262-9267.
Rawlins SC, Martinez R, Wiltshire S and Legall G (1998) A comparison of surveillance systems for the dengue vector
Aedes aegypti in port of Spain, Trinidad. J Am Mosq Control Assoc 14: 131-136.
Regis L, Monteiro AM, Varjal de Melo-Santo MA, Silveira Jr JC, Freire Furtado A, Veiga Aciolis R, Santo GM, Nkazawa M, Sa
Carvalho M, Ribeiro Jr PJ, Viera de Souza W (2008) Developing new approaches for detecting and preventing Aedes
aegypti population outbreaks: basis for surveillance, alert and control system. Mem Inst Oswaldo Cruz 103: 50-59.
Reiter P (1983) A portable, battery-powered trap for collecting gravid Culex mosquitoes. Mosq News 4: 496-498.
Reiter P (1986) A standardized procedure for the quantitative surveillance of certain Culex mosquitoes by egg raft
collection. J Am Mosq Control Assoc 3: 494-501.
Reiter P, Amador MA and Colon N (1991) Enhancement of the CDC ovitrap with hay infusion for daily monitoring of A.
aegypti populations. J Am Mosq Control Assoc 7: 52-55.
Reiter P, Nathan MB (2001) Guidelines for assessing the efficacy of insecticide space sprays for the control of the Dengue
vector Aedes aegypti. WHO/CDS/CPE/PVC/2001.1, WHO, Geneva, Switzerland.
Ritchie SA, Long S, Hart A, Webb CE and Russell RC (2003) An adulticidal sticky ovitrap for sampling container-breeding
mosquitoes. J Am Mosq Control Assoc 19: 235-242.
Ritchie SA (1984) Hay infusion and isopropyl alcohol-baited CDC light trap; a simple effective trap for gravid Culex
mosquitoes. Mosq News 44: 404-407.
Ritchie SA, Moore P, Carruthers M (2006) Discovery of a widespread infestation of Aedes albopictus in the Torres Strait,
Australia. J Am Mosq Control Assoc 22: 358-365.
Ritchie SA, Long S, Smith G, Pike A and Knox TB (2004) Entomological investigations in a focus of dengue transmission
in Cairns, Queesland, Australia, by using the sticky ovitraps. J Med Entomol 41: 1-4.
Roque RA (2007) Formulação e avaliação de atraentes sintéticos de oviposição, identificados em infusões do capim
colonião (Panicum maximum) para fêmeas grávidas do mosquito Aedes (Stegomyia) aegypti (Linnaeus, 1762)
(Diptera: Culicidae) em condições de campo e semi-campo. Dissertation, Universidade Federal de Minas Gerais,
Brazil.
Roque RA and Eiras AE (2008) Calibration and evaluation of field cage for oviposition study with Aedes (Stegomyia)
aegypti female (L.) (Diptera: Culicidae). Neotrop Entomol 37: 478-485.
Rössler HP (1961) Versuche zur geruchlichen Anlockung weiblicher Stechmücken (Aedes aegypti L. (Culicidae). Z Vgl
Physiol 44: 184-231.
Russell RC (2004) The relative attractiveness of carbon dioxide and octenol in CDC- and EVS-type light traps for sampling
the mosquitoes Aedes aegypti (L.), Aedes polynesiensis Marks, and Culex quinquefasciatus Say in Moorea, French
Polynesia. J Vector Ecol 29: 309-314.
Russel RC, Webb CE, Williams CR and Ritchie SA (2005) Mark-release-recapture study to measure dispersal of the
mosquito Aedes aegypti in Cairns, Queensland, Australia. Med Vet Entomol 19: 1-7.
Russell RC and Ritchie SA (2004) Surveillance and behavioral investigations of Aedes aegypti and Aedes polynesisensis in
Moorea, French Polynesia, using a sticky ovitrap. J Am Mosq Control Assoc 20: 370-375.

Olfaction in vector-host interactions 397

Alvaro E. Eiras, Martin Geier, Andreas Rose and Owen Jones

Sant’Ana AL (2003) Avaliação, extração, identificação e estudos eletrofisiológicos dos voláteis presentes em infusões
de Panicum maximum que estimulam e/ou atraem fêmeas de Aedes (Stegomyia) aegypti Linnaeus, 1762 (Diptera:
Culicidae) para oviposição. Dissertation, Universidade Federal de Minas Gerais, Minas Gerais, Brazil.
Sant’Ana AL, Roque RA and Eiras AE (2006) Characteristics of grass infusions as oviposition attractants to Aedes
(Stegomyia) (Diptera: Culicidae). J Med Entomol 43: 214-20.
Service MW (1993) Mosquito ecology: field sampling methods. Second edition. Chapman & Hall, London, UK.
Schoeler GB, Schleich SS, Manweiler SA and Sifuentes VA (2004) Evaluation of surveillance devices for monitoring Aedes
aegypti in an urban area of northeastern Peru. J Am Mosq Control Assoc 20: 6-11.
Schmied WH, Takken W, Killen GF, Knols BGJ and Smallegange (2008) Evaluation of two counterflow traps for testing
behaviour-mediating compounds for the malaria vector Anopheles gambiae s.s. under semi-field conditions in
Tanzania. Malaria J 7: 230.
Scott TW and Morrison AC (2002) Aedes aegypti density and the risk of dengue-virus transmission. In: Takken W and
Scott TW (eds) Ecological aspects for application of genetically modified mosquitoes. Proceedings of the Frontis
workshop on ecological challenges concerning the use of genetically modified mosquitoes for disease control.
Wageningen, the Netherlands 26-29 June 2002. Wageningen UR Frontis Series, pp 187-206.
Silver JB 2008. Mosquito ecology: field sampling methods. 3rd edition. Springer Co., New York, USA.
Sithprasasna R, Linthicum KJ, Lerdthusnee K and Brewer TG (1997) Use of Geographical information system to study
the epidemiology of dengue haemorrhagic fever in Thailand. Dengue Bull 21: 68-72.
Smith CN, Smith N, Gouck HK, Weidhaas DE, Gilbert IH, Mayer MS, Smittle BJ and Hofbauer A (1970) L-Lactic acid as a
factor in the attraction of Aedes aegypti (Diptera: Culicidae) to human hosts. Ann Entomol Soc Am 7: 99-117.
Solberg VB, Sithiprasasna R and Fansiri T (2007) Efficacy testing of eight unique or standard mosquito surveillance traps/
methods in Thailand mouses. Presentation at the ESA Annual Meeting, San Diego, California, USA, 9-12 Dec 2007.
Suárez Ramírez N and Colás Bonne M (2008) Afectación de la sensibilidad del sistema de vigilancia entomológica
mediante larvitrampas en un área de salud. Medisan 12(4). available at: http://bvs.sld.cu/revistas/san/vol12_4_08/
san11408.htm.
Takken W and Mboera LEG (2000) Effects of chemical stimuli on oviposition of Culex quinquefasciatus (Diptera: Culicidae)
in Tanzania. Proc Exp Appl Entomol 11: 182-187.
Taneja J. and Guerin PM (1997) Ammonia attracts the haematophagus bug Triatoma infestans: behavioural and
neurophysiological data on nymphs. J Comp Physiol 181: 21-34.
Trexler JD, Apperson CS, Gemeno C, Perich MJ, Carlson D and Schal C (2003) Field and laboratory evaluations of potential
oviposition attractants for Aedes albopictus (Diptera: Culicidae). J Am Mosq Control Assoc 19: 228-34.
Tsuda Y, Suwonkerd W, Chawprom S, Prajakwong S and Takagi M. (2006) Different spatial distribution of Aedes aegypti
and Aedes albopictus along an urban-rural gradient and the relating environmental factors examined in three
villages in northern Thailand. J Am Mosq Control Assoc 22: 222-228.
WHO (1999) Strengthening implementation of the global strategy for dengue fever/ dengue haemorrhagic fever
prevention and control. Report of the Informal Consultation. WHO, Geneva, Switzerland.
WHO (2009) Dengue and dengue hemorrhagic fever. Available at: http://www.who.int/csr/disease/dengue/impact/en/.
Williams CR, Long SA, Russel RC and Ritchie SA (2006) Field efficacy of the BG-Sentinel compared with the CDC Backpack
Aspirator and CO2-baited EVS trap for collection of adult Aedes aegypti in Cairns, Queensland, Australia. J Am Mosq
Control Assoc 22: 296- 300.
Williams CR, Long SA, Webb CE, Bitzhenner M, Geier M, Russel RC and Ritchie SA (2007) Aedes aegypti population
sampling using BG-Sentinel traps in north Queensland Australia: statistical considerations for trap deployment
and sampling strategy. J Med Entomol 44: 345-350.
Wilton DP (1985) Preliminary evaluation of a black cylinder suction trap for Aedes aegypti and Culex quinquefasciatus
(Diptera: Culicidae). J Med Entomol 22: 113-114.

398  Olfaction in vector-host interactions

18. Exploitation of olfactory-mediated behaviour
Willem Takken and Bart G.J. Knols

Abstract

Intra- and interspecific communication between arthropods and their blood hosts is to a large
extent mediated by chemical cues. The information provided in this book shows the many and
significant advancements in our knowledge of these interactions, from DNA regulation to natural
interactions of wild arthropod populations and their hosts. Recent information from molecular
genetics, neurobiology, behavioural ecology and chemistry have resulted in understanding how
these organisms are affected by small molecules, mostly produced by the vertebrate host, and
that the insects and ticks respond to blends of odorants rather than single cues. Whereas CO2
is a universal kairomone for these blood-feeding arthropods, other host-derived odorants are
often required to attain behavioural responses similar to those observed with natural hosts. The
collective data described in this book provide opportunities and challenges for integration of
olfactory tools as novel interventions for the control of these arthropods, many of which are vectors
of infectious diseases or nuisance insects. Examples of such a strategy are provided, notably with
mosquitoes, as these are among the group of arthropods for which alternative control strategies
are most urgently required. Finally, the gaps in research that need to be addressed are discussed,
focusing on identification and formulation of semiochemicals, sampling technologies and the
integration of this in existing control programmes.

Keywords: semiochemicals, blood-feeding arthropods, chemical communication, challenge,

vector control, attractant, repellent

Introduction

Chemical communication is an important mechanism of intra- and interspecific communication

of arthropods. This book, which reviews the role of olfaction in vector-host interactions, shows
progress made in our understanding of this process in blood feeding insects and ticks, from cellular
level to population. Whereas the role of olfactory cues in host seeking of disease vectors was
known for many decades, the regulatory mechanisms of this behaviour remained largely hidden
until recent advancements in the fields of molecular genetics, analytical chemistry and visual
technology. These technologies allowed for the discovery of olfactory genes and examination
of the genetic regulation of gene expression in several disease vectors. In sensory physiology
advances in technology and chemistry have made detailed studies possible on the activation
of olfactory processing from the periphery to the insect brain. Advanced knowledge on insect
behaviour led to the development of tools with which flight behaviour and responses to olfactory
cues can be accurately assessed in laboratory and field. This book provides numerous examples
of these advances over a wide range of insects and ticks. The collected information should be
used for the development of tools that can be exploited for the control of vector-borne disease
transmission and reduction of biting nuisance by blood-feeding arthropods, and several examples
of these will be discussed in this concluding chapter.

Molecular regulation of olfactory behaviour

The recent publication of the genomes of several disease vectors (2002: Anopheles gambiae Giles
sensu stricto; 2006: Culex pipiens quinquefasciatus Say; 2007: Aedes aegypti (L.); 2007: Pediculus

Olfaction in vector-host interactions 399

Willem Takken and Bart G.J. Knols

humanus Linnaeus; 2008: Ixodes scapularis Say (see www.vectorbase.org for all genomes) creates
unprecedented opportunities for studying the genetic and cellular regulation of insect behaviour.
For example, mechanisms that affect the expression of olfactory genes in mosquitoes can now be
studied in detail, allowing for a comprehensive understanding of the insect response to selected
odorants (Touhara and Vosshall 2009, Chapter 2 in this volume). In addition, recent data on odorant
binding proteins (OBPs) demonstrate the significance of these molecules in transporting volatile
chemicals through the sensillum fluid to the receptor membrane, but also how gene silencing,
using RNAi technologies, can fundamentally alter these processes, suggesting behavioural
changes that went hitherto unexplained (Pelletier and Leal 2009, Sengul and Tu 2008, Zhou et al.
2008). The challenge is how this information can be maximally exploited for vector-borne disease
control. For example, it has been suggested that silencing of selected olfactory genes might lead
to mosquitoes that no longer recognise human odours, or not any vertebrate indeed. Exploitation
of this knowledge would require the stable transformation of genetic traits to produce lines of
mosquitoes with altered host-seeking behaviour and/or host-seeking strategies, so that fewer
bites would occur on humans. We can then envisage the release of such mosquitoes in natural
ecosystems to replace wild sibling populations as a contribution to disease control. In recent years,
the use of such genetically modified vectors has been debated intensively as the socio-ethical and
legal aspects of this technology remain unresolved (Benedict et al. 2008, Knols and Louis 2004,
Scott et al. 2002, 2008). The World Health Organization, through its Tropical Disease Research
(TDR) programme, discussed this topic at length during a meeting in May 2009, and it is expected
that the organisation will provide guidance on the use of this technology in disease-endemic
countries (Mumford et al. 2009). As the focus of genetic modification for vector-borne disease
control appears to be more on the alteration of vector competence then on other biological
traits (Catteruccia et al. 2009), the development of mosquitoes with altered biting behaviour is at
present remote.

Neurophysiology

Recent advances in the neurophysiology of olfaction of disease vectors focused mostly on

mosquitoes (Chapters 3 and 4) although studies on sandflies (Chapter 9) and kissing bugs (Chapter
14) also provided new insights in chemoreception of these insects. Drosophila melanogaster Meigen
serves as a universal model insect for our understanding of processing of olfactory information
in insects. Studies on Ae. aegypti, An. gambiae sensu stricto and Cx. quinquefasciatus demonstrate
many similarities with D. melanogaster in olfactory processing, confirming the validity of using this
species as a model for our understanding of insect olfaction. Whereas olfactory receptor cells can
be broadly tuned to a wide range of chemical cues (Chapter 3, this volume and Carey et al. 2010),
internal processing of these signals, perhaps governed by olfactory binding proteins (OBPs) and
interaction at the neural membrane cause a first selective barrier for the action of semiochemicals,
which is further regulated at the primary olfactory centre in the brain (Chapter 4). Functional
dimorphism has been described for insects from different orders, but for mosquitoes this still
has to be described. If so, this might explain the observed differences in olfactory responses of
mosquitoes with respect to mating, host seeking and oviposition. It is interesting that odours, and
specifically odour quality, ultimately are encoded by spatial patterns of activation in defined sets
of glomeruli (Chapter 4). Such information needs to be linked to behavioural studies in which the
behavioural response of insects to qualitatively different blends of odorants are being examined
(Bernier et al. 2007, Jawara et al. 2009, Okumu et al. 2010a, Smallegange et al. 2009, Syed and Leal
2009, Williams et al. 2006). Whereas in recent years significant advances have been made in our
understanding of the olfactory processing of odour cues of vectors, there is a lack of congruence
between these results and those of the behavioural essays reported elsewhere in this volume

400  Olfaction in vector-host interactions

 18. Exploitation of olfactory-mediated behaviour

(Chapters 7 and 8). For rapid assessment of the quality of candidate odorant cues as behaviourally
relevant, we propose the development of a model that would predict the behavioural outcome of
the signal-transduction cascade of semiochemicals.

Flight behaviour

It is now generally accepted that blood-feeding insects respond to host odours, which as
kairomones serve to guide the insects to the host. In this volume mosquitoes were selected as
an example for reviewing upwind flight behaviour (Chapter 6). During the flight, mosquitoes
use both chemical and visual cues, while temperature is detected only within a short range
from the host. The effect of carbon dioxide during upwind flight has become better understood
in recent years but the overall effect of blends of kairomones on in-flight behaviour, however,
remains to be clarified. For example, CO2 is sensed from a long distance, but we do not know over
which distance other chemical cues are being detected and cause a behavioural effect. This is
puzzling, as the principles of odour dispersal predict that packages of odour move downwind in a
concentration similar to that at the source (Murlis et al. 2000). In a wind tunnel, Dekker et al. (2005)
demonstrated that the yellow fever mosquito Ae. aegypti responded to homogeneous plumes of
host odour and to turbulent plumes of CO2. A similar phenomenon was shown to occur in the
malaria mosquito An. gambiae s.s. (Spitzen et al. 2008). Recent semi-field and field studies showed
that host odours alone attracted few mosquitoes to odour-baited traps, and that addition of CO2
increased the catch significantly (Qiu et al. 2007, Schmied et al. 2008), providing support for these
laboratory findings. These results suggest that mosquitoes sense a host from a distance with CO2
as a principle component and that close-range host orientation is guided by other host odours,
principally derived from the skin. It remains a challenge to elucidate whether such host odours
induce upwind flight behaviour from a distance (Chapter 6). Tools such as windtunnels provided
with imaging recorders, capable to record flights at 6,000 images per second (Anonymous, 2010)
and equipped with on-line analysis software, as well as the use of electric nets for recording
host-seeking flights in a natural environment (Chapters 6 and 12) suggest that details or odour-
modulated flight behaviour of insects will rapidly advance. This, in turn, provides opportunities
to better understand the role of specific host cues in the host-seeking strategy of disease vectors.
Indoor host-seeking behaviour has remained largely unexplored, yet given the tendency of many
vectors to bite indoors (endophagy) this deserves more focus. Exactly how host-seeking insects
behave in a ‘no-wind’ environment could serve as a means to interfere with odour-modulated
behaviour indoors and optimise methods to divert, trap, or repel insects close to hosts.

Behavioural assays

The study of odour-modulated behaviour of flying insects is, by nature, complex as the target
organism moves fast through the natural environment such that direct observations are highly
unreliable. This complexity is even enhanced as we consider that odour plumes are highly motile
and affected by wind (Chapter 6, this volume). A consensus seems to have been reached among
scientists that this behaviour should be studied at different levels: (a) under controlled conditions
in the laboratory, (b) in semi-field enclosures, where the insects can be observed in a simulated
near-natural environment and (c) in the field. Examples of these studies are reported in this
volume, with mosquitoes, sandflies, stable flies, blackflies, tsetse flies, kissing bugs and ticks as
target arthropods. The latter two groups require different observation tools as they move by
walking and, hence, are more easily observable by the researcher. In the cascade of behavioural
observations, the order of laboratory – semi field – field is desirable, but as the use of semi field
systems is of recent origin (Ferguson et al. 2008, Knols et al. 2002), many studies move directly from

Olfaction in vector-host interactions 401

Willem Takken and Bart G.J. Knols

the laboratory to the field. Video recording can be used for detailed studies of in-flight behaviour
(Beeuwkes et al. 2008, Cooperband and Cardé 2006, Dekker et al. 2005, Gibson and Brady 1985,
1988), and may reveal behavioural manoeuvres associated with particular olfactory cues (Dekker
et al. 2005, Spitzen et al. 2008) and advanced software is available that enables detailed analysis
of flight paths according to essential behavioural parameters (e.g. see www.noldus.com).

Several standard assays for laboratory studies have emerged, which provide robust tools for
examining the behavioural response of blood-feeding insects and ticks to sensory cues such as
odour, heat, humidity and visual objects. For the study of mosquito behaviour, a wind tunnel or
olfactometer is widely used, and using these devices, semiochemicals that affect the behaviour
of mosquitoes have been identified. The most recent studies report odour blends of chemicals
identified from the natural volatiles of blood hosts (Bernier et al. 2007, Okumu et al. 2010a,
Chapter 7 in this volume). Although this work initially focused on tsetse flies, it is encouraging to
note that these studies are rapidly expanding to other blood-feeding species, leading to several
promising odour blends with which these insects can be manipulated. Laboratory studies now
produce blends of odorants to which the insects are attracted or by which they are repelled,
depending on the odour composition. The challenge now is to provide proof that these blends
are also effective in the field, as was done for tsetse flies. Only then can these studies be justified
and perhaps improved to provide a large range of odorant cues with which nuisance insects and
disease vectors can be effectively controlled. The recent study by Okumu et al. (2010a) in Tanzania
and field studies in Brazil (Chapter 17, this volume) as well as a field study in Scotland (Logan et
al. 2009) demonstrate that such cues are effective against different mosquito and biting midge
species on several continents. Ongoing studies are expected to lead to a rapid expansion of the
available odour cues affecting these insects.

Semiochemicals

For many years, odorant baits for mosquitoes consisted of carbon dioxide (CO2) or live hosts
(Takken and Knols 1999). Detailed research on tsetse flies and screwworm flies, however, resulted
in the identification of chemical cues to which these insects respond by attraction or aversion
(Cork and Hall 2007, Vale 1993, Chapter 12 in this volume). Inspired by these developments,
studies on odour baits for other vectors were actively initiated, and in the preceding chapters of
this volume the state-of-the art of these studies are reviewed. Carbon dioxide acts as a universal
kairomone (activator and attractant) for blood-feeding arthropods (Chapter 5, this volume), which
suggests an evolutionary-safe adaptation to vector-host interactions, as all vertebrates exhale this
compound. Any response to CO2 is expected to lead the arthropod to a live host. However, CO2 is a
not a kairomone that arthropods can exploit to satisfy their intrinsic preference for biting specific
host species. For several mosquito species, which express a strong anthropophilic behaviour,
studies focused on the chemical identification of human odorants and on selection of those
compounds that cause a behavioural response in the mosquito (Bernier et al. 2000, 2002, Logan
et al. 2009). This has led to the publication of blends of odorant chemicals that cause attraction
of Ae. aegypti (Bernier et al. 2003) and An. gambiae (Okumu et al. 2010a, Smallegange et al. 2009).
Both species are important vectors of arboviruses such as dengue, yellow fever, and Chikungunya
(Ae. aegypti) and malaria (An. gambiae), and the availability of attractive blends opens the way
for improved surveillance of these vectors, as well as for the development of removal trapping
systems. In Chapter 17 (this volume) a practical application for the surveillance of Ae. aegypti
in Brazil is described, which is the immediate result of the detailed chemical, physiological and
behavioural studies reviewed in other chapters of this book. Compared to mosquitoes, the
identification of kairomones for other vectors has been less intense, but recent data on biting

402  Olfaction in vector-host interactions

 18. Exploitation of olfactory-mediated behaviour

midges (Chapter 10, this volume), kissing bugs (Chapter 14, this volume) and sandflies (Chapter
9, this volume) suggest that these insects can soon also be sampled using attractive odour baits.
It has been suggested that only baits that are as attractive as natural host odours can be applied
for the successful manipulation of vectors, such as for example the control of tsetse populations
in southern Africa (Hargrove 2005), and that baits that are less attractive than natural hosts are not
relevant for vector control. We argue that a constant removal of fractions of vector populations
over time will cause a significant reduction in density irrespective of the degree of attractiveness,
although the speed of population reduction will be much higher when the baits are powerful. This
of course on condition that the target vector population does not develop a change in behaviour
by selection. We are not aware of publications of behavioural resistance as a result of olfactory
manipulation, and suggest that the probability of this to occur is small as the insects can usually
feed on alternative hosts (Killeen et al. 2001), which express a different odour profile.

Pathogen-induced effects on olfactory behaviour

The few reports on manipulation of vector behaviour by pathogens remains intriguing, as it is

evolutionary advantageous for these microorganisms to control their transmission from host
to vector and vice versa. Although there is much evidence for the manipulation of feeding
behaviour by pathogens, the olfactory evidence of these interactions remains limited to studies
on Plasmodium and Leishmania (Chapter 16, this volume). Recent work on vector manipulation
by viruses in plants demonstrated convincingly that the pathogens affect the composition of
semiochemicals that guide their insect vectors to the plant (Mauck et al. 2010) and match earlier
work on malaria vectors (Lacroix et al. 2005). It is, however, not known whether these effects
apply to a broad range of vector-borne microorganisms and of those where such behavioural
manipulation was found, the nature of the semiochemicals involved is unknown to date. Yet,
detailed knowledge of these interactions would provide insight in specific cues that attract
vectors to their hosts, and can be used for vector manipulation.

Future challenges and prospects

Our knowledge on the role of olfactory cues in host-seeking behaviour of blood-feeding

arthropods has made significant advances in the last decade (this volume). Tools for the detailed
observation and analysis of this behaviour are now being used by a large group of scientists, which
is expected to answer questions that have eluded us this far. The field of research has broadened,
to include molecular genetics, cellular regulation of signal transduction, analytical chemistry,
behavioural ecology and parasite-vector interactions. While these studies answer fundamental
questions about the role of odorants in arthropod behaviour, several challenges need to be
resolved. For example, the sampling of blood-feeding insects relies currently on crude tools such
as traps, knock-down sprays, resting collections or the human biting catch (Chapters 8, 11 and
12, this volume). The rationale for trap design is often unknown, and studies as described by
Cooperband et al. (2006) show the importance of understanding the way insects approach a trap
while responding to host volatiles. All mosquito traps currently marketed operate a strong suction
airflow to draw insects into a collection bag. The disadvantage of such traps is that they require a
power source to operate the fan, which is often lacking in rural areas where such traps need to be
operated. Research into passive trapping systems or even targets should receive more attention.
Recently, for the first time, odour-baited stations for mosquitoes have been field tested (Okumu
et al. 2010b). Although still not practical and in need of a CO2 source provided from a pressurised
gas cylinder, this study marks the first step towards effective use of attractive kairomone blends
for operational use in developing countries. Such systems can be used as a ‘lure and kill’ tactic,

Olfaction in vector-host interactions 403

Willem Takken and Bart G.J. Knols

whereby insects are attracted to an insecticide-treated resting site. Alternatively, insects can be
contaminated with biological control agents such as entomopathogenic fungi, through a ‘lure and
contaminate’ strategy (Lwetoijera et al., in press).

An important goal of research on olfactory behaviour of arthropods is to understand how odour

cues can be used operationally. Many blood-feeding arthropods are disease vectors, and hence
are targeted for control. The work on tsetse flies has convincingly shown that odour cues can be
used to control disease vectors (Torr et al. 2005), but currently there are no examples of other
disease vectors that can be controlled in this way.

Much work reviewed in this book concerns the role of kairomones in vector-host interactions, and
for some vector species effective strategies are available that are used for their control (Chapter
12, this volume). An older, and more widely applied strategy of vector manipulation is the use
of repellents such as N,N-diethyl-3-methylbenzamide (DEET) (Katz et al. 2008). DEET has been in
use for more than 50 years, and is effective against mosquitoes, sand flies and ticks. Because of
its relatively long persistence and high efficacy, it is not only used as protection against nuisance
mosquitoes but also as personal protection against malaria and dengue vectors. Other compounds
like IR3535((R)) and Picaridin have been shown to provide a similar degree of protection as DEET.
Recently, olfactory receptors for DEET were reported, providing exciting prospects for behavioural
manipulation for other candidate chemicals with repellent action (Ditzen et al. 2008, Syed and
Leal 2008). For instance, it is known that at low concentrations 1-octen-3-ol is attractive to
some mosquito species and tsetse flies, but that higher concentrations cause a repellent effect.
A different, but repellent effect was observed with carboxylic acids in anopheline mosquitoes,
which was overcome by the addition of ammonia + lactic acid to the acid mixture (Smallegange
et al. 2005). From these studies it is suggested that semiochemicals can be strategically applied
for behavioural manipulation of disease vectors, opening the way for push-pull strategies as have
been developed for herbivorous insects (Agelopoulos et al. 1999, Khan et al. 2008).

The emergence and spread of infectious diseases is studied using models and for vector-borne
diseases these include the interaction between vectors, parasites and hosts (Anderson and May
1992). The vector-component in these models is, however, surprisingly inaccurate as vector
behaviour is often not taken into account. For example, vectors exhibit different biting behaviour
depending on physiological stage, age and infectious stage. The preceding chapters provide
evidence that behavioural responses to blood hosts can be much affected by these factors, and
therefore should be included in these models. Models can also be used to assess the impact of
behavioural manipulation, for instance by removal trapping or the impact of a push-pull system.
Agent-based models have recently been developed to assess the efficacy of bed net interventions
on a micro scale, and such models can similarly be developed for removal trapping (Gu and Novak
2009). We suggest that field studies on the risk of vector-borne disease incorporate a behavioural
component, so that better use can be made of behavioural manipulation through vector control
or through personal protection.

This book focuses on advances in our knowledge on olfactory mediation of arthropod-host

interactions. To study these interactions in natural situations in the field, sampling systems are
needed to assess the behavioural response to semiochemicals. Although the efficacy of the many
tools used for the collection of arthropods is not discussed here, it is obvious that such tools are
essential in behavioural studies. A growing number of semiochemicals is becoming available for
behavioural manipulation of nuisance insects and disease vectors and proof of principle of their
operational use is becoming available (Okumu et al. 2010b). We suggest that the private pest

404  Olfaction in vector-host interactions

 18. Exploitation of olfactory-mediated behaviour

control sector becomes more closely involved in this work, so that the concept of arthropod
manipulation with semiochemicals can be further developed. One successful example of such a
development is that of the insecticide-treated bed nets, where the private sector became closely
involved in disease control (Lengeler et al. 2007, Noor et al. 2009), allowing for a massive surge
in technological advances using a private/public partnership. The control of tsetse flies using
semiochemicals has received commercial interest, but to date only on a limited scale. On a larger
scale the control of nuisance mosquitoes using semiochemical technology was undertaken
by American Biophysics Corp. The latter company, however, dissolved after only a few years of
successful operation. Although the reasons for the breakup of this company remain elusive, the
results of their activities show that there was a broad private and public interest in this method
of pest control. Now that better and more advanced methods of semiochemical production have
become available, it seems worthwhile to examine the possible integration of this technology into
the control of nuisance insects and disease vectors.

The reviews in this book demonstrate the importance of chemical ecology in vector-host
interactions. The rapid advances in this field made in recent years suggest that we can now
begin to consider applying the collective knowledge gained for manipulation of these vectors,
introducing a novel, sustainable tool for disease control. The example of the successful control of
tsetse flies using semiochemicals remains the best proof that this strategy can be highly effective.
Semiochemicals also provide a solid base for studies of population densities and regulation of
disease vectors, so that more accurate and objective measurements of transmission risk as well
as the impact of vector control can be obtained.

References

Agelopoulos N, Birkett MA, Hick AJ, Hooper AM, Pickett JA, Pow EM, Smart LE, Smiley DWM, Wadhams LJ and Woodcock
CM (1999) Exploiting semiochemicals in insect control. Pest Sci 55: 225-235.
Anderson RM and May RM (1992) Infectious diseases of humans - Dynamics and control. Oxford University Press, Oxford,
UK. 757 pp.
Anonymous (2010) Malaria. Available at: http://intellectualventureslab.com/?page_id=563. Accessed 23 Feb 2010.
Beeuwkes J, Spitzen J, Spoor CW, Van Leeuwen JL and Takken W (2008) 3-D flight behaviour of the malaria mosquito
Anopheles gambiae s.s. inside an odour plume. Proc Exp Appl Entomol 19: 137-146.
Benedict M, D’Abbs P, Dobson S, Gottlieb M, Harrington L, Higgs S, James A, James S, Knols B, Lavery J, O’Neill S, Scott T,
Takken W and Toure Y (2008) Guidance for contained field trials of vector mosquitoes engineered to contain a gene
drive system: recommendations of a scientific working group. Vector Borne Zoonotic Dis 8: 127-166.
Bernier UR, Kline DL, Allan SA and Barnard DR (2007) Laboratory comparison of Aedes aegypti attraction to human odors
and to synthetic human odor compounds and blends. J Am Mosq Control Assoc 23: 288-293.
Bernier UR, Kline DL, Barnard DR, Schreck CE and Yost RA (2000) Analysis of human skin emanations by gas
chromatography/mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for
yellow fever mosquito (Aedes aegypti). Anal Chem A 72: 747-756.
Bernier UR, Kline DL, Posey KH, Booth MM, Yost RA and Barnard DR (2003) Synergistic attraction of Aedes aegypti (L.)
to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl Disulfide. J Med Entomol 40: 653-656.
Bernier UR, Kline DL, Schreck CE, Yost RA and Barnard DR (2002) Chemical analysis of human skin emanations:
comparison of volatiles from humans that differ in attraction of Aedes aegypti (Diptera: Culicidae). J Am Mosq
Contr Ass 18: 186-195.
Carey AF, Wang G, Su CY, Zwiebel LJ and Carlson JR (2010) Odorant reception in the malaria mosquito Anopheles
gambiae. Nature (in press). Doi:10.1038/nature08834.
Catteruccia F, Crisanti A and Wimmer EA (2009) Transgenic technologies to induce sterility. Malaria J 8 (Suppl 2): S7.

Olfaction in vector-host interactions 405

Willem Takken and Bart G.J. Knols

Cooperband MF and Cardé RT (2006) Orientation of Culex mosquitoes to carbon dioxide-baited traps: flight manoeuvres
and trapping efficiency. Med Vet Entomol 20: 11-26.
Cork A and Hall MJR (2007) Development of an odour-baited target for female New World screwworm, Cochliomyia
hominivorax: studies with host baits and synthetic wound fluids. Med Vet Entomol 21: 85-92.
Dekker T, Geier M and Cardé RT (2005) Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human
skin odours. J Exp Biol 208: 2963-2972.
Ditzen M, Pellegrino M and Vosshall LB (2008) Insect odorant receptors are molecular targets of the insect repellent
DEET. Science 319: 1838-1842.
Ferguson HM, Ng’habi KR, Walder T, Kadungula D, Moore SJ, Lyimo I, Russell TL, Urassa H, Mshinda H, Killeen GF and
Knols BG (2008) Establishment of a large semi-field system for experimental study of African malaria vector ecology
and control in Tanzania. Malar J 7: 158.
Gibson G and Brady J (1985) ‘Anemotactic’ flight paths of tsetse flies in relation to host odour: a preliminary video study
in nature of the response to loss of odour. Physiol Entomol 10: 395-406.
Gibson G and Brady J (1988) Flight behaviour of tsetse flies in host odour plumes: the initial response to leaving or
entering odour. Physiol Entomol 13: 29-42.
Gu W and Novak RJ (2009) Predicting the impact of insecticide-treated bed nets on malaria transmission: The devil is
in the detail. Malaria J 8: 256.
Hargrove JW (2005) Extinction probabilities and times to extinction for populations of tsetse flies Glossina spp. (Diptera:
Glossinidae) subjected to various control measures. Bull Entomol Res 95: 13-21.
Jawara M, Smallegange RC, Jeffries D, Nwakanma DC, Awolola TS, Knols BGJ, Takken W and Conway DJ (2009) Optimizing
odor-baited trap methods for collecting mosquitoes during the malaria season in The Gambia. PLoS ONE 4: e8167.
Katz TM, Miller JH and Hebert AA (2008) Insect repellents: historical perspectives and new developments. J Am Acad
Dermatol 58: 865-871.
Khan ZR, Midega CAO, Amudavi DM, Hassanali A and Pickett JA (2008) On-farm evaluation of the ‘push-pull’ technology
for the control of stemborers and striga weed on maize in western Kenya. Field Crops Res 106: 224-233.
Killeen GF, McKenzie FE, Foy BD, Bogh C and Beier JC (2001) The availability of potential hosts as a determinant of
feeding behaviours and malaria transmission by African mosquito populations. Trans Roy Soc Trop Med Hygiene
95: 469-476.
Knols BG, Njiru BN, Mathenge EM, Mukabana WR, Beier JC and Killeen GF (2002) MalariaSphere: A greenhouse-enclosed
simulation of a natural Anopheles gambiae (Diptera: Culicidae) ecosystem in western Kenya. Malar J 1: 19.
Knols BGJ and Louis C (2004) Bridging laboratory and field research for genetic control of disease vectors. In: Knols BGJ
and Louis C (eds) Proceedings of the joint WHO/TDR, NIAID, IAEA and Frontis Workshop, Nairobi, Kenya 14-16 July
2004. Wageningen UR Frontis series Vol. 11. Springer Science + Business Media, Berlin, pp 225.
Lacroix R, Mukabana WR, Gouagna LC and Koella JC (2005) Malaria infection increases attractiveness of humans to
mosquitoes. PLoS Biol 3: e298.
Lengeler C, Grabowsky M, McGuire D and Desavigny D (2007) Quick wins versus sustainability: options for the upscaling
of insecticide-treated nets. Am J Trop Med Hygiene 77 Suppl S: 222-226.
Logan JG, Seal NJ, Cook JI, Stanczyk NM, Birkett MA, Clark SJ, Gezan SA, Wadhams LJ, Pickett JA and Mordue J (2009)
Identification of human-derived volatile chemicals that interfere with attraction of the Scottish biting midge and
their potential use as repellents. J Med Entomol 46: 208-219.
Lwetoijera DW, Sumaye RD, Madumla EP, Kavishe DR, Mnyone LL, Russell TL and Okumu FO (2010) An extra-domiciliary
method for delivering entomopathogenic fungi, Metharizium anisopliae IP 46 against malaria vectors, Anopheles
arabiensis. Parasites and Vectors (in press).
Mauck KE, De Moraes CM and Mescher MC (2010) Deceptive chemical signals induced by a plant virus attract insect
vectors to inferior hosts. PNAS 107: 3600-3605.
Mumford J, Quinlan MM, Beech C, Alphey L, Bayard V, Capurro ML, Kittayapong P, Knight JD, Marrelli MT, Ombongi K,
Ramsey JM and Reuben R (2009) MosqGuide: A project to develop best practice guidance for the deployment
of innovative genetic vector control strategies for malaria and dengue. Asia-Pacific J Mol Biol Biotech 17: 93-95.

406  Olfaction in vector-host interactions

 18. Exploitation of olfactory-mediated behaviour

Murlis J, Willis MA and Cardé RT (2000) Spatial and temporal structures of pheromone plumes in fields and forests.
Physiol Entomol 25: 211-222.
Noor AM, Mutheu JJ, Tatem AJ, Hay SI and Snow RW (2009) Insecticide-treated net coverage in Africa: mapping progress
in 2000-07. Lancet 373: 58-67.
Okumu FO, Killeen GF, Ogoma S, Biswaro L, Smallegange RC, Mbeyela E, Titus E, Munk C, Ngonyani H, Takken W, Mshinda
H, Mukabana WR and Moore SJ (2010a) Development and field evaluation of a synthetic mosquito lure that is more
attractive than humans. PLoS ONE 5: e8951.
Okumu FO, Madumla EP, John AN, Dickson W, Lwetoijera DW and Sumaye RD (2010b) Attracting, trapping and killing
disease-transmitting mosquitoes using odor-baited stations -The Ifakara odor-baited stations. Parasites and
Vectors 3: 12.
Pelletier J and Leal WS (2009) Genome analysis and expression patterns of odorant-binding proteins from the southern
house mosquito Culex pipiens quinquefasciatus. PLoS One 4: e6237.
Qiu YT, Smallegange RC, Ter Braak CJF, Spitzen J, Van Loon JJA, Jawara M, Milligan P, Galimard AM, Van Beek TA, Knols BGJ
and Takken W (2007) Attractiveness of MM-X traps baited with human or synthetic odor to mosquitoes (Diptera:
Culicidae) in The Gambia. J Med Entomol 44: 970-983.
Schmied WH, Takken W, Killeen GF, Knols BG and Smallegange RC (2008) Evaluation of two counterflow traps for testing
behaviour-mediating compounds for the malaria vector Anopheles gambiae s.s. under semi-field conditions in
Tanzania. Malar J 7: 230.
Scott TW, Harrington LC, Knols BGJ and Takken W (2008) Applications of mosquito ecology for successful insect
transgenesis-based disease prevention programs. In: Aksoy S (ed) Transgenesis and the management of vector-
borne disease. Advances in experimental medicine and biology. Springer-Verlag, Berlin, Germany, pp 151-168.
Scott TW, Takken W, Knols BGJ and Boëte C (2002) The ecology of genetically modified mosquitoes. Science 298: 117-
119.
Sengul MS and Tu Z (2008) Characterization and expression of the odorant-binding protein 7 gene in Anopheles
stephensi and comparative analysis among five mosquito species. Insect Mol Biol.
Smallegange RC, Qiu YT, Bukovinszkine-Kiss G, Van Loon JJA and Takken W (2009) The effect of aliphatic carboxylic acids
on olfaction-based host-seeking of the malaria mosquito Anopheles gambiae sensu stricto. J Chem Ecol 35: 933-943.
Smallegange RC, Qiu YT, Van Loon JJ and Takken W (2005) Synergism between ammonia, lactic acid and carboxylic acids
as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto (Diptera:
Culicidae). Chem Senses 30: 145-152.
Spitzen J, Smallegange RC and Takken W (2008) Effect of human odours and positioning of CO2 release point on trap
catches of the malaria mosquito Anopheles gambiae sensu stricto in an olfactometer. Physiol Entomol 33: 116-122.
Syed Z and Leal WS (2008) Mosquitoes smell and avoid the insect repellent DEET. PNAS 105: 13598-13603.
Syed Z and Leal WS (2009) Acute olfactory response of Culex mosquitoes to a human- and bird-derived attractant.
PNAS 106: 18803-18808.
Takken W and Knols BGJ (1999) Odor-mediated behavior of afrotropical malaria mosquitoes. Ann Rev Entomol 44:
131-157.
Torr SJ, Hargrove JW and Vale GA (2005) Towards a rational policy for dealing with tsetse. Trends Parasitol 21: 537-541.
Touhara K and Vosshall LB (2009) Sensing odorants and pheromones with chemosensory receptors. Ann Rev Physiol
71: 307-332.
Vale GA (1993) Development of baits for tsetse flies (Diptera: Glossinidae) in Zimbabwe. J Med Entomol 30: 831-842.
Williams CR, Bergbauer R, Geier M, Kline DL, Bernier UR, Russell RC and Ritchie SA (2006) Laboratory and field assessment
of some kairomone blends for host-seeking Aedes aegypti. J Am Mosq Control Assoc 22: 641-647.
Zhou JJ, He XL, Pickett JA and Field LM (2008) Identification of odorant-binding proteins of the yellow fever mosquito
Aedes aegypti: genome annotation and comparative analyses. Insect Mol Biol 17: 147-163.

Olfaction in vector-host interactions 407



About the editors

Willem Takken (1951) is Professor in Medical and Veterinary
Entomology at Wageningen University, the Netherlands. He
studied in Wageningen and obtained his PhD degree in 1980 based
on research on the biology and feeding behaviour of tsetse flies.
He worked in several African countries on the control of animal
trypanosomiasis. Upon his return to Europe, he was appointed as
lecturer at the Laboratory of Entomology in Wageningen, where he
introduced Medical and Veterinary Entomology to the Wageningen
academic society. His work involved mosquito-host interactions,
in particular the host-seeking behaviour of malaria mosquitoes.
He later expanded this work to include field research in Tanzania,
Kenya and other tropical countries. Today, his work involves the
ecology of mosquitoes, biological control of mosquitoes and the impact of environmental change
on malaria vectors. In Europe, Willem studies the ecology of Lyme disease vectors and vectors of
other, emerging infectious diseases. He emphasises collaboration with other institutions, and
has an extensive network of national and international collaborators. Willem is a member of
the Council of the International Congress of Entomology and serves on several editorial boards
and advisory committees. He has (co)authored more than 160 scientific publications and book
chapters.

Bart G.J. Knols (1965) is Visiting Scientist at the Division of

Infectious Diseases, Tropical Medicine & AIDS of the Academic
Medical Centre in Amsterdam, the Netherlands and holds an
Honorary Professorship in medical entomology at the University
of Witwatersrand, South Africa. He studied biology in Wageningen
where he also obtained his PhD degree in Medical Entomology in
1996 based on studies on odour-mediated host seeking behaviour
of the African malaria vector Anopheles gambiae. He has lived and
worked in East and Southern Africa for eleven years. In Zambia
he was involved in large-scale tsetse fly control campaigns using
odour-baited targets. In Tanzania he studied the behavioural
ecology of malaria mosquitoes for several years, before working
at the International Centre of Insect Physiology and Ecology (ICIPE) in Kenya from 1997-2002.
There he led a research team of thirty students and scientists on the shores of Lake Victoria (at
Mbita Point). Bart worked for the United Nations from 2003-2006 (IAEA in Vienna) to develop
genetic control strategies for mosquitoes, and was Assistant Professor at Wageningen University
until 2009. His career is marked by the discovery of Limburger cheese as a surrogate attractant
for An. gambiae (Ig Nobel Prize 2006), besides his pioneering work in semi-field systems in Africa
and the discovery of a fungal entomopathogen for adult malaria vector control. Bart obtained his
MBA degree (Open University, UK) in 2006 and is a member of the Royal Dutch Academy of Arts
and Sciences (KNAW-DJA). He has (co)authored more than 130 scientific publications and book
chapters. He currently directs K&S Consulting, a company engaged in the global dissemination
of medical information (www.malariaworld.org), organisation building, and consultancies in
infectious disease control.

Willem and Bart were the joint recipients of the Eijkman medal in 2007, which is the highest award
in the field of tropical medicine and international health in the Netherlands.

Olfaction in vector-host interactions 409

Contributors
Sandra A. Allan (USA)
Center for Medical, Agricultural and Veterinary Entomology, ARS/USDA, Gainesville, FL 32608, USA.
Email: [email protected]

Jonathan D. Bohbot (USA)

Plant Sciences, USDA-ARS, Room 024, 10300 Baltimore avenue, Bldg 007 Barc-West, Beltsville, MD 20705,
USA. Email: [email protected]

Daniel P. Bray (United Kingdom)

Centre for Applied Entomology and Parasitology, Institute for Science and Technology in Medicine,
Keele University, Keele, Staffordshire, ST5 5BG, UK. Email: [email protected]

Ring T. Cardé (USA)

Department of Entomology, University of California, Riverside, CA 92521, USA. Email:
[email protected]

James I. Cook (United Kingdom)

Rothamsted Research, Harpenden, Hertfordshire, AL5 2JQ, UK. Email: [email protected]

Alvaro E. Eiras (Brazil)

Lab. Ecologia Química de Insetos Vetores, Depto. Parasitologia,Instituto de Ciências Biológicas. Univ.
Federal de Minas Gerais, Av. Presidente Antônio Carlos 6627, Pampulha, P.O. Box 486, 31270-901 Belo
Horizonte, MG, Brazil. Email: [email protected]

Martin Geier (Germany)

Institut für Zoologie, Universität Regensburg, 93053 Regensburg, Germany.
Biogents AG, Bruderwöhrdstr 15b, 93053 Regensburg, Germany. Email:
[email protected]

Gabriella Gibson (United Kingdom)

Natural Resources Institute, University of Greenwich at Medway, Chatham, Kent ME4 4TB, UK.
School of Life Sciences, University of Sussex, Falmer, Brighton, BN1 9 QG, UK. Email: [email protected]

Alan J. Grant (USA)

Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, USA. Email: [email protected]

Pablo G. Guerenstein (Argentina)

Centro de Investigaciones Científicas y Transferencia de Tecnología a la Producción, CONICET, Matteri 49
(esq. España), 3105 Diamante, Entre Ríos, Argentina. Email: [email protected]

James G.C. Hamilton (United Kingdom)

Centre for Applied Entomology and Parasitology, Institute for Science and Technology in Medicine,
Keele University, Keele, Staffordshire, ST5 5BG, UK. Email: [email protected]

Bill S. Hansson (Germany)

Max Planck Institute for Chemical Ecology, Department of Evolutionary Neuroethology, Hans-Knoell-
Strasse 8, D-07745 Jena, Germany. Email: [email protected]

Olfaction in vector-host interactions 411

Contributors

Sharon R. Hill (Sweden)

SLU, Division of Chemical Ecology, Department of Plant Protection Biology, P.O. Box 102, Sundsvägen
14, 230 53 Alnarp, Sweden. Email: [email protected]

Hilary Hurd (United Kingdom)

Centre for Applied Entomology and Parasitology, School of Life Sciences, Keele University, Keele,
Staffordshire, ST5 5BG, UK. Email: [email protected]

Rickard Ignell (Sweden)

SLU, Division of Chemical Ecology, Department of Plant Protection Biology, P.O. Box 102, Sundsvägen
14, 230 53 Alnarp, Sweden. Email: [email protected]

Owen Jones (United Kingdom)

AgriSense BCS Ltd, Treforest Industrial Estate Pontypridd, South Wales CF37 5SU, UK. Email:
[email protected]

Dan L. Kline (USA)

USDA/ARS, Center for Medical, Agricultural and Veterinary Entomology, P.O. Box 14565, 1600/1700 SW
23rd Dr., Gainesville, FL 32608, USA. Email: [email protected]

Bart G.J. Knols (the Netherlands)

Division of Infectious Diseases, Tropical Medicine & AIDS, Academic Medical Centre, Amsterdam, the
Netherlands.
K&S Consulting, Kalkestraat 20, 6669 CP Dodewaard, the Netherlands. Email: [email protected]

Claudio R. Lazzari (France)

Institut de Recherche sur la Biologie de l’Insecte, UMR 6035 CNRS, Université François Rabelais, Institut
de Recherche sur la Biologie de l’Insecte, UMR 6035 CNRS, Université François Rabelais, Av. Monge, Parc
Grandmont, 37200 Tours, France. Email: [email protected]

James G. Logan (United Kingdom)

Rothamsted Research, Harpenden, Hertfordshire, AL5 2JQ, UK. Email: [email protected]

Tan Lu (USA)
Vanderbilt University/Vanderbilt University Medical Center, Institutes for Chemical Biology & Global
Health, Center for Molecular Neuroscience, VU Station B 351634, Nashville, TN 37235-3582, USA. Email:
[email protected]

A. Jennifer Mordue (Luntz) (United Kingdom)

Aberdeen University, School of Biological Sciences, Zoology Building, Tillydrone Avenue, Aberdeen
AB24 2TZ, UK. Email: [email protected]

Wolfgang R. Mukabana (Kenya)

School of Biological Sciences, University of Nairobi, P.O. Box 30197, 00100 GPO, Nairobi, Kenya and
International Centre of Insect Physiology and Ecology, P.O. Box 30772, 00100, GPO, Nairobi, Kenya. Email:
[email protected]

412  Olfaction in vector-host interactions

 Contributors

Robert J. O’Connell (USA)

Brudnick Neuropsychiatric Research Institute, University of Massachusetts Medical School, 303 Belmont
Street, Worcester, MA 01604, USA. Email: [email protected]

Evelyn A. Olanga (Kenya)

International Centre of Insect Physiology and Ecology, P.O. Box 30772, 00100, GPO, Nairobi, Kenya and
Department of Zoological Sciences, Kenyatta University, P.O. Box 43844, 00100 GPO, Nairobi, Kenya.
Email: [email protected]

Yu Tong Qiu (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, PO Box 8031, 6700 EH
Wageningen, the Netherlands. Email: [email protected]

Andreas Rose (Germany)

Institut für Zoologie, Universität Regensburg, 93053 Regensburg, Germany.
Biogents AG, Bruderwöhrdstr 15b, 93053 Regensburg, Germany. Email: [email protected]

Steve Schofield (Canada)

Force Health Protection, Communicable Disease Control Program, CF H Svcs Gp HQ Ottawa, Department
of National Defence, 1745 Alta Vista Drive, Ottawa, Ontario K1A 0K6, Canada. Email:
[email protected]

Meryem S. Sengul (Sweden)

SLU, Division of Chemical Ecology, Department of Plant Protection Biology, P.O. Box 102, Sundsvägen
14, 230 53 Alnarp, Sweden. Email: [email protected]

Renate C. Smallegange (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, P.O. Box 8031, 6700 EH
Wageningen, the Netherlands. Email: [email protected]

Philippe Solano (Burkina Faso)

Institut de Recherche pour le Développement (IRD), UMR 177 IRD-CIRAD/ Centre International de
Recherche-Développement sur l’Elevage en zone Subhumide(CIRDES), Bobo Dioulasso, Burkina Faso.
Email: [email protected]

James F. Sutcliffe (Canada)

Department of Biology, Trent University, 1600 West Bank Drive, Peterborough, Ontario, K9J 2B6, Canada.
Email: [email protected]

Willem Takken (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, P.O. Box 8031, 6700 EH
Wageningen, the Netherlands. Email: [email protected]

Steve J.Torr (United Kingdom)

Natural Resources Institute (NRI), University of Greenwich, Natural Resources Institute, University of
Greenwich at Medway, Central Avenue, Chatham Maritime, Chatham, Kent ME4, 4TB, UK. Email:
[email protected]

Olfaction in vector-host interactions 413

Contributors

Joop J.A. van Loon (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, PO Box 8031, 6700 EH
Wageningen, the Netherlands. Email: [email protected]

Richard D. Ward (United Kingdom)

Centre for Applied Entomology and Parasitology, Institute for Science and Technology in Medicine,
Keele University, Keele, Staffordshire, ST5 5BG, UK. Email: [email protected]

Laurence J. Zwiebel (USA)

Vanderbilt University/Vanderbilt University Medical Center, Institutes for Chemical Biology & Global
Health, Center for Molecular Neuroscience, VU Station B 351634, Nashville, TN 37235-3582, USA. Email:
[email protected]

414  Olfaction in vector-host interactions

Reviewers
Prof. J.R. Carlson (USA)
Molecular, Cellular, and Developmental Biology, KBT 1132, Yale University, P.O. Box 208103, New Haven,
CT 06520-8103, USA

Dr. L. Cruz López (Mexico)

Departamento de Entomología, El Colegio de la Frontera Sur Unidad Tapachula, Tapachula, Mexico

Dr. H.M. Ferguson (United Kingdom)

Division of Immunity and Infection & Division of Environmental and Evolutionary Biology, Glasgow
Biomedical Research Centre, University of Glasgow, 120 University Place, Glasgow G12 8TA, UK

Prof. B.S. Hansson (Germany)

Max Planck Institute for Chemical Ecology, Hans-Knöll-Straße 8, D-07745 Jena, Germany

Dr. J.W. Hargrove (South Africa)

SACEMA Director, DST/NRF Centre of Excellence in Epidemiological Modelling and Analysis, c/o STIAS,
19 Jonkershoekweg, Stellenbosch 7600, South Africa

Prof. J.G. Hildebrand (USA)

ARL Division of Neurobiology, The University of Arizona, PO Box 210077, Tucson, AZ 85721-0077, USA

Dr. B.G.J. Knols (the Netherlands)

Division of Infectious Diseases, Tropical Medicine & AIDS, Academic Medical Centre, Amsterdam, the
Netherlands

Prof. M.J. Klowden (USA)

University of Idaho Campus, Ag Science 233, Moscow, ID 83844-2339, USA

Prof. B.A. Mullens (USA)

Department of Entomology, University of California, Riverside, Ca 92506, USA

Prof. D.E. Sonenshine (USA)

Old Dominion University, Department of Biological Sciences, Norfolk, VA 23529 -0266, USA

Prof. Dr. W. Takken (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, PO Box 8031, 6700 EH
Wageningen, the Netherlands

Prof. S.J. Torr (United Kingdom)

Natural Resources Institute, University of Greenwich at Medway, Central Avenue, Chatham Maritime,
Chatham, Kent ME4, 4TB, UK

Dr. J.J.A. van Loon (the Netherlands)

Laboratory of Entomology, Wageningen University and Research Centre, PO Box 8031, 6700 EH
Wageningen, the Netherlands

Olfaction in vector-host interactions 415

 Index

Keyword index –– aegypti genome 68

–– albopictus 156, 366, 370, 371, 372, 373,
A 376, 383
AAA pheromone – See: attraction- –– caspius 189
aggregation-attachment pheromone –– dalzieli 189
AaOR7 27 –– hirsutus 189
ABC PRO suction trap 233 –– polynesiensis 355
acaricide 341 –– taeniorhynchus 94
acaricide tags 341 –– triseriatus 47, 371
acetic acid 48, 52, 151, 153, 158, 161, 163, Aethia cristatella 156, 337
165, 333, 336 Africa 248, 266, 279, 281, 283, 301, 365, 403
–– ethyl lactate 48 African horse sickness 220
acetone 17, 51, 143, 154, 155, 160, 161, 167, age 53
169, 192, 195, 223, 228, 230, 236, 252, agent-based model 404
253, 254, 265, 270, 271, 272, 274, 276, AgGR 28, 29, 53
277, 278, 282, 293, 295, 297, 298, 301, 335 aggregation 69, 224
acetophenone 26, 28 aggregation pheromone 67, 312, 319
acetothiophene 52 AgOR 28, 29, 30, 32
acetylphenone 52 AgOR7 26
acetylpyridine 52 AgOR8 28, 29, 53, 55
acetylthiazole 52 Agrotis segetum 70
across-fibre patterning 54 AHSV 234, 238
action potential 96, 97 aim-then-shoot 125
activation 115, 118, 119, 121, 250, 258, 268, air entrainment 228
271, 294, 295, 301, 309, 315, 328 air pollutant 238
activator 146, 194, 205, 320, 402 Akabane virus 220
active alarm pheromone 316
–– orientation 250 alcohol 51, 53, 163, 167, 191
–– phase 122 aldehyde 46, 163, 167, 168
–– space 132 –– aromatic 330
activity rhythm 115 –– C8 293
adenosine triphosphate 66 aliphatic acid 27
adenotrophic viviparity 265 n-aliphatic aldehyde 318
adenyl cyclase 19 aliphatic carboxylic acid 26, 47, 50, 143, 157,
adenylyl cyclase 66, 68 158, 163, 168, 169, 195
Aedes 25, 32, 39, 44, 127, 130, 132, 193, 238 n-aliphatic carboxylic acid 317
–– aegypti 17, 18, 21, 22, 23, 24, 25, 27, 28, allergic reaction 248
29, 30, 31, 39, 41, 44, 47, 49, 52, 53, 54, 63, allo-aroma-dendr-9-ene 208
65, 67, 69, 70, 72, 73, 74, 75, 91, 92, 93, 94, allomone 147, 165, 166, 272, 273, 328, 338
97, 99, 100, 101, 102, 105, 118, 120, 121, alpha-isomethylionone 230, 232
122, 124, 126, 127, 128, 130, 133, 136, Amazon basin 311
143, 144, 145, 146, 147, 148, 150, 151, Amblyomma
152, 153, 154, 155, 156, 157, 165, 167, –– americanum 328, 330, 334, 335, 336,
168, 169, 228, 231, 351, 352, 354, 356, 337, 340, 342
358, 365, 366, 367, 368, 369, 370, 371, –– cajennense 339
372, 373, 374, 375, 376, 377, 378, 379, –– cohaerens 333
380, 382, 384, 385, 386, 388, 390, 391, –– hebraeum 328, 332, 333, 334, 335, 336,
399, 400, 401, 402 338, 339, 341

Olfaction in vector-host interactions 417

Index

–– maculatum 330, 339, 341 –– parensis 182

–– variegatum 327, 328, 330, 331, 332, 333, –– pharoensis 186, 187, 188, 189, 193, 194
334, 335, 336, 338, 339, 340, 341, 342 –– punctulatus 352
ammonia (NH3) 17, 47, 48, 49, 50, 51, 52, 53, –– quadriannulatus 51, 128, 129, 130, 132,
54, 55, 91, 109, 130, 143, 151, 152, 155, 162, 182, 184, 187, 189, 192, 194
157, 158, 159, 160, 161, 167, 168, 187, –– quadriannulatus species A 181
190, 191, 193, 194, 228, 237, 313, 316, –– quadriannulatus species B 181
317, 318, 332, 335, 379, 381, 404 –– rivulorum 182
ammonium bicarbonate 237 –– rufipes 189
amyl acetate 318 –– sergenti 187, 189, 194
anaphylactic shock 248 –– squamosus 187, 189
anaplasmosis 327 –– stephensi 41, 43, 50, 93, 94, 118, 356
anautogenous 104, 116 –– vaneedeni 182
anemotactic response 280 –– ziemanni 44, 187, 194
anemotaxis 116, 123, 126, 232 Antarctica 219
Angola 183, 185, 186, 274 antenna 21, 23, 24, 27, 30, 31, 39, 40, 41, 42,
angular velocity 335 44, 45, 64, 65, 67, 74, 92, 93, 94, 96, 125,
animal extract 223 126, 146, 206, 229, 230, 312
Anopheles 25, 32, 39, 127, 130, 181, 197 antennal lobe 26, 27, 29, 30, 31, 45, 63, 74,
–– arabiensis 122, 128, 132, 133, 181, 182, 79, 80
183, 184, 187, 188, 189, 191, 192, 193, antennal pedicelus 314
194, 195 Antheraea polyphemus 21, 23, 65
–– aruni 182 anthropophagic 133, 206
–– brucei 182 anthropophilic 43, 122, 128, 132, 133, 144,
–– bwambae 181, 182 145, 151, 182, 183, 185, 190, 191, 250
–– confusus 182 –– behaviour 402
–– coustani 184, 187, 189 anticoagulant 353
–– funestus 129, 181, 182, 185, 187, 188, anti-coagulation factor 358
189, 190, 193, 194, 355 anti-platelet compound 353
–– fuscivenosus 182 anti-platelet factor 358
–– gambiae 17, 18, 20, 21, 22, 23, 24, 25, 26, Apis mellifera 20, 76, 77, 78
27, 28, 29, 30, 31, 39, 41, 44, 45, 47, 48, 49, apocrine sweat 255
51, 52, 53, 54, 63, 67, 69, 70, 74, 94, 106, apyrase 349, 358, 359
107, 118, 120, 121, 122, 126, 128, 129, Arachnidae 92
145, 146, 147, 148, 150, 157, 161, 165, arbovirus 220
167, 168, 181, 182, 183, 184, 188, 189, Argas
190, 191, 192, 193, 194, 195, 260, 318, –– persicus 333, 334, 341
352, 355, 358, 399, 400, 401, 402 –– walkerae 333, 335
–– gambiae genome 68 Argasidae 327
–– gambiae sensu stricto 120, 122, 128, 129, Argentina 311, 368
133, 136, 143, 144, 158, 159, 181, 187, armadillo 310
195, 211 arrestant 146
–– leesoni 182 arrestin 20
–– maculipennis 41, 44 arrestment 115
–– melas 181, 182, 183 –– (assembly) pheromone 338
–– merus 181, 182, 184 –– pheromone 332, 333, 334, 339
–– moucheti 185 asexual parasitaemia 351
–– nili 185, 187, 190 Asia 365

418  Olfaction in vector-host interactions

 Index

assembly pheromone 339, 340 –– response, defensive 353

AtrAedes lure 374, 375, 378 Belgium 220
attractant 135, 143, 144, 145, 146, 150, 151, Benin 183, 274
156, 168, 181, 182, 187, 193, 194, 195, benzaldehyde 163, 164, 165, 230, 232, 331,
196, 197, 203, 204, 205, 239, 252, 320, 402 332, 333
attraction 132, 133, 135, 136, 144, 166, 189, benzene 46, 155
205, 206, 207, 296, 309, 328, 355 benzeneethanol 160, 162
attraction-aggregation-attachment (AAA) benzoic acid 154, 164, 333, 337
pheromone 328, 333, 334, 338, 339, 340, benzothiazole 160, 161, 372
341, 342, 343 benzyl alcohol 70, 72, 163
–– component 331 beta trap 300
attractiveness 187, 188, 208, 210 BG
audio signals 209, 210 –– Lure trap 380, 381, 383
Australia 221, 222, 226, 235, 310, 327, 376, –– Mesh lure trap 237
378, 383, 386 –– Sentinel trap 367, 381, 382, 383, 391
autogenous 104, 116, 248 biased random walk 270
autogeny 218 bioassay 122, 127, 130, 135, 136, 143, 145,
Autospike 95 148, 165, 167, 168, 205, 206, 208, 231,
axillary 331 272, 273, 338, 374
axillary odour 47, 53 biting
axon 42, 45, 80, 92 –– inhibitor 255
Axonopus compressus 371 –– midge 91, 94, 95, 96, 97, 98, 101, 102,
8-azaguanine 339 124, 144, 402
–– periodicity 369
B –– site, preferential 223
babesiosis 327, 336 –– site selection 251
Bacillus thuringiensis var. israelensis 384 blend 48, 55, 128, 135, 143, 150, 155, 157,
back pack aspirator 382 165, 167, 169, 182, 187, 191, 193, 195,
bait 131, 144, 147, 167, 182, 187, 189, 190, 206, 207, 231, 270, 276, 316, 337, 400, 402
191, 192, 193, 194, 195, 196, 197, 205 –– natural and synthetic 13
baited trap 320 –– volatile 55
baker yeast 321 blood feeding 94, 99, 100, 103, 105, 109,
bancroftian filariasis 144 118, 119, 124, 191, 211, 239, 247, 248,
β-barbatene 208 291, 293, 294, 327, 350
basic reproductive number 358 –– parasite-induced changes in behaviour
Bayrepel 235 351
Beauveria bassiana 341 –– persistence 357
behaviour 20, 30, 53, 65, 68, 73, 74, 80, 91, blood meal 53, 68, 91, 92, 99, 104, 115, 133,
94, 99, 106, 115, 117, 118, 120, 122, 123, 145, 146, 150, 205, 206, 236, 259, 273,
125, 129, 136, 182, 184, 203, 204, 205, 276, 296, 311, 319, 358
207, 209 bluetongue virus (BTV) 13, 218, 220, 228,
–– defensive 304 234, 238, 239
–– foraging 53 body odour 30, 188, 236
–– manipulation 349 Bolivia 311
behavioural Bombyx mori 20, 80
–– package 258 Borrelia burgdorferi 353
–– resistance 403 Botswana 183, 185
–– response 54, 55, 79 brain 26, 45, 55, 74, 76, 78, 79, 92, 96

Olfaction in vector-host interactions 419

Index

Brazil 204, 205, 311, 365, 366, 368, 375, 376, 181, 187, 188, 189, 190, 191, 192, 193,
377, 378, 382, 384, 386, 388, 390, 391, 194, 195, 204, 205, 206, 207, 223, 228,
392, 402 229, 230, 231, 233, 234, 236, 237, 238,
breath 46, 47, 122, 135, 151, 155, 156, 161, 239, 247, 250, 251, 252, 253, 254, 255,
187, 188, 189, 205, 235, 253, 333, 379 256, 257, 258, 259, 260, 265, 270, 271,
–– exhaled 351 272, 273, 276, 277, 281, 282, 291, 293,
Breteau Index 366, 368 294, 295, 296, 297, 298, 299, 301, 309,
Brugia malayi 350 315, 316, 317, 318, 319, 320, 321, 327,
BTV – See: bluetongue virus 328, 330, 335, 336, 339, 340, 341, 343,
buffalo 268, 272, 273 367, 370, 379, 380, 381, 382, 383, 399,
Burkina Faso 183, 185, 186, 188, 189, 266, 401, 402, 403
275, 277, 280, 282 –– plume 135
Burundi 183, 185, 186 –– receptors 54
bushbuck 275, 276 carbon disulfide 154, 156, 164
–– urine 276 carboxylic acid 43, 51, 52, 55, 72, 151, 157,
bushpig 272 167, 191, 193, 194, 195, 293, 294, 379, 404
1,3-butanediamine 155 –– aromatic 154, 164
2,3-butanediol 164, 165 Caribbean 222
2,3-butanedione 160, 162 (E)-β-caryophyllene 208
butanoic acid 70, 72, 151, 153, 158, 160, cation channel 69
163, 295, 296, 333, 336 cattle 187, 191, 192, 195, 268, 273, 277
1-butanol 160, 162 –– faeces 293
butanone 154, 156, 223, 265 –– odour 277, 278, 293
2-butoxyethanol 48, 50, 70, 72 –– urine 301
1-butylamine 52 CDC
2-butylamine 313 –– backpack aspirator 369, 376
butylamine 26, 27, 51, 52, 313 –– gravid trap 370, 376
butyric acid 313 –– light trap 14, 382, 383
8-butyrolactone 331 cembrene 203, 207
Central African Republic 183, 185, 275, 276
C Central America 247
cabbage root fly 125 central nervous system (CNS) 71, 74, 97, 98,
calf 188, 189 106, 229
Cameroon 183, 185, 186, 250, 260, 266 Ceratopogonidae 101, 217
Canada 247, 248 Cercopithecus aethiops 190
canine skin 336 Chad 183, 185, 186
Cape Verde 183 Chagas disease 14, 309, 310, 311, 321, 350
capitate peg 27 Chaitophorus stevensis 238
–– sensilla 29, 44, 48, 52, 53, 54 Chelydra serpentina 248
caproic acid 379, 381 chemical
6-caprolactone 331, 332 –– cue 22, 64, 143, 204
carbohydrate 46 –– ecology 55
carbon dioxide (CO2) 14, 17, 27, 28, 29, 30, –– signal 18, 209, 211
46, 52, 53, 73, 91, 92, 93, 94, 95, 96, 97, 98, –– stimulus 92, 99
99, 100, 101, 103, 105, 106, 107, 108, 109, chemokinetic responses 232
115, 117, 118, 120, 122, 124, 127, 128, chemoreception 18, 30, 67
130, 131, 132, 133, 134, 135, 136, 143, chemoreceptor 94
144, 151, 157, 158, 161, 162, 163, 169, –– neuron 92

420  Olfaction in vector-host interactions

 Index

chemosensory copulation 209, 210, 225

–– information 74 co-receptor 24, 52
–– protein 21 Côte d’Ivoire 183, 185, 186, 275, 280
–– sensilla 23, 25, 27 counterflow geometry 234
–– system 109 coupled gas chromatography-
chemotactic response 232 electroantennography 231
chemotaxis 232 cow 21, 51, 190, 191
chicken 162, 166, 311, 314 –– odour 254
Chikungunya 13, 17, 144, 402 –– urine 254, 255
Chile 311 crepuscular 223
China 221, 310 m-cresol 293
cholesterol 155, 163 p-cresol 47, 293, 295, 296, 372
cholinergic connection 77 crested auklet 337
cibarial pump 354 Crimean Congo Haemorrhagic fever 327
circadian crosstalk 76
–– clock 115, 309, 311, 319 Cuba 369
–– rhythm 117, 118, 148, 150, 223, 355 Culex 39, 44, 47, 122, 127, 132, 193, 194
cis-vaccenyl acetate 23 –– annulirostris 223
citral 70, 72 –– antennatus 189
citronellol 48, 51, 293 –– duttoni 189
climate change 218, 237 –– molestus 104, 105
climatic condition 226 –– nebulosus 189
cluster analysis 388 –– nigripalpus 165, 371
Cnephia pecuarum 254 –– pipiens 41, 54, 92, 93, 100, 103, 104, 105,
CNS – See: central nervous system 136, 371
CO2 – See: carbon dioxide –– pipiens quinquefasciatus 17, 399
cockroach 74, 76, 78, 319 –– quinquefasciatus 25, 27, 28, 30, 31, 50,
coding pathway 78 55, 63, 69, 70, 72, 73, 74, 93, 94, 107, 118,
Cola laurifolia 274 131, 143, 144, 146, 147, 150, 162, 163,
Colophospermum mopane 268 165, 166, 168, 169, 192, 211, 352, 355,
Columbia 369 369, 371, 372, 373, 400
common crane 259 –– tarsalis 131, 165, 372
community-based –– territans 41
–– dengue control 392 Culicidae 366
–– intervention 278 Culicoides 102, 103, 217
Congo 183, 185 –– albicans 223
–– -Brazzaville 275, 277 –– barbosai 222
–– Democratic Republic of 183, 185, 186, –– bolitinos 225
275, 280 –– brevitarsis 218, 223, 224, 226
contact pheromone 204 –– chiopterus 220, 225
Container Index 368 –– darlingtonae 230
continuous feeding model 360 –– delta 223
controlled-release device 373 –– dewulfi 220, 225
convection 136 –– flavivenula 230
–– current 117, 127, 134 –– floridensis 222
–– heat 42 –– furens 102, 222, 223, 231, 233, 234, 236
convention –– glabellus 230
–– current 126 –– heliconiae 230

Olfaction in vector-host interactions 421

Index

–– henryi 221 dendrite 18, 24, 31, 32, 39, 42, 43, 44, 64, 69
–– histrio 224 dengue 13, 17, 39, 144, 291, 365, 367, 376,
–– hollensis 222, 223 384, 385, 388, 390, 391, 402, 404
–– imicola 218, 219, 220, 226 –– control 377, 378
–– immaculatus 222 –– fever 350, 368
–– impunctatus 218, 219, 220, 221, 223, –– haemorrhagic fever 365
224, 227, 228, 230, 231, 232, 235, 236, –– risk 366
238, 239 –– transmission risk 368
–– insinuates 230 –– virus 366
–– longior 221 Dermacentor variabilis 329, 330, 331, 334,
–– lupicaris 223 335, 336, 337
–– marmoratus 221 desiccation 237
–– melleus 222, 223, 224 deterrent 15
–– mississippiensis 102, 103, 222, 234, 236 diapause 54, 103
–– molestus 221, 224 dichloromethane 155, 156, 167, 338
–– nubeculosus 224, 225, 226, 230, 235, 238, 2,6-dichlorophenol 339, 341
239 Dictyoptera 74
–– obsoletus 218, 220, 225 differential attractiveness 227
–– ornatus 221 dihydromyrcenol 230, 232
–– paraensis 230 dimethyl
–– pseudodiabolicus 230 –– amine 313, 318
–– pulicaris 220, 223, 225 –– disulfide 143, 154, 156, 164, 167, 169
–– punctatus 225 –– sulfide 331
–– scoticus 220, 225 –– trisulfide 155, 156, 164, 165, 294, 295,
–– sonorensis 239 296
–– stellifer 102 diones 167
–– subimmaculatus 221, 222, 224 Dipetalogaster maxima 314
–– tissoti 222 Diptera 74, 92
–– variipennis 226, 236 Dirofilaria
–– variipennis sonorensis 227, 238 –– immitis 350
–– vexans 223 –– ursi 249
Culiseta disability-adjusted life years 203
–– inornata 25, 26 disease
–– melanura 94 –– transmission 13, 14, 98, 143, 204
cuticular hydrocarbons 210 –– vector 13, 23, 56, 63, 69, 70, 80, 110, 169
cyclic adenosine monophosphate 19, 66 dispersion 133
cyclic feeding model 360 diurnal 120, 150
cysteine 22 Djibouti 183
2,3-docosanediol 164, 165
D dodecanal 165
decanal 72, 152, 156, 230, 232, 372 dodecanoic acid 151, 153, 159, 161, 163
(E)-2-decanal 165 n-dodecanoic acid 204
decanoic acid 151, 153, 159, 161, 163 1-dodecanol 47, 158
decanol 163 doe urine 337
2-decanone 154 dog 351
deer 223 –– ears 336
DEET – See: N,N-diethyl-meta-toluamide DOR67d 21
dehydrolinalool 235 dose

422  Olfaction in vector-host interactions

 Index

–– -effect 252 erhlichiosis 327, 336

–– -response curve 188, 252, 295 Eritrea 183, 185
–– -response relationship 294 ethanol 162
Drosophila 18, 20, 21, 22, 23, 24, 25, 26, 29, Ethiopia 183, 184, 185, 186, 191, 301
30, 31, 45, 53, 106, 117, 121, 130, 146, ethylamine 313, 318
210, 211 ethyl butyrate 48, 70, 72, 156
–– melanogaster 17, 44, 45, 67, 68, 71, 76, ethyl disulfide 154, 156, 164
77, 78, 79, 80, 106, 211, 238, 400 ethylene glycol 163
–– melanogastor OR83b 69 2-ethylhexanoic acid 333, 337
dry ice 251, 321, 370, 379 2-ethylhexanol 230, 232
dual-choice ethyl lactate 48
–– assay 187 ethyl-L-lactate 48, 70, 72
–– bioassay 372 ethyl mercaptan 331
–– olfactometer 232 4-ethylphenol 50, 158, 224
–– test 189 ethyl propanoate 48
dual-port olfactometer – See: olfactometer, ethyl propionate 48
dual-port Eucalyptus citriodora 235
dung eugenol 48, 51, 70, 72
–– odour 294 Europe 247
–– volatile 295, 296 evolution 17, 24, 30
EVS trap 370, 382
E excitation 39
E-2-heptenal 331 exophagic 186, 224, 225
EAG – See: electroantennogram exophilic 184, 186, 224
East Coast fever 341 extrinsic incubation period 227
ectoparasite 327
ectopic expression 71 F
Egypt 186, 189 F3 trap 300
Ehrlichia (Cowdria) ruminantium 340 faeces 165, 167, 319
electric net 132, 189, 268, 269, 271, 276, fatty acid 17, 48, 130, 228, 379
277, 401 –– derivative 46
electroantennogram (EAG) 165, 167, 229 Fay-Prince trap 367, 370, 382
electroantennography 146, 228, 270 feather 336
electrophysiological 206 –– odour 314
–– response 23, 30 feeding
–– study 26, 48, 54, 55, 95, 251, 293, 294 –– aggregation 338
ELISA 259 –– persistence 349, 356, 361
emanation 162 –– site preference 257, 338
encephalitis 17 filament 124, 128, 151
endogenous filiariasis 17
–– activity 293 Flaviviridae 365
–– physiological factor 318 flea 359
–– rhythm 269 flight
endophagic 186 –– path 119, 122, 132
endophilic 182 –– periodicity 294
epizootic haemorrhagic disease 220, 221 –– simulator 130
Equatorial Guinea 183, 185 floating larval trap 369
equine enchephalosis 221 fluoro-lactate 48

Olfaction in vector-host interactions 423

Index

fly 22 glomeruli 27, 29, 31, 74, 75, 76, 77, 78, 79,
foot odour 190, 191, 193 80
fox 206 Glossina 303
France 220, 225 –– austeni 268, 273, 279, 280
French Polynesia 376 –– brevipalpis 274, 280
fruit fly 319 –– fuscipes 265, 275, 276, 280
functional –– fuscipes fuscipes 277, 280, 282
–– genomics 63 –– longipalpis 273
–– group 45 –– longipennis 274, 279, 280
–– specificity 20 –– medicorum 274
–– type 49, 55 –– morsitans centralis 267, 273
furan 314, 318 –– morsitans morsitans 265, 267, 268, 273,
furfural 331 278, 280, 296
Fusca group 266, 267, 274, 279 –– morsitans submorsitans 267, 279, 282
–– pallidipes 252, 253, 265, 267, 268, 273,
G 274, 278, 279, 280, 282, 352, 355
GABA receptor antagonists 77 –– palpalis 265, 275, 281
Gabon 183, 185, 186, 383 –– palpalis gambiensis 275, 277, 280
GAL4/UAS expression system 67 –– palpalis palpalis 274, 277, 280
Gambia 183, 185, 186 –– swynnertoni 268, 273
game elimination 266 –– tachinoides 274, 275, 277, 280, 282
gametocytes 351, 352 Glossinidae 292
gas chromatography 145, 228, 270, 312 glycerol 152, 163
–– -mass spectometry analysis 46, 47 goat 187, 191, 272, 273
–– -mass spectrometry 228 gonotrophic cycle 53, 115, 119, 136, 248,
–– -SSR 73 318
gene 17, 20, 21, 23, 24, 29, 31, 39, 44, 45, Gp – See: G-protein
55, 67 GPCR – See: G-protein coupled receptor
–– sequence 146 G-protein (Gp) 19, 20, 25, 66
–– silencing 358, 400 –– coupled receptor (GPCR) 23, 45, 68
genetically modified vector 400 GR – See: gustatory receptor gene
genetics 14 grape berry moth 238
genetic variation 280 grass infusion 47, 370, 371, 372, 373, 375,
genome 17, 22, 24, 45, 136 376
–– sequencing 63, 80 green leaf volatiles 204
genomic studies 64 GRN – See: gustatory receptor neuron
geographic information system (GIS) 367, grooming 273
384, 390 –– behaviour 224
–– map 386, 391 grooved peg 41, 44, 64, 309, 312, 313
–– technology 239 –– sensilla 42, 48, 49, 51, 52, 53, 54, 65
geranyl acetone 47, 50, 143, 154, 156, 160, Grus grus 259
161, 169, 228, 230, 232, 235 Guadeloupe 340, 341
Germany 220 guanine 333, 339
Ghana 183, 185, 186 Guinea 183, 185
GIS: – See geographic information system Guinea Bissau 183
glandular secretion 337, 338 guinea pig 295, 311
global warming 105 gustatory
–– neuron 44

424  Olfaction in vector-host interactions

 Index

–– organ 26, 64 hexane 165

–– receptor cell 332 hexanoic acid 68, 153, 155, 156, 158, 163,
–– receptor gene (GR) 146 169, 190, 332, 333, 336, 337, 379
–– receptor neuron (GRN) 27 1-hexen-3-ol 50
–– sensilla 251 4-hexen-1-ol 152, 156
high-throughput 32, 165
H Hippopotamus 276
Haemaphysalis leachi 338 homeostasis 45
haematin 333, 339 homogeneous 129, 130, 132, 151
Haematobia irritans 293, 302 homolog 24, 68
Haematobosca 301 honeybee 22
haematophagy 291 host
haemostasis 353 –– cue 115, 127, 128, 132, 136, 143, 206
haemostatic defence 358 –– defensive behaviour 257
Haller’s organ 329, 330, 331, 332, 335 –– finding behaviour 133
hamster 205, 206, 351 –– lethargy 353
hard tick 327, 328 –– location 117, 129, 134, 135, 217
Hawaii 219 –– location behaviour 239
hay infusion 370, 371, 376 –– manipulation by parasites 359
HBC – See: human biting catch –– odour 14, 117, 122, 127, 131, 134
headspace extract 312 –– orientation 136
heartwater 327, 336 –– -parasite relationships 299
heat 46, 91, 109, 115, 127, 134, 136, 150, –– preference 43, 128, 146, 249, 259
234, 317, 320, 335, 340 –– selection 272
heat/thermal stimuli 257 –– stimuli 119, 133
Heliothis virescens 20, 23 host-seeking 20, 26, 99, 100, 101, 103, 125
heptadecanal 165 –– behaviour 69, 91, 92, 98, 100, 103, 105,
heptadecane 333 106, 109, 143, 145, 147, 148, 150, 157,
n-heptadecane 224, 226 158, 162, 163, 165, 166, 167, 168, 169,
heptadecanoic acid 151, 154, 164 181, 182, 187, 188, 195, 197, 204, 247
heptanal 72, 152, 163, 314, 331 –– behaviour, indoor 401
heptane 155 –– strategy 259
heptanoic acid 70, 72, 151, 153, 159, 160, house-entering 132, 133, 135
163, 332, 333 house improvement 311
1-hepten-3-ol 50, 152, 156 hovering 134
heterodimer 24, 45, 66 human
heterogeneous odour plumes 381 –– biting catch (HBC) 403
heterologous –– body odour 283
–– expression 29, 71 –– breath 293, 294, 331, 335
–– platform 55 –– effluvia 314
–– system 21 –– emanation 145
heteromeric complex 20 –– feet 47
(E, Z)-6,11-hexadecadienyl acetate 65 –– landing catch 191, 192, 196
hexadecanal 165 –– landing collection 369, 382
hexadecanoic acid 151, 153, 159, 161, 164, –– odour 255
165 –– skin 46, 151, 156, 157, 162, 379
hexanal 152, 156, 163, 164, 165, 204, 331, –– skin emanation 26
333, 337, 338 –– skin extract 379

Olfaction in vector-host interactions 425

Index

–– skin odour 380, 381 interneuron 63

–– sweat 250, 313, 317 intrinsic attractiveness 352
–– sweat odour 316 intron 22
–– urine odour 314 invitation effect 255
humidity 134, 136 ionotropic 69
hunter strategy 328 –– response 20
Hyalomma dromedarii 328, 342 IPCC – See: Intergovernmental Panel on
hydrogen sulfide 330, 333 Climate Change
Hydrotaea irritans 228 IR3535 404
3-hydroxy-2-butanone 160, 162 isobutanoic acid 333
2-hydroxybenzaldehyde 331 isobutylamine 313, 318
3-hydroxybenzaldehyde 331 isobutyric acid 313, 316, 317, 318
hydroxybenzaldehyde 332 isopentanoic acid 336
2-hydroxybenzoic acid 154, 164 isopentylamine 313
2-hydroxypentanoic acid 160 isoprene 335
hygroreceptor 44 isovaleric acid 313
–– cell 313 Israel 186
Hymenoptera 74, 92 Italy 376, 383
hypoxanthine 333, 339 Ivory Coast 278
Ixodes
I –– angustus 327
iguanas 311 –– canisuga 336
Ilex vomitoria 236 –– neitzi 337
immunomodulator 353 –– persulcatus 336
impala 272, 273 –– ricinus 328, 331, 332, 336, 337, 339, 340,
increased 342
–– attractiveness of pregnant women 352 –– rubicundus 340
–– biting behaviour 349 –– scapularis 333, 334, 336, 337, 339, 340,
India 310 341, 342, 343, 353, 400
Indian Ocean 187 –– texanus 329
1H-indole 155 –– uriae 334, 337, 339
indole 30, 47, 49, 50, 52, 53, 55, 70, 72, 160, Ixodidae 327
161, 336, 372
Indonesia 310 J
infectious diseases 17, 39 Johnston’s organ 75, 125, 130
infochemical 48
infrared radiation 314 K
inhibition 39 kairomone 28, 69, 70, 71, 115, 134, 136, 143,
inosine 333, 339 147, 151, 157, 161, 165, 166, 167, 169,
insecticide-treated 181, 189, 194, 195, 196, 203, 205, 207,
–– bed net 405 235, 270, 271, 272, 273, 280, 281, 293,
–– cattle 278, 304 297, 299, 302, 304, 328, 330, 337, 340,
–– target 265 399, 401, 402, 403, 404
in situ hybridisation 52, 67 Kenya 183, 184, 185, 186, 190, 270, 273, 275,
interception trap 252 278, 280, 301, 352, 355
interdigital gland substance 337 Kenyon cells 79
Intergovernmental Panel on Climate Change ketone 26, 27, 52, 161, 167, 168, 191, 194,
(IPCC) 237 293

426  Olfaction in vector-host interactions

 Index

kissing bug 403 lemon eucalyptus 235

klinokinesis 115, 116, 126, 127, 134, 232 lemongrass 191
–– longitudinal 126 Lepidoptera 65, 69, 74, 92
–– transverse 126 Leucocytozoon 249
klinokinetic response 280, 296 –– simondi 249
klipspringer antelope 337 –– smithi 249
knock-down spray 403 Liberia 183, 185, 274, 277
Kobus ellipsiprymnus 272 light trap 370
kudu 268, 272 Limburger cheese 47, 51, 190, 194
limonene 70, 72, 230, 232, 372
L linalool 70, 72, 73, 74, 230, 232, 235, 293
labellar lipid extract 337
–– lobe 44, 64, 65 lipocalin 21
–– sensilla 52 lizard 281
labellum 19, 25, 26, 27 –– monitor 276
laboratory assay 295 –– odour 277, 278
La Crosse encephalitis 291 –– urine 276
lactic acid 44, 47, 49, 51, 52, 53, 54, 55, 91, locomotion compensator 315
109, 143, 151, 152, 155, 156, 157, 158, locomotor stimulant 335
159, 160, 161, 162, 163, 167, 168, 223, long-range
228, 237, 379, 381, 404 –– dispersal 374
–– L- 17, 92, 130, 187, 190, 191, 193, 194, –– orientation 270
195 –– response 268, 269
–– L(+) 309, 317, 318, 380 loon 249, 259, 260
lactone 330, 332 lure 206, 207, 209
lambdacyhalothrin 233 –– and contaminate’ strategy 404
landing 117, 131, 135, 148, 151, 159, 161, –– and kill 196
167, 168, 181, 192, 207, 251, 272, 273, –– and kill tactic 403
355, 361, 370 Lurex 237
–– assay 156 Lurex3 237
–– response 159, 206 LUSH 21, 22, 67
–– site 257 Lutzomyia
larval –– carmelinoi 208
–– source reduction 366 –– evandroi 208
–– trap 368 –– evansi 205
larvicide 384 –– intermedia 205
lateral –– lenti 208
–– accessory lobe 75 –– lichyi 210
–– protocerebrum 75 –– longipalpis 203, 204, 205, 206, 207, 208,
Latin America 309 209, 210, 211, 351, 355
learning 148 –– migonei 208
Leishmania 349, 350, 357, 358, 360, 403 –– neivai 208
–– chagasi 203 –– tejadai 208
–– donovani 205 –– vexator 210
–– infantum 204, 351, 357 –– whitmani 205, 206, 208
–– mexicana 355, 357 Luxembourg 220
leishmaniasis 13, 17, 204, 206, 207, 209, 211 Lyme disease 13, 327, 336, 338, 353
lekking site 206 lymphatic filariasis 350

Olfaction in vector-host interactions 427

Index

M menthone 70, 72
macroglomerulus 76 meso-2,3-butanediol 164, 165
Madagascar 183, 184, 185, 190 metabotropic 69
MAG protein 118 metacyclic promastigote 357
Malaise trap 299 Metarhizium anisopliae 341
malaria 13, 17, 39, 63, 133, 144, 181, 182, methanoic acid 152
183, 185, 187, 195, 196, 291, 350, 351, 2-methoxyphenol 273
352, 355, 357, 358, 360, 361, 384, 402, 404 2-methyl-1-butanol 160, 162
Malawi 183, 184, 185, 186 3-methyl-1-butanol 50, 158, 160, 162
Malaysia 205, 310 2-methyl-2-butanol 204
Mali 183, 185 3-methyl-2-butylamine 313
mammalophilic 249, 256 3-methyl-2-hexanoic acid 236
Manduca 68 3-methyl-2-hexenoic acid 50, 189
–– sexta 76, 77 (E/Z)-3-methyl-2-hexenoic acid 159, 161
mangrove 222, 274 4-methyl-2-methoxyphenol 273, 283
Mansonia 127, 188, 193, 194 4-methyl-2-nitrophenol 331, 336
–– africana 41, 189 6-methyl-5-hepten-2-one 47, 53, 143, 154,
–– uniformis 41, 129, 188, 189 156, 160, 161, 169, 228, 230, 232, 235
mark-release-recapture 122, 133 methyl-amine 313
mass-spectrometry 145 2-methylbutanal 160, 162
mass trapping 196 3-methylbutanal 160, 162
mating 118, 209 2-methylbutanoic acid 160, 162
–– aggregation 206, 207 2-oxo-3-methylbutanoic acid 159, 161
–– behaviour 210, 211 3-methylbutanoic acid 151, 153, 158, 160,
–– disruption 209 162, 163, 187, 191, 194, 332
–– swarm 236 methylbutyrate 48
Mauritania 183, 185, 186 2-methylbutyric acid 313
Mauritius 183, 184 4-methylcyclohexanol 48, 50, 72
maxillary palp 19, 24, 27, 28, 29, 30, 39, 40, 2-methyldocosane 224
44, 48, 52, 54, 64, 65, 67, 91, 92, 93, 94, 8-methyldocosane 224
102, 103, 104, 127, 146, 229, 230 9-methylgermacrene 203, 204, 208
–– sensilla 96, 97, 106 9-methylgermacrene-B 203, 208
Mean Female Aedes Index (MFAI) 386, 387, S-9-methylgermacrene-B 204, 207
388 3-methylindole 47, 53, 70, 72, 334, 336, 372,
mechanical transmission 350 373
mechanism 4-methylnitrophenol 332
–– rate-sensitive 97 2-oxo-3-methylpentanoic acid 160, 161
–– tonic response 97 2-oxo-4-methylpentanoic acid 160, 161
mechanoreception 116 2-methylphenol 332, 336
mechanoreceptive anemotaxis 125 3-methylphenol 70, 72, 223, 274, 276, 336
mechanoreceptor 125 4-methylphenol 47, 50, 51, 53, 192, 195,
mechanosensory cue 130 223, 265, 270, 273, 274, 276, 277, 278,
membrane potential 229 334, 336
memory methylphenol 30
–– formation 79 1S,2S’-2-methylpiperidinyl-3-cyclohexene-1-
–– system 80 carboxamide 342
p-menthane-3 235 2-methylpropanoic acid 158, 161, 332, 334
menthol 230, 232 methyl propyl disulfide 155, 156, 164, 165

428  Olfaction in vector-host interactions

 Index

methyl salicylate 230, 334, 339, 340, 341 moth 74, 76, 78, 117, 122, 123, 124, 126,
methyl sesquiterpenes 208 128, 135, 298
methyl sulfide 154, 156, 164 mouse 317, 320
9-methyltricosane 224 mouthpart 40, 74
3-methyl-α-himachalene 203, 207 –– sensilla 251
Mexico 376 Mozambique 183, 184
MFAI – See: Mean Female Aedes Index Murray Valley encephalitis 291
microbiota 145, 162 Muscidae 101
microfilariae 251 mushroom body 75, 79, 80
microfilarial periodicity 355 β-myrcene 164, 165
microsatellite DNA 281
microvasculature 355 N
midge 304 nagana 265
Midgeater trap 237 Namibia 183
Midg-IT trap 237 naphthalene 230, 232
migratory flight 133 Nasci aspirator 369
Mimosa pigra 274 neem 235
‘miombo’ woodland 267 nervous system 97
miR-279 29 neural
Mitragyna inermis 274 –– circuit 29
MM-X trap 149 –– connection 63
moisture 46, 91, 109, 115, 150 –– network 76
molecular genetics 30 neuroethology 64
monkey 187, 190 neuron 21, 26, 27, 28, 29, 30, 41, 42, 43, 44,
Morelia senegalensis 274 45, 46, 50, 52, 53, 54, 69, 72, 78, 94, 96, 97,
Morsitans group 265, 266, 267, 273, 274, 98, 99, 101, 102, 105, 146, 206
275, 278, 279 –– empty 53, 71
mosquito 14, 20, 22, 23, 24, 25, 26, 27, 28, –– postsynaptic 74
29, 30, 31, 32, 39, 40, 41, 42, 44, 46, 47, 48, –– presynaptic 74
51, 52, 53, 54, 55, 63, 64, 68, 69, 70, 72, 73, –– projection 63, 76, 78, 79
74, 76, 78, 79, 80, 91, 92, 93, 94, 95, 96, 97, neuronal signal filtering 78
98, 100, 101, 102, 103, 104, 106, 107, 109, neurophysiology 165, 166
115, 117, 118, 119, 120, 121, 122, 124, New Guinea 310
125, 126, 127, 131, 132, 134, 136, 143, New Zealand 219
144, 145, 146, 147, 151, 155, 156, 157, NH3 – See: ammonia
162, 165, 166, 167, 168, 169, 181, 187, Nicaragua 368
188, 189, 190, 191, 192, 193, 194, 195, nidicolous 327
196, 197, 205, 239, 294, 298, 304, 316, Niger 183, 186
317, 318, 319, 351, 352, 353, 357, 358, Nigeria 183, 185, 186
361, 368, 390, 400, 405 nitric oxide 334, 335
–– behaviour 17 2-nitrophenol 331, 332, 334, 336, 337, 339
–– genes 18 o-nitrophenol 340, 341
–– olfactory gene 21 N,N-diethyl-meta-toluamide (DEET) 27, 30,
Mosquito Magnet trap 233, 234, 237, 382 31, 48, 54, 55, 92, 106, 107, 168, 233, 235,
MosquitoTRAP 377 342, 404
MosquiTRAP 365, 367, 371, 372, 375, 376, nocturnal 121, 125, 150
378, 384, 385, 388, 391

Olfaction in vector-host interactions 429

Index

nonanal 70, 72, 143, 147, 152, 156, 162, 163, –– neuron (ORN) 19, 20, 23, 24, 26, 27, 28,
164, 165, 230, 232, 314, 316, 317, 372, 29, 30, 31, 32, 40, 41, 42, 43, 44, 45, 46, 47,
373, 374, 375 48, 50, 51, 52, 53, 55, 64, 66, 68, 69, 70, 71,
nonanoic acid 151, 153, 159, 161, 163, 332, 72, 73, 74, 76, 77, 78, 231
334, 339, 340 –– protein 17, 25
2-nonanone 50 odour 48, 51, 55, 71, 77, 79, 115, 116, 118,
(E)-2-nonenal 230, 232 123, 125, 129, 133, 162, 165
2-nonene 155 –– -baited entry trap (OBET) 129, 132, 133,
non-nidicolous 327, 328 136, 188, 189, 190, 191, 192, 194, 196
non-random feeding 352 –– -baited station 403
non-vector biting fly 292 –– -baited target 282
nuisance insect 13 –– -baited trap 14, 217, 239, 278, 283, 309,
nutritional state 118, 150 401
Nzi trap 301 –– blend 17, 27, 46, 69, 71, 144, 145, 157,
166, 169, 236
O –– blend, natural 315
OBET – See: odour-baited entry trap –– coding 31, 39, 45, 46, 63, 64, 78
obligate haematophagy 319 –– cue 121, 122, 134
OBP – See: odorant-binding protein –– -degrading enzyme 343
Obp genes 68 –– discrimination 79
octadecanal 165 –– -mediated upwind anemotaxis 166
octadecanoic acid 151, 154, 164, 165 –– -modulated anemotaxis 314
octan-1-ol 293 –– -modulated behaviour 401
octanal 72, 143, 152, 156, 230, 232, 314 –– plume 71, 73, 117, 119, 120, 121, 122,
octanoic acid 68, 153, 159, 160, 163, 253 123, 124, 125, 126, 127, 129, 132, 133,
1- and 2-octanol 253 134, 135, 136, 269, 296
1-octen-3-ol 17, 27, 28, 29, 43, 48, 50, 52, 54, –– profile 147
55, 70, 72, 91, 92, 106, 109, 132, 156, 162, –– receptor cell (ORC) 312, 313, 314, 316
187, 189, 190, 192, 194, 195, 207, 223, –– source 124, 130
228, 230, 232, 233, 234, 236, 237, 253, –– space 24, 30
265, 270, 293, 294, 295, 297, 298, 300, –– stimulus 45, 73, 124
301, 316, 317, 318, 334, 336, 337, 339, 404 olfactometer 95, 127, 129, 143, 146, 148,
octenal 337 149, 150, 151, 155, 156, 158, 159, 160,
2-octene 155 161, 162, 163, 166, 167, 168, 208, 223,
7-octenoic acid 50, 53, 159, 161, 187, 189, 226, 258, 272, 273, 291, 294, 295, 315,
194, 236 320, 335, 375, 402
octenol 167, 253, 254, 271, 272, 274, 276, –– dual-port 152, 158, 163, 374
277, 278, 282 –– Y-tube 152, 223, 230, 232, 379
Odocoileus virginianus 336, 337 olfactory
odorant 14, 30, 39, 44, 46, 76 –– behaviour 14, 15, 80
–– -binding protein (OBP) 17, 20, 21, 22, 24, –– coding 80
25, 39, 44, 63, 64, 65, 66, 67, 80, 211, 343, –– cue 47, 119, 125, 146, 150, 165, 168, 182
400 –– gene 20, 399, 400
–– degrading enzymes 20 –– lobe 55, 74
–– stimuli 26 –– neuron 42, 106
odorant receptor (OR) 17, 19, 20, 23, 29, 30, –– organ 52
46, 63, 64, 66, 76, 80, 343 –– physiology 24, 39
–– gene 45 –– processing 77, 80

430  Olfaction in vector-host interactions

 Index

–– protein 18, 23 –– attractant 365, 367

–– receptor 31, 39, 45, 68, 330 –– behaviour 26
–– receptor cell 309 –– cue 375
–– receptor gene 146 –– site 371
–– receptor neuron 39, 92, 95, 230 –– site, artificial 370
–– receptor neuron (ORN) 18 –– skip- 378
–– receptor protein 44 –– trap 367
–– regulation 14 ovitrap 368, 377, 378
–– response 31 ox
–– sensilla 27, 41, 43, 44, 48, 49, 312 –– blood 355
–– sensilla type 229 –– odour 270, 271, 272, 274, 277, 301
–– signal 78, 91 –– odour, natural 269
–– signalling 23, 29, 69 –– urine 272
–– signalling pathways 21 2-oxobutanoic acid 159, 161
–– signal transduction 29, 32, 66 oxocarboxylic acid 47, 50, 52
–– stimuli 144 2-oxohexanoic acid 160, 161
–– transduction 19, 20 2-oxooctanoic acid 160
ommatidia 256 2-oxopentanoic acid 159, 161
Onchocerca 2-oxopropanoic acid 159
–– cervicalis 220 ozone 238
–– volvulus 248, 261, 350
opossum 310 P
opportunistic 128, 147, 156, 183, 205 packed cell volume 354
optic lobe 26 Palpalis group 265, 266, 267, 274, 275, 278,
optomotor 279, 281, 282, 283
–– anemotaxis 115, 125, 134, 269, 271 Panama 205
–– response 123, 126 Panicum maximum 371, 372, 373, 375
OR – See: odorant receptor Panstrongylus megistus 310, 321
OR7 24 Papua New Guinea 352
OR83b 24, 44, 66, 68, 69 Paraguay 311
ORC – See: odour receptor cell Paralobesia viteana 238
Oreotragus oreotragus 337 parasite periodicity 355
orientation 91, 92, 109, 115, 116, 117, 119, parrafin oil 70, 72
120, 121, 122, 124, 125, 127, 128, 129, passive ambush tactic 328
130, 132, 134, 160 Patagonia 219
–– behaviour 78 pathway 21
ORN – See: odorant receptor neuron PBP – See: pheromone-binding protein
Ornithodoros erraticus 333, 335 Pediculus humanus 400
Ornithodoros turicata 340 penguin 340
ornithophilic 147, 249, 257 pentacosane 155
–– species 254 pentadecanal 165
orthokinesis 126, 127, 232 pentadecanoic acid 151, 153, 164
orthokinetic response 280, 296 pentanoic acid 151, 153, 158, 160, 334, 336,
ortholog 29, 69, 74 337
oviposition 47, 48, 49, 50, 51, 53, 54, 55, 68, 2-pentanone 154, 156
69, 80, 99, 118, 133, 136, 203, 204, 205, 3-pentanone 154, 332
208, 236, 238, 239, 319, 372, 373, 374, peptide 22, 64
375, 376, 384, 400 peridomestic 311

Olfaction in vector-host interactions 431

Index

peripheral plume 101, 109, 115, 116, 119, 123, 124,

–– olfactory system 53, 63, 64, 74 127, 131, 132, 134, 135, 151, 236, 252,
–– sensory neuron 103 253, 270, 294, 380, 381 – See also: odour
–– sensory system 101 plume
Periplaneta americana 76 –– filamentous 73
Peru 311, 368 –– following 115, 117, 128, 129
phagostimulant 251 –– homogeneous 73
Phantom Midge 94 –– intermittent 130, 132
pharangeal pump 354 –– structure 117, 135, 136
phasic response 73 –– temperature 234
phasic-tonic discharge pattern 102 Porapak 331
phenol 46, 50, 53, 55, 163, 194, 223, 230, positive anemotaxis 314
236, 272, 273, 276, 277, 282, 330, 336 Premise Index 366, 368
phenylacetaldehyde 341 pre-orbital gland 272
pheromone 20, 21, 23, 27, 32, 46, 65, 69, 70, presynaptic inhibition 78
71, 80, 122, 123, 124, 126, 135, 204, 205, Principe 266
206, 210, 211, 224, 226, 236, 238, 328 probing time 354
–– -binding protein (PBP) 65 proboscis 24, 25, 39, 40, 44, 52, 64, 65, 67,
–– -mediated behaviour 238 93, 104
–– plume 125, 128 promastigote secretory gel 357, 358
–– receptor 68 propanal 163
–– receptor neuron 96 propanoic acid 151, 153, 158, 160, 163, 334
–– tag 341 propionic acid 48, 70, 72, 237
Phlebotomus 204 3-n-propylphenol 192, 195, 223, 233, 265,
–– argentipes 205, 207, 208, 210 270, 273, 274, 277, 278
–– duboscqi 208, 210 Prosimulium decemarticulatum 260
–– longipes 210 Prostomoxys 301
–– martini 210 proteophosphoglycan 357
–– orientalis 208, 210 prothoracic tarsi 251
–– papatasi 208 protocerebrum 79, 80
physical cue 150 Pterocarpus santalinoides 274
physiological state 53, 117, 118, 135 pulsed stimulation 315
Phytoseiulus persimilis 74 push-pull
Picaridin 404 –– approach 341, 342
pig 187, 191 –– effect 338
pigeon 162, 311 –– strategy 235, 239, 404
α-pinene 48, 164, 165 –– system 234
α(+) pinene 314 4-pyridinamine 155
plague 359 pyridine 155, 314, 318
Plasmodium 181, 350, 353, 403 pyruvate 334, 337
–– chabaudi 351, 354, 356 pyruvic acid 48
–– falciparum 181, 352, 353, 355
–– gallinaceum 351, 356, 358 Q
–– knowlesi 181 quality coding 77
–– malariae 181 questing 328
–– ovale 181 –– behaviour 335, 337
–– vivax 181
–– yoelii nigeriensis 354, 356

432  Olfaction in vector-host interactions

 Index

R River Blindness 248

R0 13, 358, 359, 360 RNAi 400
rabbit 311 –– knock-down 211
–– odour 313, 331 –– strategy 68
ranging rolling fulcrum model 258
–– flight 119, 121, 124 Russia 221
–– strategy 121 Rwanda 183, 185, 186
receptor 19, 23, 28, 29, 69, 70, 71, 98, 127,
211 S
–– neuron 46, 52, 53, 63, 91, 92, 95, 97, 98, Sabethes cayaneus 121
99, 100, 101, 102, 103, 104, 105, 106, 107 Saccharomyces cerevisiae 320
–– potential 229 salicylaldehyde 334
Reduviidae 309 saliva 251
refuge 311 salivary
removal trapping 233, 404 –– antigen 248
repellent 15, 30, 54, 143, 144, 145, 146, 148, –– gland protein 353
150, 156, 157, 162, 167, 168, 169, 182, sandfly 69, 124, 203, 204, 205, 206, 207, 208,
191, 228, 233, 235, 239, 283, 304, 328, 209, 211, 357, 358, 400, 403
337, 341, 343, 404 savanna 256
–– spatial 235 scanning electron micrograph 92, 93
reproductive fitness 350 scent-marking behaviour 337
reptile 275 Sclerica bractaeda 371
–– faeces 340 Scotland 221, 226, 227, 231, 232, 402
response 29, 30, 39, 45, 48, 73, 95, 97, 98, scotophase 150, 319
99, 100, 101, 102, 103, 106, 115, 122, 128, screenhouse 190
135, 146, 148, 150, 159, 161, 162, 168, screwworm fly 144, 402
187, 192, 197, 205 second messenger 66
–– electrophysiological 103 secretion 331
–– excitation 46 semi-field 187, 188, 191, 193, 195, 197, 375,
–– pattern 73 378, 401
–– short-range 268, 270 –– enclosure 401
–– spectra 29 –– study 383
–– threshold 102 –– test 374
–– tonic 46 semiochemical 21, 47, 107, 115, 117, 143,
Rhinomusca 301 144, 145, 146, 148, 150, 165, 166, 168,
Rhipicephalus 209, 211, 231, 235, 239, 341, 402, 404
–– appendiculatus 333, 338, 339, 341 Senegal 183, 185, 186, 190, 274
–– evertsi 338, 341 sensillum 18, 21, 24, 26, 30, 40, 42, 44, 46,
–– microplus 328, 329, 331, 332, 333, 334, 47, 52, 55, 63, 64, 66, 67, 68, 69, 71, 74, 92,
337 93, 94, 95, 99, 146, 206, 211, 229, 329, 330
–– sanguineus 333, 335, 336 –– ampullacea 41, 42
Rhodnius prolixus 309, 312, 314, 319, 320, –– basiconica 41, 52, 64, 67, 92, 93, 94, 99,
321 102, 104, 229, 230, 309, 312, 314
Rift Valley fever 17, 352, 354 –– chaetica 41
rinderpest 266 –– coeloconica 41, 42, 43, 44, 64, 229
risk –– coeloconica, large 42
–– assessment 13 –– trichodea 23, 31, 41, 42, 43, 44, 48, 49,
–– model 220 53, 54, 64, 65, 67, 69, 70, 72, 73, 107, 312

Olfaction in vector-host interactions 433

Index

sensory –– electrophysiology 330, 332

–– cue 92, 259 –– recording (SSR) 26, 27, 45, 69, 146, 206,
–– neuron 64, 105, 106 228, 312
–– neuron membrane protein (SNMP) 19, singleton 24
20, 21, 23 sit-and-wait strategy 121, 268
–– organ 64 skin 205, 206
–– periphery 80 –– emanation 127, 129, 143, 145, 155, 157,
–– physiology 399 158, 162, 235
–– stimuli 30 –– microbial flora 351
–– system 115, 146 –– odour 120, 128, 130, 317, 335
Sergentomyia 204 –– secretion 255
–– fallax 208 SkinLure 383
–– minuta 208 sleeping sickness 278
servosphere 315, 316, 317, 335 smell 39
sesquiterpene 208 snapping turtle 248
sex pheromone 203, 207, 208, 209, 341 SNMP – See: sensory neuron membrane
sexual dimorphism 74, 80 protein
sexually dimorphic macroglomeruli 78 sodium hydroxide 335
Seychelles 184 soft tick 327
sheep 187, 191 solid phase micro extraction (SPME) 228,
–– odour 314 372
short-chain solvent washings 228
–– aliphatic amine 313 Somalia 183, 184, 185, 186, 268, 274, 279,
–– carboxylic acid 313 280, 282
–– fatty acid 330 sound 91
Sierra Leone 183, 185, 186 South Africa 183, 184, 185, 186, 266, 268,
signal 91 280
–– transduction 17, 18, 24, 63 South America 249
signalling 204 spider mite 74
–– molecule 68 SPME – See: solid phase micro extraction
–– pathway 20, 66 Spodoptera littoralis 76
Simuliidae 101 squalene 155, 334, 337
Simulium Sri Lanka 207
–– anatinum 250, 254, 260 SSR 71, 73 – See: single sensillum recording
–– annulus 259 stable fly 291, 292, 294, 296, 297, 299, 302,
–– arcticum 247, 248, 249, 252, 254, 256, 304
257 Staphylococcus epidermidis 145
–– aureum 260 steer hair extract 332
–– damnosum 250, 255, 256, 260 Stegomyia 368
–– euryadminiculum 249, 250, 254, 257, sticky panel 266
259, 260 sticky trap 252, 367, 372, 376, 377, 378, 391
–– luggeri 248 Stomoxyinae 291, 299
–– meridionale 254 Stomoxys 291, 293, 298, 301, 303
–– rugglesi 250, 254, 257, 259, 260 –– calcitrans 236, 292, 300
–– venustum 248, 251, 252, 255 –– niger niger 300
–– vittatum 249, 251 –– nigra 350
Singapore 384 –– sitiens 300
single sensillum Stygeromyia 301

434  Olfaction in vector-host interactions

 Index

styrene 155 toluene 155

suboesophageal ganglion 26, 27 tonic spiking pattern 74
suction trap 121 tourism 221
Sudan 183, 185, 186 Toxorhynchites brevipalpis 121
sugar feeding 236 transduction 20
sulcatone 50 –– mechanism 68, 80
sulfide 164 –– olfactory 18, 44
surveillance 14, 309, 340 –– pathway 19
–– system 299 transgene 29, 67
–– tool 299 transmission 106
Swaziland 184 –– risk 405
sweat 47, 51, 54, 130, 157, 161, 195 trap 14, 108, 117, 129, 131, 132, 144, 157,
Sweden 259 162, 181, 187, 188, 189, 190, 191, 192,
Sweetscent dispenser 380 193, 194, 195, 196, 205, 207, 208, 232,
synergism 55, 152, 158, 159, 169, 194, 209 233, 236, 237, 261
synergistic effect 47, 167, 379 Triatoma
Syria 186 –– brasiliensis 321
Syzygium guineense 274 –– dimidiata 309, 321
–– infectans 350
T –– infestans 309, 310, 312, 313, 314, 315,
Tabanidae 101, 121 317, 319, 320, 321
Tanzania 183, 184, 185, 186, 188, 189, 191, –– pallidipennis 321
192, 193, 402 –– pseudomaculata 321
targeted on-host control 341 Triatominae 309, 310
temporal coding 73 triatomine 69, 315, 317, 319, 320, 321, 356
tent trap 254 –– bug 144, 309, 311
terpene 46, 207, 208, 314 –– trap 321
terpinen-4-ol 314, 318 tridecanoic acid 153, 159, 161, 163
terpineol 51 2,4,5-trimethylthiazole 48
–– α(+) 314 Trypanosoma 265, 277
tetradecanal 165 –– brucei 282
tetradecanoic acid 151, 153, 159, 160, 164, –– brucei gambiense 266, 282, 350
165 –– brucei rhodesiense 282, 350
Thailand 368 –– congolense 282, 352, 355
The Gambia 191, 193 –– cruzi 309, 310, 318, 350, 356
theileriosis 327, 336 –– vivax 282, 355
The Netherlands 220 trypanosomiasis 13, 161, 265, 266, 275, 278,
thermal cue 117 279, 282, 283, 291
thermoreceptor 126, 340 tsetse 14, 117, 118, 119, 121, 123, 124, 125,
thigmotaxis 311, 312 132, 135, 144, 161, 167, 207, 230, 236,
α-thujone 48, 51, 70, 72 237, 253, 261, 265, 269, 270, 291, 292,
tick 14, 69, 74, 327 296, 297, 298, 299, 301, 303, 304, 336,
–– -borne disease 13 355, 357, 361, 402, 403, 404, 405
–– -borne encephalitis 327 –– control 282
–– climbing bioassay 338 tuning 69, 70, 71
–– repellent 342 –– curve 71, 77
tissue, olfactory 21 –– profile 71, 79
Togo 183, 185, 186 Tunisia 208

Olfaction in vector-host interactions 435

Index

turbulence 116, 122, 125 –– contact 256

turbulent –– cue 46, 117, 121, 125, 126, 132, 134, 136,
–– diffusion 120 150, 292
–– flow 124 –– resolution 121
–– plume 157 –– stimulus 117, 130, 247
–– presentation 129 voltinism 238
–– stream 109 Vulpes vulpes 206

U W
Uganda 183, 185, 186, 275, 282 Wahlund effect 281
Ulomyia fuliginosa 203 warthog 268, 269, 272, 276
undecanal 165, 338 water 48
undecanoic acid 151, 153, 163 waterbuck 272, 273
2-undecanone 70, 72 water buffalo 223, 230
United Kingdom 220, 235 waterfowl 259
upwind weather
–– anemotaxis 134, 252, 271, 295, 296, 298, –– effect of 226
301 West Africa 247
–– flight 381 West Nile 13, 17, 39, 103, 144, 146, 165, 384
–– flight behaviour 401 white-tailed deer 336, 337
–– movement 295 Wilhelmia equina 257
uric acid 334, 340 Wilton trap 367, 370
urine 192, 235, 336, 351 wind
–– odour 313 –– speed 226
uropygial –– tunnel 120, 121, 122, 126, 128, 131, 135,
–– gland 254, 260 136, 143, 148, 152, 157, 159, 232, 258,
–– odour 250, 259 269, 294, 295, 296, 335, 381, 401, 402
Ursus americanus 249 Wuchereria bancrofti 350, 355
Uruguay 311 Wyomia smithii 41
USA 221, 222, 223, 234, 235
X
V xanthine 334, 339
11-cis vaccenyl acetate 67 Xenopus 27, 29, 31, 146
vacuum distillation 228
γ-valerolactone 331, 332 Y
Varanus niloticus niloticus 276 yeast-baited trap 321
vasodilator 353, 358 yeast odour 320
Vavoua trap 299, 301 yellow fever 17, 39, 354, 365, 368, 401, 402
vector Yersinia pestis 357, 359
–– behaviour, changes in 356 Y-tube – See: olfactometer, Y-tube
–– -borne disease 106, 143, 144, 181
–– control 14, 32, 405 Z
–– longevity 350 z-4-decanal 152, 156
–– manipulation 359 Zambia 183, 185, 186, 268, 278
vectorial capacity 182, 203, 360 Zanzibar 184
ventilated pits 268 zebra 273
Vesicular Stomatitis Virus 249 Zimbabwe 183, 184, 185, 186, 192, 268, 270,
visual 273, 278, 279, 280, 300, 301, 303, 341

436  Olfaction in vector-host interactions

 Index

zoonosis 310
zoophagic 184
zoophilic 43, 129, 162, 183, 184, 186

Olfaction in vector-host interactions 437