Advances in Experimental Medicine and Biology 1117

Katsumi Matsuzaki Editor

Antimicrobial
Peptides
Basics for Clinical Application
Advances in Experimental Medicine
and Biology

Volume 1117

Editorial Board
IRUN R. COHEN, The Weizmann Institute of Science, Rehovot, Israel
ABEL LAJTHA, N.S. Kline Institute for Psychiatric Research, Orangeburg,
NY, USA
JOHN D. LAMBRIS, University of Pennsylvania, Philadelphia, PA, USA
RODOLFO PAOLETTI, University of Milan, Milan, Italy
NIMA REZAEI, Tehran University of Medical Sciences, Children’s Medical
Center Hospital, Tehran, Iran
More information about this series at http://www.springer.com/series/5584
Katsumi Matsuzaki
Editor

Antimicrobial Peptides
Basics for Clinical Application
Editor
Katsumi Matsuzaki
Graduate School of Pharmaceutical Sciences
Kyoto University
Sakyo-ku, Kyoto, Japan

ISSN 0065-2598     ISSN 2214-8019 (electronic)

Advances in Experimental Medicine and Biology
ISBN 978-981-13-3587-7    ISBN 978-981-13-3588-4 (eBook)
https://doi.org/10.1007/978-981-13-3588-4

Library of Congress Control Number: 2019933895

© Springer Nature Singapore Pte Ltd. 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore
189721, Singapore
Preface

Antimicrobial peptides (AMPs) discovered about 30 years ago are responsi-

ble for part of the innate immunity of animals and plants, and almost 3000
peptides have been reported so far. Previously, AMPs were thought to possess
only antimicrobial activity, but subsequent research revealed that they also
exhibit various properties including immune-modulating, anticancer, antibio-
film, and cell-penetrating activities.
The emergence of multidrug-resistant bacteria and new pathogens is a
threat to human health. Therefore, the development of novel antimicrobial
agents is a pressing need. AMPs have been considered promising candidates
for new therapeutics to combat this problem because their antimicrobial spec-
tra are broad and the development of bacterial resistance against them is dif-
ficult compared to conventional antibiotics.
This book of 15 chapters gives an overview of AMPs, how they work,
what activities they have other than antimicrobial activity, how to design and
discover them, what cautions should be taken into consideration before com-
mercialization, and examples of clinical applications. Each chapter is written
by leading scientists in that field. I appreciate their contributions very much.
Now that commercial development of AMPs has been reignited, it is a
good time to publish this book to celebrate the 30-year anniversary of the
discovery of AMPs. I hope this book will provide scientists in both academia
and industry with the basic and comprehensive knowledge needed to develop
AMPs of clinical use.

Kyoto, Japan Katsumi Matsuzaki

v
Contents

Part I Introduction

1 Antimicrobial Peptides of Multicellular Organisms:

My Perspective ������������������������������������������������������������������������������    3
Michael Zasloff

Part II Mechanisms of Antimicrobial Action

2 Membrane Permeabilization Mechanisms����������������������������������    9

Katsumi Matsuzaki
3 Elementary Processes and Mechanisms of Interactions
of Antimicrobial Peptides with Membranes—Single
Giant Unilamellar Vesicle Studies—��������������������������������������������   17
Moynul Hasan and Masahito Yamazaki
4 The Mechanisms of Action of Cationic Antimicrobial
Peptides Refined by Novel Concepts
from Biophysical Investigations����������������������������������������������������   33
Christopher Aisenbrey, Arnaud Marquette,
and Burkhard Bechinger
5 Anionic Lipid Clustering Model ��������������������������������������������������   65
Richard M. Epand
6 Intracellular Antimicrobial Peptides Targeting
the Protein Synthesis Machinery��������������������������������������������������   73
Michael Graf and Daniel N. Wilson

Part III Other Activities of AMPs

7 Antimicrobial and Cell-Penetrating Peptides:

How to Understand Two Distinct Functions
Despite Similar Physicochemical Properties ������������������������������   93
Ines Neundorf

vii
viii Contents

8 Synthetic Anti-lipopolysaccharide Peptides (SALPs)

as Effective Inhibitors of Pathogen-Associated
Molecular Patterns (PAMPs)�������������������������������������������������������� 111
Wilmar Correa, Lena Heinbockel,
Guillermo Martinez -de- Tejada, Susana Sánchez,
Patrick Garidel, Tobias Schürholz, Walter Mier,
Aline Dupont, Mathias Hornef, Thomas Gutsmann,
Karl Mauss, Günther Weindl, and Klaus Brandenburg
9 Anticancer Activities of Natural and Synthetic Peptides ���������� 131
A. L. Hilchie, D. W. Hoskin, and M. R. Power Coombs
10 Antimicrobial Host Defence Peptides: Immunomodulatory
Functions and Translational Prospects���������������������������������������� 149
Anne M. van der Does, Pieter S. Hiemstra,
and Neeloffer Mookherjee

Part IV Towards Clinical Applications

11 Selectivity of Antimicrobial Peptides:

A Complex Interplay of Multiple Equilibria������������������������������ 175
Sara Bobone and Lorenzo Stella
12 Design of Antimicrobial Peptides:
Progress Made with Human Cathelicidin LL-37������������������������  215
Guangshun Wang, Jayaram Lakshmaiah Narayana,
Biswajit Mishra, Yingxia Zhang, Fangyu Wang,
Chunfeng Wang, D. Zarena, Tamara Lushnikova,
and Xiuqing Wang
13 Application of Synthetic Molecular Evolution
to the Discovery of Antimicrobial Peptides���������������������������������� 241
William C. Wimley
14 AMPs as Anti-biofilm Agents for Human Therapy
and Prophylaxis������������������������������������������������������������������������������ 257
Hawraa Shahrour, Raquel Ferrer-Espada, Israa Dandache,
Sergio Bárcena-Varela, Susana Sánchez-Gómez, Ali Chokr,
and Guillermo Martínez-de-Tejada
15 Clinical Application of AMPs�������������������������������������������������������� 281
Fabíola Costa, Cátia Teixeira, Paula Gomes,
and M. Cristina L. Martins

Index�������������������������������������������������������������������������������������������������������� 299
 Part I
Introduction
 Antimicrobial Peptides
of Multicellular Organisms: My 1
Perspective

Michael Zasloff

Antimicrobial peptides of multicellular organ- erty of being amphiphilic: they are soluble in
isms were first characterized in the 1980s by aqueous environments but also can partition into
investigators who felt that known systems of lipid environments, such as membranes (Zasloff
immunity could not explain what they observed: 2002).
the resistance to bacterial infection of a Cecropia What makes them antimicrobial is the surpris-
moth pupa lacking antibodies or lymphocytes ing difference in the structure of the membranes
(cecropins (Steiner 1981)), the potent microbici- that surround microbes (most species of bacteria
dal activity of neutrophils from a rabbit (defen- and some species of fungi and protozoa) and the
sins (Selsted et al. 1985)), and the healing of a cells of multicellular plants and animals. For rea-
wound on the skin of the African clawed frog sons that remain unclear, the membranes that sur-
without infection in a non-sterile aquarium round many species of microbe display negatively
(magainins (Zasloff 1987)). Since then AMPs charged phospholipid headgroups (i.e., phospha-
have been discovered in diverse species of fungi, tidylglycerol, cardiolipin) to their outside world.
plants, and animals (Seshadri Sundararajan et al. In contrast, the membranes of the cells that com-
2012; Fan et al. 2016; Waghu et al. 2016; Wang prise the multicellular organism, in general, place
et al. 2016). It is likely that we will discover that zwitterionic phospholipids (i.e., phosphatidyl-
every multicellular organism expresses antimi- choline) on the outer leaflet of their plasma mem-
crobial peptides as a key element of their immune brane and segregate phospholipids with anionic
system. Why are antimicrobial peptides so popu- headgroups (i.e., phosphatidylserine, phosphati-
lar in Nature? dylinositol polyphosphates) on the inner face of
The basic design of an AMP is exceedingly the plasma membrane. AMPs are attracted elec-
simple: a short peptide synthesized from the 20 trostatically to an accessible membrane with a
usual amino acids that can organize into a sec- strong negative surface charge, a feature that
ondary structure in which hydrophilic and hydro- characterizes microbes. However, this selectivity
phobic amino acids are segregated spatially and is only relative, in the sense that AMPs will, due
in which the peptide has a net cationic charge. to their fundamental detergent-like properties,
Peptides of this type have the remarkable prop- eventually damage the membranes of their host’s
cells if the concentrations are increased suffi-
M. Zasloff (*) ciently. This becomes an issue in the develop-
MedStar Georgetown Transplant Institute, ment of AMPs as anti-infective therapeutics.
Georgetown University School of Medicine, In the case of Gram-negative bacteria, which
Washington, DC, USA are enclosed by an outer membrane in addition to
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 3

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_1
4 M. Zasloff

an inner membrane surrounding the cytoplasm, into phagolysosomes in which microbes have
AMPs initially interact electrostatically with the been sequestered. Neuropeptides produced by
anionic lipopolysaccharide outer membrane, dis- the nervous system are amphiphilic, since they,
rupt its structure, and gain access to the inner. In like AMPs, are secreted from nerve endings and
the case of Gram-positive bacteria, AMPs are then captured onto membrane receptors of neigh-
attracted electrostatically to the anionic teichoic boring cells. It isn’t surprising then that certain
acids that decorate the cellular proteoglycan neuropeptides exhibit antimicrobial activity, and
envelope, accumulate on the surface, and then that by this means, the nervous system can pro-
diffuse inward onto the inner cytoplasmic mem- vide immune support in certain contexts (Brogden
brane. Following the initial interaction with the et al. 2005; Augustin et al. 2017).
inner membrane, AMPs organize into simple Because there are no specific sequence con-
polymeric structures. In most cases the function straints on the design of an AMP, other than the
of the membrane is irreversibly disturbed, lead- requirement to adopt an amphiphilic secondary
ing to loss of cellular contents, dissipation of structure, an enormous diversity of AMPs is seen
membrane potential, and rapid cellular death across Nature. That being said, a single amino
(Brogden 2005). Certain AMPs exhibit an affin- acid substitution in an AMP, while not profoundly
ity for a specific membrane lipid, such as lipids I altering the biophysical properties of the mole-
and II in the case of Gram-positive bacteria, and cule, can dramatically alter its antimicrobial
inhibit proteoglycan synthesis (Schneider et al. spectrum (Seshadri Sundararajan et al. 2012; Fan
2010), or galactosylceramide, in the case of et al. 2016; Waghu et al. 2016; Wang et al. 2016).
fungi, and induce apoptosis (Thevissen et al. In most instances, the mechanism underlying the
2004). In addition, some AMPs flip from the change in specificity is not understood and likely
outer to the inner face of the plasma membrane involves specific interactions between the AMP
and then into the cytoplasm where they disturb and the microbial membrane that cannot be deci-
the health of the cell by binding electrostatically phered using current biophysical techniques. As a
to a vital target (Park et al. 2000). consequence when organisms find themselves
By virtue of their secondary structure and facing microbes against which their existing
their targeting of the microbial membrane AMPs are inadequate, effective AMPs can often
through electrostatic interaction, AMPs generally evolve within the threatened population through
exhibit effective microbicidal concentrations in simple amino acid substitutions.
the low micromolar range. To achieve these con- The simple mechanism by which AMPs kill
centrations, AMPs are generally directed to pro- their targets has another striking benefit: It is very
vide antimicrobial defense in close proximity to difficult for a microbe to evade the attack by
the cell from which they are secreted. On dry skin AMPs on its membrane. Such a mutation would
surfaces, they create an antimicrobial barrier that require a change in lipid composition or organi-
can help shape the epidermal microbiome. On zation. Since the AMP is constructed from ordi-
wet mucosal surfaces, they are secreted from the nary amino acids with no specific primary
epithelium into the thin aqueous biofilm in direct sequence signature, the microbe would have a
contact with the epithelium and underneath the tough task of protecting itself with a protease that
overriding mucous layer in contact with the could cleave the AMP, but not any of its own pro-
lumen or cavity. In this setting, AMPs serve to teins. Furthermore, all multicellular organisms
prevent microbes from gaining access to the epi- deploy a cocktail of AMPs in each physiological
thelial layer, killing them rapidly should they context, tuned to cover the full spectrum of
penetrate the physical barrier posed by the microbes likely to be encountered in the particu-
mucous layer. AMPs that are produced by hema- lar niche.
topoietic cells are at the service of the particular One should appreciate that a healthy multicel-
cell in which they are carried, which either lular organism presents a variety of ecological
releases them at a site of infection or directs them niches to the microbes in its environment. These
1 Antimicrobial Peptides of Multicellular Organisms: My Perspective 5

niches have relatively stable characteristics, AMPs at the microbe, and the AMPs released can
maintained by active homeostatic physiological be tuned to exhibit selectivity for the microbes
mechanisms, and within an environment certain likely to be encountered in that niche as well as
microbes will gain a foothold over others. For maintain an AMP concentration that will not
example, in the absence of an effective antimi- damage the host’s cells within that niche. From
crobial defense, Pseudomonas aeruginosa will this perspective, the development of AMP thera-
become the dominant Gram-negative bacterial peutics should be informed by the physiological
species in the moist mucus-lined bronchial tubes context of the infection to be treated.
of the human airway, and it is against this organ- The discovery of AMPs has led to a deeper
ism that the AMPs normally secreted from the appreciation of the importance of innate immu-
bronchial epithelium of a healthy person are nity as an arm of vertebrate defense and provided
directed. This defense fails in cystic fibrosis due insight into the immune defenses of the vast
to a failure of the epithelium to maintain a pH in majority of plants and animals that now exist.
which the cocktail of AMPs can work effectively Much remains to be explored.
(Pezzulo et al. 2012). Pseudomonas then invades
the bronchial epithelium, provoking the body to
direct all of its backup inflammatory immune References
defenses to prevent further tissue invasion. A pro-
found destructive inflammatory process unfolds. Augustin R et al (2017) A secreted antibacterial neuropep-
tide shapes the microbiome of Hydra. Nat Commun
In the human ileum, the most distal segment of 8(1):698
the small intestine, thousands of species of bacte- Brogden KA (2005) Antimicrobial peptides: pore formers
ria are effectively contained in the lumen, sepa- or metabolic inhibitors in bacteria? Nat Rev Microbiol
rated from the interior wall of the bowel by a 3(3):238–250
Brogden KA et al (2005) The nervous system and innate
single-celled epithelium. AMPs secreted from immunity: the neuropeptide connection. Nat Immunol
the Paneth cells and from the enterocytes, along 6(6):558–564
with mucous secreted from goblet cells, deter the Fan L et al (2016) DRAMP: a comprehensive data reposi-
luminal microbes from attaching to the epithe- tory of antimicrobial peptides. Sci Rep 6:24482
Park CB et al (2000) Structure-activity analysis of buforin
lium. In Crohn’s disease Paneth cells fail to func- II, a histone H2A-derived antimicrobial peptide: the
tion properly, weakening the antimicrobial proline hinge is responsible for the cell-­penetrating
barrier and permitting certain microorganisms to ability of buforin II. Proc Natl Acad Sci U S A
establish a foothold within the fluid layer cover- 97(15):8245–8250
Pezzulo AA et al (2012) Reduced airway surface pH
ing the epithelium, eventually invading the layer impairs bacterial killing in the porcine cystic fibrosis
and provoking an inflammatory process as lung. Nature 487(7405):109–113
described in cystic fibrosis (Wehkamp et al. Schneider T et al (2010) Plectasin, a fungal defensin, tar-
2005). gets the bacterial cell wall precursor Lipid II. Science
328(5982):1168–1172
Finally, a word about the development of Selsted ME et al (1985) Primary structures of six antimi-
AMPs as therapeutics. Because of their simple crobial peptides of rabbit peritoneal neutrophils. J Biol
design, many AMPs have been synthesized dif- Chem 260(8):4579–4584
fering in sequence, length, and amino acid com- Seshadri Sundararajan V et al (2012) DAMPD: a manu-
ally curated antimicrobial peptide database. Nucleic
position. Molecules that exhibit amphiphilic Acids Res 40(Database issue):D1108–D1112
properties and antimicrobial activity but are not Steiner H et al (1981) Sequence and specificity of two
themselves peptides have been designed. Very antibacterial proteins involved in insect immunity.
few of these molecules have been effective as Nature 292(5820):246–248
Thevissen K et al (2004) Defensins from insects and
systemically administered anti-infectives. AMPs plants interact with fungal glucosylceramides. J Biol
evolved, in most multicellular organisms, to Chem 279(6):3900–3905
defend a microenvironment and not to be deliv- Waghu FH et al (2016) CAMPR3: a database on
ered into the systemic circulation. The organism sequences, structures and signatures of antimicrobial
peptides. Nucleic Acids Res 44(D1):D1094–D1097
can direct a high concentration of a cocktail of
6 M. Zasloff

Wang G et al (2016) APD3: the antimicrobial peptide Zasloff M (1987) Magainins, a class of antimicrobial pep-
database as a tool for research and education. Nucleic tides from Xenopus skin: isolation, characterization of
Acids Res 44(D1):D1087–D1093 two active forms, and partial cDNA sequence of a pre-
Wehkamp J et al (2005) Reduced Paneth cell alpha-­ cursor. Proc Natl Acad Sci U S A 84(15):5449–5453
defensins in ileal Crohn’s disease. Proc Natl Acad Sci Zasloff M (2002) Antimicrobial peptides of multicellular
U S A 102(50):18129–18134 organisms. Nature 415(6870):389–395
 Part II
Mechanisms of Antimicrobial Action
 Membrane Permeabilization
Mechanisms 2
Katsumi Matsuzaki

Abstract bility in minutes (Matsuzaki et al. 1997a). Rapid

Many antimicrobial peptides are considered to killing is an important self-defense action of
kill microbes by permeabilizing cell mem- AMPs. Observations that enantiomeric peptides
branes. This chapter summarizes the driving composed of all D-amino acids are equipotent to
force of peptide binding to membranes; various the parent L-peptides indicate that proteins
mechanisms of lipid bilayer permeabilization requiring chiral recognition such as receptors and
including the barrel-stave, toroidal pore, and enzymes are not involved in the membrane per-
carpet models; and modes of permeabilization meabilization process (Bessalle et al. 1990; Wade
of bacterial and mammalian membranes. et al. 1990). Furthermore, magainins induce the
leakage of water-soluble dyes entrapped in artifi-
Keywords cial lipid vesicles (Matsuzaki et al. 1989, 1991a),
Membrane permeabilization · Membrane suggesting that the lipid matrix of membranes is
binding · Membrane curvature · Barrel-stave a target of the peptides. Many AMPs such as
model · Toroidal pore model · Carpet model tachyplesin I (Matsuzaki et al. 1991b) and LL-37
(Lee et al. 2011) also permeabilize membranes.

2.1 Introduction 2.2 Membrane Binding

Membrane permeabilization as a mechanism for The first step in the membrane permeabilization
bacterial killing has already been suggested in a process is the binding of peptides to membranes.
paper on the discovery of magainins from the AMPs are generally polycationic and amphipa-
African clawed frog Xenopus laevis, an arche- thic. In many cases, linear peptides take unor-
typical antimicrobial peptide (AMP) isolated dered structures in an aqueous solution, whereas
from a vertebrate for the first time (Zasloff 1987). they are conformed to be amphipathic secondary
As shown in Fig. 2.1, the addition of magainin 2 structures, typically α-helices or β-strands, upon
to E. coli cells induced an efflux of intracellular membrane binding (Matsuzaki et al. 1989,
K+ ions and concomitantly a decrease in cell via- 1991a). Such amphipathic structures fit the mem-
brane–water interface. Even conformationally
K. Matsuzaki (*) restricted cyclic peptides also change their
Graduate School of Pharmaceutical Sciences, ­structures to accommodate themselves to mem-
Kyoto University, Sakyo-ku, Kyoto, Japan brane environments (Imura et al. 2007).
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 9

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_2
10 K. Matsuzaki

The main driving forces for membrane binding dylserine, are essentially sequestered on the cyto-
are electrostatic attraction and hydrophobic inter- plasmic face of the membrane, although cancer
action (Matsuzaki et al. 1995a, 2009). Positively cells show an increased exposure of phosphatidyl-
charged peptides preferentially interact with nega- serine (Utsugi et al. 1991). However, gangliosides
tively charged membranes. Bacterial membranes containing sialic acid residues have some negative
are rich in acidic phospholipids (phosphatidylg- charges on the cell surface, which are a target of
lycerol and cardiolipin). Furthermore, cell walls AMPs (Miyazaki et al. 2012). Interestingly, acidic
containing lipopolysaccharides (LPS) and pepti- phospholipids and gangliosides interact differently
doglycans are also negatively charged. In contrast, with AMPs (Fig. 2.2). Cationic peptides specifi-
mammalian cell membranes are less negatively cally bind to gangliosides containing anionic sialic
charged. Acidic phospholipids, such as phosphati- acid residues so that the charge is neutralized. The
binding is described by the Langmuir-­type equa-
tion. Fluorescent resonance transfer experiments
clearly revealed that AMPs preferentially inter-
acted with monosialoganglioside GM1 compared
with phosphatidylcholine in a GM1/phosphatidyl-
choline mixed bilayer mimicking mammalian cell
membranes. In contrast, peptides equally interact
with anionic phosphatidylglycerol and zwitter-
ionic phosphatidylcholine in a phosphatidylglyc-
erol/phosphatidylcholine mixed membrane, a
model for bacterial cell membranes (Miyazaki
et al. 2012). The interaction is theoretically
explained by a combination of the Gouy-Chapman
Fig. 2.1 Correlation between membrane permeabiliza- theory (electrostatic c­oncentration immediately
tion and bacterial death. Magainin 2 (50 nmol) was added above the membrane surface) and a partition equi-
to E. coli cells (5 × 108 CFU/mL). K+ efflux (open circles,
left axis) and the percent cell viability (closed circles,
librium (Wenk and Seelig 1998; Wieprecht et al.
right axis) are plotted as a function of time 1999).

Fig. 2.2 Different binding modes of AMPs to (a) mam- electrostatically concentrated immediately above the
malian and (b) bacterial model membranes. (a) Cationic membrane surface according to the Gouy-Chapman the-
AMPs specifically bind to gangliosides containing anionic ory and then partitioned into the membrane. There is no
sialic acid residues. The binding is described by the specific interaction between AMPs and acidic
Langmuir-type equation. (b) In contrast, peptides are phospholipids
2 Membrane Permeabilization Mechanisms 11

2.3 Permeabilization of Model lipids by cationic AMPs and is described in Chap.

Membranes 5. The “barrel-stave channel,” “toroidal pore,”
and original “carpet” mechanisms are catego-
Membrane-bound AMPs change membrane rized as the former mechanism (Fig. 2.3). Refer
structures and organizations leading to mem- to Chap. 3 and a review by Huang (2006) for
brane permeabilization. Using model membranes theoretical treatments.
such as liposomes, several mechanisms so far Amphipathic secondary structures on the sur-
proposed for this can be roughly classified into face of membranes can modulate membrane cur-
two categories, i.e., membrane curvature modula- vature. The peptides expand the interfacial
tion and phase separation. The recently proposed region, making a void in the hydrocarbon region
latter mechanism includes clustering of acidic of the membrane. Consequently, membrane-­

Fig. 2.3 Major mechanisms of membrane permeabiliza- wall. The pore is not stable (the lifetime is typically ms),
tion induced by AMPs. (a) AMPs form amphipathic sec- and upon its disintegration a fraction of peptide molecules
ondary structures (α-helix in this case) on the surface of translocates across bilayers. At much higher peptide-to-
membranes. The peptides expand the interfacial region lipid ratios, membranes are solubilized into micelles. (b)
making a void in the hydrocarbon region of the membrane. When membranes have a negative-­curvature (convex) ten-
Consequently, membrane-thinning and positive curvature dency, it counteracts the positive curvature induced by the
(concave) strain are induced (broken line). At threshold peptides, stabilizing the peptide–lipid system and allowing
peptide-to-lipid ratios, typically below ~1:100, either the accumulation of large amounts of peptides. Eventually
“barrel-stave channel” or the “toroidal pore” is formed. (e.g., peptide-to-lipid ratios above 1:10), membrane dis-
The former is solely composed of peptides, making a ruption occurs. This mechanism corresponds to the origi-
water-filled channel. The ionic current is discrete and the nal “carpet model.” It should be noted that the modified
conductance depends on the number of peptides involved carpet model also includes toroidal pore formation,
in a single channel. A typical example of this class of pep- although its scientific validity needs careful consideration.
tide is the peptaibol alamethicin. However, most AMPs Similar phenomena can happen in zero-curvature bilayers
form a toroidal pore, in which both the polar faces of the with peptides having a large hydrophobic surface capable
amphiphilic structures and the polar headgroups of lipids of expanding the hydrocarbon core of the membrane.
constitute the pore wall. This unique structure allows not Thus, the mechanism of membrane permeabilization is not
only the passage of ions and small molecules through the unique to peptides but also depends on the physicochemi-
pore but also the rapid flip–flop of lipids along the pore cal properties of membranes
12 K. Matsuzaki

thinning and positive curvature (concave) strain number of pores decreased with increasing time
are induced (Fig. 2.3a left). At threshold peptide-­ (Fig. 2.5). We hypothesized that pores were
to-­
lipid ratios, typically below ~1:100, the unstable and a fraction of peptide molecules was
surface-­lying peptides are cooperatively inserted translocated into the inner leaflet. Thus, the pep-
into the membrane, forming a water-filled pore tide density in the outer leaflet decreased, decel-
(Fig. 2.3a right). Peptaibols, peptides containing erating pore formation. Indeed, the translocation
Aib (aminoisobutyric acid) residues with the was coupled to the leakage. Kinetic analysis sug-
C-terminal alcohol, have been known to form a gested that a pore is composed of ~5 magainin
barrel-stave channel (Sansom 1991). These pep- molecules, which appeared to be too small to
tides are electrostatically almost neutral. A typi- allow the passage of calcein (Matsuzaki et al.
cal example is the antibiotic alamethicin produced 1995b). Thus, we hypothesized that lipid mole-
by the fungus Trichoderma viride. The channel is cules were also involved in the pore structure and
solely composed of helical peptides. The ionic examined the flip–flop of lipids. The flop was
current is discrete and the conductance depends again coupled to leakage and translocation
on the number of peptides involved in a single (Fig. 2.5). The flip rate was identical to the flop
channel. Magainins were also considered to form rate and did not depend on the type of lipid, sug-
this type of channel at the time of its discovery. gesting that all lipid molecules in the membrane
However, we noticed that this was not the case were randomized (Matsuzaki et al. 1996a). Based
and proposed the toroidal pore model in 1996 on these observations, we proposed the toroidal
(Matsuzaki et al. 1996a). pore model, in which both the polar faces of the
First, in the case of the barrel-stave channel, amphiphilic helices and the polar headgroups of
the size of the pore depends on the peptide-to-­ lipids constitute the pore wall. This unique struc-
lipid molar ratio. However, as shown in Fig. 2.4, ture allows not only the passage of ions and small
the small I− ion and the medium-sized calcein molecules through the pore but also the rapid
dye leaked out of liposomes in the same range of flip–flop of lipids along the pore wall. Both types
peptide-to-lipid ratio, and larger dye-labeled dex- of events seem to contribute to bactericidal activ-
tran molecules were retained even at higher ity. The flop of phosphatidylserine was also
peptide-­to-lipid ratios, suggesting that a pore of observed in mammalian cells (Imura et al. 2008).
defined size (diameter 2–3 nm) was formed. The structure of the toroidal pore was confirmed
Second, the leakage kinetics were unique. The by neutron scattering experiments by Huang’s
percent leakage value appeared to reach a plateau group (Ludtke et al. 1996).
instead of complete leakage, indicating that the

Fig. 2.4 Estimation of the pore size formed by magainin Fig. 2.5 Coupling between leakage, peptide transloca-
2. The percent leakage values are plotted as a function of tion, and lipid flop. The percent leakage of calcein (solid
the peptide-to-lipid ratio for I− ions (open circles), calcein line), percent peptide translocation (open circles), and
(MW 623, closed circles), and FITC-dextran (MW 4400, percent flop of NBD-phosphatidylethanolamine (closed
closed squares) circles) are plotted as a function of time
2 Membrane Permeabilization Mechanisms 13

Not only magainins but also other AMPs and carpet and disrupt the bilayer organization (Shai
membrane-acting peptides form toroidal pores, 1995) (Fig. 2.3b). This model was proposed fol-
including helical PGLa (Matsuzaki et al. 1998a), lowing studies on the interaction between cecropins
mastoparan X (Matsuzaki et al. 1996b), melittin and dermaseptins with phosphatidylserine-­
(Yang et al. 2001), buforin 2 (Kobayashi et al. containing bilayers. We found that even magainins
2004), and cyclic β-sheet tachyplesin I (Imura do not form toroidal pores, but accumulate on the
et al. 1768). The pore size depends on the type of membrane surface and eventually disrupt it at pep-
peptide. The two substitutions F5Y and F16W tide-to-lipid ratios above 1:10 in membranes con-
for magainin 2 enlarged the pore size to 4–7 nm taining phosphatidylserine (Fig. 2.6), phosphatidic
(Hara et al. 2001). The bacteriocin lacticin Q also acid, or cardiolipin (Matsuzaki et al. 1998b). These
forms a large toroidal pore of similar size lipids form hexagonal-II phases under charge-neu-
(Yoneyama et al. 2009). tralizing conditions and thus tend to induce nega-
The mode of dye leakage is classified into an tive curvature strain on the membrane, counteracting
all-or-none or graded mode. In the former, some the positive curvature strain imposed by magainins
vesicles are empty, whereas the rest are intact. In and allowing accumulation of huge amounts of
the latter, all vesicles partially lose their contents. peptides on the membrane. Thus, the mechanism
Although these modes are sometimes considered of membrane permeabilization is not unique to
to originate from different leakage mechanisms, peptides, but also depends on the physicochemical
earlier theoretical studies showed that the pore properties of membranes. It should be noted that
lifetime τ determines the mode of leakage the current modified carpet model also includes
(Schwarz and Arbuzova 1995; Schwarz and toroidal pore formation (Oren and Shai 1998),
Robert 1992). The all-or-none and the graded although its scientific validity needs careful consid-
mode correspond to τ >> τ0 (the time necessary eration. Recently, the detergent-like model
for a 1/e reduction in the intravesicular dye con- (Bechinger and Lohner 2006) and the interfacial
centration) and τ << τ0, respectively. activity model (Wimley 2010) have been proposed
The pore lifetime is governed by electrostatics for the mechanism of AMPs. These more compre-
within the pore and is therefore modulated by hensive models are similar to the curvature modu-
both peptide charge and lipid composition lation mechanism described above in that the
because the toroidal pore is composed of pep- incorporation of amphipathic structures in lipid
tides and lipids. An increase in peptide positive bilayers modifies the bilayer organization.
charge destabilizes the pore because of enhanced
electrostatic repulsion between closely spaced
peptides, facilitating peptide translocation
(Matsuzaki et al. 1997b). Note that translocation
occurs only upon the disintegration of the pore.
For example, an increase in the positive charge of
magainin from +4 to +6 reduced the τ value from
9τ0 to 0.1τ0. Buforin 2 with +6 charges is translo-
cated across liposomal (Kobayashi et al. 2000)
and bacterial (Park et al. 1998) membranes with-
out permeabilizing them. Similarly, an increase
in acidic lipid content stabilizes the pore by
reducing electrostatic repulsion between peptides
in a pore (Kobayashi et al. 2004). Dye leakage Fig. 2.6 Dependence of magainin 2 leakage activity on
and lipid flip–flop are observed in this case. lipid species. The percent leakage values are plotted as a
function of the peptide-to-lipid ratios in a logarithmic
Another well-known mechanism of membrane scale for palmitoyloleoylphosphatidylglycerol (POPG,
permeabilization is the original “carpet model,” in closed circles) and palmitoyloleoylphosphatidylserine
which peptides cover the membrane surface like a (POPS, open circles)
14 K. Matsuzaki

2.4 Permeabilization of Bacterial more susceptible populations. The experimental

Membranes conditions in Fig. 2.1 may be appropriate for dis-
cussion because about 50% of bacteria are killed.
AMPs have been proposed to disrupt and perme- Assuming that (1) peptide molecules are intact
abilize the outer membrane of Gram-negative and completely bound to both leaflets of the outer
bacteria by the “self-promoted pathway” and inner membranes, (2) the binding to other
(Hancock and Chapple 1999). Cationic peptides bacterial components is negligible, and (3) the
compete with Mg2+ ions that bridge between number of lipids per cell is 4 × 107 (Roversi et al.
adjacent phosphates of LPS. Magainin forms a 2014), the peptide-to-lipid ratio in the inner
helix upon binding to the lipid A moiety of LPS membrane is ~1. To examine the direct interac-
(Matsuzaki et al. 1999) and induces blebs on the tion of magainin 2 with inner membranes, we
outer membrane, while the presence of Mg2+ ions used E. coli spheroplasts lacking outer mem-
inhibits its antimicrobial activity (Matsuzaki branes (Matsuzaki et al. 1997a). The peptide
et al. 1997a). The permeabilization of the outer lysed spheroplasts in a peptide-to-lipid ratio
membrane can be detected by the permeation of range of 0.1–1, in accordance with the rough esti-
impermeable substances such as nonionic deter- mate above. The lipid composition of the sphero-
gents (Matsuzaki et al. 1997a) and plasts was 68% phosphatidylethenolamine/24%
1-N-phenylnaphthylamine (NPN) (Zhang et al. phosphatidylglycerol/8% cardiolipin. Asymmetry
1999). Recently, a method based on the leakage of lipid distribution in the inner membrane has
of green fluorescent protein (GFP) expressed in not yet been observed (Furse and Scott 2016).
the periplasmic space was developed (Sochacki Nevertheless, even if the inner leaflet is com-
et al. 2011). Note that the diameter of the protein posed of phosphatidylethenolamine alone, the
(4.5 nm) is larger than that of the magainin toroi- outer leaflet is composed of ~50% phosphatidyl-
dal pore (2–3 nm). glycerol and ~50% lipids with a negative-­
The permeabilization of cytoplasmic mem- curvature tendency (cardiolipin and
branes can be monitored by several techniques. phosphatidylethenolamine). Thus, the original
The efflux of intracellular K+ ions is detected by carpet mechanism is more likely, although fur-
a K+-selective electrode (Matsuzaki et al. 1997a). ther studies are needed to confirm this.
Potential sensitive dyes, such as diSC35, detect Transmission electron microscopy revealed that
the dissipation of transmembrane potential LL-37 caused local perturbations and breaks
(Patrzykat et al. 2002). The hydrolysis of along P. aeruginosa cell membranes (Andersson
o-nitrophenyl-β-D-galactoside (ONPG) by cyto- et al. 2004). However, the mechanism of mem-
plasmic β-galactosidase is also often utilized for brane permeabilization may depend on the type
E. coli ML-35 (lacI−, lacY−, lacZ+). More conve- of AMPs, as well as bacterial strains. Magainin
nient methods include observation of the cell appears to form a toroidal pore against the Gram-­
entry of water-soluble dyes such as calcein positive B. megaterium because the peptide forms
(Imura et al. 2008) and Sytox Green (Sochacki membrane lesions of definite size (~2.8 nm,
et al. 2011) by fluorescent microscopy. <6.6 nm) (Imura et al. 2008).
Which mechanism works better for the inter-
action of AMPs with bacterial membranes is a
matter of attention (Wimley 2010; Roversi et al. 2.5 Permeabilization
2014). Generally, the amount of peptide bound to of Mammalian Cell
bacterial cells at its minimum inhibitory or bacte- Membranes
ricidal concentration (MIC or MBC) is consid-
ered. However, it should be noted that MIC or Relatively little is known about how AMPs per-
MBC is the concentration needed to inhibit the meabilize and are translocated across mamma-
growth of or to kill the most resistant population lian cell membranes. Our earlier work suggested
of bacteria and is thus too strong for the other that prototypical membrane-permeabilizing
2 Membrane Permeabilization Mechanisms 15

magainin 2 and intracellular-targeting buforin 2 Hancock REW, Chapple DS (1999) Peptide antibiotics.
Antimicrob Agents Chemother 43:1317–1323
interact with human cells differently (Takeshima Hara T, Kodama H, Kondo M, Wakamatsu K, Takeda A,
et al. 2003). Dye-labeled magainin was translo- Tachi T et al (2001) Effect of peptide dimerization
cated across HeLa cells via both energy-­ on pore formation: antiparallel disulfide-dimerized
dependent and energy-independent mechanisms, magainin 2 analog. Biopolymers 58:437–446
Huang HW (2006) Molecular mechanism of antimicro-
and the internalization was accompanied by cyto- bial peptides: the origin of cooperativity. Biochim
toxicity. In contrast, dye-labeled buforin pene- Biophys Acta 1758:1292–1302
trates cells in an energy-independent fashion and Imura Y, Nishida M, Ogawa Y, Takakura Y, Matsuzaki
exerts little toxicity. Zanetti’s group reported that K (2007) Action mechanism of tachyplesin I and
effects of PEGylation. Biochim Biophys Acta
the Pro-rich peptide Bac7 (1–35) is taken up by 1768:1160–1169
murine and human cells through a nontoxic Imura Y, Choda N, Matsuzaki K (2008) Magainin 2 in
energy- and temperature-dependent process action: distinct modes of membrane permeabiliza-
(Tomasinsig et al. 2006). Details on so-called tion in living bacterial and mammalian cells. Biophys
J 95:5757–5765
cell-penetrating peptide (CPP) are described in Kobayashi S, Takeshima K, Park CB, Kim SC, Matsuzaki
Chap. 7. K (2000) Interactions of the novel antimicrobial pep-
tide buforin 2 with lipid bilayers: proline as a translo-
cation promoting factor. Biochemistry 39:8648–8654
Kobayashi S, Chikushi A, Tougu S, Imura Y, Nishida
2.6 Conclusion M, Yano Y et al (2004) Membrane translocation
mechanism of the antimicrobial peptide buforin 2.
Thirty years of extensive studies, both experi- Biochemistry 43:15610–15616
mental and theoretical, have revealed several Lee C-C, Sun Y, Qian S, Huang HW (2011) Transmembrane
pores formed by human antimicrobial peptide LL-37.
basic mechanisms for the permeabilization of Biophys J 100:1688–1696
lipid bilayers by AMPs, as described above. It Ludtke SJ, He K, Heller WT, Harroun TA, Yang L, Huang
depends on both peptide sequences and lipid HW (1996) Membrane pores induced by magainin.
compositions. However, the molecular details are Biochemistry 35:13723–13728
Matsuzaki K (2009) Control of cell selectivity of
difficult to understand because initial membrane antimicrobial peptides. Biochim Biophys Acta
permeabilization is often dynamic and transient 1788:1687–1692
in nature, whereas spectroscopic methods usually Matsuzaki K, Harada M, Handa T, Funakoshi S, Fujii N,
detect equilibrium states after the permeabiliza- Yajima H et al (1989) Magainin 1-induced leakage of
entrapped calcein out of negatively-charged lipid vesi-
tion ceases. In contrast to model membrane stud- cles. Biochim Biophys Acta 981:130–134
ies, little is known on the permeabilization Matsuzaki K, Harada M, Funakoshi S, Fujii N, Miyajima
mechanisms of bacterial and mammalian cells, K (1991a) Physicochemical determinants for the inter-
which remain subjects of further study. actions of magainins 1 and 2 with acidic lipid bilayers.
Biochim Biophys Acta 1063:162–170
Matsuzaki K, Fukui M, Fujii N, Miyajima K (1991b)
Interactions of an antimicrobial peptide, tachyple-
References sin I, with lipid membranes. Biochim Biophys Acta
1070:259–264
Andersson E, Rydengard V, Sonesson A, Morgelin M, Matsuzaki K, Sugishita K, Fujii N, Miyajima K (1995a)
Bjorck L, Schmidtchen A (2004) Antimicrobial Molecular basis for membrane selectivity of an antimi-
activities of heparin-binding peptides. Eur J Biochem crobial peptide, magainin 2. Biochemistry 34:3423–3429
271:1219–1226 Matsuzaki K, Murase O, Miyajima K (1995b) Kinetics of
Bechinger B, Lohner K (2006) Detergent-like actions of pore formation induced by an antimicrobial peptide,
linear amphipathic cationic antimicrobial peptides. magainin 2. Biochemistry 34:12553–12559
Biochim Biophys Acta 1758:1529–1539 Matsuzaki K, Murase O, Fujii N, Miyajima K (1996a) An
Bessalle R, Kapitkovsky A, Gorea A, Shalit I, Fridkin M antimicrobial peptide, magainin 2, induced rapid flip-­
(1990) All-D-magainin: chirality, antimicrobial activ- flop of phospholipids coupled with pore formation and
ity and proteolytic resistance. FEBS Lett 274:151–155 peptide translocation. Biochemistry 35:11361–11368
Furse S, Scott DJ (2016) Three-dimensional distribu- Matsuzaki K, Yoneyama S, Murase O, Miyajima K
tion of phospholipids in gram negative bacteria. (1996b) Transbilayer transport of ions and lipids cou-
Biochemistry 55:4742–4747 pled with mastoparan X translocation. Biochemistry
35:8450–8456
16 K. Matsuzaki

Matsuzaki K, Sugishita K, Harada M, Fujii N, Miyajima Shai Y (1995) Molecular recognition between membrane-­
K (1997a) Interactions of an antimicrobial pep- spanning polypeptides. Trends Biol Sci 20:460–465
tide, magainin 2, with outer and inner membranes Sochacki KA, Barns KJ, Bucki R, Weisshaar JC (2011)
of gram-negative bacteria. Biochim Biophys Acta Real-time attack on single Escherichia coli cells by
1327:119–130 the human antimicrobial peptide LL-37. Proc Natl
Matsuzaki K, Nakamura A, Murase O, Sugishita K, Fujii Acad Sci U S A 108:E77–E81
N, Miyajima K (1997b) Modulation of magainin Takeshima K, Chikushi A, Lee K-K, Yonehara S,
2–lipid bilayer interactions by peptide charge. Matsuzaki K (2003) Translocation of analogues of the
Biochemistry 36:2104–2111 antimicrobial peptides magainin and buforin across
Matsuzaki K, Mitani Y, Akada K, Murase O, Yoneyama human cell membranes. J Biol Chem 278:1310–1315
S, Zasloff M et al (1998a) Mechanism of synergism Tomasinsig L, Skerlavaj B, Papo N, Giabbai B, Shai
between antimicrobial peptides magainin 2 and PGLa. Y, Zanetti M (2006) Mechanistic and functional
Biochemistry 37:15144–15153 studies of the interaction of a proline-rich antimi-
Matsuzaki K, Sugishita K, Ishibe N, Ueha M, Nakata S, crobial peptide with mammalian cells. J Biol Chem
Miyajima K et al (1998b) Relationship of membrane 281:383–391
curvature to the formation of pores by magainin. Utsugi T, Schroit AJ, Connor J, Bucana CD, Fidler IJ
Biochemistry 37:11856–11863 (1991) Elevated expression of phosphatidylserine
Matsuzaki K, Sugishita K, Miyajima K (1999) in the outer membrane leaflet of human tumor cells
Interactions of an antimicrobial peptide, magainin 2 and recognition by activated human blood monocytes.
with lipopolysaccharide-­ containing liposomes as a Cancer Res 51:3062–3066
model for outer membranes of gram-negative bacteria. Wade D, Boman A, Wåhlin B, Drain CM, Andreu D,
FEBS Lett 449:221–224 Boman HG et al (1990) All-D amino acid-containing
Miyazaki Y, Aoki M, Yano Y, Matsuzaki K (2012) channel forming antibiotic peptides. Proc Natl Acad
Interaction of antimicrobial peptide magainin 2 with Sci U S A 87:4761–4765
gangliosides as a target for human cell binding. Wenk MR, Seelig J (1998) Magainin 2 amide interac-
Biochemistry 51:10229–10235 tion with lipid membranes: calorimetric detection of
Oren Z, Shai Y (1998) Mode of action of linear amphipa- peptide binding and pore formation. Biochemistry
thic alpha-helical antimicrobial peptides. Biopolymers 37:3909–3916
47:451–463 Wieprecht T, Beyermann M, Seelig J (1999) Binding
Park CB, Kim HS, Kim SC (1998) Mechanism of action of antibacterial magainin peptides to electrically
of the antimicrobial peptide buforin II: buforin II kills neutral membranes: thermodynamics and structure.
microorganisms by penetrating the cell membrane and Biochemistry 38:10377–10387
inhibiting cellular functions. Biochem Biophys Res Wimley WC (2010) Describing the mechanism of anti-
Commun 244:253–257 microbial peptide action with the interfacial activity
Patrzykat A, Friedrich CL, Zhang L, Mendoza V, Hancock model. ACS Chem Biol 5:905–917
RE (2002) Sublethal concentrations of pleurocidin-­ Yang L, Harroun TA, Weiss TM, Ding L, Huang HW
derived antimicrobial peptides inhibit macromolecu- (2001) Barrel-stave model or toroidal model? A case
lar synthesis in Escherichia coli. Antimicrob Agents study on melittin pores. Biophys J 81:1475–1485
Chemother 46:605–614 Yoneyama F, Imura Y, Ohno K, Zendo T, Nakayama J,
Roversi D, Luca V, Aureli S, Park Y, Mangoni ML, Stella Matsuzaki K et al (2009) Peptide-lipid huge toroidal
L (2014) How many antimicrobial peptide molecules pore, a new antimicrobial mechanism mediated by a
kill a bacterium? The case of PMAP-23. ACS Chem lactococcal bacteriocin, lacticin Q. Antimicrob Agents
Biol 9:2003–2007 Chemother 53:3211–3217
Sansom MSP (1991) The biophysics of peptide models Zasloff M (1987) Magainins, a class of antimicrobial pep-
of ion channels. Prog Biophys Mol Biol 55:139–235 tides from Xenopus skin: isolation, characterization
Schwarz G, Arbuzova A (1995) Pore kinetics reflected in of two active forms, and partial cDNA sequence of a
the dequenching of a lipid vesicle entrapped fluores- precursor. Proc Natl Acad Sci U S A 84:5449–5453
cent dye. Biochim Biophys Acta 1239:51–57 Zhang L, Benz R, Hancock REW (1999) Influence of
Schwarz G, Robert CH (1992) Kinetics of pore-mediated proline residues on the antibacterial and synergis-
release of marker molecules from liposomes or cells. tic activities of α-helical peptides. Biochemistry
Biophys Chem 42:291–296 38:8102–8111
 Elementary Processes
and Mechanisms of Interactions 3
of Antimicrobial Peptides
with Membranes—Single Giant
Unilamellar Vesicle Studies—

Moynul Hasan and Masahito Yamazaki

Abstract rate constants of AMP-induced pore forma-

To elucidate the mechanisms of action of anti- tion or local rupture and membrane perme-
microbial peptides (AMPs) and to develop de ation of internal contents through the pore or
novo designed peptides with activities similar the local rupture, the transbilayer movement
to those of AMPs, it is essential to elucidate of lipids, and the relationship between the
the detailed processes of AMP interactions location of AMPs and pore formation. Effects
with plasma membranes of bacterial and fun- of membrane tension and of asymmetric lipid
gal cells and model membranes (lipid bilay- packing in the bilayer on AMP-induced pore
ers). In this mini-review, we summarize the formation also are described. On the basis of
present state of knowledge of the interactions these data, we discuss the present state of
of AMPs with lipid vesicles obtained using the understanding of the interaction of AMPs with
single giant unilamellar vesicle (GUV) lipid bilayers and future prospects for AMP
method. Currently, three modes of action of studies.
AMPs on GUVs have been defined. The ele-
mentary processes of interactions of AMPs Keywords
with lipid vesicles revealed by the single GUV Antimicrobial peptides · Giant unilamellar
method, and the advantages of this technique, vesicle · Pore formation · Local rupture ·
are described and discussed. For example, the Translocation across membranes · Lipid
single GUV method can be used to determine bilayers · Elementary process · Rate constant
· Membrane tension

M. Hasan
Integrated Bioscience Section, Graduate School of
Science and Technology, Shizuoka University,
Shizuoka, Japan 3.1 Introduction
M. Yamazaki (*)
Integrated Bioscience Section, Graduate School of Antimicrobial peptides (AMPs) and antimicro-
Science and Technology, Shizuoka University, bial proteins with bactericidal and fungicidal
Shizuoka, Japan activities are produced by various organisms
Nanomaterials Research Division, Research Institute (e.g., animals including humans, plants, and
of Electronics, Shizuoka University, Shizuoka, Japan insects) to defend themselves against microbes.
Department of Physics, Graduate School of Science, Most AMPs target bacterial and fungal plasma
Shizuoka University, Shizuoka, Japan membranes and induce damage in the m
­ embranes,
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 17

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_3
18 M. Hasan and M. Yamazaki

including pore formation that increases mem- We recently developed the single GUV
brane permeability (Zasloff 2002; Melo et al. method to investigate interactions of compounds
2009; Hwang and Vogel 1998; Propheter et al. such as peptides/proteins with lipid bilayers
2017). Some AMP-induced damage to the plasma (Tamba and Yamazaki 2005, 2009; Yamazaki
membrane is considered to reflect the interaction 2008; Tamba et al. 2010; Islam et al. 2014a, b,
of these AMPs with the lipid membrane regions 2018). In this method, performed under various
of bacterial and fungal plasma membranes, as optical microscopes, a compound solution is
indicated by the observation that synthetic AMP delivered continuously to the vicinity of a single
analogues composed of all-D amino acids have GUV through a micropipette to induce the inter-
the same antibacterial activity as wild-type AMPs action of the compound with the single
composed of all-L amino acids (Wade et al. 1990; GUV. During the interaction, changes in the
Wakabayashi et al. 1999). structure and physical properties of the single
It has been established that the physicochemi- GUV are measured as a function of time and
cal properties of lipid bilayer vesicles are similar spatial coordinates using fluorescence micros-
to those of native plasma membranes, and lipid copy coupled with an electron-multiplying
bilayers are therefore excellent models of plasma charge-­coupled device (EM-CCD) camera and
membranes (Lipowsky and Sackmann 1995; confocal laser scanning microscopy (CLSM).
Baumgart et al. 2003; McLaughlin and Murray We perform the same experiment using many
2005; Bigay and Antonny 2012). To investigate “single GUVs” under the same conditions and
the interaction of AMPs with the lipid membrane then analyze the results statistically (Yamazaki
regions of microbial plasma membranes, various 2008; Islam et al. 2018). The single GUV
kinds of vesicles composed of lipid bilayers have method enables us to distinguish the elementary
been extensively used. To date, most research on processes of individual events occurring in a
the interactions of AMPs with lipid bilayers has single GUV during the interaction of an AMP
been performed using suspensions of large unila- with the GUV, and statistical analysis permits us
mellar vesicles (LUVs) with a diameter of to derive the rate constants of each elementary
50 nm–1 μm (100–300 nm for most uses) (i.e., the process.
LUV suspension method). Given their small size, In this mini-review, we focus on research on
we cannot observe the structure and the physical the interactions of AMPs with lipid vesicles using
quantities of each LUV in buffer; hence, using the the single GUV method. We review what kinds of
LUV suspension method, we can obtain only new information on the interaction of AMPs with
ensemble averages of physical quantities of many lipid bilayers have been obtained using the single
LUVs in a suspension, as measured by various GUV method, compared with the information
biophysical and physicochemical techniques. obtained using the LUV suspension method.
Therefore, the LUV suspension technique does Here, as the examples of AMPs, we use magainin
not permit observation of each elementary pro- 2, lactoferricin B (LfcinB), LfcinB (4–9) (which
cess in the interaction of peptides such as AMPs has the amino acid sequence RRWQWR, making
with lipid bilayers (Yamazaki 2008; Islam et al. this AMP one of the shortest versions of LfcinB),
2018). In contrast, the shape and physical proper- and transportan 10 (TP10). Among these exam-
ties of each giant unilamellar vesicle (GUV) with ples, TP10 is not a native AMP because this mol-
a diameter larger than 1 μm (10–40 μm for most ecule is a synthetic, cell-penetrating peptide
uses) in buffer can be observed in real time using (CPP). However, TP10 has been shown to exhibit
various optical microscopy techniques. Using this strong antimicrobial activity while being able to
advantage, GUVs have been employed to investi- enter eukaryotic cells without affecting viability
gate shape changes, the elastic modulus of lipid (Nekhotiaeva et al. 2004). Hence, we can reason-
bilayers, phase separation, and tension-induced ably infer that TP10 has an activity similar to
pore formation (Sandre et al. 1999; Rawictz et al. AMPs, given that TP10 induces preferential dam-
2000; Baumgart et al. 2003; Tanaka et al. 2004). age in microbes.
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 19

3.2 Mode of Action of AMPs magainin 2 formed pores in the membrane

through which the membrane permeation of cal-
To date, three distinct modes of action (MoAs) of cein occurred. Hence, the MoA of magainin 2 is
AMPs on lipid bilayers have been observed using type A. Similarly, lysenin, a pore-forming toxin
the single GUV method: (A) small pore forma- (PFT) protein, induced gradual membrane per-
tion, (B) local rupture and complete rupture (or meation of calcein from GUVs of lipid bilayers
burst), and (C) no damage to the bilayers. In the composed of sphingomyelin (SM), PC, and cho-
type A MoA, AMPs induce small pores in lipid lesterol (chol) (Alam et al. 2012); the MoA of
bilayers; these pores cannot be observed using lysenin also is type A.
optical microscopy but can be detected by the In the type B MoA, AMPs induce local rup-
membrane permeation (i.e., leakage) of water-­ ture or complete rupture of single GUVs, through
soluble fluorescent probes from the lumen of which fluorescent probes rapidly leak away.
single GUVs to the outside. After the leakage, the After the leakage, the size of GUVs decreases a
same spherical GUVs without any detectable little, and sometimes several partially thickened
breaks remain, and their sizes do not change sig- membranes appear at the rim of the GUVs (cor-
nificantly. First, we consider the results for responding to the GUV membranes). First, we
magainin 2. The magainin 2-induced leakage of consider the results for LfcinB. Figure 3.2a
internal contents has been extensively investi- shows the result of LfcinB-induced leakage of
gated using the LUV suspension method calcein from single PG/PC-GUVs
(Matsuzaki et al. 1995, 1998; Gregory et al. (Moniruzzaman et al. 2015). The calcein began
2009). Those studies indicated that magainin 2 to leak from the lumen suddenly after an
induces leakage of water-soluble fluorescent extended time of the interaction, and the leakage
probes from the lumen of LUVs, suggesting that then was completed within several seconds
magainin 2 induces pores in lipid membranes. (Fig. 3.2a, c), after which the spherical GUV
However, there are several factors (such as pore structure remained, albeit with a decreased
formation, membrane fusion, large change diameter. These observations indicate the occur-
in local curvature, rupture, and fragmentation of rence of a local rupture or a large pore in the
the liposomes) that induce leakage of the internal GUV membrane, and hence, LfcinB belongs to
contents from vesicle lumens, and it is difficult to type B. In the case of the interaction of TP10
clearly identify the cause of leakage using the with single GUVs, a transient local rupture was
LUV suspension method (Yamazaki 2008). sometimes observed, although the rate of leak-
Figure 3.1a shows a result for the magainin age was slower than that observed with LfcinB
2-induced leakage of a water-soluble fluorescent (Islam et al. 2014a). On the other hand, epigal-
probe, calcein, from single GUVs composed of locatechin gallate (EGCg), a tea catechin (anti-
dioleoylphosphatidylglycerol (DOPG; hereafter microbial substance), induced rapid leakage of
PG) and dioleoylphosphatidylcholine (DOPC; calcein from single PC-GUVs within a few sec-
hereafter PC) (Tamba and Yamazaki 2005, 2009). onds, and after this leakage, the spherical GUVs
The fluorescence intensity (FI) of the GUV were converted to single smaller lumps com-
lumen in the fluorescence microscopic images posed of lipid and EGCg (Tamba et al. 2007).
(Fig. 3.1a (II)) was proportional to calcein con- During the conversion, we observed a large pore
centration in the lumen. The time course of the in a GUV membrane using optical microscopy.
change in FI indicated that the calcein began to In this case, no spherical GUVs remained after
leak from the lumen suddenly after an extended the leakage, and therefore, we can infer that
time of interaction, and this leakage continued EGCg induces rupture of GUVs.
gradually (Fig. 3.1b). During this leakage, local In the type C MoA, AMPs do not induce any
disruption, rupture of the GUVs, changes in local significant damage in lipid bilayers, and hence
curvature of the membrane, membrane fusion, no leakage of fluorescent probes occurs; how-
and vesicle fission did not occur, indicating that ever, AMPs translocate across lipid bilayers to
20 M. Hasan and M. Yamazaki

Fig. 3.1 Magainin 2-induced membrane permeation of tions as in (a). (d) Time course of Pintact of PG/PC (6/4)-
calcein from single GUVs. Buffer A (10 mM PIPES, pH GUV in the presence of various concentrations of
7.0, 150 mM NaCl, 1 mM EGTA) was used at 25 °C. (a) magainin 2: (○) 5, (■) 3, (∆) 2.5, and (●) 1 μM. The
Fluorescence images (II) show the time course of the FI solid lines represent the best fit curves of Eq. 3.1. (e)
due to calcein in a PG/PC (6/4)-GUV lumen during the Dependence of the rate constant of pore formation, kP, on
interaction with 3 μM magainin 2. The numbers above magainin 2 concentration in buffer, C. (■), PG/PC (6/4)-;
each image show the time in seconds after the magainin 2 (green ●), PG/PC (5/5)-; (blue ●), PG/PC (4/6)-; and (red
addition was started. Also shown are phase-contrast ▲), PG/PC (3/7)-GUVs. (f) Dependence of kP on X (a–e
images of the GUV (I, III). The bar corresponds to 10 μm. and f are reprinted from Tamba and Yamazaki 2009; Karal
(b) Time course of the change in the normalized FI of the et al. 2015, respectively, with permission from the
GUV shown in (a). (c) Other examples of the time course American Chemical Society)
of the change in FI of single GUVs under the same condi-
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 21

Fig. 3.2 LfcinB-induced membrane permeation of cal- change in the FI of single GUVs. (d) The LfcinB concen-
cein from single GUVs. Buffer A was used at 25 °C. (a) tration dependence of the rate constant of pore formation,
Fluorescence images (II) show the time course of the FI kP, in PG/PC (1/4)-GUV (□) and PG/PC (1/1)-GUV (○).
due to calcein in a PG/PC (1/1)-GUV lumen during the (E) The LfcinB concentration dependence of the kP in PG/
interaction with 2.0 μM LfcinB. The numbers above each PC (1/1)-GUV in buffer A without NaCl (10 mM PIPES,
image show the time in seconds after the LfcinB addition pH 7.0, 1.0 mM EGTA) (●) and in buffer A containing
was started. Also shown are phase-contrast images of the 150 mM NaCl (○) (a−e are reprinted from Moniruzzaman
GUV (I, III). The bar corresponds to 20 μm. (b) Time et al. 2015 with permission from the American Chemical
course of the change in the normalized FI of the GUV Society)
shown in (a). (c) Other examples of the time course of the

enter the vesicle lumen. Here, we consider the age of AF647 (Fig. 3.3a) (Moniruzzaman et al.
results for LfcinB (4-9). The interaction of 2017). Moreover, Rh-LfcinB (4-9) entered the
LfcinB (4-9) with single PG/PC-GUVs did not cytoplasm of E. coli cells without leakage of
induce any leakage of calcein or AF647 (another calcein. These data indicated that the antimicro-
water-­soluble fluorescent probe). Nonetheless, bial activity of LfcinB (4-9) is not the result of
Rh-LfcinB (4-9), which is LfcinB (4-9) labeled damage to plasma membranes. Since AMPs
with a fluorescent probe (lissamine rhodamine with a type C MoA exhibit cell penetration
B red), was translocated across a GUV mem- activity, they are similar to cell-penetrating pep-
brane and entered the GUV lumen without leak- tides (CPPs).
22 M. Hasan and M. Yamazaki

Fig. 3.3 Entry of Rh-LfcinB (4-9) into single PG/PC Rh-LfcinB (4-9) shown in (a). Red and green points cor-
(1/1)-GUVs containing small GUVs. Buffer A was used at respond to the FI due to AF647 in the GUV lumen and due
25 °C. (a) CLSM images of (I) AF647 and (II) Rh-LfcinB to Rh-LfcinB (4-9) in the rim of the GUV, respectively. (c)
(4-9). The numbers above each image show the time in Dependence of Pentry (10 min) on the Rh-LfcinB (4-9) con-
seconds after the addition of 5.0 μM Rh-LfcinB (4-9) was centration (a–c are reprinted from Moniruzzaman et al.
started. The bar is 20 μm. (b) Time course of the change in 2017 with permission from the American Chemical
the normalized FI of the GUV during the interaction of Society)

3.3 Advantages of the Single images of a PG/PC (6/4)-GUV containing cal-

GUV Method cein in its lumen during the interaction with
magainin 2. The time course of the FI of the GUV
3.3.1 Separation of Elementary lumen due to calcein (Fig. 3.1b) provides two
Processes and Determination pieces of information: one is the time when pore
of the Rate Constant of Each formation starts in the membrane (based on the
Elementary Process time at which the FI starts to decrease), and the
other is the characteristics of membrane perme-
AMPs of types A and B induce membrane per- ation of the fluorescent probe such as its rate con-
meation of the internal contents via a mechanism stant (based on the time course of the decrease in
that consists of at least two elementary processes. FI after pore formation). If we perform the same
The first process is the AMP-induced damage in experiments using multiple individual “single
the membrane such as pore formation, local rup- GUVs,” we can obtain more information. In the
ture, and burst; the second process is membrane case of magainin 2, the membrane permeation of
permeation through the damage in the membrane. calcein from the single GUVs began stochasti-
One of the largest disadvantages of the LUV sus- cally (Fig. 3.1c), indicating that pore formation
pension method is the inability to distinguish started stochastically in the lipid bilayers.
these two steps (Yamazaki 2008). On the other Membrane permeation in each GUV was com-
hand, the single GUV method enables us to sepa- plete within a short time (~30 s under these con-
rate these elementary processes of AMP-induced ditions) after the start of membrane permeation
pore formation in the lipid bilayers and (local) (Fig. 3.1c). These results indicated that pore for-
rupture of GUVs. mation in the membrane is the rate-determining
First, we show the results for magainin 2. step in the total reaction of the magainin
Figure 3.1a (II) shows fluorescence microscopic 2-induced leakage, because the other elementary
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 23

process (i.e., membrane permeation through the where IN (t) is the normalized FI of the GUV
pore) is completed rapidly. This may correspond lumen, I(t) is the FI of the GUV lumen at time t
to the concept of the “all-or-none” mode of leak- after initiation of membrane permeation, and I(0)
age in the LUV suspension method (Gregory is I(t) at t = 0. If we plot the log of IN (t) as a func-
et al. 2009). To analyze this type of stochastic tion of time, the kmp (t) can be obtained quantita-
phenomenon, the concept of a probability can be tively from the slope of the curve. The value of
used; in this case, the probability is assessed as kmp depends on the radius of the GUVs (r), but the
Pintact (t), the fraction (among all examined GUVs) membrane permeability coefficient, P (t), which
of intact GUVs where no membrane permeation is derived from kmp (t) (i.e., P (t) = r kmp (t)/3),
of fluorescent probes occurs (Tamba and does not depend on the value of r (Alam et al.
Yamazaki 2005, 2009). Pintact (t) decreased mono- 2012).
tonically over time. If we apply an irreversible
two-state transition model from the intact state to
the pore state (i.e., the state of the GUV with 3.3.2  ate Constant of Pore
R
pores), we can obtain the theoretical equation of Formation and Local Rupture
the time course of Pintact (t) as follows:
The rate constants of pore formation or local rup-
{
Pintact ( t ) = exp −kP ( t − teq ) } (3.1) ture greatly depend not only on AMP concentra-
tion in buffer but also on lipid composition and
where kP is the rate constant of the two-state tran- buffer conditions. As the concentration of the
sition, which can be considered as the rate con- negatively charged lipid (such as PG) in the
stant of magainin 2-induced pore formation at the bilayers (i.e., the surface charge density of the
initial state, and teq is a fitting parameter and membrane) increases, the kP greatly increases
implies the time required to achieve the binding (Figs. 3.1e and 3.2d). Moreover, as the concen-
equilibrium of magainin 2 from aqueous solution tration of the salt (e.g., NaCl) decreases, the kP
to the GUV membrane. All the curves of the time also greatly increases (Fig. 3.2e) (Tamba and
courses of Pintact (t) were well fit to Eq. 3.1, pro- Yamazaki 2009; Moniruzzaman et al. 2015). It is
viding the values of kP (Fig. 3.1d). The rate con- well recognized that electrostatic interactions due
stant kP increased with an increase in magainin 2 to the surface charges of lipid membranes in buf-
concentration (Fig. 3.1e). Similarly, for AMPs fer increase with an increase in surface charge
with a type B MoA, we can obtain the rate con- density or with a decrease in salt concentration
stant of the two-state transition from the intact (Israelachvili 1992). Therefore, these results
state to the rupture state (i.e., the rate constant of clearly indicate that for magainin 2 and LfcinB,
local rupture or complete rupture) from analysis the values of kP increase with an increase in elec-
of the time course of Pintact (t) using Eq. 3.1 trostatic interactions.
(Moniruzzaman et al. 2015; Tamba et al. 2007). Using the binding constant of magainin 2 to
These rate constants also increased with increas- lipid bilayers, we can convert the magainin 2 con-
ing LfcinB and EGCg concentrations. centration in the buffer, C, to the magainin 2 sur-
On the other hand, if we analyze the time face concentration in the membrane, X (mol/mol;
course of the decrease in FI of the GUV lumen the molar ratio of magainin 2 bound to the mem-
due to fluorescent probe after pore formation, we brane interface to lipid in the outer monolayer of
can obtain information on the AMP-induced a GUV). Figure 3.1f shows the relationship
membrane permeation such as the rate constant between kP and X. The dependences of kP on X for
kmp (Tamba and Yamazaki 2005; Tamba et al. GUVs with two different lipid compositions are
2010). The value of kmp is determined as follows: largely superimposable, indicating that X deter-
mines the rate constant of magainin 2-induced
I N ( t ) = I ( t ) / I ( 0 ) = exp ( −kmp ( t ) t ) (3.2) pore formation (Tamba and Yamazaki 2009).
24 M. Hasan and M. Yamazaki

3.3.3  ate Constant of Membrane

R
Permeation

Generally, two factors (the size and the number

of pores) determine the rate constant of mem-
brane permeation of fluorescent probes through
pores, kmp. If we use various fluorescent probes of
different sizes (Stokes-Einstein radius, RSE) for
the AMP-induced membrane permeation experi-
ment, we can determine the size of the AMP-­
induced pore and obtain information on the
temporal change in pore size and the number of
pores by analyzing the time course of kmp (Tamba
et al. 2010). In the case of magainin 2-induced
pore formation in PG/PC (4/6)-GUVs, rapid
membrane permeation of larger fluorescent
probes [such as Texas-Red dextran 3000
(TRD-3k) (RSE = 1.4 nm) and Texas-Red dextran
10,000 (TRD-10k) (RSE = 2.7 nm)] occurred at
the initial time, but the rate of membrane perme-
ation subsequently decreased. Figure 3.4a shows
more quantitative data for TRD-3k and TRD-­
10k. Using the two linear regions of these curves,
we can obtain the values of kmp at the initial stage
( kmp )
initial

initial
( )
steady
and the final steady stage kmp . The Fig. 3.4 Time course of membrane permeation. (a)
value of kmp is 20–40 times greater than that of Magainin 2-induced membrane permeation of fluorescent
steady probes from a PG/PC (1/1)-GUV. Time course of the loga-
kmp under these conditions. In contrast, larger
rithm of the normalized FI of the GUV containing
fluorescent probes [such as TRD-40k
TRD-3 k or TRD-10 k during the interaction with 4 μM
(RSE = 5.0 nm) and FITC-BSA (RSE = 3.6 nm)] magainin 2 in buffer A at 25 °C. (b) Lysenin-induced
leaked only at the initial stage of magainin membrane permeation of calcein from a SM/PC/chol
2-induced pore formation. Based on these results, (42/30/28)-GUV. Time course of the normalized FI of the
GUV (log-10 scale) during the interaction with 40 ng/mL
we can reasonably conclude that magainin 2 ini-
lysenin in PBS buffer at 37 °C (a, b are reprinted from
tially induces a large, transient pore in lipid bilay- Tamba et al. 2010; Alam et al. 2012, respectively, with
ers; within several minutes, the radius of the pore permission from the American Chemical Society)
then decreases to a stable smaller size at the final
steady stage. The result of no membrane perme-
ation of Alexa-Fluor trypsin inhibitor (AF-SBTI) tide-/protein-induced pores have been found
at the final stage with 7 μM magainin 2 indicates using the GUV suspension method (Fuertes et al.
that the radius of the pore at the final stage is 2010; Bleicken et al. 2013).
smaller than 2.8 nm (RSE of AF-SBTI) but is Contrast these results with those obtained
larger than 1.4 nm (RSE of TRD-3k). This value with lysenin-induced membrane permeation of a
agrees with the 1.9 nm radius of magainin small fluorescent probe, calcein (RSE = 0.7 nm),
2-induced pores in multilayer membranes at in single SM/PC/chol (42/30/28)-GUVs. After
equilibrium, as determined by neutron in-plane membrane permeation was initiated, the kmp grad-
scattering (Ludtke et al. 1996). These data pro- ually increased with time before reaching a
vide the first information concerning the kinetic steady, maximum value, which persisted for an
pathway of AMP-induced pore formation in lipid extended time (e.g., 280 s) (Fig. 3.4b) (Alam
bilayers (i.e., evolution of pores). Subsequently, et al. 2012). This result indicated that after
other examples of the change in the size of pep- lysenin-­induced pore formation began, the value
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 25

of P (t) increased with time before reaching a of GUVs into which AMPs enter before a specific
steady, maximum value of Ps, which was main- time t. Figure 3.3c indicates that Pentry (10 min)
tained for an extended time. Based on the experi- increased with an increase in Rh-LfcinB (4-9)
mental results that lysenin molecules form a concentration. It is also possible to estimate the
specific oligomer in SM/PC/chol bilayers, we rate constant of the binding of AMPs to the mono-
infer that these oligomers form pores with the layer and the rate constant of unbinding by ana-
same diameter. Therefore, the result of Fig. 3.4b lyzing the rim intensity (Islam et al. 2014a, 2018).
indicated that the number of pores increased with
time to reach the steady, maximum number.
3.3.5 AMP-Induced Transbilayer
Movement of Lipids
3.3.4 Entry of AMPs Without Pore
Formation The studies using the LUV suspension method
suggested that the interactions of AMPs and
To detect the entry of AMPs into a GUV lumen venom peptides with lipid vesicles yielded an
with high sensitivity, we can use the single GUV increase in kFF, the rate constant of the transbi-
method for CPPs. In this instance, we used CLSM layer movement (i.e., flip-flop) of lipids
to investigate the interaction of fluorescent probe- (Matsuzaki et al. 1996; Müller et al. 2000).
labeled AMPs with single GUVs containing Matsuzaki et al. (1996) indicated a strong corre-
smaller GUVs or LUVs and water-soluble fluo- lation between the magainin 2-induced leakage
rescent probes such as AF647 in their lumen of internal contents and the flip-flop of lipids,
(Islam et al. 2014a, 2018; Moghal et al. 2018). suggesting the toroidal model of magainin
The binding of the fluorescent probe-labeled 2-induced pore formation. Since the peptide-­
AMPs with bilayers of the small GUVs and LUVs induced transbilayer movement of lipids is com-
increases their FI, and hence, we can detect the posed of several elementary steps, it was difficult
entry of the AMPs into the GUV lumen with to determine directly the main process for growth
higher sensitivity. The comparison between the FI of kFF in the framework of the LUV suspension
of the GUV lumen due to water-soluble fluores- method. On the other hand, the use of the single
cent probe and that due to fluorescent probe- GUV method to measure the kFF has an advantage
labeled AMPs provides information on the (Hasan et al. 2018a, b). In this method, we use
relationship between the entry of AMPs and pore single GUVs with asymmetric lipid compositions
formation. Figure 3.3 shows a result for the inter- in two monolayers, where a low concentration of
action of Rh-LfcinB (4-9) with single PG/PC fluorescent probe-labeled lipid [such as
(1/1)-GUVs (Moniruzzaman et al. 2017). The FI 18:1-NBD-lyso-PE (hereafter NBD-LPE)] exists
of the GUV lumen due to AF647 remained con- only in their inner leaflet (e.g., PG/PC/NBD-LPE
stant (Fig. 3.3a (I) and red line in Fig. 3.3b) for (40/59/1; inner leaflet)-PG/PC (40/59; outer
10 min, indicating no pore formation. On the leaflet)-GUVs). The analysis of the time course
other hand, the FI of small vesicles in the GUV of the decrease in rim intensity due to NBD-LPE
lumen was observed after 143 s (Fig. 3.3a (II)), (measured using CLSM) provides the value of
and the FI of the GUV membrane (i.e., the rim kFF, because the rate of transfer of NBD-lipids
intensity) due to Rh-LfcinB (4-9) increased rap- from the outer leaflet to aqueous solution is large,
idly with time to reach a steady value at 100 s and hence the rate-determining step of the
(Fig. 3.3b). These results indicated that Rh-LfcinB decrease in rim intensity is the transbilayer
(4-9) first bound to the GUV membrane, then ­movement of lipids from the inner leaflet to the
translocated across the bilayer, and finally entered outer one (Hasan et al. 2018b). In the interaction
into the GUV lumen without pore formation. of magainin 2 with single GUVs with this asym-
At present, the rate of entry of AMPs into the metric composition (Fig. 3.5), rim intensity
vesicle lumen is estimated as Pentry (t), the fraction decreased rapidly after the start of magainin
26 M. Hasan and M. Yamazaki

Fig. 3.5 Relationship between the transbilayer move- bers above each image show the time in seconds after
ment of lipids and magainin 2-induced pore formation. (a) magainin 2 addition was started. The bar corresponds to
Relationship between the rim intensity of the GUVs due 20 μm. (b) Time course of the change in FI of the GUV
to NBD-LPE and the FI of the GUV lumen due to AF647 shown in (a). Red and green points correspond to the nor-
during the interaction of 31 μM magainin 2 with PG/PC/ malized FI of AF647 in the GUV lumen and the FI of
NBD-LPE (40/59/1; inner)-PG/PC (40/59; outer)-GUVs. NBD-LPE in the rim of the GUV, respectively
CLSM images of (I) NBD-LPE and (II) AF647. The num-

2-induced membrane permeation of AF647, indi- other in a toroidal fashion to form a pore in which
cating that the transbilayer movement of NBD- the inner wall is composed of peptides and lipid
LPE occurred rapidly after pore formation, head groups (Matsuzaki et al. 1996; Ludtke et al.
whereas before pore formation, the kFF value was 1996; Yang et al. 2000; Qian et al. 2008).
similar to that in the absence of magainin 2. This Therefore, lipid molecules can rapidly diffuse
result clearly indicated that the occurrence of along this integrated lipid monolayer from the
pores induced by magainin 2 increased the kFF of inner leaflet to the outer one through the wall of
NBD-LPE, whereas the binding of magainin 2 the pore, thereby inducing rapid transbilayer
only to the membrane did not increase the kFF movement.
value. The kFF after pore formation was much
larger than that in the normal lipid bilayer, indi-
cating that the mechanism of the transbilayer 3.4 Relationship Between the
movement of lipids after pore formation is differ- Location of AMPs and Pore
ent from that of normal transbilayer movement. Formation
If we assume that the structure of the magainin
2-induced pore at the initial stage is a toroidal To elucidate the mechanism of AMP-induced
pore, we can reasonably explain this great pore formation, it is important to know the loca-
increase in kFF. In a toroidal pore, the outer and tion of peptides in a lipid bilayer during pore for-
inner monolayers bend and merge with each mation. To elucidate the relationship between the
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 27

AMP-induced pore formation and the location of membrane permeation of AF647 started through
AMP in a GUV, interactions of fluorescent probe-­ CF-TP10-induced pores. After pore formation,
labeled AMPs (e.g., carboxyfluorescein (CF)- the rim intensity did not change. This result indi-
labeled AMP) with single GUVs containing cated that CF-TP10 translocates from the outer
AF647 in their lumen have been investigated leaflet to the inner one from the beginning of the
using CLSM. At present, we have defined two interaction, without pore formation; the steady
kinds of relationships: (A) asymmetric trans- binding of TP10 in both leaflets is achieved rap-
membrane distribution of AMPs before pore for- idly, and then after an extended time, pore forma-
mation (e.g., as with magainin 2) and (B) tion occurs. Therefore, the symmetric
symmetric transmembrane distribution of AMPs transmembrane distribution of CF-TP10 (i.e.,
before pore formation (e.g., as with TP10). localization of CF-TP10 in both monolayers)
First, we consider the case of magainin 2. induces pore formation. The AMPs with a type C
Figure 3.6a shows the time course of FI in the MoA (e.g., LfcinB (4-9)) also can translocate
interaction of CF-magainin 2/magainin 2 with a across lipid bilayers without pore formation, but
single PG/PC (4/6)-GUV containing AF647 in the symmetric distribution of these AMPs does
its lumen (Karal et al. 2015). The FI of the GUV not induce pore formation (Moniruzzaman et al.
lumen due to AF647 started to decrease at 149 s, 2017).
indicating that the membrane permeation of The different relationship between the AMP-­
AF647 from the GUV started through a induced pore formation and the location of AMP
magainin 2-induced pore. On the other hand, the in the membrane is due to the activity of translo-
rim intensity due to CF-magainin 2 rapidly cation of AMPs across the lipid bilayer. This
increased to a steady value, I1, at t = 50 s, activity is more important for the CPPs, and vari-
remained constant for a long time (~90 s) until ous research have been performed on this topic
the intensity started to increase at a time similar (Madani et al. 2011; Islam et al. 2018). However,
to that of the start of membrane permeation of at present, the molecular mechanism of the trans-
AF647, and then rapidly reached a final steady location of peptides across the lipid bilayer is not
value, I2. The average value of I2/I1 is approxi- clear, although several mechanisms have been
mately 2.0. This result indicates that the steady proposed (Islam et al. 2018). The characteristics
binding of magainin 2 to the outer monolayer of of peptides (e.g., hydrophobicity and amino acid
the GUV membrane is attained at ~50 s and the sequence) required for translocation activity
surface concentration of magainin 2 in the outer without pore formation also are not clear.
monolayer, X, remains constant until just before Notably, the translocation activity depends not
pore formation. Therefore, we can reasonably only on peptide properties but also on lipid
infer that the asymmetric transmembrane distri- bilayer properties. The presence of a high con-
bution of magainin 2 (i.e., localization of centration of cholesterol in the bilayer greatly
magainin 2 only in the outer monolayer) induces suppressed the translocation of CF-TP10 from
pore formation. the outer leaflet to the inner one, but higher con-
In contrast, the result for TP10 showed a dif- centrations of CF-TP10 induced pore formation
ferent pattern. Figure 3.6b shows the time course in the same bilayer, through which CF-TP10 rap-
of FI in the interaction of CF-TP10 with a single idly translocated across the bilayer (Islam et al.
PG/PC-GUV containing AF647 in its lumen 2017). Therefore, the relationship between the
(Islam et al. 2014a). The rim intensity due to CF-TP10-induced pore formation and the loca-
CF-TP10 gradually increased and reached a final, tion of CF-TP10 in the bilayers containing high
steady value at 125 s (Fig. 3.6b). On the other concentrations of cholesterol was almost the
hand, the FI of the GUV lumen due to AF647 same as that of magainin 2 in PG/PC (4/6) mem-
started to decrease at 212 s, indicating that the brane (Fig. 3.6a).
28 M. Hasan and M. Yamazaki

Fig. 3.6 Relationship between the location of AMPs and of AF647 and location of CF-TP10 in a PG/PC (2/8)-
pore formation. (a) Membrane permeation of AF647 and GUV. Time course of the change in FI due to AF647 and
location of CF-magainin 2 in a PG/PC (4/6)-GUV. Time CF-TP10 of the GUV interacting with 1.9 μM CF-TP10 is
course of the change in FI due to AF647 and CF-magainin shown. The red and green points correspond to the nor-
2 of the GUV interacting with 31 μM CF-magainin 2/ malized FI of AF647 inside the GUV and of CF-TP10 in
magainin 2 is shown. The red squares and green triangles the rim of the GUV, respectively (a, b are reprinted from
correspond to the normalized FI of AF647 inside the GUV Karal et al. 2015; Islam et al. 2014a, respectively, with
(left axis) and the FI of CF-magainin 2 in the rim of the permission from the American Chemical Society)
GUV (right axis), respectively. (b) Membrane permeation

3.5 Effect of Membrane Tension aspiration pressure, producing a small tension

(or Stretching (σ ~ 0.5 mN/m) in the GUV membrane. In the
of Membranes) on AMP-­ case of magainin 2, the fractional change in the
Induced Pore Formation area of a GUV membrane (δ) increased rapidly
with time after the interaction of magainin 2 with
3.5.1 AMP-Induced Stretching the GUV started, reaching a steady value of δ
of Lipid Bilayers (δST) within a short time (~30 s). The δST increased
and Membrane Tension with an increase in magainin 2 concentration in
the buffer, C (Fig. 3.7a) (Karal et al. 2015). After
Generally, there is a possibility that the interac- conversion of C to the surface concentration of
tion with peptides/proteins such as AMPs affects magainin 2, X (mol/mol), the result indicated that
the mechanical properties of plasma membranes δST ∝ X (Fig. 3.7b). The combination of Figs. 3.1f
and lipid bilayers. To measure the change in area and 3.7b indicated that the rate constant of
of a GUV, we can use the micropipette aspiration magainin 2-induced pore formation (kp) had a
method (Rawictz et al. 2000), where the GUV is positive correlation with δST (Fig. 3.7c). At a high
fixed at the tip of the micropipette by a small concentration of magainin 2, during the steady
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 29

Fig. 3.7 Effect of binding of AMP to a GUV on the GUV lumen and of CF-magainin 2 at its rim, respectively.
membrane area. (a) The dependence of the fractional Open triangles represent δ (right axis). (e) Time course of
change in area of single GUVs, δ, on the magainin 2 con- the FI of a PG/PC (2/8)-GUV and δ interacting with
centration in buffer A. (b) Dependence of δ on the 0.3 μM CF-TP10 at σ = 1.0 mN/m. Red line and green
magainin 2 surface concentration, X (mmol/mol). (●) PG/ squares correspond to FI of AF647 in the GUV lumen and
PC (4/6)- and (▲) PG/PC (3/7)-GUV. (c) Dependence of of CF-TP10 at the rim of the GUV, respectively. Open
kP on δ for magainin 2. (d) Time course of the FI of a PG/ squares represent δ (a, b, d, and e are reprinted from Karal
PC (4/6)-GUV and δ interacting with 15 μM CF-magainin et al. 2015; Islam et al. 2017, respectively, with permis-
2/magainin 2 at σ = 0.5 mN/m. Red line and green solid sion from the American Chemical Society)
triangles represent the normalized FI of AF647 in the

state of stretching (i.e., δ = δST), aspiration of ment of the magainin 2-induced change in the
GUV (which is induced by pore formation in the area, the rim intensity due to CF-magainin 2, and
lipid bilayer) occurred stochastically. Figure 3.7d the membrane permeation of AF647 using
shows the result of the simultaneous measure- CLSM. Both the rim intensity and δ rapidly
30 M. Hasan and M. Yamazaki

increased to steady values within 50 s and then et al. 2017). It is well known that the mechano-
remained constant for a long time (~200 s) until sensitive ion channels are activated when plasma
pore formation. We could not detect any increase membranes are stretched (Sukharev et al. 1994;
in the rim intensity just before pore formation (as Sachs 2010). Therefore, it will be useful to com-
shown in Fig. 3.6a), because the tension caused pare AMP-induced pore formation with the acti-
by the aspiration of a GUV by a micropipette vation of the mechanosensitive channels.
increases the rate of opening (i.e., the increase in
the radius) of the magainin 2-induced pore, and
hence, the GUV is immediately aspirated into the 3.5.3  ffect of Asymmetric Lipid
E
micropipette after pore formation starts. This Packing
result indicated that the translocation of magainin
2 from the outer leaflet to the inner one occurred The magainin 2-related results described in pre-
after pore formation, consistent with the infer- vious sections suggested that the stretching of the
ence that the asymmetric distribution of magainin inner leaflet is a main driving force of magainin
2 in the bilayer induces pore formation (Fig. 3.6a). 2-induced pore formation (Karal et al. 2015).
In contrast, in the case of TP10, δ increased However, there was no direct experimental evi-
with time after the interaction of TP10 with the dence for this hypothesis. Recently, Hasan et al.
GUV started until aspiration of the GUV (2018a) developed a new method to prepare
(Fig. 3.7e) (Islam et al. 2017). As the transloca- spherical GUVs with asymmetric lipid composi-
tion of CF-TP10 from the outer leaflet to the tions in the two monolayers, such that a low con-
inner one proceeds, δ increases with time until centration of lyso-PC (hereafter LPC) exists only
pore formation. This result supports the idea that in the inner leaflet (e.g., PG/PC/LPC (40/59/1;
CF-TP10 translocates across the bilayer and that inner)-PG/PC (40/59; outer)-GUVs). In these
the binding of CF-TP10 in both monolayers GUVs, the lipid packing in the inner leaflet was
induces their stretching. larger than that in the outer one. Magainin
2-induced area increase (δST) and the rate con-
stant of magainin 2-induced pore formation (kp)
3.5.2 Effect of Application decreased with increasing LPC concentration in
of Tension Due to an External the inner leaflet (i.e., the increase in asymmetric
Force packing) (Hasan et al. 2018a). These results
clearly support the idea that the stretching of the
We can control the membrane tension of single inner leaflet plays an important role in magainin
GUVs by applying an external force with micro- 2-induced pore formation. Based on these results,
pipette aspiration (Evans et al. 2003; Levadny we formulated a new quantitative theory on the
et al. 2013). This aspect is also one of the advan- initial stage of pore formation induced by
tages of experimentation using single GUVs. magainin 2 (Hasan et al. 2018a). A theoretical
Magainin 2-induced pore formation is affected equation of kp as a function of magainin 2 surface
greatly by membrane tension due to an external concentration (X) provided a good fit to the
force; the rate constant of magainin 2-induced experimental results (Fig. 3.1f). This theory can
pore formation, kp, increased with an increase in explain the effect on kp of the LPC concentration
membrane tension (Karal et al. 2015). This result, in the inner leaflet (Hasan et al. 2018a).
as well as the result that the kp increased with an
increase in magainin 2-induced stretching of the
bilayer, indicated that the magainin 2-induced 3.6 Concluding Remarks
pore is a stretch-activated pore (Karal et al. 2015).
The membrane tension also increased the rate of In this mini-review, we summarized the present
translocation of CF-TP10 across the bilayer and state of knowledge on the interactions of AMPs
that of CF-TP10-induced pore formation (Islam with lipid vesicles that were obtained using the
3 Elementary Processes and Mechanisms of Interactions of Antimicrobial Peptides… 31

single GUV method. AMPs exhibit three MoAs on form stable protein-permeable pores of tunable size.
J Biol Chem 2013(288):33241–33252
lipid bilayers. For a limited number of AMPs, we Evans E, Heinrich V, Ludwig F, Rawicz W (2003)
have obtained detailed information on the elemen- Dynamic tension spectroscopy and strength of bio-
tary processes of these MoAs, indicating that vari- membranes. Biophys J 85:2342–2350
ous processes are mediated by AMPs. On the basis Fuertes G, Garcia-Sáez A, Esteban-Martin S, Giménez
D, Sánchez-Muñoz OL, Schwille P, Salgado J (2010)
of this information, we have proposed a mecha- Pores formed by Baxα5 relax to a smaller size and
nism for the initial stage of magainin 2-induced keep at equilibrium. Biophys J 99:2917–2925
pore formation (Hasan et al. 2018a). However, we Gregory SM, Pokorny A, Almeida PFF (2009) Magainin
do not know the mechanism of the AMP-induced 2 revisited: a test of the quantitative model for the all-­
or-­
none permeabilization of phospholipid vesicles.
local rupture (type B MoA) and that of the entry of Biophys J 96:116–131
AMPs into the vesicle lumen (or bacterial cyto- Hasan M, Karal MAS, Levadnyy V, Yamazaki M (2018a)
plasm) without pore formation (type C MoA). If Mechanism of initial stage of pore formation induced
we could examine other AMPs using single GUV by antimicrobial peptide magainin 2. Langmuir
34:3349–3362
methods, we might identify other kinds of MoAs Hasan M, Saha SK, Yamazaki M (2018b) Effect of mem-
and other elementary processes within the present brane tension on transbilayer movement of lipids.
MoAs. Further investigations of the MoAs of J Chem Phys 148:245101
AMPs and the elementary processes of interaction Hwang PM, Vogel HJ (1998) Structure-function relation-
ships of antimicrobial peptides. Biochem Cell Biol
of AMPs with lipid bilayers will be essential. 76:235–246
Even in the case of AMPs (such as magainin 2 Islam MZ, Ariyama H, Alam JM, Yamazaki M (2014a)
and LfcinB) that attack the lipid membrane Entry of cell-penetrating peptide transportan 10 into a
regions of bacterial and fungal plasma mem- single vesicle by translocating across lipid membrane
and its induced pores. Biochemistry 53:386–396
branes, other factors of these microbial cells may Islam MZ, Alam JM, Tamba Y, Karal MAS, Yamazaki
affect the interaction of AMPs with the lipid M (2014b) The single GUV method for revealing
membrane regions of the plasma membranes. the functions of antimicrobial, pore-forming toxin,
However, we believe that these factors are addi- and cell-penetrating peptides or proteins. Phys Chem
Chem Phys 16:15752–15767
tional to the fundamental characteristics of the Islam MZ, Sharmin S, Levadnyy V, Shibly SUA,
interactions between AMPs and lipid bilayers. A Yamazaki M (2017) Effects of mechanical properties
close comparison between the results of interac- of lipid bilayers on entry of cell-penetrating peptides
tions of AMPs with microbial cells and those of into single vesicles. Langmuir 33:2433–2443
Islam MZ, Sharmin S, Moniruzzaman M, Yamazaki
GUVs of lipid bilayers is expected to provide M (2018) Elementary processes for the entry of
deeper insight into the elementary processes of cell-­
penetrating peptides into lipid bilayer vesicles
action of AMPs and the mechanism of their anti- and bacterial cells. Appl Microbiol Biotechnol
microbial activity. 102:3879–3892
Israelachvili JN (1992) Intermolecular & surface forces,
2nd edn. Academic, New York
Karal MAS, Alam JM, Takahashi T, Levadny V, Yamazaki
References M (2015) Stretch-activated pore of antimicrobial pep-
tide magainin 2. Langmuir 31:3391–3401
Alam JM, Kobayashi T, Yamazaki M (2012) The single Levadny V, Tsuboi T, Belaya M, Yamazaki M (2013) Rate
giant unilamellar vesicle method reveals lysenin constant of tension-induced pore formation in lipid
induced pore formation in lipid membranes containing membranes. Langmuir 29:3848–3852
sphingomyelin. Biochemistry 51:5160–5172 Lipowsky R, Sackmann E (eds) (1995) Structure and
Baumgart T, Hess ST, Webb WW (2003) Imaging coex- dynamics of membranes. Elsevier Science BV,
isting fluid domains in biomembrane models coupling Amsterdam
curvature and line tension. Nature 425:821–824 Ludtke SJ, He K, Heller KH, Harroun TA, Yang L, Huang
Bigay J, Antonny B (2012) Curvature, lipid packing, and HW (1996) Membrane pores induced by magainin.
electrostatics of membrane organelles: defining cel- Biochemistry 35:13723–13728
lular territories in determining specificity. Dev Cell Madani F, Lindberg S, Langel Ű, Futaki S, Gräslund
23:886–895 A (2011) Mechanisms of cellular uptake of cell-­
Bleicken S, Landeta O, Landajuela L, Basañez G, García-­ penetrating peptides. J Biophys 2011:414729
Sáez AJ (2013) Proapoptotic Bax and Bak proteins Matsuzaki K, Murase K, Fujii N, Miyajima K (1995)
Translocation of a channel-forming antimicrobial pep-
32 M. Hasan and M. Yamazaki

tide, magainin 2, across lipid bilayers by forming a Rawictz W, Olbrich KC, McIntosh T, Needham D, Evans
pore. Biochemistry 34:6521–6526 E (2000) Effect of chain length and unsaturation on
Matsuzaki K, Murase O, Fujii N, Miyajima K (1996) elasticity of lipid bilayers. Biophys J 79:328–339
An antimicrobial peptide, magainin 2, induced Sachs F (2010) Stretch-activated ion channels: what are
rapid flip-­flop of phospholipids coupled with pore they? Physiology 25:50–56
formation and peptide translocation. Biochemistry Sandre O, Moreaux L, Brochard-Wyard F (1999)
35:11361–11368 Dynamics of transient pores in stretched vesicles. Proc
Matsuzaki K, Sugishita K, Ishibe N, Ueha M, Nakata S, Natl Acad Sci U S A 96:10591–10596
Miyajima K, Epand RM (1998) Relationship of mem- Sukharev SI, Blount P, Martinac B, Blattner FR, Kung
brane curvature to the formation of pores by magainin C (1994) A large-conductance mechanosensitive
2. Biochemistry 37:11856–11863 channel in E. coli encoded by mscL alone. Nature
McLaughlin S, Murray D (2005) Plasma membrane phos- 368:265–268
phoinositide organization by protein electrostatics. Tamba Y, Yamazaki M (2005) Single giant unilamellar
Nature 438:605–611 vesicle method reveals effect of antimicrobial peptide,
Melo MN, Ferre R, Castanho ARB (2009) Antimicrobial magainin 2, on membrane permeability. Biochemistry
peptides: linking partition, activity and high 44:15823–15833
membrane-­bound concentrations. Nat Rev Microbiol Tamba Y, Yamazaki M (2009) Magainin 2-induced pore
8:1–5 formation in membrane depends on its concentration
Moghal MMR, Islam MZ, Sharmin S, Levadnyy V, in membrane interface. J Phys Chem B 113:4846–4852
Moniruzzaman M, Yamazaki M (2018) Continuous Tamba Y, Ohba S, Kubota M, Yoshioka H, Yoshioka H,
detection of entry of cell-penetrating peptide trans- Yamazaki M (2007) Single GUV method reveals inter-
portan 10 into single vesicles. Chem Phys Lipids action of tea catechin (-)-epigallocatechin gallate with
212:120–129 lipid membranes. Biophys J 92:3178–3194
Moniruzzaman M, Alam JM, Dohra H, Yamazaki M Tamba Y, Ariyama H, Levadny V, Yamazaki M (2010)
(2015) Antimicrobial peptide lactoferricin B-induced Kinetic pathway of antimicrobial peptide magainin
rapid leakage of internal contents from single giant 2-induced pore formation in lipid membranes. J Phys
unilamellar vesicles. Biochemistry 54:5802–5814 Chem B 114:12018–12026
Moniruzzaman M, Islam MZ, Sharmin S, Dohra H, Tanaka T, Sano R, Yamashita Y, Yamazaki M (2004)
Yamazaki M (2017) Entry of a six-residue antimicro- Shape changes and vesicle fission of giant unilamellar
bial peptide derived from lactoferricin B into single vesicles of liquid-ordered phase membrane induced by
vesicles and Escherichia coli cells without damaging lysophosphatidylcholine. Langmuir 20:9526–9534
their membranes. Biochemistry 56:4419–4431 Wade D, Boman A, Wahlin B, Drain CM, Andreu A,
Müller P, Schiller S, Wieprecht T, Dathe M, Herrmann A Boman HG, Merrifield RB (1990) All-D amino acid-­
(2000) Continuous measurement of rapid transbilayer containing channel-forming antibiotic peptides. Proc
movement of a pyrene-labeled phospholipid analogue. Natl Acad Sci U S A 87:4761–4765
Chem Phys Lipids 106:89–99 Wakabayashi H, Matsumoto H, Hashimoto K, Teraguchi
Nekhotiaeva N, Elmquist A, Rajarao GK, Hällbrink M, S, Takase M, Hayasawa H (1999) N-acylated and D
Langel C, Good L (2004) Cell entry and antimicro- enantiomer derivatives of a nonamer core peptide of
bial properties of eukaryotic cell-penetrating peptides. lactoferricin B showing improved antimicrobial activ-
FASEB J 18:394–396 ity. Antimicrob Agents Chemother 43:1267–1269
Propheter DC, Chara AL, Harris TA, Ruhn KA, Hooper Yamazaki M (2008) The single GUV method to reveal
LV (2017) Resistin-like molecule β is a bactericidal elementary processes of leakage of internal contents
protein that promotes spatial segregation of the micro- from liposomes induced by antimicrobial substances.
biota and the colonic epithelium. Proc Natl Acad Sci Adv Planar Lipid Bilayers Liposomes 7:121–142
U S A 114:11027–11033 Yang LT, Weiss M, Lehrer RI, Huang HW (2000)
Qian S, Wang W, Yang L, Huang HW (2008) Structure of Crystallization of antimicrobial pores in membranes:
transmembrane pore induced by Bax-derived peptide: magainin and protegrin. Biophys J 79:2002–2009
evidence for lipidic pores. Proc Natl Acad Sci U S A Zasloff M (2002) Antimicrobial peptides of multicellular
105:17379–17383 organisms. Nature 415:389–395
 The Mechanisms of Action
of Cationic Antimicrobial Peptides 4
Refined by Novel Concepts
from Biophysical Investigations

Christopher Aisenbrey, Arnaud Marquette,

and Burkhard Bechinger

Abstract morphology, locally and globally, and in par-

Even 30 years after the discovery of magain- ticular of cationic amphipathic helices that
ins, biophysical and structural investigations partition into the membrane interface. This
on how these peptides interact with mem- concept has been extended in this review to
branes can still bear surprises and add new include more recent ideas on soft membranes
interesting detail to how these peptides exert that can adapt to external stimuli including
their antimicrobial action. Early on, using ori- membrane-disruptive molecules. In this man-
ented solid-state NMR spectroscopy, it was ner, the lipids can change their shape in the
found that the amphipathic helices formed by presence of low peptide concentrations,
magainins are active when being oriented par- thereby maintaining the bilayer properties. At
allel to the membrane surface. More recent higher peptide concentrations, phase transi-
investigations indicate that this in-planar tions occur which lead to the formation of
alignment is also found when PGLa and pores and membrane lytic processes. In the
magainin in combination exert synergistic context of the molecular shape concept, the
pore-forming activities, where studies on the properties of lipopeptides, including surfac-
mechanism of synergistic interaction are tins, are shortly presented, and comparisons
ongoing. In a related manner, the investigation with the hydrophobic alamethicin sequence
of dimeric antimicrobial peptide sequences are made.
has become an interesting topic of research
which bears promise to refine our views how Keywords
antimicrobial action occurs. The molecular Magainin · PGLa · Cecropin · LL37 ·
shape concept has been introduced to explain Surfactin · Alamethicin · Membrane topology
the effects of lipids and peptides on membrane · Membrane pore · Membrane macroscopic
phase · SMART model · Carpet model ·
Toroidal pore model · Peptide-lipid interac-
C. Aisenbrey · A. Marquette tions · Molecular shape concept
Université de Strasbourg/CNRS, UMR7177, Institut
de Chimie, Strasbourg, France
B. Bechinger (*)
Université de Strasbourg/CNRS, UMR7177, Institut
de Chimie, Strasbourg, France
Faculté de chimie, Institut le Bel, Strasbourg, France
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 33

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_4
34 C. Aisenbrey et al.

Abbreviations and fast response to fight bacterial and fungal

infections (Boman 1995; Zasloff 2002). A multi-
Aib α-Aminobutyric acid tude of AMPs belonging to different classes have
AMP Antimicrobial peptide been detected in the plant and animal kingdom,
ATR FTIR Attenuated total reflection Fourier including in the human body (Agerberth et al.
transform infrared 1995). These complement a multitude of antibi-
CD Circular dichroism otic peptides produced by microorganisms that
CL Cardiolipin have early on been identified and investigated
CMC Critical micelle concentration (Leitgeb et al. 2007; Rautenbach et al. 2016a, b).
DLPC 1 , 2 - L a u r o y l - s n - g l y c e r o - 3 - The corresponding databases encompass thou-
phosphocholine sands of sequences (Pirtskhalava et al. 2016;
DMPC 1,2-Dimyristoyl-sn-glycero-3- Wang et al. 2016), and early on several of them
zhosphocholine have been investigated by cell biological and bio-
DMPG 1,2-Dimyristoyl-sn-glycero-3-­ physical approaches in order to reveal their
phospho-(1′-rac-glycerol) mechanisms of action (Sansom 1991; Bechinger
DOPC 1 , 2 - D i o l e o y l - s n - g l y c e r o - 3 -­ 1997). Magainins from the African clawed frog
phosphocholine were among the first of those peptides found in
DOPG 1 , 2 - D i o l e o y l - s n - g l y c e r o - 3 -­ the animal kingdom, where the antimicrobial
phospho-(1′-rac-glycerol) activity has been described about 30 years ago
DPC Dodecylphosphocholine (Zasloff 1987), thereby adding a new exciting
GUV Giant unilamellar vesicle perspective to earlier publications describing
ITC Isothermal titration calorimetry peptides from toads and frogs (Kiss and Michl
LUV Large unilamellar vesicle 1962; Giovannini et al. 1987).
MD Molecular dynamics The very first attempts to develop these pep-
MIC Minimal inhibitory concentration tides into commercial products were unsuccessful
NMR Nuclear magnetic resonance which resulted in a temporary decline in interest
PC Phosphatidylcholine on this topic around the turn of the millennium.
PE Phosphatidylethanolamine However, the rapid increase in multiresistant
PG Phosphatidylglycerol pathogens (Chang et al. 2003) has again stimu-
POPC 1-Palmitoyl-2-oleoyl-sn-glycero- lated research activities on AMPs because they
3-phosphocholine promise to be paradigms for new classes of antibi-
POPE 1-Palmitoyl-2-oleoyl-sn-glycero- otics with mechanisms of action that are less
3-phosphoethanolamine prone to be neutralized by microbial resistance
POPG 1-Palmitoyl-2-oleoyl-sn-glycero- (Zasloff 2002). To protect such polypeptides from
3-phospho-(1′-rac-­glycerol) proteases, special formulations inside nanostruc-
POPS 1-Palmitoyl-2-oleoyl-sn-glycero- tures, attachment to surfaces, or the use of unnatu-
3-phosphoserine ral amino acids is explored (Yang et al. 2014;
SMART Soft Membranes Adapt and Yuksel and Karakecili 2014; Reijmar et al. 2016).
Respond, also Transiently Furthermore, mimetics in the shape of small mol-
ecules or foldamers have also been developed
(Violette et al. 2006; Arnusch et al. 2012; Ghosh
and Haldar 2015; Ghosh et al. 2018).
4.1 Introduction Magainin and other antimicrobial peptides
discussed in this paper are thought to interfere
The innate immune system of higher organisms with the barrier function of bacterial membranes
provides a first line of defense against a multitude rather than specific membrane receptors
of pathogenic microorganisms, where the release (Bechinger 2015). Whereas molecules whose
of antimicrobial peptides (AMP) is a powerful interactions are with proteinaceous receptors can
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 35

be made inefficient by one or a few changes in Table 4.1 Amino acid sequences of selected antimicro-
bial peptides
amino acid sequence, polypeptides that act by
disrupting the lipid bilayer physicochemical Magainin 2 GIGKF LHSAK KFGKA FVGEI
MNS
properties are less likely to become inactivated
PGLa GMASK AGAIA GKIAK VALKA
by resistance (Rollins-Smith et al. 2002). Indeed L-NH2
the amphipathic nature of AMPs has been found LL37 LLGDF FRKSK EKIGK EFKRI
essential and can be achieved by helical (Sansom VQRIK DFLRN LVPRT ES
1991; Bechinger 1997), cyclic (Cao et al. 2018; Cecropin P1 SWLSK TAKKL ENSAK KRISE
Laurencin et al. 2018; Tsutsumi et al. 2018; Zhao GIAIA IQGGP R
Cecropin A KWKLF KKIEK VGQNI RDGII
et al. 2018), and/or β-sheet arrangements (Hong
KAGPA VAVVG QATQI AK-NH2
and Su 2011; Salnikov et al. 2011; Rautenbach LAH4 KKALL ALALH HLAHL ALHLA
et al. 2016a, b; Sychev et al. 2018; Usachev et al. LALKK A-NH2
2017). Thus the insights gained from the studies Melittin GIGAV LKVLT TGLPA LISWI
of cationic amphipathic antimicrobial peptides KRKRQ Q-NH2
have stimulated the design of a number of small Alamethicin Ac-Aib-Pro-Aib-Ala-Aib-Aib-Gln-­
(F50/7) Aib-Val-Aib-Gly-Leu-Aib-Pro-Val-
amphipathic molecules (Arnusch et al. 2012; Aib-­Aib-Gln-Gln-Phl
Ghosh et al. 2014), pseudopeptides (Porter et al. The one-letter code is used for peptides made from con-
2002; Patch and Barron 2003; Kuroda and ventional amino acids. For the alamethicin sequence, the
DeGrado 2005; Violette et al. 2006; Makovitzki three-letter code with the following nonstandard residues
et al. 2008; Scott et al. 2008; Rotem and Mor is used: Aib α-aminoisobutyric acid, Phl L-phenylalaninol,
Ac- for acetyl- and -NH2 for the carboxamide termini,
2009; Palermo and Kuroda 2010; Laurencin et al. respectively
2018), and polymers (Rank et al. 2017) with
potent antimicrobial properties.
In this review, some of the laboratory-based the cell interior (Roversi et al. 2014). This results
biophysical techniques that have provided valu- in complex patterns of metabolic reactions by the
able information on the interactions of peptides bacterial cells (Kozlowska et al. 2014; Cardoso
shall be introduced and the corresponding data et al. 2017). Because many peptides have been
presented (Lear et al. 1988; Killian et al. 1998; shown to also modulate the immune response of
Harzer and Bechinger 2000; Bechinger 2011) the host organisms, they are also called “host
including work on the synergistic activities defense peptides” (McCafferty et al. 1999’ Holzl
between PGLa and magainin 2 two members of et al. 2008; Diamond et al. 2009; Steinstraesser
the magainin family (see Table 4.1 for amino acid et al. 2010). Such cell biological studies shall not
sequences). Even after 30 years of biophysical be part of this review.
studies on how these AMPs interact with lipid
bilayers, new details are revealed, and as a conse-
quence, exiting new research directions open up 4.2 Electrophysiological
(Salnikov et al. 2010; Hong and Su 2011). It is Recordings
now clear that such amphipathic membrane-­
active peptides are very dynamic in nature and Electrophysiological experiments provide an
can adopt a large diversity of conformations and interesting method to observe and characterize
topologies whose exchange and interactions are membrane interactions of peptides and proteins.
governed by multiple equilibria (Bechinger One approach is to measure the ionic conductivi-
2015). These cationic linear peptides are random ties across small patches of a lipid bilayer (rang-
coil in solution and adopt their three-dimensional ing from several micrometers up to hundreds of
amphipathic helical structure only when interact- micrometers) separating two electrodes connected
ing with membranes. By disrupting the integrity to a voltage-clamp amplifier (Montal and Mueller
of bacterial and fungal membranes, they inhibit 1972). The electrodes can be positioned in differ-
the growth of microorganisms and/or enter into ent chambers or constitute the inside and outside
36 C. Aisenbrey et al.

of dedicated pipettes. When lipids alone are used, such as calcein. At a concentration of some tens
a tight electric seal (gigaseal) is established; thus of millimolar, the dye molecules are close to each
no ions can flow across the lipid bilayer. In the other resulting in the quenching of their radiative
presence of small amounts of pore-­forming poly- emission. Once unilamellar vesicles have been
peptides, a decrease in ohmic resistance occurs formed, the fluorophore on the outside is
(“membrane openings”), and ionic current can be exchanged by passage through a gel filtration col-
recorded, when at the same time the transmem- umn. Using this method, LUVs and the encapsu-
brane voltage is electronically regulated to remain lated dye can be easily separated from the outside
constant (“voltage clamp”). Ideally, single events buffer. Care should be taken that the osmolarities
are observed and characterized in frequency, of the buffer solutions at the interior and exterior
duration, and conductivity. More recent chip tech- of the liposomes match each other to assure that
nology working with arrays of freestanding lipid there is no pressure gradient across the mem-
bilayers has simplified the handling and provides brane (Marquette et al. 2008). Under these pre-
an increased throughput for such single-channel cautions, vesicles are stable for days or even
measurements (del Rio Martinez et al. 2015). weeks keeping the dye encapsulated and ready
Notably, discrete single-channel openings/ for future experiments.
events are rather difficult to observe for magain- Upon addition of peptide, the formation of
ins and cecropins because most of the time, these pores can be measured by monitoring the increase
peptides lyse the membranes. However, in some in the fluorescence intensity due to the escape of
electrophysiological experiments, discrete multi- the dye from the vesicles. Fluorescence is contin-
level conductivities have been recorded uously measured while irradiating the sample
(Christensen et al. 1988; Duclohier et al. 1989; with monochromatic light tuned to the most
Cruciani et al. 1991; Watanabe and Kawano intense absorption band of the chromophore.
2016), which early on was taken as an indicator Variants of the technique have been developed to
for transmembrane helical bundle formation measure the kinetics in more detail and to distin-
(Tieleman et al. 2002). However, unlike the well-­ guish between all-or-nothing and graded release.
defined ion channels formed by alamethicin, For example, in the fluorescence quenching assay,
those recorded in the presence of cationic the fluorophore is encapsulated with a quencher.
amphipathic peptides are erratic and character- Upon all-or-nothing release, both leak out without
ized by large variations (Christensen et al. 1988; changes in fluorescence inside the vesicles. In
Duclohier et al. 1989; Cruciani et al. 1991). contrast, a graded release results in a fluorescence
Therefore, despite some similarities, electro- increase when the quencher inside the vesicles is
physiological recordings between these peptides diluted (Ladokhin et al. 1995; Clark et al. 2011).
exhibit rather pronounced differences as do the Alternatively, due to an about tenfold difference
physicochemical properties of their primary in fluorescence lifetime between entrapped and
sequences, i.e., number of charges and overall free dye, their respective proportions can be esti-
hydrophobicity (Table 4.1). mated using biexponential fits. This has been used
to develop an approach to distinguish all-or-noth-
ing and graded release (Patel et al. 2014). Yet
4.3 Fluorescence Spectroscopy another method is based on the direct observation
of giant unilamellar vesicles under the micro-
4.3.1 Fluorophore Release scope which can thus be studied one-by-one indi-
vidually (Tamba and Yamazaki 2005).
An alternative method to monitor the formation Using fluorescence spectroscopy, the forma-
of membrane openings has been established tion of magainin pores was investigated, and the
based on fluorescence spectroscopic approaches. kinetics of calcein release from individual giant
In these experiments vesicles are prepared in the unilamellar vesicles (GUVs) made from DOPC/
presence of high concentrations of fluorophore DOPG at different molar ratios was monitored
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 37

(Islam et al. 2014). Membrane pore formation tryptophan absorb and emit light in the near-UV
from DOPC/DOPG GUVs sets in at peptide con- region. Intensity modulation of emission spectra
centrations of 0.7 mole% (Tamba and Yamazaki and/or wavelength shift is generally observed
2009). After the addition of peptide, it takes min- when the amino acids change environments, for
utes before the release of fluorophores sets in, but example, when they move from aqueous buffers
then the vesicles, which are several micrometers into a lipid membrane. This fluorescence-based
in diameter, empty within only 30 s. The fast approach has the great advantage of being nonde-
release is suggestive that magainin pore forma- structive for biological samples and require only
tion in GUVs follows an all-or-nothing mecha- small amounts of material.
nism rather than gradual diffusion through the Among the three naturally occurring amino
pore. Similar conclusions were drawn from cal- acid chromophores, tryptophan emission has the
cein release experiments from suspensions of longest wavelength emission and the highest
large unilamellar vesicles made from POPC and absorption and emission yields, thereby being
POPG (Gregory et al. 2008). However, the topic most sensitive to measure. Several groups have
turns out more complex because the mechanism taken advantage of measuring the blue shift and/
is lipid dependent and a graded release is observed or intensity changes in the emission spectrum to
when the POPG content is reduced from 50 to quantify an apparent binding constant between
20 mole% (Gregory et al. 2008). Once the pore peptide and a large variety of model membranes
has formed, the subsequent fluorophore release is (Matos et al. 2010; Zanin et al. 2013; Michalek
a two-stage process where an initial fast release is et al. 2014). To monitor the membrane-induced
due to an unbalance of the bilayer because as a changes in fluorescence, a solution of peptide is
first step magainin solely associates with the out- continuously excited at a fixed wavelength, while
side monolayer (Tamba et al. 2010). The move- the dispersed emission spectra are recorded after
ment of the peptide from the outer to the inner each addition of aliquot of vesicle suspension.
leaflet results in the transient formation of very The wavelength of the maximum emission inten-
large pores concomitant with an equilibration of sity and/or its amplitude is then plotted against
the peptide density (Tamba et al. 2010). In the the lipid concentration, and a fitting procedure is
following, a slower release of fluorophore sets in, applied to extract apparent peptide-to-lipid bind-
but even these persistent openings are large ing constants. Antimicrobial peptide association
enough to allow for the passage of molecules to membranes of different composition was used
with a hydrodynamic radius of 3 nm (Tamba using this method (Vogt and Bechinger 1999),
et al. 2010). In follow-up investigations, it was including for magainin 2 and its interactions with
shown that antimicrobial action was associated PGLa within lipid bilayers (Matsuzaki et al.
with a preferential interfacial localization (rather 1998).
than insertion into the hydrophobic core of the
membrane) and correlated to the Gibbs free
energy of membrane association rather than 4.3.3 Fluorescence: Depth
membrane insertion (Clark et al. 2011). of Membrane Insertion

Acrylamide and free radicals are known to

4.3.2 Fluorescence: Natural quench tryptophan fluorescence, and this prop-
Chromophores erty has been used to monitor the depth of inser-
and Membrane Partitioning tion of AMPs into membranes (Caputo and
London 2003). Collisional quenching is enhanced
The intrinsic emission properties of AMPs con- with the contact frequency between the fluoro-
taining chromophores naturally, by mutagenesis phore and its environment and so with the dis-
or by chemical attachment, can be exploited to tance between the fluorescent amino acids and
gain insight into peptide-lipid interactions. The the position of the quencher. Whereas acrylamide
amino acid residues phenylalanine, tyrosine, and is water soluble, the hydrophobic alkyl chain-­
38 C. Aisenbrey et al.

bound doxyl radicals are positioned within the minated at either positions 6 and 7, 9 and 10, or
membrane interior. This method has been cali- 11 and 12 (Michalek et al. 2014). Fluorescence
brated by studying the emission properties of quenching in the presence of paramagnetic agents
tryptophan residues of different mutants of a allowed the determination of an in-planar align-
membrane-spanning helix (Caputo and London ment of Phe to Trp mutants of magainin 2, where
2003). The quenching efficiency of the emission all three tryptophans investigated are localized at
by acrylamide molecules in solution was found to about 1 nm from the bilayer center (Matsuzaki
be decreasing with the distance from the mem- et al. 1994).
brane surface; the one by the membrane-inserted
radical depends on the penetration depth of the
tryptophan. The quenching efficiencies can thus 4.3.4 Fluorescence Self-Quenching
serve as an atomics-scale ruler to localize the
position of the tryptophan relative to the mem- A fluorescence self-quenching effect has been
brane surface. developed to quantitatively evaluate if peptides
Brome has also been reported as an efficient distribute randomly at the membrane surface or
short-range fluorescence quencher. Brominated in a more heterogeneous manner (Fig. 4.1). Many
phospholipids have been introduced from the models of peptide organization in membranes
beginning of the 1990s to estimate the position of assume that there are oligomeric structures of
tryptophan residues within the membrane through peptide monomers. For example, the toroidal
the use of different labeling positions along the pore model assumes the formation of well-­
phospholipid fatty acyl chains (Bolen and defined transmembrane multimers (Ludtke et al.
Holloway 1990). As an example of the method, 1996; Matsuzaki 1998). In contrast, the carpet
the membrane penetration depth of tryptophans (Shai 1999) or the SMART model (Bechinger
within the amino-terminal domain of huntingtin 2015) postulate a looser and less well-defined
was investigated taking advantage of lipids dibro- arrangement of the peptides at the membrane sur-

Fig. 4.1 Fluorescence self-quenching to analyze the pep- increase in fluorescence is observed (black data points and
tide distribution. The fluorescence signal is measured at a line in a). In the presence of peptide aggregates or when
fixed peptide-to-lipid ratio, where the fraction of peptides are localized in more confined domains (c), add-
fluorophore-­carrying peptide is increased in a stepwise ing more of the labeled peptide is accompanied in self-­
manner (a). When the fluorophore-labeled peptides are quenching of the fluorescence and a nonlinear fluorescence
distributed randomly (illustrated in panel b), a linear increase (red data points and line in a)
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 39

face. Nevertheless, these concepts do not exclude cients of some well-chosen dyes with almost
that high local order and/or nonrandom peptide unity emission yields, single molecules can be
assemblies play an important role in enabling and detected when highly sensitive instruments are
modulating the membrane-disruptive function. used.
Indeed, for the designed antibiotic peptide LAH4, The interaction of PGLa with magainin 2
the formation of nematic phases at the membrane inside membranes was probed by FRET
surface has been demonstrated which can be (Marquette et al. 2015). The peptides were
modulated by the charge of the lipids and the salt labeled with NBD, and rhodamine fluorophores
concentration of the surrounding buffer and LUVs made of POPC/POPS or of POPE/
(Fig. 4.1c) (Aisenbrey and Bechinger 2014). POPG were chosen as model bilayers. At high
Self-quenching of fluorescent molecules acts on peptide-to-lipid ratio, FRET is indeed observed
a length scale of nanometers and is therefore sen- because even their random distribution assures
sitive to the packing of peptides on the membrane that they find neighbors regularly within the
surface at high concentration. Self-quenching is Förster radius (≈56 Å for the NBD-rhodamine
explored by dilution of the label (replacing pair, Medintz and Hildebrandt 2013). However,
labeled peptide by unlabeled peptide, Fig. 4.1a), upon dilution of the peptides in the lipid bilayer,
and distances are estimated by fitting the data the FRET effect disappears indicating that the
with a Poisson distribution. Ongoing research peptides do not interact strongly and specifically
indicates an important role of mesophases and but distribute in a more stochastic manner along
clustering for the function of magainin 2 and the membrane surface (Marquette et al. 2015).
PGLa (manuscript in preparation).

4.3.6 Fluorescence Imaging

4.3.5  örster Resonance Energy
F
Transfer (FRET) Fluorescence imaging aims to get a deeper under-
standing in the mechanisms of AMP action by
FRET is another fluorescence technique widely making direct observations on microscopic living
used in the field of biophysics that can reveal the or nonliving systems. The main advantage of
proximity between two systems of interest. Most imaging techniques lies in the fact that temporal
often the proteins and/or peptides have to be and spatial information are made accessible at
labeled with soluble and non-perturbative dyes, the same time and sometimes on very dilute
so-called acceptor and donor for those who are ­systems as low as single fluorescent molecules.
prone to absorb and emit light, respectively. When probing bacteria, imaging techniques can
When they encounter in close vicinity, the excited provide subcellular spatial resolution and/or dis-
states of both fluorescent molecules mix resulting cern heterogeneities between cells within a sam-
in the emission of the acceptor when the donor is ple population. Over the last decades,
excited with light. Because the emission wave- time-resolved/high-resolution imaging assays
lengths of the acceptor are longer than the ones of have brought a new dimension to the description
the donor, this phenomenon is sometimes associ- and understanding mode of action of AMPs
ated with a considerable shift in the emission (reviewed in Choi et al. 2016).
spectrum. Detecting the quenching of the emis- To perform time-resolved imaging on single
sion of the donor and/or the enhancement of the cells with resolution at the millisecond time
fluorescence of the acceptor reveals the proxim- scale, a time-resolved microscope equipped with
ity of the dyes. These kinds of experiments can high-magnification objectives and phase-contrast
be performed within a conventional spectrofluo- detection system has been used. Different
rometer or under the lens of a microscope which schemes of chromophore labeling including visu-
adds spatial resolution to the spectroscopic infor- alizing the AMPs themselves have proven their
mation. Because of the high absorption coeffi- potential in revealing when and where key mech-
40 C. Aisenbrey et al.

anistic events occur within individual cells (Choi position of the bilayers under investigation
et al. 2016). It is believed that in the near future (Gregory et al. 2008; Cheng et al. 2011).
improvement of time resolution combined with
superresolution imaging systems could provide
an even more comprehensive picture of how 4.4 Circular Dichroism
AMPs halt growth and/or kill bacteria. Spectroscopy
The binding of antimicrobial peptides to live
bacteria was monitored using microscopic imag- Peptides exhibit multiple chiral centers that are
ing revealing a spatiotemporal sequence of events optically active in the mid-ultraviolet spectral
including membrane permeabilization. These range. This can be investigated by circular dichro-
images reveal that the human LL37 peptide ism where the absorption spectra exhibit pro-
(Table 4.1) attacks septating E. coli cells. The nounced features correlating to the dihedral
peptide distributes unevenly and preferentially angles of the polypeptide backbone and thereby
binds to the septum and the curved regions of the the secondary structure (Sreerama and Woody
outer membrane (Barns and Weisshaar 2013). In 2000; Miles and Wallace 2006). For example,
non-septating cells, it is found to associate with α-helical folds exhibit two characteristic minima
one of the endcaps. When the AMPs enter the at 208 and 222 nm, while the spectral intensities
periplasmic space, the cells shrink which is prob- of random coil structures show a single minimum
ably due to an osmotic effect. The outer mem- at 195 nm. Thus, the CD line shape of polypep-
brane loses the barrier function first, and after a tides can be deconvoluted into contributions from
short delay, permeabilization of the cytoplas- helices, sheets, turns, and random coil conforma-
matic membrane is observed. The openings of the tions (Sreerama and Woody 2000; Miles and
outer and cytoplasmic membranes monitored in Wallace 2006). When peptides are reconstituted
the presence of LL37 are localized and persistent into oriented membranes, different CD line
rather than global and transient (Rangarajan et al. shapes are obtained that provide information on
2013). Although many events observed in this the peptide alignment in membranes (Wu et al.
manner follow a related schedule, important 1990; Perrone et al. 2014). In this context, it
details vary with the antimicrobial compound should be mentioned that ATR FTIR spectra of
when cationic polymers and longer or shorter oriented membranes were also analyzed to study
peptides such as LL37, cecropin A, or melittin, the bilayer topologies of antimicrobial peptides
are compared to each other (Yang et al. 2018). including magainin 2 (Bechinger et al. 1999).
From mutagenesis experiments and a comparison A refined analysis of CD spectra has even
of E. coli cells grown under aerobic or anaerobic been reported to probe the monomer-dimer equi-
conditions, it has been concluded that LL37 spe- librium of a transmembrane helix (Loudet et al.
cifically affects the electron transport chain (Choi 2005); however, when reproducing such experi-
et al. 2017). On the other hand, human β-defensins ments, care should be taken during analysis
tend to concentrate in a few foci that localize at because related changes can also be due to light
the septum of cell division sites of Enterococcus scattering.
faecalis where they colocalize with PG, CL, Sec Because membrane polypeptides exhibit a
A, and sortases, interfering with the activities of high local concentration and reside in vesicles
the latter proteins (Kandaswamy et al. 2013). In that are close to the size of the absorption wave-
contrast, alamethicin causes a different series of lengths, distortions due to light scattering and
permeabilization events (Barns and Weisshaar absorption flattening artifacts have to be taken
2016). Thus, the detailed cellular response into consideration (Wallace and Moa 1984; Miles
depends on the peptide, the bacterial species, and and Wallace 2016). Recently a new approach has
the growth conditions. Notably, this reflects been presented which allows one to quantify and
experiments with membrane model systems correct for light scattering artifacts (Vermeer
where the very details varied with the lipid com- et al. 2016). Finally, uncertainties in the determi-
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 41

nation of peptide concentration due to the pres- ratio. Based on a simple membrane insertion
ence of counterions and salts or errors in weighing model (Vogt and Bechinger 1999), an apparent
the sample propagate into errors in the secondary association constant of Kass ≈ 1680 M−1 can be
structure determination by these algorithms. calculated, but clearly the experimental data do
Therefore, relative intensities have also been not correlate well with such an asymptotic bind-
used to estimate structural transitions between ing isotherm. Indeed, it should be noted that this
random coil and helical secondary structure apparent value covers the electrostatic interaction
(Bruch et al. 1991). to the membrane surface and the hydrophobic
Because pronounced spectral changes occur partitioning into the bilayer interface. Therefore,
when membrane association is accompanied by the membrane surface concentration of the posi-
structural transitions, it is possible to quantita- tively charged PGLa (nominal charge +4 to +5) is
tively determine membrane association constants much increased in the proximity of the anionic
by titrating vesicles step by step to a peptide solu- POPE/POPG bilayer. However, membrane asso-
tion, also of magainin AMPs (Wieprecht et al. ciation of the cationic peptide neutralizes the
2000a, b; Voievoda et al. 2015). Furthermore, charges of the PG lipids and can even result in
valuable information on the association kinetics repulsive interactions. A more refined analysis
of magainin to whole cells and lipopolysaccha- separating electrostatic and hydrophobic contri-
rides has been obtained using CD spectroscopy butions has been reported in the literature for the
(Avitabile et al. 2014). binding of PGLa to POPC/POPG 3/1 mole/mole
As an illustration of the technique, we show in SUVs in the presence of 100 mM NaCl (Wieprecht
Fig. 4.2 the CD spectra and the corresponding et al. 2000a, b). Indeed, whereas the apparent
association isotherm of the PGLa peptide to SUVs membrane association is 50-fold increased for the
made of POPE/POPG 3/1 mole/mole at neutral anionic lipid mixture when compared to pure
pH. A fitting procedure was applied to each spec- POPC, a hydrophobic surface partition equilib-
trum to estimate the percentage of peptides bound rium with Kp = 800–1500 M−1 was obtained for
to the membrane as a function of peptide-to-lipid both membranes.

Fig. 4.2 Titration of

Ellipticity signal (a.u.)

50 μM PGLa with 100 PGLa

increasing amounts of
SUVs (POPE/POPG 3/1)
in 5 mM Tris-HCl, 50
pH = 7, at 25 °C. The
changes in the circular
0
dichroism spectra upon
membrane association A
(a) have been used to
quantify the percentage 190 200 210 220 230 240 250
of membrane-associated Wavelength (nm)
peptide and are reported
on the vertical axis of the 100
lower frame panel (b) T = 25 °C
PGLa bound (%)

80
60
40
20
B
0
0 1 2 3 4 5 6x10
-3

Lipid concentration (mol/L)

42 C. Aisenbrey et al.

Therefore, an alternative analysis was per- reactant is placed in an isolated reactor, while the
formed for the data shown in Fig. 4.2 where focus ligand is added in a stepwise manner, allowing
is on electrostatic interactions. Under these con- the determination of enthalpy, binding constants,
ditions, it was assumed that the amount of pep- stoichiometry, and entropy changes, thereby pro-
tide that binds to the POPE/POPG 3/1 mole/mole viding a complete thermodynamic picture of the
membranes is governed by electrostatic interac- system. The technique is particularly well suited
tions and stops when charge neutrality is reached. to study water-soluble molecules, but the method
Indeed, Fig. 4.2b exhibits a linear increase in was also developed for the determination of the
bound peptide up to about 1.4 mM total lipid. binding parameters between membranes and
Under these conditions, the negative charge con- lipophilic biomolecules such as membrane pep-
tribution corresponds to 350 μM POPG; how- tides (Seelig 2004).
ever, almost half of the charges reside in the inner The association of magainin 2 and/or PGLa
leaflet of the vesicles and during the titration may with membranes has been studied by ITC under
not or only partially be accessible to the PGLa different conditions and provided valuable insight
peptide. At this lipid concentration, 50 μM pep- into their reversible interaction with phospho-
tides are used up in the binding reaction, contrib- lipid bilayers (Wenk and Seelig 1998; Wieprecht
uting 200–250 μM in positive charges. The P/L et al. 1999a, b, 2000a, b, 2002). The ITC data
ratio at this lipid concentration is about reveal apparent partitioning constants for
3.6 mole%. magainin 2 in the 12,000 M–1 range for POPC/
Such structural investigations show that the POPG 3/1 mole/mole. When separated from the
random coil structure of magainins in aqueous electrostatic attractive term, a hydrophobic parti-
solution becomes helical once the peptide inserts tioning of 55 M−1 remains (Wenk and Seelig
into membrane environments (Bechinger 1999). 1998). For pure POPC, values of 2000 M−1 were
This conformational transition has been identi- obtained at 30 °C (Wieprecht et al. 1999b).
fied to be a driving force of membrane associa- Notably, when 100 nm LUVs and SUVs were
tion (Wieprecht et al. 1999a). Oriented CD compared to each other, the enthalpic and entro-
spectroscopy confirmed solid-state NMR data pic contributions were much different, whereas
obtained at peptide-to-lipid ratios <3 mole% the Gibbs free energy and therefore the binding
showing that the helix is oriented parallel to the constants hardly changed (Wieprecht et al. 2000a,
membrane surface (cf. below) and suggested an b). By investigating double D-amino acid replace-
insertion when the concentration is increased ment sequences of magainin 2, the random coil to
(Ludtke et al. 1994). However, it should be noted helix transition at the membrane surface was
that at higher concentrations the membrane found to contribute about −0.6 kJ/mole per resi-
supramolecular architecture exhibits pronounced due and thereby about 50% of the driving force of
changes probably by a transition to bicellar frag- the magainin 2 membrane association (Wieprecht
ments (Bechinger 2005). et al. 1999a). The hydrophobic contribution of
PGLa membrane association is 800–1500 M−1
when at the same time the apparent partitioning
4.5 Isothermal Titration to POPC/POPG 3/1 mole/mole membranes is
Calorimetry 50-fold increased (Wieprecht et al. 2000a, b).
When PGLa or magainin was titrated into
Isothermal titration calorimetry (ITC) is a widely LUV suspensions made of POPE/POPG 3/1 at
used technique when thermodynamic parameters pH 7, the peptide membrane association was
describing the interactions between two systems characterized by endothermic reaction enthalpies
need to be investigated. The method is based (ΔH) (Marquette and Bechinger 2018). These are
upon the measurement of the heat absorbed or relatively small when compared to previous
released during a reaction such as binding investigations with 30 nm SUVs of different lipid
between a ligand and a reactant molecule. The compositions (Wenk and Seelig 1998; Wieprecht
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 43

et al. 1999a, b, 2000a, 2002). In the presence of 2015; Jaipuria et al. 2017; Visscher et al. 2017;
both peptides, an additional enthalpy contribu- Naito et al. 2018).
tion of −8 kJ/mole was observed which corre- A second approach relies on exploiting the
lates with the formation of larger complexes, anisotropies of interactions inherent to solid-state
probably vesicle agglutination (Marquette and NMR spectra rather than averaging them
Bechinger 2018). (Fig. 4.3). When membranes are uniaxially ori-
ented relative to the magnetic field of the NMR
spectrometer, a unique molecular alignment is
4.6 Solid-State NMR retained and spectral resolution is recovered. The
Spectroscopy resulting anisotropic chemical shifts, dipolar and
quadrupolar interactions provide valuable infor-
4.6.1 Solid-State NMR mation about the orientation of bonds, protein
Investigations of Polypeptides domains, and polypeptides as a whole. Thus, the
corresponding spectra can be used to analyze the
Nuclear magnetic resonance is a powerful tech- structure, dynamics, and topology of membrane-­
nique to investigate the structure, topology, associated polypeptides (Das et al. 2015;
dynamics, and interactions of biomolecules. Gopinath et al. 2015; Itkin et al. 2017; Salnikov
Whereas peptides have been investigated by mul- et al. 2018).
tidimensional 1H-1H solution NMR spectroscopy Whereas the 15N chemical shift alone provides
in membrane-mimetic micelles made from deu- an approximate tilt angle of helical domains
terated DPC (Brown 1979; Georgescu et al. (Bechinger and Sizun 2003), the combination
2010), this technique is unsuitable for the investi- with 2H solid-state NMR spectra from methyl-­
gation of peptides associated with large peptide-­ deuterated alanines results in accurate tilt and
bilayer complexes. However, NMR spectroscopy pitch angle information (Fig. 4.3) (Bechinger
becomes much more powerful when polypep- et al. 2011; Salnikov et al. 2018).
tides can be prepared either by bacterial overex- Only few years after the discovery of magain-
pression and biochemical purification or by ins solid-state NMR experiments on these pep-
chemical synthesis such that stable isotopic tides reconstituted into uniaxially oriented lipid
labels such as 15N, 13C, and/or 2H were bilayers indicated for the very first time that they
introduced. adopt stable alignments parallel to the membrane
For polypeptides associated with large com- surface (Bechinger et al. 1990, 1991a, b, 1992,
plexes that rotate slowly on the NMR time scales, 1993). This topology has been confirmed for
solid-state NMR spectroscopy has been devel- magainin 2 in all lipid compositions investigated
oped. Because chemical shifts, dipolar and quad- so far (Bechinger 2011), for magainin analogues
rupolar interactions are all dependent on the (Ramamoorthy et al. 2006; Mason et al. 2009),
molecular alignment relative to the magnetic and for several other linear cationic antimicrobial
field and the dipolar interactions between nuclei peptides (Resende et al. 2009, 2014; Hayden et al.
can be strong for immobilized molecules, in a 2015; Bechinger and Gorr 2017; Sani and
static sample, broad overlapping line shapes are Separovic 2018). Furthermore, oriented CD spec-
observed that hamper a detailed analysis. tra agree with such an alignment of the magainin
One approach to obtain well-resolved solid-­ helix (Ludtke et al. 1994). Fluorescence quench-
state NMR spectra is to subject the sample to fast ing experiments not only confirm the alignment
rotation around the magic angle. This method parallel to the surface but also reveal an interfacial
results in NMR spectra where only the isotropic localization of the magainin 2 helix (Matsuzaki
chemical shifts remain which resemble those et al. 1994). A parallel alignment has also been
observed in solution, and similar concepts for detected for cecropin P1 (Table 4.1) using ATR
assignment and structural analysis are used (Das FTIR, a topology which was associated with the
et al. 2015; Eddy et al. 2015; Gopinath and Veglia term “carpet model” (Gazit et al. 1996).
44 C. Aisenbrey et al.

Fig. 4.3 The figure shows the 2H (a) and 15N solid-state ture (c). This alignment dependence can be used to obtain
NMR spectra (b) of an amphipathic helical model peptide orientational constraints for the peptide. (d) The tilt/pitch
(Aisenbrey and Bechinger 2004) labeled with 2H3-alanine angular pairs of the helix that agree with the 2H quadrupo-
and a 15N label at a single peptide bond (c). The 2H quad- lar splitting are shown in red and those that agree with the
rupolar splitting and 15N chemical shift obtained from 15
N chemical shift in blue. Both NMR parameters have to
these solid-state NMR spectra are a function of the align- agree with the real peptide alignment which leaves only
ment of the alanine Cα-Cβ bond and of a vector close to five possible topologies where the two restraints intersect.
parallel to the amide 15N-1H bond, respectively, relative to Data from well-chosen additional positions results in a
the magnetic field of the NMR spectrometer (Bo). The unique solution for the peptide topology (Bechinger et al.
bonds are highlighted in red and blue in the helical struc- 2011)

This topology assures that the peptide asso- composition (Matsuzaki et al. 1994; Bechinger
ciation is reversible (Bechinger 2011) and con- 2011), its relative PGLa (Table 4.1) adopts a
trasts findings for alamethicin which is much much wider range of alignments but only when
more hydrophobic and forms stable transmem- investigated in bilayers composed of fully satu-
brane helical bundles when investigated in rated fatty acyl chains (Tremouilhac et al.
DMPC or POPC membranes (North et al. 1995; 2006a, b; Salnikov and Bechinger 2011). In
Bak et al. 2001; Milov et al. 2009; Salnikov DMPC the PGLa tilt angle depends on peptide-­
et al. 2009b, 2016b). Notably, alamethicin also to-­
lipid ratio and membrane hydration
exhibits pronounced differences in electrophysi- (Tremouilhac et al. 2006a, b; Salnikov and
ological recordings where the single-channel Bechinger 2011). A continuous range of tilt
events are well defined and reproducible angles was observed as a function of hydropho-
(Sansom 1993; Bechinger 1997). Notably, even bic thickness in fully saturated PC bilayers
for alamethicin, conditions can be found where (Tremouilhac et al. 2006a, b; Salnikov and
it adopts in-plane alignments (He et al. 1996; Bechinger 2011). However, when studied in
Salnikov et al. 2010) which underlines the phospholipid bilayers carrying unsaturations
dynamic nature of antimicrobial peptide-lipid (such as palmitoyl-oleoyl-phospholipids), also
interactions involving multiple equilibria this peptide remains stably aligned parallel to
(Bechinger 2015). Whereas the alignment of the bilayer surface (Bechinger et al. 1991a,
magainin 2 has been found parallel to the mem- 1998; Bechinger 2011; Salnikov and Bechinger
brane surface regardless of membrane lipid 2011; Strandberg et al. 2012).
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 45

4.6.2  olid-State NMR Spectroscopy

S characterize the order parameters (Fig. 4.4),
of Lipids hydrophobic thickness, and packing of lipid
bilayers (Harmouche and Bechinger 2018).
Notably, solid-state NMR spectroscopy has been Finally, conformational changes of the phospho-
used not only to investigate the structure, dynam- lipid head groups have been monitored by 2H and
ics, and topology of membrane-associated poly- 31
P solid-state NMR spectroscopy and thereby
peptides (see ultra) but also provides valuable allowed to quantitatively follow electrostatic
information on the lipids (Bechinger and Salnikov interactions at the membrane interface (Scherer
2012). Because the lipid packing and phase prop- and Seelig 1989).
erties are modulated by interactions with pep- An amphipathic helix that inserts into the
tides, this information has been particularly bilayer interface with an alignment parallel to the
valuable when the mechanism of antimicrobial membrane surface needs to expand the surface at
peptides has been investigated. 31P solid-state the level of the lipid head groups and in the glyc-
NMR spectra provide information on the macro- erol region (Matsuzaki et al. 1994). This is paral-
scopic phase properties of phospholipid mem- leled by a loosening of the packing of the
branes or the orientational order of lipid hydrophobic region and an increased disorder of
bilayers. They are particularly straightforward to the fatty acyl chains (Salnikov et al. 2009a, b;
obatin due to 100% natural abundance of this Bortolus et al. 2014). This effect can be even
nucleus (Bechinger and Salnikov 2012). more pronounced when the peptides initially
Furthermore, 2H solid-state NMR spectroscopy associate with only the outer monolayer, which
of deuterated fatty acyl chains has been used to subsequently allows flip-flop of lipids and pep-

Fig. 4.4 (a) Phosphatidylcholine (POPC) where all 1H of from individual CD2 functional groups that sum up into a
the palmitoyl chain have been exchanged with 2H. Upon spectrum of superimposed resonances. (c) The mobility of
hydration, the lipid self-assembles into liposomes that tum- the CD2 segments increases from the bilayer interface to the
ble slowly in the NMR magnetic field. (b) The 2H solid- hydrophobic membrane interior and results in a decrease in
state NMR spectra are composed of quadrupolar splittings quadrupolar splitting and the order parameters SCD
46 C. Aisenbrey et al.

tides across the membrane, thereby relieving the (Kmiecik et al. 2016; Harpole and Delemotte
asymmetry-related tension of the outer mono- 2018).
layer (Matsuzaki et al. 1996; Karal et al. 2015; Molecular dynamics simulations can therefore
Hasan et al. 2018). The increased disorder at the provide atomistic views of how the molecules
level of the lipid fatty acyl chains results in the change conformation over time. With increasing
reduction of the membrane thickness (Ludtke computer power, lower resolution (coarse grain),
et al. 1995; Kim et al. 2009). In other words, and better algorithms for exploiting today’s com-
while the total volume of the lipid remains con- puter architectures, they also reveal how mole-
stant, in average it expands over an increased area cules interact and how supramolecular complexes
of the bilayer which results in a reduction in form and evolve. MD simulations have been
hydrophobic thickness. used, for example, to follow the insertion of
Pronounced decreases in order parameters hydrophobic peptides into the membrane, where
especially in the bilayer interior have indeed been they assemble into transmembrane helical bun-
observed upon addition of magainin 2, PGLa, dles (Tieleman et al. 2002) and how magainin
and other amphipathic peptides using 2H solid-­ interacts with lipopolysaccharides (Smart et al.
state NMR of deuterated lipids (Hallock et al. 2017). Furthermore, they permitted the visualiza-
2002; Salnikov et al. 2009a, b; Grage et al. 2016; tion of the deformation of the lipid bilayer in the
Harmouche and Bechinger 2018). Such bilayer presence of in-plane oriented amphipathic pep-
disruptive properties have been estimated to tides. From the conformational details given by
extend over 10 nm in diameter (Chen et al. 2003; the simulations, it was shown that the positioning
Mecke et al. 2005). of side chains is important for reaching across a
bilayer leaflet and to contribute to the formation
of water-filled openings (Vacha and Frenkel
4.7 Molecular Dynamics 2014; Farrotti et al. 2015; Pino-Angeles et al.
Simulations 2016).
A 5–9 μs all-atom MD calculation shows that
Molecular dynamics (MD) simulations is a the starting structure of tetrameric transmem-
computer-­based numerical method for calculat- brane helical bundles of magainin or PGLa is
ing the motions of atoms and molecules. Based unstable when simulated in 80–120 lipids of
on an empirical potential energy function, the DMPC or DMPC/DMPG 3/1(P/L ratio of
atoms and molecules are allowed to interact for 3.3–5 mole%) (Pino-Angeles et al. 2016). The
short time intervals and change positions as a peptides exhibit tilted configurations, thereby
result of the instantaneous forces. The trajecto- better representing the topologies and aggrega-
ries are calculated by numerically solving tion states found during various biophysical
Newton’s equations of motion for a given system experiments (Matsuzaki et al. 1994; Bechinger
of interacting particles. The empirical potential 2011). In a related manner, all-atom 100 ns simu-
energy function, also known as a molecular lations of 1, 2, or 8 peptides in 512 lipids (POPE/
mechanics force field, uses physics-based models POPG 3/1; P/L, 0.5–1.6 mole%) confirm stable
to represent the forces that act between the parti- in-plane topologies of magainin and pleurocidin
cles, including bonding (bonds, angle, dihedrals) (Amos et al. 2016). Although some oligomeriza-
and nonbonded (van der Waals and electrostatic) tion occurs, the formation of pores or supramo-
terms. Using this approach, the conformational lecular rearrangement could not be observed
space and the time evolution of the system can be within this relatively short time frame.
visualized at atomic resolution if a force field Finally, the possibility of a double-belt
representing all atoms is used. An alternative is to arrangement of peptides oriented parallel to the
use a coarse-grained representation, where atoms bilayer plane has been suggested from coarse-­
are grouped into larger entities. This approach grain MD simulations of schematic amphipathic
permits long simulations of very large systems helices (Vacha and Frenkel 2014). Such a model
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 47

resembles an inversion of the double-belt model (Ines and Dhouha 2015; Otzen 2017; Wu et al.
that is discussed for apolipoproteins AI particles 2017; Zhao et al. 2017). These surface-active
or related nanoparticles (Gogonea 2015). compounds are produced by a wide variety of
From the molecular dynamics point of view, bacteria, fungi, and yeast and exhibit a high
the magainin membrane interactions are charac- structural diversity (Ines and Dhouha 2015).
terized by peptides that adopt many different Furthermore, following their natural templates,
conformations and membrane alignments. ultrashort lipopeptides with antimicrobial activi-
Transmembrane alignments have been found ties have been engineered (Mangoni and Shai
unstable, and little direct interactions between the 2011). In particular, the lipopeptides produced by
peptides have been observed. Openings can form Bacillus are small cyclic structures of 7–10 amino
by stochastic rearrangements of peptides and lip- acids and a β-hydroxy fatty acid with 13–19 car-
ids rather than through well-defined supramolec- bon atoms (Zhao et al. 2017). According to the
ular assemblies, which could explain the early peptidic ring structure, the Bacillus lipopeptides
observations during electrophysiological record- are divided into the surfactin, fengycin, and iturin
ings (Christensen et al. 1988; Duclohier et al. family. They are widely used in agriculture, food,
1989; Cruciani et al. 1991; Watanabe and Kawano medicine, and feed production due to their anti-
2016). In order to visualize local or global fungal, antibacterial, antitumor, antiviral, and
changes in the membrane macroscopic phase or anti-inflammatory activities (Wu et al. 2017;
membrane lysis, including the all-or-nothing Zhao et al. 2017). Furthermore, they interact with
release observed in dye release experiments biofilms, have been suggested to be useful in
(Gregory et al. 2008; Tamba et al. 2010), larger thrombolytic and Alzheimer therapies or to be
systems would need to be simulated over longer used to create nanostructures for drug delivery
time scales. (Wu et al. 2017; Zhao et al. 2017). Many of these
Furthermore, a symmetric antiparallel dimer activities are thought to be due to the interactions
of PGLa has been preassembled and simulated of these biosurfactants with biological mem-
for up to 2 μs using all-atom MD (Ulmschneider branes (Carrillo et al. 2003; Heerklotz and Seelig
et al. 2012). Indeed, dimer formation of 2007). In the context of the molecular shape
membrane-­ associated antimicrobial peptides is model, we will focus to present some biophysical
an interesting hypothesis but so far lacks experi- work obtained on surfactins and related
mental proof, for example, by solid-state NMR lipopeptides.
distance measurements. Although the GxxxG Surfactin is made of seven amino acids (with
sequence within the PGLa sequence is suggestive L- or D-conformation) linked to a fatty acyl chain
for peptide-peptide interactions, this motif of C12–C14 that closes the peptide ring by a
requires a hydrophobic environment to drive β-lactone (Wu et al. 2017). Many lipopeptides
dimerization (Russ and Engelman 2000) rather have been shown to exhibit antimicrobial activi-
than its experimental orientation along the mem- ties against a range of bacteria (Ines and Dhouha
brane interface or within a water-filled channel. 2015). Importantly, they are some of the most
potent and most popular antifungal agents and
have been investigated for their anticancer activi-
4.8 Lipopeptide Biosurfactants ties (Ines and Dhouha 2015; Wu et al. 2017).
with Antimicrobial The hydrophobic alkyl chain and the more
Properties polar peptidic portion confer an amphiphilic
character and thus, in cases where the fatty acyl
Before establishing a more elaborate model for chain adopts an extended conformation, a pro-
the action of antimicrobial peptides, it is interest- nounced cone shape to the molecule (Otzen
ing to mention a range of lipopeptides whose 2017). Therefore, these compounds form micelles
amphiphilic character is assured by a long fatty in aqueous buffer and exhibit surfactant activities
acyl chain attached to a polar peptidic structures such as monolayer formation at the air-water
48 C. Aisenbrey et al.

interface (Maget-Dana and Peypoux 1994; Otzen Finally, membrane solubilization and micelle
2017). It should be noted that in contrast to chem- formation are observed between Rb = 0.22 and
ical detergents, biosurfactants exhibit a more 0.42. The same experiments provide a membrane
mosaic-like amphipathic structure (Otzen 2017). partitioning coefficient of 20,000 M−1 (Heerklotz
Calcein release from POPC vesicles by sur- and Seelig 2007). Furthermore, fengycins, lipo-
factin is a cooperative process (index 1.82) peptides sold together with iturins and surfactin
(Carrillo et al. 2003). The membrane-perturbing for agricultural applications, were investigated in
effect of surfactin, measured by calcein release, fluorescence lifetime efflux measurements (Patel
is attenuated in POPE or cholesterol-containing et al. 2011).
membranes but accentuated when 25% of DPPC The length of the fatty acyl chain, hydropho-
has been mixed into the POPC bilayers (at 25 °C) bicity, and membrane association of surfactin are
(Carrillo et al. 2003). Notably as with detergents, directly correlated to its anticancer activities (Wu
the surfactin behavior in aqueous solution has et al. 2017). Using measurement on lipid mono-
been characterized in terms of CMC (7.5 μM), layers and CD, spectroscopic investigations
onset of membrane solubilization (comparatively reveal changes in the membrane insertion process
low when compared to other detergents, and the proteic structure in the presence of Ca2+
Heerklotz and Seelig 2001), and aggregation (Maget-Dana and Ptak 1995). Furthermore, these
number (20) (Otzen 2017). The imbalance in the investigations demonstrate that electrostatic con-
lateral pressure profile at the interface and the tributions and the space occupied by the head
hydrophobic portion of the lipid bilayer in the group play an important role in surfactin mem-
presence of surfactin and chemical detergents has brane insertion (Maget-Dana and Ptak 1995)
been investigated by 2H solid-state NMR of spe- although the hydrophobic interactions of the fatty
cifically deuterated POPC membranes (Heerklotz acyl chains as well as other factors remain of
et al. 2004). Whereas C12EO6 and C12EO8 cause major importance (Wu et al. 2017). The gel-to-­
the expected disordering of the membrane fatty liquid phase transition of DMPG has been shown
acyl chains, in the presence of surfactin, the fatty to be significantly broadened by surfactin an
acyl chains tilt, and the head group reorients to effect that is enhanced in the presence of Ca2+
accommodate the bulky heptapeptide ring. This (Grau et al. 1999). Because of this observation, a
difference is probably related to the amphipathic deeper penetration into the membrane upon com-
but predominantly hydrophobic character of the plexation of Ca2+ with the Glu-1 and Asp-5 resi-
peptide moiety which results in a deeper inser- dues of the peptide sequence has been suggested
tion of the peptide ring when compared to the (Grau et al. 1999). Surfactin has been incorpo-
polar groups of the detergents (Heerklotz et al. rated in a wide variety of nanoformulations for
2004). To resolve ambiguities about the mecha- materials and biomedical applications (Wu et al.
nism how surfactin causes membrane leakage, a 2017).
series of experiments was performed at non-lytic Surfactin, iturins, and the closely related bac-
concentrations correlating data from ITC, 31P illomycins and mycosubtilin are of similar built.
solid-state NMR, and leakage assays (Heerklotz Iturins are made of circular peptides made of
and Seelig 2007). This systematic analysis seven L- and D-amino acids connected to a
reveals three different mechanisms depending on β-amino fatty acid of 14–17 C-atoms. They
the surfactin-to-lipid ratio Rb. Leakage starts at exhibit strong antifungal but little antibacterial
Rb = 0.05 probably by a mechanism where the activities. Interestingly, iturin A has been shown
surfactin accumulates at the outer membrane to specifically interact with cholesterol and sug-
leaflet and opens pores transiently to equilibrate gested to form specific phospholipid/peptide/
with the inner side of the bilayer (bilayer couple cholesterol complexes that are responsible for the
mechanism). At Rb = 0.15, it is suggested that measured electrophysiological properties. The
surfactin-rich clusters form which cause leaks latter are strongly dependent on the lipid compo-
and stabilize the hydrophobic etches of those. sition, the physical state of the lipids, and the
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 49

detailed peptide structure. An interesting obser- phases which are a function of peptide concentra-
vation is that the CMC and the MIC follow the tion in the membrane and other environmental
same trend suggesting that large iturin A aggre- parameters (Bechinger 2009). The concept has
gates are the active component in biological been developed early on to rationalize the macro-
membranes (Maget-Dana and Peypoux 1994). scopic phase transitions of lipids (Israelachvili
When traces of iturin A are added to black lipid et al. 1980). Because the phosphatidylcho-
membranes, stepwise conductances, whose char- line head groups and two fatty acyl chains expand
acteristics evolve over time, are observed. At P/L laterally over an area that is about equivalent at
ratios above 10−7, the membranes break. Notably the level of the membrane interface and the
iturin A is coproduced with surfactin, and both hydrophobic interior, the PC molecules when
peptides together exhibit synergistic activities in part of a membrane can be described by a cylin-
biological assays (citations in Maget-Dana and drical shape (Fig. 4.5a). In contrast, the head
Peypoux 1994). group of phosphatidylethanolamine is much
Owing to their membrane activities, several smaller which results in a truncated inverted
surfactants including iturin A have been shown to cone-shaped molecule (Fig. 4.5b), whereas lyso-
exhibit hemolytic and anti-clot-forming activities lipids with only one fatty acyl chain (or deter-
(Ines and Dhouha 2015). An overview over many gents) are best represented by a cone (Fig. 4.5c).
more membrane-active surfactants’ interesting For a more quantitative treatment, the critical
biomedical and technical activities is presented packing parameter comparing the optimal sur-
in Ines and Dhouha (2015). face area at the carbon-water interface (ao), the
optimal chain length (lc), and the hydrocarbon
volume (v) is related by the packing parameter
4.9  he Molecular Shape
T v/lcao (Israelachvili et al. 1980). When these geo-
Concept Explains the Many metrical shapes are assembled into supramolecu-
Different Supramolecular lar aggregates, the PC cylinders line up side by
Arrangements of AMPs side in a phospholipid bilayer (Fig. 4.5f), the
and Lipids cones form micellar assemblies (Fig. 4.5i), and
the PE lipids at higher temperatures tend to form
A number of seemingly contradictory models hexagonal II phases (Fig. 4.5h). When PE or
have been suggested to explain the mechanism of detergents are forced to be part of a planar lipid
action and the interaction of antimicrobial pep- bilayer, they are under curvature elastic stress
tides with membranes. The most cited are the for- where interactions with the opposite monolayer
mation of toroidal pores (Ludtke et al. 1996; maintain the bilayer arrangement. However, a
Matsuzaki 1998), a dense “carpet” of peptides spontaneous curvature has been defined for each
covering the membrane surface that causes lysis lipid which takes into account the differences in
(Shai 1999), or small aggregates without specific lateral cross section at the membrane interface
structure within the membrane (Jenssen et al. when compared to the hydrophobic interior
2006). Whereas on the one hand, occasionally (Kollmitzer et al. 2013). Thus, the intrinsic cur-
channel-like events are recorded in electrophysi- vature of POPC is around 0, of POPE it is −0.32,
ological experiment (Christensen et al. 1988; and for lyso-PE, a value of +0.18 has been deter-
Duclohier et al. 1989; Cruciani et al. 1991), at mined (Kollmitzer et al. 2013; Leber et al. 2018).
high peptide concentrations, worm-like struc- It should be noted that the shape is not only deter-
tures, disk-shaped particles, or micelles have mined by the van der Waals contacts but can be
been observed (Hallock et al. 2002; Bechinger modulated by other interactions. For example,
and Lohner 2006; Wolf et al. 2017). This wide repulsive electrostatic interactions at the head
variety of observations can be taken into consid- group level increase its optimal surface area,
eration by the differential shape of lipids resulting in a more cone-shaped molecule and a
(Fig. 4.5) and the resulting supramolecular positive curvature (Israelachvili et al. 1980).
50 C. Aisenbrey et al.

Fig. 4.5 The molecular shape concept and how mole- amphipathic peptide, the lipids adjust their shape by add-
cules assemble into supramolecular arrangements. When ing gauche conformations within the alkyl chain, thereby
the lateral cross sections of the lipid head groups and at increasing the disorder of their fatty acyl chains. The com-
the level of the fatty acyl chains are compared to each parison with unperturbed cylinders illustrates how these
other, phosphatidylcholines and phosphatidylglycerol conformational changes are accompanied by membrane
adopt cylindrical shapes (a), phosphatidylethanolamine thinning. (f) Cylindrical lipids self-assemble into stable
with its reduced head group size corresponds to a trun- bilayers where by being soft they can adjust to peptides or
cated inverted cone (b), and lysolipids and detergents other external stimuli (Bechinger 2015). (g) When the
resemble a cone (c). The corresponding intrinsic curva- concentration of cone-shaped molecules increases locally,
tures Jo of these lipids are zero, negative, and positive, phase transitions and membrane openings occur. (h)
respectively (Kollmitzer et al. 2013; Leber et al. 2018). Inverted cone-shaped molecules self-assemble into hex-
(d) A highly charged amphipathic peptide partitions into agonal II phases. (i) Cone shapes self-assemble into
the head group region without filling the hydrophobic micelles. In panels D–G, the helices are schematized as
region. Thereby it resembles a truncated cone and exerts yellow/orange cylinders (side view) and circles (front
pronounced positive curvature strain. (e) Next to such an view)

An amphipathic helix that resides at the mem- Thus, in contrast to the transmembrane
brane interface with the long axis approximately alamethicin helix, which is modeled by a cylin-
parallel to the membrane surface (Bechinger der, the shape of these amphipathic cationic
2009) or surfactin, a cyclic peptide with a long peptides has the properties of a large truncated
fatty acyl chain (Zhao et al. 2018), occupies a cone (Fig. 4.5d). Notably, the presence of PE,
much larger area at the level of the interface with- which has intrinsically an inverted cone-shaped
out filling the corresponding space at the level of and negative curvature (Fig. 4.5b), compen-
the fatty acyl chain (Fig. 4.5d). This induces con- sates for the wedge-like properties of the in-
siderable positive curvature strain and requires plane oriented helices, thereby stabilizing the
major rearrangement of the lipids in compensa- lipid bilayer arrangement (Batenburg et al.
tion. This shape helps to create the line tension 1988; Hallock et al. 2002). Recently the differ-
required for membrane pores to form (Hall et al. ence in “molecular shape” of magainin and
2014; Henderson et al. 2016). melittin has been included in a model that
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 51

explains the distinct activities of these peptides composition (Bechinger and Lohner 2006;
(Paterson et al. 2017). Bechinger 2009), which are in a delicate balance
making predictions on how the macroscopic
ensemble behaves rather difficult.
4.10 The “Molecular Shape”
of Membrane Constituents
Adapts 4.11  oft Membranes Adapt
S
and Respond, Also
When interacting with the membranes, many Transiently
amphipathic peptides, including magainins, fold
into helical conformations and intercalate into Not only the peptides are highly dynamic with
the lipid head group region. In this manner, they considerable conformational and topological
act as a spacer pushing apart the lipids. This freedom (Cheng et al. 2009, 2011), but also the
results in positive membrane curvature strain and liquid crystalline lipid bilayer has the capacity to
has been rationalized by a cone molecular shape compensate for external influences before their
of the in-plane oriented amphipathic helix barrier function breaks down. Together they form
(Fig. 4.5d) (Bechinger 2009). However, the pep- soft supramolecular assemblies which can change
tide does not fill the hydrophobic region com- thickness, morphology, and macroscopic phase
pletely; thus the lipids respond by increased properties globally or locally. Therefore, in the
trans-gauche isomerization or chain interdigita- presence of antimicrobial peptides, Soft
tion (Bechinger and Lohner 2006). The details of Membranes Adapt and Respond, also Transiently,
the conformational changes of the lipids are a a concept that makes up the SMART model for
function of the detailed peptide alignment and antimicrobial or other membrane-active peptides
penetration depth. The latter depend on the three-­ (Bechinger 2015). The model thereby extends on
dimensional distribution of charges, hydrophobic the molecular shape concept by taking into
side chains, i.e., the resulting hydrophobic account that the lipids have the capacity to
moment and amino acid distribution. Because respond to the membrane-disruptive properties of
both the peptide and the lipids exhibit consider- the peptides which themselves adjust their prop-
able conformational flexibility, the molecular erties when associating with lipid membranes.
shape of these membrane constituents is not fully However, once a critical concentration is reached,
determined, but they adjust to external forces phase transitions of the membrane are observed
within the supramolecular assembly. Thus, in the locally or globally. Due to lateral diffusion, weak
neighborhood of the cone of an amphipathic pep- peptide-peptide interactions, transient lipid phase
tide, a PC which is cylindrical in a pure lipid separation, etc., the local peptide concentrations
bilayer adopts a more inverted cone-shaped can vary which can explain how stochastic fluc-
structure, thereby stabilizing the bilayer tuations of peptide density result in the transient
(Fig. 4.5e). Furthermore, the peptides are not stiff pore events observed in, for example, electro-
helices and they can adjust their penetration physiological recordings (Christensen et al.
depths. Their alignment relative to the membrane 1988; Duclohier et al. 1989; Cruciani et al. 1991;
surface can be modulated by a wide range of Watanabe and Kawano 2016). Transient open-
interaction contributions (Bechinger 1996; ings also occur during membrane crossing when
Harmouche and Bechinger 2018). All these the peptide density equilibrates between the outer
adjustments correspond to modulations of their and the inner leaflet of the bilayer (Matsuzaki
“molecular shape” (Fig. 4.5e). The very details of et al. 1995a, b; Tamba et al. 2010; Wheaten et al.
the peptide-lipid supramolecular arrangement 2013). The different supramolecular morpholo-
thus depend on the peptide sequence, its confor- gies of the SMART model can be nicely repre-
mation and resulting hydrophobic moment, the sented by phase diagrams where regions
peptide-to-lipid ratio, as well as the detailed lipid corresponding to bilayer, wormholes, tubular
52 C. Aisenbrey et al.

structures, bicelle, micelle, or hexagonal phases model” (Shai 1999), whereas stochastic and tran-
are represented as a function of the peptide-to-­ sient openings occur when peptides orient along
lipid ratio, the detailed membrane composition, the membrane surface at low peptide-to-lipid
temperature, hydration, salt, pH, and other envi- ratios (Christensen et al. 1988; Duclohier et al.
ronmental factors (Bechinger and Lohner 2006; 1989; Cruciani et al. 1991).
Bechinger 2011). Notably, when bacteria are It should be noted that the peptide concentra-
exposed to AMPs, the peptides diffuse to the cell, tion at the membrane surface may be different by
across the cell wall and other cellular barriers, orders of magnitude from those in bulk solution
before they interact with the cellular membranes. because the cationic peptides are attracted to the
Therefore, the peptide local concentrations vary surface of negatively charged membranes (Wenk
over time, and intermediate states have been and Seelig 1998; Wieprecht et al. 1999a, b).
observed for magainin 2 where membranes tem- Thereby, the apparent partitioning coefficients
porarily lyse and recover (Hall et al. 2014). are also much increased, which provides one
At low peptide concentrations, the bilayer explanation why these AMPs kill bacteria which
structure is maintained; however, transient open- expose a highly negative surface charge. In con-
ings may form stochastically due to the lateral trast, the eukaryotic host cells are neutral in
diffusion of peptides and lipids concomitant with charge at their outside monolayer (Matsuzaki
density alterations. Notably, whereas in most et al. 1991; Wenk and Seelig 1998; Bechinger
models the membranes mechanically break or 2004; Klocek and Seelig 2008; Lohner 2009).
water-filled pores form, it seems also possible Electrostatic interactions also play a role when
that the physicochemical properties change in AMPs cause the lateral phase separation of lipids
such a manner to allow diffusion of ions through (Mason et al. 2006; Voievoda 2014), when the
a less densely packed lipid phase or along phase peptides arrange in mesophase structures along
boundaries which form when the peptides are the membrane surface (Aisenbrey and Bechinger
responsible for the lateral phase separation of 2014), or when peripheral membrane proteins are
membrane constituents (Cruzeiro-Hansson and repelled from the bilayer surface due to a more
Mouritsen 1988; Jean-Francois et al. 2008; positive surface charge in the presence of cationic
Gallaher et al. 2010; Aisenbrey and Bechinger AMPs (Wenzel et al. 2014).
2014). Furthermore, it has been pointed out that Notably the molecular shape concept and the
membrane morphological changes, including the SMART model have allowed the design of novel
thinning of the lipid bilayer, which has been peptide and non-peptidic mimetics of AMPs
shown to occur in the presence of magainin 2 with high antimicrobial efficiency that are cat-
(Ludtke et al. 1995), are associated with changes ionic or amphipathic and partition into the inter-
in membrane capacity and concomitantly electri- face without filling the hydrophobic volume to
cal currents (Heimburg 2012; Laub et al. 2012). the same extent such as short peptides (Oyston
At higher peptide concentrations, the system et al. 2009; Schweizer 2009; Hadley and
enters non-bilayer phases (at least locally) which Hancock 2010; Kindrachuk and Napper 2010;
come along with more stable openings (Gregory Liu et al. 2010; Mangoni and Shai 2011; Chou
et al. 2008) and macroscopic phase transitions of et al. 2016; Ahn et al. 2017), foldamers (Porter
the membranes (Fig. 4.5g–i) (Bechinger 2009). et al. 2002; Patch and Barron 2003; Kuroda and
The threshold concentrations that have been DeGrado 2005; Violette et al. 2006; Makovitzki
identified in some biophysical studies thus repre- et al. 2008; Scott et al. 2008; Rotem and Mor
sent boundaries in phase diagrams (Bechinger 2009; Palermo and Kuroda 2010), polymers
and Lohner 2006; Bechinger 2011). The loss of (Rank et al. 2017), and small organic molecules
bilayer integrity has been proposed by the “carpet (Ghosh et al. 2014).
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 53

Fig. 4.6 The figure shows how amphipathic helix dimers tions. (d) The two helices of heterodimers can exhibit very
potentially interact with membranes. (a) The disruptive different properties and affect the membrane similar to a
properties of two side-by-side helices should be more pro- monomer. Helices are schematized as cylinders and cir-
nounced and/or (b) result in the approach of the opposite cles, where the combination of different colors is indica-
monolayer to interact with the hydrophobic face of the tive of a heterodimer
peptides. (c) Membrane openings form by bilayer disrup-

4.12 Dimers of Antimicrobial shown to extend over a radius of 5 nm (Chen et al.

Peptides 2003; Mecke et al. 2005) which involves many
lipids of an estimated diameter of about 0.8 nm.
In the context of the SMART model, an interest- Within a mesophase supramolecular arrangement,
ing question arises how the oligomerization of the close proximity of helices thus results in a
peptides along the membrane surface would affect concerted destabilization of the membrane
their activity (Bechinger 1999). Figure 4.6a, b (Fig. 4.1c).
shows possible mechanisms how the membrane Only a few dimers have been identified in
can adjust to two tightly packed side-by-side heli- nature and have been investigated by biophysical
ces. On the one hand, one could imagine a situa- methods. Distinctin is composed of two different
tion where the lipids of the peptide-­ bearing polypeptide chains that are linked near their
monolayer compensate for the packing deficiency carboxy-­terminus by a cystine bond. In solution,
underneath in-plane oriented helices. Because the peptide forms a compact four-helix dimer of
they are excluded from the peptide-­peptide inter- heterodimers (Raimondo et al. 2005). When
face, they have to move in exclusively from the interacting with membranes, the solution
opposite helical side affecting the lipids that are ­structure unfolds. Whereas the 25-residue chain 2
neighboring the peptide even more significantly partitions into the membrane in a manner similar
(Fig. 4.6a). On the other hand, the opposite leaflet to magainin with a stable alignment parallel to
of the bilayer could be used to cover the hydro- the membrane surface, chain 1 encompassing 22
phobic face of the peptide dimer (Fig. 4.6b). In residues associates more loosely with the lipid
yet another situation, a mesophase structure of bilayer (Fig. 4.6d) (Resende et al. 2009; Verardi
many helices may form along the bilayer surface et al. 2011). The slightly increased antimicrobial
(Aisenbrey and Bechinger 2014). In this case the activity of the dimer when compared to the
peptides are separated by one or few lipids only monomers (Dalla Serra et al. 2008) is thought to
(Fig. 4.1c). Notably, previously the membrane- be due to the better resistance to proteolysis
disruptive properties of magainin 2 have been (Raimondo et al. 2005).
54 C. Aisenbrey et al.

More recently a natural homodimer of twice peptides were tested (Nishida et al. 2007).
24 residues has been described and its structure Whereas the antimicrobial activities of the het-
investigated in solution by NMR spectroscopy. erodimer were about the same as those of an
Interestingly doubling of some NMR cross peaks equimolar mixture of monomers, the calcein
are indicative of slight asymmetries of the fold release activity was somewhat stronger for the
(Verly et al. 2017). The global structure shows a dimers. However, both the hybrid and the mix-
tightly packed coiled coil where the homodimer ture are much more potent that the individual
is stabilized by hydrophobic interactions encap- peptides alone. In a recent extension of this previ-
sulating a hydrophilic cluster made up from the ous study, the (PGLa-GGC)2 and (magainin-­
two chains (Verly et al. 2017). The dimer is con- GGC)2 homodimers as well as the magainin-GGC/
siderably more active than the monomers (Verly PGLa-GGC heterodimer showed increased cal-
et al. 2017) suggesting that the homodimer is cein release activities from POPE/POPG
more membrane-disruptive than two monomers 3/1 mole/mole liposomes when compared to
(illustrated in Fig. 4.6c). Additional investiga- unmodified peptides in mixtures (Leber et al.
tions of the homotarsinin membrane interactions 2018). In POPC/cholesterol 3/1 mole/mole mix-
are ongoing. tures, only the PGLa-homodimer and the PGLa-­
Notably a cystine-linked magainin 2 dimer magainin heterodimer but not unlinked peptides
was also found to exhibit enhanced membrane showed significant release activities at all (Leber
permeabilization and antimicrobial activities et al. 2018). These findings are in line with
(Dempsey et al. 2003). The higher efficiency was increased membrane-perturbing properties of
correlated to an enhanced association with nega- larger peptide aggregates (illustrated in Fig. 4.6a–
tively charged bilayers and a reduction in the c) (Dempsey et al. 2003; Lorenzon et al. 2016;
concentration dependence in the dimer when Verly et al. 2017).
compared to the monomer. On the one hand, the
formation of magainin oligomers had been sug-
gested early on; FRET measurements in the pres- 4.13 Synergistic Enhancement
ence of membranes did not yield evidence for of the Activities
dimer or higher oligomer formation (Clark et al. of Antimicrobial Peptides
2011). On the other hand, NMR structural inves-
tigations of 5 mM magainin in the presence of Synergistic enhancement of antimicrobial activi-
0.5 mM DLPC showed a dimer arrangement of ties has been described (McCafferty et al. 1999;
the peptide (Wakamatsu et al. 2002). In this Acar 2000) and includes mixtures of peptides
densely packed peptide-lipid supramolecular with conventional antibiotics (Chou et al. 2016;
aggregate, aromatic interactions involve mostly Bolosov et al. 2017; Kim et al. 2017; Payne et al.
the F5Y and F16W sites, which had been intro- 2017; Rank et al. 2017; Sakoulas et al. 2017).
duced artificially for better assignment, and these Furthermore, combinations of different peptides
amino acids are involved in the dimer interface from the dermaseptin or the bacteriocin
(Wakamatsu et al. 2002). In combination with ­family (Mor et al. 1994; McCafferty et al. 1999),
magainin monomers, the dimer stabilizes the magainin 2 and PGLa (Vaz Gomes et al. 1993),
pore formation in egg-PG membranes (Hara et al. or magainin and the cyclic β-sheet tachyplesin I
2001). Notably whereas a dimer linked through a sequence have been shown to interact in a syner-
carboxy-terminal lysine extension was consider- gistic manner (Kobayashi et al. 2001). Here we
ably more active than the monomer, the amino-­ will shortly summarize findings made with
terminal linkage through glutamic acid has no magainin 2 and PGLa which have recently been
effect (Lorenzon et al. 2016). reviewed in more detail (Marquette and Bechinger
Furthermore, the activities of homodimers of 2018). This mixture is of particular interest
magainin 2 or PGLa carrying GGC extensions because synergism has been observed in antimi-
and of a heterodimeric hybrid made of the two crobial assays but also when model membranes
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 55

are tested (Westerhoff et al. 1995; Matsuzaki terminal residues of magainin showed some
et al. 1998; Leber et al. 2018), suggesting that the effect in reducing the synergistic activity, whereas
mode of action should reveal itself when study- modifying F5W was neutral (Matsuzaki et al.
ing the membrane interactions. In cellular assays, 1998). Thus, when replacing the negative E19
it is also possible that one component helps to and the carboxy-terminus of magainin 2, syner-
path the way of the second component to its gism is abolished (Zerweck et al. 2017). Notably,
active site. In this context it is noteworthy that amidation of the magainin 2 carboxyterminus has
synergism in calcein release experiments was been shown to increase the activity of the peptide
more pronounced for membranes with high nega- (Cuervo et al. 1988), but within experimental
tive intrinsic curvature such as POPE/POPG error, the combination with PGLa exhibits a simi-
3/1 mole/mole (Leber et al. 2018). Furthermore, lar degree of synergism (Marquette et al. 2015;
these peptides naturally occur as a mixture in the Glattard et al. 2016).
skin of Xenopus laevis frogs, suggesting that our When the PGLa sequence was modified, the
conventional approach to separate such compo- positively charged K15 and K19 sites had a favor-
nents to analyze each one individually should be able synergistic effect (Zerweck et al. 2017).
reconsidered. Furthermore, the G7, G11, and L18 positions of
Early on Matsuzaki et al. suggested that the PGLa are important for the synergistic enhance-
formation of magainin pores is slow, but once ment of activities between the two peptides
formed, they are more stable than those of PGLa (Zerweck et al. 2017). Cross-linking experiments
(Matsuzaki et al. 1998). It was also suggested with PGLa and magainin 2 both carrying a GGC
that synergistic vesicle leakage results from opti- extensions indicate that in egg-PC/PG 1/1 mole/
mizing both pore size and distribution among the mole lipid membranes, parallel dimers preferen-
liposomes present in the suspension (Patel et al. tially form (Hara et al. 2001).
2014). The openings should be large enough and From fluorescence binding experiments, ener-
at the same time abundant in order for all the gies were derived that suggest favorable interac-
entrapped dye being released. Such more general tions when magainin and PGLa are added to
concepts are nicely complemented by experi- egg-PG membranes (Matsuzaki et al. 1998).
ments aiming to reveal the mechanisms of syner- However, FRET experiments do not reveal
gism at a molecular, even at an atomistic level. strong, long-lasting contacts between the pep-
Solid-state NMR investigations show that in tides when associated with POPE/POPG 3/1 or
equimolar mixtures and in bilayers whose lipid POPC/POPS 3/1 membranes (Marquette et al.
composition resembles that of membranes that 2015). When investigated by a combination of
occur in nature, PGLa and magainin exhibit an ITC and dynamic light scattering, an additional
alignment parallel to the membrane surface exothermic contribution in the presence of both
(Salnikov and Bechinger 2011; Strandberg et al. peptides was attributed to the agglutination of the
2013; Glattard et al. 2016; Salnikov et al. 2016a). liposomes (Marquette and Bechinger 2018) simi-
Thus, the helix topology does not change much lar to studies with other amphipathic helical pep-
when compared to investigations of magainin or tides (Marquette et al. 2010; Vermeer et al. 2016).
PGLa individually (Bechinger 2011; Salnikov Such processes involving the lipids and/or
and Bechinger 2011). It should be mentioned that changes in the supramolecular assembly of pep-
the PGLa behavior is somewhat different in fully tides and lipids also contribute to the total inter-
saturated membranes (Salnikov and Bechinger action energies (Bechinger 1996; Harmouche
2011; Strandberg et al. 2013; Harmouche and and Bechinger 2018). Considering such supra-
Bechinger 2018). However, here we focus on molecular changes, it should also be mentioned
studies that were obtained with more biological that a reduced bilayer repeat distance has been
lipid compositions, carrying an unsaturation. In observed when both peptides are present but not
dye release experiments, mutating the carboxy- with PGLa alone (Grage et al. 2016).
56 C. Aisenbrey et al.

4.14 Conclusions Ahn M, Gunasekaran P, Rajasekaran G, Kim EY, Lee SJ,

Bang G, Cho K, Hyun JK, Lee HJ, Jeon YH, Kim NH,
Ryu EK, Shin SY, Bang JK (2017) Pyrazole derived
During three decades of research, we have moved ultra-short antimicrobial peptidomimetics with potent
from a textbook view where AMPs form pores in anti-biofilm activity. Eur J Med Chem 125:551–564
the shape of transmembrane helical bundles to a Aisenbrey C, Bechinger B (2004) Tilt and rotational pitch
angles of membrane-inserted polypeptides from com-
more complex vision where lipids play an essen- bined 15N and 2H solid-state NMR spectroscopy.
tial role (Bechinger et al. 1991a, b; Pouny et al. Biochemistry 43:10502–10512
1992). Thus, in order to understand the mecha- Aisenbrey C, Bechinger B (2014) Molecular packing of
nism of action not only of AMPs alone but also of amphipathic peptides on the surface of lipid mem-
branes. Langmuir 30:10374–10383
their synergism, an in-depth understanding of the Amos ST, Vermeer LS, Ferguson PM, Kozlowska J, Davy
interactions within the supramolecular lipid-­ M, Bui TT, Drake AF, Lorenz CD, Mason AJ (2016)
peptide architectures is required (Bechinger and Antimicrobial peptide potency is facilitated by greater
Lohner 2006; Imura et al. 2008; Kim et al. 2009; conformational flexibility when binding to gram-­
negative bacterial inner membranes. Sci Rep 6:37639
Aisenbrey and Bechinger 2014). The molecular Arnusch CJ, Albada HB, van Vaardegem M, Liskamp
shape and SMART models (Bechinger 2009; RMJ, Sahl HG, Shadkchan Y, Osherov N, Shai Y
2015) provide valuable concepts from where to (2012) Trivalent ultrashort lipopeptides are potent
further explore these activities, where new tech- pH dependent antifungal agents. J Med Chem
55:1296–1302
nological approaches reveal how structure, topol- Avitabile C, D’Andrea LD, Romanelli A (2014) Circular
ogy, dynamics, and interactions evolve in a dichroism studies on the interactions of antimicrobial
spatiotemporal manner. peptides with bacterial cells. Sci Rep 4:4293
Bak M, Bywater RP, Hohwy M, Thomsen JK, Adelhorst
K, Jakobsen HJ, Sorensen OW, Nielsen NC (2001)
Acknowledgments We gratefully acknowledge the dis-
Conformation of alamethicin in oriented phospholipid
cussion with and contributions from many coworkers and
bilayers determined by N-15 solid-state nuclear mag-
colleagues from our own team and from outside. We are
netic resonance. Biophys J 81:1684–1698
grateful to Ekaterina Zaitseva for her feedback on electro-
Barns KJ, Weisshaar JC (2013) Real-time attack of LL-37
physiology and Roland Stote for much helping to refine
on single Bacillus subtilis cells. Biochim Biophys
the text on molecular dynamics. Over the years, the
Acta 1828:1511–1520
Agence Nationale de la Recherche (projects TRANSPEP
Barns KJ, Weisshaar JC (2016) Single-cell, time-resolved
07-PCV-0018, ProLipIn 10-BLAN-731, membraneDNP
study of the effects of the antimicrobial peptide alam-
12-BSV5-0012, MemPepSyn 14-CE34-0001-01,
ethicin on Bacillus subtilis. Biochim Biophys Acta
InMembrane 15-CE11-0017-01, Biosupramol 17-CE18-­
1858:725–732
0033-3, and the LabEx Chemistry of Complex Systems
Batenburg AM, van Esch JH, De Kruijff B (1988)
10-LABX-0026_CSC), the IRTG Soft Matter Science
Melittin-­induced changes of the macroscopic struc-
(Freiburg, Strasbourg), the Marie-Curie Research and
ture of phosphatidylethanolamines. Biochemistry
Training Network 33439 of the European Commission
27:2324–2331
BIOCONTROL, the University of Strasbourg, the CNRS,
Bechinger B (1996) Towards membrane protein design:
the Région Alsace, the RTRA International Center of
pH-sensitive topology of histidine-containing poly-
Frontier Research in Chemistry, and the French
peptides. J Mol Biol 263:768–775
Foundation for Medical Research (FRM) have provided
Bechinger B (1997) Structure and functions of channel-­
financial support. BB thanks the Institut Universitaire de
forming polypeptides: magainins, cecropins, melittin
France for providing additional time to be dedicated to
and alamethicin. J Membr Biol 156:197–211
research.
Bechinger B (1999) The structure, dynamics and orienta-
tion of antimicrobial peptides in membranes by mul-
tidimensional solid-state NMR spectroscopy. Biochim
References Biophys Acta 1462:157–183
Bechinger B (2004) Membrane-lytic peptides. Crit Rev
Plant Sci 23:271–292
Acar JF (2000) Antibiotic synergy and antagonism. Med
Bechinger B (2005) Detergent-like properties of magainin
Clin N Am 84:1391–1406
antibiotic peptides: a 31P solid-state NMR study.
Agerberth B, Gunne H, Odeberg J, Kogner P, Boman HG,
Biochim Biophys Acta 1712:101–108
Gudmundsson GH (1995) FALL-39, a putative human
Bechinger B (2009) Rationalizing the membrane interac-
peptide antibiotic, is cysteine-free and expressed in
tions of cationic amphipathic antimicrobial peptides
bone marrow and testis. Proc Natl Acad Sci U S A
by their molecular shape. Cur Opin Colloid Interface
92:195–199
Sci Surfactant 14:349–355
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 57

Bechinger B (2011) Insights into the mechanisms of Bortolus M, Dalzini A, Toniolo C, Hahm KS, Maniero AL
action of host defence peptides from biophysical and (2014) Interaction of hydrophobic and amphipathic
structural investigations. J Pept Sci 17:306–314 antimicrobial peptides with lipid bicelles. J Pept Sci
Bechinger B (2015) The SMART model: soft membranes 20:517–525
adapt and respond, also transiently, to external stimuli. Brown LR (1979) Use of fully deuterated micelles for
J Pept Sci 21:346–355 conformational studies of membrane proteins by high
Bechinger B, Gorr SU (2017) Antimicrobial peptides: resolution 1H nuclear magnetic resonance. Biochim
mechanisms of action and resistance. J Dent Res Biophys Acta 557:135–148
96:254–260 Bruch MD, Dhingra MM, Gierasch LM (1991) Side
Bechinger B, Lohner K (2006) Detergent-like action of chain-backbone hydrogen bonding contributes to helix
linear cationic membrane-active antibiotic peptides. stability in peptides derived from an alpha-helical
Biochim Biophys Acta 1758:1529–1539 region of carboxypeptidase A. Proteins Struct Funct
Bechinger B, Salnikov ES (2012) The membrane interac- Genet 10:130–139
tions of antimicrobial peptides revealed by solid-state Cao P, Yang Y, Uche FI, Hart SR, Li WW, Yuan C (2018)
NMR spectroscopy. Chem Phys Lipids 165:282–301 Coupling plant-derived cyclotides to metal surfaces:
Bechinger B, Sizun C (2003) Alignment and structural an antibacterial and antibiofilm study. Int J Mol Sci
analysis of membrane polypeptides by 15N and 31P 19:E793
solid-state NMR spectroscopy. Concepts Magn Reson Caputo GA, London E (2003) Using a novel dual fluores-
18A:130–145 cence quenching assay for measurement of tryptophan
Bechinger B, Shon K, Eck H, Zasloff M, Opella SJ (1990) depth within lipid bilayers to determine hydrophobic
NMR studies of magainin peptide antibiotics in mem- alpha-helix locations within membranes. Biochemistry
branes. Biol Chem Hoppe Seyler 371:758–758 42:3265–3274
Bechinger B, Kim Y, Chirlian LE, Gesell J, Neumann JM, Cardoso MH, de Almeida KC, Candido ES, Murad
Montal M, Tomich J, Zasloff M, Opella SJ (1991a) AM, Dias SC, Franco OL (2017) Comparative
Orientations of amphipathic helical peptides in mem- NanoUPLC-­ MS(E) analysis between magainin
brane bilayers determined by solid- state NMR spec- I-susceptible and -resistant Escherichia coli strains.
troscopy. J Biomol NMR 1:167–173 Sci Rep 7:4197
Bechinger B, Zasloff M, Opella SJ (1991b) Solid state Carrillo C, Teruel JA, Aranda FJ, Ortiz A (2003)
NMR of magainin and PGLa peptide antibiotics in Molecular mechanism of membrane permeabilization
bilayers. J Cell Biochem Suppl 15G:84 by the peptide antibiotic surfactin. Biochim Biophys
Bechinger B, Zasloff M, Opella SJ (1992) Structure and Acta 1611:91–97
interactions of magainin antibiotic peptides in lipid Chang S, Sievert DM, Hageman JC, Boulton ML, Tenover
bilayers: a solid-state NMR investigation. BiophysJ FC, Downes FP, Shah S, Rudrik JT, Pupp GR, Brown
62:12–14 WJ, Cardo D, Fridkin SK, Staphylococcus V-R (2003)
Bechinger B, Zasloff M, Opella SJ (1993) Structure and Infection with vancomycin-resistant Staphylococcus
orientation of the antibiotic peptide magainin in mem- aureus containing the vanA resistance gene. N Engl
branes by solid-state NMR spectroscopy. Protein Sci J Med 348:1342–1347
2:2077–2084 Chen FY, Lee MT, Huang HW (2003) Evidence for mem-
Bechinger B, Zasloff M, Opella SJ (1998) Structure and brane thinning effect as the mechanism for peptide-­
dynamics of the antibiotic peptide PGLa in mem- induced pore formation. Biophys J 84:3751–3758
branes by multidimensional solution and solid-state Cheng JTJ, Hale JD, Elliot M, Hancock REW, Straus SK
NMR spectroscopy. Biophys J 74:981–987 (2009) Effect of membrane composition on antimi-
Bechinger B, Ruysschaert JM, Goormaghtigh E (1999) crobial peptides aurein 2.2 and 2.3 from Australian
Membrane helix orientation from linear dichroism of Southern Bell Frogs. Biophys J 96:552–565
infrared attenuated Total reflection spectra. Biophys Cheng JTJ, Hale JD, Elliott M, Hancock REW, Straus
J 76:552–563 SK (2011) The importance of bacterial membrane
Bechinger B, Resende JM, Aisenbrey C (2011) The struc- composition in the structure and function of aurein
tural and topological analysis of membrane-associated 2.2 and selected variants. Biochim Biophys Acta
polypeptides by oriented solid-state NMR spectros- 1808:622–633
copy: established concepts and novel developments. Choi H, Rangarajan N, Weisshaar JC (2016) Lights,
Biophys Chem 153:115–125 camera, action! Antimicrobial peptide mecha-
Bolen EJ, Holloway PW (1990) Quenching of tryp- nisms imaged in space and time. Trends Microbiol
tophan fluorescence by brominated phospholipid. 24:111–122
Biochemistry 29:9638–9643 Choi H, Yang Z, Weisshaar JC (2017) Oxidative stress
Bolosov IA, Kalashnikov AA, Panteleev PV, Ovchinnikova induced in E. coli by the human antimicrobial peptide
TV (2017) Analysis of synergistic effects of antimi- LL-37. PLoS Pathog 13:e1006481
crobial peptide arenicin-1 and conventional antibiot- Chou S, Shao C, Wang J, Shan A, Xu L, Dong N, Li Z
ics. Bull Exp Biol Med 162:765–768 (2016) Short, multiple-stranded beta-hairpin peptides
Boman HG (1995) Peptide antibiotics and their role in have antimicrobial potency with high selectivity and
innate immunity. Annu Rev Immunol 13:61–92 salt resistance. Acta Biomater 30:78–93
58 C. Aisenbrey et al.

Christensen B, Fink J, Merrifield RB, Mauzerall D (1988) Gazit E, Miller IR, Biggin PC, Sansom MSP, Shai Y
Channel-forming properties of cecropins and related (1996) Structure and orientation of the mammalian
model compounds incorporated into planar lipid mem- antibacterial peptide cecropin P1 within phospholipid
branes. Proc Natl Acad Sci USA 85:5072–5076 membranes. J Mol Biol 258:860–870
Clark KS, Svetlovics J, McKeown AN, Huskins L, Georgescu J, Munhoz VHO, Bechinger B (2010) NMR
Almeida PF (2011) What determines the activity of structures of the histidine-rich peptide LAH4 in micel-
antimicrobial and cytolytic peptides in model mem- lar environments: membrane insertion, pH-dependent
branes. Biochemistry 50:7919–7932 mode of antimicrobial action and DNA transfection.
Cruciani RA, Barker JL, Zasloff M, Chen HC, Colamonici Biophys J 99:2507–2515
O (1991) Antibiotic magainins exert cytolytic activ- Ghosh C, Haldar J (2015) Membrane-active small mol-
ity transformed cell lines through channel formation. ecules: designs inspired by antimicrobial peptides.
Proc Natl Acad Sci U S A 88:3792–3796 ChemMedChem 10:1606–1624
Cruzeiro-Hansson L, Mouritsen OG (1988) Passive Ghosh C, Manjunath GB, Akkapeddi P, Yarlagadda V,
ion permeability of lipid membranes modelled via Hoque J, Uppu DSSM, Konai MM, Haldar J (2014)
lipid-domain interfacial area. Biochim Biophys Acta Small molecular antibacterial peptoid mimics: the
944:63–72 simpler the better! J Med Chem 57:1428–1436
Cuervo JH, Rodriguez B, Houghten RA (1988) The Ghosh C, Harmouche N, Bechinger B, Haldar J (2018)
magainins: sequence factors relevant to increased anti- Aryl-alkyl-lysines interact with anionic lipid com-
microbial activity and decreased hemolytic activity. ponents of bacterial cell envelope eliciting anti-­
Pept Res 1:81–86 inflammatory and anti-biofilm properties. ACS Omega
Dalla Serra M, Cirioni O, Vitale RM, Renzone G, 3:9182–9190
Coraiola M, Giacometti A, Potrich C, Baroni E, Giovannini MG, Poulter L, Gibson BW, Williams DH
Guella G, Sanseverino M, De LS, Scalise G, Amodeo (1987) Biosynthesis and degradation of peptides
P, Scaloni A (2008) Structural features of distinctin derived from Xenopus laevis prohormones. Biochem
affecting peptide biological and biochemical proper- J 243:113–120
ties. Biochemistry 47:7888–7899 Glattard E, Salnikov ES, Aisenbrey C, Bechinger B
Das N, Dai J, Hung I, Rajagopalan MR, Zhou HX, Cross (2016) Investigations of the synergistic enhancement
TA (2015) Structure of CrgA, a cell division structural of antimicrobial activity in mixtures of magainin 2 and
and regulatory protein from Mycobacterium tuber- PGLa. Biophys Chem 210:35–44
culosis, in lipid bilayers. Proc Natl Acad Sci U S A Gogonea V (2015) Structural insights into high den-
112:E119–E126 sity lipoprotein: old models and new facts. Front
del Rio Martinez JM, Zaitseva E, Petersen S, Baaken G, Pharmacol 6:318
Behrends JC (2015) Automated formation of lipid Gopinath T, Veglia G (2015) Multiple acquisition of
membrane microarrays for ionic single-molecule magic angle spinning solid-state NMR experiments
sensing with protein nanopores. Small 11:119–125 using one receiver: application to microcrystalline
Dempsey CE, Ueno S, Avison MB (2003) Enhanced mem- and membrane protein preparations. J Magn Reson
brane permeabilization and antibacterial activity of a 253:143–153
disulfide-dimerized magainin analogue. Biochemistry Gopinath T, Mote KR, Veglia G (2015) Simultaneous
42:402–409 acquisition of 2D and 3D solid-state NMR experi-
Diamond G, Beckloff N, Weinberg A, Kisich KO (2009) ments for sequential assignment of oriented mem-
The roles of antimicrobial peptides in innate host brane protein samples. J Biomol NMR 62:53–61
defense. Curr Pharm Des 15:2377–2392 Grage SL, Afonin S, Kara S, Buth G, Ulrich AS (2016)
Duclohier H, Molle G, Spach G (1989) Antimicrobial Membrane thinning and thickening induced by
peptide magainin I from xenopus skin forms anion-­ membrane-­ active amphipathic peptides. Front Cell
permeable channels in planar lipid bilayers. Biophys Dev Biol 4:65
J 56:1017–1021 Grau A, Gomez Fernandez JC, Peypoux F, Ortiz A (1999)
Eddy MT, Su Y, Silvers R, Andreas L, Clark L, Wagner A study on the interactions of surfactin with phospho-
G, Pintacuda G, Emsley L, Griffin RG (2015) Lipid lipid vesicles. Biochim Biophys Acta 1418:307–319
bilayer-bound conformation of an integral membrane Gregory SM, Cavenaugh A, Journigan V, Pokorny A,
beta barrel protein by multidimensional MAS NMR. J Almeida PFF (2008) A quantitative model for the
Biomol NMR 61:299–310 all-or-none permeabilization of phospholipid vesicles
Farrotti A, Bocchinfuso G, Palleschi A, Rosato N, by the antimicrobial peptide cecropin A. Biophys
Salnikov ES, Voievoda N, Bechinger B, Stella L J 94:1667–1680
(2015) Molecular dynamics methods to predict pep- Hadley EB, Hancock RE (2010) Strategies for the discov-
tide location in membranes: LAH4 as a stringent test ery and advancement of novel cationic antimicrobial
case. Biochim Biophys Acta 1848:581–592 peptides. Curr Top Med Chem 10:1872–1881
Gallaher J, Wodzinska K, Heimburg T, Bier M (2010) Hall K, Lee TH, Mechler AI, Swann MJ, Aguilar MI
Ion-channel-like behavior in lipid bilayer membranes (2014) Real-time measurement of membrane confor-
at the melting transition. Phys Rev E Stat Nonlinear mational states induced by antimicrobial peptides: bal-
Soft Matter Phys 81:061925 ance between recovery and lysis. Sci Rep 4:5479
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 59

Hallock KJ, Lee DK, Omnaas J, Mosberg HI, Hong M, Su Y (2011) Structure and dynamics of cationic
Ramamoorthy A (2002) Membrane composition membrane peptides and proteins: insights from solid-­
determines pardaxin’s mechanism of lipid bilayer dis- state NMR. Protein Sci 20:641–655
ruption. Biophys J 83:1004–1013 Imura Y, Choda N, Matsuzaki K (2008) Magainin 2 in
Hara T, Kodama H, Kondo M, Wakamatsu K, Takeda action: distinct modes of membrane permeabiliza-
A, Tachi T, Matsuzaki K (2001) Effects of peptide tion in living bacterial and mammalian cells. Biophys
dimerization on pore formation: antiparallel disulfide-­ J 95:5757–5765
dimerized magainin 2 analogue. Biopolymers Ines M, Dhouha G (2015) Lipopeptide surfactants: pro-
58:437–446 duction, recovery and pore forming capacity. Peptides
Harmouche N, Bechinger B (2018) Lipid-mediated inter- 71:100–112
actions between the amphipathic antimicrobial pep- Islam MZ, Alam JM, Tamba Y, Karal MAS, Yamazaki M
tides magainin 2 and PGLa in phospholipid bilayers. (2014) The single GUV method for revealing the func-
Biophys J 115:1033-1044 tions of antimicrobial, pore-forming toxin, and cell-­
Harpole TJ, Delemotte L (2018) Conformational land- penetrating peptides or proteins. Phys Chem Chem
scapes of membrane proteins delineated by enhanced Phys 16:15752–15767
sampling molecular dynamics simulations. Biochim Israelachvili JN, Marcelja S, Horn RG (1980) Physical
Biophys Acta 1860:909–926 principles of membrane organization. Q Rev Biophys
Harzer U, Bechinger B (2000) The alignment of lysine-­ 13:121–200
anchored membrane peptides under conditions of Itkin A, Salnikov ES, Aisenbrey C, Raya J, Raussens V,
hydrophobic mismatch: a CD, 15 N and 31 P solid-­ Ruysschaert JM, Bechinger B (2017) Evidence for
state NMR spectroscopy investigation. Biochemistry heterogeneous conformations of the gamma cleavage
39:13106–13114 site within the amyloid precursor proteins transmem-
Hasan M, Karal MAS, Levadnyy V, Yamazaki M (2018) brane domain. ACS Omega 2:6525–6534
Mechanism of initial stage of pore formation induced Jaipuria G, Leonov A, Giller K, Vasa SK, Jaremko L,
by antimicrobial peptide magainin 2. Langmuir Jaremko M, Linser R, Becker S, Zweckstetter M
34:3349–3362 (2017) Cholesterol-mediated allosteric regulation of
Hayden RM, Goldberg GK, Ferguson BM, Schoeneck the mitochondrial translocator protein structure. Nat
MW, Libardo MD, Mayeux SE, Shrestha A, Commun 8:14893
Bogardus KA, Hammer J, Pryshchep S, Lehman Jean-Francois F, Castano S, Desbat B, Odaert B, Roux M,
HK, McCormick ML, Blazyk J, Angeles-Boza AM, Metz-Boutigue MH, Dufourc EJ (2008) Aggregation
Fu R, Cotten ML (2015) Complementary effects of of cateslytin beta-sheets on negatively charged lip-
host defense peptides piscidin 1 and piscidin 3 on ids promotes rigid membrane domains. A new mode
DNA and lipid membranes: biophysical insights of action for antimicrobial peptides? Biochemistry
into contrasting biological activities. J Phys Chem B 47:6394–6402
119:15235–15246 Jenssen H, Hamill P, Hancock RE (2006) Peptide antimi-
He K, Ludtke SJ, Heller WT, Huang HW (1996) crobial agents. Clin Microbiol Rev 19:491–511
Mechanism of alamethicin insertion into lipid bilay- Kandaswamy K, Liew TH, Wang CY, Huston-Warren E,
ers. Biophys J 71:2669–2679 Meyer-Hoffert U, Hultenby K, Schroder JM, Caparon
Heerklotz H, Seelig J (2001) Detergent-like action of MG, Normark S, Henriques-Normark B, Hultgren
the antibiotic peptide surfactin on lipid membranes. SJ, Kline KA (2013) Focal targeting by human beta-­
Biophys J 81:1547–1554 defensin 2 disrupts localized virulence factor assem-
Heerklotz H, Seelig J (2007) Leakage and lysis of lipid bly sites in Enterococcus faecalis. Proc Natl Acad Sci
membranes induced by the lipopeptide surfactin. Eur U S A 110:20230–20235
Biophys J 36:305–314 Karal MA, Alam JM, Takahashi T, Levadny V, Yamazaki
Heerklotz H, Wieprecht T, Seelig J (2004) Membrane M (2015) Stretch-activated pore of the antimicrobial
perturbation by the lipopeptide surfactin and deter- peptide, magainin 2. Langmuir 31:3391–3401
gents as studied by deuterium. J Phys Chem B Killian JA, de Planque MRR, van der Wel PCA, Salemink
108:4909–4915 I, De Kruijff B, Greathouse DV, Koeppe RE (1998)
Heimburg T (2012) The capacitance and electrome- Modulation of membrane structure and function by
chanical coupling of lipid membranes close to hydrophobic mismatch between proteins and lipids.
transitions: the effect of electrostriction. Biophys Pure Appl Chem 70:75–82
J 103:918–929 Kim C, Spano J, Park EK, Wi S (2009) Evidence of pores
Henderson JM, Waring AJ, Separovic F, Lee KYC (2016) and thinned lipid bilayers induced in oriented lipid
Antimicrobial peptides share a common interaction membranes interacting with the antimicrobial pep-
driven by membrane line tension reduction. Biophys tides, magainin-2 and aurein-3.3. Biochim Biophys
J 111:2176–2189 Acta 1788:1482–1496
Holzl MA, Hofer J, Steinberger P, Pfistershammer K, Kim EY, Rajasekaran G, Shin SY (2017) LL-37-derived
Zlabinger GJ (2008) Host antimicrobial proteins short antimicrobial peptide KR-12-a5 and its d-amino
as endogenous immunomodulators. Immunol Lett acid substituted analogs with cell selectivity, anti-­
119:4–11 biofilm activity, synergistic effect with conventional
60 C. Aisenbrey et al.

antibiotics, and anti-inflammatory activity. Eur J Med Lohner K (2009) New strategies for novel antibiotics:
Chem 136:428–441 peptides targeting bacterial cell membranes. Gen
Kindrachuk J, Napper S (2010) Structure-activity rela- Physiol Biophys 28:105–116
tionships of multifunctional host defence peptides. Lorenzon EN, Santos-Filho NA, Ramos MA, Bauab TM,
Mini-Rev Med Chem 10:596–614 Camargo IL, Cilli EM (2016) C-terminal lysine-linked
Kiss G, Michl H (1962) Über das Giftsekret der magainin 2 with increased activity against multidrug-­
Gelbbauchunke Bombina variegata L. Toxicon resistant bacteria. Protein Pept Lett 23:738–747
(Oxford) 1:33–39 Loudet C, Khemtemourian L, Aussenac F, Gineste S,
Klocek G, Seelig J (2008) Melittin interaction Achard MF, Dufourc EJ (2005) Bicelle membranes
with sulfated cell surface sugars. Biochemistry and their use for hydrophobic peptide studies by circu-
47:2841–2849 lar dichroism and solid state NMR. Biochim Biophys
Kmiecik S, Gront D, Kolinski M, Wieteska L, Dawid AE, Acta 1724:315–323
Kolinski A (2016) Coarse-grained protein models and Ludtke SJ, He K, Wu Y, Huang HW (1994) Cooperative
their applications. Chem Rev 116:7898–7936 membrane insertion of magainin correlated with
Kobayashi S, Hirakura Y, Matsuzaki K (2001) Bacteria-­ its cytolytic activity. Biochim Biophys Acta
selective synergism between the antimicrobial pep- 1190:181–184
tides alpha-helical magainin 2 and cyclic beta-sheet Ludtke S, He K, Huang H (1995) Membrane thinning
tachyplesin I: toward cocktail therapy. Biochemistry caused by magainin 2. Biochemistry 34:16764–16769
40:14330–14335 Ludtke SJ, He K, Heller WT, Harroun TA, Yang L, Huang
Kollmitzer B, Heftberger P, Rappolt M, Pabst G (2013) HW (1996) Membrane pores induced by magainin.
Monolayer spontaneous curvature of raft-forming Biochemistry 35:13723–13728
membrane lipids. Soft Matter 9:10877–10884 Maget-Dana R, Peypoux F (1994) Iturins, a special class
Kozlowska J, Vermeer LS, Rogers GB, Rehnnuma of pore-forming lipopeptides: biological and physico-
N, Amos SB, Koller G, McArthur M, Bruce KD, chemical properties. Toxicology 87:151–174
Mason AJ (2014) Combined systems approaches Maget-Dana R, Ptak M (1995) Interactions of surfactin
reveal highly plastic responses to antimicrobial with membrane models. Biophys J 68:1937–1943
peptide challenge in Escherichia coli. PLoS Pathog Makovitzki A, Baram J, Shai Y (2008) Antimicrobial
10:e1004104 lipopolypeptides composed of palmitoyl Di- and tri-
Kuroda K, DeGrado WF (2005) Amphiphilic polymethac- cationic peptides: in vitro and in vivo activities, self-­
rylate derivatives as antimicrobial agents. J Am Chem assembly to nanostructures, and a plausible mode of
Soc 127:4128–4129 action. Biochemistry 47:10630–10636
Ladokhin AS, Wimley WC, White SH (1995) Leakage Mangoni ML, Shai Y (2011) Short native antimicrobial
of membrane vesicle contents – determination of peptides and engineered ultrashort lipopeptides: simi-
mechanisms using fluorescence requenching. Biophys larities and differences in cell specificities and modes
J 69:1964–1971 of action. Cell Mol Life Sci 68:2267–2280
Laub KR, Witschas K, Blicher A, Madsen SB, Luckhoff Marquette A, Bechinger B (2018) Biophysical investiga-
A, Heimburg T (2012) Comparing ion conductance tions elucidating the mechanisms of action of anti-
recordings of synthetic lipid bilayers with cell mem- microbial peptides and their synergism. Biomol Ther
branes containing TRP channels. Biochim Biophys 8:E18
Acta 1818:1123–1134 Marquette A, Mason AJ, Bechinger B (2008) Aggregation
Laurencin M, Simon M, Fleury Y, Baudy-Floc’h M, and membrane permeabilizing properties of designed
Bondon A, Legrand B (2018) Selectivity modulation histidine-containing cationic linear peptide antibiot-
and structure of alpha/aza-beta(3) cyclic antimicrobial ics. J Pept Sci 14:488–495
peptides. Chemistry 24:6191–6201 Marquette A, Lorber B, Bechinger B (2010) Reversible
Lear JD, Wasserman ZR, DeGrado WF (1988) Synthetic liposome association induced by LAH4: a peptide
amphiphilic peptide models for protein ion channels. with potent antimicrobial and nucleic acid transfection
Science 240:1177–1181 activities. Biophys J 98:2544–2553
Leber R, Pachler M, Kabelka I, Svoboda I, Enkoller D, Marquette A, Salnikov E, Glattard E, Aisenbrey C,
Vácha R, Lohner K, Pabst G (2018) Synergism of anti- Bechinger B (2015) Magainin 2-PGLa interactions
microbial frog peptides couples to membrane intrinsic in membranes – two peptides that exhibit synergistic
curvature strain. Biophys J 114:1945–1954 enhancement of antimicrobial activity. Curr Top Med
Leitgeb B, Szekeres A, Manczinger L, Vagvolgyi C, Chem 16:65–75
Kredics L (2007) The history of alamethicin: a review Mason AJ, Martinez A, Glaubitz C, Danos O, Kichler
of the most extensively studied peptaibol. Chem A, Bechinger B (2006) The antibiotic and DNA-­
Biodivers 4:1027–1051 transfecting peptide LAH4 selectively associates with,
Liu SP, Zhou L, Lakshminarayanan R, Beuerman RW and disorders, anionic lipids in mixed membranes.
(2010) Multivalent antimicrobial peptides as thera- FASEB J 20:320–322
peutics: design principles and structural diversities. Int Mason AJ, Moussaoui W, Abdelrhaman T, Boukhari A,
J Pept Res Ther 16:199–213 Bertani P, Marquette A, Shooshtarizaheh P, Moulay G,
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 61

Boehm N, Guerold B, Sawers RJH, Kichler A, Metz- Milov AD, Samoilova RI, Tsvetkov YD, De Zotti M,
Boutigue MH, Candolfi E, Prevost G, Bechinger B (2009) Formaggio F, Toniolo C, Handgraaf JW, Raap J (2009)
Structural determinants of antimicrobial and antiplasmo- Structure of self-aggregated alamethicin in ePC mem-
dial activity and selectivity in histidine rich amphipathic branes detected by pulsed electron-electron double
cationic peptides. J Biol Chem 284:119–133 resonance and electron spin Echo envelope modula-
Matos PM, Franquelim HG, Castanho MA, Santos NC tion spectroscopies. Biophys J 96:3197–3209
(2010) Quantitative assessment of peptide-lipid inter- Montal M, Mueller P (1972) Formation of bimolecular
actions. Ubiquitous fluorescence methodologies. membranes from lipid monolayers and a study of
Biochim Biophys Acta 1798:1999–2012 their electrical properties. Proc Natl Acad Sci U S A
Matsuzaki K (1998) Magainins as paradigm for the mode 69:3561–3566
of action of pore forming polypeptides. Biochim Mor A, Hani K, Nicolas P (1994) The vertebrate peptide
Biophys Acta 1376:391–400 antibiotics dermaseptins have overlapping structural
Matsuzaki K, Harada M, Funakoshi S, Fujii N, Miyajima features but target specific microorganisms. J Biol
K (1991) Physicochemical determinants for the inter- Chem 269:31635–31641
actions of magainins 1 and 2 with acidic lipid bilayers. Naito A, Matsumori N, Ramamoorthy A (2018) Dynamic
Biochim Biophys Acta 1063:162–170 membrane interactions of antibacterial and antifun-
Matsuzaki K, Murase O, Tokuda H, Funakoshi S, Fujii gal biomolecules, and amyloid peptides, revealed by
N, Miyajima K (1994) Orientational and aggrega- solid-state NMR spectroscopy. Biochim Biophys Acta
tional states of magainin 2 in phospholipid bilayers. 1862:307–323
Biochemistry 33:3342–3349 Nishida M, Imura Y, Yamamoto M, Kobayashi S, Yano Y,
Matsuzaki K, Murase O, Fujii N, Miyajima K (1995a) Matsuzaki K (2007) Interaction of a magainin-PGLa
Translocation of a channel-forming antimicrobial hybrid peptide with membranes: insight into the mech-
peptide, magainin2, across lipid bilayers by forming a anism of synergism. Biochemistry 46:14284–14290
pore. Biochemistry 34:6521–6526 North CL, Barranger-Mathys M, Cafiso DS (1995)
Matsuzaki K, Murase O, Miyajima K (1995b) Kinetics Membrane orientation of the N-terminal segment
of pore formation by an antimicrobial peptide, of alamethicin determined by solid-state 15 N
magainin 2, in phospholipid bilayers. Biochemistry NMR. Biophys J 69:2392–2397
34:12553–12559 Otzen DE (2017) Biosurfactants and surfactants interact-
Matsuzaki K, Murase O, Fujii N, Miyajima K (1996) An ing with membranes and proteins: same but different?
antimicrobial peptide, magainin 2, induced flip-flop of Biochim Biophys Acta 1859:639–649
phospholipid coupled with pore formation and peptide Oyston PC, Fox MA, Richards SJ, Clark GC (2009) Novel
translocation. Biochemistry 35:11361–11368 peptide therapeutics for treatment of infections. J Med
Matsuzaki K, Mitani Y, Akada K, Murase O, Yoneyama S, Microbiol 58:977–987
Zasloff M, Miyajima K (1998) Mechanism of syner- Palermo EF, Kuroda K (2010) Structural determinants
gism between antimicrobial peptides magainin 2 and of antimicrobial activity in polymers which mimic
PGLa. Biochemistry 37:15144–15153 host defense peptides. Appl Microbiol Biotechnol
McCafferty DG, Cudic P, Yu MK, Behenna DC, Kruger 87:1605–1615
R (1999) Synergy and duality in peptide antibiotic Patch JA, Barron AE (2003) Helical peptoid mimics of
mechanisms. Curr Opin Chem Biol 3:672–680 magainin-2 amide. J Am Chem Soc 125:12092–12093
Mecke A, Lee DK, Ramamoorthy A, Orr BG, Banaszak Patel H, Tscheka C, Edwards K, Karlsson G, Heerklotz
Holl MM (2005) Membrane thinning due to anti- H (2011) All-or-none membrane permeabilization
microbial peptide binding: an atomic force micros- by fengycin-type lipopeptides from Bacillus subtilis
copy study of MSI-78 in lipid bilayers. Biophys QST713. Biochim Biophys Acta 1808:2000–2008
J 89:4043–4050 Patel H, Huynh Q, Barlehner D, Heerklotz H (2014)
Medintz IL, Hildebrandt N (2013) Förster resonance Additive and synergistic membrane permeabiliza-
energy transfer: from theory to applications. Wiley-­ tion by antimicrobial (lipo)peptides and detergents.
VCH, Weinheim Biophys J 106:2115–2125
Michalek M, Aisenbrey C, Bechinger B (2014) Paterson DJ, Tassieri M, Reboud J, Wilson R, Cooper JM
Investigation of membrane penetration depth and (2017) Lipid topology and electrostatic interactions
interactions of the amino-terminal domain of hunting- underpin lytic activity of linear cationic antimicrobial
tin: refined analysis by tryptophan fluorescence mea- peptides in membranes. Proc Natl Acad Sci U S A
surement. Eur Biophys J 43:347–360 114:E8324–E8332
Miles AJ, Wallace BA (2006) Synchrotron radiation cir- Payne JE, Dubois AV, Ingram RJ, Weldon S, Taggart CC,
cular dichroism spectroscopy of proteins and applica- Elborn JS, Tunney MM (2017) Activity of innate anti-
tions in structural and functional genomics. Chem Soc microbial peptides and ivacaftor against clinical cys-
Rev 35:39–51 tic fibrosis respiratory pathogens. Int J Antimicrob
Miles AJ, Wallace BA (2016) Circular dichroism spec- Agents 50:427–435
troscopy of membrane proteins. Chem Soc Rev Perrone B, Miles AJ, Salnikov ES, Wallace B, Bechinger
45:4859–4872 B (2014) Lipid- interactions of the LAH4, a peptide
62 C. Aisenbrey et al.

with antimicrobial and nucleic transfection activities. by oriented solid-state NMR spectroscopy. Biophys
Eur Biophys J 43:499–507 J 107:901–911
Pino-Angeles A, Leveritt JM 3rd, Lazaridis T (2016) Pore Rollins-Smith LA, Doersam JK, Longcore JE, Taylor
structure and synergy in antimicrobial peptides of the SK, Shamblin JC, Carey C, Zasloff MA (2002)
magainin family. PLoS Comput Biol 12:e1004570 Antimicrobial peptide defenses against pathogens
Pirtskhalava M, Gabrielian A, Cruz P, Griggs HL, Squires associated with global amphibian declines. Dev Comp
RB, Hurt DE, Grigolava M, Chubinidze M, Gogoladze Immunol 26:63–72
G, Vishnepolsky B, Alekseyev V, Rosenthal A, Rotem S, Mor A (2009) Antimicrobial peptide mimics
Tartakovsky M (2016) DBAASP v.2: an enhanced for improved therapeutic properties. Biochim Biophys
database of structure and antimicrobial/cytotoxic Acta 1788:1582–1592
activity of natural and synthetic peptides. Nucleic Roversi D, Luca V, Aureli S, Park Y, Mangoni ML, Stella
Acids Res 44:D1104–D1112 L (2014) How many AMP molecules kill a bacterium?
Porter EA, Weisblum B, Gellman SH (2002) Mimicry of Spectroscopic determination of PMAP-23 binding to
host-defense peptides by unnatural oligomers: antimi- E. Coli. ACS Chem Biol 9:2003–2007
crobial beta-peptides. J Am Chem Soc 124:7324–7330 Russ WP, Engelman DM (2000) The GxxxG motif: a
Pouny Y, Rapaport D, Mor A, Nicolas P, Shai Y (1992) framework for transmembrane helix-helix association.
Interaction of antimicrobial dermaseptin and its fluo- J Mol Biol 296:911–919
rescently labeled analogues with phospholipid mem- Sakoulas G, Kumaraswamy M, Kousha A, Nizet V (2017)
branes. Biochemistry 31:12416–12423 Interaction of antibiotics with innate host defense fac-
Raimondo D, Andreotti G, Saint N, Amodeo P, Renzone tors against Salmonella enterica serotype Newport.
G, Sanseverino M, Zocchi I, Molle G, Motta A, Scaloni mSphere 2:e00410–e00417
A (2005) A folding-dependent mechanism of antimi- Salnikov E, Bechinger B (2011) Lipid-controlled pep-
crobial peptide resistance to degradation unveiled by tide topology and interactions in bilayers: structural
solution structure of distinctin. Proc Natl Acad Sci U insights into the synergistic enhancement of the anti-
S A 102:6309–6314 microbial activities of PGLa and magainin 2. Biophys
Ramamoorthy A, Thennarasu S, Lee DK, Tan A, Maloy J 100:1473–1480
L (2006) Solid-state NMR investigation of the Salnikov ES, Friedrich H, Li X, Bertani P, Reissmann S,
membrane-­ disrupting mechanism of antimicrobial Hertweck C, O’Neil JD, Raap J, Bechinger B (2009a)
peptides MSI-78 and MSI-594 derived from magainin Structure and alignment of the membrane-associated
2 and melittin. Biophys J 91:206–216 peptaibols ampullosporin A and alamethicin by ori-
Rangarajan N, Bakshi S, Weisshaar JC (2013) Localized ented 15 N and 31 P solid-state NMR spectroscopy.
permeabilization of E. coli membranes by the Biophys J 96:86–100
antimicrobial peptide cecropin A. Biochemistry Salnikov ES, Mason AJ, Bechinger B (2009b) Membrane
52:6584–6594 order perturbation in the presence of antimicro-
Rank LA, Walsh NM, Liu R, Lim FY, Bok JW, Huang bial peptides by 2H solid-state NMR spectroscopy.
M, Keller NP, Gellman SH, Hull CM (2017) A cat- Biochimie 91:734–743
ionic polymer that shows high antifungal activity Salnikov E, Aisenbrey C, Vidovic V, Bechinger B (2010)
against diverse human pathogens. Antimicrob Agents Solid-state NMR approaches to measure topological
Chemother 61:e00204–e00217 equilibria and dynamics of membrane polypeptides.
Rautenbach M, Troskie AM, Vosloo JA (2016a) Biochim Biophys Acta 1798:258–265
Antifungal peptides: to be or not to be membrane Salnikov E, Aisenbrey C, Balandin SV, Zhmak MN,
active. Biochimie 130:132–145 Ovchinnikova AY, Bechinger B (2011) Structure and
Rautenbach M, Troskie AM, Vosloo JA, Dathe ME alignment of the membrane-associated antimicrobial
(2016b) Antifungal membranolytic activity of the peptide arenicin by oriented solid-state NMR spec-
tyrocidines against filamentous plant fungi. Biochimie troscopy. Biochemistry 50:3784–3795
130:122–131 Salnikov ES, Aisenbrey C, Aussenac F, Ouari O, Sarrouj
Reijmar K, Edwards K, Andersson K, Agmo Hernandez H, Reiter C, Tordo P, Engelke F, Bechinger B (2016a)
V (2016) Characterizing and controlling the load- Membrane topologies of the PGLa antimicrobial
ing and release of cationic amphiphilic peptides peptide and a transmembrane anchor sequence by
onto and from PEG-stabilized lipodisks. Langmuir Dynamic Nuclear Polarization/solid-state NMR spec-
32:12091–12099 troscopy. Sci Rep 6:20895
Resende JM, Moraes CM, Munhoz VHDO, Aisenbrey Salnikov ES, Raya J, De Zotti M, Zaitseva E, Peggion C,
C, Verly RM, Bertani P, Cesar A, Pilo-Veloso D, Ballano G, Toniolo C, Raap J, Bechinger B (2016b)
Bechinger B (2009) Membrane structure and confor- Alamethicin supramolecular organization in lipid
mational changes of the antibiotic heterodimeric pep- membranes from 19F solid- state NMR. Biophys
tide distinctin by solid-state NMR spectroscopy. Proc J 111:2450–2459
Natl Acad Sci U S A 106:16639–16644 Salnikov ES, Anantharamaiah GM, Bechinger B (2018)
Resende JM, Verly RM, Aisenbrey C, Amary C, Bertani Supramolecular organization of apolipoprotein A-I –
P, Pilo-Veloso D, Bechinger B (2014) Membrane derived peptides within disc-like arrangements.
interactions of phylloseptin-1, −2, and −3 peptides Biophys J 115:467–477 in press
4 The Mechanisms of Action of Cationic Antimicrobial Peptides Refined by Novel Concepts… 63

Sani MA, Separovic F (2018) Antimicrobial peptide struc- 2-induced pore formation in lipid membranes. J Phys
tures: from model membranes to live cells. Chemistry Chem B 114:12018–12026
24:286–291 Tieleman DP, Hess B, Sansom MS (2002) Analysis and
Sansom MSP (1991) The biophysics of peptide models of evaluation of channel models: simulations of alam-
ion channels. Prog Biophysmol Biol 55:139–235 ethicin. Biophys J 83:2393–2407
Sansom MS (1993) Alamethicin and related peptaibols – Tremouilhac P, Strandberg E, Wadhwani P, Ulrich AS
model ion channels. Eur Biophys J 22:105–124 (2006a) Conditions affecting the re-alignment of the
Scherer PG, Seelig J (1989) Electric charge effects on antimicrobial peptide PGLa in membranes as moni-
phospholipid headgroups. Phosphatidylcholine in tored by solid state 2H-NMR. Biochim Biophys Acta
mixtures with cationic and anionic amphiphiles. 1758:1330–1342
Biochemistry 28:7720–7727 Tremouilhac P, Strandberg E, Wadhwani P, Ulrich AS
Schweizer F (2009) Cationic amphiphilic peptides (2006b) Synergistic transmembrane alignment of the
with cancer-selective toxicity. Eur J Pharmacol antimicrobial heterodimer PGLa/magainin. J Biol
625:190–194 Chem 281:32089–32094
Scott RW, DeGrado WF, Tew GN (2008) De novo Tsutsumi LS, Elmore JM, Dang UT, Wallace MJ,
designed synthetic mimics of antimicrobial peptides. Marreddy R, Lee RB, Tan GT, Hurdle JG, Lee RE,
Curr Opin Biotechnol 19:620–627 Sun D (2018) Solid-phase synthesis and antibacterial
Seelig J (2004) Thermodynamics of lipid-peptide interac- activity of cyclohexapeptide wollamide B analogs.
tions. Biochim Biophys Acta 1666:40–50 ACS Comb Sci 20:172–185
Shai Y (1999) Mechanism of the binding, insertion, and Ulmschneider JP, Smith JC, Ulmschneider MB, Ulrich
destabilization of phospholipid bilayer membranes by AS, Strandberg E (2012) Reorientation and dimeriza-
alpha-helical antimicrobial and cell non-selective lytic tion of the membrane-bound antimicrobial peptide
peptides. Biochim Biophys Acta 1462:55–70 PGLa from microsecond all-atom MD simulations.
Smart M, Rajagopal A, Liu WK, Ha BY (2017) Opposing Biophys J 103:472–482
effects of cationic antimicrobial peptides and divalent Usachev KS, Kolosova OA, Klochkova EA, Yulmetov AR,
cations on bacterial lipopolysaccharides. Phys Rev E Aganov AV, Klochkov VV (2017) Oligomerization of
96:042405 the antimicrobial peptide protegrin-5 in a membrane-­
Sreerama N, Woody RW (2000) Estimation of protein mimicking environment. Structural studies by
secondary structure from circular dichroism spectra: high-resolution NMR spectroscopy. Eur Biophys
comparison of CONTIN, SELCON, and CDSSTR J 46:293–300
methods with an expanded reference set. Anal Vacha R, Frenkel D (2014) Simulations suggest pos-
Biochem 287:252–260 sible novel membrane pore structure. Langmuir
Steinstraesser L, Kraneburg U, Jacobsen F, Al-Benna S 30:1304–1310
(2010) Host defense peptides and their antimicrobial-­ Vaz Gomes A, de Waal A, Berden JA, Westerhoff HV
immunomodulatory duality. Immunobiology (1993) Electric potentiation, cooperativity, and syner-
216:322–333 gism of magainin peptides in protein-free liposomes.
Strandberg E, Tiltak D, Ehni S, Wadhwani P, Ulrich AS Biochemistry 32:5365–5372
(2012) Lipid shape is a key factor for membrane Verardi R, Traaseth NJ, Shi L, Porcelli F, Monfregola L,
interactions of amphipathic helical peptides. Biochim De Luca S, Amodeo P, Veglia G, Scaloni A (2011)
Biophys Acta 1818:1764–1776 Probing membrane topology of the antimicrobial
Strandberg E, Zerweck J, Wadhwani P, Ulrich AS (2013) peptide distinctin by solid-state NMR spectroscopy
Synergistic insertion of antimicrobial magainin-family in zwitterionic and charged lipid bilayers. Biochim
peptides in membranes depends on the lipid spontane- Biophys Acta 1808:34–40
ous curvature. Biophys J 104:L09–L11 Verly RM, Resende JM, Junior EFC, de Magalhães MTC,
Sychev SV, Sukhanov SV, Panteleev PV, Shenkarev Guimarães CFCR, Munhoz VHO, Bemquerer MP,
ZO, Ovchinnikova TV (2018) Marine antimicro- Almeida FCL, Santoro MM, Piló-Veloso D, Bechinger
bial peptide arenicin adopts a monomeric twisted B (2017) Structure and membrane interactions of the
beta-hairpin structure and forms low conductivity homodimeric antibiotic peptide homotarsinin. Sci Rep
pores in zwitterionic lipid bilayers. Pept Science 7:40854
110:e23093 Vermeer LS, Marquette A, Schoup M, Fenard D, Galy A,
Tamba Y, Yamazaki M (2005) Single giant unilamellar Bechinger B (2016) Simultaneous analysis of second-
vesicle method reveals effect of antimicrobial peptide ary structure and light scattering from circular dichro-
magainin 2 on membrane permeability. Biochemistry ism titrations: application to vectofusin-1. Sci Rep
44:15823–15833 6:39450
Tamba Y, Yamazaki M (2009) Magainin 2-induced pore Violette A, Fournel S, Lamour K, Chaloin O, Frisch B,
formation in the lipid membranes depends on its con- Briand JP, Monteil H, Guichard G (2006) Mimicking
centration in the membrane interface. J Phys Chem B helical antibacterial peptides with nonpeptidic folding
113:4846–4852 oligomers. Chem Biol 13:531–538
Tamba Y, Ariyama H, Levadny V, Yamazaki M (2010) Visscher KM, Medeiros-Silva J, Mance D, Rodrigues J,
Kinetic pathway of antimicrobial peptide magainin Daniels M, Bonvin A, Baldus M, Weingarth M (2017)
64 C. Aisenbrey et al.

Supramolecular organization and functional implica- Wieprecht T, Beyermann M, Seelig J (1999b) Binding
tions of K+ channel clusters in membranes. Angew of antibacterial magainin peptides to electrically
Chem Int Ed Eng 56:13222–13227 neutral membranes: thermodynamics and structure.
Vogt TCB, Bechinger B (1999) The interactions of Biochemistry 38:10377–10378
histidine-­containing amphipathic helical peptide anti- Wieprecht T, Apostolov O, Beyermann M, Seelig
biotics with lipid bilayers: the effects of charges and J (2000a) Membrane binding and pore formation of
pH. J Biol Chem 274:29115–29121 the antibacterial peptide PGLa: thermodynamic and
Voievoda N (2014) Biophysical investigations of the mechanistic aspects. Biochemistry 39:442–452
membrane and nucleic acids interactions of the trans- Wieprecht T, Apostolov O, Seelig J (2000b) Binding of the
fection peptide LAH4-L1. PhD thesis, University of antibacterial peptide magainin 2 amide to small and
Strasbourg PhD large unilamellar vesicles. Biophys Chem 85:187–198
Voievoda N, Schulthess T, Bechinger B, Seelig J (2015) Wieprecht T, Beyermann M, Seelig J (2002)
Thermodynamic and biophysical analysis of the Thermodynamics of the coil-alpha-helix transition of
membrane-­ association of a histidine-rich peptide amphipathic peptides in a membrane environment: the
with efficient antimicrobial and transfection activities. role of vesicle curvature. Biophys Chem 96:191–201
J Phys Chem B 119:9678–9687 Wolf J, Aisenbrey C, Harmouche N, Raya J, Bertani P,
Wakamatsu K, Takeda A, Tachi T, Matsuzaki K (2002) Voievoda N, Süss R, Bechinger B (2017) pH-­dependent
Dimer structure of magainin 2 bound to phospholipid membrane interactions of the histidine-­rich cell pen-
vesicles. Biopolymers 64:314–327 etrating peptide LAH4-L1. Biophys J 113:1290–1300
Wallace BA, Moa B (1984) Circular dichroism analyses Wu Y, Huang HW, Olah GA (1990) Method of oriented
of membrane proteins: an examination of differential circular dichroism. Biophys J 57:797–806
light scattering and absorption flattening effects in Wu YS, Ngai SC, Goh BH, Chan KG, Lee LH, Chuah LH
large membrane vesicles and membrane sheets. Anal (2017) Anticancer activities of surfactin and potential
Biochem 142:317–328 application of nanotechnology assisted surfactin deliv-
Wang G, Li X, Wang Z (2016) APD3: the antimicrobial ery. Front Pharmacol 8:761
peptide database as a tool for research and education. Yang D, Zou R, Zhu Y, Liu B, Yao D, Jiang J, Wu J, Tian H
Nucleic Acids Res 44:D1087–D1093 (2014) Magainin II modified polydiacetylene micelles
Watanabe H, Kawano R (2016) Channel current analysis for cancer therapy. Nanoscale 6:14772–14783
for pore-forming properties of an antimicrobial pep- Yang Z, Choi H, Weisshaar JC (2018) Melittin-induced
tide, magainin 1, using the droplet contact method. permeabilization, re-sealing, and re-permeabilization
Anal Sci 32:57–60 of E. coli membranes. Biophys J 114:368–379
Wenk M, Seelig J (1998) Magainin 2 amide interac- Yuksel E, Karakecili A (2014) Antibacterial activity on
tion with lipid membranes: calorimetric detection of electrospun poly(lactide-co-glycolide) based mem-
peptide binding and pore formation. Biochemistry branes via magainin II grafting. Mater Sci Eng C
37:3909–3916 Mater Biol Appl 45:510–518
Wenzel M, Chiriac AI, Otto A, Zweytick D, May C, Zanin LM, Dos Santos Alvares D, Juliano MA, Pazin WM,
Schumacher C, Gust R, Albada HB, Penkova M, Ito AS, Ruggiero Neto J (2013) Interaction of a syn-
Kramer U, Erdmann R, Metzler-Nolte N, Straus SK, thetic antimicrobial peptide with model membrane by
Bremer E, Becher D, Brotz-Oesterhelt H, Sahl HG, fluorescence spectroscopy. Eur Biophys J 42:819–831
Bandow JE (2014) Small cationic antimicrobial pep- Zasloff M (1987) Magainins, a class of antimicrobial pep-
tides delocalize peripheral membrane proteins. Proc tides from Xenopus skin: isolation, characterization
Natl Acad Sci USA 111:E1409–E1418 of two active forms, and partial cDNA sequence of a
Westerhoff HV, Zasloff M, Rosner JL, Hendler RW, de precursor. Proc Natl Acad Sci U S A 84:5449–5453
Waal A, Vaz G, Jongsma PM, Riethorst A, Juretic D Zasloff M (2002) Antimicrobial peptides of multicellular
(1995) Functional synergism of the magainins PGLa organisms. Nature 415:389–395
and magainin-2 in Escherichia coli, tumor cells and Zerweck J, Strandberg E, Kukharenko O, Reichert J,
liposomes. Eur J Biochem 228:257–264 Burck J, Wadhwani P, Ulrich AS (2017) Molecular
Wheaten SA, Ablan FD, Spaller BL, Trieu JM, Almeida mechanism of synergy between the antimicrobial pep-
PF (2013) Translocation of cationic amphipathic pep- tides PGLa and magainin 2. Sci Rep 7:13153
tides across the membranes of pure phospholipid giant Zhao H, Shao D, Jiang C, Shi J, Li Q, Huang Q, Rajoka
vesicles. J Am Chem Soc 135:16517–16525 MSR, Yang H, Jin M (2017) Biological activity of
Wieprecht T, Apostolov O, Beyermann M, Seelig ­lipopeptides from Bacillus. Appl Microbiol Biotechnol
J (1999a) Thermodynamics of the alpha-helix-coil 101:5951–5960
transition of amphipathic peptides in a membrane Zhao P, Xue Y, Gao W, Li J, Zu X, Fu D, Bai X, Zuo Y, Hu
environment: implications for the peptide-membrane Z, Zhang F (2018) Bacillaceae-derived peptide antibi-
binding equilibrium. J Mol Biol 294:785–794 otics since 2000. Peptides 101:10–16
 Anionic Lipid Clustering Model
5
Richard M. Epand

Abstract Keywords
Many molecular features contribute to the anti- Antimicrobial peptides · Anionic lipid
microbial activity of peptides. One aspect that clustering · Bacterial membrane lipids ·
contributes to the antimicrobial activity of a Cardiolipin · Phosphatidylethanolamine ·
peptide, in many cases, results from the fact that Lipid domain formation
many antimicrobial peptides are polycationic
and the lipids on the surface of bacteria are often
anionic. In certain cases this can result in the
clustering of anionic lipids as a result of the 5.1 Description of Model
binding of the cationic peptide to the surface of
the bacterial membrane. This lipid clustering Gram-negative bacteria have an outer membrane
can be detrimental to the viability of the bacteria that is negatively charged, but most antimicrobial
to which the peptide binds. Several factors, agents have bactericidal or bacteriostatic action
including the charge, size, and conformational as a result of interactions with the cell membrane,
flexibility of the peptide, will determine the effi- i.e., the inner membrane of Gram-negative bacte-
ciency of lipid clustering. In addition, the lipid ria and the cell membrane of Gram-positive bac-
composition of the bacterial membrane is very teria. In both cases these membranes are rich in
variable, and it plays a critical role in this mech- anionic lipids phosphatidylglycerol and cardio-
anism. As a result, one can test the importance lipin. Antimicrobial peptides have a large range
of this factor by determining the species speci- of structures, but the majority of these peptides
ficity of the antimicrobial activity of the peptide. have several positive charges. This would allow
The molecular mechanism by which lipid clus- each peptide to bind to several lipids and cluster
tering affects bacterial viability is uncertain in them in the same location in the membrane. For
many cases. This phenomenon can be used to membranes that also contain zwitterionic or
increase the antimicrobial potency of peptides uncharged lipids, this would produce an area in
in some case and can also predict the bacterial the membrane that is depleted of anionic lipids. A
species specificity of some agents. schematic model of this phenomenon is shown in
Fig. 5.1.
R. M. Epand (*)
Department of Biochemistry and Biomedical
Sciences, McMaster University,
Hamilton, ON, Canada
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 65

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_5
66 R. M. Epand

Fig. 5.1 Clustering of anionic lipid in a bilayer com- causes defects in the surrounding area of the bilayer,
posed of anionic (red headgroups) and zwitterionic lipids increasing permeability (Figure taken from Epand et al.
(green headgroups) in the presence of a cationic agent 2011)

Fig. 5.2 Maximum temperature shift of the main transi- The short peptides BP100 and HIV-TAT fall out of line.
tion obtained by DSC in the presence of peptide, with The linear relationship holds for the antimicrobial peptides
respect to the lipid mixture in the absence of peptide, Δλ, (solid squares) as well as for the cell-penetrating peptides
plotted as a function of the peptide’s net positive charge. (open squares) (Taken from Wadhwani et al. 2012)

5.2 Observations Supporting Fig. 5.2, the majority of the peptides followed a
Model simple relationship in which the clustering of
anionic lipid was proportional to the number of
Several experimental observations support this net positive charges on the peptide.
simple model.

5.2.2 Bacterial Species Specificity

5.2.1 Potency Related to Charge
Our model requires that anionic lipids are clus-
There are several examples showing a correlation tered away from the remaining lipids in a mem-
between overall charge and the ability to cluster brane, i.e., uncharged or zwitterionic lipids. This
anionic lipids. I draw attention to a particular cannot occur with bacteria that have little
study that compared 16 different antimicrobial uncharged or zwitterionic lipid. Hence certain
and cell-penetrating peptides. Anionic lipid clus- bacterial species will be more susceptible to
tering was evaluated by differential scanning anionic lipid clustering. This was first demon-
calorimetry (Wadhwani et al. 2012). As seen in strated in our study of a synthetic o­ ligo-acyl-­lysine
5 Anionic Lipid Clustering Model 67

(OAK) (Epand et al. 2008b). In this work it was tion of factors. For a subclass of antimicrobial
demonstrated that most Gram-negative bacteria agents, the clustering of anionic lipids provides
had high mole fractions of phosphatidylethanol- an important factor contributing to their potency.
amine, a zwitterionic lipid, in their membranes. Whether a particular antimicrobial agent falls
Hence they could cluster anionic lipid in a back- into this subclass can be tested experimentally by
ground of zwitterionic lipid, and therefore they determining if the bacterial species that are sus-
exhibited a high sensitivity (low MIC) to this ceptible to the agent have surface membranes
OAK (Epand et al. 2008b). In contrast, many spe- that contain both anionic and either zwitterionic
cies of Gram-positive bacteria were largely or uncharged lipids, while bacteria that are com-
devoid of phosphatidylethanolamine and required prised mainly of anionic lipids are resistant to
tenfold or higher concentrations of OAK to these agents. Another test can be performed using
inhibit cell growth. An exception was bacteria of NMR or DSC to test if the agent induces cluster-
the genus Bacillus that in many cases had high ing of anionic lipid, as we have described (Epand
concentrations of phosphatidylethanolamine and and Epand 2009).
exhibited sensitivity to OAKs comparable to that It should also be taken into account that there
of Gram-negative bacteria (Epand et al. 2008b). are limitations in the data for the lipid composi-
The phenomenon of anionic lipid clustering by tion of bacterial membranes. The lipid composi-
OAK in lipid mixtures of anionic cardiolipin and tion will be modified by growth conditions
zwitterionic phosphatidylethanolamine was dem- including nutrients, salts and pH, stage of the
onstrated by differential scanning calorimetry growth cycle, nature of the acyl chains that are
and by 31P-MAS/NMR (Epand et al. 2008b). For incorporated into lipids, and other factors.
another family of synthetic antimicrobial com-
pounds, the ceragenins, it was also found that
their toxicity to bacteria was related to the con- 5.3.1 Conformational Flexibility
tent of phosphatidylethanolamine in the bacteria.
In this study, not only were different strains of In addition to charge, conformational flexibility
bacteria utilized, but the loss of susceptibility to improved antimicrobial potency, possibly allow-
this cationic antimicrobial was demonstrated ing the antimicrobial agent to match the distance
using a mutant form of E. coli that could not syn- between positive charge to the distance between
thesize phosphatidylethanolamine (Epand et al. negatively charged lipids on the membrane
2007). In wild-type strains of E. coli, phosphati- (Epand and Epand 2009). One example of this is
dylethanolamine comprises 80% of the total a pair of α-/β-peptides having different charge
lipid. However, phosphatidylethanolamine is not distributions and different conformation flexibili-
the only bacterial lipid with no overall charge. ties, but with identical chemical composition.
Some bacteria contain a significant amount of The more flexible peptide had a higher activity
uncharged glycosyl diglycerides. One example is against E. coli (Schmitt et al. 2004). In addition,
S. pyogenes. We have found that S. pyogenes has a sequence-random cationic polymer that was
a MIC that is comparable to Bacilli that have a designed to be unstructured in solution was
high phosphatidylethanolamine mole fraction shown to cluster anionic lipids (Epand et al.
(Epand et al. 2010). 2008a), indicating that a specific conformation of
the antimicrobial agent was not required.

5.3 Limitation of Model

5.3.2 Other Factors
The proposed charge clustering model is not
intended to be a universal model that applies to 5.3.2.1 Hydrophobicity
all antimicrobial agents. Generally antimicrobial Since the site of action of antimicrobial agents
agents are toxic to bacteria because of a combina- that cluster anionic lipids is the membrane, the
68 R. M. Epand

antimicrobial agent has to be sufficiently hydro- 5.3.2.4 Pore Formation

phobic to partition into a membrane. Partitioning The charge clustering model is one of several mod-
of cationic polymers onto the surface of anionic els to describe the action of antimicrobial agents.
bacterial membranes is expected. However, the Perhaps the most studied model is that of the mem-
strength of electrostatic binding alone and lack of brane pore (Imura et al. 2008) and variations of that
significant penetration of these peptides into a model (Shai 2002). There are several observations
membrane make such polymers weakly antimi- that support these models which do not assume
crobial. An example is polycationic amino acids anything about the rearrangement of membrane
that have antimicrobial activity in the approxi- components. Hence the lipid composition of the
mate range of 50–250 μg/mL (Venkatesh et al. bacteria will not correlate with the potency of the
2017). Adding acyl groups to a polymer of Lys antimicrobial that acts through a pore mechanism.
can increase the membrane partitioning and the In general, there does not have to be a choice
antimicrobial potency (Mor 2016). among different proposed mechanisms of antimi-
crobial action. In most cases the potency of a par-
5.3.2.2 Specific Binding to Component ticular agent will be determined by several
Of course hydrophobicity is only one factor that properties, some of which will be interrelated and
will determine the partitioning of a compound to others not. An attractive feature of a pore model is
a membrane (Andreev et al. 2018). One addi- that it directly provides a mechanism for toxicity to
tional mechanism is the binding of an antimicro- bacteria through depolarization of the membrane
bial molecule to a specific membrane component electrical potential as well as loss of other trans-
(Epand et al. 2016; Phoenix et al. 2015; Guder membrane gradients of ions and small molecules.
et al. 2000; Schmitt and Rosa 2016). This prop-
erty is often exhibited by cyclical compounds 5.3.2.5 Phospholipid Acyl Chains
that have a more restricted conformation. The acyl chains of phospholipids of bacteria differ
from those of mammalian systems. In particular
5.3.2.3 Access to Membrane there is very little unsaturation in the acyl chains of
The charge cluster model depends on the inter- bacterial lipids and no polyunsaturated chains. In
action of the antimicrobial substance with the addition, bacterial lipid acyl chains contain single
phospholipids of the cell membrane of the bac- double bonds in the trans configuration, which is
teria. For Gram-positive bacteria, the cell mem- virtually nonexistent in higher organisms. Bacterial
brane is surrounded by a cell wall. This cell lipids are also unique in having lipid acyl chains
wall is generally porous to small molecules, with cyclopropyl groups as well as terminal methyl
although it does contain some negative charge. branching. In methicillin-bacterial lipids are also
There is one example in which the toxicity of unique in containing lipid acyl chains with cyclo-
an antimicrobial results from it become propyl-containing as well as terminally methyl-
entrapped in the cell wall and blocking branched acyl chains. In methicillin-resistant
exchange of materials across the cell membrane strains of E. coli, the fraction of anteiso-branched
(Epand et al. 2008a). Generally Gram-negative acyl chains is elevated relative to susceptible
bacteria are more resistant to antimicrobial strains (Mitchell et al. 2016). Also, the action of
agents because they present an additional bar- the cytolytic peptide, δ-lysin, strongly depends on
rier of the outer membrane to protect the cell the structure of acyl chains in microbial lipids
membrane from access by antimicrobial agents. (Pokorny et al. 2008). Model liposome studies
A major component of the outer membrane of have shown that anteiso-­branched acyl chains in
Gram-negative bacteria is lipopolysaccharide phospholipids make liposomes more susceptible
that is anionic and can potentially sequester to lysis by antimicrobial peptides and these lipids
cationic drugs. have lower gel-to-liquid crystalline phase transi-
The question of antibiotic resistance is an tion temperatures (Mitchell et al. 2016).
important separate issue. There are several syn- Not only is membrane permeability an impor-
thetic agents that have been shown to reverse tant factor for bacterial toxicity, but it is also rel-
microbial resistance (Molchanova et al. 2017). evant for the access of antimicrobial agents that
5 Anionic Lipid Clustering Model 69

have intracellular targets that require that they by which anionic lipid clustering can be detri-
pass through the cell membrane to access these mental to bacteria.
targets (Chopra 1988). An example of an antimi-
crobial that has an intracellular ribosomal target
is apidaecin 1b (Schmidt et al. 2018). Bacterial 5.4.1 Phase Boundary Defect
lipidomic analysis showed that strains of E. coli
resistant to apidaecin 1b had lower levels of acyl A possible mechanism by which antimicrobial
chains with cyclopropyl groups. Furthermore, compounds that function by charge clustering
knocking out the enzyme responsible for the bio- can affect bacterial viability is by forming phase
synthesis of the cyclopropyl group also resulted boundary defects between the clustered anionic
in greater resistance to apidaecin. lipids and the remainder of the membrane (Jean-­
The charge clustering model considers only Francois et al. 2008). However, it is not clear if
the interaction of the antimicrobial agent with the the formation of these new lipid domains would
headgroups of the phospholipids. Clearly there greatly increase the membrane defects on the
are also other factors affecting sensitivity to anti- bacterial cell surface. There is evidence that bac-
microbial agents, and one of these is the nature of terial membranes naturally possess domains
the acyl chains, particularly the structures of the (Nickels et al. 2017) and hence have phase
acyl chains that are unique to bacteria. One such boundary defects before the addition of any anti-
structure is the trans carbon-carbon double bond. microbial compound. The role of phase boundary
To our knowledge the role of this chemical group- defects, in comparison with other membrane
ing to bacterial resistance has not been evaluated. defects, as a cause of membrane leakage or insta-
bility is difficult to assess quantitatively.
5.3.2.6 Non-membrane Targeting
There is evidence that the bacterial membrane is
the site of action of many antimicrobial agents. 5.4.2  epletion of Anionic Lipids
D
However, there are also antimicrobials that have from Regions
intracellular targets. The membrane still plays a of the Membrane
role as a barrier of the drug to the cell interior,
and the lipid cluster mechanism can play a role in If an antimicrobial agent clusters anionic lipids
this access. However, the potency of such an anti- into a smaller zone in the membrane, of necessity,
microbial will also depend on the affinity and it will deplete other regions of the membrane
specificity it has for its intracellular target. This from this lipid class. Lipids can modulate the
will likely have a larger effect on the potency of activity of membrane-bound proteins either by
the antimicrobial and be completely independent modulating the properties of the environment
of actions at the membrane. around the protein in a solvent-like effect or by
binding to specific sites on the protein (Salas-­
Estrada et al. 2017; Lee 2003; Yeagle et al. 1988).
5.4 Molecular Mechanism Both of these mechanisms for changing the
behavior of a membrane protein will be affected
We described above the charge cluster model and by the clustering of anionic lipid. This will not
how certain antimicrobial agents could rearrange happen with many Gram-positive bacteria whose
bacterial membrane components so as to cluster membrane surface is composed largely of anionic
anionic lipids. However, we did not discuss how lipid; however, with bacteria having in addition to
this clustering of anionic lipid could result in anionic lipid a high fraction of zwitterionic or
antibacterial action. This is generally the situa- uncharged lipid, there will be an area depleted of
tion for most antimicrobial agents that are not anionic lipids. Anionic lipids often play important
specific for a particular mechanism and often roles in membrane function and/or in binding to
have contributing factors from several sources. membrane proteins. These roles cannot be effi-
We will consider some of the likely mechanisms ciently fulfilled if the anionic lipids are clustered
70 R. M. Epand

together with an antimicrobial agent. This mecha- 5.5 Summary

nism appears likely to contribute to the mecha-
nism of action of these anionic lipid clustering The charge cluster model is one of a number of
agents, but to prove it as a mechanism requires mechanisms that can provide antimicrobial activ-
knowledge of the specific roles of anionic lipids in ity to a substance. One criterion to test the impor-
the membrane and how these functions are lost as tance of this factor is to determine the dependence
a consequence of lipid clustering. of antimicrobial potency on the nature of the lipid
composition of bacterial membranes. Often the
only criterion taken into account in evaluating the
5.4.3  hase Separation Resulting
P potency of antimicrobial agents is to compare
in Redistribution Gram-positive with Gram-negative bacteria.
of Membrane Proteins However, as illustrated in this review and refer-
ences therein, bacteria vary widely in terms of the
Recent studies of the mechanism of action of a nature of their lipid composition. One important
synthetic cyclical hexapeptide, cWFW, having example being phosphatidylethanolamine that is
the structure cyclo(RRRWFW), has bactericidal the most abundant lipid in the membranes of
activity as a result of membrane domain forma- some bacteria but is present in only trace amounts
tion (Scheinpflug et al. 2017). Although this pep- in other bacteria. Phosphatidylethanolamine and
tide has been shown to cluster anionic lipids in other zwitterionic or uncharged lipids are not
model membranes (Arouri et al. 2009; Finger clustered by polycationic antimicrobial agents,
et al. 2015), the mechanism of cell death in vivo but they are required to facilitate the clustering of
is caused by redistribution of proteins in the bac- anionic lipids. Thus, discovering agents that
terial membrane caused by domain formation function primarily by the charge cluster mecha-
(Scheinpflug et al. 2017). It was found that nism will allow the design of antimicrobials that
cWFW does not depolarize the membrane or are most suitable for a particular bacterial target.
affect the energy state of the cell, but it does
cause a large-scale phase separation of the mem-
brane. It was suggested that the cause of this References
phase separation was a result of changes in mem-
brane fluidity, rather than clustering of anionic Andreev K, Martynowycz MW, Huang ML, Kuzmenko
I, Bu W, Kirshenbaum K, Gidalevitz D (2018)
lipids. However, domains in biological mem- Hydrophobic interactions modulate antimicrobial
branes are often smaller than they are in model peptoid selectivity towards anionic lipid membranes.
membranes and may therefore be more difficult Biochim Biophys Acta Biomembr 1860:1414–1423
to detect. However, it was shown that in bacteria Arouri A, Dathe M, Blume A (2009) Peptide induced
demixing in PG/PE lipid mixtures: a mechanism for the
the antimicrobial activity was not a consequence specificity of antimicrobial peptides towards bacterial
of permeation of the membrane but rather a trig- membranes? Biochim Biophys Acta 1788:650–659
gering of a stress response resulting in an inhibi- Chopra I (1988) Molecular mechanisms involved in the
tion of lipid II and cell wall synthesis as well as transport of antibiotics into bacteria. Parasitology
96:S25–S44
through increased autolysis triggered by cWFW Epand RM, Epand RF (2009) Lipid domains in bacterial
(Scheinpflug et al. 2017). Future studies will membranes and the action of antimicrobial agents.
determine how general this mechanism is appli- Biochim Biophys Acta Biomembr 1788:289–294
cable to other antimicrobial peptides. Epand RF, Savage PB, Epand RM (2007) Bacterial lipid
composition and the antimicrobial efficacy of cationic
steroid compounds (Ceragenins). Biochim Biophys
Acta 1768:2500–2509
5 Anionic Lipid Clustering Model 71

Epand RF, Mowery BP, Lee SE, Stahl SS, Lehrer RI, Host defense peptides and their potential as therapeu-
Gellman SH, Epand RM (2008a) Dual mechanism tic agents. Springer, Switzerland, pp 205–226
of bacterial lethality for a cationic sequence-random Nickels JD, Chatterjee S, Stanley CB, Qian S, Cheng
copolymer that mimics host-defense antimicrobial X, Myles DAA, Standaert RF, Elkins JG, Katsaras
peptides. J Mol Biol 379:38–50 J (2017) The in vivo structure of biological mem-
Epand RM, Rotem S, Mor A, Berno B, Epand RF (2008b) branes and evidence for lipid domains. PLoS Biol
Bacterial membranes as predictors of antimicrobial 15:e2002214
potency. J Am Chem Soc 130:14346–14352 Phoenix DA, Harris F, Mura M, Dennison SR (2015) The
Epand RM, Epand RF, Arnusch CJ, Papahadjopoulos-­ increasing role of phosphatidylethanolamine as a lipid
Sternberg B, Wang G, Shai Y (2010) Lipid cluster- receptor in the action of host defence peptides. Prog
ing by three homologous arginine-rich antimicrobial Lipid Res 59:26–37
peptides is insensitive to amino acid arrangement and Pokorny A, Kilelee EM, Wu D, Almeida PFF (2008) The
induced secondary structure. Biochim Biophys Acta activity of the amphipathic peptide δ-lysin correlates
1798:1272–1280 with phospholipid acyl chain structure and bilayer
Epand RF, Mor A, Epand RM (2011) Lipid complexes elastic properties. Biophys J 95:4748–4755
with cationic peptides and OAKs; their role in anti- Salas-Estrada LA, Leioatts N, Romo TD, Grossfield A
microbial action and in the delivery of antimicrobial (2017) Lipids alter rhodopsin function via ligand-like
agents. Cell Mol Life Sci 68:2177–2188 and solvent-like interactions. Biophys J 114:355–367
Epand RM, Walker C, Epand RF, Magarvey NA (2016) Scheinpflug K, Wenzel M, Krylova O, Bandow JE, Dathe
Molecular mechanisms of membrane targeting antibi- M, Strahl H (2017) Antimicrobial peptide cWFW kills
otics. Biochim Biophys Acta Biomembr 1858:980–987 by combining lipid phase separation with autolysis.
Finger S, Kerth A, Dathe M, Blume A (2015) The efficacy Sci Rep 7:44332
of trivalent cyclic hexapeptides to induce lipid clus- Schmidt R, Yonghong D, Hoffmann R (2018)
tering in PG/PE membranes correlates with their anti- Phospholipid composition of the outer membrane of
microbial activity. Biochim Biophys Acta Biomembr Escherichia coli influences its susceptibility against
1848:2998–3006 antimicrobial peptide apidaecin 1b. Diagn Microbiol
Guder A, Wiedemann I, Sahl HG (2000) Infect Dis 90(4):316–323
Posttranslationally modified bacteriocins – the lanti- Schmitt MA, Weisblum B, Gellman SH (2004) Unexpected
biotics. Biopolym Pept Sci Sect 55:62–73 relationships between structure and function in α,β-­
Imura Y, Choda N, Matsuzaki K (2008) Magainin 2 in peptides: antimicrobial foldamers with heterogeneous
action: distinct modes of membrane permeabiliza- backbones. J Am Chem Soc 126:6848–6849
tion in living bacterial and mammalian cells. Biophys Schmitt P, Rosa RD, Destoumieux-Garzón D (2016) An
J 95:5757–5765 intimate link between antimicrobial peptide sequence
Jean-Francois F, Castano S, Desbat B, Odaert B, Roux M, diversity and binding to essential components of bac-
Metz-Boutigue MH, Dufourc EJ (2008) Aggregation terial membranes. Biochim Biophys Acta Biomembr
of cateslytin β-sheets on negatively charged lipids 1858:958–970
promotes rigid membrane domains. A new mode Shai Y (2002) Mode of action of membrane active antimi-
of action for antimicrobial peptides? Biochemistry crobial peptides. Biopolym Pept Sci Sect 66:236–248
47:6394–6402 Venkatesh M, Barathi VA, Goh ETL, Anggara R, Fazil
Lee AG (2003) Lipid-protein interactions in biologi- MHUT, Ng AJY, Harini S, Aung TT, Fox SJ, Liu S,
cal membranes: a structural perspective. Biochim Yang L, Barkham TMS, Loh XJ, Verma NK, Beuerman
Biophys Acta Biomembr 1612:1–40 RW, Lakshminarayanan R (2017) Antimicrobial activ-
Mitchell NJ, Seaton P, Pokorny A (2016) Branched phos- ity and cell selectivity of synthetic and biosynthetic
pholipids render lipid vesicles more susceptible to cationic polymers. Antimicrob Agents Chemother
membrane-active peptides. Biochim Biophys Acta 61:e00469–e00417
Biomembr 1858:988–994 Wadhwani P, Epand RF, Heidenreich N, Burck J, Ulrich
Molchanova N, Hansen PR, Franzyk H (2017) Advances AS, Epand RM (2012) Membrane-active peptides and
in development of antimicrobial peptidomimetics as the clustering of anionic lipids. Biophys J 103:265–274
potential drugs. Molecules 22(9):E1430 Yeagle PL, Young J, Rice D (1988) Effects of cholesterol
Mor A (2016) Engineered OAKs against antibiotic resis- on sodium-potassium ATPase ATP hydrolyzing activ-
tance and for bacterial detection. In: Epand RM (ed) ity in bovine kidney. Biochemistry 27:6449–6452
 Intracellular Antimicrobial
Peptides Targeting the Protein 6
Synthesis Machinery

Michael Graf and Daniel N. Wilson

Abstract PrAMPs. Specifically, class I PrAMPs, such

While antimicrobial peptides (AMPs) are as Bac7, Onc112, pyrrhocoricin, and metal-
well-known for their disruptive effects on bac- nikowin, block the delivery of aa-tRNA by
terial membranes, the mechanism of many EF-Tu to the ribosomal A-site, whereas the
intracellular AMPs is still being elucidated. In class II PrAMPs, such as apidaecin 1b and
the recent years, it has been demonstrated that Api137, act during translation termination and
the subclass of proline-rich AMPs (PrAMPs) inhibit protein synthesis by trapping of release
can pass through the bacterial membrane and factors on the 70S ribosome following hydro-
kill bacteria by inhibiting protein synthesis. lysis of the nascent polypeptide chain.
PrAMPs are a product of the innate immune
system and are secreted in response to bacte- Keywords
rial infection. So far PrAMPs have been iden- Proline-rich antimicrobial peptides · Protein
tified in many arthropods, such as beetles, synthesis · Translation · Ribosome
wasps, and flies, as well as some mammals,
such as sheep, cows, and goats. PrAMPs show
high potency against Gram-negative bacteria,
while exhibiting low toxicity in eukaryotes, 6.1 Diversity of Antimicrobial
suggesting that they may represent a promis- Peptide Targets
ing avenue for the development of future anti-
microbial agents to combat the increase of While AMPs are well-known for their action to
multidrug-resistant bacterial pathogens. kill bacteria via membrane disruption, there is an
Structural and biochemical data have revealed increasing number of AMPs that have been char-
the PrAMP binding sites on the ribosome as acterized to pass through the bacterial cell mem-
well as insight into their mechanisms of brane and target intracellular processes (Brogden
action. While the binding site of all so far 2005). Intracellular targets encompass many
investigated PrAMPs is situated within fundamental processes for the bacterial propaga-
nascent polypeptide exit tunnel, the mecha- tion, such as DNA and RNA replication, mRNA
nism of action is distinct between class I and II transcription, and protein synthesis (Brogden
2005; Graf et al. 2017). The intrinsic diversity of
M. Graf · D. N. Wilson (*) peptides as well as their ease of synthesis, in
Institute for Biochemistry and Molecular Biology, combination with the urgent need for novel com-
University of Hamburg, Hamburg, Germany pounds to fight bacterial infections, makes AMPs
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 73

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_6
74 M. Graf and D. N. Wilson

a v­ aluable and promising source for future anti- 2018; Schnapp et al. 1996; Schneider and Dorn
biotics to be employed within the clinical 2001; Shamova et al. 1999; Stensvag et al. 2008).
setting. Among mammals, PrAMPs have been identified
in cows (Gennaro et al. 1989), pigs (Agerberth
et al. 1991), sheep (Huttner et al. 1998; Shamova
6.1.1 Sources of Proline-Rich et al. 1999), goats (Shamova et al. 1999), and
Antimicrobial Peptides more recently dolphins (Mardirossian et al. 2018)
(Fig. 6.1) but so far not in primates or humans.
The proline-rich AMPs (PrAMPs) are a subclass Arthropod PrAMPs have been found in crabs
of cationic peptides, which as their name sug- (Stensvag et al. 2008) and numerous insects,
gests are characterized by a high content of pro- including wasps, bees, flies, and beetles (Fig. 6.1)
line, as well as arginine residues (Graf et al. (Bulet et al. 1993; Casteels et al. 1989, 1990;
2017). The first PrAMP was identified in the late Chernysh et al. 1996; Cociancich et al. 1994;
1980s by HPLC, namely, apidaecin, from the Knappe et al. 2010; Schneider and Dorn 2001).
honey bee Apis mellifera (Casteels et al. 1989), Analogously to apidaecin (Casteels et al. 1989),
which was followed by the identification of many the nomenclature of PrAMPs predominantly cor-
other PrAMPs from higher eukaryotes, including responds to the origin of identification.
some mammals and many arthropods (Fig. 6.1) Accordingly, PrAMPs from Tursiops truncatus
(Agerberth et al. 1991; Bulet et al. 1993; Casteels (bottlenose dolphin) are termed Tur1A and Tur1B
et al. 1990; Chernysh et al. 1996; Cociancich (Mardirossian et al. 2018), peptides from Bos
et al. 1994; Gennaro et al. 1989; Huttner et al. taurus (cattle) are called bactenecins (Bac)
1998; Knappe et al. 2010; Mardirossian et al. (Gennaro et al. 1989), and Oncopeltus fasciatus

Fig. 6.1 Sequences of natural and synthetic PrAMPs respectively. The O-glycosylation of drosocin is indicated
derived from arthropods (insects and crustaceans, salmon) in blue (Thr11). Position 11 of oncocin is unknown and
and mammals (purple). The central PrAMPs were aligned indicated with a bold letter “X.” The number of amino
based on ribosome-bound structures of Onc112, Pyr, Met, acids (aa) representing the mature peptide and the origin
Tur1A, and Bac7 and then on sequence similarity. Similar (organism) is stated on the right (Figure adapted from
and identical residues are shown in gray and black, Graf et al. 2017)
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 75

(milkweed bug)-derived PrAMPs are termed peptide, and second, it allows for a fast pathogen
oncocins (Knappe et al. 2010; Schneider and response by avoiding time-consuming protein
Dorn 2001). Noteworthy, peptides with high synthesis. Activation of the inactive precursors
sequence similarity are commonly named after occurs at the target site by proteolytic cleavage
the source in which the PrAMP was first discov- (Zanetti et al. 1990, 1991). The respective prote-
ered. Thus, PrAMPs from Capra hircus (goat) ases are stored separately from the inactive pre-
and Ovis aries (sheep) (Huttner et al. 1998; cursors in a different set of granules (Fig. 6.2). In
Shamova et al. 1999), which show similarity to mammals, PrAMP-containing granules are called
Bos taurus peptides, are also referred to as large granules, and the protease-harboring gran-
bactenecins. ules are termed azurophil granules (Fig. 6.2).
Activation of precursors is induced by simultane-
ous exocytosis into extracellular space or
6.1.2 Synthesis of PrAMPs bacteria-­containing phagosomes (Fig. 6.2).
However, the overall structure of PrAMP precur-
PrAMPs are primarily produced upon bacterial sors, as well as the resulting path of activation,
infection by cells of the innate immune system, differs between organisms and PrAMPs (Graf
called phagocytes (Zasloff 2002). Phagocyte pro- et al. 2017). Although not all mechanisms of pep-
genitors, immature myeloid cells in mammals, tide activation have been clarified, a single-­
synthesize PrAMPs as inactive precursors and peptide activation mechanism can be
store these in granules (Fig. 6.2) (Graf et al. 2017; distinguished from a multi-peptide activation
Zanetti et al. 1990, 1991). Storage of PrAMPs as mechanism (Graf et al. 2017). All mammalian
inactive precursors in granules serves different PrAMPs seem to be activated by the single-­
functions. First, it potentially protects the ribo- peptide mechanism, whereas insect PrAMPs are
somes of the producing cell from harm by the activated through a multi-peptide activation

Fig. 6.2 Mammalian PrAMPs are synthesized as pre-/ membrane or the phagosome. Elastase activates the
pro-sequences and targeted to large granules. The PrAMPs mature PrAMP by removal of the pro-sequence. The acti-
are activated upon bacterial infection by fusion of pro-­ vated PrAMP is transported via SbmA (to less extent by
PrAMP containing large granules with the elastase-­ MdtM) into the bacterial cell (Figure adapted from Graf
containing azurophil granules and either the plasma et al. 2017)
76 M. Graf and D. N. Wilson

mechanism, as described for apidaecin (from PrAMP uptake, namely, SbmA (Mattiuzzo et al.
Apis mellifera) (Casteels-Josson et al. 1993) and 2007) and more recently MdtM (Fig. 6.2)
riptocin (from Riptortus pedestris) (Graf et al. (Krizsan et al. 2015). While SbmA seems to be
2017). The single-peptide mechanism involves the major transporter for PrAMP uptake, the
an mRNA coding for a polypeptide sequence har- inner membrane protein MdtM appears to also
boring a pre-sequence (~20 to 30 aa), a pro-­ play an auxiliary role in antimicrobial peptide
sequence (~20 to 100 aa), and a Pro-Arg-rich internalization (Krizsan et al. 2015), especially
sequence (Bulet et al. 1993; Graf et al. 2017; at higher peptide concentrations. SbmA is a
Storici and Zanetti 1993; Zanetti et al. 1993). The 46.5 kDa protein localized in the inner mem-
pro-sequence is placed either before (e.g., bacte- brane of Gram-negative bacteria (Mattiuzzo
necin and PR-39) (Storici and Zanetti 1993; et al. 2007) that internalizes PrAMPs, as well as
Zanetti et al. 1993) or after the PrAMP (e.g., dro- other antimicrobial agents, using an electro-
socin) (Bulet et al. 1993). The multi-peptide chemical proton gradient (Runti et al. 2013).
mechanism comprises an mRNA coding for mul- Besides peptide uptake, a general function for
tiple PrAMPs that are interrupted by an inactivat- SbmA remains unclear. Among Gram-­negative
ing oligopeptide linker, such as “RR-EAEPEAEP” bacteria, homologs of SbmA can be found in
from Apis mellifera (Casteels et al. 1989; Enterobacteriaceae, such as Escherichia and
Casteels-Josson et al. 1993). Proteolytic cleavage Salmonella species, and Pseudomonadales,
sites at the N- and C-terminus of the linker liber- such as Acinetobacter baumannii, as well as in
ate the peptide (Casteels et al. 1989; Casteels-­ α- and ε-proteobacteria, such as Neisseria men-
Josson et al. 1993). Similar to single ingitis and Campylobacter species, respectively
PrAMP-encoding mRNAs, the multi-coding (Graf et al. 2017).
messages harbor a pre- (~15 to 20 aa) and pro-­
sequence (~13 to 16 aa) at the very beginning of
the polypeptide (Casteels et al. 1989; Casteels-­ 6.1.4 Discovery of the Intracellular
Josson et al. 1993). Noteworthy, the amino acid Target of PrAMPs
sequences of the PrAMPs encoded by the multi-­
peptide messages can vary significantly. In accor- Biochemical studies within the last 50 years
dance, the multi-peptide activation mechanism raised some controversy as to the primary intra-
leads to the liberation of different isoforms of the cellular target of PrAMPs. Although translation
PrAMP. was suggested as the putative target for PrAMP
inhibition at first (Castle et al. 1999), subsequent
co-immunoprecipitation experiments contra-
6.1.3 Uptake Pathways of PrAMPs dicted this theory by identifying chaperone-­
assisted folding as the primary target of PrAMPs
The vast majority of AMPs utilized by the innate (Otvos et al. 2000). In this study, biotinylated
immune system target the bacterial membrane derivatives of pyrrhocoricin (Pyr), drosocin
and thereby facilitate lysis of the pathogenic cell (Dro), and apidaecin (Api) analogs were purified
(Brogden 2005). This contrasts with most with anti-biotin antibodies and observed to co-­
PrAMPs that were shown to inhibit bacterial purify with the chaperone DnaK (Otvos et al.
growth at low concentrations using a non-lytic 2000). In a similar fashion to Pyr, Dro, and Api,
mechanism of action (Casteels and Tempst coupling of the N-terminal 35 amino acids of
1994; Castle et al. 1999; Otvos 2002; Scocchi bactnecin-7 (Bac7) to 2-chlorotrityl resins also
et al. 2011). The utilization of a non-lytic mech- resulted in co-purification of DnaK (Scocchi
anism initially raised the question as to which et al. 2009). The direct visualization of PrAMP
uptake pathways are used by PrAMPs. binding to DnaK was achieved using X-ray crys-
Subsequent mutagenesis experiments led to the tallography, where PrAMPs were observed to
identification of two key players responsible for bind to the substrate cleft of DnaK (Knappe et al.
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 77

2011; Zahn et al. 2013, 2014). However, soon tide exit tunnel in an extended conformation,
after structures of PrAMPs in complex with which overlaps with the path of a nascent chain
DnaK had been obtained, questions were raised (Florin et al. 2017; Gagnon et al. 2016;
about the validity of the working model since Mardirossian et al. 2018; Roy et al. 2015; Seefeldt
minimal inhibitory concentration (MIC) experi- et al. 2015, 2016). The extended conformation
ments with E. coli dnaK deletion strains yielded probably results from the high content of inter-
susceptibilities comparable to the wild-type E. spaced proline residues that prevent helix forma-
coli strains (Krizsan et al. 2014; Scocchi et al. tion (Graf et al. 2017). As shown by structural
2009). Therefore, co-immunoprecipitation exper- studies, binding of PrAMPs involves a series of
iments with labeled PrAMPs and E. coli lysate polar contacts, as well as stacking interactions,
were again undertaken to search for alternative between the peptide and the ribosome (Graf et al.
targets. This led to the identification of ribosomal 2017). Interactions of Bac7 and Tur1A with the
proteins, suggesting that ribosomes and transla- ribosome predominantly involve arginine side
tion may represent an alternative target of chains (Gagnon et al. 2016; Mardirossian et al.
PrAMPs (Krizsan et al. 2014; Mardirossian et al. 2018; Seefeldt et al. 2016). These structures
2014). Subsequent binding assays, as well as revealed that two distinct binding modes and
in vitro translation experiments, indicated that mechanisms of action exist for PrAMPs: the type
protein synthesis is most likely the primary pro- I PrAMPs, including Bac7, Onc, Pyr, Met, and
cess inhibited by PrAMPs (Krizsan et al. 2014; Tur1A, all act as inhibitors of the first elongation
Mardirossian et al. 2014). step, whereas type II PrAMPs, such as Apidaecin
1b and Api137, act predominantly as translation
termination inhibitors.
6.1.5 Determination
of the Ribosomal Binding Site
of PrAMPs 6.1.6 Interactions of PrAMPs
with the Nascent Polypeptide
As consequence of the newly discovered ribo- Exit Tunnel
some binding and inhibition capacity of PrAMPs,
it became interesting to resolve structures of 70S Although class I and class II PrAMPs all bind
ribosomes in complex with different peptides and within the polypeptide tunnel of the ribosome
to determine the mode of action for each peptide and utilize similar modes of interaction between
member, for example, to determine whether all aromatic and arginine side chains with rRNA
PrAMPs bind in the same position on the ribo- nucleotides, the overall orientation of the pep-
some and whether they exhibit an identical mech- tides within the tunnel is different as well as the
anism of action. The first PrAMP to be resolved extent to which they encroach into the ribosomal
on the bacterial ribosome was a derivative of A-site at the peptidyltransferase center (PTC).
oncocin (Onc) from milkweed bugs (Oncopeltus
fasciatus), termed Onc112 (Roy et al. 2015; 6.1.6.1 Binding of Class I PrAMPs
Seefeldt et al. 2015). This was followed by Bac7 Class I PrAMPs bind to the ribosome with an
from cows (Bos taurus), metalnikowin (Met) inverted orientation compared to a nascent poly-
from the green shield bug (Palomena prasina), peptide, i.e., with the N-terminus located in the
and Pyr from the firebug (Pyrrhocoris apterus) A-site and the C-terminus extending down the
(Gagnon et al. 2016; Seefeldt et al. 2016). The polypeptide exit tunnel (Fig. 6.3a). The PrAMP
latest resolved structures of PrAMPs are Api137 binding site on the ribosome can be divided into
from the honey bee (Apis mellifera) (Florin et al. sections located within the A-site binding pocket,
2017) and Tur1A from bottlenose dolphins the A-site crevice, and the upper region of the
(Tursiops truncatus) (Mardirossian et al. 2018). polypeptide exit tunnel. The N-terminal residues
All PrAMPs were found to bind to the polypep- reaching into the ribosomal A-site determine the
78 M. Graf and D. N. Wilson

Fig. 6.3 (a) Overview showing the binding site of Tur1A NPET. Hydrogen bonds are indicated by yellow dashed
within the LSU (gray) polypeptide exit tunnel (NPET) lines, whereas stacking interactions are indicated by black
with P-tRNA (green) shown for reference (Mardirossian two-headed arrows. (b) Pyr (cyan) establishes interactions
et al. 2018). The N- and C-termini of Tur1A are indicated solely with 23S rRNA residues (gray) (Gagnon et al.
by the letters N and C, respectively. (b, c) Established 2016; Seefeldt et al. 2016). (c) Tur1A (yellow) establishes
interactions of (b) Pyr (Gagnon et al. 2016; Seefeldt et al. interactions with 23S rRNA residues (gray), as well as
2016) and (c) Tur1A (Mardirossian et al. 2018) within the protein L4 (purple) (Mardirossian et al. 2018)

mechanism of inhibition (Fig. 6.3a). The through a multitude of polar contacts and stack-
C-terminal residues reaching into the polypeptide ing interactions with residues of the polypeptide
exit tunnel seem to be less crucial for binding and exit tunnel. The most striking similarity between
the inhibitory activity of PrAMPs. Noteworthy, all class I PrAMPs is a conserved core, which
in none of the available structures were the resi- harbors a PRP motif. In all resolved class I
dues at the very C-terminus resolved (Gagnon PrAMP 70S complexes so far, the PRP motif of
et al. 2016; Mardirossian et al. 2018; Roy et al. each PrAMP was located in exactly the same
2015; Seefeldt et al. 2015, 2016). In biochemical position with exactly the same conformation
studies investigating Bac7 fragments, deletion of (Fig. 6.5g) (Gagnon et al. 2016; Mardirossian
up to 19 amino acids at the C-terminus did not et al. 2018; Roy et al. 2015; Seefeldt et al. 2015,
abolish PrAMP activity (Benincasa et al. 2004). 2016). However, the amino acid composition, the
In contrast, deletion of the amino acids valine and number of residues, and the number of estab-
aspartate from the N-terminus of oncocins, which lished contacts within the A-site binding pocket
are positioned in the A-site binding pocket, leads are the most obvious differences among class I
to strongly reduced activity and underlines the PrAMPs (Gagnon et al. 2016; Mardirossian et al.
crucial function of the N-terminus for antimicro- 2018; Roy et al. 2015; Seefeldt et al. 2015, 2016).
bial activity (Gagnon et al. 2016). Binding of Consistently, while just four N-terminal amino
PrAMPs to the ribosome in general is facilitated acids of insect Onc112, Met, and Pyr reach into
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 79

the A-site binding pocket (Fig. 6.3b) (Gagnon 5 and 7, which seem to be less critical for binding.
et al. 2016; Roy et al. 2015; Seefeldt et al. 2015, While Onc112 harbors a proline residue in posi-
2016), seven additional amino acids are present tion 5, Met and Pyr have in the same position an
in mammalian Bac7 and Tur1A (Fig. 6.3c) aspartate and a serine residue, respectively
(Gagnon et al. 2016; Mardirossian et al. 2018; (Fig. 6.1) (Seefeldt et al. 2015, 2016). Furthermore,
Seefeldt et al. 2016). The additional residues the leucine residue in position 7 of Onc112 and
present in Bac7 and Tur1A form a short loop, Pyr is an arginine residue in Met. The alteration of
which acts like an A-site anchor and seems to be residues 5 and 7 results in different hydrogen
specific for mammalian PrAMPs (Fig. 6.3c) bonding patterns within the A-site crevice. Asp5 in
(Gagnon et al. 2016; Mardirossian et al. 2018; Met establishes hydrogen bonds with U2584 and
Seefeldt et al. 2016). U2585. Arg7 of Met contacts A2503 and the back-
bone phosphate of G2505. Ser5, which is present
Binding of Insect PrAMPs Onc112, in Pyr, interacts with U2584 (Fig. 6.3b). Leu7 does
Metalnikowin, and Pyrrhocoricin The amino not establish interactions, either in Onc112 or in
acid sequence of the shorter N-terminus of insect Pyr. Hydrogen bonds in the A-site crevice, which
PrAMPs is conserved between Onc112, Pyr, and are common for all insect PrAMPs, involve the
Met (Fig. 6.1) (Graf et al. 2017). Interactions peptide backbone. Accordingly, the backbone
within the A-site binding pocket and A-site crev- α-carbonyl oxygen, the α-amine of Tyr6, and the
ice involve polar contacts as well as stacking α-amine of Leu7/Arg7 contact the U2506 nucleo-
interactions (Fig. 6.3b) (Gagnon et al. 2016; Roy base (Fig. 6.3b).
et al. 2015; Seefeldt et al. 2015, 2016). Polar con- As mentioned before, the number of resolved
tacts are established by the peptide backbone and residues at the C-terminus differs between
side chains of the amino acid residues in the sec- PrAMPs. With respect to insect PrAMPs, 6 aa of
ond and third positions (Fig. 6.3b) (Seefeldt et al. Onc112, 5 aa of Met, and 4 aa of Pyr were not
2015, 2016). Consistently, Asp2 of insect resolved in the reported structures (Seefeldt et al.
PrAMPs contacts the 2′-OH group of C2507 and 2015, 2016). However, the visualized residues at
the base of G2553 (Fig. 6.3b) (Seefeldt et al. the C-terminus of insect PrAMPs were observed
2015, 2016). The side chain of Lys3 establishes a to establish just a single hydrogen bond and a
hydrogen bond with the phosphate-oxygen of the single stacking pair within the upper polypeptide
23S rRNA nucleotide A2453 (Fig. 6.3b) (Seefeldt exit tunnel (Fig. 6.3b). Both interactions involve
et al. 2015, 2016). Examples for backbone inter- Arg9 of the conserved PRP motif (Fig. 6.3b). The
actions are provided by Val1 (Fig. 6.3b). While hydrogen bond is established with 23S nucleo-
the α-carbonyl oxygen of Val1 interacts with the tide U2584. The stacking interaction is observed
nucleobase of C2573, the α-amine of Val1 con- with C2610 (Fig. 6.3b). The presence of only two
tacts the ribose of C2507 (Seefeldt et al. 2015, interactions underlines the minor importance of
2016). the C-terminus of insect PrAMPs in binding.

Residues five to seven of insect PrAMPs are Binding of Mammalian PrAMPs Bac7 and
located within the A-site crevice (Seefeldt et al. Tur1A Binding of mammalian Bac7 and Tur1A
2015, 2016). Among those residues, Tyr6 is the to the A-site binding pocket and the A-site crev-
most conserved and therefore present in all insect ice of eubacterial 70S ribosomes also involves
PrAMPs (Graf et al. 2017; Seefeldt et al. 2015, polar contacts of the peptide backbone and the
2016). Tyr6 contributes to A-site crevice binding, amino acid side chains (Fig. 6.3c) (Mardirossian
as well as antimicrobial activity, by establishing a et al. 2018; Seefeldt et al. 2016). Four of eight
stacking interaction with C2452 (Fig. 6.3b) positions in Bac7 and Tur1A harbor arginine
(Seefeldt et al. 2015, 2016). Replacement of Tyr6 residues in the LSU A-site binding pocket which
by Ala results in a 32-fold decreased activity establish an extensive hydrogen bonding net-
(Knappe et al. 2011). This contrasts with positions work as well as a single stacking interaction.
80 M. Graf and D. N. Wilson

The stacking interaction is observed between Due to the enrichment in arginine residues
Arg2 and C2573 (Fig. 6.3c). Compared to insect compared to insect PrAMPs, Bac7 and Tur1A
PrAMPs, this stacking interaction replaces the establish more interactions within the upper
backbone contact of Val1 with the nucleobase of polypeptide exit tunnel (Mardirossian et al. 2018;
C2573 (Fig. 6.3c) (Mardirossian et al. 2018; Seefeldt et al. 2016). The majority of these con-
Seefeldt et al. 2015, 2016). Hydrogen bonds are tacts involve stacking interactions. In the case of
evident for all arginine residues in the A-site Tur1A, two stacking interactions are observed
binding pocket (Fig. 6.3c) (Mardirossian et al. (Fig. 6.3c), such that Arg15 and Arg16 stack
2018; Seefeldt et al. 2016). The side chain of upon U2586 and His69 of L4, respectively
Arg1 interacts with the nucleobase of U2555. (Fig. 6.3c) (Mardirossian et al. 2018). The
Hydrogen bonds with the 23S rRNA backbone C-terminus of Bac7 establishes three stacking
are established by the side chains of Arg2 interactions with the surrounding tunnel (Seefeldt
(G2454 and A2453), Arg4 (C2452 and G2494), et al. 2016). Namely, Arg12, Arg14, and Arg16
and Arg6 (A2453 and U2493). Interactions of form stacking interactions with C2610, U2586,
the peptide backbone are present for the and A2062, respectively. The stacking interac-
α-amines of Arg1, Arg2, Ile3, and Arg4 which tions of Arg12 with C2610 are analogous to the
contact U2491, C2573, U2492, and U2493, tunnel interaction of Arg9 in insect PrAMPs
respectively. (Seefeldt et al. 2015, 2016).

The overall contacts of Bac7 and Tur1A 6.1.6.2 B inding of Class II PrAMP
(Fig. 6.3c) with the A-site crevice are comparable Api137
to those of insect PrAMPs (Fig. 6.3b) Api137, which is derived from wild-type apidae-
(Mardirossian et al. 2018; Seefeldt et al. 2015, cin 1b from Apis mellifera (Fig. 6.1), is a class II
2016). Insect PrAMP residues five to seven cor- PrAMP. In contrast to other class I PrAMPs,
respond to mammalian PrAMP residues eight to Api137 binds to the polypeptide tunnel with a
ten. Backbone interactions involve the α-carbonyl similar orientation to a nascent chain (Fig. 6.4a,
oxygen and α-amine of residue nine as well as b) (Florin et al. 2017). The C-terminal residues
the α-amine of residue ten that contacts the Arg17 and Leu18 are located in the A-site crev-
U2506 nucleobase (Fig. 6.3c) (Mardirossian ice but do not reach into the A-site binding
et al. 2018; Seefeldt et al. 2016). Similar to pocket of the PTC. Placement of Arg17 and
Onc112, positions eight and ten harbor proline Leu18 in the A-site crevice is crucial for the
and leucine residues (Fig. 6.1) (Mardirossian mechanism of action of Api137. For instance,
et al. 2018; Seefeldt et al. 2015, 2016). Except for exchange of Arg17 by Ala leads to decreased
the described backbone interactions, the side activity against E. coli ribosomes (Castle et al.
chains of both residues do not establish hydrogen 1999). The N-terminal residues pass down the
bonds with the surrounding tunnel (Fig. 6.3c) polypeptide tunnel (Fig. 6.4a, c) (Florin et al.
(Mardirossian et al. 2018; Seefeldt et al. 2016). 2017). In the available structure of Api137, the
Nonetheless, similar to Tyr6 which is present in last four residues at the N-terminus are not
all insect PrAMPs, residue nine of Bac7 and resolved. Nonetheless, compared to class I
Tur1A establishes a stacking interaction with PrAMPs, binding of Api137 to the polypeptide
C2452 (Fig. 6.3c) (Mardirossian et al. 2018; tunnel is primarily facilitated by stacking inter-
Seefeldt et al. 2015, 2016). As observed for insect actions (Fig. 6.4c) (Florin et al. 2017; Seefeldt
PrAMPs, Tur1A base stacking involves a tyro- et al. 2015, 2016). Stacking interactions are
sine residue as well (Fig. 6.3c). Bac7 instead har- observed between Tyr7 and A751, Arg12 and
bors an arginine residue in this position that C2611, as well as His15 and the nucleobase of
stacks upon C2452 (Fig. 6.1) (Seefeldt et al. G2505 (Fig. 6.4c) (Florin et al. 2017). In addi-
2015, 2016). tion to stacking interactions, polar contacts are
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 81

Fig. 6.4 (a) Overview showing the binding site of Api137 shown for reference. (c) Interactions of Api137 (slate)
(slate) within the LSU (gray) polypeptide exit tunnel with 23S rRNA nucleotides (gray) located in the NPET
(NPET) in the presence of P-tRNA (green) and RF1 (Florin et al. 2017). Stacking interactions are indicated by
(orange) (Florin et al. 2017). The GGQ motif is high- black two-headed arrows. (d, e) Interactions of Api137 in
lighted in red. The N- and C-termini of Api137 are indi- the presence of (d) RF1 (orange) and (e) P-tRNA (green)
cated by the letters N and C, respectively. (b) (Florin et al. 2017). (d) Arg17 contacts 23S rRNA nucleo-
Superimposition of Api137 (slate) and Pyr (left; cyan) tides and Gln235 of RF1. (e) Leu18 of Api137 contacts
(Gagnon et al. 2016; Seefeldt et al. 2016) or Tur1A (right; A76 of the deacylated tRNA located in the P-site
yellow) (Mardirossian et al. 2018). P-tRNA (green) is

established in the PTC with RF1 and the deacyl- 6.1.7  echanism of Action of Class
M
ated P-site tRNA (Fig. 6.4d, e). The side chain of I and Class II PrAMPs
the penultimate residue Arg17 is coordinated
between 23S rRNA residues and Gln235 of the Under normal conditions ribosomal protein syn-
RF1 GGQ motif (Fig. 6.4d). The 23S rRNA con- thesis passes through the phases of translation ini-
tacts comprise hydrogen bonds with the nucleo- tiation, translation elongation, and translation
base of C2452, the 2′- and the 3′-oxygen of termination. Translation initiation involves the
G2505, and the ribose of U2506 (Fig. 6.4d). placement of an fMet-tRNAfMet over an AUG start
Gln235 interacts with Arg17 through the car- codon in the P-site of the 70S ribosome. The
bonyl oxygen of the side chain (Fig. 6.4d). CCA-end that carries the fMet moiety is placed in
Leu18 appears to establish two interactions via the PTC, and the 3′-end establishes contacts with
the N-terminal OH group (Fig. 6.4e). These the P-loop of the LSU. Translation elongation
interactions include the 2′-OH and the 3′-OH of comprises the EF-Tu-mediated delivery and sub-
A76 of deacyl-P-tRNA (Fig. 6.4e). The overall sequent accommodation of an aa-tRNA to the
altered interaction network compared to class I ribosomal A-site, transpeptidation in the PTC,
PrAMPs forms the basis for the unique mecha- and translocation catalyzed by
nism of action of Api137. EF-G. Accommodation coincides with the estab-
82 M. Graf and D. N. Wilson

Fig. 6.5 (a–c) Canonical translation in the absence of 2016; Seefeldt et al. 2016) and (f) mammalian Tur1A
protein synthesis inhibitors, showing (a) translation initia- (yellow) (Mardirossian et al. 2018) with accommodated
tion with initiator P-tRNA (green) bound to the ribosomal aa-tRNA (salmon). (g) Superimposition of mammalian
P-site. (b) Delivery of aa-tRNA (light blue) by EF-Tu Tur1A (yellow) (Mardirossian et al. 2018) and Bac7(1-­16)
(salmon) to the A-site, followed by (c) tRNA accommoda- (teal) and insect-derived PrAMPs Onc112 (slate), metal-
tion into the A-site on the large subunit and subsequent nikowin-­1 (Met, purple), and Pyr (cyan) (Gagnon et al.
departure of EF-Tu. (d) In the presence of class I PrAMPs 2016; Roy et al. 2015; Seefeldt et al. 2015, 2016), with the
(yellow), such as Tur1A, aa-tRNA delivery can occur; conserved PRP motif highlighted. (Figure adapted from
however the aa-tRNA accommodation is blocked. (e, f) Graf et al. 2017)
Superimposition of (e) insect Pyr (cyan) (Gagnon et al.

lishment of interactions between the CCA-end of but perturbs delivery of the first aa-tRNA by
the A-site tRNA and the A-loop within the 23S EF-Tu (Fig. 6.5d) (Mardirossian et al. 2018;
rRNA. During translation termination the nascent Seefeldt et al. 2015, 2016). Although decoding
chain is released from the ribosome with the aid on the SSU is in principle possible, subsequent
of the peptide chain release factor RF1 or RF2. accommodation following release from EF-Tu is
Both classes of PrAMPs inhibit protein synthesis blocked due to the steric hindrance generated by
by binding to the polypeptide tunnel of bacterial the N-terminal residues of class I PrAMPs
70S ribosomes, yet binding of each class of (Fig. 6.5d–g). Accordingly, superimposition with
PrAMPs results in inhibition of different steps of pre-attack 70S ribosomes carrying an accommo-
the translation cycle (Florin et al. 2017; Gagnon dated aa-tRNA in the A-site shows incompatibil-
et al. 2016; Roy et al. 2015; Seefeldt et al. 2015, ity of the A-tRNA CCA-end and the N-terminal
2016). In the case of class I PrAMPs, the residues residues of PrAMPs (Fig. 6.5e, f) (Graf et al.
crucial for translation interference are located in 2017). The same steric clash also prevents bind-
the A-site binding pocket (Fig. 6.5) (Seefeldt et al. ing of PrAMPs during translation elongation. To
2015, 2016). In the case of the class II PrAMP avoid steric clashes with a peptidyl-tRNA, bind-
Api137, the important residues are located in the ing of class I PrAMPs has to occur between
PTC (Figs. 6.4 and 6.6) (Florin et al. 2017). translation termination and initiation (Graf et al.
In the presence of class I PrAMPs, the A-site 2017). In general, class I PrAMPs were observed
crevice and the A-site binding pocket are blocked to inhibit the transition from translation initiation
by the N-terminal residues of the PrAMP to translation elongation as confirmed by bio-
(Fig. 6.5) (Gagnon et al. 2016; Mardirossian chemical experiments (Mardirossian et al. 2018;
et al. 2018; Roy et al. 2015; Seefeldt et al. 2015, Seefeldt et al. 2015, 2016). In accordance, toe-
2016). This allows placement of an fMet-­ printing assays showed that translation in the
tRNAfMet during translation initiation (Fig. 6.5a) presence of class I PrAMPs indeed leads to
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 83

Fig. 6.6 (a) Translation terminates when the bacterial tRNA. (c) Api137 binds to post-hydrolysis complexes and
70S ribosome encounters a stop codon. (b) RF1 is (d) prevents dissociation of release factor from the 70S
recruited to the terminating ribosome and mediates the ribosome
hydrolysis of the nascent chain (NC, green) from P-site

r­ ibosomal stalling at the AUG start codon. Further to trap the majority of ribosomes during termina-
translation elongation is therefore not possible. tion due to the absence of free RF1 and RF2.
Biochemical toeprinting assays showed that in Because of the small time window in which
contrast to class I PrAMPs, class II PrAMPs do Api137 has to evoke its inhibition, it seems rela-
not inhibit translation initiation or elongation but tively clear how the binding site of Api137 in the
rather trap ribosomes with a stop codon in the polypeptide exit tunnel is accessed. Although the
A-site, suggesting that translation termination is 70S ribosome constitutes a porous complex (Voss
affected (Florin et al. 2017). Subsequent biophys- et al. 2006), due to the relatively large size of
ical assays demonstrated that Api137 does not Api137, access to its binding site within the exit
interfere RF binding to the ribosome nor with the tunnel must be gained either from the PTC or via
RF-mediated release of the nascent chain but the tunnel exit rather than by diffusion through the
instead prevents the dissociation of the RF from ribosomal core. Given that Api137 traps the ter-
the ribosome (Florin et al. 2017). Cryo-EM struc- minating ribosome with a deacylated tRNA in the
tures revealed the interactions that are crucial for P-site and one of the decoding RFs in the A-site
RF trapping, encompassing the C-terminus of (Florin et al. 2017), it seems unlikely that Api137
Api137 where Leu18 interacts with the ribose of can gain access to its binding site via the PTC. This
A76 of the deacyl-tRNA in the P-site and Arg17 suggests that Api137 must access its binding site
establishes a number of hydrogen bonds with the by entering from the tunnel exit site. Considering
23S rRNA and Gln235 of the GGQ motif of RF1 the length of ~100 Å and the width of the poly-
(Florin et al. 2017). As consequence, dissociation peptide tunnel with a diameter of less than 15 Å at
of RF1 from the post-hydrolysis 70S ribosome is the narrowest section (Nissen et al. 2000), one
inhibited, even in the presence of RF3, which is role of the proline residues may be to maintain an
known to recycle RF1 and RF2 from the ribo- extended conformation of the peptides to facili-
some. Moreover, inhibition of bacterial cell tate their diffusion up to the ribosomal tunnel.
growth by Api137 is not only caused by trapping With respect to the timing of binding, Api137
of RF1 or RF2 on post-hydrolysis ribosomes. cannot access the binding site at all stages during
Because the number of RFs is 20–100x lower termination. First, the presence of a nascent chain
than the number of ribosomes in the cell, Api137 in the polypeptide exit tunnel during translation
also has an indirect effect to deplete the endoge- elongation represents a steric block that is incom-
nous pools of free RF1 and RF2, causing ribo- patible with simultaneous binding of Api137.
somes to become stalled during translation Second, Api137 action requires 70S-bound RF1
termination due to lack of RFs (Florin et al. 2017). and P-tRNA, which interact via Gln235 of the
As consequence of this RF depletion, it was GGQ motif and A76 with Arg17 and Leu18,
shown that Api137 also promotes stop codon read respectively (Florin et al. 2017). As consequence,
through. Hence, Api137 has a dual mode of inhi- binding of Api137 has to occur after peptide chain
bition, namely, (i) to directly trap RF1 and RF2 on release and before RF1/2 departure from the post-
a small minority of terminating ribosomes and (ii) hydrolysis 70S ribosome.
84 M. Graf and D. N. Wilson

6.1.8 Mutagenesis Studies residues, which are crucial for stable binding of
with PrAMPs PrAMPs. Consistently, the published structural
data show that A2503 and A2059 are situated in
Although the recent studies show inhibition and close proximity to Api137 (Florin et al. 2017).
binding of PrAMPs to 70S ribosomes, further Although no hydrogen bonds are observed for
biochemical evidence was required to confirm both nucleotides, the close proximity to Api137
that the translation machinery is the primary tar- possibly allows van der Waals interactions. Upon
get of PrAMPs. A tool of choice for this valida- mutation these interactions are presumably per-
tion is the selection of mutant strains that exhibit turbed and result in a decreased binding affinity
resistance to the PrAMP. Bacterial cells exposed of Api137 toward the 70S ribosome. This con-
to subinhibitory concentrations of a compound trasts the effect of A2503 and A2059 mutation
in culture acquire rRNA or protein sequence on Onc112 binding (Gagnon et al. 2016). In the
alterations that allow bacteria to escape the presence of Onc112, A2503 hydrogen bonds
inhibitory action. Protein or rRNA alterations in with A2062 and stabilizes the nucleotide in a 90°
general can affect inhibition either more directly, rotated conformation, which is crucial for
e.g., by decreasing the affinity to the binding accommodation of Onc112 in its binding site
site, or more indirectly, e.g., by preventing com- (Roy et al. 2015; Seefeldt et al. 2015). The non-
pound uptake into the cell. Problematic for the rotated conformation of A2062 represents a ste-
identification of rRNA mutations is the presence ric block that prevents binding of Onc112 in the
of several rRNA operons in E. coli. Thus, exit tunnel. However, independent of the effect
acquirement of a single mutation within one of the A2503 and A2059 mutation on neighbor-
rRNA operon does not affect the complete pool ing rRNA residues, the decreased susceptibility
of rRNAs incorporated into ribosomes. To avoid against Onc112 and Api137 is a clear evidence
this issue, the identification of rRNA mutations, for ribosomes as major target for the inhibition
which confer resistance to PrAMPs, requires the by PrAMPs.
use of special E. coli strains, called Squires (SQ) With respect to protein alterations, most of the
strains (Asai et al. 1999). The SQ strains exhibit initially identified mutations, conferring resis-
a single rRNA allele that is encoded by an extra- tance to PrAMPs, were detected in the gene cod-
genomic plasmid. The genomically encoded ing for the SbmA transporter. The primary reason
operons, coding for 5S, 16S, and 23S rRNA, are for this is possibly the lower cost in fitness for
absent. By exposing an SQ strain in culture to bacteria grown under lab conditions. Accordingly,
PrAMPs, it is possible to identify rRNA muta- sbmA-deficient E. coli cells stay viable and
tions conferring resistance against antibiot- exhibit lower susceptibility against PrAMPs,
ics. This is exemplified macrolide erythromycin such as Bac7 (Mattiuzzo et al. 2007). In contrast,
resistance mutants, which also exhibit some sequence alterations within functionally impor-
cross-resistance with the insect PrAMPs Onc112 tant regions are not well tolerated (Sato et al.
(Gagnon et al. 2016) and Api137 (Florin et al. 2006). Nonetheless, for Api137 it was possible to
2017), including 23S rRNA nucleotides A2503 identify mutations within proteins by culturing E.
and A2059. Single or double mutation of A2503 coli cells that express multiple copies of an sbmA
and/or A2059 was shown to result in a 4- to coding plasmid (Florin et al. 2017). The presence
16-fold decreased susceptibility of E. coli cells of multiple copies of the sbmA gene makes the
against Onc112 and Api137 (Florin et al. 2017; acquirement of the same resistance mutation in
Gagnon et al. 2016). Normally, A2503 and all plasmids highly unlikely and ensures continu-
A2059 stack upon each other and stabilize adja- ous uptake of PrAMPs into the bacterial cell.
cent nucleotides, such as A2062. The decreased Consequently, the bacterial cell is forced to
susceptibility against both PrAMPs is presum- acquire alternative resistance mechanisms, which
ably caused by conformational changes of nucle- help to validate the ribosome as primary target
otides A2503 and A2059 as well as neighboring for PrAMP action. Overall, it was possible to
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 85

identify Api137 resistance mutations within ribo- a P-tRNA. As consequence, the binding affinity
somal proteins L3 and L16 as well as in RF1 and of Api137 toward the ribosome is reduced.
RF2. Additionally, mutations within ribosomal In addition to ribosomal protein mutations,
proteins L4 and L22 also confer cross-resistance more resistance mutations against Api137 were
against Api137, specifically, mutation of L4 resi- identified within RF1 and RF2. In this context,
due Lys63 to Glu as well as deletion of L22 resi- mutation of Arg262 to Cys and Gln280 to Leu in
dues 82–84 (Florin et al. 2017). RF2 as well as Asp241 to Gly in RF1 leads to
Ribosomal proteins L4 and L22 harbor each significantly reduced susceptibility of E. coli
an extended loop that form together the central cells against Api137 (Florin et al. 2017). The
constriction within the polypeptide exit tunnel. cause of the decreased susceptibility is most
The observed mutations within L4 and L22 pre- likely a destabilization of the binding of RF1 and
sumably lead to reordering of these loops, which RF2 which overcomes the stabilizing interactions
were shown in our structure to directly or indi- of Api137 with the GGQ motif of RF1 and RF2.
rectly stabilize binding of Api137 in the polypep- Accordingly, RF1 binding to the 70S ribosome is
tide tunnel (Florin et al. 2017). Thus, Arg61 of L4 stabilized by a single polar contact of 23S nucle-
provides a putative hydrogen bond to binding of otide C2573 with Asp241. Mutation of Asp241 to
Api137. Mutation of Lys63 could reorder the Gly prevents an interaction with C2573 and
extension, which contains the interacting Arg61, hence could promote dissociation of RF1. Similar
and hence possibly destabilize binding of Api137. to RF1, mutation of RF2 residues Arg262 to Cys
In contrast, deletion of L22 residues 82–84 pre- and Gln280 to Leu resolves interactions with 23S
sumably destabilizes Api137 binding more indi- rRNA nucleotides C2556 and U2492, respec-
rectly by contraction of the extended loop of L22. tively. This could promote dissociation of RF2
The extended loop of L22 contacts the 23S rRNA since the interactions of Arg17 of Api137 are pre-
backbone and stabilizes H34. H34 harbors nucle- sumably insufficient to trap RF2 on the
otide A751 that stacks upon Api137 residue Tyr6 ribosome.
and is crucial for stable binding. The contraction It is noteworthy that none of the rRNA or pro-
of the extended loop of L22 most likely causes tein mutations conferred complete resistance
reordering of H34 and, hence, destabilization of against Onc112 and Api137. Nonetheless, all
Api137 by the lacking stacking interaction mutations taken together are rather strong argu-
between Tyr6 and 23S rRNA residue A751. ments favoring the ribosome as the major target
Arg81 of L16, which can confer resistance to of these PrAMPs. A major problem for identify-
Api137 upon mutation to Cys, stabilizes P-loop ing more resistance mutations against PrAMPs is
nucleotide G2251 via two hydrogen bonds the high conservation of rRNA residues involved
(Florin et al. 2017). One hydrogen bond is estab- in PrAMP binding (RNAcentral 2017). As a con-
lished with the phosphate-oxygen backbone, and sequence, mutation of interacting residues would
another established with the nucleobase of result in dramatically decreased PTC activity and
G2251. P-loop nucleotide G2251 Watson-Crick therefore in reduced viability of bacterial cells
base-pairs with C75 of a P-site tRNA and stabi- (Sato et al. 2006; Thompson et al. 2001). An
lizes the CCA-end in the PTC for peptide bond example is C2452, which is located within the
formation. In the presence of Api137, the A2451 region (23S nts 2448 to 2554). C2452
C-terminal residue Leu18 contacts A76 of the establishes a stacking interaction with Tur1A
P-tRNA. The interaction between Leu18 and the (Mardirossian et al. 2018) and Pyr (Gagnon et al.
3′-end of the P-tRNA stabilizes Api137 binding 2016; Seefeldt et al. 2016) via a Tyr residue and a
to the ribosome. Upon mutation of Arg81 of L16 polar contact with Arg17 of Api137 (Florin et al.
to Cys, both polar interactions with nucleotide 2017). Sato and colleagues (Sato et al. 2006)
G2251 are resolved and possibly lead to higher showed that except for A2448 and A2453, all
flexibility within the P-loop and the CCA-end of nucleotides within the A2451 region are essential
86 M. Graf and D. N. Wilson

for ribosomal function. Consistently, it is difficult Scocchi et al. 2011). Since PrAMPs target the
to select for viable C2452 mutants. 70S ribosome and inhibit protein synthesis of
With regard to mammalian PrAMPs, such as eubacteria, it needs clarification whether PrAMPs
Bac7 and Tur1A, no rRNA or ribosomal protein inhibit human translation as well. To have an
resistance mutations have been identified so far. inhibitory effect on human cells, it requires, first,
In general, the identification of mutations confer- the uptake of PrAMPs into the cytoplasm and,
ring resistance to mammalian PrAMPs is more second, binding to the eukaryotic ribosome.
difficult. In contrast to insect PrAMPs, mamma- Because Bac7 and Tur1A are expressed as a pre-­
lian PrAMP Bac7 and Tur1A are considerably pro-­peptide, active peptide is not likely to be
longer (Graf et al. 2017). Mammalian PrAMPs present in the cytosol of eukaryotic cells.
show a full length of 39–60 aa, while insect Nevertheless, even if Bac7 and Tur1A gain access
PrAMPs show a full length between 15 and 34 aa. to the cytosol, these peptides appear to have a low
Although both insect and mammalian PrAMPs affinity for the eukaryotic ribosome since the
target ribosomes and inhibit translation, it was peptides inhibited protein synthesis in an E. coli
shown for full-length Bac7, which is very similar lysate more efficiently than in rabbit reticulocyte
to Tur1A (Mardirossian et al. 2018), that a lytic system (Mardirossian et al. 2018; Seefeldt et al.
mode of action is present (Podda et al. 2006). 2016). Indeed, the bacteriostatic and even bacte-
Only truncated derivatives of Bac7, like ricidal concentrations of Bac5 fragments were
Bac7(1-­16), solely inhibit protein synthesis with- shown not to be significantly toxic toward
out permeabilizing the bacterial membrane eukaryotic cells (Mardirossian et al. 2018).
(Podda et al. 2006; Seefeldt et al. 2016). Selection Similarly, it is necessary to validate in the
for resistance mutations is in principle possible, future whether Api137 can bind and inhibit mam-
but it is difficult for bacteria to acquire mutations malian 80S ribosomes. The available cryo-EM
that affect both modes of action (lytic and non-­ structure of 70S ribosomes in complex with RF1
lytic) of mammalian full-length PrAMPs (Lai and Api137 shows that Api137 binding primarily
and Gallo 2009; Peschel and Sahl 2006). As a involves interactions with universally conserved
consequence, bacteria remain susceptible against rRNA residues (RNAcentral 2017) and the GGQ
full-length AMPs in culture. With respect to the motif at the tip of RF1 domain III (Florin et al.
previously described rRNA mutations conferring 2017). The latter GGQ motif is conserved within
resistance to Onc112 and Api137 (A2503C and bacteria as well as evolutionary unrelated eukary-
A2059G), no elevated MIC was observed in pres- otic eRF1 (Frolova et al. 1999; Mora et al. 2003;
ence of Tur1A and Bac7 upon mutation of these Seit-Nebi et al. 2001; Shaw and Green 2007;
residues (Gagnon et al. 2016; Mardirossian et al. Zavialov et al. 2002). Assuming that Api137 is
2018). Regardless, as described before, the selec- taken up by human cells, the conservation of both
tion for more rRNA resistance mutations is rather rRNA residues and the GGQ motif of the class I
difficult, since nucleotide alterations within the release factors implies that Api137 is active
exit tunnel that are involved in binding of mam- against human 80S ribosomes and hence would
malian PrAMPs could lead to strongly perturbed be unsuitable for use as a therapeutic.
PTC activity (Sato et al. 2006). Superimposition of a mammalian 80S ribosome
in complex with eRF1 (Shao et al. 2016) and the
70S E. coli ribosome in complex with RF1 and
6.1.9 Toxicity of PrAMPs Api137 (Florin et al. 2017) shows that binding of
in Eukaryotes Api137 is in principle compatible with binding to
both eubacterial and eukaryotic ribosomes. All
PrAMPs have already been tested using different observed contacts between Api137 and the 70S-­
animal models of infection and shown that they RF1 complex could be established with an 80S-­
are well tolerated and exert a promising healing eRF1 complex as well. Therefore, inhibition of
activity (Berthold et al. 2013; Knappe et al. 2015; human translation termination by Api137 seems
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 87

to be likely. Nonetheless, inhibition of human Casteels P, Tempst P (1994) Apidaecin-type peptide anti-
biotics function through a non-poreforming mecha-
80S ribosomes would still require uptake of nism involving stereospecificity. Biochem Biophys
Api137 from the extracellular space. Gram-­ Res Commun 199:339–345. https://doi.org/10.1006/
negative bacteria, like E. coli, actively uptake bbrc.1994.1234
Api137, even at low concentrations, with the Casteels P, Ampe C, Jacobs F, Vaeck M, Tempst P (1989)
Apidaecins: antibacterial peptides from honeybees.
SbmA transporter. This transporter is absent from EMBO J 8:2387–2391
eukaryotic cells. Consistently, uptake of PrAMPs Casteels P et al (1990) Isolation and characterization
into the cytoplasm of human cells would require of abaecin, a major antibacterial response peptide
diffusion through the plasma membrane, but this in the honeybee (Apis mellifera). Eur J Biochem
187:381–386
is unlikely due to the properties of Api137. Casteels-Josson K, Capaci T, Casteels P, Tempst P (1993)
Similar to most of the AMPs, Api137 exhibits a Apidaecin multipeptide precursor structure: a putative
positive net charge that has been suggested to mechanism for amplification of the insect antibacterial
interfere with the lytic action or uptake of AMPs response. EMBO J 12:1569–1578
Castle M, Nazarian A, Yi SS, Tempst P (1999) Lethal
into eukaryotic cells (Yeaman and Yount 2003). effects of apidaecin on Escherichia coli involve
Bacterial cells exhibit a negatively charged cyto- sequential molecular interactions with diverse targets.
plasmic membrane surface, which promotes J Biol Chem 274:32555–32564
adsorption of positively charged AMPs and sub- Chernysh S, Cociancich S, Briand JP, Hetru C, Bulet P
(1996) The inducible antibacterial peptides of the
sequent insertion into the membrane or uptake Hemipteran insect Palomena prasina: identification of
into the cell. Eukaryotic membranes, by contrast, a unique family of prolinerich peptides and of a novel
exhibit a more neutral surface that does not pro- insect defensin. J Insect Physiol 42:81–89. https://doi.
mote adsorption of AMPs. Future experiments org/10.1016/0022-1910(95)00085-2
Cociancich S et al (1994) Novel inducible antibacterial
will be required to determine in vivo and in vitro peptides from a hemipteran insect, the sap-sucking bug
whether Api137 is taken up by human cells and Pyrrhocoris apterus. Biochem J 300(Pt 2):567–575
whether it is toxic or even teratogenic. Florin T et al (2017) An antimicrobial peptide that inhib-
its translation by trapping release factors on the ribo-
some. Nat Struct Mol Biol 24:752–757. https://doi.
org/10.1038/nsmb.3439
References Frolova L et al (1999) Mutations in the highly conserved
GGQ motif of class 1 polypeptide release factors
Agerberth B, Lee JY, Bergman T, Carlquist M, Boman abolish the ability of human eRF1 to trigger peptidyl-­
HG, Mutt V, Jornvall H (1991) Amino acid sequence tRNA hydrolysis. RNA 5:1014–1020
of PR-39. Isolation from pig intestine of a new mem- Gagnon MG, Roy RN, Lomakin IB, Florin T, Mankin AS,
ber of the family of proline-arginine-rich antibacterial Steitz TA (2016) Structures of proline-rich peptides
peptides. Eur J Biochem 202:849–854 bound to the ribosome reveal a common mechanism
Asai T, Zaporojets D, Squires C, Squires CL (1999) An of protein synthesis inhibition. Nucleic Acids Res
Escherichia coli strain with all chromosomal rRNA 44:2439–2450. https://doi.org/10.1093/nar/gkw018
operons inactivated: complete exchange of rRNA Gennaro R, Skerlavaj B, Romeo D (1989) Purification,
genes between bacteria. Proc Natl Acad Sci U S A composition, and activity of two bactenecins, antibac-
96:1971–1976 terial peptides of bovine neutrophils. Infect Immun
Benincasa M, Scocchi M, Podda E, Skerlavaj B, Dolzani 57:3142–3146
L, Gennaro R (2004) Antimicrobial activity of Bac7 Graf M et al (2017) Proline-rich antimicrobial peptides
fragments against drug-resistant clinical isolates. targeting protein synthesis. Nat Prod Rep 34:702–711.
Peptides 25:2055–2061. https://doi.org/10.1016/j. https://doi.org/10.1039/c7np00020k
peptides.2004.08.004 Huttner KM, Lambeth MR, Burkin HR, Burkin DJ, Broad
Berthold N et al (2013) Novel apidaecin 1b analogs with TE (1998) Localization and genomic organization of
superior serum stabilities for treatment of infections sheep antimicrobial peptide genes. Gene 206:85–91
by gram-negative pathogens. Antimicrob Agents Knappe D et al (2010) Oncocin
Chemother 57:402–409. https://doi.org/10.1128/ (VDKPPYLPRPRPPRRIYNR-NH2): a novel antibac-
AAC.01923-12 terial peptide optimized against gram-negative human
Brogden KA (2005) Antimicrobial peptides: pore formers pathogens. J Med Chem 53:5240–5247. https://doi.
or metabolic inhibitors in bacteria? Nat Rev Microbiol org/10.1021/jm100378b
3:238–250. https://doi.org/10.1038/nrmicro1098 Knappe D, Zahn M, Sauer U, Schiffer G, Strater N,
Bulet P et al (1993) A novel inducible antibacterial pep- Hoffmann R (2011) Rational design of oncocin deriva-
tide of Drosophila carries an O-glycosylated substitu- tives with superior protease stabilities and antibacterial
tion. J Biol Chem 268:14893–14897 activities based on the high-resolution structure of the
88 M. Graf and D. N. Wilson

oncocin-DnaK complex. Chembiochem 12:874–876. RNAcentral (2017) RNAcentral: a comprehensive data-

https://doi.org/10.1002/cbic.201000792 base of non-coding RNA sequences. Nucleic Acids
Knappe D, Adermann K, Hoffmann R (2015) Oncocin Res 45:D128–D134. https://doi.org/10.1093/nar/
Onc72 is efficacious against antibiotic-susceptible gkw1008
Klebsiella pneumoniae ATCC 43816 in a murine thigh Roy RN, Lomakin IB, Gagnon MG, Steitz TA (2015)
infection model. Biopolymers 104:707–711. https:// The mechanism of inhibition of protein synthesis by
doi.org/10.1002/bip.22668 the proline-rich peptide oncocin. Nat Struct Mol Biol
Krizsan A, Volke D, Weinert S, Strater N, Knappe D, 22:466–469. https://doi.org/10.1038/nsmb.3031
Hoffmann R (2014) Insect-derived proline-rich anti- Runti G et al (2013) Functional characterization of SbmA,
microbial peptides kill bacteria by inhibiting bacterial a bacterial inner membrane transporter required for
protein translation at the 70S ribosome. Angew Chem importing the antimicrobial peptide Bac7(1-35).
Int Ed Eng 53:12236–12239. https://doi.org/10.1002/ J Bacteriol 195:5343–5351. https://doi.org/10.1128/
anie.201407145 JB.00818-13
Krizsan A, Prahl C, Goldbach T, Knappe D, Hoffmann Sato NS, Hirabayashi N, Agmon I, Yonath A, Suzuki T
R (2015) Short proline-rich antimicrobial peptides (2006) Comprehensive genetic selection revealed
inhibit either the bacterial 70S ribosome or the assem- essential bases in the peptidyl-transferase center. Proc
bly of its large 50S subunit. Chembiochem 16:2304– Natl Acad Sci U S A 103:15386–15391. https://doi.
2308. https://doi.org/10.1002/cbic.201500375 org/10.1073/pnas.0605970103
Lai Y, Gallo RL (2009) AMPed up immunity: how anti- Schnapp D, Kemp GD, Smith VJ (1996) Purification and
microbial peptides have multiple roles in immune characterization of a proline-rich antibacterial peptide,
defense. Trends Immunol 30:131–141. https://doi. with sequence similarity to bactenecin-7, from the
org/10.1016/j.it.2008.12.003 haemocytes of the shore crab, Carcinus maenas. Eur
Mardirossian M, Grzela R, Giglione C, Meinnel T, J Biochem 240:532–539
Gennaro R, Mergaert P, Scocchi M (2014) The host Schneider M, Dorn A (2001) Differential infectivity of
antimicrobial peptide Bac71-35 binds to bacterial two Pseudomonas species and the immune response
ribosomal proteins and inhibits protein synthesis. in the milkweed bug, Oncopeltus fasciatus (Insecta:
Chem Biol 21:1639–1647. https://doi.org/10.1016/j. Hemiptera). J Invertebr Pathol 78:135–140. https://
chembiol.2014.10.009 doi.org/10.1006/jipa.2001.5054
Mardirossian M et al (2018) The dolphin proline-rich Scocchi M, Lüthy C, Decarli P, Mignogna G, Christen P,
antimicrobial peptide Tur1A inhibits protein syn- Gennaro R (2009) The proline-rich antibacterial pep-
thesis by targeting the bacterial ribosome. Cell tide Bac7 binds to and inhibits in vitro the molecular
Chem Biol 25(5):530–539. https://doi.org/10.1016/j. chaperone DnaK. Int J Pept Res Ther 15:147–155.
chembiol.2018.02.004 https://doi.org/10.1007/s10989-009-9182-3
Mattiuzzo M, Bandiera A, Gennaro R, Benincasa M, Scocchi M, Tossi A, Gennaro R (2011) Proline-rich anti-
Pacor S, Antcheva N, Scocchi M (2007) Role of the microbial peptides: converging to a non-lytic mecha-
Escherichia coli SbmA in the antimicrobial activity nism of action. Cell Mol Life Sci 68:2317–2330.
of proline-rich peptides. Mol Microbiol 66:151–163. https://doi.org/10.1007/s00018-011-0721-7
https://doi.org/10.1111/j.1365-2958.2007.05903.x Seefeldt AC et al (2015) The proline-rich antimicrobial
Mora L, Heurgue-Hamard V, Champ S, Ehrenberg M, peptide Onc112 inhibits translation by blocking
Kisselev LL, Buckingham RH (2003) The essential and destabilizing the initiation complex. Nat Struct
role of the invariant GGQ motif in the function and Mol Biol 22:470–475. https://doi.org/10.1038/
stability in vivo of bacterial release factors RF1 and nsmb.3034
RF2. Mol Microbiol 47:267–275 Seefeldt AC et al (2016) Structure of the mammalian
Nissen P, Hansen J, Ban N, Moore PB, Steitz TA (2000) antimicrobial peptide Bac7(1-16) bound within the
The structural basis of ribosome activity in peptide exit tunnel of a bacterial ribosome. Nucleic Acids Res
bond synthesis. Science 289:920–930 44:2429–2438. https://doi.org/10.1093/nar/gkv1545
Otvos L Jr (2002) The short proline-rich antibacterial pep- Seit-Nebi A, Frolova L, Justesen J, Kisselev L (2001)
tide family. Cell Mol Life Sci 59:1138–1150 Class-1 translation termination factors: invariant GGQ
Otvos L Jr et al (2000) Interaction between heat shock minidomain is essential for release activity and ribo-
proteins and antimicrobial peptides. Biochemistry some binding but not for stop codon recognition.
39:14150–14159 Nucleic Acids Res 29:3982–3987
Peschel A, Sahl HG (2006) The co-evolution of host Shamova O, Brogden KA, Zhao C, Nguyen T, Kokryakov
cationic antimicrobial peptides and microbial resis- VN, Lehrer RI (1999) Purification and properties of
tance. Nat Rev Microbiol 4:529–536. https://doi. proline-rich antimicrobial peptides from sheep and
org/10.1038/nrmicro1441 goat leukocytes. Infect Immun 67:4106–4111
Podda E, Benincasa M, Pacor S, Micali F, Mattiuzzo Shao S, Murray J, Brown A, Taunton J, Ramakrishnan V,
M, Gennaro R, Scocchi M (2006) Dual mode of Hegde RS (2016) Decoding mammalian ribosome-­
action of Bac7, a proline-rich antibacterial peptide. mRNA states by translational GTPase complexes.
Biochim Biophys Acta 1760:1732–1740. https://doi. Cell 167:1229–1240.e1215. https://doi.org/10.1016/j.
org/10.1016/j.bbagen.2006.09.006 cell.2016.10.046
6 Intracellular Antimicrobial Peptides Targeting the Protein Synthesis Machinery 89

Shaw JJ, Green R (2007) Two distinct components of erone DnaK. J Mol Biol 425:2463–2479. https://doi.
release factor function uncovered by nucleophile par- org/10.1016/j.jmb.2013.03.041
titioning analysis. Mol Cell 28:458–467. https://doi. Zahn M, Kieslich B, Berthold N, Knappe D, Hoffmann
org/10.1016/j.molcel.2007.09.007 R, Strater N (2014) Structural identification of DnaK
Stensvag K, Haug T, Sperstad SV, Rekdal O, Indrevoll binding sites within bovine and sheep bactenecin
B, Styrvold OB (2008) Arasin 1, a proline-arginine-­ Bac7. Protein Pept Lett 21:407–412
rich antimicrobial peptide isolated from the spider Zanetti M, Litteri L, Gennaro R, Horstmann H, Romeo
crab, Hyas araneus. Dev Comp Immunol 32:275–285. D (1990) Bactenecins, defense polypeptides of
https://doi.org/10.1016/j.dci.2007.06.002 bovine neutrophils, are generated from precursor
Storici P, Zanetti M (1993) A cDNA derived from pig molecules stored in the large granules. J Cell Biol
bone marrow cells predicts a sequence identical to 111:1363–1371
the intestinal antibacterial peptide PR-39. Biochem Zanetti M, Litteri L, Griffiths G, Gennaro R, Romeo D
Biophys Res Commun 196:1058–1065. https://doi. (1991) Stimulus-induced maturation of probactene-
org/10.1006/bbrc.1993.2358 cins, precursors of neutrophil antimicrobial polypep-
Thompson J et al (2001) Analysis of mutations at residues tides. J Immunol 146:4295–4300
A2451 and G2447 of 23S rRNA in the peptidyltrans- Zanetti M, Del Sal G, Storici P, Schneider C, Romeo D
ferase active site of the 50S ribosomal subunit. Proc (1993) The cDNA of the neutrophil antibiotic Bac5
Natl Acad Sci U S A 98:9002–9007 predicts a pro-sequence homologous to a cysteine pro-
Voss NR, Gerstein M, Steitz TA, Moore PB (2006) The teinase inhibitor that is common to other neutrophil
geometry of the ribosomal polypeptide exit tunnel. antibiotics. J Biol Chem 268:522–526
J Mol Biol 360:893–906 Zasloff M (2002) Antimicrobial peptides in health and
Yeaman MR, Yount NY (2003) Mechanisms of antimi- disease. N Engl J Med 347:1199–1200. https://doi.
crobial peptide action and resistance. Pharmacol Rev org/10.1056/NEJMe020106
55:27–55. https://doi.org/10.1124/pr.55.1.2 Zavialov AV, Mora L, Buckingham RH, Ehrenberg M
Zahn M, Berthold N, Kieslich B, Knappe D, Hoffmann (2002) Release of peptide promoted by the GGQ motif
R, Strater N (2013) Structural studies on the forward of class 1 release factors regulates the GTPase activity
and reverse binding modes of peptides to the chap- of RF3. Mol Cell 10:789–798
 Part III
Other Activities of AMPs
 Antimicrobial and Cell-Penetrating
Peptides: How to Understand Two 7
Distinct Functions Despite Similar
Physicochemical Properties

Ines Neundorf

Abstract Keywords
Antimicrobial and cell-penetrating peptides Antimicrobial peptides · Cell-penetrating
are both classes of membrane-active peptides peptides · Plasma membranes · Drug delivery
sharing similar physicochemical properties. · Lipid-peptide interaction
Both kinds of peptides have attracted much
attention owing to their specific features.
AMPs disrupt cell membranes of bacteria and
display urgently needed antibiotic substances
with alternative modes of action. Since the
multidrug resistance of bacterial pathogens is Abbreviations
a more and more raising concern, AMPs have
gained much interest during the past years. On AMP Antimicrobial peptide
the other side, CPPs enter eukaryotic cells CD Circular dichroism
without substantially affecting the plasma CPP Cell-penetrating peptide
membrane. They can be used as drug delivery CS Chondroitin sulfate
platforms and have proven their usefulness in DSC Differential scanning calorimetry
various applications. However, although both EM Electron microscopy
groups of peptides are quite similar, their EPR Electron paramagnetic resonance
intrinsic activity is often different, and respon- FDA Food and Drug Administration
sible factors are still in discussion. The aim of FMM Functional membrane microdomain
this chapter is to summarize and shed light on GAG Glycosaminoglycan
recent findings and concepts dealing with dif- GPMV Giant plasma membrane vesicle
ferences and similarities of AMPs and CPPs GUV Giant unilamellar vesicle
and to understand these different functions. HS Heparan sulfate

I. Neundorf (*)
Department of Chemistry, Institute for Biochemistry,
University of Cologne, Cologne, Germany
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 93

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_7
94 I. Neundorf

IR Infrared tions arguing that the activity of these membrane-­

Ld Liquid disordered active families simply represents different facets
Lo Liquid ordered of what is a shared energy landscape (Last et al.
LTA Lipoteichoic acid 2013). Therefore it is not contradictory to notice
LUV Large unilamellar vesicle AMPs that cross plasma membranes and CPPs
MALDI 
Matrix-assisted laser desorption/ that offer antimicrobial activity. Moreover, it has
ionization been shown in many studies that by careful amino
MS Mass spectrometry acid substitutions, some CPPs can be turned into
NMR Nuclear magnetic resonance AMPs and vice versa, leading to the often posed
OBOC One bead one compound question: how different are they? The goal of this
PC Phosphatidylcholine chapter is to underpin the recently emerged
PE Phosphatidylethanolamine hypothesis that activity of AMPs and CPPs is
PG Phosphatidylglycerol mainly related to cooperative effects emerging
PI Phosphatidylinositol during binding of peptides to the lipid bilayer.
PS Phosphatidylserine Thus, the membrane is not any more seen as a
QSAR Quantitative structure-activity relation- passive layer that is simply affected when pep-
ship tides approach but actively contributes by its cur-
ROS Reactive oxygen species vature and partitioning into microdomains to the
STED Stimulated emission depletion entire peptide-lipid interaction process. In this
SUV Small unilamellar vesicle way it might be possible to explain these different
systems by common physicochemical processes.
After a short introduction to both groups of pep-
tides, their interplay with biological membranes
7.1 Introduction and resultant biological effects, methods to mea-
sure those activities, as well as ways to predict
Nature has breeded a group of fantastic mole- membrane activity and to design novel AMP or
cules, namely peptides, which are able to interact CPP sequences will be summarized.
with membranes and operate in processes of fun-
damental importance such as viral fusion, antimi-
crobial defense mechanisms, membrane poration, 7.1.1 Cell-Penetrating Peptides
delivery across membranes, and hormone-­
receptor interactions, to name only few. Two Cell-penetrating peptides constitute a family of
groups of peptides fall into focus of this chapter: natural or synthetically generated peptides that
antimicrobial peptides (AMPs) and cell-­are able to translocate across cellular membranes.
penetrating peptides (CPPs). While AMPs are Usually these peptides are relatively short (smaller
key components of the innate immune system than 35 amino acids) and mediate the transport of
and act against many pathogens, CPPs have been cargos into cells, which can be either covalently
emerged as valuable tools to translocate cargos or non-covalently attached to the CPP. A vast
across biological barriers. Despite their different sequence variety exists making classification of
functions, both groups share a lot of similarities CPPs difficult. One possibility is to group them
in structure, sequence, and, particularly, mem- depending on their origin in protein-­derived, chi-
brane activity rising the question how these dis- meric, or synthetic peptides. Otherwise, they can
crepancies in function can be explained by a be ordered based on their physicochemical char-
common set of physicochemical characteristics. acteristics into three main classes such as cationic,
In fact, both groups of peptides are strongly hydrophobic, and amphipathic. Whereby, this lat-
membrane-active and induce membrane fluctua- ter classification is more helpful in view of the
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 95

current understanding of CPPs’ cellular entry concentrations endocytosis and at high concen-
mechanism. CPPs have emerged as a powerful trations, direct entry processes are usually
technique to deliver various types of molecules observed.
into cells such as proteins, peptides, siRNA, One pitfall when working with CPPs is their
DNA, liposomes, nanoparticles, or small organic mainly endocytotic uptake. Once taken up via
drugs (Kalafatovic and Giralt 2017). Additionally, endocytosis, the peptide-cargo complex resides
some CPPs have made it already in clinical appli- in endosomes, from which it has to be released
cations (Feni and Neundorf 2017). for reaching its target site. Thus, the endocytosis
CPPs enter cells using different modes of mechanism represents one of the major disadvan-
action such as energy-dependent (endocytotic) or tages for the further development of CPPs.
energy-independent (direct) entry pathways. The Problems concerning the efficient escape of CPPs
latter might be accompanied by toxic membrane from the endosomes still persist and are in any
activity based on the ability of the CPP to disrupt case present when large molecules are delivered
membranes. Endocytosis on the other side is by CPPs. To circumvent these problems, CPPs
mainly observed when large cargos are attached may be equipped with fusogenic sequences or
to CPPs. Still it is not clear which factors provoke other endosomolytic molecules (Neundorf et al.
the one or other pathway. Notably, in both cases 2009). Still many efforts are made to develop
peptides adsorb at the membrane surface, where more selective and efficient CPP sequences.
they interact with negatively charged surface
components, probably glycoconjugates, anionic
lipids, or membrane proteins. This adsorption is 7.1.2 Antimicrobial Peptides
the result of a structural rearrangement, and
sometimes it leads also to an interaction with the Antimicrobial peptides are usually short peptides
interfacial zone of the lipid bilayer. After a criti- (<50 amino acids) and present in all forms of life,
cal (threshold) concentration is reached, mem- where they play a major role in the innate immune
brane deformation occurs as a result of elastic system and act as the “first line of defense”
stress and mass imbalance (Alvares et al. 2017). against invading pathogens. This class of pep-
Likely, these process parameters are unique for tides is structurally highly divers and of amphipa-
each CPP-cargo complex/conjugate. Of note is thic or cationic nature. AMPs share widespread
that most CPPs are unstructured in aqueous solu- toxicity against bacteria, yeasts, and fungi but are
tion but rapidly adopt defined secondary struc- relatively inactive toward host eukaryotic cells at
tures when coming in contact with the membrane bactericidal concentrations. Membrane permea-
lipid phase. Frequently, the formation of bilization is their main mechanism of action, but
amphipathic helices is observed in this case (Di additional mechanisms have been supposed,
Pisa et al. 2015). Although the intracellular including membrane destabilization, intracellular
uptake of most CPP is not fully understood, it is translocation, and inhibition of protein or nucleic
of wide acceptance that CPP enter cells by using acid synthesis. Often they display a polycationic
probably both pathways, direct entry and endocy- nature supporting electrostatic interaction with
tosis, simultaneously and/or depending on physi- negatively charged bacterial surface structures
cochemical properties, concentration, charge, such as lipoteichoic acids (LTA). Classical mech-
and length of the CPP, as well as characteristics anisms of antibiotic action include their penetra-
of the cargo. Additional influences on the uptake tion of the plasma membrane or cell wall, thus
mechanism of CPPs derive from properties, lipid resulting in lysis or disruption of ionic gradients.
composition, and protein content of the cell Specifically, they gain access to the cytoplasmic
membrane. Moreover, CPP concentration is sug- membrane and interact with lipid bilayers, form-
gested to have an important role, while at low ing transmembrane pores that disrupt the cell
96 I. Neundorf

membrane, finally leading to cell death. Several 7.2 Constitution of Biological

models have been proposed that explain how Lipid Membranes and Its
AMPs induce these membrane-disrupting pro- Relevance to the Activity
cesses, including the barrel-stave, toroidal pore of AMPs or CPPs
and carpet model (Sierra et al. 2017).
Although they act preferentially on the mem- Cellular membranes regulate the in- and outflow
brane level, AMPs may affect multiple biochemi- of nutrients, give the cell its shape, and are
cal processes in the pathogen. An increasing responsible for many other important cellular
body of evidence has demonstrated that AMPs functions like cell-cell communication and sig-
have also intracellular targets (Le et al. 2017). naling processes. Furthermore, membranes con-
For instance, blocking of RNA or protein synthe- stitute an impermeable barrier for large, charged,
sis and inhibiting enzymes necessary for linking or hydrophilic exogenous molecules, like thera-
cell wall structural proteins are further mecha- peutic oligonucleotides, proteins, or peptides.
nisms of AMPs leading to cell death. Some AMPs Still, it is one of the major challenges in pharma-
have the ability to translocate in eukaryotic cells ceutical industry to find powerful techniques to
without damage of the plasma membrane. overcome this barrier for an efficient drug deliv-
Notably, they target cancer cells and find intracel- ery. The membrane itself is built up of a lipid
lular targets like mitochondria, where they bilayer that is composed of various lipids, pro-
inhibit, e.g., cellular respiration and induce reac- teins, and sugars. Different phospholipid classes
tive oxygen species (ROS) formation (da Costa are the most abundant molecules, and besides
et al. 2015). their structural function, they play important
Since toxicity of AMPs is in most cases medi- roles in regulating and controlling processes
ated by a non-specific process, bacteria have dif- occurring throughout the membrane (Jobin and
ficulties to develop resistance against AMPs. Alves 2014). Glycosaminoglycans (GAGs) are
However, development of AMP-resistant strains characteristic for mammalian cells, and espe-
is of course inevitable once they have been put cially the presence of heparan sulfate (HS) is
into clinic. Indeed, processes by which microor- thought to be important for CPP-cell interaction.
ganisms have produced resistance mechanisms Consistently it has been noticed that AMPs and
against AMPs have been already reported. CPPs act on two main membrane classes: mam-
Moreover, to survive the bactericidal action of malian (eukaryotic) and prokaryotic ones.
AMPs, bacteria must sense the presence and Moreover, distinct activities for both groups of
adapt accordingly by controlling the expression peptides have been found for tumor cells. All
of genes involved in AMP resistance. Generally, these cells are characterized by certain heteroge-
bacteria try to change the composition of the neities with differences in lipid bilayer composi-
outer or inner membrane, or to modify their cell tion, expression of various specific markers,
wall composition, thus making principal AMP glycosylation profiles at the outer surface layer,
targets less susceptible. In fact, bacterial defense or the presence of cell walls in the case of bacte-
mechanisms often rely on cell wall modifica- ria and will be discussed briefly.
tions, which usually alter the ionic cell wall
potential leading to a reduced AMP binding
(Maria-Neto et al. 2015). Although clear efforts 7.2.1 Eukaryotic Cell Membranes
have been made, more techniques are needed to
fully understand bacterial resistance strategies. Eukaryotic cell membranes are mainly composed
Nevertheless, owing to their remarkable proper- of phospholipids, glycosphingolipids, and choles-
ties AMPs are one of the most promising drug terol. The different lipid species are segregated
candidates in a foreseeable future to overcome into different domains within the lipid membrane.
the alarming rise in microbial drug resistance (da Moreover, whereas the outer phase of the lipid
Costa et al. 2015). bilayer is usually characterized by the presence of
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 97

Fig. 7.1 Sketch of different plasma membrane organiza- peptides by their phase separation and membrane curva-
tions in bacterial, mammalian, and cancer cells. Bacterial ture. However, for all membrane systems, the effects can
outer layers, as well as that of cancer cells, contain more be somehow interchangeable. Peptide orientation and
negatively charged phospholipids and attract cationic aggregation are further important factors for following
AMPs or CPPs. Mammalian outer layers present zwitter- peptide association and partitioning within the membrane
ionic phospholipids and attract cationic or amphipathic

zwitterionic phosphatidylcholine (PC), the inner teins are recruited and clustered. Cholesterol and
layer contains essentially more negatively charged sphingolipids are main components of such
phosphatidylserine (PS) contributing to a charac- domains. With the presence of such lipid raft
teristic asymmetry between outer and inner layer microdomains, it is hypothesized that lipids do
(Fig. 7.1). Additionally, cholesterol is a major not simply act as passive solvent but play a regu-
constituent of eukaryotic plasma membranes and latory role in protein membrane assembly.
regulates, among other things, plasma membrane However, still the raft hypothesis is highly dis-
fluidity and endocytosis. Together with sphingo- cussed owing to the lack of methods for direct
myelin, it occurs in so-­ called liquid-ordered observation of such domains (Sezgin et al. 2017).
domains, a liquid crystalline phase of a lipid For many CPPs an involvement of lipid rafts and
bilayer. Nearly any eukaryotic cell contains a dis- the relevance of cholesterol for their uptake have
tinct membrane organization in membrane been demonstrated (Pae et al. 2014; Watkins et al.
domains, which plays a role in many functional 2009). Cholesterol depletion with methyl-β-
processes related to polarization, ­signal transduc- cyclodextrin significantly affects translocation of
tion, and membrane trafficking. Moreover, this many CPPs across the plasma membrane, either
temporal and spatial compartmentalization of by affecting endocytosis pathways such as mac-
membrane molecules in assemblies is crucial for ropinocytosis or clathrin-/caveolin-­ mediated
membrane function (Simons and Vaz 2004). A uptake or by influencing the overall membrane
number of observations corroborated the idea of fluidity and, thus, direct translocation.
the existence of so-called lipid rafts, relatively Other critical molecules that are exposed at the
ordered subdomains, in which lipids and/or pro- outer surface of the lipid bilayer and important for
98 I. Neundorf

CPP cell entry are glycosaminoglycans and pro- enriched lipid rafts. Therefore, it is likely that
teoglycans. Particularly, heparan sulfate and other cholesterol is not as important in determining the
sulfated GAGs attract cationic CPPs by their neg- selectivity of AMPs toward bacterial membranes
ative charges, thus acting as primary binding site once supposed. However, which exact role cho-
for CPPs. Generally, it is hypothesized that argi- lesterol plays in the toxicity of AMPs or if
nine-rich CPPs are able to associate at the mem- unknown additional factors are necessary has to
brane by bidentate-dependent binding at binding be elucidated in more detail (Brender et al. 2012;
sites or partners, possibly represented by sulfate McHenry et al. 2012). Nonetheless, also prokary-
groups of HS. Furthermore, it has been recently otic cells contain functional membrane microdo-
shown that heparan sulfate proteoglycans or syn- mains (FMMs) such as the lipid rafts in eukaryotic
decans, another group of transmembrane proteo- cells (Lopez 2015). For instance, it has been
glycans, may act as CPP receptors (Chen et al. demonstrated that membrane-bound sensor
2015; Letoha et al. 2010; Kawaguchi et al. 2016). kinases are organized in polyisoprenoid lipid
containing lipid phases. As is with lipid rafts,
Interestingly, although it is relatively clear that
electrostatic interaction is one of the key events these FMMs are able to resist detergent disaggre-
for CPP membrane association, cell surface tar- gation when using a mixture of nonionic deter-
gets mediating this process remain surprisingly gents (Brown 2002). However, the existence of
unknown. Therefore, it is of undisputable need to such lipid subdomains suggests an important role
find suitable techniques and to further elucidate of lipid organization in all domains in life.
the role that distinct membrane constituents play External to the lipid membrane of Gram-­
in CPP uptake mechanisms. positive bacteria exists a thick peptidoglycan
layer, whose major constituent is lipoteichoic
acid (LTA). The structure of LTA varies between
7.2.2 Bacterial Cell Membranes the different species. Owing to its anionic nature
caused by the presence of carboxyl and phosphate
The bacterial cell wall is a complex polymeric groups of the LTA as well as carboxyl groups of
structure with essential roles in defense, survival, the muramyl peptides, a first contact with cationic
and pathogenesis. The outer leaflet of the cyto- AMPs is feasible. In fact, this electrostatic inter-
plasmic membrane of both Gram-positive and action is thought to be the primary mechanism for
Gram-negative bacteria is surrounded by a mesh-­ antimicrobial activity. Following, AMPs promote
like peptidoglycan sacculus (Caveney et al. membrane damage and cell lysis either by mem-
2018). The lipid bilayer of the plasma membrane brane thinning, pore formation, or bilayer disrup-
includes proteins, associated RNA, and the com- tion (Pushpanathan et al. 2013). Gram-negative
mon phospholipids phosphatidylethanolamine bacteria, on the other side, present another outer
(PE), phosphatidylglycerol (PG), and cardiolipin, lipid membrane, which is separated by the peri-
while PC and phosphatidylinositols (PIs) are less plasm. It consists of phospholipids, lipopolysac-
frequent. Cardiolipin and PG are negatively charides, integral membrane proteins, and
charged and contribute to the negative charge of lipoproteins. In a multistep process, AMPs are
the membrane. One additional major difference first attracted by negatively charged groups of the
between eukaryotic and prokaryotic cell mem- lipopolysaccharides, following disruption of the
branes is the existence of cholesterol in eukary- outer membrane to gain access to the periplasmic
otic cell membranes and its complete absence in space. As is the case with Gram-positive bacteria,
bacterial cell membranes (Fig. 7.1). Although it electrostatic attraction by negatively charged
has been supposed that the presence of choles- groups lead then in a further step to AMP mem-
terol suppresses the activity of AMPs, recently, brane binding. However, owing to the more com-
this effect was put into perspective when it was plex structure, Gram-negative bacteria are usually
shown that AMPs significantly disrupt heteroge- more difficult to target, and only few exceptional
neous lipid structures that contain cholesterol-­ AMPs exist that show activity against Gram-
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 99

negative bacteria. Additionally, also Gram- concerning the development of new anticancer
negative bacteria have developed various therapeutics, the field is still in problems owing
mechanisms to resist AMPs, like proteolytic deg- to upcoming resistances and low specificity of
radation of AMPs, shielding of the bacterial sur- currently available drugs. One alternative
face, modification of the bacterial outer approach that has come up during the last years is
membrane, and pumping AMPs in or out of the to use anticancer peptides. Many of these
cell (Gruenheid and Le Moual 2012). sequences derive originally from host-defense
peptides (i.e., AMPs), but also for cell-­penetrating
peptides, an inhibition of cancer cells has been
7.2.3 Tumor Cells observed. Because of their alternative mode of
action including their specificity to cell mem-
Tumor cells are characterized by a phenotypic branes as their primary target site, resistance and
distinct membrane organization compared to cytotoxicity are less likely to occur.
healthy cells. In fact, compared to healthy cells, For AMPs it has become more and more evi-
tumor cells display a higher overall net negative denced that they are more than just alternative
charge on their outer cell surface. This specific antibiotic weapons. Indeed, a lot of studies dem-
phenotype results from a number of key factors onstrate a clear anticancer activity, and many
making cationic and amphipathic peptides sus- anticancer peptides are often based on AMPs
ceptible to tumor cells. One is the presence of an (Patel and Akhtar 2017; Deslouches and Di 2017;
increased amount of anionic phospholipids in the Roudi et al. 2017). Efforts are being made in
outer layer of the plasma membrane (e.g., PS) order to understand the targeting mechanism of
(Fig. 7.1) (Ran et al. 2002). Thus, the natural antimicrobial peptides, which would enable an
asymmetry between the inner and outer mem- improved design. Again, certainly structure plays
brane leaflets is lost in tumor cells. Elevated reac- a central role in their activity, while AMPs adopt
tive oxygen species (ROS) and hypoxia lead to a defined, often alpha-helical, structure when in
dysregulation of phospholipid transporters, sup- presence of cancer cell membranes (Felicio et al.
porting this imbalance in the maintenance of the 2017). Moreover, the activity of anticancer pep-
plasma asymmetry (Baxter et al. 2017). This tides is in most cases driven by their cationic
involves activation of a putative scramblase and charge, while the presence of an amphipathic
inactivation of a putative ATP-dependent phos- helix, i.e., spatial segregation of cationic and
pholipid translocase. Moreover, several anionic hydrophobic residues, seems also to be essential
cell surface glycoproteins are highly expressed in (Roudi et al. 2017). Notably, AMPs trigger dis-
cancers and additionally contribute to an tinct killing mechanisms based on membrane-­
increased level of negative surface charge (Utsugi lytic events or such without membrane-lytic
et al. 1991; Ran et al. 2002). Examples include events. In this way, necrosis occurs after cell
mucins and heparan sulfate proteoglycans, which membrane lysis and apoptosis in the case of
both promote electrostatic interaction of posi- mitochondrial membrane lysis. In both cases the
tively charged peptides at the outer surface of presence of anionic lipids such as PS or cardio-
tumor cells. Furthermore, changes in membrane lipin is indispensable. Binding of AMPs to
fluidity and pH, increased surface area and trans- surface-­exposed PS leads probably to membrane
membrane potential, as well as a higher number depolarization and cell death. Additionally, those
of microvilli contribute to this specific character- anticancer peptides can present other intracellu-
istic of tumor cells. lar targets, either targeting essential cell proteins,
inhibiting angiogenesis, or recruiting immune
7.2.3.1 Anticancer Activity cells to attack cancer cells (Wu et al. 2014).
of Membrane-Active Peptides On the other side, CPPs are usually not cell
Cancer is one of the major causes of death world- selective and, thus, controlled targeting strategies
wide, and although many efforts have been made using internal or external stimuli to selectively
100 I. Neundorf

increase the activity of CPPs at the target site are adverse effects such as high toxicity to healthy
necessary. With respect to tumor targeting, sev- cells and immune response. For this reason it is
eral such strategies have been successfully devel- still necessary to dissociate the toxicity to mam-
oped making use of activatable CPPs, attachment malian cells from antimicrobial/anticancer activ-
of ligands to CPPs that act as address labels, or ity. Moreover, as is the case with all possible
localized hyperthermia, to name only few peptide drug candidates, their high susceptibility
(Raucher and Ryu 2015; Bergmann et al. 2017; to proteases is a major challenge that has to be
Splith et al. 2012). One other alternative targeting tackled. One solution could be the design of
approach for tumor cells is based on the intrinsic modified peptides and peptide conjugates to
properties of cationic CPPs. Their targeting increase selectivity and lower proteolytic degra-
mechanisms rely on the same parameters as those dation (Reinhardt and Neundorf 2016; Feni and
explained for AMP-derived anticancer peptides Neundorf 2017). In this way, some cancer-­
that act by their positive charge on the negatively targeting peptides are now in clinical trials, and
charged surface of tumor cells. Several efforts the future will show if approvals will arise during
have been made, in which such cationic antican- the next years.
cer peptides were successfully used to target and
affect cancerous cells, also in vivo (Szczepanski
et al. 2014; Gronewold et al. 2017). Moreover, 7.3 Experimental Methods
such anticancer CPPs may also have intracellular to Classify Membrane-Active
targets. Thus, pore formation in the presence of Peptides
high electrical potential at the mitochondrial
membrane might be the basis for their activity Studying peptide-membrane interactions con-
(Rodriguez Plaza et al. 2014). stitutes a challenging topic up to now, and
However, for such anticancer peptides, differ- besides all processes that have been already
ent activity levels might exist, and a careful selec- uncovered, their detailed understanding is still
tion, depending, e.g., on the tumor to be treated, elusive. This is partially based on the complex-
has to be performed. For instance, for many anti- ity of the membrane composition, and addi-
cancer peptides, it is shown that the interaction tionally dependent on the favored arrangement
with glycosaminoglycans, HS, and chondroitin of the peptide (also the cargo in the case of
sulfate (CS), which are present on the outer sur- CPPs) when in the presence of the lipid phase,
face, is one of the key steps during their action. and its following membrane insertion at the
However, it was recently demonstrated that HS at same time. Regardless, it is of singular impor-
the outer surface of cancer cells sequesters anti- tance to reveal the biophysical and biochemi-
cancer peptides, in this case bovine lactoferricin, cal processes behind the function of
away from the phospholipid bilayer and thereby membrane-active peptides helping to design
impede their ability to induce cell lysis (Fadnes more potent molecules with tailored function-
et al. 2009). The results let further conclude that alities that may be applied as active therapeu-
poorly differentiated tumors, with low expression tics. Several techniques have been developed
of HS, are more susceptible to treatment with and applied to biological samples or artificial
anticancer peptides, an interesting hypothesis membrane systems in combination with pep-
that should be investigated in in vivo studies. tides. They differ in their sensitivity, resolu-
Additionally, by generating modified versions of tion, and sample preparation and are often
bovine lactoferricin, the cytotoxic activity against combined to gain complementary information.
HS- and CS-expressing tumor cells was regained, Roughly one can probably divide those meth-
demonstrating the need of such detailed structure-­ ods into the three following categories: (i)
activity relationship studies (Fadnes et al. 2011). quantification methods to unravel, e.g., peptide
Other obstacles that have to be faced with for content in cells, (ii) methods to determine
a future application of anticancer peptides are binding constants and affinities when peptides
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 101

interact with lipids, and (iii) visualization tech- Table 7.1 Methods to classify membrane-active
peptides
niques and methods yielding structural infor-
mation to track membrane-­deforming events Method Applied in context of…
after peptide binding. Determination of binding constants, affinities of
peptides to lipids
One focus has to be set on the choice of suit- Isothermal Information about binding
able membrane models to examine the specific calorimetry (ITC) constants, affinity,
lipid-peptide interaction. Various different such thermodynamic parameters
artificial membrane systems are presented, among Differential Phase transition measurements to
those are unilamellar vesicles, mainly giant unila- scanning yield thermodynamic parameters
calorimetry
mellar vesicles (GUVs) and large unilamellar (DSC)
vesicles (LUVs), which represent the most rele- Fluorescence Peptide insertion into membranes,
vant model systems to study membrane structures spectroscopy dye-release assays
and dynamics. Herein, different lipid composi- Surface plasmon Kinetic profiles of membrane
tions are usually tested that mimic either bacterial resonance (SPR) lipid interaction
Quantification methods
membranes (containing mainly PG), healthy
Flow cytometry Association of peptides to
human cell membranes (containing a mixture out membranes, uptake into cells
of PC:PE), or cancer cell membranes (containing Mass Quantification of peptide cellular
a mixture out of PC:PE:PG). By adding sphingo- spectrometry uptake or uptake into liposomes
myelin or cholesterol, the membrane fluidity can (MS)
be further modulated, and by including appropri- Structural studies and visualization methods
Infrared (IR) Secondary structure of peptides in
ately labeled phospholipids, the vesicle mem-
spectroscopy lipid phases
brane can be stained. As such it is possible to Circular Secondary structure of peptides in
visualize membrane deformation processes or dichroism (CD) lipid phase; possible in the
disruption after peptide addition by using fluores- spectroscopy presence of membrane vesicles or
cence microscopy. Furthermore, encapsulating bacteria
Nuclear magnetic Topology and three-dimensional
dyes within the vesicles allows for performing
resonance (NMR) structure of peptides in lipid
dye-release assays, which are easily conducted spectroscopy phases; either solid or solution
using flow cytometry or fluorescence spectros- NMR in combination with
copy. In addition to that, many researchers have artificial membranes or membrane
lipids
made use of giant plasma membrane vesicles
Fluorescence Cellular uptake; interaction with
(GPMVs). GPMVs comprise a biologically more microscopy membrane vesicles
complex model and display a versatile tool to Electron Membrane organization, peptide
study membrane translocation of CPPs in condi- microscopy (EM) distribution at membranes
tions lacking endocytosis processes (Pae et al. Atomic force In-depth membrane structures
2014). Since GPMVs are released from cells after microscopy
(AFM)
chemical induction, their lipid and protein content
resembles that of the plasma membrane of living
cells. Moreover, by applying low temperature, it hand, arginine-rich CPPs translocated dependent
is possible to segregate the membrane of GPMVs on membrane proteinaceous components (Pae
into different co-existing phases, namely, liquid- et al. 2014).
ordered (Lo) and liquid-­disordered (Ld) membrane There are several biophysical techniques
microdomains. By cholesterol depletion it is thus available that help to study membrane-active
possible to determine the influence of these phases peptides and their adsorption, location, and ori-
(ordered/disordered) to the lipid interaction of entation relative to a lipid bilayer (Table 7.1).1
membrane-active peptide. For instance, Pae et al.
demonstrated that amphiphilic CPP crosses more 1
A few examples are listed in the text and in Table 7.1;
efficiently membranes that are partially depleted however, this list is not exhaustive, and also alternative
from cholesterol or are less ordered. On the other methods have been used.
102 I. Neundorf

Circular dichroism (CD) and infrared (IR) spec- For quantifying the intracellular peptide con-
troscopy give information about secondary struc- tent, or the amount of peptides that has crossed
tures of peptides. Nuclear magnetic resonance the membranes of liposomes, mainly fluorescent
(NMR) spectroscopy yields three-dimensional methods, such as flow cytometry, or fluorescent
structure parameters. Particularly with solid-state microscopy or spectroscopy have been applied. A
NMR, it is possible to gain important insights in smart alternative method was offered recently
peptide depth and positioning within model and uses matrix-assisted laser desorption/ioniza-
membranes. Also solution-phase NMR is fre- tion (MALDI) mass spectrometry (MS) (Walrant
quently applied to analyze peptides in the pres- et al. 2013). Since MS is not a quantitative
ence of lipid micelles or other membrane method, an internal standard labeled with a stable
mimetics. Herein, small unilamellar vesicles isotope is added to the sample. Therefore,
(SUVs) are often used as versatile models, since deuterium-­labeled groups are introduced into the
owing to their small size, they possess a large sur- peptide sequence.
face curvature and, thus, differ in their membrane Another method is to perform electrical mea-
topology from those found in natural membranes. surements on bilayers providing a tool for evalu-
Complementary, electron paramagnetic reso- ating the increase in conductance due to the
nance (EPR) spectroscopy is useful to obtain formation of pore structures at low peptide con-
measures about peptide-lipid interaction on the centrations. By determining the ionic current dis-
molecular level (Galdiero et al. 2013; Alves et al. tribution, it is thus possible to analyze the
2010). pore-forming activity according to membrane
Effects on phase transitions, membrane affini- phospholipid composition, for instance, with
ties, and binding of peptides can be determined respect to cholesterol and sphingomyelin con-
using differential scanning calorimetry (DSC), tent. In this way, direct evidence for the binding
isothermal calorimetry (ITC), and surface plas- and pore-forming activity of AMPs in either
mon resonance (SPR). Indeed, SPR has been anionic or zwitterionic bilayers can be delivered
applied in many studies, in which membrane-like (dos Santos Cabrera et al. 2012).
structures are immobilized on sensor chips to
measure molecular interaction with peptides. In
another method internal tryptophan residues of 7.4  rediction of AMP or CPP
P
peptides or fluorophore-labeled peptides or lipids Activity by Bioinformatics
are investigated by fluorescence spectroscopy for and Library Screenings
determining binding constants and partition coef-
ficients. Moreover, membrane leakage or fusion Recently, several methods have been developed
can be followed by this technique. to predict antimicrobial as well as cell-­penetrating
Microscopical techniques, such as electron peptides and to establish a link to their physico-
microscopy, atomic force microscopy, or fluores- chemical properties (Brand et al. 2018; Lee et al.
cence microscopy, are useful to track morpho- 2018). In silico approaches have further contrib-
logical changes of cells or vesicles after binding uted to the design of highly effective engineered
of peptides. Whereas electron and atomic force peptides with cell-penetrating, antimicrobial, or
microscopies live from their high spatial resolu- even anticancer activity (Gautam et al. 2013;
tion, classical fluorescence spectroscopy might Tyagi et al. 2013; Deslouches et al. 2013). For
be limited by the sensitivity of the used dyes and instance, based on the examination of structure-­
low resolution. However, several recent improve- function relationship studies of synthetic pep-
ments like stimulated emission depletion (STED) tides, a library of Arg and Val or Trp composed
super-resolution microscopy have pushed appli- motifs that fold into helical amphipathic struc-
cations in this latter field forward (Vicidomini tures in the presence of lipid membranes were
et al. 2018). developed (Deslouches et al. 2005; Deslouches
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 103

et al. 2013). Additionally, within a recent study, might be helpful for own rational peptide design
physical properties of around 750 CPPs were strategies. Some have a particular focus on clini-
analyzed, and it was concluded that they are cal or patent information or are provided as a
median 14 residues in length and mainly cationic, tool to design novel AMP or CPP sequences. To
the latter promoting their interaction with nega- date only one extensive database dedicated to
tively charged outer plasma membrane CPPs has been developed (Gautam et al. 2013).
constituents. CPPsite has more than 1700 entries, and pep-
To facilitate the understanding of membrane tides can be searched by their sequence, name,
interaction and the discovery of cell-penetrating and source or defined by delivered cargo, modi-
and endosomolytic peptides, Carney et al. fication, or used cell lines. Furthermore, also
reported on a combinatorial library screen with cyclized peptides are included in this collection.
liposomes (Carney et al. 2017). The authors It has to be pointed out that increasing interest
employed an OBOC (one-bead one-compound) lies in the development of cell-permeable, small,
approach that was easily adapted to a variety of often cyclized peptides, which, after cellular
buffer and liposome compositions mimicking entry, interact selectively with proteins. Actually,
cellular biomembranes. By choosing the right inhibitors of protein-protein interactions in cells
lipid/sterol composition and by measuring the indeed count to a rapid developing field in phar-
binding to zwitterionic liposomes, it was thus maceutical research (Kauffman et al. 2015).
possible to discover several novel peptides and to Thus, researchers have to find and determine the
successfully test them for their siRNA delivery physical properties for a peptide that provides it
ability. The presented method is easy to expand with adequate membrane activity promoting
not only in regard of library size but also with binding at the lipid interface and favorable per-
respect to specifying the uptake pathways of the turbance of the membrane structure. Such
identified peptides. parameters may be found within this database
Machine learning enabled antimicrobial pep- and used for further peptide design. Notably, the
tide discovery, and design has been summarized same authors also offer another database includ-
recently by Lee et al. (2016). For realization, the ing FDA-approved proteins and peptides
authors trained computational models on large (THPdb) (Usmani et al. 2017). Herein, some
and high-quality data sets to perform high-­ membrane-active peptides can be recovered, as,
throughput virtual screenings. In principle, such for instance, from the gramicidin family, and
machine learning models fall all under the thus, this database might be a worthwhile exten-
umbrella of quantitative structure-activity rela- sion. The number of available AMP databases is
tionship (QSAR) models. Nicely, they help to relatively large, and only some of them are listed
design novel antimicrobial peptides and to dis- in Table 7.2. They differ mainly in their collec-
cover membrane activity in diverse peptide fami- tion of AMPs concerning source (APD offers
lies. Introduction of interpretable QSAR models mainly natural AMP sequences) or additional
permits also mechanistic understanding of under- including information about, e.g., pharmacoki-
lying processes and provides predictable data netic parameters and therapeutic index. Actually,
about membrane activity. it seems that these databases belonging to AMPs
On the other hand, several databases have do not act as a network but rather in competition.
sprouted during the last years that specifically It is a pity, since by merging and by updating all
compile sequence, structure, function, activity, these data, differences and redundancies in
etc. of CPPs and AMPs and give more compre- entries would be diminished. As a consequence,
hensive information (Table 7.2). Thus, the vast researchers would greatly benefit by a more
majority of those databases offer additional use- comprehensive picture about structure and func-
ful physicochemical parameters, like charge, tion of given peptides or peptide families and
isoelectric point, hydrophobicity, etc., which their cellular mechanisms.
104 I. Neundorf

Table 7.2 Useful databases for membrane-active peptides

Database Description Link Reference
Cell-penetrating peptides
CPPsite Contains around 1700 CPPs along with their http://crdd.osdd.net/ Gautam et al.
structures, delivered cargos, and used cell raghava/cppsite/ (2013)
lines
CPC Scientific Commercial supplier that provides tool to https://www. Gautam et al.
assist in custom CPP design. Adapted to cpcscientific.com/ (2013) and
THPdb and CPPsite resources/ Usmani et al.
cell-penetrating- (2017)
peptide-database/
Antimicrobial peptides
The Antimicrobial Contains 2987 peptide entries from six http://aps.unmc.edu/ Wang et al.
Peptide Database kingdoms, mainly natural AMPs AP/main.php (2016)
(APD) https://omictools.
com/apd-tool
Collection of Design of new AMPs, links databases for http://www.camp. Waghu et al.
Antimicrobial sequence alignment bicnirrh.res.in/ (2016)
Peptides (CAMP)
Database of Manually curated database providing https://dbaasp.org/ Pirtskhalava
Antimicrobial information and analytical resources to et al. (2016)
Activity and Structure develop antimicrobial compounds with high
of Peptides therapeutic index
(DBAASP)
Data repository of Collection of AMPs with special focus on http://dramp. Liu et al. (2017)
antimicrobial peptides patent and clinical information containing cpu-bioinfor.org/
(DRAMP) 17,608 entries (thereof 4833 general AMPs,
12,704 patents)
Other useful databases
THPdb Collection of FDA-approved therapeutic http://crdd.osdd.net/ Usmani et al.
peptides and proteins providing information raghava/thpdb/index. (2017)
on sequence, indication, mechanism of action, html
pharmacodynamics, toxicity, metabolism,
absorption, half-life, etc.

To mention is further that the accuracy of 7.5 Functional and Mechanistic

these in silico tools highly depends on the applied Redundancy of CPPs
template structure, quality of sequence align- and AMPs
ment, and prediction method. Hence, careful
analysis of the data is necessary to allow a better Although CPPs and AMPs share physicochemi-
understanding of physicochemical and functional cal properties, like their often alpha-helical struc-
properties. Nonetheless, these tools are highly ture, cationic charge, and amphipathicity, it is
valuable to expand the knowledge about AMPs still not clear why these peptides display different
and CPPs and to provide new leads for the trans- activities. CPPs have only limited toxicity to
lational design of new-generation antibiotics or eukaryotic cells and the ability to cross cellular
drug transporters. Particularly in view of produc- plasma membranes in both energy-independent
tion costs, concerning future clinical applica- and endocytosis processes. Although many of
tions, such in silico tools offer valuable these share also amphipathic characteristics, the
alternatives to modulate and improve natural overall interfacial hydrophobicity of most CPPs
peptide sequences. is not favorable for spontaneous partitioning into
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 105

membranes. Concluding that the observed activi- study of Wadhwani et al., fusion activity could
ties are far more dependent on the chemical not be correlated with the obtained secondary
structure of a CPP, and that membrane lipid com- structure of membrane-bound peptides. Thus, it
position, applied concentration of CPPs but also was hypothesized that not the particular type of
lipid concentration and other bilayer physical secondary structure might be crucial but rather
properties play important factors for peptide-­ the extent of the conformational change upon
membrane interaction. Indeed, certain threshold membrane binding that correlates with the fusion
values are indispensable for cellular uptake, even activity. It was concluded that the driving force
for direct translocation or for endocytosis. On the for fusion is the energy released when a formerly
other hand, AMPs target and effectively kill bac- disordered, soluble peptide binds to a vesicle and
terial cells. Arguing that the presence of bacterial acquires a secondary structure by H-bond forma-
cells might switch the activity of a CPP, several tion (Wadhwani et al. 2012). Although pre-folded
CPPs have indeed been tested and demonstrated peptides are suggested not to promote much ves-
to have antibacterial activity, too (Splith and icle fusion, this well-defined fold might be none-
Neundorf 2011). Interestingly, the same mecha- theless supportive for other functions of CPPs,
nistic models leading to pore formation, which like lipid-phase interaction and cargo delivery of
have been proposed for AMPs, are postulated for cyclic peptides (Horn et al. 2016; Lattig-­
CPPs when acting on bacterial cells. In many Tunnemann et al. 2011). Also charge distribution
cases, helical structure or hydrophobic content is patterns were found to play a significant role for
tuned by small sequence modifications leading to peptide-membrane interaction, particularly for
the one or other activity. Hereby, substitutions membrane insertion of lytic and cell-penetrating
with positively charged arginine or aromatic resi- peptides (Chen et al. 2017). In fact, charge distri-
dues play an important role (Piotrowska et al. bution alongside the amphipathic helix might be
2017). However, considering their therapeutic the key factor affecting peptide insertion ability
potential, both groups share some disadvantages and thus bilayer disruption. Moreover, surface
that often come along when using peptides such pressure is likely increased by both groups of
as toxicity, stability, and cost issues for their clin- peptides, CPPs and lytic peptides, whereas for
ical and commercial development. the latter the pressure declines owing to molecule
In spite of this observation, one might ask if rearrangements after peptide insertion into the
the distinct membranes (prokaryotic versus membrane.
eukaryotic) modulate their function. In fact, stud- On the other side, many AMPs have mainly
ies using artificial membrane vesicles illustrate deforming effects on the lipid phase, leading
the need of certain phospholipids present at the often to total membrane rupture. The response of
water-lipid interface directing the activity of the AMPs to synthetic lipid compositions can be eas-
peptides in the one or other direction. Typically, ily modulated by the presence of distinct mole-
CPPs and AMPs have only weak interaction with cules like cholesterol, sphingomyelin, or
zwitterionic, synthetic membranes and strongly cardiolipin within the lipid phase. By the pres-
interact with anionic membranes. In addition, ence of cholesterol, eukaryotic membranes can
dependent on the CPP sequence, different effects be mimicked, and usually the activity of AMPs is
after membrane binding can occur, such as vesi- decreased (Reinhardt et al. 2014). More recent
cle aggregation and fusion, curvature formation, findings have already demonstrated that for some
lipid flip-flop, membrane disruption, and many AMPs their lack in cell selectivity might be based
others. Indeed, some AMPs and CPPs share lipid-­ on aggregation behavior in solution and propen-
mixing abilities, concluding that they might be sity to proteolytic degradation. For instance,
capable of triggering membrane fusion in vitro LL-37 exists in equilibrium between monomers
(Wadhwani et al. 2012). However, within this and oligomers in solution and is highly resistant
106 I. Neundorf

to proteolytic degradation when bound to both present at the outer monolayer of eukaryotic
zwitterionic and negatively charged membranes. cells, the selectivity of such peptides targeting
The oligomerized state might support a detergent-­ membrane curvature can be nicely explained.
like effect also in the presence of zwitterionic
membranes (Oren et al. 1999).
One more parameter that is strongly linked to 7.6 Conclusions
membrane insertion activity is the spontaneous
lipid curvature triggering response of membrane-­ Membrane-active peptides are a class of peptides
active peptides. Recently, Koller and Lohner per- with undisputable relevant functions.
fectly discussed how interfacially active peptides Antimicrobial peptides are negotiated as highly
induce membrane curvature and which factors promising new weapons against bacterial infec-
facilitate its formation (Koller and Lohner 2014). tions, while cell-penetrating peptides act as ver-
Membrane curvature is a result of a complex satile delivery tools. Although both groups of
interplay between membrane proteins, lipids, and peptides share physicochemical properties, their
physical forces that are applied to the membrane intrinsic function and mechanism of action can
surface. Nevertheless, membrane phospholipids only hardly be predicted. It is relatively certain
itself have an intrinsic property to adopt planar or that in both cases lipid composition and constitu-
curved lipid molecular shapes leading to the for- tion play major roles in peptide interaction and
mation of positive or negative curvature. Since that the nature of the lipid phase somehow defines
the bilayer contains an asymmetrical lipid distri- the orientation of the peptides. So, size of the
bution, the different physical parameters of, e.g., phospholipid headgroups and bilayer elastic
PE at the inner monolayer and PC at the outer properties and dynamics are important determi-
monolayer, induce curvature to a different extent. nants. On the other side, it is clearly demonstrated
Membrane-active peptides induce curvature after that charge is a further important key to attract
incorporation in a bilayer; however, curvature cationic peptides to membrane surfaces. One way
induction by a peptide may differ for different to achieve peptide selectivity would thus be to
lipid systems. Several parameters play a role, like equip it with positive charge, making binding to
H-bonding, electrostatic repulsion, monolayer anionic bacterial membranes (or to that of
surface area, and lateral pressure. Otherwise tumoral cells) more probable. Additionally, struc-
membrane curvature triggers peptide response, tural rearrangements during peptide-lipid inter-
what becomes noticeable in peptide secondary play that help to bind and to position the peptide
structure formation. Thus, not only is membrane within the lipid bilayer are certainly essential. As
curvature spontaneously changed upon incorpo- already mentioned, folding into amphipathic
ration of a guest peptide, but lipid curvature can helices might stabilize the membrane-bound state
also influence the preference of a given peptide to and increase the probability of the peptide to par-
insert into the host lipid matrix. For instance, the tition into the bilayer. Following, the interaction
orientation of a peptide in the membrane is a key of AMPs or CPPs with lipid bilayers causes either
parameter determining the subsequent mecha- local rupture (AMP) or transient permeation
nism of action (toroidal pore, barrel stave, etc.). (CPP). How one can predict and exactly deter-
Hence, different modes of peptide-lipid interac- mine these two functions has to be one of the
tion can be expected depending on the different main foci within this research field in the future.
cell types, which differ in terms of lipid composi- However, several tools that link those different
tion and as a result in the amount of lamellar and functions are nowadays available, like databases,
non-lamellar phase-forming lipids present in the molecular dynamic studies, and other experimen-
target membrane. PE exhibits a negative sponta- tal classification systems. In this regard, it might
neous curvature and is more prone to membrane be possible to use AMPs as a constructive tem-
disruption by interfacial active peptides. Given plate to design CPPs and, of course, vice versa.
that PE is more abundant in the cytoplasmic Therefore, it is hoped that in the near future, a
membrane of Gram-negative bacteria and less more obvious picture can be drawn to unravel the
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 107

differences of AMPs and CPPs and that this or cell-penetrating? Int J Biochem Cell Biol 83:71–75.
https://doi.org/10.1016/j.biocel.2016.12.011
knowledge will help in the engineering and dis- da Costa JP, Cova M, Ferreira R, Vitorino R (2015)
covery of more efficient and safer peptide Antimicrobial peptides: an alternative for innovative
sequences. medicines? Appl Microbiol Biotechnol 99(5):2023–
2040. https://doi.org/10.1007/s00253-015-6375-x
Deslouches B, Di YP (2017) Antimicrobial peptides with
selective antitumor mechanisms: prospect for anti-
References cancer applications. Oncotarget 8(28):46635–46651.
https://doi.org/10.18632/oncotarget.16743
Alvares DS, Viegas TG, Ruggiero Neto J (2017) Lipid-­ Deslouches B, Phadke SM, Lazarevic V, Cascio M, Islam
packing perturbation of model membranes by pH- K, Montelaro RC, Mietzner TA (2005) De novo gen-
responsive antimicrobial peptides. Biophys Rev eration of cationic antimicrobial peptides: influence of
9(5):669–682. https://doi.org/10.1007/s12551-017- length and tryptophan substitution on antimicrobial
0296-0 activity. Antimicrob Agents Chemother 49(1):316–
Alves ID, Jiao CY, Aubry S, Aussedat B, Burlina F, 322. https://doi.org/10.1128/AAC.49.1.316-322.2005
Chassaing G, Sagan S (2010) Cell biology meets Deslouches B, Steckbeck JD, Craigo JK, Doi Y, Mietzner
biophysics to unveil the different mechanisms of TA, Montelaro RC (2013) Rational design of engi-
penetration internalization in cells. Biochim Biophys neered cationic antimicrobial peptides consisting
Acta 1798(12):2231–2239. https://doi.org/10.1016/j. exclusively of arginine and tryptophan, and their activ-
bbamem.2010.02.009 ity against multidrug-resistant pathogens. Antimicrob
Baxter AA, Lay FT, Poon IKH, Kvansakul M, Hulett Agents Chemother 57(6):2511–2521. https://doi.
MD (2017) Tumor cell membrane-targeting cationic org/10.1128/AAC.02218-12
antimicrobial peptides: novel insights into mecha- Di Pisa M, Chassaing G, Swiecicki JM (2015)
nisms of action and therapeutic prospects. Cell Mol Translocation mechanism(s) of cell-penetrating pep-
Life Sci 74(20):3809–3825. https://doi.org/10.1007/ tides: biophysical studies using artificial membrane
s00018-017-2604-z bilayers. Biochemistry 54(2):194–207. https://doi.
Bergmann R, Splith K, Pietzsch J, Bachmann M, org/10.1021/bi501392n
Neundorf I (2017) Biological characterization of dos Santos Cabrera MP, Arcisio-Miranda M, Gorjao R,
novel nitroimidazole-peptide conjugates in vitro and Leite NB, de Souza BM, Curi R, Procopio J, Ruggiero
in vivo. J Pep Sci Off Publ Eur Pep Soc 23(7–8):597– Neto J, Palma MS (2012) Influence of the bilayer
609. https://doi.org/10.1002/psc.2995 composition on the binding and membrane disrupting
Brand GD, Ramada MHS, Genaro-Mattos TC, Bloch C effect of Polybia-MP1, an antimicrobial mastoparan
Jr (2018) Towards an experimental classification sys- peptide with leukemic T-lymphocyte cell selectiv-
tem for membrane active peptides. Sci Rep 8(1):1194. ity. Biochemistry 51(24):4898–4908. https://doi.
https://doi.org/10.1038/s41598-018-19566-w org/10.1021/bi201608d
Brender JR, McHenry AJ, Ramamoorthy A (2012) Does Fadnes B, Rekdal O, Uhlin-Hansen L (2009) The antican-
cholesterol play a role in the bacterial selectivity of cer activity of lytic peptides is inhibited by heparan
antimicrobial peptides? Front Immunol 3:195. https:// sulfate on the surface of the tumor cells. BMC Cancer
doi.org/10.3389/fimmu.2012.00195 9:183. https://doi.org/10.1186/1471-2407-9-183
Brown DA (2002) Isolation and use of rafts. Curr Fadnes B, Uhlin-Hansen L, Lindin I, Rekdal O (2011)
Protoc Immunol Chapter 11:Unit 11:10. https://doi. Small lytic peptides escape the inhibitory effect of hep-
org/10.1002/0471142735.im1110s51 aran sulfate on the surface of cancer cells. BMC Cancer
Carney RP, Thillier Y, Kiss Z, Sahabi A, Heleno Campos 11:116. https://doi.org/10.1186/1471-2407-11-116
JC, Knudson A, Liu R, Olivos D, Saunders M, Tian Felicio MR, Silva ON, Goncalves S, Santos NC, Franco
L, Lam KS (2017) Combinatorial library screening OL (2017) Peptides with dual antimicrobial and
with liposomes for discovery of membrane active anticancer activities. Front Chem 5:5. https://doi.
peptides. ACS Comb Sci 19(5):299–307. https://doi. org/10.3389/fchem.2017.00005
org/10.1021/acscombsci.6b00182 Feni L, Neundorf I (2017) The current role of
Caveney NA, Li FK, Strynadka NC (2018) Enzyme struc- cell-penetrating peptides in cancer therapy.
tures of the bacterial peptidoglycan and wall teichoic Adv Exp Med Biol 1030:279–295. https://doi.
acid biogenesis pathways. Curr Opin Struct Biol org/10.1007/978-3-319-66095-0_13
53:45–58. https://doi.org/10.1016/j.sbi.2018.05.002 Galdiero S, Falanga A, Cantisani M, Vitiello M, Morelli G,
Chen CJ, Tsai KC, Kuo PH, Chang PL, Wang WC, Galdiero M (2013) Peptide-lipid interactions: experi-
Chuang YJ, Chang MDT (2015) A heparan sulfate-­ ments and applications. Int J Mol Sci 14(9):18758–
binding cell penetrating peptide for tumor targeting 18789. https://doi.org/10.3390/ijms140918758
and migration inhibition. Biomed Res Int:237969. Gautam A, Chaudhary K, Kumar R, Sharma A, Kapoor
https://doi.org/10.1155/2015/237969 P, Tyagi A, Open Source Drug Discovery Consortium,
Chen L, Zhang Q, Yuan X, Cao Y, Yuan Y, Yin H, Ding X, Raghava GP (2013) In silico approaches for design-
Zhu Z, Luo SZ (2017) How charge distribution influ- ing highly effective cell penetrating peptides. J Transl
ences the function of membrane-active peptides: lytic Med 11:74. https://doi.org/10.1186/1479-5876-11-74
108 I. Neundorf

Gronewold A, Horn M, Randelovic I, Tovari J, Munoz active peptides. Bioorg Med Chem 26(10):2708–2718.
Vazquez S, Schomacker K, Neundorf I (2017) https://doi.org/10.1016/j.bmc.2017.07.012
Characterization of a cell-penetrating peptide with Letoha T, Keller-Pinter A, Kusz E, Kolozsi C, Bozso
potential anticancer activity. ChemMedChem Z, Toth G, Vizler C, Olah Z, Szilak L (2010) Cell-­
12(1):42–49. https://doi.org/10.1002/cmdc.201600498 penetrating peptide exploited syndecans. BBA-­
Gruenheid S, Le Moual H (2012) Resistance to anti- Biomembranes 1798(12):2258–2265. https://doi.
microbial peptides in Gram-negative bacteria. org/10.1016/j.bbamem.2010.01.022
FEMS Microbiol Lett 330(2):81–89. https://doi. Liu S, Fan L, Sun J, Lao X, Zheng H (2017) Computational
org/10.1111/j.1574-6968.2012.02528.x resources and tools for antimicrobial peptides. J Pep
Horn M, Reichart F, Natividad-Tietz S, Diaz D, Sci Off Publ Eur Pep Soc 23(1):4–12. https://doi.
Neundorf I (2016) Tuning the properties of a novel org/10.1002/psc.2947
short cell-­ penetrating peptide by intramolecular Lopez D (2015) Molecular composition of functional
cyclization with a triazole bridge. Chem Commun microdomains in bacterial membranes. Chem
(Camb) 52(11):2261–2264. https://doi.org/10.1039/ Phys Lipids 192:3–11. https://doi.org/10.1016/j.
c5cc08938g chemphyslip.2015.08.015
Jobin ML, Alves ID (2014) On the importance of elec- Maria-Neto S, de Almeida KC, Macedo ML, Franco OL
trostatic interactions between cell penetrating pep- (2015) Understanding bacterial resistance to anti-
tides and membranes: a pathway toward tumor cell microbial peptides: from the surface to deep inside.
selectivity? Biochimie 107(Pt A):154–159. https://doi. Biochim Biophys Acta 1848(11 Pt B):3078–3088.
org/10.1016/j.biochi.2014.07.022 https://doi.org/10.1016/j.bbamem.2015.02.017
Kalafatovic D, Giralt E (2017) Cell-penetrating pep- McHenry AJ, Sciacca MF, Brender JR, Ramamoorthy A
tides: design strategies beyond primary structure and (2012) Does cholesterol suppress the antimicrobial
amphipathicity. Molecules 22(11):1929. https://doi. peptide induced disruption of lipid raft containing
org/10.3390/molecules22111929 membranes? Biochim Biophys Acta 1818(12):3019–
Kauffman WB, Fuselier T, He J, Wimley WC (2015) 3024. https://doi.org/10.1016/j.bbamem.2012.07.021
Mechanism matters: a taxonomy of cell penetrat- Neundorf I, Rennert R, Hoyer J, Schramm F, Lobner K,
ing peptides. Trends Biochem Sci 40(12):749–764. Kitanovic I, Wolfl S (2009) Fusion of a short HA2-­
https://doi.org/10.1016/j.tibs.2015.10.004 derived peptide sequence to cell-penetrating peptides
Kawaguchi Y, Takeuchi T, Kuwata K, Chiba J, Hatanaka improves cytosolic uptake, but enhances cytotoxic
Y, Nakase I, Futaki S (2016) Syndecan-4 is a recep- activity. Pharmaceuticals (Basel) 2(2):49–65. https://
tor for Clathrin-mediated endocytosis of arginine-­ doi.org/10.3390/ph2020049
rich cell-penetrating peptides. Bioconjug Chem Oren Z, Lerman JC, Gudmundsson GH, Agerberth B,
27(4):1119–1130. https://doi.org/10.1021/acs. Shai Y (1999) Structure and organization of the
bioconjchem.6b00082 human antimicrobial peptide LL-37 in phospho-
Koller D, Lohner K (2014) The role of spontaneous lipid membranes: relevance to the molecular basis
lipid curvature in the interaction of interfacially for its non-­cell-­selective activity. Biochem J 341(Pt
active peptides with membranes. Biochim Biophys 3):501–513
Acta 1838(9):2250–2259. https://doi.org/10.1016/j. Pae J, Saalik P, Liivamagi L, Lubenets D, Arukuusk P,
bbamem.2014.05.013 Langel U, Pooga M (2014) Translocation of cell-­
Last NB, Schlamadinger DE, Miranker AD (2013) A penetrating peptides across the plasma membrane
common landscape for membrane-active peptides. is controlled by cholesterol and microenviron-
Protein Sci 22(7):870–882. https://doi.org/10.1002/ ment created by membranous proteins. J Control
pro.2274 Release 192:103–113. https://doi.org/10.1016/j.
Lattig-Tunnemann G, Prinz M, Hoffmann D, Behlke J, jconrel.2014.07.002
Palm-Apergi C, Morano I, Herce HD, Cardoso MC Patel S, Akhtar N (2017) Antimicrobial peptides (AMPs):
(2011) Backbone rigidity and static presentation the quintessential ‘offense and defense’ molecules
of guanidinium groups increases cellular uptake of are more than antimicrobials. Biomed Pharmacother
arginine-rich cell-penetrating peptides. Nat Commun 95:1276–1283. https://doi.org/10.1016/j.biopha.2017.
2:453. https://doi.org/10.1038/ncomms1459 09.042
Le CF, Fang CM, Sekaran SD (2017) Intracellular target- Piotrowska U, Sobczak M, Oledzka E (2017) Current
ing mechanisms by antimicrobial peptides. Antimicrob state of a dual behaviour of antimicrobial peptides-­
Agents Chemother 61(4). https://doi.org/10.1128/ therapeutic agents and promising delivery vectors.
AAC.02340-16 Chem Biol Drug Des 90(6):1079–1093. https://doi.
Lee EY, Fulan BM, Wong GC, Ferguson AL (2016) org/10.1111/cbdd.13031
Mapping membrane activity in undiscovered peptide Pirtskhalava M, Gabrielian A, Cruz P, Griggs HL, Squires
sequence space using machine learning. Proc Natl RB, Hurt DE, Grigolava M, Chubinidze M, Gogoladze
Acad Sci U S A 113(48):13588–13593. https://doi. G, Vishnepolsky B, Alekseyev V, Rosenthal A,
org/10.1073/pnas.1609893113 Tartakovsky M (2016) DBAASP v.2: an enhanced
Lee EY, Wong GCL, Ferguson AL (2018) Machine database of structure and antimicrobial/cytotoxic
learning-­enabled discovery and design of membrane-­ activity of natural and synthetic peptides. Nucleic
7 Antimicrobial and Cell-Penetrating Peptides: How to Understand Two Distinct Functions Despite… 109

Acids Res 44(13):6503. https://doi.org/10.1093/nar/ Szczepanski C, Tenstad O, Baumann A, Martinez

gkw243 A, Myklebust R, Bjerkvig R, Prestegarden L
Pushpanathan M, Gunasekaran P, Rajendhran J (2013) (2014) Identification of a novel lytic peptide for
Antimicrobial peptides: versatile biological the treatment of solid tumours. Genes Cancer
properties. Int J Pept 2013:675391. https://doi. 5(5–6):186–200
org/10.1155/2013/675391 Tyagi A, Kapoor P, Kumar R, Chaudhary K, Gautam A,
Ran S, Downes A, Thorpe PE (2002) Increased exposure Raghava GP (2013) In silico models for designing
of anionic phospholipids on the surface of tumor blood and discovering novel anticancer peptides. Sci Rep
vessels. Cancer Res 62(21):6132–6140 3:2984. https://doi.org/10.1038/srep02984
Raucher D, Ryu JS (2015) Cell-penetrating peptides: Usmani SS, Bedi G, Samuel JS, Singh S, Kalra
strategies for anticancer treatment. Trends Mol Med. S, Kumar P, Ahuja AA, Sharma M, Gautam A,
https://doi.org/10.1016/j.molmed.2015.06.005 Raghava GPS (2017) THPdb: database of FDA-
Reinhardt A, Neundorf I (2016) Design and application approved peptide and protein therapeutics. PLoS
of antimicrobial peptide conjugates. Int J Mol Sci One 12(7):e0181748. https://doi.org/10.1371/jour-
17(5):701. https://doi.org/10.3390/ijms17050701 nal.pone.0181748
Reinhardt A, Horn M, Schmauck JP, Brohl A, Giernoth Utsugi T, Schroit AJ, Connor J, Bucana CD, Fidler IJ
R, Oelkrug C, Schubert A, Neundorf I (2014) Novel (1991) Elevated expression of phosphatidylserine
imidazolium salt – peptide conjugates and their anti- in the outer membrane leaflet of human tumor cells
microbial activity. Bioconjug Chem 25(12):2166– and recognition by activated human blood monocytes.
2174. https://doi.org/10.1021/bc500510c Cancer Res 51(11):3062–3066
Rodriguez Plaza JG, Morales-Nava R, Diener C, Schreiber Vicidomini G, Bianchini P, Diaspro A (2018) STED
G, Gonzalez ZD, Lara Ortiz MT, Ortega Blake I, super-resolved microscopy. Nat Methods 15(3):173–
Pantoja O, Volkmer R, Klipp E, Herrmann A, Del Rio 182. https://doi.org/10.1038/nmeth.4593
G (2014) Cell penetrating peptides and cationic anti- Wadhwani P, Reichert J, Burck J, Ulrich AS (2012)
bacterial peptides: two sides of the same coin. J Biol Antimicrobial and cell-penetrating peptides induce
Chem 289(21):14448–14457. https://doi.org/10.1074/ lipid vesicle fusion by folding and aggregation.
jbc.M113.515023 Eur Biophys J EBJ 41(2):177–187. https://doi.
Roudi R, Syn NL, Roudbary M (2017) Antimicrobial pep- org/10.1007/s00249-011-0771-7
tides as biologic and immunotherapeutic agents against Waghu FH, Barai RS, Gurung P, Idicula-Thomas S (2016)
cancer: a comprehensive overview. Front Immunol CAMPR3: a database on sequences, structures and
8:1320. https://doi.org/10.3389/fimmu.2017.01320 signatures of antimicrobial peptides. Nucleic Acids
Sezgin E, Levental I, Mayor S, Eggeling C (2017) The Res 44(D1):D1094–D1097. https://doi.org/10.1093/
mystery of membrane organization: composition, reg- nar/gkv1051
ulation and roles of lipid rafts. Nat Rev Mol Cell Biol Walrant A, Matheron L, Cribier S, Chaignepain S, Jobin
18(6):361–374. https://doi.org/10.1038/nrm.2017.16 ML, Sagan S, Alves ID (2013) Direct translocation of
Sierra JM, Fuste E, Rabanal F, Vinuesa T, Vinas M (2017) cell-penetrating peptides in liposomes: a combined
An overview of antimicrobial peptides and the latest mass spectrometry quantification and fluorescence
advances in their development. Expert Opin Biol Ther detection study. Anal Biochem 438(1):1–10. https://
17(6):663–676. https://doi.org/10.1080/14712598.201 doi.org/10.1016/j.ab.2013.03.009
7.1315402 Wang G, Li X, Wang Z (2016) APD3: the antimicrobial
Simons K, Vaz WL (2004) Model systems, lipid rafts, peptide database as a tool for research and education.
and cell membranes. Annu Rev Biophys Biomol Nucleic Acids Res 44(D1):D1087–D1093. https://doi.
Struct 33:269–295. https://doi.org/10.1146/annurev. org/10.1093/nar/gkv1278
biophys.32.110601.141803 Watkins CL, Schmaljohann D, Futaki S, Jones AT (2009)
Splith K, Neundorf I (2011) Antimicrobial peptides Low concentration thresholds of plasma membranes
with cell-penetrating peptide properties and vice for rapid energy-independent translocation of a cell-­
versa. Eur Biophys J EBJ 40(4):387–397. https://doi. penetrating peptide. Biochem J 420(2):179–189.
org/10.1007/s00249-011-0682-7 https://doi.org/10.1042/BJ20090042
Splith K, Bergmann R, Pietzsch J, Neundorf I (2012) Wu D, Gao Y, Qi Y, Chen L, Ma Y, Li Y (2014) Peptide-­
Specific targeting of hypoxic tumor tissue with based cancer therapy: opportunity and challenge.
nitroimidazole-peptide conjugates. ChemMedChem Cancer Lett 351(1):13–22. https://doi.org/10.1016/j.
7(1):57–61. https://doi.org/10.1002/cmdc.201100401 canlet.2014.05.002
 Synthetic Anti-lipopolysaccharide
Peptides (SALPs) as Effective 8
Inhibitors of Pathogen-Associated
Molecular Patterns (PAMPs)

Wilmar Correa, Lena Heinbockel,

Guillermo Martinez -de- Tejada, Susana Sánchez,
Patrick Garidel, Tobias Schürholz, Walter Mier,
Aline Dupont, Mathias Hornef,
Thomas Gutsmann, Karl Mauss, Günther Weindl,
and Klaus Brandenburg

Abstract Our approach of constructing AMP, called

Antimicrobial peptides (AMPs) are in the synthetic anti-lipopolysaccharide peptides
focus of scientific research since the 1990s. (SALPs), on the basis of inhibiting the inflam-
In most cases, the main aim was laid on the matory action of lipopolysaccharide (LPS,
design of AMP to kill bacteria effectively, endotoxin) from Gram-negative bacteria was
with particular emphasis on broadband focused on the neutralization of the decisive
action and independency on antibiotic toxins. These are, beside LPS from Gram-­
resistance. However, so far no approved negative bacteria, the lipoproteins (LP) from
drug on the basis of AMP has entered the Gram-positive origin. Although some of the
market. SALPs have an antibacterial action, the most

W. Correa · T. Gutsmann W. Mier

LG Biophysik, Forschungszentrum Borstel, Leibniz Radiopharmazeutische Chemie, Universitätsklinikum
Lungenzentrum, Borstel, Germany Heidelberg, Heidelberg, Germany

L. Heinbockel A. Dupont · M. Hornef

Forschungszentrum Borstel, Leibniz Lungenzentrum, Universitätsklinik Aachen, Institute of Medical
Klinische und Experimentelle Pathologie, Microbiology, Aachen, Germany
Borstel, Germany K. Mauss
G. M. -de- Tejada · S. Sánchez Asklepios-Klinik Hamburg-Altona,
Dep. de Microbiologia, Universidad de Navarra, Hamburg, Germany
Pamplona, Spain G. Weindl
P. Garidel Freie Universität Berlin, Institut für Pharmazie,
Universität Halle-Wittenberg, Halle/Saale, Germany Berlin, Germany

T. Schürholz K. Brandenburg (*)

Klinik&Poliklinik für Anästhesiologie und Brandenburg Antiinfektiva GmbH,
Intensivtherapie, Universitätsmedizin Rostock, c/o Forschungszentrum Borstel, Borstel, Germany
Rostock, Germany e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 111

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_8
112 W. Correa et al.

important property is the high-affinity binding have already shown the opposite, namely, that
to LPS and LP, whether as constituent of the bacteria have developed resistance against anti-
bacteria or in free form which prevents the microbial peptides, such as the resistance
damaging inflammation, that could otherwise observed for S. aureus against pexiganan (Fox
lead to life-threatening septic shock. Most 2013; Pfalzgraff et al. 2018b and references cited
importantly, the SALP may inhibit inflamma- therein). The potential mechanisms that might
tion independently of the resistance status of lead to bacterial resistance to AMP are listed in
the bacteria, and so far the repeated use of the Table 8.2.
peptides apparently does not cause resistance These bacterial resistance mechanisms
of the attacking pathogens. include:
In this chapter, an overview is given over
the variety of possible applications in the field • Surface charge modification
of fighting against severe bacterial infections, • Cell surface structure modification
from the use in systemic infection/inflamma- • Active efflux
tion up to various topical applications such as • Protease release
anti-biofilm action and severe skin and soft • AMP sequestering
tissue infections. • Host AMP downregulation

Although a few commercialized AMP or pep-

tide analogous are already available, the pro-
8.1 General Remarks gresses made in the last decades have been
modest (Table 8.3) (Mishra et al. 2017). In fact,
According to the WHO, about 700,000 people the first commercialized AMP was put on the
die every year due to an infection induced by market more than half a century ago.
resistant bacteria. These bacteria have developed One of the most critical clinical pathologies is
resistance against most commonly used antibiot- sepsis, triggered by microbial infection and leads
ics (Molchanova et al. 2017). A number of bacte- to malfunctions of the lungs, kidneys, or other
ria have developed a strong resistance; the WHO vital organs. It is a quite common, potentially
has recently established a priority list, classifying fatal systemic illness with high mortalities (close
bacteria against which antibiotics have to be to 50%) (Andra et al. 2006).
developed (Global priority list of antibiotic-­ The development of therapeutic agents for the
resistant bacteria to guide research, discovery, treatment of sepsis has been fostered for the last
and development of new antibiotics) (Tacconelli four decades with very limited therapeutic suc-
et al. 2018). Table 8.1 lists the most critical bac- cess so far. With few exceptions, the results of
teria and their drug resistance. these clinical trials have been disappointing, and
Since millions of years, host defense peptides, no specific therapeutic agent is currently approved
known as antimicrobial peptides (AMPs), are for the treatment of sepsis. Strategies to improve
present in the animal, plant, and bacterial world, drug development for sepsis were reviewed
to protect the organism. The strengths of these recently by Fink and Warren (2014). Their con-
naturally occurring AMP are (i) the broad spec- clusions to improve upon this dismal record were
trum of biological activity, (ii) the quasi-absence focused on (i) the need for investigators to iden-
of antimicrobial resistance development, (iii) the tify more suitable therapeutic targets, (ii) improve-
possibility of a “cocktail” of AMP with improved ment of the strategies and approaches for selecting
activity, and (iv) an additional LPS-neutralizing appropriate candidate compounds for clinical
activity of some AMP (Boto et al. 2018; development, and (iii) adoption and adjustment of
Brandenburg et al. 2011; Brandenburg et al. the clinical trial design and treatments.
2016). Although it is described that bacterial Therefore, it was necessary to characterize the
resistance against AMP is low, different studies target structures (Brandenburg et al. 2003) and
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 113

Table 8.1 WHO priority pathogens for research and development of new antibiotics (Tacconelli et al. 2018)
Priority classification Pathogen Resistance
1. Criticala Acinetobacter baumannii Carbapenem-resistant
Pseudomonas aeruginosa Carbapenem-resistant
Enterobacteriaceaeb Carbapenem-resistant
Third-generation cephalosporin-resistant
2. High Enterococcus faecium Vancomycin-resistant
Staphylococcus aureus Methicillin-resistant
Vancomycin intermediate and resistant
Helicobacter pylori Clarithromycin-resistant
Campylobacter Fluoroquinolone-resistant
Salmonella spp. Fluoroquinolone-resistant
Neisseria gonorrhoeae Third-generation cephalosporin-resistant,
fluoroquinolone-resistant
3. Medium Streptococcus pneumoniae Penicillin-non-susceptible
Haemophilus influenzae Ampicillin-resistant
Shigella spp. Fluoroquinolone-resistant
a
Mycobacteria (including Mycobacterium tuberculosis, the cause of human tuberculosis) were not subjected to review
for inclusion in this prioritization exercise as they are already a globally established priority for which innovative new
treatments are urgently needed
b
Enterobacteriaceae include Klebsiella pneumoniae, Escherichia coli, Enterobacter spp., Serratia spp., Proteus spp.,
Providencia spp., and Morganella spp.

Table 8.2 Summary of potential bacterial resistance mechanisms to reduce the therapeutic effect of AMP (Sierra et al.
2017)
Bacterial resistance mechanisms against antimicrobial peptides
Possible resistance mechanisms Consequence
Net surface charge modification with an increase of the Reduction AMP attraction, especially relevant for
positive charge potential electrostatic interactions between the host and the
Gram-negative bacteria: main modification of LPS peptide
headgroup charge
Gram-positive bacteria: modification of lipoteichoic acid
Protease: release of proteases, such as metalloproteases or Increased proteolytic degradation of the peptides
serine endopeptidases and/or increase of protease Inactivation of AMP by bacterial proteases strongly
concentration depends on the peptide structure, given a higher
susceptibility to degradation of linear peptides
compared to cyclic peptides containing disulfide
bonds
Modification of cell surface structure: changes the Masking the anionic phosphate group with cationic
phospholipid composition and thus alters the cytoplasmic primary amines and decreasing AMP attraction and
membrane structure and packing characteristics by chemical membrane insertion
modification of specific lipids
For example, amino acylation of phosphatidylglycerol
headgroups
General resistance to AMP Can further reduce susceptibility to antibiotics
Active efflux modulation: efflux pumps belonging to the Increase the AMP transport out of the cell and
RND (resistance-­nodulation cell division) family reduce the AMP concentration inside
transporters in Gram-negative bacteria
Downregulation of host AMP production: by interference Reduction of host AMP concentration
with or suppression of AMP production
AMP sequestration: e.g., of HNP-1 and HNP-2 Reduce the active AMP concentration
sequestration by S. aureus staphylokinase
AMP antimicrobial peptide, LPS lipopolysaccharide
114 W. Correa et al.

Table 8.3 Selected examples of commercialized AMP

INN and molar
mass/gmol−1 Brand names Admin. Therapeutic indication and relevant information
Anidulafungin Eraxis i.v. Semisynthetic echinocandin used as an antifungal drug
MM: 1140 Ecalta Treatment of invasive Aspergillus infections
Pfizer FDA approved in 2006
Bacitracin Baciim top. Mixture of related cyclic peptides disrupting Gram-positive
bacteria by interfering with the call wall and peptidoglycan
synthesis
MM: 1423 Neosporin i.m. Used against skin infections
Available as bacitracin zinc salt and used in “triple” ointment
composed of additionally PMB and neomycin
Used in humans and chickens/turkey
FDA approved: 1948
Colistin Colomycin top. Acute/chronic infections due to sensitive strains of specific
(polymyxin E) gram-negative microorganism
MM: 1155 Coly-Mycin oral Last-in-line treatment
Promixin i.v. Highly toxic
ColistiFlex inhal. Available since 1959
Dalbavancin Dalvance (US) i.v. Second-generation lipoglycopeptide antibiotic; semisynthetic,
belongs to the same class as vancomycin
MM: 1817 Xydalba (EU) One of the few treatments available to patients infected with
MRSA
FDA approved in 2014 for the treatment of ABSSIs, including
MRSA and Streptococcus pyogenes infections
EMA approved in 2006
Daptomycin Cubicin First marketed cyclic lipopeptide
MM: 1621 Solely active against Gram-positive bacteria
Complex skin and skin structure infections caused by
susceptible strains of Gram-positive bacteria
Enfuvirtide Fuzeon s.c. Antiretroviral drugs for the treatment of AIDS/HIV-1 infection
MM: 4492 Roche FDA approved in 2003 as the first HIV fusion inhibitor
Oritavancin Orbactiv i.v. Semisynthetic glycopeptide for the treatment of serious
Gram-positive bacterial infections
MM: 1793 Eli Lilly FDA approved in 2014 for treatments of skin infections
EMA approved in 2015 for the treatment of acute bacterial
skin and skin structure infections in adults
Teicoplanin Targocid (Sanofi) i.v. A semisynthetic glycopeptide antibiotic used in the
MM: 1564–1908 Ticocin (Cipla) i.m. prophylaxis and treatment of serious infections caused by
Gram-positive bacteria, incl. MRSA and Enterococcus
Telavancin Vibativ i.v Bactericidal lipoglycolipid for use in MRSA or other
Gram-­positive infections
MM: 1756 Telavancin: semisynthetic derivative of vancomycin
FDA approved in 2009 for cSSI
Vancomycin Vancocin (as i.v. Glycopeptide antibiotic
MM: 1449 Hydro chlorid, Eli top. Treatment for complicated skin, bloodstream, endocarditis
Lilly) bone and joint infections, and meningitis caused by susceptible
strains of methicillin-resistant (beta-lactam-resistant)
Staphylococci
First sold in 1954
ABSSSIs acute bacterial skin and skin structure infections, Admin. administration, cSSI complicated skin and skin struc-
ture infections, EMA European Medicines Agency, FDA Food and Drug Administration, i.m. intramuscular, inhal.
inhalation, i.v. intravenous, MM molar mass expressed in g/mol, MRSA Methicillin-resistant Staphylococcus aureus,
PMB polymyxin B, s.c. subcutaneous, top. topical
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 115

Table 8.4 Pharmaceutical challenges for AMP binding structure of the Limulus anti-LPS factor
Pharmaceutical challenges (LALF), also called endotoxin-neutralizing pro-
Physical and chemical degradation tein (ENP) in recombinant form. The LPS-­
Biological and enzymatic degradation binding polypeptide of LALF22 (LALF31–52), a
Colloidal instabilities and the formation of associates
polypeptide with the sequence
and aggregates
HYRIKPTFRRLKWKYKGKFWCG, was taken
Solubility
Ability to generate conformers as starting template. Multiple structural variants,
Limited half-life all in a cyclated form bridged by N- and
Fast elimination C-terminal cysteine linkages, have been analyzed
Reduced membrane permeability with respect to their antimicrobial and anti-­
Limited or complex drug delivery and overall drug inflammatory activity, respectively, all in a
product stability cyclized form bridged by the N- and C-terminal
Synthesis/expression, purification and manufacturing cysteine linkages (Andra et al. 2004, 2007).
(CMC issues)
Costs
Some of these peptides were able to significantly
inhibit the LPS-induced cytokine production in
human mononuclear cells. The most active of
understand the biophysics of the target-AMP these compounds, however, needed a 100- to
interaction (Andra et al. 2005; Howe et al. 2007; 1000-fold excess molar ratio [Peptide]:[LPS] to
Brandenburg et al. 2005, 2010, b; Kaconis et al. act with high affinity and thus did not fulfill the
2011). A deeper understanding of the mechanism claim of extreme selectivity.
of action of current peptide and new candidates is These findings were the starting point of sys-
still necessary (Andra et al. 2007; Garidel and tematic analysis of the structure-activity relation-
Brandenburg 2009; Mirski et al. 2017). ships with sequence variations of the N- and
Optimization of naturally occurring host defense C-terminal ends of the peptides, in a way that the
AMP is performed via, e.g., structure-activity N-terminal end was assigned to have a polar/posi-
relationship studies (Mishra et al. 2017; Sierra tive charge and the C-terminal end a hydrophobic
et al. 2017). character. In fact, it was assumed that binding of
Peptide molecules are in general quite labile the peptide’s N-terminal end to the negatively
structures. In order to develop a therapeutic drug, charged backbone of the lipid should be the most
a number of pharmaceutical challenges, such as adequate way to counteract the lipid A portion of
stability, formulation, and drug delivery, have to LPS (Gutsmann et al. 2010). For an optimal bind-
be addressed (Katarzyna and Małgorzata 2017; ing and inhibition of the LPS-­induced cytokine
Nordström and Malmsten 2017; Kuhlmann et al. production, the exact number and position of each
2018). Table 8.4 summarizes the most prominent amino acid (aa) within the polypeptide plays a
pharmaceutical development challenges that decisive role for optimizing the neutralizing
have to be undertaken in order to develop a suc- activity. In Fig. 8.1, the SALP Pep19-2
cessful therapeutic drug. (GCKKYRRFRWKFKGKFWFWCG), Pep19-8
In order to reduce resistance, synthetic antimi- (GRRYKKFRWK FKGRWFWFG), Pep19-
crobial peptides are under investigations as anti-­ 2.5KO (KFGKWRFGKYRFCWKFRGWK), and
lipopolysaccharide peptides (Habersetzer et al. Pep19-2.5 (GCKKYRRFRWKFKGKFWFWG)
2013; Garidel et al. 2007). were compared in their ability to inhibit the inflam-
mation signal. Cleary, the compound Pep19-2.5,
called Aspidasept®, has by far the strongest anti-
8.2 Development of SALP LPS activity. It is remarkable that the omittance of
only the cysteine at the C-terminal end of Pep19-2,
The development of synthetic anti-­leading to Pep19-2.5, was responsible for the dra-
lipopolysaccharide peptides (SALPs) was based matic increase in neutralizing action. In the same
on a systematic sequence variation of the LPS-­ way, it becomes clear that the exact sequence is
116 W. Correa et al.

Fig. 8.1 LPS-induced

secretion of tumor
necrosis factor α at three
different concentrations
100, 10, and 1 ng/ml in
the presence of the four
peptides Pep19-2,
Pep19-8, Pep19-2.5KO,
and Pep19-2.5 at
different weight ratios.
(From Gutsmann et al.
2010, reproduced with
permission)

important, since the scrambled version of Pep19- analysis of the LPS aggregate structure by small-
2.5, Pep19-­2.5KO, with the same aa but in an arbi- angle X-ray scattering (SAXS) with synchrotron
trary sequence order, shows only activity at high radiation as well as with freeze-fracture transmis-
excess weight ratio. Similarly, Pep19-8, which has sion electron microscopy (FFTEM), the use of a
a similar physicochemical characteristic as Pep19- Zetasizer for determination of the surface charges
2.5, loses dramatically the ability to reduce the (Zeta potential) of LPS layers, the analysis of the
inflammation signal in comparison to the latter. gel to liquid crystalline phase transition of the
Therefore, from these and other results, it was con- methylene groups of the lipid A portion of LPS by
cluded that Pep19-2.5 represents the optimum Fourier-transform infrared spectroscopy (FTIR),
sequence for LPS neutralization and its binding and the secondary structure of the peptide using
constant to LPS was determined by isothermal also FTIR. It was found that LPS in the presence
titration calorimetry (ITC) to 2.8 × 108 M−1, of the peptide is converted from a non-­lamellar
(Fig. 8.2). Further sequence variations did not lead bilayered aggregate structure into a multilamellar
to further improvements (Kaconis et al. 2011). form, as evidenced by SAXS and FFTEM, con-
Besides ITC, further techniques were applied nected with an exothermic reaction between LPS
to get a complete characterization of the LPS-­ and peptide with saturation characteristic (Fig. 8.2)
peptide binding. These investigations comprise the and increase of the Zeta potential of LPS from
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 117

Fig. 8.2 Isothermal Time (min)

calorimetric titration of
LPS from Salmonella 0.05 mM HL 165 R60 + 25x3µl 1mM Lpep 19-2.5(neu) 37°C
minnesota strain R60 0
with Pep19-2.5

Power (ucavsec)
(Aspidasept) indicating
an exothermic reaction
of the two compounds -1
with saturation, which
takes place at a molar
ratio [Pep19-2.5]:[LPS]
0.3, i.e., three peptide -2
molecules bind and
neutralize ten LPS 5
(k cal/mole of injectant)

molecules. (Adapted
Enthalpy change

from Kaconis et al. 0

2011)
-5

-10

-15

-20
0.0 0.2 0.4 0.6 0.8 1.0
[Pep19-2.5]:[LPS Ra] / molar ratio

−45 mV to +10 mV in saturation. The presence of polymicrobial sepsis using cecal ligation and
the peptide leads to a fluidization of the lipid A puncture (CLP) to induce septic cardiomyopathy
acyl chains as seen by FTIR. In contrast, the ß-like and (ii) an in vitro model of cardiomyocytes
structure of the peptide’s secondary structure is not exposed to human sepsis serum for a translational
changed during the binding process. approach. After discovering a Pep19-2.5-induced
Interestingly, much less active compounds weakening of cardiomyopathy caused by sepsis,
such as Pep19-8 neutralize the surface charges in the effects of the peptide on SERCA2 expression
the same way or even better than Pep19-2.5, and were then analyzed (Martin et al. 2015, 2016). It
this is similarly true for the fluidization of the was assumed that the administration of Pep19-­2.5
acyl chains of lipid A. From these data, it can be may attenuate the cardiac dysfunction in murine
concluded that as physical determinants of anti-­ polymicrobial sepsis through regulating SERCA2
LPS activity, a high-affinity exothermic reaction expression. Actually, the infusion of Pep19-2.5
associated with a change of the supramolecular reduced the impaired systolic and diastolic con-
aggregate structures of LPS into a multilamellar tractility and improved the survival time in poly-
organization is necessary and sufficient for its microbial sepsis (Fig. 8.3).
neutralization. It is important to note that the survival benefit
was obtained without any antibiotics by continu-
ous intravenous administration of Pep19-2.5.
8.3  ALPs Inhibit Sepsis-Induced
S Furthermore, preservation of cardiac function in
Cardiac Dysfunction sepsis by Pep19-2.5 was associated with inhibi-
tion of the activation of NF-κB and of Akt/eNOS
An impairment of cardiac function is a key fea- survival pathways. Most interestingly, the pep-
ture of the cardiovascular failure associated with tide prevented the downregulation of SERCA2
sepsis. A study was designed to evaluate the expression in (i) murine heart samples from mice
effects of Pep19-2.5 in (i) a murine model of with sepsis and (ii) in cardiomyocytes exposed to
118 W. Correa et al.

intestinal tissue inflammation in the context of

chronic inflammatory bowel diseases or infection
with enteropathogens. This was suggested by the
observation that cryptdins, endogenous antimi-
crobial peptides secreted into the gut lumen by
small intestinal Paneth cells, were concentrated in
the enteric mucus layer, where they bound to the
surface of colonizing commensal bacteria pre-
sumably reducing their inflammatory potential
(Dupont et al. 2014, 2015). Similar to endogenous
cryptdins, the synthetic peptide Pep19-2.5 was
shown to bind to the bacterial cell surface of a
number of both, commensal and pathogenic, bac-
Fig. 8.3 Effect of cecal ligation and puncture (CLP) and
teria and significantly reduce their viability
treatment with Pep19-2.5 on survival rate of 2-month-old in vitro. In addition, it efficiently reduced the
male NMRI mice (2.0 μg/h in saline 0.9%) or vehicle immune stimulatory potential of both heat-killed
(100 μl/h saline 0.9%). The following groups were stud- and viable Gram-negative and Gram-positive bac-
ied: CLP + vehicle (n = 12); CLP + Pep19-2.5 (n = 12).
Results were globally analyzed by means of Kaplan-­
teria on small intestinal epithelial cells. Following
Meier survival analysis for n number of observations. oral application, rhodamine-labeled Pep19-2.5
(Reproduced from Martin et al. 2015) could be visualized within the proximal small
intestinal mucus layer where it colocalized with
serum from septic shock patients. From these LPS aggregates. Using small-angle X-ray scatter-
data, it can be anticipated that Pep19-2.5 is able ing (SAXS) analysis, we noted that Pep19-2.5
to prevent downregulation of cardiac SERCA2 bound to and converted LPS to a multilamellar
expression in patients with sepsis. This can in structure. This phenomenon was also observed in
turn improve cardiac function and outcome in a mucus matrix suggesting a similar effect also
these patients. in vivo. We therefore hypothesized that Pep19-2.5
would reduce the concentration of active LPS
molecules present within the intestinal mucus
8.4 SALPs Inhibit Inflammation layer, in close proximity of the intestinal epithe-
in Intestinal Cells lial cell surface, and reduce the overall inflamma-
tory potential (Dupont et al. 2015). Therefore, we
In addition to the systemic route, also local next tested the influence of oral Pep19-2.5 admin-
application of synthetic anti-lipopolysaccharide istration on mice kept under homeostatic condi-
peptides to outer or inner body surfaces might tions. Whereas the total bacterial load was
provide significant health benefits. Indeed, the unaffected in both the small and the large intestine
local application of SALP derivatives to the skin following daily oral administration of 100 μg of
was already successfully applied (see later para- Pep19-2.5 for 4 days, a significant alteration of
graph). Likewise, Pep19-2.5 was shown to the colonic microbiota composition was noted.
inhibit lipoprotein- and lipopolysaccharide- The oral administration of Pep19-2.5 to mice
induced cytokine secretion by keratinocytes, infected with Salmonella enterica sv.
fibroblasts, and dendritic cells and to promote Typhimurium (S. typhimurium) had no effect on
wound healing (Pfalzgraff et al. 2016). Further bacterial organ counts or the total mouse body
analysis showed that Pep19-2.5 retained its anti-­ weight. However, consistent with its potent anti-
lipopolysaccharide activity in a cream formula- inflammatory activity, oral Pep19-2.5 administra-
tion (Kuhlmann et al. 2018). tion reduced the local mucosal pro-­inflammatory
In principle, the potent anti-inflammatory and response, as witnessed by a lower, albeit not sig-
antibacterial activity of SALP might also reduce nificant, epithelial Cxcl2 mRNA expression in the
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 119

small intestine. Peptide degradation and binding peptides could protect mice when their adminis-
to bacterial surfaces within the gut lumen might tration was delayed 30 min or 60 min with respect
have reduced the fraction of bioactive peptide at to the LPS challenge (Heinbockel et al. 2013).
the mucosal surface. Future work should aim at These experiments were performed with Pep
improving peptide stability in the intestine and its 19-2.5, the peptide showing the highest LPS-­
targeted release at a close proximity to the muco- binding activity in vitro. Interestingly, this com-
sal surface. Oral Pep19-2.5 administration might pound conferred a very significant level of
then be able to greatly influence mucosal inflam- protection when given 30 min after LPS chal-
mation, wound healing, and microbiota composi- lenge, whereas it exhibited no therapeutic activity
tion and thereby exert a beneficial effect on the when administered 60 min post-challenge. In the
infected host. same report, Pep 19-2.5 appeared to possess a low
half-life in vivo, since a single dose of the com-
pound administered 30 min prior to the endotoxin
8.5  ALPs Efficiently Neutralize
S challenge gave no protection to mice. However,
LPS In Vivo keeping in mind that antisepsis treatments are not
given as bolus but as intravenous infusion, this
The first evidence that SALPs could neutralize observation does not detract from the clinical
LPS in vivo was reported by Gutsmann and col- application of SALPs. Actually, Schuerholz and
laborators (Gutsmann et al. 2010) using a mouse collaborators demonstrated that Pep 19-2.5 effi-
model of endotoxemia induced by intraperitoneal ciently protects against sepsis when given as infu-
(i.p.) co-inoculation of LPS and galactosamine. sion using a mouse model of cecal ligation and
Immediately after this endotoxin challenge, a puncture (Schuerholz et al. 2013).
single dose of the treatment (i.e., the peptide) was
administered as bolus by the same route.
Although this model seems rather unsophisti- 8.6  ALPs Neutralize PAMPs
S
cated, it allowed the authors to rank SALPs from Gram-Positive Bacteria
according to their endotoxin-neutralizing ability. In Vivo
Thus, whereas Pep 19-2, Pep 19-2.5, Pep 19-12,
and Pep 19-2.2 were as potent as polymyxin B Since only approximately 50% of sepsis cases
(PMB), Pep 19-5 and Pep 19-8 granted no detect- result from Gram-negative origin, we extended
able protection and a third group of peptides, the investigations also to the interaction of
exemplified by Pep19-4, and conferred an inter- Pep19-­2.5 with various germs of Gram-positive
mediate level of protection against endotoxemia. origin (Tabah et al. 2012). Surprisingly, Pep19-­2.5
Using the same mouse model, Martinez de exhibited a better antimicrobial activity against a
Tejada and collaborators confirmed the high anti- variety of Gram-positive source (e.g., Bacillus
endotoxic potency of Pep 19-2, Pep 19-12, and megaterium, Streptococcus agalactiae,
Pep 19-2.2, while showing that shorter deriva- Clostridium perfringens) as compared to Gram-­
tives of this peptide (Pep 17-1 and Pep 17-2) negative source, such as E. coli (EPEC),
granted almost no protection (Martínez de Tejada Salmonella enterica, and Stenotrophomonas
et al. 2012). Since Pep 17-1 and Pep 17-2 lack the maltophilia (Dupont et al. 2015). These data sug-
N-terminal polar residue characteristic of Pep 19 gest a hint that Pep19-2.5 could also be efficient
series, this observation suggests the importance against the Gram-positive inflammation-inducing
of this structural feature for anti-LPS activity. An toxins, which were described to be peptidogly-
alternative (although compatible) explanation is cans (PG), lipoteichoic acids (LTA), and lipopro-
that peptides must have a minimum length to teins/peptides (LP). In a systematic study, the
guarantee appropriate binding to LPS. influence of Pep19-2.5 on the stimulatory capac-
To better characterize the therapeutic potential ity of these compounds was analyzed. Out of all
of SALPs, Heinbockel and collaborators tested if these compounds, only LP induced a strong cyto-
120 W. Correa et al.

kine activity in human mononuclear cells, and the amoxicillin, erythromycin, ciprofloxacin, imipe-
peptide inhibited the cytokine secretion consider- nem, ceftriaxone, and tetracycline, with
ably. Then, using the above-mentioned animal Pep19-­2.5 was checked in the mouse model for
model, Martinez de Tejada and collaborators bacteremia. For all combinations the animals had
showed that Pep19-2.5 efficiently neutralizes a high level of TNFα in the absence of the pep-
FSL-1 (fibroblast-stimulating lipopeptide) in vivo tide, and this level decreased considerably when
(Fig. 8.4) (Martinez de Tejada et al. 2015). This Pep19-2.5 was added. This means that for anti-
synthetic compound is homolog to the N-terminal sepsis therapies, antibiotics alone are not suffi-
part of a lipoprotein from Mycoplasma salivar- cient to control the inflammation. This may be
ium. Notably, the level of protection against lethal the reason that the treatment of sepsis patients
toxemia (80%) was achieved with a single bolus with antibiotics alone frequently fails due to the
of Pep 19-2.5 and lasted for the entire time of inability to reduce inflammation.
monitoring (at least 4 days). Researchers had pre-
viously shown that the inoculation of FSL-1 into
animals caused the characteristic symptoms of 8.8  ode of Action of SALP
M
septic shock in a dose-dependent manner. In the Against PAMPs
same report, this group demonstrated that lethal-
ity due to FSL-1 was associated with high serum Summarizing these data, the mode of interaction
levels of TNF-­ alpha and IL-6. Importantly, of the peptide with LPS can be described as a
administration of Pep 19-2.5 to animals chal- Coulomb interaction of the positively charged aa
lenged with the lipopeptide completely inhibited of the positively charged residues (R and K) of
secretion of TNF-­alpha and caused a slight reduc- the N-terminal region of the peptide with the neg-
tion of IL-6. It is worth noting that animals only atively charged phosphate and carboxylates
received the lipopeptide (i.e., without galactos- within the LPS head group in a first step, fol-
amine) in this set of experiments. lowed by a folding of the nonpolar hydrophobic
interaction of the C-terminal region of the pep-
tide into the lipid A acyl chain moiety. It should
8.7 SALPs Block Cell-Bound be mentioned that the latter step is absolutely
Toxins necessary, since a shortened variant of the pep-
tide lacking the last 5 aa of the C-terminal end
Heinbockel et al. (2013) extended the preclinical was unable to bind and neutralize LPS effec-
investigations of compound Pep19-2.5 to the tively. The final result of this two-step mecha-
direct interaction of this peptide with Gram-­ nism is a kind of sequestering of LPS, similarly
negative (Salmonella) and Gram-positive what has been observed for lipopolyamines (Sil
(methicillin-­resistant Staphylococcus aureus) et al. 2013). This leads to a blocking of the bind-
bacteria (Heinbockel et al. 2013). They found ing of LPS to their cell receptors as outlined in
similar exothermic reactions of Pep19-2.5 with Fig. 8.5. As illustrated, SALPs are able to block
both bacterial species including saturation char- the bacterial toxins as constituents of the bacte-
acter, which was indicative that a binding of rial cells or in free form and thus inhibit the over-
Pep19-2.5 took place also with bound bacterial bordering immune reaction.
toxins. The bacterial binding was correlated with
the suppression of an immune reaction, i.e., a
decrease of cytokine secretion. Also, the addition 8.9 SALP Synergizes
of Pep19-2.5 to surgically remove human lung with Antibiotics In Vivo
tissue stimulated with LPS and with MRSA was
again able to completely inhibit this immune Candidate therapies against sepsis are not
reaction. Finally, in the study the therapeutic effi- expected to replace antibiotics but to work in tan-
cacy of combinations of various antibiotics, i.e., dem with these latter drugs. In this context, any
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 121

Fig. 8.4 (a) Serum levels of tumor necrosis factor α test). (b) Survival rate of mice intraperitoneally inocu-
(upper panel) and interleukin-6 (lower panel) in a group lated with FSL-1 (open circles; n = 6) or receiving FSL-1
of BALB/c mice (n = 4) intraperitoneally (i.p.) inoculated and then treated with Pep19-2.5 (solid squares; n = 6). All
with 40 μg of FSL-1 (central bars) compared to another the animals received FSL-1 (4 μg/mouse) co-adminis-
group receiving i.p. 400 μg of Pep19-2.5 (Aspidasept®) tered with D-galactosamine (18 mg/mouse) intraperito-
immediately after an identical FSL-1 challenge (right neally. A group of animals was treated with Pep19-2.5
bars; n = 5) and a third group administered only with (200 μg/mouse) immediately after FSL-1 challenge at a
pyrogen-free saline i.p. (vehicle; left bars; n = 5). TNFα different site of the peritoneum, and animal mortality was
(top) and IL-6 (bottom) levels were measured at 1.5 h or monitored every 4 h for 4 days. Results were globally
4 h after challenge, respectively. Double asterisks denote analyzed by means of a Kaplan-Meier survival analysis.
significant statistical differences between the two groups Asterisk denotes significant statistical differences
indicated by the bracket (p < 0.01; Mann-Whitney U (p < 0.05)
122 W. Correa et al.

Fig. 8.5 Summary of the therapeutic action of Pep19-2.5 The SALPs have no direct influence on the cytokine pro-
by blocking the PAMPs of bacteria, LPS from Gram-­ duction of the immune cells and thus do not impair their
negative and LP from Gram-positive origin, and convert- function
ing the excessive immune response into a normal reaction.

mutual enhancement of activity between antimi- uninfected animals. Interestingly, the peptide was
crobials and antisepsis compounds would be shown to efficiently neutralize the endotoxin
highly advantageous. Precisely, this is what induced by ceftriaxone in vivo and to reduce
Barcena-Varela and collaborators recently serum levels of IL-6 and TNF-alpha.
showed using a rabbit model of bacteremia Prior to this report, Heinbockel and collabo-
induced by the intravenous inoculation of live rators made similar observations using a mouse
Salmonella enterica serovar Minnesota cells model of peritonitis induced by i.p. injection of
(Barcena-Varela et al. 2017). This rabbit model live Salmonella enterica serovar Minnesota
reproduces symptoms associated with human cells. Specifically, these researchers showed
sepsis including leukopenia, hyperlactatemia, that administration of Pep19-2.5 combined
hyperglycemia, hypothermia, splenomegaly, and with antibiotics efficiently protected mice
hyperproduction of TNF-alpha and IL-6. against sepsis (Heinbockel et al. 2013). Notably,
Barcena-Varela and collaborators demon- neither the antibiotic nor the peptide by them-
strated that a combination of Pep19-2.5 and cef- selves had any therapeutic efficacy. Their
triaxone administered intravenously to the rabbits experimental setting allowed the authors to
killed bacteria and eliminated bacteremia more conclude that the components responsible for
rapidly than any of the components of the combi- the bactericidal and PAMP-­neutralizing activi-
nation when given alone. In addition, the com- ties were the antibiotic and the peptide, respec-
bined treatment was the only one capable of tively. However, synergism between the SALP
reverting hypothermia and giving rise to temper- and the antibiotics was not observed, contrary
ature values indistinguishable from levels in to the previous case.
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 123

8.10  ALPs Synergize with Anti-­

S tion of Pep 19.2-5, antibiotics, and NSAIDs in
inflammatory Agents In Vivo human sepsis therapy.

Although the use of nonsteroidal anti-­

inflammatory drugs (NSAIDs) as part of sepsis 8.11 SALPs Inhibit Intracellular
treatment is under discussion, a survival benefit LPS Responses
after NSAD therapy has been reported in several
animal models of endotoxemia. These anti-­ The LPS-induced sepsis via TLR4 was thought
inflammatory drugs inhibit the activity of to be the critical signaling pathway to drive sep-
prostaglandin-­synthesizing enzyme cyclooxy- sis. However, a novel intracellular LPS sensing
genase-­2 (COX-2) and most of them also COX-1. mechanism, which proved to be highly relevant
On the other hand, due to its ability to neutralize for LPS-triggered sepsis independently of TLR4,
PAMPs, Pep19-2.5 is known to prevent Toll-like has been identified, providing a new paradigm
receptor TLR-4- and TLR-2-dependent activa- for sepsis development. Inflammatory caspases,
tion (Martinez de Tejada et al. 2015). such as caspase-11 in mice and caspase-4/cas-
To test whether these two pro-inflammatory pase-5 in humans, recognize LPS in the cytosol
pathways could be simultaneously inhibited, (Shi et al. 2014; Hagar et al. 2013). Binding of
Heinbockel and collaborators evaluated the anti- LPS to inflammatory caspases induces their
sepsis potential of a combined SALP-NSAID oligomerization, followed by pyroptosis, an
treatment (Heinbockel et al. 2015). For this pur- inflammatory form of cell death, and NLRP3
pose, they used a mouse model of endotoxemia inflammasome- and caspase-1-dependent IL-1β
induced by the inoculation of a high amount of and IL-18 release. LPS-triggered pyroptosis was
LPS without galactosamine. Their experiments found to be the main driver for sepsis develop-
demonstrated that ibuprofen and Pep 19.2-5 acted ment and could explain, at least partially, why
in synergy to protect mice against lethal toxemia. TLR4 inhibitors failed in clinical trials as anti-
Specifically, they showed that neither ibuprofen sepsis drugs (Opal et al. 2013; Rice et al. 2010).
nor Pep19.2-5 protected mice when administered In line with this, the TLR4 inhibitor TAK-242
alone 1 h after the LPS challenge. In contrast, the reduced intracellular LPS-mediated IL-1β
combination of both compounds afforded a high release but failed to reduce pyroptosis (Pfalzgraff
level of protection that lasted for the entire exper- et al. 2017). In contrast, Pep19-2.5 inhibited
imental duration (at least 4 days). Interestingly, intracellular LPS-­induced caspase-1 activation,
authors showed that animals receiving the com- IL-1β production, as well as pyroptosis as dem-
bined treatment had reduced concentrations of onstrated by reduced HMGB (high mobility
both TNF-α and PGE2, compared to mice treated group box) 1 secretion and LDH (lactate dehy-
with either compound. drogenase) release. Since Pep19-2.5 neutralizes
To better characterize this cooperativity at also LP from Gram-­positive bacteria, a similar
the molecular level, the same research group mode of action can be expected for LP such as
performed a transcriptome analysis of human fibroblast-stimulating lipid 1 (FSL-1) which
monocytes exposed to the combined treatment activates the NLRP7 inflammasome in the cyto-
and confirmed downregulation of TNF-α and sol (Khare et al. 2012). Indeed, transfection of
PGE2 coding transcripts. Finally, this experi- THP-1 monocytes with FSL-1 leads to IL-1β
ment showed that other transcripts involved in release which is inhibited by Pep19-2.5 (manu-
inflammatory cascades were affected by the script in revision).
­
combination of ibuprofen and Pep 19-2.5, such While the intracellular detection of LPS by the
as those related with innate immune response, noncanonical inflammasome pathway is now
pattern recognition receptor networks, and well established, it has only recently been discov-
TLR signaling pathways. These observations ered how LPS gains access to inflammatory cas-
highlight the potential of the triple combina- pases in the cytosol since most Gram-negative
124 W. Correa et al.

bacteria are not cytosolic. LPS is delivered into against common antibiotics including topical
the cytosol via outer membrane vesicles (OMVs) agents such as mupirocin and fusidic acid (Serra
(Vanaja et al. 2016) that are secreted by all Gram-­ et al. 2015). Consequently, new treatment options
negative bacteria during cell growth or as an are urgently needed not only for systemic but also
adaptive response to stress such as antibiotic for non-systemic infections. AMPs show impor-
treatment or when exposed to the host. OMVs tant wound healing-promoting activities which
contain several PAMPs found within the parent render them promising therapeutic options for
bacterium including LPS and lipoproteins indi- the treatment of wounds and skin and soft tissue
cating that OMVs potentially mediate pro-­ infections (SSTIs) (Pfalzgraff et al. 2018b). Yet,
inflammatory effects and may contribute to sepsis for topical treatment of noninfected and infected
(Park et al. 2010). SALPs are also able to sup- wounds, AMPs should have low cytotoxicity and
press OMV-induced responses. In macrophages, high stability in the host environment with high
Pep19-2.5 as well as polymyxin B reduced IL-1β salt concentrations and proteases present at the
and TNF release as well as pyroptosis induced by wound site.
OMVs, while TAK-242 suppressed OMV-­ Besides a broad spectrum of antimicrobial
induced TNF and IL-1β secretion, but not pyrop- activity, AMPs may also show immunomodula-
tosis (manuscript in revision). The exact tory properties such as anti-inflammatory actions
underlying mechanisms are not fully understood; which are often physiologically more relevant
however, Pep19.2-5 may act extracellularly by (Hilchie et al. 2013). As mentioned earlier, neu-
directly binding to OMVs or intracellularly by tralization of TLR2- and TLR4-mediated
neutralizing LPS after internalization of OMVs responses induced by cell wall-derived inflam-
and Pep19-2.5. Thus, given the neutralizing mode matory toxins of Gram-positive and Gram-­
of action of SALPs, the peptides inhibit not only negative bacteria has also been confirmed for
extracellular TLR2/TLR4 signaling but also Pep19-2.5 and Pep19-4LF in skin cells such as
intracellular signaling cascades mediated by keratinocytes, dermal fibroblasts, and dendritic
inflammasomes (Fig. 8.6). This holds even true cells (Pfalzgraff et al. 2016).
for more physiological conditions as demon- In addition, natural and synthetic AMPs have
strated by inhibition of OMV-induced activation been demonstrated to induce pivotal processes of
of the inflammasome/IL-1 axis by Pep19-2.5. wound healing such as cell migration and prolif-
eration as well as angiogenesis. Various peptides
promote cell migration, an important mechanism
8.12 SALP for Topical Treatment during reepithelialization through transactivation
of Wounds and Skin of the epidermal growth factor receptor (EGFR).
Infections Pep19-2.5 and Pep19-4LF potently accelerated
artificial wound closure in keratinocytes via puri-
Endogenous AMPs play an important role during nergic P2X7 receptors (P2X7R) leading to cal-
the different stages of wound healing which can cium influx and mitochondrial reactive oxygen
be divided into an inflammatory, a proliferative species (ROS) release followed by
and remodeling phase (Werner and Grose 2003; metalloprotease-­dependent transactivation of the
Mangoni et al. 2016). Inadequate wound healing EGFR and downstream activation of ERK1/
due to infection or bacterial colonization can lead ERK2 (Pfalzgraff et al. 2016, 2018a) (Fig. 8.7).
to the formation of chronic wounds which are Pep19-2.5-mediated keratinocyte migration was
challenging to heal and often remain in the reduced in the presence of the ATPase hexoki-
inflammatory stage (Frykberg and Banks 2015). nase; however, Pep19-2.5 failed to increase extra-
Chronic wounds are mainly polymicrobial and cellular levels of the P2X7R ligand, adenosine
most frequently caused by S. aureus and triphosphate (ATP). This indicates that Pep19-2.5
Pseudomonas aeruginosa which represent a ther- either indirectly activates the P2X7R or increases
apeutic challenge given the increasing resistance the sensitivity of the P2X7 receptor to its ligand
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 125

Fig. 8.6 Pep19-2.5 inhibits signaling pathways induced membrane vesicle. (Adapted from Pfalzgraff et al. (2018a,
by LP and LPS via transmembrane and cytosolic PRRs. b). Use permitted under the Creative Commons Attribution
The signaling cascades lead to inflammation and pyropto- License CC BY 4.0)
sis. LP lipopeptides, LPS lipopolysaccharides, OMV outer

Fig. 8.7 Potential mode of action of Pep19-2.5-induced influx and subsequently mitochondrial ROS is released
keratinocyte migration. Pep19-2.5 may activate the which triggers metalloprotease-mediated EGFR transacti-
P2X7R indirectly to increase sensitivity of the extracellu- vation and ERK1/ERK2 activation. (From Pfalzgraff et al.
lar ligand ATP. Activation of P2X7R leads to calcium (2018a). Reproduced with permission)

ATP at least at low ATP concentrations. Whether via EGFR transactivation, SALPs promote kera-
Pep19-2.5 acts as an allosteric modulator of the tinocyte migration but not proliferation
P2X7R similar to polymyxin B is currently not (Pfalzgraff et al. 2016). It remains to be deter-
known (Ferrari et al. 2004). In contrast to other mined whether Pep19.2-5 modulates other
AMPs such as human beta defensins and melittin wound healing-promoting activities such as
which stimulate cell migration and proliferation induction of angiogenesis.
126 W. Correa et al.

Importantly, Pep19-2.5 accelerated wound mucosa (male patient, 31 years old), which origi-
closure in vivo of noninfected as well as nated from a herpes simplex I virus together with
methicillin-­resistant S. aureus-infected wounds mouth bacteria such as Streptococcus sanguis
in mice (Pfalzgraff et al. 2018a). Thus, our and Streptococcus oralis. In a former report, it
in vitro and in vivo studies indicate that SALPs was found that Pep19-2.5 and some derivatives
may be beneficial for the treatment of nonin- are also active against a particular virus which
fected wounds and polymicrobial wound infec- enters the cells via the heparan sulfate proteogly-
tions. For the latter, combination with an can (HIV, Hep B, and herpes simplex viruses I
antibiotic may further increase therapeutic effi- and II) (Krepstakies et al. 2012). The application
cacy by a synergistic effect with respect to the showed considerable healing already after 2 days,
anti-inflammatory and reepithelialization pro- i.e., much earlier than the normal healing due to
moting activity of SALPs and the direct antimi- body’s own immune system.
crobial effect of the antibiotic. In a second case (female patient, 68 years
old), erythema exsudativum multiforme was
treated successfully. Already 24 h after applica-
8.13  ALPs Are Effective
S tion, a significant relief of symptoms was
Therapeutics in Human observed, and 7 days after treatment, the wound
“Healing Attempts” lesion had disappeared. A therapy with antihista-
minic and cortisone preparations, which had pre-
Finally, for dermal application, a cream formula- viously been used, showed no effect (data to be
tion of Pep19-2.5 and the second lead structure published elsewhere).
Pep19-4LF were manufactured by dispersing the
lyophilized peptide into a DAC base cream
(Deutscher Apotheker Codex) (Kuhlmann et al. 8.14 Conclusions
2018). The investigations comprised data about
the stability of the peptides in the cream in regard In this chapter, an overview is given over the
to time, which included the evaluation of the development of a particular class of AMP, the
extraction procedure, the quantitative analysis of SALP, originally designed to effectively neutral-
the peptides after extraction, its sensitivity to pro- ize bacterial Gram-negative LPS but later shown
tease degradation, and its ability to maintain to act in a broadband manner also against Gram-­
activity against LPS-induced inflammation positive LP. In contrast to other approaches, the
in vitro. The authors showed that Pep19-2.5 was inhibition of the toxin-induced inflammation and
present as a dimer after extraction from the not the direct antimicrobial activity is the basis of
cream, whereas Pep19-4LF retained its mono- SALP action. Therefore, the clinical application
meric form. Both peptides did not show any deg- of the lead structure Pep19-2.5 to combat sepsis
radation by chymotrypsin after extraction, in from polymicrobial origin seems to be a great
contrast to the peptides dissolved in buffer. The chance for the development of an effective medi-
formation of Pep19-2.5 into a dimer structure, cament. This is even strengthened by the results
however, did not decline its ability to inhibit the from the intracellular action of the SALP indicat-
LPS-induced inflammation reaction in human ing also an inhibition of the TLR-4 independent
mononuclear cells. toxin-induced cellular stimulation, with which a
The formulation in DAC base cream was gap is closed which may be responsible for for-
applied in “healing attempts” (Heilversuch mer failed antisepsis approaches using pure
according to German Arzneimittelgesetz §4b). TLR-4 antagonists. Furthermore, the suitability
As example for this, the base cream formulation of the SALP for fighting against chronical
of Pep19-2.5 was applied in a 0.01% crème wounds including severe skin and tissue infec-
­formulation against a coinfection of the mouth tions (SSTIs) seems to represent a big chance for
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 127

their topical administration to resolve an unmet derived from the alpha helical cationic core region of
NK lysin. Biophys Chem 150(1–3):80–87. https://doi.
medical need worldwide, also considering that org/10.1016/j.bpc.2010.01.009
the development of resistance seems to be Brandenburg K, Howe J, Sanchez-Gomez S, Garidel P,
improbable. Roessle M, Andrä J, Jerala R, Zweytick D, Lohner
K, Rappolt M, Blondelle S, Moriyon I, Martinez de
Tejada GM (2010b) Effective antimicrobial and anti-­
endotoxin activity of cationic peptides based on lac-
References toferricin: a biophysical and microbiological study.
Anti-Infect Agents Med Chem 9(1):9–22. https://doi.
Andrä J, Lamata M, Martinez de Tejada G, Bartels R, org/10.2174/187152110790886736
Koch MHJ, Brandenburg K (2004) Cyclic antimicro- Brandenburg K, Andrä J, Garidel P, Gutsmann T (2011)
bial peptides based on Limulus anti lipopolysaccha- Peptide- based treatment of sepsis. Appl Microbiol
ride factor for neutralization of lipopolysaccharide. Biotechol 90(3):799–808. https://doi.org/10.1007/
Biochem Pharmacol 68(7):1297–1307. https://doi. s00253-011-3185-7
org/10.1016/j.bcp.2004.05.054 Brandenburg K, Heinbockel L, Correa W, Lohner K
Andrä J, Lohner K, Blondelle SE, Jerala R, Moriyon (2016) Peptides with dual mode of action: killing
I, Koch MH, Garidel P, Brandenburg K (2005) bacteria and preventing endotoxin-induced sepsis.
Enhancement of endotoxin neutralization by cou- Biochim Biophys Acta 1858(5):971–979. https://doi.
pling of a C12-alkyl chain to a lactoferricin-derived org/10.1016/j.bbamem.2016.01.011
peptide. Biochem J 385(Pt 1):135–143. https://doi. Dupont A, Heinbockel L, Brandenburg K, Hornef MW
org/10.1042/BJ20041270 (2014) Antimicrobial peptides and the enteric mucus
Andrä J, Gutsmann T, Garidel P, Brandenburg K (2006) layer act in concert to protect the intestinal mucosa.
Mechanisms of endotoxin neutralization by synthetic Gut Microbes 5(6):761–765. https://doi.org/10.4161/1
cationic compounds. J Endotoxin Res 12(5):261–277. 9490976.2014.972238
https://doi.org/10.1179/096805106X118852 Dupont A, Kaconis Y, Yang I, Albers T, Woltemate
Andrä J, Howe J, Garidel P, Rössle M, Richter W, S, Heinbockel L, Andersson M, Suerbaum S,
Leiva-Leon J, Moriyon I, Bartels R, Gutsmann T, Brandenburg K, Hornef MW (2015) Intestinal mucus
Brandenburg K (2007) Mechanism of interaction of affinity and biological activity of an orally admin-
optimized limulus derived cyclic peptides with endo- istered antibacterial and anti-inflammatory pep-
toxins: thermodynamic, biophysical and microbiolog- tide. Gut 64(2):222–232. https://doi.org/10.1136/
ical analysis. Biochem J 406(2):297–307. https://doi. gutjnl-2014-307150
org/10.1042/BJ20070279 Ferrari D, Pizzirani C, Adinolfi E, Forchap S, Sitta B,
Barcena-Varela S, Martinez de Tejada G, Martin L, Turchet L, Falzoni S, Minelli M, Baricordi R, Di
Schuerholz T, Gil-Royo AG, Fukuoka S, Goldmann Virgilio F (2004) The antibiotic polymyxin B modu-
T, Droemann D, Correa W, Gutsmann T, Brandenburg lates P2X7 receptor function. J Immunol 173(7):4652–
K, Heinbockel L (2017) Coupling killing to neu- 4660. https://doi.org/10.4049/jimmunol.173.7.4652
tralization: combined therapy with ceftriaxone/ Fink MP, Warren HS (2014) Strategies to improve drug
Pep19-­2.5 counteracts sepsis in rabbits. Exp Mol Med development for sepsis. Nat Rev Drug Discov 13:741.
49(6):e345. https://doi.org/10.1038/emm.2017.75 https://doi.org/10.1038/nrd4368
Boto A, Perez de la Lastra JM, Gonzalez CC (2018) The Fox JL (2013) Antimicrobial peptides stage a come-
road from host-defense peptides to a new generation back. Nat Biotechnol 31:379. https://doi.org/10.1038/
of antimicrobial drugs. Molecules 23(2). https://doi. nbt.2572
org/10.3390/molecules23020311 Frykberg RG, Banks J (2015) Challenges in the treat-
Brandenburg K, Andrä J, Müller M, Koch MHJ, Garidel ment of chronic wounds. Adv Wound Care (New
P (2003) Physicochemical properties of bacterial Rochelle) 4(9):560–582. https://doi.org/10.1089/
glycopolymers in relation to bioactivity. Carbohydr wound.2015.0635
Res 338(23):2477–2489. https://doi.org/10.1016/j. Garidel P, Brandenburg K (2009) Current understanding
carres.2003.08.008 of polymyxin B applications in bacteraemia/sepsis
Brandenburg K, David A, Howe J, Koch MHJ, Andrä therapy prevention: clinical, pharmaceutical, struc-
J, Garidel P (2005) Temperature dependence of tural and mechanistic aspects. Antinfect Agents Med
the binding of endotoxins to the polycationic pep- Chem 8(4):367–385
tides polymyxin B and its nonapeptide. Biophys Garidel P, Andrä J, Howe J, Gutsmann T, Brandenburg
J 88(3):1845–1858. https://doi.org/10.1529/ K (2007) Novel antiinflammatory and antiinfective
biophysj.104.047944 agents. Anti-infect Agents Med Chem 6(3):185–200.
Brandenburg K, Garidel P, Fukuoka S, Howe J, Koch https://doi.org/10.2174/187152107781023647
MHJ, Gutsmann T, Andrä J (2010) Molecular basis Gutsmann T, Razquin-Olazaran I, Kowalski I, Kaconis
for endotoxin neutralization by amphipathic peptides Y, Howe J, Bartels R, Hornef M, Schürholz T, Rössle
128 W. Correa et al.

M, Sanchez-Gomez S, Moriyon I, Martinez de Tejada Krepstakies M, Lucifora J, Nagel CH, Zeisel MB,
G, Brandenburg K (2010) New antiseptic peptides to Holstermann B, Hohenberg H, Kowalski I, Gutsmann
protect against endotoxin-mediated shock. Antimicrob T, Baumert TF, Brandenburg K, Hauber J, Protzer U
Agents Chemother 54(9):3817–3824. https://doi. (2012) A new class of synthetic peptide inhibitors
org/10.1128/AAC.00534-10 blocks attachment and entry of human pathogenic
Habersetzer F, Leboeuf C, Doffoël M, Zeisel MB, viruses. J Infect Dis 205(11):1654–1664. https://doi.
Baumert TF (2013) Synthetic anti-lipopolysaccharide org/10.1093/infdis/jis273
peptides and hepatitis C virus infection. Exp Opin Kuhlmann N, Heinbockel L, Correa W, Gutsmann T,
Invest Drugs 22(7):853–862. https://doi.org/10.1517/ Goldmann T, Englisch U, Brandenburg K (2018)
13543784.2013.794218 Peptide drug stability: the anti-inflammatory drugs
Hagar JA, Powell DA, Aachoui Y, Ernst RK, Miao Pep19-2.5 and Pep19-4LF in cream formulation. Eur
EA (2013) Cytoplasmic LPS activates caspase-11: J Pharm Sci 115:240–247. https://doi.org/10.1016/j.
implications in TLR4-independent endotoxic ejps.2018.01.022
shock. Science 341(6151):1250–1253. https://doi. Mangoni ML, McDermott AM, Zasloff M (2016)
org/10.1126/science.1240988 Antimicrobial peptides and wound healing: biologi-
Heinbockel L, Sanchez-Gomez S, Martinez de Tejada cal and therapeutic considerations. Exp Dermatol
G, Dömming S, Brandenburg J, Kaconis Y, Hornef 25(3):167–173. https://doi.org/10.1111/exd.12929
M, Dupont A, Marwitz S, Goldmann T, Ernst M, Martin L, De Santis R, Koczera P, Simons N, Haase H,
Gutsmann T, Schürholz T, Brandenburg K (2013) Heinbockel L, Brandenburg K, Marx G, Schuerholz
Preclinical investigations reveal the broad-spectrum T (2015) The synthetic antimicrobial peptide 19-2.5
neutralizing activity of peptide Pep19-2.5 on bac- interacts with heparanase and heparan sulfate in
terial pathogenicity factors. Antimicrob Agents murine and human sepsis. PLoS One 10(11):e0143583.
Chemother 57(3):1480–1487. https://doi.org/10.1128/ https://doi.org/10.1371/journal.pone.0143583
AAC.02066-12 Martin L, Horst K, Chiazza F, Oggero S, Collino M,
Heinbockel L, Marwitz S, Barcena Varela S, Ferrer-­ Brandenburg K, Hildebrand F, Marx G, Thiemermann
Espada R, Reiling N, Goldmann T, Gutsmann T, C, Schuerholz T (2016) The synthetic antimicrobial
Mier W, Schürholz T, Drömann D, Brandenburg K, peptide 19-2.5 attenuates septic cardiomyopathy and
Martinez de Tejada G (2015) Therapeutical admin- prevents down-regulation of SERCA2 in polymicro-
istration of peptide Pep19-2.5 and ibuprofen reduces bial sepsis. Sci Rep 6:37277. https://doi.org/10.1038/
inflammation and prevents lethal sepsis. PLoS One srep37277
10(7):e0133291. https://doi.org/10.1371/journal. Martínez de Tejada GM, Sanchez-Gomez S, Razquin-­
pone.0133291 Olazaran I, Kowalski I, Kaconis Y, Heinbockel L,
Hilchie AL, Wuerth K, Hancock RE (2013) Immune mod- Andrä J, Schürholz T, Hornef M, Dupont A, Garidel
ulation by multifaceted cationic host defense (anti- P, Lohner K, Gutsmann T, David SA, Brandenburg
microbial) peptides. Nat Chem Biol 9(12):761–768. K (2012) Bacterial cell wall compounds as promis-
https://doi.org/10.1038/nchembio.1393 ing targets of antimicrobial agents I. Antimicrobial
Howe J, Andrä J, Conde R, Iriarte M, Garidel P, Koch Peptides and Lipopolyamines. Curr Drug Targets
MHJ, Gutsmann T, Moriyon I, Brandenburg K (2007) 13(9):1121–1130
Thermodynamic analysis of the lipopolysaccharide Martinez de Tejada G, Heinbockel L, Ferrer-Espada R,
dependent resistance of gram-negative bacteria against Heine H, Alexander C, Barcena-Varela S, Goldmann
polymyxin B. Biophys J 92(8):2796–2805. https://doi. T, Correa W, Wiesmüller KH, Gisch N, Sanchez-­
org/10.1529/biophysj.106.095711 Gomez S, Fukuoka S, Schürholz T, Gutsmann T,
Kaconis Y, Kowalski I, Howe J, Brauser A, Richter Brandenburg K (2015) Lipoproteins/peptides are
W, Razquin-Olazaran I, Inigo-Pestana M, Garidel sepsis-inducing toxins from bacteria that can be neu-
P, Rössle M, Martinez de Tejada G, Gutsmann T, tralized by synthetic anti-endotoxin peptides. Sci Rep
Brandenburg K (2011) Biophysical mechanisms of 5:14292. https://doi.org/10.1038/srep14292
endotoxin neutralization by cationic amphiphilic Mirski T, Niemcewicz M, Bartoszcze M, Gryko R,
peptides. Biophys J 100(11):2652–2661. https://doi. Michalski A (2017) Utilisation of peptides against
org/10.1016/j.bpj.2011.04.041 microbial infections – a review. Ann Agric Environ
Katarzyna EG, Małgorzata D (2017) Antimicrobial Med 25(2):205–210. https://doi.org/10.26444/
peptides under clinical trials. Curr Top Med Chem aaem/74471
17(5):620–628. https://doi.org/10.2174/15680266166 Mishra B, Reiling S, Zarena D, Wang G (2017) Host
66160713143331 defense antimicrobial peptides as antibiotics: design
Khare S, Dorfleutner A, Bryan NB, Yun C, Radian AD, and application strategies. Curr Opin Chem Biol
de Almeida L, Rojanasakul Y, Stehlik C (2012) An 38:87–96. https://doi.org/10.1016/j.cbpa.2017.03.014
NLRP7-containing inflammasome mediates recogni- Molchanova N, Hansen PR, Franzyk H (2017) Advances
tion of microbial lipopeptides in human macrophages. in development of antimicrobial peptidomimet-
Immunity 36(3):464–476. https://doi.org/10.1016/j. ics as potential drugs. Molecules 22(9). https://doi.
immuni.2012.02.001 org/10.3390/molecules22091430
8 Synthetic Anti-lipopolysaccharide Peptides (SALPs) as Effective Inhibitors of Pathogen-Associated… 129

Nordström R, Malmsten M (2017) Delivery systems for receptors for intracellular LPS. Nature 514(7521):187–
antimicrobial peptides. Adv Colloid Interf Sci 242:17– 192. https://doi.org/10.1038/nature13683
34. https://doi.org/10.1016/j.cis.2017.01.005 Sierra JM, Fusté E, Rabanal F, Vinuesa T, Viñas M (2017)
Opal SM, Laterre P, Francois B et al (2013) Effect of erito- An overview of antimicrobial peptides and the latest
ran, an antagonist of md2-tlr4, on mortality in patients advances in their development. Expert Opin Biol Ther
with severe sepsis: the access randomized trial. 17(6):663–676. https://doi.org/10.1080/14712598.201
JAMA 309(11):1154–1162. https://doi.org/10.1001/ 7.1315402
jama.2013.2194 Sil D, Heinbockel L, Kaconis Y, Rössle M, Garidel
Park KS, Choi KH, Kim YS, Hong BS, Kim OY, Kim JH, P, Gutsmann T, David SA, Brandenburg K (2013)
Yoon CM, Koh GY, Kim YK, Gho YS (2010) Outer Biophysical mechanisms of the neutralization of endo-
membrane vesicles derived from Escherichia coli toxins by lipopolyamines. Open Biochem J 7:82–93.
induce systemic inflammatory response syndrome. https://doi.org/10.2174/1874091X01307010082
PLoS One 5(6):e11334. https://doi.org/10.1371/jour- Tabah A, Koulenti D, Laupland K, Misset B, Valles
nal.pone.0011334 J, Bruzzi de Carvalho F, Paiva JA, Cakar N, Ma
Pfalzgraff A, Heinbockel L, Su Q, Gutsmann T, X, Eggimann P, Antonelli M, Bonten MJ, Csomos
Brandenburg K, Weindl G (2016) Synthetic antimicro- A, Krueger WA, Mikstacki A, Lipman J, Depuydt
bial and LPS-neutralising peptides suppress inflamma- P, Vesin A, Garrouste-Orgeas M, Zahar JR, Blot
tory and immune responses in skin cells and promote S, Carlet J, Brun-Buisson C, Martin C, Rello J,
keratinocyte migration. Sci Rep 6:31577. https://doi. Dimopoulos G, Timsit JF (2012) Characteristics
org/10.1038/srep31577 and determinants of outcome of hospital-acquired
Pfalzgraff A, Heinbockel L, Su Q, Brandenburg K, Weindl bloodstream infections in intensive care units:
G (2017) Synthetic anti-endotoxin peptides inhibit the EUROBACT international cohort study.
cytoplasmic LPS-mediated responses. Biochem Intensive Care Med 38(12):1930–1945. https://doi.
Pharmacol 140:64–72. https://doi.org/10.1016/j. org/10.1007/s00134-012-2695-9
bcp.2017.05.015 Tacconelli E, Carrara E, Savoldi A, Harbarth S,
Pfalzgraff A, Barcena-Varela S, Heinbockel L, Gutsmann Mendelson M, Monnet DL, Pulcini C, Kahlmeter
T, Brandenburg K, Martinez de Tejada G, Weindl G G, Kluytmans J, Carmeli Y, Ouellette M, Outterson
(2018a) Antimicrobial endotoxin-neutralizing pep- K, Patel J, Cavaleri M, Cox EM, Houchens CR,
tides promote keratinocyte migration via P2X7 recep- Grayson ML, Hansen P, Singh N, Theuretzbacher U,
tor activation and accelerate wound healing in vivo. Br Magrini N, Aboderin AO, Al-Abri SS, Awang Jalil
J Pharmacol. https://doi.org/10.1111/bph.14425 N, Benzonana N, Bhattacharya S, Brink AJ, Burkert
Pfalzgraff A, Brandenburg K, Weindl G (2018b) FR, Cars O, Cornaglia G, Dyar OJ, Friedrich AW,
Antimicrobial peptides and their therapeutic poten- Gales AC, Gandra S, Giske CG, Goff DA, Goossens
tial for bacterial skin infections and wounds. H, Gottlieb T, Guzman Blanco M, Hryniewicz W,
Front Pharmacol 9:281. https://doi.org/10.3389/ Kattula D, Jinks T, Kanj SS, Kerr L, Kieny M-P,
fphar.2018.00281 Kim YS, Kozlov RS, Labarca J, Laxminarayan R,
Rice TW, Wheeler AP, Bernard GR, Vincent JL, Angus Leder K, Leibovici L, Levy-Hara G, Littman J,
DC, Aikawa N, Demeyer I, Sainati S, Amlot N, Cao Malhotra-Kumar S, Manchanda V, Moja L, Ndoye
C, Ii M, Matsuda H, Mouri K, Cohen J (2010) A B, Pan A, Paterson DL, Paul M, Qiu H, Ramon-­
randomized, double-blind, placebo-controlled trial Pardo P, Rodríguez-Baño J, Sanguinetti M, Sengupta
of TAK-242 for the treatment of severe sepsis. Crit S, Sharland M, Si-Mehand M, Silver LL, Song W,
Care Med 38(8):1685–1694. https://doi.org/10.1097/ Steinbakk M, Thomsen J, Thwaites GE, van der Meer
CCM.0b013e3181e7c5c9 JWM, Van Kinh N, Vega S, Villegas MV, Wechsler-­
Schuerholz T, Doemming S, Hornef M, Martin L, Simon Fördös A, Wertheim HFL, Wesangula E, Woodford
TP, Heinbockel L, Brandenburg K, Marx G (2013) N, Yilmaz FO, Zorzet A (2018) Discovery, research,
The anti inflammatory effect of the synthetic antimi- and development of new antibiotics: the WHO prior-
crobial peptide 19-2.5 in a murine sepsis model: a pro- ity list of antibiotic-resistant bacteria and tubercu-
spective randomized study. Crit Care 17(1):R3. https:// losis. Lancet Infect Dis 18(3):318–327. https://doi.
doi.org/10.1186/cc11920 org/10.1016/s1473-3099(17)30753-3
Serra R, Grande R, Butrico L, Rossi A, Settimio UF, Vanaja SK, Russo AJ, Behl B, Banerjee I, Yankova M,
Caroleo B, Amato B, Gallelli L, de Franciscis S (2015) Deshmukh SD, Rathinam VA (2016) Bacterial outer
Chronic wound infections: the role of Pseudomonas membrane vesicles mediate cytosolic localization of
aeruginosa and Staphylococcus aureus. Expert Rev LPS and caspase-11 activation. Cell 165(5):1106–
Anti-Infect Ther 13(5):605–613. https://doi.org/10.15 1119. https://doi.org/10.1016/j.cell.2016.04.015
86/14787210.2015.1023291 Werner S, Grose R (2003) Regulation of wound healing by
Shi J, Zhao Y, Wang Y, Gao W, Ding J, Li P, Hu L, Shao growth factors and cytokines. Physiol Rev 83(3):835–
F (2014) Inflammatory caspases are innate immune 870. https://doi.org/10.1152/physrev.2003.83.3.835
 Anticancer Activities of Natural
and Synthetic Peptides 9
A. L. Hilchie, D. W. Hoskin,
and M. R. Power Coombs

Abstract patient survival. Packaging ACPs in nanopar-

Anticancer peptides (ACPs) are cationic ticles or fusogenic liposomes may be benefi-
amphipathic peptides that bind to and kill can- cial for increasing ACP half-life and enhancing
cer cells either by a direct- or indirect-acting the delivery of ACPs to tumor target cells.
mechanism. ACPs provide a novel treatment Additionally, engineering ACP-producing
strategy, and selected ACPs are currently in oncolytic viruses may be an effective future
phase I clinical trials to examine their safety treatment strategy. Overall research in this
and overall benefit in cancer patients. area has been slow to progress, but with ongo-
Increasing the selectivity of ACPs is important ing ACP-based clinical trials, the potential for
so that these peptides kill cancer cells without ACPs in cancer treatments is closer to being
harming normal cells. Peptide sequence modi- realized. The integration of basic research
fications may help to improve ACP selectivity. with computer modeling of ACPs is predicted
ACPs also have immune-modulatory effects, to substantially advance this field of research.
including the release of danger signals from
dying cancer cells, induction of chemokine Keywords
genes, increasing T-cell immune responses, Anticancer peptides · Cytotoxicity · Immune
and inhibiting T regulatory cells. These effects modulation · Nanoparticles · Selectivity ·
ultimately increase the potential for an effec- Therapeutic
tive anticancer immune response that may
contribute to long-term benefits and increased

9.1 Introduction
A. L. Hilchie
Department of Microbiology and Immunology,
Dalhousie University, Halifax, Nova Scotia, Canada 9.1.1 Anticancer Therapies
D. W. Hoskin (*) and the Need for Alternative
Department of Microbiology and Immunology, Treatment Strategies
Dalhousie University, Halifax, Nova Scotia, Canada
Department of Pathology, Dalhousie University, Despite decades of research and progress in the
Halifax, Nova Scotia, Canada field of cancer therapy, conventional chemother-
e-mail: [email protected] apy remains the most commonly used treatment
M. R. Power Coombs modality for most cancers. Chemotherapy func-
Department of Biology, Acadia University, tions by indiscriminately killing rapidly dividing
Wolfville, Nova Scotia, Canada

© Springer Nature Singapore Pte Ltd. 2019 131

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_9
132 A. L. Hilchie et al.

cells. As a consequence of this mechanism of respectively. However, these therapies are not
action, chemotherapy cannot discriminate nor- without side effects (e.g., severe diarrhea, colitis,
mal proliferating cells from cancer cells, and as a inflammation pneumonitis), and patients with
result, it is unable to target indolent or dormant advanced disease often do not respond to treat-
cancers (Donnelly 2004; Naumov et al. 2003). ment or relapse thereafter (Seidel et al. 2018;
Furthermore, the acquisition by cancer cells of a Jean et al. 2017; Furue et al. 2018; Pillai et al.
chemo-resistant phenotype further reduces the 2018). Collectively, these issues highlight the
therapeutic value of chemotherapeutic com- ongoing need for novel, broad-spectrum antican-
pounds (Bush and Li 2002). Importantly, certain cer compounds capable of selectively killing can-
chemotherapeutic compounds (e.g., cyclophos- cer cells. Ideally, these new therapies would also
phamide) are associated with the development of harness the power of the immune system by initi-
secondary malignancies (Choi et al. 2014). This ating protective antitumor immune responses in
issue is particularly problematic in pediatric can- patients.
cers in which secondary malignancies, as well as
lifelong consequences of toxicities, represent the
most severe long-term complications of chemo- 9.1.2 Anticancer Peptides
therapy (Kebudi and Ozdemir 2017). For exam-
ple, alkylating agents (e.g., cyclophosphamide Cationic anticancer peptides (ACPs) represent a
and ifosfamide) are commonly used to treat pedi- promising alternative to conventional chemother-
atric hematologic malignancies and solid tumors apy. ACPs are small peptides that contain several
and as preconditioning treatment regimens for cationic and hydrophobic amino acids, giving
hematopoietic stem cell transplantation (Choi them an overall positive charge and amphipathic
et al. 2014). However, these same drugs are structure (Hoskin and Ramamoorthy 2008). Most
known to cause therapy-related acute myeloge- ACPs are inherently antimicrobial in nature. In
nous leukemia (Thirman and Larson 1996). To fact, cationic peptides isolated from various
address the many limitations of chemotherapy, organisms were historically assessed for antimi-
significant research efforts over the last decade crobial activities and were studied as such prior
led to the identification of “targeted therapies” to their first being described as potent anticancer
(e.g., trastuzumab) that function by selectively agents in 1985 (Sheu et al. 1985). In addition to
targeting and killing cancer cells while sparing their antimicrobial and anticancer activities,
normal healthy cells, regardless of their rate of these so-called host defense peptides (HDPs)
growth. Unfortunately, cancer cell resistance to exhibit many other biological properties, includ-
these targeted therapies was reported shortly after ing antiviral (Wang et al. 2008; Bergman et al.
their introduction to the clinic (Nagy et al. 2005). 2007), anti-biofilm (Overhage et al. 2008; de la
Researchers and clinicians alike are now rec- Fuente-Núñez et al. 2012), wound healing
ognizing that novel treatment strategies harness- (Steinstraesser et al. 2012), anti-parasitic (Couto
ing the power of the immune system may lead to et al. 2018), adjuvant (Kindrachuk et al. 2009),
improved clinical outcomes. Indeed, the use of and immune-modulatory activities (Madera and
neutralizing antibodies targeting the immune Hancock 2012; Nijnik et al. 2010) (Fig. 9.1).
checkpoints cytotoxic T-lymphocyte-associated Peptides are ideal drug candidates due to their
protein 4 (CTLA-4; e.g., ipilimumab) and pro- low cost of production, the ease with which they
grammed cell death protein 1 (PD-1; e.g., pem- can be modified, and relatively high tissue pene-
brolizumab and nivolumab) has enjoyed tration (e.g., compared to antibody-based thera-
considerable clinical success (Seidel et al. 2018; pies) (Soman et al. 2009; Richardson et al. 2009;
Jean et al. 2017; Furue et al. 2018). These thera- Hilchie et al. 2012, 2013a, 2015).
pies are now used as first- and second-line thera- ACPs are often classified based on the struc-
pies for the treatment of inoperable advanced ture that they adopt upon contact with a biologi-
melanoma and non-small cell lung cancer, cal membrane. Three main classes exist, namely,
9 Anticancer Activities of Natural and Synthetic Peptides 133

resistant (MDR) cancer cells (Hilchie et al. 2011;

Kim et al. 2003; Johnstone et al. 2000). Several
different peptides, including the pleurocidin NRC-
Anti-microbial 03, act as chemosensitizing agents by reducing the
EC50 of several different chemotherapeutic drugs
(Hilchie et al. 2011; Kim et al. 2003; Johnstone
et al. 2000; Hui et al. 2002). These chemosensitiz-
ing activities suggest that ACPs may work in a
synergistic fashion with conventional anticancer
Immune- drugs. Indeed, we recently showed that the wasp
Anti-cancer venom peptide mastoparan synergizes with che-
modulatory
motherapeutic compounds both in vitro and
in vivo (Hilchie et al. 2016). Many ACPs, includ-
ing DAAs, destroy primary tumors and their
metastases without causing undue harm to normal
Fig. 9.1 Biological activity of cationic amphipathic tissues (Hansel et al. 2007). Moreover, preclinical
peptides. Cationic amphipathic peptides may exhibit any studies show that DAAs exert antitumor effects
combination of anti-microbial, anti-cancer, or immune-­
when delivered by intratumoral, intraperitoneal, or
modulatory properties. While many still refer to anti-­
cancer peptides as cationic antimicrobial peptides, or host intravenous injection (Hilchie et al. 2016; Berge
defense peptides (i.e., immune-modulatory peptides), it is et al. 2010; Camilio et al. 2014a). Importantly, the
important to appreciate that these biological activities work of others shows that, in addition to their abil-
may be completely independent of each other, and thus
ity to destroy the primary tumor, certain DAAs
should be examined on an individual basis
initiate an antitumor immune response that pro-
α-helical (e.g., magainin - Baker et al. 1993; tects the mouse from tumor rechallenge (Berge
Nguyen et al. 2011), β-sheet (e.g., lactoferri- et al. 2010; Camilio et al. 2014a). These activities
cin - Nguyen et al. 2011; Mader et al. 2005), and will be discussed in more detail in Sect. 9.3.
extended (e.g., LfcinB6 - Richardson et al. 2009; Furthermore, tumor resistance to DAAs is pre-
Nguyen et al. 2011). These structures, which are dicted to be difficult to achieve because DAAs do
all amphipathic in nature, typically consist of a not rely on unique receptors or a specific signal
predominantly cationic face and a hydrophobic transduction pathway for their action. Indeed, we
face. This is necessary to facilitate peptide inter- investigated cancer cell resistance to DAAs and
action with the target cell. ACPs can also be clas- found that continuous exposure (i.e., more than
sified on the basis of their mechanism of action, 1 year) to increasing concentrations of ACPs only
of which two classes exist: direct-acting (i.e., generated cancer cells with low-level resistance to
lytic) or indirect-acting (i.e., apoptosis-inducing) lytic peptides (manuscript in preparation).
(Hilchie and Hoskin 2010), both of which will be Importantly, these peptide-resistant cancer cells
discussed in further detail in Sect. 9.2. maintained susceptibility to chemotherapeutic
drugs and, to our surprise, were unable to establish
tumors in immune-deficient mice.
9.1.3 Advantages of ACPs
Over Conventional
Chemotherapy 9.1.4  imitations to the Clinical Use
L
of ACPs
Due to their unique mechanism of action, ACPs,
and particularly direct-acting ACPs (DAAs), have Until recently, the clinical use of ACPs was lim-
many advantages over conventional chemother- ited by their high cost of production. However,
apy. Unlike conventional chemotherapy, many since their discovery as novel anticancer agents
ACPs kill slow-growing as well as multidrug-­ (Sheu et al. 1985), the cost of producing peptides
134 A. L. Hilchie et al.

at high purity (i.e., >95%) by high-performance directly proportional to peptide length, and
liquid chromatography (HPLC) has undergone a requires that the tumor cells maintain expression
substantial decline. Moreover, the cost of synthe- of the receptor with which the targeted peptide
sizing large amounts of good manufacturing interacts. As an alternative strategy, amino acid
practice (GMP)-grade peptide is declining as substitution has been used to reduce peptide tox-
more and more peptide synthesis companies icity to normal cells (Dennison et al. 2006; Yang
enter the marketplace. The use of recombinant et al. 2003; Eliassen et al. 2003). This approach
technology, which is a useful method for synthe- typically involves modifying simple peptide
sizing large amounts of peptides, has to date been characteristics, such as charge and/or hydropho-
very difficult because most ACPs exhibit antimi- bicity, as these are known to be required for tox-
crobial activities (Greenshields et al. 2008). To icity to tumor cells; however, the structural basis
address this issue, Ishida et al. recently developed for selective cancer cell killing by ACPs is still
a unique method whereby calmodulin is used as a poorly understood. We will further discuss this
carrier protein to express several different antimi- approach in Sect. 9.5.2.
crobial peptides (Ishida et al. 2016). In this The in vivo stability of peptides is a significant
approach, the toxic (i.e., antimicrobial) activities shortcoming of many ACP-based therapeutics.
are masked, and the peptide is protected from Unpublished work by the Hancock group sug-
degradation during peptide expression and purifi- gests that small cationic peptides rapidly distrib-
cation. Others have taken an alternate approach ute to all tissues in the body and possess a half-life
of identifying truncated forms of the parent pep- of approximately 2 min in the blood (discussed in
tide that maintain their biological activities Hilchie et al. 2013a). While many see this as an
(Richardson et al. 2009; Mader et al. 2005). issue, others argue that this problem is negated by
Moreover, identifying combinations of ACPs and the speed with which many different ACPs exert
chemotherapeutic compounds that synergize their toxic effects to cancer cells and that, by lim-
in vivo is expected to reduce the dose of each iting peptide half-life, the likelihood of off-target
compound that is required for a biological effect, toxicities is also reduced. Nevertheless, there are
thereby reducing any treatment-related toxicities several reports that peptidomimetics show
and overall treatment cost. Collectively, these improved stability in vitro. Moreover, various
research endeavors, as well as a competitive mar- nanoparticle-based delivery strategies show con-
ketplace, significantly reduce the financial bur- siderable promise. These strategies will be dis-
den of novel peptide-based therapies. cussed in further detail in Sects. 9.4.1 and 9.4.2.
One of the most significant shortcomings of In our opinion, the biggest issue facing ACP-­
ACP-based therapies is their toxicity to normal based therapies is the loss of momentum that this
cells at high peptide concentrations. Many field of research is experiencing. Time and time
research groups have attempted to reduce off-­ again, researchers identify new ACPs and
target toxicity by adding a targeting sequence to describe their mechanism of action and perhaps
their peptide of choice (Liu et al. 2011; Zitzmann their spectrum of activity (i.e., which cancer cell
et al. 2002; Leuschner and Hansel 2004). To this types are susceptible to peptide-mediated kill-
end, small targeting moieties that interact with ing); however, there is little follow-up work.
specific cell surface molecules overexpressed on Thus, with few exceptions, little has been done to
cancer cells are added to the peptide of interest, thoughtfully address the shortcomings of ACP-­
typically using a glycine-glycine linker. To date, based therapies. Here, our aim is to describe the
this strategy, which will be discussed in more anticancer potential of these molecules and their
detail in Sect. 9.5.1, has shown mixed results. It mechanism(s) of action and discuss ways in
is important to note that this strategy increases which momentum can be regained in an other-
the cost of production, as synthesis costs are wise promising field of research.
9 Anticancer Activities of Natural and Synthetic Peptides 135

9.2 Direct-Acting both are initiated by the selective binding of

Versus Indirect-Acting ACPs ACPs to cancer cell membranes.

ACPs are classified as direct- or indirect-acting

based on their mechanism of action (Hilchie and 9.2.1 Factors That Contribute
Hoskin 2010). DAAs bind to and kill cancer cells to Selective Peptide Binding
by causing irreparable membrane damage fol- to Cancer Cell Membranes
lowed by cell lysis (Fig. 9.2). In contrast, expo-
sure to indirect-acting ACPs results in cell death ACPs are thought to selectively bind to cancer cell
by apoptosis, which occurs in the absence of membranes because of differences in membrane
extensive membrane damage. composition (i.e., charge), surface area, trans-
DAAs do not require access to the cytosol in membrane potential, and membrane fluidity
order to kill the cell. As a consequence of this, (reviewed in (Hoskin and Ramamoorthy 2008;
DAAs have many advantages over conventional Hilchie and Hoskin 2010; Mader and Hoskin
chemotherapy (see Sect. 9.1.3). DAAs tend to be 2006; Yeaman and Yount 2003; Bhutia and Maiti
highly potent and maintain a relatively broad 2008; Giuliani et al. 2007)). To our knowledge, no
spectrum of activity (i.e., kill a wide variety of study has definitively elucidated the mechanism
cancer cells) in comparison to indirect-acting by which ACPs selectively bind to cancer cell
ACPs. Indirect-acting ACPs tend to be less potent membranes. However, experts agree that mem-
than DAAs, and they often target mitochondria, brane composition (i.e., charge) appears to be the
thereby killing cancer cells by initiating most significant factor in this process. Thus, we
mitochondrial-­dependent (i.e., intrinsic) apopto- will limit our discussion to the importance of
sis (Mader et al. 2005; de Azevedo et al. 2015). membrane composition, as the other factors have
While these two mechanisms vary considerably, been reviewed elsewhere (Hoskin and

Fig. 9.2 Direct-acting

ACPs rapidly lyse
human multiple
myeloma cells.
MPLfcinB6 (50 μM) or
its vehicle control were
added to U226 human
multiple myeloma cells
for 2 h. The cells were
subsequently fixed,
processed, and
visualized by scanning
electron microscopy.
The top and bottom
images were captured
under 7000 and 40,000×
magnification,
respectively
136 A. L. Hilchie et al.

Ramamoorthy 2008; Hilchie and Hoskin 2010; anionic surface molecules. Our own work revealed
Mader and Hoskin 2006; Yeaman and Yount 2003; that hundreds of genes are differentially expressed
Bhutia and Maiti 2008; Giuliani et al. 2007). in cancer cells that are refractory to these DAAs
Owing to the presence of zwitterionic phos- (manuscript in preparation). Importantly, these
phatidylcholine, phosphatidylethanolamine, and factors appear to influence the toxicity of several
sphingomyelin, normal cell membranes are neu- DAAs, suggesting a common mechanism of
tral in charge (Zachowski 1993). In contrast, the membrane perturbation. Further to this, decreased
outer membrane leaflet of cancer cells carries a susceptibility to these DAAs impacted the tumori-
net negative charge due to increased levels of genicity of the malignant cells. This work also
anionic phosphatidylserine, O-glycosylated suggests that dozens of components of the extra-
mucins, heparan and chondroitin sulfate proteo- cellular matrix are likely involved in peptide bind-
glycans, and sialylated glycoproteins (Utsugi ing to, and disruption of, the target cell membrane.
et al. 1991; Bafna et al. 2010; Koo et al. 2008; It is clear that we are only beginning to compre-
van Beek et al. 1973; Iida et al. 1996). Collectively, hend the complexity of this process.
these differences are thought to contribute to the
selective attraction of ACPs to cancer cell mem-
branes. Following the initial stages of peptide 9.2.2 Factors That Influence
binding, ACPs are thought to anchor to the mem- the Mechanism of ACP-­
brane via insertion of hydrophobic residues into Mediated Anticancer Activity
the hydrophobic core of the plasma membrane
(Hoskin and Ramamoorthy 2008; Hilchie and To our knowledge, there is no evidence to suggest
Hoskin 2010; Mader and Hoskin 2006; Yeaman that specific structural determinants are respon-
and Yount 2003; Bhutia and Maiti 2008; Giuliani sible for rendering an ACP direct-acting or
et al. 2007). Once the peptide is securely bound indirect-­acting. Interestingly, in select cases, the
to the membrane, it either causes membrane mechanism of peptide-mediated cytotoxicity is
instability followed by pore formation and cell dependent on the cancer cell line under investiga-
lysis (DAAs), or it penetrates into the cytoplasm tion. For instance, bovine lactoferricin induces
without substantially damaging the membrane, apoptosis in human leukemia, lymphoma, and
wherein the peptide initiates apoptosis (indirect-­ breast cancer cells (Mader et al. 2005; Furlong
acting ACPs). There are several different models et al. 2010; Furlong et al. 2006), whereas the
to describe how ACPs cause membrane instabil- same ACP is lytic to fibrosarcoma, melanoma,
ity. These models are thoughtfully described colorectal cancer, and neuroblastoma cells
elsewhere (Nguyen et al. 2011). (Eliassen et al. 2002; Eliassen et al. 2006). In
Many studies have used artificial membranes other cases, ACPs may be selectively toxic for
as model systems to show that peptide binding one cancer cell type but devoid of effects on
and membrane perturbation are influenced by the another cancer cell type. For example, the ACP
lipid content of membranes (Gazit et al. 1995; MPLfcinB6 selectively lyses leukemia and lym-
Matsuzaki et al. 1989). However, it is consider- phoma cells (Hilchie et al. 2013b) but is not cyto-
ably more difficult to determine the factors that toxic for breast cancer cells (unpublished). This
are involved in ACP binding to eukaryotic mem- may be the result of many fundamental differ-
branes due to the complexity of the membrane. ences in the complexity of the membranes of
We demonstrated that the DAAs NRC-03 and these different types of cancer cells, as we consis-
NRC-07 exhibit 100- and 50-fold, respectively, tently note that cancer cells in suspension (e.g.,
greater binding to breast cancer cells than to nor- Jurkat T leukemia cells) are much more suscep-
mal untransformed fibroblasts (Hilchie et al. tible to killing by ACPs than are cancer cells
2011). In this case, peptide binding was influ- grown as monolayers (e.g., MDA-MB-231 breast
enced by, but not dependent on, several different carcinoma cells). For instance, the pleurocidins
9 Anticancer Activities of Natural and Synthetic Peptides 137

NRC-03 and NRC-07, as well as the wasp venom the immune-modulatory functions of these pep-
peptide mastoparan, are roughly two- to fourfold tides are maintained under physiologically rele-
more toxic to leukemia and myeloma cells than vant conditions (Hilchie et al. 2013a).
they are to breast carcinoma cells (Hilchie et al. Some synthetic peptides that have the ability
2011, 2013c, 2016). These findings are further to modulate the immune system are known as
supported by our ongoing quantitative structure/ innate defense regulators (IDR). Although some
activity relationship studies, which are discussed of the antibacterial properties of these peptides
briefly in Sect. 9.5.2. are lost under physiological conditions, these
As ACPs are small amphipathic molecules peptides are still bioactive through immune-­
with defined secondary structures, it stands to modulating effects such as increasing chemokine
reason that alterations in the amino acid compo- production (Hilchie et al. 2013a). Monocyte
sition of ACPs may affect their potency and migration in response to chemokines shows a fur-
mechanism of action. While this has not been ther increase in the presence of the peptide IDR-­
studied extensively, we recently noted a striking 1002 via a mechanism involving integrins and
difference in the mechanism of action of masto- AKT signaling (Madera and Hancock 2012).
paran by simple C-terminal amidation. Others IDR-1002 also activates the immune response by
have shown that unamidated mastoparan kills increasing chemokine production and by recruit-
cancer cells by induction of apoptosis; in con- ing leukocytes to the site of infection (Nijnik
trast, we showed that mastoparan that incorpo- et al. 2010).
rates a C-terminal amide is much more potent Some peptides have the ability to promote
and kills cancer cells by inducing cell lysis antibody production. Immunizing mice with
(Hilchie et al. 2016; de Azevedo et al. 2015). the peptide HH2 along with pertussis toxoid
This finding not only demonstrates the impor- and CpG 10101 significantly increases the titer
tance of the primary amino acid sequence in of toxoid-specific antibodies, indicating that
determining the mechanism of action of a given this peptide enhances antibody production in
ACP but also provides hope that detailed struc- this mouse vaccination model (Kindrachuk
ture/activity relationship studies may reveal et al. 2009).
next-generation peptides with improved selec- Despite the many actions of ACPs that pro-
tivity for cancer cells, thereby addressing one of mote immune responses, the bioactive peptide
the most significant limitations to peptide-based lactoferricin B decreases superantigen-mediated
therapeutics. interleukin-2 production by mouse splenocytes
(Hayworth et al. 2009). This finding indicates
that certain ACPs modulate immune function;
9.3 ACP-Mediated Immune however, the nature of that modulation may be
Activation dependent on the variables present in a given
situation.
9.3.1 Cationic Amphipathic Mast cells are prominent within tissues and
Peptides as Modulators exert multiple effects on the vasculature as a
of Immune Function result of their degranulation. Pleurocidins, IDR-­
1018, and other HDPs induce mast cell degranu-
Many cationic peptides were initially character- lation, intracellular calcium mobilization, and the
ized for their antimicrobial activities. More recent release of prostaglandins (Pundir et al. 2014;
research shows that many of these same peptides Yanashima et al. 2017; Gupta et al. 2016). These
exhibit immune-modulatory activities, as previ- peptides may therefore act on mast cells to pro-
ously reviewed (Hilchie et al. 2013a). Importantly, mote vascular permeability and vasodilation,
the antimicrobial activities of many of these pep- subsequently shaping the developing immune
tides are lost in the presence of serum, whereas response.
138 A. L. Hilchie et al.

ACP-induced
Intratumoral cell lysis
administration
HMGB1
of ACP
ATP
Release of
DAMPs
and tumor Immature
antigens DC

Antigen
engulfment

Activation of
tumor-
specific CTLs

Antigen
Tumor cell presentation
killing by CTLs Mature
DC

Fig. 9.3 Activation of a protective anti-tumor cytotoxic pattern molecules (DAMPs) such as ATP and high mobility
T lymphocyte (CTL) response by ACPs. In vivo data sug- group box protein 1 (HMGB1) that promote tumor antigen
gests that a protective immune response develops after intra- uptake by dendritic cells (DCs), which then mature and
tumoral administration of an ACP. The ACP kills tumor present antigen to T cells. Tumor-specific cytotoxic T lym-
cells, resulting in the release of danger-­associated molecular phocytes (CTLs) are generated that kill tumor cells

9.3.2 Induction of Antitumor increase in HMGB1 release from these cells

Immune Responses (Berge et al. 2010). Taken together, these findings
by Immunogenic Cell Death suggest that ACP-mediated lysis of malignant
cells induces anticancer immunity.
Some ACPs release danger signals from the cell The study of B16 melanoma-bearing mice
that are thought to be immunogenic. The release showed that intratumoral treatment with DAA
of danger-associated molecular patterns LTX-315 results in tumor regression and signifi-
(DAMPs) like calreticulin, ATP, and high mobil- cantly increased survival following tumor rechal-
ity group box protein 1 (HMGB1) from dying lenge (Camilio et al. 2014a). In these animals, T
cancer cells results in the induction of an immune cells are recruited to LTX-315-treated tumors,
response to tumor antigens (Fig. 9.3) (Camilio and inflammatory cytokine gene expression is
et al. 2014b). elevated following LTX-315 treatment. Mice that
Intratumoral administration of the bovine were previously cured of palpable melanoma
lactoferricin-­derived ACP LTX-302 to A20 with LTX-315 treatment are protected from
lymphoma-­ bearing immune-competent mice rechallenge with B16 melanoma cells (Camilio
results in tumor necrosis and inflammatory cell et al. 2014a). In vitro LTX-315 treatment of mel-
infiltration, followed by complete tumor regres- anoma cells releases DAMPs that include ATP,
sion as well as tumor-specific protection against cytochrome C, reactive oxygen species (ROS),
tumor rechallenge. In vitro treatment of these and HMGB1 (Camilio et al. 2014a; Eike et al.
lymphoma cells with LTX-302 results in an 2015).
9 Anticancer Activities of Natural and Synthetic Peptides 139

ACPs that induce a local immune response genetically modified type 1 herpes simplex virus
in tumors may also trigger a systemic immune (HSV) has been approved for use and shows ther-
response that removes all neoplastic cells, apeutic benefit to melanoma patients (Andtbacka
including those that have spread to other parts et al. 2015). However, only a moderate increase
of the body. This immune response is activated in survival is reported with this oncolytic virus
by the ACP-induced release of DAMPs. therapy, indicating other treatments are needed.
Inhibition of local regulatory T cells (Tregs) at Another oncolytic virus, vaccinia JX-594, has
the tumor site is another aspect to consider been used to treat liver cancer patients, in which
since inhibiting these cellular regulators of the the virus was shown to be oncolytic and increase
immune response is known to promote antican- patient survival with some evidence of an
cer immune responses. In tumor beds, LTX-315 immune-activating mechanism (Heo et al. 2013).
increased the number of CD4+ (Th1 and Th17) Therefore, evidence exists that treatment with
and CD8+ T cells while decreasing Treg num- certain oncolytic viruses is able to increase the
bers (Yamazaki et al. 2016). The cationic pep- survival of cancer patients.
tide LL-37, which has both pro- and anticancer Oncolytic peptides, as a result of their short
effects depending on the cancer (Chen et al. half-life, may provide a safer alternative for
2018), also inhibits CD25+CD4+FOXP3+ T reg- patients in comparison to oncolytic viruses.
ulatory cells and so may be helpful in promot- There are safety concerns when patients are
ing an anticancer immune response (Mader administered a virus that may persist long term
et al. 2011). and has the potential to mutate into a harmful
Administration of LTX-315 increased variant. Some ACPs are active against drug-­
CTLA-4 expression on CD8+ T cells while resistant cancer cells and are not lytic for red
decreasing PD-1 expression, suggesting that blood cells, making them potential candidates
using this ACP in combination with an inhibitor for development as future treatments for cancer.
of CTLA-4 (ipilimumab) may improve treatment Since some ACPs kill cancer cells and induce
outcome (Yamazaki et al. 2016). Initial experi- an anticancer immune response (Haug et al.
ments with immune checkpoint inhibitors such as 2016), injection of LTX-315 into transdermally
ipilimumab suggest that timing of the treatments accessible tumors is currently in phase I clini-
may be critical as administration of the CTLA-4 cal trials to assess safety, dosing, pharmacoki-
neutralizing agent prior to the treatment with the netics, and immune response development
ACP, LTX-315, may be needed to achieve a ther- (ClinicalTrials.gov (NCT01058616) 2010).
apeutic benefit. The need for treatment with ipili- LTX-315 is also now being assessed in multiple
mumab in advance of ACP administration is cancers as a monotherapy or in combination
explained by the fact that CTLA-4 is involved in with ipilimumab or pembrolizumab
down-regulating T-cell activation. (ClinicalTrials.gov (NCT01986426) 2013).
Discovery research has identified these onco-
lytic peptides and has revealed their in vitro and
9.3.3 Comparison of ACPs in vivo activities. These ACPs are now being
to Oncolytic Virus Therapy examined for clinical efficacy. Even in phase I
clinical trials, there is assessment of ACP anti-
Oncolytic viruses are another class of novel ther- tumor activity as indicated by complete and
apeutics being investigated for the management partial response rates, overall response rate,
of various cancers. Oncolytic viruses may fail to and progression-free survival. The results of
kill tumor cells in an individual if the virus is these trials will begin to answer questions
quickly eliminated as the result of triggering an regarding the effectiveness of ACPs and the
innate immune response (Chiocca and Rabkin potential benefit of enhanced delivery of these
2014). The first oncolytic virus derived from a oncolytic peptides.
140 A. L. Hilchie et al.

9.4  trategies to Enhance ACP

S arginine-­ glycine-­
aspartic acid (RGD) peptide
Delivery on its surface is taken up by breast cancer cells
via integrin receptor-mediated endocytosis and
Since peptides typically undergo rapid degrada- has been engineered to subsequently degrade in
tion in the body, a delivery platform may be the acidic endosomal compartment, resulting in
needed to ensure that ACPs get to their desired the intracellular release of cytotoxic anticancer
target. This may not be necessary for fast-acting drugs (Murugan et al. 2016). It should be
peptides; nevertheless, methods to package ­possible to use a similar approach to deliver
ACPs so that they reach the tumor microenvi- ACPs directly into the acidic tumor
ronment include the use of nanoparticles and microenvironment.
fusogenic liposomes. Peptide modification Some nanocarriers are toxic on their own,
strategies can also be used to promote tumor especially those made with polyacrylic acids,
cell targeting; however, this approach will be indicating the need for nontoxic nanocarriers. In
discussed in Sect. 9.5. this regard, doxorubicin has been encapsulated
within a nanocarrier made of 30% oxidized
starch and decorated with an integrin-targeting
9.4.1 Nanoparticles peptide attached with a polyethylene glycol
(PEG) linker in order to selectively target
Nanoparticles provide a mechanism for drug integrin-­overexpressing cancer cells (Jiang et al.
delivery to the correct location in patients, includ- 2018). Such a starch-based nanocarrier is likely
ing those with drug-resistant cancers. Many dif- to be less toxic than other nanoparticle
ferent nanoparticle formulations have been formulations.
considered. The nanoparticles themselves need to Another strategy for delivering ACPs to can-
be stable and nontoxic, and they must be targeta- cer cells is through the use of fusogenic lipo-
ble in order to deliver the drug of interest to the somes, which are able to deliver hydrophobic or
correct cell/tissue. hydrophilic drugs directly into a target cell with-
Perfluorocarbon nanoparticles have been of out risking degradation by the endocytic pathway
particular interest for drug delivery because (Kube et al. 2017). Fusogenic liposomes that
these nanoparticles are biologically inert, stable, effect membrane fusion have been used to deliver
nontoxic, and can be monitored using different LfcinB6 into the cytoplasmic compartment of
imaging platforms. Perfluorocarbon nanoparti- both leukemia cells and breast cancer cells,
cles can carry large quantities of drugs, and their resulting in rapid cytotoxicity (Richardson et al.
delivery to target sites can be observed in vivo 2009). In addition to the potential for tumor cell
(Winter 2014; Chen et al. 2013). Since ACPs are targeting, fusogenic liposomes are expected to
small, it is feasible to put ACPs on/in these per- protect ACPs from proteolysis long enough for
fluorocarbon nanoparticles to enhance their them to reach effective concentrations in the
delivery to a primary tumor and metastatic tumor site.
lesions. Studies that used perfluorocarbon Clearly, there are multiple strategies that can
nanoparticles loaded with melittin, a cytolytic be employed to effectively deliver ACPs to can-
peptide from bee venom, have revealed that the cer cells. The next critical step will be to evaluate
combination of an ACP and nanoparticle deliv- the safest approach in phase I clinical trials.
ery system is able to significantly decrease B16 Indeed, multiple clinical trials in which nanopar-
melanoma tumor volumes in vivo (Soman et al. ticles are being used to deliver different chemo-
2009; Pan et al. 2011). therapeutic agents (e.g., paclitaxel) are underway.
Since the microenvironment of many solid Since there are already ongoing clinical trials
tumors is acidic (Tannock and Rotin 1989), with ACPs, such as LTX-315, future use of
some nanoparticle delivery approaches have nanoparticles as delivery vehicles for ACPs may
been engineered to function best at low pH. For be an effective strategy to increase the half-life of
example, a CPMSN nanocarrier bearing the oncolytic peptides.
9 Anticancer Activities of Natural and Synthetic Peptides 141

9.4.2 Peptides with Altered limited by their toxicity to normal human cells at
Stereochemistry high peptide concentrations. Several strategies
have been used to improve ACP selectivity for
One potential problem with ACP-based treat- cancer cells. Many of these strategies involve
ments is that these peptides can be easily degraded optimizing the delivery of peptide to tumor cells
by proteolytic enzymes present in the digestive through the use of nanoparticle-based delivery
system and blood plasma (Vlieghe et al. 2010). systems, as discussed in Sect. 9.4.1. Here, we
Susceptibility to degradation is dependent on the will briefly review how alterations in the primary
peptide sequence (e.g., trypsin cleaves arginine amino acid sequence influence the selectivity of
and lysine); however, altering the stereochemis- ACPs for cancer cells.
try of an ACP may render it unrecognizable by
proteolytic enzymes. In this regard, since amino
acids occur naturally as an “L” stereoisomer, 9.5.1 Generating Tumor-Specific
D-isomers are not susceptible to proteolytic deg- ACPs Through the Addition
radation (Hilchie et al. 2015). For example, an of Peptide-Targeting Motifs
all-D-amino acid variant of pleurocidin that is
based on the L form of the cationic antimicrobial As noted in Sect. 9.1.4, ACP selectivity for can-
peptide from winter flounder resists degradation cer cells can be enhanced through the addition of
by trypsin, plasmin, and carboxypeptidase (Jung so-called targeting sequences. This strategy
et al. 2007). Findings such as this indicate that involves the use of a glycine-glycine linker to
the stereochemistry of a peptide is relevant with conjugate the ACP with a peptide sequence that
respect to its susceptibility to degradation. recognizes specific molecules that are overex-
pressed by cancer cells. The targeting motif then
promotes ACP binding to the tumor cell, after
9.4.3 Potential for ACP-Expressing which the cytotoxic portion of the peptide trig-
Oncolytic Virus Therapy gers cell death. There are dozens of examples of
targeting sequences and many instances in which
If clinical trials continue to show that oncolytic this strategy has been used to improve ACP selec-
viruses are safe and effective anticancer agents, it tivity – some have been successful, whereas oth-
may be advantageous to engineer an oncolytic ers have not enjoyed success. Here, we will
virus that also expresses a direct- or indirect-­acting provide an example of each strategy for the pur-
ACP. Since the mechanism of oncolysis is differ- pose of illustration.
ent between oncolytic viruses and ACPs, the Bombesin is a 14-residue tumor-homing pep-
potential exists for an enhanced cytotoxic effect by tide that binds several receptors that are overex-
an ACP-expressing oncolytic virus. Administration pressed by many cancer cell types (Anastasi
of an oncolytic virus that also codes for an onco- et al. 1971; Reubi et al. 2002; Cornelio et al.
lytic peptide is predicted to increase the likelihood 2007). Significant improvements in tumor cell
of killing all cancer cells in a given tissue, includ- killing were noted when magainin 2 was conju-
ing cancer stem cells, and activating a long-lasting gated to bombesin (Liu et al. 2011). In compari-
anticancer immune response that will protect son to the parental peptide (magainin 2), the IC50
against cancer recurrence. of the hybrid peptide for cancer cells was at least
tenfold lower, which was substantially lower
than the IC50 for normal cells. This finding sug-
9.5  trategies to Enhance ACP
S gests that the increase in potency of the hybrid
Selectivity for Cancer Cells peptide was not at the expense of cancer cell
selectivity.
ACPs, particularly those that are direct-acting, Phage display libraries can be used to identify
have many advantages over conventional chemo- novel targeting sequences for ACPs. For exam-
therapeutic agents; however, ACPs continue to be ple, a screen of phage display libraries was used
142 A. L. Hilchie et al.

to identify the sequence LTVSPWY, which has structure of the ACP (Yang et al. 2002). In this
been successfully used to deliver oligonucle- study, helical wheel diagrams of the parent pep-
otides to SKBR3 breast cancer cells (Shadidi and tide were used to show that positively charged
Sioud 2003). However, this sequence did not amino acids cluster into two spatially separated
improve the cytotoxicity of LfcinB6 for a differ- regions, termed the major and minor sector, that
ent breast cancer cell line (unpublished), indicat- contain four and two cationic amino acids,
ing the need to screen for broad applicability of respectively. Moving the two cationic amino
targeting sequences in a particular type of acids from the minor sector to the major sector
cancer. increased cancer cell killing at the expense of
cancer cell selectivity, suggesting that the pres-
ence of a minor sector may reduce ACP toxicity
9.5.2 Enhancing ACP Selectivity to normal cells. The authors also noted that
Through Amino Acid increasing the overall charge of the ACP by the
Substitution/Modification addition of two additional cationic amino acids to
the major sector resulted in reduced potency;
It is no secret that slight alterations to the primary however, selectivity for malignant cells was
amino acid sequence can drastically affect the maintained, most likely because the addition of
potency of an ACP. In some cases, a minor altera- these two amino acids occurred at the expense of
tion may even change the mechanism of action of two hydrophobic amino acids.
the peptide. For example, the addition of a In spite of numerous efforts to generate next-­
C-terminal amide causes the wasp venom peptide generation ACPs with improved selectivity for
mastoparan to become lytic, whereas in its cancer cells, we still do not really understand the
unamidated form, mastoparan induces structural basis for cancer cell selectivity. It is our
mitochondria-­dependent apoptosis (Hilchie et al. opinion that this is due to the lack of available
2016; de Azevedo et al. 2015). datasets that are sufficiently large to conduct
Many groups have attempted to improve pep- thorough structure/activity relationship (SAR)
tide selectivity through amino acid substitution. studies. To this end, we have used SPOT array
The vast majority of these studies use hypothesis-­ technologies to create a massive peptide library
driven, small-scale approaches, whereby charge (n = 210), which we then screened for cytotoxic
and/or hydrophobicity of the parent ACP is modi- activity against cancer cells and normal cells
fied, based on the knowledge that these features (manuscript in preparation). We found that single
are required for cancer cell killing (Hoskin and amino acid substitutions may eliminate cytotox-
Ramamoorthy 2008). Such studies generate a icity for both cancer cells and normal cells, elimi-
very small peptide library that is subsequently nate selectivity for cancer cells, and/or improve
screened for cytotoxic activity against cancer selectivity for leukemia and/or breast cancer
cells and normal cells. This approach has been cells. Our goal is to use an artificial intelligence
used to identify novel peptides with slightly approach to predict highly selective next-­
increased selectivity for cancer cells (Dennison generation ACPs through computer modeling of
et al. 2006; Yang et al. 2003; Eliassen et al. 2003; quantitative structure/activity relationships
Arias et al. 2017). Often, many incremental (QSAR), which yielded hundreds of peptides
improvements are needed before one obtains an predicted to be more selective for cancer cells
ACP with significantly improved selectivity rela- than the parent ACP. Efforts to screen this new
tive to the parent peptide, likely because we still peptide library are underway. While this study is
do not understand how the overall structure of the in its infancy, we are confident that highly selec-
ACP affects its selectivity for cancer cells. To our tive ACPs will be identified as this approach has
knowledge, only one study has examined the successfully delivered novel peptides with
effect of altered charge and hydrophobicity on improved antimicrobial and anti-biofilm activi-
cancer cell selectivity in the context of the overall ties (Hilpert et al. 2005; Haney et al. 2018).
9 Anticancer Activities of Natural and Synthetic Peptides 143

9.5.3 Improving Tumor Selectivity the 1980s without any significant progress to
Through Histidine clinical trials. Areas in which further study is
Substitution essential include ascertaining the immuno-mod-
ulatory properties of ACPs and improving their
It is well established that the microenvironment selectivity for cancer cells under physiologically
of solid tumors is acidic in comparison to most relevant conditions. Indeed, ACP-mediated
normal tissues due to lactic acid buildup coupled induction of danger signals and the subsequent
with inadequate washout of acidic products as a development of anticancer immune responses
result of inadequate vascularization (Tannock may be essential for long-term benefit and
and Rotin 1989; Newell et al. 1993; Vaupel et al. increased patient survival. Moreover, improved
1989). Yechiel Shai’s group has used these differ- selectivity is essential for future clinical trials to
ences in tumor microenvironment to optimize ensure the safety and efficacy of ACP adminis-
ACP selectivity through the use of histidine sub- tration to cancer patients. ACPs must be able to
stitutions (Makovitzki et al. 2009). In this innova- kill cancer cells without adverse toxicities.
tive approach, three or six lysine residues in the Although it is possible that a single ACP may be
ACP [D]-K6L9 (pKa ~10.5) were replaced with effective against all cancer types, it is more
histidine residues (pKa ~6.1), generating likely that different ACPs will be needed to treat
[D]-K3H3L9 and [D]-H6L9, respectively. Unlike different cancers. Computer modeling may help
[D]-K6L9, neither [D]-K3H3L9 nor [D]-H6L9 had to advance this area of research so that future
adverse toxic side effects when delivered to mice treatments can be identified and assessed at a
via intravenous injection, and both ACPs caused more rapid pace. In addition, computer models
a reduction in the growth of prostate tumor xeno- may predict highly selective ACPs that also acti-
grafts in mice. These results provide the intrigu- vate antitumor immune responses.
ing possibility of customizing peptides for Treating patients with combinational thera-
selective targeting of the solid tumor microenvi- pies including novel drugs like ACPs is going to
ronment, thereby sparing healthy tissues from be essential to combat multidrug-resistant can-
potential adverse side effects. Despite these cers. Generation of ACP-producing oncolytic
exciting results, to our knowledge, this proof of viruses may be an additional area for future study
concept work has not been replicated with other to combat recurrence and multidrug-resistant
ACPs. However, it is worth noting that our QSAR cancers. Finally, the use of tumoricidal ACPs in
analysis predicts that peptide selectivity for combination with conventional cytotoxic drugs is
breast carcinoma cells often involves histidine likely to improve patient survival by more effec-
substitutions (manuscript in preparation). tive lysis of tumor cells with reduced treatment-­
related toxicities and a reduction in the risk of
tumor recurrence as the result of the generation
9.6 The Future of ACP Research of long-lasting tumor-specific immunity.

Cancer cells are increasingly resistant to conven-

tional treatment modalities. As patient survival References
increases, so does the risk of recurrent disease in
a form that is resistant to previously used drugs. Anastasi A, Erspamer V, Bucci M (1971) Isolation and
structure of bombesin and alytesin, 2 analogous active
Peptide-based therapies have the potential to peptides from the skin of the European amphibians
treat many different cancers, including those that Bombina and Alytes. Experientia 27:166–167
are multidrug resistant or slow growing and Andtbacka RHI et al (2015) Talimogene Laherparepvec
therefore not susceptible to conventional chemo- improves durable response rate in patients with
advanced melanoma. J Clin Oncol 33:2780–2788
therapy. Unfortunately, research in this area has Arias M et al (2017) Anticancer activities of bovine and
been slow to progress. The anticancer potential human lactoferricin-derived peptides. Biochem Cell
of magainins has been appreciated since at least Biol 95:91–98
144 A. L. Hilchie et al.

Bafna S, Kaur S, Batra SK (2010) Membrane-bound Dennison SR, Whittaker M, Harris F, Phoenix DA (2006)
mucins: the mechanistic basis for alterations in Anticancer alpha-helical peptides and structure/func-
the growth and survival of cancer cells. Oncogene tion relationships underpinning their interactions
29:2893–2904 with tumour cell membranes. Curr Protein Pept Sci
Baker MA, Maloy WL, Zasloff M, Jacob LS (1993) 7:487–499
Anticancer efficacy of magainin2 and analogue pep- Donnelly JG (2004) Pharmacogenetics in cancer chemo-
tides. Cancer Res 53:3052–3057 therapy: balancing toxicity and response. Ther Drug
Berge G et al (2010) Therapeutic vaccination against a Monit 26:231–235
murine lymphoma by intratumoral injection of a cat- Eike L-M, Yang N, Rekdal Ø, Sveinbjørnsson B (2015)
ionic anticancer peptide. Cancer Immunol Immunother The oncolytic peptide LTX-315 induces cell death and
59:1285–1294 DAMP release by mitochondria distortion in human
Bergman P, Walter-Jallow L, Broliden K, Agerberth B, melanoma cells. Oncotarget 6:34910–34923
Söderlund J (2007) The antimicrobial peptide LL-37 Eliassen LT et al (2002) Evidence for a direct antitu-
inhibits HIV-1 replication. Curr HIV Res 5:410–415 mor mechanism of action of bovine lactoferricin.
Bhutia SK, Maiti TK (2008) Targeting tumors with Anticancer Res 22:2703–2710
peptides from natural sources. Trends Biotechnol Eliassen LT, Haug BE, Berge G, Rekdal O (2003)
26:210–217 Enhanced antitumour activity of 15-residue bovine
Bush JA, Li G (2002) Cancer chemoresistance: the rela- lactoferricin derivatives containing bulky aromatic
tionship between p53 and multidrug transporters. Int amino acids and lipophilic N-terminal modifications.
J Cancer 98:323–330 J Pept Sci 9:510–517
Camilio KA, Berge G, Ravuri CS, Rekdal O, Eliassen LT et al (2006) The antimicrobial peptide, lacto-
Sveinbjørnsson B (2014a) Complete regression and ferricin B, is cytotoxic to neuroblastoma cells in vitro
systemic protective immune responses obtained in and inhibits xenograft growth in vivo. Int J Cancer
B16 melanomas after treatment with LTX-315. Cancer 119:493–500
Immunol Immunother 63:601–613 Furlong SJ, Mader JS, Hoskin DW (2006) Lactoferricin-­
Camilio KA, Rekdal O, Sveinbjörnsson B (2014b) induced apoptosis in estrogen-nonresponsive
LTX-315 (Oncopore™): a short synthetic antican- MDA-MB-435 breast cancer cells is enhanced by C6
cer peptide and novel immunotherapeutic agent. ceramide or tamoxifen. Oncol Rep 15:1385–1390
Oncoimmunology 3:e29181 Furlong SJ, Mader JS, Hoskin DW (2010) Bovine lac-
Chen J, Pan H, Lanza GM, Wickline SA (2013) toferricin induces caspase-independent apoptosis in
Perfluorocarbon nanoparticles for physiological and human B-lymphoma cells and extends the survival of
molecular imaging and therapy. Adv Chronic Kidney immune-deficient mice bearing B-lymphoma xeno-
Dis 20:466–478 grafts. Exp Mol Pathol 88:371–375
Chen X et al (2018) Roles and mechanisms of human Furue M et al (2018) Melanoma and immune checkpoint
cathelicidin LL-37 in cancer. Cell Physiol Biochem inhibitors. Curr Oncol Rep 20:29
47:1060–1073 Gazit E, Boman A, Boman HG, Shai Y (1995) Interaction
Chiocca EA, Rabkin SD (2014) Oncolytic viruses and of the mammalian antibacterial peptide cecropin P1
their application to cancer immunotherapy. Cancer with phospholipid vesicles. Biochemistry (Mosc)
Immunol Res 2:295–300 34:11479–11488
Choi DK, Helenowski I, Hijiya N (2014) Secondary malig- Giuliani A, Pirri G, Nicoletto S (2007) Antimicrobial pep-
nancies in pediatric cancer survivors: perspectives and tides: an overview of a promising class of therapeutics.
review of the literature. Int J Cancer 135:1764–1773 Cent Eur J Biol 2:1–33
ClinicalTrials.gov (NCT01058616) (2010) Greenshields AL, Knickle LC, Syvitski R, Douglas SE
ClinicalTrials.gov (NCT01986426) (2013) (2008) Strategies for recombinant expression of small,
Cornelio DB, Roesler R, Schwartsmann G (2007) highly disulphide-bonded, cationic antimicrobial pep-
Gastrin-releasing peptide receptor as a molecular tides. Protein Pept Lett 15:985–994
target in experimental anticancer therapy. Ann Oncol Gupta K, Subramanian H, Ali H (2016) Modulation of
18:1457–1466 host defense peptide-mediated human mast cell acti-
Couto J et al (2018) Antiplasmodial activity of tick defen- vation by LPS. Innate Immun 22:21–30
sins in a mouse model of malaria. Ticks Tick-Borne Haney EF et al (2018) Computer-aided discovery of pep-
Dis 9:844–849 tides that specifically attack bacterial biofilms. Sci Rep
de Azevedo RA et al (2015) Mastoparan induces apopto- 8:1871
sis in B16F10-Nex2 melanoma cells via the intrinsic Hansel W, Enright F, Leuschner C (2007) Destruction of
mitochondrial pathway and displays antitumor activity breast cancers and their metastases by lytic peptide
in vivo. Peptides 68:113–119 conjugates in vitro and in vivo. Mol Cell Endocrinol
de la Fuente-Núñez C et al (2012) Inhibition of bacte- 260–262:183–189
rial biofilm formation and swarming motility by a Haug BE et al (2016) Discovery of a 9-mer cationic pep-
small synthetic cationic peptide. Antimicrob Agents tide (LTX-315) as a potential first in class oncolytic
Chemother 56:2696–2704 peptide. J Med Chem 59:2918–2927
9 Anticancer Activities of Natural and Synthetic Peptides 145

Hayworth JL et al (2009) Attenuation of massive cytokine a calmodulin-peptide fusion system. J Am Chem Soc
response to the staphylococcal enterotoxin B superan- 138:11318–11326
tigen by the innate immunomodulatory protein lacto- Jean F, Tomasini P, Barlesi F (2017) Atezolizumab: fea-
ferrin. Clin Exp Immunol 157:60–70 sible second-line therapy for patients with non-small
Heo J et al (2013) Randomized dose-finding clinical trial cell lung cancer? A review of efficacy, safety and place
of oncolytic immunotherapeutic vaccinia JX-594 in in therapy. Ther Adv Med Oncol 9:769–779
liver cancer. Nat Med 19:329–336 Jiang Y et al (2018) TEMPO-oxidized starch nanoassem-
Hilchie AL, Hoskin DW (2010) Application of cationic blies of negligible toxicity compared with polyacrylic
antimicrobial peptides in cancer treatment: laboratory acids for high performance anti-cancer therapy. Int
investigations and clinical potential. In: Emerging can- J Pharm 547:520–529
cer therapy: microbial approaches and biotechnologi- Johnstone SA, Gelmon K, Mayer LD, Hancock RE, Bally
cal tools. Wiley, Hoboken, pp 309–332 MB (2000) In vitro characterization of the antican-
Hilchie AL et al (2011) Pleurocidin-family cationic anti- cer activity of membrane-active cationic peptides.
microbial peptides are cytolytic for breast carcinoma I. Peptide-mediated cytotoxicity and peptide-enhanced
cells and prevent growth of tumor xenografts. Breast cytotoxic activity of doxorubicin against wild-type
Cancer Res 13:R102 and p-glycoprotein over-expressing tumor cell lines.
Hilchie AL, Power Coombs MR, Hoskin DW (2012) Anticancer Drug Des 15:151–160
Obstacles and solutions to the use of cationic anti- Jung HJ et al (2007) Fungicidal effect of pleurocidin by
microbial peptides in the treatment of cancer. membrane-active mechanism and design of enantio-
In: Small wonders: peptides for disease control. meric analogue for proteolytic resistance. Biochim
American Chemical Society/Oxford University Press, Biophys Acta 1768:1400–1405
Washington, DC/Oxford, pp 61–78 Kebudi R, Ozdemir GN (2017) Secondary neoplasms in
Hilchie AL, Wuerth K, Hancock REW (2013a) Immune children treated for cancer. Curr Pediatr Rev 13:34–41
modulation by multifaceted cationic host defense Kim S, Kim SS, Bang YJ, Kim SJ, Lee BJ (2003) In vitro
(antimicrobial) peptides. Nat Chem Biol 9:761–768 activities of native and designed peptide antibiotics
Hilchie AL, Vale R, Zemlak TS, Hoskin DW (2013b) against drug sensitive and resistant tumor cell lines.
Generation of a hematologic malignancy-selective Peptides 24:945–953
membranolytic peptide from the antimicrobial core Kindrachuk J et al (2009) A novel vaccine adjuvant com-
(RRWQWR) of bovine lactoferricin. Exp Mol Pathol prised of a synthetic innate defence regulator peptide
95:192–198 and CpG oligonucleotide links innate and adaptive
Hilchie AL et al (2013c) Pleurocidin-family cationic anti- immunity. Vaccine 27:4662–4671
microbial peptides mediate lysis of multiple myeloma Koo C-Y, Sen Y-P, Bay B-H, Yip GW (2008) Targeting
cells and impair the growth of multiple myeloma heparan sulfate proteoglycans in breast cancer treat-
xenografts. Leuk Lymphoma 54:2255–2262 ment. Recent Pat Anticancer Drug Discov 3:151–158
Hilchie AL, Haney EF, Pinto DM, Hancock REW, Hoskin Kube S et al (2017) Fusogenic liposomes as nanocarri-
DW (2015) Enhanced killing of breast cancer cells by ers for the delivery of intracellular proteins. Langmuir
a d-amino acid analog of the winter flounder-derived 33:1051–1059
pleurocidin NRC-03. Exp Mol Pathol 99:426–434 Leuschner C, Hansel W (2004) Membrane disrupting
Hilchie AL et al (2016) Mastoparan is a membranolytic lytic peptides for cancer treatments. Curr Pharm Des
anti-cancer peptide that works synergistically with 10:2299–2310
gemcitabine in a mouse model of mammary carci- Liu S et al (2011) Enhancement of cytotoxicity of anti-
noma. Biochim Biophys Acta 1858:3195–3204 microbial peptide magainin II in tumor cells by
Hilpert K, Volkmer-Engert R, Walter T, Hancock REW bombesin-­ targeted delivery. Acta Pharmacol Sin
(2005) High-throughput generation of small antibac- 32:79–88
terial peptides with improved activity. Nat Biotechnol Mader JS, Hoskin DW (2006) Cationic antimicrobial pep-
23:1008–1012 tides as novel cytotoxic agents for cancer treatment.
Hoskin DW, Ramamoorthy A (2008) Studies on anti- Expert Opin Investig Drugs 15:933–946
cancer activities of antimicrobial peptides. Biochim Mader JS, Salsman J, Conrad DM, Hoskin DW (2005)
Biophys Acta 1778:357–375 Bovine lactoferricin selectively induces apoptosis in
Hui L, Leung K, Chen HM (2002) The combined effects of human leukemia and carcinoma cell lines. Mol Cancer
antibacterial peptide cecropin A and anti-cancer agents Ther 4:612–624
on leukemia cells. Anticancer Res 22:2811–2816 Mader JS, Ewen C, Hancock REW, Bleackley RC (2011)
Iida J, Meijne AM, Knutson JR, Furcht LT, McCarthy JB The human cathelicidin, LL-37, induces granzyme-­
(1996) Cell surface chondroitin sulfate proteoglycans mediated apoptosis in regulatory T cells. J Immunother
in tumor cell adhesion, motility and invasion. Semin 34:229–235
Cancer Biol 7:155–162 Madera L, Hancock REW (2012) Synthetic immunomod-
Ishida H, Nguyen LT, Gopal R, Aizawa T, Vogel HJ (2016) ulatory peptide IDR-1002 enhances monocyte migra-
Overexpression of antimicrobial, anticancer, and tion and adhesion on fibronectin. J Innate Immun
transmembrane peptides in Escherichia coli through 4:553–568
146 A. L. Hilchie et al.

Makovitzki A, Fink A, Shai Y (2009) Suppression of Seidel JA, Otsuka A, Kabashima K (2018) Anti-PD-1
human solid tumor growth in mice by intratumor and anti-CTLA-4 therapies in cancer: mechanisms of
and systemic inoculation of histidine-rich and pH-­ action, efficacy, and limitations. Front Oncol 8:86
dependent host defense-like lytic peptides. Cancer Res Shadidi M, Sioud M (2003) Identification of novel carrier
69:3458–3463 peptides for the specific delivery of therapeutics into
Matsuzaki K et al (1989) Magainin 1-induced leakage cancer cells. FASEB J 17:256–258
of entrapped calcein out of negatively-charged lipid Sheu MJ, Baldwin WW, Brunson KW (1985) Cytotoxicity
vesicles. Biochim Biophys Acta 981:130–134 of rabbit macrophage peptides MCP-1 and MCP-2 for
Murugan C et al (2016) Combinatorial nanocarrier based mouse tumor cells. Antimicrob Agents Chemother
drug delivery approach for amalgamation of anti-­ 28:626–629
tumor agents in breast cancer cells: an improved nano- Soman NR et al (2009) Molecularly targeted nanocarri-
medicine strategies. Sci Rep 6:34053 ers deliver the cytolytic peptide melittin specifically
Nagy P et al (2005) Decreased accessibility and lack of to tumor cells in mice, reducing tumor growth. J Clin
activation of ErbB2 in JIMT-1, a herceptin-resistant, Invest 119:2830–2842
MUC4-expressing breast cancer cell line. Cancer Res Steinstraesser L et al (2012) Innate defense regulator
65:473–482 peptide 1018 in wound healing and wound infection.
Naumov GN et al (2003) Ineffectiveness of doxorubicin PLoS One 7:e39373
treatment on solitary dormant mammary carcinoma Tannock IF, Rotin D (1989) Acid pH in tumors and its
cells or late-developing metastases. Breast Cancer Res potential for therapeutic exploitation. Cancer Res
Treat 82:199–206 49:4373–4384
Newell K, Franchi A, Pouysségur J, Tannock I (1993) Thirman MJ, Larson RA (1996) Therapy-related myeloid
Studies with glycolysis-deficient cells suggest that leukemia. Hematol Oncol Clin North Am 10:293–320
production of lactic acid is not the only cause of tumor Utsugi T, Schroit AJ, Connor J, Bucana CD, Fidler IJ
acidity. Proc Natl Acad Sci U S A 90:1127–1131 (1991) Elevated expression of phosphatidylserine
Nguyen LT, Haney EF, Vogel HJ (2011) The expanding in the outer membrane leaflet of human tumor cells
scope of antimicrobial peptide structures and their and recognition by activated human blood monocytes.
modes of action. Trends Biotechnol 29:464–472 Cancer Res 51:3062–3066
Nijnik A et al (2010) Synthetic cationic peptide IDR-­ van Beek WP, Smets LA, Emmelot P (1973) Increased
1002 provides protection against bacterial infections sialic acid density in surface glycoprotein of trans-
through chemokine induction and enhanced leukocyte formed and malignant cells – a general phenomenon?
recruitment. J Immunol 184:2539–2550 Cancer Res 33:2913–2922
Overhage J et al (2008) Human host defense peptide Vaupel P, Kallinowski F, Okunieff P (1989) Blood flow,
LL-37 prevents bacterial biofilm formation. Infect oxygen and nutrient supply, and metabolic microen-
Immun 76:4176–4182 vironment of human tumors: a review. Cancer Res
Pan H, Soman NR, Schlesinger PH, Lanza GM, 49:6449–6465
Wickline SA (2011) Cytolytic peptide nanoparticles Vlieghe P, Lisowski V, Martinez J, Khrestchatisky M
(‘NanoBees’) for cancer therapy. Wiley Interdiscip (2010) Synthetic therapeutic peptides: science and
Rev Nanomed Nanobiotechnol 3:318–327 market. Drug Discov Today 15:40–56
Pillai RN et al (2018) Comparison of the toxicity profile Wang G, Watson KM, Buckheit RW (2008) Anti-human
of PD-1 versus PD-L1 inhibitors in non-small cell immunodeficiency virus type 1 activities of antimicro-
lung cancer: a systematic analysis of the literature. bial peptides derived from human and bovine catheli-
Cancer 124:271–277 cidins. Antimicrob Agents Chemother 52:3438–3440
Pundir P, Catalli A, Leggiadro C, Douglas SE, Kulka Winter PM (2014) Perfluorocarbon nanoparticles: evolu-
M (2014) Pleurocidin, a novel antimicrobial pep- tion of a multimodality and multifunctional imaging
tide, induces human mast cell activation through the agent. Scientifica 2014:746574
FPRL1 receptor. Mucosal Immunol 7:177–187 Yamazaki T et al (2016) The oncolytic peptide LTX-315
Reubi JC, Wenger S, Schmuckli-Maurer J, Schaer J-C, overcomes resistance of cancers to immunotherapy
Gugger M (2002) Bombesin receptor subtypes in with CTLA4 checkpoint blockade. Cell Death Differ
human cancers: detection with the universal radio- 23:1004–1015
ligand 125I-[D-TYR(6), β-ALA(11), PHE(13), Yanashima K et al (2017) Innate defense regulator IDR-­
NLE(14)] bombesin(6-14). Clin Cancer Res 1018 activates human mast cells through G protein-,
8:1139–1146 phospholipase C-, MAPK- and NF-ĸB-sensitive path-
Richardson A, de Antueno R, Duncan R, Hoskin DW ways. Immunol Res 65:920–931
(2009) Intracellular delivery of bovine lactoferricin’s Yang N, Stensen W, Svendsen JS, Rekdal Ø (2002)
antimicrobial core (RRWQWR) kills T-leukemia cells. Enhanced antitumor activity and selectivity of
Biochem Biophys Res Commun 388:736–741 lactoferrin-­derived peptides. J Pept Res 60:187–197
9 Anticancer Activities of Natural and Synthetic Peptides 147

Yang N, Lejon T, Rekdal O (2003) Antitumour activity Zachowski A (1993) Phospholipids in animal eukaryotic
and specificity as a function of substitutions in the membranes: transverse asymmetry and movement.
lipophilic sector of helical lactoferrin-derived peptide. Biochem J 294(Pt 1):1–14
J Pept Sci 9:300–311 Zitzmann S, Ehemann V, Schwab M (2002) Arginine-­
Yeaman MR, Yount NY (2003) Mechanisms of antimi- glycine-­
aspartic acid (RGD)-peptide binds to both
crobial peptide action and resistance. Pharmacol Rev tumor and tumor-endothelial cells in vivo. Cancer Res
55:27–55 62:5139–5143
 Antimicrobial Host Defence
Peptides: Immunomodulatory 10
Functions and Translational
Prospects

Anne M. van der Does, Pieter S. Hiemstra,

and Neeloffer Mookherjee

Abstract important role in the control of infections,

Cationic host defence peptides (CHDPs), also regulation of inflammation, and maintaining
known as antimicrobial peptides, exhibit a immune homeostasis. It is thus not surprising
wide range of activities contributing to that dysregulation of expression of CHDPs is
immune responses and resolution of infec- implicated in the susceptibility, pathology,
tions. CHDPs are expressed across diverse and progression of various diseases. In this
species, are generally amphipathic with less chapter, we summarize the immunomodula-
than 50 amino acids in length, and differ sig- tory functions of CHDPs, its clinical rele-
nificantly in sequence and structure. This vance, and the translational opportunities that
chapter focuses on the role of these peptides in these peptides provide for the development of
immunity. CHDPs are known to function in new therapies.
both innate and adaptive immune responses.
These peptides exert both pro- and anti-­ Keywords
inflammatory properties, which are likely con- Inflammation · Immunomodulation ·
text dependent based on cell and tissue type, Cathelicidin · Defensin · LL-37 · IDR
concentration of the peptides, and its interac- peptides
tion with other factors in the microenviron-
ment. Furthermore, the crosstalk between
CHDPs and the microbiome and how this may
influence mucosal immunity is a rapidly 10.1 Introduction
emerging field of research. Overall, the immu-
nomodulatory functions of CHDPs play an Cationic host defence peptides (CHDPs), also
known as antimicrobial peptides (AMPs), are
immune effector molecules with antimicrobial
A. M. van der Does · P. S. Hiemstra functions. These natural peptides are critical for
Department of Pulmonology, Leiden University resolution of infections and exhibit a wide range
Medical Center, Leiden, The Netherlands
e-mail: [email protected]; of immune functions. CHDPs are typically
[email protected] amphipathic, less than 50 amino acids in length
N. Mookherjee (*) with a net positive charge of +2 to +9 at physio-
Manitoba Centre for Proteomics and Systems logical pH 7.4 (Powers and Hancock 2003;
Biology, Department of Internal Medicine, University Epand and Vogel 1999; Choi and Mookherjee
of Manitoba, Winnipeg, Canada 2012). CHDPs are expressed across diverse
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 149

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_10
150 A. M. van der Does et al.

s­ pecies including microbes, plants, insects, crus- CHDPs is now widely used to encompass both
taceans, amphibians, reptiles, and mammals the direct antimicrobial functions and immuno-
(Ganz 2003a, b; Nakamura et al. 1988; Cowland modulatory properties exhibited by these natural
et al. 1995; Fernandez de Caleya et al. 1972; cationic peptides. The fundamental functions and
Simmaco et al. 1993, 1996; Hultmark et al. emerging themes of clinical applications of
1982). The first report on AMPs was by Kiss and CHDPs, with focus on mammalian cathelicidins
Michl in the 1960s with the description of the and defensins, are reviewed in this chapter.
peptide bombinin from speckled frog Bombina
variegata (Simmaco et al. 2009). In the 1980s,
cecropin peptide was isolated from moths by 10.1.1 Expression and Processing
Boman’s group (Steiner et al. 1981) followed by of CHDPs
the characterization of defensins from human
neutrophils by Ganz and Lehrer (Ganz et al. CHDPs differ significantly in sequence and struc-
1985) and magainins from amphibians by Zasloff tures and can be broadly classified into four
et al. (Zasloff 1987). Over the last four decades, groups based on conformational structures: those
various studies have defined broad-spectrum with linear α-helical structure (e.g., cathelicidin),
antimicrobial activity of natural cationic peptides β-sheet structure with disulfide bridges (e.g.,
against bacteria, viruses, fungi, and parasites α-defensin), cyclic structures (e.g., catestatin),
(Powers and Hancock 2003; Epand and Vogel and extended and flexible loop structures (e.g.,
1999; Ganz 2003a, b; Hancock and Lehrer 1998; indolicidin) (Epand and Vogel 1999; Ganz 2003b;
Barlow et al. 2011; Larrick et al. 1995). With the Hancock and Lehrer 1998; Hancock et al. 2016;
rise of microbial resistance to antibiotics and Wang et al. 2016). The two most well-­
simultaneous lack of development of new antibi- characterized families of CHDPs in mammals are
otics, there has been an intense focus on the the cathelicidins and defensins. Cathelicidins are
development of new antimicrobials using cat- characterized by a conserved cathelin-like
ionic antimicrobial peptides. Several mecha- N-terminal domain which forms a random-coil
nisms underlying the direct antimicrobial conformation in hydrophilic environment,
functions of AMPs have been proposed over the whereas the C-terminal domain corresponding to
years, including direct electrostatic interaction of the mature peptide exhibits diversity in structures
the peptide with microbial membranes leading to which is predominantly α-helical or can adopt
membrane permeabilization and destabilization, elongated or beta-hairpin conformations (De
intracellular translocation of the peptide and inhi- Smet and Contreras 2005; Tomasinsig and
bition of microbial DNA/RNA and protein syn- Zanetti 2005). Defensins are peptides with six
thesis, and triggering autolysis (Seil et al. 2010; cysteine residues that form intramolecular disul-
Wang et al. 2014; Sierra et al. 2017). However, fide bridges and β-sheet conformation (De Smet
studies in the last two decades have demonstrated and Contreras 2005). CHDPs can be expressed
that the direct microbicidal functions of certain constitutively or induced depending on the cell
cationic peptides, e.g., the human cathelicidin type, tissue, and extracellular stimulus. These
peptide LL-37 and β-defensin, are compromised peptides are expressed throughout the body such
in the presence of host factors such as physiologi- as in different cell types of hematopoietic origin
cal salt concentrations (divalent cations Mg2+, (e.g., lymphocytes, macrophages, neutrophils)
Ca2+) and anionic polysaccharides (Scott et al. and in structural cells (e.g., airway epithelial cells
2002; Bowdish et al. 2005). Despite the antago- and keratinocytes) and can be found in various
nizing properties of physiological host factors, body fluids such as sweat, breast milk, plasma,
these peptides are essential for resolution of and saliva (Doss et al. 2010).
infections. Concomitant with this, studies have Humans express only one cathelicidin, hCAP-­
demonstrated effects of endogenous cationic 18/LL-37, which is encoded by the camp gene,
peptides on the host immune system and wound transcription of which results in the generation of
repair (discussed below). Therefore, the term a pre-propeptide and an inactive precursor hCAP-­
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 151

18 (Cowland et al. 1995; Agerberth et al. 1995). β-Defensins are encoded by distinct genes and
hCAP-18 undergoes proteolytic cleavage by pro- similar to cathelicidins are also produced as pre-­
teases such as proteinase 3 in neutrophils propeptides, which require one or more proteo-
(Sorensen et al. 2001) or kallikrein 5 and 7 in lytic step to generate the mature peptides. Various
skin (Yamasaki et al. 2006) which leads to the proteases have been reported to be involved in the
loss of the N-terminal cathelin domain and the processing of the pre-propeptide for the genera-
release of the active, mature 37 amino acid pep- tion of the mature defensins such as neutrophil-­
tide LL-37 from the C-terminus. LL-37 can be derived proteinase 3 and elastase process HNP1
further cleaved by proteases and unknown mole- (Valore and Ganz 1992; Tongaonkar et al. 2012),
cules (also derived from the microbiome trypsin is involved in the processing of HD5
(Sieprawska-Lupa et al. 2004)) to smaller frag- (Ghosh et al. 2002), whereas in mice α-defensins
ments, some of which exhibit increased antimi- are processed by metalloproteinase 7 or matrily-
crobial activity but reduced immunomodulatory sin (Wilson et al. 1999). α-Defensin expression
properties (Murakami et al. 2004). hCAP-18/ seems to be cell-type specific; HNP1–4 are pri-
LL-37 levels are highest in neutrophils where the marily expressed by human neutrophils, where
inactive pro-form is stored in secondary or spe- they localize in primary or azurophilic granules
cific granules (Cowland et al. 1995; Gudmundsson (Ganz 2003a). Other immune cell types, for
et al. 1996). LL-37 is also expressed by other example, NK cells and T-cell subsets, also
immune cells including monocytes, NK cells, express HNP1–4 (Obata-Onai et al. 2002). HD5
and lymphocytes and by several types of epithe- and HD6 are expressed by Paneth cells of the
lial cells (Agerberth et al. 2000; Frohm et al. small intestine (Jones and Bevins 1992; Jones
1997; Chromek et al. 2006; Bals et al. 1998). and Bevins 1993), and HD5 is also expressed in
Camp expression in these cell types is usually the endometrium and fallopian tubes (Quayle
relatively low during homeostasis, but transcrip- et al. 1998). β-Defensins are expressed more
tion can be strongly increased upon activation by ubiquitously relative to α-defensin, although
pathogens such as Salmonella enterica serovar mostly by epithelial cells, hBD3 and hBD4 are
Dublin or E. coli (Hase et al. 2002) or by pathogen-­ also expressed in the endometrium (King et al.
derived molecules like LPS or LTA (Nell et al. 2003), hBD5 and 6 can be found in the epididy-
2004). Furthermore, signals not associated with mis (Yamaguchi et al. 2002), and hBD25–29 can
“danger” can also regulate camp expression, be found primarily in the male genital tract
including vitamin D3 (Yim et al. 2007), curcumin (Rodriguez-Jimenez et al. 2003). Expression lev-
(Guo et al. 2013), and several histone deacetylase els of defensins are differently regulated; for
inhibitors (HDACi) (Schauber et al. 2004). In example, hBD1 is expressed in both a constitu-
addition, specific levels of ­endoplasmic reticulum tive and inducible manner, whereas hBD2 and 3
stress can promote transcription of LL-37 (Park are usually expressed in low levels during homeo-
et al. 2011). Camp gene expression is not always stasis but transcription increases in response to,
increased upon a­ ctivation of the cell with patho- e.g., microbial exposure or inflammation (Harder
gens or associated ­molecules, for example, infec- et al. 2004; Kuwano et al. 2006; Harder et al.
tions with Shigella (Islam et al. 2001) or Neisseria 2001). A recent study has shown that levels of
(Bergman et al. 2005) result in a decreased expres- α-defensin can be increased in the lungs in
sion in infected cells. response to inhaled air pollution (Piyadasa et al.
Defensins can be classified, based on the loca- 2018b). In contrast, virulence factors of pulmo-
tion of their disulfide bonds, into α-, β-, and nary pathogens and air pollution-related
θ-defensins, of which only α- and β-defensins are ­particulate matter can decrease the expression of
expressed by humans (Ganz 2003a) and defensin genes (Laube et al. 2006).
θ-defensins are only expressed in nonhuman pri- Physiological concentrations of CHDPs vary
mates. Six α-defensins (HNP1–4 and HD5–6) within the body, with higher concentrations (in
and 11 β-defensins (hBD1–6 and hBD25–29) the range of mg/mL) found in the granules of leu-
have so far been identified in humans. α- and kocytes and at the bottom of intestinal crypts
152 A. M. van der Does et al.

where Paneth cells are located and lower concen- et al. 2016; Hiemstra 2015; Choi et al. 2012). The
trations (in the range of ng/mL to μg/mL) found findings on the impact of CHPDs on the immune
in the mucosa and in circulation (Murakami et al. system have raised discussions related to the pri-
2002, 2004; Frohm et al. 1997; Gordon et al. mary function of these peptides and the relative
2005; Agerberth et al. 1999; Woo et al. 2003). contribution of these functions versus direct anti-
Likely these concentrations are higher in the close microbial activity of the peptides in host defense
vicinity of the cells that produce these peptides, against infections. Furthermore, the cellular
such as in the pericellular layer overlying the air- receptor systems employed by CHDPs to medi-
way epithelial cells. As mentioned above, concen- ate the diverse range of effects on the host is also
trations of these peptides are typically increased an area of intense investigation in recent years.
in response to infectious challenge and inflamma- Nevertheless, the pleiotropic activity of these
tory signals. Epithelial wound repair has also been peptides results in both pro- and anti-­inflammatory
shown to increase the expression of CHDPs roles and enhances immune activation as well as
(Sorensen et al. 2003). A wide range of immuno- contributes to immune regulation, thus compli-
modulatory effects of CHDPs contribute to regu- cating the interpretation of the role of CHDPs. It
lation of inflammation and immunity and thus is likely that the exact contribution of CHDPs to
resolution of infections. Immunity-related func- the regulation of inflammation and immunity is
tions of these peptides include induction of cyto- context dependent.
kine and chemokine production, promotion of
leukocyte recruitment to sites of infection and 10.1.2.1 Pro- and Anti-inflammatory
inflammation, and modulation of dendritic cell Role of CHDPs
and lymphocyte proliferation and differentiation in the Regulation of Innate
(Choi and Mookherjee 2012; Hancock et al. 2016; Immunity and Inflammation
Bucki et al. 2010; Hemshekhar et al. 2016; Neutrophils are a rich source of various CHDPs,
Hiemstra 2015). These immunomodulatory activ- including the human cathelicidin hCAP-18/LL-37
ities of CHDPs are discussed in the next section. and α-defensins. Degranulation and other forms of
release from neutrophils may therefore generate
high local concentrations of CHDPs that can either
10.1.2 CHDPs in Inflammation be released into the environment or become immo-
and Immunity bilized in neutrophil extracellular traps (NETs). In
addition, CHDPs, such as LL-37, can also facili-
Most of the initial CHDPs identified were iso- tate the formation of NETs (Neumann et al. 2014).
lated using activity-guided purification proce- CHDPs from other cellular sources can accumu-
dures and assessment of antimicrobial activity to late locally at high concentrations, especially if
identify active components. Later on in silico these are stored in granules such as in the intestinal
approaches were also used to identify novel Paneth cells or mast cells. At high concentrations
CHDPs. This explains the initial focus in studies such peptides are found to display cytotoxic activ-
on the biology of CHDPs on their antimicrobial ity toward cultured cells (Okrent et al. 1990).
activity against a wide range of microorganisms Subsequent in vitro and in vivo studies showed
and parasites. However, relatively early on other that this may also contribute to accumulation of
activities were also discovered, including their inflammatory cells by direct and/or indirect che-
ability to modulate inflammation and immunity moattraction (Rehaume and Hancock 2008;
(Chertov et al. 1996; Van Wetering et al. 1997). Hosoda et al. 2017). Furthermore, it was shown
These and subsequent studies demonstrated that that CHDPs display a range of pro-inflammatory
CHDPs may regulate inflammation and immu- and anti-­ inflammatory activities toward both
nity through their direct action on inflammatory recruited leukocytes and tissue resident cells
and immune cells but also indirectly via induc- (Hancock et al. 2016; Suarez-Carmona et al. 2015;
tion of mediators in other cell types (Hancock Agier et al. 2015). Following the characterization
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 153

of human neutrophil α-defensins as T-cell che- M1 phenotype, which preferentially expresses

moattractants (Chertov et al. 1996), other studies this CHDP when compared to the anti-­
reported chemotactic activity toward additional inflammatory pro-repair M2 cells (van der Does
cell types including macrophages and mast cells et al. 2010). In addition, LL-37 displays a range
(Suarez-Carmona et al. 2015). These defensins of anti-inflammatory activities; it neutralizes the
were also found to indirectly induce inflammatory pro-inflammatory activity of selected TLR-­
cell accumulation, e.g., by increasing expression ligands such as lipopolysaccharide (LPS) which
of the neutrophil chemoattractant IL-8/CXCL8 in was initially thought to be mediated in part by
epithelial cells (Van Wetering et al. 1997). The rel- direct LPS binding (Scott et al. 2000). The direct
evance of these activities was demonstrated in neutralizing interaction of CHDPs such as LL-37
various reports of in vivo animal models, includ- with pro-inflammatory microbial components,
ing one demonstrating that neutrophil α-defensins most notably LPS, has prompted the search for
may aggravate experimental colitis in a mouse CHDPs or their mimetics as anti-endotoxin com-
model (Hashimoto et al. 2012). Following the pio- pounds. However, subsequent studies have shown
neering work of Yang et al. demonstrating that that LPS neutralization by direct interaction
β-defensins also display direct chemotactic activi- between LL-37 and LPS alone does not fully
ties (Yang et al. 1999), a variety of studies showed explain the inhibitory effect of LL-37 on LPS-­
that β-defensins display similar activities as out- induced cell activation, and that LL-37-mediated
lined above for α-defensins (Hancock et al. 2016; modulation of intracellular pathways such as the
Suarez-Carmona et al. 2015). TLR-to-NFκB pathway likely contributes to the
The activities of the human cathelicidin anti-endotoxin effects of the peptide (Mookherjee
LL-37 in inflammation and innate immunity dis- et al. 2006). At high concentrations (~20 μg/ml),
play a remarkable overlap with those outlined for LL-37 can contribute to the local regulation of
α- and β-defensins, although the anti-­inflammation by inducing the expression of the
inflammatory activities of LL-37 have been bet- anti-inflammatory cytokine IL-10 in monocytes/
ter defined (Hancock et al. 2016; Agier et al. macrophages, dendritic cells, and B and T cells
2015). LL-37 is chemotactic for neutrophils, (Mookherjee et al. 2009). In line with these data,
eosinophils, monocytes, T cells, and mast cells LL-37 has also been shown to increase the pro-
(Yang et al. 2000; Tjabringa et al. 2006), which in duction of IL-1 receptor antagonist (IL-1RA) in
part is mediated through formyl peptide receptors monocytes and neutrophils (Choi et al. 2014;
to modulate the activity of these cells (reviewed Zhang et al. 2008). These studies demonstrate
in Hancock et al. 2016; Agier et al. 2015). that CHDPs may affect a range of cell types that
However, at the site of inflammation, LL-37 may play crucial roles in inflammation and innate
dampen further chemotaxis by internalization of immunity. As CHDPs also influence related sys-
CXCR2 on neutrophils and monocytes (Zhang tems involved in these processes including the
et al. 2009), regulate tissue half-life and clear- complement system (Hiemstra 2015) and
ance of neutrophils by suppressing apoptosis antigen-­presenting cells (APCs), this broadens
(Nagaoka et al. 2006), and induce non-­ the scope of immunological effects of CHDPs
inflammatory secondary necrosis of apoptotic beyond innate immunity.
neutrophils (Li et al. 2009). The ability of LL-37
to regulate neutrophil recruitment through direct 10.1.2.2 Role of CHPDs in Shaping
neutrophil chemotactic activity and induction of Adaptive Immunity
neutrophil attracting chemokines was found to CHDPs play an important role in the link between
contribute to host defense against respiratory innate and adaptive immunity. This function is
tract infection with Pseudomonas aeruginosa primarily mediated by the ability of CHDPs to
(Beaumont et al. 2014). Interestingly, LL-37 was recruit APCs such as monocyte/macrophages and
also found to redirect macrophage polarization dendritic cells (DCs) to the site of infection and/
toward macrophages with a pro-inflammatory or inflammation. As mentioned above, Yang et al.
154 A. M. van der Does et al.

in 1999 demonstrated that human β-defensin inducing both mucosal and systemic antigen-­
hBD2 is directly chemotactic to immature den- specific responses (Kim et al. 2017).
dritic cells (iDCs) and T cells, thus establishing Overall, CHDPs are involved in the interplay
the concept that CHDPs function in the interplay between many cell types of the human body, and
between the innate and adaptive arms of the this is crucial for the ability of CHDPs to influence
immune system (Yang et al. 1999). Subsequent and regulate immunity. Additional discoveries that
studies have established that CHDPs such as these peptides are expressed by intestinal Paneth
defensins and cathelicidins can recruit APCs cells which play a role in maintaining host-micro-
(Suarez-Carmona et al. 2015; Kim et al. 2015). biome homeostasis have raised interest in explor-
The direct chemotactic activity of CHDPs is ing the crosstalk between CHDPs and the
mediated by several immune receptors such as microbiome, as discussed in the next section.
chemokine receptors (e.g., CCR6 and CCR2),
pattern recognition receptors (TLR4 and
TLR1/2), and G-protein-coupled receptors 10.1.3 CHDPs and the Microbiome
(Suarez-Carmona et al. 2015; De et al. 2000;
Kurosaka et al. 2005). Apart from promoting The role of CHDPs in microbiome composition
migration and recruitment of APCs to local sites is a relatively new emerging field of research.
of infection and/or inflammation, CHDPs also Several studies have demonstrated that CHDPs
influence adaptive immune response by activat- are important contributors to the host-­microbiome
ing APCs and modulating the differentiation of interactions. For example, enteric α-defensins
lymphocytes. Human defensin hBD3 induces the expressed by Paneth cells may be essential in
expression of co-stimulatory molecules CD80, shaping the murine gut microbiome (Salzman
CD86, and CD40 on monocytes and myeloid et al. 2010), as transgenic mice expressing a
DCs via interaction with TLRs, thereby enhanc- human defensin, HD5, exhibit reduced Firmicutes
ing adaptive immune response (Funderburg et al. and increased Bacteroidetes gut microbiome lev-
2007). Similarly, hBD2 and hBD3 can promote els compared to wild-type controls, whereas defi-
the production of IFN-α from plasmacytoid DCs ciency of matrilysin which is essential for murine
in a TLR-9-dependent manner and the conse- α-defensin processing results in the opposite
quent initiation of a T-cell immune response effect on murine gut microbiome composition
(Tewary et al. 2013). Likewise, LL-37 promotes (Salzman et al. 2010). Further indications that
the maturation of iDCs, enhances their expres- CHDPs play an important role in shaping the
sion of co-stimulatory molecules and phagocytic microbiome is derived from a study demonstrat-
activity, and influences polarization of T -cells ing that differences in copy number of a specific
toward a Th1-skewed response (Davidson et al. β-defensin gene cluster correlate with alterations
2004). A recent study has also shown that LL-37 in the microbiota of the nasopharynx associated
can activate follicular DCs of Peyer’s patches pri- with otitis media infection-related pathogens
marily by engaging formyl peptide receptors and (Jones et al. 2014). Cullen and co-workers
can enhance the proliferation and activation of B showed that dominant members of the gut micro-
cells (Kim et al. 2017). The ability of CHDPs to biota develop a selective resistance against
enhance adaptive immunity indicates the poten- ­specific CHDPs that are enhanced during inflam-
tial to use these peptides as adjuvants. Indeed, mation, suggesting that co-evolvement of this
CHDPs have been shown to function as adjuvants microbial resistance mechanism with the host
exerting antigen-specific immune responses. For may have allowed stabilization of the gut micro-
example, hBD3 and murine cathelicidin CRAMP biome during inflammation (Cullen et al. 2015).
exhibited adjuvant effects in ovalbumin-­challenge Apart from CHDPs influencing the composition
models (Kurosaka et al. 2005; Tewary et al. of the microbiome, recent studies have also dem-
2013). Similarly, LL-37 was shown to mediate onstrated that the host microbiome can in turn
adjuvant effect in an oral vaccine formulation impact CHDP expression. Direct effects of
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 155

microbial signaling through pattern recognition between CHDPs and the microbiome in main-
receptors have been shown to enhance expression taining immune homeostasis and the control of
of CHDPs such as defensins in the gut (Ostaff pathogens (Fig. 10.1). Therefore, it is not surpris-
et al. 2013). In addition, bacterial metabolites ing that disturbance of the balance between
also contribute to the regulation of expression of microbiome and CHDPs can have severe conse-
CHDPs. For example, several short-chain fatty quences for the host (Sun et al. 2015). The recog-
acids such as butyrate produced by gut bacteria nition of the importance of the microbiome in
upon fermentation of dietary fibers can regulate human health has opened up a new avenue of
the expression of specific CHDPs like cathelici- research to examine the therapeutic possibility of
dins (Schauber et al. 2004; Zhao et al. 2018), and targeting the microbiome with synthetic deriva-
commensal bacteria can promote CHDP expres- tives or mimetics of CHDPs (Stone and Xu 2017;
sion in the skin (Meisel et al. 2018). Furthermore, Kaplan et al. 2011).
host-derived CHDPs such as cathelicidins can
exhibit synergistic functions with the microbi-
ome. For example, antimicrobials produced by 10.1.4 Clinical Relevance
the skin microbiome such as Staphylococcus and Therapeutic Potential
hominis Sh-lantibiotic-α and Sh-­lantibiotic-­β can of CHDPs
synergistically enhance antimicrobial activity
against Staphylococcus aureus with LL-37 As discussed above, CHDPs elicit antimicrobial
(Nakatsuji et al. 2017). Overall, recent reports functions and exhibit a wide range of roles in
clearly demonstrate the complex interplay immunity (summarized in Fig. 10.2). Thus, there

B. CHDPs selectively kill

C. Synergism
A. Microbiome- pathogenic microbiome
between
derived bacteria
commensals and
SCFA promotes (resistance by
CHDPs against
CHDP commensals)
pathogens

Microbiome
Epithelial Cells

Homeostasis

Fig. 10.1 CHDP-microbiome interaction potentially sion could aid in controlling pathogenic bacterial strains.
contributes to homeostasis. Cationic host defence pep- (B) Resistance was demonstrated by commensal bacte-
tides (CHDPs) may contribute to host-microbiome rial strains against CHDPs that are specifically expressed
homeostasis. (A) Short-chain fatty acids (SCFA) pro- during inflammation, which could potentially stabilize
duced by microbiome bacteria upon fermentation of the microbiome during inflammation. (C) CHDPs
dietary fibers were demonstrated to promote HDP exhibit synergistic antimicrobial activities with skin
expression by epithelium, as were commensal bacteria microbiome-­derived antimicrobials against pathogenic
and microbial products. These effects on CHDP expres- bacterial strains
156 A. M. van der Does et al.

Anti-microbial / anti-biofilm activity

Wound healing

Promotion of
Pathogens NETs formation
Neutralization of LPS

CHDP Recruitment and

polarization of T-cells
Induction of
chemokines

Differentiation of
dendritic cells

Altering Induction of
endotoxin-induced anti-inflammatory
signaling cytokines

Endotoxin Suppression of
(LPS) pro-inflammatory Altering the
cytokines cytokine milieu
Recruitment of
monocytes / macrophages

Fig. 10.2 CHDP-mediated activity controls infections ing endotoxin-mediated signaling, suppression of pro-­
and inflammation. CHDPs exhibit antimicrobial activi- inflammatory cytokines, induction of anti-inflammatory
ties, including toward bacterial biofilms, to control a vari- cytokines and chemokines, promoting neutrophil extra-
ety of infections. CHDPs mediate a wide range of cellular traps (NETs), influencing the differentiation of
immunity-related functions which includes but not lim- dendritic cells, and polarization of T -cells. The functions
ited to recruitment of immune cells to site of infections of CHDPs link innate and adaptive immunity to contribute
which contributes to enhance clearance of microbes, alter- in the resolution of infections and immune regulation

is a keen interest in using these peptides and their 2011; Rivas-Santiago et al. 2013; Niyonsaba
synthetic mimics for the development of novel et al. 2013). Some of the advances made in the
therapies for a wide range of diseases, from understanding of the role of CHDPs in disease
infectious disease to autoimmunity. The diversity pathology and in the development of CHDPs and
of natural CHDPs with more than 2600 peptides IDR peptide-based therapeutic strategies are
defined to date (http://aps.unmc.edu/AP/main. summarized as follows.
php) has propelled the design of synthetic pep-
tides based on the natural CHDPs. 10.1.4.1 Infectious Disease
Immunomodulatory short synthetic peptides Since the discovery of CHDPs, research in the
designed from either internal fragments or by field has been focused on the antimicrobial func-
systematic amino acid substitution of CHDPs are tions of CHDPs with impetus for the develop-
known as innate defence regulator (IDR) pep- ment of a new class of antibiotics. Whereas the
tides (Hilpert et al. 2005, 2006). The cost of pro- (modest) direct killing activities of natural
duction of shorter IDR peptides is less compared CHDPs show some possibilities for this develop-
to CHDPs (Scott et al. 2007; Nijnik et al. 2010; ment, the expanding insight into the immuno-
Cherkasov et al. 2009). Moreover, IDR peptides modulatory activities of CHDPs gives rise to
have negligible toxicity and are not immuno- many more therapeutic opportunities. Current
genic, thus making these valuable therapeutic paradigm suggests that CHDPs modulate the
candidates (Scott et al. 2007; Nijnik et al. 2010; host immune response to facilitate clearance of
Achtman et al. 2012; Molhoek et al. 2009; infections under physiological conditions.
Steinstraesser et al. 2012; Turner-Brannen et al. However, it should be noted that the immuno-
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 157

modulatory activities of CHDPs that aid in the et al. 2005). In line with this, adjunct therapy
resolution of infections are extensive and depend with (phenyl)butyrate has shown beneficial
on peptide concentrations, source, and target effects in humans infected with tuberculosis
cell. Several studies have demonstrated the (Mily et al. 2015) and Shigella (Raqib et al. 2012)
potential of CHDPs in controlling infections in and in rabbits infected with E. coli (Al-Mamun
various models; α-defensins were shown to pro- et al. 2013). However, in these studies it remains
mote recruitment of macrophages and lympho- unclear if the pathogens are cleared as a conse-
cytes to the peritoneal cavity upon Klebsiella quence of the microbicidal activity of the pep-
pneumoniae infection in mice (Welling et al. tides or is it due to CHDP-induced immune
1998). LL-37 was demonstrated to augment responses in the host. Furthermore, it should be
phagocytosis by human macrophages (Wan et al. noted that inducers of CHDPs such as phenylbu-
2014) and facilitate airway epithelial barrier tyrate and vitamin D may also control infections
function to prevent Pseudomonas aeruginosa independent of CHDPs by influencing the
invasion (Byfield et al. 2011). Davidson et al. immune system based on in vitro studies demon-
showed that administration of LL-37 promoted strating that these compounds can promote
clearance of P. aeruginosa from infected mice autophagy and intracellular killing of bacteria
by enhancing the early neutrophil response (Rekha et al. 2015).
(Beaumont et al. 2014). Similar protective Exogenous administration of synthetic IDR
effects were obtained by overexpression of peptides has also been shown to control infec-
endogenous cathelicidin in the lungs of mice tions in various models, including antibiotic-­
(Bals et al. 1999). LL-37 was also shown to resistant MRSA (Scott et al. 2007; Nijnik et al.
diminish Bacillus anthracis spore-­induced death 2010; Rivas-Santiago et al. 2013; Hirsch et al.
of mice by promoting the recruitment of neutro- 2008; Mookherjee and Hancock 2007; Hou
phils to the site of infection (Lisanby et al. 2008). et al. 2013). Recent studies have demonstrated
Administration of the neutrophil α-defensin the use of IDR peptides on the control of bio-
HNP1 was demonstrated to protect against M. film infections that are typically recalcitrant to
tuberculosis in a murine mouse model, with antibiotics (Haney et al. 2015). The recent
in vitro mechanistic studies clearly demonstrat- advances in the development of IDR peptides
ing beneficial effects to support the possible use for treatment of infectious diseases are encour-
as anti-infective in tuberculosis (Sharma et al. aging as some of these peptide therapies are
2001). Similarly, beneficial effects of exogenous currently in phase II/III clinical trials
CHDPs administration have also been achieved (Cherkasov et al. 2009; Yeung et al. 2011;
against viral pathogens. For example, adminis- Hancock et al. 2012). A distinct advantage of
tration of LL-37 was demonstrated to be effec- developing CHDP-based a­ntimicrobial thera-
tive against influenza A virus and respiratory peutics is thought to be the potential to avoid
syncytial virus in a murine model (Barlow et al. emergence of microbial resistance. This is
2011; Currie et al. 2016). speculated primarily due to two reasons: (i) the
Another promising strategy being explored for selective pressure on pathogens is reduced as
anti-infective therapy is by promoting the expres- CHDPs may resolve infections by modulating
sion of endogenous CHDP expression using host immunity rather than directly targeting the
inducers such as vitamin D3 (Gombart et al. microbe, and (ii) when applicable the direct
2005), sodium (phenyl)butyrate (Steinmann et al. microbicidal activity of these peptides is
2009; Schauber et al. 2003), and other recently directed at microbial targets that are indispens-
discovered compounds (Miraglia et al. 2016; able for microbial survival. However, recent
Fischer et al. 2016). Inducers of CHDPs are being studies have indicated several bacterial resis-
explored to restore pathogen-induced downregu- tance mechanisms to antimicrobial peptides
lation of CHDPs, which is a virulence strategy of (reviewed in Joo et al. 2016), and so it remains
pathogens such as Shigella flexneri (Islam et al. unclear if CHDP-based therapy can indeed
2001) and Neisseria gonorrhoeae (Bergman avoid the emergence of microbial resistance.
158 A. M. van der Does et al.

10.1.4.2  hronic Inflammatory Lung

C increased levels of hCAP-18/LL-37 and
Disease decreased levels of the antimicrobial proteinase
The lung is exposed daily to large numbers of inhibitor secretory leukocyte proteinase inhibitor
inhaled pathogens, and CHDPs play a central (SLPI), and that higher hCAP-18/LL-37 levels in
role in the protection of the host from such infec- stable COPD disease are associated with an
tions. CHDPs are produced by a range of cells in increased risk of exacerbations of the disease
the lungs, including the airway epithelial cells (Persson et al. 2017). It is thus tempting to specu-
and neutrophils. As discussed above, these pep- late that specific CHDPs may contribute to dys-
tides exert prominent effects on host immune regulated innate and adaptive immune responses
responses, and therefore it is not surprising that in inflammatory lung disease due to their wide
specific CHDPs are implicated in chronic inflam- range of effects on immunity.
matory lung diseases. In this context, it should be Enhancing CHDP levels in the lungs may be
noted that a relative deficiency of CHDPs result- an attractive strategy to counteract deficient
ing from decreased expression, degradation, or endogenous expression and combat the respira-
impairment of function, as well as their excessive tory infections that play a detrimental role in
activity, both have been demonstrated to contrib- patients with CF, COPD, and asthma (Hiemstra
ute to disease pathology in the lung. et al. 2016). However, control of inflammation
Deficient expression or activity of CHDP has and restoration of dysregulated immune responses
been linked to the frequent respiratory infections is also a major aim of treatment. The dual face of
observed in smokers (Herr et al. 2009), patients CHDP in providing direct antimicrobial activity
with chronic obstructive pulmonary disease and regulating inflammation and immunity has to
(COPD) (Pace et al. 2012; Amatngalim et al. be taken into account when considering the use
2017), and those with cystic fibrosis (CF) (Chen of CHDP as therapeutics in the treatment of
et al. 2004; Pezzulo et al. 2012). Previous studies chronic inflammatory lung diseases. Therefore,
have used airway epithelial cell cultures to show vitamin D treatment aimed to increase endoge-
that smoking, the main risk factor for COPD in nous expression of CHDPs such as hCAP-18/
industrialized societies, reduces the expression of LL-37 while at the same time dampening
specific CHDP (Herr et al. 2009; Amatngalim unwanted immune responses is an attractive
et al. 2017). Furthermore, the expression of some strategy. Indeed, results of two recent trials in
of these peptides is also decreased in airway epi- COPD have shown that vitamin D treatment may
thelial cell cultures from COPD patients serve to decrease exacerbations in patients with a
(Amatngalim et al. 2017). In addition, expression vitamin D deficiency (Lehouck et al. 2012;
of specific CHDPs is also decreased in allergic Martineau et al. 2015). Other therapeutic strate-
airway inflammation which may be linked to gies may include direct administration of syn-
decreased pulmonary host defenses (Beisswenger thetic IDR peptides with an optimal balance of
et al. 2006). Furthermore, a recent study demon- activities, to facilitate both antimicrobial activi-
strates that specific CHDPs are decreased in the ties and immunomodulatory functions in the
lungs in response to inhaled air pollutants known lungs. For example, administration of a bovine
to exacerbate asthma (Piyadasa et al. 2018b). In cathelicidin-based IDR peptide may alleviate air-
contrast, it is also evident that neutrophil-derived way inflammation and pathophysiology of
CHDPs such as the neutrophil α-defensins and asthma (Piyadasa et al. 2018a).
hCAP-18/LL-37 may be increased in neutrophil-­
dominated airway inflammation in COPD and 10.1.4.3 Skin Disease
CF and during rhinovirus infection in asthma The skin provides a protective interface between
(Chen et al. 2004; Paone et al. 2011; Merkel et al. internal organs and environmental factors includ-
2005; Rohde et al. 2014). It was demonstrated ing pathogens. Apart from being a physical bar-
that COPD exacerbations are associated with rier, it is also an active immune organ. Thus,
changes in the pattern of CHDP in sputum, with substantial research has focused on trying to
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 159

understand the role of CHDPs in antimicrobial is a matter of debate, as CHDPs also play an
defenses of the skin and in unraveling the role of important pro-inflammatory and immunoregula-
these peptides in chronic skin diseases such as tory role in atopic dermatitis (Kopfnagel et al.
psoriasis and atopic dermatitis. The variety of 2013). Furthermore, studies in various models
constitutively expressed and inducible CHDPs and in atopic dermatitis and psoriasis have dem-
produced by keratinocytes and other cells in the onstrated a role for CHDPs in wound repair
skin provides an important layer of defense (Mangoni et al. 2016). This resulted in a clinical
(Niyonsaba et al. 2017). Whereas most studies trial exploring the topical use of LL-37 to treat
have focused on the role of β-defensins and venous leg ulcers. Topical application of LL-37
hCAP-18/LL-37, other CHDPs such as S100A7/ to these chronic wounds was found to be safe and
psoriasin, the RNase family member RNase 7, well tolerated, and initial results suggested a ben-
and dermcidin are likely also important for local eficial effect on wound healing (Gronberg et al.
defense. Various experimental models have been 2014). The observation that high β-defensin gene
used to uncover the role of CHDPs in host copy numbers are associated with an increased
defense, including knockout models showing risk of developing psoriasis also indicates that
increased susceptibility to necrotizing skin infec- CHDPs such as β-defensins are not merely pro-
tions with group A streptococci of mice deficient tective against skin infections in inflammatory
in CRAMP, the mouse homologue of the gene skin diseases (Hollox et al. 2008). Aligned with
encoding the human cathelicidin hCAP-18/ this, it has been shown that genetic knockout of
LL-37 (Nizet et al. 2001). Also, treatment with a β-defensins in a bioengineered skin-humanized
synthetic CHDP-inspired peptide (SAAP-148) mouse model for psoriasis has beneficial effects
was found to eradicate acute and established bio- (Bracke et al. 2014). Other CHDPs, including
film infections with antibiotic-resistant hCAP18/LL-37, have been shown to contribute
Staphylococcus aureus and Acinetobacter bau- to skin inflammation in psoriasis, and in contrast
mannii in wounded human skin ex vivo and protective roles for CHDPs in skin barrier func-
mouse skin in vivo (de Breij et al. 2018). tion and wound healing and in reducing inflam-
Moreover, patient studies in atopic dermatitis and mation have also been reported (Niyonsaba et al.
psoriasis, two frequent chronic inflammatory 2017). Collectively these studies suggest a com-
skin diseases, indicate that the role of CHDP in plex role of CHDPs in skin disease, with both
diseases may be more complex. protective antimicrobial, immunomodulatory,
Initial studies showed an association of defi- and wound repair enhancing functions and
cient expression of selected CHDPs in lesional disease-­promoting inflammatory activities with
skin of patients with atopic dermatitis who suffer dysregulation of innate and adaptive immune
from frequent bacterial skin infections (Ong et al. response. Therefore, a better insight into the role
2002). The ability of Th2 cytokines, key in aller- of CHDPs is essential for the development of
gic inflammation in diseases such as atopic der- novel CHDP-based therapies for skin diseases.
matitis, to suppress expression of CHDPs offered
an attractive explanation for the frequently 10.1.4.4 Autoimmune Disease
observed skin infections with Staphylococcus Previous studies provide contradictory evidences
aureus in patients with atopic dermatitis. demonstrating CHDPs as both effectors and reg-
Furthermore, these studies suggested that the ulators of autoimmune diseases such as in sys-
relative deficiency of CHDPs in atopic dermatitis temic lupus erythematosus (SLE) and rheumatoid
and high expression in psoriasis might help to arthritis (RA). Serum levels of CHDPs, e.g.,
explain the differential occurrence of skin infec- defensins (hBD-1, hBD-2, HNP1–3) and psoria-
tions in these patients (Ong et al. 2002). However, sin, are increased, with the circulating levels of
whether decreased levels of CHDPs are a com- these defensins associated with the disease activ-
mon characteristic and main factor in determin- ity in SLE (Kahlenberg and Kaplan 2013; Kreuter
ing susceptibility of the skin in atopic dermatitis et al. 2011; Sthoeger et al. 2009; Vordenbaumen
160 A. M. van der Does et al.

et al. 2010). Release of nuclear autoantigens examining the role of CHDPs in other autoim-
within neutrophil extracellular traps (NETs) is a mune conditions, which also provide contradic-
characteristic feature of SLE (Radic 2014). As tory evidence. Increased circulating levels of
LL-37 can complex with nuclear autoantigens α-defensins have been demonstrated in type I
within NETs, it has been suggested that LL-37 diabetes, suggesting that these peptides may con-
may facilitate subsequent pro-inflammatory tribute to the pathogenesis of this disease
responses (Lande et al. 2011; Kahlenberg et al. (Saraheimo et al. 2008; Joseph et al. 2008;
2013; Favilli et al. 2009), as well as prevent the Nemeth et al. 2014). In contrast, levels of
degradation of NETs, and facilitate the formation β-defensins hBD-1 and hBD-2 were shown to be
of NETs (Neumann et al. 2014), thus collectively decreased in salivary glands of patients with
promoting autoantibody formation and genera- Sjogren’s syndrome (Kaneda et al. 2009), and
tion of immune complexes in SLE (Lande et al. CHDPs can limit inflammation and prevent infec-
2011). Similarly, levels of other CHDPs such as tions in this disease (Luciano et al. 2015).
defensins and S100 peptides are increased in the Similarly, LL-37 was shown to be beneficial in
synovial fluid of patients with RA (Baillet et al. the control of keratinopathy and polymicrobial
2010; Bokarewa et al. 2003). In vitro studies infected corneal wounds, common in Sjogren’s
using cells such as synoviocytes, osteoclasts, and syndrome (Yin and Yu 2010; Duplantier and van
granulocytes isolated from RA patients also show Hoek 2013). Similar to what was discussed above
an increased expression of specific CHDPs such for other diseases, there are contradictory evi-
as hBD-2, HNP-1, and LL-37 (Hoffmann et al. dences in various studies examining the role of
2013; Varoga et al. 2006, 2009; Ahn et al. 2013; CHDPs in autoimmunity. It is plausible that the
Kienhofer et al. 2014). This is corroborated by function of these peptides may be disease spe-
animal model studies demonstrating a correlation cific and/or depend on the disease stage.
with increased cathelicidin expression and dis- Nevertheless, it is important to explore the role of
ease pathology in arthritis (Hoffmann et al. CHDPs in autoimmunity, which may contribute
2013). These studies indicate that specific CHDPs to the development of new therapies for the spec-
may contribute to the disease pathology and pro- trum of diseases in this category.
gression in autoimmune diseases such as SLE
and RA. In contrast, recent studies suggest that 10.1.4.5 Cancer
CHDPs may function in the control of inflamma- The involvement of CHDPs in cancer develop-
tion and tissue damage in autoimmune disease; ment and progression is increasingly being
Miles et al. showed anti-inflammatory mecha- examined (Deslouches and Di 2017; Roudi et al.
nisms mediated by the release of α-defensins 2017). Reports demonstrate that CHDPs such as
HNP1–3 from neutrophils (Miles et al. 2009), defensins and cathelicidins exert both antican-
thus suggesting that the release of specific cer and pro-tumorigenic activity. For example,
CHDPs from apoptotic or necrotic neutrophils in LL-37 was demonstrated to be overexpressed in
SLE may contribute to the control of inflamma- melanoma (Kim et al. 2010), breast cancer
tion. Similarly, LL-37 and its synthetic derivative (Heilborn et al. 2005), prostate cancer (Hensel
peptide can inhibit pro-inflammatory responses et al. 2011), and some lung cancers, but not in
and signaling mechanisms related to RA (Choi small cell lung cancers and carcinoid tumors
et al. 2014; Xu et al. 2013). Moreover, LL-37 was (von Haussen et al. 2008). It was also shown in
shown to prevent osteoclastogenesis (Supanchart in vitro studies that LL-37 can promote prolif-
et al. 2012) thus indicating a role of LL-37 in pre- eration and/or migration of malignant cell lines
venting bone loss in RA. Aligned with this, a such as melanoma (Jia et al. 2017), keratino-
LL-37-derived synthetic peptide was shown to cytes (Weber et al. 2009), kidney cells (Weber
prevent disease progression, inflammation, and et al. 2009), prostate cancer cells (Hensel et al.
articular tissue damage in a murine model of RA 2011), and several lung cancer cells (von
(Chow et al. 2014). There are limited studies Haussen et al. 2008), via different signaling
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 161

pathways such as MAPK, EGFR, Y-box binding et al. 2009). Interestingly, it has been suggested
protein 1, and NF-κB. Indirect effects have been that CHDPs may also influence cancer develop-
demonstrated in the case of ovarian cancer, ment indirectly by interacting with bacteria or
where LL-37 mediated the recruitment of mes- viruses. Bacteria associated with the develop-
enchymal stem cells to promote tumor develop- ment of oral cancers actively dysregulate tran-
ment (Coffelt et al. 2009). Conversely, anticancer script levels of α- and β-defensins in oral tumor
effects have also been suggested for CHDPs in cells, which in its turn has been suggested to
colorectal and gastric cancer; loss of LL-37 promote the proliferation of these cells (Hoppe
expression was associated with tumor growth, et al. 2016). In contrast, a possible protective
thus suggesting a protective function of LL-37 in role for α-defensins in cancer was suggested by
these cancers (Cheng et al. 2015; Ren et al. the antiviral activities of HNP1–3 and HD-5
2012). Wu et al. further demonstrated that LL-37 against human papillomavirus (Buck et al.
can inhibit the proliferation of gastric cell lines 2006). CHDPs are also being examined for a
in vitro and gastric cancer xenografts in vivo possible role as biomarkers due to their dysreg-
(Wu et al. 2010), and pro-apoptotic effects of ulated expression in cancers, often as a result of
this peptide were shown in Jurkat T-leukemia a significant loss of expression in tumors (Hong
cells (Mader et al. 2009). Similarly, several et al. 2017), except for α-defensins which are
studies demonstrated loss of β-defensins at often highly expressed by tumors, e.g., in
tumor sites including in salivary gland tumor colorectal cancer (Melle et al. 2005), renal cell
(Kesting et al. 2012), renal and prostatic cancer carcinomas (Muller et al. 2002), and bladder
(Donald et al. 2003), and oral squamous cell cancer (Holterman et al. 2006). Overall, it
carcinoma (Joly et al. 2009), suggesting a pro- remains unclear whether CHDPs are protective
tective role for these defensins in cancer. This or contribute to the pathology of cancer. It is
was further highlighted by a study examining likely that this is dependent on the type of can-
defensin expression in 29 different tumor- cer, the specific CHDP, and concentration of the
derived cell lines, which showed a low or no specific peptide. Nevertheless, the therapeutic
transcript expression of β-defensins, while potential of HDPs for cancer is being examined
α-defensin expression was maintained in these (Roudi et al. 2017). An hBD-2-based gene ther-
cell lines (Winter et al. 2016). Aligned with this, apy was suggested based on a study demonstrat-
several studies have demonstrated direct protec- ing that hBD-2 can promote recruitment of DCs
tive effects for β-defensins. For example, hBD-1 to the tumor site resulting in enhanced antitu-
was shown to inhibit cell migration and invasion mor immunity and a reduction in tumor size
in oral squamous cell carcinoma (Han et al. (Malinovschi et al. 2014). A combination of
2014); the sequence of hBD-3 was found to con- LL-37 and CpG-­ oligodeoxynucleotides was
tain an oncolytic binding motif promoting cytol- shown to be s­uccessful in a murine model of
ysis of tumor cells (Phan et al. 2016); hBD-3 ovarian cancer (Chuang et al. 2009). Vragniau
was found to be cytotoxic to human alveolar and co-authors have showed high expression of
basal epithelial cell line A549; the murine ortho- HD5 in ovary, endometrium, and lung cancer
logue of hBD-3, Defb14, was cytotoxic to Lewis and suggested a role of HD5 in preventing onco-
lung carcinoma cell line-derived tumors in mice lytic adenovirus replication and spread
(Hanaoka et al. 2016); and hBD-3 was demon- (Vragniau et al. 2017). Even though the diverse
strated to suppress the migration of head and activities of CHDPs provide potential for their
neck cancer cells (Wang et al. 2012) and several use in cancer therapy, contradictory effects of
colon cancer cell lines (Uraki et al. 2015). these peptides complicate the development of
However, similar to LL-37, contradictory results CHDP-based therapies for cancers. It is possible
have also been shown for defensins, for exam- that future studies using synthetic IDR peptides
ple, high expression of hBD-3 was associated designed with defined activity could be explored
with oral squamous cell carcinoma (Kesting to target specific tumors.
162 A. M. van der Does et al.

10.1.5 Challenges 10.2 Summary

in the Development of CHDP-­
Based Therapy The wide repertoire of functions of CHDPs from
its role in the control of infections to diverse func-
Despite the impressive and wide repertoire of tions in immunity underscores the importance of
activities, most natural CHDP may not be ide- these peptides in health and disease. It is now well
ally suited as drugs that can be used for direct appreciated that CHDPs exhibit both innate and
application. Direct antimicrobial activity adaptive immune functions, and influence regula-
requires high local concentrations, and many tion of inflammation by direct effects on immune
CHDPs display pro-inflammatory and cytotoxic cells and via induction of immune mediators by
activities at such concentrations. This needs to other cell types. Impact of CHDPs on the immune
be taken into account when developing thera- system has significantly broadened the scope of
pies using CHDPs to target infections requiring research in this field beyond antimicrobial activ-
local concentrations high enough to display ity. However, the duality of immunomodulatory
direct antimicrobial activity. The route of functions exhibited by these peptides, by both
administration is also an important consider- pro- and anti-inflammatory functions contributing
ation, and so far most studies using CHDP- to immune activation and regulation, and the
based therapies have employed topical interplay of CHDPs and microbiome in mucosal
application, e.g., on skin, or local administration immunity and immune homeostasis, highlights
in the respiratory tract. Another challenge is the the complexity of interpreting the exact role of
local activity of CHDPs, since especially at sites these peptides in the context of immune response.
of intense inflammation and infection the local It is thus also not surprising that the dysregulation
microenvironment may not allow full activity of of expression of these peptides is associated with
the peptides. For example, local mucosal pH disease pathology from infectious disease to auto-
(Pezzulo et al. 2012), degradation by microbial immunity and various cancers. Diversity of natu-
and host proteases (Mallia et al. 2012), and ral CHDPs and its repertoire of functions have
impairment of CHDP-mediated antimicrobial propelled interest in the use of these peptides and
activity by salt, F-actin, DNA, mucus, and their synthetic mimics in the development of new
microbial saccharides (Bucki et al. 2007) are therapies for various diseases. Despite challenges
among the factors that can alter/impair CHDP associated with the development of CHDP-based
activity. Thus stability and bioavailability are therapy, the applicability of CHDP-based
among the limiting factors in the use of CHDPs ­immunomodulatory therapies offers an entirely
as therapeutics. However, this can potentially be new therapeutic approach for the resolution of
mitigated by exploring delivery systems with infections by enhancing host immune response
formulations using liposomes, polymeric and on the other hand the control of inflammatory
nanoparticles, carbon nanotubes, and other diseases without compromising the patients’ abil-
materials. Another challenge in the develop- ity to resolve infections. In this context, current
ment of CHDP therapy is the cost of production. research indicates that synthetic variants of
Based on these challenges, alternative strategies CHDPs may be promising for future clinical use.
are being explored to develop CHDP-based
treatments. These include (i) stimulation of Acknowledgment AD is supported by an EU Marie
endogenous CHDP production by inducers such Curie Global Fellowship (#748569). Studies in the labo-
ratory of PSH on CHDPs are supported by grants from
as vitamin D or butyrate, (ii) improving the the Lung Foundation Netherlands, the Eurostars pro-
local conditions that may otherwise impair the gram, The Netherlands Organisation for Health
activity of CHDPs, and (iii) the development of Research and Development (ZonMw), and Galapagos
novel synthetic peptides inspired by the struc- NV. NM is supported by Canadian Institutes of Health
Research (CIHR) and Natural Sciences and Engineering
ture and/or sequence of natural CHDPs, such as Research Council of Canada (NSERC) for peptide
IDR peptides. research.
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 163

References host defense against influenza infection elicited by the

human cathelicidin LL-37. PLoS One 6(10):e25333
Beaumont PE, McHugh B, Gwyer Findlay E, Mackellar
Achtman AH, Pilat S, Law CW, Lynn DJ, Janot L,
A, Mackenzie KJ, Gallo RL et al (2014) Cathelicidin
Mayer ML et al (2012) Effective adjunctive therapy
host defence peptide augments clearance of pulmonary
by an innate defense regulatory Peptide in a pre-
Pseudomonas aeruginosa infection by its influence on
clinical model of severe malaria. Sci Transl Med
neutrophil function in vivo. PLoS One 9(6):e99029
4(135):135ra64
Beisswenger C, Kandler K, Hess C, Garn H, Felgentreff
Agerberth B, Gunne H, Odeberg J, Kogner P, Boman HG,
K, Wegmann M et al (2006) Allergic airway inflam-
Gudmundsson GH (1995) FALL-39, a putative human
mation inhibits pulmonary antibacterial host defense.
peptide antibiotic, is cysteine-free and expressed in
J Immunol 177(3):1833–1837
bone marrow and testis. Proc Natl Acad Sci U S A
Bergman P, Johansson L, Asp V, Plant L, Gudmundsson
92(1):195–199
GH, Jonsson AB et al (2005) Neisseria gonorrhoeae
Agerberth B, Grunewald J, Castanos-Velez E, Olsson B,
downregulates expression of the human antimicrobial
Jornvall H, Wigzell H et al (1999) Antibacterial com-
peptide LL-37. Cell Microbiol 7(7):1009–1017
ponents in bronchoalveolar lavage fluid from healthy
Bokarewa MI, Jin T, Tarkowski A (2003) Intraarticular
individuals and sarcoidosis patients. Am J Respir Crit
release and accumulation of defensins and bacteri-
Care Med 160(1):283–290
cidal/permeability-increasing protein in patients with
Agerberth B, Charo J, Werr J, Olsson B, Idali F, Lindbom
rheumatoid arthritis. J Rheumatol 30(8):1719–1724
L et al (2000) The human antimicrobial and chemotac-
Bowdish DM, Davidson DJ, Scott MG, Hancock RE
tic peptides LL-37 and alpha-defensins are expressed
(2005) Immunomodulatory activities of small host
by specific lymphocyte and monocyte populations.
defense peptides. Antimicrob Agents Chemother
Blood 96(9):3086–3093
49(5):1727–1732
Agier J, Efenberger M, Brzezinska-Blaszczyk E (2015)
Bracke S, Carretero M, Guerrero-Aspizua S, Desmet
Cathelicidin impact on inflammatory cells. Cent Eur
E, Illera N, Navarro M et al (2014) Targeted silenc-
J Immunol 40(2):225–235
ing of DEFB4 in a bioengineered skin-humanized
Ahn JK, Huang B, Bae EK, Park EJ, Hwang JW, Lee
mouse model for psoriasis: development of siRNA
J et al (2013) The role of alpha-defensin-1 and
SECosome-based novel therapies. Exp Dermatol
related signal transduction mechanisms in the pro-
23(3):199–201
duction of IL-6, IL-8 and MMPs in rheumatoid
Buck CB, Day PM, Thompson CD, Lubkowski J, Lu W,
fibroblast-like synoviocytes. Rheumatology (Oxford)
Lowy DR et al (2006) Human alpha-defensins block
52(8):1368–1376
papillomavirus infection. Proc Natl Acad Sci U S A
Al-Mamun A, Mily A, Sarker P, Tiash S, Navarro A, Akter
103(5):1516–1521
M et al (2013) Treatment with phenylbutyrate in a
Bucki R, Sostarecz AG, Byfield FJ, Savage PB, Janmey
pre-clinical trial reduces diarrhea due to enteropatho-
PA (2007) Resistance of the antibacterial agent cer-
genic Escherichia coli: link to cathelicidin induction.
agenin CSA-13 to inactivation by DNA or F-actin
Microbes Infect 15(13):939–950
and its activity in cystic fibrosis sputum. J Antimicrob
Amatngalim GD, Schrumpf JA, Henic A, Dronkers E,
Chemother 60(3):535–545
Verhoosel RM, Ordonez SR et al (2017) Antibacterial
Bucki R, Leszczynska K, Namiot A, Sokolowski W
defense of human airway epithelial cells from chronic
(2010) Cathelicidin LL-37: a multitask antimicrobial
obstructive pulmonary disease patients induced by
peptide. Arch Immunol Ther Exp 58(1):15–25
acute exposure to nontypeable haemophilus influen-
Byfield FJ, Kowalski M, Cruz K, Leszczynska K,
zae: modulation by cigarette smoke. J Innate Immun
Namiot A, Savage PB et al (2011) Cathelicidin LL-37
9(4):359
increases lung epithelial cell stiffness, decreases
Baillet A, Trocme C, Berthier S, Arlotto M, Grange L,
transepithelial permeability, and prevents epithelial
Chenau J et al (2010) Synovial fluid proteomic fin-
invasion by Pseudomonas aeruginosa. J Immunol
gerprint: S100A8, S100A9 and S100A12 proteins
187(12):6402–6409
discriminate rheumatoid arthritis from other inflam-
Chen CI, Schaller-Bals S, Paul KP, Wahn U, Bals R
matory joint diseases. Rheumatology (Oxford)
(2004) Beta-defensins and LL-37 in bronchoalveo-
49(4):671–682
lar lavage fluid of patients with cystic fibrosis. J Cyst
Bals R, Wang X, Zasloff M, Wilson JM (1998) The pep-
Fibros 3(1):45–50
tide antibiotic LL-37/hCAP-18 is expressed in epithe-
Cheng M, Ho S, Yoo JH, Tran DH, Bakirtzi K, Su B et al
lia of the human lung where it has broad antimicrobial
(2015) Cathelicidin suppresses colon cancer develop-
activity at the airway surface. Proc Natl Acad Sci U S
ment by inhibition of cancer associated fibroblasts.
A 95(16):9541–9546
Clin Exp Gastroenterol 8:13–29
Bals R, Weiner DJ, Moscioni AD, Meegalla RL, Wilson
Cherkasov A, Hilpert K, Jenssen H, Fjell CD, Waldbrook
JM (1999) Augmentation of innate host defense by
M, Mullaly SC et al (2009) Use of artificial intel-
expression of a cathelicidin antimicrobial peptide.
ligence in the design of small peptide antibiotics
Infect Immun 67(11):6084–6089
effective against a broad spectrum of highly antibiotic-­
Barlow PG, Svoboda P, Mackellar A, Nash AA, York IA,
resistant superbugs. ACS Chem Biol 4(1):65–74
Pohl J et al (2011) Antiviral activity and increased
164 A. M. van der Does et al.

Chertov O, Michiel DF, Xu L, Wang JM, Tani K, Murphy De Y, Chen Q, Schmidt AP, Anderson GM, Wang JM,
WJ et al (1996) Identification of defensin-1, defensin- Wooters J et al (2000) LL-37, the neutrophil gran-
2 and CAP37/azurocidin as T-cell chemoattractant
­ ule- and epithelial cell-derived cathelicidin, utilizes
proteins released from interleukin-8-stimulated neu- formyl peptide receptor-like 1 (FPRL1) as a receptor
trophils. J Biol Chem 271:2935–2940 to chemoattract human peripheral blood neutrophils,
Choi KY, Mookherjee N (2012) Multiple immune-­ monocytes, and T cells. J Exp Med 192(7):1069–1074
modulatory functions of cathelicidin host defense Deslouches B, Di YP (2017) Antimicrobial peptides with
peptides. Front Immunol 3:149 selective antitumor mechanisms: prospect for antican-
Choi KY, Chow LN, Mookherjee N (2012) Cationic cer applications. Oncotarget 8(28):46635–46651
host defence peptides: multifaceted role in immune Donald CD, Sun CQ, Lim SD, Macoska J, Cohen C, Amin
modulation and inflammation. J Innate Immun MB et al (2003) Cancer-specific loss of beta-­defensin
4(4):361–370 1 in renal and prostatic carcinomas. Lab Investig
Choi KY, Napper S, Mookherjee N (2014) Human cat- 83(4):501–505
helicidin LL-37 and its derivative IG-19 regulate Doss M, White MR, Tecle T, Hartshorn KL (2010) Human
interleukin-32-induced inflammation. Immunology defensins and LL-37 in mucosal immunity. J Leukoc
143(1):68–80 Biol 87(1):79–92
Chow LN, Choi KY, Piyadasa H, Bossert M, Uzonna J, Duplantier AJ, van Hoek ML (2013) The human cathe-
Klonisch T et al (2014) Human cathelicidin LL-37-­ licidin antimicrobial peptide LL-37 as a potential
derived peptide IG-19 confers protection in a murine treatment for polymicrobial infected wounds. Front
model of collagen-induced arthritis. Mol Immunol Immunol 4:143
57(2):86–92 Epand RM, Vogel HJ (1999) Diversity of antimicrobial
Chromek M, Slamova Z, Bergman P, Kovacs L, Podracka peptides and their mechanisms of action. Biochim
L, Ehren I et al (2006) The antimicrobial peptide cat- Biophys Acta 1462(1–2):11–28
helicidin protects the urinary tract against invasive Favilli F, Anzilotti C, Martinelli L, Quattroni P, De
bacterial infection. Nat Med 12(6):636–641 Martino S, Pratesi F et al (2009) IL-18 activity in
Chuang CM, Monie A, Wu A, Mao CP, Hung CF (2009) systemic lupus erythematosus. Ann N Y Acad Sci
Treatment with LL-37 peptide enhances antitumor 1173:301–309
effects induced by CpG oligodeoxynucleotides against Fernandez de Caleya R, Gonzalez-Pascual B, Garcia-­
ovarian cancer. Hum Gene Ther 20(4):303–313 Olmedo F, Carbonero P (1972) Susceptibility of phy-
Coffelt SB, Marini FC, Watson K, Zwezdaryk KJ, topathogenic bacteria to wheat purothionins in vitro.
Dembinski JL, LaMarca HL et al (2009) The pro-­ Appl Microbiol 23(5):998–1000
inflammatory peptide LL-37 promotes ovarian tumor Fischer N, Sechet E, Friedman R, Amiot A, Sobhani I,
progression through recruitment of multipotent mes- Nigro G et al (2016) Histone deacetylase inhibition
enchymal stromal cells. Proc Natl Acad Sci U S A enhances antimicrobial peptide but not inflammatory
106(10):3806–3811 cytokine expression upon bacterial challenge. Proc
Cowland JB, Johnsen AH, Borregaard N (1995) hCAP-18, Natl Acad Sci U S A 113(21):E2993–E3001
a cathelin/pro-bactenecin-like protein of human neu- Frohm M, Agerberth B, Ahangari G, Stahle-Backdahl
trophil specific granules. FEBS Lett 368(1):173–176 M, Liden S, Wigzell H et al (1997) The expression of
Cullen TW, Schofield WB, Barry NA, Putnam EE, the gene coding for the antibacterial peptide LL-37 is
Rundell EA, Trent MS et al (2015) Gut microbiota. induced in human keratinocytes during inflammatory
Antimicrobial peptide resistance mediates resilience disorders. J Biol Chem 272(24):15258–15263
of prominent gut commensals during inflammation. Funderburg N, Lederman MM, Feng Z, Drage MG,
Science 347(6218):170–175 Jadlowsky J, Harding CV et al (2007) Human -defen-
Currie SM, Gwyer Findlay E, McFarlane AJ, Fitch PM, sin-­3 activates professional antigen-presenting cells
Bottcher B, Colegrave N et al (2016) Cathelicidins via toll-like receptors 1 and 2. Proc Natl Acad Sci U S
have direct antiviral activity against respiratory syn- A 104(47):18631–18635
cytial virus in vitro and protective function in vivo in Ganz T (2003a) Defensins: antimicrobial peptides of
mice and humans. J Immunol 196(6):2699–2710 innate immunity. Nat Rev Immunol 3(9):710–720
Davidson DJ, Currie AJ, Reid GS, Bowdish DM, Ganz T (2003b) The role of antimicrobial peptides in
MacDonald KL, Ma RC et al (2004) The cationic innate immunity. Integr Comp Biol 43(2):300–304
antimicrobial peptide LL-37 modulates dendritic cell Ganz T, Selsted ME, Szklarek D, Harwig SS, Daher
differentiation and dendritic cell-induced T cell polar- K, Bainton DF et al (1985) Defensins. Natural pep-
ization. J Immunol 172(2):1146–1156 tide antibiotics of human neutrophils. J Clin Invest
de Breij A, Riool M, Cordfunke RA, Malanovic N, de 76(4):1427–1435
Boer L, Koning RI et al (2018) The antimicrobial pep- Ghosh D, Porter E, Shen B, Lee SK, Wilk D, Drazba J et al
tide SAAP-148 combats drug-resistant bacteria and (2002) Paneth cell trypsin is the processing enzyme
biofilms. Sci Transl Med 10(423):eaan4044 for human defensin-5. Nat Immunol 3(6):583–590
De Smet K, Contreras R (2005) Human antimicro- Gombart AF, Borregaard N, Koeffler HP (2005) Human
bial peptides: defensins, cathelicidins and histatins. cathelicidin antimicrobial peptide (CAMP) gene is a
Biotechnol Lett 27(18):1337–1347 direct target of the vitamin D receptor and is strongly up-
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 165

regulated in myeloid cells by 1,25-­dihydroxyvitamin aggravates dextran sulfate sodium-induced colitis.

D3. FASEB J 19(9):1067–1077 Inflamm Bowel Dis 18(4):667–675
Gordon YJ, Huang LC, Romanowski EG, Yates KA, Heilborn JD, Nilsson MF, Jimenez CI, Sandstedt B,
Proske RJ, McDermott AM (2005) Human cathelici- Borregaard N, Tham E et al (2005) Antimicrobial pro-
din (LL-37), a multifunctional peptide, is expressed tein hCAP18/LL-37 is highly expressed in breast can-
by ocular surface epithelia and has potent antibacterial cer and is a putative growth factor for epithelial cells.
and antiviral activity. Curr Eye Res 30(5):385–394 Int J Cancer 114(5):713–719
Gronberg A, Mahlapuu M, Stahle M, Whately-Smith C, Hemshekhar M, Anaparti V, Mookherjee N (2016)
Rollman O (2014) Treatment with LL-37 is safe and Functions of cationic host defense peptides in immu-
effective in enhancing healing of hard-to-heal venous nity. Pharmaceuticals (Basel) 9(3):40
leg ulcers: a randomized, placebo-controlled clinical Hensel JA, Chanda D, Kumar S, Sawant A, Grizzle WE,
trial. Wound Repair Regen 22(5):613–621 Siegal GP et al (2011) LL-37 as a therapeutic target
Gudmundsson GH, Agerberth B, Odeberg J, Bergman T, for late stage prostate cancer. Prostate 71(6):659–670
Olsson B, Salcedo R (1996) The human gene FALL39 Herr C, Beisswenger C, Hess C, Kandler K, Suttorp N,
and processing of the cathelin precursor to the antibac- Welte T et al (2009) Suppression of pulmonary innate
terial peptide LL-37 in granulocytes. Eur J Biochem host defence in smokers. Thorax 64(2):144–149
238(2):325–332 Hiemstra PS (2015) Parallel activities and interactions
Guo C, Rosoha E, Lowry MB, Borregaard N, Gombart between antimicrobial peptides and complement in
AF (2013) Curcumin induces human cathelicidin anti- host defense at the airway epithelial surface. Mol
microbial peptide gene expression through a vitamin Immunol 68(1):28–30
D receptor-independent pathway. J Nutr Biochem Hiemstra PS, Amatngalim GD, van der Does AM,
24(5):754–759 Taube C (2016) Antimicrobial peptides and innate
Han Q, Wang R, Sun C, Jin X, Liu D, Zhao X et al (2014) lung defenses: role in infectious and noninfectious
Human beta-defensin-1 suppresses tumor migration lung diseases and therapeutic applications. Chest
and invasion and is an independent predictor for sur- 149(2):545–551
vival of oral squamous cell carcinoma patients. PLoS Hilpert K, Volkmer-Engert R, Walter T, Hancock RE
One 9(3):e91867 (2005) High-throughput generation of small antibac-
Hanaoka Y, Yamaguchi Y, Yamamoto H, Ishii M, Nagase terial peptides with improved activity. Nat Biotechnol
T, Kurihara H et al (2016) In vitro and in vivo antican- 23(8):1008–1012
cer activity of human beta-defensin-3 and its mouse Hilpert K, Elliott MR, Volkmer-Engert R, Henklein P,
homolog. Anticancer Res 36(11):5999–6004 Donini O, Zhou Q et al (2006) Sequence requirements
Hancock RE, Lehrer R (1998) Cationic peptides: a new and an optimization strategy for short antimicrobial
source of antibiotics. Trends Biotechnol 16(2):82–88 peptides. Chem Biol 13(10):1101–1107
Hancock RE, Nijnik A, Philpott DJ (2012) Modulating Hirsch T, Jacobsen F, Steinau HU, Steinstraesser L (2008)
immunity as a therapy for bacterial infections. Nat Rev Host defense peptides and the new line of defence
Microbiol 10(4):243–254 against multiresistant infections. Protein Pept Lett
Hancock RE, Haney EF, Gill EE (2016) The immunology 15(3):238–243
of host defence peptides: beyond antimicrobial activ- Hoffmann MH, Bruns H, Backdahl L, Neregard P,
ity. Nat Rev Immunol 16(5):321–334 Niederreiter B, Herrmann M et al (2013) The cat-
Haney EF, Mansour SC, Hilchie AL, de la Fuente-Nunez helicidins LL-37 and rCRAMP are associated with
C, Hancock RE (2015) High throughput screen- pathogenic events of arthritis in humans and rats. Ann
ing methods for assessing antibiofilm and immuno- Rheum Dis 72(7):1239–1248
modulatory activities of synthetic peptides. Peptides Hollox EJ, Huffmeier U, Zeeuwen PL, Palla R, Lascorz J,
71:276–285 Rodijk-Olthuis D et al (2008) Psoriasis is associated
Harder J, Bartels J, Christophers E, Schroder JM (2001) with increased beta-defensin genomic copy number.
Isolation and characterization of human beta -defen- Nat Genet 40(1):23–25
sin-­3, a novel human inducible peptide antibiotic. Holterman DA, Diaz JI, Blackmore PF, Davis
J Biol Chem 276(8):5707–5713 JW, Schellhammer PF, Corica A et al (2006)
Harder J, Meyer-Hoffert U, Wehkamp K, Schwichtenberg Overexpression of alpha-defensin is associated
L, Schroder JM (2004) Differential gene induction of with bladder cancer invasiveness. Urol Oncol
human beta-defensins (hBD-1, -2, -3, and -4) in kerati- 24(2):97–108
nocytes is inhibited by retinoic acid. J Invest Dermatol Hong SA, Kim KH, Lee TJ, Park ES, Kim MK, Myung
123(3):522–529 SC (2017) A role of human beta defensin-1 in pre-
Hase K, Eckmann L, Leopard JD, Varki N, Kagnoff MF dicting prostatic adenocarcinoma in cases of false-­
(2002) Cell differentiation is a key determinant of negative biopsy. APMIS 125(12):1063–1069
cathelicidin LL-37/human cationic antimicrobial pro- Hoppe T, Kraus D, Novak N, Probstmeier R, Frentzen M,
tein 18 expression by human colon epithelium. Infect Wenghoefer M et al (2016) Oral pathogens change
Immun 70(2):953–963 proliferation properties of oral tumor cells by affect-
Hashimoto S, Uto H, Kanmura S, Sakiyama T, Oku M, ing gene expression of human defensins. Tumour Biol
Iwashita Y et al (2012) Human neutrophil peptide-1 37(10):13789–13798
166 A. M. van der Does et al.

Hosoda H, Nakamura K, Hu Z, Tamura H, Reich J, Kaplan CW, Sim JH, Shah KR, Kolesnikova-Kaplan A,
Kuwahara-Arai K et al (2017) Antimicrobial catheli- Shi W, Eckert R (2011) Selective membrane disrup-
cidin peptide LL37 induces NET formation and sup- tion: mode of action of C16G2, a specifically targeted
presses the inflammatory response in a mouse septic antimicrobial peptide. Antimicrob Agents Chemother
model. Mol Med Rep 16(4):5618–5626 55(7):3446–3452
Hou M, Zhang N, Yang J, Meng X, Yang R, Li J et al Kesting MR, Loeffelbein DJ, Hasler RJ, Wolff KD, Rittig
(2013) Antimicrobial peptide LL-37 and IDR-1 A, Schulte M et al (2009) Expression profile of human
ameliorate MRSA pneumonia in vivo. Cell Physiol beta-defensin 3 in oral squamous cell carcinoma.
Biochem 32(3):614–623 Cancer Investig 27(5):575–581
Hultmark D, Engstrom A, Bennich H, Kapur R, Boman Kesting MR, Stoeckelhuber M, Kuppek A, Hasler R,
HG (1982) Insect immunity: isolation and struc- Rohleder N, Wolff KD et al (2012) Human beta-­
ture of cecropin D and four minor antibacterial defensins and psoriasin/S100A7 expression in salivary
components from Cecropia pupae. Eur J Biochem glands: anti-oncogenic molecules for potential thera-
127(1):207–217 peutic approaches. BioDrugs 26(1):33–42
Islam D, Bandholtz L, Nilsson J, Wigzell H, Christensson Kienhofer D, Hahn J, Schubert I, Reinwald C, Ipseiz
B, Agerberth B et al (2001) Downregulation of bacte- N, Lang SC et al (2014) No evidence of pathogenic
ricidal peptides in enteric infections: a novel immune involvement of cathelicidins in patient cohorts and
escape mechanism with bacterial DNA as a potential mouse models of lupus and arthritis. PLoS One
regulator. Nat Med 7(2):180–185 9(12):e115474
Jia J, Zheng Y, Wang W, Shao Y, Li Z, Wang Q et al (2017) Kim JE, Kim HJ, Choi JM, Lee KH, Kim TY, Cho BK
Antimicrobial peptide LL-37 promotes YB-1 expres- et al (2010) The antimicrobial peptide human cat-
sion, and the viability, migration and invasion of malig- ionic antimicrobial protein-18/cathelicidin LL-37 as
nant melanoma cells. Mol Med Rep 15(1):240–248 a putative growth factor for malignant melanoma. Br
Joly S, Compton LM, Pujol C, Kurago ZB, Guthmiller J Dermatol 163(5):959–967
JM (2009) Loss of human beta-defensin 1, 2, and 3 Kim SH, Yang IY, Kim J, Lee KY, Jang YS (2015)
expression in oral squamous cell carcinoma. Oral Antimicrobial peptide LL-37 promotes antigen-­
Microbiol Immunol 24(5):353–360 specific immune responses in mice by enhancing
Jones DE, Bevins CL (1992) Paneth cells of the human Th17-skewed mucosal and systemic immunities. Eur
small intestine express an antimicrobial peptide gene. J Immunol 45(5):1402–1413
J Biol Chem 267(32):23216–23225 Kim SH, Kim YN, Jang YS (2017) Cutting edge: LL-37-­
Jones DE, Bevins CL (1993) Defensin-6 mRNA in human mediated formyl peptide Receptor-2 signaling in
Paneth cells: implications for antimicrobial pep- follicular dendritic cells contributes to B cell activa-
tides in host defense of the human bowel. FEBS Lett tion in Peyer’s patch germinal centers. J Immunol
315(2):187–192 198(2):629–633
Jones EA, Kananurak A, Bevins CL, Hollox EJ, Bakaletz King AE, Fleming DC, Critchley HO, Kelly RW (2003)
LO (2014) Copy number variation of the beta defensin Differential expression of the natural antimicrobi-
gene cluster on chromosome 8p influences the bacte- als, beta-defensins 3 and 4, in human endometrium.
rial microbiota within the nasopharynx of otitis-prone J Reprod Immunol 59(1):1–16
children. PLoS One 9(5):e98269 Kopfnagel V, Harder J, Werfel T (2013) Expression of
Joo HS, Fu CI, Otto M (2016) Bacterial strategies of resis- antimicrobial peptides in atopic dermatitis and pos-
tance to antimicrobial peptides. Philos Trans R Soc sible immunoregulatory functions. Curr Opin Allergy
Lond B Biol Sci 371(1695):20150292 Clin Immunol 13(5):531–536
Joseph G, Tarnow L, Astrup AS, Hansen TK, Parving Kreuter A, Jaouhar M, Skrygan M, Tigges C, Stucker
HH, Flyvbjerg A et al (2008) Plasma alpha-defensin is M, Altmeyer P et al (2011) Expression of antimi-
associated with cardiovascular morbidity and mortal- crobial peptides in different subtypes of cutane-
ity in type 1 diabetic patients. J Clin Endocrinol Metab ous lupus erythematosus. J Am Acad Dermatol
93(4):1470–1475 65(1):125–133
Kahlenberg JM, Kaplan MJ (2013) Little peptide, big Kurosaka K, Chen Q, Yarovinsky F, Oppenheim JJ, Yang
effects: the role of LL-37 in inflammation and autoim- D (2005) Mouse cathelin-related antimicrobial pep-
mune disease. J Immunol 191(10):4895–4901 tide chemoattracts leukocytes using formyl peptide
Kahlenberg JM, Carmona-Rivera C, Smith CK, Kaplan receptor-­like 1/mouse formyl peptide receptor-like
MJ (2013) Neutrophil extracellular trap-­ associated 2 as the receptor and acts as an immune adjuvant.
protein activation of the NLRP3 inflammasome J Immunol 174(10):6257–6265
is enhanced in lupus macrophages. J Immunol Kuwano K, Tanaka N, Shimizu T, Kida Y (2006)
190(3):1217–1226 Antimicrobial activity of inducible human beta
Kaneda Y, Yamaai T, Mizukawa N, Nagatsuka H, defensin-2 against mycoplasma pneumoniae. Curr
Yamachika E, Gunduz M et al (2009) Localization of Microbiol 52(6):435–438
antimicrobial peptides human beta-defensins in minor Lande R, Ganguly D, Facchinetti V, Frasca L, Conrad C,
salivary glands with Sjogren’s syndrome. Eur J Oral Gregorio J et al (2011) Neutrophils activate plasma-
Sci 117(5):506–510 cytoid dendritic cells by releasing self-DNA-peptide
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 167

complexes in systemic lupus erythematosus. Sci Merkel D, Rist W, Seither P, Weith A, Lenter MC (2005)
Transl Med 3(73):73ra19 Proteomic study of human bronchoalveolar lavage flu-
Larrick JW, Hirata M, Zhong J, Wright SC (1995) ids from smokers with chronic obstructive pulmonary
Anti-microbial activity of human CAP18 peptides. disease by combining surface-enhanced laser desorp-
Immunotechnology 1(1):65–72 tion/ionization-mass spectrometry profiling with
Laube DM, Yim S, Ryan LK, Kisich KO, Diamond G mass spectrometric protein identification. Proteomics
(2006) Antimicrobial peptides in the airway. Curr Top 5(11):2972–2980
Microbiol Immunol 306:153–182 Miles K, Clarke DJ, Lu W, Sibinska Z, Beaumont PE,
Lehouck A, Mathieu C, Carremans C, Baeke F, Verhaegen Davidson DJ et al (2009) Dying and necrotic neutro-
J, Van Eldere J et al (2012) High doses of vitamin D phils are anti-inflammatory secondary to the release of
to reduce exacerbations in chronic obstructive pulmo- alpha-defensins. J Immunol 183(3):2122–2132
nary disease. Ann Intern Med 156(2):105–114 Mily A, Rekha RS, Kamal SM, Arifuzzaman AS, Rahim
Li HN, Barlow PG, Bylund J, Mackellar A, Bjorstad A, Z, Khan L et al (2015) Significant effects of oral phen-
Conlon J et al (2009) Secondary necrosis of apoptotic ylbutyrate and vitamin D3 adjunctive therapy in pul-
neutrophils induced by the human cathelicidin LL-37 monary tuberculosis: a randomized controlled trial.
is not proinflammatory to phagocytosing macro- PLoS One 10(9):e0138340
phages. J Leukoc Biol 86(4):891–902 Miraglia E, Nylen F, Johansson K, Arner E, Cebula M,
Lisanby MW, Swiecki MK, Dizon BL, Pflughoeft KJ, Farmand S et al (2016) Entinostat up-regulates the
Koehler TM, Kearney JF (2008) Cathelicidin admin- CAMP gene encoding LL-37 via activation of STAT3
istration protects mice from Bacillus anthracis spore and HIF-1alpha transcription factors. Sci Rep 6:33274
challenge. J Immunol 181(7):4989–5000 Molhoek EM, den Hertog AL, de Vries AM, Nazmi K,
Luciano N, Valentini V, Calabro A, Elefante E, Vitale Veerman EC, Hartgers FC et al (2009) Structure-­
A, Baldini C et al (2015) One year in review function relationship of the human antimicrobial pep-
2015: Sjogren’s syndrome. Clin Exp Rheumatol tide LL-37 and LL-37 fragments in the modulation of
33(2):259–271 TLR responses. Biol Chem 390(4):295–303
Mader JS, Mookherjee N, Hancock RE, Bleackley RC Mookherjee N, Hancock RE (2007) Cationic host defence
(2009) The human host defense peptide LL-37 induces peptides: innate immune regulatory peptides as a
apoptosis in a calpain- and apoptosis-inducing factor-­ novel approach for treating infections. Cell Mol Life
dependent manner involving Bax activity. Mol Cancer Sci 64(7–8):922–933
Res 7(5):689–702 Mookherjee N, Brown KL, Bowdish DM, Doria S,
Malinovschi A, Masoero M, Bellocchia M, Ciuffreda A, Falsafi R, Hokamp K et al (2006) Modulation of the
Solidoro P, Mattei A et al (2014) Severe vitamin D TLR-mediated inflammatory response by the endog-
deficiency is associated with frequent exacerbations enous human host defense peptide LL-37. J Immunol
and hospitalization in COPD patients. Respir Res 176(4):2455–2464
15:131 Mookherjee N, Hamill P, Gardy J, Blimkie D, Falsafi R,
Mallia P, Footitt J, Sotero R, Jepson A, Contoli M, Chikatamarla A et al (2009) Systems biology evalu-
Trujillo-Torralbo MB et al (2012) Rhinovirus infec- ation of immune responses induced by human host
tion induces degradation of antimicrobial peptides defence peptide LL-37 in mononuclear cells. Mol
and secondary bacterial infection in chronic obstruc- BioSyst 5(5):483–496
tive pulmonary disease. Am J Respir Crit Care Med Muller CA, Markovic-Lipkovski J, Klatt T, Gamper
186(11):1117–1124 J, Schwarz G, Beck H et al (2002) Human alpha-­
Mangoni ML, McDermott AM, Zasloff M (2016) defensins HNPs-1, -2, and -3 in renal cell carcinoma:
Antimicrobial peptides and wound healing: biologi- influences on tumor cell proliferation. Am J Pathol
cal and therapeutic considerations. Exp Dermatol 160(4):1311–1324
25(3):167–173 Murakami M, Ohtake T, Dorschner RA, Schittek B, Garbe
Martineau AR, James WY, Hooper RL, Barnes NC, C, Gallo RL (2002) Cathelicidin anti-microbial pep-
Jolliffe DA, Greiller CL et al (2015) Vitamin D3 tide expression in sweat, an innate defense system for
supplementation in patients with chronic obstructive the skin. J Invest Dermatol 119(5):1090–1095
pulmonary disease (ViDiCO): a multicentre, double-­ Murakami M, Lopez-Garcia B, Braff M, Dorschner RA,
blind, randomised controlled trial. Lancet Respir Med Gallo RL (2004) Postsecretory processing generates
3(2):120–130 multiple cathelicidins for enhanced topical antimicro-
Meisel JS, Sfyroera G, Bartow-McKenney C, Gimblet bial defense. J Immunol 172(5):3070–3077
C, Bugayev J, Horwinski J et al (2018) Commensal Nagaoka I, Tamura H, Hirata M (2006) An antimicro-
microbiota modulate gene expression in the skin. bial cathelicidin peptide, human CAP18/LL-37, sup-
Microbiome 6(1):20 presses neutrophil apoptosis via the activation of
Melle C, Ernst G, Schimmel B, Bleul A, Thieme H, formyl-peptide receptor-like 1 and P2X7. J Immunol
Kaufmann R et al (2005) Discovery and identification 176(5):3044–3052
of alpha-defensins as low abundant, tumor-derived Nakamura T, Furunaka H, Miyata T, Tokunaga F, Muta T,
serum markers in colorectal cancer. Gastroenterology Iwanaga S et al (1988) Tachyplesin, a class of antimi-
129(1):66–73 crobial peptide from the hemocytes of the horseshoe
168 A. M. van der Does et al.

crab (Tachypleus tridentatus). Isolation and chemical in symptomatic smokers and COPD patients. Dis
structure. J Biol Chem 263(32):16709–16713 Markers 31(2):91–100
Nakatsuji T, Chen TH, Narala S, Chun KA, Two AM, Yun Park K, Elias PM, Oda Y, Mackenzie D, Mauro T, Holleran
T et al (2017) Antimicrobials from human skin com- WM et al (2011) Regulation of cathelicidin antimicro-
mensal bacteria protect against Staphylococcus aureus bial peptide expression by an endoplasmic reticulum
and are deficient in atopic dermatitis. Sci Transl Med (ER) stress signaling, vitamin D receptor-independent
9(378):eaah4680 pathway. J Biol Chem 286(39):34121–34130
Nell MJ, Tjabringa GS, Vonk MJ, Hiemstra PS, Grote Persson LJ, Aanerud M, Hardie JA, Miodini Nilsen R,
JJ (2004) Bacterial products increase expression of Bakke PS, Eagan TM et al (2017) Antimicrobial pep-
the human cathelicidin hCAP-18/LL-37 in cultured tide levels are linked to airway inflammation, bacterial
human sinus epithelial cells. FEMS Immunol Med colonisation and exacerbations in chronic obstructive
Microbiol 42(2):225–231 pulmonary disease. Eur Respir J 49(3):1601328
Nemeth BC, Varkonyi T, Somogyvari F, Lengyel C, Pezzulo AA, Tang XX, Hoegger MJ, Abou Alaiwa MH,
Fehertemplomi K, Nyiraty S et al (2014) Relevance of Ramachandran S, Moninger TO et al (2012) Reduced
alpha-defensins (HNP1-3) and defensin beta-1 in dia- airway surface pH impairs bacterial killing in the por-
betes. World J Gastroenterol 20(27):9128–9137 cine cystic fibrosis lung. Nature 487(7405):109–113
Neumann A, Berends ET, Nerlich A, Molhoek EM, Gallo Phan TK, Lay FT, Poon IK, Hinds MG, Kvansakul
RL, Meerloo T et al (2014) The antimicrobial peptide M, Hulett MD (2016) Human beta-defensin 3 con-
LL-37 facilitates the formation of neutrophil extracel- tains an oncolytic motif that binds PI(4,5)P2 to
lular traps. Biochem J 464(1):3–11 mediate tumour cell permeabilisation. Oncotarget
Nijnik A, Madera L, Ma S, Waldbrook M, Elliott MR, 7(2):2054–2069
Easton DM et al (2010) Synthetic cationic peptide Piyadasa H, Hemshekhar M, Altieri A, Basu S, van der Does
IDR-1002 provides protection against bacterial infec- AM, Halayko AJ et al (2018a) Immunomodulatory
tions through chemokine induction and enhanced leu- innate defence regulator (IDR) peptide alleviates air-
kocyte recruitment. J Immunol 184(5):2539–2550 way inflammation and hyper-responsiveness. Thorax.
Niyonsaba F, Madera L, Afacan N, Okumura K, Ogawa https://doi.org/10.1136/thoraxjnl-2017-210739. [Epub
H, Hancock RE (2013) The innate defense regula- ahead of print]
tor peptides IDR-HH2, IDR-1002, and IDR-1018 Piyadasa H, Hemshekhar M, Carlsten C, Mookherjee N
modulate human neutrophil functions. J Leukoc Biol (2018b) Inhaled diesel exhaust decreases the anti-
94(1):159–170 microbial peptides alpha-Defensin and S100A7 in
Niyonsaba F, Kiatsurayanon C, Chieosilapatham P, human bronchial secretions. Am J Respir Crit Care
Ogawa H (2017) Friends or foes? Host defense (anti- Med 197(10):1358–1361
microbial) peptides and proteins in human skin dis- Powers JP, Hancock RE (2003) The relationship between
eases. Exp Dermatol 26(11):989–998 peptide structure and antibacterial activity. Peptides
Nizet V, Ohtake T, Lauth X, Trowbridge J, Rudisill J, 24(11):1681–1691
Dorschner RA et al (2001) Innate antimicrobial pep- Quayle AJ, Porter EM, Nussbaum AA, Wang YM, Brabec
tide protects the skin from invasive bacterial infection. C, Yip KP et al (1998) Gene expression, immuno-
Nature 414(6862):454–457 localization, and secretion of human defensin-5 in
Obata-Onai A, Hashimoto S, Onai N, Kurachi M, Nagai human female reproductive tract. Am J Pathol
S, Shizuno K et al (2002) Comprehensive gene expres- 152(5):1247–1258
sion analysis of human NK cells and CD8(+) T lym- Radic M (2014) Clearance of apoptotic bodies, NETs, and
phocytes. Int Immunol 14(10):1085–1098 biofilm DNA: implications for autoimmunity. Front
Okrent DG, Lichtenstein AK, Ganz T (1990) Direct cyto- Immunol 5:365
toxicity of polymorphonuclear leukocyte granule Raqib R, Sarker P, Mily A, Alam NH, Arifuzzaman AS,
proteins to human lung-derived cells and endothelial Rekha RS et al (2012) Efficacy of sodium butyrate
cells. Am Rev Respir Dis 141(1):179–185 adjunct therapy in shigellosis: a randomized, double-­
Ong PY, Ohtake T, Brandt C, Strickland I, Boguniewicz blind, placebo-controlled clinical trial. BMC Infect
M, Ganz T et al (2002) Endogenous antimicrobial Dis 12:111
peptides and skin infections in atopic dermatitis. N Rehaume LM, Hancock RE (2008) Neutrophil-derived
Engl J Med 347(15):1151–1160 defensins as modulators of innate immune function.
Ostaff MJ, Stange EF, Wehkamp J (2013) Antimicrobial Crit Rev Immunol 28(3):185–200
peptides and gut microbiota in homeostasis and Rekha RS, Rao Muvva SS, Wan M, Raqib R, Bergman
pathology. EMBO Mol Med 5(10):1465–1483 P, Brighenti S et al (2015) Phenylbutyrate induces
Pace E, Ferraro M, Minervini MI, Vitulo P, Pipitone L, LL-37-dependent autophagy and intracellular kill-
Chiappara G et al (2012) Beta defensin-2 is reduced ing of Mycobacterium tuberculosis in human macro-
in central but not in distal airways of smoker COPD phages. Autophagy 11(9):1688–1699
patients. PLoS One 7(3):e33601 Ren SX, Cheng AS, To KF, Tong JH, Li MS, Shen J et al
Paone G, Conti V, Vestri A, Leone A, Puglisi G, Benassi F (2012) Host immune defense peptide LL-37 activates
et al (2011) Analysis of sputum markers in the evalu- caspase-independent apoptosis and suppresses colon
ation of lung inflammation and functional impairment cancer. Cancer Res 72(24):6512–6523
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 169

Rivas-Santiago B, Castaneda-Delgado JE, Rivas Santiago Sieprawska-Lupa M, Mydel P, Krawczyk K, Wojcik K,

CE, Waldbrook M, Gonzalez-Curiel I, Leon-Contreras Puklo M, Lupa B et al (2004) Degradation of human
JC et al (2013) Ability of innate defence regulator pep- antimicrobial peptide LL-37 by Staphylococcus
tides IDR-1002, IDR-HH2 and IDR-1018 to protect aureus-derived proteinases. Antimicrob Agents
against Mycobacterium tuberculosis infections in ani- Chemother 48(12):4673–4679
mal models. PLoS One 8(3):e59119 Sierra JM, Fuste E, Rabanal F, Vinuesa T, Vinas M (2017)
Rodriguez-Jimenez FJ, Krause A, Schulz S, Forssmann An overview of antimicrobial peptides and the latest
WG, Conejo-Garcia JR, Schreeb R et al (2003) advances in their development. Expert Opin Biol Ther
Distribution of new human beta-defensin genes clus- 17(6):663–676
tered on chromosome 20 in functionally different seg- Simmaco M, Mignogna G, Barra D, Bossa F (1993)
ments of epididymis. Genomics 81(2):175–183 Novel antimicrobial peptides from skin secretion
Rohde G, Message SD, Haas JJ, Kebadze T, Parker H, of the European frog Rana esculenta. FEBS Lett
Laza-Stanca V et al (2014) CXC chemokines and 324(2):159–161
antimicrobial peptides in rhinovirus-induced experi- Simmaco M, Mignogna G, Canofeni S, Miele R, Mangoni
mental asthma exacerbations. Clin Exp Allergy ML, Barra D (1996) Temporins, antimicrobial pep-
44(7):930–939 tides from the European red frog Rana temporaria.
Roudi R, Syn NL, Roudbary M (2017) Antimicrobial Eur J Biochem 242(3):788–792
peptides as biologic and immunotherapeutic agents Simmaco M, Kreil G, Barra D (2009) Bombinins, anti-
against cancer: a comprehensive overview. Front microbial peptides from Bombina species. Biochim
Immunol 8:1320 Biophys Acta 1788(8):1551–1555
Salzman NH, Hung K, Haribhai D, Chu H, Karlsson-­ Sorensen OE, Follin P, Johnsen AH, Calafat J, Tjabringa
Sjoberg J, Amir E et al (2010) Enteric defensins are GS, Hiemstra PS et al (2001) Human cathelicidin,
essential regulators of intestinal microbial ecology. hCAP-18, is processed to the antimicrobial peptide
Nat Immunol 11(1):76–83 LL-37 by extracellular cleavage with proteinase 3.
Saraheimo M, Forsblom C, Pettersson-Fernholm K, Blood 97(12):3951–3959
Flyvbjerg A, Groop PH, Frystyk J (2008) Increased Sorensen OE, Cowland JB, Theilgaard-Monch K, Liu
levels of alpha-defensin (−1, −2 and −3) in type 1 L, Ganz T, Borregaard N (2003) Wound healing and
diabetic patients with nephropathy. Nephrol Dial expression of antimicrobial peptides/polypeptides
Transplant 23(3):914–918 in human keratinocytes, a consequence of common
Schauber J, Svanholm C, Termen S, Iffland K, Menzel T, growth factors. J Immunol 170(11):5583–5589
Scheppach W et al (2003) Expression of the catheli- Steiner H, Hultmark D, Engstrom A, Bennich H, Boman
cidin LL-37 is modulated by short chain fatty acids HG (1981) Sequence and specificity of two antibac-
in colonocytes: relevance of signalling pathways. Gut terial proteins involved in insect immunity. Nature
52(5):735–741 292(5820):246–248
Schauber J, Iffland K, Frisch S, Kudlich T, Schmausser Steinmann J, Halldorsson S, Agerberth B, Gudmundsson
B, Eck M et al (2004) Histone-deacetylase inhibitors GH (2009) Phenylbutyrate induces antimicrobial
induce the cathelicidin LL-37 in gastrointestinal cells. peptide expression. Antimicrob Agents Chemother
Mol Immunol 41(9):847–854 53(12):5127–5133
Scott MG, Vreugdenhil AC, Buurman WA, Hancock Steinstraesser L, Hirsch T, Schulte M, Kueckelhaus M,
RE, Gold MR (2000) Cutting edge: cationic antimi- Jacobsen F, Mersch EA et al (2012) Innate defense
crobial peptides block the binding of lipopolysac- regulator peptide 1018 in wound healing and wound
charide (LPS) to LPS binding protein. J Immunol infection. PLoS One 7(8):e39373
164(2):549–553 Sthoeger ZM, Bezalel S, Chapnik N, Asher I, Froy
Scott MG, Davidson DJ, Gold MR, Bowdish D, Hancock O (2009) High alpha-defensin levels in patients
RE (2002) The human antimicrobial peptide LL-37 with systemic lupus erythematosus. Immunology
is a multifunctional modulator of innate immune 127(1):116–122
responses. J Immunol 169(7):3883–3891 Stone VN, Xu P (2017) Targeted antimicrobial ther-
Scott MG, Dullaghan E, Mookherjee N, Glavas N, apy in the microbiome era. Mol Oral Microbiol
Waldbrook M, Thompson A et al (2007) An anti-­ 32(6):446–454
infective peptide that selectively modulates the innate Suarez-Carmona M, Hubert P, Delvenne P, Herfs M
immune response. Nat Biotechnol 25(4):465–472 (2015) Defensins: “simple” antimicrobial peptides or
Seil MN, Nagant C, Dehaye J-P, Vandenbranden M, broad-spectrum molecules? Cytokine Growth Factor
Lensink MF (2010) Spotlight on human LL-37, an Rev 26(3):361–370
immunomodulatory peptide with promising cell-­ Sun J, Furio L, Mecheri R, van der Does AM, Lundeberg
penetrating properties. Pharmaceuticals (Basel) E, Saveanu L et al (2015) Pancreatic beta-cells limit
3(11):3435–3460 autoimmune diabetes via an immunoregulatory anti-
Sharma S, Verma I, Khuller GK (2001) Therapeutic microbial peptide expressed under the influence of the
potential of human neutrophil peptide 1 against exper- gut microbiota. Immunity 43(2):304–317
imental tuberculosis. Antimicrob Agents Chemother Supanchart C, Thawanaphong S, Makeudom A, Bolscher
45(2):639–640 JG, Nazmi K, Kornak U et al (2012) The antimicrobial
170 A. M. van der Does et al.

peptide, LL-37, inhibits in vitro osteoclastogenesis. peptides in systemic lupus erythematosus. Lupus
J Dent Res 91(11):1071–1077 19(14):1648–1653
Tewary P, de la Rosa G, Sharma N, Rodriguez LG, Tarasov Vragniau C, Hubner JM, Beidler P, Gil S, Saydaminova
SG, Howard OM et al (2013) Beta-Defensin 2 and 3 K, Lu ZZ et al (2017) Studies on the interaction of
promote the uptake of self or CpG DNA, enhance tumor-derived HD5 alpha defensins with adenoviruses
IFN-alpha production by human plasmacytoid den- and implications for oncolytic adenovirus therapy.
dritic cells, and promote inflammation. J Immunol J Virol 91(6):e02030–e02016
191(2):865–874 Wan M, van der Does AM, Tang X, Lindbom L, Agerberth
Tjabringa GS, Ninaber DK, Drijfhout JW, Rabe KF, B, Haeggstrom JZ (2014) Antimicrobial peptide
Hiemstra PS (2006) Human cathelicidin LL-37 is a LL-37 promotes bacterial phagocytosis by human
chemoattractant for eosinophils and neutrophils that macrophages. J Leukoc Biol 95(6):971–981
acts via formyl-peptide receptors. Int Arch Allergy Wang K, Wang JH, Baskaran H, Wang R, Jurevic R
Immunol 140(2):103–112 (2012) Effect of human beta-defensin-3 on head and
Tomasinsig L, Zanetti M (2005) The cathelicidins – struc- neck cancer cell migration using micro-fabricated cell
ture, function and evolution. Curr Protein Pept Sci islands. Head Neck Oncol 4:41
6(1):23–34 Wang G, Mishra B, Epand RF, Epand RM (2014) High-­
Tongaonkar P, Golji AE, Tran P, Ouellette AJ, Selsted ME quality 3D structures shine light on antibacterial,
(2012) High fidelity processing and activation of the anti-biofilm and antiviral activities of human catheli-
human alpha-defensin HNP1 precursor by neutrophil cidin LL-37 and its fragments. Biochim Biophys Acta
elastase and proteinase 3. PLoS One 7(3):e32469 1838(9):2160–2172
Turner-Brannen E, Choi KY, Lippert DN, Cortens JP, Wang G, Li X, Wang Z (2016) APD3: the antimicrobial
Hancock RE, El-Gabalawy H et al (2011) Modulation peptide database as a tool for research and education.
of interleukin-1beta-induced inflammatory responses Nucleic Acids Res 44(D1):D1087–D1093
by a synthetic cationic innate defence regulator pep- Weber G, Chamorro CI, Granath F, Liljegren A, Zreika
tide, IDR-1002, in synovial fibroblasts. Arthritis Res S, Saidak Z et al (2009) Human antimicrobial protein
Ther 13(4):R129 hCAP18/LL-37 promotes a metastatic phenotype in
Uraki S, Sugimoto K, Shiraki K, Tameda M, Inagaki breast cancer. Breast Cancer Res: BCR 11(1):R6
Y, Ogura S et al (2015) Corrigendum: human beta-­ Welling MM, Hiemstra PS, van den Barselaar MT,
defensin-­3 inhibits migration of colon cancer cells Paulusma-Annema A, Nibbering PH, Pauwels EK
via downregulation of metastasis-associated 1 family, et al (1998) Antibacterial activity of human neutrophil
member 2 expression. Int J Oncol 46(4):1858 defensins in experimental infections in mice is accom-
Valore EV, Ganz T (1992) Posttranslational processing panied by increased leukocyte accumulation. J Clin
of defensins in immature human myeloid cells. Blood Invest 102(8):1583–1590
79(6):1538–1544 Wilson CL, Ouellette AJ, Satchell DP, Ayabe T, Lopez-­
van der Does AM, Beekhuizen H, Ravensbergen B, Vos Boado YS, Stratman JL et al (1999) Regulation of
T, Ottenhoff THM, van Dissel JT et al (2010) LL-37 intestinal alpha-defensin activation by the metallo-
directs macrophage differentiation toward macro- proteinase matrilysin in innate host defense. Science
phages with a Proinflammatory signature. J Immunol 286(5437):113–117
185(3):1442–1449 Winter J, Kraus D, Reckenbeil J, Probstmeier R (2016)
Van Wetering S, Mannesse-Lazeroms SPG, Van Oncogenic relevant defensins: expression pattern and
Sterkenburg MAJA, Daha MR, Dijkman JH, Hiemstra proliferation characteristics of human tumor cell lines.
PS (1997) Effect of defensins on IL-8 synthesis in Tumour Biol 37(6):7959–7966
airway epithelial cells. Am J Physiol (Lung Cell Mol Woo JS, Jeong JY, Hwang YJ, Chae SW, Hwang SJ,
Physiol) 272(16):L888–LL96 Lee HM (2003) Expression of cathelicidin in human
Varoga D, Paulsen FP, Kohrs S, Grohmann S, Lippross salivary glands. Arch Otolaryngol Head Neck Surg
S, Mentlein R et al (2006) Expression and regulation 129(2):211–214
of human beta-defensin-2 in osteoarthritic cartilage. Wu WK, Sung JJ, To KF, Yu L, Li HT, Li ZJ et al (2010)
J Pathol 209(2):166–173 The host defense peptide LL-37 activates the tumor-­
Varoga D, Klostermeier E, Paulsen F, Wruck C, Lippross suppressing bone morphogenetic protein signaling via
S, Brandenburg LO et al (2009) The antimicrobial inhibition of proteasome in gastric cancer cells. J Cell
peptide HBD-2 and the toll-like receptors-2 and -4 Physiol 223(1):178–186
are induced in synovial membranes in case of septic Xu WD, Zhang M, Feng CC, Yang XK, Pan HF, Ye DQ
arthritis. Virchows Arch 454(6):685–694 (2013) IL-32 with potential insights into rheumatoid
von Haussen J, Koczulla R, Shaykhiev R, Herr C, arthritis. Clin Immunol 147(2):89–94
Pinkenburg O, Reimer D et al (2008) The host defence Yamaguchi Y, Nagase T, Makita R, Fukuhara S, Tomita
peptide LL-37/hCAP-18 is a growth factor for lung T, Tominaga T et al (2002) Identification of multiple
cancer cells. Lung Cancer 59(1):12–23 novel epididymis-specific beta-defensin isoforms in
Vordenbaumen S, Fischer-Betz R, Timm D, Sander humans and mice. J Immunol 169(5):2516–2523
O, Chehab G, Richter J et al (2010) Elevated lev- Yamasaki K, Schauber J, Coda A, Lin H, Dorschner RA,
els of human beta-defensin 2 and human neutrophil Schechter NM et al (2006) Kallikrein-mediated prote-
10 Antimicrobial Host Defence Peptides: Immunomodulatory Functions and Translational Prospects 171

olysis regulates the antimicrobial effects of cathelici- Yin J, Yu FS (2010) LL-37 via EGFR transactivation to
dins in skin. FASEB J 20(12):2068–2080 promote high glucose-attenuated epithelial wound
Yang D, Chertov O, Bykovskaia SN, Chen Q, Buffo MJ, healing in organ-cultured corneas. Invest Ophthalmol
Shogan J et al (1999) Beta-defensins: linking innate Vis Sci 51(4):1891–1897
and adaptive immunity through dendritic and T cell Zasloff M (1987) Magainins, a class of antimicrobial pep-
CCR6. Science 286(5439):525–528 tides from Xenopus skin: isolation, characterization of
Yang D, Chen Q, Schmidt AP, Anderson GM, Wang two active forms, and partial cDNA sequence of a pre-
JM, Wooters J et al (2000) LL-37, the neutrophil cursor. Proc Natl Acad Sci U S A 84(15):5449–5453
granule- and epithelial cell-derived cathelicidin, Zhang Z, Cherryholmes G, Shively JE (2008) Neutrophil
utlizes formyl peptide receptor-like 1 (FPRL1) as secondary necrosis is induced by LL-37 derived from
a receptor to chemoattract human peripheral blood cathelicidin. J Leukoc Biol 84(3):780–788
neutrophils, monocytes, and T cells. J Exp Med Zhang Z, Cherryholmes G, Chang F, Rose DM,
192(7):1069–1074 Schraufstatter I, Shively JE (2009) Evidence that cat-
Yeung AT, Gellatly SL, Hancock RE (2011) helicidin peptide LL-37 may act as a functional ligand
Multifunctional cationic host defence peptides for CXCR2 on human neutrophils. Eur J Immunol
and their clinical applications. Cell Mol Life Sci 39(11):3181–3194
68(13):2161–2176 Zhao Y, Chen F, Wu W, Sun M, Bilotta AJ, Yao S et al
Yim S, Dhawan P, Ragunath C, Christakos S, Diamond (2018) GPR43 mediates microbiota metabolite SCFA
G (2007) Induction of cathelicidin in normal and CF regulation of antimicrobial peptide expression in
bronchial epithelial cells by 1,25-dihydroxyvitamin intestinal epithelial cells via activation of mTOR and
D(3). J Cyst Fibros 6(6):403–410 STAT3. Mucosal Immunol 11(3):752–762
 Part IV
Towards Clinical Applications
 Selectivity of Antimicrobial
Peptides: A Complex Interplay 11
of Multiple Equilibria

Sara Bobone and Lorenzo Stella

Abstract Keywords
Antimicrobial peptides (AMPs) attack bacte- Antimicrobial peptides · Host defense
rial membranes selectively, killing microbes at peptides · Selectivity · Toxicity · Peptide-­
concentrations that cause no toxicity to the membrane association · Aggregation ·
host cells. This selectivity is not due to inter- Hydrophobicity · Amphipathicity
action with specific receptors but is deter-
mined by the different lipid compositions of
the membranes of the two cell types and by
the peculiar physicochemical properties of 11.1 Introduction
AMPs, particularly their cationic and
amphipathic character. However, the available The scientific and medical interest for antimicro-
data, including recent studies of peptide-cell bial peptides (AMPs), short peptides produced by
association, indicate that this picture is exces- most organisms as part of their innate immune
sively simplistic, because selectivity is modu- defenses, derives from their wide-spectrum bac-
lated by a complex interplay of several tericidal properties and their possible application
interconnected phenomena. For instance, con- to fight drug-resistant bacteria. However, in view
formational transitions and self-assembly of clinical applications, the absence of significant
equilibria modulate the effective peptide toxicity is almost as important as a good activity.
hydrophobicity, the electrostatic and hydro- In this respect, one of the appealing properties of
phobic contributions to the membrane-­binding many AMPs is their cell selectivity, i.e., the abil-
driving force are nonadditive, and kinetic pro- ity to kill bacterial cells at concentrations signifi-
cesses can play an important role in selective cantly lower than those causing damage to cells
bacterial killing in the presence of host cells. of the host organism, at least in in vitro tests.
All these phenomena and their bearing on the Still, potential toxicity is commonly listed as one
final activity and toxicity of AMPs must be of the challenges limiting the clinical application
considered in the definition of design princi- of AMPs as systemic drugs (Hancock and Sahl
ples to optimize peptide selectivity. 2006; Eckert 2011; Yeung et al. 2011; Seo et al.
2012; Carneiro et al. 2015; Pachón-Ibáñez et al.
2017), and therefore, several research efforts are
S. Bobone · L. Stella (*) devoted to understand and further improve AMP
Department of Chemical Science and Technologies, selectivity.
University of Rome Tor Vergata, Rome, Italy
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 175

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_11
176 S. Bobone and L. Stella

This chapter discusses AMP selectivity, the reviewed by Phoenix and coworkers (2012) and
origin and the structural determinants of this Harris et al. (2013).
property, the design strategies available to
improve it, and the results of recent studies on
the quantitative determination of peptide-cell 11.2  MPs Are Selective
A
association. Overall, the available data indicate for Microbial Cells
that selectivity is the result of a complex inter-
play of several interconnected phenomena, AMPs have been isolated from natural sources
including peptide association to target and host based on their antimicrobial activity. The mini-
cells, peptide conformational equilibria, and mum inhibitory concentration (MIC, i.e., the
AMP aggregation. Any modification to the pep- lowest concentration of antimicrobial agent that
tide sequence and structure necessarily affects inhibits the visible growth of a microorganism)
all of these processes, which must therefore be (Wiegand et al. 2008) or the minimum bacteri-
fully understood and considered in the rational cidal concentration (MBC, i.e., the minimal drug
design of new peptide or peptidomimetic mole- dosage killing at least 99.9% of the bacterial
cules with improved selectivity properties. Our cells) (Lorian 2005) for AMPs is usually in the
attempt is to give a critical overview of the low μM range (Giacometti et al. 1998). AMPs are
available evidences, in order to provide a ratio- typically bactericidal, and therefore, the MIC and
nale for future efforts in this area. To this end, MBC values are usually similar (Giacometti et al.
we have strived to derive, whenever possible, 1998).
generalizations of the findings reported in the The active concentration of a bioactive, thera-
literature, but we have to stress from the begin- peutically useful molecule must be much lower
ning that the presence of exceptions to every than the concentration causing toxic effects to the
rule is the norm, in such a diverse set as AMPs, host cells. This property is quantified by the ther-
also as a consequence of the complications apeutic index (TI), i.e., the ratio of the active con-
mentioned above. centration to the toxic concentration (see the
The different aspects of AMP selectivity legend to Table 11.1 for a detailed definition of
have last been reviewed by Matsuzaki in 2009 these parameters). In the case of AMPs, whose
(Matsuzaki 2009). Selectivity or toxicity has main mechanism of bactericidal action is mem-
often been considered in general review articles branolytic (as discussed in Sect. 11.3), toxicity is
on AMPs (Alba et al. 2012; Teixeira et al. 2012; most commonly assessed by measuring the lysis
Oddo and Hansen 2017; Hollmann et al. 2018). of erythrocytes (Fig. 11.1). Table 11.1 summa-
Some reviews have summarized our current rizes some TI values of natural and artificial
knowledge on the structural determinants of AMPs, which are typically in the range 10–1000.
AMP activity and selectivity (Takahashi et al. However, it should be considered that unfortu-
2010; Huang et al. 2010; Strömstedt et al. 2010; nately a strong variability is present in the litera-
Tossi 2011; Ruiz et al. 2014; Ebenhan et al. ture regarding the definition of the toxic
2014a). Finally, for a recent discussion on how concentration, because different thresholds of
the interaction of AMPs with target and host lysed red blood cells (RBCs) are utilized to define
cells determines their selectivity, see Savini the minimum hemolytic concentration (MHC),
et al. (2018). ranging from barely detectable to full hemolysis
As illustrated in other chapters of this book, (see references cited in Table 11.1) (Bacalum and
AMPs have multiple functions, including anti- Radu 2015). In addition, MIC values depend on
cancer, antifungal, and antiviral activities. For the specific strains tested in the assay.
the sake of brevity and simplicity, in this chap- When toxicity is assayed on other human
ter, we will essentially limit ourselves to discuss cells, the results are generally not very different
selectivity for bacterial versus host cells. from those obtained using hemolysis (Table 11.2),
Selectivity of anticancer peptides has been but combining the two toxicity tests obviously
 11

Table 11.1 Therapeutic index (TI) of natural and artificial AMPs

Name Sequence TI Calculated as References
Natural Peptides
Temporin-L FVQWFSKFLGRIL-NH2 1 HC50/MIC (Gram + &) Mangoni (2011)
Magainin-1 GIGKFLHSAGKFGKAFVGEIMKS 2.54 HC5/MIC (E. coli) Bacalum and Radu (2015)
PMAP-36 GRFRRLRKKTRKRLKKIGKVLKWIPPIVGSIPLGCG-NH2 3 HC5/MIC (Gram + & and C. albicans) Lyu et al. (2016)
Mastoparan X INWKGIAAMAKKLL-NH2 3.4 HC5/MIC (E. coli) Henriksen et al. (2014)
Arenicin-1 RWCVYAYVRVRGVLVRYRRCW 5 HC50/MIC (Gram + &) Panteleev et al. (2015)
Polistes Mastoparan VDWKKIGQHILSVL-NH2 6 HC5/MIC (E. coli) Bacalum and Radu (2015)
Magainin 2B GIGKFLHAAKKFAKAFVAEIMNS 9 HC50/IC50 (B. anthracis) Dawson et al (2011)
Dermaseptin 1 ALWKTMLKKLGTMALHAGKAALGAAADTISQGTQ 0.83–7 HC5/MIC (E. coli) Bacalum and Radu (2015)
LL-37 LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES 12 HC5/MIC (E. coli, S. aureus) Luo et al. (2017)
Indolicidin ILPWKWPWWPWRR-NH2 13 HC50/MIC (Gram + &) Nan et al. (2009)
4–11 HC5/MIC (E. coli) Bacalum and Radu (2015)
Gramicidin-S VOLFPVOLFP 1–18 HC50/MIC (Gram + &) Swierstra et al. (2016)
2 HC100/MIC (Gram + &) Kondejewski (1999)
XT-7 GLLGPLLKIAAKVGSNLL 18 HC50/MIC (E. coli) Kamech et al. (2012)
PGLa GMASKAGAIAGKIAKVALKAL-NH2 24 HC10/MIC (Gram + &) Strandberg et al. (2007)
Tachyplesin I KWCFRVCYRGICYRRCR-NH2 2–24 HC5/MIC (E. coli) Bacalum and Radu (2015)
Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria

SMAP-29 RGLRRLGRKIAHGVKKYGPTVLRIIRIAG 26 HC50/IC50 (B. anthracis) Dawson and Liu (2011)

Ascaphin-8 GFKDLLKGAAKALVKTVLF 37 HC50/MIC (E. coli) Kamech et al. (2012)
Papillosin GFWKKVGSAAWGGVKAAAKGAAVGGLNALAKHIQ 34 HC50/IC50 (B. anthracis) Dawson and Liu (2011)
PMAP-23 RIIDLLWRVRRPQKPKFVTVWVR 23 HC5/MIC (Gram+ &) Velduhizen (2017)
>57 HC=0/MIC (Gram + &) Kang et al. (1999)
Ascaphin-1 GIRDVLKGAAKAFVKTVAGHIAN-NH2 >60 HC50/MIC (E. coli) Juretic et al. (2009)
Magainin 2 GIGKFLHSAKKFGKAFVGEIMNS 3–82 HC5/MIC (E. coli) Bacalum and Radu (2015)
Cecropin B KWKVFKKIEKMGRNIRNGIVKAGPAIAVLGEAKAL-NH2 20–130 HC5/MIC (E. coli) Bacalum and Radu (2015)
Cecropin A KWKLFKKIEKVGQNIRDGIIKAGPAVAVVGQATQIAK-NH2 177–579 HC5/MIC (E. coli) Bacalum and Radu (2015)
177

(continued)
Table 11.1 (continued)
178

Modified analogues of natural peptides

Mastoparan-X Ala1 ANWKGIAAMAKKLL-NH2 7 HC5/MIC (E. coli) Henriksen et al. (2014)
PMAP-36 T115 TRKRLKKIGKVLKWI-NH2 29 HC5/MIC (Gram + & and C. albicans) Lyu et al. (2016)
Gramicidin GS14K4 VKLkVyPLKVKLyP 31 HC100/MIC (Gram + &) Kondejewski (1999)
D-Piscidin1 19K ffhhifrgkvhvgktihrlvtg-NH2 33 HC50/MIC (A. baumannii) Jiang (2014)
Arenicin-1 V8R RWCVYAYRRVRGVLVRYRRCW 80 HC50/MIC (Gram + &) Panteleev et al. (2015)
[K2, K16] XT-7 GKLGPLLKIAAKVGSKLL >130 HC50/MIC (E. coli) Kamech et al. (2012)
Indolicidin-A7 ILKWKWKWWKWRR-NH2 190 HC50/MIC (Gram + &) Nan et al. (2009)
D-Dermaseptin S4 alwmtlkkkvlkakakalnavlvgana-NH2 219 HC50/MIC (A. baumannii) Jiang (2014)
L7K,A14K
[I2, K19] ascaphin-8 GIKDLLKGAAKALVKTVLK >480 HC50/MIC (E. coli) Kamech et al. (2012)
Gramicidin V3/A3 AKLkAyPLKAKLyP 520 HC100/MIC (C. xerosis) Kondejewski (2002 )
Designed peptides
V13K Ac-KWKSFLKTFKSAKKTVLHTALKAISS-NH2 163 HC>0/MIC (Gram + &) Chen et al. (2005)
P5 KWKKLLKKPLLKKLLKKL-NH2 >150 HC=0/MIC (Gram + &) Park et al. (2003)
Pep-1-K KKTWWKTWWTKWSQPKKKRKV 174 HC>0/MIC (Gram + &) Zhu et al. (2009)
PK-12-KKP KKPWWKPWWPKWKK 200 HC>0/MIC (Gram + &) Zhu et al. (2009)
D16 Ac-klksllktlskakkkklktllkalsk-NH2 890 HC50/MIC (P. aeruginosa) Jiang (2011)
3355 HC50/MIC (A. baumannii)
The TI is defined as the ratio of hemolytic to inhibitory peptide concentration. HCx is defined as the peptide concentration causing the x% of hemolysis. HC>0 is the minimal peptide
concentration that produces detectable hemolysis; HC=0 is the highest peptide concentration that causes no detectable release of hemoglobin. HC100 is the minimal concentration
causing total lysis. MIC is defined as the minimum concentration that inhibits bacterial growth; when calculated on more than one strain, the geometric mean of the values is
reported. Data presented as a range result from different MIC values reported in the literature for the same peptide. –NH2 in the peptide sequence indicates amidation at the
C-terminus, while Ac- indicates acylation at the N-terminus. O is pyrrolysine. Amino acids are colored in blue, red, orange, and green based on their cationic, anionic, polar, and
hydrophobic character, respectively. Lowercase letters indicate d-amino acids
S. Bobone and L. Stella
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 179

Fig. 11.1 Selective cytotoxicity in in vitro assays. for P5 in the peptide concentration range investigated
Minimal inhibitory concentrations against different (up to 100 μM). Adapted, with permission, from research
Gram- and Gram+ bacterial strains and minimal hemo- originally published in Bobone et al. 2013, published
lytic concentration for the designed artificial AMP P5. by the European Peptide Society and John Wiley &
The asterisk indicates that no hemolysis was observed Sons, Ltd.

provides a clearer picture of the selectivity of thanks to their specific binding to bacterial cells
AMPs (Bacalum and Radu 2015). (Welling et al. 2000) (Fig. 11.2b, c). Among the
AMP activity and toxicity are usually mea- peptides used for this purpose, there are defen-
sured in separate assays performed under rather sins, cathelicidins, lactoferricins, histatins, artifi-
different conditions (for instance, regarding cell cial peptoids, and particularly sequences derived
density, see Sect. 11.9) (Matsuzaki 2009). We from ubiquicidin (Lupetti et al. 2003; Brouwer
have argued that experiments on bacterial and et al. 2008; Akhtar et al. 2012; Ebenhan 2014a, b;
human cells in co-culture would provide a more Dutta et al. 2017; Lei et al. 2018). Several imag-
stringent test of peptide selectivity (Savini et al. ing studies employed ubiquicidin 29–41
2017, 2018). However, this approach has been (Meléndez-Alafort et al. 2004; Akhtar et al. 2005;
employed only in a few cases. These studies, dis- Vallejo et al. 2008; Gandomkar et al. 2009; de
cussed in detail in Sect. 9.3, demonstrated that Murphy et al. 2010; Assadi et al. 2011; Ostovar
AMPs are selective even when acting on bacteria et al. 2013; Saeed et al. 2013; Kahrom et al. 2014;
co-cultured with mammalian cells (Fig. 11.2a). Ebenhan et al. 2018; Bhatt et al. 2018), which has
Several evidences indicate that AMPs are moderate activity and selectivity in the standard
selective also in vivo. For instance, a large body assays (MIC 40 μM, TI > 5) (Brouwer et al. 2006;
of studies starting in 1999 (Welling) has shown Lupetti et al. 2008) but accumulates at the site of
that radiolabeled (Lupetti et al. 2003; Brouwer infection. For instance, one study reported over-
et al. 2008; Akhtar et al. 2012; Ebenhan 2014a) all values of sensitivity, specificity, and accuracy
or fluorescent (Akram et al. 2015) AMPs can be for infection detection of 100%, 80%, and 94%
used to image infections in vivo and can even dis- (Akhtar et al. 2005). Analogues of the ubiquici-
criminate between infection and inflammation, din peptide have been used also for targeted
 180

Table 11.2 Comparison of TI values determined with RBCs or with other eukaryotic cells
Name Sequence TI (cells) Calculated as TI (RBCs) Calculated as References
Natural peptides
Magainin2-NH2 GIGKFLHSAKKFGKAFVGEIMNS-NH2 6 LC99 (3T3)/MIC (Gram+ &) >130 HC100 /MIC (Gram+ &) Javadpour et al. (1996)
Polistes Mastoparan VDWKKIGQHILSVL-NH2 7 LC50(PBMC)MIC (E. coli) 6 HC5/MIC (E. coli) Bacalum and Radu (2015)
Dermaseptin 1 ALWKTMLKKLGTMALHAGKAALGAAADTISQGTQ 17 LC50 (PBMC)/MIC (E. coli) 110 HC5/MIC (E. coli) Bacalum and Radu (2015)
Lasioglossin III VNWKKILGKIIKVVK-NH2 >8 LC50 (HUVEC)/MIC (Gram+ &) >105 HC50/MIC (Gram+ &) Slaninová et al. (2012)
Magainin I GIGKFLHSAGKFGKAFVGEIMKS 712 LC50 (PBMC)/MIC (E. coli) 2.54 HC5/MIC (E. coli) Bacalum and Radu (2015)
Indolicidin ILPWKWPWWPWRR-NH2 617 LC50 (PBMC)/MIC (E. coli) 411 HC5/MIC (E. coli) Bacalum and Radu (2015)
BMAP-28 GGLRSLGRKILRAWKKYGPIIVPIIRIG 17 LC95 (HN)/MIC(Gram+ &) 55 HC90/MIC (Gram+ &) Skerlavaj et al. (1996)
BMAP-27 GRFKRFRKKFKKLFKKLSPVIPLLHLG 18 LC80 (HN)/MIC (Gram+ &) 59 HC30/MIC (Gram+ &) Skerlavaj et al. (1996)
PMAP-23 RIIDLLWRVRRPQKPKFVTVWVR 23 LC30 (IPEC-J2)/MIC (Gram+ &) 23 HC5/MIC (Gram+ &) Velduhizen (2017) and
>57 HC=0/MIC (Gram + &) Kang et al. (1999)
Cecropin B-NH2 KWKVFKKIEKMGRNIRNGIVKAGPAIAVLGEAKAL-NH2 29 LC99 (3T3)/MIC (Gram+ &) >86 HC100 /MIC (Gram+ &) Javadpour et al. (1996)
Tachyplesin I KWCFRVCYRGICYRRCR-NH2 330 LC50 (PBMC)/MIC (E. coli) 224 HC5/MIC (E. coli) Bacalum and Radu (2015)
Magainin2 GIGKFLHSAKKFGKAFVGEIMNS 382 LC50 (PBMC)/MIC (E. coli) 382 HC5/MIC (E. coli) Bacalum and Radu (2015)
Cecropin A KWKLFKKIEKVGQNIRDGIIKAGPAVAVVGQATQIAK-NH2 27180 LC50 (PBMC)/MIC (E. coli) 20130 HC5/MIC (E. coli) Bacalum and Radu (2015)
Modified analogues of natural peptides
A(A1R A8R I17K) RIGSILGRLAKGLPTLKSWIKNR-NH2 1.53 LC50 (L929)/MIC (Gram+ &) 1016 HC10/MIC (Gram+ &) Zhang (2016)
Designed peptides
D-LAK120 AP13 kklalalakkwlplakklalalakk-NH2 4 LC50 (RAW 264.7)/MIC (Gram) 67 HC50/MIC (Gram) Vermeer (2012)
WK12 KWWKWWKKWWKK >10 LC20 (PBMC)/MIC (Gram) >10 HC10/MIC (Gram) Deslouches et al. (2016)
WR12 RWWRWWRRWWRR >20 LC20 (PBMC)/MIC (Gram) >20 HC10/MIC (Gram) Deslouches et al. (2016)
(KLAKKLA)2 KLAKKLAKLAKKLA >45 LC99 (3T3)/MIC (Gram+ &) >125 HC100 /MIC (Gram+ &) Javadpour et al. (1996)
(KLAKLAK)2 KLAKLAKKLAKLAK >86 LC99 (3T3)/MIC (Gram+ &) >125 HC100 /MIC (Gram+ &) Javadpour et al. (1996)
LC is the lethal peptide concentration. LCx is defined as the peptide concentration killing x% of the cells. For HC and MIC definitions and for the color code used for the sequences,
please see Table 11.1. When data are presented as a range, they result from different MIC values reported in the literature for the same peptide. HBRC human red blood cells,
PBMC peripheral blood mononuclear cells (3 · 106/mL), 3T3 cells murine fibroblast cell line (2 · 105/mL), HUVEC cells human umbilical vein endothelial cells (2 · 104/mL), HN
human neutrophils (4 · 106/mL), IPEC-J2 porcine intestinal epithelial cells (1.5 · 105/mL), L929 mouse fibroblast cells (1.5 · 104/mL), RAW 264.7 murine macrophage cells (3 · 105/
mL). The density of RBCs used in the hemolytic activity assays (when specified in the original reference) ranged from 106 to 109 cells/mL. Lowercase letters indicate d-amino
acids
S. Bobone and L. Stella
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 181

Fig. 11.2 Selective targeting of bacterial cells by AMPs 30 min after tracer administration (Reproduced, with per-
in vitro and in vivo. (a) Optical and fluorescence micros- mission, from research originally published in Akhtar et al.
copy image of a labeled ubiquicidin analogue (visible by 2012). (c) Visualization of in vivo targeting of human
the green fluorescence) selectively binding to S. aureus α-defensin 5 (HD5) toward E. coli cells. The mesenteric
bacteria, in co-culture with isolated human neutrophils vein was imaged intravitally in mice by two-photon laser
(blue arrows) Adapted, with permission, from research scanning microscopy, 30 min after injection of E.coli cells
originally published in Akram et al. 2015, published by expressing a green fluorescent protein (visualized in the left
The Royal Society of Chemistry. (b) Positron emission panel), and treatment with HD5 labeled with a red fluores-
tomography image of a patient with an infection in the left cent probe (imaged in the center panel). Colocalization is
hand (indicated by the arrow), traced with a radiolabeled demonstrated by the overlapped images (right panel). Scale
ubiquicidin analogue. No significant peptide uptake in the bars, 50 μm. Adapted, with permission, from Lei et al.
contralateral hand was noted. The image was obtained 2018. Copyright (2018) American Chemical Society

delivery of traditional antibiotics to the infection have been exploited in sensing elements that can
site (Chen et al. 2015). detect infection (Mannoor et al. 2010; Shriver-­Lake
Although in vivo studies of the activity of et al. 2012; Silva et al. 2014; Hoyos-Nogués et al.
AMPs abound, similar investigations character- 2018), even in whole blood (Shi et al. 2017) or other
izing their toxicity are more sparse (Mahlapuu complex biological samples (Qiao et al. 2017).
et al. 2016). Some TIs derived from animal stud- Overall, the results collected in the literature
ies are summarized in Table 11.3, and the range support an interesting selectivity of AMPs for tar-
of values is similar to that obtained in vitro. get versus host cells. The origin of this property
The selectivity of AMPs for bacterial cells is is necessarily related to the mechanism of action
demonstrated also by the fact that these peptides of AMPs.
 182

Table 11.3 TI values obtained from in vivo studies

Name Sequence TI Calculated as Bacterial strain References
Natural pepdes
P3 VNFKLLSHSLLVTLASHL 3 LD50/ED60 E.coli Zhang (2015)
OH-CATH30 KFFKKLKNSVKKRAKKFFKKPRVIGVSIPF 12 LD50/ED80 E.coli Li (2012)
BMAP-27 GRFKRFRKKFKKLFKKLSPVIPLLHLG 4755 LD50/ED100 P.aeruginosa Benincasa (2003)
190220 LD50/ED100 E.coli
BMAP-28 GGLRSLGRKILRAWKKYGPIIVPIIRIG 2427 LD50/ED100 E.coli Benincasa (2003)
4755 LD50/ED100 S.aureus
Modified analogues of natural pepdes
HD5-myr ATCYCRTGRCATRESLSGVCEISGRLYRLCCR-myr >3 LD=0/ED90 E. coli Lei (2018)
RN7-IN8 FLGGLIKWPWWPWRR-NH2 4 LD100/ED50 S. pneumoniae Jindal (2017)
JH3 RRFKLLSHSLLVTLASHL 4.5 LD50/ED90 E.coli Zhang (2015)
TP3 FIHHIIGGLFSVGKHIHSLIHGH 7 MTD/ED60 A.baumannii Pan (2015)
7 MTD/ED70 K pneumoniae Pan (2015)
>24 MTD/ED100 MRSA Huang (2015a)
D-OH-CATH30 kffkklknsvkkrakkffkkprvigvsipf 8 LD50/ED100 E.coli Li (2012)
OH-CM6 KFFKKLKKAVKKGFKKFAKV 10 LD50/ED70 E.coli Li (2012)
TP4 FIHHIIGGLFSAGKAIHRLIRRRRR 20 MTD/ED90 A.baumannii Pan (2015)
20 MTD/ED90 K pneumoniae Pan (2015)
>24 MTD/ED100 MRSA Huang (2015b)
Designed pepdes
A3-APO (Chex-RPEKPRPYLPRPRPPRPVR)2-Dab-NH2 2.5 LD50/ED100 E.coli Szabo (2010 )
Onc72 VDKPPYLPRPRPPROIYNO-NH2 >20 LD50/ED50 E.coli Knappe (2012)
(LLKK)2C LLKKLLKKC 28 LD50/ED50 A.baumannii Huang (2012)
C(LLKK)2C CLLKKLLKKC 34 LD50/ED50 A.baumannii Huang (2012)
TI was calculated as the ratio of the toxic or lethal dose to the effective dose of peptides in mice. LDx is the minimum dose that was lethal for at least x% of animals; LD=0 is the
highest peptide concentration that caused no deaths; MTD is minimum dose that caused toxicity (narrowing of the eyes was taken as symptom), MLD is the minimum lethal dose,
and EDx is the effective dose, i.e., the minimum dose causing the survival of at least x% of the animals. Bacteria and peptides were injected intraperitoneally, except for Huang
(2015a, b) and Pan (2015), where toxicity tests were performed by intramuscular injection. -myr indicates myristoylation at the C-terminus. Chex 1-amino-cyclohexane carbox-
ylic acid, Dab 2,4-diamino-butyric acid, O pyrrolysine. MRSA is methicillin-resistant S. aureus. For the color code used for the sequences, please see Table 11.1. Lowercase
letters indicate d-amino acids
S. Bobone and L. Stella
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 183

11.3  ellular Membranes Are

C dients are dissipated (Tiozzo et al. 1998;
the Main Target of AMPs Arcidiacono et al. 2009; Hartmann et al. 2010;
Agrawal and Weisshaar 2018) (Fig. 11.4).
Selectivity is not surprising when a biomolecule Usually, membrane perturbation and bacterial
associates to a specific receptor or protein (Le killing are correlated, further supporting mem-
Joncour and Laakkonen 2018). However, this is brane disruption as the main bactericidal mecha-
not the case for most AMPs. In general, natural nism. However, also in this case, exceptions exist,
AMPs and their enantiomers comprising all indicating that a subclass of AMPs might act
D-amino acids have a comparable antimicrobial through different killing mechanisms (He et al.
activity (Fig. 11.3), while any interaction of a 2014; Friedrich et al. 2000). Finally, it is worth
peptide with a protein, due to the chirality of both mentioning that AMPs usually are able to perturb
systems, would be favored for one enantiomer the permeability of artificial membranes, com-
over the other. As is often the case for AMPs, prising only phospholipids (Fig. 11.4c) (Orioni
exceptions to this rule have been reported (Otvos et al. 2009; Bocchinfuso et al. 2011; Braun et al.
et al. 2000; Bulet and Stocklin 2005; de la Fuente-­ 2017; Savini et al. 2018). This observation dem-
Núñez et al. 2015), showing that the mechanism onstrates that the membrane-perturbing activity
of action of a minority of AMPs could be recep- is purely the result of physicochemical interac-
tor mediated. On the other hand, microbiological tion between the peptide and the lipid bilayer,
assays of membrane permeability and micro- and not the consequence of some biological
scopic imaging of bacteria treated with AMPs process.
clearly show that cell membranes are damaged Overall, literature data clearly demonstrate
and that, as a consequence, transmembrane gra- that membrane perturbation is the main mecha-

Fig. 11.3 Activity of natural AMPs and their enantio- values), camel 48 (Oh et al. 2000) (red), V681 and ana-
meric analogues. Comparison of the antibacterial activi- logues (Chen et al. 2006) (dark red; data refer to LC50 val-
ties (circles for MBC, squares for MIC) of enantiomeric ues), lactoferricin B analogues (Wakabayashi et al. 1999)
peptides. Data refer to magainin 2 (Bessalle et al. 1990) (silver), and cecropin B (Bland et al. 2001) (black). The
(violet), cecropin A (Wade et al. 1990) (blue), melittin blue line is the diagonal of the plot (corresponding to
(Juvvadi et al. 1996) (dark green), LL-37 (Dean et al. identical activity for D and L enantiomers) and not a fit. A
2011) (light green), KSLK (Hong et al. 1999) (yellow), version of this figure with a more limited set of data has
temporin A (Wade et al. 2000) (orange; data refer to IC50 been published previously (Savini et al. 2018)
184 S. Bobone and L. Stella

Fig. 11.4 Fluorescence and electron microscopy images uptake of the DNA stain Sytox Orange (red fluorescence)
of the effects of AMPs on bacterial and artificial mem- demonstrates pore formation in the plasma membrane
branes. (a) Scanning electron micrographs of E. coli (top) (Adapted, with permission, from research originally pub-
and S. aureus (bottom) before (images on the left) and after lished in Agrawal 2018 © Elsevier). (c) Fluorescence
(images on the right) treatment with the synthetic AMP microscopy images of perturbation of a giant unilamellar
PGYa (30 min, 10 μM). The images show a considerable vesicle by the AMP PMAP-23. The top panels report the
roughening of the bacterial membranes and formation of green fluorescence emission from carboxyfluorescein mol-
blebs on the cell surface, in contrast to the smooth surfaces ecules entrapped inside the GUV, which were completely
of untreated bacteria, providing a strong indication that the released after peptide addition (right). By contrast, the
membrane is being considerably altered by the peptide. vesicle was still present after peptide addition, as indicated
Adapted, with permission, from research originally pub- by the red fluorescence of rhodamine-­labeled phospholip-
lished in Tiozzo et al. 1998 © Elsevier. (b) Fluorescence ids located in the GUV bilayer (bottom panels). Taken
microscopy images of an E. coli cell attacked by the AMP together, these images demonstrate pore formation by the
cecropin A (0.5 μM). The leakage of periplasmic green AMP. The vesicle diameter is about 20 μm. Adapted, with
fluorescent protein (GFP), shown by the green fluores- permission, from research originally published in Orioni
cence, indicates perturbation of the outer membrane, while et al. 2009 © Elsevier

nism of direct bacterial killing for most AMPs. Incidentally, the fact that AMPs target micro-
Even for those AMPs that act through a different bial membranes determines their broad-spectrum
antibacterial mechanism (Nicolas 2009; Otvos activity, their bactericidal, rather than bacterio-
2017), the cell envelope is the first cell compo- static, mechanism of action and also the higher
nent that the peptides encounter, and they have to difficulty for bacteria in developing resistance
cross the extracellular membrane (when pres- against them (compared to resistance against
ent) and the cell wall to reach the plasma mem- conventional antibiotics acting on a protein tar-
brane and eventually the cell interior. get) (Perron et al. 2006; Otvos 2017).
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 185

11.4  acterial and Host Cells Have

B negatives, both membranes contain phosphati-
Different Membrane dylglycerol (PG, ~20% overall) and cardiolipin
Structure and Composition (CL, ~5% overall); in Gram-positive bacteria, the
content of anionic lipids is much higher, again
If membranes are the target, then it is conceivable with PG and CL being the most important com-
that selectivity arises from a difference in mem- ponents (Malanovic and Lohner 2016). However,
brane composition of the various cell types. the membranes of these cells can also contain the
Indeed, bacterial and eukaryotic cells have very positively charged l-lysyl-PG (LPG). In both
different cell envelopes (Wang 2017). Bacteria cases, the main zwitterionic component is phos-
can be divided into Gram-positive and Gram-­ phatidylethanolamine (PE), and no sterols are
negative, depending on whether they are colored present. Of course, the values for the composition
by the Gram stain or not. This assay reflects dif- reported in Table 11.4 are only approximate,
ferences in the composition of the cell envelope. since they change with the specific strain and
In both cases, the plasma membrane is sur- growth conditions. In addition, lipid composition
rounded by a cell wall. However, in Gram-­ is not homogeneous over the cell surface (Renner
positive bacteria, this is formed by a thick and Weibel 2011; Oliver et al. 2014).
peptidoglycan and lipoteichoic acid layer (40– Human cells contain cholesterol and have no
80 nm). By contrast, in Gram negatives, a thin anionic phospholipids in the outer leaflet of their
peptidoglycan layer (8 nm thick) is contained in a cell membrane. Some negatively charged glyco-
second (outer) membrane, with asymmetric com- lipids, such as gangliosides, are present on the
position: phospholipids are the main components cell surface (Miyazaki et al. 2012), but they are
of the inner leaflet, while the outer layer is mainly minor components in most cell membranes (with
formed by lipopolysaccharides (LPS). On the the exception of nerve cells) (Storch and Kleinfeld
other hand, eukaryotic cells only have the plasma 1985). These properties are exemplified by
membrane, with asymmetric lipid composition in RBCs, which are commonly used to test toxicity
the two leaflets of the bilayer (Fig. 11.5). and selectivity (Table 11.5). For eukaryotes, the
In addition to the different structures of the main zwitterionic components are phosphatidyl-
cell envelope, important differences are present choline (PC), sphingomyelin (SM), and PE.
in the lipid composition of the cellular mem- Overall, we can translate these differences in
branes. Tables 11.4 and 11.5 summarize the lipid lipid composition in distinct physicochemical
content of bacterial and RBC membranes. properties. Bacterial membranes contain more
Bacterial membranes contain a significant anionic lipids in the outer surface of their bilayers
fraction of negatively charged lipids: in Gram than eukaryotic cells. This difference combines

Fig. 11.5 Schematic depiction of the structure of the cel- acids (in Gram+ bacteria) have been omitted, for the sake
lular envelope in different cell types. The three panels, of clarity. LPS, light gray; peptidoglycan, dark gray; PC,
from left to right, schematize the structure of the cellular light green; PE, dark green; SM, light brown; PI, light
envelope in Gram- and Gram+ bacteria and in human gray; PG, red; PS, dark red; CL, orange; L-lysyl PG, blue;
cells, respectively. Proteins, glycolipids, and lipoteichoic cholesterol, beige
186 S. Bobone and L. Stella

Table 11.4 Phospholipid composition of the membranes of Gram-negative (name highlighted in red) and Gram-­
positive (name highlighted in blue) bacteria
Phospholipid composion of bacterial membranes
PE PG CL PA L-lysyl PG
Total charge 0 1 2 (1) 1 +1
Intrinsic curvature  0   +
E. coli
(both membranes)
(Ames 1968) 69 19 6.5
(Raetz 1986) 7585 1020 515
(Morein et al. 1996) 79 17 4
(Rowlett et al. 2017) 78 12 6 <3
E. coli
(cell membrane)
(Morein et al. 1996) 75 19 6
S. typhimurium
(both membranes)
(Osborn 1972) 86 13 1
(Ames 1968) 78 18 3
S. typhimurium
(cell membrane)
(Osborn 1972) 76 21 3
B. subtilis
(Op den Kamp et al. 1969)* 30 36 12 22
(Bishop et al. 1967)** 34 49 11
S. aureus
(Hayami 1979):
(strain Newman) 49 13 0.6 34
(strain Tazaki) 47 10 1 38
Phospholipid composition data are expressed as molar percentages. In the case of Hayami et al. (1979), data were cal-
culated converting the % of phosphorus to molar % by considering two P atoms per CL molecule and one for the other
lipids. In the case of Osborn et al. (1972), data were calculated converting the % of [2-3H] glycerol to molar %, by
considering three 3H atoms per CL molecule, two per PG, and one for the other lipids. Data on total charge at physio-
logic pH and intrinsic curvature were taken from Marsh (1990), McMahon and Boucrot (2015), Malanovic and Lohner
(2016), and Boyd et al. (2017)
PE phosphatidylethanolamine, PG phosphatidylglycerol, CL cardiolipin, PA phosphatidic acid, L-lysyl PG L-lysyl
phosphatidylglycerol
*Indicates a growth condition without glucose and sulfate
**Indicates that 6% of lipo-amino acids were also recovered.

with the additional negative charges conferred to addition, they contain larger amounts of “non-­
bacterial cells by teichoic and teichuronic acids bilayer” lipids, with negative or positive values
and LPS. Furthermore, the transmembrane poten- for the “intrinsic curvature,” such as PE, CL, and
tial of bacterial cells is more inside-negative than PA, or LPG, respectively (McMahon and Boucrot
that of normal mammalian cells (Yeaman and 2015; Malanovic and Lohner 2016). This prop-
Yount 2003). For all these reasons, bacteria have erty depends on the relative sizes of the phospho-
stronger electrostatic interactions with positively lipid head-groups and acyl chains. Lipids where
charged molecules than eukaryotic cells. Another the cross-sectional area occupied by head-groups
difference is that bacterial membranes are more and tails is similar (e.g., PC, PG, PS) are said to
disordered and less well packed than those of have a cylindrical shape and pack well in locally
eukaryotes, due to the lack of cholesterol. In flat bilayer structures (zero intrinsic curvature).
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 187

Table 11.5 Phospholipid and cholesterol content of human erythrocyte membrane

Phospholipid composion of human RBC membranes
PC PE SM PI PS PA
Total charge 0 0 0 1 1 1 (2)
Intrinsic curvature 0  0 + 0 
Both leaflets
(Dodge and Phillips 1967) 29.2±1.5 27.5±1.5 25.4±1.4 0.6±0.5 14.8±1.7 1.1±0.5
(Broekhuyse 1969) 28.3±2.1 26.7±1.0 25.8±1.7 1.9±0.6 12.7±1.3
(White 1973) 34.7 28.0 20.1 14.3
(Verkeleij 1973) 28 26 24 13
(Van Meer 1981) 29.5 25.9 25.3 12.2
Outer leaflet
(Verkeleij 1973) 42 10 40 0
(Virtanen et al. 1998) 44.8 11.1 42.1
Inner leaflet
(Verkeleij 1973) 14 42 8 26
(Virtanen et al. 1998) 14.0 43.9 9.1 1.2 29.6 2.2
Asymmetry of distribution
(% in the outer leaflet)
(Verkleij 1973) 76 20 82 0
(Zwaal et al. 1973) 62 83
(Gordesky and Marinetti 1973) 15 0
(Zwaal et al. 1975) 75
(Gordesky et al. 1975) 33 0
(Van Meer 1981) 78 20 80
(Bütikofer et al. 1990) 24
(Gascard et al. 1991) 20

Cholesterol content as cholesterol/phospholipid molar rao

(Cooper 1975) 0.95
(Ballas and Krasnow 1980) 0.75
(Chabanel 1983) 0.80
Phospholipid composition data are expressed as molar percentages. Data from Verkeleij (1973) were derived from a
figure in the cited reference. Data on total charge at physiologic pH and intrinsic curvature were taken from Marsh
(1990) and McMahon and Boucrot (2015)
PC phosphatidylcholine, PE phosphatidylethanolamine, SM sphingomyelin, PI phosphatidylinositol, PS phosphatidyl-
serine, PA phosphatidic acid

By contrast, lipids where the head-group is membrane composition (particularly regarding

smaller than the tails (e.g., PE or PA) favor con- the content of anionic lipids and sterols) have
cave shapes of the monolayer (negative curva- been proposed to explain the selectivity for can-
ture). The opposite is true for lipids with cer cells (Hoskin and Ramamoorthy 2008;
comparatively larger polar heads (e.g., LPG) Schweizer 2009; Phoenix et al. 2012; Gaspar
which have a positive curvature (Koller and et al. 2013), fungi (van der Weerden et al. 2013;
Lohner 2014). Rautenbach et al. 2016), protozoa (Rivas et al.
The differences in lipid composition and in 2009), and enveloped viruses (Aloia et al. 1993;
physical properties between bacterial and human Findlay et al. 2013), since in all cases the lipid
cell membranes are considered to be the origin distribution is different from that of a normal
of AMP selectivity. Similar considerations on eukaryotic cell.
188 S. Bobone and L. Stella

11.5 Lipid Composition (Matsuzaki et al. 1995; Tytler et al. 1995; Hallock
Determines the Affinity et al. 2002; Sood et al. 2008; Sood and Kinnunen
of AMPs for Lipid Bilayers 2008; Gonçalves et al. 2012; Verly et al. 2008;
Wu et al. 2010; McHenry et al. 2012). The
The hypothesis of a selectivity based on differ- membrane-­ordering effects of cholesterol in fluid
ences in lipid composition has been tested by bilayers are well established: insertion of the
studying the interaction of AMPs with model rigid ring structure of the sterol limits the possi-
membranes mimicking the composition of the bility for trans-gauche isomerization for adjacent
natural bilayers. With liposomes, it is possible to phospholipid tails, leading to an increase in
vary the lipid composition at will and to measure bilayer order, packing, thickness, and rigidity
both peptide-membrane association and peptide-­ (Henriksen et al. 2006; Mouritsen and
induced membrane permeability (Bocchinfuso Zuckermann 2004). All these effects could con-
et al. 2011; Savini et al. 2018). The role of vari- tribute to reduce peptide binding and membrane
ous membrane properties in AMP selectivity is perturbation (McIntosh et al. 2002). However,
summarized in the following sections. the relevance of cholesterol for AMP selectivity
has been recently questioned. While all the inves-
tigations listed above were performed on simple
11.5.1 Membrane Charge lipid mixtures, a comprehensive study by
Ramamoorthy and coworkers on more realistic
In model membranes, the presence of anionic lip- lipid compositions showed that cholesterol’s pro-
ids increases peptide association to the bilayer tective effect against AMPs does not occur in
and, as a consequence, peptide-induced leakage lipid systems containing raft domains and pre-
(Matsuzaki et al. 1989, 1995; Gazit et al. 1995; senting phase separation (McHenry et al. 2012;
Abraham et al. 2005; Sood et al. 2008; Russell Brender et al. 2012). There are examples where
et al. 2010; Bobone et al. 2013; Golbek et al. the activity of AMPs is not affected by the pres-
2017; Maturana et al. 2017). On the other hand, ence of cholesterol even in simple lipid mixtures
the positively charged lipid lysyl-PG, present in (Bobone et al. 2013). On the other hand,
Gram+ bacteria, inhibits AMP activity (Nishi Matsuzaki et al. (1995) demonstrated an AMP-­
et al. 2004; Andra et al. 2011). These findings are inhibiting effect of cholesterol in real cells, by
a straightforward consequence of electrostatic artificially varying the cholesterol content of
interaction of the membranes with the positively RBCs.
charged AMPs (see Sect. 11.6). Regarding the It is worth mentioning that the inhibitory
anionic gangliosides present in the outer leaflet of effect is specific of cholesterol, while ergosterol,
eukaryotic membranes, Matsuzaki and cowork- present in fungal membranes, does not appear to
ers (2012) demonstrated that, although their inhibit peptide binding and activity to the same
acidic moieties favor the association of AMPs to extent, in agreement with the specific activity of
model membranes, this interaction does not lead antifungal peptides (Sood and Kinnunen 2008;
to strong membrane perturbation, since the pep- Gonçalves et al. 2012) and with the compara-
tides remain trapped in the sugar region. tively smaller effects of ergosterol on membrane
order (Henriksen et al. 2006).

11.5.2 Cholesterol Content

11.5.3 Intrinsic Curvature
Several studies also reported an AMP inhibitory
effect of cholesterol. For instance, the presence The situation is less clear regarding the effect of
of cholesterol inhibits the membrane-perturbing the presence of negative curvature lipids (PE) in
activity of magainin, pardaxin, LL-37, temporin bacterial membranes. PE has been shown to
L, human defensin HNP1, and other AMPs inhibit pore formation by magainin, melittin,
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 189

alamethicin, PMAP-23, and mastoparan X 11.6 Thermodynamics of Peptide-­

(Matsuzaki et al. 1998; Allende et al. 2005; Lee Membrane Association
et al. 2005; Bobone et al. 2012). On the other
hand, the activity of some AMPs is favored by the In principle, selectivity for different membrane
presence of PE (Schröder-Borm et al. 2003; compositions could result from two effects.
Epand et al. 2006; Leite et al. 2015). Inhibition of AMPs could have a higher affinity for bacterial
pore formation by PE can be understood by con- membranes than for human bilayers, or they
sidering that the peptides act by inserting in the could be more effective in perturbing the former,
head-group region of the membrane, thus impos- once inserted (Wimley and Hristova 2011). The
ing a positive curvature strain, which is released data on model membranes presented above
after a threshold of membrane-bound peptide clearly indicate that differential binding is an
concentration is reached, through the formation important aspect of AMP selectivity.
of membrane defects or pores. The presence of AMPs are usually short (about 10–50 residues
lipids with negative intrinsic curvature would in length), and their sequences and structures
counteract this mechanism (Matsuzaki et al. have no common features, except for the cationic
1998, Lee et al. 2005). Similar considerations, on charge (most AMPs fall in the range of +2 to +4
the other hand, suggest that PE can favor mem- e), and amphipathic character, with an overall
brane binding of AMPs, by reducing the intrinsic content of about 50% hydrophobic residues
curvature strain needed for peptide insertion in (Wang 2017). The role of these properties is eas-
the polar region of the bilayer; an increased bind- ily rationalized: charge imparts selectivity toward
ing to PE-containing membranes has been bacterial versus eukaryotic membranes, and apo-
reported for some AMPs (Schröder-Borm et al. lar residues provide a hydrophobic driving force
2003; Phoenix et al. 2015). However, reasoning for binding and insertion into membranes, lead-
only in terms of intrinsic curvature might be mis- ing to perturbation of bilayer integrity. Indeed,
leading. For instance, the peculiar lipid-lipid peptides interacting only electrostatically usually
interactions made possible by the structure of PE do not cause significant membrane leakage,
could inhibit peptide insertion: this phospholipid because their depth of insertion is too shallow
contains a primary amine (lacking in PC), which (Wimley 2010a).
allows it to form strong hydrogen bonds with
phosphate or CO groups in other lipids (Lewis
and McElhaney 2005). This H-bond network is 11.6.1 Hydrophobic and Electrostatic
responsible for the melting temperatures of PE Driving Forces Are
lipids being higher than those of their corre- Nonadditive
sponding PC analogues (Lewis and McElhaney
2005). This difference might lead to an increase Different treatments are used in the literature to
in the energy needed to insert a peptide in the describe peptide-membrane interactions, and
bilayer, or to open a pore, in the presence of comprehensive reviews are available on this topic
PE. Finally, an additional mechanism invoked to (White and Wimley 1999; Wieprecht and Seelig
explain different activities on vesicles lacking or 2002; Simon and McIntosh 2002; Santos et al.
containing PE involves peptide-induced forma- 2003; Seelig 2004; Wimley 2010a). Here we will
tion of lipid domains (Epand et al. 2006). Overall, just briefly mention that, since peptide-­membrane
these considerations can explain why different association does not have a specific stoichiome-
final effects are observed on the membrane-­ try, it is not correctly described by a binding equi-
perturbing activity of AMPs, depending on which librium, and it is better treated as a partition
of the various phenomena predominates in each equilibrium between the water and the membrane
specific case. In any case, it is difficult to ascribe phase (White and Wimley 1999; Wieprecht and
a well-defined role in AMP selectivity to the PE Seelig 2002; Santos et al. 2003; Wimley 2010a).
content. In this view, the main effect of Coulombic inter-
190 S. Bobone and L. Stella

actions can be described as an increase in local toxic (Dathe et al. 1996; Cornut et al. 1994;
peptide concentration in the vicinity of the Bobone et al. 2013). These findings demon-
bilayer, according to Gouy-Chapman theory strated that a cationic charge is not sufficient for
(Beschiaschvili and Seelig 1990; Wieprecht and specificity and provided a first indication of the
Seelig 2002; Seelig 2004). However, the thermo- complexities of peptide-membrane interaction
dynamic contributions of electrostatic and hydro- discussed above. Peptides in solution can
phobic effects to the driving force of peptide assume different conformations and aggregation
water-membrane partition are not simply additive states, and once membrane-bound, they can
(Ladokhin and White 2001). This finding is due change conformation, orientation, insertion
mainly to the different depths of polar and ali- depth, and aggregation state (Fig. 11.6). All
phatic moieties of phospholipids in the bilayer: these phenomena are regulated by intercon-
charged groups are located on the surface of the nected equilibria, and therefore they contribute
membrane, well separated from the hydrocarbon in determining the final membrane-­perturbing
core, and the physicochemical properties of the activity (Stella et al. 2004; Mazzuca et al. 2005;
bilayer vary steeply in the head-group region Gatto et al. 2006; Bobone et al. 2013). Every
(Wimley 2010a). As a consequence, the depth of modification in peptide properties can affect all
insertion of a peptide in the membrane is deter- these processes (Gatto et al. 2006). In our opin-
mined by the interplay between hydrophobic ion, this is the reason why the rational design of
effect and Coulombic forces (Wimley 2010a). peptides with improved selectivity has met with
Highly charged, hydrophilic molecules sit on the limited success, and it has progressed through a
membrane surface, and strongly hydrophobic trial-and-error process. Even so, several useful
peptides insert into the hydrocarbon core, while principles for the optimization of AMP selectiv-
cationic, amphipathic peptides are located at an ity have been defined.
intermediate position, which depends on their
specific properties (Bocchinfuso et al. 2009;
Farrotti et al. 2015). In turn, the depth of insertion 11.7  electivity of AMPs Is
S
in the bilayer modulates the intensity of electro- Determined by Their
static and hydrophobic contributions: strongly Physicochemical Properties
simplifying, one could say that the position of a
peptide in the membrane determines the average Helical peptides are the most abundant and best-­
distance between the peptide and the charged characterized class of AMPs. Investigations on
lipid moieties and the degree of insertion of the the structural determinants of AMP selectivity
peptide in the water-free hydrocarbon core. An have mostly focused on this type of peptides.
increase in peptide hydrophobicity ultimately They are usually disordered in solution but
reduces the effect of electrostatic interactions; on attain a helical conformation when membrane-
the other hand, augmenting the Coulombic forces bound, with a spatially amphipathic distribution
diminishes the hydrophobic contribution to the of the side chains, where most of the hydropho-
binding free energy (Ladokhin and White 2001). bic residues face toward the membrane center
and the polar and charged residues are oriented
toward the water phase. From a physicochemi-
11.6.2 Multiple Interconnected cal point of view, they can be characterized by
Equilibria Modulate Peptide several parameters, such as charge, hydropho-
Activity and Selectivity bicity, or amphipathicity. In addition to the con-
siderations discussed in the previous section, on
Several artificial peptides were designed having the multiple processes involved in peptide-
the required characteristics of cationic charge membrane interactions, investigations on the
and amphipathic character. However, in many role of each of these parameters are complicated
cases, such peptides turned out to be highly by the fact that varying the sequence by a single
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 191

Fig. 11.6 Schematic depiction of the phenomena and aggregation equilibria are at play in determining the
involved in peptide-membrane interaction. In addition to final peptide activity and selectivity
water-membrane partition, conformational, orientational,

amino acid substitution usually causes a varia- ity and selectivity has often been described
tion in multiple physicochemical properties of (Bessalle et al. 1992; Matsuzaki et al. 1997;
the peptides (Wieprecht et al. 1997a; Dathe Dathe et al. 2001; Giangaspero et al. 2001;
et al. 2001). For instance, inserting an additional Zelezetsky and Tossi 2006; Bobone et al. 2011).
cationic residue does not vary only the peptide However, several studies reported that increas-
charge but also its hydrophobicity and amphipa- ing cationicity above a certain level (+5, +8, or
thicity. However, some systematic studies have +9 depending on the specific case) is not benefi-
been performed where the authors tried to com- cial and might even cause a decrease in activity
pare peptide sequences where multiple substitu- or selectivity (Dathe et al. 2001; Giangaspero
tions were inserted to cause the significant et al. 2001; Zelezetsky and Tossi 2006; Jiang
variation of one parameter only, while the others et al. 2008). This last finding might be due to an
were kept as constant as possible (Dathe et al. overly shallow insertion of the peptide in the
1997, 2001, 2002; Wieprecht et al. 1997a, b, c; bilayer (Wimley 2010a) and to the nonadditivity
Dathe and Wieprecht 1999; Giangaspero et al. of electrostatic and hydrophobic effects, dis-
2001; Zelezetsky et al. 2005; Zelezetsky and cussed in Sect. 11.6.
Tossi 2006). One of the ways to increase the total positive
charge of the peptide is C-terminal amidation,
which is frequent in natural sequences and has
11.7.1 C
 ationic Charges Favor the additional advantage of reducing susceptibil-
Selectivity ity to proteolytic degradation (Huang et al. 2010;
Mura et al. 2016). However, this approach to
Based on the results on the importance of anionic increase peptide selectivity is not generally valid,
lipids for the membrane activity of AMPs, it is possibly because it also affects the stability of
not surprising that a positive correlation between helical conformations in solution (Dennison et al.
peptide positive charge and antimicrobial activ- 2009) (see Sect. 11.8).
192 S. Bobone and L. Stella

11.7.2 H
 ydrophobicity Is Necessary based on the nonadditivity of electrostatic and
for Activity but Correlates hydrophobic effects (Sect. 11.6).
with Toxicity: The Two In the case of highly hydrophilic peptides,
Thresholds modifications that increase hydrophobicity can
enhance the antimicrobial activity (first thresh-
The other main parameter influencing peptide old), without inducing strong toxicity (second
affinity for membranes is hydrophobicity. Several threshold). Malmsten and coworkers have
studies concur to support the view that two reported addition of hydrophobic oligopeptide
hydrophobicity thresholds exist (Dathe et al. stretches to the N- or C-terminus of the sequence
1997; Kondejewski et al. 1999, 2002; Stark et al. as a way to improve peptide activity and selec-
2002; Chen et al. 2007; Glukhov et al. 2008; tivity (Pasupuleti et al. 2009; Schmidtchen et al.
Mojsoska et al. 2015; Uggerhøj et al. 2015). A 2009, 2011, 2014). Comparison of different
first threshold hydrophobicity value must be hydrophobic modifications indicated that tag-
reached to obtain peptides with significant mem- ging by oligo-Trp sequences at the C-terminus
brane binding and insertion and thus endowed is the most effective one, leading to a substantial
with antimicrobial activity. However, if hydro- increase in activity, without significant enhance-
phobicity surpasses a second, higher threshold, ment of toxicity. Trp residues have peculiar
toxicity is observed, because binding to neutral properties, since they are known to have an
membranes becomes significant. The difference affinity for membrane interfaces, thanks to their
between these two thresholds is due to the elec- ability to interact both with hydrophobic moi-
trostatic contributions to peptide binding to bac- eties and with charged groups (through cation-
terial membranes. Therefore, an optimal range of aromatic interactions) (Yau et al. 1998). It has
hydrophobicity values exists, in which peptides been speculated that the specificity-enhancing
exhibit antimicrobial activity, but no significant effect of Trp might be linked to the difficulty of
toxicity. Above a third, even higher threshold, inserting such a bulky residue in the tightly
activity decreases, because of peptide aggrega- packed, cholesterol-­ containing membranes of
tion and lack of solubility (Gatto et al. 2006; eukaryotes (Pasupuleti et al. 2009; Schmidtchen
Chen et al. 2007; Chu-Kung et al. 2010; Wimley et al. 2009, 2011, 2014). However, preferential
2010a). It is difficult to provide quantitative val- interaction of Trp with cholesterol has also been
ues for these thresholds, since different hydro- hypothesized (de Kruijff 1990), although it is
phobicity scales are used in the literature. Just as disputed (Holt et al. 2008), and in some cases,
an example, Deber and coworkers identified val- introduction of Trp residues has been linked to
ues of 0.4 and approximately 2 in the Liu-Deber enhanced peptide toxicity (Oddo and Hansen
scale for the activity and toxicity thresholds, 2017; Matsuzaki et al. 1997). Another common
respectively, for the hydrophobicity of the core approach to increase the hydrophobicity of
segment of a series of model peptides (Glukhov highly hydrophilic peptides is lipidation (Gatto
et al. 2008). et al. 2006). Shai’s group demonstrated that
It is interesting to note that hydrophobicity highly polar peptides, originally devoid of anti-
affects binding to neutral membranes more than microbial activity, can become antimicrobial,
to charged bilayers and hemolysis more than but not toxic, after this modification (Avrahami
bactericidal activity (Wieprecht et al. 1997a; and Shai 2004; Malina and Shai 2005;
Dathe et al. 2002). The rationale underlying this Makovitzki et al. 2006, 2008). However, in
finding is not immediately obvious, since the other cases, lipidation led to strong toxicity
hydrophobic driving force is present for both (Chu-Kung et al. 2004; Laverty et al. 2010), or
membrane types, and therefore any variation in even to loss of activity, when it compromised
hydrophobicity should affect both antimicrobial peptide solubility (Toniolo et al. 1996; Gatto
activity and toxicity to the same extent. The et al. 2006; Chu-Kung et al. 2010). These find-
experimental observations can be explained ings highlight the fine-­tuning of AMP hydro-
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 193

phobicity needed for optimal activity and phobicity, also in this case, it is difficult to pro-
selectivity properties. vide a quantitative, generally valid value for this
threshold.
Another measure of the distribution of polar
11.7.3 Excessive Amphipathicity and hydrophobic residues is the angle subtended
Causes Toxicity by the polar face of the amphipathic helix, again
assuming an ideal conformation and looking
The total quantities of charged and hydrophobic along the helix axis (Uematsu and Matsuzaki
residues provide only a very rough measure of 2000). The available data on the role of this
peptide properties, since also their position in property in selectivity are limited, but a compre-
the sequence and structure are obviously impor- hensive study by Dathe and coworkers (2002)
tant. Amphipathicity measures the degree of provided some indications. As discussed above,
asymmetry in the distribution of polar and hydrophobicity and hydrophobic moment
hydrophobic residues. This property can be mostly affect the affinity for neutral membranes
quantified by the hydrophobic moment. This and thus the toxic activity. By contrast, in model
quantity is usually defined assuming an ideal membranes, the polar angle affects AMP ability
helical structure and summing the vectors indi- to perturb the bilayer, after membrane binding:
cating the position of each residue with respect in charged bilayers, peptide-induced membrane
to the helix axis, multiplied by their respective leakage decreases with increasing polar angle,
hydrophobicity values (in analogy with the defi- while it is essentially unaffected in neutral
nition of an electric dipole). To compare membranes (Dathe et al. 2002). However, the
sequences of different lengths, the mean hydro- effects of the polar angle in cellular assays of
phobic moment can be obtained by normalizing activity and toxicity are more limited (Dathe
for the number of amino acids (Eisenberg et al. et al. 2002).
1982; Phoenix and Harris 2002). Peptide
amphipathicity is a very important parameter
for determining the free energy of membrane 11.8 Conformational
binding (Fernández-Vidal et al. 2007). As early and Aggregation Equilibria
as 1981, De Grado demonstrated that amphipa- Play an Important Role
thicity is sufficient to induce lytic activity in a in Membrane Selectivity:
helical peptide (De Grado et al. 1981). The spe- The Concept of Effective
cific value of the hydrophobic moment becomes Hydrophobicity
particularly important for selectivity in an inter-
mediate range of hydrophobicity values, when All the considerations reported in the previous
the hydrophilic or hydrophobic components of section are based on hydrophobicity values deter-
the peptide do not predominate in determining mined from the peptide amino acid composition
its behavior (Dathe and Wieprecht 1999; Dathe and on amphipathicity calculated assuming an
et al. 2002). Similar to what has been reported ideal helical conformation. In addition, a mono-
for hydrophobicity, an increased hydrophobic meric peptide state is always considered.
moment affects the activity on neutral mem- However, as discussed in Sect. 11.6, peptides in
branes more than that on charged bilayers solution and in the membrane attain specific
(Wieprecht et al. 1997b; Dathe et al. 2002). ensembles of conformations, which can deviate
Increasing amphipathicity above a critical significantly from an ideal alpha-helix. In addi-
threshold results in strong interaction with neu- tion, amphipathic peptides have a strong ten-
tral membranes, leading to toxicity (Wieprecht dency to aggregate (Fig. 11.6). Conformational
et al. 1997b; Dathe and Wieprecht 1999; equilibria and self-assembly affect the degree to
Fernández-­Vidal et al. 2007; Kindrachuk and which the hydrophobic moieties of AMPs are
Napper 2010). As discussed above for hydro- exposed to the water phase and therefore modu-
194 S. Bobone and L. Stella

late the hydrophobic driving force for membrane et al. 2013; Wang et al. 2015). An increase in
binding (Bobone et al. 2013). Similarly, water-­ selectivity with a reduction in helical structure
membrane partition is affected by the peptide has been reported for other helix-breaking strate-
conformation, orientation, and depth of insertion gies, such as the insertion of d-amino acids (Shai
in the bilayer. Based on these considerations, in and Oren 1996, 2001; Oren and Shai 1997; Papo
our opinion, AMP selectivity is not determined et al. 2002; Chen et al. 2005; Zhu et al. 2007c;
by the “ideal” peptide hydrophobicity or Kaminski and Feix 2011; Nan et al. 2012; Huang
amphipathicity but by what we call “effective” et al. 2014) or peptoid residues (N-substituted
hydrophobicity and amphipathicity, i.e., the value glycines, which lack a H-bonding proton on the
these parameters assume in the actual conforma- N backbone atom and comprise a flexible main-­
tion and aggregation state attained by the peptide chain methylene group) (Song et al. 2005; Zhu
in solution and in the bilayer (Bobone et al. 2013; et al. 2007a, b; Kim et al. 2010). Incidentally,
Uggerhøj et al. 2015). these non-proteinogenic residues have the added
If peptide conformation and aggregation influ- advantage of reducing peptide susceptibility to
ence peptide hydrophobicity, the opposite is also proteolysis (Papo et al. 2002; Kim et al. 2010).
true: high hydrophobicity and amphipathicity The correlation between helicity and toxic-
values favor a stable secondary structure by ity was tentatively explained by proposing that
allowing the formation of intramolecular interac- a stable helical conformation enhances the pep-
tions between apolar residues (Fernández-Vidal tide propensity to aggregate (Kindrachuk and
et al. 2007). Peptide structure is influenced by Napper 2010; Vermeer et al. 2012): in helical
self-assembly processes, too (Sal-Man et al. amphipathic peptides, the hydrophobic face of
2002). Therefore, in order to fully understand the the helix is totally exposed to the aqueous
determinants of peptide selectivity, conforma- phase, and therefore aggregation is hydropho-
tional and self-assembly equilibria should be bically favored. Aggregation, in turn, would
considered. inhibit crossing of the LPS layer and cell wall
and thus access to the plasma membrane of bac-
teria (see below). However, helicity normally
11.8.1 Helicity Correlates affects toxicity, rather than antibacterial activ-
with Toxicity ity. In addition, in the peptides we investigated,
aggregation was significant only at concentra-
A correlation between peptide helicity and toxic- tions higher than the membrane-perturbing val-
ity has been reported in many studies (Tossi et al. ues, and therefore it was not relevant for activity
2000; Giangaspero et al. 2001; Zelezetsky et al. (Bobone et al. 2013). Probably a higher ten-
2005; Chen et al. 2005; Khandelia and Kaznessis dency to aggregate and an enhanced toxicity are
2006; Zhang et al. 2011; Mangoni et al. 2011; just two independent consequences of the
Chapuis et al. 2012; Bobone et al. 2013; Cherry hydrophobicity induced by a stable helical
et al. 2014). In addition, helix-destabilizing Gly structure, but the lack of selectivity in helical
or Pro residues are often present close to the cen- peptides is not caused by peptide aggregation
ter of the sequence of natural, selective AMPs (see also below). Based on a systematic study
that attain a helical conformation in membranes in which a central proline residue was moved
(Tossi et al. 2000, Bobone et al. 2013). These along the sequence or deleted, we obtained data
amino acids are important for peptide selectivity, supporting an alternative explanation (Bobone
since their deletion, substitution, or insertion sig- et al. 2013). In a perfectly helical, amphipathic
nificantly affects toxicity, through the perturba- structure, the apolar residues are completely
tion of the secondary structure (Thennarasu and exposed on the hydrophobic face of the helix.
Nagaraj 1996; Zhang et al. 1999; Shin et al. 2001; Even though short peptides are often unstruc-
Yang et al. 2002, 2006a, b; Lee et al. 2004, 2007; tured in water, helical conformations can be at
Song et al. 2004; Carotenuto et al. 2008; Bobone least partially populated, also thanks to the sta-
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 195

bilization due to the interaction between hydro- 11.8.2 Imperfect Amphipathicity

phobic side chains aligned along the helix (a Optimizes Selectivity
motif often called “leucine zipper” (Asthana
et al. 2004)). White and coworkers reported a Another idea related to effective hydrophobicity is
strong correlation between the amphiphilicity imperfect amphipathicity. Insertion of a polar/
of a peptide sequence in an ideal helical confor- charged residue in the hydrophobic face of an
mation and both the degree of helicity in solu- amphipathic helix has proven to be a reliable
tion and the affinity for neutral membranes method to increase AMP selectivity (Asthana et al.
(Fernández-Vidal et al. 2007). Destabilization 2004; Chen et al. 2005; Ahmad et al. 2006, 2009a,
of the helical conformation allows the peptide b; Hawrani et al. 2008; Pandey et al. 2010, 2011;
to fold onto itself, hiding the apolar side chains Jiang et al. 2011, 2014, 2018; Son et al. 2013;
from the water phase, and reducing the effec- Dalzini et al. 2016; Zhang et al. 2016). Hodges and
tive hydrophobicity of the peptide, and thus the coworkers even termed these misplaced polar resi-
driving force for binding neutral membranes dues “specificity determinants” (Jiang et al. 2011,
(Bobone et al. 2013; Büttner et al. 1992). 2014, 2018), even though exceptions to the selec-
Destabilization of the helix also increases the tivity-improving effect of this approach have been
entropic cost of membrane binding, since asso- reported (Wang et al. 2018). Interestingly, an
ciation to the bilayer is normally followed by imperfectly amphipathic structure is a common
peptide structuring (Zelezetsky et al. 2005). property of many natural, selective AMPs (Wimley
Interestingly, once membrane-bound, the helix- 2010b; Orioni et al. 2009). Again, this approach
destabilizing modifications do not preclude the reduces the hydrophobic driving force for binding
attainment of an amphipathic helical conforma- neutral membranes. On the other hand, antimicro-
tion (Bobone et al. 2013; Orioni et al. 2009; bial activity is usually not affected significantly by
Oren and Shai 2000). Therefore, variations in these changes: in charged bilayers, binding takes
the free energy of membrane binding are deter- place all the same, thanks to the electrostatic
mined essentially by changes in the solution attraction; once membrane-bound, the peptide is
conformation. Unstructured conformations able to attain a (possibly distorted) helical confor-
would inhibit binding to neutral membranes but mation, as demonstrated by spectroscopic and
would affect only marginally the affinity for simulative studies (Hawrani et al. 2008; Orioni
charged bilayers, leading to enhanced selectiv- et al. 2009). In the bilayer, imperfect amphipathic-
ity. Related to this interpretation is the concept ity will contribute to membrane disruption, by
of position-dependent hydrophobicity: hydro- driving some polar head-groups in the hydropho-
phobic residues in unstructured regions of the bic core of the membrane, as we observed for
peptide contribute to the effective hydrophobic- PMAP-23 (Orioni et al. 2009) (Fig. 11.7). This
ity and to toxicity less than those in helical seg- type of membrane activity has been termed “inter-
ments (Tachi et al. 2002). facial activity” by Wimley (2010b). Finally, it is
The effective, conformation-dependent hydro- worth mentioning that we recently observed an
phobicity can be calculated from peptide struc- effect of imperfect amphipathicity on toxicity also
tures (Gaillard et al. 1994), but it can also be in the case of peptidomimetic antimicrobial mole-
determined experimentally. Reversed-phase cules (Konai et al. 2018). Two small amphipathic,
chromatography retention times have resulted to cationic molecules were characterized by the same
be an accurate measure of the effective hydro- compositional hydrophobicity but had very differ-
phobicity of peptides (Krause et al. 1995; Zhou ent selectivity. By combining molecular dynamics
et al. 1990; Kim et al. 2005). Interestingly, strong simulations and RP-HPLC retention times, we
correlation between RP-HPLC retention times demonstrated that this was due to imperfect
and the hemolytic activity of AMPs has been amphipathicity and lower effective ­hydrophobicity
reported (Blondelle and Houghten 1991; of the selective analogue compared to the toxic
Kondejewski et al. 1999; Tachi et al. 2002). compound.
196 S. Bobone and L. Stella

Fig. 11.7 Interfacial activity of an imperfectly amphipa- rus atoms as yellow spheres. The peptide backbone is
thic AMP. Effects of the imperfectly amphipathic AMP shown in gray, charged side chains in red, polar amino
PMAP-23 on the structure of a lipid bilayer, as observed acids in orange, apolar residues in blue, and prolines in
in MD simulations. Two charged residues located on the green. The lipid composition was POPG/POPC (1:3 mol/
hydrophobic side of the helix drive three phospholipid mol).The bottom panel reports the density map of the lipid
head-groups and some water molecules into the hydro- phosphorus atoms (red) and of the peptide backbone
phobic core of the membrane. Water is represented in atoms (blue). Adapted, with permission, from research
cyan, phospholipids in gray, and phospholipids’ phospho- originally published in Orioni et al. 2009 © Elsevier

11.8.3 Effects of Peptide others form micellar structures (Liu et al. 2009;
Aggregation in the Aqueous Wang et al. 2010; Joshi et al. 2015; Lin and
Phase on Activity and Toxicity Grossfield 2015; Haney et al. 2017; Lei et al.
Are System Dependent 2018), fibrils (Tu et al. 2007; Chen et al. 2010;
Chen and Liang 2013; Shankar et al. 2013;
AMPs, due to their amphipathic nature, are sus- Chairatana and Nolan 2014; Ravi et al. 2015), or
ceptible to aggregation in water (Tian et al. 2015). even hydrogels (Veiga et al. 2012; McCloskey
Some peptides oligomerize through the interac- et al. 2014; Haney et al. 2017). Often the aggre-
tion of the apolar sides of their amphipathic heli- gates disassemble into monomers once
ces (Oren et al. 1999; Asthana et al. 2004; membrane-­bound (Ghosh et al. 1997). The criti-
Raimondo et al. 2005; Ahmad et al. 2006), while cal concentration for self-assembly can vary sig-
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 197

nificantly from one specific case to the other, also (Stella et al. 2004; Mazzuca et al. 2005; Gatto
depending on the experimental conditions and et al. 2006; Chen et al. 2007; Chu-­Kung et al.
particularly on salt concentration. 2010; Farrotti et al. 2017). Considering the vari-
The results of experimental and theoretical ous hydrophobicity thresholds discussed above,
studies on the effects of aggregation on peptide aggregation could therefore lead to a reduced tox-
activity and selectivity are extremely contradic- icity. At the same time, preassembly of AMPs
tory. Both negative (Feder et al. 2000; Kustanovich causes a local release of a high concentration at a
et al. 2002; Chen et al. 2006, 2007; Daschbach single site in the membrane, and this could cause
et al. 2012; Lin and Grossfield 2015; Farrotti higher activity (Ravi et al. 2015). In addition,
et al. 2017; Haney et al. 2017; Bagheri et al. computational studies suggested that self-assem-
2018; Zou et al. 2018) and positive (Sal-Man bly could lead to membrane selectivity also by
et al. 2002; Avrahami and Shai 2002; Liu et al. affecting the kinetics of membrane binding (Lin
2009; Chen et al. 2010; Joshi et al. 2015; Ravi and Grossfield 2015): binding to host mammalian
et al. 2015; Lei et al. 2018) correlations between membranes will be slow and inefficient as long as
aggregation and activity have been reported, as the lipopeptides are micellized in solution, while
well as lack of activity changes following aggre- binding to the bacterial surface will still be effi-
gation (Chen and Liang 2013). Similarly, some cient, thanks to electrostatic interactions and to
studies found that toxicity was not significantly the higher fluidity of the membrane. On the other
affected by aggregation (Lei et al. 2018), while hand, in cellular assays, the large size of the
others reported an increase (Chen and Liang aggregates, compared to monomers, could impair
2013; Lin and Grossfield 2015) or a decrease selectivity: preassembled AMPs might be unable
(Kustanovich et al. 2002; Chen et al. 2006; Chen to cross the LPS layer or the cell wall and thus to
et al. 2007) in selectivity upon self-assembly. reach the plasma membrane of bacteria. At the
Shankar et al. (2013) suggested that toxicity of same time, they would still be able to interact with
self-assembled lipopeptides depends on the spe- the “naked” membrane of host cells (Oren and
cific structure of the fibrillar aggregates. Shai 2000; Kustanovich et al. 2002; Sal-Man
The reported discrepancies are most likely due et al. 2002; Mangoni and Shai 2009).
to the fact that peptide aggregation is usually Discussing aggregation, it is important to note
controlled by varying the peptide properties or by that this phenomenon reduces susceptibility to
modulating electrostatic interactions by changing proteolytic degradation and affects the pharma-
the ionic strength of the solution. It is therefore cokinetics and pharmacodynamics in vivo
difficult to discriminate between the direct effects (Raimondo et al. 2005; Tu et al. 2007; Chen and
of these changes (e.g., an increase in hydropho- Liang 2013; Lei et al. 2018). It is also worth men-
bicity) and the consequence of the variations they tioning that human α-defensin 6 (HD6) has negli-
induce in aggregation. One approach to solve this gible direct killing activity but prevents infections
problem is covalent linking of the monomers by self-assembling into a network of fibrils that
(Sal-Man et al. 2002; Dempsey et al. 2003), but it capture pathogens and thus contrast microbial
does not exactly mimic self-assembly driven by invasion (Chairatana and Nolan 2014).
hydrophobic interactions.
Thermodynamic considerations on the inter-
connected equilibria involved in AMP activity 11.9 AMP Binding to Cells
indicate a possible positive role of aggregation in
enhancing peptide selectivity. Aggregation, which As discussed in the previous sections, AMP
is hydrophobically driven, reduces the effective selectivity is usually interpreted essentially on
peptide hydrophobicity by hiding the apolar moi- the basis of the different affinities observed in
eties in the molecule from the aqueous phase. As liposome studies for bilayers mimicking the
a consequence, the hydrophobic driving force for membranes of bacteria or eukaryotes. However,
membrane binding is reduced in the aggregates quite surprisingly, peptide affinities toward the
198 S. Bobone and L. Stella

two types of cells are largely uncharacterized. In more efficiently to bacteria than to mammalian
addition, if peptide activity is modulated by a cells, even though the latter are much bigger.
cell-binding equilibrium, it should depend on the Similarly, Ferro-Flores et al. (2003) reported
density of cells, but antimicrobial activity and that, in the presence of 2 · 107 cell/mL, an ubiqui-
toxicity assays are usually carried out using stan- cidin analogue was 35% bound in the case of
dardized, fixed cell densities, which are not nec- bacteria (S. aureus) while less than 4% in the
essarily representative of the cell concentrations case of human tumor cell lines LS174T and
present in a typical infection site (Savini et al. ACHN (which, again, are significantly bigger
2018). Finally, bactericidal and hemolytic activi- than bacteria). Comparable results have been
ties are routinely determined in separate assays, reported for two ubiquicidin analogues (approxi-
but when the two cell populations are present at mately 45–100% binding to S. aureus while only
the same time, they compete for peptide associa- 10% to leukocytes, in the presence of 2 · 105 cell/
tion. All these aspects have received limited mL), although in this case, selectivity was sur-
attention, until quite recently. Biophysical studies prisingly observed also for an anionic peptide
on model membranes allow the determination of used as negative control (Ebenhan et al. 2014b).
both membrane-binding and bilayer-perturbing Wimley and coworkers (Starr et al. 2016) mea-
activities, while microbiological studies usually sured the binding of the artificial AMP ARVA to
report activities only in terms of total peptide E. coli, S. aureus, and RBCs. In all cases, asso-
concentration. We recently reviewed the few ciation to bacteria was more favorable than to
studies that are trying to apply to cellular experi- RBCs. After accounting for the differences in
ments the same quantitative approaches normally cell size, the authors estimated that the affinity
used with model systems (Savini et al. 2018). for bacterial membranes was more than two
Here, only the aspects relevant to AMP selectiv- orders of magnitude higher than for erythro-
ity are summarized. cytes. Finally, an analogue of LL37 was reported
to bind E. coli, S. aureus, and M. smegmatis, but
not to hepatic cells, under conditions of compa-
11.9.1 A
 MPs Have a Higher Affinity rable cell numbers (Dutta et al. 2017). Overall,
for Bacterial Than for these data indicate that the differential affinity
Eukaryotic Cells routinely observed with model bilayers is pres-
ent also for the membranes of real cells.
Only a handful of studies reported data on AMP
binding to bacterial and eukaryotic cells. As
soon as 1988, Bruce Merrifield and his group 11.9.2 A
 ctivity and Toxicity Are Cell
(Steiner et al. 1988) measured binding of cecro- Density Dependent
pin A and some of its analogues to Escherichia
coli, B. megaterium, B. thuringiensis, and P. Another aspect that has remained essentially
aeruginosa cells and to erythrocytes. While uncharacterized until very recently is whether the
binding to the bacteria was significant (between activities of AMPs depend on the density of cells
70% and 80% for the natural peptide, under the present in the assays. Based on a partition equi-
conditions studied), no detectable association librium, the fraction of membrane-bound peptide
was observed for RBCs, at a cell density corre- obviously depends on the concentration of cells
sponding to a membrane area similar to that in the sample. Therefore, it is to be expected that
present in the experiments with bacteria. Welling MIC/MBC/MHC values depend on the concen-
et al. (2000) measured the binding of defensins tration of cells used in the assays. In broth dilu-
1–3, ubiquicidin, and human lactoferrin to bacte- tion assays of antimicrobial activity, the
ria and activated murine peritoneal leucocytes. recommended value for the initial cell density
In the presence of the same cell density (inoculum) is 5 · 105 cell/mL (Patel et al. 2012),
(2 · 107 cell/mL), the peptides bound 5–500 times which was selected for minimizing false-positive
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 199

and false-negative results in the clinical practice to form pores (Melo et al. 2009), it is obvious that
(Wiegand et al. 2008). However, bacterial cell more membrane-bound peptide molecules are
densities in clinically relevant infections range needed to kill a higher number of cells (Savini
from 1 to 109 cell/mL. Similarly, hemolytic activ- et al. 2017, 2018). Therefore, under conditions of
ity assays are normally performed with relatively high cell densities, where the peptide is
5 · 108 cell/mL, which is 1/10 of the cell density completely bound to the cells, a strong cell den-
in whole blood (Savini et al. 2018 and references sity dependence of the activity is expected. Data
therein). Matsuzaki (2009) pointed out that the by Jepson et al. (2016) and Snoussi et al. (2018)
cell densities in the two assays are very different, indicate that this effect might be due also to pep-
also considering that the membrane area of an tide sequestration by strong binding to killed bac-
erythrocyte is approximately ten times bigger terial cells. We tentatively explained the plateau
than that of a typical bacterium. Therefore, he in the low cell density regime as due to the cell-­
wondered if TI values such as those reported in binding equilibrium. At low cell densities, some
Table 11.1 are an experimental artifact due sim- of the peptide remains free in solution, and this
ply to the fact that more peptide is probably fraction increases with decreasing concentrations
needed to kill a higher number of bigger cells. of cells. As a consequence, the two effects (when
In the case of traditional antibiotics, it is well there are less cells to kill, also a lower fraction of
known that the MIC often depends on the size of peptide is cell-bound) cancel each other, leading
the bacterial inoculum (“inoculum effect”). By to a plateau in the total peptide concentration
contrast, in the case of AMPs, this possible needed in the sample to kill the bacteria (Savini
dependence has been investigated only in very et al. 2017, 2018).
few studies (Savini et al. 2018). In the 1990s, Interestingly, we observed a cell density
Levison et al. (1993) reported that the bacteri- dependence also for the hemolytic activity, with a
cidal activity of magainins against P. aeruginosa plateau at densities below 107 cell/mL (Savini
was inoculum dependent above 3 · 105 cell/mL, et al. 2017).
but it did not vary if the inoculum was reduced
below this value. Similarly, Jones et al. (1994)
observed an inoculum effect for lactoferricin B 11.9.3 C
 ompetition for Cell Binding
against E. coli, with a plateau at inoculum densi- in Co-culture Experiments
ties below 106 cell/mL. Ulrich and coworkers Might Be Regulated by Kinetic
measured MIC values for gramicidin S and PGLa Phenomena
at two cell density values and observed a cell
density dependence (Hartmann et al. 2010). More Since both antimicrobial activity and toxicity
recently, we measured the MBC values for a fluo- seem to depend on the density of cells, the effec-
rescent analogue of PMAP-23 in the presence of tive selectivity also depends on the value of this
different E. coli cell densities (Savini et al. 2017). parameter used in the two assays (MIC and
Also in our case, the MBC increased with inocu- MHC). The quantity of peptide that binds to a
lum size but reached a plateau at densities of type of cell or to the other is determined by the
5 · 106 cell/mL and below. A similar trend was respective affinities but also by the concentration
reported by Poon and coworkers for pexiganan of cells of each type. In the end, the peptide is
(Jepson et al. 2016). Finally, a recent study active and/or toxic if the respective threshold of
reported an inoculum effect also for LL-37 bound peptide needed for membrane perturbation
(Snoussi et al. 2018). is reached. In principle, under conditions where
Overall, these studies show that AMP activity the host cells are in large excess (as in the case of
is strongly dependent on the density of cells in systemic treatment of an infection), lack of
the assay, with a linear (Savini et al. 2017) or sub- ­toxicity is expected even if the affinities for the
linear (Jepson et al. 2016) trend. If a threshold two cell types are similar (Matsuzaki 2009;
concentration must be reached in the membrane Savini et al. 2017).
200 S. Bobone and L. Stella

A limited number of studies have tested pep- tion for peptide binding should take place: the
tide activity and toxicity in assays where both antimicrobial activity and/or the toxicity of the
bacteria and mammalian cells were present. Mor peptides should be inhibited by sequestration of a
and coworkers showed that AMPs bound to fraction of the peptide molecules due to binding
RBCs are able to transfer to microbial cells, to the other cell population. Very surprisingly, we
exerting their activity (a phenomenon they termed observed that this is not the case (Savini et al.
“affinity-driven molecular transfer”) (Feder et al. 2017). We measured both bacterial killing and
2001). Derivatives of lentivirus lytic peptides hemolysis for analogues of PMAP-23 and escu-
killed P. aeruginosa bacteria interacting with cul- lentin, in a mixed population of E. coli and eryth-
tured human airway epithelial cells, at peptide rocytes, and compared these results with the
concentrations that only moderately affected the traditional assays performed on the two cell types
cell monolayer (Phadke et al. 2003). Fluorescence separately (Fig. 11.8). The activities on bacteria
microscopy images showed that in a co-culture of and on RBCs were essentially unaffected by the
S. aureus and human cells (endothelial cells or presence of the other cell population, in contradic-
neutrophils), AMPs concentrate on the bacterial tion with the predictions based on binding equi-
cells (Matsuzaki 2009; Akram et al. 2015) libria. Data from Wimley’s lab provide further
(Fig. 11.2a). Chen and Liang (2013) showed that support to the conclusion that out of equilibrium,
the artificial AMP CL-1 is able to kill selectively kinetic phenomena are at play when two cell pop-
S. aureus in co-culture with human cells, and ulations are present at the same time. They showed
Malmsten and coworkers reported activity with- that the results of such experiments depend on the
out hemoliticity in bacteria-supplemented blood order of addition of the different components: like
for an engineered AMP (Schmidtchen et al. in our case, no change in the MIC was observed
2011). A striking evidence of selectivity is pro- when AMPs were added to a mixture of bacteria
vided by the fact that some AMPs (e.g., LL-37) and RBCs or to bacteria alone. However, the anti-
are able to kill S. aureus bacteria internalized into microbial activity was significantly inhibited
mammalian cells (Noore et al. 2012). Selectivity when the peptide was incubated with RBCs first,
in co-culture has been reported also for AMP-­ and then both were added to the bacterial culture
inspired systems, such as peptidomimetics, cat- (Starr et al. 2016) (Fig. 11.8). Actually, while
ionic peptidopolysaccharides, and peptide equilibrium processes can be invoked before the
hydrogels (Salick et al. 2007; Li et al. 2012; cell membranes are perturbed, it is easy to realize
Konai et al. 2018). Some co-culture data have that this approach is too simplistic in the case of
been reported also for antifungal, antiprotozoan, bilayer disruption, which allows access to multi-
and anticancer activities. For instance, an ana- ple additional binding targets (e.g., inside the cell)
logue of the antifungal peptide PAF26 concen- (Snoussi et al. 2018).
trated on fungal cells in co-culture with human Overall, quantitative measurements of pep-
lung epithelial cells (Mendive-Tapia et al. 2016); tide interactions with cells confirmed that AMPs
an artificial anticancer peptide concentrated in have a higher affinity for bacterial than for host
cancer cells, co-cultured with primary cells cells. Experiments performed with varying cell
(Chen et al. 2014). Some AMPs (e.g., derma- densities indicated that both activity and toxicity
septins or NK-lysin) are able to kill protozoan depend on this parameter (even though a plateau
parasites such as P. falciparum or T. cruzi inside is observed at low cell densities) and that there-
human cells, disrupting the plasma membrane of fore the measured selectivity depends on the spe-
the intracellular parasites without harming that of cific conditions of the experiments. Finally,
the host (Ghosh et al. 1997; Krugliak et al. 2000; experiments with mixed bacterial and eukaryotic
Jacobs et al. 2003; Gelhaus et al. 2008). cells showed that, contrary to expectations based
Based on association equilibria, when the two on equilibrium considerations, competition for
cell types are present at the same time, competi- peptide binding does not lead to a loss in activity.
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 201

Fig. 11.8 Antimicrobial and hemolytic activity in represents measurements done in the absence of RBCs.
assays with both bacterial and erythrocytes. Left panel, All other experiments include 109 human RBC/mL. Time
bactericidal (blue) and hemolytic (red) activities of the zero represents the experiments in which RBC and bac-
AMP DNS-PMAP23 in the presence of both bacteria teria were first mixed, followed by peptide addition, i.e.,
and erythrocytes (squares) or of one cell type only (cir- no preincubation with either cell type. Negative times
cles). Both activities are only slightly affected by the represent peptide preincubation with bacteria before the
presence of the other cell population. 4.5 × 107 E. coli addition of RBCs. Positive times represent peptide pre-
cells/mL, 4.5 × 108 RBCs/mL. (Reproduced with per- incubation with RBCs, followed by addition of bacteria.
mission from Savini et al. 2017 https://pubs.acs.org/doi/ Points plotted at 20 μM had MIC values
abs/10.1021/acschembio.6b00910 (Copyright 2017, ≥20 μM. Significant inhibition of peptide antimicrobial
American Chemical Society)). Further permissions activity due to the presence of RBCs was observed only
related to the material excerpted should be directed to in the case of preincubation with erythrocytes.
the ACS. Right panel: effect of incubation time on MIC Proteolytic degradation effects can be ruled out, since
values in assays with both bacteria and erythrocytes. The ARVA-D is a peptide comprising all d-amino acids.
artificial AMP ARVA-D was tested against E. coli (red) (Adapted with permission from Starr 2016 (Copyright
and S. aureus (blue) under various conditions. “MIC” 2016, American Chemical Society))

This finding definitely warrants further co-­ side of the peptide structure, leads to a better
culture studies. selectivity. However, an excessive increase in
positive charge, above a threshold that depends
on each specific case, might be ineffective or
11.10 Concluding Remarks even detrimental.
• Reducing hydrophobicity, amphipathicity, and
A large body of studies has been devoted to char- helicity is an effective strategy. These proper-
acterize, understand, and improve the selectivity ties are necessary for activity, since they are
of AMPs. These data support the view that selec- responsible for the hydrophobic driving force
tivity arises due to the different lipid composition for membrane binding and for insertion in the
of bacterial and host cell membranes. AMPs are bilayer. However, several studies have demon-
able to discriminate between the two types of strated that they affect toxicity more than
bilayers, thanks to their physicochemical proper- activity. This finding is probably a conse-
ties. We can summarize here the main guidelines quence of the nonadditivity of electrostatic
for optimization of peptide selectivity: and hydrophobic effects. Therefore, an
intermediate range of hydrophobicity and
­
• Increasing the cationic charge, by C-terminal amphipathicity values optimizes selectivity.
amidation, substitution of anionic residues, or • The real determinants of selectivity are the
insertion of cationic amino acids in the polar effective, conformation-dependent, hydropho-
202 S. Bobone and L. Stella

bicity and amphipathicity values (rather than References

the parameters determined based on an ideal-
ized conformation). They can be optimized by Abraham T, Lewis RN, Hodges RS, McElhaney RN
(2005) Isothermal titration calorimetry studies of
introducing: the binding of the antimicrobial peptide gramicidin
–– Polar residues in the hydrophobic side of the S to phospholipid bilayer membranes. Biochemistry
peptide helix (imperfect amphipathicity) 44(33):11279–11285
–– Helix-breaking residues, such as Pro, Gly, Agrawal A, Weisshaar JC (2018) Effects of alterations of
the E. coli lipopolysaccharide layer on membrane per-
d-amino acids, or peptoids meabilization events induced by Cecropin A. Biochim
• Aggregation in the aqueous phase modulates Biophys Acta 1860(7):1470–1479
selectivity, too. Ahmad A, Yadav SP, Asthana N, Mitra K, Srivastava
SP, Ghosh JK (2006) Utilization of an amphipa-
thic leucine zipper sequence to design antibacte-
In this chapter, we tried to generalize and sim- rial peptides with simultaneous modulation of toxic
plify as much as possible the results collected activity against human red blood cells. J Biol Chem
over many years of studies. However, the readers 281(31):22029–22038
that had the patience to follow us until here must Ahmad A, Asthana N, Azmi S, Srivastava RM, Pandey
BK, Yadav V, Ghosh JK (2009a) Structure–
have realized that the literature on AMP selectiv- function study of cathelicidin-derived bovine
ity is full of exceptions and contradictions. antimicrobial peptide BMAP-28: design of its cell-
Peptide association to bacterial and host cells is selective analogs by amino acid substitutions in the
modulated (in a nonadditive way) by electrostatic heptad repeat sequences. Biochim Biophys Acta
1788(11):2411–2420
and hydrophobic interactions. The effective pep- Ahmad A, Azmi S, Srivastava RM, Srivastava S, Pandey
tide hydrophobicity is determined by peptide BK, Saxena R, Bajpai VK, Ghosh JK (2009b) Design
conformation and aggregation state. Therefore, of nontoxic analogues of cathelicidin-derived bovine
AMP selectivity is finely regulated by intercon- antimicrobial peptide BMAP-27: the role of leucine as
well as phenylalanine zipper sequences in determining
nected binding, aggregation and conformational its toxicity. Biochemistry 48(46):10905–10917
equilibria. Any variation in peptide property will Akhtar MS, Qaisar A, Irfanullah J, Iqbal J (2005)
affect all of these phenomena. Therefore, in order Antimicrobial peptide 99mTc-ubiquicidin 29-41 as
to predict, or at least to understand, the effect of human infection-imaging agent: clinical trial. J Nucl
Med 46(4):567–573
peptide modifications on the final selectivity, all Akhtar MS, Imran MB, Nadeem MA, Shahid A (2012)
the possible processes involved in peptide behav- Antimicrobial peptides as infection imaging agents:
ior in the aqueous and membrane phases must be better than radiolabeled antibiotics. Int J Pept. https://
considered. doi.org/10.1155/2012/965238
Akram AR, Avlonitis N, Lilienkampf A, Perez-Lopez
We still do not fully understand what happens AM, McDonald N, Chankeshwara SV, Scholefield E,
when AMPs act on bacteria and human cells Haslett C, Bradley M, Dhaliwal K (2015) A labelled-­
together. Recent data indicate that in this case ubiquicidin antimicrobial peptide for immediate in
even the complex scenario outlined above is situ optical detection of live bacteria in human alveolar
lung tissue. Chem Sci 6(12):6971–6979
overly simplified, since kinetic phenomena prob- Alba A, López-Abarrategui C, Otero-González AJ (2012)
ably have to be taken into account. Further stud- Host defense peptides: an alternative as antiinfective
ies in this area are definitely warranted and hold and immunomodulatory therapeutics. Biopolymers
the promise to provide a better understanding of 98(4):251–267
Allende D, Simon SA, McIntosh TJ (2005) Melittin-­induced
AMP selectivity. bilayer leakage depends on lipid material properties: evi-
dence for toroidal pores. Biophys J 88(3):1828–1837
Acknowledgments The authors gratefully thank Dr. Aloia RC, Tian H, Jensen FC (1993) Lipid composition
F. Savini and Dr. A. Papi for their help with Figs. 11.3 and and fluidity of the human immunodeficiency virus
11.5. Research in our lab is currently supported by the envelope and host cell plasma membranes. Proc Natl
Italian Ministry for Education, University and Research Acad Sci U S A 90(11):5181–5185
(grant PRIN 20157WW5EH_007), and by the Italian Ames GF (1968) Lipids of Salmonella typhimurium and
Association for Cancer Research (AIRC grant IG 2016 Escherichia coli: structure and metabolism. J Bacteriol
19171). 95(3):833–843
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 203

Andra J, Goldmann T, Ernst CM, Peschel A, Gutsmann as a potential infection imaging agent. Nucl Med Biol
T (2011) Multiple peptide resistance factor (MprF)- 62–63:47–53
mediated resistance of Staphylococcus aureus against Bishop DG, Rutberg L, Samuelsson B (1967) The chemi-
antimicrobial peptides coincides with a modulated cal composition of the cytoplasmic membrane of
peptide interaction with artificial membranes com- Bacillus subtilis. Eur J Biochem 2(4):448–453
prising lysyl-phosphatidylglycerol. J Biol Chem Bland JM, De Lucca AJ, Jacks TJ, Vigo CB (2001) All-­
286(21):18692–18700 D-­cecropin B: synthesis, conformation, lipopolysac-
Arcidiacono S, Soares JW, Meehan AM, Marek P, Kirby charide binding, and antibacterial activity. Mol Cell
R (2009) Membrane permeability and antimicrobial Biochem 218(1–2):105–111
kinetics of cecropin P1 against Escherichia coli. J Pept Blondelle SE, Houghten RA (1991) Hemolytic and
Sci 15(6):398–403 antimicrobial activities of the twenty-four indi-
Assadi M, Vahdat K, Nabipour I, Sehhat MR, Hadavand vidual omission analogs of melittin. Biochemistry
F, Javadi H, Tavakoli A, Saberifard J, Kalantarhormozi 30(19):4671–4678
MR, Zakani A, Eftekhari M (2011) Diagnostic value Bobone S, Piazzon A, Orioni B, Pedersen JZ, Nan YH,
of 99mTc-ubiquicidin scintigraphy for osteomyelitis Hahm KS, Shin SH, Stella L (2011) The thin line
and comparisons with 99mTc-methylene diphospho- between cell-penetrating and antimicrobial pep-
nate scintigraphy and magnetic resonance imaging. tides: the case of Pep-1 and Pep-1-K. J Pept Sci
Nucl Med Commun 32(8):716–723 17(5):335–341
Asthana N, Yadav SP, Ghosh JK (2004) Dissection of Bobone S, Roversi D, Giordano L, De Zotti M, Formaggio
antibacterial and toxic activity of Melittin: a leucine F, Toniolo C, Park Y, Stella L (2012) The lipid depen-
zipper motif plays a crucial role in determining its dence of antimicrobial peptide activity is an unre-
hemolytic activity but not antibacterial activity. J Biol liable experimental test for different pore models.
Chem 279(53):55042–55050 Biochemistry 51(51):10124–10126
Avrahami D, Shai Y (2002) Conjugation of a magainin Bobone S, Bocchinfuso G, Park Y, Palleschi A, Hahm
analogue with lipophilic acids controls hydro- KS, Stella L (2013) The importance of being
phobicity, solution assembly, and cell selectivity. kinked: role of Pro residues in the selectivity of
Biochemistry 41(7):2254–2263 the helical antimicrobial peptide P5. J Pept Sci
Avrahami D, Shai Y (2004) A new group of antifungal and 19(12):758–769
antibacterial lipopeptides derived from non-membrane Bocchinfuso G, Palleschi A, Orioni B, Grande G,
active peptides conjugated to palmitic acid. J Biol Formaggio F, Toniolo C et al (2009) Different mecha-
Chem 279(13):12277–12285 nisms of action of antimicrobial peptides: insights
Bacalum M, Radu M (2015) Cationic antimicrobial pep- from fluorescence spectroscopy experiments and
tides cytotoxicity on mammalian cells: an analysis molecular dynamics simulations. J Pept Sci Off Publ
using therapeutic index integrative concept. Int J Pept Eur Pept Soc 15(9):550–558
Res Ther 21:47–55 Bocchinfuso G, Bobone S, Mazzuca C, Palleschi A, Stella
Bagheri M, Amininasab M, Dathe M (2018) Arg/ L (2011) Fluorescence spectroscopy and molecular
Trp-­rich cyclic α/β-antimicrobial peptides: the dynamics simulations in studies on the mechanism of
roles of H-bonding and hydrophobic/hydro- membrane destabilization by antimicrobial peptides.
philic solvent accessible surface areas upon the Cell Mol Life Sci 68(13):2281–2301
activity and membrane selectivity. Chem Eur Boyd KJ, Alder NN, May ER (2017) Buckling under
J. 24(53):14242–14253 pressure: curvature-based lipid segregation and sta-
Ballas SK, Krasnow SH (1980) Structure of erythrocyte bility modulation in cardiolipin-containing bilayers.
membrane and its transport functions. Ann Clin Lab Langmuir 33(27):6937–6946
Sci 10(3):209–219 Braun S, Pokorná S, Šachl R, Hof M, Heerklotz H,
Beschiaschvili G, Seelig J (1990) Melittin binding to Hoernke M (2017) Biomembrane permeabiliza-
mixed phosphatidylglycerol/phosphatidylcholine tion: statistics of individual leakage events harmo-
membranes. Biochemistry 29(1):52–58 nize the interpretation of vesicle leakage. ACS Nano
Bessalle R, Kapitkovsky A, Gorea A, Shalit I, Fridkin 12(1):813–819
M (1990) All-D-magainin: chirality, antimicro- Brender JR, McHenry AJ, Ramamoorthy A (2012) Does
bial activity and proteolytic resistance. FEBS Lett cholesterol play a role in the bacterial selectivity of
274(1–2):151–155 antimicrobial peptides? Front Immunol 3:195
Bessalle R, Haas H, Goria A, Shalit I, Fridkin M (1992) Broekhuyse RM (1969) Quantitative two-dimensional
Augmentation of the antibacterial activity of magainin thin-layer chromatography of blood phospholipids.
by positive-charge chain extension. Antimicrob Clin Chim Acta 23(3):457–461
Agents Chemother 36(2):313–317 Brouwer CP, Bogaards SJ, Wulferink M, Velders MP,
Bhatt J, Mukherjee A, Shinto A, Karuppusamy KK, Korde Welling MM (2006) Synthetic peptides derived from
A, Kumar M, Sarma HD, Repaka K, Dash A (2018) human antimicrobial peptide ubiquicidin accumu-
Gallium-68 labeled Ubiquicidin derived octapeptide late at sites of infections and eradicate (multi-drug
204 S. Bobone and L. Stella

resistant) Staphylococcus aureus in mice. Peptides designer peptides: a link between propensity for nano-
27(11):2585–2591 structuring and capacity for membrane destabilization.
Brouwer CPJM, Sarda-Mantel L, Meulemans A, Guludec Biomacromolecules 11(2):402–411
DL, Welling MM (2008) The use of technetium-­ Chen C, Hu J, Zeng P, Pan F, Yaseen M, Xu H, Lu JR
99m radiolabeled human antimicrobial peptides for (2014) Molecular mechanisms of anticancer action
infection specific imaging. Mini-Rev Med Chem and cell selectivity of short α-helical peptides.
8(10):1039–1052 Biomaterials 35(5):1552–1561
Bulet P, Stocklin R (2005) Insect antimicrobial peptides: Chen H, Liu C, Chen D, Madrid K, Peng S, Dong X,
structures, properties and gene regulation. Protein Zhang M, Gu Y (2015) Bacteria-targeting conjugates
Pept Lett 12(1):3–11 based on antimicrobial peptide for bacteria diagnosis
Bütikofer P, Lin ZW, Chiu DT, Lubin B, Kuypers FA and therapy. Mol Pharm 12(7):2505–2516
(1990) Transbilayer distribution and mobility of phos- Cherry MA, Higgins SK, Melroy H, Lee HS, Pokorny A
phatidylinositol in human red blood cells. J Biol Chem (2014) Peptides with the same composition, hydro-
265(27):16035–16038 phobicity, and hydrophobic moment bind to phospho-
Büttner K, Blondelle SE, Ostresh JM, Houghten RA lipid bilayers with different affinities. J Phys Chem B
(1992) Perturbation of peptide conformations 118(43):12462–12470
induced in anisotropic environments. Biopolymers Chu-Kung AF, Bozzelli KN, Lockwood NA, Haseman
32(6):575–583 JR, Mayo KH, Tirrell MV (2004) Promotion of pep-
Carneiro VA, Duarte HS, Prado MGV, Silva ML, Teixeira tide antimicrobial activity by fatty acid conjugation.
M, dos Santos YM, Vasconcelos IB, Cunha RMS Bioconjug Chem 15(3):530–535
(2015) Antimicrobial peptides: from synthesis to Chu-Kung AF, Nguyen R, Bozzelli KN, Tirrell M (2010)
clinical perspectives. In: The battle against microbial Chain length dependence of antimicrobial peptide-­
pathogens: basic science, technological advances and fatty acid conjugate activity. J Colloid Interface Sci
educational programs, 1st edn. Formatex Research 345(2):160–167
Center, Spain, pp 81–90 Cornut I, Büttner K, Dasseux JL, Dufourcq J (1994) The
Carotenuto A, Malfi S, Saviello MR, Campiglia P, Gomez-­ amphipathic α-helix concept: application to the de
Monterrey I, Mangoni ML, Marcellini Hercolani novo design of ideally amphipathic Leu, Lys peptides
Gaddi L, Novellino E, Grieco P (2008) A different with hemolytic activity higher than that of melittin.
molecular mechanism underlying antimicrobial and FEBS Lett 349(1):29–33
hemolytic actions of temporins A and L. J Med Chem Dalzini A, Bergamini C, Biondi B, De Zotti M, Panighel
51(8):2354–2362 G, Fato R, Peggion C, Bortolus M, Maniero AL (2016)
Chairatana P, Nolan EM (2014) Molecular basis for The rational search for selective anticancer derivatives
self-assembly of a human host-defense peptide of the peptide trichogin GA IV: a multi-technique bio-
that entraps bacterial pathogens. J Am Chem Soc physical approach. Sci Rep 6:24000
136(38):13267–13276 Daschbach MM, Negin S, You L, Walsh M, Gokel GW
Chapuis H, Slaninová J, Bednárová L, Monincová L, (2012) Aggregation and supramolecular mem-
Buděšínský M, Čeřovský V (2012) Effect of hydrocar- brane interactions that influence anion transport in
bon stapling on the properties of α-helical antimicro- tryptophan-­containing synthetic peptides. Chem Eur
bial peptides isolated from the venom of hymenoptera. J 18(24):7608–7623
Amino Acids 43(5):2047–2058 Dathe M, Wieprecht T (1999) Structural features of heli-
Chen L, Liang JF (2013) Peptide fibrils with altered sta- cal antimicrobial peptides: their potential to modulate
bility, activity, and cell selectivity. Biomacromolecules activity on model membranes and biological cells.
14(7):2326–2331 Biochim Biophys Acta 1462(1):71–87
Chen Y, Mant CT, Farmer SW, Hancock REW, Michael Dathe M, Schümann M, Wieprecht T, Winkler A,
L, Vasil ML, Hodges RS (2005) Rational design of Beyermann M, Krause E, Matsuzatki K, Murase
alpha-helical antimicrobial peptides with enhanced O, Bienert M (1996) Peptide helicity and mem-
activities and specificity/therapeutic index. J Biol brane surface charge modulate the balance of elec-
Chem 280:12316–12329 trostatic and hydrophobic interactions with lipid
Chen Y, Vasil AI, Rehaume L, Mant CT, Burns JL, Vasil bilayers and biological membranes. Biochemistry
ML, Hancock RE, Hodges RS (2006) Comparison of 35(38):12612–12622
biophysical and biologic properties of α-helical enan- Dathe M, Wieprecht T, Nikolenko H, Handel L, Maloy
tiomeric antimicrobial peptides. Chem Biol Drug Des WL, MacDonald DL, Beyermann M, Bienert M
67(2):162–173 (1997) Hydrophobicity, hydrophobic moment and
Chen Y, Guarnieri MT, Vasil AI, Vasil ML, Mant CT, angle subtended by charged residues modulate anti-
Hodges RS (2007) Role of peptide hydrophobic- bacterial and haemolytic activity of amphipathic heli-
ity in the mechanism of action of α-helical antimi- cal peptides. FEBS Lett 403(2):208–212
crobial peptides. Antimicrob Agents Chemother Dathe M, Nikolenko H, Meyer J, Beyermann M, Bienert
51(4):1398–1406 M (2001) Optimization of the antimicrobial activity of
Chen C, Pan F, Zhang S, Hu J, Cao M, Wang J, Xu H, magainin peptides by modification of charge. FEBS
Zhao X, Lu JR (2010) Antibacterial activities of short Lett 501(2–3):146–150
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 205

Dathe M, Meyer J, Beyermann M, Maul B, Hoischen C, tion and 68Ga-radiolabeling of NOTA-conjugated

Bienert M (2002) General aspects of peptide selectivity ubiquicidin fragments for prospective infection imag-
towards lipid bilayers and cell membranes studied by vari- ing with PET/CT. Nucl Med Biol 41(5):390–400
ation of the structural parameters of amphipathic helical Ebenhan T, Sathekge M, Lenagana T, Koole M, Gheysens
model peptides. Biochim Biophys Acta 1558(2):171–186 O, Govender T, Zeevaart JR (2018) 68Ga-NOTA-­
Dawson RM, Liu CQ (2011) Analogues of peptide SMAP-­ functionalized ubiquicidin: cytotoxicity, biodistribu-
29 with comparable antimicrobial potency and reduced tion, radiation dosimetry and first-in-human positron
toxicity. Int J Antimicrob Agents 37(5):432–437 emission tomography/computed tomography imaging
Dawson RM, Fox MA, Atkins HS, Liu CQ (2011) Potent of infections. J Nucl Med 59(2):334–339
antimicrobial peptides with selectivity for Bacillus Eckert R (2011) Road to clinical efficacy: challenges and
anthracis over human erythrocytes. Int J Antimicrob novel strategies for antimicrobial peptide develop-
Agents 38(3):237–242 ment. Future Microbiol 6(6):635–651
de Kruijff B (1990) Cholesterol as a target for toxins. Eisenberg D, Weiss RM, Terwilliger TC (1982) The heli-
Biosci Rep 10(2):127–130 cal hydrophobic moment: a measure of the amphiphi-
de la Fuente-Núñez C, Reffuveille F, Mansour SC, licity of a helix. Nature 299(5881):371–374
Reckseidler-Zenteno SL, Hernández D, Brackman G, Epand RF, Schmitt MA, Gellman SH, Epand RM (2006)
Coenye T, Hancock REW (2015) D-enantiomeric pep- Role of membrane lipids in the mechanism of bacte-
tides that eradicate wild-type and multidrug-resistant rial species selective toxicity by two α/β-antimicrobial
biofilms and protect against lethal Pseudomonas aeru- peptides. Biochim Biophys Acta 1758(9):1343–1350
ginosa infections. Chem Biol 22(2):196–205 Farrotti A, Bocchinfuso G, Palleschi A, Rosato N,
de Murphy CA, Gemmel F, Balter J (2010) Clinical trial Salnikov ES, Voievoda N, Bechinger B, Stella L
of specific imaging of infections. Nucl Med Commun (2015) Molecular dynamics methods to predict pep-
31(8):726–733 tide locations in membranes: LAH4 as a stringent test
Dean SN, Bishop BM, Van Hoek ML (2011) Susceptibility case. Biochim Biophys Acta 1848(2):581–592
of Pseudomonas aeruginosa biofilm to alpha-helical pep- Farrotti A, Conflitti P, Srivastava S, Ghosh JK, Palleschi
tides: D-enantiomer of LL-37. Front Microbiol 2:128 A, Stella L, Bocchinfuso G (2017) Molecular dynam-
DeGrado WF, Kezdy FJ, Kaiser ET (1981) Design, syn- ics simulations of the host defense peptide tempo-
thesis, and characterization of a cytotoxic peptide with rin L and its Q3K derivative: an atomic level view
melittin-like activity. J Am Chem Soc 103(3):679–681 from aggregation in water to bilayer perturbation.
Dempsey CE, Ueno S, Avison MB (2003) Enhanced Molecules 22(7):1235
membrane permeabilization and antibacterial activ- Feder R, Dagan A, Mor A (2000) Structure-activity rela-
ity of a disulfide-dimerized magainin analogue. tionship study of antimicrobial dermaseptin S4 show-
Biochemistry 42(2):402–409 ing the consequences of peptide oligomerization on
Dennison SR, Harris F, Bhatt T, Singh J, Phoenix DA selective cytotoxicity. J Biol Chem 275(6):4230–4238
(2009) The effect of C-terminal amidation on the effi- Feder R, Nehushtai R, Mor A (2001) Affinity driven
cacy and selectivity of antimicrobial and anticancer molecular transfer from erythrocyte membrane to tar-
peptides. Mol Cell Biochem 332(1–2):43–50 get cells. Peptides 22(10):1683–1690
Deslouches B, Hasek ML, Craigo JK, Steckbeck JD, Fernández-Vidal M, Jayasinghe S, Ladokhin AS, White
Montelar RC (2016) Comparative functional proper- SH (2007) Folding amphipathic helices into mem-
ties of engineered cationic antimicrobial peptides con- branes: amphiphilicity trumps hydrophobicity. J Mol
sisting exclusively of tryptophan and either lysine or Biol 370(3):459–470
arginine. J Med Microbiol 65(6):554–565 Ferro-Flores G, de Murphy CA, Pedraza-López M,
Dodge JT, Phillips GB (1967) Composition of phospho- Meléndez-Alafort L, Zhang YM, Rusckowski M,
lipids and of phospholipid fatty acids and aldehydes in Hnatowich DJ (2003) In vitro and in vivo assessment
human red cells. J Lipid Res 8(6):667–675 of 99mTc-UBI specificity for bacteria. Nucl Med Biol
Dutta J, Baijnath S, Somboro AM, Nagiah S, Albericio 30(6):597–603
F, de la Torre BG, Marjanovic-Painter B, Zeevaart Findlay EG, Currie SM, Davidson DJ (2013) Cationic
JR, Sathekge M, Kruger HG, Chuturgoon A, Naicker host defence peptides: potential as antiviral therapeu-
T, Ebenhan T, Govender T (2017) Synthesis, in vitro tics. BioDrugs 27(5):479–493
evaluation, and 68Ga-radiolabeling of CDP1 toward Friedrich CL, Moyles D, Beveridge TJ, Hancock RE
PET/CT imaging of bacterial infection. Chem Biol (2000) Antibacterial action of structurally diverse cat-
Drug Des 90(4):572–579 ionic peptides on gram-positive bacteria. Antimicrob
Ebenhan T, Gheysens O, Kruger HG, Zeevaart Agents Chemother 44(8):2086–2092
JR, Sathekge MM (2014a) Antimicrobial pep- Gaillard P, Carrupt PA, Testa B, Boudon A (1994) Molecular
tides: their role as infection-selective tracers for lipophilicity potential, a tool in 3D QSAR: method and
molecular imaging. Biomed Res Int. https://doi. applications. J Comput Aided Mol Des 8(2):83–96
org/10.1155/2014/867381 Gandomkar M, Najafi R, Shafiei M, Mazidi M, Goudarzi
Ebenhan T, Chadwick N, Sathekge MM, Govender P, M, Mirfallah SH, Ebrahimi F, Heydarpor HR, Abdie
Govender T, Kruger HG, Marjanovic-Painter B, N (2009) Clinical evaluation of antimicrobial pep-
Zeevaart JR (2014b) Peptide synthesis, characteriza- tide [99mTc/Tricine/HYNIC0] ubiquicidin 29–41 as
206 S. Bobone and L. Stella

a human-specific infection imaging agent. Nucl Med determines pardaxin’s mechanism of lipid bilayer dis-
Biol 36(2):199–205 ruption. Biophys J 83(2):1004–1013
Gascard P, Tran D, Sauvage M, Sulpice JC, Fukami K, Hancock RE, Sahl HG (2006) Antimicrobial and host-­
Takenawa T, Claret M, Giraud F (1991) Asymmetric defense peptides as new anti-infective therapeutic
distribution of phosphoinositides and phosphatidic strategies. Nat Biotechnol 24(12):1551
acid in the human erythrocyte membrane. Biochim Haney EF, Wu BC, Lee K, Hilchie AL, Hancock RE
Biophys Acta 1069(1):27–36 (2017) Aggregation and its influence on the immuno-
Gaspar D, Veiga AS, Castanho MA (2013) From anti- modulatory activity of synthetic innate defense regula-
microbial to anticancer peptides. A review. Front tor peptides. Cell Chem Biol 24(8):969–980
Microbiol 4:294 Harris F, Dennison SR, Singh J, Phoenix DA (2013)
Gatto E, Mazzuca C, Stella L, Venanzi M, Toniolo On the selectivity and efficacy of defense pep-
C, Pispisa B (2006) Effect of peptide lipidation tides with respect to cancer cells. Med Res Rev
on membrane perturbing activity: a comparative 33(1):190–234
study on two trichogin analogues. J Phys Chem B Hartmann M, Berditsch M, Hawecker J, Ardakani MF,
110(45):22813–22818 Gerthsen D, Ulrich AS (2010) Damage of the bacterial
Gazit E, Boman A, Boman HG, Shai Y (1995) Interaction cell envelope by antimicrobial peptides gramicidin S
of the mammalian antibacterial peptide cecro- and PGLa as revealed by transmission and scanning
pin P1 with phospholipid vesicles. Biochemistry electron microscopy. Antimicrob Agents Chemother
34(36):11479–11488 54(8):3132–3142
Gelhaus C, Jacobs T, Andrä J, Leippe M (2008) The Hawrani A, Howe RA, Walsh TR, Dempsey CE (2008)
antimicrobial peptide NK-2, the core region of mam- Origin of low mammalian cell toxicity in a class of
malian NK-lysin, kills intraerythrocytic Plasmodium highly active antimicrobial amphipathic helical pep-
falciparum. Antimicrob Agents Chemother tides. J Biol Chem 283(27):18636–18645
52(5):1713–1720 Hayami M, Okabe A, Kariyama R, Abe M, Kanemasa
Ghosh JK, Shaool D, Guillaud P, Cicéron L, Mazier D, Y (1979) Lipid composition of Staphylococcus
Kustanovich I, Shai Y, Mor A (1997) Selective cyto- aureus and its derived L-forms. Microbiol Immunol
toxicity of dermaseptin S3 toward intraerythrocytic 23(6):435–442
plasmodium falciparum and the underlying molecular He J, Krauson AJ, Wimley WC (2014) Toward the de novo
basis. J Biol Chem 272(50):31609–31616 design of antimicrobial peptides: lack of correlation
Giacometti A, Cirioni O, Greganti G, Quarta M, Scalise between peptide permeabilization of lipid vesicles and
G (1998) In vitro activities of membrane-active antimicrobial, cytolytic, or cytotoxic activity in living
peptides against gram-positive and gram-negative cells. Biopolymers 102(1):1–6
aerobic bacteria. Antimicrob Agents Chemother Henriksen J, Rowat AC, Brief E, Hsueh YW, Thewalt
42(12):3320–3324 JL, Zuckermann MJ, Ipsen JH (2006) Universal
Giangaspero A, Sandri L, Tossi A (2001) Amphipathic behavior of membranes with sterols. Biophys
α helical antimicrobial peptides. A systematic J 90(5):1639–1649
study of the effects of structural and physical Henriksen JR, Etzerodt T, Gjetting T, Andresen TL (2014)
properties on biological activity. Eur J Biochem Side chain hydrophobicity modulates therapeutic
268(21):5589–5600 activity and membrane selectivity of antimicrobial
Glukhov E, Burrows LL, Deber CM (2008) Membrane peptide mastoparan-X. PLoS One 9:e91007
interactions of designed cationic antimicro- Hollmann A, Martinez M, Maturana P, Semorile LC,
bial peptides: the two thresholds. Biopolymers Maffia PC (2018) Antimicrobial peptides: interaction
89(5):360–371 with model and biological membranes and synergism
Golbek TW, Franz J, Elliott Fowler J, Schilke KF, with chemical antibiotics. Front Chem 6:204
Weidner T, Baio JE (2017) Identifying the selectivity Holt A, de Almeida RF, Nyholm TK, Loura LM, Daily
of antimicrobial peptides to cell membranes by sum AE, Staffhorst RW, Rijkers DTS, Koeppe RE II, Prieto
frequency generation spectroscopy. Biointerphases M, Killian JA (2008) Is there a preferential interaction
12(2):02D406 between cholesterol and tryptophan residues in mem-
Gonçalves S, Abade J, Teixeira A, Santos NC (2012) brane proteins? Biochemistry 47(8):2638–2649
Lipid composition is a determinant for human defen- Hong SY, Oh JE, Lee KH (1999) Effect of D-amino acid
sin HNP1 selectivity. Biopolymers 98(4):313–321 substitution on the stability, the secondary structure,
Gordesky SE, Marinetti GV (1973) The asymmetric and the activity of membrane-active peptide. Biochem
arrangement of phospholipids in the human eryth- Pharmacol 58(11):1775–1780
rocyte membrane. Biochem Biophys Res Commun Hoskin DW, Ramamoorthy A (2008) Studies on anti-
50(4):1027–1031 cancer activities of antimicrobial peptides. Biochim
Gordesky SE, Marinetti GV, Love R (1975) The reaction Biophys Acta 1778(2):357–375
of chemical probes with the erythrocyte membrane. Hoyos-Nogués M, Gil FJ, Mas-Moruno C (2018)
J Membr Biol 20(1–2):111–132 Antimicrobial peptides: powerful biorecognition
Hallock KJ, Lee DK, Omnaas J, Mosberg HI, ­elements to detect bacteria in biosensing technologies.
Ramamoorthy A (2002) Membrane composition Molecules 23(7):1683
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 207

Huang Y, Huang J, Chen Y (2010) Alpha-helical cationic Kahrom M, Bahar MM, Jangjoo A, Erfani M, Sadeghi R,
antimicrobial peptides: relationships of structure and Zakavi SR (2014) Poor sensitivity of 99mTc-labeled
function. Protein Cell 1(2):143–152 ubiquicidin scintigraphy in diagnosis of acute appen-
Huang Y, He L, Li G, Zhai N, Jiang H, Chen Y (2014) dicitis. Eur Surg 46(4):173–176
Role of helicity of α-helical antimicrobial peptides to Kamech N, Vukičević D, Ladram A, Piesse C, Vasseur J,
improve specificity. Protein Cell 5(8):631–642 Bojović V, Simunić J, Juretić D (2012) Improving the
Jacobs T, Bruhn H, Gaworski I, Fleischer B, Leippe M selectivity of antimicrobial peptides from anuran skin.
(2003) NK-lysin and its shortened analog NK-2 J Chem Inf Model 52(12):3341–3351
exhibit potent activities against Trypanosoma cruzi. Kaminski HM, Feix JB (2011) Effects of D-lysine substi-
Antimicrob Agents Chemother 47(2):607–613 tutions on the activity and selectivity of antimicrobial
Javadpour MM, Juban MM, Lo WCJ, Bishop SM, Alberty peptide CM15. Polymers 3(4):2088–2106
BJ, Cowell SM, Becker CL, McLaughlin ML (1996) Kang JH, Shin SY, Jang SY, Kim KL, Hahm KS (1999)
De novo antimicrobial peptides with low mammalian Effects of tryptophan residues of porcine myeloid
cell toxicity. J Med Chem 39:3107–3113 antibacterial peptide PMAP-23 on antibiotic activity.
Jepson AK, Schwarz-Linek J, Ryan L, Ryadnov MG, Biochem Biophys Res Commun 264(1):281–286
Poon WC (2016) What is the ‘Minimum Inhibitory Khandelia H, Kaznessis YN (2006) Molecular dynamics
Concentration’ (MIC) of Pexiganan acting on investigation of the influence of anionic and zwitter-
Escherichia coli?-A cautionary case study. In: Leake ionic interfaces on antimicrobial peptides’ structure:
MC (ed) Biophysics of infection. Springer, Cham, implications for peptide toxicity and activity. Peptides
pp 33–48 27(6):1192–1200
Jiang Z, Vasil AI, Hale JD, Hancock RE, Vasil ML, Kim S, Kim SS, Lee BJ (2005) Correlation between
Hodges RS (2008) Effects of net charge and the num- the activities of α-helical antimicrobial peptides and
ber of positively charged residues on the biological hydrophobicities represented as RP HPLC retention
activity of amphipathic α-helical cationic antimicro- times. Peptides 26(11):2050–2056
bial peptides. Biopolymers 90(3):369–383 Kim JK, Lee SA, Shin S, Lee JY, Jeong KW, Nan YH,
Jiang Z, Vasil AI, Gera L, Vasil M, Hodges RS (2011) Park YS, Shin SY, Kim Y (2010) Structural flexibility
Rational design of a-helical antimicrobial peptides to and the positive charges are the key factors in bacterial
target Gram-negative pathogens, Acinetobacter bau- cell selectivity and membrane penetration of peptoid-­
mannii and Pseudomonas aeruginosa: utilization of substituted analog of Piscidin 1. Biochim Biophys
charge, ‘specificity determinants,’ total hydrophobic- Acta 1798(10):1913–1925
ity, hydrophobe type and location as design param- Kindrachuk J, Napper S (2010) Structure-activity rela-
eters to improve the therapeutic ratio. Chem Biol Drug tionships of multifunctional host defence peptides.
Des 77(4):225–240 Mini-Rev Med Chem 10(7):596–614
Jiang Z, Vasil AI, Vasil ML, Hodges RS (2014) Koller D, Lohner K (2014) The role of spontaneous lipid
“Specificity Determinants” improve therapeutic curvature in the interaction of interfacially active
indices of two antimicrobial peptides piscidin 1 and peptides with membranes. Biochim Biophys Acta
dermaseptin s4 against the Gram-negative pathogens 1838(9):2250–2259
Acinetobacter baumannii and Pseudomonas aerugi- Konai MM, Samaddar S, Bocchinfuso G, Santucci V,
nosa. Pharmaceuticals 7(4):366–391 Stella L, Haldar J (2018) Selectively targeting bac-
Jiang Z, Mant CT, Vasil M, Hodges RS (2018) Role of teria by tuning the molecular design of membrane-­
positively charged residues on the polar and non- polar active peptidomimetic amphiphiles. Chem Commun
faces of amphipathic α- helical antimicrobial peptides 54(39):4943–4946
on specificity and selectivity for Gram-negative patho- Kondejewski LH, Jelokhani-Niaraki M, Farmer SW, Lix
gens. Chem Biol Drug Des 91(1):75–92 B, Kay CM, Sykes BD, Hancock RE, Hodges RS
Jones EM, Smart A, Bloomberg G, Burgess L, Millar MR (1999) Dissociation of antimicrobial and hemolytic
(1994) Lactoferricin, a new antimicrobial peptide. activities in cyclic peptide diastereomers by sys-
J Appl Bacteriol 77(2):208–214 tematic alterations in amphipathicity. J Biol Chem
Joshi S, Dewangan RP, Yar MS, Rawat DS, Pasha S 274(19):13181–13192
(2015) N-terminal aromatic tag induced self assem- Kondejewski LH, Lee DL, Jelokhani-Niaraki M, Farmer
bly of tryptophan–arginine rich ultra short sequences SW, Hancock REW, Hodges RS (2002) Optimization
and their potent antibacterial activity. RSC Adv of microbial specificity in cyclic peptides by modu-
5(84):68610–68620 lation of hydrophobicity within a defined structural
Juretic D, Vukicevic D, Ilic N, Antcheva N, Tossi A (2009) framework. J Biol Chem 277(1):67–74
Computational design of highly selective antimicro- Krause E, Beyermann M, Dathe M, Rothemund S,
bial peptides. J Chem Inf Model 49(12):2873–2882 Bienert M (1995) Location of an amphipathic. alpha-­
Juvvadi P, Vunnam S, Merrifield RB (1996) Synthetic Helix in peptides using reversed-phase HPLC reten-
melittin, its enantio, retro, and retroenantio isomers, tion behavior of D-amino acid analogs. Anal Chem
and selected chimeric analogs: their antibacterial, 67(2):252–258
hemolytic, and lipid bilayer action. J Am Chem Soc Krugliak M, Feder R, Zolotarev VY, Gaidukov L, Dagan
118(38):8989–8997 A, Ginsburg H, Mor A (2000) Antimalarial activities
208 S. Bobone and L. Stella

of dermaseptin S4 derivatives. Antimicrob Agents Lorian V (2005) Antibiotics in laboratory medicine, 5th
Chemother 44(9):2442–2451 edn. Lippincott Williams & Wilkins, Philadelphia
Kustanovich I, Shalev DE, Mikhlin M, Gaidukov L, Mor Luo Y, McLean DT, Linden GJ, McAuley DF, McMullan
A (2002) Structural requirements for potent versus R, Lundy FT (2017) The naturally occurring host
selective cytotoxicity for antimicrobial dermaseptin defense peptide, LL-37, and its truncated mimet-
S4 derivatives. J Biol Chem 277(19):16941–16951 ics KE-18 and KR-12 have selected biocidal and
Ladokhin AS, White SH (2001) Protein chemistry at mem- antibiofilm activities against Candida albicans,
brane interfaces: non-additivity of electrostatic and Staphylococcus aureus, and Escherichia coli in vitro.
hydrophobic interactions. J Mol Biol 309(3):543–552 Front Microbiol 8:544
Laverty G, McLaughlin M, Shaw C, Gorman SP, Gilmore Lupetti A, Welling MM, Pauwels EK, Nibbering PH
BF (2010) Antimicrobial activity of short, syn- (2003) Radiolabelled antimicrobial peptides for infec-
thetic cationic lipopeptides. Chem Biol Drug Des tion detection. Lancet Infect Dis 3(4):223–229
75(6):563–569 Lupetti A, Van Dissel JT, Brouwer CP, Nibbering PH
Le Joncour V, Laakkonen P (2018) Seek & destroy, use of (2008) Human antimicrobial peptides’ antifungal
targeting peptides for cancer detection and drug deliv- activity against Aspergillus fumigatus. Eur J Clin
ery. Bioorg Med Chem 26(10):2797–2806 Microbiol Infect Dis 27(11):1125–1129
Lee K, Shin SY, Kim K, Lim SS, Hahm KS, Kim Y Lyu Y, Yang Y, Lyu X, Dong N, Shan A (2016)
(2004) Antibiotic activity and structural analysis Antimicrobial activity, improved cell selectivity and
of the scorpion-­ derived antimicrobial peptide IsCT mode of action of short PMAP-36-derived peptides
and its analogs. Biochem Biophys Res Commun against bacteria and Candida. Sci Rep 6:27258
323(2):712–719 Mahlapuu M, Håkansson J, Ringstad L, Björn C (2016)
Lee MT, Hung WC, Chen FY, Huang HW (2005) Many-­ Antimicrobial peptides: an emerging category of ther-
body effect of antimicrobial peptides: on the correla- apeutic agents. Front Cell Infect Microbiol 6:194
tion between lipid’s spontaneous curvature and pore Makovitzki A, Avrahami D, Shai Y (2006) Ultrashort anti-
formation. Biophys J 89(6):4006–4016 bacterial and antifungal lipopeptides. Proc Natl Acad
Lee SA, Kim YK, Lim SS, Zhu WL, Ko H, Shin SY, Hahm Sci U S A 103(43):15997–16002
KS, Kim Y (2007) Solution structure and cell selec- Makovitzki A, Baram J, Shai Y (2008) Antimicrobial lipo-
tivity of piscidin 1 and its analogues. Biochemistry polypeptides composed of palmitoyl di-and tricationic
46(12):3653–3663 peptides: in vitro and in vivo activities, self-assembly
Lei R, Hou J, Chen Q, Yuan W, Cheng B, Sun Y, Jin Y, to nanostructures, and a plausible mode of action.
Ge L, Ben-Sasson SA, Chen J, Wang H, Lu W, Fang Biochemistry 47(40):10630–10636
X (2018) Self-assembling myristoylated human Malanovic N, Lohner K (2016) Antimicrobial peptides
α-defensin 5 as a next-generation nanobiotics poten- targeting gram-positive bacteria. Pharmaceuticals
tiates therapeutic efficacy in bacterial infection. ACS 9(3):59
Nano 12(6):5284–5296 Malina A, Shai Y (2005) Conjugation of fatty acids with
Leite NB, Aufderhorst-Roberts A, Palma MS, Connell different lengths modulates the antibacterial and anti-
SD, Neto JR, Beales PA (2015) PE and PS lip- fungal activity of a cationic biologically inactive pep-
ids synergistically enhance membrane poration tide. Biochem J 390(3):695–702
by a peptide with anticancer properties. Biophys Mangoni ML, Shai Y (2009) Temporins and their syn-
J 109(5):936–947 ergism against Gram-negative bacteria and in lipo-
Levison ME, Pitsakis PG, May PL, Johnson CC (1993) polysaccharide detoxification. Biochim Biophys Acta
The bactericidal activity of magainins against 1788(8):1610–1619
Pseudomonas aeruginosa and Enterococcus faecium. Mangoni ML, Carotenuto A, Auriemma L, Saviello MR,
J Antimicrob Chemother 32(4):577–585 Campiglia P, Gomez-Monterrey I, Malfi S, Marcellini
Lewis RNAH, McElhaney RN (2005) The mesomorphic L, Barra D, Novellino E, Grieco P (2011) Structure-­
phase behavior of lipids. In: Yeagle PL (ed) The struc- activity relationship, conformational and biologi-
ture of biological membranes, 2nd edn. CRC Press, cal studies of Temporin L analogues. J Med Chem
Boca Raton, pp 53–120 54(5):1298–1307
Li P, Zhou C, Rayatpisheh S, Ye K, Poon YF, Hammond Mannoor MS, Zhang S, Link AJ, McAlpine MC (2010)
PT, Duan H, Chan-Park MB (2012) Cationic peptido- Electrical detection of pathogenic bacteria via immo-
polysaccharides show excellent broad-spectrum anti- bilized antimicrobial peptides. Proc Natl Acad Sci U S
microbial activities and high selectivity. Adv Mater A 107(45):19207–19212
24(30):4130–4137 Marsh D (1990) CRC handbook of lipid bilayers. CRC
Lin D, Grossfield A (2015) Thermodynamics of micelle Press, Boca Raton
formation and membrane fusion modulate antimicro- Matsuzaki K (2009) Control of cell selectivity of
bial lipopeptide activity. Biophys J 109(4):750–759 antimicrobial peptides. Biochim Biophys Acta
Liu L, Xu K, Wang H, Tan PJ, Fan W, Venkatraman SS, 1788(8):1687–1692
Li L, Yang YY (2009) Self-assembled cationic peptide Matsuzaki K, Harada M, Handa T, Funakoshi S, Fujii N,
nanoparticles as an efficient antimicrobial agent. Nat Yajima H, Miyajima K (1989) Magainin 1-induced
Nanotechnol 4(7):457–463 leakage of entrapped calcein out of negatively-
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 209

charged lipid vesicles. Biochim Biophys Acta that mimic the structure of antimicrobial peptides.
981(1):130–134 Antimicrob Agents Chemother 59(7):4112–4120
Matsuzaki K, Sugishita K, Fujii N, Miyajima K (1995) Morein S, Andersson AS, Rilfors L, Lindblom G (1996)
Molecular basis for membrane selectivity of an Wild-type Escherichia coli cells regulate the membrane
antimicrobial peptide, magainin 2. Biochemistry lipid composition in a window between gel and non-­
34(10):3423–3429 lamellar structures. J Biol Chem 271(12):6801–6809
Matsuzaki K, Sugishita KI, Harada M, Fujii N, Miyajima Mouritsen OG, Zuckermann MJ (2004) What’s so special
K (1997) Interactions of an antimicrobial pep- about cholesterol? Lipids 39(11):1101–1113
tide, magainin 2, with outer and inner membranes Mura M, Wang J, Zhou Y, Pinna M, Zvelindovsky AV,
of Gram-negative bacteria. Biochim Biophys Acta Dennison SR, Phoenix DA (2016) The effect of ami-
1327(1):119–130 dation on the behaviour of antimicrobial peptides. Eur
Matsuzaki K, Sugishita KI, Ishibe N, Ueha M, Nakata S, Biophys J 45(3):195–207
Miyajima K, Epand RM (1998) Relationship of mem- Nan YH, Bang JK, Shin SY (2009) Design of novel
brane curvature to the formation of pores by magainin indolicidin-­derived antimicrobial peptides with
2. Biochemistry 37(34):11856–11863 enhanced cell specificity and potent anti-inflammatory
Maturana P, Martinez M, Noguera ME, Santos NC, activity. Peptides 30(5):832–838
Disalvo EA, Semorile L, Maffia PC, Hollmann A Nan YH, Lee BJ, Shin SY (2012) Prokaryotic selectivity,
(2017) Lipid selectivity in novel antimicrobial pep- anti-endotoxic activity and protease stability of diastereo-
tides: implication on antimicrobial and hemolytic meric and enantiomeric analogs of human antimicrobial
activity. Colloids Surf B Biointerfaces 153:152–159 peptide LL-37. Bull Kor Chem Soc 33(9):2883–2889
Mazzuca C, Stella L, Venanzi M, Formaggio F, Toniolo Nicolas P (2009) Multifunctional host defense peptides:
C, Pispisa B (2005) Mechanism of membrane activity intracellular-targeting antimicrobial peptides. FEBS
of the antibiotic trichogin GA IV: a two-state transi- J 276(22):6483–6496
tion controlled by peptide concentration. Biophys Nishi H, Komatsuzawa H, Fujiwara T, McCallum
J 88(5):3411–3421 N, Sugai M (2004) Reduced content of lysyl-­
McCloskey AP, Gilmore BF, Laverty G (2014) Evolution phosphatidylglycerol in the cytoplasmic membrane
of antimicrobial peptides to self-assembled peptides affects susceptibility to moenomycin, as well as
for biomaterial applications. Pathogens 3(4):791–821 vancomycin, gentamicin, and antimicrobial pep-
McHenry AJ, Sciacca MF, Brender JR, Ramamoorthy tides, in Staphylococcus aureus. Antimicrob Agents
A (2012) Does cholesterol suppress the antimi- Chemother 48(12):4800–4807
crobial peptide induced disruption of lipid raft Noore J, Noore A, Li B (2012) Cationic antimicrobial
containing membranes? Biochim Biophys Acta peptide LL-37 is effective against both extra-and intra-­
1818(12):3019–3024 cellular staphylococcus aureus. Antimicrob Agents
McIntosh TJ, Vidal A, Simon SA (2002) The energetics Chemother 57(3):1283–1290
of peptide-lipid interactions: modulation by interfa- Oddo A, Hansen PR (2017) Hemolytic activity of anti-
cial dipoles and cholesterol. In: Peptide-lipid interac- microbial peptides. In: Hansen PR (ed) Antimicrobial
tions, current topics in membranes, vol 52. Elsevier, peptides: methods and protocols. Methods in molec-
Amsterdam, pp 309–338 ular biology, vol 1548. Humana Press, New York,
McMahon HT, Boucrot E (2015) Membrane curvature at pp 427–435
a glance. J Cell Sci 128(6):1065–1070 Oh H, Hedberg M, Wade D, Edlund C (2000) Activities
Meléndez-Alafort L, Rodríguez-Cortés J, Ferro-Flores G, of synthetic hybrid peptides against anaerobic bacte-
De Murphy CA, Herrera-Rodríguez R, Mitsoura E, ria: aspects of methodology and stability. Antimicrob
Martínez-Duncker C (2004) Biokinetics of 99mTc-­ Agents Chemother 44(1):68–72
UBI 29-41 in humans. Nucl Med Biol 31(3):373–379 Oliver PM, Crooks JA, Leidl M, Yoon EJ, Saghatelian
Melo MN, Ferre R, Castanho MA (2009) Antimicrobial A, Weibel DB (2014) Localization of anionic phos-
peptides: linking partition, activity and high pholipids in Escherichia coli cells. J Bacteriol
membrane-­bound concentrations. Nat Rev Microbiol 196(19):3386–3398
7(3):245–250 Op den Kamp JAF, Redai I, van Deenen LLM (1969)
Mendive-Tapia L, Zhao C, Akram AR, Preciado S, Phospholipid composition of Bacillus subtilis.
Albericio F, Lee M, Serrels A, Kielland N, Read ND, J Bacteriol 99(1):298–303
Lavilla R, Vendrell M (2016) Spacer-free BODIPY Oren Z, Shai Y (1997) Selective lysis of bacte-
fluorogens in antimicrobial peptides for direct imag- ria but not mammalian cells by diastereomers of
ing of fungal infection in human tissue. Nat Commun melittin: structure-­ function study. Biochemistry
7:10940 36(7):1826–1835
Miyazaki Y, Aoki M, Yano Y, Matsuzaki K (2012) Oren Z, Shai Y (2000) Cyclization of a cytolytic amphipa-
Interaction of antimicrobial peptide magainin 2 with thic α-helical peptide and its diastereomer: effect on
gangliosides as a target for human cell binding. structure, interaction with model membranes, and bio-
Biochemistry 51(51):10229–10235 logical function. Biochemistry 39(20):6103–6114
Mojsoska B, Zuckermann RN, Jenssen H (2015) Oren Z, Lerman JC, Gudmundsson GH, Agerberth
Structure-activity relationship study of novel peptoids B, Shai Y (1999) Structure and organization of
210 S. Bobone and L. Stella

the human antimicrobial peptide LL-37 in phos- BL (2012) Methods for dilution antimicrobial sus-
pholipid membranes: relevance to the molecular ceptibility tests for bacteria that grow aerobically,
basis for its non-­ cell-­
selective activity. Biochem approved standard, 10th edn. Clinical and Laboratory
J 341(3):501–513 Standards Institute CLSI document M07-A10
Orioni B, Bocchinfuso G, Kim JY, Palleschi A, Grande Perron GG, Zasloff M, Bell G (2006) Experimental evo-
G, Bobone S, Park Y, Kim JI, Hahm KS, Stella L lution of resistance to an antimicrobial peptide. Proc
(2009) Membrane perturbation by the antimi- Biol Sci 273(1583):251–256
crobial peptide PMAP-23: a fluorescence and Phadke SM, Islam K, Deslouches B, Kapoor SA, Stolz
molecular dynamics study. Biochim Biophys Acta DB, Watkins SC, Montelaro RC, Pilewski JM,
1788(7):1523–1533 Mietzner TA (2003) Selective toxicity of engineered
Osborn MJ, Gander JE, Parisi E, Carson J (1972) lentivirus lytic peptides in a CF airway cell model.
Mechanism of assembly of the outer membrane of Peptides 24(8):1099–1107
Salmonella typhimurium isolation and characteriza- Phoenix DA, Harris F (2002) The hydrophobic moment
tion of cytoplasmic and outer membrane. J Biol Chem and its use in the classification of amphiphilic struc-
247(12):3962–3972 tures. Mol Membr Biol 19(1):1–10
Ostovar A, Assadi M, Vahdat K, Nabipour I, Javadi H, Phoenix DA, Dennison SR, Harris F (eds) (2012)
Eftekhari M, Assadi M (2013) A pooled analysis of Antimicrobial peptides. Wiley, New York
diagnostic value of 99mTc-ubiquicidin (UBI) scintig- Phoenix DA, Harris F, Mura M, Dennison SR (2015) The
raphy in detection of an infectious process. Clin Nucl increasing role of phosphatidylethanolamine as a lipid
Med 38(6):413–416 receptor in the action of host defence peptides. Prog
Otvos L (2017) Racing on the wrong track. Front Chem Lipid Res 59:26–37
5:42 Qiao Z, Lei C, Fu Y, Li Y (2017) Rapid and sensitive
Otvos L Jr, Bokonyi K, Varga I, Otvos BI, Hoffmann R, detection of E. coli O157: H7 based on antimicrobial
Ertl HC, Wade JD, McManus AM, Craik DJ, Bulet peptide functionalized magnetic nanoparticles and
P (2000) Insect peptides with improved protease-­ urease-catalyzed signal amplification. Anal Methods
resistance protect mice against bacterial infection. 9(35):5204–5210
Protein Sci 9(4):742–749 Raetz CR (1986) Molecular genetics of membrane phos-
Pachón-Ibáñez ME, Smani Y, Pachón J, Sánchez-Céspedes pholipid synthesis. Annu Rev Genet 20(1):253–291
J (2017) Perspectives for clinical use of engineered Raimondo D, Andreotti G, Saint N, Amodeo P, Renzone
human host defense antimicrobial peptides. FEMS G, Sanseverino M, Zocchi I, Molle G, Motta A, Scaloni
Microbiol Rev 41(3):323–342 A (2005) A folding-dependent mechanism of antimi-
Pandey BK, Ahmad A, Asthana N, Azmi S, Srivastava crobial peptide resistance to degradation unveiled by
RM, Srivastava S, Verma R, Vishwakarma AL, Ghosh solution structure of distinction. Proc Natl Acad Sci U
JK (2010) Cell-selective lysis by novel analogues of S A 102(18):6309–6314
melittin against human red blood cells and Escherichia Rautenbach M, Troskie AM, Vosloo JA (2016) Antifungal
coli. Biochemistry 49(36):7920–7929 peptides: to be or not to be membrane active.
Pandey BK, Srivastava S, Singh M, Ghosh JK (2011) Biochimie 130:132–145
Inducing toxicity by introducing a leucine-zipper-­ Ravi J, Bella A, Correia AJ, Lamarre B, Ryadnov MG
like motif in frog antimicrobial peptide, magainin 2. (2015) Supramolecular amphipathicity for probing
Biochem J 436(3):609–620 antimicrobial propensity of host defence peptides.
Panteleev P, Bolosov IA, Balandin SV, Ovchinnikova Phys Chem Chem Phys 17(24):15608–15614
TV (2015) Design of antimicrobial peptide arecinin Renner LD, Weibel DB (2011) Cardiolipin micro-
analogs with improved therapeutic indices. J Pept Sci domains localize to negatively curved regions of
21(2):105–113 Escherichia coli membranes. Proc Natl Acad Sci U S
Papo N, Oren Z, Pag U, Sahl HG, Shai Y (2002) The con- A 108(15):6264–6269
sequence of sequence alteration of an amphipathic Rivas L, Luque-Ortega JR, Andreu D (2009) Amphibian
α-helical antimicrobial peptide and its diastereomers. antimicrobial peptides and Protozoa: lessons from
J Biol Chem 277(37):33913–33921 parasites. Biochim Biophys Acta 1788(8):1570–1581
Park Y, Lee DG, Jang SH, Woo EH, Jeong HG, Choi CH, Rowlett VW, Mallampalli VK, Karlstaedt A, Dowhan
Hahm KS (2003) A Leu-Lys-rich antimicrobial pep- W, Taegtmeyer H, Margolin W, Vitrac H (2017) The
tide: activity and mechanism. Biochim Biophys Acta impact of membrane phospholipid alterations in
1645(2):172–182 Escherichia coli on cellular function and bacterial
Pasupuleti M, Schmidtchen A, Chalupka A, Ringstad L, stress adaptation. J Bacteriol 199(13). https://doi.
Malmsten M (2009) End-tagging of ultra-short antimi- org/10.1128/JB.00849-16
crobial peptides by W/F stretches to facilitate bacterial Ruiz J, Calderon J, Rondón-Villarreal P, Torres R (2014)
killing. PLoS One 4(4):e5285 Analysis of structure and hemolytic activity relation-
Patel JB, Cockerill FR, Bradford PA, Eliopulos GM, ships of antimicrobial peptides (AMPs). In: Advances
Hindler JA, Jenkins SG, Lewis II JS, Limbago B, in computational biology. Springer, Cham, pp 253–258
Miller LA, Nicolau DP, Powell DP, Swenson JM, Russell AL, Kennedy AM, Spuches AM, Venugopal D,
Traczewski MM, Turnidge JD, Weistein MP, Zimmer Bhonsle JB, Hicks RP (2010) Spectroscopic and ther-
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 211

modynamic evidence for antimicrobial peptide mem- function: design, synthesis, and broad spectrum anti-
brane selectivity. Chem Phys Lipids 163(6):488–497 microbial properties of short hybrid E-vinylogous
Saeed S, Zafar J, Khan B, Akhtar A, Qurieshi S, Fatima lipopeptides. J Med Chem 56(21):8468–8474
S, Ahmad N, Irfanullah J (2013) Utility of 99mTc-­ Shi X, Zhang X, Yao Q, He F (2017) A novel method for
labelled antimicrobial peptide ubiquicidin (29-41) in the rapid detection of microbes in blood using pleuro-
the diagnosis of diabetic foot infection. Eur J Nucl cidin antimicrobial peptide functionalized piezoelec-
Med Mol Imaging 40(5):737–743 tric sensor. J Microbiol Methods 133:69–75
Salick DA, Kretsinger JK, Pochan DJ, Schneider Shin SY, Yang ST, Park EJ, Eom SH, Song WK, Kim JI,
JP (2007) Inherent antibacterial activity of a Lee SH, Lee MK, Lee DG, Hahm KS, Kim Y (2001)
peptide-­based β-hairpin hydrogel. J Am Chem Soc Antibacterial, antitumor and hemolytic activities of
129(47):14793–14799 α-helical antibiotic peptide, P18 and its analogs. J Pept
Sal-Man N, Oren Z, Shai Y (2002) Preassembly of Res 58(6):504–514
membrane-active peptides is an important factor in Shriver-Lake LC, North SH, Dean SN, Taitt CR (2012)
their selectivity toward target cells. Biochemistry Antimicrobial peptides for detection and diagnostic
41(39):11921–11930 assays. In: Designing receptors for the next generation
Santos NC, Prieto M, Castanho MA (2003) Quantifying of biosensors. Springer, Berlin, pp 85–104
molecular partition into model systems of biomem- Silva RR, Avelino KY, Ribeiro KL, Franco OL, Oliveira
branes: an emphasis on optical spectroscopic methods. MD, Andrade CA (2014) Optical and dielectric sen-
Biochim Biophys Acta 1612(2):123–135 sors based on antimicrobial peptides for microorgan-
Savini F, Luca V, Bocedi A, Massoud R, Park Y, Mangoni ism diagnosis. Front Microbiol 5:443. https://doi.
ML, Stella L (2017) Cell-density dependence of host-­ org/10.3389/fmicb.2014.00443
defense peptide activity and selectivity in the presence Simon SA, McIntosh TJ (2002) Peptide-lipid interac-
of host cells. ACS Chem Biol 12(1):52–56 tions, current topics in membranes, vol 52. Elsevier,
Savini F, Bobone S, Roversi D, Mangoni ML, Stella Amsterdam
L (2018) From liposomes to cells: filling the gap Skerlavaj B, Renato Gennaro R, Luigi Bagella L, Laura
between physicochemical and microbiological studies Merluzzi L, Angela Risso A, Zanetti M (1996)
of the activity and selectivity of host-defense peptides. Biological characterization of two novel cathelicidin-­
Pept Sci 110(5):e24041 derived peptides and identification of structural
Schmidtchen A, Pasupuleti M, M&oumlrgelin M, requirements for their antimicrobial and cell lytic
Davoudi M, Alenfall J, Chalupka A, Malmsten activities. J Biol Chem 271(45):28375–28381
M (2009) Boosting antimicrobial peptides by Slaninová J, Mlsová V, Kroupová H, Alán L, Tůmová T,
hydrophobic oligopeptide end-tags. J Biol Chem Monincová L, Borovičková L, Fučík V, Ceřovský V
284(26):17584–17594 (2012) Toxicity study of antimicrobial peptides from
Schmidtchen A, Ringstad L, Kasetty G, Mizuno H, wild bee venom and their analogs toward mammalian
Rutland MW, Malmsten M (2011) Membrane selectiv- normal and cancer cells. Peptides 33(1):18–26
ity by W-tagging of antimicrobial peptides. Biochim Snoussi M, Talledo JP, Del Rosario NA, Ha BY, Kosmrlj
Biophys Acta 1808(4):1081–1091 A, Taheri-Araghi S (2018) Heterogeneous absorption
Schmidtchen A, Pasupuleti M, Malmsten M (2014) Effect of antimicrobial peptide LL37 in Escherichia coli cells
of hydrophobic modifications in antimicrobial pep- enhances population survivability. eLife 7:e38174
tides. Adv Colloid Interf Sci 205:265–274 Son M, Lee Y, Hwang H, Hyun S, Yu J (2013) Disruption
Schröder-Borm H, Willumeit R, Brandenburg K, Andrä of interactions between hydrophobic residues on
J (2003) Molecular basis for membrane selectivity nonpolar faces is a key determinant in decreasing
of NK-2, a potent peptide antibiotic derived from hemolysis and increasing antimicrobial activities
NK-lysin. Biochim Biophys Acta 1612(2):164–171 of α-helical amphipathic peptides. ChemMedChem
Schweizer F (2009) Cationic amphiphilic peptides 8(10):1638–1642
with cancer-selective toxicity. Eur J Pharmacol Song YM, Yang ST, Lim SS, Kim Y, Hahm KS, Kim JI,
625(1–3):190–194 Shin SY (2004) Effects of L-or D-Pro incorpora-
Seelig J (2004) Thermodynamics of lipid–peptide interac- tion into hydrophobic or hydrophilic helix face of
tions. Biochim Biophys Acta 1666(1):40–50 amphipathic α-helical model peptide on structure
Seo MD, Won HS, Kim JH, Mishig-Ochir T, Lee BJ and cell selectivity. Biochem Biophys Res Commun
(2012) Antimicrobial peptides for therapeutic applica- 314(2):615–621
tions: a review. Molecules 17(10):12276–12286 Song YM, Park Y, Lim SS, Yang ST, Woo ER, Park IS,
Shai Y, Oren Z (1996) Diastereomers of cytolysins, a Lee JS, Kim JI, Hahm KS, Kim Y, Shin SY (2005)
novel class of potent antibacterial peptides. J Biol Cell selectivity and mechanism of action of antimi-
Chem 271(13):7305–7308 crobial model peptides containing peptoid residues.
Shai Y, Oren Z (2001) From “carpet” mechanism to de-­ Biochemistry 44(36):12094–12106
novo designed diastereomeric cell-selective antimi- Sood R, Kinnunen PK (2008) Cholesterol, lanosterol,
crobial peptides. Peptides 22(10):1629–1641 and ergosterol attenuate the membrane association
Shankar SS, Benke SN, Nagendra N, Srivastava PL, of LL-37 (W27F) and temporin L. Biochim Biophys
Thulasiram HV, Gopi HN (2013) Self-assembly to Acta 1778(6):1460–1466
212 S. Bobone and L. Stella

Sood R, Domanov Y, Pietiäinen M, Kontinen VP, Effect of N α-acyl chain length on the membrane-
Kinnunen PK (2008) Binding of LL-37 to model bio- modifying properties of synthetic analogs of the
membranes: insight into target vs host cell recogni- lipopeptaibol trichogin GA IV. J Am Chem Soc
tion. Biochim Biophys Acta 1778(4):983–996 118(21):4952–4958
Stark M, Liu LP, Deber CM (2002) Cationic hydropho- Tossi A (2011) Design and engineering strategies for syn-
bic peptides with antimicrobial activity. Antimicrob thetic antimicrobial peptides. In: Prokaryotic antimi-
Agents Chemother 46(11):3585–3590 crobial peptides. Springer, New York, pp 81–98
Starr CG, He J, Wimley WC (2016) Host cell interactions Tossi A, Sandri L, Giangaspero A (2000) Amphipathic,
are a significant barrier to the clinical utility of peptide α-helical antimicrobial peptides. Biopolymers
antibiotics. ACS Chem Biol 11(12):3391–3399 55(1):4–30
Steiner H, Andreu D, Merrifield RB (1988) Binding and Tu Z, Hao J, Kharidia R, Meng XG, Liang JF (2007)
action of cecropin and cecropin analogues: antibac- Improved stability and selectivity of lytic pep-
terial peptides from insects. Biochim Biophys Acta tides through self-assembly. Biochem Biophys Res
939(2):260–266 Commun 361(3):712–717
Stella L, Mazzuca C, Venanzi M, Palleschi A, Didone M, Tytler EM, Anantharamaiah GM, Walker DE, Mishra VK,
Formaggio F, Toniolo C, Pispisa B (2004) Aggregation Palgunachari MN, Segrest JP (1995) Molecular basis
and water-membrane partition as major determinants for prokaryotic specificity of magainin-induced lysis.
of the activity of the antibiotic peptide trichogin GA Biochemistry 34(13):4393–4401
IV. Biophys J 86(2):936–945 Uematsu N, Matsuzaki K (2000) Polar angle as a deter-
Storch J, Kleinfeld AM (1985) The lipid structure minant of amphipathic α-helix-lipid interactions: a
of biological membranes. Trends Biochem Sci model peptide study. Biophys J 79(4):2075–2083
10(11):418–421 Uggerhøj LE, Poulsen TJ, Munk JK, Fredborg M,
Strandberg E, Tiltak D, Ieronimo M, Kanithasen N, Sondergaard TE, Frimodt-Moller N, Hansen
Wadhwani P, Ulrich AS (2007) Influence of C-terminal PR, Wimmer R (2015) Rational design of alpha-­
amidation on the antimicrobial and hemolytic activi- helical antimicrobial peptides: do’s and don’ts.
ties of cationic -helical peptides. Pure Appl Chem ChemBioChem 16(2):242–253
79(4):717–728 Vallejo E, Martinez I, Tejero A, Hernandez S, Jimenez L,
Strömstedt AA, Ringstad L, Schmidtchen A, Malmsten M Bialostozky D, Sanchez G, Ilarraza H, Ferro-Flores
(2010) Interaction between amphiphilic peptides and G (2008) Clinical utility of 99mTc-labeled ubiquici-
phospholipid membranes. Curr Opin Colloid Interface din 29–41 antimicrobial peptide for the scintigraphic
Sci 15(6):467–478 detection of mediastinitis after cardiac surgery. Arch
Swierstra J, Kapoerchan V, Knijnenburg A, van Belkum Med Res 39(8):768–774
A, Overhand M (2016) Structure, toxicity and antibi- van der Weerden NL, Bleackley MR, Anderson MA
otic activity of gramicidin S and derivatives. Eur J Clin (2013) Properties and mechanisms of action of natu-
Microbiol Infect Dis 35(5):763–769 rally occurring antifungal peptides. Cell Mol Life Sci
Tachi T, Epand RF, Epand RM, Matsuzaki K (2002) 70(19):3545–3570
Position-dependent hydrophobicity of the antimicro- Veiga AS, Sinthuvanich C, Gaspar D, Franquelim HG,
bial magainin peptide affects the mode of peptide-­ Castanho MA, Schneider JP (2012) Arginine-rich
lipid interactions and selective toxicity. Biochemistry self-assembling peptides as potent antibacterial gels.
41(34):10723–10731 Biomaterials 33(35):8907–8916
Takahashi D, Shukla SK, Prakash O, Zhang G (2010) Velduhizen EJA, Scheenstra MR, Tjeerdsma-van
Structural determinants of host defense peptides Bokhoven JLM, Coorens M, Schneider VAF, Bikker
for antimicrobial activity and target cell selectivity. FJ, van Dijk A, Haagsman HP (2017) Antimicrobial
Biochimie 92(9):1236–1241 and immunomodulatory activity of PMAP-23 derived
Teixeira V, Feio MJ, Bastos M (2012) Role of lipids in the peptides. Protein Pept Lett 24(7):609–616
interaction of antimicrobial peptides with membranes. Verkleij AJ, Zwaal RFA, Roelofsen B, Comfurius P,
Prog Lipid Res 51(2):149–177 Kastelijn D, van Deenen LLM (1973) The asymmet-
Thennarasu S, Nagaraj R (1996) Specific antimicrobial ric distribution of phospholipids in the human red
and hemolytic activities of 18-residue peptides derived cell membrane. A combined study using phospholi-
from the amino terminal region of the toxin pardaxin. pases and freeze-etch electron microscopy. Biochim
Protein Eng Des Sel 9(12):1219–1224 Biophys Acta 323(2):178–193
Tian X, Sun F, Zhou XR, Luo SZ, Chen L (2015) Role of Verly RM, Rodrigues MA, Daghastanli KRP, Denadai
peptide self-assembly in antimicrobial peptides. J Pept AML, Cuccovia IM, Bloch C Jr, Frezard F, Santoro
Sci 21(7):530–539 MM, Pilo-Veloso D, Bemquerer MP (2008) Effect of
Tiozzo E, Rocco G, Tossi A, Romeo D (1998) Wide-­ cholesterol on the interaction of the amphibian anti-
spectrum antibiotic activity of synthetic, amphipa- microbial peptide DD K with liposomes. Peptides
thic peptides. Biochem Biophys Res Commun 29(1):15–24
249(1):202–206 Vermeer LS, Lan Y, Abbate V, Ruh E, Bui TT, Wilkinson
Toniolo C, Crisma M, Formaggio F, Peggion C, L, Jumagulova E, Kozlowska J, Patel J, McIntyre
Monaco V, Goulard C, Rebuffat S, Bodo B (1996) CA, Yam WC, Siu GKH, Atkinson RA, Lam JKW,
11 Selectivity of Antimicrobial Peptides: A Complex Interplay of Multiple Equilibria 213

Bansal SS, Drake AF, Mitchell GH, Mason AJ (2012) Wieprecht T, Dathe M, Beyermann M, Krause E, Maloy
Conformational flexibility determines selectivity WL, MacDonald DL, Bienert M (1997a) Peptide
and Antibacterial, Antiplasmodium, and Anticancer hydrophobicity controls the activity and selectivity
potency of cationic α-helical peptides. J Biol Chem of magainin 2 amide in interaction with membranes.
287(41):34120–34133 Biochemistry 36(20):6124–6132
Virtanen JA, Cheng KH, Somerharju P (1998) Wieprecht T, Dathe M, Krause E, Beyermann M, Maloy
Phospholipid composition of the mammalian red cell WL, MacDonald DL, Bienert M (1997b) Modulation
membrane can be rationalized by a superlattice model. of membrane activity of amphipathic, antibacterial
Proc Natl Acad Sci U S A 95(9):4964–4969 peptides by slight modifications of the hydrophobic
Wade D, Boman A, Wåhlin B, Drain CM, Andreu D, moment. FEBS Lett 417(1):135–140
Boman HG, Merrifield RB (1990) All-D amino acid-­ Wieprecht T, Dathe M, Epand RM, Beyermann M,
containing channel-forming antibiotic peptides. Proc Krause E, Maloy WL, MacDonald DL, Bienert M
Natl Acad Sci U S A 87(12):4761–4765 (1997c) Influence of the angle subtended by the posi-
Wade D, Silberring J, Soliymani R, Heikkinen S, Kilpeläinen tively charged helix face on the membrane activity
I, Lankinen H, Kuusela P (2000) Antibacterial activities of amphipathic, antibacterial peptides. Biochemistry
of temporin A analogs. FEBS Lett 479(1–2):6–9 36(42):12869–12880
Wakabayashi H, Matsumoto H, Hashimoto K, Teraguchi Wimley WC (2010a) Energetics of peptide and protein
S, Takase M, Hayasawa H (1999) N-Acylated and D binding to lipid membranes. In: Proteins membrane
enantiomer derivatives of a nonamer core peptide of binding and pore formation. Springer, New York,
lactoferricin B showing improved antimicrobial activ- pp 14–23
ity. Antimicrob Agents Chemother 43(5):1267–1269 Wimley WC (2010b) Describing the mechanism of anti-
Wang G (ed) (2017) Antimicrobial peptides: discov- microbial peptide action with the interfacial activity
ery, design and novel therapeutic strategies. CABI model. ACS Chem Biol 5(10):905–917
Publishing, Cambridge, MA Wimley WC, Hristova K (2011) Antimicrobial peptides:
Wang H, Xu K, Liu L, Tan JP, Chen Y, Li Y, Fan W, Wei successes, challenges and unanswered questions.
Z, Sheng J, Yang YY, Li L (2010) The efficacy of self-­ J Membr Biol 239(1–2):27–34
assembled cationic antimicrobial peptide nanoparticles Wu G, Wu H, Fan X, Zhao R, Li X, Wang S, Ma Y, Shen
against Cryptococcus neoformans for the treatment of Z, Xi T (2010) Selective toxicity of antimicrobial pep-
meningitis. Biomaterials 31(10):2874–2881 tide S-thanatin on bacteria. Peptides 31(9):1669–1673
Wang J, Chou S, Xu L, Zhu X, Dong N, Shan A, Chen Z Yang ST, Shin SY, Kim YC, Kim Y, Hahm KS, Kim JI
(2015) High specific selectivity and membrane-active (2002) Conformation-dependent antibiotic activity of
mechanism of the synthetic centrosymmetric α-helical tritrpticin, a cathelicidin-derived antimicrobial peptide.
peptides with Gly-Gly pairs. Sci Rep 5:15963 Biochem Biophys Res Commun 296(5):1044–1050
Wang J, Chou S, Yang Z, Yang Y, Wang Z, Song J, Dou Yang ST, Lee JY, Kim HJ, Eu YJ, Shin SY, Hahm KS, Kim
X, Shan A (2018) Combating drug-resistant fungi with JI (2006a) Contribution of a central proline in model
novel imperfectly amphipathic palindromic peptides. amphipathic α-helical peptides to self-­ association,
J Med Chem 61(9):3889–3907 interaction with phospholipids, and antimicrobial
Welling MM, Nibbering PH, Paulusma-Annema A, mode of action. FEBS J 273(17):4040–4054
Hiemstra PS, Pauwels EK, Calame W (1999) Imaging Yang ST, Jeon JH, Kim Y, Shin SY, Hahm KS, Kim JI
of bacterial infections with 99mTc-labeled human neu- (2006b) Possible role of a PXXP central hinge in
trophil peptide-1. J Nucl Med 40(12):2073–2080 the antibacterial activity and membrane interac-
Welling MM, Paulusma-Annema A, Balter HS, Pauwels tion of PMAP-23, a member of cathelicidin family.
EK, Nibbering PH (2000) Technetium-99m labelled Biochemistry 45(6):1775–1784
antimicrobial peptides discriminate between bacterial Yau WM, Wimley WC, Gawrisch K, White SH (1998)
infections and sterile inflammations. Eur J Nucl Med The preference of tryptophan for membrane inter-
27(3):292–301 faces. Biochemistry 37(42):14713–14718
White DA (1973) The phospholipid composition of mam- Yeaman MR, Yount NY (2003) Mechanisms of antimi-
malian tissues. In: Ansell GB, Hawthorne JN, Dawson crobial peptide action and resistance. Pharmacol Rev
RMC (eds) Form and function of phospholipids. 55(1):27–55
Elsevier, Amsterdam, pp 441–482 Yeung AT, Gellatly SL, Hancock REW (2011)
White SH, Wimley WC (1999) Membrane protein folding Multifunctional cationic host defence peptides
and stability: physical principles. Annu Rev Biophys and their clinical applications. Cell Mol Life Sci
Biomol Struct 28(1):319–365 68(13):2161
Wiegand I, Hilpert K, Hancock REW (2008) Agar and Zelezetsky I, Tossi A (2006) Alpha-helical antimicrobial
broth dilution methods to determine the minimal peptides-using a sequence template to guide structure–
inhibitory concentration (MIC) of antimicrobial sub- activity relationship studies. Biochim Biophys Acta
stances. Nat Protoc 3(2):163–175 1758(9):1436–1449
Wieprecht T, Seelig J (2002) Isothermal titration calorim- Zelezetsky I, Pacor S, Pag U, Papo N, Shai Y, Sahl HG,
etry for studying interactions between peptides and Tossi A (2005) Controlled alteration of the shape and
lipid membranes. Curr Top Membr 52:31–56 conformational stability of α-helical cell-lytic pep-
214 S. Bobone and L. Stella

tides: effect on mode of action and cell specificity. of the leucine zipper sequence in melittin with pep-
Biochem J 390(1):177–188 toid residues affects self-association, cell selectiv-
Zhang L, Benz R, Hancock REW (1999) Influence of ity, and mode of action. Biochim Biophys Acta
proline residues on the antibacterial and synergis- 1768(6):1506–1517
tic activities of R-helical peptides. Biochemistry Zhu WL, Nan YH, Hahm K, Shin SY (2007c) Cell selec-
38(25):8102–8111 tivity of an antimicrobial peptide melittin diastereo-
Zhang Y, Lu H, Lin Y, Cheng J (2011) Water-soluble poly- mer with D-amino acid in the leucine zipper sequence.
peptides with elongated, charged side chains adopt J Biochem Mol Biol 40(6):1090–1094
ultrastable helical conformations. Macromolecules Zhu WL, Hahm KS, Shin SY (2009) Cell selectivity and
44(17):6641–6644 mechanism of action of short antimicrobial peptides
Zhang SK, Song JW, Gong F, Li SB, Chang HY, Xie HM, designed from the cell-penetrating peptide Pep-1.
Gao HW, Tan YX, Ji SP (2016) Design of an α-helical J Pept Sci 15(9):569–575
antimicrobial peptide with improved cell-selective and Zou R, Zhu X, Tu Y, Wu J, Landry MP (2018) Activity
potent anti-biofilm activity. Sci Rep 6:27394 of antimicrobial peptide aggregates decreases with
Zhou NE, Mant CT, Hodges RS (1990) Effect of preferred increased cell membrane embedding free energy cost.
binding domains on peptide retention behavior in Biochemistry 57(18):2606–2610
reversed-phase chromatography: amphipathic alpha-­ Zwaal RFA, Roelofsen B, Colley CM (1973) Localization
helices. Pept Res 3(1):8–20 of red cell membrane constituents. Biochim Biophys
Zhu WL, Hahm K, Shin SY (2007a) Cathelicidin-derived Acta 300(2):159–182
Trp/Pro-rich antimicrobial peptides with lysine pep- Zwaal RFA, Roelsfsen B, Comfurius P, van Deenen LLM
toid residue (Nlys): therapeutic index and plausible (1975) Organization of phospholipids in human red
mode of action. J Pept Sci 13(8):529–535 cell membranes as detected by the action of vari-
Zhu WL, Song YM, Park Y, Park KH, Yang ST, Kim JI, ous purified phospholipases. Biochim Biophys Acta
Park IS, Hahm KS, Shin SY (2007b) Substitution 406(1):83–96
 Design of Antimicrobial Peptides:
Progress Made with Human 12
Cathelicidin LL-37

Guangshun Wang, Jayaram Lakshmaiah Narayana,

Biswajit Mishra, Yingxia Zhang, Fangyu Wang,
Chunfeng Wang, D. Zarena, Tamara Lushnikova,
and Xiuqing Wang

Abstract has been engineered into 17BIPHE2, a stable,

The incorporation of the innate immune sys- selective, and potent antimicrobial, antibio-
tem into humans is essential for survival and film, and anticancer peptide. Both 17BIPHE2
health due to the rapid replication of invading and SAAP-148 can eliminate the ESKAPE
microbes and the delayed action of the adap- pathogens and show topical in vivo antibio-
tive immune system. Antimicrobial peptides film efficacy. Also discussed are other applica-
are important components of human innate tion strategies, including peptide formulation,
immunity. Over 100 such peptides have been antimicrobial implants, and peptide-inducing
identified in various human tissues. Human factors such as vitamin D and sunlight. Finally,
cathelicidin LL-37 is best studied, and there we summarize what we learned from peptide
has been a growing interest in designing new design based on human LL-37.
peptides based on LL-37. This chapter
describes the alternative processing of the Keywords
human cathelicidin precursor, protease diges- Anticancer peptides · Antimicrobial peptides ·
tion, and lab cutting of LL-37. Both a syn- Antiviral peptides · Cathelicidins · LL-37 ·
thetic peptide library and structure-based Peptide design
design are utilized to identify the active
regions. Although challenging, the determina-
tion of the 3D structure of LL-37 enabled the
identification of the core antimicrobial region. 12.1 Introduction
The minimal region of LL-37 can be function-­
dependent. We discuss the design and poten- The discovery of penicillin in 1928 is one of the
tial applications of LL-37 into antibacterial, greatest achievements in human history. However,
antibiofilm, antiviral, antifungal, immune overreliance on antimicrobial drugs against infec-
modulating, and anticancer peptides. LL-37 tious diseases led to a rapid development of resis-
tance in bacteria. Antimicrobial resistance posts a
great risk to the human health-care system. To
G. Wang (*) · J. L. Narayana · B. Mishra · Y. Zhang counteract the impacts of emerging multidrug
F. Wang · C. Wang · D. Zarena · T. Lushnikova
X. Wang resistance, alternatives are in urgent need.
Department of Pathology and Microbiology, College Antimicrobial peptides (AMPs) have been recog-
of Medicine, University of Nebraska Medical Center, nized as one such alternative because bacterial
Omaha, NE, USA resistance development is rare or not yet observed
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 215

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_12
216 G. Wang et al.

(Zasloff 2002; Boman 2003; Jayaram and Chen cells, including monocytes, neutrophils, mast
2015; Wang et al. 2015; Mishra et al. 2017b). They cells, stem cells, NK cells, and B and T cells. As
are ancient molecules optimized through their an innate immune peptide, it is upregulated upon
coevolution with bacteria over millions of years pathogen invasion or by immune stimuli such as
(Peschel and Sahl 2006; Spencer et al. 2014). vitamin D and sunlight. LL-37 has a broad-­
According to the antimicrobial peptide data- spectrum activity at a micromolar concentration
base (APD; http://aps.unmc.edu/AP), naturally against bacteria, fungi, viruses, and fungi at least
occurring AMPs (<100 amino acids) have been in vitro (Durr et al. 2006; Vandamme et al. 2012).
discovered in the six life kingdoms, including The observation that the concentration of
bacteria, archaea, protists, fungi, plants, and ani- LL-37 in certain human tissues is below the mini-
mals (Wang and Wang 2004; Wang et al. 2009, mal inhibitory concentration (MIC) required to
2016). In invertebrates, innate immune systems kill pathogens, however, led to an emphasis on
are the only defense weapon against microbial immune modulation (Scott et al. 2002). It is now
infection. The discovery of cecropins, magainins, accepted that human cathelicidin LL-37 is a
and defensins in the 1980s laid the foundation for moonlighting peptide with multiple functional
deciphering the pathogen-specific innate immune roles, ranging from antimicrobial to immune reg-
response pathways (Boman 2003; Lehrer and Lu ulation. The moonlighting properties of LL-37
2012; Meister et al. 1997; Zasloff 1987). In mam- provide a basis for its involvement in a variety of
mals, several defense mechanisms guard against human diseases such as infection, diabetes, can-
the threat of infection, ranging from the innate to cer metastasis, and atherosclerosis (Scott et al.
adaptive immune systems, including skin barriers 2002; Vandamme et al. 2012).
and physical factors such as urine flow, pH, and This chapter summarizes the discovery,
ionic composition. In humans, 124 AMPs were design, and potential applications of new AMPs
identified in various tissues as of June 2018 (Wang based on the LL-37 template. Both structure-­
et al. 2016). Typical examples include lysozyme, based design and library screening are covered. It
defensins, histatins, cathelicidins, lactoferricin, is useful to mention the challenges in structural
kinocidins, ribonuclease, and dermcidin (reviewed determination of LL-37 by NMR spectroscopy
in Wang 2014). On average, these molecules have because of the importance of this structure for a
a chain length of 44.6 amino acids with a net better understanding of the antimicrobial activity
charge of +4.8. Such properties allow them to of the peptide as well as for structure-based pep-
adopt unique structures for host defense. tide discovery. To facilitate a connection between
Unlike other animals, there is only one cathe- LL-37 and its fragments, we use the nomencla-
licidin gene in humans encoded on chromosome ture and numbering of LL-37 throughout the text.
3p21.3 (Frohm et al. 1997). Cathelicidins are For example, KR-12 (Table 12.1) means a
synthesized as preproproteins with a highly con- 12-­residue peptide corresponding to residues
served N-terminal domain and a highly variable 18–29 of LL-37; it starts with amino acids
C-terminal antimicrobial domain. The N-terminal K18R19 at the N-terminus and ends with R29 at
domain usually consists of 94–114 amino acids the C-terminus. Subsequent sections are devoted
and shares sequence homology with cathelin, a to the potential applications of these peptides in
cysteine protease inhibitor derived from porcine antibiofilm, antiviral, antifungal, anticancer, sur-
neutrophils, hence the name cathelin-like domain face immobilization, and immune modulation
(CLD). At present, there are 113 mature catheli- studies with a focus on the relationship between
cidin peptides in the APD, ranging from hagfish intact LL-37 and its fragments. It seems a slightly
to humans (Wang et al. 2016). Human cathelici- different region can be utilized for peptide design
din LL-37 is one of the best-studied AMPs. It is depending on the targeted pathogen or disease,
widely distributed in the human saliva, sperm, including laboratory preference. Some LL-37
skin, gastrointestine, urinary tract, and respira- designer peptides have been shown to have antib-
tory airways. LL-37 is expressed by a number of iofilm efficacy in animal models.
 12

Table 12.1 Amino acid sequences and physical properties of select peptides derived from LL-37
Name Amino acid sequence Net charge Pho% HP Boman index GRAVY
1
LL-37 LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES37 +6 35% 47.79 2.99 −0.72
1
LL-31 LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNL31 +6 38% 50.98 2.81 −0.64
7
RK-31 RKSKEKIGKEFKRIVQRIKDFLRNLVPRTES37 +7 31% 38.23 3.83 −1.16
8
KS-30 KSKEKIGKEFKRIVQRIKDFLRNLVPRTES37 +6 30% 37.39 3.47 −1.05
1
LL-25 LLGDFFRKSKEKIGKEFKRIVQRIK25 +6 36% 38.4 2.79 −0.75
7
RK-25 RKSKEKIGKEFKRIVQRIKDFLRNL31 +7 32% 38.69 3.81 −1.17
13
IG-25 IGKEFKRIVQRIKDFLRNLVPRTES37 +4 36% 39.37 3.08 −0.62
13
IG-24 (P60) IGKEFKRIVQRIKDFLRNLVPRTE36 +4 37% 40.22 3.07 −0.62
OP-145 Acetyl13IGKEFKRIVERIKRFLRELVRPLR36 +6 41% 45.31 3.42 −0.51
1
LL-23 LLGDFFRKSKEKIGKEFKRIVQR23 +5 34% 38.61 3.01 −0.84
12
KR-22 KIGKEFKRIVQRIKDFLRNLVP33 +5 40% 41.18 2.5 −0.45
14
GKE-21 GKEFKRIVQRIKDFLRNLVPR34 +5 38% 40.2 3.3 −0.72
17
FK-21 FKRIVQRIKDFLRNLVPRTES37 +4 38% 38.67 3.36 −0.59
18
KR-20 KRIVQRIKDFLRNLVPRTES37 +4 35% 37.88 3.68 −0.76
GI-20 13GIKEFKRIVQRIKDFLRNLV32 +4 45% 45.07 2.47 −0.225
13
IG-19 IGKEFKRIVQRIKDFLRNL31 +4 42% 41.61 2.82 −0.46
GF-17 G17FKRIVQRIKDFLRNLV32 +5 47% 41.28 2.47 −0.094
17
FK-16 FKRIVQRIKDFLRNLV32 +5 50% 40.37 2.69 −0.075
17
FK-13 FKRIVQRIKDFLR29 +5 46% 35.69 3.48 −0.44
17
FK-12 FKRIVQRIKDFL28 +4 50% 33.54 2.53 −0.1
18
KR-12 KRIVQRIKDFLR29 +5 41% 32.11 4.02
Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37

−0.71
1
LL-12 LLGDFFRKSKEK12 +2 33% 26.35 2.81 −0.93
9
SK-12 SKEKIGKEFKRI20 +3 25% 19.97 3.36 −1.38
26
DF-12 DFLRNLVPRTES37 0 33% 26.96 3.42 −0.67
19
RI-10 RIVQRIKDFL28 +2 50% 31.1 2.78 −0.01
Calculated based on the APD website (http://aps.unmc.edu/AP/prediction/prediciton_main.php). Pho%, hydrophobic content; HP, hydrophobicity calculated using the Fisher
peptide-analyzing tool website https://www.thermofisher.com/us/en/home/life-science/protein-biology/peptides-proteins/custom-peptide-synthesis-services/peptide-analyzing-
tool.html; Boman index, a term renamed in the APD database based on protein-binding potential (Boman 2003); GRAVY, grand average of hydropathy, which is calculated by
summing the hydropathy value of each residue and divided by the length of the peptide (Kyte and Doolittle 1982); underlined residues are either artificial or order swapped
217
218 G. Wang et al.

12.2 Processing of the Human protect against the S. aureus infection (Zhang
Cathelicidin Gene Product et al. 2015). However, the exact peptide sequence
has not been elucidated.
12.2.1 Mature Peptide LL-37

Human cathelicidin is expressed as an 18 kDa 12.2.4 T

 LN-58 from a Diseased Skin
precursor protein (hCAP-18). It was discovered State
in 1995 by three laboratories based on the
highly conserved “cathelin” domain (Cowland It is fascinating that hCAP-18 can also be cleaved
et al. 1995; Larrick et al. 1995; Agerberth et al. upstream of the LL-37 sequence into another lon-
1995). One of the groups predicted the mature ger mature peptide with 58 amino acids
form as FALL-39 by comparison with the pig (Murakami et al. 2017). TLN-58 is isolated from
cathelicidin PR-39, a 39-residue peptide rich in the lesion vesicle of palmoplantar pustulosis, a
amino acids P and R (Agerberth et al., 1995). diseased skin condition on the palms and soles.
The mature form was established as a 37-resi- Similar to human LL-37, this peptide form can
due peptide LL-37 after its isolation from gran- upregulate IL-17C, IL-8, IL-23, IL-1α, and IL-1β
ulocytes (Gudmundsson et al. 1996). In mRNA and protein expression in normal human
neutrophils, hCAP-18 is processed to the anti- keratinocytes.
microbial peptide LL-37 (a 37-residue peptide These findings provide additional evidence for
starting with two leucines) by extracellular the idea of one cathelicidin gene and multiple
cleavage with proteinase 3 (Sorensen et al. peptides (Wang 2014). It is likely that the single
2001). Thus, the LL-37 pathway is composed cathelicidin gene can be translated and processed
of the precursor hCAP-18, the cathelin-like in other manners yet to be discovered. Once a
domain, LL-37, and its further cleaved frag- link is established between the cathelicidin form
ments under natural conditions. Interestingly, and disease, our detection of a particular form of
the precursor of human LL-37 can also be human cathelicidin may serve as a biomarker for
cleaved into alternative forms below. that disease.

12.2.2 ALL-38 from the Human 12.3  hysical and Structural Basis
P
Reproductive System of LL-37 in Targeting
Bacterial Membranes
AMPs appear to play a role in sperm fertilization.
During sexual intercourse, hCAP-18, along with 12.3.1 Bacterial Recognition
sperm, is injected into the vagina where it is via Electrostatic Interactions
cleaved into ALL-38 under acidic conditions by
gastricsin. ALL-38, with one more alanine at the Cationic LL-37 is active against a broad range of
N-terminus than LL-37, showed similar antibac- Gram-positive and Gram-negative pathogens
terial activity (Sorensen et al. 2003). (updated in Fig. 12.1). It is proposed that this lin-
ear peptide, like other cationic AMPs, targets
anionic bacterial membranes via the carpet or
12.2.3 An Uncharacterized toroidal pore model (Oren et al. 1999; Henzler
Alternative Form from Fat et al. 2003; Lee et al. 2011). The membrane tar-
Cells geting of LL-37 is determined by its sequence
property. According to the APD (Wang and Wang
A recent exciting discovery is that fat cells also 2004), LL-37 (M. Wt. 4493.312 and molecular
participate in host defense. Adipocytes can formula C205H341N59O53) has a net charge of +6
release a mature peptide longer than LL-37 to (i.e., sum of 6 Lys, 5 Arg, 2 Asp, and 3 Glu),
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 219

Fig. 12.1 Antibacterial activity of human antimicrobial peptide LL-37. Updated based on Vandamme et al. (2012) and
Durr et al. (2006), the antimicrobial peptide database (http://aps.unmc.edu/AP) and the PubMed

allowing it to recognize the negative charge on electrostatic attraction is the first step of peptide-­
bacterial surfaces, rather than mammalian cells. bacteria interaction. This is because bacterial
While bacterial membranes are rich in anionic surface modification to decrease negative charge
PGs, mammalian cell membranes contain mainly can reduce peptide activity. For instance, modifi-
zwitterionic phosphorylcholines (PCs) (Wang cation of lipid A with glucosamine confers resis-
et al. 2014a, b). This preference of bacterial tance to LL-37 (Shah et al. 2014). In the case of
membranes is supported by in vitro studies using S. aureus, knockout of the gene for alanylation
model membranes (Sevcsik et al. 2008). of teichoic acid of the cell wall or lysylation of
The negative charge on the bacterial surface phosphatidylglycerols (PG) in the inner mem-
originates from a different basis: an outer mem- brane makes the bacterium more susceptible to
brane for Gram-negative bacteria but a cell wall cationic LL-37 (Saar-Dover et al. 2012; Peschel
for Gram-positive bacteria. In the outer mem- and Sahl 2006). Likewise, citrullination of
branes of Gram-negative bacteria, anionic lipo- LL-37 can make it less positively charged and
polysaccharides (LPS) are dominant. They are reduces its LPS neutralization (Koziel et al.
composed of a polysaccharide moiety and a lipid 2014). Host Toll-like receptors (TLRs) play an
A. The cell wall of Gram-positive bacteria com- important role in recognizing these bacterial
prises peptidoglycan and lipoteichoic acids components. TLR2 is required to respond to cell
(LTA) that maintain cell integrity. These outer wall preparations of Gram-positive pathogens,
layers confer negative charges to the surface and whereas TLR4 is involved in recognition of LPS
are important targets for cationic AMPs. Such an (Takeuchi et al. 1999).
220 G. Wang et al.

12.3.2 Challenges for Structural protein by enzyme digestion. The aggregation of

Studies of LL-37 LL-37 might have blocked the enzyme cleavage
site. Fortunately, the fusion protein was success-
Positive charge alone, however, may not be suffi- fully cleaved by formic acid (Li et al. 2006b). By
cient for membrane binding of AMPs. The mem- making use of the oligomerization property of
brane targeting of LL-37 is determined by its LL-37 (Li et al. 2007), we improved the peptide
three-dimensional structure. Circular dichroism yield and obtained additional 15N-labeled or 15N,
(CD) and FT-IR analysis indicate a helical confor- 13
C-labeled peptides. These labeled peptides in
mation, leading to an immediate classification of complex with deuterated SDS micelles provided
this human peptide into the helical family. In par- the needed sample stability for recording a suite of
ticular, an amphipathic helix of this cationic pep- triple-resonance NMR spectra, such as HNCACB,
tide, with distinct hydrophobic and hydrophilic CBCA(CO)NH, HNCO, and HNCA (Kay et al.
surfaces, facilitates its interaction with anionic 2011). The 3D structure of LL-37 determined by
bacterial membranes (Oren et al. 1999). However, 3D NMR reveals a long amphipathic helix (resi-
it has to wait for the solution nuclear magnetic dues 2–31) and a C-terminal tail (residues 32–37)
resonance (NMR) spectroscopy to determine the (Fig. 12.2). To validate the structure, we measured
atomic structure of LL-37 that informs us where the ps-ns peptide backbone dynamics using a
the helix starts and ends and whether there is a 15
N-labeled LL-37 sample. The dynamics data
helix break upon membrane binding. It took years indicate the helical region is rigid, while the
of work to complete the 3D structure of LL-37 due C-terminal tail is mobile, fully consistent with the
to multiple challenges. The first challenge was the structure (Wang 2008). There are also other struc-
complex nature of bacterial membranes. tures for LL-37 and its fragments. Additional
Consequently, solution NMR studies were con- information can be found in other articles (Wang
ducted in the presence of membrane-mimetic 2008; Wang et al. 2014b).
micelles such as anionic sodium dodecyl sulfate
(SDS). The deuteration of micelles simplified the
NMR spectra and allowed us to focus only on the 12.3.3 S
 tructural Basis of Bacterial
peptide signals. As SDS has a head group different Membrane Binding of LL-37
from that of the anionic lipid in bacteria, we also
explored the possible use of a series of short-chain We also asked whether the high-resolution struc-
PGs for structural studies of AMPs (Wang et al. ture determined in SDS micelles could be applied
2004; Keifer et al. 2004; Wang 2006, 2007, 2008). to bacterial membranes. First, we used dioctanoyl
The second challenge was spectral resolution. The phosphatidylglycerol (D8PG) for NMR studies
first solution NMR studies revealed the need of 3D because it has the same head group as the major
NMR for structural determination of LL-37 bound anionic lipid in bacteria (Wang et al. 2004; Keifer
to SDS micelles (Li et al. 2006a). The separation et al. 2004). Based on all the NMR data, the
of the overlapped cross peaks onto numerous 2D structure of LL-37 in complex with D8PG is the
planes along the 15N or 13C dimension of 3D NMR same as that bound to SDS micelles, implying
spectra yielded the needed spectral resolution for a that the detergent/lipid head groups play little
complete assignment of the LL-37 signals and role (Wang 2008). Because D8PG is not deuter-
subsequent NOE assignments. The third challenge ated, it enabled us to observe intermolecular
was the requirement of establishing a bacterial NOE cross peaks between peptide and lipid as
expression system to produce stable isotope- well. In particular, aromatic F5, F6, F17, and F27
labeled LL-37 required for 3D NMR studies. The as well as basic R23 of LL-37 show direct inter-
toxicity of LL-37 to the expression host made it actions with the phosphatidylglycerol, confirm-
necessary to express the peptide as a fusion pro- ing their important role in bacterial membrane
tein. Another challenge was the difficulty to binding (Wang 2008). The significance of R23 in
release the recombinant peptide from the fusion interaction with bacteria was initially observed
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 221

Fig. 12.2 3D structure of human innate immune peptide stick view of the LL-37 structure with hydrophobic and
LL-37 bound to bacteria membrane-mimetic micelles hydrophilic residues selectively labeled. In both views,
determined by 3D NMR (PDB ID: 2K6O). (a) A back- hydrophobic amino acids are in green. A discontinuation
bone view of the long helix corresponding to residues of the hydrophobic surface at S9, rather than the helix, is
2-31 of LL-37 with the C-terminal tail disordered. (b) A thus evident (Wang 2008)

using the central fragment of LL-37 (Wang 2007) LPS (Wang 2008). In addition, S9 is also impor-
and later confirmed by alanine scan (Wang et al. tant for antibacterial activity since an effort to
2012b) as well as the lysine/arginine swap (Wang render the hydrophobic surface continuous
et al. 2018). Second, we also studied the interac- reduced the peptide activity (Wang et al. 2012a).
tion of LL-37 with E. coli LPS. While the signals The 3D structure determined by 3D NMR also
for the helical region are absent, the signals for laid a foundation for discovering active frag-
the C-terminal tail are evident. Thus, the disor- ments within LL-37.
dered tail of LL-37 observed in micelles recurs
when bound to LPS. We conclude that human
LL-37 utilizes the hydrophobic surface of a long 12.4 Discovery of Antibacterial
amphipathic helix (residues 2–31) to interact Regions Within LL-37
with bacterial membranes for antimicrobial kill-
ing without the need of the C-terminal tail. This section highlights LL-37 fragments gener-
Because the SDS-bound structure reflects the ated under natural conditions and in laboratories.
bacterial membrane-bound form, the 3D NMR-­ The natural conditions include protease ­processing
derived structure can be used to explain peptide from both the host and pathogens. Peptides are
LPS-binding and activity data. The synergistic also chemically synthesized in laboratories to
binding to LPS by an ovine cathelicidin SMAP-­ understand sequence-activity relationship.
29 can be explained by a helix break caused by a
proline (Tack et al. 2002). This is not the case for
LL-37 (Turner et al. 1998) because of a lack of 12.4.1 S
 kin Processing of Human
proline in the middle and the continuity of the LL-37
helix. However, S9 separates the hydrophobic
surface of the long helix of LL-37 into two The Gallo lab studied the protease cleavage of
domains (Fig. 12.2), providing a novel mecha- LL-37 in human skin by kallikreins. They found
nism for synergistic binding of human LL-37 to multiple fragments of LL-37, including both active
222 G. Wang et al.

and inactive forms. Some examples are KR-20, long with 37 amino acids. To reduce the synthesis
LL-23, KS-30, and RK-31 (Table 12.1). cost, it is necessary to trim unwanted regions.
Interestingly, KS-30 and RK-31 are more active Two major methods are used to locate the active
than their parent LL-37 in bacterial killing, indi- regions of LL-37. The first method is library
cating the N-terminal region is less important screening. There are numerous studies with a
(Murakami et al. 2004). This processing may be a goal of locating the active region of LL-37. The
natural regulatory mechanism that enhances pep- sequence relationship of these peptides with
tide antimicrobial potency but reduces unwanted LL-37 is summarized in Table 12.1. For example,
immune responses via the intact molecule. This Braff et al. (2005) made a peptide library to iden-
mechanism further enriches the molecule reservoir tify the antibacterial, antifungal, and antiviral
of human cathelicidin-based defense line. Future regions. Two long peptides, KS-30 and RK-31
studies may elucidate the details of this skin regu- (Table 12.1), are more potent than LL-37. Nell
latory mechanism of human cathelicidin. et al. (2006) identified a 24mer peptide P60 (i.e.,
residues 13-36 of LL-37 in Fig. 12.3; IG-24 in
Table 12.1). A peptide P60.4 with the C-terminal
12.4.2 Pathogen Degradation four-residues changed from PRTE to RPLR has
of LL-37 antimicrobial activity, LPS and LTA neutraliza-
tion ability comparable to LL-37. Kanthawong
As a resistance mechanism, there are multiple et al. (2010) found IG-19, RK-25, and LL-31
pathogen proteases that can degrade human (Table 12.1) against Gram-negative Burkholderia
defense peptide LL-37. However, only in a few pseudomallei both in planktonic and biofilm
cases were the resultant fragments documented. forms. Among these peptides, LL-31 is most
Sieprawska-Lupa et al. (2004) studied the digested potent.
products of S. aureus proteases such as aureolysin The second method is structure-based. Based
and V8 proteases. Aureolysin cleaved LL-37 on the helix-forming propensity of each amino
between R19-I20, R23-I24, and L31-V32 peptide acid along the sequence, Sigurdardottir et al.
bonds, leading to the instant inactivation of LL-37. (2006) found GKE-21 with reduced cytotoxicity
In contrast, the V8 protease cleaves the E16-F17 than LL-37. Li et al. (2006a) identified multiple
peptide bond of LL-37 (Fig. 12.3), leading to the LL-37 peptides based on NMR structural studies.
accumulation and isolation of FK-21 (Table 12.1), In that study, LL-37 was initially split into LL-12
a fragment corresponding to residues 17-37 of and IG-25 for structural studies (Table 12.1).
LL-37. Interestingly, FK-21 is more potent than While LL-12 is inactive, IG-25 is antibacterial.
LL-37 in killing E. coli, B. subtilis, P. aeruginosa, The study of a micelle-bound form of IG-25 led
and E. faecalis. This is understandable since to the discovery of the major antimicrobial pep-
FK-21 contains the major antimicrobial region tide FK-16 of LL-37 by using the NMR-trim
FK-16 discovered by NMR (Li et al. 2006a). In technology that removes nonessential membrane-­
another study, Rapala-Kozik et al. (2015) isolated binding regions, such as the disordered C-terminal
an intermediate peptide LL-25 (Table 12.1) from tail (Li et al. 2006a). Compared to the library
the C. albicans cleavage of LL-37. LL-25 appears approach that requires the synthesis of at least
to have a different immune modulating role com- dozens of peptides, this NMR technology is effi-
pared to LL-37 (see below). cient since it arrived at the major antimicrobial
region using only a couple of synthetic peptides.
In particular, IG-25 is only one residue longer
12.4.3 Identification of Active than IG-24 found above via library screening
Regions Within Human LL–37 (Nell et al. 2006). An N-terminally glycine
in Laboratories appended version of FK-16 is called GF-17.
FK-16 and GF-17 have very similar antibacterial
One of the major barriers toward peptide thera- activity (Table 12.2). A comparison of the anti-
peutics is production cost. LL-37 is relatively bacterial activity of multiple peptides described
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 223

Table 12.2 Comparison of antibacterial activity of the LL-37-derived antimicrobial peptides FK-16, GF-17, and
17BIPHE2
MIC (μM)
Peptide S. aureus USA300 E. coli ATCC 25922 K. pneumonia ATCC 13883 P. aeruginosa PAO1
FK-16 3.1 ≥3.1 3.1 25
GF-17 3.1 6.25 3.1 25
17BIPHE2 3.1 6.25 3.1 6.25
Peptide sequences in Table 12.1. 17BIPHE2 is a peptide antibiotic designed based on GF-17 (Wang et al. 2014a)

Fig. 12.3 Host-pathogen interaction at the peptide level. core antimicrobial region of human LL-37 is also recog-
Listed are select LL-37 peptides discovered from struc- nized by bacteria for degradation or to release virulence
ture and library approaches (Li et al. 2006a, b; Nell et al. factors (see the text)
2006; Wang 2008; Wang et al. 2008). It is amazing that the

above confirmed GF-17 is most potent against DF-12 (net charge = 0, Table 12.1). This is not
methicillin-resistant Staphylococcus aureus surprising since most of these fragments do not
(MRSA) USA300 (Wang et al. 2014a). This have the minimal hydrophobicity set by KR-12,
result may be due to a sufficient hydrophobicity the minimal antibacterial peptide we found
of the major antimicrobial peptide of LL-37 (Wang 2008). In contrast, weak antibacterial
(Table 12.1). In addition, a balance in charged activity (MIC 50 μg/mL) was observed for cen-
and hydrophobic amino acids in the sequence led tral fragments KR-12 and FK-12 against S. epi-
to a near zero GRAVY value in Table 12.1. Our dermidis. Note that both KR-12 and FK-12
NMR studies also led to the identification of (Saporito et al. 2018) were derived from FK-13
FK-13, the core antibacterial region of LL-37 (Table 12.1), further validating the antimicrobial
corresponding to residues 17–29 (Li et al. 2006a). core region (Fig. 12.3) of LL-37 (Li et al. 2006a).
Using a series of shorter peptides, it was found
that the antibacterial region of FK-13 could be
further shortened to 12-residue KR-12, the small- 12.5  eptide Design Based
P
est antibacterial peptide of LL-37 (Wang, 2008). on Select LL-37 Peptides
KR-12 is active against Gram-negative E. coli
K12 (MIC 40 μM) but not Gram-positive S. The ESKAPE pathogens, including Enterococcus
aureus USA300. In 2018, Jessen lab made sev- faecium, Staphylococcus aureus, Klebsiella
eral 12-residue peptides along the LL-37 pneumoniae, Acinetobacter baumannii,
sequence. Their study showed no activity Pseudomonas aeruginosa, and Enterobacter spe-
(>100 μg/mL) for N-terminal fragments LL-12, cies, are responsible for the majority of hospital-­
SK-12, C-terminal fragments VQ-12, IK-12, and acquired infections (Boucher et al. 2009). For
224 G. Wang et al.

MRSA alone, the estimated total deaths are search for candidates with reduced binding to
already comparable to those caused by HIV-1/ plasma. SAAP-148, with such a property, is
AIDS (Klevens et al. 2007). Therefore, new anti- found to have antimicrobial and antibiofilm abil-
microbials are needed. There is also a high inter- ity against the ESKAPE pathogens, including
est in developing AMPs into therapeutic persisters, which cannot be killed by traditional
molecules (Zasloff 2002). Being linear, human antibiotics.
cathelicidin LL-37 is an important template. The
therapeutic implication of LL-37 is evident. First,
a lack of LL-37 in neutrophils may be responsi- 12.5.2 T
 he FK-16-Derived GF-17
ble for periodontal disease in patients with mor- Template
bus Kostmann (Pütsep et al. 2002). Second,
knockout of the homologous cathelicidin gene Using GF-17 as a template, we studied the role of
CRAMP makes the mice more susceptible to basic amino acids by alanine scan (Wang et al.
infection (Nizet et al. 2001). As a complement 2012b). One important finding is that the five
strategy, expression of additional cathelicidin basic amino acids in this peptide are not equal,
protects the skin from infection (Lee et al. 2005) but all involved in lipid clustering (Epand et al.
and restores bacterial killing in a cystic fibrosis 2009). In the case of membrane permeation, the
xenograft model (Bals et al. 1999). Third, topical R23A variant of GF-17 failed to cross either the
treatment of chronic wounds such as venous leg outer or inner membranes of E. coli, indicative of
ulcers with LL-37 (0.5 or 1.6 mg/mL) markedly an essential role of R23 for membrane damage
reduces the mean ulcer area without safety con- (Wang et al. 2012b), consistent with the observa-
cerns (Grönberg et al. 2014). Furthermore, LL-37 tion of intermolecular NOE cross peaks between
can inhibit biofilm formation probably by opso- R23 and D8PG by NMR spectroscopy (Wang
nization that enhances bacterial clearance 2007). As a conservative change, we also
(Overhage et al. 2008; Sol et al. 2013). In addi- swapped the positions between a pair of lysine
tion, LL-37 boosts immune response to clear and arginine in GF-17. Interestingly, the charged
pathogens (Overhage et al., 2014). Fourth, LL-37 swapped peptides showed reduced killing effi-
may also be induced to help eliminate pathogens ciency and increased cytotoxicity to human cells,
(van der Does et al. 2012; Jiang et al. 2013; revealing the evolutional significance of the
Schögler et al. 2016). These observations laid a native sequence of the peptide (Wang et al. 2017).
solid basis for peptide design based on LL-37 and Cytotoxicity also limits the applications of
its derived fragments. cationic AMPs. It is commonly observed that
peptide cytotoxicity results from high hydropho-
bicity (Zasloff 2002; Boman 2003). In the case of
12.5.1 IG-24 Derived P60.4 GF-17, the major antimicrobial region of LL-37
(Fig. 12.3), two approaches were utilized to
Serum/plasma binding is believed to be a key fac- reduce the peptide hydrophobicity. The first
tor that might have limited the systemic use of method is peptide truncation that led to the dis-
cationic AMPs. Based on the P60.4 template covery of KR-12 (Wang 2008). The second
originally identified from a library screen (Nell method is to incorporate D-amino acids. The
et al. 2006), OP-145 or P60.4Ac (Table 12.1) was structural basis for this has been elucidated for
initially obtained via further amino acid modifi- GF-17d3, which contains D-amino acids at posi-
cation, including N-terminal acetylation. OP-145 tions 20, 24, and 28 of GF-17 (Table 12.1). The
is effective against S. aureus clinical strains, but D-amino acids distorted the backbone of GF-17
its activity can be reduced in the presence of and caused non-coherent packing of side chains,
plasma (de Breij et al. 2016). Recently, de Breij leading to hydrophobic defects (Li et al. 2006a).
et al. (2018) made additional peptide mutants to A hydrophobic gap exists on the hydrophobic
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 225

surface of LL-23, explaining its poor antibacte- B28-16, but not Gram-positive pathogens we
rial activity (Wang et al. 2012a). tested (Wang et al. 2017). Alternatively, hydro-
Another hurdle in peptide development is pro- phobic truncation of GF-17 led to KR-12 that is
tease stability. Any peptide that is cleaved prior to only active against E. coli K12. Second,
bacterial killing is not useful. We used a library 17BIPHE2-3RA, where three arginines are
screen method to identify a peptide template with changed to alanines, is only active against Gram-­
stability to chymotrypsin. This was initially positive staphylococcal strains such as S. aureus
accomplished during antibacterial assays in USA300 and S. epidermidis 1457. 17BIPHE2-­
duplicated wells containing proteases; a stable 3RA, with a net charge of +2 and hydrophobic
peptide remains bacterial inhibition even in the content of 64%, resembles the database designed
presence of proteases. In this experiment, anti-MRSA peptide DFTamP1 that kills only
GF-17d3, with 3 D-amino acids incorporated, is Gram-positive pathogens (Mishra and Wang
active against E. coli K12, but not GF-17d1 (one 2012). These results underscore the importance
D-amino acid at position 20) or GF-17d2 (two of amino acid composition in determining the
D-amino acids at positions 20 and 24) containing peptide activity spectrum (Wang et al. 2018).
1-2 D-amino acids. This chymotrypsin-resistant
template, GF-17d3, was used to design peptide
analogs to combat resistant pathogens. One of the 12.5.3 FK-13 and KR-12 Templates
peptides, 17BIPHE2, with both F17 and F27
replaced with biphenylalanines, is potent, selec- NMR studies established a helical structure for
tive, and stable (Wang et al. 2014a). This is the KR-12 in complex with D8PG (Wang 2008).
first LL-37-derived peptide illustrated to kill the Understanding the role of each residue in KR-12
ESKAPE pathogens (MIC 3.1–6.2 μM) and can is important for peptide design. We investigated
inhibit biofilm formation in vivo. 17BIPHE2 the effect of a single alanine substitution of basic
showed a 50% hemolytic concentration (HL50) amino acids on KR-12 activity (Mishra et al.
between 150 and 220 μM depending on the types 2013). Consistent with our finding for GF-17 that
of blood cells used. Our further investigation of R23 and K25 are important for antimicrobial
17BIPHE2 confirmed its stability to chymotryp- action, KR-12R23A and KR-12K25A are less
sin digestion. In addition, it is not degraded after effective in binding to anionic D8PG (Mishra
digestion with pathogen protease S. aureus V8 or et al. 2013). We did not replace hydrophobic
fungal protease K, but can be cleaved by trypsin. amino acids because KR-12 is already the small-
In contrast, our recently designed Trp-rich pep- est, and such substitutions would lead to inactive
tide TetraF2W-RK with eight amino acids is peptides. This is indeed the case when we substi-
inherently stable to trypsin and S. aureus V8 but tuted I20, I24, or L28 of KR-12 into an alanine
can be cleaved by chymotrypsin (Mishra et al. (Wang 2010). Recently, Gunasekera et al. (2018)
2017a). However, TetraF2W-RK, synthesized in made a systematic amino acid substitution within
D-amino acids, is stable to both trypsin and chy- KR-12. It seems the culture media and bacterial
motrypsin. Peptide stability to such proteases strains used for antimicrobial assays can substan-
from the digestive system can be important for tially influence the MIC values and make the
future engineering peptides as oral drugs. results incomparable with the previously pub-
With the idea of personalized medicine, it may lished results (Mishra et al. 2013).
be necessary to design peptides with a desired Because of its low cytotoxicity and short
activity spectrum. Recently, we found it possible length, KR-12 (Fig. 12.3) becomes an attractive
to convert the broad-spectrum GF-17 into narrow-­ template for peptide engineering. Jacob et al.
spectrum AMPs. First, by partially incorporating (2013) generated several variants of KR-12 by
D-amino acids, we obtained GF-17d3 (Li et al. increasing basic and hydrophobic residues at
2006a), which is only active against Gram-­ positions 27, 26, 22, 23, 25, and 18. KR-12-a1 to
negative E. coli ATCC 25922 and A. baumannii KR-12-a6 contains 1–6 changes of the listed resi-
226 G. Wang et al.

dues of KR-12. KR-12a1 (a F27W variant) strated the antibiofilm ability of the engineered
showed MIC values in the range of 1–8 μM using peptide 17BIPHE2 in a mouse catheter model
six Gram-positive and negative bacteria. Of note, (Wang et al. 2014a). As an independent path, a
additional substitutions did not increase peptide European group demonstrated antibiofilm ability
activity substantially against S. aureus, S. of OP-145 and SAAP-148 (de Breij et al. 2016,
typhimurium, B. subtilis, and S. epidermidis, and 2018). However, combination treatment may be
in the case of KR-12-a6 with all six changes necessary and a natural choice for preformed bio-
(K18L, Q22K, R23L, K25L, D26K, and F27W), films that are difficult to remove. Our recent
antibacterial activity of the peptide against E. coli antipseudomonal biofilm results are summarized
and P. aeruginosa actually reduced. Meanwhile, here: (1) Antibiotics are not effective, while
KR-12-a5 and KR-12-a6 became more hemo- AMPs are active, underscoring the importance of
lytic, indicating the K25L change is not favor- such peptides as a new antibiofilm agent. (2)
able. Of note, a topical use of KR-12-a2 to treat Combined use of 17BIPHE2 with antibiotics is
MRSA otorrhea found no hearing loss or cochlear more effective to remove the P. aeruginosa bio-
damage in guinea pigs (Sung et al. 2017), while films. (3) Early treatment prior to biofilm matura-
KR-12-a5 is potent against oral pathogens tion (<10 h) is advantageous since monotherapy
(Caiaffa et al. 2017). A similar mutational study of either antibiotic or peptide is effective (Mishra
was also conducted using FK-13 (Rajasekaran and Wang 2017). Based on these observations,
et al. 2017). Da Silva et al. (2017) also utilized there is now a growing interest in developing bio-
KR-12 as a template to design antibiofilm pep- film prevention approaches.
tides against oral pathogens such as Streptococcus
mutans. Substitution of I20 with W and append-
ing KAEK at the C-terminus of KR-12 led to a 12.6.2 A
 ntibiofilm Coating of LL-37
more potent peptide against biofilms. These stud- Peptides
ies indicate a great potential of KR-12 as a tem-
plate or a sequence unit for peptide design. An attractive antimicrobial strategy is to coat
human cathelicidin LL-37 to the surfaces of med-
ical devices to prevent infection, especially when
12.6 Potential Applications biofilms form. In particular, peptide surface
of LL-37 Peptides immobilization confers some advantages such as
a local high peptide concentration, reduced cyto-
12.6.1 Antibiofilm Peptides toxicity, and increased host cell adhesion. This
section describes covalent coupling of LL-37 and
Many pathogens are able to form biofilms, where its peptides (Table 12.3). There are multiple fac-
a surface-anchored bacterial community shares a tors that affect the antimicrobial activity of
tower-like structure decorated with polymers immobilized peptides. These include the types of
(e.g., polysaccharides and DNA). Such biofilms, materials (metals or polymers), the linker or
once formed on medical devices, are very chal- spacer, and the site for peptide coupling ­(terminus
lenging to remove. Therefore, much research vs. side chain). There is no consensus as to the
effort of AMPs was oriented toward the develop- nature of linkers (short vs. long).
ment of potent antibiofilm agents and infection-­ The full-length LL-37 molecule, with a cyste-
free implants to combat these pathogens. Biofilm ine added at the N-terminus, was first immobi-
inhibition by human LL-37 at a low concentra- lized on a titanium surface via the maleimide
tion was initially observed by the Hancock Group chemistry. Gabriel et al. (2006) found that site-­
(Overhage et al. 2008). As one advantage of pep- specific coupling of LL-37 with a flexible hydro-
tide design, however, both GF-17 and 17BIPHE2 philic poly(ethylene glycol) (PEG) spacer is
possess antibiofilm capability superior to LL-37 better than randomized coating. The LL-37-­
(Mishra et al. 2016). In addition, we have demon- coated surface shows antimicrobial activity
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 227

Table 12.3 Covalent immobilization of LL-37 and its fragments onto various substrates
Substrate/ Activity
Study Peptide surface Chemistry Spacer organism Ref
1 LL-­37 Titanium Surface hydroxylation, PEG EC (K12) Gabriel et al.
APTES amine generation, (2006)
and peptide coupling via
maleimide chemistry
2 LL-­37 Nickel NPs Dielectric barrier Polyacrylic acid EC Chen et al.
discharge glow plasma (DH5α) (2009)
fluidized bed (GPFB) for
deposition of nanolayer
polyacrylic acid (PAA) on
NPs. And peptide reaction
using EDC/NHS
3 LL-­37 Etafilcon A Carbodiimide reaction Carboxylic acid PA Dutta et al.
contact lens linking surface acidic (2016)
(pHEMA) group to the peptide
amino group
4 LL-­37 Amino saline LL-37 was reacted with APTMS CA, CG Niemirowicz
coated APTMS-coated NPs via and CT et al. (2017)
magnetic amidation
nanoparticle
5 IG-­25 Fluorous Perfluorocarbon chains Perfluorocarbon chains PA Santos et al.
slide and with alkynyl group (C8F17) (PAO1) (2013)
Fluoroperm reacted with peptide-­
60 contact containing azido group
lens via Cu-catalyzed
azide-alkyne
cycloaddition reaction
(click chemistry)
6 FK-­16 Titanium Similar to LL-37/Ti 6-Maleimidohexanoic ESKAPE Mishra et al.
acid pathogens (2017a)
7 KR-­12 Silk fibroin Carboxylic acid on the SF Aminoethyl maleimide SA, SE, Song et al.
(SF) nanofiber reacted with EC, and (2016)
nanofiber aminoethyl maleimide via PA
the EDC/NHS reaction.
Peptide was conjugated
with the maleimide-Cys
reaction
8 KR-­12 Titanium The hydroxylated surface 3-(2-aminoethylamino) SE Nie et al.
was amine-functionalized propyltrimethoxysilane (2016)
and reacted with Asp of
peptide for coupling
EC Escherichia coli, SA Staphylococcus aureus, PA Pseudomonas aeruginosa, SE Staphylococcus epidermidis, CA C.
albicans, CG C. glabrata, CT C. tropicalis

against E. coli. Recently, LL-37 has also been latory activity such cell migration, and (6) treat-
immobilized on other types of surfaces, including ment of wounds in a mouse model.
nanoparticles (Gustafsson et al. 2010; Cassin Several fragments (IG-25, FK-16, and KR-12)
et al. 2016; Niemirowicz et al. 2017; Comune initially reported by us (Li et al. 2006a; Wang
et al. 2017). The chemistries are summarized in 2008) have also been surface immobilized.
Table 12.3. These studies illustrated (1) antifun- Santos et al. (2013) immobilized IG-25
gal activity of LL-37 against C. albicans, (2) (Table 12.1) on fluorous thin films and fluorosili-
anti-adhesion properties, (3) LPS binding, (4) no cone contact lens with activity against an ocular
toxicity to mammalian cells, (5) immune modu- pathogen P. aeruginosa (Table 12.3). KR-12 was
228 G. Wang et al.

also covalently immobilized via a site-specific can only be sent to a containment facility and
maleimide chemistry on electrospun silk fibroin wait for miracle medicine. The development of
nanofiber. The surface could inhibit S. aureus, S. new vaccines is slow. Therefore, alternative anti-
epidermidis, E. coli, and P. aeruginosa. Compared viral strategies are clearly needed. Current rise in
to the free peptide described above, this surface-­ HIV patients with coinfections has required
coated KR-12 appeared to have gained antibacte- researchers to revisit AMPs with a hope to over-
rial activity. In addition, the immobilized peptide come ever-increasing cases of antibiotic resis-
suppressed the LPS-induced TNF-α expression tance and to increase the immune response status.
of monocytes (RAW264.7), helped the prolifera- It is interesting to explore how nature has devised
tion of fibroblasts and keratinocytes, and pro- AMPs for protection against various viruses. As
moted the differentiation of keratinocytes with of June 2018, the APD registered 182 antiviral
enhanced cell-cell attachment (Song et al. 2016). peptides (Wang et al. 2016). While a systematic
Likewise, KR-12 was also immobilized on the review on HIV inhibitory AMPs (Wang 2012)
titanium surface (Nie et al. 2016), which shows can be found online, this section focuses on anti-
antimicrobial and antibiofilm activities against S. viral effects of human cathelicidin LL-37. Up to
epidermidis. Importantly, the peptide-coated sur- date, LL-37 has been demonstrated to have inhib-
face is able to increase the adhesion and prolif- itory effects against at least ten types of viruses,
eration of human bone marrow mesenchymal including herpes simplex virus (HSV; Yasin et al.
stem cells (Nie et al. 2016). To obtain a surface 2000), smallpox vaccinia virus (Howell et al.
with broad-spectrum activity, we recently immo- 2004), HIV-1 (Bergman et al. 2007), respiratory
bilized FK-16, the major antimicrobial peptide of syncytial virus (RSV; Tian et al. 2011), varicella
LL-37, on the titanium surface using a similar zoster virus (VZV; Crack et al. 2012), human
maleimide chemistry (Mishra and Wang 2017). adenovirus (Uchio et al. 2013), influenza A virus
FK-16, with a higher coating density than LL-37 (IAV; Tripathi et al. 2013), dengue virus type 2
on the titanium surface, shows potent activity (DENV-2; Alagarasu et al. 2017), human rhinovi-
against the ESKAPE pathogens but is nontoxic to rus (HRV; Sousa et al. 2017), and Zika virus (He
human erythrocytes and epidermal keratinocytes et al. 2018). To facilitate our discussion, these
HaCaT cells. Significantly, the FK-16-coated LL-37 inhibited viruses are classified in
surface is able to inhibit biofilm formation of S. Table 12.4 based on the type of nucleic acids
aureus USA300 (initial inoculation at 103 CFU) (DNA/RNA) and whether they are enveloped.
for up to 72 h. A lower CFU (103) used here Thus, LL-37 has an inhibitory effect on both
should be medically more relevant considering enveloped and non-enveloped viruses.
the sterile condition of the surgery room. This It is natural to ask whether the antibacterial
study provides a proof-of-concept example for segments derived from LL-37 also effectively
generating a broad-spectrum antimicrobial sur- inhibit viruses. While fragments from either the
face to prevent bacterial adhesion and biofilm N-terminus or C-terminus are inactive, a central
formation on medical implants. fragment of LL-37 is active against HIV-1 (Wang
et al. 2008). Different from the anti-MRSA case,
GI-20 (Fig. 12.3) has the highest therapeutic
12.6.3 Antiviral Peptides index, indicating that the central region of LL-37
is also important to inhibit viruses. In collabora-
Viral infection is life threatening and has raised tion with ImQuest Biosciences, we also identi-
concerns from the public. This ranges from the fied the minimal anti-HIV peptide of LL-37.
well-known human immunodeficiency virus type Different from the bacteria case, the LL-37 core
1 (HIV-1) to the recent Zika and Ebola viruses in peptide FK-13 retains anti-HIV activity but not
the news. Often, we do not have therapeutic mol- KR-12. This fact implies a significant role of
ecules to treat such outbreaks. Infected patients F17 in inhibiting HIV-1. There are also other dif-
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 229

Table 12.4 Classification of LL-37-inhibited viruses

RNA viruses DNA viruses
Naked Enveloped Naked Enveloped
Rhinoviruses Human immunodeficiency virus type 1 (HIV-1) Adenovirus Herpes simplex virus (HSV)
Influenza A virus (IAV) Vaccinia virus (smallpox)
Respiratory syncytial virus (RSV) Varicella zoster virus (VZV)
Zika virus (ZIKV)
Dengue virus type 2 (DENV-2)

ferences from the antibacterial case. In particu- Langerhans cells (the mucosal epithelium resi-
lar, after reversal of the FK-13 sequence or dent dendritic cells), thereby enhancing HIV
incorporation of D-amino acids into GF-17, the infection (Ogawa et al. 2013).
peptides remain active against bacteria but not LL-37 peptides also have inhibitory effects on
the virus, implying a different molecular target. other RNA viruses (Table 12.4). LL-37 can
It appears that LL-37 and its peptides IG-25 and directly disrupt enveloped influenza A virus
FK-16 are able to inhibit HIV-1 reverse tran- (Tripathi et al. 2013). In the case of the pandemic
scriptase in vitro at IC50 of 15, 7, and 70 μM, H1N1 strain of 2009 (A/California/04/09/H1N1
respectively (Wong et al. 2011). The weak or “Cal09”), LL-37 is inactive. However, GI-20,
enzyme inhibition activity of these peptides a central LL-37 fragment, retains anti-IAV activ-
observed here (20–30% inhibition at 100 μM) ity against this strain (Tripathi et al. 2015b). This
does not sufficiently explain anti-­HIV activity observation indicates the advantage of LL-37
observed below 1 μM. Therefore, additional reengineering.
mechanistic studies are needed to better under- Mosquito-borne Zika is another enveloped
stand peptide activity. It is clear that the sequence RNA virus first isolated in Uganda in 1947
requirement for inhibiting HIV-1 differs from near the Zika forest. After its outbreak in 2007,
that for bacterial inhibition. Other cathelicidins it became a rapidly emerging public health
can also be useful. While BMAP-27 is toxic, a threat. Although clinical infection is frequently
C-terminal truncated peptide is not (Skerlavaj mild, significant neurological manifestations
et al. 1996). BMAP-18 is also demonstrated to have been demonstrated in infants born to Zika
be inhibitory to HIV-1 (Wang et al. 2008). Future virus (ZIKV)-infected mothers (McArthur
studies may further define the mechanism of 2017). Currently, there is no drug to treat ZIKV
action and evaluate the therapeutic potential of infection. Although vaccines are under active
these anti-HIV peptides in animal models. development, other effective countermeasures
Considering other benefits of LL-37 such as may also be considered. Recently, the efficacy
spermicidal effects and selective killing of invad- of LL-37 and its derived peptides (e.g., GI-20
ing pathogens without damaging commensal and GF-17) against ZIKA has been demon-
bacteria (Tanphaichitr et al. 2016), these LL-37 strated in vitro, whereas RI-10 (Fig. 12.3) is
peptides may be promising candidates as topical ineffective (He et al. 2018). Further character-
spermicides/microbicides. However, further ization reveals that GF-17 (Fig. 12.3) can
studies are required to resolve the complication directly inactivate this virus and work via the
from coinfected viruses such as HSV-2 that interferon pathway. In addition, 17BIPHE2, an
appears to alter the defense role of LL-37 into an engineered version of GF-17, also inhibits
offense role. The mechanistic studies revealed ZIKA effectively. However, its antiviral effect
that LL-37 produced by HSV-2-infected epithe- is reduced when R23 is altered to ornithine,
lial cells upregulates HIV receptors (CD4 and indicating the important role of this arginine in
CCR5) on the surface of monocyte-derived inhibiting ZIKA.
230 G. Wang et al.

Other labs also searched active antiviral liposomes (size 106.8 ± 10.1 nm, shelf-life stabil-
regions of LL-37 against respiratory syncytial ity >1 year) are found to be superior to its free
virus (RSV), which is responsible for lower form in protecting keratinocytes from RSV infec-
respiratory tract infections of children. A low tion without displaying cytotoxicity even at
level of human cathelicidin is directly correlated 400 μM (Ron-Doitch et al. 2016). In future stud-
with human RSV infection. Harcourt et al. (2016) ies, the central fragment of LL-37 may be formu-
found a better anti-RSV effect when LL-37 is lated in the same manner. Other preventative
used prophylactically (treat before infection) measures may include the administration of vita-
than therapeutically (treat after infection). Currie min D and increased exposure to sunlight to
et al. (2016) also compared the effect of treat- boost the expression of LL-37. However, the
ment time. They found that co-administration of interaction of LL-37 with combustion-derived
LL-37 with virus clearly protects animals better carbon nanoparticles (especially in winter) can
against infection either when the peptide is compromise peptide antiviral and antibacterial
treated before or after viral colonization. It is activity, making immunocompromised people
established that LL-37 can directly damage the more susceptible to infection in highly polluted
viral envelope, disrupt virus particles, and inhibit environments (Findlay et al. 2017).
infection of epithelial cells in vitro (Currie et al.
2016). The direct anti-RSV effect of LL-37 stim-
ulates the interest in identifying the active region 12.6.4 Antifungal Peptides
of LL-37. As we observed in the antibacterial
study (Li et al. 2006a), Tian et al. (2011) found High mortality and morbidity rates due to inva-
that the N-terminal fragment LL-12 is inactive sive mycosis have been increasing over the last
while the C-terminal fragment IG-25 is. 20 years. Medically significant pathogenic fungi
Moreover, among the four 22mer LL-37 peptides (~300 species) are almost always molds.
corresponding to residues 13–34 (IG-22), 14–35 Opportunistic fungal infections create therapeu-
(GK-22), 15–36 (KE-22), and 16–37 (EF-22), tic challenges, particularly in high-risk immuno-
EF-22 shows an anti-RSV effect similar to LL-37. compromised patients with AIDS, cancer, and
Currie et al. (2013) also showed that, while nei- those undergoing transplantation. In light of
ther the N-terminal fragment LL-22 nor the growing resistance to antifungal drugs, novel
C-terminal fragment EF-22 is inhibitory, a cen- medicine and treatment approaches are required.
tral peptide KI-22 (Table 12.1) is effective against The current APD registered 1067 antifungal
viral particles. Despite the conflicting results peptides (Wang et al. 2016). Cathelicidin
between Tian et al. (2011) and Currie et al. (2013) α-helical peptides have shown activity (BMAP-­
regarding EF-22, these studies also point at the 27 and BMAP-28 from cows) against Candida
important antiviral role of the central fragment of spp. and C. neoformans, but they are less active
LL-37 previously found based on bacteria (Li against filamentous fungi. Both LL-37 and a
et al. 2006a; Nell et al. 2006) and later demon- close mouse analog mCRAMP have a similar
strated against HIV-1 (Wang et al. 2008). It seems MIC range (15–20 μM) against C. albicans. In
that the central region of LL-37 plays a general one study, mCRAMP was induced by C. albi-
protective role against both bacterial and viral cans at the skin surface in a mouse model, dem-
infection (Fig. 12.3). onstrating that these peptides provide a natural
Because peptides have a relatively short half-­ barrier to fungal infection. In addition, LL-37
life, potential cytotoxicity, and non-specific inter- can also regulate the immune response to better
actions with cells, one possible way to improve clear fungal infection. Gallo and colleagues
the treatment outcome is to combine LL-37 with tested antifungal activity of LL-37 and its frag-
nanoparticles, which can be internalized, increas- ments against C. albicans and found that KS-30
ing the uptake of the peptide. LL-37 containing and RK-31 are more active (Braff et al. 2005).
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 231

Using fl
­ uorescein-­labeled peptides, den Hertog Chemotaxis is a recognized role of LL-37.
et al. (2006) investigated the cell location of There is no correlation between immune modu-
LL-37-derived peptides. While LL-37 and its lation (e.g., IL-8 release from keratinocytes)
C-terminally truncated peptide LL-31 and antimicrobial activity (Braff et al. 2005).
(Table 12.1) are found at the perimeter of C. Nell et al. (2006) also found that, although
albicans, the N-terminally truncated peptide P60.4 has antimicrobial activity comparable to
RK-31 (Table 12.1) enters the cytoplasm within LL-37, it loses its chemotactic ability, while P60
30 min. LL-25 can enter cells faster than RK-31. and LL-37 are nearly equivalent in inducing
It seems that both the N-terminal helix and the neutrophil migration. Thus, an alteration of the
central helix play a role in determining the pep- follow-­ up sequence behind the central helix
tide location on the cell perimeter since all the (Fig. 12.3) regulates chemotaxis. Chemotaxis of
phenylalanines are important bacterial mem- LL-37 is important for host defense against
brane anchors (Wang 2008). In addition, these pathogen invasion. As a counteracting strategy,
peptides can all induce lipid phase separation in fungi can cleave LL-37 using aspartic proteases.
fungal membranes. LL-25 is identified as an intermediate peptide
that shows a lower chemotactic activity to neu-
trophils than LL-37, reducing the recruitment of
12.6.5 Immune Modulating Peptides neutrophils to the infection sites (Rapala-Kozik
et al. 2015).
Human LL-37 also plays an important role in Interestingly, LL-37 can also be cleaved into
regulating the immune response (Scott et al. LL-23 in human skin (Murakami et al. 2004).
2002; Choi et al. 2012; Kahlenberg and Kapplan There is a clear difference between LL-37 and its
2013). This section highlights some differences N-terminal fragment LL-23 in immune regula-
between LL-37 and its derived peptides. The LPS tion (Wang et al. 2012a). Immune modulation by
neutralization property of LL-37 forms the basis LL-37 also plays a role in viral control. For
for its use to treat sepsis. Arginines are important example, LL-37 can inhibit the release of IL-8
in this neutralization as citrullination of LL-37 from neutrophils induced by IAV (Tripathi et al.
makes it inactive in a sepsis mouse model (Koziel 2014). Of note, the central peptide GI-20 is
et al. 2014). Using a designed peptide 17BIPHE2, equally effective in reducing IL-8 (Tripathi et al.
it is shown that an alteration of R23 to ornithine 2015a). LL-37 can also stimulate immune
slightly reduces LPS neutralization (Wang et al. response by binding to nucleic acids. In this pro-
2018). In addition, synergistic LPS binding of cess, LL-37 forms oligomers and serves as a car-
LL-37 requires two domains (See Fig. 12.2). This rier. A comparison of LL-37 with its fragments
explains the reduced LPS-binding ability of reveals that both the N- and C-terminal regions
LL-37 peptides without the N-terminal domain, are required in this process (Singh et al. 2014).
including the 24mer peptide P60 found by Nell This sequence requirement agrees with the NMR
et al. (2006) and IG-19 (Nan et al. 2012) study of free LL-37 at pH 7 that the entire region
(sequences in Table 12.1). Increasing basic/ of LL-37 is involved in the oligomerization into
hydrophobic amino acids and changing F17 and tetramers (Wang 2017).
F27 to W, however, enhance the LPS-binding It is interesting to note that there is also a
ability of IG-19. By binding to LPS, LL-37 sup- sequence difference required for antimicrobial
presses the LPS binding to receptors such as action and pathogen response. While FK-13
CD14 and TLR-4, thereby reducing the apoptosis (Table 12.1) is the minimal peptide to inhibit
of liver endothelial cells (Suzuki et al., 2011). In HIV-1, KR-12 (obtained by deleting the
addition, LL-37 also plays a role in LPS clear- N-terminal F17 of FK-13) retains antibacterial
ance. In this process, LL-37 enhances LPS uptake activity against E. coli (Wang 2008). Interestingly,
by liver cells via endocytosis (Suzuki et al. 2016). RI-10 (Table 12.1), obtained by deleting one resi-
232 G. Wang et al.

due from the N- and C-termini of KR-12, lost 12.6.6.1 Colon Cancers
antimicrobial activity (Wang 2008; He et al. Human cathelicidin LL-37 is expressed strongly
2018). However, RI-10 retains the minimal in normal colon mucosa but downregulated in
sequence information of LL-37 that triggers the colon cancer tissues. Kuroda et al. (2012) showed
bacterial two-component system. A direct inter- that LL-37 and a peptide analog FF/CAP18 sup-
action between LL-37 and the CSrRS receptor presses colon cancer cell (HCT116) prolifera-
increases the virulence factor expression of group tion. FF/CAP18 corresponds to residues 5–32 of
A streptococcus (GAS) (Velarde et al. 2014). LL-37 with residues E16 and K25 changed to
With the expansion in our investigation, other F. The peptide works by depolarization of the
functional sequence motifs of human LL-37 may mitochondrial membrane independent of the p53
emerge, further enriching our understanding of pathway. Ren et al. (2012) showed that treatment
this innate immune peptide. of colon cancer cells with LL-37 induces anionic
Taken together, there are different sequence phosphatidylserine (PS) exposure and DNA frag-
requirements for antimicrobial action and cell mentation, indicative of apoptosis. Previously,
response from both the host and pathogen sides. FK-16 is found to be the major antimicrobial and
Such sequence differences are determined by dif- anticancer peptide of LL-37 (Li et al. 2006a). It is
ferent molecular targets: usually bacterial mem- interesting that FK-16 shows a similar anti-colon
branes for antimicrobial activity but cell receptors cancer effect independent of caspase activation
for host immune stimulation and pathogen (Ren et al. 2013). Mechanistically, FK-16 causes
response. the upregulation of Bax and downregulation of
Bcl-2 by activating p53. As an alternative anti-
cancer strategy, Cheng et al. (2014) observed
12.6.6 Anticancer Peptides tumor size shrinking when cathelicidin-­
expressing adeno-associated virus was adminis-
Anticancer activity of AMPs was demonstrated tered intravenously into HT-29-derived
rather early (Ohsaki et al. 1992). AMPs with anti- subcutaneous tumors in nude mice.
cancer activity are discussed elsewhere in this
book. An updated list of anticancer peptides can 12.6.6.2 Gastric Cancer
be found in the APD (Wang et al. 2016). Human Helicobacter pylori is linked with gastric cancers
cathelicidin LL-37 is linked to cancers in differ- (Li and Perez Perez 2018). Hase et al. (2003)
ent manners. Its level is increased in ovarian, noticed that the level of LL-37 in various types of
breast, and lung cancers, but LL-37 suppresses gastric cancers is significantly reduced. During
colon and gastric cancers (Wu et al. 2010b). Helicobacter pylori infection, the level of LL-37/
Whether and how LL-37 promotes cancer and hCAP-18 secreted into gastric juice is increased.
metastasis deserves further studies. This section This is understandable since the H. pylori-­
discusses anticancer activity of LL-37 as a basis induced expression of LL-37 exerts bactericidal
for developing alternative cancer treatment effects. Wu et al. (2010a) showed that LL-37 sup-
approaches. Such methods can be important for presses gastric cancer by increasing the tumor-­
cancers that are resistant to existing anticancer suppressing bone morphogenetic protein
drugs. However, the poor cell selectivity between signaling via inhibiting proteasome. Schauber
normal and malignant cells may make it chal- et al. (2004) detected upregulation of LL-37 in
lenging to put them into practical use directly both colonic and gastric cells when histone-­
(Ohsaki et al. 1992; Li et al. 2006a; Mishra et al. deacetylase (HDAC) inhibitors (e.g., butyrate
2018). Advanced engineering strategies dis- and trichostatin A) were administered. Induction
cussed elsewhere may be helpful (Mishra et al. of LL-37 appears to be promising also for treat-
2017b). ment of colon cancer.
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 233

12.7 Concluding Remarks GI-20 are all derived from this region
(Fig. 12.3). The sequence of IG-24 (Nell
Human cathelicidin LL-37 is an interesting et al. 2006), the template for SAAP-148,
moonlighting peptide with multiple functional however, extends beyond this central region
roles and involvement in numerous diseases. The by including four residues from the non-­
wide functions of this peptide provide a scientific membrane-­ targeting C-terminal tail of
basis for developing its potential applications. LL-37.
The high interest in the therapeutic potential of 3. An alteration of the amino acid composition
antimicrobial peptides originates from their affects the peptide activity spectrum. The
potency against drug-resistant bacteria, RNA wide-spectrum GF-17 (i.e., killing both
viruses, and cancer. The design of AMPs based Gram-positive and Gram-negative bacteria)
on LL-37 starts from the identification of its has been converted to narrow-spectrum pep-
active regions (Table 12.1 and Fig. 12.3). Both tides that kill only either Gram-positive or
library screen and structure-based approaches are Gram-negative bacteria. These results under-
utilized. It is also possible to combine the library score the importance of basic amino acids
screen with structure-based design (e.g., Wang for antimicrobial activity against Gram-­
et al. 2004). What we have learned to date from negative bacteria and hydrophobic residues
LL-37 peptide design can be summarized below: for killing Gram-positive pathogens, consis-
tent with the database findings (Wang et al.
1. Through multiple studies, the benefits of 2018).
LL-37 engineering emerge. While the activ- 4. Consistent with the classic view, hydropho-
ity of full-length LL-37 is media dependent, bic amino acids of LL-37 peptides are impor-
the designer peptides such as GF-17 and tant for membrane anchoring. The protruding
17BIPHE2 are not (Li et al. 2006a; Wang aromatic rings of F17 and F27 imply their
et al. 2014, 2018; Mishra et al. 2016). While interdigitating membranes. While all basic
LL-37 is poor in preformed biofilm disrup- amino acids participate in lipid clustering
tion, both GF-17 and 17BIPHE2 work well (Epand et al. 2009), they are not equal. R23
(Mishra et al. 2016). While LL-37 is inactive of LL-37 is essential for pathogen recogni-
against a seasonal flu virus, GI-20 remains tion, LPS neutralization, membrane perme-
active (Tripathi et al. 2015a, b). While the ation, pathogen killing, and antibiofilm
effect of LL-37 on cancer is controversial, effects (Wang 2007; Wang et al. 2012b,
FK-16 is anticancer and works superior to 2017, 2018; He et al. 2018). Even a lysine-­
LL-37 (Li et al. 2006a; Ren et al. 2013). arginine positional swap can affect such
These examples underscore the medical sig- properties of LL-37 peptides, revealing the
nificance of LL-37 fragments as well as pep- evolutional significance of the native
tide engineering. sequence (Wang et al. 2017). As an unwanted
2. There is a consensus that the antimicrobial effect, such an interfacial arginine also con-
action of LL-37 is achieved primarily via its tributes to hemolysis and an change to orni-
central region. This central antimicrobial thine improves peptide selectivity (Wang
region (residues 13–32) becomes remark- et al. 2018).
ably evident in the 3D structure of LL-37 5. To reduce peptide production cost, there has
(Fig. 12.2) because it is sandwiched between been a desire to identify the minimally active
two punctuation signals of the LL-37 regions of LL-37. Interestingly, the shortest
sequence: hydrophilic S9 (which splits the active peptide varies with biological activity
hydrophobic surface into two domains) and (i.e., RI-10 for receptor binding, KR-12
P33 (which ends the helical region) (Wang against bacteria, FK-13 against HIV-1, and
2008). Peptides KR-12, FK-13, GF-17, and GF-17 against MRSA).
234 G. Wang et al.

6. While NMR studies have accurately mapped ability, production cost, and potential
the core antimicrobial region of LL-37 (Li cytotoxicity. The template for engineering
et al. 2006a), bacteria also counteract on this 17BIPHE2 is obtained based on structure,
region. S. aureus can secrete aureolysin to whereas the template for SAAP-148 origi-
cut this region of LL-37 (Sieprawska-Lupa nates from peptide library screen (Wang
et al. 2004). In addition, another Gram-­ et al. 2014a; de Breij et al. 2018). Both pep-
positive pathogen GAS can recognize RI-10 tides show effects in vitro and in vivo on
via the CsrRS receptor (Velarde et al. 2014) resistant pathogens and biofilms.
to increase the expression of virulence fac- 11. It is preferred to prevent biofilm formation
tors (Fig. 12.3). Such bacterial recognition because preformed biofilms (e.g., P. aerugi-
regions shed new light on the significance of nosa) are notoriously difficult to get rid of.
the core antimicrobial region of LL-37 as Combined treatment using 17BIPHE2 and
well as host-pathogen interactions at the antibiotics give better results. This practical
molecular level. approach not only potentiates the effect of
7. While an increase of excessive basic amino traditional antibiotics but also reduces the
acids can make the peptide more toxic (e.g., peptide needed, reducing cost and potential
Jacob et al. 2013), a general and classic wis- cytotoxicity (Wang 2017).
dom to improve cell selectivity of peptides is 12. Antimicrobial activity and immune modula-
to decrease hydrophobicity. Sequence muta- tion are the two faces of the same coin. These
tion, deletion, and truncation are routinely innate immune peptides use different molec-
utilized for this purpose. Partial incorpora- ular targets (e.g., bacterial membranes for
tion of D-amino acids can alter peptide con- antimicrobial effects and host receptors for
formation and generate incoherent side chain signal transduction and immune modula-
packing, reducing hydrophobicity (Li et al. tion). In a catheter-associated mouse bio-
2006a). The incoherent packing of side film model, 17BIPHE2 is not only
chains leads to hydrophobic defects in the antibacterial but also chemotactic to mono-
structure, the basis for cell selectivity. The cytes (Wang et al. 2014). GF-17 can directly
importance of peptide hydrophobicity for inactivate Zika virus. It also acts via the
targeting bacterial membranes sets a limit on interferon pathway (He et al. 2018). Through
the degree of hydrophobic reduction for cell peptide design, it is possible to retain anti-
selectivity. bacterial activity and tune host immune
8. Usually, peptides made of D-amino acids are response (Nell et al. 2006).
more resistant to proteases (Boman 2003).
Protease-stable peptides can also be screened We anticipate that research interest in human cat-
from a peptide library followed by structure-­ helicidin will continue to expand, and new func-
guided peptide design to enhance activity tions of LL-37 may be discovered. Both
against the ESKAPE pathogens. 17BIPHE2 prevention and treatment strategies are under
can kill bacteria in the presence of host chy- active development. To avoid infection, we antic-
motrypsin, S. aureus V8 protease, or fungal ipate continued research interest in covalent
proteinase K (Wang et al. 2014). immobilization of AMPs on medical implants.
9. Likewise, plasma binding to SAAP-148 has The basis for this is that peptide injection or non-­
been minimized (de Breij et al. 2018). This is covalent coating of LL-37 peptides can reduce S.
regarded as a main reason for the failure of aureus infection (Wang et al. 2014a; de Breij
AMPs in vivo. Whether this is a gen- et al. 2016). Efforts will also continue to find the
eral requirement remains to be validated. optimal approach for LL-37 induction at a right
10. To date, the therapeutic use of LL-37 is lim- time and location. In winter, sunlight and vitamin
ited to topical treatment due to poor bioavail- D supplement can be beneficial. Our results indi-
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 235

cate the importance of early treatment. In addi- agent with potent antimicrobial effects against oral
pathogens. Biofouling 33(10):807–818
tion, the anti-infective potential of LL-37-derived Cassin ME, Ford AJ, Orbach SM, Saverot SE, Rajagopalan
peptides may be expanded by combining them P (2016) The design of antimicrobial LL37-modified
with traditional antibiotics or other approved collagen-hyaluronic acid detachable multilayers. Acta
simple compounds. As a milestone, topical treat- Biomater 40:119–129
Chen G, Zhou M, Chen S, Lv G, Yao J (2009) Nanolayer
ment of infections using LL-37-derived peptides biofilm coated on magnetic nanoparticles by using
has been demonstrated in animal models (e.g., a dielectric barrier discharge glow plasma fluid-
Wang et al. 2014a; de Breij et al. 2018), and ized bed for immobilizing an antimicrobial peptide.
OP-145 was used to treat chronic otitis media in Nanotechnology 20(46):465706
Cheng M, Ho S, Yoo JH, Tran DH, Bakirtzi K, Su B,
a clinical phase 2 trial in 2009 (de Breij et al. Tran DH, Kubota Y, Ichikawa R, Koon HW (2014)
2018). These achievements will reignite the hope Cathelicidin suppresses colon cancer development by
to develop the peptide into a new systemic antibi- inhibition of cancer associated fibroblasts. Clin Exp
otic, the holy grail of future LL-37 engineering. Gastroenterol 8:13–29
Choi KY, Chow LN, Mookherjee N (2012) Cationic host
defence peptides: multifaceted role in immune modu-
Acknowledgments This study is supported by the lation and inflammation. J Innate Immun 4(4):361–370
NIAID/NIH grant R01 AI105147 and AI128230 to Comune M, Rai A, Chereddy KK, Pinto S, Aday S,
GW. This chapter reflects the point of view of the authors Ferreira AF, Zonari A, Blersch J, Cunha R, Rodrigues
and may not represent the funding agency. R, Lerma J, Simões PN, Préat V, Ferreira L (2017)
Antimicrobial peptide-gold nanoscale therapeutic
formulation with high skin regenerative potential.
J Control Release 262:58–71
References Cowland JB, Johnsen AH, Borregaard N (1995) hCAP-18,
a cathelin/probactenecin-like protein of human neutro-
Agerberth B, Gunne H, Odeberg J, Kogner P, Boman HG, phil specific granules. FEBS Lett 368(1):173–176
Gudmundsson GH (1995) FALL-39, a putative human Crack LR, Jones L, Malavige GN, Patel V, Ogg GS (2012)
peptide antibiotic, is cysteine-free and expressed in Human antimicrobial peptides LL-37 and human
bone marrow and testis. Proc Natl Acad Sci U S A β-defensin-2 reduce viral replication in keratinocytes
92(1):195–199 infected with varicella zoster virus. Clin Exp Dermatol
Alagarasu K, Patil PS, Shil P, Seervi M, Kakade MB, Tillu 37(5):534–543
H, Salunke A (2017) In-vitro effect of human cathe- Currie SM, Findlay EG, McHugh BJ, Mackellar A, Man
licidin antimicrobial peptide LL-37 on dengue virus T, Macmillan D, Wang H, Fitch PM, Schwarze J,
type 2. Peptides 92:23–30 Davidson DJ (2013) The human cathelicidin LL-37
Bals R, Weiner DJ, Meegalla RL, Wilson JM (1999) has antiviral activity against respiratory syncytial
Transfer of a cathelicidin peptide antibiotic gene virus. PLoS One 8(8):e73659
restores bacterial killing in a cystic fibrosis xenograft Currie SM, Gwyer Findlay E, McFarlane AJ, Fitch PM,
model. J Clin Invest 103:1113–1117 Böttcher B, Colegrave N, Paras A, Jozwik A, Chiu C,
Bergman P, Walter-Jallow L, Broliden K, Agerberth B, Schwarze J, Davidson DJ (2016) Cathelicidins have
Söderlund J (2007) The antimicrobial peptide LL-37 direct antiviral activity against respiratory syncytial
inhibits HIV-1 replication. Curr HIV Res 5(4):410–415 virus in vitro and protective function in vivo in mice
Boman HG (2003) Antibacterial peptides: basic facts and and humans. J Immunol 196(6):2699–2710
emerging concepts. J Intern Med 254(3):197–215 da Silva BR, Conrado AJS, Pereira AL, Evaristo FFV,
Boucher HW, Talbot GH, Bradley JS, Edwards JE, Gilbert Arruda FVS, Vasconcelos MA, Lorenzón EN, Cilli EM,
D, Rice LB, Scheld M, Spellberg B, Bartlett J (2009) Teixeira EH (2017) Antibacterial activity of a novel
Bad bugs, no drugs: no ESKAPE! An update from the antimicrobial peptide [W7]KR12-KAEK derived from
Infectious Diseases Society of America. Clin Infect KR-12 against Streptococcus mutans planktonic cells
Dis 48:1–12 and biofilms. Biofouling 33(10):835–846
Braff MH, Hawkins MA, Di Nardo A, Lopez-Garcia B, de Breij A, Riool M, Kwakman PH, de Boer L, Cordfunke
Howell MD, Wong C, Lin K, Streib JE, Dorschner R, RA, Drijfhout JW, Cohen O, Emanuel N, Zaat SA,
Leung DY, Gallo RL (2005) Structure-function rela- Nibbering PH, Moriarty TF (2016) Prevention of
tionships among human cathelicidin peptides: disso Staphylococcus aureus biomaterial-associated infec-
ciation of antimicrobial properties from host immuno- tions using a polymer-lipid coating containing the anti-
stimulatory activities. J Immunol 174(7):4271–4278 microbial peptide OP-145. J Control Release 222:1–8
Caiaffa KS, Massunari L, Danelon M, Abuna GF, Bedran de Breij A, Riool M, Cordfunke RA, Malanovic N, de
TBL, Santos-Filho NA, Spolidorio DMP, Vizoto NL, Boer L, Koning RI, Ravensbergen E, Franken M, van
Cilli EM, Duque C (2017) KR-12-a5 is a non-­cytotoxic der Heijde T, Boekema BK, Kwakman PHS, Kamp N,
236 G. Wang et al.

El Ghalbzouri A, Lohner K, Zaat SAJ, Drijfhout JW, Hase K, Murakami M, Iimura M, Cole SP, Horibe Y,
Nibbering PH. (2018) Sci Transl Med. 10(423). pii: Ohtake T, Obonyo M, Gallo RL, Eckmann L, Kagnoff
eaan4044 MF (2003) Expression of LL-37 by human gastric
den Hertog AL, van Marle J, Veerman EC, Valentijn-­ epithelial cells as a potential host defense mecha-
Benz M, Nazmi K, Kalay H, Grün CH, Van’t Hof nism against Helicobacter pylori. Gastroenterology
W, Bolscher JG, Nieuw Amerongen AV (2006) The 125(6):1613–1625
human cathelicidin peptide LL-37 and truncated He M, Zhang H, Li Y, Wang G, Tang B, Zhao J, Huang
variants induce segregation of lipids and proteins
­ Y, Zheng J (2018) Cathelicidin-derived antimicrobial
in the plasma membrane of Candida albicans. Biol peptides inhibit Zika virus through direct inactivation
Chem 387(10–11):1495–1502 and interferon pathway. Front Immunol 9:722
Dürr UH, Sudheendra US, Ramamoorthy A (2006) Henzler Wildman KA, Lee DK, Ramamoorthy A
Membrane fragmentation by an amyloidogenic frag- (2003) Mechanism of lipid bilayer disruption by the
ment of human islet amyloid polypeptide detected by human antimicrobial peptide, LL-37. Biochemistry
solid-state NMR spectroscopy of membrane nano- 42(21):6545–6558
tubes. Biochim Biophys Acta 1768(9):2026–2029 Howell MD, Jones JF, Kisich KO, Streib JE, Gallo RL,
Dutta D, Kumar N, D P Willcox M (2016) Antimicrobial Leung DY (2004) Selective killing of vaccinia virus
activity of four cationic peptides immobilised to poly-­ by LL-37: implications for eczema vaccinatum.
hydroxyethylmethacrylate. Biofouling 32(4):429–438 J Immunol 172(3):1763–1767
Epand RF, Wang G, Berno B, Epand RM (2009) Lipid Jacob B, Park IS, Bang JK, Shin SY (2013) Short KR-12
segregation explains selective toxicity of a series of analogs designed from human cathelicidin LL-37
fragments derived from the human cathelicidin LL-37. possessing both antimicrobial and antiendotoxic
Antimicrob Agents Chemother 53(9):3705–3714 activities without mammalian cell toxicity. J Pept Sci
Findlay F, Pohl J, Svoboda P, Shakamuri P, McLean K, 19(11):700–707
Inglis NF, Proudfoot L, Barlow PG (2017) Carbon Jayaram LN, Chen JY (2015) Antimicrobial peptides:
nanoparticles inhibit the antimicrobial activities of the possible anti-infective agents. Peptides 72:88–94
human cathelicidin LL-37 through structural altera- Jiang W, Sunkara LT, Zeng X, Deng Z, Myers SM, Zhang
tion. J Immunol 199(7):2483–2490 G (2013) Differential regulation of human cathelicidin
Frohm M, Agerberth B, Ahangari G, Stâhle-Bäckdahl LL-37 by free fatty acids and their analogs. Peptides
M, Lidén S, Wigzell H, Gudmundsson GH (1997) 50:129–138
The expression of the gene coding for the antibac- Kahlenberg JM, Kaplan MJ (2013) Little peptide, big
terial peptide LL-37 is induced in human keratino- effects: the role of LL-37 in inflammation and autoim-
cytes during inflammatory disorders. J Biol Chem mune disease. J Immunol 191(10):4895–4901
272(24):15258–15263 Kanthawong S, Bolscher JG, Veerman EC, van Marle J,
Gabriel M, Nazmi K, Veerman EC, Nieuw Amerongen Nazmi K, Wongratanacheewin S, Taweechaisupapong
AV, Zentner A (2006) Preparation of LL-37-grafted S (2010) Antimicrobial activities of LL-37 and its
titanium surfaces with bactericidal activity. Bioconjug truncated variants against Burkholderia thailandensis.
Chem 17(2):548–550 Int J Antimicrob Agents 36(5):447–452
Grönberg A, Mahlapuu M, Ståhle M, Whately-Smith C, Kay LE, Ikura M, Tschudin R, Bax A (2011) Three-­
Rollman O (2014) Treatment with LL-37 is safe and dimensional triple-resonance NMR Spectroscopy of
effective in enhancing healing of hard-to-heal venous isotopically enriched proteins. 1990. J Magn Reson
leg ulcers: a randomized, placebo-controlled clinical 213(2):423–441
trial. Wound Repair Regen 22:613–621 Keifer PA, Peterkofsky A, Wang G (2004) Effects of
Gudmundsson GH, Agerberth B, Odeberg J, Bergman T, detergent alkyl chain length and chemical structure on
Olsson B, Salcedo R (1996) The human gene FALL39 the properties of a micelle-bound bacterial membrane
and processing of the cathelin precursor to the antibac- targeting peptide. Anal Biochem 331:33–39
terial peptide LL-37 in granulocytes. Eur J Biochem Klevens RM, Morrison MA, Nadle J et al (2007) Invasive
238:325–332 methicillin-resistant Staphylococcus aureus infections
Gunasekera S, Muhammad T, Strömstedt AA, Rosengren in the United States. JAMA 298:1763–1771
KJ, Göransson U (2018) Alanine and lysine scans of Koziel J, Bryzek D, Sroka A, Maresz K, Glowczyk I,
the LL-37-derived peptide fragment KR-12 reveal Bielecka E, Kantyka T, Pyrć K, Svoboda P, Pohl
key residues for antimicrobial activity. Chembiochem J, Potempa J (2014) Citrullination alters immuno-
19(9):931–939 modulatory function of LL-37 essential for pre-
Gustafsson A, Olin AI, Ljunggren L (2010) LPS interac- vention of endotoxin-induced sepsis. J Immunol
tions with immobilized and soluble antimicrobial pep- 192(11):5363–5372
tides. Scand J Clin Lab Invest 70(3):194–200 Kuroda K, Fukuda T, Yoneyama H, Katayama M, Isogai
Harcourt JL, McDonald M, Svoboda P, Pohl J, Tatti K, H, Okumura K, Isogai E (2012) Anti-proliferative
Haynes LM (2016) Human cathelicidin, LL-37, inhib- effect of an analogue of the LL-37 peptide in the colon
its respiratory syncytial virus infection in polarized cancer derived cell line HCT116 p53+/+ and p53−/−.
airway epithelial cells. BMC Res Notes 9:11 Oncol Rep 28(3):829–834
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 237

Kyte J, Doolittle RF (1982) A simple method for display- and application strategies. Curr Opin Chem Biol
ing the hydropathic character of a protein. J Mol Biol 38:87–96
157:105–132 Mishra B, Wang X, Lushnikova T, Zhang Y, Golla RM,
Larrick JW, Hirata M, Balint RF, Lee J, Zhong J, Wright Narayana JL, Wang C, McGuire TR, Wang G (2018)
SC (1995) Human CAP18: a novel antimicrobial Antibacterial, antifungal, anticancer activities and
lipopolysaccharide-binding protein. Infect Immun structural bioinformatics analysis of six naturally
63(4):1291–1297 occurring temporins. Peptides 106:9–20
Lee PH, Ohtake T, Zaiou M, Murakami M, Rudisill JA, Murakami M, Lopez-Garcia B, Braff M, Dorschner RA,
Lin KH, Gallo RL (2005) Expression of an additional Gallo RL (2004) Postsecretory processing generates
cathelicidin antimicrobial peptide protects against multiple cathelicidins for enhanced topical antimicro-
bacterial skin infection. Proc Natl Acad Sci U S A bial defense. J Immunol 172:3070–3077
102:3750–3755 Murakami M, Kameda K, Tsumoto H, Tsuda T, Masuda
Lee CC, Sun Y, Qian S, Huang HW (2011) Transmembrane K, Utsunomiya R, Mori H, Miura Y, Sayama K (2017)
pores formed by human antimicrobial peptide LL-37. TLN-58, an additional hCAP18 processing form,
Biophys J 100(7):1688–1696 found in the lesion vesicle of palmoplantar pustulosis
Lehrer RI, Lu W (2012) α-Defensins in human innate in the skin. J Invest Dermatol 137(2):322–331
immunity. Immunol Rev 245(1):84–112 Nan YH, Bang JK, Jacob B, Park IS, Shin SY (2012)
Li J, Perez Perez GI. (2018) Is there a role for the non-­ Prokaryotic selectivity and LPS-neutralizing activ-
helicobacter pylori bacteria in the risk of developing ity of short antimicrobial peptides designed from
gastric cancer? Int J Mol Sci 19(5) pii: E1353 the human antimicrobial peptide LL-37. Peptides
Li X, Li Y, Han H, Miller DW, Wang G (2006a) Solution 35(2):239–247
structures of human LL-37 fragments and NMR-­ Nell MJ, Tjabringa GS, Wafelman AR, Verrijk R, Hiemstra
based identification of a minimal membrane-targeting PS, Drijfhout JW, Grote JJ (2006) Development of
antimicrobial and anticancer region. J Am Chem Soc novel LL-37 derived antimicrobial peptides with LPS
128:5776–5785 and LTA neutralizing and antimicrobial activities for
Li Y, Li X, Wang G (2006b) Cloning, expression, isotope therapeutic application. Peptides 27(4):649–660
labeling, and purification of human antimicrobial Nie B, Ao H, Chen C, Xie K, Zhou J, Long T, Tang T,
peptide LL-37 in Escherichia coli for NMR studies. Yue B (2016) Covalent immobilization of KR-12 pep-
Protein Expr Purif 47:498–505 tide onto a titanium surface for decreasing infection
Li Y, Li X, Li H, Lockridge O, Wang G (2007) A novel and promoting osteogenic differentiation. RSC Adv
method for purifying recombinant human host defense 6(52):46733–46743
cathelicidin LL-37 by utilizing its inherent property of Niemirowicz K, Durnaś B, Tokajuk G, Piktel E, Michalak
aggregation. Protein Expr Purif 54:157–165 G, Gu X, Kułakowska A, Savage PB, Bucki R (2017)
McArthur MA (2017) Zika virus: Recent advances toward Formulation and candidacidal activity of magnetic
the development of vaccines and therapeutics. Viruses nanoparticles coated with cathelicidin LL-37 and cer-
9(6) pii: E143 agenin CSA-13. Sci Rep 7(1):4610
Meister M, Lemaitre B, Hoffmann JA (1997) Nizet V, Ohtake T, Lauth X, Trowbridge J, Rudisill J,
Antimicrobial peptide defense in Drosophila. Dorschner RA, Pestonjamasp V, Piraino J, Huttner
BioEssays 19(11):1019–1026 K, Gallo RL (2001) Innate antimicrobial peptide pro-
Mishra B, Wang G (2012) Ab initio design of potent anti-­ tects the skin from invasive bacterial infection. Nature
MRSA peptides based on database filtering technol- 414(6862):454–457
ogy. J Am Chem Soc 134:12426–12429 Ogawa Y, Kawamura T, Matsuzawa T, Aoki R, Gee P,
Mishra B, Wang G (2017) Individual and combined Yamashita A, Moriishi K, Yamasaki K, Koyanagi Y,
effects of engineered peptides and antibiotics on the Blauvelt A, Shimada S (2013) Antimicrobial peptide
Pseudomonas aeruginosa biofilms. Pharmaceuticals LL-37 produced by HSV-2-infected keratinocytes
10(3):58 enhances HIV infection of Langerhans cells. Cell Host
Mishra B, Epand RF, Epand RM, Wang G (2013) Microbe 13(1):77–86
Structural location determines functional roles of the Ohsaki Y, Gazdar AF, Chen HC, Johnson BE (1992)
basic amino acids of KR-12, the smallest antimicro- Antitumor activity of magainin analogues
bial peptide from human cathelicidin LL-37. RSC Adv against human lung cancer cell lines. Cancer Res
3:19560–19571 52(13):3534–3538
Mishra B, Golla R, Lau K, Lushnikova T, Wang G (2016) Oren Z, Lerman JC, Gudmundsson GH, Agerberth
Anti-Staphylococcal biofilm effects of human catheli- B, Shai Y (1999) Structure and organization of the
cidin peptides. ACS Med Chem Lett 7:117–121 human antimicrobial peptide LL-37 in phospho-
Mishra B, Lushnikova T, Golla RM, Wang X, Wang G lipid membranes: relevance to the molecular basis
(2017a) Design and surface immobilization of short for its non-­cell-­selective activity. Biochem J 341(Pt
anti-biofilm peptides. Acta Biomater 49:316–328 3):501–513
Mishra B, Reiling S, Zarena D, Wang G (2017b) Host Overhage J, Campisano A, Bains M, Torfs EC, Rehm
defense antimicrobial peptide as antibiotics: design BH, Hancock RE (2008) Human host defense peptide
238 G. Wang et al.

LL-37 prevents bacterial biofilm formation. Infect wound healing potential in HaCaT cell line. J Pept Sci
Immun 76(9):4176–4182 8:e3080
Overhage PE, McHugh B, Gwyer Findlay E, Mackellar Schauber J, Iffland K, Frisch S, Kudlich T, Schmausser B,
A, Mackenzie KJ, Gallo RL, Govan JR, Simpson AJ, Eck M, Menzel T, Gostner A, Lührs H, Scheppach W
Davidson DJ (2014) Cathelicidin host defence peptide (2004) Histone-deacetylase inhibitors induce the cat-
augments clearance of pulmonary Pseudomonas aeru- helicidin LL-37 in gastrointestinal cells. Mol Immunol
ginosa infection by its influence on neutrophil func- 41(9):847–854
tion in vivo. PLoS One 9:e99029 Schögler A, Muster RJ, Kieninger E, Casaulta C, Tapparel
Peschel A, Sahl HG (2006) The co-evolution of host cat- C, Jung A, Moeller A, Geiser T, Regamey N, Alves
ionic antimicrobial peptides and microbial resistance. MP (2016) Vitamin D represses rhinovirus replication
Nat Rev Microbiol 4:529–536 in cystic fibrosis cells by inducing LL-37. Eur Respir
Pütsep K, Carlsson G, Boman HG, Andersson M (2002) J 47:520–530
Deficiency of antibacterial peptides in patients with Scott MG, Davidson DJ, Gold MR, Bowdish D, Hancock
morbus kostmann: an observation study. Lancet RE (2002) The human antimicrobial peptide LL-37
360:1144–1149 is a multifunctional modulator of innate immune
Rajasekaran G, Kim EY, Shin SY (2017) LL-37-derived responses. J Immunol 169(7):3883–3891
membrane-active FK-13 analogs possessing cell selec- Sevcsik E, Pabst G, Richter W, Danner S, Amenitsch H,
tivity, anti-biofilm activity and synergy with chlor- Lohner K (2008) Interaction of LL-37 with model
amphenicol and anti-inflammatory activity. Biochim membrane systems of different complexity: influence
Biophys Acta 1859(5):722–733 of the lipid matrix. Biophys J 94:4688–4699
Rapala-Kozik M, Bochenska O, Zawrotniak M, Wolak Shah NR, Hancock RE, Fernandez RC (2014) Bordetella
N, Trebacz G, Gogol M, Ostrowska D, Aoki W, Ueda pertussis lipid A glucosamine modification confers
M, Kozik A (2015) Inactivation of the antifungal resistance to cationic antimicrobial peptides and
and immunomodulatory properties of human cathe- increases resistance to outer membrane perturbation.
licidin LL-37 by aspartic proteases produced by the Antimicrob Agents Chemother 58(8):4931–4934
pathogenic yeast Candida albicans. Infect Immun Sieprawska-Lupa M, Mydel P, Krawczyk K, Wójcik K,
83(6):2518–2530 Puklo M, Lupa B, Suder P, Silberring J, Reed M, Pohl
Ren SX, Cheng AS, To KF, Tong JH, Li MS, Shen J, J, Shafer W, McAleese F, Foster T, Travis J, Potempa
Wong CC, Zhang L, Chan RL, Wang XJ, Ng SS, J (2004) Degradation of human antimicrobial peptide
Chiu LC, Marquez VE, Gallo RL, Chan FK, Yu J, LL-37 by Staphylococcus aureus-derived proteinases.
Sung JJ, Wu WK, Cho CH (2012) Host immune Antimicrob Agents Chemother 48(12):4673–4679
defense peptide LL-37 activates caspase-independent Sigurdardottir T, Andersson P, Davoudi M, Malmsten M,
apoptosis and suppresses colon cancer. Cancer Res Schmidtchen A, Bodelsson M (2006) In silico identifi-
72(24):6512–6523 cation and biological evaluation of antimicrobial pep-
Ren SX, Shen J, Cheng AS, Lu L, Chan RL, Li ZJ, Wang tides based on human cathelicidin LL-37. Antimicrob
XJ, Wong CC, Zhang L, Ng SS, Chan FL, Chan FK, Agents Chemother 50(9):2983–2989
Yu J, Sung JJ, Wu WK, Cho CH (2013) FK-16 derived Singh D, Vaughan R, Kao CC (2014) LL-37 pep-
from the anticancer peptide LL-37 induces caspase-­ tide enhancement of signal transduction by Toll-­
independent apoptosis and autophagic cell death in like receptor 3 is regulated by pH: identification
colon cancer cells. PLoS One 8(5):e63641 of a peptide antagonist of LL-37. J Biol Chem
Ron-Doitch S, Sawodny B, Kühbacher A, David MMN, 289(40):27614–27624
Samanta A, Phopase J, Burger-Kentischer A, Griffith Skerlavaj B, Gennaro R, Bagella L, Merluzzi L, Risso A,
M, Golomb G, Rupp S (2016) Reduced cytotoxicity Zanetti M (1996) Biological characterization of two
and enhanced bioactivity of cationic antimicrobial novel cathelicidin-derived peptides and identification
peptides liposomes in cell cultures and 3D epidermis of structural requirements for their antimicrobial and
model against HSV. J Control Release 229:163–171 cell lytic activities. J Biol Chem 271(45):28375–28381
Saar-Dover R, Bitler A, Nezer R, Shmuel-Galia L, Firon A, Sol A, Ginesin O, Chaushu S, Karra L, Coppenhagen-­
Shimoni E, Trieu-Cuot P, Shai Y (2012) D-alanylation Glazer S, Ginsburg I, Bachrach G (2013) LL-37 opso-
of lipoteichoic acids confers resistance to cationic nizes and inhibits biofilm formation of Aggregatibacter
peptides in group B streptococcus by increasing the actinomycetemcomitans at subbactericidal concentra-
cell wall density. PLoS Pathog 8:e1002891 tions. Infect Immun 81:3577–3585
Santos CM, Kumar A, Kolar SS, Contreras-Caceres R, Song DW, Kim SH, Kim HH, Lee KH, Ki CS, Park YH
McDermott A, Cai C (2013) Immobilization of anti- (2016) Multi-biofunction of antimicrobial peptide-­
microbial peptide IG-25 onto fluoropolymers via fluo- immobilized silk fibroin nanofiber membrane:
rous interactions and click chemistry. ACS Appl Mater implications for wound healing. Acta Biomater
Interfaces 5(24):12789–12793 39:146–155
Saporito P, Vang Mouritzen M, Løbner-Olesen A, Jenssen Sorensen OE, Follin P, Johnsen AH, Calafat J, Tjabringa
H (2018) LL-37 fragments have antimicrobial activ- GS, Hiemstra PS et al (2001) Human cathelicidin,
ity against Staphylococcus epidermidis biofilms and hCAP-18, is processed to the antimicrobial peptide
12 Design of Antimicrobial Peptides: Progress Made with Human Cathelicidin LL-37 239

LL-37 by extracellular cleavage with proteinase 3. the presence of influenza virus: functional and struc-
Blood 97(12):3951–3959 tural analysis. PLoS One 10(8):e0133454
Sørensen OE, Gram L, Johnsen AH, Andersson E, Tripathi S, Wang G, White M, Qi L, Taubenberger J,
Bangsbøll S, Tjabringa GS, Hiemstra PS, Malm J, Hartshorn KL (2015b) Antiviral activity of the human
Egesten A, Borregaard N (2003) Processing of semi- cathelicidin, LL-37, and derived peptides on sea-
nal plasma hCAP-18 to ALL-38 by gastricsin: a novel sonal and pandemic influenza A viruses. PLoS One
mechanism of generating antimicrobial peptides in 10:e0124706
vagina. J Biol Chem 278(31):28540–28546 Turner J, Cho Y, Dinh NN, Waring AJ, Lehrer RI (1998)
Sousa FH, Casanova V, Findlay F, Stevens C, Svoboda P, Activities of LL-37, a cathelin-associated antimicro-
Pohl J, Proudfoot L, Barlow PG (2017) Cathelicidins bial peptide of human neutrophils. Antimicrob Agents
display conserved direct antiviral activity towards rhi- Chemother 42(9):2206–2214
novirus. Peptides 95:76–83 Uchio E, Inoue H, Kadonosono K (2013) Anti-adenoviral
Spencer JD, Schwaderer AL, Becknell B, Watson J, Hains effects of human cationic antimicrobial protein-18/
DS (2014) The innate immune response d­ uring urinary LL-37, an antimicrobial peptide, by quantitative
tract infection and pyelonephritis. Pediatr Nephrol polymerase chain reaction. Korean J Ophthalmol
29(7):1139–1149 27(3):199–203
Sung CM, Kim HC, Cho YB, Shin SY, Jang CH (2017) van der Does AM, Bergman P, Agerberth B, Lindbom L
Evaluating the ototoxicity of an anti-MRSA pep- (2012) Induction of the human cathelicidin LL-37 as a
tide KR-12-a2. Braz J Otorhinolaryngol pii: novel treatment against bacterial infections. J Leukoc
S1808-8694(17)30078–2 Biol 92(4):735–742
Suzuki K, Murakami T, Kuwahara-Arai K, Tamura H, Vandamme D, Landuyt B, Luyten W, Schoofs L (2012)
Hiramatsu K, Nagaoka I (2011) Human anti-­microbial A comprehensive summary of LL-37, the facto-
cathelicidin peptide LL-37 suppresses the LPS-­ tum human cathelicidin peptide. Cell Immunol
induced apoptosis of endothelial cells. Int Immunol 280(1):22–35
23(3):185–193 Velarde JJ, Ashbaugh M, Wessels MR (2014) The human
Suzuki K, Murakami T, Hu Z, Tamura H, Kuwahara-Arai antimicrobial peptide LL-37 binds directly to CsrS, a
K, Iba T, Nagaoka I (2016) Human host defense cathe- sensor histidine kinase of group A Streptococcus, to
licidin peptide LL-37 enhances the lipopolysaccharide activate expression of virulence factors. J Biol Chem
uptake by liver sinusoidal endothelial cells without 289(52):36315–36324
cell activation. J Immunol 196(3):1338–1347 Wang G (2006) Structural biology of antimicrobial
Tack BF, Sawai MV, Kearney WR, Robertson AD, peptides by NMR spectroscopy. Curr Org Chem
Sherman MA, Wang W, Hong T, Boo LM, Wu H, 10:569–581
Waring AJ, Lehrer RI (2002) SMAP-29 has two Wang G (2007) Determination of solution structure and
LPS-binding sites and a central hinge. Eur J Biochem lipid micelle location of an engineered membrane pep-
269(4):1181–1189 tide by using one NMR experiment and one sample.
Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Biochim Biophys Acta 1768:3271–3281
Ogawa T, Takeda K, Akira S (1999) Differential roles Wang G (2008) Structures of human host defense
of TLR2 and TLR4 in recognition of gram-negative cathelicidin LL-37 and its smallest antimicro-
and gram-positive bacterial cell wall components. bial peptide KR-12 in lipid micelles. J Biol Chem
Immunity 11(4):443–451 283:32637–32643
Tanphaichitr N, Srakaew N, Alonzi R, Kiattiburut Wang G (2010) Structure, dynamics and mapping of
W, Kongmanas K, Zhi R, Li W, Baker M, Wang membrane-binding residues of micelle-bound antimi-
G, Hickling D (2016) Potential use of antimicro- crobial peptides by natural abundance 13C NMR spec-
bial peptides as vaginal spermicides/microbicides. troscopy. Biochim Biophys Acta 1798:114–121
Pharmaceuticals 9:13 Wang G (2012) Natural antimicrobial peptides as promis-
Tian M, Zhao DY, Wang HW (2011) Studies on the core ing anti-HIV candidates. Curr Top Peptide Protein Res
functional region of antimicrobial peptide LL-37 for 13:93–110
inhibition of RSV replication. Zhonghua Shi Yan He Lin Wang G (2014) Human antimicrobial peptides and pro-
Chuang Bing Du Xue Za Zhi 25(5):355–357. In Chinese teins. Pharmaceuticals 7:545–594
Tripathi S, Tecle T, Verma A, Crouch E, White M, Wang G (2017) In: Wang G (ed) Antimicrobial peptides.
Hartshorn KL (2013) The human cathelicidin LL-37 CABI, UK, pp 1–19. and 169–187
inhibits influenza A viruses through a mechanism dis- Wang Z, Wang G (2004) APD: the antimicrobial peptide
tinct from that of surfactant protein D or defensins. database. Nucleic Acids Res 32:D590–D592
J Gen Virol 94(Pt 1):40–49 Wang G, Keifer PA, Peterkofsky A (2004) Short-
Tripathi S, Verma A, Kim EJ, White MR, Hartshorn KL chain diacyl phosphatidylglycerols: which one
(2014) LL-37 modulates human neutrophil responses to choose for NMR structural determination of a
to influenza A virus. J Leukoc Biol 96(5):931–938 membrane-­associated peptide from Escherichia coli?
Tripathi S, Wang G, White M, Rynkiewicz M, Seaton B, Spectroscopy 18:257–264
Hartshorn K (2015a) Identifying the critical domain of Wang G, Waston K, Buckheit R Jr (2008) Anti-human
LL-37 involved in mediating neutrophil activation in immunodeficiency virus type 1 (HIV-1) activities
240 G. Wang et al.

of antimicrobial peptides derived from human and of the native sequence and leads to bacterial probes.
bovine cathelicidins. Antimicrob Agents Chemother Biochim Biophys Acta 1859:1350–1361
52:3438–3440 Wang X, Mishra B, Lushinikova T, Narayana JL, Wang
Wang G, Li X, Wang Z (2009) APD2: the updated antimi- G (2018) Amino acid composition determines peptide
crobial peptide database and its application in peptide activity spectrum and hot spot-based design of mereci-
design. Nucleic Acids Res 37:D933–D937 din. Adv Biosyst 2:1700259
Wang G, Elliott M, Cogen AL, Ezell EL, Gallo RL, Wong JH, Legowska A, Rolka K, Ng TB, Hui M, Cho CH,
Hancock REW (2012a) Structure, dynamics, antimi- Lam WW, Au SW, Gu OW, Wan DC (2011) Effects of
crobial and immune modulatory activities of human cathelicidin and its fragments on three key enzymes of
LL-23 and its single residue variants mutated based HIV-1. Peptides 32(6):1117–1122
on homologous primate cathelicidins. Biochemistry Wu WK, Sung JJ, To KF, Yu L, Li HT, Li ZJ, Chu KM,
51:653–664 Yu J, Cho CH (2010a) The host defense peptide LL-37
Wang G, Epand RF, Mishra B, Lushnikova T, Thomas VC, activates the tumor-suppressing bone morphogenetic
Bayles KW, Epand R (2012b) Decoding the functional protein signaling via inhibition of proteasome in gas-
roles of cationic side chains of the major antimicro- tric cancer cells. J Cell Physiol 223(1):178–186
bial region of human cathelicidin LL-37. Antimicrob Wu WK, Wang G, Coffelt SB, Betancourt AM, Lee
Agents Chemother 56:845–856 CW, Fan D, Wu K, Yu J, Sung JJ, Cho CH (2010b)
Wang G, Hanke ML, Mishra B, Lushnikova T, Heim CE, Emerging roles of the host defense peptide LL-37 in
Chittezham Thomas V, Bayles KW, Kielian T (2014a) human cancer and its potential therapeutic applica-
Transformation of human cathelicidin LL-37 into tions. Int J Cancer 127(8):1741–1747
selective, stable, and potent antimicrobial compounds. Yasin B, Pang M, Turner JS, Cho Y, Dinh NN, Waring
ACS Chem Biol 9:1997–2002 AJ, Lehrer RI, Wagar EA (2000) Evaluation of the
Wang G, Mishra B, Epand RF, Epand RM (2014b) High-­ inactivation of infectious Herpes simplex virus by
quality 3D Structure shine light on antibacterial, host-defense peptides. Eur J Clin Microbiol Infect Dis
anti-biofilm and antiviral activities of human catheli- 19(3):187–194
cidin LL-37 and its fragments. Biochim Biophys Acta Zasloff M (1987) Magainins, a class of antimicrobial
1838:2160–2172 peptides from Xenopus skin: isolation, charac-
Wang G, Mishra B, Lau K, Lushnikova T, Golla R, terization of two active forms, and partial cDNA
Wang X (2015) Antimicrobial peptides in 2014. sequence of a precursor. Proc Natl Acad Sci U S A
Pharmaceuticals 8:123–150 84(15):5449–5453
Wang G, Li X, Wang Z (2016) APD3: the antimicrobial Zasloff M (2002) Antimicrobial peptides of multicellular
peptide database as a tool for research and education. organisms. Nature 415(6870):389–395
Nucleic Acids Res 44:D1087–D1093 Zhang LJ, Guerrero-Juarez CF, Hata T, Bapat SP, Ramos R,
Wang X, Junior JCB, Mishra B, Lushnikova T, Epand Plikus MV, Gallo RL (2015) Innate immunity. Dermal
RM, Wang G (2017) Arginine-lysine positional swap adipocytes protect against invasive Staphylococcus
of the LL-37 peptides reveals evolutional advantages aureus skin infection. Science 347:67–71
 Application of Synthetic Molecular
Evolution to the Discovery 13
of Antimicrobial Peptides

William C. Wimley

Abstract Wang et al. 2016). Many more synthetic AMPs

Despite long-standing promise and many have been also created, often by mimicking natu-
known examples, antimicrobial peptides ral sequences in combination with trial and error
(AMPs) have failed, with few exceptions, to experimentation (Fjell et al. 2007; Wang et al.
significantly impact human medicine. 2016) and sometimes by screening or computer-­
Impediments to the systemic activity of AMPs aided design (Rathinakumar and Wimley 2010;
include proteolysis, host cell interactions, andRathinakumar et al. 2009; Moy et al. 2009;
serum protein binding, factors that are not Kulagina et al. 2006, 2007; Hilpert et al. 2005).
often considered in the early stages of AMP Since the beginning, AMPs have been promoted
development. Here we discuss how synthetic as novel antibiotics that might improve human
molecular evolution, iterative cycles of libraryhealth and well-being. Yet, at the time of their ini-
design, and physiologically relevant screening tial discovery, there was little urgency to the
can be used to evolve AMPs that do not have translational applications of AMPs. It was not
these impediments. known at the time that drug-resistant bacterial
infections would grow over the next 30 years to
become a global health crisis in morbidity and
mortality (Arias and Murray 2009; Boucher et al.
13.1 Introduction 2009; Otto 2012). We are now urgently and ever
increasingly in need of novel antibiotic treatment
Antimicrobial peptides were first described in the options against drug-resistant bacteria. While
1980s, having been found in insect hemolymph some AMPs have been developed into potential
(Steiner et al. 1981; Okada and Natori 1983), topical drugs, and some are nearing clinical tri-
mammalian neutrophil granules (Patterson-­ als, or are in clinical trials (Fox 2013), AMPs
Delafield et al. 1981), and frog skin secretions have not succeeded, recently or in past decades
(Zasloff, 1987). Subsequently, more than a thou- (Gordon et al. 2005), to have any real impact on
sand natural AMPs have been found in many tis- the systemic treatment options for drug-resistant
sues of many different species (Fjell et al. 2007; bacteria.
Many of the known AMPs have good antibi-
W. C. Wimley (*) otic activity in the culture tube, microwell plate,
Department of Biochemistry and Molecular Biology, and petri dish, i.e., under standard laboratory
Tulane University School of Medicine, conditions. Many known AMPs have potent, ster-
New Orleans, LA, USA ilizing activity at low μM concentrations against
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 241

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_13
242 W. C. Wimley

multiple species of bacteria (Easton et al. 2009; human body (Joly et al. 2010) despite the lack of
Hamill et al. 2008), often including both Gram-­ specific modifications to increase bioavailability
positive and Gram-negative species. AMP activ- or decrease proteolytic sensitivity. The peptide is
ity is observed at the same concentration range, an amphipathic α-helical peptide (Rapaport et al.
1–10 μM, at which many conventional antibiotics 1995) that likely binds to cells and serum pro-
are active under laboratory conditions. AMPs teins, and this may help it to remain intact and in
often have equally potent activity against drug-­ circulation. Despite this, enfuvirtide has low tox-
susceptible, drug-resistant, and multidrug-­icity and is able to effectively inhibit the fusion of
resistant bacteria (de Breij et al. 2018; HIV viruses with cell membranes in vivo, prob-
Schlusselhuber et al. 2014; Mechkarska et al. ably by interfering with the structure-function
2013; Park et al. 2011) showing that the conven- relationships of the GP41 fusion protein, from
tional mechanisms of drug resistance do not which enfuvirtide was obtained (Qureshi et al.
apply to AMPs. AMPs can also act against bio- 1990).
films (de Breij et al. 2018; Wolfmeier et al. 2017), If there are no insurmountable barriers to sys-
and they can create antibacterial surfaces by temically active AMPs and we have thousands of
covalent tethering or physical adsorption known AMPs with activity in the laboratory, it is
(Kazemzadeh-Narbat et al. 2010). Importantly, reasonable to ask: Why are there no systemically
AMPs may be less likely to induce resistance active AMP drugs? Why does there seem to be
than conventional antibiotics (Dobson et al. 2013; few in the development pipeline? In this chapter
Pollard et al. 2012; Fedders et al. 2010) although we hypothesize that the number of potential sys-
resistance to AMPs does occur (Perron et al. temically active AMP drugs in the drug develop-
2006). Despite their potent, broad-spectrum ment pipeline is small because (i) the thousands
activity in the laboratory at low concentration, of AMPs known have not evolved to be systemic
AMPs have not reached the ultimate goal: devel- AMPs or have not been discovered under the
opment into novel antibiotics that can be used most relevant conditions and because (ii) rational
systemically to prevent or treat drug-resistant engineering of AMP properties is not possible
bacterial infections. due to the fact that we do not have a sufficient
This dearth of systemically active AMPs has knowledge of sequence-function relationships
many causes but may be due in part to impedi- for any of these impediments or for antibacterial
ments to bioavailability and dosing such as host activity. Below, we discuss an approach that may
cell inhibition (Starr et al. 2016), serum inhibi- be especially well suited in this situation: discov-
tion (de Breij et al. 2018), residual toxicity ery of novel AMPs by iteratively screening small
(Yeaman and Yount 2003), and proteolytic degra- peptide libraries under experimental conditions
dation (Starr and Wimley 2017). Systemically that are increasingly relevant to physiological
active AMPs are almost certainly possible, as conditions. We have referred to this approach as
none of these impediments seems to be unsur- “synthetic molecular evolution” (Krauson et al.
mountable on its own. In fact, a systemically 2013; Kauffman et al. 2018; Li et al. 2018). By
active, lifesaving, anti-infective peptide drug these means we suggest approaching the most
already exists. The anti-HIV peptide enfuvirtide, clinically relevant conditions in a stepwise man-
which has been approved for human use in the ner and doing so as a first stage in the preclinical
USA and Europe since 2003 (Poveda et al. 2005), identification of peptide antibiotic drugs to feed a
is a linear 36-residue peptide drug with over $1 larger number of relevant candidates into the
billion in net sales (Poveda et al. 2005; Joly et al. development pipeline. Below we detail some of
2010; LaBonte et al. 2003). This peptide is the major impediments to systemic activity of
administered subcutaneously in 90 mg doses and AMPs and then describe how screening for AMP
has extended the lives of many patients infected activity can be done under conditions that much
with HIV that had become resistant to other more closely mimic in vivo conditions to identify
drugs. Enfuvirtide has a long half-life in the peptides without these impediments.
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 243

13.2 Impediments to Systemic action of AMPs may involve the entire cell archi-
Activity tecture. When resistance is observed, it is usually
due to changes in LPS or cell wall components
Compared to conventional antibiotic drugs, (Peschel and Sahl 2006; Peschel 2002), making
AMPs have a fundamentally different mode of these structures less anionic so they do not accu-
action on bacteria. Most conventional antibiotics mulate large amounts of cationic AMPs.
inhibit a critical biochemical process by targeting
one molecule (e.g., an enzyme or ribosome). The
number of functional molecules decreases until 13.3  erum and Host Cell
S
the microbe loses so much of that essential func- Inhibition
tion that it cannot replicate or survive. The rate or
degree to which critical activity is lost depends The need for accumulation of peptide means that
on local drug concentration. AMPs on the other systemic activity of AMPs faces different chal-
hand kill microbes in a cooperative, saturation-­ lenges than conventional drugs (Fig. 13.1) and
like process in which the peptides must massively will exhibit threshold behavior. There may be
accumulate on bacteria to levels that essentially only a narrow window between saturation/killing
saturate the cell in order to kill them through the of bacteria and survivable accumulation of an
effect of their interfacial activity (Wimley 2010) AMP (Fig. 13.2). Thus, any factor that competes
on membrane integrity. Various measurements in for bacterial binding has the capacity to decrease
the literature (Starr et al. 2016; Savini et al. 2018, accumulation on bacteria. Contrary to the com-
2017; Tran et al. 2002; Steiner et al. 1988) have monly stated belief, external eukaryotic mem-
shown that the number of bound AMPs required branes are highly anionic overall, due to the large
to kill a bacterial cell is extremely high, from 107 amount of anionic glycoconjugates attached to
to 108 peptides per cell. Since an E. coli cell can lipids and proteins. Thus, cationic AMPs bind to
be expected to have perhaps 2 × 107 total lipids, eukaryotic cells and tissues, at least moderately.
this means that killing does not occur until there Even weak competition can be problematic
is around one peptide bound for every bacterial because host cells and tissue will always be
lipid, which is more peptide than can possibly orders of magnitude more abundant than patho-
bind to the cytoplasmic membrane alone. This gens. Using a set of 12 natural and synthetic
lethal amount of peptide is also equivalent to AMPs, we have shown that even a few minutes of
roughly one peptide per DNA base. If we assume preincubation of AMPs with human erythrocytes
that 108 peptides are evenly distributed in the vol- strongly reduced the activity of most of them
ume of the cell, the local concentration of AMP is (Starr et al. 2016) (see Fig. 13.3) through a com-
about 80 mM (Starr et al. 2016). In reality, AMPs bination of host cell binding and proteolysis by
will specifically accumulate on anionic structures the cytosolic proteases of RBCs (Starr and
such as cell wall, LPS, cytoplasmic membrane, Wimley 2017). This effect is not always observed,
and DNA and could have local concentrations indicating that it is a surmountable impediment.
that approach molar. As a result, the mode of For example, the insect peptide cecropin A was
action of AMPs is a successive attack on the cell not affected in our study by human RBCs (Starr
architecture. Membranes are permeabilized et al. 2016). Similarly Stella and colleagues care-
within minutes (Rathinakumar and Wimley 2010; fully examined the effect of RBCs on the activity
Rathinakumar et al. 2009), followed by leakage of an AMP and found little inhibition (Savini
of macromolecules, including DNA. Within et al. 2017).
60 min of treatment of bacteria with AMPs, the Serum proteins, especially serum albumin,
entire cell architecture is compromised, and indi- can also bind cationic AMPs and are also highly
vidual cells are sometimes not discernable. It is concentrated in the body (35–50 mg/mL in
thus reasonable that resistance would be more blood), further potentially reducing the effective
difficult to evolve, given that mechanism of concentration of peptide available to bind to
244 W. C. Wimley

Fig. 13.1 Some

impediments to the
bioavailability and
systemic activity of
antimicrobial peptides.
Antimicrobial peptides
must accumulate
significantly on bacteria
to have bactericidal
activity. In the body,
interactions with host
cells and tissue,
interactions with serum
proteins, and proteolytic
degradation can decrease
accumulation and
decrease activity

b­ acteria (de Breij et al. 2018). As we discuss in that these factors be included during initial dis-
detail below, host cell and serum protein binding covery of AMPs by screening. This will presum-
are rarely considered in the early stages of novel ably give rise to a large number of relevant AMPs
AMP discovery or design. If they are tested at all, that can enter the pipeline. Loading the front of
it is determined how much these factors interfere the development pipeline with better candidates,
with AMP activity only very late in the preclini- in turn, will increase the probability of finding a
cal development pipeline. Here we are proposing few that can be developed into systemic drugs.
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 245

Fig. 13.2 Saturation-dependent activity of an antimicro- assuming simple competition between E. coli and RBC
bial peptide and its inhibition by host cells. The protease-­ show that the measured binding accounts for the loss of
resistant AMP D-ARVA (rrgwalrlvlay-NH2) was used in activity when host cells are present. (d) Survival of differ-
these experiments (Starr et al. 2016; Rathinakumar and ent innocula of E. coli incubated with 20 μM D-ARVA in
Wimley 2010; Rathinakumar et al. 2009). (a) Measured conjunction with the binding curve in panel A enables
binding of D-ARVA to E. coli cells. (b) Measured binding comparison of peptide lethality and the number of pep-
of D-ARVA to human RBCs. (c) Experimental measure- tides bound to each bacterial cell. More than 2 × 108 pep-
ments compared to simulation of a mixed experiment tides bound per cell are required for sterilization

13.4 Toxicity Against Mammalian nificantly among known AMPs, yet those with
Cells relatively little toxicity still have relatively poor
therapeutic indices compared to conventional
AMPs bind to anionic mammalian cells through antibiotics. Some AMPs may become less toxic
electrostatic interactions (Starr et al. 2016; Riedl in the presence of serum or when host cells are
et al. 2011a; Selsted et al. 1985; Agawa et al. highly concentrated, although this effect is rarely
1991) and have interfacial activity (Wimley tested.
2010). As a result, many have at least some acute In the abundant literature on AMPs, research-
toxicity due to permeabilization of the plasma ers often measure lysis of erythrocytes (hemoly-
membrane of the host cells. Toxicities vary sig- sis) as a surrogate for eukaryotic cell toxicity.
246 W. C. Wimley

Fig. 13.3 Synthetic molecular evolution of peptides. As we practice it, SME utilizes multiple small libraries (genera-
tions) which are iteratively screened for gain-of-function daughter sequences

This is an easy assay and its widespread use Wimley 2017; Werle and Bernkop-Schnurch
­provides some uniformity in the AMP literature. 2006; Molhoek et al. 2011). Sensitivity to serum
However for the maximum sensitivity to toxicity, proteolysis is partially predictable based on
nucleated cells in culture may be a more sensitive sequence. In peptide drug development, serum
and more informative model system. Nucleated stability competes with synthetic complexity
cells can also respond to mechanisms of toxicity (i.e., manufacturing cost). Shorter, linear,
other than acute cytolysis. While some research- L-amino acid peptides are most economical to
ers have discussed how some AMPs may have produce but are susceptible to rapid proteolysis,
useful selective activity against cancer cells while cyclic, cross-linked, or chemically modi-
(Riedl et al. 2011a, b), we argue that cancer cells fied peptides are more costly to produce but are
would make an especially stringent test system also more protease resistant. Proteolysis is a per-
for selecting against toxicity because they are vasive threat to AMPs. For example, we have
especially sensitive to AMPs. In other words, a shown that washed human RBCs contain a very
synthetically evolved AMP that has potent anti- high concentration of multiple proteases in their
bacterial activity under relevant conditions and cytosol (Starr and Wimley 2017). Incubation of a
no toxicity against cultured mammalian cancer set of natural and synthetic linear AMPs with
cells, such as HeLa cells, would seem to be an dilute RBCs leads to rapid degradation of peptide
ideal candidate for development into a systemic if there is even a small amount of hemolysis,
drug. We discuss how this can be done in a screen which is almost always true. Cytosolic amino-
below. and carboxy-exopeptidases removed amino acids
one or two at a time from both termini (Starr and
Wimley 2017). For this reason, even standard
13.5 Proteolytic Degradation hemolysis assays in the laboratory may be
strongly affected by the proteolytic sensitivity of
Chemical stability of peptides, i.e., resistance to peptides.
proteolysis, which does not come into play in It is likely that host cell binding and serum
laboratory assays, is also a critical consideration protein binding will decrease susceptibility to
for systemic activity of peptides. Some peptides degradation, but as stated above they may also
are degraded rapidly by serum exopeptidases, interfere with activity. Cyclization or cross-­
dipeptidases, and other proteases (Starr and linking, as found in many natural AMPs, will
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 247

reduce proteolysis. For linear AMPs chemical variations in library members. SME is especially
modification of the termini (Werle and Bernkop-­ useful for the development of AMPs when
Schnurch 2006; Starr et al. 2018; Nguyen et al. screening is done under conditions that mimic
2010) with nonnatural terminal amino acids or the environment in which a systemically active
selective substitution with D-amino acids can peptide must function. We call it “evolution”
increase stability (Starr et al. 2018). Perhaps the because it is most economical to screen itera-
simplest approach is to replace all residues with tively such that each “generation” of gain-of-­
D-amino acids, as this will provide complete function AMP is selected from a library built
resistance to proteolysis while not changing around an active sequence from the previous gen-
activity (Starr et al. 2016; Rathinakumar and eration, and each iterative screen further refines
Wimley 2010; Savini et al. 2018) the selected sequences to have the properties that
are sought. In this case we seek bactericidal
activity at low μM concentration in the presence
13.6 Synthetic Molecular of concentrated host cells and serum, without
Evolution toxicity.

To create an AMP that could have useful, sys-

temic (in vivo) antibacterial activity, the factors 13.7 Design and Synthesis
described above will need to be simultaneously of Combinatorial Peptide
optimized. Specifically it will be necessary to (i) Libraries
maximize selectivity for binding to bacteria over
serum proteins and host cells, (ii) maximize bac- Although there are many ways to synthesize and
tericidal activity of bound peptide, (iii) minimize screen peptide libraries (Hilpert et al. 2005; Lam
susceptibility to proteolytic degradation, (iv) et al. 1991; Dooley et al. 1994; Chen et al. 1996;
minimize residual cytotoxicity, and (v) maximize Frank 2002; Humet et al. 2003; Rathinakumar
solubility under physiological conditions. Thus, and Wimley 2008; Deuss et al. 2013; Wiedman
the design of a systemically active AMP is like a et al. 2016), we focus here on the approach we
puzzle in which each of these coupled factors have taken recently to identify membrane-active
must be simultaneously minimized or maximized peptides with specific properties (Rathinakumar
without negatively affecting the others. Yet, other and Wimley 2010; Rathinakumar et al. 2009;
than proteolytic susceptibility, the sequence-­ Krauson et al. 2013; Kauffman et al. 2018; Li
structure-­function relationships for none of these et al. 2018; Wiedman et al. 2016; Krauson et al.
factors are understood well enough to make use- 2012, 2015; Marks et al. 2011; Rausch et al.
ful predictions or to enable rational engineering. 2005). We design small, iterative libraries of
This is why most new AMPs described in the lit- 10–30,000 members that are based on a template
erature are either identified from natural sources sequence with known activity. Using a library of
or are discovered in the laboratory by simple trial this level of diversity means that we can design a
and error under standard conditions. library synthesis scheme that provides a rela-
How can one simultaneously optimize these tively large amount of each library member to
various factors when they are incompletely cou- work with. This, in turn, enables us to use much
pled and when the molecular mechanisms are not more complex screens, and it allows us to screen
understood in enough molecular detail to enable the same library member in multiple parallel
rational design? In this chapter, we discuss how assays, which is needed to screen for bactericidal
this can be done using synthetic molecular evolu- activity against multiple microbes as well as tox-
tion (SME). By this we mean iterative screening icity against host cells.
of rationally designed peptide libraries that are Our approach to library synthesis and quality
based on known AMPs and are designed using control is well described in many papers
known physical principles to choose rational (Rathinakumar and Wimley 2008, 2010;
248 W. C. Wimley

Rathinakumar et al. 2009; Krauson et al. 2012, (i) cytosolic RBC proteases, released by back-
2013, 2015; Wiedman et al. 2016; Marks et al. ground autolysis or by a small amount of direct
2011; Rausch et al. 2005; He et al. 2011, 2013). hemolysis, and (ii) direct interactions of AMPs
In short, a photocleavable linker is added to with the host cells that reduce the pool of avail-
Tentagel-NH2 Megabeads, followed by library able AMP by competition. Some peptides are
construction using the split and recombine more susceptible to the former and some are
approach (Chen et al. 1996). Quality control is more susceptible to the latter (Starr et al. 2016).
assured with HPLC, mass spec, and sequencing Any AMP that will maintain antibacterial activity
performed on multiple individual beads. Each in vivo will need to be resistant to both proteoly-
bead contains about 1 nmol of peptide which is sis and direct host cell inhibition.
released by UV light providing 100 μL of a
10 μM solution. This is sufficient peptide to per-
form multiple parallel assays on each library 13.8.1 Radial Diffusion
member. Screening of such small iterative librar-
ies has long been routine in the laboratory There are a number of antimicrobial assays that
(Rathinakumar and Wimley 2008, 2010; can be used in a high-throughput screen format.
Rathinakumar et al. 2009; Krauson et al. 2012, Here we will discuss the strengths and weak-
2013, 2015; Wiedman et al. 2016; Marks et al. nesses of some of them in the context of
2011; Rausch et al. 2005; He et al. 2011, 2013) SME. Radial diffusion is an assay in which a thin
and requires no special robotic instrumentation. bacteria-seeded agar layer is overlaid with a ster-
ile, nutrient-rich agar enabling a lawn of bacteria
to grow between the layers, except where growth
13.8 High-Throughput Screening is inhibited in a zone around a locally applied
for Antibacterial Activity antibiotic. For peptides to be tested in the pres-
ence of host cells and serum, a small hole can be
As we envision the discovery of systemically made in the lower agar layer and a small volume
active AMPs, screening must be done under con- of peptide mixture can be introduced. After
ditions that most closely mimic the conditions allowing some time for peptide diffusion into the
experienced by a peptide antibiotic in vivo. At a agar, the nutrient overlay is added, and the plate
minimum, there must be a high concentration of is allowed to grow overnight. The following day,
host cells and a high concentration of serum that the lawn of bacteria is visible, and the zones of
has been heat inactivated to eliminate activated inhibition are readily observed and quantitated,
complement proteins. RBCs also contain a very as shown in Fig. 13.5. Radial diffusion has the
high concentration of proteases that are somewhat advantage that it is readily adapted to high
different than serum proteases (Starr and Wimley throughput and that it is quantitative: A larger
2017), so any screen with RBCs and/or serum will zone of inhibition is related to better activity.
contain many realistic proteases. We have been However, the size of the zone of inhibition may
experimenting with using human red blood cells also be strongly affected by the ability of the anti-
(RBCs) as a host tissue analog. This is advanta- biotic to diffuse in agar rather than by its inherent
geous because it is easy to procure large amounts antibacterial activity. MIC-based quantitation by
of fresh, concentrated human RBCs (Fig. 13.4). radial diffusion using serial dilution circumvents
In Fig. 13.5 we show the effect of preincuba- this problem, but it cannot be accomplished in a
tion of 1 × 109 RBC/mL on the antimicrobial screen. Further, radial diffusion reports on inhibi-
activity of a set of natural and synthetic AMPs. In tion of bacterial growth, which does not neces-
many cases, but not all, RBCs inhibit the antimi- sarily indicate sterilization. In Fig. 13.5 we show
crobial activity when AMPs are preincubated for some examples of radial diffusion-based screens
a few minutes with RBCs. We subsequently of a peptide library tested simultaneously against
showed that this inhibition is due to two factors: the Gram-negative Escherichia coli and against
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 249

Fig. 13.4 Human red blood cells (RBCs) inhibit antimi- tion) causes inhibition of most. We have shown that such
crobial peptides. As described elsewhere (Starr et al. host cell inhibition is the result of both direct RBC bind-
2016), preincubation of natural and synthetic AMPs with ing and also proteolysis of the AMP by the cytosolic pro-
1 × 109 human RBC/mL (2% of physiological concentra- teases found in human RBCs (Starr and Wimley 2017)

the Gram-positive Staphylococcus aureus. Some have developed screens based on broth steriliza-
library members inhibit only one or the other spe- tion, which we describe next.
cies, while some inhibit both.
Incorporation of concentrated host cells and
serum is easily accomplished in radial diffusion, 13.8.2 Broth Sterilization
as they can be mixed with peptide prior to intro-
duction of the whole mixture into the well in the Broth sterilization is an unambiguous, all-or-­
agar layer. However the environment in the gel none assay for sterilization. Broth assays are
mimics physiological conditions less well done in liquid media inoculated with bacteria and
because peptides can diffuse into the agar/aga- treated with antibiotic. After overnight incuba-
rose, while the host cells cannot. On the other tion, cultures are assayed for absence or presence
hand, strong binding of the peptide to the host of live bacteria. Survival of any bacteria gener-
cell will prevent diffusion of peptide and will ally means that they will grow to a high density
inhibit activity. In Fig. 13.6 we show the effect of after overnight incubation, while sterilized cul-
concentrated human RBCs on radial diffusion tures will remain sterile. These two outcomes can
against S. aureus. While powerful and quantita- easily be measured with optical density, and ste-
tive, we have found that the most significant fault rility can be verified by plating the sterile culture
of radial diffusion as a screening method is that it on nutrient agar and noting the presence or
does not necessarily select for peptides with ster- absence of colony-forming units (CFUs). Broth
ilizing activity. To overcome this barrier, we also sterilization assays can be modified by the
250 W. C. Wimley

a­ddition of concentrated host cells and heat-­

inactivated serum to test for sterilization under
physiologically relevant conditions. When con-
centrated host cells, such as concentrated human
RBCs, are used, the growth of bacteria cannot be
measured directly by turbidity, so a secondary
plate can be inoculated and allowed to grow over-
night. Alternately, aliquots can be spotted on
nutrient agar and CFUs can be counted.
Broth dilution assays are not quantitative in a
high-throughput screen. They are binary tests in
which library members will either be positive or
negative for sterilization under the conditions of
the screen. Typically screens can only be done
under one condition for each library member
tested. We propose testing antibiotic activity
Fig. 13.5 Example radial diffusion-based screening of against multiple species simultaneously limiting
members of a peptide library in parallel against two the amount of each library member available.
organisms using radial diffusion. The same set of peptide Thus it is critical to adjust the stringency such
library members were screened against the Gram-negative
E. coli and the Gram-positive S. aureus with radial diffu- that a small number of leads are identified. The
sion. Zones of inhibition are observed for some library stringency of a broth sterilization assay can be
members. Some active library members inhibit both bac- modified by adjusting the inoculum size, antibi-
teria (yellow), while others inhibit only one of the two otic concentration, incubation time, or other fac-
tors. A possible scheme for a broth dilution
screen is shown in Fig. 13.7, along with the
results of screening members of an AMP library
using broth dilution.

13.8.3 Reduction of Colony-Forming

Units

CFU reduction is a hybrid assay that enables

counting of live bacteria remaining in a solution
after antibacterial treatment. It essentially reports
on the same phenomenon as broth dilution, yet
can be done with less labor and in less time. In
this assay, which is readily adapted to high
throughput, bacteria and antibiotic, with host cells
and/or serum, are incubated together for an
amount of time that enables killing and then are
Fig. 13.6 The effect of human RBCs on the activity of
peptide library members against S. aureus using radial dif-
spotted on a nutrient agar plate at high nominal
fusion. Example screening of a peptide library S. aureus CFU counts and grown overnight. In the absence
with radial diffusion. Both plates were screened with the of bactericidal activity, a dense mat of bacteria
same library members. The samples added to the right will grow. However, when only a small fraction of
plate had been incubated with 1x109 human RBC/mL
prior to use. Most of the active library members are inhib-
bacteria survive, or none at all, a countable num-
ited by RBCs. A few library members are not inhibited by ber of colonies will grow, a quantitative result that
RBCs can be used to rank order AMPs in a screen. CFU
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 251

Fig. 13.7 Peptide library screening using broth sterilization. 96-well plate after screening a library for sterilizing activity
Left: One possible scheme for screening a peptide library for against E. coli. Wells are either transparent or opaque.
solubility, for physiologically relevant broad-­spectrum bacte- Transparent wells have no colony-­forming units on nutrient
ricidal activity, and for lack of toxicity. Right: Example agar, confirming sterility. Bottom row is for controls

reduction assays can readily be modified by the serum which can be protective. Here we suggest
addition of host cells and serum, as above. In that toxicity be measured in parallel using sensi-
Fig. 13.8 we show a possible scheme for SME tive human cancer cell lines, such as HeLa cells
using CFU counts along with the results of a test such that peptides with low toxicity can be identi-
screen of AMPs from a library. Note that positive, fied during the screen.
sterilizing sequences can readily be identified by
the absence of CFUs, which amounts to more
than four logs of CFU reduction. 13.9 ESKAPE Pathogens

While many bacteria can infect humans and har-

13.8.4 Cytotoxicity and Hemolysis bor drug resistance, there are a small set that
account for the majority of morbidity and mortal-
In a screen for antimicrobial assays under physi- ity (Boucher et al. 2009; Centers For Disease
ological conditions, toxicity must also be mea- Control 2014). These include Clostridium difficile,
sured simultaneously. This can be accomplished often associated with gastrointestinal infections,
in assays that are done in the presence of RBCs as and the ESKAPE pathogens, whose acronym indi-
host cells by also measuring hemolysis. However, cates Enterococcus faecalis, Staphylococcus
hemolysis may not be sensitive enough to be use- aureus, Klebsiella pneumonia, Acinetobacter bau-
ful, especially in the presence of concentrated mannii, Pseudomonas aeruginosa, and
252 W. C. Wimley

Fig. 13.8 Peptide library screening using the reduction activity, and for lack of toxicity. Right: Example nutrient
in colony-forming units (CFU). Left: One possible agar plate after spotting library members mixed with bac-
scheme for screening a peptide library for solubility, for teria for 1 h. Clear spots with no colonies have been
physiologically relevant broad-spectrum bactericidal sterilized

Enterobacteriaceae, which includes Escherichia, 13.10 Future Prospects

Salmonella, Vibrio, and Shigella species, among
others. Since some AMPs have variable potencies The physical chemistry-based action of AMPs on
against these different organisms, screening for the bacteria leads to broad-spectrum activity and
broadest activity must be done against multiple more difficulty in evolving resistance, which
species simultaneously. We previously screened accounts for some of the appeal of AMPs as
against two ESKAPE bacteria, E. coli and S. potential drugs. Yet, these same properties also
aureus, and a fungus, Cryptococcus neoformans, drive nonspecific interactions with serum protein
simultaneously and found very low overlap in and host cells that reduce the effectiveness of
activities. This enabled the identification of the AMPs. In this chapter, we have presented the
rare peptides with broad-­ spectrum activity concept of synthetic molecular evolution as a tool
(Rathinakumar and Wimley 2010). We suggest to enable the discovery of AMPs, early in the
screening in parallel against S. aureus two Gram- development pipeline, that are less affected by
negative ESKAPE pathogens, P. aeruginosa and host cell and serum protein binding. We remain
either K. pneumonia or A. baumannii. hopeful that this new approach will finally enable
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 253

AMP researchers to bridge the gap between the tic potential as anti-infective drugs. Curr Eye Res
30:505–515
laboratory bench and the clinic. Hamill P, Brown K, Jenssen H, Hancock RE (2008) Novel
anti-infectives: is host defence the answer? Curr Opin
Biotechnol 19:628–636
References He L, Hoffmann AR, Serrano C, Hristova K, Wimley WC
(2011) High-throughput selection of transmembrane
sequences that enhance receptor tyrosine kinase acti-
Agawa Y et al (1991) Interaction with phospholipid
vation. J Mol Biol 412:43–54
bilayers, ion channel formation, and antimicro-
He J et al (2013) Direct cytosolic delivery of polar cargo
bial activity of basic amphipathic a-helical model
to cells by spontaneous membrane-translocating pep-
peptides of various chain lengths. J Biol Chem
tides. J Biol Chem 288:29974–29986
266:20218–20222
Hilpert K, Volkmer-Engert R, Walter T, Hancock RE
Arias CA, Murray BE (2009) Antibiotic-resistant bugs in
(2005) High-throughput generation of small antibac-
the 21st century – a clinical super-challenge. N Engl
terial peptides with improved activity. Nat Biotechnol
J Med 360:439–443
23:1008–1012
Boucher HW et al (2009) Bad bugs, no drugs: no
Humet M et al (2003) A positional scanning combinato-
ESKAPE! An update from the Infectious Diseases
rial library of peptoids as a source of biological active
Society of America. Clin Infect Dis 48:1–12
molecules: identification of antimicrobials. J Comb
Centers For Disease Control (2014) Antibiotic resistance
Chem 5:597–605
threats in the United States, 2013. http://www.cdc.
Joly V, Jidar K, Tatay M, Yeni P (2010) Enfuvirtide: from
gov/drugresistance/pdf/ar-threats-2013-508.pdf.
basic investigations to current clinical use. Expert
Chen CL, Strop P, Lebl M, Lam KS (1996) One bead-­
Opin Pharmacother 11:2701–2713
one compound combinatorial peptide library: different
Kauffman WB, Guha S, Wimley WC (2018) Synthetic
types of screening. Methods Enzymol 267:211–219
molecular evolution of hybrid cell penetrating pep-
de Breij A et al (2018) The antimicrobial peptide SAAP-­
tides. Nat Commun 9:2568
148 combats drug-resistant bacteria and biofilms. Sci
Kazemzadeh-Narbat M et al (2010) Antimicrobial pep-
Transl Med 10(423):eaan4044
tides on calcium phosphate-coated titanium for the pre-
Deuss PJ, Arzumanov A, Williams DL, Gait MJ (2013)
vention of implant-associated infections. Biomaterials
Parallel synthesis and splicing redirection activity of
31:9519–9526
cell-penetrating peptide conjugate libraries of a PNA
Krauson AJ, He J, Wimley WC (2012) Gain-of-function
cargo. Org Biomol Chem 11:7621–7630
analogues of the pore-forming peptide melittin
Dobson AJ, Purves J, Kamysz W, Rolff J (2013)
selected by orthogonal high-throughput screening.
Comparing selection on S. aureus between antimi-
J Am Chem Soc 134:12732–12741
crobial peptides and common antibiotics. PLoS One
Krauson AJ, He J, Hoffmann AR, Wimley AW, Wimley
8:e76521
WC (2013) Synthetic molecular evolution of pore-­
Dooley CT et al (1994) An all D-amino acid opioid pep-
forming peptides by iterative combinatorial library
tide with central analgesic activity from a combinato-
screening. ACS Chem Biol 8:823–831
rial library. Science 266:2019–2022
Krauson AJ et al (2015) Conformational fine-tuning of
Easton DM, Nijnik A, Mayer ML, Hancock RE (2009)
pore-forming peptide potency and selectivity. J Am
Potential of immunomodulatory host defense pep-
Chem Soc 137:16144–16152
tides as novel anti-infectives. Trends Biotechnol
Kulagina NV, Shaffer KM, Anderson GP, Ligler FS, Taitt
27:582–590
CR (2006) Antimicrobial peptide-based array for
Fedders H, Podschun R, Leippe M (2010) The antimi-
Escherichia coli and Salmonella screening. Anal Chim
crobial peptide Ci-MAM-A24 is highly active against
Acta 575:9–15
multidrug-resistant and anaerobic bacteria pathogenic
Kulagina NV, Shaffer KM, Ligler FS, Taitt CR (2007)
for humans. Int J Antimicrob Agents 36:264–266
Antimicrobial peptides as new recognition molecules
Fjell CD, Hancock RE, Cherkasov A (2007) AMPer: a
for screening challenging species. Sensors Actuators
database and an automated discovery tool for antimi-
B Chem 121:150–157
crobial peptides. Bioinformatics 23:1148–1155
LaBonte J, Lebbos J, Kirkpatrick P (2003) Enfuvirtide.
Fox JL (2013) Antimicrobial peptides stage a comeback.
Nat Rev Drug Discov 2:345–346
Nat Biotechnol 31:379–382
Lam KS et al (1991) A new type of synthetic peptide
Frank R (2002) The SPOT-synthesis technique. synthetic
library for identifying ligand-binding activity. Nature
peptide arrays on membrane supports – principles and
(London) 354:82–84
applications. J Immunol Methods 267:13–26
Li S et al (2018) Potent macromolecule-sized poration
Gordon YJ, Romanowski EG, McDermott AM (2005) A
of lipid bilayers by the Macrolittins, a synthetically
review of antimicrobial peptides and their therapeu-
254 W. C. Wimley

evolved family of pore-forming peptides. J Am Chem Rathinakumar R, Wimley WC (2008) Biomolecular engi-
Soc 140:6441–6447 neering by combinatorial design and high-throughput
Marks JR, Placone J, Hristova K, Wimley WC (2011) screening: small, soluble peptides that permeabilize
Spontaneous membrane-translocating peptides by membranes. J Am Chem Soc 130:9849–9858
orthogonal high-throughput screening. J Am Chem Rathinakumar R, Wimley WC (2010) High-throughput
Soc 133:8995–9004 discovery of broad-spectrum peptide antibiotics.
Mechkarska M et al (2013) An analog of the host-defense FASEB J 24:3232–3238
peptide hymenochirin-1B with potent broad-­spectrum Rathinakumar R, Walkenhorst WF, Wimley WC (2009)
activity against multidrug-resistant bacteria and Broad-spectrum antimicrobial peptides by rational
immunomodulatory properties. Peptides 50:153–159 combinatorial design and high-throughput screening:
Molhoek EM, van Dijk A, Veldhuizen EJ, Haagsman the importance of interfacial activity. J Am Chem Soc
HP, Bikker FJ (2011) Improved proteolytic stabil- 131:7609–7617
ity of chicken cathelicidin-2 derived peptides by Rausch JM, Marks JR, Wimley WC (2005) Rational com-
D-amino acid substitutions and cyclization. Peptides binatorial design of pore-forming beta-sheet peptides.
32:875–880 Proc Natl Acad Sci U S A 102:10511–10515
Moy TI et al (2009) High-throughput screen for novel Riedl S, Zweytick D, Lohner K (2011a) Membrane-active
antimicrobials using a whole animal infection model. host defense peptides – challenges and perspectives
ACS Chem Biol 4:527–533 for the development of novel anticancer drugs. Chem
Nguyen LT et al (2010) Serum stabilities of short trypto- Phys Lipids 164:766–781
phan- and arginine-rich antimicrobial peptide analogs. Riedl S et al (2011b) In search of a novel target – phos-
PLoS One 5:e12684 phatidylserine exposed by non-apoptotic tumor cells
Okada M, Natori S (1983) Purification and characteriza- and metastases of malignancies with poor treatment
tion of an antibacterial protein from haemolymph of efficacy. Biochim Biophys Acta 1808:2638–2645
Sarcophaga peregrina (flesh-fly) larvae. Biochem Savini F et al (2017) Cell-density dependence of host-­
J 211:727–734 defense peptide activity and selectivity in the presence
Otto M (2012) MRSA virulence and spread. Cell of host cells. ACS Chem Biol 12:52–56
Microbiol 14:1513–1521 Savini F, Bobone S, Roversi D, Mangoni M, Stella
Park SC et al (2011) Synthetic diastereomeric-­ L (2018) From liposomes to cells: filling the gap
antimicrobial peptide: antibacterial activity against between physicochemical and microbiological studies
multiple drug-resistant clinical isolates. Biopolymers of the activity and selectivity of host-defense peptides.
96:130–136 Pept Sci 110:e24041
Patterson-Delafield J, Szklarek D, Martinez RJ, Lehrer RI Schlusselhuber M et al (2014) In vitro effectiveness of
(1981) Microbicidal cationic proteins of rabbit alveo- the antimicrobial peptide eCATH1 against antibiotic-­
lar macrophages: amino acid composition and func- resistant bacterial pathogens of horses. FEMS
tional attributes. Infect Immun 31:723–731 Microbiol Lett 350:216–222
Perron GG, Zasloff M, Bell G (2006) Experimental evo- Selsted ME, Brown DM, DeLange RJ, Harwig SSL,
lution of resistance to an antimicrobial peptide. Proc Lehrer RI (1985) Primary structures of six antimicro-
Biol Sci 273:251–256 bial peptides of rabbit peritoneal neutrophils. J Biol
Peschel A (2002) How do bacteria resist human antimi- Chem 260:4579–4584
crobial peptides? Trends Microbiol 10:179–186 Starr CG, Wimley WC (2017) Antimicrobial peptides are
Peschel A, Sahl HG (2006) The co-evolution of host cat- degraded by the cytosolic proteases of human erythro-
ionic antimicrobial peptides and microbial resistance. cytes. Biochim Biophys Acta 1859:2319–2326
Nat Rev Microbiol 4:529–536 Starr CG, He J, Wimley WC (2016) Host cell interactions
Pollard JE et al (2012) In vitro evaluation of the poten- are a significant barrier to the clinical utility of peptide
tial for resistance development to ceragenin CSA-13. antibiotics. ACS Chem Biol 11:3391–3399
J Antimicrob Chemother 67:2665–2672 Starr CG, Maderdrut JL, He J, Coy DH, Wimley WC
Poveda E, Briz V, Soriano V (2005) Enfuvirtide, the first (2018) Pituitary adenylate cyclase-activating poly-
fusion inhibitor to treat HIV infection. AIDS Rev peptide is a potent broad-spectrum antimicrobial
7:139–147 peptide: structure-activity relationships. Peptides
Qureshi NM, Coy DH, Garry RF, Henderson LA (1990) 104:35–40
Characterization of a putative cellular receptor for Steiner H, Hultmark D, Engstrom A, Bennich H, Boman
HIV-1 transmembrane glycoprotein using synthetic HG (1981) Sequence and specificity of two antibac-
peptides. AIDS 4:553–558 terial proteins involved in insect immunity. Nature
Rapaport D, Ovadia M, Shai Y (1995) A synthetic peptide (London) 292:246–248
corresponding to a conserved heptad repeat domain Steiner H, Andreu D, Merrifield RB (1988) Binding and
is a potent inhibitor of Sendai virus-cell fusion: an action of cecropin and cecropin analogues: antibac-
emerging similarity with functional domains of other terial peptides from insects. Biochim Biophys Acta
viruses. EMBO J 14:5524–5531 939:260–266
13 Application of Synthetic Molecular Evolution to the Discovery of Antimicrobial Peptides 255

Tran D et al (2002) Homodimeric theta-defensins from Wimley WC (2010) Describing the mechanism of anti-
rhesus macaque leukocytes: isolation, synthesis, anti- microbial peptide action with the interfacial activity
microbial activities, and bacterial binding properties model. ACS Chem Biol 5:905–917
of the cyclic peptides. J Biol Chem 277:3079–3084 Wolfmeier H, Pletzer D, Mansour SC, Hancock REW
Wang G, Li X, Wang Z (2016) APD3: the antimicrobial (2017) New perspectives in biofilm eradication. ACS
peptide database as a tool for research and education. Infect Dis 4:93–106
Nucleic Acids Res 44:D1087–D1093 Yeaman MR, Yount NY (2003) Mechanisms of antimi-
Werle M, Bernkop-Schnurch A (2006) Strategies to crobial peptide action and resistance. Pharmacol Rev
improve plasma half life time of peptide and protein 55:27–55
drugs. Amino Acids 30:351–367 Zasloff M (1987) Magainins, a class of antimicrobial
Wiedman G, Kim SY, Zapata-Mercado E, Wimley WC, peptides from Xenopus skin: isolation, charac-
Hristova K (2016) PH-triggered, macromolecule-sized terization of two active forms, and partial cDNA
poration of lipid bilayers by synthetically evolved pep- sequence of a precursor. Proc Natl Acad Sci U S A
tides. J Am Chem Soc 139:937–945 84:5449–5453
 AMPs as Anti-biofilm Agents
for Human Therapy 14
and Prophylaxis

Hawraa Shahrour, Raquel Ferrer-Espada,

Israa Dandache, Sergio Bárcena-Varela,
Susana Sánchez-Gómez, Ali Chokr,
and Guillermo Martínez-de-Tejada

Abstract infective endocarditis, pneumonia, wound

Microbial cells show a strong natural tendency infections, dental caries, infections of indwell-
to adhere to surfaces and to colonize them by ing devices, etc. AMPs are well suited to com-
forming complex communities called bio- bat biofilms because of their potent bactericidal
films. In this growth mode, biofilm-forming activity of broad spectrum (including resting
cells encase themselves inside a dense matrix cells and persisters) and their ability to first
which efficiently protects them against anti- penetrate and then to disorganize these struc-
microbial agents and effectors of the immune tures. In addition, AMPs frequently synergize
system. Moreover, at the physiological level, with antimicrobial compounds and were
biofilms contain a very heterogeneous cell recently reported to repress the molecular
population including metabolically inactive pathways leading to biofilm formation.
organisms and persisters, which are highly Finally, there is a very active research to
tolerant to antibiotics. The majority of human develop AMP-containing coatings that can
infectious diseases are caused by biofilm-­ prevent biofilm formation by killing microbial
forming microorganisms which are responsi- cells on contact or by locally releasing their
ble for pathologies such as cystic fibrosis, active principle. In this chapter we will

H. Shahrour
Department of Microbiology and Parasitology,
I. Dandache · A. Chokr
University of Navarra, Pamplona, Spain
Laboratory of Microbiology, Department of Life &
Laboratory of Microbiology, Department of Life & Earth Sciences, Faculty of Sciences I, Lebanese
Earth Sciences, Faculty of Sciences I, Lebanese University, Hadat campus, Beirut, Lebanon
University, Hadat campus, Beirut, Lebanon
Platform of Research and Analysis in Environmental
Platform of Research and Analysis in Environmental Sciences (PRASE), Doctoral School of Sciences and
Sciences (PRASE), Doctoral School of Sciences and Technologies, Lebanese University, Hadat Campus,
Technologies, Lebanese University, Hadat Campus, Beirut, Lebanon
Beirut, Lebanon
S. Bárcena-Varela · G. Martínez-de-Tejada (*)
R. Ferrer-Espada Department of Microbiology and Parasitology,
Department of Microbiology and Parasitology, University of Navarra, Pamplona, Spain
University of Navarra, Pamplona, Spain e-mail: [email protected]
Wellman Center for Photomedicine, Massachusetts S. Sánchez-Gómez
General Hospital, Harvard Medical School, Bionanoplus S.L. Polígono Mocholí,
Boston, MA, USA Noain, Navarra, Spain

© Springer Nature Singapore Pte Ltd. 2019 257

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_14
258 H. Shahrour et al.

describe these strategies and discuss the per- beneficial to microorganisms which have a gen-
spectives of the use of AMPs as anti-biofilm eral tendency to colonize all types of environ-
agents for human therapy and prophylaxis. ments using this growth mode (Jefferson 2004).
The development of a biofilm is a multi-step
Keywords process starting with an initial reversible attach-
Biofilm · Antimicrobial peptide · Host-­ ment of planktonic cells to a surface followed by
defense peptide · Antibiotic lock therapy · a maturation phase. Attachment can occur both
Medical implant on abiotic and biotic surfaces including living tis-
sues (Palmer et al. 2007). Initial attachment is
mediated by electrostatic or hydrophobic interac-
tions between planktonic cells and a particular
14.1 Introduction to Microbial surface. Depending on conditions such as steric
Biofilms hindrance, temperature, and hydrodynamic
forces at the site of attachment, this initial inter-
Biofilms are complex aggregates made up of action can lead to irreversible adhesion to that
cells adhered to each other and to a solid surface surface (Fig. 14.1).
via an extracellular matrix (Stanley and Lazazzera Following adhesion, a community of micro-
2004; Mielich-Süss and Lopez 2015). Frequently, bial cells (or “microcolony”) develops as a result
they originate from a single bacterial or fungal of clonal growth and stable cell-cell interactions.
clone, although sometimes these structures con- This phase is usually accompanied with secretion
tain mixtures of different organisms including of EPS matrix components that stabilize the bio-
bacteria, fungi, algae, and protozoa. Within the film structure. Each microcolony is separated
biofilm, cells are embedded in a self-produced from the others by water channels that allow the
highly hydrated extracellular matrix mediating diffusion of nutrients, oxygen, and other sub-
cell-to-cell and cell-to-surface interactions stances. Subsequent growth of microcolonies
(Fig. 14.1). gives rise to a mature biofilm (or “macrocolony”)
The matrix accounts for 50–90% of the total that acquires the typical mushroom-like structure
organic content of the biofilm mass and is (Lehner et al. 2005). A final step involves the
composed of a complex mixture of extracellular detachment or dispersal of single cells or cell
polymeric substances (EPS) including exopoly- clusters that have the potential to colonize sur-
saccharides, proteins, and extracellular DNA rounding sites either in planktonic form or by
(Donlan 2002; Kostakioti et al. 2013; Flemming establishing new sessile communities (Landini
and Wingender 2010). EPS acts as a physical bar- et al. 2010) (Fig. 14.1).
rier against external threats and traps exogenous Biofilm-forming cells are much better suited
substances such as nucleic acids, proteins, miner- than their planktonic counterparts to colonize and
als, nutrients, and cell wall components found in persist inside a host (Jamal et al. 2018). In
the local environment. Thus, biofilm formation is humans, the former cells cause serious medical

Fig. 14.1 Life cycle of biofilm-forming organisms (Based on Lebeaux et al. 2014). Persister cells are depicted in red
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 259

complications such as chronic infections that are reported to play an important role in the persis-
very difficult to eradicate. Biofilms have been tence of infections, especially in immunocom-
reported to be involved in 80% of all human promised patients, a more detailed understanding
infections including native valve endocarditis of these phenomena at the molecular level is nec-
(Giamarellou 2002), osteomyelitis (Gbejuade et al. essary to design new agents that target biofilms.
2014), dental caries (Karygianni et al. 2016), The high doses of antimicrobials and lengthy
middle ear infections (Akyıldız et al. 2013), regimes needed to treat biofilm-associated infec-
ocular implant infections (Bispo et al. 2015), and tions increase the risk of adverse reactions and
chronic lung infections in cystic fibrosis patients facilitate the emergence of multidrug-resistant
(Høiby et al. 2010). The presence of inert surfaces strains. In this context, it has been estimated that
like those of medical devices (e.g., catheters, drug-resistant infections lead to the death of at
prostheses, valves, pacemakers) greatly facilitates least 25,000 people per year in Europe (ECDC
biofilm formation and the emergence of severe 2009) and cause 23,000 deaths in the USA (CDC
implant-associated infections (Davey and O’Toole 2013). Incidence of infections caused by
2000; Guggenbichler et al. 2011). Efficient clear- antimicrobial-­resistant pathogens is in constant
ance of these infections often requires implant expansion, and it has been estimated that mortal-
removal which is associated with high morbidity ity due to these diseases will reach an annual toll
and economic losses. of ten million deaths by 2050 if no breakthroughs
Biofilm-forming cells have markedly different in antimicrobial therapy occur (Hancock 2015).
gene expression profiles, physiology, and mor- This situation represents a major medical chal-
phology compared to planktonic cells. In part, lenge highlighting the need for new therapeutic
this explains the increased antibiotic resistance of options with specific anti-biofilm activity (Wu
the former (up to 1000 times, according to some et al. 2015).
reports) compared to the latter (Hall and Mah
2017). Moreover, it has been proposed that the
EPS can slow down or completely block the pen- 14.2  MPs as Anti-biofilm Agents:
A
etration of some antimicrobials into the biofilm Mechanisms of Action
cells. Finally, a fraction of cells, named persisters
or dormant cells, was shown to differentiate into In the last years, numerous biofilm control strate-
a metabolically inactive state that renders them gies have been proposed but no clinically useful
completely tolerant to antibiotics (Mah and therapy has yet emerged. The problem is particu-
O’Toole 2001; Stewart and Costerton 2001; larly severe in hospitals, where patients are often
Harms et al. 2016). Globally, the main changes immunocompromised and are confronted with
occurring in biofilm-forming cells include induc- increased chances of infection with antimicrobial-­
tion of the general stress response, increasing resistant pathogens. Numerous characteristics
expression of multiple drug resistance (MDR) make AMPs very attractive candidates for the
pumps, activation of quorum sensing (QS) sys- development of anti-biofilm therapies. Among
tems, and reorganization of outer membrane pro- other features, many of these compounds have
teins (OMP) (Rabin et al. 2008). broad-spectrum of microbicidal activity, reduced
The combination of the abovementioned tendency to induce resistance, ability to kill met-
mechanisms is the basis for the increased resis- abolically inactive cells, and synergistic activity
tance of biofilms, not only to conventional anti- when combined with commonly used drugs
microbials but also to phagocytosis and other (Pletzer et al. 2016). In addition, some AMPs dis-
effectors of the immune system (Roilides et al. play multiple mechanisms of action which enable
2015). Moreover, horizontal transfer of resis- them to interfere with various stages of biofilm
tance and virulence genes takes place with high formation. These include the ability to (i) inhibit
efficiency in the dense cell population of the bio- the growth of planktonic cells, (ii) prevent the
film community. While these mechanisms are initial adhesion of microbial cells to surfaces,
260 H. Shahrour et al.

(iii) disorganize mature biofilms, and (iv) kill gen Pseudomonas aeruginosa (Singh et al. 2002)
biofilm-embedded organisms (Fig. 14.2). (Table 14.1). Notably, this activity seemed to be
However, of all AMPs developed in the last dependent on the ability of the peptide to activate
decades, only a few exhibit a significant anti-­ twitching motility which triggers the daughter
biofilm activity below their minimal inhibitory cells to move away from the point of parental cell
concentration (MIC) (Pletzer and Hancock division.
2016). On the other hand, AMPs were shown by other
Some prominent examples of AMPs display- authors to kill preformed biofilms at subinhibi-
ing biofilm preventive activity include the human tory concentrations. For instance, De la Fuente-­
cathelicidin peptide LL-37, a compound pro- Nuñez and collaborators developed synthetic
duced by mucosal epithelial cells and several cathelicidin-derived peptides (DJK-5 and DJK-6)
cells of the immune system (Table 14.1). LL-37 which eradicated biofilms established by
was reported to inhibit biofilm formation at con- multidrug-­ resistant (MDR) organisms (De la
centrations much lower than its MIC (Overhage Fuente-Núñez et al. 2015) (Table 14.1). Notably,
et al. 2008; De la Fuente-Núñez et al. 2012). In a Anunthawan and collaborators reported that the
different report, this activity was detectable two tryptophan-rich cationic amphipathic pep-
against urinary tract isolates of Staphylococcus tides KT2 and RT2 killed biofilm-forming cells
aureus and Escherichia coli at 1/32 to 1/2 of the of E. coli O157:H7 at sub-MBIC (minimal bio-
peptide MIC (Luo et al. 2017). Other human pep- film inhibition concentration) levels (Anunthawan
tides such as lactoferrin were shown to display et al. 2015). Other researchers found that 4 chi-
potent biofilm preventive activity at 20 μg/mL (a meric AMPs developed by them exerted potent
sub-MIC value) against the opportunistic patho- antibacterial and anti-biofilm activity against 19

Fig. 14.2 Mechanisms of anti-biofilm action displayed by AMPs

 14

Table 14.1 Selected examples of antimicrobial peptides with anti-biofilm activity

Active anti-
biofilm
Peptide Source MIC concentration Microbial species Anti-biofilm mechanism Reference
LL-37 Human 32 μg/mL 1–16 μg/mL P. aeruginosa; S. epidermidis; Decreased attachment of De la Fuente-Núñez et al. (2012),
E. coli; S. aureus; bacterial cells, stimulation of Kanthawong et al. (2012), Gabriel
Burkholderia spp.; L. twitching motility, et al. (2006), Blower et al. (2015),
monocytogenes; Candida spp. influences Las and Rhl Scarsini et al. (2015), and Dean
systems et al. (2011)
Lactoferrin Human – 20 μg/mL S. mutans; S. gordonii; P. Binding and sequestering Singh et al. (2002) and Ammons
gingivalis; F. nucleatum iron in the environment and Copié (2013)
Oritavancin Semisynthetic 2–8 μg/ml 0.5–8 μg/ml S. aureus Affects membrane integrity, Belley et al. (2009)
lipoglycopeptide kills slow-growing cells
DD13-RIP Chimeric peptide 2 μg/ml 10–20 μg/ml S. aureus, S. epidermidis Interrupts QS mechanisms Balaban et al. (2004)
(IDR-)1018 Derivative of 8–128 μg/ 2–10 μg/ml A. baumannii; E. coli; K. Degradation of (p)ppGpp De la Fuente-Núñez et al. (2014)
bactenicin mL pneumonia; P. aeruginosa; S.
enterica; S. aureus
DJK-5 Synthetic analog of 1.6– 0.8–4 μg/ml P. aeruginosa; E. coli; A. Degradation of (p)ppGpp De la Fuente-Núñez et al. (2015)
AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis

active anti-biofilm 16 μg/mL baumannii; K. pneumonia; S.

peptides enterica
DJK-6 Synthetic analog of 1.6– 0.5–8 μg/ml P. aeruginosa; E. coli; A. Degradation of (p)ppGpp De la Fuente-Núñez et al. (2015)
active anti-biofilm 16 μg/mL baumannii; K. pneumonia; S.
peptides enterica
1037 Derivative of LL-37 304 μg/ 5–10 μg/mL P. aeruginosa; L. Inhibition of swimming and De la Fuente-Núñez et al. (2012)
mL monocytogenes swarming motilities and
stimulation of twitching
motility
261
262 H. Shahrour et al.

MDR-resistant clinical Acinetobacter baumannii adhesion and stimulated twitching motility of P.

isolates. These peptides were also shown to syn- aeruginosa leading to potent inhibition of biofilm
ergize with conventional antibiotics and exhib- development (Overhage et al. 2008). Other
ited low cytotoxicity against human skin cells authors reported that LL-37 was effective against
(Gopal et al. 2014). Finally, lactoferricin-derived a slime-producing strain of S. epidermidis by
peptides and lipopeptides were reported by inhibiting the initial attachment stage and subse-
Sánchez-Gómez and collaborators to cause a quent biofilm formation (Dean et al. 2011).
10,000-fold reduction in the viability of P. aeru- Other peptides exploit totally different modes
ginosa biofilms after 1 h of treatment at 10 times of action. For instance, the synthetic compound
the peptide MIC. In addition to this bactericidal IDR-1018 prevented biofilm formation and erad-
activity, some of these compounds were shown to icated preformed biofilms of clinically relevant
have a potent biofilm disorganizing and remov- bacterial species including P. aeruginosa, E.
ing activity (Sánchez-Gómez et al. 2015). coli, A. baumannii, Klebsiella pneumoniae,
Although the most prevalent mechanism of methicillin-­
resistant S. aureus (MRSA),
AMP activity against biofilms seems to involve Salmonella typhimurium, and Burkholderia
their well-known ability to bind and disturb cenocepacia at sub-MIC concentrations (De la
microbial membranes, these compounds can also Fuente-Núñez et al. 2014) (Table 14.1). Authors
exploit other modes of action. Thus, some AMPs found that IDR-1018 binds to the second mes-
have been reported to interfere with mechanisms senger ppGpp and stimulates its degradation
necessary for proper biofilm formation including inside the cell. Endogenous ppGpp is synthesized
EPS biosynthesis, QS cell communication, and in response to environmental signals as part of
regulation of genes involved in motility, biofilm the bacterial stringent stress response and seems
maturation, and persister cell generation to play an important role in biofilm formation and
(Kanthawong et al. 2012; De la Fuente-Núñez in regulating the formation of persister cells (De
et al. 2014). Moreover, endogenous peptides are la Fuente-Núñez et al. 2014). On the other hand,
potent immune modulators which can control Belley and coworkers reported that oritavancin, a
biofilms indirectly by enhancing the activities of semisynthetic lipoglycopeptide in clinical devel-
immune cells against infecting microbes (Lai and opment for the treatment of serious Gram-­
Gallo 2009). Importantly, these alternative mech- positive infections, displays multiple mechanisms
anisms are detectable at AMPs concentrations of anti-biofilm action (Table 14.1). This com-
well below their MIC, and this is a clear indica- pound not only inhibited cell wall and RNA syn-
tion that selection of candidates for anti-biofilm thesis but also disrupted the membrane potential
treatment should not be based only on the posses- and increased the membrane permeability affect-
sion of a significant antimicrobial activity. ing exponential and stationary-phase S. aureus
The mechanism of P. aeruginosa biofilm inhi- cells (Belley et al. 2009).
bition by the human cathelicidin LL-37 has been Another example of unique mechanisms of
intensively studied. Using microarray technol- action is that of peptides hepcidin 20 and HBD3
ogy, Overhage and collaborators demonstrated which appear to interfere with the production of
that the peptide was able to downregulate more the EPS of S. epidermidis strains (Brancatisano
than 50 QS-controlled genes at subinhibitory et al. 2014; Zhu et al. 2013). Authors of these
concentrations. These genes include lasI and reports found that the peptides upregulated the
rhlR which code for the QS autoinducer synthe- expression of icaR (a transcriptional repressor of
sis protein LasI and the QS regulator RhlR, the ica operon), thereby inhibiting the expression
respectively. Other genes were involved in fla- of icaA and icaD, two genes of the ica operon
gella biosynthesis, a process interfering with the responsible for the synthesis of the major extra-
flagellar-mediated motility that the cells need cellular polysaccharide PIA (polysaccharide-­
prior to its attachment to a solid surface. Globally, intercellular-­adhesin). Additionally, hepcidin 20
these changes in gene expression decreased cell was reported to destabilize biofilm structure by
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 263

binding to negatively charged bacterial cells and Microbial adhesion and subsequent prolifera-
extracellular DNA, thus hindering proper interac- tion can be prevented by modifying surface prop-
tions between the components of the extracellu- erties such as charge, hydrophobicity/
lar matrix. A summary of selected examples of hydrophilicity, or surface chemistry in a way that
AMPs exerting anti-biofilm properties in vitro or precludes or hinders microbial colonization. One
in vivo is reported in Table 14.1. example of this strategy is the use of hydrophilic
Up to the present time, more than 1200 AMPs polymer coatings like those based on immobi-
have been isolated and tested for their biological lized polyethylene glycol (PEG). This approach
activity against biofilms, and there is a special- has been reported to greatly reduce colonization
ized database gathering data from those com- of contact lenses, shunts, endotracheal tubes, and
pounds that display the highest potency in this urinary catheters (Busscher et al. 2012; Banerjee
regard (Biofilm Active Antimicrobial Peptide et al. 2011). Other researchers successfully used
Database http://www.baamps.it/) (Di Luca et al. surfaces functionalized with a dense layer of
2015). polymer chains to develop the so-called polymer
brush coatings (Yu et al. 2017a).
Another strategy aimed at preventing implant
14.3 Prevention of Biofilm colonization involves the development of sur-
Growth Using AMPs faces that incorporate immobilized AMPs.
Regardless of the method used to immobilize the
An ideal strategy to combat infections due to bio- peptide (Silva et al. 2016), it is of critical impor-
films involves prevention of microbial growth at tance that the compound retains its activity after
the site of initial colonization. This implies the attachment (Costa et al. 2011). For this purpose,
development of surfaces resistant to biofilm for- different parameters should be taken into consid-
mation or able to kill microbes on contact. A sur- eration including length, flexibility, and the type
face of this type would be particularly important of spacer connecting the peptide to the surface. In
and necessary for medical devices that are addition, the orientation of the immobilized pep-
implanted or inserted into patients (i.e., intrave- tide and the AMP surface density can affect the
nous catheters, shunts, prostheses) (Rabin et al. successful tethering of the molecule to the bio-
2008). Despite a very intensive research effort in material (Li et al. 2015a).
this field, so far, a truly efficient biofilm-resistant A wide variety of AMPs, like GZ3.27 (De
surface has not been developed yet. Zoysa and Sarojini 2017), GL13K (Chen et al.
In 1987, Gristina proposed the concept of “race 2014), SESB2V (Tan et al. 2012), bacitracin (Nie
for the surface” suggesting that host cells and et al. 2017), hLF1-11 (Godoy-Gallardo et al.
microbial cells compete for a spot on the implant 2014), chimeric peptides (Yazici et al. 2016),
surface. If microorganisms win this race, initial LL-37 (Gabriel et al. 2006), melamine (Willcox
colonization will result in a biofilm-­ associated et al. 2008), lactoferricin (Yoshinari et al. 2010),
infection (Gristina 1987). This concept also high- and Mel-4 (Dutta et al. 2016), have been cova-
lights the risk of microbial colonization of the tis- lently coupled onto several surfaces, such as
sue surrounding the implant which constitutes glass, silicon, and titanium using various immo-
another mechanism of infection (Riool et al. bilization strategies. Cleophas and collaborators
2017). For the prevention of implant-associated designed a contact-killing hydrogel containing a
infections, several strategies have been explored covalently attached inverso-CysHHC10 peptide
including the development of surfaces: (i) endowed that showed a good in vitro antimicrobial activity
with physicochemical modifications aimed at pre- against S. aureus, S. epidermidis, and E. coli
venting biofilm adhesion or growth (Bryers and (Cleophas et al. 2014). Other authors developed a
Ratner 2004), (ii) that incorporate immobilized brush coating polymer conjugated with the E6
antimicrobials, and (iii) which release antimicro- peptide that reduced bacterial adhesion to cathe-
bials to the surrounding area. ters in a mouse model of urinary catheter infec-
264 H. Shahrour et al.

tion (Yu et al. 2017b). Hoyos-Nogués and broad-spectrum antimicrobials unrelated to con-
collaborators utilized a promising strategy based ventional antibiotics (Alt et al. 2011), and AMPs
on combining the RGD cell-adhesive sequence can be ideal drugs for this purpose.
with the lactoferrin-derived AMP LF1-11 and Applications of AMPs in different types of
developed a multifunctional coating that inhib- release coatings have been described, including
ited bacterial colonization by S. aureus and hydrogels, nanotubes, microporous calcium
Streptococcus sanguinis (Hoyos-Nogués et al. phosphate coatings, and polymer coatings. For
2017). Similar antimicrobial efficacy was instance, Cheng and collaborators developed a
obtained when the unrelated peptide CWR11 was gelatin-based hydrogel on titanium surfaces lead-
immobilized on commercial catheters (Lim et al. ing to a controlled release of the AMP HHC36
2013, 2015). These devices are normally made of that prevented biofilm formation by S. aureus, S.
a widely used silicone-like material called epidermidis, E. coli, and P. aeruginosa (Cheng
polydimethylsiloxane (PDMS). et al. 2017). Similarly, Kazemzadeh-Narbat and
It should be noted that biomaterials containing collaborators reported high in vitro bactericidal
immobilized peptides do not exert anti-biofilm activity of Tet213 loaded microporous calcium
preventive activity in the surrounding tissue but phosphate coatings applied on titanium against S.
only kill microbes that are in direct contact with aureus and P. aeruginosa (Kazemzadeh-Narbat
their surface. In addition, efficiency of these coat- et al. 2010). Using a combination approach,
ings is greatly compromised after insertion into Forbes and collaborators studied the anti-biofilm
the patient, due to the formation of a layer of host potential of the human apolipoprotein E peptide
proteins covering the implanted device (Sánchez-­ (apoEdp) and its tryptophan-rich analog
Gómez and Martínez-de-Tejada 2017). The only (apoEdpL-W) along with other antimicrobials
strategy preventing both the colonization of the when incorporated onto hydrogels, polyethylene
implant and that of the surrounding tissue involves glycol, and nonporous polymers [polyurethane
the use of drug release systems. The advantage of (PU) and PDMS]. It was observed that the coated
this approach is that it allows achieving high con- surfaces could entirely eradicate S. aureus and P.
centrations of the antimicrobial in the vicinity of aeruginosa planktonic and biofilm cells (Forbes
the implant. In turn, local delivery of the drug et al. 2013).
facilitates microbial killing while minimizing tox- Other authors preferred to encapsulate AMPs
icity concerns associated with systemic adminis- into drug delivery vehicles such as degradable
tration of the antimicrobial. hydrogels, nanoparticles, and liposomes that
In this respect, catheters coated with antibiotic-­ allow for a controlled and more progressive
releasing layers are already approved for medical release of the active principle. A study by
application although their use is still limited d’Angelo and collaborators evaluated the ability
(Singha et al. 2017). This is probably due not of engineered poly(lactide-co-glycolide) (PLGA)
only to their higher cost compared to regular nanoparticles (NPs) carrying the cationic peptide
catheters but also to the increased likelihood of polymyxin E (colistin) to eradicate preformed P.
the coated devices to promote antibiotic resis- aeruginosa biofilms (d’Angelo et al. 2015). The
tance. This is a realistic scenario in the case of authors showed that those NPs were able to
drug-releasing biomaterials, because of the anti- release 50% of the encapsulated colistin in 6 h
microbial gradient generated and the possibility and sustain its release during ~15 days.
of reaching subinhibitory levels at a certain dis- Interestingly, an inhalable solution containing
tance of the surface. In addition, the increasing those NPs proved to have a good efficiency
proliferation of multidrug-resistant strains com- against P. aeruginosa biofilms in the airways of
promises more and more the efficiency of these lung-infected patients. This strategy represents a
devices. All these facts underscore the impor- promising alternative for inhaled-based treatment
tance of developing release systems based on (d’Angelo et al. 2015).
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 265

14.4 Combination Strategies tory concentrations. This leads to membrane

destabilization and disruption of the cell perme-
One attractive approach to strengthen the anti-­ ability barrier. Besides combination with antibi-
biofilm activity of AMPs is to combine these otics, AMPs have been tested in combination
agents with other drugs that target biofilms at a with other AMPs (see previous section; Forbes
different level. Some of the drugs that have been et al. 2013) and as hybrid chimeric peptides. An
used successfully for this purpose include con- example of a chimeric peptide is DD13-RIP
ventional antibiotics, compounds targeting the which is composed of dermaseptin derivative
extracellular matrix, inhibitors of QS, and/or (DD13) and an RNA III-inhibiting peptide (RIP)
other signaling pathways implicated in biofilm (Table 14.1). Whereas the dermaseptin moiety
formation or dispersal. In addition, AMPs have allows the peptide to target the bacterial cytoplas-
been administered concomitantly with other mic membranes, the RIP segment interrupts QS
AMPs (or other peptide-based molecules) mechanisms of staphylococcal cells. This hybrid
(Orlando et al. 2008) or even combined with compound has been reported to prevent biofilm
physical methods such as ultrasounds. formation by S. aureus and S. epidermidis in rat
Alternatively other researchers developed hybrid graft models (Balaban et al. 2004).
peptides by fusing two different compounds or An additional advantage of these types of ther-
by combining functional domains with distinct apies is that they allow reducing the total dose of
function in the same molecule (Xu et al. 2014). drug administered because concentrations of
Some of these strategies showed that it is pos- components when associated are lower than those
sible not only to complement the activities of the given in mono-dose. In turn, this favors a decrease
combination components but also to generate in toxicity and lowers the chances of inducing
synergistic effects between them (Grassi et al. resistance to treatments (Grassi et al. 2017a). In
2017a). For example, peptide IDR-1018 was Table 14.2, we show selected examples of combi-
capable of enhancing the activities of different nation strategies published to date.
classes of conventional antibiotics against
biofilm-­forming bacterial strains. Thus, at sub-­
MBIC (minimal biofilm inhibition concentra- 14.5 AMPs in Antibiotic Lock
tion) levels, IDR-1018 caused a 64-fold decrease Therapy
in the respective MBIC of several antibiotics.
Combination of this peptide with ceftazidime Antibiotic lock therapy (ALT) was developed in
completely eradicated mature biofilms formed by the late 1980s as an alternative to catheter removal
A. baumannii and was able to disrupt MRSA in patients with catheter-associated bacteremia in
mature biofilms leading to cell death at low con- which withdrawal represented a serious threat
centrations. Furthermore, IDR-1018 in combina- (Messing et al. 1988). The technique consists of
tion with tobramycin led to killing of filling the lumen of the contaminated catheter
biofilm-forming cells of both K. pneumoniae and with a solution containing antimicrobial agents at
E. coli O157 (Reffuveille et al. 2014). Similar very high concentrations. Generally, antimicrobi-
results were obtained with the D-enantiomeric als are used at concentrations ranging from 100
peptides DJK-5 and DJK-6 which exhibited syn- to 1000 times their planktonic MICs in the pres-
ergistic interactions with different classes of anti- ence of an anticoagulant such as heparin (Justo
biotics to prevent formation or promote and Bookstaver 2014). The solution remains
eradication of P. aeruginosa, E. coli, A. bauman- inside the catheter for the necessary time to
nii, K. pneumonia, and S. enterica biofilms (De la achieve catheter sterilization, and this often
Fuente-Núñez et al. 2015). requires adding freshly prepared solution to the
The molecular basis of peptide-based potenti- catheter every 24 h for several days.
ation appears to depend on the peptide ability to ALT is frequently applied to catheters
bind to microbial membranes even at subinhibi- implanted for long-term treatments (e.g., central
266 H. Shahrour et al.

Table 14.2 Selected examples of combination strategies involving AMPs

Combined
with AMP Combined compound Bacterial specie(s) Reference
Antibiotics 17BIPHE2 and Colistin, doripenem, Pseudomonas aeruginosa Mishra and
DASamP2 tobramycin, tigecycline Wang (2017)
AMP38 Imipenem Pseudomonas aeruginosa Rudilla et al.
(2016)
BMAP-28 Vancomycin Enterococcus faecalis, Orlando et al.
Staphylococcus aureus, (2008)
Salmonella enterica serovar
Typhimurium
Colistin Doripenem Pseudomonas aeruginosa Lora-Tamayo
et al. (2014)
Colistin Tobramycin Pseudomonas aeruginosa Herrmann et al.
(2010)
Colistin Ciprofloxacin, cotrimoxazole, Pseudomonas aeruginosa Hill et al.
ceftazidime, azithromycin (2005)
Coprisin Ampicillin, vancomycin, and Enterococcus faecium, Hwang et al.
chloramphenicol Staphylococcus aureus, (2013)
Escherichia coli,
Pseudomonas aeruginosa
and Streptococcus mutans
CRAMP Vancomycin Salmonella enterica serovar Mishra et al.
Typhimurium (2015)
DJK-5 and Ciprofloxacin, ceftazidime, Acinetobacter baumannii, De La
DJK-6 tobramycin Klebsiella pneumoniae, Fuente-Núñez
Escherichia coli, et al. (2015)
Pseudomonas aeruginosa
DJK-6 Imipenem, meropenem Klebsiella pneumoniae Carmona-­
Ribeiro (2000)
FLIP7 Meropenem, amikacin, Staphylococcus aureus, Chernysh et al.
kanamycin, ampicillin, Escherichia coli, (2018)
vancomycin, cefotaxime, Pseudomonas aeruginosa,
clindamycin, erythromycin, Klebsiella pneumoniae, and
chloramphenicol, oxacillin, Acinetobacter baumannii
tetracycline, ciprofloxacin,
gentamicin, polymyxin B
FK-13-a1 and Chloramphenicol Pseudomonas aeruginosa Rajasekaran
FK-13-a7 et al. (2017)
G10KHc Tobramycin Pseudomonas aeruginosa Eckert et al.
(2006)
GL13NH2 Tobramycin Pseudomonas aeruginosa Hirt and Gorr
(2013)
HPMA Ciprofloxacin Acinetobacter baumannii Gopal et al.
(2014)
IDR-1018 Ciprofloxacin Pseudomonas aeruginosa Reffuveille
et al. (2014)
IDR-1018 Ciprofloxacin, tobramycin, Pseudomonas aeruginosa, Mansour et al.
ceftazidime Escherichia coli, Klebsiella (2015)
pneumoniae,
Staphylococcus aureus,
Salmonella enterica,
Acinetobacter baumannii
Lactoferrin Ciprofloxacin Porphyromonas gingivalis Wakabayashi
et al. (2009)
(continued)
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 267

Table 14.2 (continued)

Combined
with AMP Combined compound Bacterial specie(s) Reference
Nisin Penicillin Enterococcus faecalis Tong et al.
(2014a)
Nisin V or Penicillin, chloramphenicol Staphylococcus aureus and Field et al.
Nisin I4V Staphylococcus (2016a)
pseudintermedius
PMBN Vancomycin Pseudomonas aeruginosa, Ferrer-Espada
Escherichia coli, Klebsiella and
pneumonia collaborators
manuscript in
preparation
r(P)ApoBL and Antibiotics, EDTA Staphylococcus aureus, Gaglione et al.
r(P)ApoBS Escherichia coli, (2017)
Pseudomonas aeruginosa
RNAIII-­ Tigecycline Staphylococcus aureus Simonetti et al.
inhibiting (2016)
peptide
UP-5 Levofloxacin, Staphylococcus aureus, Almaaytah
chloramphenicol, rifampicin, Enterococcus faecalis, et al. (2018)
ampicillin, and erythromycin Escherichia coli, and
Pseudomonas aeruginosa
WLBU2 Tobramycin, ciprofloxacin, Pseudomonas aeruginosa Lashua et al.
ceftazidime, and meropenem (2016)
Antifungals DS6 Amphotericin B, fluconazole Candida tropicalis Singh et al.
(2017)
HsLin06_18 Caspofungin Candida albicans Cools et al.
(2017)
β-peptides Fluconazole or ketoconazole Candida albicans Mora-Navarro
et al. (2015)
Chelators Nisin DHBA Staphylococcus aureus Ahire and
Dicks (2014)
Temporin 1 Tb EDTA Staphylococcus epidermidis Maisetta et al.
(2016)
TB_KKG6A EDTA Pseudomonas aeruginosa Grassi et al.
and TB-L1FK (2017b)
Lin-SB056–1 EDTA Pseudomonas aeruginosa Maisetta et al.
(2017)
AMP Citropil 1.1, Colistin Pseudomonas aeruginosa, Jorge et al.
temporin A, Staphylococcus aureus (2017)
analog of
tachyplesin I
DD13 RNA-inhibiting peptide Staphylococcus aureus, Balaban et al.
Staphylococcus epidermidis (2004)
Gramicidin S PMB Pseudomonas aeruginosa Berditsch et al.
(2015)
Nisin Colistin, PMB Pseudomonas aeruginosa Field et al.
(2016b)
Nisin MTAD Enterococcus faecalis Tong et al.
(2013)
PPMO PMBN Pseudomonas aeruginosa Howard et al.
(2017)
RNAIII-­ Bismuth ethanedithiol Staphylococci Domenico et al.
inhibiting (BisEDT), rifampin (2004)
peptide (RIP)
(continued)
268 H. Shahrour et al.

Table 14.2 (continued)

Combined
with AMP Combined compound Bacterial specie(s) Reference
Disinfectant β-defensin-3 Calcium hydroxide (CH) and Streptococcus mutans Ahn et al.
(HBD3) chlorhexidine digluconate (2017)
(CHX)
Peptide 1018 Chlorhexidine Oral multispecies Wang et al.
(2015)
DJK-5 Chlorhexidine Oral multispecies, Zhang et al.
Streptococcus mutans, and (2016)
Enterococcus faecalis
Amino acids Nisin D-cysteine (Cys), D- or Streptococcus mutans Tong et al.
L-aspartic acid (Asp), and (2014b)
D- or L-glutamic acid (Glu)
Platelet-derived Two arginine-tryptophan Staphylococcus epidermidis Alabdullatif
peptide (PD4) repeats (RW3 and RW4) et al. (2018)
Temporin 1 Tb L-cysteine Staphylococcus epidermidis Maisetta et al.
(2016)
Enzymes Human DN-ase I Haemophilus influenzae Jones et al.
β-defensin-3 (2013)
TN-5 Alginate lyase Pseudomonas aeruginosa Bahar et al.
(2015)
KSL-W Dispersin B Acinetobacter baumannii, Gawande et al.
Klebsiella pneumoniae, (2014)
Staphylococcus aureus,
Staphylococcus epidermidis
Gas Human Nitric oxide Pseudomonas aeruginosa Ren et al.
β-defensin-2 (2016)
Ultrasound Human Low-frequency ultrasound-­ Staphylococcus Li et al.
β-defensin-3 targeted microbubble (2015b)
(HBD-3) destruction (UTMD)
Fatty acids PMB and Polyunsaturated fatty acids Klebsiella pneumoniae Hobby et al.
colistin (2018)
Triple Colistin Vancomycin and low-­ Acinetobacter baumannii Liu et al.
combinations frequency ultrasound (2016)
CSA-13 and Core-shell magnetic Pseudomonas aeruginosa Niemirowicz
CSA-131 nanoparticles (MNPs), and Staphylococcus aureus et al. (2016)
vancomycin and colistin
PMBN Efflux pump inhibitors, Pseudomonas aeruginosa Ferrer-
ceftazidime, aztreonam, Espada and
doxycycline, azithromycin, collaborators
levofloxacin, piperacillin unpublished
data
PMBN β-lactams and β-lactamase Pseudomonas aeruginosa Ferrer-Espada
inhibitors and
collaborators
unpublished
data
Tachyplesin III Piperacillin/tazobactam Pseudomonas aeruginosa Minardi et al.
(2007)
Bacitracin Octyl gallate Staphylococcus aureus Oh et al. (2018)
GT peptide 10 Triethyl citrate (TEC) Propionibacterium acnes Eilers and
Alexeyev
(2016)
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 269

venous catheters), because they are more likely to 14.6 Challenges to the Use
get contaminated. Central venous catheters are of AMPs
used for hemodialysis, parenteral nutrition and
for the administration of chemotherapy or other As mentioned above, AMPs possess many of the
treatments. It has been estimated that patients characteristics of an ideal antimicrobial including
may suffer an episode of bacteremia per a broad spectrum of bactericidal activity, low ten-
1000 days of catheter use (Dudeck et al. 2013). dency to promote antimicrobial resistance and
These episodes are the cause of early withdrawal ability to kill even metabolically inactive micro-
of these devices in 25% of patients. However, organisms (Batoni et al. 2011). Despite all these
catheter salvage is indicated in cases of reduced attractive characteristics and their potent anti-­
venous access or coagulation disorders (Lebeaux biofilm activity, AMPs have also drawbacks that
et al. 2014). constraint their use as drugs. On the one hand,
ALT has several advantages, including (i) the many of these compounds show reduced activity
possibility of avoiding the removal of the device, in the presence of physiological concentrations
thus deriving benefits both for the patient and for of salts or other biological fluids. In addition,
the economy, (ii) the reduction of the adverse AMPs are often very sensitive to degradation by
effects associated with a systemic antimicrobial serum proteases, and this greatly diminishes their
treatment and the possibility of administering a half-life in vivo. Finally, due to their unspecific
drug at levels that are not viable through other mechanism of action, these compounds fre-
routes, and (iii) the fact that this therapy can be quently have a low therapeutic index and exhibit
carried out on an outpatient basis. On the other certain level of cytotoxicity near their therapeutic
hand, ALT may have some disadvantages such as concentration.
(i) the possibility of promoting catheter blockade, Regardless of their clinical performance, these
especially if the ALT solution does not incorpo- molecules are still expensive to produce at indus-
rate an anticoagulant, (ii) the risk of exposing the trial scale (see below). In addition, some AMPs
patient to high concentrations of antibiotics and display poor physical-chemical properties (e.g.,
anticoagulants if the procedure is not properly tendency to aggregate or self-associate) that
performed, and (iii) the possible delay in achiev- require further improvement (Shire et al. 2004).
ing infection clearance because of opting for a All these limitations probably explain why many
non-systemic treatment along with the risk of promising AMPs have not passed clinical trials.
more severe complications if ALT fails (Justo and In other instances, use of these compounds has
Bookstaver 2014; Fernández-Hidalgo and been restricted to topical application (Heinbockel
Almirante 2014). For these reasons, it is recom- et al. 2014). However, in the last years, several
mended to supplement the ALT treatment with a strategies have been devised to overcome these
systemic antibiotic therapy (Fernández-Hidalgo downsides, and, as a consequence, the list of
and Almirante 2014). AMPs reaching clinical trials is growing very
The use of AMPs in ALT is being increasingly significantly.
favored not only for the broad spectrum of micro- Concerning safety, several studies have
bicidal activity of these compounds but also for focused their efforts in determining the m­ olecular
the biofilm disaggregating efficiency displayed level and the structural features that govern the
by some of them (Di Luca et al. 2014). interaction of AMPs with prokaryotic and eukary-
Currently, there are several experimental lock otic membranes (Gupta et al. 2013). The modifi-
solutions based on AMPs under study that include cations of some physicochemical properties such
(i) the administration of a single compound, (ii) as their hydrophobicity, amphipathicity, and
therapies combining peptides and chelators that helicity successfully enhanced their selectivity
destabilize the biofilm matrix, or (iii) synergistic (Chen et al. 2005).
combinations of AMPs with different antimicro- Other chemical modifications applied to
bials (Table 14.3). AMPs to extend their half-life once administered
Table 14.3 Experimental lock solutions based on antimicrobial peptides
270

Age of the Type of

AMP Bacterial strain biofilm Time of exposure Concentration Result investigation Reference
Temporin S. epidermidis 24 h in 96-well 24 h 50 µg/mL Prophylaxis 3 log red In vitro Maisetta et al.
1 Tb microplate 100 µg/mL Treatment 0.5–2 log red bioactivity (2016)
100 µg/mL + EDTA Treatment 3 log red study
4 mg/mL
CST P. aeruginosa 24 h in 1-cm 24, 48, 72 and 400 or 200 µg/mL 4 log red In vitro Ozbek and
segments of 96 h 200 mg/ml CLR 6 log red bioactivity Mataraci-Kara
7-French, 1000 mg/mL HEP 3.5 log red study (2016)
triple-lumen,
CVCs
CST A. baumannii 24 h in 1-cm 24, 48, 72 and 800 µg/mL 5–6 log red In vitro Ozbek and
segments of 96 h 800 µg/mL + 200 mg/ 5.5–6 log red bioactivity Mataraci
7-French, ml CLR study (2013)
triple-lumen, 800 µg/mL + 3–6 log red
CVCs 1000 mg/mL HEP
800 µg/mL + 200 mg/ 5–6 log red
ml CLR + 1000 mg/
mL HEP
CST K. pneumoniae 48 h in 1-cm 2, 4, 6 and 24 h 50 or 100 µg/mL 0.5–5 log red In vitro Mataraci Kara
segments of 50 or 100 µg/mL + 1–4 log red bioactivity and Ozbek
7-French, 200 mg/ml CLR study Celik (2018)
triple-lumen, 50 or 100 µg/mL + 2–3.5 log red
CVCs 0.25 mM ESO
Tachyplesin P. aeruginosa Coated ureteral 120 h 10 µg/mL tachyplesin Prophylaxis 3 log red In vivo study Minardi et al.
III stent III-coat (2007)
120 mg/kg Prophylaxis 3 log red
intraperitoneal TZP
120 mg/kg Prophylaxis 5 log red
intraperitoneal TZP +
10 µg/mL tachyplesin
III-coat
H. Shahrour et al.
 Age of the Type of
AMP Bacterial strain biofilm Time of exposure Concentration Result investigation Reference
14

RIP S. aureus Rat model of 30 min RIP RIP 1 mg/mL 3 log red In vivo study Cirioni et al.
central venous prophylaxis RIP 1 mg/mL + 3 log red (2006)
catheter +1 h Ab 16 µg/mL
treatment vancomycin
RIP 1 mg/mL + 5 log red
1024 µg/mL
vancomycin
RIP 1 mg/mL + 3 log red
16 µg/mL
ciprofloxacin
RIP 1 mg/mL + 5 log red
1024 µg/mL
ciprofloxacin
RIP 1 mg/mL + 8 µg/ 3 log red
mL imipenem
RIP 1 mg/mL + 5 log red
1024 µg/mL
imipenem
TLF S. epidermidis and 24 h in 96-well 0, 24, 48, and TLF 8 mg/mL 1 log red S. epidermidis In vitro Venkatesh
C. albicans microplate 72 h 0.5 log red C. albicans bioactivity et al. (2009)
polymicrobial study
biofilm
CST 0.1 mg/mL, Compatibility and stability confirmed: Stability Vincentelli
AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis

vancomycin 0.1 mg/ 10% degradation at 60 days at 4 °C study et al. (1997)

mL + heparin 100 and 25 °C; vancomycin had ∼25%
units/mL decrease in concentrations after day
15 at room temperature
CST colistin, CLR clarithromycin, HEP heparin, ESO esomeprazole, TZP piperacillin-tazobactam, RIP RNAIII-inhibiting peptide, TLF talactoferrin, CVC Central venous
catheter
271
272 H. Shahrour et al.

include strategies to avoid their proteolytic deg- trum of activity including biofilm-forming cells
radation such as the substitution of standard and persisters. Finally, AMPs display multiple
amino acids with nonnatural amino acids (e.g., mechanisms of action which enable them to
D-amino acids) (Miller et al. 1994; Strömstedt interfere with various stages of biofilm forma-
et al. 2009), peptide cyclization or terminal modi- tion. Importantly, some of these mechanisms are
fication via amidation, alkylation or acetylation independent of the peptide antimicrobial activity
(Strömstedt et al. 2009; Nguyen et al. 2010), pep- (e.g., inhibition of extracellular matrix biosyn-
tide mimetics, (Fjell et al. 2011), or end-tagging thesis, blocking of QS cell communication, or
by hydrophobic oligo amino acid stretches downregulation of genes involved in biofilm
(Malmsten et al. 2011), among others. formation).
Alternatively, to increase their in vivo stability Although AMPs have some clear drawbacks,
and lower their toxicity, AMPs can be either many strategies are being actively pursued to
immobilized on solid surfaces to prevent biofilm speed up the incorporation of these molecules
colonization or encapsulated into natural or syn- into the clinical practice. Simultaneously, other
thetic polymeric carriers and delivery systems approaches are being explored, including the use
(Kazemzadeh-Narbat et al. 2010; Mokkaphan AMPs as enhancing agents (Grassi et al. 2017a).
et al. 2014) (see Sect. 14.3 of this chapter). Thus, co-administration of AMPs with antibiot-
The large-scale production of peptides is still ics and anti-biofilm compounds offer an attrac-
a challenge that needs to be faced. AMP manu- tive strategy that allows killing a broad spectrum
facturing is difficult and expensive mainly of microbial species at low antibiotic
because it requires complex extraction and purifi- concentrations.
cations steps (Beckloff et al. 2007). Indeed, the Medical device-associated infections are of
production cost of a 5,000 Da molecular mass great concern for public health, being one of the
peptide is higher than that of a typical 500 Da causes of recurrent surgeries, prolonged adminis-
molecule (at a comparable scale of operation) tration of antibiotics, and eventually patient
(Bray 2003). One possible solution to address death. The use of AMPs to coat the surfaces of
this issue entails the production of AMPs in implanted devices appears as a promising way to
genetically manipulated microorganisms carry- address this problem. However, a truly efficient
ing expression vectors (Mygind et al. 2005; Cao biofilm-resistant surface has not been developed
et al. 2018). As the technology improves, the cost and more research is needed in this field.
of large-scale peptide manufacturing is expected Specifically, efforts should be directed to design
to decrease significantly in the next decade. more efficient systems of peptide immobilization
in biomaterials and to develop nano- and
micro-­
­ systems able to sustain the release of
14.7 Conclusions and Future AMPs at the infection sites.
Directions Finally, a new arising field of study is the
rational design of AMPs. This methodology is
Currently, one of the major challenges in medi- focused first on unraveling the complex interac-
cine is the treatment of biofilm-associated infec- tion between AMPs and microbial cells and then
tions. We cannot rely solely on antibiotics, since to use this knowledge to design optimized mole-
efficiency of these drugs is being increasingly cules. This is a complex field of study, since
compromised by the emergence of multiresistant many features of AMPs influence their activity
strains. In this context, AMPs appear as a very such as size, sequence, charge, helicity, hydro-
promising alternative, since these agents target phobicity, among others. New bioinformatic
highly conserved molecular motifs that are essen- tools are being used and developed to compile the
tial for the microbes and hence less likely to be enormous information generated and to extract
modified by them. In addition, AMPs are potent valid conclusions from these analyses. Current
bactericidal agents endowed with a broad spec- approaches aim at the screening and in silico
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 273

modeling of novel AMPs to accelerate the dis- Ammons MC, Copié V (2013) Mini-review: lactoferrin:
a bioinspired, anti-biofilm therapeutic. Biofouling
covery of new safe and effective compounds 29:443–455. https://doi.org/10.1080/08927014.2013.
(Brancatisano et al. 2014; Jorge et al. 2012; 773317
Sharma et al. 2016). Anunthawan T, De La Fuente-Núñez C, Hancock REW,
Biofilm-associated infections are a critical Klaynongsruang S (2015) Cationic amphipathic pep-
tides KT2 and RT2 are taken up into bacterial cells
health problem that needs to be addressed. and kill planktonic and biofilm bacteria. https://doi.
Although further research is needed, AMPs have org/10.1016/j.bbamem.2015.02.021
the potential to become efficient anti-biofilm Bahar AA, Liu Z, Garafalo M et al (2015) Controlling
agents alone or combined with other drugs. persister and biofilm cells of gram-negative
bacteria with a new 1,3,5-triazine deriva-
tive. Pharmaceuticals 8:696–710. https://doi.
Acknowledgments This work was supported by the org/10.3390/ph8040696
Proyectos de Investigación Universidad de Navarra, Spain Balaban N, Gov Y, Giacometti A et al (2004) A chimeric
(PIUNA-P2011-17 and P2015-14 to G. M. T.). R. F. E and peptide composed of a dermaseptin derivative and an
S.B.V. are recipients of doctoral fellowships granted by RNA III-inhibiting peptide prevents graft-­associated
Gobierno Vasco, Spain (BFI-2011-9) and Asociación de infections by antibiotic-resistant staphylococci.
Amigos de la Universidad de Navarra (Spain), Antimicrob Agents Chemother 48:2544–2550. https://
respectively. doi.org/10.1128/AAC.48.7.2544–2550.2004
H.S. is grateful to financial support received from the Banerjee I, Pangule RC, Kane RS (2011) Antifouling
Lebanese National Council for Scientific Research coatings: recent developments in the design of sur-
(CNRS) and by the Lebanese University. faces that prevent fouling by proteins, bacteria, and
marine organisms. Adv Mater 23:690–718. https://doi.
org/10.1002/adma.201001215
Batoni G, Maisetta G, Lisa Brancatisano F et al (2011)
References Use of antimicrobial peptides against microbial bio-
films: advantages and limits. Curr Med Chem 18:256–
Ahire JJ, Dicks LMT (2014) Nisin incorporated with 279. https://doi.org/10.2174/092986711794088399
2,3-Dihydroxybenzoic acid in nanofibers inhibits Beckloff N, Laube D, Castro T et al (2007) Activity of an
biofilm formation by a methicillin-resistant strain of antimicrobial peptide mimetic against planktonic and
Staphylococcus aureus. Probiotics Antimicrob Proteins biofilm cultures of oral pathogens. Antimicrob Agents
7:52–59. https://doi.org/10.1007/s12602-014-9171-5 Chemother 51:4125–4132. https://doi.org/10.1128/
Ahn KB, Kim AR, Kum KY et al (2017) The synthetic AAC.00208-07
human beta-defensin-3 C15 peptide exhibits antimi- Belley A, Neesham-Grenon E, McKay G et al (2009)
crobial activity against Streptococcus mutans, both Oritavancin kills stationary-phase and biofilm
alone and in combination with dental disinfectants. Staphylococcus aureus cells in vitro. Antimicrob
J Microbiol 55:830–836. https://doi.org/10.1007/ Agents Chemother 53:918–925. https://doi.
s12275-017-7362-y org/10.1128/AAC.00766-08
Akyıldız I, Take G, Uygur K et al (2013) Bacterial bio- Berditsch M, Jäger T, Strempel N et al (2015) Synergistic
film formation in the middle-ear mucosa of chronic effect of membrane-active peptides polymyxin B
otitis media patients. Indian J Otolaryngol Head and gramicidin s on multidrug-resistant strains and
Neck Surg 65:557–561. https://doi.org/10.1007/ biofilms of Pseudomonas aeruginosa. Antimicrob
­
s12070-012-0513-x Agents Chemother 59:5288–5296. https://doi.
Alabdullatif M, Atreya CD, Ramirez-Arcos S (2018) org/10.1128/AAC.00682-15
Antimicrobial peptides: an effective approach to pre- Bispo P, Haas W, Gilmore M (2015) Biofilms in infec-
vent bacterial biofilm formation in platelet concen- tions of the eye. Pathogens 4:111–136. https://doi.
trates. Transfusion 00:1–9. https://doi.org/10.1111/ org/10.3390/pathogens4010111
trf.14646 Blower RJ, Barksdale SM, van Hoek ML (2015) Snake
Almaaytah A, Qaoud MT, Mohammed GK et al (2018) cathelicidin NA-CATH and smaller helical antimicro-
Antimicrobial and antibiofilm activity of UP-5, an bial peptides are effective against Burkholderia thai-
ultrashort antimicrobial peptide designed using only landensis. PLoS Negl Trop Dis 9:e0003862. https://
arginine and biphenylalanine. Pharmaceuticals 11:1– doi.org/10.1371/journal.pntd.0003862
18. https://doi.org/10.3390/ph11010003 Brancatisano FL, Maisetta G, Di Luca M et al (2014)
Alt V, Bitschnau A, Böhner F et al (2011) Effects of Inhibitory effect of the human liver-derived antimicro-
gentamicin and gentamicin–RGD coatings on bone bial peptide hepcidin 20 on biofilms of polysaccharide
ingrowth and biocompatibility of cementless joint intercellular adhesin (PIA)-positive and PIA-negative
prostheses: an experimental study in rabbits. Acta strains of Staphylococcus epidermidis. Biofouling
Biomater 7:1274–1280. https://doi.org/10.1016/J. 30:435–446. https://doi.org/10.1080/08927014.2014.
ACTBIO.2010.11.012 888062
274 H. Shahrour et al.

Bray BL (2003) Large-scale manufacture of peptide ther- films. Front Microbiol 8:1–14. https://doi.org/10.3389/
apeutics by chemical synthesis. Nat Rev Drug Discov fmicb.2017.02051
2:587–593. https://doi.org/10.1038/nrd1133 Costa F, Carvalho IF, Montelaro RC et al (2011) Covalent
Bryers JD, Ratner BD (2004) Bioinspired implant immobilization of antimicrobial peptides (AMPs) onto
materials befuddle bacteria. ASM News 70:232– biomaterial surfaces. Acta Biomater 7:1431–1440.
237. https://www.asm.org/index.php/fellows-info/ https://doi.org/10.1016/j.actbio.2010.11.005
fellows-elected-in-2016/5464-gerard-wong d’Angelo I, Casciaro B, Miro A et al (2015) Overcoming
Busscher HJ, van der Mei HC, Subbiahdoss G et al barriers in Pseudomonas aeruginosa lung infec-
(2012) Biomaterial-associated infection: locat- tions: engineered nanoparticles for local delivery
ing the finish line in the race for the surface. Sci of a cationic antimicrobial peptide. Colloids Surf B
Transl Med 4:153rv10. https://doi.org/10.1126/ Biointerfaces 135:717–725. https://doi.org/10.1016/j.
scitranslmed.3004528 colsurfb.2015.08.027
Cao J, De La Fuente-Nunez C, Ou RW et al (2018) Davey ME, O’Toole GA (2000) Microbial biofilms:
Yeast-based synthetic biology platform for antimicro- from ecology to molecular genetics. Microbiol
bial peptide production. ACS Synth Biol 7:896–902. Mol Biol Rev 64:847–867. https://doi.org/10.1128/
https://doi.org/10.1021/acssynbio.7b00396 MMBR.64.4.847-867.2000
Carmona-Ribeiro (2000) Interactions between cationic De la Fuente-Núñez C, Korolik V, Bains M et al (2012)
liposomes and drugs or biomolecules. An Acad Inhibition of bacterial biofilm formation and swarm-
Bras Cienc. 72(1):39–43. https://doi.org/10.1590/ ing motility by a small synthetic cationic peptide.
S0001-37652000000100005 Antimicrob Agents Chemother 56:2696–2704. https://
CDC (2013) Antibiotic resistance threats in the United doi.org/10.1128/AAC.00064-12
States, 2013. https://www.cdc.gov/drugresistance/pdf/ De la Fuente-Núñez C, Reffuveille F, Haney EF et al (2014)
ar-threats-2013-508.pdf Broad-spectrum anti-biofilm peptide that targets a
Chen Y, Mant CT, Farmer SW et al (2005) Rational design cellular stress response. PLoS Pathog 10:e1004152.
of alpha-helical antimicrobial peptides with enhanced https://doi.org/10.1371/journal.ppat.1004152
activities and specificity/therapeutic index. J Biol De La Fuente-Núñez C, Reffuveille F, Mansour SC et al
Chem 280:12316–12329. https://doi.org/10.1074/jbc. (2015) D-enantiomeric peptides that eradicate wild-­
M413406200 type and multidrug-resistant biofilms and protect
Chen X, Hirt H, Li Y et al (2014) Antimicrobial GL13K against lethal Pseudomonas aeruginosa infections.
peptide coatings killed and ruptured the wall of Chem Biol 22:196–205. https://doi.org/10.1016/j.
Streptococcus gordonii and prevented formation and chembiol.2015.01.002
growth of biofilms. PLoS One 9:e111579. https://doi. De Zoysa GH, Sarojini V (2017) Feasibility study explor-
org/10.1371/journal.pone.0111579 ing the potential of novel Battacin Lipopeptides
Cheng H, Yue K, Kazemzadeh-Narbat M et al (2017) as antimicrobial coatings. ACS Appl Mater
Mussel-inspired multifunctional hydrogel coating for Interfaces 9:1373–1383. https://doi.org/10.1021/
prevention of infections and enhanced osteogenesis. acsami.6b15859
ACS Appl Mater Interfaces 9:11428–11439. https:// Dean SN, Bishop BM, van Hoek ML (2011) Natural
doi.org/10.1021/acsami.6b16779 and synthetic cathelicidin peptides with anti-­
Chernysh S, Gordya N, Tulin D, Yakovlev A (2018) microbial and anti-biofilm activity against
Biofilm infections between Scylla and Charybdis: Staphylococcus aureus. BMC Microbiol 11. https://
interplay of host antimicrobial peptides and antibi- doi.org/10.1186/1471-2180-11-114
otics. Infect Drug Resist 11:501–514. https://doi. Dutta D, Kumar N, Willcox MDP (2016) Antimicrobial
org/10.2147/IDR.S157847 activity of four cationic peptides immobilised
Cirioni O, Giacometti A, Ghiselli R et al (2006) RNAIII-­ to poly-hydroxyethylmethacrylate. Biofouling
inhibiting peptide significantly reduces bacterial load 32(4):429–438
and enhances the effect of antibiotics in the treatment Di Luca M, Maccari G, Nifosì R (2014) Treatment of micro-
of central venous catheter-associated Staphylococcus bial biofilms in the post-antibiotic era: prophylactic
aureus infections. J Infect Dis 193:180–186. https:// and therapeutic use of antimicrobial peptides and their
doi.org/10.1086/498914 design by bioinformatics tools. Pathog Dis 70:257–
Cleophas RTC, Riool M, Quarles van Ufford H, Linda 270. https://doi.org/10.1111/2049-632X.12151
C et al (2014) Convenient preparation of bactericidal Di Luca M, Maccari G, Maisetta G, Batoni G (2015)
hydrogels by covalent attachment of stabilized anti- BaAMPs: the database of biofilm-active antimicrobial
microbial peptides using thiol–ene click chemistry. peptides. Biofouling 31:193–199. https://doi.org/10.1
ACS Macro Lett 3:477–480. https://doi.org/10.1021/ 080/08927014.2015.1021340
mz5001465 Domenico P, Gurzenda E, Giacometti A et al (2004)
Cools TL, Struyfs C, Drijfhout JW et al (2017) A linear BisEDT and RIP act in synergy to prevent graft infec-
19-mer plant defensin-derived peptide acts synergisti- tions by resistant staphylococci. Peptides 25:2047–
cally with caspofungin against Candida albicans bio- 2053. https://doi.org/10.1016/j.peptides.2004.08.005
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 275

Donlan RM (2002) Biofilms: microbial life on surfaces. Gawande PV, Leung KP, Madhyastha S (2014) Antibiofilm
Emerg Infect Dis 8:881–890. https://doi.org/10.3201/ and antimicrobial efficacy of Dispersinb®-KSL-w
eid0809.020063 peptide-based wound gel against chronic wound infec-
Dudeck MA, Weiner LM, Allen-Bridson K et al (2013) tion associated bacteria. Curr Microbiol 68:635–641.
National Healthcare Safety Network (NHSN) report, https://doi.org/10.1007/s00284-014-0519-6
data summary for 2012, device-associated mod- Gbejuade HO, Lovering AM, Webb JC (2014) The role of
ule. Am J Infect Control 41:1148–1166. https://doi. microbial biofilms in prosthetic joint infections. Acta
org/10.1016/j.ajic.2013.09.002 Orthop 86:1–12. https://doi.org/10.3109/17453674.20
ECDC (2009) The bacterial challenge: time to react. A call 14.966290
to narrow the gap between multidrug-resistant bacteria Giamarellou H (2002) Nosocomial cardiac infections.
in the EU and the development of new antibacterial J Hosp Infect 50:91–105. https://doi.org/10.1053/
agents. https://www.ema.europa.eu/documents/report/ jhin.2001.1144
bacterial-challenge-time-react_en.pdf Godoy-Gallardo M, Mas-Moruno C, Fernández-Calderón
Eckert R, Brady KM, Greenberg EP et al (2006) Enhan- MC et al (2014) Covalent immobilization of hLf1-11
cement of antimicrobial activity against Pseudomonas peptide on a titanium surface reduces bacterial adhe-
aeruginosa by coadministration of G10KHc and tobra- sion and biofilm formation. Acta Biomater 10:3522–
mycin. Antimicrob Agents Chemother 50:3833–3838. 3534. https://doi.org/10.1016/j.actbio.2014.03.026
https://doi.org/10.1128/AAC.00509-06 Gopal R, Kim YG, Lee JH et al (2014) Synergistic effects
Eilers H, Alexeyev OA (2016) Effect of GT-peptide 10 and antibiofilm properties of chimeric peptides against
and triethyl citrate on P. acnes biofilm formation, multidrug-resistant Acinetobacter baumannii strains.
viability, and dispersion. J Drugs Dermatol 15:778– Antimicrob Agents Chemother 58:1622–1629. https://
781. http://jddonline.com/articles/dermatology/ doi.org/10.1128/AAC.02473-13
S1545961616P0778X Grassi L, Maisetta G, Esin S, Batoni G (2017a)
Fernández-Hidalgo N, Almirante B (2014) Antibiotic-­ Combination strategies to enhance the efficacy of
lock therapy: a clinical viewpoint. Expert Rev Anti-­ antimicrobial peptides against bacterial biofilms.
Infect Ther 12:117–129. https://doi.org/10.1586/1478 Front Microbiol 8:2409. https://doi.org/10.3389/
7210.2014.863148 fmicb.2017.02409
Field D, O’Connor R, Cotter PD et al (2016a) In vitro Grassi L, Maisetta G, Maccari G et al (2017b) Analogs
activities of nisin and nisin derivatives alone and in of the frog-skin antimicrobial peptide temporin 1Tb
combination with antibiotics against staphylococ- exhibit a wider spectrum of activity and a stronger
cus biofilms. Front Microbiol 7:1–11. https://doi. antibiofilm potential as compared to the parental
org/10.3389/fmicb.2016.00508 peptide. Front Chem 5:1–13. https://doi.org/10.3389/
Field D, Seisling N, Cotter PD et al (2016b) Synergistic fchem.2017.00024
nisin-polymyxin combinations for the control of pseu- Gristina AG (1987) Biomaterial-centered infec-
domonas biofilm formation. Front Microbiol 7:1–7. tion: microbial adhesion versus tissue integration.
https://doi.org/10.3389/fmicb.2016.01713 Science 237:1588–1595. https://doi.org/10.1126/
Fjell CD, Hiss JA, Hancock REW, Schneider G (2011) science.3629258
Designing antimicrobial peptides: form follows func- Guggenbichler JP, Assadian O, Boeswald M, Kramer A
tion. Nat Rev Drug Discov 11:37–51. https://doi. (2011) Incidence and clinical implication of nosoco-
org/10.1038/nrd3591 mial infections associated with implantable bioma-
Flemming H-C, Wingender J (2010) The biofilm terials – catheters, ventilator-associated pneumonia,
matrix. Nat Rev Microbiol 8:623–633. https://doi. urinary tract infections. GMS Krankenhhyg Interdiszip
org/10.1038/nrmicro2415 6:Doc18. https://doi.org/10.3205/dgkh000175
Forbes S, McBain AJ, Felton-Smith S et al (2013) Gupta S, Kapoor P, Chaudhary K et al (2013) In silico
Comparative surface antimicrobial properties of approach for predicting toxicity of peptides and pro-
synthetic biocides and novel human apolipopro- teins. PLoS One 8:e73957. https://doi.org/10.1371/
tein E derived antimicrobial peptides. Biomaterials journal.pone.0073957
34:5453–5464. https://doi.org/10.1016/J. Hall CW, Mah T-F (2017) Molecular mechanisms of
BIOMATERIALS.2013.03.087 biofilm-­based antibiotic resistance and tolerance in
Gabriel M, Nazmi K, Veerman EC et al (2006) Preparation pathogenic bacteria. FEMS Microbiol Rev 41:276–
of LL-37-grafted titanium surfaces with bactericidal 301. https://doi.org/10.1093/femsre/fux010
activity. Bioconjug Chem 17:548–550. https://doi. Hancock REW (2015) Rethinking the antibiotic dis-
org/10.1021/bc050091v covery paradigm. EBIOM 2:629–630. https://doi.
Gaglione R, Dell’Olmo E, Bosso A et al (2017) Novel org/10.1016/j.ebiom.2015.07.010
human bioactive peptides identified in apolipoprotein Harms A, Maisonneuve E, Gerdes K (2016) Mechanisms
B: evaluation of their therapeutic potential. Biochem of bacterial persistence during stress and antibi-
Pharmacol 130:34–50. https://doi.org/10.1016/j. otic exposure. Science 354:aaf4268. https://doi.
bcp.2017.01.009 org/10.1126/science.aaf4268
276 H. Shahrour et al.

Heinbockel L, de Tejada G, Sánchez-Gómez S, et al Biofouling 28:1033–1061. https://doi.org/10.1080/08

(2014) Anti-infective polypeptides for combating 927014.2012.728210
bacterial and viral infections. In: Frontiers in clini- Jorge P, Grzywacz D, Kamysz W et al (2017) Searching
cal drug research: anti-infectives. Bentham Science for new strategies against biofilm infections: colistin-­
Publishers, p 3–31. https://doi.org/10.2174/9781608 AMP combinations against pseudomonas aeruginosa
058549114010003 and staphylococcus aureus single- and double-species
Herrmann G, Yang L, Wu H et al (2010) Colistin-­ biofilms. PLoS One 12:1–21. https://doi.org/10.1371/
tobramycin combinations are superior to monother- journal.pone.0174654
apy concerning the killing of biofilm Pseudomonas Justo JA, Bookstaver PB (2014) Antibiotic lock therapy:
aeruginosa. J Infect Dis 202:1585–1592. https://doi. review of technique and logistical challenges. Infect
org/10.1086/656788 Drug Resist 7:343–363. https://doi.org/10.2147/IDR.
Hill D, Hill D, Rose B et al (2005) Antibiotic susceptibil- S51388
ity of Pseudomonas aeruginosa isolates derived from Kanthawong S, Bolscher JGM, Veerman ECI et al (2012)
patients with cystic fibrosis under aerobic, anaerobic, Antimicrobial and antibiofilm activity of LL-37 and
and biofilm conditions. J Clin Microbiol 43:5085– its truncated variants against Burkholderia pseudom-
5090. https://doi.org/10.1128/JCM.43.10.5085 allei. Int J Antimicrob Agents 39:39–44. https://doi.
Hirt H, Gorr SU (2013) Antimicrobial peptide GL13K is org/10.1016/j.ijantimicag.2011.09.010
effective in reducing biofilms of Pseudomonas aeru- Karygianni L, Al-Ahmad A, Argyropoulou A et al (2016)
ginosa. Antimicrob Agents Chemother 57:4903–4910. Natural antimicrobials and oral microorganisms: a
https://doi.org/10.1128/AAC.00311-13 systematic review on herbal interventions for the erad-
Hobby CR, Herndon JL, Morrow CA et al (2018) ication of multispecies oral biofilms. Front Microbiol
Exogenous fatty acids alter phospholipid composi- 6:1–17. https://doi.org/10.3389/fmicb.2015.01529
tion, membrane permeability, capacity for biofilm Kazemzadeh-Narbat M, Kindrachuk J, Duan K et al (2010)
formation, and antimicrobial peptide susceptibility Antimicrobial peptides on calcium phosphate-­coated
in Klebsiella pneumoniae. Microbiology 635:1–11. titanium for the prevention of implant-­ associated
https://doi.org/10.1002/mbo3.635 infections. Biomaterials 31:9519–9526. https://doi.
Høiby N, Ciofu O, Bjarnsholt T (2010) Pseudomonas org/10.1016/j.biomaterials.2010.08.035
aeruginosa biofilms in cystic fibrosis. Future Microbiol Kostakioti M, Hadjifrangiskou M, Hultgren SJ (2013)
5:1663–1674. https://doi.org/10.2217/fmb.10.125 Bacterial biofilms: development, dispersal, and thera-
Howard JJ, Sturge CR, Moustafa DA et al (2017) peutic strategies in the dawn of the postantibiotic era.
Inhibition of Pseudomonas aeruginosa by peptide-­ Cold Spring Harb Perspect Med 3:1–23. https://doi.
conjugated phosphorodiamidate morpholino org/10.1101/cshperspect.a010306
oligomers. Antimicrob Agents Chemother 61:e01938– Lai Y, Gallo RL (2009) AMPed up immunity: how anti-
e01916. https://doi.org/10.1128/AAC.01938-16 microbial peptides have multiple roles in immune
Hoyos-Nogués M, Velasco F, Ginebra M-P et al (2017) defense. Trends Immunol 30:131–141. https://doi.
Regenerating bone via multifunctional coatings: org/10.1016/j.it.2008.12.003
the blending of cell integration and bacterial inhibi- Landini P, Antoniani D, Burgess JG, Nijland R (2010)
tion properties on the surface of biomaterials. ACS Molecular mechanisms of compounds affecting bacte-
Appl Mater Interfaces 9:21618–21630. https://doi. rial biofilm formation and dispersal. Appl Microbiol
org/10.1021/acsami.7b03127 Biotechnol 86:813–823. https://doi.org/10.1007/
Hwang IS, Hwang JS, Hwang JH et al (2013) Synergistic s00253-010-2468-8
effect and antibiofilm activity between the antimi- Lashua LP, Melvin JA, Deslouches B et al (2016)
crobial peptide coprisin and conventional antibiotics Engineered cationic antimicrobial peptide (eCAP)
against opportunistic bacteria. Curr Microbiol 66:56– prevents Pseudomonas aeruginosa biofilm growth
60. https://doi.org/10.1007/s00284-012-0239-8 on airway epithelial cells. J Antimicrob Chemother
Jamal M, Ahmad W, Andleeb S et al (2018) Bacterial 71:2200–2207. https://doi.org/10.1093/jac/dkw143
biofilm and associated infections. J Chin Med Assoc Lebeaux D, Ghigo J-M, Beloin C (2014) Biofilm-related
81:7–11. https://doi.org/10.1016/j.jcma.2017.07.012 infections: bridging the gap between clinical manage-
Jefferson KK (2004) What drives bacteria to produce a ment and fundamental aspects of recalcitrance toward
biofilm? FEMS Microbiol Lett 236:163–173. https:// antibiotics. Microbiol Mol Biol Rev 78:510–543.
doi.org/10.1016/j.femsle.2004.06.005 https://doi.org/10.1128/MMBR.00013-14
Jones EA, McGillivary G, Bakaletz LO (2013) Lehner A, Riedel K, Eberl L et al (2005) Biofilm for-
Extracellular DNA within a nontypeable haemophilus mation, extracellular polysaccharide production,
influenzae-induced biofilm binds human beta defen- and cell-to-cell signaling in various Enterobacter
sin-­3 and reduces its antimicrobial activity. J Innate sakazakii strains: aspects promoting environmen-
Immun 5:24–38. https://doi.org/10.1159/000339961 tal persistence. J Food Prot 68:2287–2294. https://
Jorge P, Lourenço A, Pereira MO (2012) New trends in www.researchgate.net/publication/7468730_
peptide-based anti-biofilm strategies: a review of Biofilm_Formation_Extracellular_Polysaccharide_
recent achievements and bioinformatic approaches. Production_and_Cell-­to-­Cell_Signaling_in_Various_
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 277

Enterobacter_sakazakii_Strains_Aspects_Promoting_ derived from PRELP. PLoS One 6:e16400. https://doi.

Environmental_Persistence org/10.1371/journal.pone.0016400
Li Y, Wei S, Wu J et al (2015a) Effects of peptide immo- Mansour SC, De La Fuente-Núñez C, Hancock REW
bilization sites on the structure and activity of surface-­ (2015) Peptide IDR-1018: modulating the immune
tethered antimicrobial peptides. J Phys Chem C system and targeting bacterial biofilms to treat
119:7146–7155. https://doi.org/10.1021/jp5125487 antibiotic-­resistant bacterial infections. J Pept Sci
Li S, Zhu C, Fang S et al (2015b) Ultrasound microbub- 21:323–329. https://doi.org/10.1002/psc.2708
bles enhance human β-defensin 3 against biofilms. Mataraci Kara E, Ozbek Celik B (2018) Investigation of
J Surg Res 199:458–469. https://doi.org/10.1016/j. the effects of various antibiotics against Klebsiella
jss.2015.05.030 pneumoniae biofilms on in vitro catheter model.
Lim K, Chua RRY, Saravanan R et al (2013) Immobilization J Chemother 30:82–88. https://doi.org/10.1080/1120
studies of an engineered arginine–tryptophan-rich 009X.2017.1390633
peptide on a silicone surface with antimicrobial and Messing B, Peitra-Cohen S, Debure A et al (1988)
antibiofilm activity. ACS Appl Mater Interfaces Antibiotic-lock technique: a new approach to optimal
5:6412–6422. https://doi.org/10.1021/am401629p therapy for catheter-related sepsis in home-parenteral
Lim K, Chua RRY, Ho B et al (2015) Development of nutrition patients. J Parenter Enter Nutr 12:185–189.
a catheter functionalized by a polydopamine peptide https://doi.org/10.1177/0148607188012002185
coating with antimicrobial and antibiofilm properties. Mielich-Süss B, Lopez D (2015) Molecular mecha-
Acta Biomater 15:127–138. https://doi.org/10.1016/j. nisms involved in Bacillus subtilis biofilm forma-
actbio.2014.12.015 tion. Environ Microbiol 17:555–565. https://doi.
Liu X, Yin H, Weng CX, Cai Y (2016) Low-frequency org/10.1111/1462-2920.12527
ultrasound enhances antimicrobial activity of colistin-­ Miller SM, Simon RJ, Ng S et al (1994) Proteolytic studies
vancomycin combination against pan-resistant of homologous peptide and N-substituted glycine pep-
biofilm of acinetobacter baumannii. Ultrasound toid oligomers. Bioorg Med Chem Lett 4:2657–2662.
Med Biol 42:1968–1975. https://doi.org/10.1016/j. https://doi.org/10.1016/S0960-894X(01)80691-0
ultrasmedbio.2016.03.016 Minardi D, Ghiselli R, Cirioni O et al (2007) The antimicro-
Lora-Tamayo J, Murillo O, Bergen PJ et al (2014) bial peptide Tachyplesin III coated alone and in com-
Activity of colistin combined with doripenem at bination with intraperitoneal piperacillin-­tazobactam
clinically relevant concentrations against mul- prevents ureteral stent pseudomonas infection in a rat
tidrug-resistant pseudomonas aeruginosa in an subcutaneous pouch model. Peptides 28:2293–2298.
in vitro dynamic biofilm model. J Antimicrob https://doi.org/10.1016/j.peptides.2007.10.001
Chemother 69:2434–2442. https://doi.org/10.1093/ Mishra B, Wang G (2017) Individual and combined
jac/dku151 effects of engineered peptides and antibiotics on
Luo Y, McLean DTF, Linden GJ et al (2017) The natu- Pseudomonas aeruginosa biofilms. Pharmaceuticals
rally occurring host defense peptide, LL-37, and its 10. https://doi.org/10.3390/ph10030058
truncated mimetics KE-18 and KR-12 have selected Mishra NM, Briers Y, Lamberigts C et al (2015) Evaluation
biocidal and antibiofilm activities against Candida of the antibacterial and antibiofilm activities of novel
albicans, Staphylococcus aureus, and Escherichia CRAMP–vancomycin conjugates with diverse link-
coli in vitro. Front Microbiol 8:544. https://doi. ers. Org Biomol Chem 13:7477–7486. https://doi.
org/10.3389/fmicb.2017.00544 org/10.1039/C5OB00830A
Mah TFC, O’Toole GA (2001) Mechanisms of bio- Mokkaphan J, Banlunara W, Palaga T et al (2014) Silicone
film resistance to antimicrobial agents. Trends surface with drug nanodepots for medical devices.
Microbiol 9:34–39. https://doi.org/10.1016/ ACS Appl Mater Interfaces 6:20188–20196. https://
S0966-842X(00)01913-2 doi.org/10.1021/am505566m
Maisetta G, Grassi L, Di Luca M et al (2016) Anti-biofilm Mora-Navarro C, Caraballo-Leõn J, Torres-Lugo M,
properties of the antimicrobial peptide temporin 1Tb Ortiz-Bermúdez P (2015) Synthetic antimicrobial
and its ability, in combination with EDTA, to eradicate β-peptide in dual-treatment with fluconazole or keto-
Staphylococcus epidermidis biofilms on silicone cath- conazole enhances the in vitro inhibition of planktonic
eters. Biofouling 32:787–800. https://doi.org/10.1080/ and biofilm Candida albicans. J Pept Sci 21:853–861.
08927014.2016.1194401 https://doi.org/10.1002/psc.2827
Maisetta G, Grassi L, Esin S et al (2017) The semi-­ Mygind PH, Fischer RL, Schnorr KM et al (2005)
synthetic peptide Lin-SB056-1 in combination with Plectasin is a peptide antibiotic with therapeutic poten-
EDTA exerts strong antimicrobial and antibiofilm tial from a saprophytic fungus. Nature 437:975–980.
activity against pseudomonas aeruginosa in conditions https://doi.org/10.1038/nature04051
mimicking cystic fibrosis sputum. Int J Mol Sci 18:1– Nguyen LT, Chau JK, Perry NA et al (2010) Serum sta-
18. https://doi.org/10.3390/ijms18091994 bilities of short tryptophan- and arginine-rich antimi-
Malmsten M, Kasetty G, Pasupuleti M et al (2011) crobial peptide analogs. PLoS One 5:e12684. https://
Highly selective end-tagged antimicrobial peptides doi.org/10.1371/journal.pone.0012684
278 H. Shahrour et al.

Nie B, Ao H, Long T et al (2017) Immobilizing bacitra- Reffuveille F, de la Fuente-Núñez C, Mansour S, Hancock

cin on titanium for prophylaxis of infections and for REW (2014) A broad-spectrum antibiofilm peptide
improving osteoinductivity: an in vivo study. Colloids enhances antibiotic action against bacterial biofilms.
Surf B Biointerfaces 150:183–191. https://doi. Antimicrob Agents Chemother 58:5363–5371. https://
org/10.1016/j.colsurfb.2016.11.034 doi.org/10.1128/AAC.03163-14
Niemirowicz K, Piktel E, Wilczewska AZ et al (2016) Ren H, Wu J, Colletta A et al (2016) Efficient eradication
Core–shell magnetic nanoparticles display synergis- of mature Pseudomonas aeruginosa biofilm via con-
tic antibacterial effects against Pseudomonas aeru- trolled delivery of nitric oxide combined with antimi-
ginosa and Staphylococcus aureus when combined crobial peptide and antibiotics. Front Microbiol 7:1–8.
with cathelicidin LL-37 or selected ceragenins. https://doi.org/10.3389/fmicb.2016.01260
Int J Nanomedicine 11:5443–5455. https://doi. Riool M, de Breij A, Drijfhout JW et al (2017)
org/10.2147/IJN.S113706 Antimicrobial peptides in biomedical device manu-
Oh E, Bae J, Kumar A et al (2018) Antioxidant-based facturing. Front Chem 5:63. https://doi.org/10.3389/
synergistic eradication of methicillin-resistant fchem.2017.00063
Staphylococcus aureus (MRSA) biofilms with bacitra- Roilides E, Simitsopoulou M, Katragkou A, Walsh TJ
cin. Int J Antimicrob Agents. https://doi.org/10.1016/j. (2015) How biofilms evade host defenses. Microbiol
ijantimicag.2018.03.006 Spectr 3:287–300. https://doi.org/10.1128/microbiol-
Orlando F, Ghiselli R, Cirioni O et al (2008) BMAP-28 spec.MB-0012-2014
improves the efficacy of vancomycin in rat mod- Rudilla H, Fusté E, Cajal Y et al (2016) Synergistic
els of gram-positive cocci ureteral stent infection. antipseudomonal effects of synthetic peptide AMP38
Peptides 29:1118–1123. https://doi.org/10.1016/j. and carbapenems. Molecules 21:1–12. https://doi.
peptides.2008.03.005 org/10.3390/molecules21091223
Overhage J, Campisano A, Bains M et al (2008) Human Sánchez-Gómez S, Martínez-de-Tejada G (2017)
host defense peptide LL-37 prevents bacterial biofilm Antimicrobial peptides as anti-biofilm agents in medi-
formation. Infect Immun 76:4176–4182. https://doi. cal implants. Curr Top Med Chem 17:590–603. https://
org/10.1128/IAI.00318-08 doi.org/10.2174/1568026616666160713141439
Ozbek B, Mataraci E (2013) In vitro effectiveness of Sánchez-Gómez S, Ferrer-Espada R, Stewart PS et al
colistin, tigecycline and levofloxacin alone and com- (2015) Antimicrobial activity of synthetic cationic
bined with clarithromycin and/or heparin as lock peptides and lipopeptides derived from human lacto-
solutions against embedded Acinetobacter baumannii ferricin against Pseudomonas aeruginosa planktonic
strains. J Antimicrob Chemother 68:827–830. https:// cultures and biofilms. BMC Microbiol 15:137. https://
doi.org/10.1093/jac/dks472 doi.org/10.1186/s12866-015-0473-x
Ozbek B, Mataraci-Kara E (2016) Comparative in vitro Scarsini M, Tomasinsig L, Arzese A et al (2015)
efficacies of various antipseudomonal antibiotics based Antifungal activity of cathelicidin peptides against
catheter lock solutions on eradication of Pseudomonas planktonic and biofilm cultures of Candida species
aeruginosa biofilms. J Chemother 28:20–24. https:// isolated from vaginal infections. Peptides 71:211–221.
doi.org/10.1179/1973947814Y.0000000212 https://doi.org/10.1016/J.PEPTIDES.2015.07.023
Palmer J, Flint S, Brooks J (2007) Bacterial cell attach- Sharma A, Gupta P, Kumar R, Bhardwaj A (2016)
ment, the beginning of a biofilm. J Ind Microbiol dPABBs: a novel in silico approach for predicting
Biotechnol 34:577–588. https://doi.org/10.1007/ and designing anti-biofilm peptides. Sci Rep 6:21839.
s10295-007-0234-4 https://doi.org/10.1038/srep21839
Pletzer D, Hancock REW (2016) Antibiofilm peptides: Shire SJ, Shahrokh Z, Liu J (2004) Challenges in the
potential as broad-spectrum agents. J Bacteriol development of high protein concentration formu-
198:2572–2578. https://doi.org/10.1128/JB.00017-16 lations. J Pharm Sci 93:1390–1402. https://doi.
Pletzer D, Coleman SR, Hancock REW (2016) Anti-­ org/10.1002/jps.20079
biofilm peptides as a new weapon in antimicrobial Silva RR, Avelino KYPS, Ribeiro KL et al (2016)
warfare. Curr Opin Microbiol 33:35–40. https://doi. Chemical immobilization of antimicrobial peptides on
org/10.1016/j.mib.2016.05.016 biomaterial surfaces. Front Biosci (Schol Ed) 8:129–
Rabin N, Zheng Y, Opoku-Temeng C et al (2008) Biofilm 142. https://doi.org/10.2741/s453
formation mechanisms and targets for developing anti- Simonetti O, Cirioni O, Cacciatore I et al (2016) Efficacy
biofilm agents. Future Med Chem 7:493–512. https:// of the quorum sensing inhibitor FS10 alone and in
doi.org/10.4155/FMC.15.6 combination with tigecycline in an animal model of
Rajasekaran G, Kim EY, Shin SY (2017) LL-37-derived staphylococcal infected wound. PLoS One 11:1–12.
membrane-active FK-13 analogs possessing cell selec- https://doi.org/10.1371/journal.pone.0151956
tivity, anti-biofilm activity and synergy with chlor- Singh PK, Parsek MR, Greenberg EP, Welsh MJ (2002)
amphenicol and anti-inflammatory activity. Biochim A component of innate immunity prevents bacterial
Biophys Acta Biomembr 1859:722–733. https://doi. biofilm development. Nature 417:552–555. https://
­
org/10.1016/j.bbamem.2017.01.037 doi.org/10.1038/417552a
14 AMPs as Anti-biofilm Agents for Human Therapy and Prophylaxis 279

Singh K, Shekhar S, Yadav Y et al (2017) DS6: antican- Wakabayashi H, Yamauchi K, Kobayashi T et al (2009)
didal, antibiofilm peptide against Candida tropicalis Inhibitory effects of lactoferrin on growth and biofilm
and exhibit synergy with commercial drug. J Pept Sci formation of Porphyromonas gingivalis and Prevotella
23:228–235. https://doi.org/10.1002/psc.2973 intermedia. Antimicrob Agents Chemother 53:3308–
Singha P, Locklin J, Handa H (2017) A review of the 3316. https://doi.org/10.1128/AAC.01688-08
recent advances in antimicrobial coatings for uri- Wang Z, De La Fuente-Núñez C, Shen Y et al (2015)
nary catheters. Acta Biomater 50:20–40. https://doi. Treatment of oral multispecies biofilms by an anti-­
org/10.1016/j.actbio.2016.11.070 biofilm peptide. PLoS One 10:1–16. https://doi.
Stanley NR, Lazazzera BA (2004) Environmental sig- org/10.1371/journal.pone.0132512
nals and regulatory pathways that influence biofilm Willcox MDP, Hume EBH, Aliwarga Y et al (2008)
formation. Mol Microbiol 52:917–924. https://doi. A novel cationic-peptide coating for the preven-
org/10.1111/j.1365-2958.2004.04036.x tion of microbial colonization on contact lenses.
Stewart PS, Costerton JW (2001) Antibiotic resistance of J Appl Microbiol 105:1817–1825. https://doi.
bacteria in biofilms. Lancet 358:135–138. https://doi. org/10.1111/j.1365-2672.2008.03942.x
org/10.1016/S0140-6736(01)05321-1 Wu H, Moser C, Wang H-Z et al (2015) Strategies for
Strömstedt AA, Pasupuleti M, Schmidtchen A, Malmsten combating bacterial biofilm infections. Int J Oral Sci
M (2009) Evaluation of strategies for improving pro- 7:1–7. https://doi.org/10.1038/ijos.2014.65
teolytic resistance of antimicrobial peptides by using Xu W, Zhu X, Tan T et al (2014) Design of embedded-­
variants of EFK17, an internal segment of LL-37. hybrid antimicrobial peptides with enhanced
Antimicrob Agents Chemother 53:593–602. https:// cell selectivity and anti-biofilm activity. PLoS
doi.org/10.1128/AAC.00477-08 One 9:e98935. https://doi.org/10.1371/journal.
Tan XW, Lakshminarayanan R, Liu SP et al (2012) Dual pone.0098935
functionalization of titanium with vascular endothe- Yazici H, O’Neill MB, Kacar T et al (2016) Engineered
lial growth factor and β-defensin analog for poten- chimeric peptides as antimicrobial surface coat-
tial application in keratoprosthesis. J Biomed Mater ing agents toward infection-free implants. ACS
Res B Appl Biomater 100B:2090–2100. https://doi. Appl Mater Interfaces 8:5070–5081. https://doi.
org/10.1002/jbm.b.32774 org/10.1021/acsami.5b03697
Tong Z, Ling J, Lin Z et al (2013) The effect of MTADN Yoshinari M, Kato T, Matsuzaka K et al (2010) Prevention
on 10 enterococcus faecalis isolates and biofilm: of biofilm formation on titanium surfaces modified
an in vitro study. J Endod 39:674–678. https://doi. with conjugated molecules comprised of antimicrobial
org/10.1016/j.joen.2012.12.010 and titanium-binding peptides. Biofouling 26:103–
Tong Z, Zhang Y, Ling J, et al (2014a) An in vitro study 110. https://doi.org/10.1080/08927010903216572
on the effects of nisin on the antibacterial activi- Yu L, Hou Y, Cheng C et al (2017a) High-antifouling
ties of 18 antibiotics against enterococcus faeca- polymer brush coatings on nonpolar surfaces via
lis. PLoS One 9. https://doi.org/10.1371/journal. adsorption-cross-linking strategy. ACS Appl Mater
pone.0089209 Interfaces 9:44281–44292. https://doi.org/10.1021/
Tong Z, Zhang L, Ling J et al (2014b) An in vitro study on acsami.7b13515
the effect of free amino acids alone or in combination Yu K, Lo JCY, Yan M et al (2017b) Anti-adhesive anti-
with nisin on biofilms as well as on planktonic bacteria microbial peptide coating prevents catheter associ-
of Streptococcus mutans. PLoS One 9:3–10. https:// ated infection in a mouse urinary infection model.
doi.org/10.1371/journal.pone.0099513 Biomaterials 116:69–81. https://doi.org/10.1016/j.
Venkatesh M, Rong L, Raad I, Versalovic J (2009) Novel biomaterials.2016.11.047
synergistic antibiofilm combinations for salvage of Zhang T, Wang Z, Hancock REW et al (2016) Treatment
infected catheters. J Med Microbiol 58:936–944. of oral biofilms by a D-enantiomeric peptide. PLoS
https://doi.org/10.1099/jmm.0.009761-0 One 11:1–16. https://doi.org/10.1371/journal.
Vincentelli J, Braguer D, Guillet P et al (1997) pone.0166997
Formulation of a flush solution of heparin, vanco- Zhu C, Tan H, Cheng T et al (2013) Human β-defensin
mycin, and colistin for implantable access systems in 3 inhibits antibiotic-resistant staphylococcus bio-
oncology. J Oncol Pharm Pract 3:18–23. https://doi. film formation. J Surg Res 183:204–213. https://doi.
org/10.1177/107815529700300103 org/10.1016/j.jss.2012.11.048
 Clinical Application of AMPs
15
Fabíola Costa, Cátia Teixeira, Paula Gomes,
and M. Cristina L. Martins

Abstract still somewhat expensive production of AMP

Antimicrobial peptides (AMPs) have been along with regulatory obstacles. This chapter
described as one of the most promising com- offers an overview of selected AMPs that are
pounds able to address one of the main health currently in the market or under clinical trials.
threats of the twenty-first century that is the Strategies for assisting AMP industrial trans-
continuous rise of multidrug-resistant micro- lation and major regulatory difficulties associ-
organisms. However, despite the clear advan- ated with AMP approval for clinical evaluation
tages of AMPs as a new class of antimicrobials, will be also discussed.
such as broad spectrum of activity, high selec-
tivity, low toxicity and low propensity to
induce resistance, only a small fraction of
AMPs reported thus far have been able to suc- 15.1 Introduction
cessfully complete all phases of clinical trials
and become accessible to patients. This is The increasing reports on antibiotic resistance
mainly related to the low bioavailability and among pathogenic microorganisms have alerted
authorities to what has been named as the ‘post-­
antibiotic era’ (WHO 2018). Indeed, there are
F. Costa
only a limited number of chemical classes within
i3S, Instituto de Investigação e Inovação em Saúde,
Universidade do Porto, Porto, Portugal the currently used antibiotics, which share strong
similarities regarding both spectrum of activity
INEB, Instituto de Engenharia Biomédica,
Universidade do Porto, Porto, Portugal and mode of action, altogether favouring the
spread of cross-resistance between pathogens
C. Teixeira · P. Gomes
LAQV-REQUIMTE, Departamento de Química e (Martinez 2014). As there are only a few new
Bioquímica, Faculdade de Ciências, Universidade do antibiotics in the pharmaceutical industry pipe-
Porto, Porto, Portugal line, it is imperative to find new therapeutic
M. C. L. Martins (*) classes to tackle the growing menace of
i3S, Instituto de Investigação e Inovação em Saúde, multidrug-­ resistant microorganisms (Coates
Universidade do Porto, Porto, Portugal
et al. 2011; WHO 2018).
INEB, Instituto de Engenharia Biomédica, Antimicrobial peptides (AMPs) have been
Universidade do Porto, Porto, Portugal
suggested as a new therapeutic class with poten-
Instituto de Ciências Biomédicas Abel Salazar, tial to overcome some of the major disadvantages
Universidade do Porto, Porto, Portugal
associated with classical antibiotics, namely,
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2019 281

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4_15
282 F. Costa et al.

high prevalence of resistant microorganisms regarded as a ‘no go’ area for many pharmaceuti-
emerging only a few years after market entry, cal companies, given their poor pharmacokinet-
narrow scope of activity and significant systemic ics (PK), which arises from, among other factors,
toxicity (Zasloff 2002; Brooks and Brooks 2014). low absorption and short half-lives in vivo due to
As already explained in previous chapters, AMPs metabolic transformations and/or fast clearance
are a distinct and diverse class of molecules, most from the organism (Henninot et al. 2018). While
of which share some fundamental similarities this can be sometimes an advantage, as peptides
such as having cationic and hydrophobic domains do not tend to accumulate in the body and hence
and amphipathic behaviour (Zasloff 2002). have reduced toxicity risks, it also constitutes the
Despite the structural heterogeneity found within major issue that has to be dealt with when devel-
AMPs, they share some very important functions oping peptide therapeutics (Fernandes and
that include direct and broad-spectrum microbial Martens 2017; Henninot et al. 2018). In fact, as
killing (including of antibiotic-resistant microor- herein reviewed, shortcomings, such as inade-
ganisms), endotoxin neutralization, favourable quate in vitro to in vivo translation, high produc-
immune modulation, enhancement of adaptive tion costs and, mainly, poor oral bioavailability,
immune responses and, most importantly, very have been major barriers in AMP progression
low propensity to promote microorganism resis- into the clinics.
tance (Gordon et al. 2005; Wang et al. 2015). In
fact, AMPs constitute a new hope in the difficult
fight against the so-called ESKAPE bacteria, i.e. 15.2 AMPs in Clinical Use
multidrug-resistant strains of Enterococcus fae-
cium, Staphylococcus aureus, Klebsiella pneu- Given their high potency and selectivity, broad
moniae, Acinetobacter baumannii, Pseudomonas range of targets, potentially low toxicity, low
aeruginosa and Enterobacter species responsible accumulation in tissues and no propensity to trig-
for grave nosocomial infections (Giuliani et al. ger resistance, AMPs have been considered as
2007; Santajit and Indrawattana 2016; Gomes promising anti-infection candidates especially
et al. 2017). AMPs are also involved in the for polymicrobial infections as in the case of skin
intracellular processes of angiogenesis, wound and soft tissue infections (SSTI) (Mahlapuu et al.
healing and cell signalling, which make them 2016; Pfalzgraff et al. 2018). Still, despite the
especially interesting for the development of myriad of AMP-related patents filed, only a very
novel drugs (Afacan et al. 2012). These remark- narrow proportion of the AMPs have already
able properties have driven researchers to make received the certification, either by US Food and
considerable efforts towards the use of AMPs Drug Administration (FDA) or EU European
as commercially available drugs, which have Medicines Agency (EMA), mandatory for entry
been translated to an ever-increasing number of into market and clinical use. FDA-approved
submissions of AMP-related patents. In AMPs are listed on Table 15.1.
PATENTSCOPE of World Intellectual Property
Organization (WIPO), thousands of hits appear
when searching for AMPs (WIPO 2018). Most of 15.3 AMPs and Analogues
these patents are related to the description of Under Clinical Trials
newly found AMPs either from natural or syn-
thetic origin, as well as AMP derivatives with Clinical trials are aimed at testing potential treat-
increased activity, stability and/or lower toxicity. ments in human volunteers, to assess if they should
Interestingly, more recently submitted patents be approved for wider use in the general popula-
also include a focus on both an effective pharma- tion. This means that both efficacy and safety of
ceutical composition and an optimal administra- the proposed treatment should surpass the existing
tion route for AMPs (Kosikowska and Lesner drugs. Therefore, clinical trials are a key moment
2016; WIPO 2018). However, AMPs are still in the new drug development process, representing
15 Clinical Application of AMPs 283

Table 15.1 FDA-approved AMPs in clinical use (DrugDataBase n.d.; Usmani et al. 2017)
Route of
AMPa Sequenceb Indication administration
Bacitracin ICLeI[εKoIfHdN]c Skin and eye infections Topical application
(Mixture of side chain-to-tail cyclic peptides Parenteral Intramuscular
from Bacillus subtilis var Tracy; the sequence of (intramuscular)
bacitracin A is shown, which possesses a administration
modification of the Cys residue – the side chain restricted to infants
thiol of Cys has undergone an intramolecular with staphylococcal
condensation with the preceding peptide bond, pneumonia and
forming a thiazolidine ring; this modified Cys is empyema when due to
noted by C in the above sequence) organisms shown to be
susceptible to
bacitracin
Polymyxin B 6-mo-DabTDabDab[γDabfLDabDabT]d Last-line treatment Parenteral
(Mixture of side chain-to-tail cyclic lipopeptides option for multidrug-­ administration
from Bacillus polymyxa; sequence shown for resistant gram-­negative (intramuscular,
polymyxin B1 that is N-terminally acylated with bacterial infections intravenous,
6-methyloctanoic acid, 6-mo) intrathecal
ophthalmic)
Polymyxin E 6-mh-DabTDab[γDablLDabDabT]d Gastrointestinal tract Parenteral
(colistins) infections caused by administration
Escherichia coli and (intramuscular,
Salmonella spp. intravenous)
(Mixture of side chain-to-tail cyclic lipopeptides Multidrug-resistant
from Bacillus colistinus; sequence shown for gram-negative bacteria
colistin A, which is N-terminally acylated with (e.g. Pseudomonas
6-methylheptanoic acid, 6-mh) aeruginosa; Klebsiella
pneumoniae;
Acinetobacter;
unresponsive to other
antibiotics)
Tyrothricin [fPFfNQYVOL] Local treatment of Topical application
infected skin and only, as it is very
infected oropharyngeal toxic when
mucous membranes administrated
(Mixture of head-to-tail cyclic polypeptides from Effective against parenterally
Bacillus brevis; the sequence of tyrothricin A is gram-positive bacteria
shown)
Gramicidin D f
VGALAVVVWLWLWLW-ea Skin lesions, surface External use only
(or, simply, (Mixture of linear pentadecapeptides from wounds and eye
gramicidin) Bacillus brevis whose N-terminal residue (V or infections
L) is N-formylated (fV or fL) and whose
C-terminal tryptophan (W) is forming an
additional peptide bond to ethanolamine (ea); the
sequence of gramicidin A is represented)
Gramicidin S [fPVOLfPVOL] Potent against Topical application
(Cyclic peptide biosynthesized from gramicidin gram-negative and only, as it is highly
in Bacillus brevis; comprises two identical gram-positive bacteria haemolytic
pentapeptides coupled head to tail) and fungi restricted use
as spermicide and to
treat genital ulcers
caused by STD
(continued)
284 F. Costa et al.

Table 15.1 (continued)

Route of
AMPa Sequenceb Indication administration
Daptomycin da-WnD[βTGODaDGs3mEKyn]e (side chain-to- Complicated skin Intravenous
(lipopeptide) tail cyclic lipopeptide from Streptomyces infections caused by injection
roseosporus, which is N-terminally acylated with susceptible strains of
decanoic acid (da) and comprises the unusual gram-positive
C-terminal amino acid kynurenine (Kyn, microorganisms
metabolite of Trp)
a
Other well-known antibiotics that include amino acid building blocks, and are thus considered as AMPs by many
authors, e.g. teicoplanin or vancomycin, are outside the scope of this chapter, whose focus is on AMPs that, regardless
of having or not modified amino acid residues, retain a typical peptide backbone
b
Uppercase single-letter codes for standard L-α-amino acids have been used as defined by the IUPAC-IUB Joint
Commission on Biochemical Nomenclature; lowercase indicates the corresponding D-enantiomers (e.g. if ‘O’ stands
for L-ornithine, then ‘o’ stands for D-ornithine); amino acid residues inside brackets are those comprised in cyclic
homodetic (lactam) structures (which may cover either a part or the full peptide sequence)
cε
K stands for the Lys residue whose side chain amine (εNH2) group is forming an intramolecular lactam (amide) bridge
with the C-terminal amino acid
d
Dab stands for L-α,γ-diaminobutanoic acid, and γDab stands for the Dab residue whose side chain amine (γNH2) group
is forming an intramolecular lactam (amide) bridge with the C-terminal amino acid
eβ
T stands for a Thr residue whose side chain hydroxyl (βOH) is forming an intramolecular lactone (ester) bridge with
the C-terminal amino acid; 3mE stands for 3-methyl glutamic acid

the major struggle towards market entry, in terms (ClinicalTrialsPage n.d.). A Phase III trial aims
of both time and costs (ClinicalTrials.gov n.d.; at gathering additional information on safety and
ClinicalTrialsPage n.d.; Greber and Dawgul 2017). effectiveness, by studying different populations
Benefits may only come after competent authori- and different dosages and using the drug in com-
ties such as the FDA (US) or EMA (EU) approval, bination with other drugs (drug interactions stud-
where the patent owner/sponsor earns the right of ies); this phase mainly targets assessment of
exclusive marketing of the drug for the period of whether the benefits of the new medicine are
patent protection (usually for 10–15 years). higher than those of currently available therapeu-
Clinical trials are organized in four phases, tics for the same indication. Therefore, the num-
always in strict obedience to the standards of ber of subjects involved at this stage ranges from
Good Clinical Practice (GCP) (ClinicalTrialsPage hundreds to thousands (ClinicalTrialsPage n.d.).
n.d.). Phase I trial tests are focused in judging the If the drug obtains a positive outcome at this
drug safety and potential side-effects and in find- stage, authorities (either FDA or EMA) will grant
ing the correct drug dosage. Therefore, PK the certification required for its introduction in
(absorption, distribution, metabolism, excretion the market. Finally, Phase IV trials are conducted
and toxicity – ADMET) is the main focus of the when the new drug is already in the market, by
trial, while drug pharmacodynamics (PD – phar- monitoring effectiveness and safety in a much
macological activity of the drug in the body) is larger scale, involving diverse populations
also surveilled; often, a small group (50–100) of (ClinicalTrialsPage n.d.).
healthy individuals is involved at this stage. A Given the broad spectrum of activity of AMPs,
larger group of people is involved in Phase II tri- it is possible to find the same peptide undergoing
als (up to 600), whose focus is on the effective- clinical trials for different pathological condi-
ness rather than safety, as this phase aims at tions or for use via different administration routes
obtaining preliminary data on whether the drug (ClinicalTrials.gov n.d.). This means that failure
works in people affected by a certain disease or in a specific clinical trial of a given AMP does not
condition; comparative studies (use of a new drug necessarily rule out its potential approval, and
compared to a standard treatment and placebo) consequent clinical use, in a different scenario.
are mainly conducted in this phase As outlined by Greber et al. (2017), most AMPs
15 Clinical Application of AMPs 285

under clinical development are being tested for for treatment of oral mucositis in patients receiv-
topical application, as major challenges for pep- ing radiation therapy for head and neck cancer
tide administration by other routes have not yet but terminated at Phase III due to low effective-
been satisfactorily addressed, as further discussed ness. The same AMP was then tested in the pre-
in the next section (Greber and Dawgul 2017). As vention of ventilator-associated pneumonia,
such, the number of AMPs that have successfully having failed again at Phase II/III. Interestingly,
crossed the full clinical trial path is low, particu- an iseganan derivative, POL7080 (murepavadin),
larly when comparing with the enormous number has started a new promising path, having com-
of patented AMPs for potential therapeutic appli- pleted two Phase II studies in patients with
cations (WIPO 2018). Table 15.2 highlights the ventilator-­associated bacterial pneumonia and
selected AMPs currently in the latest stages of non-cystic fibrosis-associated bronchiectasis
clinical development. Omiganan (an indolicin (Polyphor, Lda) (WebPage n.d.; Kosikowska and
derivative, i.e. belonging to the cathelicidin fam- Lesner 2016). One final example of an AMP that
ily) was one of the first recent AMPs able to com- is a Phase III achiever is bovine lactoferrin that is
plete Phase III trials’ effectiveness tests for currently recruiting to improve neonatal survival
treatment of rosacea while being also under in low birth weight babies, by fighting neonatal
Phase II trials for evaluation of safety and effi- sepsis and necrotizing enterocolitis
cacy of an omiganan-based topical gel on female (ClinicalTrials.gov n.d.). Actually, several
subjects with moderate to severe inflammatory attempts have been made over the years to put
Acne vulgaris (Cutanea Life Sciences, Inc.) forward into the clinics different AMPs from the
(ClinicalTrials.gov n.d.; Sader et al. 2004; Melo lactoferrin family. Lactoferrin has been tested as
and Castanho 2007). Interestingly, there are sev- a dietary supplement both for HIV-1/AIDS infec-
eral clinical trials involving peptides from the tion and for nosocomial infections in critically ill
cathelicidin family, such as those addressing both patients, having completed Phase II
the immunomodulatory effects of hCAP18 and (ClinicalTrials.gov n.d.). AM-Pharma developed
LL37 peptides in improving infection outcomes the human lactoferrin-derived peptide (hLF1-11)
and the direct bactericidal effects of some of their for two potential indications: treatment of infec-
derivatives (ClinicalTrials.gov n.d.). For instance, tious complications among autologous haemato-
LL37 has been tested for the treatment of hard-­ poietic stem cell transplant (HSCT) recipients
to-­heal venous leg ulcers (Phase II), whereas its and bacteraemia due to S. epidermidis. However,
derivative OP-145 (also known as AMP60.4Ac) Phase I/II clinical trials with hLF1-11 for these
has been tested for the treatment of chronic sup- two applications were withdrawn prior to enrol-
purative otitis media with very positive Phase II ment. Nevertheless, AM-Pharma was able to take
results (OctoPlus BV) (ClinicalTrials.gov n.d.; to conclusion Phase I/II studies on the safety of a
Greber and Dawgul 2017). Notwithstanding, the single dose of 5 mg of hLF1-11 given to autolo-
best-known AMP reaching Phase III trials is pex- gous HSCT recipients (Greber and Dawgul
iganan, an analogue of magainin 2. A 0.8% cream 2017).
containing pexiganan was tested in Phase III clin- Another promising AMP is PAC-113, an ana-
ical trials for treatment of infected diabetic foot logue of histatin 5 that occurs naturally in saliva,
ulcers (Dipexium Pharmaceuticals, Inc.) which has high activity against Candida albi-
(ClinicalTrials.gov n.d.). Despite the positive cans, including drug-resistant strains isolated
outcomes, the FDA did not approve its marketing from HIV-1 carriers. PAC-113 has passed Phase
on the grounds that its efficacy was comparable II clinical trials aimed at determining the optimal
to that of the standard oral antibiotic ofloxacin peptide doses for mouth rinsing towards treat-
(Greber and Dawgul 2017). Another Phase III ment of oral candidiasis in HIV-1 carriers
achiever is iseganan, a protegrin-1-derived (ClinicalTrials.gov n.d.; Greber and Dawgul
AMP. Iseganan was submitted to clinical trials 2017).
286 F. Costa et al.

Table 15.2 Selected AMP and AMP derivatives under clinical trial as of July 2018
Clinical trial
Peptide Origin Evaluated condition phase Sponsor/collaborator
Omiganan (MBI Cathelicidin Treatment of rosacea III Cutanea Life Sciences
226 or MX 226) family (completed)
Treatment of severe II Cutanea Life Sciences
inflammatory acne
vulgaris
Preventing central venous III Mallinacckrodt
catheter infection
LL37 Cathelicidin Treatment of hard-to-heal II Lipopeptide AB
family (derived venous leg ulcers
from hCAP18)
OP-145 Cathelicidin Treatment of chronic II OctoPlus BV
(AMP60.4Ac) family (derived suppurative otitis media
from LL37)
Pexiganan Analogue of Treatment of infected III Dipexium Pharmaceuticals
magainin-2 diabetic foot ulcers (has merged with PLx
Pharma)
Iseganan Protegrin-1 Treatment of oral III (failed) IntraBiotics Pharmaceuticals
derivative mucositis in patients
receiving radiation
therapy for head and neck
cancer
Treatment for ventilator-­ II/III (failed)
associated pneumonia
POL7080 Derivative of Treatment of ventilator-­ II Polyphor, Lda
(murepavadin) iseganan associated pneumonia
Treatment of non-cystic II
fibrosis-associated
bronchiectasis
PAC-113 Histatin Mouth rinse in HIV-­ II Pacgen Biopharmaceuticals
analogue seropositive individuals Corporation/Quintiles, Inc.
with oral candidiasis
Lactoferrin HIV infection II (dietary Jason Baker/Ventri
supplement) Bioscience/Minneapolis
Medical Research Foundation
Lactoferrin Nosocomial infection in II (dietary Queen’s University
critically ill patients supplement)
Bovine Neonatal sepsis III Aga Khan University/
lactoferrin Necrotizing enterocolitis III University of Sydney/United
States Agency for International
Development
hLF1-11 Derived from Treatment of infectious I/II AM-Pharma
human complications among
lactoferrin HSCT recipients
Treatment of bacteraemia I/II
due to Staphylococcus (withdrawn)
epidermidis
Talactoferrin Oral solution for I/II Agennix/NIH
nosocomial infection in
preterm infants
15 Clinical Application of AMPs 287

15.4 Challenges in the Clinical excretion (ADME), which is further aggravated

Application of AMPs by their propensity to aggregate with plasma pro-
teins, the vast majority of natural peptides has an
The very high number of promising AMPs found oral bioavailability no higher than 1% (Di 2015).
in numerous patents and literature reports fail in Still, as compared with protein-/antibody-based
the translation into clinical trials mostly because therapeutics, peptides have much better perme-
the preclinical studies are frequently performed in ation into tissues and are considerably more sta-
absence of physiological conditions. This means ble, less prone to trigger immune/allergic
that the influence of, e.g. (i) physiological levels of reactions and significantly cheaper to produce,
relevant ions, (ii) high ionic strength (due to high- which underpins the extensive research that has
salt concentration) and (iii) presence of proteases been devoted to find strategies to improve their
and of physiological barriers is frequently over- bioavailability (Otvos and Wade 2014; Di 2015) .
looked (Fox 2013; Mahlapuu et al. 2016). This A number of chemical strategies have been
often leads to in vivo performances that are unex- developed to tackle this hurdle, mainly by
pectedly low, as compared to in vitro data. On the increasing AMP (i) metabolic stability and (ii)
other hand, the opposite sometimes happens: high- lipophilicity, towards enhanced permeability (Di
throughput analysis of an AMP library has revealed 2015; Molchanova et al. 2017; Qvit et al. 2017;
AMP with poor in vitro performance to be very Sierra et al. 2017; Henninot et al. 2018). Common
effective in vivo, even in amounts below their bac- approaches include:
tericidal concentrations, which was ascribed to
their immunomodulatory properties promoting • Cyclization to increase metabolic stability
bacterial killing via stimulation of the host immune and/or lock bioactive conformations
system (Hancock and Sahl 2006). Also, variability • Replacement of genetically encoded amino
associated to different animal models used for acids by noncanonical residues, to deliver
in vivo testing of AMPs may create results with peptidomimetics of increased stability and
poor translation into humans. permeability
Adding to the above, one of the greatest chal- • N-/C-Terminal modifications, to avoid proteo-
lenges towards effective clinical applications of lytic degradation and increase permeability
AMPs is their low oral bioavailability. Peptides are
unstable in the gastrointestinal tract (GIT) due to Backbone modifications, i.e. replacement of
both the harsh gastric pH and proteolytic enzymes, peptide bonds by enzyme-stable bioisosteres,
which is further aggravated by their poor penetra- have been also applied, but will not be addressed
tion across the intestinal epithelium (Di 2015; in this chapter.
Lundquist and Artursson 2016; Lau and Dunn
2018). Hence, the benefits that AMPs represent as 15.4.1.1 Cyclization
a potential new class of antimicrobial agents Cyclization has been proven quite effective
require that efficient strategies are found which which is supported by the fact that most orally
enable them to step out of their current almost active peptides are cyclic, including well-known
exclusive confinement to topical applications. potent peptide antibiotics like bacitracin A,
Some of the approaches that are being devised to ­colistin, gramicidins and polymyxins B1 and B2,
tackle these challenges are next addressed. just to name a few emblematic examples (Joo
2012; Tapeinou et al. 2015; Falanga et al. 2017).
Chemical strategies towards peptide cycliza-
15.4.1 Chemical Approaches tion span from heterodetic (cyclization through a
to Tackle Low Bioavailability non-amide, e.g. a disulphide), or homodetic
(cyclization through an amide, i.e. lactamization)
Given their generally poor absorption and distri- cyclic peptides, to fully unnatural enzyme-stable
bution, along with fast metabolic degradation and all-hydrocarbon tethers (Fig. 15.1) (Lambert
288 F. Costa et al.

Fig. 15.1 Classic peptide cyclization approaches. Top: chain-to-side chain lactamization to give homodetic cyclic
intramolecular oxidation of cysteine side chain thiols to peptide (Adapted from http://www.bachem.com/service-
produce cyclic disulphide (heterodetic) peptides; bottom: support/newsletter/peptide-trends-march-2016/)
(1) head-to-tail, (2) side chain-to-head/tail and (3) side

et al. 2001; Davies 2003; White and Yudin 2011; more stable than their linear precursors, even
Chu et al. 2015; Marti-Centelles et al. 2015). against endopeptidases, cyclic homodetic pep-
Heterodetic cyclization through formation of a tides may remain somewhat susceptible to meta-
single disulphide bridge via oxidation of the side bolic degradation (Bourne et al. 1996).
chain thiols of a couple of cysteines (either More recent approaches towards stable cyclic
belonging or added to the native peptide sequence) peptides are taking advantage of the wide pano-
is quite straightforward from the chemical view- ply of commercial non-natural amino acids suit-
point (Fig. 15.1, top) and has delivered α-helical ably functionalized to enable application of
AMPs inspired in tenecin 1, an insect defensin ‘click’-type intramolecular reactions. These che-
peptide (Ahn et al. 2008). Still, given that peptides moselective cyclizations include, among many
with a single disulphide bridge may be unstable in others, (i) Michael-type (thiol-ene) additions
bioreducing environments (Yang et al. 2006), (Fig. 15.2a), (ii) Huisgen’s 1,3-dipolar (azide-­
efforts have been focused on AMPs bearing mul- alkyne) cycloadditions (Fig. 15.2b) and (iii)
tiple disulphide bridges, e.g. as those inspired in ring-­
­ closure olefin metathesis (RCM), also
human defensins (Cheneval et al. 2014; Falanga known as Grubbs reaction (Fig. 15.2c) (Lambert
et al. 2017), and on development of efficient et al. 2001; Davies 2003; White and Yudin 2011;
chemical methods for the synthesis of cysteine- Chu et al. 2015; Marti-Centelles et al. 2015; Lau
rich peptide (Cheneval et al. 2014). Another and Dunn 2018).
approach towards stable cyclic AMPs formed by Peptide cyclization via ring-closure metathe-
means of ‘natural’ bonds has been lactamization sis (RCM), leading to the so-called stapled pep-
(Fig. 15.1, bottom), i.e. formation of intramolecu- tides, is actually gaining prominence towards
lar peptide bonds to produce homodetic cyclic development of synthetic AMPs and structurally
peptides (Monaim et al. 2018). Still, despite being related peptides, like cell-penetrating peptides
15 Clinical Application of AMPs 289

Fig. 15.2 Examples of chemoselective (‘click’-type) on use of (a) Grubbs catalyst and (b) non-natural amino
cyclization reactions: (a) thiol-ene and azide-alkyne con- acid building blocks, to yield (c) stapled peptides (Chu
jugations and (b) ring-closure metathesis (RCM) settling et al. 2015)

(CPPs), by presenting a restrained and 15.4.1.2  se of Noncanonical Amino

U
­metabolically stable helical conformation, widely Acids
known as relevant for the membrane interaction Another popular approach to improve peptides’
ability of most cationic amphipathic AMPs and bioavailability has been the replacement of pro-
CPPs (Walensky and Bird 2014; Lau and Dunn teinogenic amino acids in the native sequence by
2018; Migon et al. 2018). non-coded ones, such as D-amino acids,
290 F. Costa et al.

N-methylated amino acids and other noncanoni- cyclization, to produce more stable and perme-
cal residues. In this sense, the term ‘peptidomi- able AMPs (Rader et al. 2018). Remarkably, the
metic’ has arisen to classify cases where amino nature itself has been using the combination of all
acid residues present in the natural bioactive these strategies, i.e. cyclization, use of D-amino
sequence were replaced by similar building acids and N-methylation, to afford orally active
blocks, often differing only in chirality and/or peptides, like cyclosporine A (Fig. 15.3), or
other subtle structural modifications, with the AMPs of marine origin, e.g. discodermins
main purpose of enhancing bioavailability (Wu (Falanga et al. 2016).
and Gellman 2008).
Merrifield and co-workers were pioneers in 15.4.1.3 N-/C-Terminal Modifications
using D-amino acids to improve the bioavailabil- Along with N-methylation, other reasonably
ity of AMPs, by producing ‘retro’ (inverted straightforward approaches can be used to
sequence, all L-amino acids), ‘enantio’ (normal enhance peptides’ stability and circulating half-­
sequence, all D-amino acids) and ‘retro-inverso’ lives, namely, modifications/conjugations at the
(or ‘retro-enantio’) analogues of cecropin N- and/or C-termini of the bioactive sequence or,
A-melittin hybrid AMPs, and the latter were less frequently, at the side chains of trifunctional
found to be equipotent to their original membrane-­ amino acid residues like Cys, Ser, Thr or Lys.
active AMP templates while presenting markedly The simplest methods include C-terminal amida-
higher proteolytic stability (Merrifield et al. tion and/or N-terminal acylation (Di 2015;
1995). Use of retro-inverso analogues of bioac- Mahlapuu et al. 2016; Henninot et al. 2018), the
tive peptides has endured as a popular approach latter eventually using fatty acids (peptide lipida-
to increase AMPs’ stability towards proteolysis tion) (Chicharro et al. 2001; Oh et al. 2014; Trier
(Cardoso et al. 2018), as the main target of et al. 2014; Vidal and Geffard 2014; Swiecicki
amphipathic α-helical AMPs is the bacterial et al. 2015; Cochrane et al. 2016; Di Pisa et al.
membrane, not involving stereospecific interac- 2015). Alkylation of the ε-amine group of Lys
tions. Hence, resorting to retro-inverso ana- side chains is an equally simple approach that has
logues, or just discrete replacements of L- by been used in AMPs in order to preserve or even
D-amino acids, remains an interesting option to increase their antimicrobial potency (Teixeira
improve AMPs’ performance (Silva et al. 2014). et al. 2010) while possibly reducing their lability
However, though retro-inverso AMPs may be to enzymatic degradation by, e.g. trypsin (Arias
fully proteolysis resistant, their permeability is et al. 2018).
not expected to differ much from that of native Other moieties have been conjugated to pep-
AMPs. Actually, it has been reported that retro-­ tides, the most popular being polyethylene glycol
inverso peptides may have increased hydrophilic- (PEGylation) and sugars (glycosylation), in order
ity as compared to their normal templates to improve peptide’s biocompatibility and bio-
(Alminana et al. 2004), which has a negative availability (Veronese 2001; Henninot et al. 2018;
impact on the peptide’s ability to permeate across Lau and Dunn 2018). PEGylation protects pep-
physiological barriers such as, e.g. mucosae, epi- tides from proteolytic degradation, decreases
thelia and sub-epithelial tissues (Lundquist and their immunogenicity by limiting their uptake via
Artursson 2016). dendritic cells and, in some instances, increases
Given that orally available peptides often pos- their circulating half-life by over two orders of
sess molecular weights no higher than 1000 Da magnitude, partially due to reduced renal clear-
and no more than six hydrogen bond donors ance (Veronese 2001; Henninot et al. 2018; Lau
(Doak et al. 2014), one way to improve AMP per- and Dunn 2018). PEGylated AMPs were found to
meability is to reduce their number of hydrogen have increased lung biocompatibility suitable for
bond donors. This can be achieved by the use of local inhaled therapy of respiratory infections
N-methylated amino acids (Chatterjee et al. (Morris et al. 2012). Chemoselective coupling of
2013); this strategy has been combined with PEG has experienced some limitations in the
15 Clinical Application of AMPs 291

Fig. 15.3 Structure of cyclosporine A, an orally active sarcosine, Sar), N-methyl-L-valine (N-MeVal) and
peptide drug produced in nature by Tolypocladium infla- N-methyl-L-leucine (N-MeLeu); one D-amino acid
tum fungi, which exhibits almost all of the features known (D-alanine); and two additional N-methylated noncanoni-
to enhance peptides’ oral bioavailability: N-methylated cal amino acids, N-methyl-2-aminobutanoic acid and
amino acids, namely, N-methyl-glycine (best known as N-methyl-2-amino-3-hydroxy-4-methyloct-6-enoic acid

early days of peptide PEGylation (Veronese ented by Hoffman et al. (2014). Remarkably, in
2001), including activity decrease for some this case, AMP (namely, apidaecin and oncocin)
AMPs (Guiotto et al. 2003; Imura et al. 2007). side chains were PEGylated using a PEG modi-
Suitable chemoselective methods became avail- fied with a linker that includes a recognition
able in recent years (Roberts et al. 2002; Doherty sequence for trypsin-like serum proteases; in
et al. 2005; Hamley 2014), but the balance other words, a protease-sensitive linker was used
between improving PK and retaining good PD is to increase peptide stability and circulation time,
a delicate one (Fishburn 2008); one recent as it allowed for a controlled release of the AMPs
approach to bypass this limitation is based on the in their target tissues. Moreover, this strategy also
use of chemoreversible PEGylation of arginine contributed to delayed peptide renal excretion.
side chains via cleavable phenylglyoxal linkers Glycosylation may also improve peptides’ PK
(Fang et al. 2017); this produced prodrugs of and PD (Moradi et al. 2016). Naturally glycosyl-
arginine-rich AMPs, whose resistance to serum ated AMPs, namely, insect-derived Pro-rich
proteases was dramatically increased while AMPs (Lele et al. 2015), have inspired chemists
allowing the parent bioactive peptide to be slowly and bioengineers to develop artificial site-specific
released in a matter of hours to days (Gong et al. glycosylation methods delivering glycopeptides
2017). Actually, a temporary PEGylation strategy with improved stability, bioactivity and specific-
to increase stability, reduce immunogenicity and ity (Doores et al. 2006; Bednarska et al. 2017).
improve the PK of AMPs for intravenous admin- Bioactive peptides have also been conjugated to
istration had been previously described and pat- selected moieties, including antibodies or albu-
292 F. Costa et al.

Fig. 15.4 Peptide modification/conjugation toolbox: geted delivery, and conjugation to protein or
from current approaches, like lipidation, glycosylation protein-binding moieties allowing for increased peptide
and PEGylation, to more sophisticated strategies, includ- circulating half-lives
ing insertion of specific peptide segments enabling tar-

min, to increase circulating time due to reduced gold NPs have been also reported (Rai et al.
renal clearance and rescue by the FcRn recycling 2016a, b; Reinhardt and Neundorf 2016; Zong
pathway (Sockolosky and Szoka 2015; Schmidt et al. 2017). This strategy increases AMP serum
et al. 2017), and albumin-binding moieties, lead- stability and was able to improve their antimicro-
ing to identical effects as when conjugated with bial efficacy against a broad range of drug-­
large proteins (Henninot et al. 2018; Lau and resistant bacteria in comparison with its free form
Dunn 2018) (Fig. 15.4). (Rajchakit and Sarojini 2017).
NPs can also offer the possibility to selec-
tively target AMPs against bacteria (Reinhardt
15.4.2 Bioengineering Strategies and Neundorf 2016) and to focus AMP delivery
to Improve Pharmacokinetics onto specific sites, minimizing side-effects and
and Pharmacodynamics increasing efficacy (Eckert et al. 2014).
Specifically targeted antimicrobial peptides
Bioengineering approaches, such as AMP encap- (STAMP), containing a targeting peptide coupled
sulation for local/targeted delivery, have been to the AMP, were also proposed to inhibit the
investigated to improve their stability and reduce growing pathogenic bacteria in a selective way.
systemic toxicity. Nanoparticles (NPs) prepared STAMP developed by Eckert RH and co-workers
using biodegradable materials such as natural (Eckert et al. 2014) demonstrated increased kill-
and synthetic polymers (e.g. chitosan NP ing potency against targeted bacteria in mucosal
(Almaaytah et al. 2017) and PLGA-NP (d’Angelo infections, preserving the normal microflora on
et al. 2015), respectively) and lipids (Carmona-­ the affected region.
Ribeiro and Carrasco 2014) have been used for AMP attachment onto magnetic NPs was also
AMP encapsulation due to their high degree of proposed as a new nanotechnology strategy to
biocompatibility. Moreover, AMP-conjugated increase AMP efficacy, since magnetic NPs can
15 Clinical Application of AMPs 293

be directed to the infection site using an external synthesis, as well as difficulties faced in its scale-
magnetic field (Blin et al. 2011). ­up. It is true that multi-gram production of pep-
Antimicrobial dendrimer peptides, prepared tides has been a challenge for many years, but the
by multiple peptide sequences linked to an inner widely recognized benefits of peptide-based ther-
branched core, represent a different approach to apeutics have been contributing to a slow, but
potentiate the antimicrobial effect of the AMP steady, paradigm shift in the way Big Pharma
(decreasing their susceptibility to proteases due to looks at peptides as drugs. Moreover, peptide
steric hindrance) and decrease their haemolytic synthesis costs represent a small fraction of the
activity (Scorciapino et al. 2017). Antimicrobial overall costs involved in the development of a
dendrimeric peptides, designed by the attachment novel active pharmaceutical ingredient (API)
of two to eight copies of a tetrapeptide (RLYR) or from bench to the clinics (Otvos and Wade 2014).
an octapeptide (RLYRKVYG) around a core of As such, we have been ‘watching peptide drugs
lysine residues, have demonstrated very high anti- go up’ (Marx 2005), accompanied by consider-
microbial efficacy against several bacteria and able progress in chemical and biotechnological
fungi in high- and low-salt conditions (Tam et al. approaches towards peptide large-scale produc-
2002). Other types of antimicrobial dendrimeric tion (Marx 2005; Thayer 2011). Large-scale syn-
peptides have been proposed by grafting unnatu- thesis of peptides is now possible by means of
ral peptides or using organic groups as the branch- both solid- (SPPS) and liquid-phase (LPPS) pep-
ing core (Scorciapino et al. 2017). tide synthesis methods (Fig. 15.5) (Marx 2005;
An innovative strategy was developed and pat- Thayer 2011). The most popular example of a
ented by Collins et al. (2015) that uses engineered peptide-based drug produced in multi-tons is the
bacteriophages to express AMP and/or antimi- T-20 peptide (Fuzeon, Roche), which is the first
crobial polypeptides (lytic enzymes) in a broad HIV fusion-inhibitor peptide produced at such
spectrum of host bacteria. Expressed AMP will scale by a combination of solid- and solution-­
be secreted from host bacteria or alternatively, phase methodologies. The decision of Roche to
released after bacteria lysis, improving bacterial go fully synthetic, instead of doing this 26-mer
infection treatment. peptide recombinantly, and the following mas-
Targeted bioengineering approaches have been sive investment, lowered the cost of all synthesis
proposed to prevent implant-associated infections reagents, making industrial peptide synthesis
(IAI), namely, biofilm formation by AMP immo- more feasible and affordable.
bilization onto biomaterial coatings (Costa et al. One of the cheapest alternatives to synthetic
2011, 2015; Willcox et al. 2013; Rai et al. 2016a, peptides is production by recombinant expres-
b). These strategies imply covalent binding of the sion methods using microorganisms for which
AMP directly, through PEG spacers, or even at the AMP produced is innocuous. For instance,
polymeric brushes onto the surface of biomateri- plectasin, the first fungal defensin whose interest-
als. Surfaces with AMP-immobilized coatings ing therapeutic potential was unveiled, can be
were capable to decrease bacterial adhesion and/ efficiently produced in high yields, via a fungal
or affect viability of adherent bacteria. AMPs that expression system that is currently used by
were successfully covalently bound to a material Novozymes for industrial-scale production of
surface and the chemical methods used for their proteins (Mygind et al. 2005). Still, one cannot
immobilization were described in the following underestimate the technical difficulties in pro-
patent application (Willcox et al. 2013). ducing large amounts of peptides by recombinant
expression: at the current production efficiency,
for each 100 kg of recombinant peptide, 1 million
15.4.3 Dealing with Peptide litres of fermentation mixture will be necessary.
Production Costs Nonetheless, it is foreseen that peptide produc-
tion costs through this method can still be reduced
One of the arguments disfavouring peptide-based by two orders of magnitude; hence, as with many
drugs has been the alleged high cost of peptide protein production methods, large-scale produc-
294 F. Costa et al.

Fig. 15.5 Modern peptide large-scale production facility (https://www.almacgroup.com/api-chemical-development/

highly-potent-peptide-manufacture/)

tion of AMPs by recombinant means could be an optimization of AMPs’ performance in physio-

economically viable and promising alternative. logical conditions.
In this connection, biotech companies are Researchers and early biotech companies
emerging which are investing on large-scale were also caught by the intricate regulatory envi-
peptide production at highly competitive prices ronment that encases the entry of AMPs into
(Scott 2018). clinical trials. Even when this first obstacle is sur-
passed, it is important to highlight that each clini-
cal trial must be precise on the application/
15.4.4 Impact of Regulatory pathological condition, focused population and
Impediments and Future route of administration. In other words, for a
Directions given AMP to be successful in completing the
full clinical trial path, it has to undergo through a
Unlike the exuberance accompanying forays into long and expensive iterative process. This
AMP development a decade ago, current opti- explains why this is a difficult path that drains
mism is tempered by deeper recognition of the newly formed biotech firms, even when they
challenges in translating R&D leading to a new manage to partner with established Big Pharma
class of products (Fox 2013). company. Fortunately, this paradigm seems to be
In the early years of AMP discovery, research reaching a point of renewal: the ‘post-antibiotic
focused on testing for activity in simplified era’ is demanding for new solutions to enter the
in vitro settings with little or no concern about clinics in a near future; therefore, authorities are
the legal prerequisites of an actual preclinical now more attentive and receptive to new thera-
study. Therefore, many promising studies have peutics. The ever-evolving regulations are now
proceeded possibly too fast, without a full pre-­ being directed to create a clearer path for promot-
15 Clinical Application of AMPs 295

ers. It is possible that, in the near future, cases Brooks BD, Brooks AE (2014) Therapeutic strategies
to combat antibiotic resistance. Adv Drug Deliv Rev
such as that of pexiganan, which did not receive 78:14–27
positive ruling by the FDA for the treatment of Cardoso MH, Cândido ES, Oshiro KGN, Rezende SB,
infected diabetic foot ulcers simply because it Franco OL (2018) Peptides containing D-amino
was not superior to oral ofloxacin, would have a acids and retro-inverso peptides: general applica-
tions and special focus on antimicrobial peptides.
different outcome in the present scenario. In: Koutsopoulos S (ed) Peptide applications in
Moreover, it is also important to highlight that biomedicine, biotechnology and bioengineering.
some AMPs have a synergistic effect when Woodhead Publishing, Elsevier Ltd, Kidlington,
applied in combination with standard antibiotics; pp 131–155
Carmona-Ribeiro AM, Carrasco LDD (2014) Novel for-
therefore, they could be used as part of multidrug mulations for antimicrobial peptides. Int J Mol Sci
approaches to enhance the rate of disease eradi- 15(10):18040–18083
cation and prevent the development of pathogen Chatterjee J, Rechenmacher F, Kessler H (2013)
resistance to conventional therapy. N-methylation of peptides and proteins: an important
element for modulating biological functions. Angew
Altogether, the future of AMPs as a novel and Chem Int Ed 52(1):254–269
viable class of antibiotics must be regarded with Cheneval O, Schroeder CI, Durek T, Walsh P, Huang YH,
optimism. Liras S, Price DA, Craik DJ (2014) Fmoc-based syn-
thesis of disulfide-rich cyclic peptides. J Org Chem
79(12):5538–5544
Chicharro C, Granata C, Lozano R, Andreu D, Rivas L
References (2001) N-terminal fatty acid substitution increases
the leishmanicidal activity of CA(1-7)M(2-9), a
Afacan NJ, Yeung AT, Pena OM, Hancock RE (2012) cecropin-melittin hybrid peptide. Antimicrob Agents
Therapeutic potential of host defense peptides in Chemother 45(9):2441–2449
antibiotic-resistant infections. Curr Pharm Des Chu Q, Moellering RE, Hilinski GJ, Kim YW,
18(6):807–819 Grossmann TN, Yeh JTH, Verdine GL (2015) Towards
Ahn HS, Cho W, Kim JM, Joshi BP, Park JW, Lohani CR, understanding cell penetration by stapled peptides.
­
Cho H, Lee KH (2008) Design and synthesis of cyclic Medchemcomm 6(1):111–119
disulfide-bonded antibacterial peptides on the basis of ClinicalTrials.gov (n.d.) https://www.clinicaltrials.gov/
the alpha helical domain of Tenecin 1, an insect defen- ClinicalTrialsPage (n.d.) https://www.nia.nih.gov/health/
sin. Bioorg Med Chem 16(7):4127–4137 what-are-clinical-trials-and-studies. Accessed Jul
Almaaytah A, Mohammed GK, Abualhaijaa A, Al-Balas 2018.
Q (2017) Development of novel ultrashort antimicro- Coates ARM, Halls G, Hu YM (2011) Novel classes of
bial peptide nanoparticles with potent antimicrobial antibiotics or more of the same? Br J Pharmacol
and antibiofilm activities against multidrug-resistant 163(1):184–194
bacteria. Drug Des Dev Ther 11:3159–3170 Cochrane SA, Findlay B, Bakhtiary A, Acedo JZ,
Alminana N, Alsina MA, Ortiz A, Reig F (2004) Rodriguez-Lopez EM, Mercier P, Vederas JC (2016)
Comparative physicochemical study of SIKVAV pep- Antimicrobial lipopeptide tridecaptin A(1) selectively
tide and its retro and retro-enantio analogues. Colloids binds to gram-negative lipid II. Proc Natl Acad Sci U
Surf A: Physicochem Eng Asp 249(1–3):19–24 S A 113(41):11561–11566
Arias M, Piga KB, Hyndman ME, Vogel HJ (2018) Collins JJ, Koeris M, Lu TKT, et al. (2015) Bacteriophages
Improving the activity of Trp-rich antimicrobial pep- expressing antimicrobial peptides and uses thereof.
tides by Arg/Lys substitutions and changing the length US 2015/0050717 A1
of cationic residues. Biomolecules 8(2):1–17 Costa F, Carvalho IF, Montelaro RC, Gomes P, Martins
Bednarska NG, Wren BW, Willcocks SJ (2017) The MCL (2011) Covalent immobilization of antimicro-
importance of the glycosylation of antimicrobial pep- bial peptides (AMPs) onto biomaterial surfaces. Acta
tides: natural and synthetic approaches. Drug Discov Biomater 7(4):1431–1440
Today 22(6):919–926 Costa FMTA, Maia SR, Gomes PAC, Martins MCL
Blin T, Purohit V, Leprince J, Jouenne T, Glinel K (2011) (2015) Dhvar5 antimicrobial peptide (AMP) che-
Bactericidal microparticles decorated by an antimicro- moselective covalent immobilization results on higher
bial peptide for the easy disinfection of sensitive aque- antiadherence effect than simple physical adsorption.
ous solutions. Biomacromolecules 12(4):1259–1264 Biomaterials 52:531–538
Bourne DG, Jones GJ, Blakeley RL, Jones A, Negri AP, d’Angelo I, Casciaro B, Miro A, Quaglia F, Mangoni ML,
Riddles P (1996) Enzymatic pathway for the bacte- Ungaro F (2015) Overcoming barriers in Pseudomonas
rial degradation of the cyanobacterial cyclic pep- aeruginosa lung infections: engineered nanoparticles
tide toxin microcystin LR. Appl Environ Microbiol for local delivery of a cationic antimicrobial peptide.
62(11):4086–4094 Colloids Surf B:Biointerfaces 135:717–725
296 F. Costa et al.

Davies JS (2003) The cyclization of peptides and depsip- Greber KE, Dawgul M (2017) Antimicrobial pep-
eptides. J Pept Sci 9(8):471–501 tides under clinical trials. Curr Top Med Chem
Di L (2015) Strategic approaches to optimizing peptide 17(5):620–628
ADME properties. AAPS J 17(1):134–143 Guiotto A, Pozzobon M, Canevari M, Manganelli R,
Di Pisa M, Chassaing G, Swiecicki JM (2015) When Scarin M, Veronese FM (2003) PEGylation of the anti-
cationic cell-penetrating peptides meet hydrocar- microbial peptide nisin a: problems and perspectives.
bons to enhance in-cell cargo delivery. J Pept Sci Farmaco 58(1):45–50
21(5):356–369 Hamley IW (2014) PEG-peptide conjugates.
Doak BC, Over B, Giordanetto F, Kihlberg J (2014) Biomacromolecules 15(5):1543–1559
Oral druggable space beyond the rule of 5: insights Hancock RE, Sahl HG (2006) Antimicrobial and host-­
from drugs and clinical candidates. Chem Biol defense peptides as new anti-infective therapeutic
21(9):1115–1142 strategies. Nat Biotechnol 24(12):1551–1557
Doherty DH, Rosendahl MS, Smith DJ, Hughes JM, Henninot A, Collins JC, Nuss JM (2018) The current state
Chlipala EA, Cox GN (2005) Site-specific PEGylation of peptide drug discovery: back to the future? J Med
of engineered cysteine analogues of recombinant Chem 61(4):1382–1414
human granulocyte-macrophage colony-stimulating Hoffmann R, Berthold N, Nollmann F (2014). Modified
factor. Bioconjug Chem 16(5):1291–1298 antibiotic peptides having variable systemic release.
Doores KJ, Gamblin DP, Davis BG (2006) Exploring and US20140309161 A1. 2014
exploiting the therapeutic potential of glycoconju- Imura Y, Nishida M, Ogawa Y, Takakura Y, Matsuzaki K
gates. Chem Eur J 12(3):656–665 (2007) Action mechanism of tachyplesin I and effects
DrugDataBase (n.d.) http://www.drugsdb.eu/index. of PEGylation. Biochim Biophys Acta-Biomembr
php?l=c. Accessed Jul 2018. 1768(5):1160–1169
Eckert RH, Yarbrough D, Shi W, et al (2014) Selectively Joo SH (2012) Cyclic peptides as therapeutic agents and
targeted antimicrobial peptides and the use thereof. biochemical tools. Biomol Ther 20(1):19–26
EP2801368A1 2014 Kosikowska P, Lesner A (2016) Antimicrobial peptides
Falanga A, Lombardi L, Franci G, Vitiello M, Iovene MR, (AMPs) as drug candidates: a patent review (2003-­
Morelli G, Galdiero M, Galdiero S (2016) Marine 2015). Expert Opin Ther Pat 26(6):689–702
antimicrobial peptides: nature provides templates for Lambert JN, Mitchell JP, Roberts KD (2001) The syn-
the design of novel compounds against pathogenic thesis of cyclic peptides. J Chem Soc-Perkin Trans
bacteria. Int J Mol Sci 17(5):1–17 1(5):471–484
Falanga A, Nigro E, De Biasi MG, Daniele A, Morelli Lau JL, Dunn MK (2018) Therapeutic peptides:
G, Galdiero S, Scudiero O (2017) Cyclic peptides as historical perspectives, current development
novel therapeutic microbicides: engineering of human trends, and future directions. Bioorg Med Chem
defensin mimetics. Molecules 22(7):1–15 26(10):2700–2707
Fang Y, Xue JX, Gao S, Lu AQ, Yang DJ, Jiang H, He Lele DS, Talat S, Kumari S, Srivastava N, Kaur KJ
Y, Shi K (2017) Cleavable PEGylation: a strategy (2015) Understanding the importance of glycosyl-
for overcoming the “PEG dilemma” in efficient drug ated threonine and stereospecific action of Drosocin,
delivery. Drug Deliv 24(2):22–32 a proline rich antimicrobial peptide. Eur J Med Chem
Fernandes P, Martens E (2017) Antibiotics in late clinical 92:637–647
development. Biochem Pharmacol 133:152–163 Lundquist P, Artursson P (2016) Oral absorption of pep-
Fishburn CS (2008) The pharmacology of PEGylation: tides and nanoparticles across the human intestine:
balancing PD with PK to generate novel therapeutics. opportunities, limitations and studies in human tis-
J Pharm Sci 97(10):4167–4183 sues. Adv Drug Deliv Rev 106:256–276
Fox JL (2013) Antimicrobial peptides stage a Mahlapuu M, Hakansson J, Ringstad L, Bjorn C (2016)
comeback(vol 31, pg 379, 2013). Nat Biotechnol Antimicrobial peptides: an emerging category of ther-
31(12):1066–1066 apeutic agents. Front Cell Infect Microbiol 6:1–12
Giuliani A, Pirri G, Nicoletto SF (2007) Antimicrobial Marti-Centelles V, Pandey MD, Burguete MI, Luis SV
peptides: an overview of a promising class of thera- (2015) Macrocyclization reactions: the importance of
peutics. Cent Eur J Biol 2(1):1–33 conformational, configurational, and template-induced
Gomes A, Teixeira C, Ferraz R, Prudencio C, Gomes preorganization. Chem Rev 115(16):8736–8834
P (2017) Wound-healing peptides for treatment of Martinez JL (2014) General principles of antibiotic
chronic diabetic foot ulcers and other infected skin resistance in bacteria. Drug Discov Today Technol
injuries. Molecules 22(10):2–18 11:33–39
Gong Y, Andina D, Nahar S, Leroux JC, Gauthier Marx V (2005) Watching peptide drugs grow up. Chem
MA (2017) Releasable and traceless PEGylation Eng News 83(11):17–84
of arginine-rich antimicrobial peptides. Chem Sci Melo MN, Castanho MARB (2007) Omiganan interac-
8(5):4082–4086 tion with bacterial membranes and cell wall models.
Gordon YJ, Romanowski EG, McDermott AM (2005) A Assigning a biological role to saturation. Biochim
review of antimicrobial peptides and their therapeu- Biophys Acta-Biomembr 1768(5):1277–1290
tic potential as anti-infective drugs. Curr Eye Res Merrifield RB, Juvvadi P, Andreu D, Ubach J, Boman A,
30(7):505–515 Boman HG (1995) Retro and retroenantio analogs of
15 Clinical Application of AMPs 297

cecropin-melittin hybrids. Proc Natl Acad Sci U S A Reinhardt A, Neundorf I (2016) Design and application
92(8):3449–3453 of antimicrobial peptide conjugates. Int J Mol Sci
Migon D, Neubauer D, Kamysz W (2018) Hydrocarbon 17(5):1–21
stapled antimicrobial peptides. Protein J 37(1):2–12 Roberts MJ, Bentley MD, Harris JM (2002) Chemistry for
Molchanova N, Hansen PR, Franzyk H (2017) Advances peptide and protein PEGylation. Adv Drug Deliv Rev
in development of antimicrobial peptidomimetics as 54(4):459–476
potential drugs. Molecules 22(9):1–60 Sader HS, Fedler KA, Rennie RP, Stevens S, Jones RN
Monaim SAHA, Jad YE, El-Faham A, de la Torre BG, (2004) Omiganan pentahydrochloride (MBI 226),
Albericio F (2018) Teixobactin as a scaffold for unlim- a topical 12-amino-acid cationic peptide: spectrum
ited new antimicrobial peptides: SAR study. Bioorg of antimicrobial activity and measurements of bac-
Med Chem 26(10):2788–2796 tericidal activity. Antimicrob Agents Chemother
Moradi SV, Hussein WM, Varamini P, Simerska P, Toth I 48(8):3112–3118
(2016) Glycosylation, an effective synthetic strategy Santajit S, Indrawattana N (2016) Mechanisms of antimi-
to improve the bioavailability of therapeutic peptides. crobial resistance in ESKAPE pathogens. Biomed Res
Chem Sci 7(4):2492–2500 Int 2016(2475067):1–8
Morris CJ, Beck K, Fox MA, Ulaeto D, Clark GC, Schmidt EGW, Hvam ML, Antunes F, Cameron J, Viuff
Gumbleton M (2012) Pegylation of antimicrobial pep- D, Andersen B, Kristensen NN, Howard KA (2017)
tides maintains the active peptide conformation, model Direct demonstration of a neonatal Fc receptor (FcRn)-
membrane interactions, and antimicrobial activity driven endosomal sorting pathway for cellular recy-
while improving lung tissue biocompatibility follow- cling of albumin. J Biol Chem 292(32):13312–13322
ing airway delivery. Antimicrob Agents Chemother Scorciapino MA, Serra I, Manzo G, Rinaldi AC (2017)
56(6):3298–3308 Antimicrobial dendrimeric peptides: structure, activ-
Mygind PH, Fischer RL, Schnorr KM, Hansen MT, ity and new therapeutic applications. Int J Mol Sci
Sonksen CP, Ludvigsen S, Raventos D, Buskov S, 18(3):1–13
Christensen B, De Maria L, Taboureau O, Yaver D, Scott A (2018) Peptide makers face threat from biotech-
Elvig-Jorgensen SG, Sorensen MV, Christensen BE, nology how a 1-year-old German start-up could be
Kjaerulff S, Frimodt-Moller N, Lehrer RI, Zasloff M, about to shake up a $1 billion market. Chem Eng News
Kristensen HH (2005) Plectasin is a peptide antibiotic 96(5):4–4
with therapeutic potential from a saprophytic fungus. Sierra JM, Fuste E, Rabanal F, Vinuesa T, Vinas M (2017)
Nature 437(7061):975–980 An overview of antimicrobial peptides and the latest
Oh D, Shirazi AN, Northup K, Sullivan B, Tiwari RK, advances in their development. Expert Opin Biol Ther
Bisoffi M, Parang K (2014) Enhanced cellular uptake 17(6):663–676
of short polyarginine peptides through fatty acylation Silva T, Magalhaes B, Maia S, Gomes P, Nazmi K,
and cyclization. Mol Pharm 11(8):2845–2854 Bolscher JGM, Rodrigues PN, Bastos M, Gomes MS
Otvos L, Wade JD (2014) Current challenges in peptide-­ (2014) Killing of mycobacterium avium by lactofer-
based drug discovery. Front Chem 2:1–4 ricin peptides: improved activity of arginine- and
Pfalzgraff A, Brandenburg K, Weindl G (2018) D-amino-acid-containing molecules. Antimicrob
Antimicrobial peptides and their therapeutic poten- Agents Chemother 58(6):3461–3467
tial for bacterial skin infections and wounds. Front Sockolosky JT, Szoka FC (2015) The neonatal Fc recep-
Pharmacol 9:1–23 tor, FcRn, as a target for drug delivery and therapy.
Qvit N, Rubin SJS, Urban TJ, Mochly-Rosen D, Gross Adv Drug Deliv Rev 91:109–124
ER (2017) Peptidomimetic therapeutics: scientific Swiecicki JM, Di Pisa M, Lippi F, Chwetzoff S, Mansuy
approaches and opportunities. Drug Discov Today C, Trugnan G, Chassaing G, Lavielle S, Burlina F
22(2):454–462 (2015) Unsaturated acyl chains dramatically enhanced
Rader AFB, Reichart F, Weinmuller M, Kessler H (2018) cellular uptake by direct translocation of a mini-
Improving oral bioavailability of cyclic peptides by malist oligo-arginine lipopeptide. Chem Commun
N-methylation. Bioorg Med Chem 26(10):2766–2773 51(78):14656–14659
Rai A, Pinto S, Evangelista MB, Gil H, Kallip S, Ferreira Tam JP, Lu YA, Yang JL (2002) Antimicrobial dendri-
MGS, Ferreira L (2016a) High-density antimicrobial meric peptides. Eur J Biochem 269(3):923–932
peptide coating with broad activity and low cytotoxic- Tapeinou A, Matsoukas MT, Simal C, Tselios T (2015)
ity against human cells. Acta Biomater 33:64–77 Cyclic peptides on a merry-go-round; towards drug
Rai A, Pinto S, Velho TR, Ferreira AF, Moita C, Trivedi design. Biopolymers 104(5):453–461
U, Evangelista M, Comune M, Rumbaugh KP, Simoes Teixeira V, Feio MJ, Rivas L, De la Torre BG, Andreu
PN, Moita L, Ferreira L (2016b) One-step synthesis D, Coutinho A, Bastos M (2010) Influence of lysine
of high-density peptide-conjugated gold nanoparticles N-epsilon-trimethylation and lipid composition on
with antimicrobial efficacy in a systemic infection the membrane activity of the cecropin A-melittin
model. Biomaterials 85:99–110 hybrid peptide CA(1-7)M(2-9). J Phys Chem B
Rajchakit U, Sarojini V (2017) Recent developments in 114(49):16198–16208
antimicrobial-peptide-conjugated gold nanoparticles. Thayer AM (2011) Making peptides at large scale. Chem
Bioconjug Chem 28(11):2673–2686 Eng News 89(22):21–25
298 F. Costa et al.

Trier S, Linderoth L, Bjerregaard S, Andresen TL, Rahbek White CJ, Yudin AK (2011) Contemporary strategies for
UL (2014) Acylation of glucagon-like peptide-2: peptide macrocyclization. Nat Chem 3(7):509–524
interaction with lipid membranes and in vitro intesti- WHO (2018) http://www.who.int/antimicrobial-resistance/
nal permeability. PLoS One 9(10):1–10 global-action-plan/en/.
Usmani SS, Bedi G, Samuel JS, Singh S, Kalra S, Kumar Willcox MDP, Kumar N, Cole N, et al. (2013)
P, Ahuja AA, Sharma M, Gautam A, Raghava GPS Antimicrobial peptides and uses thereof.
(2017) THPdb: database of FDA-approved peptide WO2013076666 A1. 2013
and protein therapeutics. PLoS One 12(7):1–12 WIPO (2018) https://patentscope.wipo.int/search/pt/
Veronese FM (2001) Peptide and protein PEGylation: search.jsf.
a review of problems and solutions. Biomaterials Wu YD, Gellman S (2008) Peptidomimetics. Acc Chem
22(5):405–417 Res 41(10):1231–1232
Vidal L, Geffard M (2014) Lauryl-poly-L-lysine: a new Yang J, Chen H, Vlahov IR, Cheng JX, Low PS (2006)
antimicrobial agent? FEBS J 281:328–328 Evaluation of disulfide reduction during receptor-­
Walensky LD, Bird GH (2014) Hydrocarbon-stapled pep- mediated endocytosis by using FRET imaging. Proc
tides: principles, practice, and progress. J Med Chem Natl Acad Sci U S A 103(37):13872–13877
57(15):6275–6288 Zasloff M (2002) Antimicrobial peptides of multicellular
Wang G, Mishra B, Lau K, Lushnikova T, Golla R, organisms. Nature 415(6870):389–395
Wang X (2015) Antimicrobial peptides in 2014. Zong JY, Cobb SL, Cameron NR (2017) Peptide-­
Pharmaceuticals (Basel) 8(1):123–150 functionalized gold nanoparticles: versatile bioma-
WebPage, P (n.d.) http://www.polyphor.com/pol7080/. terials for diagnostic and therapeutic applications.
Accessed Jul 2018. Biomater Sci 5(5):872–886
 Index

A lactoferricin-derived peptides and lipopeptides, 262

Absorption, distribution, metabolism, excretion and mechanisms of action, 259–262
toxicity (ADMET), 284 microarray technology, 262
Acidic phospholipids, 10 minimal inhibitory concentration, 260
Acute myelogenous leukemia, 132 QS cell communication, 262
Alamethicin, 36 Antibiofilm coating, cathelicidin LL-37, 226–228
Alexa-Fluor trypsin inhibitor (AF-SBTI), 24 Antibiofilm efficacy in animal models, cathelicidin
ALL-38, human reproductive system, 218 LL-37, 216
α-defensin, 151, 153 Antibiofilm peptides, cathelicidin LL-37, 226
Amino acid sequences, AMP, 35 Antibiotic lock therapy (ALT), 265, 269–271
Amino acid substitution/modification, 142 Antibiotic ofloxacin, 285
AMP binding to cells Antibiotic-resistant microorganisms, 281, 282
activity and toxicity, cell density dependent, 198–199 Antibiotics, human health and well-being, 241
aggregation, aqueous phase, 202 Anticancer peptides (ACP)
antimicrobial and hemolytic activity in assays, advantages of, 133
200, 201 cathelicidin LL-37, 232
bacterial cells, higher affinity, 198 colon cancers, 232
cell binding in co-culture experiments, kinetic gastric cancer, 232
phenomena, 199–201 cationic amphipathic peptides, 137
AMP-inspired systems, 200 chemotherapy functions, 131–132
Anionic lipid clustering model, 65–66 conventional cytotoxic drugs, 143
bacterial species specificity, 66–67 delivery
limitation of nanoparticle, 140
access to membrane, 68 oncolytic virus therapy, 141
bacterial membrane lipids, 67 peptide modification strategies, 140
conformational flexibility, 67 peptides with altered stereochemistry, 141
hydrophobicity, 67–68 direct-acting vs. indirect-acting, 133
non-membrane targeting, 69 cancer cell membranes, selective peptide binding
phospholipid acyl chains, 68–69 to, 135–136
pore formation, 68 cell lysis, 135
specific binding to component, 68 mechanism of ACP-mediated anticancer activity,
molecular mechanism 136–137
depletion of anionic lipids, 69–70 hematopoietic stem cell transplantation, 132
phase boundary defect, 69 host defense peptides, 132
phase separation in redistribution of membrane immunogenic cell death, 138–139
proteins, 70 limitations to, 133–134
potency related to charge, 66 oncolytic virus therapy, 139
Anti-biofilm agents selectivity for cancer cells
biological activity, 263 amino acid substitution/modification, 142
combination strategies, 265–269 histidine substitution, 143
EPS biosynthesis, 262 peptide-targeting motifs, 141–142
gene regulation, 262 targeted therapies, 132
human cathelicidin peptide LL-37, 260, 261 Antifungal peptides, cathelicidin LL-37, 230–231

© Springer Nature Singapore Pte Ltd. 2019 299

K. Matsuzaki (ed.), Antimicrobial Peptides, Advances in Experimental Medicine and Biology 1117,
https://doi.org/10.1007/978-981-13-3588-4
300 Index

Antigen-presenting cells (APCs), 153, 154 anticancer peptides, 232

Anti-HIV peptide, enfuvirtide, 242 antifungal peptides, 230–231
Anti-infective peptide drugs, 242 antiviral peptides, 228–230
Anti-infective therapy, 157 covalent immobilization, 227
Antimicrobial dendrimer peptides, 293 immune modulating peptides, 231–232
Antimicrobial peptide database (APD), 216 bacterial membranes, physical and structural basis
Antimicrobial peptides (AMPs), 14 bacterial recognition, electrostatic interactions,
bioinformatics and library screenings, 102–104 218–219
biological lipid membranes structural basis, 220–221
bacterial cell membranes, 98–99 structural studies, 220
eukaryotic cell membranes, 96–98 biofilm formation, 223, 234
tumor cells, 99–100 broad-spectrum activity, micromolar
cationic, hydrophobic and amphipathic, 94 concentration, 216
direct entry, 95 distribution, 216
drug resistance, 96 evolutional significance, 233
endocytosis, 95 gene product processing
functional and mechanistic redundancy of, 104–106 ALL-38, human reproductive system, 218
membrane-active peptides, 100–102 fat cells, uncharacterized alternative form, 218
membrane permeabilization mechanisms, 9, 11 mature peptide LL-37, 218
binding modes of, 10 TLN-58, diseased skin state, 218
detergent-like model, 13 immune stimuli, 216
interfacial activity model, 13 library screen and structure-based approaches, 233
toxicity of, 96 membrane anchoring, 233
Antimicrobial proteins, 17 minimal inhibitory concentration, 216
Anti-tumor cytotoxic T lymphocyte (CTL), 138 moonlighting peptide, functional roles, 216
Antiviral peptides, cathelicidin LL-37, 228–230 NMR studies, 234
pathogen degradation, 222
pathogen invasion/immune stimuli, vitamin D and
B sunlight, 216
Bacteria and drug resistance, 112, 113 peptide design, 223–226
Bacterial and host cells, membrane structure and plasma binding to SAAP-148, 234
composition, 185–187 skin processing, 221–222
Bacterial cell membranes, 98–99 therapeutic use, 234
Bacterial resistance mechanisms, 112 Cathelicidins, see Human Cathelicidin LL-37
Bactnecin-7 (Bac7), 76, 79–80, 86 Cathelin-like domain (CLD), 216
Barrel-stave channel model, 11, 12, 96 Cationic amphipathic peptides, 133, 137
β-defensins, 153, 154, 159, 161 Cationic antimicrobial peptides
Bioengineering strategies, pharmacokinetics and circular dichroism spectroscopy, 40–42
pharmacodynamics, 292–293 dimers of antimicrobial peptides, 53–54
Biofilm-forming organisms, life cycle, 258 electrophysiological recordings, 35–36
Biofilms, 242 fluorescence spectroscopy
Bombesin, 141 depth of membrane insertion, 37–38
fluorescence imaging, 39–40
fluorophore release, 36–37
C FRET, 39
Calcein, 12, 14, 19–22, 36, 37, 48, 54 natural chromophores and membrane
Canonical translation, 82 partitioning, 37
Cardiolipin, 65, 67, 98 self-quenching of fluorescent molecules, 38–39
Carpet model, 11, 43, 52, 96 isothermal titration calorimetry, 42–43
Cathelicidin LL-37 lipopeptide biosurfactants with antimicrobial
active regions identification in laboratories, 222–223 properties, 47–48
amino acid sequences and physical properties, molecular dynamics (MD) simulations, 46–47
216, 217 molecular shape concept, 49–51
antibacterial region discovery, 221–223 molecular shape of membrane constituents
antibiofilm efficacy in animal models, 216 adapts, 51
anti-infective potential, 235 SMART model, 51–52
antimicrobial activity and immune modulation, 234 solid-state NMR spectroscopy
applications of lipids, 45–46
antibiofilm coating, 226–228 polypeptides, investigation of, 43–44
antibiofilm peptides, 226 synergistic enhancement of activities, 54–55
Index 301

Cationic host defence peptides (CHDPs), 149–150, 162 Circular dichroism (CD) spectroscopy, 40–42, 102
expression and processing, 150–152 Clinical application of AMPs
in inflammation and immunity bioengineering strategies, 292–293
antimicrobial activities, 155, 156 chemical approaches, 287–292
autoimmune disease, 159–160 immunomodulatory properties, 287
cancer, 160–161 peptide production costs, 293–294
chronic inflammatory lung disease, 158 regulatory impediments, 294–295
IDR peptides, 156 Clinical trials, AMP, 282, 284–286
infectious disease, 156–157 Collisional quenching, 37
microbiome, 154–155 Combinatorial peptide libraries, SME, 247–248
pro-and anti-inflammatory role, 152–153 Commercialized AMP, 112, 114
in shaping adaptive immunity, 153–154 Computer-aided design, 241
skin disease, 158–159 Confocal laser scanning microscopy (CLSM), 18
route of administration, 162 Conformational and aggregation equilibria, AMP
Cecropins, 36, 40, 43, 216 selectivity
Cell-penetrating peptides (CPPs), 15, 18, 21, 25, imperfect amphipathicity, 195, 196
288–289 peptide aggregation, aqueous phase on activity and
bioinformatics and library screenings, 102–104 toxicity, 196–197
biological lipid membranes peptide helicity and toxicity, 194–195
bacterial cell membranes, 98–99 Covalent immobilization, cathelicidin LL-37, 227
eukaryotic cell membranes, 96–98 CPPs, see Cell-penetrating peptides (CPPs)
tumor cells, 99–100 Crohn’s disease, 8
cationic, hydrophobic and amphipathic, 94
direct entry, 95
endocytosis, 95 D
functional and mechanistic redundancy of, 104–106 Defensins, 150, 151, 216
membrane-active peptides, 100–102 Direct-acting anticancer peptides (DAAs), 133
Cell selectivity, AMPs cancer cell membranes, selective peptide binding to,
AMP aggregation, 176 135–136
anticancer peptides, 176 cell lysis, 135
antifungal activities, 176 mechanism of ACP-mediated anticancer activity,
antiviral activities, 176 136–137
cationic charges, 191 Drug delivery, 47, 96, 115, 140
cell density dependence, 199 Drug pharmacodynamics (PD), 284
clinical applications, 175 Drug-resistant bacterial infections, 242
conformational and aggregation equilibria, 193–195
design strategies, 176
excessive amphipathicity, toxicity, 193 E
hydrophobicity, 192–193 Enfuvirtide, anti-HIV peptide, 242
mechanism of action, 181 Epigallocatechin gallate (EGCg), 19
for microbial cells, 176–182 ESKAPE pathogens, 251–252
origin and the structural determinants, 176 Eukaryotic cell membranes, 96–98
peptide association, target and host cells, 176, 181 Experimental lock solutions, AMP, 269–271
peptide conformational equilibria, 176
physicochemical properties, 190–193
potential toxicity, 175 F
properties of, 175 Fat cells, uncharacterized alternative form, 218
structural determinants, 176 FDA-approved AMPs in clinical use, 282–284
in vitro assays, 176, 179, 181 Fluorescence spectroscopy
in vivo imaging, 179, 181 depth of membrane insertion, 37–38
Cellular membranes, 183–184 fluorescence imaging, 39–40
CHDPs, see Cationic host defence peptides (CHDPs) fluorophore release, 36–37
Chemical approaches, bioavailability FRET, 39
N-/C-terminal modifications, 290–292 natural chromophores and membrane
noncanonical amino acids, 289–290 partitioning, 37
peptide cyclization approaches, 287–289 self-quenching of fluorescent molecules, 38–39
proteinogenic amino acid replacement, 289 Förster Resonance Energy Transfer (FRET), 39, 54
Chondroitin sulfate (CS), 100 Frog skin secretions, 241
Chronic obstructive pulmonary disease (COPD), 158 Fusogenic liposomes, 140
302 Index

G membrane anchoring, 233

Gene product processing, cathelicidin LL-37 minimal inhibitory concentration, 216
ALL-38, human reproductive system, 218 moonlighting peptide, functional roles, 216
fat cells, uncharacterized alternative form, 218 NMR studies, 234
mature peptide LL-37, 218 pathogen degradation, 222
TLN-58, diseased skin state, 218 pathogen invasion/immune stimuli, vitamin D and
Giant plasma membrane vesicles (GPMVs), 101 sunlight, 216
Glycosaminoglycans (GAGs), 96 peptide design, 223–226
Good Clinical Practice (GCP), 284 plasma binding to SAAP-148, 234
Good manufacturing practice (GMP)-grade peptide, 134 skin processing, 221–222
Gouy-Chapman theory, 10 therapeutic use, 234
Green fluorescent protein (GFP), 14 Human LL37 peptide, 40
Hydrophobic and electrostatic driving forces,
nonadditive, 189–190
H Hydrophobicity, 67–68
hCAP-18, 150, 151, 159
Hematopoietic stem cell transplantation, 132
Heparan sulfate (HS), 96 I
Heterodetic cyclization, 288 IL-1 receptor antagonist (IL-1RA), 153
High-throughput screening for antibacterial activity, Immune modulating peptides, cathelicidin LL-37,
synthetic molecular evolution 231–232
broth sterilization, 249–250 Immunogenic cell death, 138–139
colony-forming units reduction, 250–251 Immunomodulatory effects of hCAP18 and LL37
cytotoxicity and hemolysis, 251 peptides, 285
radial diffusion, 248–249 Immunomodulatory short synthetic peptides, 156
Histidine substitution, 143 Implant-associated infections, prevention, 263
Histone deacetylase inhibitors (HDACi), 151 Indirect-acting anticancer peptides, 133
Host defense peptides (HDPs), 132 cancer cell membranes, selective peptide binding to,
Human cathelicidin LL-37, 150 135–136
active regions identification in laboratories, 222–223 cell lysis, 135
amino acid sequences and physical properties, mechanism of ACP-mediated anticancer activity,
216, 217 136–137
antibacterial region discovery, 221–223 Infrared (IR) spectroscopy, 102
antibiofilm efficacy in animal models, 216 Innate defense regulators (IDR), 137, 156, 157
anti-infective potential, 235 Innate immune defenses, 175
antimicrobial activity and immune modulation, 234 Insect hemolymph, 241
applications Ipilimumab, 139
antibiofilm coating, 226–228 Iseganan, 285
antibiofilm peptides, 226 Isothermal titration calorimetry (ITC), 42–43
anticancer peptides, 232 Iturin A, 48, 49
antifungal peptides, 230–231
antiviral peptides, 228–230
covalent immobilization, 227 L
immune modulating peptides, 231–232 Lactoferricin B (LfcinB), 18, 19, 21, 23, 133, 142
bacterial membranes, physical and structural basis Large-scale peptide manufacturing, 272
bacterial recognition, electrostatic interactions, Large unilamellar vesicles (LUVs), 18, 19, 25, 36, 39,
218–219 42, 101
structural basis, 220–221 Lewis lung carcinoma, 161
structural studies, 220 Lipid composition, AMP affinity for lipid bilayers
biofilm formation, 223, 234 cholesterol content, 188
broad-spectrum activity, micromolar concentration, 216 intrinsic curvature, 188–189
distribution, 216 membrane charge, 188
evolutional significance, 233 Lipid domain formation, 69
gene product processing Lipid-peptide interaction, 101
ALL-38, human reproductive system, 218 Lipid rafts, 97
fat cells, uncharacterized alternative form, 218 Lipopeptide biosurfactants, 47–48
mature peptide LL-37, 218 Lipopolysaccharides (LPS), 10, 14, 153
TLN-58, diseased skin state, 218 Lipoteichoic acids (LTA), 95, 98
immune stimuli, 216 Liquid-ordered domains, 97
library screen and structure-based approaches, 233 LPS-induced secretion of tumor necrosis factor α, 116
Index 303

M N
Magainins, 12–15, 18, 19, 23, 34, 35, 39, 46, 52, 216 Nanoparticle, 140
Mammalian neutrophil granules, 241 Neuropeptides, 4
Membrane curvature, 11, 51, 106 Neutrophil extracellular traps (NETs), 152, 160
Membrane macroscopic phase, 47 Nuclear magnetic resonance (NMR) spectroscopy, 102
Membrane permeabilization, 95
AMPs, 9
and bacterial death, 9, 10 O
bacterial membranes, 14 Oligo-acyl-lysine (OAK), 66–67
barrel-stave channel, 11, 12 Onc112, 79, 80
carpet model, 11 Oncolytic virus therapy, 139, 141
detergent-like model, 13
interfacial activity model, 13
leakage kinetics, 9 P
liposomes, 11 PAC-113, 285
magainin 2 leakage activity, 13 PATENTSCOPE of World Intellectual Property
mammalian cell membranes, 14–15 Organization (WIPO), 282
membrane binding, 9–10 Pathogen-associated molecular patterns (PAMPs), see
membrane curvature, 11 Synthetic anti-lipopolysaccharide peptides
mode of dye leakage, 13 (SALPs)
pore lifetime, 13 Pathogen-specific innate immune response pathways, 216
toroidal pore model, 11, 12 Pembrolizumab, 139
Membrane pore formation, 37 Penicillin, discovery, 215
Membrane topology, 43, 44 Peptaibols, 12
Metalnikowin, 77, 79 Peptide-cell association, quantitative determination, 176
Microbial adhesion and subsequent proliferation, Peptide cyclization approaches, 287–289
prevention, 263 Peptide design, Cathelicidin LL-37 peptides, 223–224
Microbial biofilms FK-13 and KR-12 templates, 225–226
adverse reactions, 259 FK-16-derived GF-17 template, 224–225
clonal growth and stable cell-cell interactions, 258 IG-24 derived P60.4, 224
development of, 258 Peptide-induced membrane permeability, 188
horizontal transfer of resistance and virulence Peptide-lipid interactions, 38, 44
genes, 259 Peptide-mediated cytotoxicity, 136
irreversible adhesion, 258 Peptide-membrane association, 188
maturation phase, 258 thermodynamics, 189–190
multiple drug resistance, 259 Peptide modification strategies, 140
outer membrane protein reorganization, 259 Peptide-targeting motifs, 141–142
prevention of, 263–264 Peptidyltransferase center (PTC), 77, 83, 85
quorum sensing systems, 259 Perfluorocarbon nanoparticles, 140
reversible attachment of planktonic cells to PGLa, 39, 41, 46, 55
surface, 258 Phagocytes, 75
stress response, 259 Pharmaceutical development, AMP, 115
Minimal inhibitory concentration (MIC), 14, 77 Phosphatidylcholine (POPC), 45, 48
Minimum bactericidal concentration (MBC), 14 Phosphatidylethanolamine, 49, 67, 70
Modern peptide large-scale production facility, 293, 294 Phospholipid acyl chains, 68–69
Molecular dynamics (MD) simulations, 46–47 Phospholipid and cholesterol content of human
Molecular mechanics force field, 46 erythrocyte membrane, 185, 187
Molecular shape concept, 49–51 Phospholipid composition of membranes, Gram-negative
Multicellular organisms, AMP of, 3–5 and Gram-positive bacteria, 185, 186
design of, 3 Plasma binding to SAAP-148, cathelicidin LL-37, 234
in Gram-negative bacteria, 3–4 Pleurocidins, 133, 137
in Gram-positive bacteria, 4 Polymicrobial infections, 282
hematopoietic cells, 4 Pore-forming toxin (PFT) protein, 19
mechanism, 4 Proline-rich AMPs (PrAMPs), protein synthesis
Paneth cells, 5 discovery of intracellular target of, 73–74, 76–77
simple amino acid substitutions, 4 mechanism of action of Class I and Class II, 81–83
Multidrug resistance, 215 mutagenesis studies, 84–86
multidrug-resistant (MDR) cancer cells, 133 nascent polypeptide exit tunnel, 77–81
Multiple interconnected equilibria modulation, peptide Class I, binding of, 77–80
activity and selectivity, 190, 191 Class II, binding of, 80–81
304 Index

Proline-rich AMPs (PrAMPs), protein synthesis (cont.) anti-inflammatory agents in vivo, 123
ribosomal binding site, determination of, 77 candidate therapies against sepsis, 120
sources of, 74–75 cell-bound toxins, 120
synthesis of, 75–76 development of, 115–117
toxicity in eukaryotes, 86–87 inflammation in intestinal cells, 118–119
uptake pathways of, 76 intracellular LPS responses, 123–124
Pro-rich peptide Bac7 (1–35), 15 isothermal calorimetric titration, 116, 117
Pseudomonas, 5 LPS-induced sepsis, TLR4, 123
Pyrrhocoricin, 76, 79 LPS neutralization in vivo, 119
mode of action, 120, 122
PAMPs from gram-positive bacteria in vivo, 119–121
Q sepsis-induced cardiac dysfunction, 117–118
Quantitative structure/activity relationships (QSAR) therapeutics in human “healing attempts,” 126
models, 103, 142 topical treatment of wounds and skin infections,
124–126
Synthetic CHDP-inspired peptide (SAAP-148), 159
R Synthetic molecular evolution (SME)
“Race for the surface” concept, 263 biochemical process, 243
Radiation therapy for head and neck cancer, 285 cell architecture, 243
Reactive oxygen species (ROS), 96, 99 design and synthesis, combinatorial peptide libraries,
Rheumatoid arthritis (RA), 159 247–248
ESKAPE pathogens, 251–252
high-throughput screening for antibacterial activity,
S 248–251
Secretory leukocyte proteinase inhibitor (SLPI), 158 host cell inhibition, 243–246
Single giant unilamellar vesicle (GUV) method impediments, 243, 244
advantages iterative screening, 247
elementary processes, separation of, 22–23 peptide antibiotic drugs, preclinical identification, 242
entry of AMPs without pore formation, 25 proteolytic degradation, 246–247
membrane permeation, 24–25 saturation-like process, 243
pore formation and local rupture, rate serum inhibition, 243–246
constant of, 23 toxicity against mammalian cells, 245–246
rate constant, determination of, 22–23 Systemic lupus erythematosus (SLE), 159, 160
transbilayer movement of lipids, 25–26
location of AMPs and pore formation, 26–28
membrane tension, effect of T
asymmetric lipid packing, effect of, 30 Targeted bioengineering approaches, 293
lipid bilayers, 28–30 Targeting sequences, 141
modes of action (MoAs) of AMPs, 19–22 Therapeutic index (TI)
complete rupture, 19 of natural and artificial AMPs, 176–178
EGCg, 19 RBCs/eukaryotic cells, 176, 180
LfcinB-induced membrane permeation in vivo studies, 181, 182
of calcein, 21 Thermodynamics, peptide-membrane association,
local rupture, 19 189–190
magainin 2-induced membrane permeation of TLN-58, diseased skin state, 218
calcein, 20 Toroidal pore model, 11, 12, 38, 96
no damage to the bilayers, 19, 21 Transportan 10 (TP10), 18, 19, 27, 29
Rh-LfcinB (4-9), 21, 22 Trial and error experimentation, 241
small pore formation, 19 Tumor cells
Small unilamellar vesicles (SUVs), 41, 42, 102 anticancer activity of membrane-active peptides,
Soft Membranes Adapt and Respond, also Transiently 99–100
(SMART) model, 38, 51–52 characterization, 99
Solid-state NMR spectroscopy Tur1A, 79–80, 86
of lipids, 45–46
polypeptides, investigation of, 43–44
Sphingomyelin (SM), 19 W
Stimulated emission depletion (STED) super-resolution WHO priority pathogens for research and development,
microscopy, 102 antibiotics, 112, 113
Structure/activity relationship (SAR), 142
Surfactin, 47, 48
Synthetic anti-lipopolysaccharide peptides (SALPs) Z
antibiotics in vivo, 120, 122 Zwitterionic lipids, 66, 67