Applied Soil Ecology 15 (2000) 25–36

Methods for assessing the composition and diversity of

soil microbial communities
G.T. Hill a,∗ , N.A. Mitkowski a , L. Aldrich-Wolfe b , L.R. Emele a , D.D. Jurkonie a , A. Ficke a ,
S. Maldonado-Ramirez a , S.T. Lynch a , E.B. Nelson a
aDepartment of Plant Pathology, Cornell University, Ithaca, NY 14853, USA
b Department of Ecology and Systematics, Cornell University, Ithaca, NY 14853, USA

Abstract
Soil microorganisms play important roles in soil quality and plant productivity. The development of effective methods for
studying the diversity, distribution, and behavior of microorganisms in soil habitats is essential for a broader understanding
of soil health. Traditionally, the analysis of soil microbial communities has relied on culturing techniques using a variety of
culture media designed to maximize the recovery of diverse microbial populations. However, only a small fraction (<0.1%)
of the soil microbial community has been accessible with this approach. To overcome these problems, other methods such
as the analysis of phospholipid fatty acids and community-level physiological profiles have been utilized in an attempt to
access a greater proportion of the soil microbial community. In recent years, molecular methods for soil microbial community
analysis have provided a new understanding of the phylogenetic diversity of microbial communities in soil. Among the most
useful of these methods are those in which small subunit rRNA genes are amplified from soil-extracted nucleic acids. Using
these techniques, it is possible to characterize and study soil microbes that currently cannot be cultured. Microbial rRNA
genes can be detected directly from soil samples and sequenced. These sequences can then be compared with those from
other known microorganisms. Additionally, group- and taxon-specific oligonucleotide probes can be developed from these
sequences making direct visualization of microorganisms in soil habitats possible. The use of these techniques provides
new ways of assessing soil microbial diversity and ultimately, a more complete understanding of the potential impacts of
environmental processes and human activities on responses of soil microorganisms. Information gained from such studies
will have direct impacts on our understanding of the role of microbial processes in soil health. © 2000 Published by Elsevier
Science B.V.
Keywords: Soil ecology; Molecular microbial ecology; Soil health; Soil quality; PLFA analysis; Community-level physiological profiles;
FISH; SSU rRNA; rDNA

1. Introduction ing of microbial properties such as biomass, activity,

and diversity are important to scientists in furthering
Microbial characteristics of soils are being evalu- knowledge of the factors contributing to soil health,
ated increasingly as sensitive indicators of soil health results of such analyses may also be useful to exten-
because of the clear relationships between microbial sion personnel and farmers in devising practical mea-
diversity, soil and plant quality, and ecosystem sus- sures of soil quality.
tainability (Doran et al., 1994). While the understand- Studies of soil microbial properties have been com-
monly conducted at the process level, where biomass,
∗ Corresponding author. respiration rates, and enzyme activities have been

0929-1393/00/$ – see front matter © 2000 Published by Elsevier Science B.V.

PII: S 0 9 2 9 - 1 3 9 3 ( 0 0 ) 0 0 0 6 9 - X
26 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

examined. Less attention has been given to community- 2. Culture-dependent methods of community
level or organism-level responses to changes in soil analysis
properties or management. Although these process-
level measurements provide an important understand- 2.1. Dilution plating and culturing methods
ing of gross microbial processes and their potential
role in soil health, they tell us little about quali- Traditionally, the analysis of soil microbial com-
tative community-level changes because any given munities has relied on culturing techniques using a
microbial process may be carried out by diverse taxa. variety of culture media designed to maximize the
Furthermore, these process-level measurements are recovery of different microbial species. This is par-
limited in their ability to describe a particular ticularly the case for soil health studies. There are
microbial ecosystem. numerous examples where these techniques have re-
Community-level microbial interactions are com- vealed a diversity of microorganisms associated with
plex, with individual species relying on the presence, various soil quality parameters such as disease sup-
function, and interaction of many other species. There- pression and organic matter decomposition (Tunlid
fore, quantitative and qualitative changes in the com- et al., 1989; Boehm et al., 1993, 1997; de Leij et al.,
position of soil microbial communities may serve as 1993; Workneh et al., 1993; Alvarez et al., 1995; Hu
important and sensitive indicators of both short and and van Bruggen, 1997; Maloney et al., 1997). Al-
long-term changes in soil health. The analysis of soil though there have been recent attempts to devise suites
microbial communities should involve not only deter- of culture media to maximize the recovery of diverse
minations of microbial biomass and diversity, but also microbial groups from soils (Balestra and Misaghi,
determinations of microbial growth, distribution, func- 1997; Mitsui et al., 1997), it has been estimated that
tion, and, if possible, the nature of interactions among less than 0.1% of the microorganisms found in typical
species. agricultural soils are culturable using current culture
Two of the longstanding challenges in soil mi- media formulations (Torsvik et al., 1990a; Atlas and
crobiology have been the development of effective Bartha, 1998). This is based on comparisons between
methods to (1) determine which microorganisms are direct microscopic counts of microbes in soil samples
present in soil and (2) determine microbial function and recoverable colony forming units.
in situ. These challenges have been exacerbated by
the difficulties of separating microorganisms from the
soil matrix and from plant tissues, the morphological 2.2. Community-level physiological profiles
similarities among many organisms found in soils,
and changing microbial taxonomies. Furthermore, the One of the more widely used culture-dependent
microscopic size of soil microorganisms has made methods for analyzing soil microbial communities
direct visualization more difficult than with macro- has been that of community-level physiological pro-
organisms. files (Garland and Mills, 1991; Winding, 1994; Zak
Over the past 10 years, the approach to analyzing et al., 1994; Konopka et al., 1998). This technique
soil microbial communities has changed dramatically. takes advantage of the traditional methods of bacterial
Many new methods and approaches are now avail- taxonomy in which bacterial species are identified
able, allowing soil microbiologists to gain access based on their utilization of different carbon sources.
to more of the microorganisms residing in soil and Community-level physiological profiles have been
allowing for better assessments of microbial diver- facilitated by the use of a commercial taxonomic
sity. In this review, we briefly discuss some of the system, known as the BIOLOG® system, which is
more important approaches for studying soil micro- currently available and has been used extensively for
bial communities, describing their strengths as well the analysis of soil microbial communities (Winding,
as their weaknesses. Our goal is to place the newer 1994; Lehman et al., 1995; Garland, 1996b).
culture-independent methods in perspective with the This BIOLOG® system is based on the utilization
traditional culture-based approaches for assessing of a suite of 95 different carbon sources that have
microbial diversity. been described previously (Garland and Mills, 1991).
 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36 27

Utilization of each substrate is detected by the reduc- function of the proportion of organisms present in the
tion of a tetrazolium dye, which results in a color sample which are able to utilize a particular substrate
change that can be quantified spectrophotometrically. (Garland, 1997). However, this may not be valid given
The pattern of substrates that are oxidized can be com- that some strains may utilize certain substrates more
pared among different soil samples from a series of efficiently than others in the guild, predominating in
times or locations as an indication of differences in the well and resulting in proportions of strains that
the physiological functions of microbial communities. differ from the original sample (Smalla et al., 1998).
Most commonly, multivariate statistical techniques are Furthermore, the ability of different taxa in the sam-
necessary to analyze the substrate utilization profile ple to utilize the same carbon sources is generally
data (Hackett and Griffiths, 1997; Hitzl et al., 1997). unknown.
For example, in these analyses, communities are con- A third problem is that the substrates found in com-
sidered to be functionally similar if the utilization pro- mercially available BIOLOG® plates are not necessar-
file of the 95 different carbon sources from one com- ily ecologically relevant and most likely do not reflect
munity clusters with that from another community. If the diversity of substrates found in the environment
the profiles segregate, communities would be consid- (Konopka et al., 1998). This is supported by the recent
ered functionally different. As such, community-level study of Campbell et al. (1997) in which plant root
physiological profiles can be useful in assessing gross exudate compounds were included as carbon sources
functional diversity (Zak et al., 1994; Garland, 1996b; in a functional analysis of nine upland grassland sites.
Campbell et al., 1997). The carbon sources most useful in differentiating the
There are a number of important considerations in different sites were predominantly these plant root ex-
the use of this method for community analysis. First, udates. All of these compounds had particularly low
the density of the initial inoculum must be standard- utilization rates suggesting they were utilized by or-
ized because it affects the rate at which color develops ganisms that were present in the soil in low numbers.
in the wells and thus the time at which color develop- Campbell et al. (1997) hypothesized that these com-
ment should be measured (Garland and Mills, 1991; pounds have greater differentiating ability both be-
Haack et al., 1995). Visible color will not develop cause they are biochemically more diverse and be-
within a well until the total number of cells able to uti- cause they select for the more slow growing organ-
lize that substrate reaches approximately 108 cells/ml isms that are usually present in the sample in smaller
(Haack et al., 1995). Because the number of cells numbers.
directly inoculated into the wells may be well be- While community level physiological profiles may
low 108 cells/ml, there can be a substantial lag phase provide information useful for assessments of soil
while the cells grow within the well. This may lead to microbial community diversity, the method still suf-
false negatives if wells are read too soon. Inaccurate fers from the same bias problems encountered with
physiological profiles may also result if samples are culture plating methods, making data interpretation
dominated by only a few species capable of growing problematic. Future work with ecologically meaning-
on particular substrates. Furthermore, the period of ful substrates (i.e. those that are likely to be found in
microbial growth within the well may also lead to soil habitats in nature) should make the method more
competition effects which again may bias the substrate appropriate for use with soil microbial communities.
utilization profile (Haack et al., 1995). Perhaps the Despite the fact that culture-dependent techniques are
best way to standardize inoculum levels is to employ not ideal for studies of the composition of natural mi-
vital stains combined with epifluorescence microscopy crobial communities when used alone, they provide
as a means to quantify actively-respiring cells (Gar- one of the more useful means of understanding the
land, 1996a). This way, a standard population of growth habit, development, and potential function of
metabolically active cells can be introduced into each microorganisms from soil habitats. A combination of
well. culture-based and culture-independent approaches is
A second methodological consideration is that an likely to reveal more complete information regard-
analysis of functional diversity is based on the assump- ing the composition of soil microbial communities
tion that color development in each well is solely a (Liesack et al., 1997).
28 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

Table 1
Common fatty acid signatures
Common bacterial signatures i15:0, a15:0, 15:0, 16:0, 16:1␻5, 16:1␻9, i17:0, a17:0, 17:0, 18:1␻7t, 18:1␻5, i19:0, a19:0
Aerobes 16:1␻7, 16:1␻7t, 18:1␻7t
Anaerobes cy17:0, cy19:0
Sulfate-reducing bacteria 10Me16:0, i17:1␻7, 17:1␻6
Methane-oxidizing bacteria 16:1␻8c, 16:1␻8t, 16:1␻5c, 18:1␻8c, 18:1␻8t, 18:1␻6c
Barophilic/psychrophilic bacteria 20:5, 22:6
Cyanobacteria 18:2␻6
Protozoa 20:3␻6, 20:4␻6
Fungi 18:1␻9, 18:2␻6, 18:3␻6, 18:3␻3
Actinobacteria 10Me18:0
Microalgae 16:3␻3
Flavobacterium balustinum i17:1␻7, Br 2OH-15:0
Bacillus spp. Various branched chain fatty acids

3. Culture-independent methods of community and not in other parts of the cell as storage products.
analysis This is important because cell membranes are rapidly
degraded and the component phospholipid fatty acids
Because of the inherent limitations of culture-based are rapidly metabolized following cell death. Con-
methods, soil microbial ecologists are turning increas- sequently, phospholipids can serve as important in-
ingly to culture-independent methods of community dicators of active microbial biomass as opposed to
analysis. Using culture-independent methods, the non-living microbial biomass.
composition of communities can be inferred based An essential consideration in the use of these
on (1) the extraction, quantification, and identifi- molecules to describe microbial communities is that
cation of molecules from soil that are specific to unique fatty acids are indicative of specific groups of
certain microorganisms or microbial groups; or (2) organisms (Table 1). Our knowledge of such signature
advanced fluorescence microscopic techniques. Use- molecules comes from the use of fatty acid analysis
ful molecules for such studies include phospholipid for bacterial taxonomy, in which specific fatty acid
fatty acids and nucleic acids (Morgan and Winstanley, methyl esters (FAMEs) have been used as an accepted
1997) whereas the microscopic techniques involve taxonomic discriminator for species identification.
either the hybridization of fluorescent-labeled nucleic Furthermore, phospholipid fatty acids are easily ex-
acid probes with total RNA extracted from soils or tracted from microbial cells in soil (Tunlid and White,
hybridizations with cells in situ. 1992; Zelles and Bai, 1993) allowing access to a
greater proportion of the microbial community resi-
3.1. Phospholipid fatty acid analysis dent in soil than would otherwise be accessed during
culture-dependent methods of analysis.
Phospholipid fatty acid (PLFA) analysis has been The presence and abundance of these signature fatty
used as a culture-independent method of assessing the acids in soil reveals the presence and abundance of
structure of soil microbial communities and determin- particular organisms or groups of organisms in which
ing gross changes that accompany soil disturbances those signatures can be found. For example, Tunlid
such as cropping practices (Zelles et al., 1992, 1995), et al. (1989) were able to use PLFA analysis to demon-
pollution (Frostegard et al., 1993), fumigation (Macal- strate differences in microbial communities associated
ady et al., 1998), and changes in soil quality (Bardgett with Rhizoctonia damping-off. They were further able
et al., 1996; Reichardt et al., 1997; Bossio et al., to monitor the presence of the biological control or-
1998; Petersen et al., 1998). Phospholipid fatty acids ganism Flavobacterium balustinum strain 299 on cu-
are potentially useful signature molecules due to their cumber roots. In other studies, PLFA profiles were
presence in all living cells. In microorganisms, phos- generated from soils exposed to different farming sys-
pholipids are found exclusively in cell membranes tems (Bossio et al., 1998). Organically-managed soils
 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36 29

(i.e. those receiving no synthetic fertilizers and pes- cultured from the same soil. This indicates that soil
ticides) gave rise to PLFA profiles that were signifi- microbial communities are much more complex than
cantly different from those from conventionally man- we currently recognize and that the analysis of DNA
aged soils (i.e. those receiving synthetic fertilizers and sequences may provide a greater understanding of the
pesticides). Profiles from organically-managed soils microbial diversity that exists in soil than could be
were enriched with i14:0, a15:0, 16:1␻7c, 16:1␻5c, gained from culture-dependent methods.
14:0, and 18:2␻6c fatty acids indicating a greater di- Of the various nucleic acid techniques used to esti-
versity of aerobic bacteria as well as populations of mate microbial community composition and diversity
cyanobacteria and methane-oxidizing bacteria. These in complex habitats, the most useful is the determina-
studies clearly demonstrate the utility of this method tion of the sequences of 16S ribosomal RNA (rRNA)
in determining gross community changes associated genes (i.e. encoded by rDNA) in prokaryotes and 5S
with soil management practices. or 18S rRNA genes in eukaryotes (Ward et al., 1992).
Despite the usefulness of this method, there are These small subunit (SSU) rDNA molecules are par-
some important limitations (Haack et al., 1994). First, ticularly suited for such studies for a number of rea-
appropriate signature molecules are not known for all sons. First, they are found universally in all three forms
organisms in a soil sample and, in a number of cases, of life: the domains Bacteria, Archaea, and Eucarya
a specific fatty acid present in a soil sample cannot be (Woese et al., 1990). Second, these molecules are com-
linked with a specific microorganisms or group of mi- posed both of highly conserved regions and also of
croorganisms. In general, the method cannot be used to regions with considerable sequence variation (Woese,
characterize microorganisms to species. Second, since 1987). Because of these differential rates of sequence
the method relies heavily on signature fatty acids to evolution, phylogenetic relationships at several hier-
determine gross community structure, any variation in archical levels can be measured from comparative se-
these signatures would give rise to false community quence analyses. Third, the phylogenetic information
estimates created by artifacts in the methods. Third, held in the SSU rDNA molecule is further enhanced by
bacteria and fungi produce widely different amounts its relatively large size (e.g. ∼1.5 kb for the 16S rDNA
of PLFA and the types of fatty acids vary with growth molecule) and the presence of many secondary struc-
conditions and environmental stresses. Although sig- tural domains. Consequently, evolutionary changes in
nature PLFAs can be correlated with the presence of one domain do not affect the rate of change in other
some groups of organisms, they may not necessarily domains. Finally, SSU rDNA can be easily amplified
be unique to only those groups under all conditions. using polymerase chain reaction (PCR) and rapidly
Consequently, this could give rise to false community sequenced.
signatures. Perhaps the greatest advantage of the analysis of
SSU rDNA is that is that microorganisms from natu-
3.2. Nucleic acid techniques ral habitats can be studied and characterized without
culturing. Various studies have shown that rDNA from
Of all the cell component molecules tested to date, over 90% of the microorganisms that can be observed
nucleic acids have been the most useful in providing a microscopically in situ can be extracted and analyzed
new understanding of the structure of microbial com- (Steffan and Atlas, 1988; Steffan et al., 1988; Tsai
munities. For example, in recent studies of soil micro- and Olsen, 1992; More et al., 1994; Zhou et al., 1996;
bial diversity, Torsvik and colleagues (Torsvik et al., Porteous et al., 1997) as compared with less than 0.1%
1990a,b, 1996; Ovreas and Torsvik, 1998) compared of the microorganisms observed in soil that can be
the re-association kinetics of DNA isolated from soil recovered on culture media.
with that of pure cultures of microorganisms. They Numerous studies have applied these techniques to
reasoned that the greater the sequence diversity of the the study of soil microbial communities (e.g. Stacke-
DNA (and hence the microbial diversity), the greater brandt et al., 1993; Lee et al., 1996; Stephen et al.,
the DNA reannealing time. Based on these studies, 1996; Ueda et al., 1995; Borneman et al., 1996;
they estimated that the genetic diversity of soil was Rheims et al., 1996; Bintrim et al., 1997; Borne-
200 times greater than the diversity among bacteria man and Triplett, 1997; Felske et al., 1997, 1998a,b;
30 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

Heuer and Smalla, 1997; Kuske et al., 1997; Smith analyzed for similarity to other known sequences in
et al., 1997; Duineveld et al., 1998; Grosskopf et al., public-domain databases (e.g. the NCBI GeneBank
1998). In nearly all of these studies, novel microbial database [http://www.ncbi.nlm.nih.gov], the Riboso-
lineages have been discovered, confirming our lack mal Database Project [http://www.cme.msu.edu/RDP/]
of understanding of the microbial species that inhabit (Maidak et al., 1997) and the Antwerp SSU rRNA
soils and their potentially important roles in ecosys- database [http://rrna.uia.ac.be/] (Van de Peer et al.,
tem function. For example, studies have shown that 1997)). By estimating phylogenetic relatedness to
agricultural soils contain a diversity of Archaea, or- other sequences in the databases, the identity of the
ganisms previously thought to exist only in extreme microorganism from which the SSU rRNA gene
environments (Ueda et al., 1995; Bintrim et al., 1997; was derived can be determined. It is hoped that
Buckley et al., 1998). Other studies have shown that the potentially close phylogenetic relationships of
some soil microbes, which have previously not been non-culturable microorganisms with known species
cultured and described, are global in their distribution can be utilized to devise culturing techniques for
and may play important roles in soils worldwide (Lie- many of these microorganisms.
sack and Stackenbrandt, 1992; Felske et al., 1997; In recent years, a number of analyses have focused
Kuske et al., 1997). on the characterization of soil microbial communities
All DNA extraction techniques are based on meth- based on rRNA as opposed to rRNA genes encoded
ods developed over the past 20 years. Once the micro- by rDNA (e.g. Felske and Akkermans, 1998b; Hahn
bial community rDNA is amplified from soil samples et al., 1990; Moran et al., 1993; Felske et al., 1996;
using PCR, individual amplicons must be separated Purdy et al., 1996; Duarte et al., 1998). Like rDNA,
prior to sequence analysis. Methods used most com- rRNA has both conserved and highly variable regions
monly for the separation of individual amplicons have that permit the discrimination of taxa at multiple tax-
been standard cloning procedures using a variety of onomic levels. In addition, use of rRNA offers three
Escherichia coli vectors. principle advantages over rDNA techniques:
Recently, as a complement to cloning procedures, 1. Because ribosomes are the sites of protein synthe-
the use of denaturing gradient and temperature gra- sis, cellular ribosome content (and thus rRNA con-
dient gel electrophoresis (DGGE/TGGE) for separat- tent) are directly correlated with metabolic activ-
ing individual amplicons has been described (Muyzer ity and growth rate (Wagner, 1994). Therefore, a
et al., 1993; Ferris and Ward, 1997; Heuer et al., 1997; high proportion of the rRNA sequences detected
Muyzer and Smalla, 1998). This technique allows one in soil samples should correspond to metabolically
to separate mixtures of PCR products that are of the active and growing microorganisms (Felske et al.,
same length but differ only in sequence. The sepa- 1996). Results with rRNA can be readily compared
ration power of this technique rests with the melting with those for simultaneously-extracted DNA (e.g.
behavior of the double stranded DNA molecule. As Felske et al., 1996) to estimate both the dormant
DNA molecules are electrophoresed in an increasing and metabolically-active community.
gradient of denaturant or in an increasing temperature 2. Because rRNA sequences are typically present in
gradient, it remains double-stranded until it reaches cells in higher copy number than rDNA sequences,
the denaturant concentration or temperature that melts they should be easier to detect (Moran et al., 1993).
the double-stranded molecule. As the DNA melts, it 3. When ribosomes are extracted directly from soil
branches, thus reducing the mobility in the gel. Since samples, free nucleic acids and many dormant mi-
the melting behavior is largely dictated by the nu- croorganisms are excluded and only rRNA from
cleotide sequence, the separation will resolve individ- active cells is detected (Felske et al., 1997; Felske
ual bands, each corresponding to a unique sequence. and Akkermans, 1998a).
Theoretically, any SSU rRNA gene found in the mixed When rRNA amplicons are separated on a DGGE or
template DNA extracted from soils could be specifi- TGGE gel, the banding pattern serves as a fingerprint
cally amplified and resolved on a DGGE gel. of the soil microbial community. Assuming no ampli-
Once rDNA amplicons have been cloned or sepa- fication bias, the intensity of a given band indicates
rated by DGGE or TGGE, they can be sequenced and the abundance of the corresponding rRNA sequence
 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36 31

in the soil community (Felske et al., 1998b). A com- Felske and Akkermans (1998a) found no evidence of
plicating factor is that the number of rRNA operons amplification bias for the universal bacterial primers
is known to vary among taxonomic groups (Rosado they have used to amplify 16S rDNA. For a detailed
et al., 1997), so that rRNA sequence heterogeneity can review of possible sources of amplification bias, see
and does occur within cells of the same species. Con- van Winzingerode et al. (1997).
sequently, each amplification product on a gel cannot Another important limitation to this approach is that
be assumed to correspond to a different organism, and it has been applied largely to investigations of prokary-
a single organism may be represented by several am- otes. Theoretically, detection of rRNA could be used
plification products. to determine active eukaryotes, as well as prokary-
A taxon may contribute rRNA to a soil community otes, in soils. However, eukaryote ribosomes have not
in two ways: (1) by being represented by many active been well studied and their encoding genes and reg-
cells; and (2) by being represented by cells containing ulation are far more complex. Furthermore, sequence
many ribosomes (Felske et al., 1997). These scenarios representation in public databases is not as extensive,
are only reliably distinguishable by in situ hybridiza- making identifications of known eukaryotes from soil
tion (e.g. Binder and Liu, 1998). However, a compar- samples problematic.
ison of probe signal intensity for rRNA and rDNA in
DGGE or TGGE gels can provide evidence for which 3.3. Phylogenetic analysis
of the scenarios is more likely. For example, simi-
lar signal intensities for rRNA and for rDNA suggest The success of any of the preceding methods for
that activity is due to cell abundance rather than high community characterization relies on a suitable phy-
ribosome copy number per cell (Felske et al., 1997). logenetic analysis because many of the organisms that
Despite the usefulness of these nucleic acid tech- are likely to be described from soil communities have
niques for characterizing soil microbial communities, not been studied previously. A number of phylogenetic
there are a number of limitations. As with most tech- methods have been utilized in studies of microbial
niques that measure metabolic activity, storage of ecology (Woese, 1987). While rDNA and rRNA are
samples prior to processing can bias results. Shifts commonly used as characters in phylogenetic analysis,
in active functional groups of prokaryotes have been the list of characters is extensive and can range from
observed when samples are stored aerobically or left molecular to morphological traits (Olsen and Woese,
at room temperature (reviewed in van Winzingerode 1993). For microorganisms, molecular data often pro-
et al., 1997). However, presumably this bias could be vide the greatest wealth of information because mi-
removed easily by immediate processing or freezing. croorganisms such as bacteria simply do not have the
Another limitation is that comparisons of activity diversity of form to make morphological characteris-
among organisms or soil samples may be confounded tics useful in establishing phylogenies.
by several factors. Extraction efficiency differs among Aside from the derivation of taxonomies, phyloge-
soils and microorganisms, so that apparent differences netic analyses are important in identifying similarities
in activity of particular organisms across soil samples between organisms, leading to the ability to understand
can be an artifact of the extraction procedure (Moran the physiology and ecology of as yet non-culturable
et al., 1993). Some prokaryotic cells are more easily species. Unfortunately for taxonomists, phylogenetic
lysed than others, so that incomplete lysis of some analyses have at least one major drawback. The fact
species could result in underestimates of activity (van that an analysis based on a single type of molecule re-
Winzingerode et al., 1997). Sequence amplification sults in a close relationship between taxa does not nec-
and detection are only as good as the probes used; or- essarily mean that another, equally suitable molecule
ganisms with different affinities for the probes used will support these results, although this often occurs
will differ in their apparent activities (Zheng et al., (Olsen and Woese, 1993). When based on a limited
1996). Amplification bias has also been shown to oc- set of taxonomic criteria, it is difficult to say with
cur for templates that differ substantially in abundance, certainty whether or not those criteria can resolve an
with preferential amplification of more abundant se- unknown microorganism from other known microor-
quences (Suzuki and Giovannoni, 1996). However, ganisms. Therefore, microbial phylogenies should be
32 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

interpreted with caution when used in soil microbial (Christensen and Poulsen, 1994; Fischer et al., 1995;
community analyses. MacNaughton et al., 1996; Zarda et al., 1997; Felske
et al., 1998a). Strongly fluorescing dyes can be used
3.4. Fluorescent in situ hybridization (FISH) or multiple probes can be designed to target different
regions of the same 16S or 23S rRNA molecule, thus
Fluorescent in situ hybridization (FISH) has been increasing the strength of the signal (Amann et al.,
used primarily with prokaryotic communities and 1995; Ludwig et al., 1997). Probes for kingdoms (Eu-
allows the direct identification and quantification of bacteria, Archaea, Eucarya), families, genera, species,
specific and/or general taxonomic groups of microor- or sub-species can be differentially labeled and used
ganisms within their natural microhabitat (Amann in combination to view the occurrence and distribution
et al., 1995; Assmus et al., 1995; MacNaughton et al., of several taxonomic groups simultaneously within a
1996; Kenzaka et al., 1998). In FISH, whole cells single soil sample (Manz et al., 1992; Amann et al.,
are fixed, their 16S or 23S rRNA is hybridized with 1995; Zarda et al., 1997). To be detected, soil mi-
fluorescently-labeled taxon-specific oligonucleotide crobes must be metabolically active and possess cell
probes, and then the labeled cells are viewed by scan- walls sufficiently permeable to allow penetration of the
ning confocal laser microscopy (SCLM). Because probe (Christensen and Poulsen, 1994; Amann et al.,
whole cells are hybridized, artifacts arising from 1995). Penetration of cells with such probes is a prob-
biases in DNA extraction, PCR amplification, and lem in nutrient-poor soils and in soils where microor-
cloning are avoided (Ludwig et al., 1997; Wallner ganisms are dormant or quiescent (Hahn et al., 1992;
et al., 1997; Felske et al., 1998a). FISH has two ad- Fischer et al., 1995) because cells are generally smaller
vantages over immunofluorescence techniques. First, and cell walls relatively thicker under these condi-
FISH can detect microorganisms across all phyloge- tions. However, progress is being made to overcome
netic levels, whereas immunofluorescence techniques these problems with groups such as actinobacteria and
are limited to the species and sub-species levels. Bacillus spores (MacNaughton et al., 1994; Fischer
Second, FISH is more sensitive than immunofluores- et al., 1995). To address the problem of low metabolic
cence because non-specific binding to soil particles activity in soil, some researchers have added nutri-
does not typically occur (Amann et al., 1995). FISH ents to stimulate microbial activity (Hahn et al., 1992).
probes can be generated without prior isolation of the However, so as not to bias the community profile, the
microorganism, whereas pure cultures are needed in amendments should equally stimulate all members of
immunofluorescence studies for generating specific the community.
antibodies (Hahn et al., 1992). Scanning confocal FISH can be used to visualize soil microorgan-
laser microscopy (SCLM) surpasses epifluorescence isms that have not yet been cultured, and is useful
microscopy in sensitivity and has the ability to view in studying the ecological distribution of microorgan-
the distribution of several taxonomic groups simulta- isms throughout diverse habitats (Ludwig et al., 1997;
neously as a three-dimensional image (Assmus et al., Zarda et al., 1997; Wullings et al., 1998). When using
1995; Kirchhof et al., 1997). Use of distinctive flu- FISH to examine all members within a given taxon,
orescent dyes and corresponding filter sets allows one must keep in mind that the probe being used
the observer to differentiate fluorescing microbes is only as good as the representative members that
from autofluorescent soil particles and plant debris were used to generate it (Amann et al., 1995). Other,
(Assmus et al., 1995; MacNaughton et al., 1996). non-cultured organisms may not be detected with this
FISH provides a more accurate quantification of cells probe or cross-hybridization to related organisms may
as compared to the rough estimates obtained from occur (Hahn et al., 1992; MacNaughton et al., 1996;
dot-blot assays (Amann et al., 1995) in which micro- Felske et al., 1998a).
bial DNA is blotted onto a membrane than probed FISH can be combined with cultivation techniques,
with the fluorescent oligonucleotide probe. immunofluorescence, nucleotide probes targeting
The sensitivity of FISH has been greatly improved structural genes or mRNAs, reporter genes, microsen-
to afford the detection of single cells within com- sors, or flow cytometry to gain information regarding
plex environments such as rhizosphere and bulk soils the structure and function of microorganisms within
 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36 33

a complex microbial community (Amann and Kuhl, Assmus, B., Hutzler, P., Kirchhof, G., Amann, R., Lawrence,
1998). FISH is a powerful tool that can be used not J.R., Hartmann, A., 1995. In situ localization of Azospirillum
only for studying individuals within a population, brasilense in the rhizosphere of wheat with fluorescently
labeled, rRNA-targeted oligonucleotide probes and scanning
but also has potential uses for studying population confocal laser microscopy. Appl. Environ. Microbiol. 61, 1013–
dynamics, tracking microorganisms released into the 1019.
environment (e.g. for biological control or biore- Atlas, R.M., Bartha, R., 1998. Microbial Ecology: Fundamentals
mediation), epidemiology, and microbial ecology of and Applications. Benjamin/Cummings, Redwood City, CA,
economically important plant pathogens in agricul- 694 pp.
tural soils (Hahn et al., 1992; Kirchhof et al., 1997; Balestra, G.M., Misaghi, I.J., 1997. Increasing the efficiency of
the plate counting method for estimating bacterial diversity. J.
Wullings et al., 1998). Microbiol. Meth. 30, 111–117.
Bardgett, R.D., Hobbs, P.J., Frostegard, A., 1996. Changes in
soil fungal:bacterial biomass ratios following reductions in the
4. Summary intensity of management of an upland grassland. Biol. Fertil.
Soils 22, 261–264.
Binder, B.J., Liu, Y.C., 1998. Growth rate regulation of rRNA
A number of methods are currently available for content of a marine Synechococcus (cyanobacterium) strain.
studies on soil microbial communities. The use of Appl. Environ. Microbiol. 64, 3346–3351.
molecular techniques for investigating microbial di- Bintrim, S.B., Donohue, T.J., Handelsman, J., Roberts, G.P.,
versity in soil communities continues to provide Goodman, R.G., 1997. Molecular phylogeny of Archaea from
new understanding of the distribution and diver- soil. Proc. Natl. Acad. Sci. USA 94, 277–282.
Boehm, M.J., Madden, L.V., Hoitink, H.A.J., 1993. Effect of
sity of organisms in soil habitats. The use of SSU
organic matter decomposition level on bacterial species diversity
rRNA or rDNA sequences, combined with fluorescent and composition in relation to Pythium damping-off severity.
oligonucleotide probes provides a powerful approach Appl. Environ. Microbiol. 59, 4171–4179.
for studying soil microorganisms that may not be Boehm, M.J., Wu, T., Stone, A.G., Kraakman, B., Iannotti,
amenable to current culturing techniques. D.A., Wilson, E.G., Madden, L.V., Hoitink, H.A.J., 1997.
Despite the utility of culture-independent tech- Cross-polarized magic-angle spinning 13 C nuclear magnetic
resonance spectroscopic characterization of soil organic matter
niques such as SSU rRNA or rDNA analyses, there relative to culturable bacterial species composition and
remains a general need to cultivate microorganisms sustained biological control of Pythium root rot. Appl. Environ.
from soil habitats to better understand their role in soil Microbiol. 63, 162–168.
processes. Future studies of soil microbial communi- Borneman, J., Skroch, P.W., O’Sullivan, K.M., Palus, J.A.,
ties must necessarily rely on a combination of both Rumjanek, N.G., Jansen, J.L., Nienhuis, J., Triplett, E.W.,
1996. Molecular microbial diversity of an agricultural soil in
culture-dependent and culture-independent methods
Wisconsin. Appl. Environ. Microbiol. 62, 1935–1943.
and approaches. Only then will we be able to develop Borneman, J., Triplett, E.W., 1997. Molecular microbial diversity
a more complete picture of the contribution of spe- in soils from eastern Amazonia: evidence for unusual
cific microbial communities to the overall quality and microorganisms and microbial population shifts associated with
health of agricultural soils. deforestation. Appl. Environ. Microbiol. 63, 2647–2653.
Bossio, D.A., Scow, K.M., Gunapala, N., Graham, K.J., 1998.
Determinants of soil microbial communities: effects of
agricultural management, season, and soil type on phospholipid
References fatty acid profiles. Microb. Ecol. 36, 1–12.
Buckley, D.H., Graber, J.R., Schmidt, T.M., 1998. Phylogenetic
Alvarez, M.A.D., Gagne, S., Antoun, L., 1995. Effect of analysis of nonthermophilic members of the kingdom
compost on rhizosphere microflora of the tomato and on Crenarchaeota and their diversity and abundance in soils. Appl.
the incidence of plant growth-promoting rhizobacteria. Appl. Environ. Microbiol. 64, 4333–4339.
Environ. Microbiol. 61, 194–199. Campbell, C.D., Grayston, S.J., Hirst, D.J., 1997. Use of
Amann, R., Kuhl, M., 1998. In situ methods for assessment of rhizosphere carbon sources in sole carbon source tests to
microorganisms and their activities. Curr. Opinon Microbiol. 1, discriminate soil microbial communities. J. Microbiol. Meth.
352–358. 30, 33–41.
Amann, R., Ludwig, W., Schleifer, K.H., 1995. Phylogenetic Christensen, H., Poulsen, L.K., 1994. Detection of Pseudomonas in
identification and in situ detection of individual microbial cells soil by rRNA targeted in situ hybridization. Soil Biol. Biochem.
without cultivation. Microbiol. Rev. 59, 143–169. 26, 1093–1096.
34 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

de Leij, F.A.A.M., Whipps, J.M., Lynch, J.M., 1993. The use Garland, J.L., Mills, A.L., 1991. Classification and characterization
of colony development for the characterization of bacterial of heterotrophic microbial communities on the basis of
communities in soil and on roots. Microb. Ecol. 27, 81–97. patterns of community-level sole-carbon-source utilization.
Doran, J.W., Coleman, D.C., Bezdicek, D.F., Stewart, B.A., 1994. Appl. Environ. Microbiol. 57, 2351–2359.
Defining Soil Quality for a Sustainable Environment. American Grosskopf, R., Stubner, S., Liesack, W., 1998. Novel
Society of Agronomy, SSSA Special Publication No. 35, euryarchaeotal lineages detected on rice roots and in the anoxic
Madison, WI. bulk soil of flooded rice microcosms. Appl. Environ. Microbiol.
Duarte, G.F., Rosado, A.S., Seldin, L., Keijzer-Wolters, A.C., van 64, 4983–4989.
Elsas, J.D., 1998. Extraction of ribosomal RNA and genomic Haack, S.K., Garchow, H., Klug, M.J., Forney, L.J., 1995.
DNA from soil for studying the diversity of the indigenous Analysis of factors affecting the accuracy, reproducibility, and
bacterial community. J. Microbiol. Meth. 32, 21–29. interpretation of microbial community carbon source utilization
Duineveld, B.M., Rosado, A.S., van Elsas, J.D., van Veen, J.A., patterns. Appl. Environ. Microbiol. 61, 1458–1468.
1998. Analysis of the dynamics of bacterial communities in the Haack, S.K., Garchow, H., Odelson, D.A., Forney, L.J., Klug, M.J.,
rhizosphere of the chrysanthemum via denaturing gradient gel 1994. Accuracy, reproducibility, and interpretation of fatty acid
electrophoresis and substrate utilization patterns. Appl. Environ. methyl ester profiles of model bacterial communities. Appl.
Microbiol. 64, 4950–4957. Environ. Microbiol. 60, 2483–2493.
Felske, A., Akkermans, A.D.L., 1998a. Prominent occurrence Hackett, C.A., Griffiths, B.S., 1997. Statistical analysis of the time
of ribosomes from an uncultured bacterium of the course of Biolog substrate utilization. J. Microbiol. Meth. 30,
Verrucomicrobiales cluster in grassland soils. Lett. Appl. 63–69.
Microbiol. 26, 219–223. Hahn, D., Amann, R.I., Ludwig, W., Akkermans, A.D.L., Schleifer,
Felske, A., Akkermans, A.D.L., 1998b. Spatial homogeneity of K.H., 1992. Detection of microorganisms in soil after in
abundant bacterial 16S rRNA molecules in grassland soils. situ hybridization with rRNA-targeted, fluorescently labelled
Microb. Ecol. 36, 31–36. oligonucleotides. J. Gen. Microbiol. 138, 879–887.
Felske, A., Akkermans, A.D.L., De Vos, W., 1998a. In situ Hahn, D., Kester, R., Starrenburg, M.J.C., Akkermans, A.D.L.,
detection of an uncultured predominant Bacillus in Dutch 1990. Extraction of ribosomal RNA from soil for detection
grassland soils. Appl. Environ. Microbiol. 64, 4588–4590. of Frankia with oligonucleotide probes. Arch. Microbiol. 154,
Felske, A., Akkermans, A.D.L., De Vos, W., 1998b. Quantification 329–335.
of 16S rRNAs in complex bacterial communities by multiple Heuer, H., Krsek, M., Baker, P., Smalla, K., Wellington,
competitive reverse transcription-PCR in temperature gradient E.M.H., 1997. Analysis of actinomycete communities by
gel electrophoresis fingerprints. Appl. Environ. Microbiol. 64, specific amplification of genes encoding 16S rRNA and
4581–4587. gel-electrophoretic separation in denaturing gradients. Appl.
Felske, A., Engelen, B., Nubel, U., Backhaus, H., 1996. Direct Environ. Microbiol. 63, 3233–3241.
ribosome isolation from soil to extract bacterial rRNA for Heuer, H., Smalla, K., 1997. Application of denaturing gradient
community analysis. Appl. Environ. Microbiol. 62, 4162–4167. gel electrophoresis and temperature gradient gel electrophoresis
Felske, A., Rheims, H., Wolterink, A., Stackenbrandt, E., for studying soil microbial communities. In: van Elsas,
Akkermans, A.D.L., 1997. Ribosome analysis reveals prominent J.D., Trevors, J.T., Wellington, E.M.H. (Eds.), Modern Soil
activity of an uncultured member of the class Actinobacteria Microbiology. Marcel Dekker, New York, pp. 353–373.
in grassland soils. Microbiology 143, 2983–2989. Hitzl, W., Henrich, M., Kessel, M., Insam, H., 1997. Applications
Ferris, M.J., Ward, D.M., 1997. Seasonal distributions of dominant of multivariate analysis of variance and related techniques in
16S rRNA-defined populations in a hot spring microbial mat soil studies with substrate utilization tests. J. Microbiol. Meth.
examined by denaturing gradient gel electrophoresis. Appl. 30, 81–89.
Environ. Microbiol. 63, 1375–1381. Hu, S., van Bruggen, A.H.C., 1997. Microbial dynamics associated
Fischer, K., Hahn, D., Honerlage, W., Schonholzer, F., Zeyer, J., with multiphasic decomposition of 14C-labeled cellulose in
1995. In situ detection of spores and vegetative cells of Bacillus soil. Microb. Ecol. 33, 134–143.
megaterium in soil by whole cell hybridization. Syst. Appl. Kenzaka, T., Yamaguchi, N., Tani, K., Nasu, M., 1998.
Microbiol. 18, 265–273. rRNA-Targeted fluorescent in situ hybridization analysis of
Frostegard, A., Tunlid, A., Baath, E., 1993. Phospholipid fatty acid bacterial community structure in river water. Microbiology 144,
composition, biomass and activity of microbial communities 2085–2093.
from two soil types experimentally exposed to different heavy Kirchhof, G., Schloter, M., Assmus, B., Hartmann, A., 1997.
metals. Appl. Environ. Microbiol. 59, 3605–3617. Molecular microbial ecology approaches applied to diazotrophs
Garland, J.L., 1996a. Analytical approaches to the characterization associated with non-legumes. Soil Biol. Biochem. 29, 853–862.
of samples of microbial communities using patterns of potential Konopka, A., Oliver, L., Turco Jr., R.F., 1998. The use of carbon
C source utilization. Soil Biol. Biochem. 28, 213–221. substrate utilization patterns in environmental and ecological
Garland, J.L., 1996b. Patterns of potential C source utilization by microbiology. Microb. Ecol. 35, 103–115.
rhizosphere communities. Soil Biol. Biochem. 28, 223–230. Kuske, C.R., Barns, S.M., Busch, J.D., 1997. Diverse uncultivated
Garland, J.L., 1997. Analysis and interpretation of community- bacterial groups from soils of the arid southwestern United
level physiological profiles in microbial ecology. FEMS States that are present in many geographical regions. Appl.
Microbiol. Ecol. 24. Environ. Microbiol. 63, 3614–3621.
 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36 35

Lee, S.Y., Bollinger, D., Bezdicek, D., Ogram, A., 1996. Estimation Modern Soil Microbiology. Marcel Dekker, New York,
of the abundance of an uncultured soil bacterial strain pp. 331–352.
by a competitive quantitative PCR method. Appl. Environ. Muyzer, G., Dewaal, E.C., Uitterlinden, A.G., 1993. Profiling
Microbiol. 62, 3787–3793. of complex microbial populations by denaturing gradient gel
Lehman, R.M., Colwell, F.S., Ringelberg, D.B., White, D.C., electrophoresis analysis of polymerase chain reaction-amplified
1995. Combined microbial community-level analyses for quality genes coding for 16S rRNA. Appl. Environ. Microbiol. 59,
assurance of terrestrial subsurface cores. J. Microbiol. Meth. 695–700.
22, 263–281. Muyzer, G., Smalla, K., 1998. Application of denaturing gradient
Liesack, W., Janssen, P.H., Rainey, F.A., Ward-Rainey, N.L., gel electrophoresis (DGGE) and temperature gradient gel
Stackenbrandt, E., 1997. Microbial diversity in soil: the need electrophoresis (TGGE) in microbial ecology. Antonie van
for a combined approach using molecular and cultivation Leeuwenhoek 73, 127–141.
techniques. In: In: van Elsas, J.D., Trevors, J.T., Wellington, Olsen, G.J., Woese, C.R., 1993. Ribosomal RNA: a key to
E.M.H. (Eds.), Modern Soil Microbiology. Marcel Dekker, New phylogeny. FASEB J. 7, 113–123.
York, pp. 375–439. Ovreas, L., Torsvik, V., 1998. Microbial diversity and community
Liesack, W., Stackenbrandt, E., 1992. Occurrence of novel groups structure in two different agricultural soil communities. Microb.
of the domain Bacteria as revealed by analysis of genetic Ecol. 36, 303–315.
material isolated from an Australian terrestrial environment. J. Petersen, S.O., Debosz, K., Schjonning, P., Christensen, B.T.,
Bacteriol. 174, 5072–5078. Elmholt, S., 1998. Phospholipid fatty acid profiles and C
Ludwig, W., Bauer, S.H., Bauer, M., Held, I., Kirchhof, G., availability in wet-stable macro-aggregates from conventionally
Schulze, R., Huber, I., Spring, S., Hartmann, A., Schleifer, K.H., and organically farmed soils. Geoderma 78, 181–196.
1997. Detection and in situ identification of representatives of Porteous, L.A., Seidler, R.J., Watrud, L.S., 1997. An improved
a widely distributed new bacterial phylum. FEMS Microbiol. method for purifying DNA from soil for polymerase chain
Lett. 153, 181–190. reaction amplification and molecular ecology applications. Mol.
Macalady, J.L., Fuller, M.E., Scow, K.M., 1998. Effects of metan Ecol. 6, 787–791.
sodium fumigation on soil microbial activity and community Purdy, K.J., Embley, T.M., Takii, S., Nedwell, D.B., 1996. Rapid
structure. J. Environ. Qual. 27, 53–63. extraction of DNA and rRNA from sediments by a novel
MacNaughton, S.J., Booth, T., Embley, T.M., O’Donnell, A.G., hydroxyapatite spin-column method. Appl. Environ. Microbiol.
1996. Physical stabilization and confocal microscopy of 62, 3905–3907.
bacteria on roots using 16S rRNA targeted fluorescent-labeled
Reichardt, W., Mascarina, G., Padre, B., Doll, J., 1997. Microbial
oligonucleotide probes. J. Microbiol. Meth. 26, 279–285.
communities of continuously cropped, irrigated rice fields.
MacNaughton, S.J., O’Donnell, A.G., Embley, T.M., 1994.
Appl. Environ. Microbiol. 63, 233–238.
Permeabilization of mycolic acid-containing actinomycetes for
Rheims, H., Sproer, C., Rainey, F.A., Stackenbrandt, E., 1996.
in situ hybridization with fluorescently labeled oligonucleotide
Molecular biological evidence for the occurrence of uncultured
probes. Microbiology 140, 2859–2865.
members of the actinomycete line of descent in different
Maidak, B.L., Olsen, G.J., Larsen, N., Overbeck, R., McCaughey,
environments and geographical locations. Microbiology 142,
M.J., Woese, C.R., 1997. The RDP (ribosomal database project).
2863–2870.
Nucleic Acids Res. 25, 109–110.
Rosado, A.S., Duarte, G.F., Seldin, L., van Elsas, J.D., 1997.
Maloney, P.E., van Bruggen, A.H.C., Hu, S., 1997. Bacterial
Molecular microbial ecology: a mini review. Revista de
community structure in relation to the carbon environments in
Microbiologia 28, 135–147.
lettuce and tomato rhizospheres and in bulk soil. Microb. Ecol.
34, 109–117. Smalla, K., Wachtendorf, U., Heuer, H., Liu, W.T., Forney, L.,
Manz, W., Amann, R., Ludwig, W., Wagner, M., Schleifer, K.H., 1998. Analysis of BIOLOG® GN substrate utilization patterns
1992. Phylogenetic oligodeoxynucleotide probes for the major by microbial communities. Appl. Environ. Microbiol. 64, 1220–
subclasses of proteobacteria: problems and solutions. Syst. 1225.
Appl. Microbiol. 15, 593–600. Smith, E., Leeflang, P., Wernars, K., 1997. Detection of shifts
Mitsui, H., Gorlach, K., Lee, H.-J., Hattori, R., Hattori, T., 1997. in microbial community structure and diversity in soil caused
Incubation time and media requirements of culturable bacteria by copper contamination using amplified ribosomal DNA
from different phylogenetic groups. J. Microbiol. Meth. 30, restriction analysis. FEMS Microbiol. Ecol. 23, 249–261.
103–110. Stackebrandt, E., Liesack, W., Goebel, B.M., 1993. Bacterial
Moran, M.A., Torsvik, V.L., Torsvik, T., Hodson, R.E., 1993. Direct diversity in a soil sample from a subtropical Australian
extraction and purification of rRNA for ecological studies. Appl. environment as determined by 16S rDNA analysis. FASEB J.
Environ. Microbiol. 59, 915–918. 7, 232–236.
More, M.I., Herrick, J.B., Silva, M.C., Ghiorse, W.C., Madsen, Steffan, R.J., Atlas, R.M., 1988. DNA amplification to enhance
E.L., 1994. Quantitative cell lysis of indigenous microorganisms detection of genetically engineered bacteria in environmental
and rapid extraction of microbial DNA from sediment. Appl. samples. Appl. Environ. Microbiol. 54, 2185–2191.
Environ. Microbiol. 60, 1572–1580. Steffan, R.J., Goksoyr, J., Bej, A.K., Atlas, R.M., 1988. Recovery
Morgan, J.A.W., Winstanley, C., 1997. Microbial biomarkers. of DNA from soils and sediments. Appl. Environ. Microbiol.
In: van Elsas, J.D., Trevors, J.T., Wellington, E.M.H. (Eds.), 56, 776–781.
36 G.T. Hill et al. / Applied Soil Ecology 15 (2000) 25–36

Stephen, J.R., McCaig, A.E., Smith, Z., Prosser, J.I., Embley, T.M., in nature. In: Marshall, K.C. (Ed.), Advances in Microbial
1996. Molecular diversity of soil and marine 16S rRNA gene Ecology. Plenum Press, New York, pp. 219–286.
sequences related to beta-subgroup ammonia-oxidizing bacteria. Winding, A., 1994. Fingerprinting bacterial soil communities
Appl. Environ. Microbiol. 62, 4147–4154. using biolog microtitre plates. In: Ritz, K., Dighton, J., Giller,
Suzuki, M.T., Giovannoni, S.J., 1996. Bias caused by template K.E. (Eds.), Beyond the Biomass. Wiley Sayce, New York,
annealing in the amplification mixtures of 16S rRNA genes by pp. 85–94.
PCR. Appl. Environ. Microbiol. 62, 625–630. Woese, C.R., 1987. Bacterial evolution. Microbiol. Rev. 51, 221–
Torsvik, V., Golsoyr, J., Daae, F., 1990a. High diversity in 271.
DNA of soil bacteria. Appl. Environ. Microbiol. 56, 782– Woese, C.R., Kandler, O., Wheelis, M.L., 1990. Towards a
787. natural system of organisms: proposal for the domains Archaea,
Torsvik, V., Salte, K., Sorheim, R., Goksoyr, J., 1990b. Bacteria, and Eucarya. Proc. Natl. Acad. Sci. USA 87, 4576–
Comparison of phenotypic diversity and DNA heterogeneity in 4579.
a population of soil bacteria. Appl. Environ. Microbiol. 56, 776– Workneh, F., van Bruggen, A.H.C., Drinkwater, L.E., Shennan, C.,
781. 1993. Variables associated with corky root and Phytophthora
Torsvik, V., Sorheim, R., Goksoyr, J., 1996. Total bacterial diversity root rot of tomatoes in organic and conventional farms.
in soil and sediment communities — a review. J. Ind. Microbiol. Phytopathology 83, 581–589.
17, 170–178. Wullings, B.A., van Beuningen, A.R., Janse, J.D., Akkermanns,
Tsai, Y., Olsen, B.H., 1992. Detection of low numbers of bacterial A.D.L., 1998. Detection of Ralstonia solanacearum which
cells in soils and sediments by polymerase chain reaction. Appl. causes brown rot of potato, by fluorescent in situ hybridization
Environ. Microbiol. 58, 754–757. with 23S rRNA-targeted probes. Appl. Environ. Microbiol. 64,
Tunlid, A., Hoitink, H.A.J., Low, C., White, D.C., 1989. 4546–4554.
Characterization of bacteria that suppress Rhizoctonia Zak, J.C., Willig, M.R., Moorhead, D.L., Wildman, H.G., 1994.
damping-off in bark compost media by analysis of fatty acid Functional diversity of microbial communities: a quantitative
biomarkers. Appl. Environ. Microbiol. 55, 1368–1374. approach. Soil Biol. Biochem. 26, 1101–1108.
Tunlid, A., White, D.C., 1992. Biochemical analysis of biomass, Zarda, B., Hahn, D., Chatzinotas, A., Schonhuber, W., Neef, J.,
community structure, nutritional status, and metabolic activity Amann, R.I., Zeyer, J., 1997. Analysis of bacterial community
of microbial communities in soil. In: Stotzky, G., Bollag, structure in bulk soil by in situ hybridization. Arch. Microbiol.
J.-M. (Eds.), Soil Biochemistry. Marcel Dekker, New York, 168, 185–192.
pp. 229–262. Zelles, L., Bai, Q.Y., 1993. Fractionation of fatty acids derived
Ueda, T., Suga, Y., Matsuguchi, T., 1995. Molecular phylogenetic from soil lipids by solid phase extraction and their quantitative
analysis of a soil microbial community in a soybean field. Eur. analysis by GC-MS. Soil Biol. Biochem. 25, 495–507.
J. Soil Sci. 46, 415–421. Zelles, L., Bai, Q.Y., Beck, T., Beese, F., 1992. Signature fatty
Van de Peer, Y., Jansen, J., De Rijk, P., De Wachter, P., 1997. acids in phospholipids and lipopolysaccharides as indicators
Database on the structure of small ribosomal subunit RNA. of microbial biomass and community structure in agricultural
Nucleic Acids Res. 25, 111–116. soils. Soil Biol. Biochem. 24, 317–323.
van Winzingerode, F., Gobel, U.B., Stackebrandt, E., 1997. Zelles, L., Rackwitz, R., Bai, Q.Y., Beck, T., Beese, F., 1995.
Determination of microbial diversity in environmental samples: Discrimination of microbial diversity by fatty acid profiles of
pitfalls of PCR-based rRNA analysis. FEMS Microbiol. Rev. phospholipids and lipopolysaccharides in differently cultivated
21, 213–229. soils. Plant Soil 170, 115–122.
Wagner, R., 1994. The regulation of ribosomal RNA synthesis and Zheng, D., Alm, E.W., Stahl, D.A., Raskin, L., 1996.
bacterial cell growth. Arch. Microbiol. 161, 100–109. Characterization of universal small-subunit rRNA hybridization
Wallner, G., Fuchs, B., Spring, S., Beisker, W., Amann, R., 1997. probes for quantitative molecular microbial ecology studies.
Flow sorting of microorganisms for molecular analysis. Appl. Appl. Environ. Microbiol. 62, 4504–4513.
Environ. Microbiol. 63, 4223–4231. Zhou, J., Bruns, M.A., Tiedje, J.M., 1996. DNA recovery from
Ward, D.M., Bateson, M.M., Weller, R., Ruff-Roberts, A.L., 1992. soils of diverse composition. Appl. Environ. Microbiol. 62,
Ribosomal RNA analysis of microorganisms as they occur 312–322.