Applications of Nucleic Acid Hybridization in Microbial Ecology
Author(s): William E. Holben and James M. Tiedje
Source: Ecology, Vol. 69, No. 3 (Jun., 1988), pp. 561-568
Published by: Ecological Society of America
Stable URL: http://www.jstor.org/stable/1941005 .
Accessed: 06/01/2011 15:41
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=esa. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology.
http://www.jstor.org
SPECIAL FEATURE-NEW
June 1988
MICROBIAL ECOLOGY
561
Ecology, 69(3), 1988, pp. 561-568
? 1988 by the Ecological Society of America
APPLICATIONS OF NUCLEIC ACID HYBRIDIZATION
IN MICROBIAL ECOLOGY
WILLIAM E. HOLBEN
Department of Crop and Soil Sciences, and Plant Research Laboratory,
Michigan State University
AND
JAMES M. TIEDJE
Department of Crop and Soil Sciences, and Department of Microbiology and Public Health,
Michigan State University, East Lansing, Michigan 48824 USA
ABSTRACT
Nucleic acid hybridization techniques offer a new
approach to answer old, intractable questions in microbial ecology as well as new questions. These include
characterization of the predominant, yet unculturable
populations in nature, the role of the environment in
gene expression, and the extent of gene exchange among
communities in nature. The essence of this methodology is the denaturing and annealing of complementary strands of nucleic acid molecules. The specificity
of this hybridization reaction can be controlled such
that only identical, or nearly identical, sequences in a
complex mixture of nucleic acids extracted from a population or community can anneal. Labeled DNA or
RNA sequences (probes) are introduced into hybridization reactions to identify and quantitate a particular
organism containing the complementary target sequence. Methods for the recovery and purification of
DNA from soils and sediments are given, as well as
important considerations for the selection of probe and
target sequences, and for the methods of detection and
quantitation. Some advantages of this methodology
include the abilities to: (1) detect populations without
prior culturing of organisms, (2) detect specific organisms without the need for selectable markers, (3) detect
multiple populations in the same analysis, and (4) detect genetic rearrangements or gene transfer in natural
communities.
Key words: community ecology; DNA; gene probes; soil microbiology.
INTRODUCTION
Progress in microbial ecology is more dependent upon
exotic chemical and physical techniques than is macroecology; the small size of microbes rules out most of
the direct approaches available for the study of macroorganisms. For example, the classical plate count, which
is the most direct enumeration method for microbes,
is of fairly limited value. Many populations from nature do not grow on plate media, and those that do
may not be distinguishable from other members of the
community. Efficiency of recovery of species is likely
to be poor for plate counts. Furthermore, after the
many generations necessary to form plate colonies, the
organism may deviate from its physiology, and possibly even from the genotypic mix, of the population
in nature.
A great set of improvements in microbial ecology
methods comes with the fluorescent antibody technique. A newer application of antibody techniques is
discussed by K. Porter in the accompanying article.
These methods allow identification and counting of
individual cells without the requirement of prior growth
in culture media.
The nucleic acid techniques offer approaches to questions that were previously difficult or impossible to
answer in microbial ecology. DNA and information on
the identity of species can be recovered from natural
populations that cannot be identified or cultured, genes
as well as the host microbes can be traced in nature,
and many species can be studied simultaneously. In
addition, changes in gene organization and the regulation of gene expression in nature are possible with
nucleic acid hybridization techniques.
This paper is a brief overview of the basic principles
of these techniques and their applications in microbial
ecology. For details of the principles and protocols of
nucleic acid hybridization methods, the reader is referred to Berger and Kimmel (1987) and several molecular biology laboratory manuals (e.g., Davis et al.
1980, Maniatis et al. 1982, Ausubel et al. 1987).
562
SPECIAL FEATURE-NEW
::
:: :: :
Ecology Vol. 69, No. 3
MICROBIAL ECOLOGY
: ::::::::
: :
::::::
: : :::::::
PRINCIPLE
HYBRIDIZATION
The common features of nucleic acid hybridization
techniques are denaturation and selective annealing.
Double-stranded DNA or the secondary structures of
RNA molecules are first denatured into their primary,
single-stranded, linear form, which makes the molecule
accessible for hybridization with a complementary
strand of nucleic acid. Double-stranded DNA is generally denatured by treatment with base (e.g., NaOH)
and/or by heating to a temperature greater than the T>,
(the melting temperature) of the molecule. RNA molecules are generally denatured by heating in the presence of formamide. Once denatured, the single-stranded nucleic acid is available for renaturation with a
complementary strand of nucleic acid. The complementary strand may simply be that which was previously paired with it and is still present in solution, as
is the case in studies of renaturation kinetics. Alternatively, the complementary strand may be supplied
exogenously and may consist of the DNA complement
of an RNA sequence or the DNA or RNA complement
of a DNA sequence. The exogenously added nucleic
acid can be labelled with either radioactive or nonradioactive residues so that it serves as a probe for a
particular nucleic acid sequence in a heterogeneous
mixture of nucleic acid molecules.
APPLICATIONS
TRADITIONAL
One early application was heteroduplex analysis, used
to reveal similar sequences of nucleic acids. Singlestranded nucleic acids of known sequence are hybridized with sample molecules, then examined by electron
microscopy. This visualization reveals the doublestranded (homologous) and single-stranded (heterologous) regions in the hybrids, which reveal aspects of
the sample sequence. Heteroduplex analysis has identified the locales of intron sites by showing where processed messenger RNA binds to DNA of the corresponding gene (Tilghman et al. 1978), detected genomic
rearrangements (Shapiro 1983), and demonstrated
transposable DNA elements (Williamson et al. 1981).
Other applications of nucleic acid hybridization
techniques are Cot (COT) and Rot (ROT) hybridization
kinetics studies (see, e.g., Britten et al. 1974, Chelm
1982a, b). The underlying principle of COT analyses
(of DNA renaturation kinetics) is that the rate of hybridization is proportional to the concentration of
complementary DNA sequences and thus inversely
proportional to the total length of different sequences
in the sample (Chelm 1982a). The kinetics of the renaturation process, therefore, indicate the complexity
of the DNA being analyzed. The dosage of specific
sequences of DNA is related to the time required for
these sequences to encounter and bind to complements
: :::
: :: : :
5:fff
:f f
: : ::::::ffgg
f:ffffi::
:::o:::
o
ff:f
ff~~x:
f D;
in complex mixtures. This approach has been used for
counting the number of copies of chloroplast DNA in
the total DNA of Euglena gracilis (Chelm 1982a) and
for demonstrating repeated families of sequences in
eukaryotic cells (Britten et al. 1974). Similarly, with
ROT analyses, the kinetics of hybridization of the DNA
and total RNA of an organism can be used to measure
the number and relative abundance of transcripts
(Chelm 1982b).
Such methods are not yet widely used in microbial
ecology. COT analyses may prove useful for measuring
the complexity of microbial communities in soil, sediment, and aquatic systems. Sayler and his co-workers
(G. Sayler, personal communication) have initiated
studies using DNA:DNA renaturation kinetics as a
measure of community complexity in aquatic sediments. They find that exceedingly long hybridization
times are involved, because one is essentially measuring the renaturation of very long molecules, entire genomes among mixtures of genomes, rather than the
short sequences within genomes.
Another method is the use of DNA probes to quantify the level of transcription of a gene of interest. To
detect the transcription level, an excess of "probe"
DNA, identical in sequence to the gene, is bound to a
filter, then hybridized to the total RNA of an organism
grown under specific conditions. Usually the RNA of
the organism is labelled in vivo with radioactive nucleotides. The amount of labelled RNA that hybridizes
to the probe DNA is proportional to the number of
RNA molecules transcribed from that gene. This approach has been used, for example, to measure rRNA
transcription in Escherichia coli (Sharrock et al. 1985).
Though not yet explored by microbial ecologists, this
methodology could reveal how the environment controls transcription and thus gene expression.
Perhaps best known is the use of DNA and RNA
probes to detect specific genes or DNA sequences. It
is this application that has the greatest immediate potential in microbial ecology. Careful control (termed
"stringency") of conditions in the process of hybridization and subsequent washing away of unhybridized
probe allow one to probe for identical sequences (termed
homologous probing). An example is to probe for the
E. coli lacZ gene in an E. coli strain. Probes for similar,
but not identical, sequences (termed heterologous
probing) can use a cloned gene from one organism as
a probe for the same gene in a different organism, as
was done in the isolation of the glnA gene from Bradyrhizobium japonicum using the glnA gene from E.
coli (Carlson et al. 1985).
More recently, DNA probe hybridization analyses
have been used for the clinical identification of microorganisms (Echeverria and Haanstra 1982, Fitts et al.
1983, Hill et al. 1983). These assays rely on the use of
SPECIAL FEATURE-NEW
June 1988
::
species-unique probes, which hybridize only to the
DNA of one organism or group of organisms and not
to DNA of other species. These detection methods
have shown such promise that several commercially
available clinical identification kits are currently being
marketed (e.g., the Salmonella detection kits for the
food industry available from GENE-TRAK Systems,
Framingham, Massachusetts). Gene probe methods are
also being used for the detection and identification of
parasitic (Palva 1985) and genetic (Harada et al. 1987)
diseases. Nucleic-acid hybridization techniques are also
being used for the phylogenetic classification of microorganisms on the basis of similarities and differences
in the genes that encode ribosomal RNA (Pace et al.
1985).
By definition, different taxa contain at least some
unique genetic information. Thus, the ability to distinguish taxon-specific sequences in a complex mixture
of DNA holds great promise for microbial ecology. One
can readily expand from detecting a specific gene in
the genome of an organism to detecting the same gene
in a mixture of genomes from several organisms, e.g.,
in the DNA obtained from the entire bacterial population in an environmental sample (Holben et al. 1988).
ADVANTAGES AND LIMITATIONS OF
HYBRIDIZATION ANALYSES IN
MICROBIAL ECOLOGY
Some advantages of nucleic acid hybridization analyses for study of bacterial populations in environmental samples are: (1) The nucleic acids can be isolated
in situ without culturing. This ability allows direct
quantification and comparisons of microbial populations. For example, in soil only 1-5% of the total population observed by direct microscopic counts can be
cultured (Bakken 1985, Bone and Balkwill 1986). The
gene probe approach allows nucleic acids from all of
the taxa in a sample to be analyzed simultaneously,
without need for cultural conditions satisfying each
taxon. (2) A specific gene or nucleic-acid sequence can
be detected, so gene expression is not a prerequisite to
successful monitoring of bacterial populations. This
point is important when one considers that perhaps
only a small minority of taxa in an environmental
sample are growing rapidly at any given time, and thus
that only a few may be actively expressing genes that
would be relied on for the detection of the organism.
(3) Several different taxa can be monitored in a single
sample. In fact, the efficiency of detection of microorganisms actually increases with the number of taxa
being studied, up to a practical limit, with probe techniques (see below). (4) If Southern transfer techniques
(Southern 1975) are employed in conjunction with
probe analyses, it should be possible to detect genetic
~
: :
MICROBIAL ECOLOGY
:
:..
::
::
::
563
:::
. :I:
rearrangements within a population and gene transfer
to new populations. (5) Selectable phenotypes, i.e., mutant derivatives, are not required because the probe
directly detects a specific nucleic acid sequence, avoiding the problem that auxotrophic or antibiotic-resistant
derivatives of wild-type strains may be compromised
in their ecological fitness.
One limitation of nucleic acid hybridization analyses
for environmental samples is their greater sophistication and complexity than traditional methodologies.
Current protocols for the extraction of nucleic acids
from environmental matrices are the limiting step in
many analyses. Although not overly difficult, they are
tedious and involve many extraction and purification
steps. Also, it is currently difficult to process many
samples, as is often necessary for the statistical analyses
needed to resolve the heterogeneity of most ecological
communities. We can currently extract and purify DNA
from only 6-12 soil samples simultaneously, but can
analyze up to 72 DNA samples concurrently (Holben
et al. 1988). Also, methods for precise determination
of the number of gene copies in a sample have not yet
been developed, but there are several possible avenues
to be explored.
RNA, especially messenger RNA, is less stable than
DNA because RNases are stable and ubiquitous and
because the alkaline conditions employed in most DNAisolation protocols destroy RNA. Methods for the isolation of ribosomal RNA (which is less labile due to
its packaging into ribosomes) from environmental
samples for use in population measurements and phylogenetic studies have been developed (Stahl et al., in
press; R. Devereaux,
personal communication).
Simultaneous analysis of multiple taxa actually increases the efficiency of nucleic acid hybridization
techniques for identification-just the opposite of the
increasing difficulty with diversity for traditional techniques. For example, selective plating techniques require different culture conditions for each organism;
every sample must be applied to all of the selective
media before the taxa can be enumerated. By contrast,
the total bacterial DNA obtained from a single environmental sample can be digested with an appropriate
restriction endonuclease (a simple process), then the
resultant pool of restriction fragments is "sorted" by
size by electrophoresis: different genetic sequences produce fragments of different lengths. The separated DNA
fragments are then transferred to filters for hybridization with DNA or RNA probes previously developed
for each taxon, i.e., are subjected to Southern transfer
(Southern 1975). By design, the probe for each taxon
of interest will hybridize to a fragment of a different
size. Thus, multiple populations of bacteria in a single
sample are detected by discrete bands of hybridization,
each indicating the fragment length for one taxon (Fig.
564
SPECIAL FEATURE-NEW
MICROBIAL ECOLOGY
Ecology Vol. 69, No. 3
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.~~~~
1
2
3
4
5
1
2
3
4
5
B
A
FIG. 1. Single-probe
detectionof a wild-typeparentalstrain
BJ110I)and an engineeredderiv(Bradyrhizobiumjaponicum
ative (B.japonicumCRM52 [McClungand Cheim 1987])in
nonsterilesoil. (A) EcoRI digested DNA samples size fractionated on a 0.7% agarose gel and stained with ethidium
bromide;(B) autoradiogramobtainedafterSoutherntransfer
of DNA from the gel to cellulose nitrate and hybridization
with a single-strandedDNA probe specific for a sequence
presentin both organisms.The mutant has a portion of the
probedgene deleted and thus its DNA fragmentis smaller,
which allows separationof the mutantfromwild-type.Lanes
1 and 5 are standardswith 0.2 uigof DNA isolated from a
pure culture of the mutant and the wild-type, respectively.
Lanes2, 3, and 4 contain2 jig of total bacterialDNA isolated
fromnonsterilesoil inoculatedwith mutant,mutantpluswildtype, or wild-type,respectively.
1). Therefore, as the number of different organisms
being monitored increases, the efficiency of detection
(as determined by the amount of time and materials
expended) increases for hybridization-probe methods.
EXPERIMENTAL CONSIDERATIONS
Recovery of sample DNA
These techniques in microbial ecology rely upon efficient isolation of usefully intact and pure nucleic acids
from environmental samples. There are two basic approaches to recovery of total community DNA from
soils and sediments. One is to extract DNA directly
from the samples by lysis of the organismsin situ; the
other is firstto extractas many organismsas possible
from the environmental matrix so that the DNA is
protected until it is in a more defined environment.
The differencesbetween these two approachesare less
importantfor aquatic samples than for soil and sediment samples.
Direct extractionof nucleic acids is much more efficientthan attemptingto recoverwhole bacteriafrom
environmental matrices;in some protocols it also is
faster.A disadvantageof direct extractionof bacterial
DNA from soil samples is contamination by humic
acids and other soil substances,which can inhibit the
efficiency of subsequent restriction-endonuclease
digestion and hybridizationreactions. Also, DNA so
isolatedprobablyincludesthat of fungi,algae,and other organismsin the sample, as well as free DNA not
representingliving organisms.G. Saylerand co-workers have developed a direct-extractionprotocol for
aquatic sediments (Ogramet al. 1987). B. Olson and
her groupare currentlydeveloping a method in which
soil or sediment samples are embedded in an agarose
matrix, the bacteriain the sample are then lysed, and
the DNA is separatedfrom the other cellularcomponentsand soil contaminantsby electrophoresisinto the
agarose.Followingelectrophoresis,the DNA is transferredto a cellulose nitrate membranefor subsequent
hybridization analysis (B. Olson, personal communication).
The second approachfor the recoveryof DNA from
soils and sediments calls for first recoveringthe bacterial fraction (Holben et al. 1988; R. Atlas, personal
communication). Most techniques for isolating the bac-
terial fraction from soils and sediments are derived
from the differentialcentrifugationtechnique first describedby Faegriet al. (1977). The isolated bacterial
fraction is then lysed and the released DNA purified
for subsequent hybridizationanalyses. Some advantages of this approachare that the DNA is protected
from contamination with soil material and that the
DNA is less fragmentedand more uniform in length,
digestion,agarose-gel
makingrestriction-endonuclease
size-fractionation,and Southerntransferand hybridization more efficientand easier to interpret.Also, the
origin of the DNA is known. Disadvantages of this
approachare that the protocols are somewhat lengthy
and that all of the bacteria in a sample are not recovered; however, the bacterial fraction isolated by
differentialcentrifugation(Bone and Balkwill 1986),
and thus the isolated DNA, appears to be representative of the total bacterialpopulation. Currently,the
quantities of bacterial DNA isolated from soils and
sedimentsaccount for largernumbersof bacteriathan
can be detectedby viable-countenumerations,and the
nucleic acid methods are thus more likely to represent
naturalbacterialpopulationsthan arenonselectivecul-
SPECIAL FEATURE-NEW
June 1988
turing methods (Ogram et al. 1987, Holben et al. 1988;
R. Atlas, personal communication).
Alternative DNA isolation protocols for environmental samples are also being developed. J. Paul and
his co-workers have isolated free DNA from marine,
estuary, and river waters for subsequent hybridization
analyses using specific probes for sequences of ecological significance: genes such as ribosomal RNA genes
or the RUBISCO enzyme large subunit. These studies
hope to yield information on the sources of, and the
means to track, recombinant DNA in the environment.
Because the free DNA is in solution, the isolation involves filtering of the water sample to remove organisms and solids followed by ethanol precipitation of
the nucleic acids in the filtrate. The nucleic acids are
further purified and concentrated by dialysis prior to
use in hybridization studies (DeFlaun et al. 1986).
Population density is approached by a combination
of classical microbiological Most Probable Number
(MPN) analysis with DNA probe analyses for specific
microorganisms in soil (Fredrickson et al. 1988). The
classical principle is that population density is correlated with frequency of the organism among subdivided aliquots of a sample. Diluted soil samples subdivided among 96 separate wells in selective (for slowgrowing bacteria) and nonselective (for fast-growing
culturable bacteria) media are incubated for bacterial
growth. The contents of each well in the microtiter dish
are then transferred to a nylon membrane, preserving
the discreteness of the subsamples. They are lysed and
hybridized with a probe specific to the organism of
interest. MPN statistical analyses of hybridization results are used to determine the numbers of the organism of interest that were present in the original sample.
Although only one probe can be used per filter (to detect
one type of microorganism) and only readily culturable
cells can be detected, this method is quite sensitive,
being able to detect 10 Pseudomonas putida and 100
Rhizobium leguminosarum per gram of soil.
An alternative to using enrichment cultures to isolate
bacteria with metabolic phenotypes of interest is to
isolate single colonies on nonselective medium from
samples likely to have bacteria with the desired phenotype. The colonies are lysed on filters, and their DNA
hybridized with a probe for a gene in the metabolic
pathway of interest. Pettigrew and Sayler (1986) used
this method to isolate 4-chlorobiphenyl-degrading bacteria from consortia obtained from PCB-contaminated
sediments. Such methods, of course, rely on having an
appropriate gene probe for the phenotype of interest.
Choice of probe
There are several parameters and possibilities to
consider when probes are used to analyze DNA: the
sequence being probed for (i.e., the "target"), what types
MICROBIAL ECOLOGY
565
of probe to use, and what hybridization conditions will
be employed. Such choices will necessarily be made on
the basis of the nature of the experiment being performed. Below, we consider a number of such possibilities and describe conditions under which one choice
would perhaps be more appropriate than others.
Probe and target-sequence considerations
The target sequence and the probe used for hybridization can range in size from about 20 bases of DNA
to the entire chromosome of an organism. When complex DNA such as that isolated from a community is
probed, specificity is of primary importance. Thus, sequences common to different organisms must be
avoided. Common sequences that complicate specific
detection range from sequences common to all organisms (e.g., conserved regions of ribosomal RNA genes)
to regions of wide-host-range plasmids that have homology to other plasmids and the chromosomes of
some organisms (Bender and Cooksey 1986, Golub
and Low 1986, T. Barkay, personal communication).
Generally, the smaller the target sequence and corresponding probe (down to 17-20 bases), the more specific the resulting detection.
Gene-sized (i.e., 0.5-1.5 kb) target and probe sequences are often appropriate for hybridization experiments involving complex mixtures of DNA from many
organisms. The target sequence may be a wild-type
sequence or consist of a DNA sequence engineered into
the organism. This size range of target sequence provides enough specificity to detect and enumerate a particular organism present in a mixed microbial population (Fredrickson et al. 1988, Holben et al. 1988; R.
Atlas, personal communication). Probes of this size
have also proven useful in correlating the presence of
particular catabolic or resistance genes with the presence of substrates for activity of those genes (Sayler et
al. 1985, Barkay and Olson 1986; W. E. Holben and
J. M. Tiedje, personal observations). It may also be
possible to develop heterologous probes for functions
such as nitrogen fixation or denitrification based on the
sequences at the active sites of enzymes involved in
these processes, because these regions are likely to be
conserved.
Probe methodologies have been to provide community population profiles based on probing for a "universal" sequence (e.g., the 16S ribosomal RNA gene;
Stahl et al., in press; R. Devereaux, personal communication). This gene exists in all organisms, yet is
sufficiently different in each organism to allow the identification of specific populations in the community (Pace
et al. 1985). Most of the probes described above are
based on known sequences unique to specific organisms or functions. It has also been demonstrated that
probes specific for a group, genus, or even species or
566
SPECIAL FEATURE-NEW
MICROBIAL ECOLOGY
strain of bacteria can readily be constructed by random
cloning of DNA fragments from the organism of interest, and screening for those that do not cross-react
with DNA from other organisms likely to be encountered in the samples (Salyers et al. 1983, Yates et al.
1986).
Using the entire chromosome as a target and probe
for hybridization is appropriate when the similarity of
one organism to another is investigated at the DNA
level. This type of comparison can be made if the total
DNA of one organism is labelled, then denatured and
hybridized to the denatured total DNA of that same
organism and, in a separate reaction, to the denatured
total DNA of the organism being compared. Comparing the amount of hybridization to each DNA indicates
the degree of similarity between the two organisms.
Smaller target sequences are useful in attempts to
distinguish organisms that differ only by very small
regions of DNA (e.g., a mutant strain that differs from
the parent by a single base-pair change). This distinction is possible because hybridization conditions can
be established such that 100% homology is required
for hybridization to occur; i.e., such that a match of
19 out of 20 bases is not sufficient for hybridization
(Wallace and Miyada 1987).
The probes used in nucleic-acid hybridization experiments are usually labelled with radioactive (usually
32P) deoxynucleotides. Perhaps best known is labelling
of double-stranded DNA by nick translation, in which
labelled deoxynucleotides are inserted in the DNA
strands by the combined exonuclease and polymerase
activities of DNA polymerase I (Davis et al. 1980,
Maniatis et al. 1982, Ausubel et al. 1987). There now
exist a number of alternative methods of producing
probes for hybridization analyses. We have developed
single-stranded DNA probes of high specific activity,
produced in a primer extension reaction from a singlestranded DNA template using 32Pdeoxynucleotides in
a reaction analogous to the dideoxy DNA-sequencing
reaction (Holben et al. 1988). These probes are extremely specific and quite sensitive, allowing us to detect as few as 103 gene copies per gram in the soil
bacterial community. R. Atlas and his co-workers have
successfully used 32P-labelled RNA probes synthesized
with RNA polymerase to detect specific organisms in
river sediments on the basis of the detection of a DNA
sequence unique to these organisms (R. Atlas, personal
communication).
There also exist methods of labelling probes with
nonradioactive residues such as biotin or by sulfonation of cytosine residues in DNA. The advantages of
using nonradioactive probes are that they are more
stable (one can store probes for later use) and do not
entail exposure to 32P. Nonradioactive probes, however, do not currently yield the levels of specificity and
Ecology Vol. 69, No. 3
sensitivity of detection that are possible with radioactive probes (Zwadyk et al. 1986).
Varying conditions (such as salt concentration and
temperature) of hybridization and subsequent washing
away of unhybridized probe can control the specificity
of the hybridization reaction. Thus a single probe can
be used in either a homologous or a heterologous fashion to detect either identical or somewhat dissimilar
organisms. Although one can generally predict the
specificity of a given probe by controlling reaction conditions and by selecting a gene (or other sequence) that
is not widely distributed in nature, proper screening of
the probe against other organisms likely to be encountered in the samples used and the use of appropriate
positive and negative controls for hybridization experiments are important to insure specificity.
Detection and quantification of hybridization
There are two major formats for hybridization analyses. In solution hybridization, both the denatured nucleic acid being analyzed and the probe are free in
solution during the hybridization reaction. Following
hybridization, unhybridized DNA and probe (i.e., single-stranded nucleic acids) are removed either by differential binding to hydroxyapatite (Britten et al. 1974)
or by digestion with the nuclease S1 (which preferentially degrades single-stranded nucleic acids), and the
remaining double-stranded hybrids collected by trichloroacetic acid precipitation (Chelm 1982a). Results
obtained by solution hybridization with radio-labelled
probe are generally quantified by liquid scintillation
counting.
In filter hybridization, the denatured nucleic acid
being analyzed is fixed to a filter (generally either cellulose nitrate or nylon membrane) prior to hybridization with probe. The simplest format for filter hybridization is the "dot blot" or "slot blot," in which
denatured nucleic acid is applied to the filter in discrete
dot- or slot-shaped areas with a simple manifold apparatus (e.g., the "Minifold" apparatus, Schleicher and
Schuel, Keene, New Hampshire). Slightly more complicated, but analytically more powerful, filter hybridization protocols are the Southern blot (for DNA) and
the northern blot (for RNA). In Southern blot analysis,
double-stranded DNA fragments resulting from restriction endonuclease digestion are separated by size
by agarose gel electrophoresis, then denatured in the
gel and transferred to filters for subsequent hybridization; the spatial orientation of each fragment is maintained (Southern 1975, David et al. 1980, Maniatis et
al. 1982). In northern blot analysis, mixtures of RNA
molecules (e.g., total RNA or messenger RNA), are
size fractionated by electrophoresis in agarose gels under denaturing conditions, then transferred to filters
SPECIAL FEATURE-NEW
June 1988
E g
E
i
i
MICROBIAL ECOLOGY
567
i
......
.....
for hybridization (Maniatis et al. 1982, Ausubel et al.
1987).
The analytical power of these procedures derives from
the initial size fractionation step. When two or more
species contain the same gene, it will probably either
have a slightly different sequence and/or be flanked by
a different sequence. Thus, if the DNA obtained from
a mixture of organisms (as in an environmental sample) is digested with an appropriate restriction endonuclease, the different-sized fragments, of different
species, will be detected as different bands of hybridization. This concept should also allow the detection
of genetic rearrangements and horizontal gene transfer.
By combining probes specific for each organism of
interest, each of which hybridizes to a restriction fragment of a different size, one should be able to detect
multiple organisms in a single sample without the need
for a probe that hybridizes to a sequence common to
each organism. Data from filter hybridization are
quantified by densitometry or liquid scintillation
counting of areas of filters where radioactive probe has
hybridized.
The use of nucleic acid hybridization methods in
microbial ecology is still in its infancy, but when one
considers that virtually all of the ecological studies we
have described have been accomplished in the last 2
yr, the progress is impressive and, if this rate continues,
suggests that additional major advances in the immediate future are likely. At present, most of the work
has involved the development of methods and illustration of their usefulness; there has been no time yet
to reveal new ecological phenomena. The value of these
methods will only be established when the latter is
achieved.
LITERATURE CITED
Ausubel, F. M., R. Bent, R. E. Kingston, D. D. Moore, J. A.
Smith, J. G. Seidman, and K. Struhl. 1987. Current protocols in molecular biology. Greene Publishing and WileyInterscience, New York, New York, USA.
Bakken, L. R. 1985. Separation and purification of bacteria
from soil. Applied and Environmental Microbiology 49:
1482-1487.
Barkay, T., and B. H. Olson. 1986. Phenotypic and genotypic adaptations of aerobic heterotrophic sediment bacterial communities to mercury stress. Applied and Environmental Microbiology 52:403-406.
Bender, C. L., and D. A. Cooksey. 1986. Indigenous plasmids in Pseudomonas syringae pv. tomato: conjugative
transfer and role in copper resistance. Journal of Bacteriology 165:534-541.
Berger, S. L., and A. R. Kimmel, editors. 1987. Guide to
molecular cloning techniques. In S. P. Colowick and N. 0.
Kaplan, editors in chief. Methods in enzymology. Volume
152. Academic Press, New York, New York, USA.
Bone, T. L., and D. L. Balkwill. 1986. Improved flotation
technique for microscopy of in situ soil and sediment microorganisms. Applied and Environmental Microbiology
51:462-468.
ii..
iA
...E E.....
. ... ....
iEE
Britten, R. J., D. E. Graham, and B. R. Neufeld. 1974. Analysis of repeating DNA sequences by reassociation. Methods
in Enzymology 29:363-406.
Carlson, T. A., M. L. Guerinot, and B. K. Chelm. 1985.
Characterization of the gene encoding glutamine synthetase
I (glnA) from Bradyrhizobium japonicum. Journal of Bacteriology 162:698-703.
Chelm, B. K. 1982a. DNA-DNA renaturation kinetics. Pages
301-313 in M. Edelman, R. B. Hallick, and H.-H. Chua,
editors. Methods in chloroplast molecular biology. Elsevier
Biomedical, New York, New York, USA.
1982b. Analysis of chloroplast DNA transcription
by RNA driven hybridization kinetics. Pages 551-557 in
M. Edelman, R. B. Hallick, and H.-H. Chua, editors. Methods in chloroplast molecular biology. Elsevier Biomedical,
New York, New York, USA.
Davis, R. W., D. Botstein, and J. R. Roth. 1980. Advanced
bacterial genetics. Cold Spring Harbor Laboratory, Cold
Spring Harbor, New York, USA.
DeFlaun, M. F., J. H. Paul, and D. Davis. 1986. Simplified
method for dissolved DNA determination in aquatic environments. Applied and Environmental Microbiology 52:
654-659.
Echeverria, P., and L. Haanstra. 1982. Detection of enterotoxigenic Escherichia coli in water by filter hybridization
with three enterotoxin gene probes. Journal of Clinical Microbiology 16:1086-1090.
Faegri, A., V. L. Torsvik, and J. Goksoyr. 1977. Bacterial
and fungal activities in soil: separation of bacteria by a rapid
fractionated centrifugation technique. Soil Biology and Biochemistry 9:105-112.
Fitts, R., M. Diamond, C. Hamilton, and M. Neri. 1983.
DNA-DNA hybridization assay for detection of Salmonella
spp. in foods. Applied and Environmental Microbiology
46:1146-1151.
Fredrickson, J. K., D. F. Bezdick, F. E. Brockman, and S. W.
Li. 1988. Enumeration of Tn5 mutant bacteria in soil by
using a most-probable-number DNA hybridization procedure and antibiotic resistance. Applied and Environmental Microbiology 54:446-453.
Golub, E. I., and K. B. Low. 1986. Unrelated conjugative
plasmids have sequences which are homologous to the leading region of the F factor. Journal of Bacteriology 166:670672.
Harada, F., A. Kimura, T. Iwanaga, K. Shimozawa, J. Yata,
and T. Sasazuki. 1987. Gene conversion-like events cause
steroid 21 -hydroxylase deficiency in congenital adrenal hyperplasia. Proceedings of the National Academy of Sciences
(USA) 84:8091-8094.
Hill, W. E., J. M. Madden, B. A. McCordell, D. B. Shah, J.
A. Jagow, W. I. Payne, and B. K. Bouton. 1983. Foodborne enterotoxigenic Escherichia coli: detection and enumeration by DNA colony hybridization. Applied and Environmental Microbiology 45:1324-1330.
Holben, W. E., J. K. Jansson, B. K. Chelm, and J. M. Tiedje.
1988. DNA probe method for the detection of specific
microorganisms in the soil bacterial community. Applied
and Environmental Microbiology 54:703-711.
Maniatis, T., E. F. Firtsch, and J. Sambrook. 1982. Molecular cloning; a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, USA.
McClung, C. R., and B. K. Chelm. 1987. Genetic locus for
formate dependent growth of Bradyrhizobium japonicum.
Journal of Bacteriology 169:3260-3267.
Ogram, A., G. A. Sayler, and T. Barkay. 1987. The extraction and purification of microbial DNA from sediments.
Journal of Microbiological Methods 7:57-66.
568
f
Ecology Vol. 69, No. 3
MICROBIAL ECOLOGY
SPECIAL FEATURE-NEW
;
L
i
~i. FF~ if~
f:;
;
E
iE~
iE
ii
S
f
Pace, N. R., D. A. Stahl, D. J. Lane, and G. J. Olsen. 1985.
Analyzing natural microbial populations by rRNA sequences. American Society for Microbiology News 51:412.
Palva, A. 1985. Nucleic acid spot hybridization for detection
of Chlamydia trachomatis. Federation of European Microbiological Societies Letters 28:85-91.
Pettigrew, C. A., and G. S. Sayler. 1986. The use of DNA:
DNA colony hybridization in the rapid isolation of 4-chlorobiphenyl degradative bacterial phenotypes. Journal of
Microbiological Methods 5:205-213.
Salyers, A. A., S. P. Lynn, and J. F. Gardner. 1983. Use of
randomly cloned DNA fragments for identification of Bacteroides thetaiotamicron. Journal of Bacteriology 154:287293.
Sayler, G. S., M. S. Shields, E. T. Tedford, A. Breen, S. W.
Hooper, K. M. Sirotkin, and J. W. Davis. 1985. Application of DNA:DNA colony hybridization to the detection
of catabolic phenotypes in environmental samples. Applied
and Environmental Microbiology 49:1295-1303.
Shapiro, J. A., editor. 1983. Mobile genetic elements. Academic Press, New York, New York, USA.
Sharrock, R. A., R. L. Gourse, and M. Nomura. 1985. Inhibitory effect of high-level transcription of the bacteriophage lambda nutL region on transcription of rRNA in
Escherichia coli. Journal of Bacteriology 163:704-708.
.D
;f~j~i
~iiiE::fi
.iii~ff;:#i~if;
if
eoff
..fi:F~ii
.S?
.5f:EE-i;
A.
-fFf
S
i...SFF:
:f-f~f-fy
-HE
j
-vf!:!...:i
Southern, E. 1975. Detection of specific sequences among
DNA fragments separated by gel electrophoresis. Journal
of Molecular Biology 98:503-517.
Stahl, D. A., B. Flescher, H. R. Mansfield, and L. Montgomery. In press. Use of phylogenetically-based hybridization
probes for studies of ruminal microbial ecology. Applied
and Environmental Microbiology.
Tilghman, S. M., D. C. Tiemeier, J. G. Seidman, B. M. Peterlin, M. Sullivan, J. V. Maizel, and P. Leder. 1978. Intervening sequence of DNA identified in the structural portion of a mouse B-globin gene. Proceedings of the National
Academy of Sciences (USA) 75:725-729.
Wallace, R. B., and C. G. Miyada. 1987. Oligonucleotide
probes for the screening of recombinant DNA libraries.
Methods in Enzymology 152:432-442.
Williamson, V. M., E. T. Young, and M. Ciriacy. 1981.
Transposable elements associated with constitutive expression of yeast alcohol dehydrogenase II. Cell 23:605-614.
Yates, J. R., J. H. Lobos, and D. S. Holmes. 1986. The use
of genetic probes to detect microorganisms in biomining
operations. Journal of Industrial Microbiology 1:129-135.
Zwadeyk, P., Jr., R. C. Cooksey, and C. Thornsberry. 1986.
Commercial detection methods for biotinylated gene probes:
comparisons with 32P-labeled DNA probes. Current Microbiology 14:95-100.