In: Biology of Plant-Microbe Interaction;
Proceedings of the 8th International Congress of Molecular Plant-Microbe
Interactions (G. Stacey, B. Mullin and P. Gresshoff, Eds.) APS
Press, 497-502 (1996).
The utility of a recently developed method to classify
bacteria on the basis of their genomic fingerprint patterns was
investigated, using collections of both symbiotic and pathogenic
plant-associated bacteria. The genomic fingerprinting method
employed is based on the use of DNA primers corresponding to naturally
occurring sed repetitive elements in bacteria, such as
the REP, ERIC and BOX elements, and the PCR reaction (rep-PCR).
We have been able to show that rep-PCR fingerprinting is a highly
reproducible and simple method to distinguish closely related
strains, to deduce phylogenetic relationships between strains
and to study their diversity in a variety of ecosystems. We have
also used computer assisted pattern analysis programs (e.g. GelCompar)
for microbial identification and phylogenetic analysis of complex
data sets, and describe the application of these methods for the
creation of databases for bacterial diagnosis.
The identification and classification of symbiotic and pathogenic plant-associated bacteria are important both in terms of their agricultural applications, as well as for basic studies on plant-microbe interactions. A variety of phenotypic methods has been traditionally used to type these bacteria, including serotyping, phage typing, microscopic identification, substrate utilization screening (e.g. BIOLOG), multilocus enzyme electrophoresis (MLEE), fatty acid methyl ester analysis (FAME), 2-D PAGE of total proteins, and intrinsic antibiotic resistance profiling.
In addition, in the case of symbiotic bacteria,
such as rhizobia, plant nodulation tests are performed, and in
the case of pathogenic bacteria, pathogenicity tests on specific
hosts are carried out. The majority of these techniques require
purification and cultivation of the bacteria and/or can be quite
laborious and time consuming.
More recently, DNA-based (genotypic) approaches have increasingly been applied to microbial identification and classification. In fact, these "molecular" approaches have resulted in the birth of a new ecology subspecialty, Molecular Microbial Ecology (see Akkermans et al. 1995). Generally, these methods tend to be less dependent on bacterial growth variables, more 'stable', less time-consuming and are very useful for determining phylogenetic relationships between microbial isolates and for assigning strains into specific groups. Such methods include DNA-DNA hybridization studies (which constitute the basis for bacterial genus/species designations), characterization of rRNA sequences, and other methodologies, including digestions of total genomic DNA with infrequently (rare) or frequently cutting restriction enzymes, followed by pulse field gel electrophoresis (PFGE) or hybridization with specific probes (RFLP), respectively, and plasmid profiling (see Louws et al. 1996 for a discussion of the relative utilities of these techniques and references).
Clearly, the utility of DNA-based approaches has
been enhanced tremendously by the application of PCR. A most
useful, application of PCR to bacterial identification and classification
has been in the area of genomic fingerprinting using, for example,
primers corresponding to endogenous interspersed repetitive
sequences (Lupski and Weinstock, 1992; de Bruijn, 1992; Versalovic
et al. 1991). The latter approach has been named
fingerprinting and will be discussed here.
As mentioned above, naturally occurring interspersed
repetitive DNA elements, found in many (if not all) bacteria,
can serve as primer sites for genomic DNA amplification (Versalovic
et al. 1991;1994; de Bruijn 1992). Several families of repetitive
sequences are interspersed throughout the genome of diverse bacterial
species (see Lupski and Weinstock 1992). Three families of repetitive
sequences have been studied in most detail, including the 35-40
bp repetitive extragenic palindromic (REP) sequence, the 124-127
bp enterobacterial repetitive intergenic consensus (ERIC) sequence,
and the 154 bp BOX element (see Versalovic et al. 1994). These
sequences appear to be located in distinct, intergenic positions
all around the chromosome. The repetitive elements may be present
in both orientations on the chromosome, and PCR primers have been
designed to "read outward" from the inverted repeats
in REP and ERIC, and from the boxA subunit of BOX (Versalovic
et al. 1994). The use of the above primer(s) and PCR leads to
the selective amplification of distinct genomic regions located
between REP, ERIC or BOX sequences. The corresponding protocols
are referred to as REP-PCR, ERIC-PCR and BOX-PCR, respectively,
and rep-PCR collectively (Versalovic et al. 1991;1994).
Amplified bands are size fractionated through a gel matrix to
yield fingerprint patterns resembling "bar codes" (Lupski,
1993), analogous to UPC codes used in grocery stores, and function
as a signature for specific bacterial strains (see Figure 1).
Figure 1. General rep-PCR genomic fingerprinting
protocol using different templates.
For the identification and classification of rhizobial
strains from soils or from nodules induced on plants, or pathogenic
bacteria from lesions on plant leaves and fruits, it is particularly
useful to have a rapid, simple and highly reproducible method
that is not absolutely dependent on purified DNA. We, and others,
have been able to show that the rep-PCR fingerprinting patterns
of purified rhizobial DNA, single colonies from a plate, liquid
cultures or direct extracts of nodule tissues are identical in
most cases (de Bruijn et al. 1992; Nick and lindstrom 1994; Schneider
and de Bruijn 1996). The same applies to purified DNA versus whole
cells and extracts from plant lesions caused, for example, by
xanthomonads (Louws et al. 1994; 1995; 1996). Thus, rep-PCR can
be carried out with different templates (see Figure 1).
The banding patterns generated by rep-PCR genomic
fingerprinting analysis of large collections of strains are too
complex to analyse by eye. Therefore, we have tested the utility
of several computer assisted banding pattern determination and
phylogenetic analysis programs.
We previously reported the use of the AMBIS System (Scanalytics, Waltham, Mass. USA) for phylogenetic analysis of rep-PCR generated genomic fingerprints of different rhizobia (Rossbach et al. 1995; Schneider and de Bruijn 1996).
More recently, we have been using the GelCompar system (Applied Math, Kortrijk, Belgium) for our analyses, since it appears to be more user friendly, more versatile and allows the construction/screening of rep-PCR generated "bar codes" for diagnostic purposes (see below), as well as for phylogenetic analyses.
We have applied these analyses to the identification and classification of plant pathogenic xanthomonads (e.g. Louws et al. 1994;1995). In order to test our hypothesis that rep-PCR generated fingerprints directly reflect genomic structure, we carried out a cluster analysis of 19 Xanthomonas strains belonging to 6 homology groups (Vauterin et al. 1995). Using the GelCompar program, we analysed the combined REP, ERIC and BOX fingerprints of the 19 strains (kindly provided by Drs. Swings and Vauterin, University of Ghent, Belgium) in duplicate. The results of this analysis is shown in Figure 2 and reveals that: 1. Virtually all duplicates ended up alligned next to eachother; 2. The rep-PCR generated groupings correspond directly to the DNA homology groups (numbered on top); 3. Even very closely related strains can be distinguished by this method.
We are also interested in being able to use the
rep-PCR fingerprinting method as a tool for strain diagnosis.
For this purpose, we are constructing a database of rep-PCR patterns
of a large collection (600-800) of Xanthomonas isolates
using the GelCompar (library search) program, in collaboration
with the laboratory of Prof. Swings (Ghent, Belgium). A limited
database has been constructed and pilot searches with unknown
input strains conducted. The fingerprint pattern of an "unknown"
(X. cassavae) isolate was run through the library of fingerprints,
and the four most closely related patterns were found to be indeed
from X. cassavae strains present in the database.
In conclusion, we believe that rep-PCR genomic fingerprinting coupled to computer assisted phylogenetic analysis and library search programs will constitute a useful method for the identification or diagnosis of plant pathogenic, as well as symbiotic bacteria (see also Versalovic et al. 1994; Schneider and de Bruijn 1996; Louws et al. 1996 and references contained therein). This technique has already been successfully used to study the
population structure of plant pathogens, to follow
green house infections, to examine nodule occupancy, to identify
(brady)-rhizobial strains that could not be distinguished by any
other method, and to carry out phylogenetic analyses of world-wide
collections of rhizobia and plant pathogens (for references see
Versalovic et al., 1994; Louws et al. 1996). It therefore is
another useful molecular approach in molecular microbial ecology.
Figure 2. Cluster analysis of 19 Xanthomonas
strains belonging to 6 DNA homology groups using combined REP,
BOX and ERIC fingerprints (Pearson correlation; UPMGA).
Akkermans, A.D.L., van Elsas, J.D., and de Bruijn, F.J. 1995. Molecular Microbial Ecology Manual. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 1-488
de Bruijn, F.J 1992. Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergenic consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Appl. Environ. Microbiol. 58: 2180-2187
Louws, F.J., Fulbright, D.W., Stephens, C.T. and de Bruijn, F.J. 1994. Specific genomic fingerprints of phytopathogenic Xanthomonas and Pseudomonas pathovars and strains generated with repetitive sequences and PCR. Appl. Environ. Microbiol. 60:2286-2295.
Louws, F.J., Fulbright, D.W., Stephens, C.T., and de Bruijn, F.J. 1995. Differentiation of genomic structure by rep-PCR fingerprinting to rapidly classify Xanthomonas campestris pv. vesicatoria. Phytopath. 85:528-836.
Louws, F.J., Schneider, M., and de Bruijn, F.J. 1995. Assessing genetic diversity of microbes using repetitive-sequence-based PCR (rep-PCR). In: Toranzos, G., Ed., Nucleic Acid Amplification Methods for the Analysis of Environmental Samples. Technomic Publishing Co., In Press.
Lupski, J.R. 1993. Molecular epidemiology and its clinical application. JAMA 270:1363-1364.
Lupski, J.R. and Weinstock, G.M.. 1992. Short, interspersed repetitive DNA sequences in prokaryotic genomes. J. Bacteriol. 174:45254529.
Nick, G. and Lindstrom, K. 1994. Use of repetitive sequences and the polymerase chain reaction to fingerprint the genomic DNA of Rhizobium galegae strains and to identify the DNA obtained by sonicating the liquid cultures and root nodules. Syst. Appl. Microbiol. 17:265-273.
Rossbach, S.R., Rasul, G., Schneider, M., Eardley, B., and de Bruijn, F.J. (1995) Structural and functional conservation of the rhizopine catabolism (moc) locus is limited to selected Rhizobium meliloti strains and unrelated to their geographical origin. Mol. Plant-Microbe Interact. 8: 549-559
Schneider,M., and de Bruijn, F.J. 1996. Rep-PCR-mediated genomic fingerprinting of rhizobia and computer-assisted phylogenetic pattern analysis. World J. Microbiol. and Biotechnol., 12: 163-174.
Vauterin, L., Hoste, B., Kersters, K., and Swings, J. 1995. Reclassification of Xanthomonas. Intl. J. Systematic Bacteriol. 45: 472-489
Versalovic, J., Koeuth, T., and Lupski, J.R. 1991. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucl. Acids Res. 19: 6823-6831
Versalovic, J., Schneider, M., de Bruijn, F.J., and
Lupski, J.R. 1994). Genomic fingerprinting of bacteria using repetitive
sequence based PCR (rep-PCR). Meth.Cell. Mol. Biol. 5: 25-40