You are on page 1of 33

Polyphasic taxonomy, a consensus approach to bacterial systematics.

P Vandamme, B Pot, M Gillis, P de Vos, K Kersters and J Swings Microbiol. Rev. 1996, 60(2):407.

Updated information and services can be found at: http://mmbr.asm.org/content/60/2/407 Downloaded from http://mmbr.asm.org/ on January 3, 2012 by guest These include:
CONTENT ALERTS

Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite this article), more»

Information about commercial reprint orders: http://mmbr.asm.org/site/misc/reprints.xhtml To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/

MICROBIOLOGICAL REVIEWS, June 1996, p. 407–438 0146-0749/96/$04.00 0 Copyright 1996, American Society for Microbiology

Vol. 60, No. 2

Polyphasic Taxonomy, a Consensus Approach to Bacterial Systematics
P. VANDAMME, B. POT, M. GILLIS, P. DE VOS, K. KERSTERS,
AND

J. SWINGS*

Laboratorium voor Microbiologie, Universiteit Gent, Ghent, Belgium INTRODUCTION .......................................................................................................................................................408 POLYPHASIC TAXONOMY.....................................................................................................................................408 Different Types of Information Used in Bacterial Polyphasic Taxonomy ......................................................409 Genotypic Methods .................................................................................................................................................409 Determination of the DNA base ratio (moles percent G C).......................................................................409 DNA-DNA hybridization studies.......................................................................................................................409 rRNA homology studies .....................................................................................................................................410 DNA-based typing methods ...............................................................................................................................411 Phenotypic Methods ...............................................................................................................................................412 Classical phenotypic analyses ...........................................................................................................................412 Numerical analysis .............................................................................................................................................413 Automated systems .............................................................................................................................................413 Typing methods ...................................................................................................................................................413 Cell wall composition .........................................................................................................................................413 Cellular fatty acids .............................................................................................................................................413 Isoprenoid quinones ...........................................................................................................................................413 Whole-cell protein analysis................................................................................................................................413 Polyamines ...........................................................................................................................................................413 Pyrolysis mass spectrometry, Fourier transformation infrared spectroscopy, and UV resonance Raman spectroscopy .......................................................................................................................................414 EVALUATION OF POLYPHASIC TAXONOMY ..................................................................................................414 Polyphasic Taxonomy of the Genus Xanthomonas..............................................................................................414 DNA-DNA hybridization studies.......................................................................................................................414 16S rRNA sequences...........................................................................................................................................414 DNA base ratio....................................................................................................................................................415 Numerical analysis of phenotypic features .....................................................................................................415 Monoclonal antibodies .......................................................................................................................................415 Whole-cell protein analysis................................................................................................................................415 Cellular fatty acid analysis................................................................................................................................415 Xanthomonadins .................................................................................................................................................415 Polyamines ...........................................................................................................................................................415 Conclusions..........................................................................................................................................................415 Polyphasic Taxonomy of the Genus Campylobacter............................................................................................415 rRNA homology studies .....................................................................................................................................416 DNA-DNA hybridization studies.......................................................................................................................416 DNA base ratio....................................................................................................................................................417 Classical phenotypic characteristics ................................................................................................................417 Respiratory quinone components .....................................................................................................................417 Cellular fatty acid analysis................................................................................................................................417 Protein analysis...................................................................................................................................................417 Conclusion ...........................................................................................................................................................417 Polyphasic Taxonomy of Lactic Acid Bacteria....................................................................................................418 Phylogenetic analysis based on rRNA homology............................................................................................418 (i) L. delbrueckii group ...................................................................................................................................418 (ii) L. casei-Pediococcus group .......................................................................................................................418 (iii) Leuconostoc group....................................................................................................................................420 (iv) Other lactobacilli.....................................................................................................................................420 Delineation of Lactobacillus species by traditional phenotypic tests ...........................................................420 (i) Obligately homofermentative lactobacilli (group A) ............................................................................420 (ii) Facultatively heterofermentative lactobacilli (group B) .....................................................................420 (iii) Obligately heterofermentative lactobacilli (group C) ........................................................................420 Delineation of Lactobacillus species by DNA-DNA hybridization studies ...................................................420
* Corresponding author. Mailing address: Laboratorium voor Microbiologie, K. L. Ledeganckstraat 35, B-9000 Ghent, Belgium. Phone: 32.9.264.5116. Fax: 32.9.264.5346. Electronic mail address: Jean.Swings @rug.ac.be. 407

Downloaded from http://mmbr.asm.org/ on January 3, 2012 by guest

408

VANDAMME ET AL.

MICROBIOL. REV.

(i) L. delbrueckii group ...................................................................................................................................421 (ii) L. casei-Pediococcus group .......................................................................................................................421 DNA base ratio....................................................................................................................................................421 rRNA-targeted oligonucleotide probes .............................................................................................................421 Whole-cell protein analysis................................................................................................................................421 Lactate dehydrogenase .......................................................................................................................................421 Cell wall components..........................................................................................................................................422 Conclusion ...........................................................................................................................................................422 Polyphasic Taxonomy of the Family Comamonadaceae......................................................................................422 rRNA similarities................................................................................................................................................423 DNA-DNA hybridizations ..................................................................................................................................423 Amplified rDNA restriction analysis................................................................................................................424 DNA base ratio....................................................................................................................................................424 Whole-cell protein analysis................................................................................................................................424 Numerical analysis of phenotypic features .....................................................................................................424 Immunotyping .....................................................................................................................................................424 Cellular fatty acid analysis................................................................................................................................425 Polyamine patterning .........................................................................................................................................425 Other features .....................................................................................................................................................425 Conclusion ...........................................................................................................................................................425 EVOLUTION OF POLYPHASIC TAXONOMY AND PERSPECTIVES............................................................425 DNA Hybridization Studies ...................................................................................................................................425 rRNA Sequence Analysis .......................................................................................................................................426 Phenotypic Data ......................................................................................................................................................427 Whole-Cell Fatty Acid Analysis.............................................................................................................................428 Whole-Cell Protein Analysis..................................................................................................................................428 DNA-Based Typing Methods .................................................................................................................................429 Strategy in Polyphasic Taxonomy.........................................................................................................................429 Polyphasic Identification........................................................................................................................................429 Unculturable Bacteria ............................................................................................................................................429 Population Genetics................................................................................................................................................430 Perspectives and Conclusions ...............................................................................................................................430 ACKNOWLEDGMENTS ...........................................................................................................................................431 REFERENCES ............................................................................................................................................................431 INTRODUCTION For a long time, bacterial taxonomy was considered one of the dullest fields in microbiology, not immediately the preferred discipline of young or ambitious scientists. Recent developments have changed this attitude, mainly because of the spectacular developments witnessed in the last 10 years in the field of sequencing of rRNA and genes coding for rRNA (rDNA) and their contribution to bacterial phylogeny and in molecular fingerprinting techniques. These techniques revolutionized our insights in the phylogeny and taxonomy of all living organisms. Taxonomy of bacteria finally also could be assigned a place in phylogeny. Another development of bacterial taxonomy, polyphasic taxonomy, arose 25 years ago and is aiming at the integration of different kinds of data and information (phenotypic, genotypic, and phylogenetic) on microorganisms and essentially indicates a consensus type of taxonomy. The term “polyphasic taxonomy” was coined by Colwell (45) and is used for the delineation of taxa at all levels (219). Also, the terms “polyphasic classification” and “polyphasic identification” can be validly used in this context. The recent developments of polyphasic taxonomy and phylogeny clearly constitute milestones in modern bacterial taxonomy. There will never be a definitive classification of bacteria. But let us be clear: this is not meant as a “fin de siecle” pessimistic statement of postmodernists among ` the choir of jubilating optimists! It is only the illustration of a rule, valid in all experimental sciences, stating that scientific progress is linked to and made possible through technological progress. In the three parts of the present contribution on polyphasic taxonomy, we will successively (i) discuss the types of information used, (ii) illustrate the practice of polyphasic taxonomy in a selected number of cases, and (iii) discuss its problems and future developments. For overviews of modern taxonomic theory and practice, we refer to recent handbooks, e.g., by Priest and Austin (257), Goodfellow and O’Donnell (112), Towner and Cockayne (304), and Logan (187), and general works, e.g., The Prokaryotes (8) and Bergey’s Manual of Systematic Bacteriology (172). POLYPHASIC TAXONOMY Taxonomy is generally taken as a synonym of systematics or biosystematics and is traditionally divided into three parts: (i) classification, i.e., the orderly arrangement of organisms into taxonomic groups on the basis of similarity; (ii) nomenclature, i.e., the labelling of the units defined in (i); and (iii) identification of unknown organisms, i.e., the process of determining whether an organism belongs to one of the units defined in (i) and labeled in (ii) (51, 292). Two additional parts are needed to completely define modern biosystematics: phylogeny and population genetics. In the last decade, it became generally accepted that bacterial classification should reflect as closely as possible the natural relationships between bacteria, which are the phylogenetic relationships as encoded in 16S or 23S rRNA sequence data (363). The species is the basic unit of bacterial taxonomy (343) and is defined as a group of strains, including the type strain, sharing 70% or greater DNA-DNA relatedness with 5 C or

Downloaded from http://mmbr.asm.org/ on January 3, 2012 by guest

The latter view implies that classification is a frame for the condensed nodes where some isolated internodal strains must also get a (provisional) place and name. It varies between 24 and 76% in the bacterial world. It comprises the major categories of taxonomic techniques required to study bacteria at different taxonomic levels and will roughly describe their general concept and applications. it therefore remains impossible to convert a percent DNA-binding or DNA-DNA hybridization value into a percentage of whole-genome sequence similarity. but primarily because our present view on classification is that it should reflect the natural relationships as encoded in the DNA. and a wide range of other expressed features (Fig. in general.2% for each 1% of mispairing (13. 306). genus. and phylogenetic information may be incorporated in polyphasic taxonomy. Others. i. however. the optical renaturation rates method (57).. we are only substantiating our own dogma. The most similar relative binding ratios were obtained by the S1-DE81 procedure and the hydroxyapatite method when hybridized at 75 C (120. Staley and Krieg (292) defended a much more vague species concept that consisted of the type strain and all other strains considered sufficiently similar to this type to warrant their inclusion within a single species. asked about their ideas about the bacterial species. It is. whereas phenotypic information is derived from proteins and their functions. which allow us to compare and group large numbers of strains in a minimal period. Several of the methods described briefly below (e. and family. Different Types of Information Used in Bacterial Polyphasic Taxonomy In principle. and it is often not clear whether if hybridizations are performed under optimal.. these methods presently dominate modern taxonomic studies as a consequence of technological progress. whereas DNADNA hybridization studies. because they are laborious. 248) is an indirect parameter of the sequence similarity between two entire genomes. 1996 POLYPHASIC TAXONOMY 409 less Tm (Tm is the melting temperature of the hybrid as determined by stepwise denaturation. Loosening the 70% rule often allows a compromise between the two views (see below). The number of different molecules which have been applied in taxonomic studies is large. 287. The percent DNA binding (57) or the DNA-DNA hybridization value or the relative binding ratio (21. 2. were performed on a limited number of taxa only. multilocus enzyme electrophoresis. it is of primary interest to understand at which level these methods carry information and to realize their technical complexity. Undoubtedly. stringent. The designated type strain of a species serves as the name bearer of the species and as the reference specimen (292). or they visualize bacterial species as condensed nodes in a cloudy and confluent taxonomic space. because different methods are used to determine the level of DNA-DNA hybridization.VOL. The species is certainly the most important and at the same time the central element of bacterial taxonomy. The hydroxyapatite method and the two procedures of the S1 nuclease method allow determination of the Tm and have been compared (121). 1). highly debatable whether data which were obtained with short oligonucleotides and experimentally induced mispairing can be extrapolated to entire genomes. and serological analyses are not useful for phylogenetic studies.g.e. an alternative phylogenetic species concept could delineate a species in a phylogenetic framework as determined by percent 16S rRNA similarities. Different methods have been described: the hydroxyapatite method (21). different chemotaxonomic markers. however. need considerable amounts of DNA and are timeconsuming.org/ on January 3. It has been shown that the results obtained by these methods give different relative binding ratios but similar Tm values. or suboptimal conditions. Although not available. they are difficult to define and represent agglomerates of nodal species and internodal strains and agglomerates of genera. 152) and are promising to replace the classical methods provided that they are further compared with them under strictly comparable conditions. Genotypic information is derived from the nucleic acids (DNA and RNA) present in the cell. are caught between Scylla and Charybdis: either they stick to a coherent species definition without it necessarily being a biological reality. The list of methods given below is not meant to be complete or to contain a description of all of their aspects. phenotypic. The bacterial species definition given above is founded upon whole genomic DNA-DNA hybridization values (343). the percent DNA-DNA hybridization and the decrease in thermal stability of the hybrid are used to delineate species (343). but it has the inconvenience of not allowing Tm determinations and of being taxonomically insignificant below approximately 30%. Indeed. such as amino acid sequencing. Generally. typing methods such as restriction enzyme patterning. 121) are the most common. however. the amount of time and work required. In fact. respectively. Working one’s way through lists of methods. the range observed is not more than 3% within a well-defined species and not more than 10% within a well-defined genus (288). and the value of 70% DNA relatedness seems only to be indicative rather than absolute (see below). 121. Obviously. 60. for example. a large number of DNA-DNA hybridization protocols have been described. New. As mentioned above. Determination of the DNA base ratio (moles percent G C). all genotypic. time-consuming. Before this definition was generally accepted. These methods do not always give the same (quantitative) results. but the hierarchical structure of taxonomy requires us to consider at least the higher taxa of genus and family. Tm is the difference in Tm in degrees Celsius between the homologous and heterologous hybrids formed under standard conditions [343]). . It has been established that thermal stabilities decrease from 1 to 2. determinations of the moles percent G C content and DNA-DNA hybridization studies) became classic and were applied in taxonomic analyses of virtually all bacteria. At present. and the S1 nuclease method (52.asm. Phenotypic and chemotaxonomic features should agree with this definition. whereas rRNA or protein sequencing is. The taxonomic information level of some of these techniques is illustrated in Fig. DNA-DNA hybridization studies. quick methods consuming less DNA have been described (93. will be restricted to a minimal but representative set of strains. 2012 by guest Genotypic methods are those that are directed toward DNA or RNA molecules. Genotypic Methods Downloaded from http://mmbr. Much more than the species. Practical problems exist. 121). not adequate to type large numbers of strains. or technically demanding or because they were applicable to only one particular taxon. Determination of the moles percent guanosine plus cytosine is one of the classical genotypic methods and is considered part of the standard description of bacterial taxa. Within the present manuscript. Bacterial taxonomists. and their applications as markers are manifold. These classical techniques. Chemotaxonomic methods such as fatty acid analysis are fast methods. all the attention will be focused on the taxonomic ranks of species. The advantage of the optical renaturation rates method is that the DNA needs no label.

optimal conditions for hybridizations are preferred.and two-dimensional. renaturation or hybridization under optimal conditions requires a temperature of 22 to 26 C (mean. 200). and is composed of highly conserved as well as more variable domains (274. DNA-DNA hybridizations are often performed under standard conditions that are not necessarily optimal or stringent for all bacterial DNAs. Generally.org/ on January 3. 1D. randomly amplified polymorphic DNA. sequencing of the rRNA molecules gradually resulted in an rRNA sequence database of 5S .410 VANDAMME ET AL. ARDRA.6 C for each percentage of formamide added to the melting or hybridization mixture (158). PFGE. As a rule. Schematic overview of various cellular components and techniques used. It is now generally accepted that rRNA is the best target for studying phylogenetic relationships because it is present in all bacteria. Later. AFLP.asm. 2D. The stringency of the reaction is determined by the salt and formamide concentrations and by the temperature and the moles percent G C of the DNAs used. Indirect comparison by either hybridization studies (58. 240) or rRNA cataloging of RNase T1-resistant oligonucleotides of 16S rRNA (100. amplified fragment length polymorphism. pulsed-field gel electrophoresis. 2012 by guest FIG. amplified rDNA restriction analysis. see references 56 and 363). one. 24 C) below the melting temperature. REV. Downloaded from http://mmbr. is functionally constant. LMW. The gradual development of new molecular techniques enabled the microbiologist to focus on the comparative study of the rRNA molecules. 101. respectively. 1. 287. measured or calculated at equal salt concentration. 363). MICROBIOL. RFLP. because the optimal temperature curve for hybridization is rather broad (about 5 C). 290) have already revealed the natural relationships with and within a number of bacterial lineages (for reviews. rRNA homology studies. low molecular weight. The components of the ribosome (rRNA and ribosomal proteins) have been the subject of different phylogenetic studies for several decades. The melting temperature (Tm) can be calculated from the salt concentration and the DNA base ratio by using different equations (57. restriction fragment length polymorphism. RAPD. The melting temperature and the optimal hybridization temperature decrease by 0.

antibiotic susceptibility patterning. these techniques are universally applicable (the number of nontypeable strains. Taxonomic resolution of some of the currently used techniques. rRNA (365). 2. whole-genome DNA is extracted and digested with restriction enzymes. Classically. and many others (see below). They provide a phylogenetic framework which serves as the backbone for modern microbial taxonomy. Although several of the techniques listed below do not conform to this general description. 60. if any. 1. Ideally. these techniques have mostly been replaced by direct sequencing of parts or nearly entire 16S or 23S rDNA molecules by using the PCR technique and a selection of appropriate primers. and restriction fragment length polymorphisms are . genotyping has replaced classical typing in many laboratories and will most probably continue to do so (202.asm. 302). In the former. A limited number of 16S rRNA gene sequences became available by direct sequencing after cloning of the genes from the bulk of the DNA (195).org/ on January 3.VOL. During the last few years. International databases comprising all published and some unpub- lished partial or complete sequences have been constructed (66. 1996 POLYPHASIC TAXONOMY 411 Downloaded from http://mmbr. a battery of DNA-directed typing methods has been developed. and they are highly discriminatory. subtyping of species was performed by means of phenotypic analyses such as biochemical (hence biotyping) or serological (hence serotyping) tests. Sequencing of 16S rRNA with conserved primers and reverse transcriptase (176) was a very important advance in bacterial phylogeny and resulted in a spectacular increase in 16S rRNA sequences. taking into account the specific resolution of each method. although the latter method had the important advantage that multiple strains could easily be included. However. which was the first rRNA molecule to be sequenced for numerous bacteria because of its less complex primary and secondary structures. 2012 by guest FIG. DNA-based typing methods. it is obvious that the larger the conserved elements. the more information they bear and the more reliable the conclusions become. The results obtained and the dendrograms constructed with data obtained from the above methods are more or less equivalent. The resulting array of DNA fragments is separated and visualized by agarose gel electrophoresis. The cataloging method and the DNA-rRNA hybridization experiments have gradually disappeared. phage or bacteriocin typing. is mostly small). The first-generation DNA-based typing methods included whole-genome restriction fragment analysis and plasmid DNA analysis. DNA-based typing methods generally refer to techniques which allow us to subdivide species into a number of distinct types. Nowadays. Abbreviations are defined in the legend to Fig. they are reproducible and simple to perform. 226).

a battery of different typing methods was developed. 338) or to tRNA gene fragments (204) may be used as primers. The number of DNA fragments can be reduced by selecting restriction enzymes which only rarely cut DNA. simplicity. In contrast to most other DNA-based methods. it is often treated as a separate unit in taxonomic reviews. Alternatively. therefore. The classical or traditional phenotypic tests are used in identification schemes in the majority of microbiology laboratories. combined randomly. Phenotypic Methods Phenotypic methods comprise all those that are not directed toward DNA or RNA. The latter PCR-based method was reported to allow differentiation at the species (347) and infraspecific (276) levels depending on the stringency of the PCR conditions. it is necessary to establish the identity of plasmids with equal molecular weight. As the introduction of chemotaxonomy is generally considered one of the essential milestones in the development of modern bacterial classification. containing the same sequence as the adapters but extended to include one or more selective bases next to the restriction site of the primer. The fragments. Also. the complex DNA patterns generated after restriction enzyme digestion can be transferred to a membrane and then hybridized with a labeled probe. Among others. The disadvantages of plasmid analysis are obvious. which yield DNA fragments with two different types of sticky ends. Different methods in which short arbitrary sequences were used as primers in the PCR assay were described: oligonucleotides of about 20 bases are used in arbitrarily primed PCR (346). which uses rRNA as probe (119). To these ends. The amplification process results in an array of about 30 to 40 DNA fragments. The rRNA fraction of the profiles allows us to discriminate between some of the major eubacterial groups. 202. The technique has the disadvantage that often very complex patterns of DNA fragments are generated. Apart from the application of tRNA sequences in the PCRbased typing methods mentioned above. which is not unexpected considering the conserved character of the rRNA genes. oligonucleotides of about 10 bases are used in randomly amplified polymorphic DNA analysis (362). The selective amplification reaction is performed by using two different primers. generally known as pulsed-field gel electrophoresis. and most strains often belong to only a few types. established. ribosomal protein S12. which are very difficult to compare. The polymorphisms between the different rRNA operons generate simple arrays of DNA fragments with different lengths (170). These fingerprints comprise the 5S rRNA and the total tRNA pool. and rapidity. The introduction of the PCR methodology into the microbiology laboratory has opened a vast array of applications. while the tRNA fraction reveals more specific taxonomic information (147. Strains do not always contain or keep their plasmids. which appear on one-dimensional gels as a set of bands belonging to three different classes (148). some of the chemotaxonomic methods have been widely applied on vast numbers of bacteria whereas others were so specific that their application was restricted to particular taxa. The restriction is performed by using two restriction enzymes. 2012 by guest . recognizing a specific combination of six to eight bases. The rRNA probe may vary in both the labeling technique and sequence. The amplicon is subsequently digested with a selected combination of restriction enzymes. The term “chemotaxonomy” refers to the application of analytical methods to collect information on various chemical constituents of the cell to classify bacteria. some of which are group specific while others are strain specific (153). As for the other phenotypic and the genotypic techniques. The technique can therefore be used simultaneously for identification purposes and typing purposes. PCR-based DNA-typing methods attracted much interest because of their universal applicability.asm. 16S or 23S rRNA or both. The technique of low-frequency restriction fragment analysis has therefore been dependent on the development of special electrophoretic techniques. 302). The basic principle of AFLP is restriction fragment length polymorphism analysis but with a PCR-mediated amplification to select particular DNA fragments from the pool of restriction fragments. and is now often considered to be the most discriminatory DNA-based typing method (118. Only fragments which completely match the primer sequence are amplified. PCR-based DNA typing was combined with restriction enzyme analysis in the so-called amplified-rDNA restriction analysis method. MICROBIOL. 174. but the general principle has remained the same. 148). or a conserved oligonucleotide part of the rRNA can be used (17). DNA sequences corresponding to elongation factor Tu. and oligonucleotides of about 5 bases are used in DNA-amplified fingerprinting (28). REV. For example. Methods that were elaborated subsequently have reduced the number of DNA fragments compared with the former method and enhanced the reliability and discriminatory power. 331). amplified-rDNA restriction analysis generates mostly species-specific patterns (122. Restriction fragment analysis of plasmids combines the two techniques and generates more simple banding patterns. AFLP screens for amplified fragment length polymorphisms by selective amplification of restriction fragments. Classical phenotypic analyses. phenotypic consistency is required to generate useful classification systems and may therefore influence the depth of a hierarchical line (343). from species and subspecies up to genus and family. Since its initial description. The PCR product. consensus motifs complementary to fragments of repetitive elements dispersed throughout the genomes of gram-positive or gram-negative bacteria (196. The technique is referred to as low-frequency restriction fragment analysis. Alternatively. the tRNA gene pool can be used in a so-called low-molecular-weight RNA profiling method (146). with or without the spacer region. A typical example of one of these developments is the ribotyping method. PCR assays have also been used to amplify the rDNA genes (with or without spacer regions) by means of universal rDNA primers. and flagellar proteins have all been used as probes (116). 258.412 VANDAMME ET AL. are too large to be separated by conventional agarose gel electrophoresis. For such bacteria.org/ on January 3. Another combination of the PCR and restriction enzyme methodologies yielded the AFLP (amplified fragment length polymorphism) technique (372). 155. they also include the chemotaxonomic techniques. however. The paucity of phenotypic characteristics in particular bacterial groups often causes problems in describing or differentiating taxa. short oligonucleotides (adapters) are ligated to form templates for the PCR. is amplified by using universal primers located in the conserved regions of the rRNA genes. While genotypic data are used to allocate taxa on a phylogenetic tree and to draw the major borderlines in classification systems. many variants have been presented. They constitute the basis for the formal description of taxa. alternative chemotaxonomic or genotypic methods are often required to re- Downloaded from http://mmbr. being 16S or 23S rDNA or parts of both genes with or without the spacer region. A typical example concerns the genus Campylobacter and allied bacteria (see below). which allows us to reveal the hybridized fragments.

Numerous studies have revealed a correlation between high similarity in whole-cell protein content and DNA-DNA hybridization (47). possibly. Isoprenoid quinones occur in the cytoplasmic membranes of most prokaryotes and play important roles in electron transport. metabolization of compounds. Very often. allowing us to establish the overall genetic relatedness of bacterial strains. Many of the described typing techniques are suitable for only some organisms and are performed by only a few reference laboratories (202.VOL. 1996 POLYPHASIC TAXONOMY 413 liably identify strains. Gram staining) and colonial (color. 302). Phenotypic data were the first to be analyzed by means of computer-assisted numerical comparison. The lipopolysaccharides present in the outer membranes of gram-negative bacteria can be analyzed by gel electrophoresis. 283). Teichoic acids can easily be extracted and purified (96) and can be analyzed by gas-liquid chromatography (97. 60. The use of SDS-PAGE for general identification purposes is hampered by the fact that it yields only discriminative information at or below the species level. 98).g. many of these characteristics have been shown to be irrelevant as parameters for genetic relatedness. such as sphingophospholipids. The mobility is an indicator for the existence of multiple polymorphisms of the encoding gene (279). but particular fractions such as the polar lipids have also been analyzed (90). Isoprenoid quinones. flagella. Many of the cellular compounds which belong to the bacterial phenotype have been used in typing systems to characterize strains at the infraspecific level. 39). Mostly. and. The comparison of whole-cell protein patterns obtained by highly standardized sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) has proven to be extremely reliable for comparing and grouping large numbers of closely related strains (164. and biochemical features. Typing methods. The classical phenotypic characteristics of bacteria comprise morphological. The outcome of a particular test with a commercial system is sometimes different from that with a classical procedure. Cellular fatty acid methyl ester content is a stable parameter provided that highly standardized culture conditions are used. production of enzymatic activity. The morphology of a bacterium includes both cellular (shape. In the 1950s. In this technique. salt concentrations. Cell walls of gram-positive bacteria. occur in only a restricted number of taxa and were shown to be valuable within these groups (159).. The former can be further subdivided into two main types. yet as a whole. Individually. although a rapid screening method has been proposed (273).org/ on January 3. Multilocus enzyme electrophoresis has been extensively used in population genetics. the phylloquinones. such as complement fixation tests. etc. Data matrices showing the degree of similarity between each pair of strains and cluster analysis resulting in dendrograms revealed a general picture of the phenotypic consistency of a particular group of strains. and hydrogenation) can be used to characterize bacteria at different taxonomic levels (39). The results are interpreted as recommended by the manufacturer and are readily available with a minimal input of time. they provide descriptive information enabling us to recognize taxa. including simple precipitation or agglutination tests and reactions requiring one or more additional components.). pH values. Determination of the cell wall composition has traditionally been important in gram-positive bacteria. inclusion bodies. physiological. Different kinds of serological reactions. form) characteristics. contain various peptidoglycan types. The large variability of the side chains (differences in length. endospore. and addition of a standardized inoculum initiates the reaction (growth. The observation of their universal character Downloaded from http://mmbr. the phenotype of endosymbionts or unculturable bacteria is beyond the reach of our present methods. the naphthoquinones and the benzoquinones. in contrast. Obviously.asm. or atmospheric conditions. Membrane-bound teichoic acid is present in all gram-positive species (7). phenotypic tests must be performed under well-standardized conditions to obtain reproducible results. and data on the presence or activity of various enzymes. Whole-cell protein analysis. cell envelopes. which occur less commonly in bacteria. that analysis of large numbers of phenotypic characteristics was indeed taxonomically relevant. etc. growth in the presence of various substances such as antimicrobial agents. Clearly. numerical taxonomy arose in parallel with the development of computers (285) and allowed comparison of large numbers of phenotypic traits for large numbers of strains. Antigens may be proteins or carbohydrates and may be thermostable or thermolabile. Simple biotyping systems were used which involved a number of tests yielding variable results within species. Automated systems. the total cellular fatty acid fraction is extracted. Cellular fatty acids. whereas cell wall-bound teichoic acid is present in only some gram-positive species (168). by comparing the results of such cluster analyses with those of other taxonomic approaches. 255. More than 300 different chemical structures of fatty acids have been identified. Although the role of polyamines in the bacterial cell is not entirely clear. The method is cheap and rapid and has reached a high degree of automation. saturation. double-bond position. Structural components such as capsules. have been described (142). Fatty acids are the major constituents of lipids and lipopolysaccharides and have been used extensively for taxonomic purposes. Miniaturized phenotypic fingerprinting systems have been introduced and may in the future replace classical phenotypic analyses. e. are distinguished. and it soon became evident. The peptidoglycan type of gram-negative bacteria is rather uniform and provides little information. Numerical analysis. 2012 by guest . highly standardized procedures are required to obtain reproducible results within and between laboratories (see. A variety of lipids are present in bacterial cells. references 227 and 228). flagella. Cell wall composition. but the same is often true for two classical procedures in the same test. oxidative phosphorylation. native enzymes are electrophoretically separated and stained for enzyme activity and their mobilities are compared. The physiological and biochemical features include data on growth at different temperatures. which may be genus or species specific (273). One of the phenotypic typing methods which is still used for various bacteria is multilocus enzyme electrophoresis. As their application in taxonomy is restricted. These systems mostly contain a battery of dehydrated reagents. and substituent groups has proven to be very useful for the characterization of bacterial taxa (298). The procedure is time-consuming. giving typical lipopolysaccharide ladder patterns which are interpreted as variants in the O-specific side chains (74. or fimbriae and intracellular molecules or secretion products such as enzymes and toxins have all been used in serological studies. In addition. Polyamines. Serotyping is based on the presence of variability in the antigenic constituents of the cells (142). 335). and the menaquinones. dimensions. Other types of lipids. active transport (36. they will not be discussed here. Polar lipids are the major constituents of the lipid bilayer of bacterial membranes and have been studied frequently for classification and identification purposes. they seem to be important in bacterial metabolism (299). The variability in chain length. such large numbers of data reflect a considerable amount of genotypic information. Two major structural groups.

maltophilia 1995a . Within the genus Xanthomonas. Downloaded from http://mmbr. X. The classification of the genus Xanthomonas as described in Bergey’s Manual of Systematic Bacteriology (20) underwent a few changes and finally comprised six species: Xanthomonas albilineans. Xanthomonas campes- 1991 . X. on the other hand. X. X. X. This special-purpose classification was designed to meet the practical needs of plant pathologists and was adopted as a provisional solution until classification was established on more generally accepted principles. X.8% 16S rDNA sequence similarity and showed 95.asm..g... or stems (186). codiae. melonis. Pyrolysis mass spectrometry. campestris was composed of over 140 pathovars which have a more or less limited host range but which are phenotypically almost indistinguishable. The latter 16 genomic groups have been described as new species and are composed of one or more former X. X. REV. cassavae. The presumed causes of confusion are the large number of species in. populi. oryzae. represents a biochemically restricted group of bacteria which constitutes an extremely heterogeneous phylogenetic lineage.g. For the family Comamonadaceae. The pathovar name is derived from the name of the host plant. oryzae. Depending on the group of organisms studied.. hyacinthi. and UV resonance Raman spectroscopy are sophisticated analytical techniques which examine the total chemical composition of bacterial cells. During the last 10 years. These methods have been used for taxonomic studies of particular groups of bacteria (198). Impact of polyphasic examination of the taxonomy of the genus Xanthomonas Date Species and quantitative and qualitative variability turned them into a suitable chemotaxonomic marker that can be determined by gas chromatography (368) or high-performance liquid chromatography (see. cassava. (Pseudomonas) maltophilia.org/ on January 3.. and 16 are clearly not consistent with the current classification. X. . TABLE 1.. X.. pisi. X. It might erroneously suggest that polyphasic taxonomy is mainly an activity of “taxonomic splitters.... X. X. polyamine patterning has been used to trace relatedness at and above the genus level and at the species level (25. hortorum. Polyphasic Taxonomy of the Genus Xanthomonas Bacteria belonging to the genus Xanthomonas are plant pathogens for at least 124 monocotyledonous and 268 dicotyledonous plant species. X. 212). which is not a plant pathogen. an extensive examination of the genus Xanthomonas has taken place. cucurbitae. (ii) the species within this genus. EVALUATION OF POLYPHASIC TAXONOMY Polyphasic taxonomy has been applied to many bacterial groups to attain a consensus assessment on the basis of all the available genotypic and phenotypic data. crucifers.. rice.. X. oryzae. It represents a biochemically very diverse group of bacteria.. X. All members of this group have been studied by large numbers of different methods. X. DNA-DNA hybridization studies. The different Xanthomonas species delineated by DNA-DNA hybridizations showed more than 97. 2012 by guest tris. fragariae. and tomato (138). S. axonopodis and 34 former X. axonopodis.. vasicola. theicola. campestris. campestris pathovars! Extensive DNA-DNA hybridizations not only revealed the natural relationships between Xanthomonas strains but also clearly demonstrated that a number of pathovars were composed of two or more unrelated genotypes and could thus not be regarded as biological entities. e.414 VANDAMME ET AL. Xanthomonas oryzae. and vascular or parenchymatous diseases on leaves. Fourier transformation infrared spectroscopy. sugarcane. X. and UV resonance Raman spectroscopy. Extensive DNA-DNA hybridizations between Xanthomonas strains allowed the distinction of 20 DNA hybridization groups (332). and the overreliance on morphological and biochemical characteristics. vesicatoria. 124. consisted of X. illustrating in detail the usefulness of the different techniques in distinguishing taxa at various hierarchical levels. albilineans.. which forms a very tight phylogenetic lineage. although in most cases our knowledge of the host range of strains of a particular pathovar was limited as no extensive host range study had ever been performed. These studies have addressed (i) the delineation of the genus Xanthomonas. albilineans. MICROBIOL. translucens. The former X. X. X. X. fragariae. which form a tight phylogenetic cluster. X. The genus Campylobacter. X. arboricola.” but this would not give credit to the tremendous inputs of polyphasic analyses that have been performed. can be differentiated from Xanthomonas species and has recently been reclassified as Stenotrophomonas maltophilia (241). populi.g. X.. which are difficult to discriminate with a limited number of phenotypic tests.2% similarity versus Stenotrophomonas (135. campestris. using a panoply of methods.. Xanthomonas fragariae. One DNA hybridization group. X.. Fourier transform infrared spectroscopy. albilineans. on bean. X. and identification is crucial. Xanthomonas axonopodis. cotton. The lactic acid bacteria were chosen to illustrate the problems that occur if settled phenotypic classification schemes do not corroborate phylogenetic insights based on rRNA sequencing. e. citrus fruit. on which they cause a variety of disease symptoms including necrosis.. gramineae. X. The Xanthomonas diseases may cause serious economic losses.. of which 4 contain the species X. polyphasic analysis led to a transition type of taxonomy in which a compromise was formulated on the basis of the results at hand. the genus Lactobacillus.. 16S rRNA sequences. poplar. gummosis. Pyrolysis mass spectrometry. It is an example of a family exclusively delineated by using DNA-rRNA hybridization data and in which phenotypic coherence had a major impact on the delineation of genera and species. X. X. (iii) the pathovar system. The use of a variety of techniques has proven that the previous pathovar notation is an artificial system. references 30 and 271). The genus Xanthomonas is an example of a biochemically versatile group. X9. fruits. e.. maltophilia a After the introduction of polyphasic examination. and Xanthomonas populi. The application of different genotypic and phenotypic methods (including chemotaxonomic markers) has shed new light on the taxonomy of this plant pathogen and will be summarized in the following paragraphs. bromi.. and X. Table 1 demonstrates the impact of polyphasic examination of the taxonomy of the genus Xanthomonas.. The strengths and weaknesses of the approach are best demonstrated by outlining studies of four bacterial groups with which we have considerable experience.. X. populi. sacchari.. the major effort was devoted to DNA-DNA hybridizations and has resulted in a complete matrix (332). and (iv) the problem of identification of nonvirulent Xanthomonas strains. X. campestris pathovars or part of them (Table 1). 278. and it will be illustrated that the classification and identification based on host specificity were not always correct. 370). fragariae. axonopodis.

S. X. The percent G C values of the genera Xanthomonas and Stenotrophomonas range between 63 and 70% and between 65 and 68%. dieffenbachiae. this definition is in full agreement with the original criteria used by Sebald and Veron (277) to ´ separate a number of Vibrio species from the genuine vibrios and to include them in the newly created genus Campylobacter. X. phenon 7 contains X. maltophilia and Xylophilus ampelinus strains. The yellow pigment of Xanthomonas species is composed of brominated octaenes (4. campestris pv. New names were proposed for the remaining two natural clusters. 104. these groups were detected.g. The genus Campylobacter became a deposit for a wide assemblage of taxa characterized by a minimal set of phenotypic characteristics including the microaerobic growth requirements and a nonsaccharolytic metabolism. between pathovars from grasses (329). (371) has allowed the differentiation of the species defined by Vauterin et al. oryzae. do not share a single common antigen and can be identified only by a panel of monoclonal antibodies. e. Arcobacter and Helicobacter (113. but strains of other pathovars. campestris pathovars (333). Arcobacter comprises 4 species with Arcobacter nitrofigilis as the type species (326). dieffenbachiae. The most aberrant protein patterns were these of the S. As such. 295). respectively. 2012 by guest . Van den Mooter and Swings (330) have numerically analyzed 295 phenotypic features determined on 266 Xanthomonas and related strains and distinguished the following nine phena. X. campestris pv.asm. campestris pv. This lack of differential characteristics led to the widespread use of vernacular names for many isolates. The recent developments in Xanthomonas taxonomy reflect the conflicting interests of a general-purpose and a special-purpose classification. including those for the taxa X. maltophilia is the occurrence of the fatty acid 17:0 cyclopropane. This was also an indication that the latter species consists of several biological entities. populi. X. specializing as a plant pathogen. as over 65 different fatty acids have been found. manihotis and X. it was shown that extended analyses of large numbers of strains of separate pathovars could reveal differentiating phenotypic features. phenon 8 contains X. with the former probably being a recently evolved branch. maltophilia strains. 85. oryzae pv. 336). albilineans. representing most pathovars. and three major natural clusters (rRNA homology groups) were recognized (303.g. Monoclonal antibody X1 reacts with all Xanthomonas and Stenotrophomonas isolates and demonstrates the relatedness between the genera. Moreover. vesicatoria.VOL.. The cellular fatty acid composition of Xanthomonas strains (371) is very complex. and they allow the identification of large numbers of isolates. Phytopathologists hoped to find a simple way to identify pathovars by this technique. The most important and unexpected result of this study was the demonstration of the heterogeneity of many pathovars.. These protein electrophoretic clusters have been used to select strains for subsequent DNA-DNA hybridizations. solely on the basis of phenotypic similarity.g. and Helicobacter comprises 12 species with Helicobacter pylori as the type species (22. phenon 6 contains X. phenon 5 contains X. e. oryzicola. 321). Monoclonal antibodies have been generated for many Xanthomonas species and X. The situation improved when phylogenetic studies revealed a considerable genotypic heterogeneity among these species. Vauterin et al. The genera Xanthomonas and Stenotrophomonas are phenotypically and genotypically highly related. Whole-cell protein analysis.g. the Campylobacteraceae (317). X. 1996 POLYPHASIC TAXONOMY 415 DNA base ratio. fragariae. campestris pv. Phenon 3 contains X. phenon 4 contains X. 321). campestris pathovars.. (334) have applied SDS-PAGE of whole-cell proteins to 307 Xanthomonas strains and delineated 19 protein electrophoretic clusters. oryzae pv. vesicatoria and X. Polyamines. X. oryzae pv. Only 29 differentiating features were found between the phena. and also very characteristic through the occurrence of many branched and hydroxybranched fatty acids. e. with the first being based on real biological units that are defined by polyphasic taxonomy and the second being defined only by phytopathogenicity. and identified as different biotypes of existing species or as new species. Xanthomonadins. Table 2 lists the members of the genus Campylobacter and related bacteria and the modifications which resulted mainly from a polyphasic taxonomic study (321). fragariae. nonsaccharolytic bacteria with microaerobic growth requirements and a low G C content. by SDS-PAGE of whole-cell proteins) are serologically complex compared with pathovars with a narrow host range and a demonstrated homogeneity. and X. axonopodis. The strains of certain pathovars. Polyphasic Taxonomy of the Genus Campylobacter The genera Campylobacter and Arcobacter form a family of gram-negative. Monoclonal antibodies. 321). from a ubiquitously occurring Stenotrophomonas-like ancestor. X. Both Xanthomonas and Stenotrophomonas species were characterized by the occurrence of spermidine and low quantities of spermine. Conclusions. begoniae. the genus Campylobacter comprises 15 species with Campylobacter fetus as the type species (2.g. X. populi.. and X. e. are indeed each characterized by a single common antigen. described. maltophilia and 24 clusters containing X. 60.g. albilineans. Numerical analysis of phenotypic features. gastric Campylobacter-like organisms or urease-producing thermophilic campylobacters (terms reflecting the unusual isolation sources or aberrant phenotypic characteristics of the strains). e. 184. oryzicola (337). Members of the family Campylobacteraceae are encountered mainly as commensals or parasites in humans and domestic animals. campestris pv. Over a period of about 30 years. Separate clusters were found for the species X. whereas Stenotrophomonas strains have chlorinated hexaenes (156). oryzae. fragariae. and phenon 9 contains a complex of 189 X. respectively. Phenon 1 and phenon 2 comprise S. cassavae (328). campestris pv. Cluster analysis revealed 31 main clusters. the two genera share a characteristic set of nine fatty acids. Cellular fatty acid analysis. Whether Xanthomonas and Stenotrophomonas are two separate genera or should be retained in the single genus Xanthomonas might be a mere reflection of splitting or lumping opinions of taxonomists. In spite of the very high phenotypic similarity of Xanthomonas species. At present.. A reevaluation of the data generated by Yang et al. oryzae and X. pelargonii and X. between X.org/ on January 3. campestris strains. campestris pv. graminis. and between X. 321). but Stenotrophomonas species characteristically contained considerable quantities of cadaverine (370). 91. X. and a revised genus description was given. The fatty acid patterns of Xanthomonas and Stenotrophomonas species are very similar. These organisms were previously included in Xanthomonas. X. Pathovars with a broad host range and with demonstrated heterogeneity (e. oryzae pv. Typical for S. campestris pv. campestris pv. 293. 78. Classical phenotypic tests routinely used for the identification of clinically significant bacteria often yield negative results or yield variable results within species. Campylobacter was separated into three genera. campestris pv. Downloaded from http://mmbr. (332. axonopodis. axonopodis. 244. X. The taxonomic history of these bacteria has been dominated by their biochemical inertness. campestris. albilineans. 100.. 294. 316. populi.

Subsequent physiological studies revealed that none of these bacteria were genuine anaerobes (129. fetus and C. ureolyticus About 10 groups of Campylobacter-like organisms C. Species 1989–1991 Campylobacter Wolinella Bacteroidesb Others C. acinonyx. showae. H. 326). hyoilei. fetus. Below. growth characteristics. However. The putative anaerobes Wolinella recta. showae A. H. C. Arcobacter. 140. 307). A. H. C. H. and between C. C. Applied in a PCR assay. 2012 by guest a b Situation before taxonomic revisions. cinaedi. other clades have bootstrap values below 55%. concisus. c New species and genera since 1991. C. canis. skirrowii H. whereas Wolinella succinogenes appeared to be a close neighbor of the Helicobacter cluster. Finally. cellular fatty acid and respiratory quinone components. jejuni. coli. felis. C. concisus and C. 12.. bubulus. curvus. pylori. 316). upsaliensis. sputorum. H. 264–266. succinogenes were clearly differentiated from their neighbors in their respective rRNA homology groups. 325). A. helveticus. H. H. C. 91. C. Bacteroides gracilis. H. butzleri. nitrofigilis. The application of sequence comparison for studying the intrageneric structure of each of these genera has been hindered by the instability of the branching levels between closely related bacteria (i. lari. Again. mucosalis (180) does not belong to C. C. upsaliensis. 321. and enzyme capabilities (113). However. curva. pylori and Helicobacter mustelae (the only two Helicobacter species at that time) on the basis of differences in ultrastructure. Partial 23S rRNA sequences (about 800 bases) have been determined for a number of Campylobacter species.org/ on January 3. nemestrinae. Generically misclassified Bacteroides species. 270. bilis. ureolyticus remains unsettled (316). pametensis C. 319. the general topology of the phylogenetic tree confirmed the one based on 16S rRNA sequence data (303. 92. B. showae (91). nitrofigilis (type species). changing the outgroups. hyointestinalis. 308. C. 266) also demonstrated that the former C. recta B. Roop et al. As described above. H. d Present situation. 323. 134. C. C. and Bacteroides ureolyticus belonged to the same rRNA homology group as the genuine campylobacters (the homology group comprising the type species). rectus. Bootstrap analysis revealed that within the genus Campylobacter. C. C. in spite of an extensive knowledge of its genotypic and phenotypic characteristics. bilis. upsaliensis A. H. cryaerophila. W.7% in Arcobacter. C. and Helicobacter species have been included in DNA-rRNA hybridization analyses as well (315. DNA-DNA hybridization studies. 130). TABLE 2. concisus. 348). sputorum. H. Sequence comparison of 16S rRNA has been used primarily to unravel the taxonomic structure and relationships of Campylobacter and affiliated genera (177. rRNA sequence homology studies allowed us to divide the genus Campylobacter into three distinct groups. 262. A number of intrageneric relationships were revealed: a close association was found between C. H. jejuni. 5. W. the genotypic heterogeneity was considerable. sputorum subsp. C. within the genera). with sequence differences of up to 9. H. Most of the Campylobacter. (32). C. mucosalis. acinonyx. jejuni. Significant DNA hybridization was found between C. C. H. hyoilei. canis. ureolyticus 1991–1995c Campylobacter Arcobacter Helicobacter Campylobacter Arcobacter Helicobacter Wolinella Bacteroidesb July 1995d Downloaded from http://mmbr. mucosalis. cryaerophilus. lari. upsaliensis W. morphology. and between C. C. ureolyticus and W. rRNA homology studies. group-. C. sputorum. C. H. 303). the branching order easily shifts upon changing the number of taxa included. several clades of species are stable and therefore their linkage can be considered significant (316). sputorum subsp. fennelliae. The former two clusters were also found to be significant after bootstrap analysis of the 16S rRNA sequences (316). hepaticus. 91). sputorum and that C. 312. C. 215. C. Most DNA-DNA hybridization studies were performed before the phylogenetic relationships of campylobacters were established by means of rRNA-directed studies and focused on the species known since the early 1980s (15. jejuni. 78. C. butzleri. C. lari. hyointestinalis (264. gracilis. the number of Helicobacter species has increased drastically. The general topology of phylogenetic trees derived from these partial sequences is in overall agreement with that of trees based on complete 16S rRNA cistrons (313). and thus the exact branching sequence is not significant. 263. 270. several techniques and parameters used in taxonomic studies of campylobacters and related bacteria are discussed. coli. C. and “Campylobacter fecalis” should be considered biovars of a single species. nemestrinae. felis. sputorum. 32. each corresponding to a separate genus. lari. fetus (type species). helveticus. Wolinella curva. muridarum. mustelae. C. sputorum subsp. C. Within each of the three major groups. between C. rectus and C.asm. C. coli. or choosing a different set of relatives (243. W. and 8. succinogenes was originally separated from H. and C. 308). C. gracilis. 243. The genus Campylobacter and allied bacteria Date Genus a MICROBIOL.e. 321). (32. mustelae. these primers and probes offer valuable alternatives for the identification of these bacteria. C. succinogenes B. fennelliae. Sequence information derived from 16S or 23S rRNA cistrons has been used successfully to design a battery of species-. and most of these features have not been examined for the recently described species. C. 307.4% in the genus Campylobacter. pametensis. C. A. Only B. muridarum. C. H. cinaedi. 182. or genus-specific primers and probes (11. succinogenes (type species). 321). C. and C.416 VANDAMME ET AL. hyointestinalis. These data were confirmed by Chevrier et al. pylori (type species) W. 140. between C. DNA-DNA hybridizations between all other Campylobacter species yielded nonsignificant hybridization val- . C. 134. The others could not be separated from campylobacters on genotypic or on phenotypic grounds and were therefore reclassified as Campylobacter species (316. 16. since then. mucosalis. hepaticus. 236. (265. 264. For these species.6% in Helicobacter (78. fetus and C. coli. C. skirrowii H. hyointestinalis. H. H. REV. the classification of B. A.

181. 235). 316).VOL. On and Holmes (229) performed a most comprehensive study: they examined 67 characteristics for 347 strains representing all present species within this phylogenetic lineage and subjected the data to numerical comparison. 264. 323). respectively. 366). DNA base ratio. including the former W. Most campylobacters are characterized by a single polar or bipolar flagellum. ureolyticus. water-soluble proteins (95. whereas C. Interestingly. this technique is not efficient for the identification of Campylobacter species and a comprehensive study on the various Helicobacter species should be performed before general conclusions for this genus can be drawn. C. 323. An even larger diversity is found within the genus Helicobacter. strains with highly similar protein patterns share high DNA hybridization values and therefore belong to the same species (47. again had different cellular fatty acid patterns (115) whereas none of the recently described new species has been examined. the closest phylogenetic neighbor of Campylobacter species. because all taxa of the Campylobacter and Arcobacter rRNA homology groups have unsheathed flagella whereas a flagellar sheath is present in all Helicobacter species (128. In contrast. bipolar flagella. and between 35 and 44%. readily lose this morphology upon aging and cells incubated longer than 24 h will transform into bent rods or straight rods and finally may become virtually completely coccoid. which are most often found in the group of the so-called thermophilic campylobacters (C. 324). 209. because two additional Helicobacter species. these differences can no longer be considered genus specific. 217. 163). 1996 POLYPHASIC TAXONOMY 417 ues. and a huge amount of data was collected. (115) used similar procedures and performed the most comprehensive studies.plus phenol-soluble proteins (131. The part of the protein profile with these variable dense bands is situated mostly in the 39. 246.to 50. and C. The study of the respiratory quinone components of campylobacters was of major importance for the revision of their classification. Arcobacter. B. However. The number of flagella and the flagellation type are obviously not a relevant taxonomic parameter in this group of bacteria. W. in which single flagella. between 27 and 31%. curva. and huge spiral or threadlike cells (up to 20 m long) may be present in old cultures. They will not be discussed here. 78. (175) and Goodwin et al. 323–326) are separated in polyacrylamide gels and stained and may be subjected to computerassisted numerical comparison. and Helicobacter were separated primarily because of Downloaded from http://mmbr. the structure of the flagellum appears to be significant. 237. 217.org/ on January 3. Classical phenotypic characteristics. Classical phenotypic tests were required to identify some strains at the species level. Again.000-molecular-weight range. lari. Helicobacter (113). pylori and C. H. Differences in the cellular fatty acid components were used as one of the criteria for including the former C. gracilis (formerly B. growth medium and growth conditions may considerably influence the protein profile (151. Several authors studied the suit- ability of cellular fatty acid analysis for the differentiation and identification of campylobacters (18. fennelliae. and several species were present in different gas-liquid chromatography groups. unlike most gram-negative bacteria. 321). In a numerical analysis. cells of other species such as C. mustelae not in the genus Wolinella but in a newly created genus. 326). Several of these gas-liquid chromatography groups consisted of more than one species. and B. whose structure has not yet been unravelled. As with other bacteria. or even straight rods. cinaedi and H. showae. contain menaquinone-6 and the so-called thermoplasmaquinon (a methyl-substituted menaquinone-6) as major components. and whole-cell proteins (48–50. growth conditions. Clearly. 216). 175. recta. the interpretation of protein pattern similarity regularly requires considerable expertise. Since the early 1970s. 294. whole-cell protein patterns of campylobacters are stable regardless of the time and methods of preservation. gracilis. however. The precise level of DNA-DNA binding is difficult to compare because of the lack of correlation when different hybridization techniques are used (see below). The genera Campylobacter. Various taxonomic parameters of campylobacters and allied bacteria were studied. 117. 2012 by guest . Species belonging to the Arcobacter rRNA homology group. as Wolinella species are characterized by menaquinone-6 and the thermoplasmaquinon whereas Helicobacter species contain menaquinone-6 and the above-mentioned quinone with the unknown structure. 216. and it may be necessary to omit this region from the numerical analysis to obtain clusters of closely related strains (47. 141. 99. This correspondence between high correlation coefficients of protein patterns and high DNA hybridization values has been illustrated for several Campylobacter and Arcobacter species (319. are characterized by menaquinone-6 and a second menaquinone. The same situation occurs in the Wolinella-Helicobacter group. PAGE of proteins has often been used successfully to classify phenotypically aberrant campylobacters. 44. The respiratory chain of campylobacters. this technique has been widely applied for the differentiation and identification of campylobacters. Arcobacter. Conversely. a detailed study of the genus Arcobacter revealed quantitative or qualitative differences (or both) between most of its members (326).asm. Acid. 249. 316. 301. The percent G C values of Campylobacter. Recently. The taxonomic structure of the entire group is thus well understood. Protein analysis. and age of the cells (99. They included nearly all known Campylobacter taxa and subdivided them into so-called gas-liquid chromatography groups. and Helicobacter species range between 30 and 46%. 232–234. 87. Within other bacterial groups. and lateral flagella are found in different species. Cellular fatty acid analysis. 88. upsaliensis). Only a thorough visual examination of the protein electrophoretic traces of strains from each cluster will reveal the reason for such subdivisions. In addition. 318. 319. Sporadic nonmotile cells occur naturally. 38. 91. coli. Their data indicated that there was considerable correspondence between the grouping obtained and previously determined genomic relationships and that most strains were accurately identified to the species or subspecies level. The cellular morphology of campylobacters and their relatives is extremely diverse. 206. jejuni. 53. C. All species belonging to the Campylobacter rRNA homology group.000. and helicobacters. However. has polar bundles of two to five flagella. At present. Respiratory quinone components. gracilis) is aflagellate. 218. Conclusion. variation in the number or molecular weight of one or a few dense protein bands may cause strains of a single species to form distinct clusters. fetus are spiral yet not slender. 145. arcobacters. bent rods. The problems inherent in the identification of campylobacters and their relatives by means of classical phenotypic tests have been mentioned above and have been the subject of several studies (23. Lambert et al. 114. The latter is a prerequisite for comparing large numbers of strains. 342). 131. 60. 115. such cells. Cells of other species are predominantly S-shaped rods. comprises only menaquinones (37. A slender spiral (“corkscrew-like”) morphology is generally considered typical. the diversity in flagellation types is striking as well. C. 318–320. 225). This stability may be explained by the limited nutritional capabilities of the campylobacters. 293.

dairy products. New Campylobacter and Helicobacter species have been emerging rapidly. Within the Leuconostoc group. non-spore-forming cocci.418 VANDAMME ET AL. it is not our purpose to list the wide variety of techniques used in the study of the taxonomy of these organisms. e. Furthermore.. instead. the congruence between the whole-cell protein pattern similarity and DNA-DNA hybridization was clearly established. This specific phylogenetic structure has been discussed as a case of a rapidly evolving organism (369) and was one of the arguments in removing the species from the genus Leuconostoc into the genus Oenococcus (79).g. dextrinicus is more closely related to L. L. The second phylogenetic group of the lactobacilli.3 to 99.9 to 86. and P. Because of its use in food and its probiotic role in the human intestinal tract.org/ on January 3. These taxonomic reassignments were based almost exclusively on the results of rRNA sequencing or DNA-rRNA hybridizations. 9 Lactobacillus species were transferred to other Lactobacillus species and at least 14 species were transferred to new or already existing genera (Table 3).9 to 91. L. parsimony. Vagococcus. MICROBIOL. Globicatella. poor discriminatory power was found in Campylobacter species. in sewage. Lactobacillus.3% (L. (41). acetotolerans. CO2. The large number of nomenclatural revisions is a striking indication of the discrepancies between the results obtained by former traditional phenotypic tests (classical taxonomy) and the present phylogenetic insights obtained by rRNA sequencing (molecular taxonomy). At present. the genus Bifidobacterium is often listed along with the lactic acid bacteria sensu stricto. They generally lack catalase. parvulus.” It is mainly their importance in the fermentation of food and feed products (meat.g. the L. alternatively. vegetables. hamsteri. intestinal. Table 3 presents a subdivision of these lactobacilli according to their fermentation type and phylogenetic assignment. They need a fermentable carbohydrate for growth. with some related species of Leuconostoc and Pediococcus. Therefore. L. The lactobacilli can be subdivided into three major groups: the Lactobacillus delbrueckii group. bifermentans than to the other Downloaded from http://mmbr. although phylogenetically it belongs to the Actinomyces subdivision of the grampositive bacteria (comprising also Atopobium. beverages. Helicobacters were only partly examined. and. As mentioned above. Protein electrophoresis offered a valuable alternative for species identification. (ii) L. will be highlighted here. jensenii.asm. 2012 by guest . the seven species of the L. P. Alloiococcus. Many lactic acid bacteria are known to have a positive impact on human and animal health (109). pentosaceus) form a separate clade. are described. Pediococcus. and the former technique was successfully used to identify phenotypically aberrant strains.0% (L. the Lactobacillus casei-Pediococcus group. (i) L. During the last several years. L. L. damnosus. acidophilus group (see below). only the most notable species of the lactobacilli. Some pathogenic species are found among. and Saccharococcus are generally not considered to belong to the lactic acid bacteria. Dolosigranulum. Propionibacterium. to some extent. the large genotypic divergence between the three groups (321). a chemotaxonomic method such as whole-cell protein analysis may therefore become essential to reliably identify unusual or atypical campylobacters pending the discovery of additional phenotypic tests. or rods with a DNA base composition of less than 50 mol% G C.5% and 85. Four of the pediococcal species (Pediococcus acidilactici. P. The percent homologies of 16S rRNA within the two major lactobacillus groups vary from 90. Tetragenococcus. casei-Pediococcus group. Brevibacterium. Enterococcus. Atopobium. delbrueckii group. However. beer and wine and to occur in the respiratory. lactic acid bacteria are also known to be involved in spoilage of. coryniformis and L. peripherally. glucose is converted mainly to lactic acid (homofermentatives) or to lactic acid. the streptococci.8% were registered between this organism and the other species in the same group (201). 5 of which contain at least two subspecies. casei-Pediococcus group) (41). Molecular diagnostic methods with DNA probes or specific PCR assays or. delbrueckii contains three subspecies which cannot be discriminated by rRNA sequence analysis. Gemella. In a recent overview of the phylogeny of lactic acid bacteria (275). casei-Pediococcus group (41. Given the very large number of species described. The L. which belong to the Clostridium subdivision of the gram-positive bacteria. In food. and genital tracts of humans and animals. meet the definition given above. L. e. coccobacilli. At the species level.. and silage) which delineates the group of lactic acid bacteria. and although they are mostly well characterized and their phylogenetic position has been clearly established. Streptococcus. alternatively. 275). a phylogenetic tree was constructed by using distance matrix. REV. delbrueckii group) and 90. delbrueckii (the type species of the genus Lactobacillus). fruits. lactic acid bacteria are generally restricted to the genera Carnobacterium. Oenococcus. Leuconostoc. lactic acid bacteria contribute to the taste and texture of fermented products and inhibit the growth of food- spoiling bacteria by the production of growth-inhibiting substances (bacteriocins) and the production of large amounts of lactic acid. whereas most Arcobacter species were easily differentiated. The total number of taxa within this lineage of the Proteobacteria now approaches 50. Helcococcus.8 to 99. outnumbers the other groups. kefiranofaciens. the suitability of cellular fatty acid analysis for the differentiation of these species is genus dependent. The large differences in 16S rRNA sequences between and within these genera may indicate that they represent old evolutionary branches in which substantial genetic drift occurred. amylophilus. This subdivision was supported by chemotaxonomic and ultrastructural characteristics. and ethanol or acetic acid or both (heterofermentatives). Lactococcus. we will focus on those techniques which have directed some of the major taxonomic changes within the lactobacilli. However. The enormous range in DNA base ratio which is found in the genus Campylobacter is an additional support for this hypothesis. the position of Oenococcus oeni is very interesting. and maximum-likelihood analyses. their identification by means of classical phenotypic tests is cumbersome. and the Leuconostoc group. although they might. Many of these can be differentiated only by a single or a few classical phenotypic tests or. Representatives of the genera Aerococcus. and the microbacteria) and therefore is only quite distantly related to the genuine lactic acid bacteria. and in plant materials. Melissococcus. Phylogenetic analysis based on rRNA homology. the genus Lactobacillus comprises 56 species. and Weissella. P. Polyphasic Taxonomy of Lactic Acid Bacteria Lactic acid bacteria are gram-positive. The major phylogenetic groups of lactic acid bacteria sensu stricto. by using a large number of phenotypic tests (229). this definition generally covers more taxa than are normally covered by the name “lactic acid bacteria. with 37 Lactobacillus species and 5 Pediococcus species. Extremely low homology values of 85. which are largely in agreement with data published by Collins et al. although pseudocatalase was detected in cultures grown at a low sugar concentration. delbrueckii group contains L.

reuteri. Leuconostoc amelibiosum. 60. L. minor (“L. L. L. yamanashiensis”). Leuconostoc pseudomesenteroides. rhamnosus. L. L. Species in this group are no longer considered members of the genus Lactobacillus and are assigned to various fermentation groups. cellobiosus”). L. L. johnsonii. L. L. casei-Pediococcus group) L. maltaromicus”). carnis”).asm. L. mali (“L. curvatus. graminis. which is outside the range defined for the genus Lactobacillus. ruminis. viridescens (“L. L. acetotolerans. parabuchneri. intestinalis. minor”). L. L. L. L. L. W. amylovorus. helveticus (“L. torquens. rogosae was described to have a DNA base composition of 59 mol% G C. L. subsp. casei. minutus”). fructivorans (“L. malefermentans. L. W. subsp. oeni Other lactobacillib L. L. 1996 . halotolerans”). Leuconostoc argentinum. Carnobacterium piscicola (“L. kefir. paramesenteroides (“Leuconostoc paramesenteroides”). salivarius subsp. crispatus. vermiforme”). salicinus. bifermentans. sanfrancisco. murinus (“L. L. pontis. vitulinus. W. fructosus. L. collinoides. xylosus”) a Data from reference 126. no strains corresponding to the original description are presently available. hellenica. kefirgranuma L. L. Pediococcus parvulus L. L. viridescens”). suebicus. Leuconostoc lactis. lactis”). catenaformis. hilgardii (“L. L. L. delbrueckii subsp. Pediococcus damnosus. salivarius. delbrueckii. plantarum. brevis. subsp. L. confusus (“L. Carnobacterium divergens (“L. Pediococcus dextrinicum. W. O. aviarius subsp. Leuconostoc mesenteroides. pentosus. W. agilis. halotolerans (“L.a L. divergens”) (“L. L. coryniformis. bulgaricus (“L. uli”). L. coryniformis. L. hordniae”) (“L. subsp. Leuconostoc carnosum. Atopobium rimae (“L. L. W. confusus”). L. L. piscicola”) (“L. gasseri. vaginalis c (Leuconostoc group) L. 2012 by guest POLYPHASIC TAXONOMY 419 TABLE 3. W. fermentum (“L. lactis (“L. alimentarius.Downloaded from http://mmbr. L. b c VOL. parakefir. sake (“L. homohiochii. kandleri (“L. amylophilus. subsp. paracasei subsp. L. hamsteri b (L. L. paracasei. araffinosus. L. kefiranofaciens.org/ on January 3. L. rimae”). kandleri”). subsp. L. gallinarum. Leuconostoc fallax. L. Lactococcus lactis (“L. L. L. kefirgranum and L. Pediococcus pentosaceus L. acidophilus. L. L. jugurti”). L. trichodes”). L. tolerans. delbrueckii group) L. bavaricus”) Pediococcus acidilactici. rogosae. bulgaricus”). Atopobium uli (“L. jensenii. L. buchneri. Leuconostoc gelidum. parakefir have not been included in 16S rRNA sequence analysis. oris. L. animalis”). subsp. vaccinostercus. aviarius.c Atopobium minutum (“L. L. sharpae. L. farciminis. L. Subdivision of the lactobacilli according to their phenotype and phylogenetic assignmenta Species in fermentation group: Phylogenetic group A (obligately homofermentative) B (facultatively heterofermentative) C (obligately heterofermentative) a (L.

259). The position of the latter lactobacillus is particularly interesting. W. to the actinomycete branch of the gram-positive bacteria (L. Obligately homofermentative lactobacilli degrade hexoses almost exclusively to lactic acid by the Embden-Meyerhof pathway. The phenotypic differentiation of the obligately homofermentative lactobacilli is given by Hammes and Vogel (126). 2012 by guest . and C. and O. xylosus and L. halotolerans. curvatus. (iii) Leuconostoc group. minor. uli. since L. (ii) the facultatively heterofermentative species. Pentoses are fermented to lactic acid and acetic acid via an inducible pentose phosphoketolase. bavaricus. At present. both transferred to Lactococcus lactis [165]). casei-Pediococcus group contains a number of wellknown lactobacilli including the poorly characterized species L. Examination of the presence of fructose-1. In this subdivision. related to the clostridia and Erysipelothrix (Lactobacillus catenaformis. and L. A first subgroup is the so-called paramesenteroides cluster containing the former species Lactobacillus confusus. which are used in many food fermentations. and L.org/ on January 3. A first indication of its heterogeneity was obtained from serological data (86). 247). Since then. 214. or to the lactococci (L. with Downloaded from http://mmbr. The third phylogenetic group within the lactobacilli.6-bisphosphate aldolase resulted in the subdivision of these bacteria into three physiological groups: (i) the obligately homofermentative lactobacilli. the paramesenteroides group (42) was assigned to the new genus Weissella. and include the thermobac- teria and a considerable number of more recently described species. acidophilus remained confusing for a long time (106. and the Leuconostoc group.420 VANDAMME ET AL. many species have been described but either not validated or. rRNA sequence analyses revealed a considerable number of misnamed lactobacilli. hamsteri. acidophilus group (157). amylophilus. viridescens. REV. most DNA-DNA hybridization studies were performed before the phylogenetic relationships of the lactobacilli were fully established. Finally. casei-Pediococcus group. Some presumed obligately homofermentative lactobacilli of the phylogenetic group of L. sake. depending on the pH. plantarum. the species of group Cb may have been derived from the metabolically more versatile group Bb species by the loss of the aldolase and triose-phosphate isomerase of the Embden-Meyerhof pathway. which together form the second subgroup within the Leuconostoc group (275). Pentoses are fermented to lactic acid and acetic acid. and L. and also L. W. and 6 have been assigned to group Cc (obligately heterofermentative lactobacilli from the Leuconostoc branch). L. The obligately homofermentative lactobacilli are allocated to two phylogenetic subgroups. refer to fermentation types and the lowercase letters a. B. respectively). 281) and is still used today. oeni (the former Leuconostoc oenos). which is found in many habitats (dairy products. W. 133. comprises two subgroups. electrophoretic analysis of soluble cellular proteins or lactate dehydrogenases. silage. acetic acid. The L.. (iii) Obligately heterofermentative lactobacilli (group C). subgroup Aa contains the industrially important assembly of species known as the L. comprising the species Weissella confusus. (i) Obligately homofermentative lactobacilli (group A). the large number of species renders an exclusively genotypic approach quite cumbersome. bifermentans shows. and formic acid under glucose limitation. fermentation reactions (mostly examined via commercially available test systems) remain important as tools for identification and classification. L. 280). L. an aberrant type of fermentation (161. Later. Although correct classification and identification of lactic acid bacteria are difficult without the support of modern genotypic techniques (see above). caseiPediococcus branch).asm. Therefore. and Leuconostoc paramesenteroides. DNADNA hybridization studies and standardized SDS-PAGE of whole-cell proteins (see below) have been necessary to actually delineate the species present in this pool of phenotypically very similar organisms. delbrueckii (L. 160. (iv) Other lactobacilli. a pentose phosphoketolase is involved in both pathways. L. W. The latter species is quite far removed from the other leuconostocs. and group Bb (facultatively heterofermentative lactobacilli from the L. In an attempt to match the phenotypic division of the lactobacilli with the phylogenetic data obtained from rRNA sequencing. L. L. a total of 22 Lactobacillus species are spread over two phylogenetic branches: 16 species have been assigned to group Cb (obligately heterofermentative lactobacilli from the L. acetotolerans and L. The subdivision of lactobacilli into fermentation groups was maintained with some major modifications until the late 1970s. detailed cell wall studies. ethanol. kandleri. L. W. kefiranofaciens) can ferment pentoses and were therefore classified as facultatively heterofermentative lactobacilli in group B (as subgroup Ba [126]). shown to be homologous to existing species. wine. ethanol. rimae. Again. cannot use pentoses or gluconate. casei-Pediococcus subgroup (group Ab). hordniae. (127). Delineation of Lactobacillus species by traditional phenotypic tests. Despite its industrial and medical importance (136. and the human intestinal tract and mouth). acetotolerans. MICROBIOL. Delineation of Lactobacillus species by DNA-DNA hybridization studies. all transferred to the genus Atopobium by Collins and Wallbanks [43]). Lactobacillus fructosus. kandleri. delbrueckii group. and a new species. casei-Pediococcus branch) harbors 15 Lactobacillus species (126). More recently. the capital letters A. Facultatively heterofermentative lactobacilli ferment hexoses almost exclusively to lactic acid by the Embden-Meyerhof pathway or to lactic acid. the L. 231. Pediococcus species. paramesenteroides. by the loss of phosphoketolase activity (126). jensenii. L. 261. sewage. W. halotolerans. 5 of which have recently been transferred to the genus Weissella (42). (ii) Facultatively heterofermentative lactobacilli (group B). hellenica for isolates from fermented sausage (42). viridescens. To cope with the increasing number of newly described species. L. Besides L. the group includes the former streptobacteria and many newly described species. e. minutus. It has been suggested that the group A species may have evolved from metabolically more versatile facultative heterofermenters.g. In general. two subgroups of species are distinguished: group Ba (facultatively heterofermentative lactobacilli from the L. acetic acid. According to Hammes and Vogel (126). Finally. vitulinus). these species still cannot be reliably differentiated by simple phenotypic tests (105). refer in each subdivision to the three phylogenetic subgroups (the L. delbrueckii rRNA branch) comprises L. 210. hamsteri (both regarded as obligate homofermenters until recently [126]). the L. as mentioned above. and (iii) the obligately heterofermentative species which lack fructose-1. and CO2 via the phosphogluconate pathway. The genus Lactobacillus was proposed by Beijerinck in 1901 (14). paracasei. the Leuconostoc group (41). the subdivision described below was proposed by Hammes and Vogel (126).6-bisphosphate aldolase. As for the lactobacilli of group A. As with Campylobacter species. the taxonomy of L. each deserving a separate species status. L. delbrueckii subgroup (group Aa) and the L. b. delbrueckii and L. and determination of the moles percent G C content of the DNA confirmed the presence of a number of genotypic groups. The obligately heterofermentative lactobacilli ferment hexoses to lactic acid. (260. and c. the fermentation groups were consecutively redefined in 1986 by Kandler and Weiss (162) and by Hammes et al. minor.

a species which was not included in the Approved Lists of Bacterial Names (284). jugurti” and L. L. was proposed for all these organisms. gallinarum. lactococci (67. L. e.. acidophilus (157). bulgaricus”). which have a G C content higher than 50 mol%. These oligonucleotides can be labeled and used as probes in hybridization experiments with unknown isolates. requiring a certain degree of preidentification. and on the basis of nucleic acid and biochemical data. The following example shows. L. pseudoplantarum. however. acidophilus group (253). After numerical comparison. rRNA-targeted oligonucleotide probes. L. delbrueckii. and tolerans. The synonymy between the former species “L. reuteri (81). the latter species name was revived by Zanoni et al. DNA base ratio. rhamnosus and for L. Whole-cell protein analysis. 60. casei subsp. 213).. 144. e. helveticus was already established (62. 305. and L. a limited number of representative strains can be selected from every (sub)cluster for further analysis by genotypic or phenotypic methods. tolerans). and lactis (previously “L. acidophilus sensu lato. L. amylovorus (222). (ii) biochemical resemblance. This relatedness was confirmed by (i) a comparable base composition of the DNA. As an increasing number of DNA or rRNA nucleotide sequences become available. acidophilus. The electrophoretic mobility of LDH in starch gels (106) or in polyacrylamide gels (143) was used mostly for the discrimination of phenotypically very similar species (e. together with members of the subspecies alactosus. casei subsp. (i) L. Subgroup 2 is represented by L. Numerous DNA-DNA hybridization studies revealed discrepancies between species level identification by means of phenotypic tests and genotypic approaches. The name L. a new species.g. plantarum was quite heterogeneous. 251). 282). 327). 305.asm. As mentioned above. however. The value of this technique is demonstrated by the design of species-specific 23S rRNA targeted oligonucleotide probes useful for the identification of authentic L. crispatus. L. L. 139). acidophilus. since 8 strains (of the 28 strains investigated) displayed only 46 to 65% DNA relatedness to the type strain. with a G C content of 40 to 43 mol% (162). which formed a genotypically homogeneous group not related to the other members of L. casei subsp. rhamnosus. johnsonii (105). the majority of the subspecies casei strains. By using PCR techniques. Collins et al. gallinarum (105). The use of SDS-PAGE for the identification of a wide variety of lactic acid bacteria is hampered by the fact that it yields only discriminative information at or below the species level. two subgroups have been delineated. (ii) L. and (iii) 16S rRNA sequence analyses (41). 1996 POLYPHASIC TAXONOMY 421 emphasis on the species of economic interest used by the food and feed industry. Strains previously assigned to L. reuteri. in contrast to the gram-positive bacteria of the Actinomycetes branch. lactis (166).org/ on January 3. Lactate dehydrogenase (LDH) has been an important characteristic in the classification of lactic acid bacteria and has been studied in many ways. paracasei (L. Identification of lactic acid bacteria is then reduced to the preparation of a protein pattern and a subsequent comparison with the patterns available in the database (245.g. cremoris (269). 105. rhamnosus (144). casei (230) by DNA-DNA hybridization (40). (339) found that 13 of 24 presumed L. that care should be taken with the results of electrophoretic mobility tests of LDH. 137. L. the target sequences can be amplified and the detection level can be enhanced. 73. (41) showed that the two species shared more than 99% 16S rRNA sequence similarity. Therefore. brevis strains belonged to the species L. kefir and L. 250. with the three subspecies delbrueckii. helveticus and L. johnsonii [105]). confusus. crispatus (31. the comparison of homologous sequences can yield oligonucleotide stretches which are specific for the taxa being compared. The technique has proven to be extremely useful for grouping of large numbers of strains (67. Subgroup 1 includes L. or L. paracasei. delbrueckii group. tolerans were phenotypically as well as genotypically sufficiently distinct to be maintained as a separate subspecies of L. acidophilus sensu stricto (133). and Leuconostoc strains (9. casei. gasseri. L. 82). casei and L. This problem has been overcome by the creation of a database of digitized and normalized protein patterns for all known species of lactic acid bacteria (254). 154. pontis. 178). The preparation of similarity dendrograms also yields information important for the classification of lactic acid bacteria (144. johnsonii. paracasei. was elevated to the species rank as L. L. casei-Pediococcus group. gasseri (179).g. (105). casei. On the basis of DNA-DNA hybridizations. has a DNA base composition of 53 to 54 mol% G C (341) and is phylogenetically very closely related to L. kefir. and L. leichmannii”). Farrow and Collins (94) investigated the genetic interrelationships of some strains from the human oral cavity and saliva. vagococci (250). which was shown to be composed of at least six species: L. A compilation of the relative migration distances of LDH of various species has been given by Kandler and Weiss (162) and more recently by Fujisawa et al. L. paracasei. Lactate dehydrogenase. 327). L. 2012 by guest . (63) showed that L. gasseri (253). L. especially in food microbiology. the DNA base composition of lactic acid bacteria is less than 50 mol% G C. earlier assigned to L. Only a slight difference in the migration distance in starch gel electrophore- Downloaded from http://mmbr. 305). 254. a DNA-DNA similarity of 13 to 44% was found between L. L. L. Vescovo et al. Moreover. and L. brevis-like strains by carbohydrate fermentation reactions or additional simple phenotypic tests is insufficient. shared high levels of DNA relatedness but were quite distinct (10 to 20% DNA similarity) from the type strain. L. rhamnosus.. to identify the genus Lactococcus (269) or the subspecies Lactococcus lactis subsp. casei (132) was restricted to the type strain only. lactis” and “L. This species classification is a result of several DNADNA hybridization studies (31. 89. delbrueckii by simple physiological tests. This is a characteristic of the so-called Clostridium branch of the gram-positive bacteria. and Lactococcus lactis subsp. Oligonucleotide probes at the subspecies or genus level were also developed. L. The comparison of whole-cell protein patterns obtained by highly standardized SDS-PAGE has proven to be extremely reliable at the species and subspecies levels (67. L. L. pentosus (63). which share over 80% DNA similarity (344). oris. The use of probes. These strains were shown to be highly related to L. In 1987. the strains were assigned to a new species. paracasei. and L. and mostly they have served the proper description of new species or contributed to the reduction of previously heterogeneous taxa and therefore proved to be indispensable for species determination within the genus Lactobacillus. It was possible to solve specific identification problems for. Only a few examples are given below. 253). L. In 1991. helveticus and group-specific probes for L. L. the species of the L. has been reviewed by Schleifer (272). which indicates that the identification of L. hilgardii. NAD-dependent D-LDH profiles have also been used for the differentiation of Leuconostoc species (80). There are many reports on the use of this technique. 253. DNA-DNA hybridization studies performed by Dellaglio et al. 157. collinoides. When comparing the five subspecies of L. paracasei subsp. brevis (54. jensenii is indistinguishable from L. bulgaricus (previously “L. within subgroup 2.VOL. 245. (373). delbrueckii. L.

354). Cell wall components. The subdivision of lactobacilli (and other lactic acid bacteria) into three main fermentation categories (homofermenters. bulgaricus showed considerable sequence differences from representatives of the relatively well-characterized NAD-dependent L-LDHs (183). Representative species of the L. and glyceraldehyde-3-phosphate dehydrogenase (189). REV. The data were used to create dendrograms and (three-dimensional) phylogenetic maps. Only a few interesting exceptions occur: in L. facultatively heterofermentative [Bb]. Most of the taxonomic revisions were proposed during the last decade and are increasingly dependent on rRNA sequence information. (i) The first is DNA-DNA hybridizations. Variovorax. The discrepancies between phenotypic data present in traditional microbiology textbooks and phylogenetic data that have recently become available have not sufficiently been translated into new strategies for future work. delbrueckii group (groups Aa and Ba. which implemented a complete reclassification of the genus Pseudomonas (240). 193).asm. Brevundimonas (278) or parts of rRNA groups such as Burkholderia (111. especially if it is coupled to a PCR amplification step. 72) illustrated the phylogenetic diversity of named Pseudomonas members compiled in addenda I through IV of Doudoroff and Palleroni (83). which were useful and helped to select reliable reference strains for the rRNA sequence analysis. this technique is well suited for identification of even large numbers of strains. it is necessarily a combination of these techniques that may actually prove to be the most rewarding with respect to accuracy and expenditure of time and labor. The phylogenetically related group of the Comamonadaceae Downloaded from http://mmbr. Its core originated from the so-called Pseudomonas rRNA group III. clinical samples. and obligate heterofermenters) is still used to identify strains. 72). 354. Since it is fast and cheap.org/ on January 3.and intraspecific relationships which are not revealed by rRNA sequence analysis. several techniques are available. and therefore the authors advised the use of the large difference in the moles percent G C of the DNA (35 to 37 mol% for L. An overview of the updated taxonomic position of the pseudomonads will be presented elsewhere (69). monocarbonyl) amino acid type (Lys–L-Ser–LAla2 or Lys–L-Ala2 [64]). facultative heterofermenters. However. Provided that a database covering all known species is available (254). the use of phylogenetic methods has provided a highly reliable basis for the reevaluation of the phenotypic classification schemes. For the identification of unknown isolates.6-bisphosphate aldolase (188. the traditional phenotypic analysis remains important. is at present clear but is different from the physiological and chemotaxonomic classification schemes used for almost a century.422 VANDAMME ET AL. confirming the genotypic heterogeneity of this species discussed above. and infected plant material (240. 367). The phylogenetic distance between each of these rRNA groups was later (68) correctly interpreted by the inclusion of representatives of a wide variety of members of the Proteobacteria (291) in the DNA-rRNA hybridization experiments. The obligately heterofermentative organisms from the Leuconostoc group (group Cc) all have the Lys (monoamino. the Lys-Ala type is found in L. They have been isolated from a wide variety of ecological niches such as soil. An important number of these species with uncertain taxonomic affiliation could be assigned to one of the five rRNA groups mentioned above. Further investigation by comparative rRNA similarity studies (70. and III (107). respectively) all have the Lys–D-Asp type. the presence of a new taxon can be easily established. they may play an important role in the biodegradation of xenobiotics (24) and recalcitrant components such as oil-derived wastes (1) and pollutants such as nitrate (311). fermentum and L. Quantitative immunological techniques have been used with LDH as an evolutionary marker besides numerous other enzymes like fructose-1. pontis. however. and the Orn–D-Asp type is found in L. This difference could not be demonstrated by polyacrylamide disk gel electrophoresis (162). because of the widespread use of relatively fast commercialized systems for testing phenotypic characteristics and the large number of taxa. This nomenclatural rearrangement of the genus Pseudomonas entailed the creation of several new genera encompassing either a complete rRNA group. MICROBIOL. The phylogenetic distribution of different murein types in lactobacilli was comprehensively summarized by Hammes and Vogel (126). II. On the basis of results with both D. Purification and characterization of the D-LDH from L. which renders this approach for the identification of large numbers of strains quite complex.. sanfrancisco. casei-Pediococcus group (homofermentative [Ab]. the variation in hybridization procedures is large. This family is currently composed of the genera Comamonas. Because of their physiological flexibility.and L-LDH. 354). 360). belonging to existing or new taxa.g. 190. In addition. etc. lysine is replaced by the chemically similar ornithine (Orn–D-Asp type). Within the group of lactic acid bacteria. (ii) SDS-PAGE of wholecell proteins has proven to be extremely reliable in revealing relationships at the species and subspecies levels. Conclusion. They are essential to study inter. a result is the effective phenotypic description of new genera like Oenococcus and Weissella. e. malic enzymes (192. Polyphasic Taxonomy of the Family Comamonadaceae The family Comamonadaceae contains saprophytic and phytopathogenic bacteria with diverse physiological profiles (chemoorganotrophic or chemolithotrophic at the expense of H2 or CO oxidation) (353. Stenotrophomonas (241). natural and industrial environments (353. and obligately heterofermentative [Cb]) have either the Lys–D-Asp type or the diaminopimelicdirect type. delbrueckii) to differentiate the species. Hydrogenophaga (351). The phylogenetic structure of the lactic acid bacteria. L. Hydrogenophaga. Acidovorax. jensenii versus 49 to 51 mol% for L. Because of the conserved nature of the rRNA molecule. acidophilus sensu lato was divided into immunological groups I. delbrueckii subsp. Acidovorax (356. as deduced from rRNA sequencing. which is now restricted to rRNA group I organisms because the type species Pseudomonas aeruginosa belongs to it. given a general identification by classical methods. 2012 by guest . (iii) rRNA probe hybridization is a promising tool. The limitations at present are twofold. while others were scattered over the entire class of Proteobacteria (70. DNA-DNA hybridization and rRNA sequence analysis are required. sis of the D-LDH of the two species has been reported by using starch gel electrophoresis (106). which both form well-delineated phylogenetic and phenotypic entities. one of the five distinct rRNA homology groups found in Pseudomonas species by the initial rRNA studies of Palleroni et al. Homofermentative as well as facultatively heterofermentative lactobacilli from the L. 191). vaginalis. 356) and have an oxidative metabolism in which O2 is used as the terminal electron acceptor. cross-reactions between closely related species occur. For the study of the relationships at the species level. water. while some species can also use nitrates. and Xylophilus and several misclassified Aquaspirillum species. To estimate observed similarity in quantitative genotypic terms. (242).

palleronii.6 C (267)]. it was shown that this rRNA complex encompasses a variety of physiologically and morphologically diverse organisms.g. testosteroni is available. lophotrichously. giesbergeri are closely related (150. Within A. This means that A. DNArRNA and DNA-DNA hybridization data could not be confirmed by phenotypic differences. 2012 by guest . Recently. while Variovorax. and typical regulation of aromatic amino acid biosynthesis (see below) were not studied for all species. this separate lineage was proposed as a new bacterial family. and the internal similarities of the whole rRNA complex (of which the members of the Comamonadaceae constitute only one of the rRNA lineages) are shown in reference 353. 6 C (61)]. but we have reservations about the absolute application of the 70% rule to species recognition (343). suggesting that both species constitute a kind of genotypic continuum.VOL. C. 361). Because features such as meta cleavage of protocatechuate. Consequently. Ideonella (199). which were consequently assigned to two different Acidovorax species (351). 171. a single species was proposed for the C. Recent 16S rDNA sequences of more members of this family (211) are in general agreement with the DNA-rRNA hybridization results. Identification at the species level became difficult. DNA relatedness was used as the genotypic parameter to delineate species. Alcaligenes eutrophus.org/ on January 3. The strains were selected as consensus representatives of clusters obtained by different techniques (SDS-PAGE of cellular proteins. 7. it originally consisted of the misclassified pseudomonads from Pseudomonas rRNA group III (68.. As a consequence. However. has been proposed for new isolates that are capable of growing anaerobically with chlorate as an electron acceptor. 8 C (60)]. the Comamonadaceae. and Alcaligenes) of which the type species belong to other rRNA branches of the Proteobacteria. as discussed elsewhere. since several members of the Comamonadaceae form a relatively tight cluster (364). A. terrigena). facilis were shown to be phenotypically and chemotaxonomically more similar to each other than to A. sinuosum and A. showing 40 to 50% of DNA relatedness. A. Indirectly measured rRNA similarities (58) were used to unravel the phylogenetic structure of the Comamonadaceae. DNA-DNA hybridizations. acidovorans and P. testosteroni. and Xylophilus each represent a single subbranch. only the 16S rRNA sequence of C. Nomenclatural changes at the genus and species levels were proposed only when there was a consensus between the data obtained by the different techniques and when phenotypic data allowed clear-cut differentiation of the taxa to be proposed. rRNA similarities. 72. It was also shown that Comamonas terrigena belonged to a separate rRNA subbranch and thus deserved a separate generic rank. A. species were defined as groups of strains sharing at least 40% of DNA binding (356). a new species constituting a separate rRNA subbranch with the former two species. were delineated. The grouping obtained by DNArRNA hybridizations was in agreement with 16S rRNA cataloging. 5 C (59)]. the genus Comamonas and the species Comamonas terrigena were revived (71). delafieldii. It constitutes a separate lineage in the beta subclass. at least two subgroups. 360. A. the Alcaligenaceae [ Tm(e) range. delafieldii. temperans. 353– 361). delicatum. Xanthomonas. By using 15 different rRNA probes from well-selected strains. acidovorans and C. Table 4 gives an overview of the taxa belonging to the Comamonadaceae and illustrates the discriminatory power of the different methods used. A. gracile. By using the initial spectrophotometric renaturation rates method (57).asm. strain selection was enlarged to represent all nonmatching clusters. DNA-DNA hybridizations revealed separate DNA groups (DNA binding below 35%) for both Pseudomonas species. each theoretically deserved a separate generic status. delafieldii. 252. a fusion of the species seems the more appropriate response. Consequently. facilis as well as with A. is still incomplete (353). delafieldii and P. the 14 distinct rRNA subbranches constitute a separate lineage with Tm(e) values that are similar or slightly lower than those observed in several other bacterial families including the Neisseriaceae [ Tm(e) range. Together. and the composition of the groups obtained by all methods was evaluated. Aquaspirillum. 356. and P. a new genus. When the groupings obtained by the different techniques did not match. Four of the remaining rRNA subbranches each contain a misclassified Aquaspirillum species: A. Acidovorax encompasses two. and the Pasteurellaceae [ Tm(e) range. P. the species delineation was hindered by the fact that the former P. terrigena phenon (361) containing three genomic groups. or peritrichously flagellated (71. P.g. New Acidovorax isolates (110) showed 40% DNA binding with A. other members of the Comamonadaceae were wrongly assigned to genera (e. 256) and belong to the last rRNA subbranch. if not impossible. Apart from the misclassified Pseudomonas species. P. Hydrogenophaga.. In the genus Acidovorax. As mentioned above. acidovorans. anulus. facilis and A. 351. testosteroni. Additional phylogenetic data on the other members of the Comamonadaceae are needed for comparison to reveal the relationship of Ideonella and the Comamonadaceae. testosteroni. Downloaded from http://mmbr. 1996 POLYPHASIC TAXONOMY 423 is also known as the acidovorans rRNA complex. Among the various members of the Comamonadaceae. aquaticum belongs to one of them and lost its nomenclatural status. acidovorans. including rods and helical cells which are polarly. with 89 to 90% rRNA sequence similarity with C. The grouping of these apparently unrelated phenotypes into a single family was based on rRNA similarity data only. metamorphum is equally far removed from the other rRNA subbranches and constitutes the core of an additional rRNA subbranch. delafieldii must either be united into one species or can be considered to constitute a number of taxa deserving separate species status if they can be discriminated phenotypically. 14 distinct rRNA subbranches were delineated. P. Later. and phenotypic and chemotaxonomic methods) used to screen large numbers of strains. respectively (300). belonging to the same rRNA subbranch as Comamonas species. The selection of strains is important. metamorphum. The internal taxonomy of the Comamonadaceae at the generic and species levels was clarified by investigating more than 150 strains belonging to the 14 rRNA subbranches with various genotypic and phenotypic methods. cellular fatty acid analysis. with Rhodocyclus gelatinosus as its closest neighbor. In some cases (e. were transferred to Comamonas as C. 60. the Acetobacteraceae [ Tm(e) range. 242): P. They belong in the beta subclass. and only a few representative strains of the different groups have been used in DNA-DNA hybridization experiments (351. Nomenclatural changes were proposed only when consistent phenotypic evidence was available to describe and differentiate between the taxa. and Burkholderia cepacia. each with a Tm(e) range of 4 to 5 C (353). Five genera covering nine rRNA subbranches were proposed: Comamonas encompasses four rRNA subbranches. the phenotypic description of the family Comamonadaceae. In view of the high phenotypic similarity between the species. The results of each technique were analyzed numerically. terrigena. which was created merely in terms of DNA-rRNA hybridization data. The assignment at the family level was arbitrarily restricted to organisms with a Tm(e) of at least 75 C when hybridized with the rRNA from the type strain of C. and A. 357. flava. growth at 41 C. absence of denitrification.

d ARDRA. REV.. which corroborated DNA hybridization groups (256. Restriction analysis of the 16S rDNA. Within C. 351. delafieldii and A.asm. pseudoflava palleronii taeniospiralis Variovorax paradoxus Xylophilus ampelinus Aquaspirillum anulus giesbergeri. Goteborg.d DNA-DNA.d and auxanography rRNA. The range within this family is from 57 to 70%. The mean percent G C content of representative members of all genera and species of the Comamonadaceae was determined. carboxydoflava. and part of the 23S rDNA with HinfI and CfoI resulted in consistent species-specific patterns (331). 1 in reference 353 from top to bottom.g. 268. DNA-DNA. The 14 rRNA branches of the Comamonadaceae have been numbered in Fig. cattleyae. SDS-PAGE.d and auxanography ARDRA.d DNA-DNA. sinuosum metamorphum delicatum gracile psychrophylum a b 8 14 ARDRA. Genera or species belonging to different rRNA branches can be differentiated by their rRNA. SDS-PAGE. the three genotypic groups could be differentiated by the combined use of HinfI and NciI patterns. If the phena did not support the clustering obtained by other screening techniques and by DNA-DNA hybridization. P. terrigena. immunotyping. Montalieu-Vercieu. 356).d SDS-PAGE.d SDS-PAGE. pseudoflava and P.d and auxanography 1 2 3 4 sepe 6b 6a 6 7 5 9 12 11 10 13 sepe ARDRA. Amplified rDNA restriction analysis. Fairly stable groups were usually obtained. SDS-PAGE. carboxydoflava even showed an identical precipitation pattern. and 353–361. In several rRNA subbranches. DNA-DNA hybridization. suggesting that identification at the species level is possible. and subspecies rRNA branchb Methods for intrageneric species and/or infraspecific differentiationc Acidovorax delafieldii. FAME. d Allows infraspecific differentiation. 2012 by guest Classical Classical Classical Classical Classical Classical phenotypingf phenotypingf phenotypingf phenotypingf phenotypingf phenotypingf The results are from references 239. 360. amplified rDNA restriction analysis.d SDS-PAGE. facilis.d DNA-DNA. flava. and citrulli) and konjaci Comamonas acidovorans testosteroni terrigena DNA hybridization group 1 terrigena DNA hybridization group 2 terrigena DNA hybridization group 3 Hydrogenophaga flava.d ARDRA. P.d immunotyping. the groups involved were classified and the phenotypic results were used to discriminate them at the appropriate taxonomic level. and auxanography ARDRA. SDS-PAGE. avenae.org/ on January 3. Polyphasic taxonomy in the family Comamonadaceaea including results and taxonomic resolution of the techniques used Genus. individual strains showed aberrant profiles because of the presence of one or more heavy protein bands that affected the numerical interpretations. the 16S to 23S rDNA spacer region. In some genera.d ARDRA. The lowest values (57 to 58%) were found for some of the misclassified Aquaspirillum species (361). the numerical interpretation of the profiles of SDS-PAGE of whole-cell proteins allowed us to group large numbers of genotypically related strains. Phenotypic analysis based on classical physiological and morphological characteristics and a wide variety of carbon assimilation tests performed in miniaturized commercially available galleries (bio Merieux. e. facilis turned out to be the only members of the Comamonadaceae to react. species. TABLE 4. auxanography. DNA-DNA. the moles percent G C range varies from 1% in Xylophilus (359) to 9% in Comamonas (361). For Hydrogenophaga species (351). 351. analysis of fatty acid methyl ester profiles. France) allowed us to cluster the ´ strains by computer-assisted numerical interpretations into different phena (351.d and auxanography ARDRA. although to a minor degree. and auxanography Downloaded from http://mmbr. and auxanography rRNA. analysis of electrophoretic protein patterns. 300.d and auxanography rRNA. The value of amplified rDNA restriction analysis for identification of the phylogenetically and phenotypically delineated genera and species within the Comamonadaceae has been investigated (331). SDS-PAGE. 356–358. the phenotypic grouping was considered decisive for taxonomic conclusions. taeniospiralis. and auxanography ARDRA. DNA-DNA. MICROBIOL.d DNA-DNA. SDS-PAGE. DNA base ratio. P. Immunotyping. SDS-PAGE.d immunotyping. with the antiserum against Hy- . A. Furthermore. When the phena corroborated the clustering found by other techniques. Whole-cell protein analysis. Sweden) was extensively used to unravel the ¨ internal structure of Comamonas and Hydrogenophaga species.d DNA-DNA. 355. 296. The immunotyping technique (319) as designed by Falsen (Culture Collection of the University of Gote¨ borg. Xylophilus (359) and Comamonas (361). palleronii. temperans avenae subspp. immunotyping.d DNA-DNA. SDS-PAGE. and FAME ARDRA. Within the different genera. f According to Krieg (171). immunotyping confirmed the close relationships of the yellow hydrogen oxidizers originally and erroneously classified as P. pseudoflava. c ARDRA. Numerical analysis of phenotypic features.d DNA-DNA.d immunotyping. P.424 VANDAMME ET AL. e At the base of the 14 rRNA branches in the Comamonadaceae. and P. FAME. 361).

general conclusions about the suitability of this method to differentiate all members of the Comamonadaceae cannot be drawn (Table 4). terrigena).asm.” whose distribution can be traced in a phylogenetic progression. 353). Every classification should be an attempt to arrange the natural diversity among strains into a hierarchical system. Wayne et al. Different taxonomic levels are discriminated to score the extent of divergence. both interspecies and intergeneric distances. Despite their separate genotypic positions. In the genus Comamonas. and DNA hybridization is acknowledged as the reference method to establish relationships within and between species. Experience showed. 239. Strategies for polyphasic identification of unknown isolates within the Comamonadaceae can be deduced from Table 4. testosteroni by measuring the levels of myristic acid (14:0). These monophasic approaches include the ancient “form” classification and pathovar systems and also rRNA sequence analyses. Therefore. acidovorans. Taxonomic studies have often revealed that classifying bacteria by a single approach conflicted with results of polyphasic analyses. Comamonas and Acidovorax. It can be argued that this is the underlying reason for its superior stability. and different methods have been used (238. a single genus corresponds to one rRNA subbranch only. EVOLUTION OF POLYPHASIC TAXONOMY AND PERSPECTIVES The problem of classifying bacteria into orderly arranged taxa was initiated in the previous century (34. the polyamine profile was rather simple and contained only hydroxyputrescine and putrescine as major components. Low-molecular-weight RNA profiles allow us to differentiate Comamonas and Hydrogenophaga species from the rRNA group I pseudomonads (148). This species definition was based on a large amount of experience combining both DNA hybridization data and other features. (343) defined a species as an entity which included “. 300. 300. Also. Numerical analysis revealed that most species investigated form rather homogeneous clusters. and Variovorax). and the Aquaspirillum species) and one (C.VOL. 27. delafieldii. saccharophila constitute a cohesive group sharing the same distinct character states for aromatic biosynthesis (297). as measured by this technique. e. The electrophoretic pattern of different enzymes was used to characterize C. . although different fingerprinting techniques (Table 4) often allow reliable species identification. acidovorans. The major fatty acids are palmitoleic acid (16:1 cis 9).strains with approximately 70% or greater DNA-DNA relatedness and with 5 C or less Tm. The concordance of the phylogenetic tree based on 16S rRNA and the clustering observed for character states of aromatic amino acids biosynthesis has been documented (26. and phenotypic characteristics had to agree with this definition. the technique of DNA-DNA hybridization did not gain large popularity. C. For family and genus level allocation. A classification which considers all information on all known aspects of a particular organism covers essential parts of its genome. . testosteroni) containing small amounts of several 2-hydroxy acids. its ability to discriminate the species and genus levels is unclear. 356). However. Recent studies have shown that C. the bacterial species concept is generally accepted among taxonomists. The odyssey of bacterial classification is characterized by several milestones which led toward our present insights and ideas.and intrageneric similarities within this family. Although proclaimed as the “gold standard” to delineate bacterial species (343). Enzymes which participate in the biosynthesis of aromatic amino acids exhibit a diverse assemblage of qualitative “character states. because each branch is composed of only a few strains and because not enough phenotypic data are available to clearly differentiate these branches from all the other aquaspirilla (256). while for others (Xy- lophilus. Polyamine patterning. it is clear that rRNA information is needed. Tamaoka et al.org/ on January 3. Cellular fatty acid analysis. Any character which reveals part of this variability is therefore useful and can be considered. on the species level. . facilis and A. and complete triangles of hybridization values between related species are extremely rare. A complete triangle based on 790 hybridization values for Downloaded from http://mmbr.g. 238. Conclusion. (300) differentiated C. palmitic acid (16:0) and cis-vaccenic acid (18:1 cis 11) (238.g.. acidovorans and C. 2012 by guest . appeared to be uncorrelated with the genetic relatedness observed in Hydrogenophaga and Acidovorax species (351. 268. 60. Hydrogenophaga facilis. Each subbranch is genotypically sufficiently distinct to represent a separate genus. as discussed below. 239. 356). 296. which was also found by phenotypic and chemotaxonomic methods but which could not be confirmed by DNA-DNA hybridizations (356). Variability among organisms is expressed in numerous different molecules. DNA-DNA hybridizations are recommended. 25. 239.. Other features. testosteroni (300). The physiologically and morphologically diverse group of the Comamonadaceae constitutes one phylogenetic lineage as measured by rRNA similarities containing 14 rRNA subbranches. 268). however. A combination and evaluation of the data obtained by the stepwise use of techniques with complementary discriminative power were applied to further unravel the inter. we avoid proposing nomenclatural changes for the different misclassified Aquaspirillum species. For differentiation between most species. 268. together with a small amount of spermidine. because their application in polyphasic approaches is not always straightforward. testosteroni. 356). Cellular fatty acid analysis was applied only to a limited number of strains of certain species. 35). 1996 POLYPHASIC TAXONOMY 425 drogenophaga pseudoflava. Two fatty acid groups were discriminated by Stead (296) and by Sakane and Yokota (268): one without 2-hydroxy acids (C. the plant-pathogenic members. that some of these characters have superior value and are relevant parameters for particular taxonomic ranks. C. terrigena and C. . which explains why monophasic classification systems are bound to fail. Minimal numbers of phenotypic characters were replaced by large numbers of phenotypic and genotypic data which rendered classification systems more stable and which are now embedded in polyphasic classification systems. DNA Hybridization Studies At present. This suggests again a closer relatedness between A. Since polyamine patterning was not used on a representative selection of all species in the family Comamonadaceae. The comparison of fatty acid profiles was intensively used in the genera Hydrogenophaga and Acidovorax (351.” Both values were to be considered. 125). 350). All members of the -subclass of the Proteobacteria are characterized by the presence of 2-hydroxyputrescine (24. 296. Hydrogenophaga. A ubiquinone with eight isoprenoid units on the side chain was found to be the major respiratory quinone component in all members of the Comamonadaceae tested (39. A number of rRNA subbranches have been united in one genus. new taxa were not created when phenotypic differentiation was not possible (e. and P. Some of these characters are discussed in detail below. The absence of 3-hydroxypalmitic acid (16:0 3-OH) has a limited value to discriminate the members of the Comamonadaceae from other related groups (353).

and B.” Logically.g. All these species are readily differentiated by means of classical phenotypic tests and wholecell protein analysis.a distinct geno- species that cannot be differentiated from another genospecies on the basis of any known phenotypic property not be named until they can be differentiated by some phenotypic property. upsaliensis and C. Particularly B. and require great quantities of DNA (287). detected significant DNADNA hybridization values only between C. using the S1 nuclease procedure. coli (25 to 49%) and detected values up to 20% between. As discussed above for A. Preferentially. genus. and it was proposed to denote these genomovars by using numerals. Regardless of the hybridization method used. The level of 70% binding and 5 C Tm is very strict. B. parapertussis. (343). C. or genomovars (309). very different organisms. automated. upsaliensis (243. partly because different methods give different results (121) but also because hybridization data are not always obtained under strictly comparable conditions (optimal. in laboratory practice. 315. C. genomospecies. Aside from the taxonomic categories species.asm. 183 strains was given for Xanthomonas species (332). DNAbinding values for strains belonging in the same DNA hybridization group were never below 60%. coli and C. and C. Comparison of these values with the percent difference in 16S rRNA sequence homology confirms the importance of the DNA-DNA hybridization technique and the conditions used. On the other hand. since it contains three clear-cut genotypic groups constituting one phenotype. more rapid. jejuni (36 to 43%). 349): Bordetella pertussis. . DNA hybridization techniques present severe disadvantages because they depend on physicochemical parameters. which clearly is a more realistic standard. Obviously. not names (309). coli (23 to 44%). it is not easy to discriminate low but significant DNA binding from nonsignificant DNA binding. Even with this less strict definition. (264. are cumbersome. many species of different phylogenetic lineages contain genomovars. rRNA Sequence Analysis It is well documented that the deep phylogenetic relationships between living organisms can be deduced from sequence Downloaded from http://mmbr. Ursing et al. The genus Bordetella presents a striking example. finally. it is just not practical to create additional names for groups which cannot be readily identified.. miniaturized. It might therefore be safer for such comparisons to distinguish three categories of DNA-DNA relatedness. C. C. whereas they are to be considered by definition to be species that cannot be reliably differentiated by phenotypic tests. (309) considered species to be groups of strains sharing 50 to 70% DNA reassociation and 5 to 7% difference in thermal stability between the homologous and heterologous duplexes. are not cumulative. upsaliensis differs in 5% of its 16S rRNA sequence from C. and B. They often created more problems than they resolved. pertussis. the conditions for hybridizations are extremely important and only results obtained under comparable conditions can be evaluated objectively. Bordetella comprises an extensively studied group of gram-negative bacteria. 322). genospecies. although values of 50% also occurred. subspecies. . lari and several other Campylobacter species from other rRNA clusters such as C. quantitative comparisons of DNA hybridization values generated by different techniques should be handled with extreme caution. upsaliensis and C. jejuni differ only in 1. 321) and has been the subject of DNA-DNA hybridization studies. which state that phenotypic characteristics can override the phylogenetic concept of species in a few exceptional situations. jejuni. genomic groups. low but significant DNA relatedness. which are easily differentiated by several morphological. As long as DNA-DNA hybridization experiments form the cornerstone of the bacterial species definition. which would indicate that the degree of DNA hybridization is too low to be measured by the method used. Although these are sometimes initially meant as temporary tools to handle unsolved problems. Roop et al. parapertussis. Ursing et al.426 VANDAMME ET AL. In other cases. REV. However. or lack of imagination. It was recommended (343) that “. about 40% DNA-DNA binding is detected by the S1 nuclease procedure (264). closely related to the genus Alcaligenes (61) and comprising six species (322. lari and C. or suboptimal). Phenotypically similar but genotypically distinct groups of strains have been referred to as genomic species. biochemical. Cases like this were foreseen in the guidelines of Wayne et al. hinzii. The average levels of DNA binding between strains of different DNA hybridization groups were typically less than 40%. DNA hybridization values conflicted with data derived from other analyses. jejuni. again. jejuni and C.5% of their 16S rRNA sequence (294). . (308) reported low but significant DNA-binding values between C. C. 322. Analysis of such observations suggests an uneven distribution in taxonomic space. However. stringent. avium. lari and C. DNA hybridization data obtained in various laboratories do not always correlate well. and family. biovar. holmesii. a tremendous amount of information must be encoded in a limited part of the Bordetella genome. and. B. and. One of the rRNA clusters within the genus Campylobacter comprises C. B. these three species share more than 80% DNA binding (167. As stated above. with clusters of highly related strains that have more than 80% DNA-binding values (332). coli (26%). C. which would denote the range of significant hybridization values below the species border (the depth of this range depends primarily on the technique used). delafieldii and A. serious difficulties were encountered in applying the 70% DNA hybridization rule. MICROBIOL. taxonomic relationships will not be clarified by merely introducing additional categories. e. terminologies. nonsignificant DNA relatedness. there will be a need for a new. coli (294). and chemotaxonomic characteristics. 2012 by guest . sputorum. bronchiseptica. 349). e.. . B. facilis. because these taxa are given an infraspecific rank. which would denote relationships between strains of a single species. a number of additional categories have been introduced.org/ on January 3. 265). whereas about 40% DNA-DNA binding is detected by the hydroxyapatite method (270). fetus and C. biotype. The name “genomovar” may be misleading. B. or both. and most of them have clearly different whole-cell fatty acid components as well (314. jejuni (29 to 40%). conceptual weakness. lari. which can lead to confusion in the species delineation and hinder the identification. Although the presence of genomovars is obviously too important to be ignored. and genomovar.g. such as high DNA relatedness. Examples of such lack of correlation are found when comparing results of DNA hybridization studies among campylobacters. Using the hydroxyapatite method. and often phenotypic consistency would not be achieved if these recommendations were strictly applied. a number of simple and straightforward tests (at least two [10]) should endorse species determination based on DNA hybridization values. functioning more as smoke screens to hide uncertainty. bronchiseptica are. but similar values were also found between C. If there were indeed a simple linear relationship between thermal stability and DNA mispairing as described by Bautz and Bautz (13) and Ullmann and McCarthy (306). Comamonas terrigena is such an example. C. and standardized method. lari and C. coli. it is practical to differentiate species on a routine basis only if they can be readily differentiated. and C.

The variability was thought to represent interoperon variation within a single strain. and selection of sequence stretches analyzed but also on the number of organisms and the selection of reference organisms. misidentification of strains. in fact. identification of microbes in the day-to-day practice of the routine microbiology laboratory relies nearly exclusively on those differential phenotypic features. Within this wave of enthusiasm. Analysis of the available data revealed interoperon differences from 0 to 5% and interstrain differences from 0 to 16%. or other laboratory errors (33). elongation factor Tu (194). It is recommended to determine the significance of the branching levels in a phylogenetic tree by means of bootstrap analysis (173). phenotypic data are compared with literature data. behavioral. However. Unexpectedly high levels of intraspecies variation (within and between strains) of 16S rRNA sequences were found. .” Phenotypic Data In taxonomic practice. no matter which DNA-DNA hybridization method is used. Total rDNA sequences are accumulating rapidly and are accessible via international databases (66. In fact. should preferentially not be used for taxonomic conclusions unless it has been shown that they give the same degree of similarity as that obtained with full sequences (287).VOL. which would render them more suitable for differentiation of closely related organisms. More recently. strainto-strain variation within a species. 15 February 1995). a paraphyletic group in which systematic designations are so often provisional. Stackebrandt and Goebel (287) presented a taxonomic note on the place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. The closing remark of Clayton et al. Therefore. the examples of polyphasic taxonomy described above) or were substantiated by a polyphasic approach on a large number of strains. Stackebrandt and Ludwig (289) discussed the importance of choosing outgroup reference organisms in phylogenetic studies. organisms sharing more than 97% rRNA similarity may belong to a single species and that the resolution of 16S rRNA sequence analysis between closely related organisms is generally low. but for a well-studied taxon it may also be achieved by the selection and evaluation of taxon-specific primers from that sequence. . Their review of the literature data revealed that. which appears to undermine the application of rRNA sequences for phylogenetic and identification purposes and to question the ability of this method to identify unculturable organisms or to study microbial communities without the need for cultivation (3). even if this seems to be common practice. The report of the Ad Hoc Committee on the Reconciliation of Approaches to Bacterial Systematics stressed that any phylogenetically based taxonomic scheme must also show phenotypic consistency (343). and their molecular evolution rate should be comparable to or somewhat higher than that of 16S rDNA. a number of dissonant voices were recorded. In 1994. (33) covers both past and future classification systems and is indeed certainly valid for bacterial systematics: “. There is simply no threshold value of 16S rRNA homology for species recognition (287). This indicates that although the scientific rationale of rRNA sequence analysis cannot be doubted. (33) presented a detailed comparison of duplicate rRNA sequences present in the GenBank database (Release GenBank87. the conclusions based on rRNA sequence analysis confirmed results obtained by equivalent techniques such as DNA-rRNA hybridizations (in which several strains were usually included) or rRNA cataloging (cf.Assessing within-taxon variability of the characters examined has long been standard procedure in animal and plant systematics and ecology. 226). 226. 2012 by guest . 363). It is important not to restrict the selection to a large number of apparently related taxa. Analysis of partial sequences. several other macromolecules have been examined for their potential as microbiological clocks. Among others. the beta subunit of ATPase (194). one should be prudent with conclusions based on single-sequence analysis. However. In Downloaded from http://mmbr. especially in prokaryotes. organisms with less than 97% 16S rRNA sequence homology will not give a DNA reassociation of more than 60%. Three phenotypically similar Bacillus strains exhibited more than 99. RNA polymerases (374). 1996 POLYPHASIC TAXONOMY 427 comparisons of rRNA molecules and from some of their signature features (123. The depth in an rRNA dendrogram at which a given hierarchical line is to be drawn should remain flexible to achieve phenotypic consistency. and tRNAs (147. it is best to include a wide range of apparently related and apparently unrelated reference organisms.5% rRNA sequence similarity (rRNA sequences are considered identical if they differ in less than 5 to 15 positions [102]). with remarkable results. 148) were shown to be valuable molecular chronometers in bacterial systematics. The different phylogenetic positions of Rhizobium galegae obtained by partial and total 16S rDNA sequences (223. Clearly. in numerous reports. It has been proposed to refer to such taxa with (virtually) identical rRNA sequences as rRNA species complexes or rRNA superspecies (102). whereas DNA hybridization experiments indicated that they belonged to two distinct species. although valuable. provided that the rRNA similarity level is below 97% and that rRNA sequence data of all relevant taxa are available for comparison. chaperonin (340). which would avoid the need for multiple-sequence analysis and has the advantage that it is feasible to examine many strains. (102) reported that 16S rRNA sequence identity may not be sufficient to guarantee species identity. 352) support this. Fox et al. families. tempo of evolution. Frequently. Alternatively. and other higher ranks but also for species delineation and increasing numbers of new species are described without DNA hybridization studies or equivalent techniques (see below). These alternative macromolecules should be universally distributed among bacteria. In 1992. It is more than relevant to know that the branching order in phylogenetic trees depends not only on differences in base composition. However. sequencing errors. which were or might have been obtained by using other conditions or methods. phenotypic characterization became compromised and sometimes more of a burden than a useful taxonomic activity.org/ on January 3. It should come as no surprise that nucleic acid sequences require similar scrutiny. unequal rates of evolution in different regions of the rRNA genes. In 1995. their genes should not transmit horizontally. various ribosomal proteins (224).asm. 60. Clayton et al. Confirmation of the validity of an rRNA sequence requires repetitive assays. rRNA sequence analysis was no longer used exclusively to determine relationships between genera. the problem of errors in some sequences present in the EMBL database should not be underestimated. Comparisons of the 16S and 23S rRNA sequences revealed the phylogenetic framework of bacterial classification and became indispensable in polyphasic taxonomy. whether features studied are morphological. or biochemical. Therefore. in general. The presence of universal bacterial sequences in the rRNA molecule allowed us to classify unculturable organisms and to perform phylogenetic identifications and in situ detection of individual cells without cultivation (3). this is the bottom line of polyphasic taxonomy. rRNA sequence analysis may replace DNA hybridization studies as part of the description of new species.

65. chromatographic separation.. cannot be adequate. In this discussion. the characteristics of one isolate cannot represent the phenotypic variability of an entire species. a single strain can unambiguously be shown to represent a new genus or a new species within an existing genus. mala fide or not. The phenotypic description of bacterial families presents even more problems. It is important to determine the resolution level of this technique for every taxon under investigation. The differentiation of Bordetella avium and B. 6% in Arcobacter. but at present.. it can be recommended that new species containing a single strain or new genera containing a single species be validly named only if their genotype and phenotype have been thoroughly and adequately characterized. The question whether to validly name such strains has been the subject of many debates. the L. the large number of species and the overall phenotypic similarity between some closely related species (e. similar fatty acid patterns may be shown by diverse taxa representing different taxonomic ranks. For many taxonomists. The genotypic and phenotypic characteristics of this strain can be determined and compared with those of strains of its nearest neighbors. Since only few 16S rRNA sequences are available in this group. Whole-Cell Protein Analysis Numerous studies have documented the application of whole-cell protein electrophoresis to taxonomic studies (47. or both. the discriminatory power is considerably higher when strains are grown for 48 h than when they are grown for 24 h (149. polyphasic approaches combining rRNA sequence data and phenotypic data entailed the delineation of the families Campylobacteraceae at 16% (294. The same could be argued for the description of a single species which represents a new genus. Similarly. for Burkholderia species (111). grown for 24 h. and it is often necessary to prolong the incubation period to generate sufficient cells for each strain under the same growth conditions. which is at that moment represented by a single strain. different species have identical fatty acid profiles (345). when cultivated under highly standardized conditions. Prolongation of the incubation period cannot generally be considered a cause of reduced resolution. whole-cell fatty acid analysis allows differentiation and identification of individual species. Del. thus also restricting the knowledge of the phenotype. were no longer distinguishable (322). many isolates which do not represent different strains (e. e. as was illustrated above for the Comamonadaceae. However.. which is based exclusively on auxanographic results obtained by an API system which is no longer commercially available in the same format. each with advantages and disadvantages. in our experience. 255. individual species within genera are often not differentiated. Clearly. whereas the same strains. or even subspecies. The discussion will probably linger. Additional examples have been described by Welch (345). strains must be grown under highly standardized conditions to obtain reproducible results. because when large numbers of strains are examined. a maximum of 5% sequence dissimilarity is estimated indirectly.asm. and Alcaligenaceae at 5. Pasteurellaceae at 6% (77). The applications and restraints of the technique were extensively discussed and documented by Welch (345).. 335). this is not always possible. REV. Alternatively. e. The need for a continued phenotypic characterization at every taxonomic level cannot be denied. 371). 164. Newark. whereas for lactobacilli. this is sufficient to validly name this organism. Although they are accepted as necessary.g. The method therefore allows the comparison and clustering of large numbers of strains with minimal effort and yields descriptive information to characterize the organisms. However. Therefore. and 197). and therefore. the two situations cannot be validly compared. while for others. both in taxonomic studies and in identification analyses (345). The procedures for extraction. and therefore the description of the new species. 22. were easily differentiated (46). 8% in Helicobacter. nomenclatural corrections. and by the description of.g. these modifications jeopardize the credibility of taxonomists in the microbiology community. For some genera. the percent difference in 16S rRNA sequences between the genera is indirectly estimated at around 5 to 6%. differential phenotypic characters are often hard to find with a reasonable amount of effort and time. However.g. The rationale for this method is (i) that strains. it should be mentioned that the application of phenotypic fingerprinting systems (e. Within the Comamonadaceae. In light of the increasing application of phenotypic fingerprinting systems. and 5% in Bordetella. it is perfectly possible that a genus comprises only a single species. In a thorough taxonomic study. daily samples of a single source). hinzii is an example of such a loss of resolution: cells of both species. The limited consensus between results obtained by classical phenotypic growth tests and commercial systems is the origin of the decrease in the availability of reliable tests for the description and identification of new species. There are different views. It can be argued that a classification based on results obtained with a single strain cannot be stable and may warrant future emendations of descriptions.. Importantly. 317). In the case of the above-mentioned examples of genera delineated by polyphasic approaches. e.g.. Biolog) and their inclusion in official descriptions restrict the examination of the bacterial phenotype to a minimum. will be Downloaded from http://mmbr. because in other groups such as Aeromonas and Xanthomonas species. which are based largely on Biolog data. For Campylobacter and its relatives.428 VANDAMME ET AL. In the framework of polyphasic taxonomy. depending on the group of bacteria studied.org/ on January 3. Whole-Cell Fatty Acid Analysis Whole-cell fatty acid analysis is increasingly used. This problem is clearly illustrated by the newly described Xanthomonas species (332). 10% in Campylobacter. practice.. cellular fatty acid analysis is often very useful as a rapid and fairly inexpensive screening method. The variability of the genus-specific characteristics can also not be estimated if only a single species is available.g. the depth expressed as a percent difference in 16S rRNA sequences corresponded to about 4% in the genus Xanthomonas.g. Acidovorax temperans. not only to delineate taxa and appreciate their phenotypic coherence but also to evaluate their physiological and ecological functions. grown for 48 h. It is impossible to estimate the variability of the phenotype in the one strain-one species case or in the one strain-one species-one genus case. two additional points should be mentioned. Cardiobacteriaceae at 8.5% (84). some grow slowly. as. The genera in the family Comamonadaceae show a phylogenetic heterogeneity corresponding to a Tm(e) of their DNA-rRNA hybrids of between 2 and 5 C. as illustrated above for the lactobacilli. the phenotypic inertness of the bacteria has prevented the development of a solid phenotypic identification scheme. The commercial microbial identification system (Microbial ID Inc. it is often very simple to obtain. for which many recent examples exist (see. Neisseriaceae at 7% (75). 2012 by guest . acidophilus complex) did prevent a reliable method of species determination until genotypic methods were used.5% (76). references 19. A minimal phenotypic description is not only the identity card of a taxon but also a key to its biology. and data analysis are relatively simple and highly automated. MICROBIOL.) gives strict guidelines and provides databases containing species-specific fatty acid patterns.

comparison of the 16S rRNA sequences of two of these organisms with those of other helicobacters revealed only 96.. As an example. Alternatively. however. Undoubtedly. In this context. wholecell fatty acid analysis.. Capnocytophaga. For certain taxa (e. 255). In case the difference in 16S rRNA similarity with its closest neighbor is very large. Strategy in Polyphasic Taxonomy From the examples and discussion given above. The classification of new or unusual isolates will. protein electrophoresis has an important drawback: in contrast to. since closely related groups of strains can differ considerably in generation time (55). This allows large numbers of strains to be compared and grouped in clusters of closely related strains (47. Provided that highly standardized conditions are used throughout the procedures of cultivation and electrophoresis. Although these techniques are primarily applied to infraspe- cific strain comparisons. it does not supply descriptive information. and Leptospira (258) and the family Comamonadaceae (331).g. In our view. it can be argued that the unculturable organism represents a novel taxon.asm. to provide the scientific community with useful information which is also easily accessible. 1996 POLYPHASIC TAXONOMY 429 characterized by a protein composition which can be separated by electrophoretic techniques and visualized by staining procedures. 2012 by guest . Any chosen strategy may be influenced in the course of the study by the results obtained. Rhizobium and Bradyrhizobium (174). Brevibacterium (29).VOL. Clostridium (122). because in general. and Naegleria [108]). The available data reveal that this technique is useful primarily to identify strains at the species or infraspecific level. unculturable bacteria morphologically different from H. in part. e. In several of these studies. often require a polyphasic approach. at present. two different screening techniques should be available to compare and group large numbers of strains.g. 255). and (ii) that a high similarity in whole-cell protein content is usually an indication of extensive DNA hybridization (47. Polyphasic Identification Choosing a polyphasic approach to bacterial classification raises the question of identification of individual isolates. computer-assisted numerical comparisons of protein patterns are feasible and databases can be created for identification purposes (164. Databases of rRNA sequences and whole-cell fatty acid components and miniaturized batteries of phenotypic characteristics exist. Preferably. a stable entry can only be the product of a polyphasic analysis which will reduce the amount of work for subsequent monophasic identification schemes. but later rRNA analysis showed that they belonged to the Helicobacter lineage. though.. it should not be forgotten that most isolates in routine microbiology laboratories are identified without difficulty by using conventional phenotypic schemes. species-specific DNA fragments were generated (e. allowing the identification of many isolates. an obvious advantage of this technique is that once the correlation between percent similarity in wholecell protein composition and DNA relatedness for a particular group has been established. In these cases. although other macromolecules probably have a similar or even additional potential. standardizing the growth period is not always possible. they may prove to be useful in classification as well. a higher discriminatory power will normally result). Therefore. numerical analysis may be hindered by the presence of distorted protein profiles or hypervariable dense protein bands. PCR-based typing methods with random or repetitive elements as primers have been applied to strain characterization in a wide variety of bacteria (310). The scope of the study is determined primarily by the problem itself. DNA hybridization experiments. it should be obvious that there are no simple and straightforward guidelines for performing polyphasic taxonomic studies. These specific DNA fragments may be useful as probes to rapidly screen and identify other isolates. the identity of none of the protein bands is revealed. some of them have found broader taxonomic applications. as well as for the fine typing of individual strains (153).g. pylori were found in the human stomach and named “Gastrospirillum hominis” (205). For its application in polyphasic taxonomy. for species belonging to the genera Campylobacter. Classifying 100 strains without any preexisting knowledge of their possible identity and classifying a single unidentified Bordetella strain are totally different problems. rRNA or rDNA sequence analysis is the best choice.org/ on January 3. The genotypic structure of such groups may be analyzed by DNA-based typing techniques. rhizobia). Unculturable Bacteria Another remaining problem concerns the classification and nomenclature of unculturable bacteria which are only minimally characterized by morphological characteristics or by differences in a molecular sequence (220). 254. some bacteria were named just because they differed morphologically from the other bacteria present in the same habitat. the phenotype should be characterized. Although most of these methods are particularly useful in determining infraspecific relationships. The first results of the AFLP technique applied to Xanthomonas and Aeromonas species suggest its potential usefulness at the species and subspecies levels.6 Downloaded from http://mmbr. it can replace. that in some groups.5% rRNA similarity between the two sequences and values of 96. strains within a single species may have rather different whole-cell protein patterns and therefore that differences in whole-cell protein composition do not necessarily imply low levels of DNA hybridization (323). if a spacer region is included. DNA-DNA hybridization studies of a limited number of well-chosen strains are required to delineate individual species. this assumption does not take into account that a 16S rRNA or rDNA sequence is not available for all known bacteria for comparison.. the introduction of molecular biological techniques into the microbiology laboratory yielded a large variety of DNA-based typing methods. Researchers who intend to perform polyphasic analyses should be equipped with techniques which allow them to determine the phylogenetic affiliation of an unknown isolate. However. visual comparison is essential to interpret the similarity of protein patterns. However. the resolution of this method is determined primarily by the target of the PCR (e. However. Numerical analysis of the patterns generated with several restriction enzymes provides supraspecific information as well. but ultimately. 60. Finally. 318). It should be remembered. Obviously. DNA-Based Typing Methods As described above. the success of these databases depends on the exactness of the methods used and on how carefully the individual entries were delineated.g. Amplified rDNA restriction analysis has been used to examine the taxonomic structure of a growing number of genera including Streptococcus (155). and so the species was renamed “Helicobacter heilmanii” (286). For some bacteria.

and 98. additional gastrospirilla were detected in the stomachs of a lemur and of pigs. and these were named “Gastrospirillum lemur” and “Gastrospirillum suis” (208). because all these rRNA similarity values are about 97% or higher and there is. an organism which is extremely fastidious. A classification system makes sense as long as this duality exists. The science of experimental evolution might not only lead to fresh ideas on the bacterial species concept but might also eventually lead to the new field of experimental bacterial taxonomy. more bacteria will be detected (whether they can be cultivated or not). rRNA sequence comparison of the latter organism revealed more than 99% similarity to one of the “G. and numerous DNA sequences will be accumulating. each containing a different taxon. From studies conducted in this field during the last 20 years. and descriptive methods which will provide further genotypic and phenotypic information. (iii) Polyphasic classification is purely empirical. there will be more automation. from the lack of association of alleles at different loci. Perspectives and Conclusions One of the perspectives in bacterial taxonomy is that technological progress will dominate and drastically influence methodology. characterized by a certain degree of phenotypic consistency and by a significant degree of DNA-DNA hybridization and over 97% of 16S rDNA sequence homology. 6) is available. Although the idea is presently purely speculative. whereas other results suggest a much higher genome plasticity (221).org/ on January 3. More data will become avail- able. or family delineation. The bacterial species appears to be an assemblage of isolates which originated from a common ancestor population in which a steady generation of genetic diversity (5) resulted in clones with different degrees of recombination (according to the species). and appropriate software packages can represent the results of the similarity comparisons as a two. artificial or not. or is it merely an image of our current limitations to study the full range of diversity. which can only be practical to handle if it is founded in an ordered structure. derived from a common ancestor. Polyphasic taxonomy is an attempt to synthesize the real landscape and a step toward a synthetic taxonomy which will be made possible through the development of new mathematical and information strategies. hominis” strains (207). Although gene acquisition can obviously represent an interesting step in long-term evolutionary development of bacterial strains. and this is the major reason why a polyphasic approach is valuable. dense cores. Population Genetics The link of taxonomy with population genetics is essential to better understanding bacterial species determination. without chromosomal recombination. All possible methods that inform on the biological nature of strains merit equal attention in principle. Nevertheless. it should not occur too frequently. Large data sets should be numerically analyzed. In such plots. and therefore it is at present obviously more appropriate not to refer to such bacteria by the usual binominal species denomination. 2012 by guest . the evolution of the bacterial genomes and thus the process of speciation are not well documented. an insight into a vast amount of data could be the basis for a perfectly reliable and stable classification system. even the best laboratory can provide only a few sets of methods. A limited number of publications on experimental evolution (185) and on the general principles of the evolution of bacterial genomes (5. pending a more thorough characterization and classification. no other way to compare these isolates. a clone being defined as a set of genetically similar cells. often seem to be present along with a number of intermediate strains. at present. and from the comparison of gene trees. there is a huge amount of biodiversity. the following findings are relevant for the present discussion and have been summarized by Maynard Smith (203). Clear evidence of recombination in bacterial populations is available from the mosaic structure of some nucleotide sequences. which will gradually disappear with the mounting piles of information? It is worthwhile to attempt to summarize what we have learned about polyphasic taxonomy as it has been practiced during the last 20 years. It is of no use in bacterial taxonomy to try and formulate a “genetic definition” of a bacterial species. Population genetics deals with the variability of populations of bacteria and the formulation of theories that account for the variability.000 bases only. because recognized species clearly do not correspond to evolving populations that have a common gene pool. Populations of bacteria consist of a number of independently evolving clones. In bacteria. Maynard Smith (203) criticizes the bacterial species concept based on DNA hybridization for arbitrarily imposing divisions upon a continuum. Murray and Schleifer (220) recently proposed that unculturable bacteria be included in a new category. even approximately. it is already sometimes unclear whether it makes sense to order bacteria into a classification system. Although a large number of generations can be easily observed in bacteria. It is obviously not possible to conclude how many species are involved. There is no phylogenetic standard for species. (i) The contours of the polyphasic bacterial species are less clear than the ones defined in the past by Wayne et al. isolated from the gene pool from other species. This is necessary to respect the need for a reasonable degree of genetic stability. the ancestral history of all chromosomal genes is identical and the phylogenetic trees derived from sequences of different genes should be similar. However. the same meaning for all bacterial taxa. with appropriate terms for communication. The most challenging task will definitely be to process this mass of information into a useful classification concept. The different methods used in polyphasic taxonomy can be considered different windows through which the same landscape is seen. Although these names are probably only meant as working designations. and software development will need to address the combination and linking of the different databases. We will also have increasing access to the genome. genus. MICROBIOL. as it always has. with our present data. The crucial question when we have access to growing data sets is the following: will this duality remain. Candidatus. limited local recombination occurs. methods to compare and group large numbers of strains into clusters of similar bacteria. (ii) A polyphasic approach to bacterial classification includes methods to phylogenetically allocate bacteria.or three-dimensional plot.asm. REV. There is no way of defining a species that will correspond to both a phenotypically recognized entity and an evolving unit and that will have. DNA-DNA hybridizations to determine the relationships between representatives within and between each of those clusters.430 VANDAMME ET AL. Undoubtedly. follows no Downloaded from http://mmbr. nor is there a “gold standard” to identify bacterial species. they cause a lot of confusion. Without recombination. which is generally thought to be necessary to ensure the endurance of a species of organisms in ecosystems (6). respectively. In the same period. giving rise to “mosaic structures” in sequence data as a result of horizontal transfer of pieces in the range of 10 to 1. (343) and take into account more elements than before.8% toward Helicobacter felis (104).

. R. 1966. 19. J. Bacteriol.. Oyaizu.. P. Sci.-H. H. Whitaker. 1983. Belland. 1980. S. and easy to handle. Aislabie. Korolik. Xanthomonas Dawson 1939. J. and Hayward 1988) gen. and J. Syst. Rev.. De Wachter. L.) and Senior Research Associate (P. 1994.asm. 2nd ed. Syst. Mich. M. 45:61–66. and R. and the Human Capital Mobility Program Network contract CHRX-CT930194) and to the National Fund for Scientific Research (Belgium) for research and personnel grants (K. Krieg and J. Johnson. J. R. there should be an international interconnection and accessibility of these databases. J. Dworkin. Sly..V. Three supplementary diagnostic tests for Campylobacter species and related organisms. 10. El-Banna. Clin. 1. Syst. Gen. Int. J. and K. Arber. Johnson. L. Microbiol. Anal. bacteria are isolated from new sources.V. A. Tiedje. Burnens. M. Bacteriol. in the emerging theory of fuzzy logic (169). Bradbury. Auling. Mol. Int. Biochem. L. W. 1995. J. Andrewes. Campylobacter hyoilei sp. and R. Archibald. Richards. 187. R.. and H. 13. Zentralbl.-J. R. Baddiley.. (v) Most bacteria in routine diagnostic laboratories will probably continue to be identified by means of classical methods. 1994. Helicobacter nemestrinae sp. Jensen. Identification of xenobiotic-degrading isolates from the beta subclass of the Proteobacteria by a polyphasic approach including 16S rRNA partial sequencing. Description of Campylobacter laridis. The evolutionary pattern of aromatic amino acid biosynthesis and the emerging phylogeny of pseudomonad bacteria. A recent initiative on integrated microbial databases has been taken by J. S. Jensen. Anhaufungsversuche mit Ureumbakterien. Acad. P. Mol. A. Gene 135:49–56. robotization. A. Infektionskr. L. 77:291–315.. and T. For new or atypical isolates or in many research units where. J. Deoxyribonucleic acid sequence relatedness between thermophilic members of the genus Campylobacter. 1990. Gherna. and K. Schoenknecht. nov. and J. Syst. 1988. Int.. Telluria mixta (Pseudomonas mixta Bowman. A. The generation of variation in bacterial genomes. Rake. V. C. Gherna. Lapage.. The teichoic acids. Characterization of Leuconostoc lactis strains from human sources. M. J. J. Stackebrand. and M. Microbiol. Busse. C. J. Dewhirst. Clin. 1992. Bowman. Skirrow. Int. 1995.).. K.. and R. V. Goossens. 22. e. De Wachter. 31: 708–710. (vii) Some laboratories that have developed sequencing and fingerprinting technology also have access to computing facilities for storage and retrieval of fingerprints or sequences and similarity.G. Chem. East Lansing. Jenkins. Chilvers. a novel dibrominated arylpolyene pigment produced by the bacteria Xanthomonas juglandis. Wilcox. 42:19–26. Coloe. 59:143–169. 24. and W. correlation. as is already the case for. C. p. Polyphasic taxonomy is not hindered by any conceptual prejudice. Microbiol.. Bautz. E. Batch procedure for thermal elution of DNA from hydroxyapatite. Hyg. 199–210. D. S. (viii) Microbial taxonomy and biosystematics are a modern discipline that needs further financial and intellectual support to fulfil its role in diagnosis. J. K. L. and J. and results in a consensus type of classification. The necessity for reliability of the data is a conditio sine qua non. Use of ribotyping in epidemiological surveillance of nosocomial outbreaks. I. L.. Chapelle. Starr. R. A. Microbiol. 11:1–8. J. Goossens. 12. Microbiol. nov. Holt (ed. readily available. and F. 7:33–61.. The most direct approach is first to place such an isolate in the phylogenetic framework and then to determine its finer relationships by means of a polyphasic approach. Bronsdon. and M. M. T. Michigan State University. P.. Microbiol. and G.. Cellular fatty acid composition of Campylobacter fetus.-H. I. It has become common practice to represent the outcome of the latter computations by means of dendrograms. REFERENCES 1. Evol. 7:311–327. 1996 POLYPHASIC TAXONOMY 431 strict rules or guidelines. 1973.. Proc. J. 128:2515–2522. 40:555–560. Appl. S. W. Clin. G.). associated with porcine proliferative enteritis. and J. I. 18. Ludwig. P. Little.. a spiral bacterium found in the stomach of a pigtailed macaque (Macaca nemestrina). nov. J. H. Identification of bacteria by computer: identification of reference strains. and K. Bergey’s manual of systematic bacteriology. Adv. e.. J. W. Bastyns. Microbiol. 1964. A. 28:1728–1733. Fanning. P. G. S. the 16S rDNA sequence and fatty acid data. Bautz. Schleifer. satisfying most but not necessarily all users of taxonomic results.. G. A.. and the study of biodiversity and the environment. Sly. 2012 by guest . 1993. J. Syst. B. Vandamme. H. 9. ACKNOWLEDGMENTS We are indebted to the Commission of the European Communities (the Biotechnology BIOTECH-G project contracts BIO2-CT94-3055 and BIO2-CT94-3098.V. Elion.org/ on January 3. R. Polyamine pattern as a chemotaxonomic marker within the Proteobacteria. 16. Blaser. Tetrahedron Lett. Byng. Nicolet. Beijerinck. Baltimore. 3. Bakteriol. Denamur. 17:563–568. T. C. The comparison of different methods is complicated by the vastness of the data and the nonexistence of suitable objective methodology. Spiegel. P. 11. S. W. A.g. Gen. 5. II Abt. Trust.g. 1995. Bastyns. Parasitenkd. F. F. and F. and R. P. R. (vi) The automation.. H. G. R. Arber. J. Paster. Curtis. Shepherd. and support of effective taxonomic reference laboratories is essential. E. a new species comprising the nalidixic acid resistant thermophilic Campylobacter (NARTC) group. Sly. C. Benjamin.. Ideally.. 17. The Williams & Wilkins Co. J. ¨ 1992. Syst. Owen. 1982.. may integrate any significant information on the organisms. E. and B. J. Description of bacteria able to degrade isoquinoline in pure culture. 1991. Bascomb. J. Bingen. A. 45:4023–4024. 1995. 1994. J. e. H. 8. Bacteriol. Can. 21. and G.) and for positions as Postdoctoral Research Fellow (P. 21:323–375. M.S.K. Vandamme. 43: 120–124. or distance coefficient calculations and a variety of clustering techniques.. Bacteriol.D. and E. 60. Structure of xanthomonadin I. the better the outcome might reflect its biological reality. 40:7–12. Variable enzymological patterning in tyrosine biosynthesis as a means of determining Downloaded from http://mmbr. 4. J. and Telluria chitinolytica sp. 1993. Harder. nov. 2. E. This would mean that bacterial strains do not have to belong to a single cluster. The analysis and outcomes of different fingerprinting methods require the development of new methods of data fusion (or data aggregation). The prokaryotes. S. 1901. 11:448–451.. Microbiol. Leaper. G. W. 1980. C. R. A. M. Mol. which is based on the simple yet powerful idea that an object is not forced to either belong or not belong to a given set but can have a partial degree of membership in this set. Hayward. J. E. J. biotechnology... K. Springer-Verlag. The classical evolutionary tree may have horizontal shunts through which successful genetic information can be made available to other branches of the tree (5). BRIDGE project BIOT-CT91-0294. H. Moss. Specific detection of Campylobacter concisus by PCR amplification of 23S rDNA areas. Ure¨ umspaltung durch Urease und durch Katabolismus. Natl. Rev. 1976. Probes 9:247–250. 28:447–459. J. F. J. Chapelle. K. A. Microbiol. which are adequate. R. soil-dwelling organisms which actively degrade polysaccharides. 1993. In N. Goodwin. Species-specific detection of campylobacters important in veterinary medicine by PCR amplification of 23S rDNA fragments. Cell. Microbiol. 25. nov. R.VOL. J. except that the more information that can be integrated on a group of organisms. a straightforward identification of microorganisms by one single method is often not possible. Amann. Truper. One of the solutions could be found. Evol. and development of databases will further develop. Balows. S. 26. New York. Hope.D. The influence of non-complementary bases on the stability of ordered polynucleotides.g. Auling. Microbiol. 15. G. 41:148–153. 7. its validation through the sequencing of other universally occurring genes is only a matter of time. 14. Schleifer. 19:247–257. 23. Curr. E. USA 52:1476–1481. and G. R. P. (iv) The introduction of 16S rDNA sequences has definitively brought the phylogenetic dimension into bacterial taxonomy. Brenner. M. 27. D. W. Barreau. N. comb. Weaver.. Busse.. 8:231– 238. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Clin.. Wagener. J. This tendency of identification to become polyphasic is an unavoidable reality... cheap. Appl. and R. 1983. vol. Most of the computational techniques involved were developed in the 1960s and are designed for processing the data obtained by a single method. Whitaker. Byng. Y. 20. Alderton. and several methods are needed. 6. Evolution of prokaryotic genomes. 1984. 1969. Carbohydr.

and S. Int. P. 1990. 35:443–453. 55. E. C. P. 1994. B. Lizzaraga. nov. J. F. 1995. A. De Ley. Goodfellow and A. Haesebrouck. and J. Kersters. Lett. J. J. H. M. The genus Leuconostoc. Bacteriol. 1955. J. 76. A. p. 57. 77:5–12. M. nov. Vescovo. Five new Legionella species isolated from water. and J. 51. K. and J.. Pot. Phylogenetic analysis of the genus Lactobacillus and related lactic acid bacteria as determined by reverse transcriptase sequencing of 16S rRNA. and A. Hinz. 1990. De Ley. J. H. J. Devriese. and Comamonas terrigena Hugh 1962 sp. 1984. Bacteriol. 265–311. Truper. Cowan. D. 1985. Tytgat. G.. V. 1992.and intrafamilial similarities of rRNA cistrons of the Pasteurellaceae. 61. Vesey. L. and S. 1991. Syst. Microbiol.. J. D. J. p. Swings. A dictionary of microbial taxonomic usage. Compilation of small ribosomal subunit RNA sequences. 26:181–184. 9:125– 131. Adv. natural relatedness among the Pseudomonadaceae. Vandamme. vol. Goor. Syst. 1876.. “Bordetella hinzii. F. Int. 44. Beitr. A. de Lajudie. Holzapfel (ed.. ¨ W. In M.org/ on January 3. 2012 by guest . Microbiol.. Int. Comamonas Davis and Park 1962 gen. and M. P. tolerans. Colwell. A. and J. Ltd. 45. Zanoni. J.. Guesdon. J. 48. Dennis. Dellaglio. M. 1980) with the type strain of Lactobacillus crispatus (Brygoo and Aladame 1953) Moore and Holdeman 1970. W. Comparison of the of Lactococcus lactis differential medium (DCL) and SDS-PAGE of whole-cell protein extracts for the identification of lactococci to subspecies level. 2111–2140. J. 22:273–276. Bacteriol. Jann. M. and P. and H. Beitr. 66. Williams. De Vos. as Dichelobacter nodosus comb. K. Collins. M. J. J. Lactobacillus paracasei sp. 20:2075–2089.. L. nom. 1970. 58. Pflanz. B. Cavanaugh. and J. J. Bult. M. 62. P. Gillis. J.. and S. Ribosomal ribonucleic acid cistron similarities of phytopathogenic Pseudomonas species. 144:272–282. P. Bottazzi. Baltimore.). J. Syst.). 40. 46. M. Samelis. P. Mannheim. Genotypic relationships and taxonomic localization of unclassified Pseudomonas and Pseudomonas-like strains by deoxyribonucleic acid: ribosomal ribonucleic acid hybridizations. De Vos. 1987. Goor. 74. P. G. B. Chichester. Segers. Identifica´ tion and classification of Campylobacter strains by using nonradioactive DNA probes. comb. J. G. Holt (ed. Balows. 1970. and A. United Kingdom. Transfer of Kingella indologenes (Snell and Lapage 1976) to the genus Suttonella gen. D. M. De Wachter. Bacteriol. 53.. J. Microbiol. Segers. Rev. Owen. 17:459–466. Zambon. Larson. nov. MICROBIOL. Rood. Thacker. 50. Scotland. H. De Vos. J. 1. M. 1989. D. R. Berlin. Cohn. FEMS Microbiol. N. Differentiation of Helicobacter species by numerical analysis of their one-dimensional electrophoretic protein patterns. Crosa. Lugtenberg. F. 45:316–354. Deoxyribonucleic acid homology among Lactobacillus species of the subgenus Streptobacterium Orla-Jensen. FEMS Microbiol. P. 37. Comparative sequence analysis of the 16S rRNA genes of Lactobacillus minutus. Syst. Bottazzi. Syst. Falsen. 1983. Clin. Syst. F. Harder. M. 1975. P. Owen. and Suttonella to Cardiobacteriaceae fam. Gen. Eur. 28:323–327. De Ley. 1993.. G. Kersters. Curtis. Trovatelli. F. Lett. Dewhirst. 65. 31. Biol. B. E. 38. Acetobacteraceae Gillis and De Ley 1980. Dworkin. K. 33:487–509. T. and R. R. J. Unpublished data. In A. M. 52. A. P. 41:287–307. H. Oliver & Boyd.. Carlson. Pot. 43:490–499. Syst. L. 32. F. V. J. 1983. Metaxopoulos. Nucleic Acids Res. A. Funke. Hinkle. D. 56. Bisgaard. 30. Syst.. Pflanz. Ash. 36:405–414. I. Wait. 1987. 35. 1984. K. Chevrier. J. Wood and W. H. Dellaglio. Segers. Fernandez. Deoxyribonucleic acid homology studies of Lactobacillus casei. Electrophor. 39:35–49. Appl. 39. S. Costas.. M. The Williams & Wilkins Co. nov. M. B. 43:329–337. Wallbanks. Costas. A. 59. Carteni-Farina. John Wiley & Sons. 26:187–197. D. Gillis. Modern microbial methods. J. Brenner. Classification of Campylobacter sputorum and allied campylobacters based on numerical analysis of electrophoretic protein patterns. Use of a single-strand specific nuclease for analysis of bacterial and plasmid deoxyribonucleic acid Downloaded from http://mmbr. Gresshoff. Coloe. nov.asm. 1994.. 1993. p. and Lactobacillus rhamnosus sp. Bacteriol. J. FEMS Microbiol. Sci. F. and K. Menaquinone-6 and thermoplasmaquinone-6 in Wolinella succinogenes. K.. Isoprenoid quinones. Bacteriol. Larzul. 1991. 40:333–348. nov. Bacteriol. 72. Slattery. Dewhirst. Intra. Descheemaeker. P. 68. A. Int. and C. 49.. and D. FEMS Microbiol. and F. Dicks. Ross. J. Classification. Y. Moore. L. D. G. 63. P. Mannheim. P. Heft ¨ 2:249–277. Microbiol. Improvements of the membrane filter method for DNA:rRNA hybridization. Pot. R. Int. J. B. J.). FEMS Microbiol. 23. M. Lactobacillus rimae. R. Edinburgh. Carlotti. Camb. 74:289–297. Steigerwalt. 1995. J. Int. 5:351–408. The classification of lactobacilli from the human mouth. 64. R. 67. D. Bergey’s manual of systematic bacteriology. and J. M. rev.. De Vos. rev. M.-K..-H. and K. M. Dichelobacter. Polyphasic taxonomy of the genus Vibrio: numerical taxonomy of Vibrio cholerae.. P. 39:105–108. J. Cacciapuoti. identification. Costas.. E. G.. nov. Int. and J. 32:2569–2571. De Ley. F. B. 43. Bio/Technology 9:553–556. Sutton.. 104:410–433. 1975. Syst. Collins. D. and P. Fernandez. 95:235–240. Aguirre.” J. W.and intergeneric similarities of the Bordetella ribosomal ribonucleic acid cistrons: proposal for a new family. Heft 2 ¨ 1:127–224. L. A. Collins. Springer-Verlag KG. 1968. 1986. 42. M. Bacteriol. Syst. J. 96:225–229. 16:396–404. D. and D. that stimulate plant growth: composition and use for strain identification. Hill. and G. and L. Lipopolysaccharides of Pseudomonas spp. M. Segers. J. Bacteriol. 235–278. Pot. Fields. 75:595–603.. Antonie Leeuwenhoek. Lastovica. 28. 54. Lett. and J. M. emend. Int. A. Blackie Academic & Professional. W. The prokaryotes. Unpublished data. R. 1872. S. E. V. F. De Ley. J. Mutters.-M. On. P. S. Krieg and J. Syst. M. and Streptococcus parvulus: proposal for the creation of a new genus Atopobium. R. transfer of Bacteroides nodosus (Beveridge 1941) to the genus Dichelobacter gen.. Vandamme.. 1985. and F. Bacteriol. 1983. in the gamma division of Proteobacteria on the basis of 16S rRNA sequence comparisons. Brenner. 1994. H. Chemical methods in prokaryotic systematics. A. R. J. 36. DNA amplification fingerprinting using very short arbitrary oligonucleotide primers. Pot.. and S. Biol. Hyg. and M. Bacteriol. La Fontaine. R. 267–268. 1985. and A.. The lactic acid bacteria. Bacteriol. Wallbanks.. Collins. Glasgow. Serol. Distribution of isoprenoid quinone structural types in bacteria and their taxonomic implications. vol. Int. B. Phillips. J. Neefs. Appl. Syst.. Electrophoresis 11:467–474. Ledeboer. Piechulla. 40:426–433.. Casey Chen. homo. and L. Microbiol. G. and J.. Torriani. Gambacorta.. A. 1989. Cookson. 34. R. D. 75. 12:133–142. M. G. 40:126–137. and B. Rossau. Gillis. nov.. Microbiol. J. De Ley. M.. Kersters. L. REV. De Ley. 69. A. Lett. and S. Inter. G. Syst. as Suttonella indologenes comb. Tytgat. Caetano-Anolles. Collins.-L. 25:160–172. Clin. Bacteriol. and P.. Intraspecific variation in small-subunit rRNA sequences in GenBank: why single sequences may not adequately represent prokaryotic taxa. Deoxyribonucleic acid homology and base composition in some thermophilic lactobacilli. 2nd ed. 1973. J. Appl. Alcaligenaceae. J. E. Paster. 1992. Paster. Microbiol. 1995. J. 2. and related Vibrio species. J.. paracasei and subsp. Dellaglio. J. Kersters. Reynaerts. Farrow. Food Microbiol. P. Untersuchungen uber Bakterien. Clayton. Bacteriol. M. Bauwens. Rodrigues. Microbiol. 41. The cellular fatty acid composition of Campylobacter species isolated from cases of enteritis in man and animals. 29. 1993. J. Bacteremia caused by a novel Bordetella species. B. K.. P. Syst. In B. 2. Sheffield. and typing of bacteria by the analysis of their one-dimensional polyacrylamide gel electrophoretic protein patterns. vol. G. Vanhoucke. Microbiol.. L. Jones. W. Wallbanks. Bacteriol. Martinez-Murcia. 1990. W. 35:169–184. D. Cattoir. J.and intergeneric similarities of Pseudomonas and Xanthomonas ribosomal ribonucleic acid cistrons. Falkow. De Smedt. p. Cellular fatty acid profiles of campylobacters. and F. Bacteriol. The genera of lactic acid bacteria. De Ley. van Landschoot. J. C. J. Grant. K. C. De Rosa. Verrips. Kersters. nov.-H. and J. Syst. J. Int. Vibrio parahaemolyticus. 169:1441–1446. J. Kersters.. Gillis. and J. 1985. Med. E. Bassam. subsp. Phillips. W. 1973. O’Donnell (ed. J. 33. B. Johnson. Appl. Identification of Enterococcus species isolated from food products of animal origin. P. Kersters. Syst. M. P. J. Collins. Lett. T. 1992. M. Int. Syst. R. 33:426–428. Intra.and heteroduplexes. B. 1992. K. L.. 17:380–386. M. Schleifer (ed. Bacteriol. M. Biochem. Gen. 1994. 13:481–493. Benson. W. Microbiol. De Vos. Cato. 27:321–326. Vaughan. R.. F. De Ley. T. and J.432 VANDAMME ET AL. 1986. Inter-laboratory comparative study of the numerical analysis of onedimensional sodium dodecyl sulphate-polyacrylamide gel electrophoretic protein patterns of Campylobacter strains. Appl. Cohn. Untersuchungen uber Bakterien. M. D. Lab. The quantitative measurement of DNA hybridization from renaturation rates. H. S. M. T. Davis. 1989. Owen. J. C. Megraud. P. Collins. Lievens. B. De Weger. Collins. U. Rapid distinction of Brevibacterium species by restriction analysis of rDNA generated by polymerase chain reaction. Polyamines in halophilic archaebacteria. and R. Taxonomic studies on some leuconostoc-like organisms from fermented sausages: description of a new genus Weissella for the Leuconostoc paramesenteroides group of species. 73. 45:595–599. 1981. D. nov. K. Cooccurrence of menaquinone-6 and thermoplasmaquinone-6 in Bacteroides gracilis. Spach. Int. Phylogeny of species in the family Neisseriaceae isolated from human dental plaque and description of Kingella orale sp. The Proteobacteria: ribosomal RNA cistron similarities and bacterial taxonomy.). A. De Ley. Jann. 1993. J.. 60. 70. Van de Peer. M. nom. 115:904–911. In N. Bacteriol. A. Costas. nov.. B. L. M. 71. Frederiksen. and assignment of the genera Cardiobacterium. J. B. C.. Porcelli. Int. 47.. De Ley. J. Lopez-Urquijo.. Jackman.. M. R.. Int. Synonymy of strains of “Lactobacillus acidophilus” group A2 (Johnson et al. De Rijk. Van Damme.

nov. A. Fox. and C. Goodwin.-P.) 93. Vlaes. Y. Rosel. H. Rosel. G. J. Krakowka. D. 1990. S. Inst. H. Syst. Harper. Dis. E. nov. Bacteriol. Hebbar. 1990. Farrow. Gilliland. Wait. E.). and R. P. The prokaryotes. R. Clin. Curr. R. and A. Fox. W. D. Hill. Fluorometric deoxyribonucleic acid-deoxyribonucleic acid hybridization in microdilution wells as an alternative to membrane filter hybridization in which radioisotopes are used to determine genetic relatedness among bacterial strains. nov. 121. The Williams & Wilkins Co.. W. J. Sakane. Int. H. 62:223–239. J. Acidophilus milk products: a review of potential benefits to consumers. Eyers. M. Chemical methods in prokaryotic systematics. J. Gillis. P. 2. Clin. F. P. B. and DNA-DNA hybridizations. 146:467– 475. and W. Lemelin. J. nov. Transfer of Campylobacter pylori and Campylobacter mustelae to Helicobacter gen. Yamamoto. M.. G. Kersters. M. Goto. A. Van den Borre. and J. 1993. T. England. Hamana. How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity. N. Wilson. Yaeshima. 1990. Fox. Characterization of Campylobacter cinaedi and C. J. Bacteriol. and M. V. L. 97. comb. 121– 163. 122. Pot. J. Eaton.. Fox. G. Kersters. P. E. Mitsuoka. G.-P. Goodfellow and A. The genus Lactobacillus. J. Goodfellow. Gasser.VOL. Int. ¨ Dworkin. Pasteur/Microbiol. B. Can. J.asm. Syst. Bacteriol. Clin. 110:258–267. 1994. 159:635–640. S. In B. H. McConnell. M. Dis. McCullough. P. 44:553–560. V. 1974.. J.. D. J. J. Polyamine analysis of the genera Aquaspirillum. Bacteriol. Mayall. J. 31:1430–1434.. Appl. Phylogeny of Helicobacter isolates from bird and swine feces and description of Helicobacter pametensis sp. Int. Microbiol. Campylobacter showae sp. Improved preparation of lipoteichoic acids. Magnetospirillum. Marbehant. Harder.. K. Gutell. A. in press. A. 45:274–289. D. C. vol. P. 37:437–440. Morgan. Etoh. F. C. Hammes. R.. 43:631–639. 38:116–118. Microbiol. R. Singh. Int. 115. Dicks. II. The lactic acid bacteria. Structural lipids of eubacteria. F. Generation of DNA probes for detection of microorganisms by polymerase chain reaction fingerprinting. W. C. McLaughlin. Gurtler. Belcher. J. T. 89. 106. B. 1987. (Erratum. Oceanospirillum and Spirillum. C. Van den Abbeele. J. Yan.). 42:166–170. 1993. and H... J. L. p. Glasgow. 1995. as revealed by numerical analysis of total soluble cell protein patterns. Popoff. Vandamme. and R. van der Plas. 40:83–91. J. T. W. T. isolated from cheetahs with gastritis. 1535–1594. W. B. Eur. Clin. L. J. Ward. and R. W. Z. 8th ed. 1994. Dicks. Tully. Wisotzkey. and P. C. Scotland. and C. J. D. Five types of polyamine distribution patterns in thiobacilli. N.. Bacteriol. Holzapfel (ed. nov. Unpublished results. 114. 110. 82. H. 1995. and G. J. Grimont. particularly Leuconostoc oenos. Gasser. 34:541–546.). 1980. 27:44–57. H. 2012 by guest . Hansen. 88. sp. Butzler. Haas. R. T. Hofstra. Grimont. Lessons from evolving rRNA: 16S and 23S rRNA structures from a comparative perspective. G. Bronsdon. Embley.. Y. J. L. Paster. C. C. G. Helicobacter hepaticus POLYPHASIC TAXONOMY 433 104. 100. fennelliae antigens and analysis of the human immune response. Bacteriol. nov. and E. and S. L. R. The genera Lactobacillus and Carnobacterium. E. Bacteriol. and H. 95. Clin. G.. Doudoroff. 1986. Seymour. 1994. E. Ann. and A... 58:10–26. F. 124. Methods Microbiol. and D. Levy. G.. S. (Paris) 137B:165–175. P. H. 1992. 113. for N2-fixing isolates from rice in Vietnam. with recognition of Lactobacillus gallinarum sp. 1995. Taxonomic study of the Lactobacillus acidophilus group. Paster. J. Microbiol. Microbiol. and M. M. Microbiol. S. Koch. An antigenic analysis of Lactobacillus acidophilus. O’Donnell. Fraser. G. Goodwin. John Wiley & Sons. G. L. J. Fennell. nov. nov. 174:2002–2013. and F. J. 1994.. F. T. Bardin. Bacteriol. 32:1238–1245. 98. J. A. Clin. E. M.. van Vuuren. De Wachter. Pigott. M. C. and K. 1984. nov. and synonymy of Lactobacillus acidophilus group A3 (Johnson et al. M. E. 33:445–454. Koeken. F. Y. Van den Borre. Fiedler. J.. 112. p. Fox. Typing of group B streptococci: comparison of pulsed-field gel electrophoresis and conventional electrophoresis. and its temperature and growth range. P. Grimont.. J. J. respectively. H. Shames. 172:1298–1305. 1980. 92. and F. Dellaglio. Microbiol.. Chilvers. R. S. Rev.. p. Bacteriol. 1970. P. Murray. Phylogeny of 54 representative strains of species in the family Pasteurellaceae as determined by comparison of 16S rRNA sequences. Megraud.. 101. Blincow. livers. E. 125. Weiss. Syst. K. 19:406–458. J. Larsen. 1993. Ezaki. Hayward. A hand- Downloaded from http://mmbr. M. and K. 1980) with the type strain of Lactobacillus amylovorus (Nakamura 1981). Dellaglio. E. Blackie Academic & Professional. Modern microbial methods. Differentiation of Leuconostoc species by nicotinamide adenine dinucleotide-dependent D( )-lactic dehydrogenase profiles. a novel Helicobacter species isolated from bile. nov.. Gorelick. H. Chapelle. Dewhirst. 40:347–352. Alanine ester ¨ containing native lipoteichoic acids do not act as lipoteichoic acid carriers.. J. 1994. E. and A. K. Ferguson. J. Hamana. C. nov.. 91. F. 1991. Microbiol. In R. The absence of thermoplasmaquinones in Campylobacter pyloridis. W. M. D. G.. and N. Goodwin. The application of 16S rRNA cataloguing and 5S rRNA sequencing in bacterial systematics. and F. DNA base compositions. J. 1989. 13:317–322. J. Fox. Immunological relationships among lactic dehydrogenases in the genera Lactobacillus and Leuconostoc. Elharrif. Gasser. and M. Curr. R. Lactobacillus oris sp. Classification of medically important clostridia using restriction endonuclease site differences of PCRamplified 16S rDNA. Cogniau. 1988. Elharrif. Collins. G. Butzler. I. Koster. Collins. and J. Fox. K. Curr. F. Syst.. Matsuzaki. 117. Fraser. Dicks. Clin. M. R. J. 1994. Dicks... Differentiation of Lactococcus lactis and Lactococcus garvieae from humans by comparison of whole-cell protein patterns. Syst. J. A. 106:113–125. L. L. C. 1989. Int. 1986. Paster. In M. 176:3081–3084. Fujisawa. Infect. 105. J. 45:395–397. N. Reproducibility and correlation study of three deoxyribonucleic acid hybridization procedures. Flores. Biol. Coynault. M. Verhoef. 84. O’Donnell (ed. Woese. J. Syst.. 96. 39:397–405. J. 116.. 123. Van Camp. Characterization and description of “Campylobacter upsaliensis” isolated from human feces. 28:1039–1046. Handbook of new bacterial systematics. Van Naelten. 118. nov. P. U. 1991.. 1981. Armstrong. Giesendorf. T. J. 108. Helicobacter acinonyx sp. J. A. 67:9–14. Collins. Zentralbl. J. Du. 83.. 32:137–140. H. Collins. 127. Microbiol. J. 81. and P. isolated from the human oral cavity. Grimont. J. Hashimoto. J. 85. Fernandez. 99. Characterization of thermophilic ´ Campylobacter. J. 1989. J. and F. 1992. P. J. Vandamme. McCullough. J. N. 1989. Bacteriol. D. H. 133:523–530. 60. 72:2483. and W. 1977. Microbiol. and R. 109. Bacteriol. Syst. K. I. 1983. Discrimination among thermophilic Campylobacter species by polymerase chain reaction amplification of 23S rRNA gene fragments. Cellular fatty acid composition of Campylobacter pylori from primates and ferrets compared with those of other campylobacters. Collins. 111. D. 119.. Gen. Int. and G. K.. from the human oral cavity. J. 1988. 29:2731–2734. Schleifer (ed. In A. Int. 27:938–943. 43:99–106. S.. Koch.org/ on January 3. Facklam. E. W. Differentiation of Campylobacter species by protein-banding patterns in polyacrylamide slab gels. J. V. V. Bacteriol. 1989. 1993. J. 120. a microaerophilic bacterium isolated from livers and intestinal mucosal scrapings from mice. Elliot. 102. Megraud. J. 217–243. Int. Grimont. Balows. FEMS Microbiol. and Helicobacter mustelae comb. p. Collins. D. 16S ribosomal DNA sequence identities of beta-proteobacterial endosymbionts in three Crithidia species. Pechman. A. 1994. Syst. Woese. U. 1992. M. Bacteriol. and E. E. Hammes. A. and E. Chem. 40:75–82. K. E. E.-P. B. Koch. Fischer. Stackebrandt. Int.. Quint. Goossens. Baker. and intestines of aged. P. Goossens. Bacteriol. Z. J. Syst. Yokota. The genera of lactic acid bacteria. 79. L. Relatedness of homofermentative Lactobacillus species revealed by numerical analysis of total soluble cell protein patterns. and B. Microbiol. P. Yabuuchi. U. nov. Lett. Holzapfel. A. Infect. C. Enzymatic profiles. Benno. Fischer. D. van Belkum. Y. 1986. J. 78. B. Wood and W. and P. Proposal to reclassify Leuconostoc oenos as Oenococcus oeni [corrig. M. 42:487–491. V. G. Biochem. E. Helicobacter bilis. A. Willems. Dairy Sci. Dellaglio. Murphy. and J.. M. M. R. W. 4:325–330. Taylor. Yan. Microbiol. 94. 137:2673–2679. Dewhirst. J. M. M. P. G. Stamm.. 1987. C. Palleroni. J. B. Syst. inbred mice. T. Paster. van Vuuren. Fischer. S. Microbiol. 1995. Use of DNA reassociation in bacterial classification. and R. Int.. T.. and M. Paster. Kasper. Clin. V. J. 87. Olsen. 19–54. 255:4557–4562. M.. Dewhirst. Radin. A. Int. 1996 77. Taxonomy of Leuconostoc species. J. 20:453–460.. Dis.). E. Giesendorf. Microbiol. R. Genus Pseudomonas. Gordillo. Vlaes. and N.. Bakteriol. 1995. J. G. 126. Syst. Goor. Vandamme. 1962. and E. Heulin. Buchanan and N. J. 1993. and Lactobacillus johnsonii sp. B. Dewhirst. J. A. S. van Vuuren. W. 107. Van. 13: 117–122. M.. L. M. and C. T. Baltimore. L. Goossens. Academic Press Ltd. J. Dewhirst. B. Truper. Quint. W. G. J. Hanicq. Investigation of an outbreak of Campylobacter upsaliensis in day care centers in Brussels: analysis of relationships among isolates by phenotypic and genotypic typing methods. Paster. Dewhirst. Gillis. F. FEMS Microbiol. F. and T. 86. C. as Helicobacter pylori comb. 1991. A. J. Bacteriol. and W. Microbios Lett. Characterization of thermophilic ´ Campylobacter. J. London. and F. B. Microbiol... Lett. Int. 1971. M. Effect of alanine ester substi¨ tution and other structural features of lipoteichoic acids on their inhibitory activity against autolysins of Staphylococcus aureus. Microbiol. Bacteriol. and D. L. B. C. 1986. and P. Microbiol. 80. C. E. Y. Jurtshuk.. Chichester. Syst. Gen. Sly. M. Carbon-substrate utilization tests.. 103. P. J. Vogel. A. 31:3340–3343. Bergey’s manual of determinative bacteriology. Polyphasic taxonomy in the genus Burkholderia leading to an emended description of the genus and proposition of Burkholderia vietnamiensis sp. H. Int. 1990. 32:1623. T. Bacteriol. S. J.. A.] gen. F... F. Ribosomal ribonucleic acid gene restriction patterns as possible taxonomic tools. Segers. T. Gibbons (ed.. J. Chang. E. Efthymiou. Electrophoretic characterization of lactic dehydrogenases in the genus Lactobacillus. 39:224–229. Peters. Gen. N. J. B. Lambe. J. 90. K. Mendes.. Blomme. W. Infect. McConnell. and B. Syst. Comparative cataloging of 16S ribosomal ribonucleic acid: molecular approach to prokaryotic systematics.

1982. Can. Clin.. R. Jackman. Transfer RNAs as genotypic fingerprints of eubacteria. and N.. Bacteriol. P.). Gen.. Miotto. 1990. M. Moore. J. Bergey’s manual of systematic bacteriology. London. B. J. B. Bacteriol. a new species of the subgenus Thermobacterium. 20:138–140.. M. In E. Pace. Differentiation of Streptococcus uberis from Streptococcus parauberis by polymerase chain reaction and restriction fragment length polymorphism analysis of 16S ribosomal DNA. USA 82:6955–6959. B. and P. 152. M. 159. Springer-Verlag. Mayr. J. D. Fischer. J. vol. W. 20:325–327. Microbiol. L. DNA relatedness ´ among strains of Campylobacter jejuni and Campylobacter coli with divergent serogroup and hippurate reactions. J. D. Stackebrandt and M. A. Janssen. and C. Barrett. 115–128. D. H. D. Microbiol. 21:69–71. Dunn. 30:2084–2087. Jr. Wolinella recta. Parasitenkd. 1957. 1970. 165. Occurrence of sheathed flagella in Campylobacter cinaedi and Campylobacter fennelliae. nom. Infektionskr. R. Int. Microbiol. M. 173. Downloaded from http://mmbr. rev. Hyg. 133. M. A. Goodfellow (ed. C. Microbiol. and M. E. 167. Washington. R. R. J. Han. Fuzzy thinking. D. R. Syst. Neill. Sci. University Park. M. 153:299–304. P. A. Natl. and ´ C. 129.asm.org/ on January 3. Bakteriol. G. Lawson. Kandler. Wolff..C. Some features of Campylobacter sputorum subsp. In E. 1964. Syst. Cellular fatty acid composition as a chemotaxonomic marker for the differentiation of phenospecies and hybridization groups in the genus Aeromonas. J. Differentiation between Campylobacter 128. S.. New York. 2. Krieg. Kersters. 1981. Hauben. and R. p. R. C. 1981. A. p. and O.. Krieg. Evaluation of the DNA fingerprinting method AFLPTM as a new tool in bacterial taxonomy.). 1973.. H. Bacterial taxonomy based on electrophoretic whole-cell protein patterns. A. 181. Vancanneyt. Kilpper-Balz. 139. 172. ¨ Arch. and E.. 1994. Stahl. 212AL. and D. Identification and typing of bacteria by protein electrophoresis. P. Tyndall (ed. J.. M. Serotyping of bacteria. Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Kostman. The genus Spirillum: a taxonomic study. 14:463–467. P. Pot. 37:215–257. 1977. Hayward. U. L. Kumar. P. Henriksen. A. E. and J. Krieg. S. Hensel. and J. Proc. Hayward. Bacteriol. and O. R. B. Clin. 7:245– 250. 170. Falsen. J. The Williams & Wilkins Co. Wat. J. R. Amarger. Clin. Bacteriol. D. Appl. P. 168. Ellis.. J. Blaser. 29:2774– 2778.). 147. 101:43–46. Acad. B. 6:31–35. Differentiation of lactobacilli occurring in fermented milk products by using oligonucleotide probes and electrophoretic protein profiles. vol. vol. E. Identification of mesophilic lactic acid bacteria by using polymerase chain reaction-amplified variable regions of 16S rRNA and specific DNA probes. 143. Krieg. Chemical methods in bacterial systematics. Purification and kinetics of the allosteric L-lactic acid dehydrogenase from Lactobacillus casei ssp. Clin. Swings and E. B. Baltimore. 1991. R. Clin. J. helical/vibrioid gram negative bacteria. Serology and chemotaxonomy. A comparison of Lactobacillus species from human saliva with those from other natural sources. Vauterin. 1994. Kandler. 1984. Bacteriol. O. London. London. J. and A. 1991. Lipuma. Deoxyribonucleotide sequence relationships among Bordetella species. Mocquot. J. Leaper. and A.). Microbiol. C. 22:360–367.. Hofle.. and Bacteroides gracilis are microaerophils. 1983. R. P. Pace. Vauterin. Cummins. Br. p. T. G. Bacteriol. Curr. and D. Hodge. G. 15: 1065–1073. Xanthomonas. Immunological properties of teichoic acids. Laguerre. B. Appl. W. Han. Priest. Int. L. K. Holt. Matthews. Kandler. C. 148. Int. J. D. Jahnke. New York. E. C. 23:340–380. Smibert. Identification of catalase-producing Campylobacter species based on biochemical characteristics and cellular fatty acid composition. and J. 164. N. Bacteriol. J. Krieg and J. Baltimore. Ltd. Janssen. Syst.. and N. K. and K. and N. L. O. Y.. B. 21–44.. Hayward. M. Microbiol. Phelps. and their taxonomic significance. Tamura. 1971. K. and W. Rapid identification of rhizobia by restriction fragment length polymorphism analysis of PCR-amplified 16S rRNA genes. 33:275–284.. D. Syst. M. and H. Microbiol.. 102:450–451. and N. E.. 116–126. Pettigrew. 1980. Syst. and M. 1978. Gasser. N. Ludwig. W.434 VANDAMME ET AL. an organism forming CO2 and H2 from lactic acid. L. Kersters. M. Y. 178.. Microbiol. 1973. Wells. R. Mohapatra. S. 2012 by guest 142. H. Syst. Hertel. M. J.. K. U. 156. H. E. Parasitenkd. Mair. Hawley. J. Ezzell. In M. nov. 1994. and R. Rev. E. 176. The Williams & Wilkins Co. Lane. Leaper.. W. 1994. Environ. Schleifer. J. ´ Oliver. In P. Patton.. I.-D. Evaluation of the indoxyl acetate hydrolysis test for the differentiation of campylobacters. 180. Comparison of aerotolerant and reference strains of Campylobacter species by polyacrylamide gel electrophoresis. De Ley. Arch. mucosalis subsp. Altwegg. The Williams & Wilkins Co. Inst. P. K. S. N. REV... 177. L. A. DNA reassociation experiments. Differentiation of Campylobacter and Campylobacter-like organisms by cellular fatty acid composition. R. rev. and K. Bacteriol. Hansen. 1209–1234. in press. 1982. D. 1993. P. F.. W. . J. p. J. E.. 85:1849–1852. MICROBIOL. L. Baltimore. W. J. M. C. N.. M. Rowland. Heterogeneity of the species Lactobacillus acidophilus (Moro) Hansen and Moquot as revealed by biochemical characteristics and DNA/DNA hybridization. 140. L. 151. 175.-H. M. 1993. Kandler. F. Stetter. 146. 38:104–110. J. W. H.). 1987. Hanna. Microbiol. Jannasch. 1985. The Pennsylvania State University. Microbiol. Moss. J. 1991.. The isolation and classification of Lactobacillus strains from Italian saliva samples. and W. J. Witholt (ed. 144. Molkereiztg.. Goodfellow and D. Krieg. R. Clin. Wolinella recta. Jarvis. In N. and K. G.. 145. Bacteriol. Microbiol. Sogin.). nov. 150. D.-R. Johnson. Identification of bacteria by low molecular weight RNA ¨ profiles: a new chemotaxonomic approach. G. J. K. 182. Bacteroides ureolyticus. J.. and B. Unpublished data. G. Baumbach. Orig. Syst. Dobrogosz. Coopman. Debrunner-Vossbrinck. P. Harper Collins Publishers. 158. Nucleic acid techniques in bacterial systematics. G. Int. Syst.. C. Huys. Kloos. Lactobacillus bifermentans sp. P. B. Olsen. T. 174. Infektionskr. Int. 162. 1. In N. 33:143–146. 30 years of campylobacters: biochemical characteristics and a biotyping proposal for Campylobacter jejuni. and G. Vandamme. Syst. J. Bacteriol. J. T. G.. Owen. L. Edlind. 30:53–68. J. Rollins. L. Bergey’s manual of systematic bacteriology. O. K. 12:1–13. 134. 137.-H. and O. Syst. 1–95. P. J.. K. 31:385– 391. Plenum Press. J. 154. M. Factors affecting the implantation of lactobacilli in the intestine.-H.). 155. R. Krieg. R. 1985.. Holt (ed. Pasteur Microbiol. Jayaro. Revoy. 1991. Academic Press Ltd. S. M. Hansen. 1983. R. Moss. Johnson. S. 1994. 136. 1988. isolation. Microbiol. Torck. 1. R. J. H. 1983. K. M. casei and Lactobacillus curvatus. Smibert. S. p. G. 132.. In F. nov. Reihe C 1:150–168. Ramos-Cormenzana. Lambert. Grouping of lactic streptococci by gel electrophoresis of soluble cell extracts. Holt (ed. G. Appl. Int. Vancanneyt. Comparative studies of lactic acid dehydrogenases in lactic acid bacteria. Weerkamp. J. 130. Hespell. 41:218–222. L. Hofle.. The hosts of Xanthomonas. Microbiol. Vos. N. M. 135. G. 1989. United Kingdom. Dent. Microbiol. I Abt. Hebert. Microbiol. R.-H. nov. Int. motile. Relationships among catalasepositive campylobacters determined by deoxyribonucleic acid-deoxyribonucleic acid hybridization. A. Zentralbl. Br. Donell. Baltimore. and applications. Sharpe. and N. Zentralbl. Wolinella curva. and N. nov. Dore. G. G. Greenwood. Appl. H. 138. M. A. Swings. 169. 1984. B. Civerolo (ed.. A. Bacteriol. Bakteriol. R. Minnikin (ed. Nucleic acid hy¨ bridization of group N and group D streptococci. L... Starr.0. De Vos. M. Lactobacillus casei (Orla-Jensen) comb. W. D. Aerobic/microaerophilic. Kersters. nom. 179. and J. Bacteriol. Coopman. R. Int.). G. B. Bacteroides ureolyticus. 131.. 161. K. 44:651–658. M. 141. Brenner. 171. Huys. Microbiol. 28:1482–1483. Knox. Edmonds. A. W. 1993. book on the biology of bacteria: ecophysiology. Formation of halogenated arylpolyene (xanthomonadin) pigments by the type and other yellow-pigmented strains of Xanthomonas maltophilia. S. Krieg. Galli. Schleifer. Zabeau. Lactobacillus acidophilus (Moro) comb. M. N. C. A. C.. 149. R. 1992. G. Appl. 1980. and C. Ltd. Dewettinck. version 1. J. Syst. Jenkins. S. Kersters. 136B:257–264. Sogin. A. R. Bergey’s manual of systematic bacteriology. 1982.. Manclark. J.. 112:81–93. 37:391– 398. A.-L. Weiss. not anaerobes. and G. Davis. and J. I Abt. J. J. J. J... 51–66. M. A. The Williams & Wilkins Co. Methods Microbiol. 57:3390–3393. 9:231–238. London. Appl. identification. L. 15–18. Stull. Hofle. Environ.. Pseudomonas molecular biology and biotechnology. Int. 39:488–490. Colwell. 31:173–176. Klijn. Rapid genotyping of pseudomonads by using low-mo¨ lecular-weight RNA profiles. 1. Genus Lactobacillus Beijerinck 1901. G. Lambert. Syst. 1975. Hylemon. Phylogenetic diversity and position of the genus Campylobacter. 1981. Lactobacillus gasseri sp. 1990. Microbiol. Jones. E. Holt (ed. A modified method of quantitative colorometric DNADNA hybridization on membrane filters for bacterial identification. In J. L. Hyg. and L. J. Swings.. J. 157. J. Kersters. Verwendung von Lactobacillus acidophilus in Milchprodukten. S. Kosko. G. Dtsch. R. A. Chichester. and N. G. Bacteriol. 1986. and S. Sneath. Identification and grouping of bacteria by numerical analysis of their electrophoretic protein patterns. R. 60:56–63. 25:706–713. 1980. Microbiol. 1988. 153. 163. Schillinger.. Leaver. 1991. J. Pot. A. R. Bleeker. Ann. Nei. and E. Int. Microbiol. C. p. Int. Smibert. 160. 1956. John Wiley & Sons. Shepherd. Chapman & Hall.. and B. Methods 8:235–248. 1984. G. Lessel. Pillidge. and G.. Hebert. 4:408–412. Orig. Wicken. M. Taxonomy of the Lactobacillus acidophilus group. Pa. Harvey. Hollis.. Syst. and K. Weiss. O’Brien. Weaver. and T. Lauer. A. 166. Microbiology. p. U. O. Borczyk. M. 1992. p. Reihe C 1:75–78. F. Lauer. Owen. P.. Syst. Y. Methods 20:273–288. 87:333–342. and N. 71–124. W. Fox. Molecular epidemiology of Pseudomonas cepacia determined by polymerase chain reaction ribotyping. Microbiol. American Society for Microbiology. Han. Helming. Kandler. C.. Lau. Environ. Dent. J. Weather. 1979. and F. J. 1985. Allard. G. Silver. S. 1959. Krieg and J. and Bacteroides gracilis. M. Bergey’s manual of systematic bacteriology. Bacterial diversity and systematics. Curr. Cytochrome composition and oxygen-dependent respiration-driven proton translocation in Wolinella curva. MEGA: molecular evolutionary genetic analysis. Kandler. vol. R. Appl. 1984. J.

1983. Fisher.. and O. K. Smibert. Moro. Romaniuk. Lactobacillus and Bifidobacterium. W. and J. Mol. J. J. R. 203. Serol. H. Stackebrandt. Helicobacter muridarum sp. I. N. Moss.. Bacteriol. Garel. Microbiol. Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. London. M. A. S. Baumberg. O’Donnell (ed. E. p. Numerical Downloaded from http://mmbr. Fonseca. 17:58–64. Blackwell Scientific Publications. Schleifer. Academic Press Ltd. 40:213–215. Curry. M. Dis. E. Neumann. Rocha. Olsen. Dewhirst. C. 79:89–94. John Wiley & Sons Ltd. Maynard Smith. Mol. 64:285–305. E. F. 186. 37:453–478. Ford. C. and R. Chase. Unpublished data. 234. Aldolases of the lactic acid bacteria. Meyer. London. Normand.. K. Larsen. 184. M. Parasitenkd. J. Microbiol. R. J. Schønberske Forlag. J. Chichester. G. Properties of D-lactate dehydrogenase from Lactobacillus bulgaricus: a possible different evolutionary origin for the D. A. Orla-Jensen. FEMS Microbiol. Bacteriol. Bacteriol. Syst. 31:56– 63. E. Outline of clinical methods in anaerobic bacteriology. 196. M. 1916. Morris. 1993. and S.). Macy. and J. J. J. M. Seal. 216. M. Kai. Lenski. E. Swings. Johnson. 19:772–776. FEMS Microbiol. G. H. R. Assessment of enzyme detection tests useful in identification of campylobacteria. and F. J. Bacterial systematics. Fernandez. E. S. Bacterium bifidum und Thermobacterium intestinale. Propionibacterium. 33:723–737. a microaerophilic helical bacterium with a novel ultrastructure isolated from the intestinal mucosa of rodents. and G. Microbiol. Marmur. Clin. Molecular epidemiology: application of contemporary techniques to the typing of microorganisms. Clin. Morgan. II Abt.. Holdeman. 231. Philips. Jahrb. 211. and J. Magee. B. F. F. Comparative ribosomal protein sequence analyses of a phylogenetically defined genus. Dewhirst.. and M. Kulczyk. Welander. Pseudomonas. Va. 2012 by guest . Lett. 208. 230. and J. 55:253–267. D. R. T.. Gherna. 229. L. Pearson. cinaedi.. A. A.):2017–2021. N. und Huhnern. Auling.. Young. 1995. L. E. Hyg. M. Infect. DNA sequencing in bacterial systematics. Demonstration of immunological relationship among eight genera of Gram-positive bacteria using anti-pediococcal aldolase serum.. C. protein content. Maelkeri-bakteriologi. M. Kinderheilk. sp. Charac´ terization of “Campylobacter pyloridis” by culture. H.. Moura. Hyg. 222. Garnier. 189. L. D. M. Syst. and H. Orig. and “C. Moore. L. 221. and R. E. Stackebrandt and M. and J. 193. Int. Whole-organism fingerprinting. Queiroz. Syst. M. R. 12:198–207. Clin. P. A. 191. and K. G. Vergleichende Untersuchungen uber die Laktobazillen ¨ aus den Faeces von Menschen. E. L. Mitsuoka. G. C. R. R. M. Jpn. 1992. 28:395–397. E. Nucleic Acids Res. P. C. J. Appl. E. Petersen.. S. Microbiol. 1991. ed. 1976.. 1990. Watanabe. Clin. Lactobacillus amylovorus. 1994. 27:2652–2655. Uff. U. S. and C. J. M. P. 1992. 228. De Cleene. Microbiol. 201. V. 1989. Pfennig. Wilkinson. E. O’Rourke. Fox. and D. J. J. G. L. D. L. M. Hoshina. and N. J. 1995. 2nd rev. Nour.. W. 15:203–208. Watase. Dorsch. V. and S. Lambert-Fair. Lett. London. Cambridge University Press.. Syst. J. 1984. 1994. sp. 213. P. M. 1988. 108:196–201. Ternstrom. Syst. Bolton. Patton. 1995.VOL. D. R. Woese. A. P. W.. Rocha. Classification and identification of campylobacters. Microbiol. 188. 194. J. Reetz. Brenner. Gen. Moss. 174:4525–4529. England. Costas. Chase. O. W. H. nov. S. London. 1970. Orla-Jensen. Neu. and J. Phylogenetic relationships of bacteria based on comparative sequence analysis of elongation factor Tu and ATPsynthase beta subunit genes. 206. S. and G. and B.. Ludwig. Weinstock. Maslow. McNulty. J. J. Int. Evol.. Actinomyces. Bacteriol. N. R. and G. London. Arch. Syst. 193–215.). Wellington. 1990. 50:308–356. nov. M. 1993. S.. Kroupach. G. J. L. M. C. J. J. Biol. Bachleitner. J. 44:511–522. E. R. sp. 33:61–66. M.. Microbiol. I Abt. C. 1992. Moore. Vet. Cummins. and I. Bonnet. Int. G. P. 69–94. and N. Aldolase of lactic acid bacteria: a case history in the use of an enzyme as an evolutionary marker. Paster. Rev. 192. Antonie Leeuwenhoek J. L. Leyns. J. C. Park. Bacteriol. Microbiol. M. Bacteriol. D. J. Unpublished data. Y. Sloss... J. N. S. Dynamics of IS-related genetic rearrangements in resting Escherichia coli K-12. Lee. 1988. 1973. 215. and its relatives. A. 42:585–591. Dent. Morgan. M. Paster. Goodfellow (ed. I. Zentralbl. G. E. Murray. P. A. J. 212. 1991.” J. and S. M. 110:121–128. Blot. I.. Megraud. Fraser. In S. J. Nicholson. 1984. M. Owen. 1989. Relationship among lactic acid bacteria demonstrated with glyceraldehyde-3-phosphate dehydrogenase as an evolutionary probe. 190. T. 50:692–698. 210:32–51. 1981. 1991. Swings. M. 1995. London. Uber den Bacillus acidophilus n. 1990. Bot. M. W. W. R. R. J.. Sly. 19(Suppl.. Rech.. Owen. A. M. Bakteriol. fennelliae. A. Brockmann. Bacteriol. Syst. Eur. M. 44:174–176. Microbiol. S. 1991. Electrophoretic protein patterns in Campylobacter species with special reference to Campylobacter mucosalis and Campylobacter hyointestinalis. Kline. N. A. Syst. and L. Costas. Clin. Holmes. J. Int. In E. Arber. Murray. Detection of relationships between Streptococcus faecalis and Lactobacillus casei by immunological studies with two forms of malic enzymes. Kruger. 204. 106:126–137. 1994. W. N.). Infektionskr. 1994. J.. 22:1007–1010. A. J. Pathol. J. B. 205. E. Logerfo. 207. 1971. E. L. Antonie Leeuwenhoek J. J. Med. Bak¨ teriol. Nakamura.. Doty. Syst. Welsh. 4:277–283. Trust. Cambridge. D. and I. Syst. Ochi. Gastroenterol. 1975. 15–22. Moore. J. Isoprenoid quinone content and cellular fatty acid composition of Campylobacter species.). Biol. M. 17:153–164. Kulczyk. A. 89:1296... A. and D. and W. J. Mulligan. Bacteriol. 93:321–343. 78:165–178. 227. F. p. Goodfellow.. nov. R. nov. S. 226. L. Molin. A. A. 45:268–273. Saunders (ed. 5:109–118.). E. C. Blacksburg. Ludvigsen.. Int. S. Epidemiol.-M. Y. 82:55–60. In M. Clin. Taxonomic notes: a proposal for recording the properties of putative taxa of procaryotes. and POLYPHASIC TAXONOMY 435 183. Winther. J. 1987. 1988. J. J. and B. Microbiol. Length polymorphism in tRNA intergeneric spacers detected by using the polymerase chain reaction can distinguish streptococcal strains and species. 225. K. E. M. Bacteriol. Colwell. Mutai.. London.. L. Hydrolysis of indoxyl acetate by Campylobacter species. 209. W. Grimont. Zentralbl.. 185. Kline. New spiral bacterium in the gastric antrum.-R. W. and J.. B. P. J. Zavarzin. E. 1962. Lett. enzymatic profile. 224. McClelland. D. 1990. H. B. J. Cambridge. C. On. T. G. Thauera selenatis gen. 232. 220. J. Fitch. p. D. Cato. Report of the ad hoc committee on approaches to taxonomy within the Proteobacteria. 1993. C. a member of the beta subclass of Proteobacteria with a novel type of anaerobic respiration.. S. Le Bras. Queiroz. 1993. P. Cambridge University Press. J. J. hyointestinalis. J. Are pigs a reservoir host for human Helicobacter infection? J. Ultrastructure of a spiral micro-organism from pig gastric mucosa (“Gastrospirillum suis”). Mendes. R. Bacteriol. Sloss. M. Int. T. B. 25:114– 123. Chase. 198. ¨ 214. L. Weinstein. Baumberg. Cleyet-Marel. P. 383–427. 195. J. nov. N. Hill. Oligonucleotide probe for detection and identification of Campylobacter pylori. Clin. F. W. J. 199. 30: 1499–1504. and allied taxa by numerical analysis of phenotypic characters. 52:38–55. 219. Isoprenoid quinones of Campylobacter cryaerophila. W. Clin.org/ on January 3. Schachtner. Clin. abortion and swine dysentery. Microbiol. 1969. W. and K. J. C. Short. M. 29: 1785–1788. and J. Roller. Electrophoretic protein typing of Campylobacter jejuni subsp. 1985. 1989.. Int. Clin. and P. Infektionskr. R. A. C. and K. L. K. S. 197. 223. C. Appl. Kubo. M. Evolution in experimental populations of bacteria. 210. On.. A. J. Martinez-Murcia. Sci. Green. Lamouliatte. London. Int. W.. Bacteriol. 1994. De Wachter. Mendes. Ideonella dechloratans gen. 1–12.. E. and M.). R. R.. and H. G. “doylei” (nitrate-negative Campylobacter-like organisms) from human faeces and gastric mucosa. R. nov. pylori. R. Aldolase of lactic acid bacteria: immunological relationships among aldolases of streptococci and gram-positive nonsporeforming anaerobes. G. J. Costas. Rev. J. Ludwig. upsaliensis. H. consisting of strains that nodulate chickpeas (Cicer arietinum L.asm. K. On. 1971. F. 218. The ribosomal RNA database project. On. Strain variation in Campylobacter pylori detected by numerical analysis of one-dimensional electrophoretic protein patterns. Orla-Jensen. E. 18:374–390. Handbook of new bacterial systematics. M. Ohya. J. M. Holdeman. Stackebrandt. E. D. p. Microbiol. Syst.. Saunders (ed.-C. interspersed repetitive DNA sequences in prokaryotic genomes. J. E. 60. J. and C. J.. S. 200. Hauben. Goodfellow and A. a new bacterium ¨ capable of growing anaerobically with chlorate as an electron acceptor. Moore. De Ley. Owen. nov. Microbiol. and S. J. 42:27–36. I. 202. Leite. A. Morotomi.. 30:746–749. Naas. G.. A comparison using gel electrophoresis of cell proteins of campylobacters (vibrios) associated with infertility. B. Moore. L. P. N. Schleifer. Int. Mills. J. Copenhagen. Lupski. 1936. 1973. Comparative biochemical and immunological study of malic enzyme from two species of lactic acid bacteria: evolutionary implications. T. Lambert.and L-lactate dehydrogenases. C. 1991. Smith (ed. and L. Do bacteria have population genetics? p. Collins. Young. Schweinen. Wallner.. Meyer. Holmes. Virginia Polytechnic Institute Anaerobe Laboratory. W. M. R. Arbeit.. Sly. Population genetics of bacteria. a new starch-hydrolyzing species from cattle waste-corn fermentations. De Vos. 233. Fox. M. Malmqvist. J. The host range of the genus Xanthomonas. 25:1560–1561. G. and B. G. 43:135–142. W. A. 1992. G. J. Reproducibility of tolerance tests that are useful in the identification of campylobacteria. and L. 1996 jejuni and allied thermophilic campylobacters by hybridization of deoxyribonucleic acids. and R. E. A. A phylogenetic analysis of an atypical leuconostoc: description of Leuconostoc fallax sp. ¨ Timmis. J. Bacteriol. and K. FEMS Microbiol.. G. R. J. Microbiol. Klugbauer. ¨ Stenstrom. Jilg. L. V. A.. Population genetics of bacteria. S. Holmes. 187. 217. D. Rhizobium ciceri sp. helicobacters. Bacteriol. In S. Nucleic acid techniques in bacterial systematics. Gear. Guerrant. In E. Logan. Microbiol. 1995. Arachnia. Wellington. W. Parasitenkd.

30:938–951. G. Pimentel. Pot. p. Phylogeny of Helicobacter felis sp.. 1989.. J. H. Int. aerobic and microaerophilic spirilla. Johnson. In B. Owen. Appl. Sharpe. 1972. Curr. Vandamme. J. 1971. Int. 1993. and L. Ludwig. Ursing. with the species Prolinoborus fasciculus comb. Paris. 81–87. as Brevundimonas diminuta comb. K.). Use of DEAE-cellulose filters in the S1 nuclease method for bacterial deoxyribonucleic acid hybridization. Krieg and J. Bradbury. 23:333–339. R. Pette. Priest. D. M. and K. 1994. Leptospira species categorized by arbitrarily primed polymerase chain reaction (PCR) and by mapped restriction polymorphisms in PCR-amplified rRNA genes. Int. 279.. p. 58:3759–3761. J. In B. L. and Bacteroides ureolyticus by 16S ribosomal ribonucleic acid sequencing. J. and J. Identification of Campylobacter species by southern hybridization of genomic DNA using an oligonucleotide probe for 16S rRNA genes. Hansenn. J. 1994. P. and O. Bergey’s manual of determinative bacteriology. A. H. 242. Onderz. M.. 281. 1963. In M. 244. B. isolation... De Ley. D. Vandemeulebroecke. Appl. Microbiol. and P. Doudoroff. The prokaryotes. B. Identification of the lactic acid bacteria. 262. F. Krieg. Ursi. Environ. J. Schleifer. and H. R.. and Centers for Disease Control groups EF-4 and M-5 in the emended family Neisseriaceae. vol. L. In A. FEMS Microbiol. and D. M. Simsonsiella. Johnson. Holzapfel (ed. and C. and A. Van Beynum. B. Holdstock. 263.. Dworkin. Appl. Harder. Pot. Patarata. Phylogenetic relationships of lactic acid bacteria. 264. Lancet i:1349–1350. J. Int.. Jornvall (ed. Helicobacter mustelae. K. Identification and classification of Lactobacillus acidophilus. K. Syst. G. L. Pot.). N. Tech. 246. Microbiol. Lakshmanan. A. Selander. Paster. J. 1987. 1993. J. C. M. Rijkslandbauwproefstation te hoorn. M. Syst. J. 1984. Owen. Scotland.. 257. Komagata.. A. Grouping of Pseudomonas species on the basis of cellular fatty acid composition and the quinone system with special reference to the existence of 3-hydroxy fatty acids. A. Microbiol. Jackson. 1986. 241. W. 1995. Baranton. Morgan.. Leaper. J. cremoris. 28:2335–2339. 175:973–981. Dewettinck. P. Phylogeny of campylobacters. Microbiol. G.org/ on January 3. and S.. G. M. 17:128–134. and H. Gram negative bacteria of medical and public health importance: taxonomy—identification—applications. Ochman. Rev. Sandine. Nucleic acid homologies in the genus Pseudomonas. and B.. Horizon Scientific Press. 39:185–198. p. Paster. Classification of Pseudomonas diminuta Leifson and Hugh 1954 and Pseudomonas vesicularis Busing. G. Alysiella. J. Int. M.. H. size. Base composition. McClelland. 240. Bacteriol.and intergeneric relationships of the genus Aquaspirillum: Prolinoborus. and F. 1994. Millward-Sadler. Truper. a new bacterial genus for Xanthomonas maltophilia (Hugh 1980) Swings et al. FEMS Microbiol. Landbouwkd. Trust. Microbiol. Elsevier Science Publishers B. Lane. B. 1970. Rossau. Contopoulou. J. J. 248. J. Leclerc (ed.-H. M. 282. Johnson. Dewhirst. D. R. 107:382–384. Hertel. The potential role of a culture collection for identification and maintenance of lactic acid bacteria. Ireland. Syst. 1983. R. L. Norfolk. G. C. Cell wall and cell membrane antigens used in the classification of lactobacilli. Jackman. 266. 1989. Pot. Fernholm. Ferrero.. R. H. A. K. Can. 1983. 277. Kunisawa.. 1987. A. p. Int. 35:189–192. 1983. 1969. Bacteriol. and M. K. 1982. J. and T.. and Brevundimonas vesicularis comb. R. J. Microbiol. Proceedings of the First Lactic Acid Bacteria Computer Conference. Caugent.. Syst. Syst. The lactic acid bacteria. Roop. Roop. and K. Magnetospirillum and Oceanospirillum. Ser. ¨ ¨ and Freytag 1953 in Brevundimonas gen. Bacteriol. Krieg.. and W. 43:606–609. in press. Goodfellow and A. Stahl.. 493–521. Kiehlbauch. and A. Zusammensetzung und Anwendung von Bakterienkulturen fur therapeutisch Zwecke.). Rogosa. 44:499–510. Lastovica. The genus Aquaspirillum. De Ley. 169:2137–2141. Boekelscheurbacterien.. P. Taylor. B. 38:56–62. J. Ludwig. Kersters. Bacteriol. I. Stenotrophomonas. Welsh. 311–320. 113A:151–155. Bacteriol. 1974. Eikenella. J. Trust. J. J. Identification of ‘Campylobacter upsaliensis’ and other catalase-negative campylobacters from pediatric blood cultures by numerical analysis of electrophoretic protein patterns. Popovic-Uroic. H. B. Differential characteristics of catalase-positive campylobacters correlated with DNA homology groups. vol. Coynault. A. p. Gen. Pasteur (Paris) 105:897–910. Gillis. In H. M. Patton. J. H. Schleifer. Bremer. Helicobacter. Torck. and J. Kersters.. H. B. Environ. and K. Food Biotechnol. R. Balows. 139:513–517. 1994. Komagata. R. 2. DNA probes in food microbiology. J. L. 29:17–40. Society Appl. and M. S.. 1970. H. 249. Syst. nov. B. and applications. and T. Lee. Gilmour. Griffin. L. T. 271. Bacteriol. 251.). Analysis of electrophoretic whole-organism protein fingerprints. 669–676. W. W. Clin. and P. R. 1992. R. Tordoff. Romaniuk. 253. In R. 19:103–109. 490:315–346. E. H. P. Bacteriol. 1984. and K. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics.. England. Bacteriol. Lett. Palleroni. 237. Owen. Bacteriol. T. C. Mollby. 236. Polyacrylamide gel electrophoresis of spiral bacteria from the gastric antrum. Grosch. C. and A. Romaniuk. 57:1313–1318. Hawtin. Lett. 7–18. M. Pot. Costas. A. wolinellas. Bacteriol. Bacteriol. Identification of lactic acid bacteria isolated from Portuguese wines and musts by SDS-PAGE. Hoste. Genus III. 41:31–38. Walder. L. O’Donnell (ed. Bacteriol. K. Downloaded from http://mmbr. 1991. and J. Lett. and J. Fraser. Syst. J. ´ ´ Oyaizu. Microbiol. Microbiol. K. M.. 1993. 278. R. J. K. K. R. Fox... 1992. Baltimore. p. Vandamme. Gothe. G. Gen. Nicholson.. E. Leaver. L. 274. Microbiol. L. S. J. Versl. and N. and proposal of the neotype strain of Campylobacter sputorum.436 VANDAMME ET AL. nov. DNA homology studies of the catalase-negative campylobacters and “Campylobacter fecalis. Gillis. Syst. Sharpe. Scherer. Chichester. and nucleotide sequence similarities of genome deoxyribonucleic acids from species of the genus Campylobacter. N. johnsonii strains by SDS-PAGE and rRNA-targeted oligonucleotide probe hybridisations.. Ltd. Int. Buchanan and N. R. M.. Yokota. Schleifer. M. B. Ann. Veron. M. J. M. 1983. M. Pearson. 272. Segers. 36:407–477. 259. Mendes-Faia. 1984. F. B. Int. and K. J. Modern microbial methods. Janssens. In N. Sebald. F. Oyaizu. Int. Devriese. De Vos. G. Deoxyribonucleic acid hybridization among strains of lactobacilli. 250. 280. London. and N. 1990. Whittam. M. Wood and W. O’Rourke. Holt (ed. 76:288–293. B. . Genus I Pseudomonas Migula 1894. J. Vancanneyt. Ribosomal nucleic acid cistron similarities and deoxyribonucleic acid homologies of Neisseria. J.asm. and P.). 20:509–518.. J. Int. 154:1315–1322. D. V. Miry. Foo. and S. L. D. and Wolinella species. D. J. 269. P. 258. Austin. A. J. 267.). 1991.. p. B.. the spiral bacterium associated with human gastritis. A. 1992. S. Chemotaxonomic investigation of heterotrophic. and F. Pot. 31:823–831. Va. 1994.. 58:145–150. G. Campylobacter mucosalis (Lawson. Appl. Aquaspirillum fasciculus. H. 1993... Booth.. J. Pot. N. Palleroni. Syst. H. 2012 by guest 239. Thermotolerant Campylobacter with no or weak catalase activity isolated from dogs. p. W. Bifidobacterium Orla-Jensen. analysis of electrophoretic protein patterns of Campylobacter laridis and allied thermophilic campylobacters from the natural environment.. 1980. L. Krieg. Microbiol. nov. Perolat. Descheemaeker. 4:585–598. Appl.. 14:233–259. J. and M. Mannheim. The Williams & Wilkins Co. G. and K. and R. REV. 260. Peptidoglycan types of bacterial cell walls and their taxonomic implications. Chemical methods in prokaryotic systematics. Dewhirst.” an emended description of Campylobacter sputorum. Segers. N. Bacteriol. Appl. Vandenbussche. 256. 20:519–533. Bergey’s manual of systematic bacteriology. Distribution of polyamines in methanogenic bacteria. R. Microbiol. E. M. K. Syst. Heden (ed. P.. R. Giovannoni. S. Institut National de la Sante et de la Recherche Medicale. Schleifer. Bacteriol. and C. M. The similarities between Pseudomonas paucimobilis and allied bacteria derived from analysis of deoxyribonucleic acids and electrophoretic protein patterns. Inst. Bacteriol. A. 235. the genera Aquaspirillum. 1985. and M. John Wiley & Sons Ltd. The Williams & Wilkins Co. R. J. J. E. Kingella.-Forsch. Intra. M. 238. Schleifer. Reuter. J. Phylogenetic relationships of bacteria. 265. Hommez. Can.. J. Teneur en bases de l’ADN et classification ´ des vibrions. nov. and D. and J. respectively. B. 245. T. Bacteriol. FEMS Microbiol. J. F. 1988. Use of tRNA consensus primers indicate subgroups of Pseudomonas solanacearum by polymerase chain reaction. Zoltowska. J. 270. Kersters. L. Sandstedt. M. A. 65:69–78. 77:362–369. Appl. Blackie Academic & Professional Publishers. J.. R. Gen. Sakane.).. J. Arzneim. vol. Grouping of Pseudomonas species on the basis of cellular fatty acid composition and the quinone system with special reference to the existence to 3-hydroxy fatty acids. 43:331–335. Pace. J. R. identification. 8th ed. and S. 1985. Willems. Roop. 1983. 268. 1973. Campylobacter pylori. M. A handbook on the biology of bacteria: ecophysiology. Amsterdam.: emended description. New York. H. P. J. Modern bacterial taxonomy. E. E. U. Simonds. 273. P. Evaluation of the indoxyl acetate hydrolysis test for rapid differentiation of Campylobacter. Bacteroides gracilis. is not a true Campylobacter sp. 261. Kneifel. Palleroni. 243. Syst. Appl.. Development and application of oligonucleotide probes for identification of Lactococcus lactis subsp. M. J. 1979.. E. Chapman & Hall. Haesebrouck. 1943. Baltimore. 276. ¨ Schleifer (ed.. J. Falsen. nov. Microbiol. Bacteriol. Smibert. J. and Rowland 1981) comb. Salama. 247. Seal. E. Pettigrew. Characters used in the classification of lactobacilli. 1981. R. The lactic acid bacteria. K. In E. nov. P. J. P. 8:209–213. Kersters. 128: 2945–2954. J. W. D. Pot. B. K. Characterization and identification of Vagococcus fluvialis strains isolated from domestic animals. De Ley. Kandler. Microbiol. 51:873–884. and J. S. E. 1990. Smibert. D. gasseri. Environ. J. The ¨ hierarchy of life. A. Thielemans. 12:395–400. M. 141– 199. Doll. Ludwig. Syst. N. a new genus for MICROBIOL. III. The genera of lactic acid bacteria. 275. Kersters. Popoff.. Springer-Verlag. G. Ann. Gibbons (ed. N. 2nd ed. 103–117.). 2569– 2582. 252. ¨ Rogosa. J. Bacteriol. G. and related bacteria. Kersters. Walkes. 237AL. Haesebrouck. Daniels.. Pot. 254. Glasgow. and H. Ralph. 1993. Smibert. 255. Devriese. and W. and N. J. H. 1989. Bamforth. Phenotypic identification and differentiation of Lactococcus strains isolated from animals. Appl. Olsen. Musser.. Bacteriol. 42:44–57.

E. and K.. 307. Methods Microbiol. and K. R.. Syst. Syst. Academic Press Ltd. R.. A. P. Van Den Borre.). Huyghebaert. Goodfellow. poae Egli and Schmidt 1982. L. M. M. P. and K. p. De Ley. Dewhirst. a new species from dogs: an integrated study of phenotype and genotype. Microbiol.” Int. novel brominated arylpolyene pigments produced by bacteria of the genus Xanthomonas. De Wachter. 311. J.. Suzuki. D. Gen. V. Holt (ed. 13:295–303. Microbiol. Goossens. Bacteriol.. Fierens. 1990. Ledeboer. Int. G. E. Acta 294:416–424. 294. 1984. G. Syst.. W. R. C. Goebel. R. P. and reclassification of B. J. Swings. De Wachter.. and M. Owen. Gillis. arrhenateri Egli and Schmidt 1982 by numerical analysis of phenotypic features and protein gel electrophoregrams. E. P. Syst. Ursing.. 41:451–455. De Vos. G. L. Liessens. and E. Smits. P. Intraand interspecific relationships of veterinary campylobacters revealed by numerical analysis of electrophoretic protein profiles and DNA:DNA hybridizations. Syst. 168:379–385. and Arcobacter skirrowii sp. 289. Verbakel. Van den Berg. Jensen. Kersters. Kersters. Tenover. J. Benzi. J. Handbook of new bacterial systematics. by using cellular fatty acid profiles. Bacteriol. 1993. T.. F. M. 292. E.. M. Hommez. Vandamme. W. Kersters. Proposal for a new family. 1987. 330. Siverio. Higgins. Meiresonne. Acta Pathol. P. A. A. M. 113:135–156. Murray. G. E. Bergey’s manual of systematic bacteriology. Stanley. and R. Krieg. nov. D.. Taxonomic note: a pragmatic approach to the nomenclature of phenotypically similar genomic groups. and Comamonas testosteroni comb. Staley. De Ley. C. Appl. Int. Syst. De Vos. Krieg.. D. bronchiseptica by whole-cell protein electrophoresis and fatty acid analysis. 1993. J. and J. Syst. Syst. Campylobacteraceae. E. 329. F. 28:1016–1020. 291. Kersters. 301. Outbreak of recurrent abdominal cramps associated with “Arcobacter butzleri” in an Italian school. phleipratensis (ISPP list 1980) emend. vol. Rossau. C.-M. R. K. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. H.. Thompson. E. B. Walder. Bacteriol.. In M. 328. S. R. R. ¨ Bordetella hinzii sp. Swaminathan.. 18:75–107. and H. The importance of choosing outgroup reference organisms in phylogenetic studies: the Atopobium case. Skerman. Differentiation of Bordetella pertussis. H. 195–250. Owen. 1995. E. G.. D. and N. P. J. Microbiol. B. Int. K. Syst. 1986. Cockayne. S. Bussey. 17:39–43. Swings. Goodfellow and A. Vandamme.. and H. A. B.. and P. R. An uncultured gastric spiral organism is a newly identified Helicobacter in humans. Kersters. Bacteroides gracilis. 1994. 316. C. Bacteriol. Falsen. Vanbrabant. J. Swings. K. J. K. In M. E.. P. Monsieurs. Bacteriol. Sneath. McCarthy. Identification of Campylobacter cinaedi isolated from blood and feces of children and adult females. Chapman & Hall. nov. and J. J. T. J. Hoste. Costas. 323. L. Syst. Microbiol. P. Cambra. M. Gillis. Porter. Dewettinck. Bacteriol. L. G. M. Holt (ed. and A. Lee. and K. Unpublished data. Clin. B. J. E. Syst. 7:189–205. R. Dewhirst. Vancanneyt. Dewettinck. 304. Paepe. Nucleic acids and classification. 1995. A. Wirsing von Konig. 27:1775–1781. 326. Microbiol.. 1992. T. Classification of prokaryotic organisms: an overview.-P. Tabor. Hooper..VOL. E. E. M. H. D. 286. Microbiol. 1992. J. Vandamme. and A. Costas. Syst. 1996 283. an aerotolerant bacterium isolated from veterinary specimens. J. S. Bacteriol.org/ on January 3. B. and K. Baltimore. 1993. 295. The Williams & Wilkins Co. and J. H. Int. Subramaniam. W. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. B. 16:30–36. E. M. J. 314. Comparison between Xanthomonas campestris pv. protein electrophoretic. Smibert. 300. 308. Food Biotechnol. D. L. and W. G. J. Pot. 1993. P. L.. 41:88–103. Van Camp. Bhatnagar. P. E. E. Fox. On. J. P. Revision of Campylobacter. 318. W. M. screening and identification of lactic acid bacteria from traditional food fermentation processes and culture collections. 59:1805–1812. K. Kersters. 111–118. Pot. P. De Ley. Gen. Appl.. R.. Polyphasic taxonomic study of the emended genus Arcobacter with Arcobacter butzleri comb. and M. D. 320. Stackebrandt. Vandamme. 315. Vandamme. 1995. O’Donnell. 327. W. Numerical analysis of 295 phenotypic features of 266 Xanthomonas and related strains and an improved taxonomy of the genus. Microbiol... Phytopathol. P. 45:37–45. 310. a new thermophilic species from domestic animals: characterization and cloning of a species-specific DNA probe. and H. Appl. 1980. J. R.). Approved lists of bacterial names. Isolation. 1990. G. Ullmann.. C. Infect. P. 285. Tabor. Stead. B. and N. Van De Peer.. 1995. McGowan. E. Verrips. K. 37:52–59. Gorris. B. 287. The dynamic progression of evolved character states for aromatic aminoacid biosynthesis in Gram-negative bacteria. Clin. Ltd. 1986.. Goossens. J. J. nov. and R. Baltimore. ´ J. Microbiol. Vandamme. Van den Mooter. J. Syst. Syst. B. O’Donnell (ed. H.. Vandamme. The relationship between mismatched base pairs and the thermal stability of DNA duplexes. P. Falsen. M. Burnens. L. Int. J. K. Clin. Microbiol. R. Microbiology 140:3431–3440. B. Sandstedt.-M. I. J. Vandamme. Van den Mooter. gracilis as Campylobacter gracilis comb. Syst. J.. and C. M. Kersters. and W. 1993.. Vandamme. Differentiation between Xanthomonas campestris pv. Vancanneyt.. Vaneechoutte. M. and J. pv. Clin. and B. Butzler. with an emended description of the genus Comamonas. J. Kersters. Blackall. E. Handbook of new bacterial systematics. Int. Vandamme. M. Corzo. Vandamme. M. nov. 290. 303. Maertens. and J. Microbiol. nov. De Ley. and S. Kersters. J.. Microbiol. 302. De Ley. J. Bacteriol. C. Kersters. Syst.. Clin. DNA hybridization and phytopathological techniques. Persing. parapertussis. and B. Bacteriol. Environ.. 7:174–184. and J. Neefs. 293. Johnson. B. Moss. 1994. 17:444–458. Stackebrandt. and B. L. Microbiol. K. Tompkins. 44:846–849. K. J. 1991. 1993. On. 113:1–9. Pot. Pot. In N. 313. Tamaoka. F. 38: 190–200. M. Bacteriol. 297.. London. 139:2495–2504. R. Identification of enteropathogenic Campylobacter species by oligonucleotide probes and polymerase chain reaction based on 16S rRNA genes. R. M. Tsakalidou. 16S ribosomal RNA oligonucleotide cataloguing. Stackebrandt. J. F. Appl. Syst. L. S. Komagata. Int. Polyamines in microorganisms. Cell envelopes and classification. J. De Vos. J.. Towner. 1989. Int. 2012 by guest . The combined use of whole-cell protein extracts for the identification (SDS-PAGE) and enzyme activity screening of lactic acid bacteria isolated from traditional Greek dairy products. L. A. Clin. Lior. Falsen. Syst. J. Falsen. Y. 1992. and G. V. p.. G. Stackebrandt. Janssens. Hoste. Microbiol. Vancanneyt. Steenackers. Int. Isolation and identification of autotrophic and heterotrophic bacteria from an autohydrogenotrophic pilot-plant for denitrification of drinking water. Pot. Ursing. 1988.. Chemotaxonomic analyses of Bacteroides gracilis and Bacteroides ureolyticus. Microbiol. J. T. Dewhirst. Rossello-Mora. B. G. 15:402–408. Verstraete.. Campylobacter concisus. Grouping of plant-pathogenic bacteria and some other Pseudomonas spp. Dis. A. M. Pugina. Segers. and C. Syst. London. A. Arch. Linton. M. cassavae (ISPP list 1980) by means of phenotypic. Appl. pv. Bergey’s manual of systematic bacteriology. M. Hannson. vol. J. Ludwig. Chemotaxonomic significance of the xanthomonadins. 45:843–847. and Wolinella taxonomy: emendation of generic descriptions and proposal of Arcobacter gen. 1973. J. and P. nov. 1994.. M. Application of numerical analysis of electrophoretic protein profiles for the identification of thermophilic campylobacters. V. O’Donnell (ed. Vlaes. Hoste. Van Camp. 40:348–369. Academic Press Ltd. Int. Curr. Maraite. B. P. and A. 1990. J. Rossau. 42:344–356. Krieg and J. Microbiol. 1977. K. Molecular methods for microbial identification and typing. P.. D. Andrewes. A. A.. 30:2335–2337. 1992. M. Lipopolysaccharides as determinants of serological variability in Pseudomonas corrugata. 45:604.). 1994. Vlaes. B. nov. isolated from poultry and humans. Tanner. D. W.. P. J. H. Krieg and J. 14:57– 66. Genetic and phenotypic characteristics of a new group of Campylobacter isolated from pigs and cattle. Burnens. Paster. 288. 1992. Scand. B. Appl. 1993. J. 1984. manihotis (ISPP list 1980) and Xanthomonas campestris pv. Stackebrandt. P. E. and B. Mickelsen. O’Rourke. A. Pot. Helicobacter. 296. Goodfellow and A. 298. M.. 324. 1993. 38:321–325.. Microbiol. De Ley. Differentiation of campylobacters and Campylobacter-like organisms by numerical analysis of onedimensional electrophoretic protein patterns. F. Biophys. Starr. J. and pv. 1986. Structure of 16S and 23S ribosomal RNA genes in Campylobacter species: phylogenetic analysis of the genus Campylobacter and presence of internal transcribed spacers. Stanley. E. p. J. and J.). Microbiol. M. Mels. 16:361– 368. Hoste. 306. J. Numerical taxonomy. Kersters. Biochim. Vandamme. Garcia-Valdes. Int. J. 1985. 317. and R. Base composition and sequence homology of deoxyribonucleic acid of thermotolerant Campylobacter from human and animal sources. In N. C. Reclassification of Pseudomonas acidovorans den Dooren de Jong 1926 and Pseudomonas testosteroni Marcus and Talalay 1956 as Comamonas acidovorans comb. J. E. Rev. 33:2233–2239. The Williams & Wilkins Co. H.asm. Identification of EF group 22 campylobacters from gastroenteritis cases as Campylobacter concisus. London.. Kersters. Lalucat. Sneath (ed. Kalantzopoulos. Liesack. 322. 1. Gen. J. Hommez. J. 1994. Solnick. De Ley. Microbiol. J. Jenkins. 1993. 305.).. Linton. 321. P. 1984.. N. Rev.. 1–3. F. J. Vancanneyt. R. 151–194. Proteobacteria ¨ classis nov. Kersters. Van Belkum. Bacteriol. Vandamme. Manolopoulou. and K. Ursing. Paster. p. Helicobacter canis sp. nov. Vandamme. 45:145–152. and Eikenella corrodens by polyacrylamide gel electrophoresis. Microbiol. Zoidou. 1991. a name for the phylogenetic taxon that includes the “purple bacteria and their relatives. Microbiol. graminis (ISPP list 1980). H. Bacteriol. J. Microbiol. A. Appl. Bacteriol. Immunol. 30:225–420. Bacteriol. 1. D. J. Goering. S. Phylogenetic study of the genus Campylobacter. DNA fingerprinting of medically important microorganisms by use of PCR. Cookson. Characterization of Wolinella spp.. D. 312. Kersters. Syst. 1983. 284. A. Nicolai. A. Butzler. P. and L. Van den Mooter. B. 60. Kabaraki. M. Lauwers. Ludwig. 133:57–71. B. K.. 1991. Kersters. Appl. 325. A. Van Etterijck. V. G.. Downloaded from http://mmbr. 1995. POLYPHASIC TAXONOMY 437 309. Syst. 138:2293–2303. J. R. Truper. 331. Campylobacter helveticus sp. 42:281–295. P.. 1988. J. J. Appl. P. 1985. P. nov. B. 16:471–482. E. 299. 8:307–310. Arbeit.. Lopez. J.-P. 319.. J. J. and R. Sandstedt. P. Sect B 92:71–72. Pot. and J. Goossens. Syst. K. Gossele. C. K. Tytgat. and J. 24:562–565. 1993. Murray. and G. 49:81–99. Daneshvar. Int. Microbiol. Vandamme. K. Int. B. Ha.

E.. Utilization of fatty acid methyl esters for the differentiation of the Xanthomonas species.. A. Microbiol. R. J. Bacteriol. M. Kersters. Rafalski. Wyss. Versalovic. P. 339. Shinoda. Mol. A handbook on the biology of bacteria: ecophysiology. Swings and E. A. H. H.-P. Vancanneyt. Dellaglio.. Can. nov. Bacteriol. M. A. Vauterin. Swings. vol. 1991. A. Swings. M. comb. Wakamatsu. Int. Pot. Gossele. 373. Yang. and eubacteria. De Rouck. Bacteriol. Jensen. C. 42:107–119. and M. P. Gen. and J. Int. De Ley.. The phylogeny of purple bacteria: the beta subdivision. 1979. Int. F. Microbiol. E. Syst. Tingey. 343. E. B. D. Microbiol. and Alcaligenes latus. P. W. Kersters. J. M. Truper. P. G. Busse.. Ezaki. Grimont. M. Weyant. Thielemans. Peterson. a new genus for Pseudomonas facilis. De Ley. a new family encompassing the acidovorans rRNA complex. Bacteriol. and transfer of seven species of the genus Pseudomonas homology group II to the new genus. M. Bacteriol. 1995.. R.. Phylogenetic structure of the “leuconostocs”: an interesting case of a rapidly evolving organism. Whitaker. 1990. J. p. Chapman & Hall. 359. London. P. 4:552–557. T. REV. G. K. J. Arakaki.. 1987. Immunol. 1993. Klenk. M. Farrow. Polyamine patterns as chemotaxonomic markers for the genus Xanthomonas. F. I. De Ley.. Gillis. and J. 1995. A. S. Murray.. oryzicola and the bacterial “brown blotch” pathogen on rice by numerical analysis of phenotypic features and protein gel electrophoregrams. lactis comb. Syst. P. Brenner. B. and K.. nov. R. 1983. 1989. L. a new gram-negative species associated with septicaemia.). M. London. T. 45:472–489. 1988.. P. F. and R. nov. K. Ludwig. De Vos. Swings. D. 365. Syst. nov. Sarra. S. Hotta. R. 341. Kersters. Int. and K. Kersters. S. 1987.. A.). 2nd ed. 6:1673–1680. Auling.asm. G. Livak. A. Deoxyribonucleic acid homology among Lactobacillus species of the subgenus Betabacterium Orla-Jensen. Transfer of Rhodocyclus gelatinosus to Rubrivivax gelatinosus gen. E. and other clinical isolates. De Ley. Willems. Weisburg. Koops. Tech. J. 1:95–99.. Wesley. New York. Can. and S.. W. In J. 1993. Microbiol. K.. Int. nov. . Gillis. McClelland. Yang. R. and K. K.. and Hydrogenophaga taeniospiralis (formerly Pseudomonas taeniospiralis).. Pseudomonas saccharophila. P. and K. Brenner. nov. Garret. Truper. J. T. Int. Immun. Stolz. D. J. as Xylophilus ampelinus (Panagopoulos 1969) comb. A. J. Lett. 1987.. J. Willems. and B. Nucleic Acids Res. Ferreyra. Bacteriol. 351. 333. nom. H. J. Yang. R.). 1984. Kersters. R.. J. Harder. Syst. W. L. Int. De Ley. 1995.. A. M. M. Wayne. D. Towards an improved classification of Pseudomonas. Appl. A. Stackebrandt. p. P. R. FEMS Microbiol. Handbook of new bacterial systematics. Swings. Fiers. Microbiol. and P. D. Bacterial evolution. Microbiol. K..). Oyaizu. Lett. including Variovorax paradoxus gen. Vera Cruz. J. Clin. Paster. 349. Microbiol. 356. B. 2012 by guest 338. Koeuth. Balows. J. A. O’Donnell (ed. Int. Kersters. 366. Nucleic Acids Res. and applications. J. and M. Appl. Microbiol. 1991. J. 1994. G. Syst. Bacteriol. Syst. Botazzi. and Anderson) sp. 337. R. Palm. comb. 18:7213–7218. Proposal of Burkholderia gen.. In A. B. McClelland.. 1992. 89:319–333. Infect. Kersters. W. Appl. Hydrogenophaga pseudoflava (formerly Pseudomonas pseudoflava and “Pseudomonas carboxydoflava”).. 29:724–728.. Acidovorax delafieldii comb.. Goor. F. and J. 1984. Clin. 368. 371. Polyphasic taxonomic study of the emended genus Comamonas: relationships to Aquaspirillum aquaticum. Lactobacillus leichmanii and Lactobacillus bulgaricus. The genus Comamonas. C. ¨ Leffers. 41:445–450. (formerly Pseudomonas flava). G. G. and phylogenetic relationships with Leptothrix. Willems.. De Ley. L. C. M. M. European Patent Office. Vauterin. Springer-Verlag. Claeys. III. P. Daneshvar. 1991. J. and several clinical isolates. 334. Makita. 346. Steigerwalt. C. 1993. De Ley. B. 367. 3133–3136. and J. Syst. 1991. Fingerprinting genomes using PCR with arbitrary primers. P. and J. De Ley. 37:463–464. Hydrogenophaga palleronii (formerly Pseudomonas palleronii). Fanta. 1991. Jantzen. J. J. Yang. eukaryotes. Arakawa. 1989.. Nucleic Acids Res. Hydrogenophaga. Appl. Appl. Wait. and J. Y. 1993. identification. M. 364. D. K. A. for Alcaligenes paradoxus (Davis 1969). 130:2983–2999. J. 16:47–71. G. M. Moss. 360. Vogel. Amin. Int. J. Hoste. De Ley. 1994. W. Hoste. 37:422–430. Falsen. E. Applications of cellular fatty acid analysis. A. J. Balows. Moore. 251–280. D. and D.. 369. Van den Broecke.. Dworkin. A. A. nov. nov.. L. Kubelic. Lactobacillus lactis. Woese. Bordetella holmesii sp. Bacteriol. M. R. 340. 57:1380–1383. 345. Syst. Xanthomonas. N. V.. 18:6531–6535. P. nov. Kersters. Microbiol. comb. Transfer of several phytopathogenic Pseudomonas species to Acidovorax as Acidovorax avenae subsp. G. and J. Reclassification of Xanthomonas. Gillis. Lupski. nov. M. Viale. G. 1992.. 43:709–714. 1989. vol. Pseudomonas delafieldii. Swings. 41:427–444. Goodfellow and A. Willems. J. Bacteriol. Hashimoto. Yamamoto. Oligodeoxynucleotide probes for Campylobacter fetus and Campylobacter hyointestinalis based on 16S rRNA sequences. K. M.. Gropp. Syst. J. G. Syst. G. Welsh. isolation. Alvarez. S. 29:1812–1817. Microbiol. bulgaricus comb. B. K. 362. Academic Press Ltd.. J. S. and M. Kersters. avenae subsp. 335. C. H. Bocker. Willems. R. Xylophilus gen nov. 145:752–759. oryzae. 342. Kosako. O’Connor. H. Int. Pot. Collins. nov. Kersters. R. J. Int. M. Hudson. 2583–2590. identification. C. E. J. and O. Int. The phylogenetic relations of DNA-dependent RNA polymerases of archaebacteria. Vauterin. Ltd. L. Stackebrandt. Syst. Vauterin. Falsen (EF) group 13. 347. Gen. Transfer of Xanthomonas ampelina Panagopoulos 1969 to a new genus. Williams. and M. and J. isolation. The use of picolinyl esters for the characterization of microbial lipids: application to the unsaturated and cyclopropane fatty acids of Campylobacter species. nov. Weaver. 37:339–341. Downloaded from http://mmbr. Falsen. Swings. J. 19:6823–6831. and A. 1987. Welch.. and J. 157–192. I. 35:73–80. Springer-Verlag. Int.. Willems. H. 44:527– 533. J. and K. Krichevsky. cattleyae. C. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. 352. M. Bacteriol. New York.. B. M. Bacteriol. M. R. R. and W. Bacteriol. Compilation of 5S rRNA and 5S rRNA gene sequences. J. H. Y. B. M. Gillis. with the species Acidovorax facilis comb. A. Jantzen. K. In A. Falsen. V. K. Gillis. J. Clin. nov. and J. Tanner. 355. Moore.. Microbiol. Woese. Yano. MICROBIOL. Dewhirst. Rev. Soc. L. Vandamme. L. Hollis. Byng. Gherna. 40:384–398. L. J..org/ on January 3. Hoste. Kersters. E. Phillips. F. Vescovo. 348. I. A. Syst. Civerolo (ed. W. 354. Truper. 361. Report of the ad hoc ¨ committee on reconciliation of approaches to bacterial systematics. 336. Kersters. Acidovorax avenae subsp. Sphaerotilus natans. J. Microbiol. E. Dworkin. U. 1990. P. Wolters. W. A. nov. Bacteriol. L. 43:305–313. J.. 1983. 4:422–438. P. K. and C. J. Weiss. Hahn. J. Erdmann. 1985. J. Syst. J. K.. nov. rev. nov.. 1981. 1989... Syst. Campylobacter-Wolinella group organisms are the only oral bacteria that form arylsulfatase-active colonies on a synthetic indicator medium. ´ Swings. Evolutionary relationships among eubacterial groups as inferred from GroEL (chaperonin) sequence comparisons. 29:21–43. Zillig. and Acidovorax konjaci. and M. E. and J. Kawaguchi. and Acidovorax temperans sp. Schleifer. Microbiol. III. Wesley. Willems. Syst. The prokaryotes. Falsen group 10. 2nd ed. V. In M. 5:327–336. Bacteriol. 1993. S. E. A. Van den Mooter. 137: 1677–1687. Rev. 1992. L. Collins. A. Zanoni. D. 51:221–271. and H..-H. 1992. 1992. Int. Kandler. Gillis. C. K. R. P. Starr. Int. and applications. 44:223–229. Polyamine distribution in Vibrionaceae: norspermidine as a general constituent of Vibrio species. Pot. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Bacteriol.. G. The prokaryotes. Identification of lactobacilli from sourdough and description of Lactobacillus pontis sp. 1992. P.438 VANDAMME ET AL. Kersters.):r1–r85. ¨ Schleifer (ed. 350. A. Vos. and K. Yabuuchi. citrulli. E. Puhler. 1992. Harms. 363. Krieg. p. nov. Nucleic Acids Res. 372. P. publication 0534858 A1. Zabeau. Goor. A. and H. 357. Acidovorax avenae subsp. and P. De Vos. Application of fatty acid methyl esters for the taxonomic analysis of the genus Xanthomonas. Kersters. 93:227–234. and Lactobacillus delbrueckii subsp. Syst.. F. 1990. The genus Xylophilus. M. E. ¨ K. Ser. 1991. F. 1993. Segers. Bacteriol. and J.. Cardella. Syst. Vauterin. Appl.-H. Yang. Ehrmann. Vauterin.. Lactobacillus pentosus (Fred. Gillis. E. subjective synonyms of Lactobacillus delbrueckii subsp. G. 16(Suppl. with the type species Burkholderia cepacia (Palleroni and Holmes 1981) comb. Gillis. Hoste. 36:1251–1275. B. and E. 353. E. H. Grouping of Xanthomonas campestris pathovars by SDS-PAGE of proteins. 374. Xanthomonas campestris pv. Taxonomy of the genus Xanthomonas. Rapid identification of bacteria of the Comamonadaceae with amplified ribosomal DNArestriction analysis (ARDRA). J. Comparative allostery of 3-deoxy-D-arabino-heptulosonate-7-phosphate synthase as an indicator of taxonomic relatedness in pseudomonad genera. Kandler. Differentiation between Xanthomonas campestris pv. Swings. Phylogenetic analysis of rhizobia and agrobacteria based on 16S rRNA gene sequences.. Protein electrophoresis and classification. A. 370. 332. J. a new genus of hydrogen-oxidizing bacteria that includes Hydrogenophaga flava comb. 12:145–149. R. and M. A. Whitney. Kersters. Colwell. R. M. Microbiol. 358. N. D. Selective restriction fragment amplification: a general method for DNA fingerprinting. Paster. ¨ Schleifer (ed. EF group 16. and R. PCR-amplified length polymorphisms in tRNA intergenic spacers for categorizing staphylococci. Willems. and J. H. De Vos. M. Willems. Soncini. Willems.. Welsh. Syst. De Ley. P. B. Vandamme. Microbiol. and K.. S. Hammes. Microbiologica 2:317–330. 344. nov. Woese.. K. M. J. Acidovorax. Kersters. Harder. J. Comamonadaceae. and M. 1991. 41:65–73. P. G. G. Syst. p. and J. Willems. Pot. I. L. Bacteriol. Schillinger. J. M. G. 46:298–304. A handbook on the biology of bacteria: ecophysiology. M. G. R. and K. 33:1–7. Gillis.