Luc Vauterin

Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology

An ad hoc committee for the re-evaluation of the species definition in bacteriology met in Gent, Belgium, in February 2002. The committee made various recommendations regarding the species definition in the light of developments in methodologies available to systematists.

RECLASSIFICATION OF XANTHOMONAS.

A comprehensive DNA-DNA hybridization study was performed by using 183 strains of the genus Xanthomonas. This genus was shown to comprise 20 DNA homology groups which are considered genomic species. Four groups corresponded to the previously described species Xanthomonas albilineans, Xanthomonas fragariae, Xanthomonas oryzae, and Xanthomonas populi. The previously described species Xanthomonas campestris was heterogeneous and was divided into 16 DNA homology groups. One of these groups exhibited a high level of DNA homology with Xanthomonas axonopodis. The 62 pathovars represented in this study were allocated to appropriate species. Our results, together with previous taxonomic data, supported a comprehensive revision of the classification of the genus Xanthomonas. The species X. albilineans, X. fragariae, X. oryzae, and X. populi are not affected. The type species of the genus, X. campestris (Pammel 1895) Dowson 1939, is emended to include only the pathovars obtained from crucifers (i.e., X. campestris pv. aberrans, X. campestris pv. armoraciae, X. campestris pv. barbareae, X. campestris pv. campestris, X. campestris pv. incanae, and X. campestris pv. raphani). X. axonopodis Starr and Garces 1950 is emended to include 34 former X. campestris pathovars. The following species names are proposed: Xanthomonas arboricola sp. nov., including X. arboricola pv. corylina, X. arboricola pv. juglandis, X. arboricola pv. poinsettiicola (type C strains of the former X. campestris pathovar), X. arboricola pv. populi, and X. arboricola pv. pruni; Xanthomonas bromi sp. nov. for strains isolated from bromegrass; Xanthomonas cassavae (ex Wiehe and Dowson 1953) sp. nov., nom. rev.; Xanthomonas codiaei sp. nov., including type B strains of the former taxon X. campestris pv. poinsettiicola; Xanthomonas cucurbitae (ex Bryan 1926) sp. nov., nom. rev.; Xanthomonas hortorum sp. nov., including X. hortorum pv. hederae, X. hortorum pv. pelargonii, and X. hortorum pv. vitians; Xanthomonas hyacinthi (ex Wakker 1883) sp. nov., nom. rev.; Xanthomonas melonis sp. nov.; Xanthomonas pisi (ex Goto and Okabe 1958) sp. nov., nom. rev.; Xanthomonas sacchari sp. nov. for strains isolated from diseased sugarcane in Guadeloupe; Xanthomonas theicola sp. nov.; Xanthomonas translucens (ex Jones, Johnson, and Reddy 1917) sp. nov., nom. rev., including X. translucens pv. arrhenatheri, X. translucens pv. cerealis, X. translucens pv. graminis, X. translucens pv. hordei, X. translucens pv. phlei, X. translucens pv. phleipratensis, X. translucens pv. poae, X. translucens pv. secalis, X. translucens pv. translucens, and X. translucens pv. undulosa; Xanthomonas vasicola sp. nov., including X. vasicola pv. holcicola and X. vasicola pv. vasculorum (type B strains of the former taxon X. campestris pv. vasculorum); and Xanthomonas vesicatoria (ex Doidge 1920) sp. nov., nom. rev., which includes the type B strains of the former taxon X. campestris pv. vesicatoria. Differentiating characteristics were determined for the new species on the basis of metabolic activity on a range of carbon substrates by using the Biolog GN microplate system.

Comparison of AFLP and rep-PCR genomic fingerprinting with DNA-DNA homology studies: Xanthomonas as a model system.

The genus Xanthomonas contains a large number of strains, which have been characterized by a variety of phenotypic and genotypic classification methods. The Xanthomonas collection constitutes one of the largest groups of bacteria that have been characterized phylogenetically by DNA-DNA homology studies and genomic fingerprinting. Presently, a total genomic DNA-DNA homology value of 70% represents an internationally accepted criterion to define bacterial species levels. However, the complexity of DNA-DNA reassociation kinetics methods precludes the rapid analysis of large numbers of bacterial isolates, which is imperative for molecular microbial diversity studies. Therefore, the aim of this study was to compare more facile PCR-based genomic fingerprinting techniques, such as repetitive-sequence-based (rep)-PCR and AFLP genomic fingerprinting, to DNA-DNA hybridization studies. Using three different primer sets, rep-PCR genomic fingerprint patterns were generated for 178 Xanthomonas strains, belonging to all 20 previously defined DNA-DNA homology groups, and one Stenotrophomonas maltophilia strain. In addition, AFLP genomic fingerprints were produced for a subset of 80 Xanthomonas strains belonging to the 20 DNA-DNA homology groups and for the S. maltophilia strain. Similarity values derived from rep-PCR- and AFLP-generated fingerprinting analyses were calculated and used to determine the correlation between rep-PCR- or AFLP-derived relationships and DNA-DNA homology values. A high correlation was observed, suggesting that genomic fingerprinting techniques truly reveal genotypic and phylogenetic relationships of organisms. On the basis of these studies, we propose that genomic fingerprinting techniques such as rep-PCR and AFLP can be used as rapid, highly discriminatory screening techniques to determine the taxonomic diversity and phylogenetic structure of bacterial populations.

Phylogenetic position of phytopathogens within the Enterobacteriaceae

The almost complete 16S rDNA sequences of twenty nine plant-associated strains, representing species of the genera Erwinia, Pantoea and Enterobacter were determined and compared with those of other members of the Enterobacteriaceae. The species of the genus Erwinia may be divided into three phylogenetic groups. Cluster I represents the true erwinias and comprises E. amylovora, E. mallotivora, E. persicinus, E. psidii, E. rhapontici and E. tracheiphila. We propose to unite the species of cluster II, E. carotovora subsp. atroseptica, E. carotovora subsp. betavasculorum, E. carotovora subsp. carotovora, E. carotovora subsp. odorifera, E. carotovora subsp. wasabiae, E. cacticida, E. chrysanthemi and E. cypripedii in the genus Pectobacterium respectively as P. carotovorum subsp. atrosepticum comb. nov., P. carotovorum subsp. betavasculorum comb. nov., P. carotovorum subsp. carotovorum comb. nov., P. carotovorum subsp. odoriferum comb. nov., P. carotovorum subsp. wasabiae comb. nov., P. cacticidum comb. nov., P. chrysanthemi and P. cypripedii. The species E. alni, E. nigrifluens, E. paradisiaca, E. quercina, E. rubrifaciens and E. salicis, comprising cluster III, are being classified into a new genus Brenneria gen. nov. respectively as B. alni comb. nov., B. nigrifluens comb. nov., B. paradisiaca comb. nov., B. quercina comb. nov., B. rubrifaciens comb. nov. and B. salicis comb. nov. The species of the genus Pantoea, included in this study, form a monophyletic unit (cluster IV), closely related with Erwinia, whereas the three phytopathogenic species of the genus Enterobacter are scattered among the genera Citrobacter and Klebsiella.

Applicability of combined amplified ribosomal DNA restriction analysis (ARDRA) patterns in bacterial phylogeny and taxonomy

Abstract A standardized method for amplified ribosomal DNA restriction analysis (ARDRA) is described. The first step involves selection of five tetracutter restriction enzymes on the basis of theoretical digestions of known 16S rDNA (rRNA) sequences. In the second step, the experimentally obtained restriction patterns are normalized and combined by means of the pattern recognition and analysis software GelCompar. Finally, numerical analysis allows the strains to be grouped according to the similarities in their combined ARDRA patterns. Results obtained with representatives of two phylogenetic lineages, the genera Alcaligenes and Bordetella and the genera Bacillus and Paenibacillus , are presented. In general, the clustering of the strains corresponded well with known species delineations and topology of phylogenetic groupings except for the discrepant position of the Bacillus lautus type strain, which can probably be explained by non-authenticity of this strain in one of the analyses. The effect of using less than five restriction enzymes on the clustering was evaluated. The frequent occurrence of interoperon variability of the 16S rRNA gene in Bacillus and Paenibacillus was also demonstrated. Because ARDRA detects interspecies and interstrain as well as interoperon variability and enables a relatively fast multiple strain analysis per taxon, this technique is appropriate to obtain indicative phylogenetic and taxonomic information. This information can be used to select strains for further detailed taxonomic studies. ARDRA fingerprinting also allows the construction of a database for indentification purposes.

Comparison of 16s Ribosomal DNA Sequences of All Xanthomonas Species

The phylogenetic relationships of all validly described species of the genus Xanthomonas and the type strain of Stenotrophomonas maltophilia were analyzed by sequencing and comparing 16S ribosomal DNAs (rDNAs). The two genera exhibited a mean sequence similarity value of 96.6%, corresponding to differences at 50 nucleotide positions on average. The species of the genus Xanthomonas exhibited relatively high levels of overall sequence similarity; the mean similarity value was 98.2%, which corresponds to an average of 14 mutual nucleotide differences. Within the genus Xanthomonas, a group containing Xanthomonas albilineans, Xanthomonas hyacinthi, Xanthomonas theicola, and Xanthomonas translucens clustered apart from the main Xanthomonas core, whereas Xanthomonas sacchari formed a third phylogenetic lineage. Due to the very restricted variability in 16S rDNA sequences within the genus Xanthomonas, rDNA signatures that have possible diagnostic value for differentiating the Xanthomonas species could not be determined with certainty. When sequence similarities were compared with DNA-DNA pairing data determined previously, there was only a limited correlation. This illustrates the different resolving powers of the techniques for determining phylogenetic hierarchies and for species delineation.

Genomic diversity of the genus Stenotrophomonas

The clinical and environmental importance of Stenotrophomonas bacteria requires thorough, molecular studies on their epidemiology and taxonomy. In order to obtain a complete genomic profile of this genus, over 100 Stenotrophomonas maltophilia strains from various origins were investigated by AFLP fingerprinting. A subset of these strains was analysed by DNA hybridization and 16S rDNA sequencing. In contrast to their high phenotypic homogeneity, the strains were found to be very heterogeneous genotypically by AFLP fingerprinting. Nevertheless, ten cores of highly similar strains representing ten genomic groups were observed. The same groups could be retrieved by DNA hybridizations and also, partly, by 16S rDNA sequence analysis. The intergroup DNA similarities were too high to create confident species delineations, neither could the genomic groups be characterized by phenotypic features.

A Comprehensive Species to Strain Taxonomic Framework for Xanthomonas

ABSTRACT A comprehensive classification framework was developed that refines the current Xanthomonas classification scheme and provides a detailed assessment of Xanthomonas diversity at the species, subspecies, pathovar, and subpathovar levels. Polymerase chain reaction (PCR) using primers targeting the conserved repetitive sequences BOX, enterobacterial repetitive intergenic consensus (ERIC), and repetitive extragenic palindromic (REP) (rep-PCR) was used to generate genomic fingerprints of 339 Xanthomonas strains comprising 80 pathovars, 20 DNA homology groups, and a Stenotrophomonas maltophilia reference strain. Computer-assisted pattern analysis of the rep-PCR profiles permitted the clustering of strains into distinct groups, which correspond directly to the 20 DNA-DNA homology groups(genospecies) previously identified. Group 9 strains (X. axonopodis) were an exception and did not cluster together into a coherent group but comprised six subgroups. Over 160 strains not previously characterized by DNA-DNA hybridization analysis, or not previously classified, were assigned to specific genospecies based on the classification framework developed. The rep-PCR delineated subspecific groups within X. hortorum, X. arboricola, X. axonopodis, X. oryzae, X. campestris, and X. translucens. Numerous taxonomic issues with regard to the diversity, similarity, redundancy, or misnaming were resolved. This classification framework will enable the rapid identification and classification of new, novel, or unknown Xanthomonas strains that are pathogenic or are otherwise associated with plants.

APPLICATION OF FATTY-ACID METHYL-ESTERS FOR THE TAXONOMIC ANALYSIS OF THE GENUS XANTHOMONAS.

Summary Gas-liquid chromatographic analysis was applied to examine whole-cell fatty acid methyl esters (FAMEs) of a total of 975 strains which included representatives of all seven Xanthomonas species ( X. albilineans, X. axonopodis, X. campestris, X. fragariae, X. maltophilia, X. oryzae and X. populi ), 134 X. campestris pathovars and two X. oryzae pathovars as well as some related strains. At least 65 different fatty acids were found within the members of the genus Xanthomonas . A set of nine fatty acids (11:0 iso, 11:0 iso 3OH, 12:0 3OH, 13:0 iso 3OH, 15:0 iso, 16:1 cis 9, 16:0, 17:1 iso F and 17:0 iso) occurred in more than 99% of the 966 genuine Xanthomonas strains and are therefore considered as the most common fatty acids for the genus Xanthomonas . The three fatty acids 11:0 iso, 11:0 iso 3OH and 13:0 iso 3OH are characteristic for all members of the genus Xanthomonas and serve as a useful criterion to differentiate Xanthomonas from other bacteria. Cluster analysis revealed 31 major FAME clusters within the genus Xanthomonas . The Xanthomonas species X. albilineans, X. axonopodis, X. fragariae, X. maltophilia and X. populi each constituted a separate FAME cluster. Two FAME clusters were formed within X. oryzae , corresponding to pvs. oryzae and oryzicola , respectively. The species X. campestris was demonstrated to be heterogeneous and comprised 24 FAME clusters. In some cases, X. campestris pathovars which were isolated from related host plants grouped together. This was the case for the following pathovars from grasses: X. c. pvs. graminis, poae and phleipratensis (FAME cluster 7); X. c. pvs. arrhenatheri and phlei (FAME cluster 8), and also for the following X. campestris pathovars from cereals: X. c. pvs. cerealis, hordei, undulosa, secalis and translucens (FAME cluster 9). All six X. campestris pathovars from the crucifers ( X. c. pvs. aberrans, armoraciae, barbareae, campestris, incanae and raphani ) fell within FAME cluster 2. Many Xanthomonas campestris pathovars from legumes were grouped within FAME cluster 1. Although most X. campestris pathovars were homogeneous or had only one or a few atypical strains, some X. campestris pathovars were found to be heterogeneous. The greatest heterogeneity occurred in X. c. pvs. vasculorum and citri , forming 4 and 3 FAME subgroups, respectively. Other X. campestris pathovars which constituted more than one FAME subgroup included X. c. pvs. phaseoli and vignicola . The comparison of the results of FAME analysis and those of phenotypic tests, SDS-PAGE of whole-cell proteins and DNA-DNA hybridization is discussed. A database generated by the authors based on the fatty acid compositions of Xanthomonas allows rapid identification of unknown xanthomonads at the genus, species and often at pathovar level.

Grouping of Xanthomonas campestris pathovars by SDS-PAGE of proteins

Summary: A numerical analysis was performed on SDS-PAGE protein patterns of 307 Xanthomonas strains comprising all species and 27 X. campestris pathovars. The electrophoretic groupings corresponded in some, but not all cases with the existing pathovars. Six pathovars constituted distinct entities, comprising all strains investigated. These included X. campestris pv. campestris, X. campestris pv. graminis, X. campestris pv. hyacinthi, X. campestris pv. pelargonii, X. campestris pv. pruni and X. campestris pv. theicola. A great number of pathovars consisted of a homogeneous group from which only one or a few strains were aberrant. In two cases (X. campestris pv. ricini and X. campestris pv. vitians) even the pathovar reference strain was aberrant. Six pathovars from members of the plant family Fabaceae could not be differentiated from one another: X. campestris pv. phaseoli, X. campestris pv. phaseoli var. fuscans, X. campestris pv. cajani, X. campestris pv. vignicola, X. campestris pv. alfalfae and X. campestris pv. glycines. At least six X. campestris pathovars were heterogeneous, displaying two or more protein electrophoretic types: X. campestris pv. alfalfae, X. campestris pv. dieffenbachiae, X. campestris pv. juglandis, X. campestris pv. poinsettiicola, X. campestris pv. vesicatoria and X. campestris pv. vignicola. A database of SDS-protein patterns provides a valuable tool for the identification of unknown xanthomonads.