A multilocus sequence analysis of the genus Xanthomonas

https://doi.org/10.1016/j.syapm.2008.06.004Get rights and content

Abstract

A multilocus sequence analysis (MLSA) of strains representing all validly published Xanthomonas spp. (119 strains) was conducted using four genes; dnaK, fyuA, gyrB and rpoD, a total of 440 sequences. Xanthomonas spp. were divided into two groups similar to those indicated in earlier 16S rDNA comparative analyses, and they possibly represent distinct genera. The analysis clearly differentiated most species that have been established by DNA–DNA reassociation. A similarity matrix of the data indicated clear numerical differences that could form the basis for species differentiation in the future, as an alternative to DNA–DNA reassociation. Some species, X. cynarae, X. gardneri and X. hortorum, formed a single heterogeneous group that is in need of further investigation. X. gardneri appeared to be a synonym of X. cynarae. Recently proposed new species, X. alfalfae, X. citri, X. euvesicatoria, X. fuscans and X. perforans, were not clearly differentiated as species from X. axonopodis, and X. euvesicatoria and X. perforans are very probably synonyms. MLSA offers a powerful tool for further investigation of the classification of Xanthomonas. Based on the dataset produced, the method also offers a relatively simple way of identifying strains as members of known species, or of indicating their status as members of new species.

Introduction

The nomenclature of the genus Xanthomonas based on the studies of Dye [5] resulted in the publication of five species in the Approved Lists [29], with more than 110 pathogens identified at the infrasubspecific level as pathovars. Following the recommendation of Wayne et al. [36], differentiation of species has depended, by convention, on DNA–DNA hybridization studies and the application of arbitrary calibration values for DNA–DNA reassociation (70%) and ΔTm (<5 °C). The requirement that species proposals include this standard method reduced the likelihood of proposals based on different incongruent methods and the creation of many synonyms.

Although DNA–DNA reassociation offered a simple criterion for species circumscriptions, wider application in different taxonomic groups of organisms has shown that it fails to differentiate some taxa and indicates amalgamation of others that have distinct phenotypic differences that may merit species recognition [3], [10]. Furthermore, DNA–DNA reassociation studies are intensive and expensive; there is a difficulty of standardization between laboratories, and with increasing numbers of species requiring comparison, it is impractical for all but a few specialist laboratories [19], [34].

Based on DNA–DNA reassociation data, Vauterin et al. [35] proposed a revision of Xanthomonas that resulted in the recognition of 20 species, to which they reallocated 62 pathovars of the 150 then reported. The species of Vauterin et al. [35] were supported by polyacrylamide gel electrophoresis of proteins and fatty acid profiles, but there were no portable methods of identification for the routine allocation of strains to many of these species [37]. This approach to the classification and nomenclature of Xanthomonas has made the allocation of unidentified strains to taxa practically impossible. Since this work was conducted, only three additional reports of new species and new species combinations in Xanthomonas have been made [15], [27], [33].

Most of the species of Vauterin et al. [35] circumscribed distinct pathogenic populations, indicating that DNA–DNA reassociation characterized these pathogens as natural ecological groups. However, many pathogens included in a few species, namely X. arboricola, X. axonopodis, X. campestris, X. oryzae and X. translucens, differed from other members of their species by only a few phenotypic differences and required classification at subspecific or infrasubspecific levels, as subspecies or pathovars. Rademaker et al. [22] showed that the classification of Vauterin et al. [35] was supported by a combined rep-PCR analysis [17], as well as by AFLP, and Rademaker et al. [23] showed that a more refined discrimination was possible within the heterogeneous species X. axonopodis.

Maiden et al. [18] proposed that classification could be refined by generating a representation of the chromosome using multilocus sequence typing (MLST) – concatenation of a selection of suitable protein-coding genes and identification of allelic mismatches at the loci of closely related organisms. The concept has been extended to a consideration of more diverse taxa to include whole genera using sequences of protein-coding genes, called multilocus sequence analysis (MLSA) [10]. MLSA is increasingly seen as offering an alternative, more flexible way of comparing bacteria, towards the development of a species concept. Gevers et al. [10] provide a critical and comprehensive review of this approach to taxonomic differentiation. Some pathogenic species, Pseudomonas syringae [14], [25] and Xylella fastidiosa [26], have been studied without drawing taxonomic conclusions, but the method is increasingly being considered in the inference of species relationships [2], [4], [10], [12], [19], [24].

In a recent MLSA of Xanthomonas, Fargier and Manceau [7] investigated six house-keeping genes (atpD, dnaK, glnA, gyrB, rpoD and tpiA) and the structural gene, fyuA, as the basis for differentiation within Xanthomonas. Of these, four gene sequences, the chaperone protein dnaK (dnaK), tonB-dependent receptor (fyuA), DNA gyrase subunit B (gyrB) and RNA polymerase sigma factor (rpoD) were congruent in their representation of Xanthomonas spp. Partial sequences of these genes were used in the present study to consider the extent to which these genes in an MLSA might reflect the species relationships indicated by DNA–DNA reassociation and, therefore, the likelihood that this method offers an alternative to support species characterization. The classifications of Vauterin et al. [35] and Rademaker et al. [22], [23] are used as the points of departure in this study for a reconsideration of Xanthomonas species using MLSA.

Section snippets

Strains

Strains were obtained from the International Collection of Micro-organisms from Plants (ICMP), Landcare Research, Auckland. Information on these strains is given at the ICMP website (http://www.landcareresearch.co.nz/research/biodiversity/fungiprog/icmp.asp), and is listed in Table 1. All numbers in this text refer to ICMP depositions and the nomenclature of Vauterin et al. [35] is used.

DNA extraction

Strains were cultured on nutrient agar (Difco, USA) and incubated at 27 °C for 2–4 days. Genomic DNA was

Results

All four gene sequences were amplified for most strains of X. arboricola, X. axonopodis, X. bromi, X. campestris, X. cassavae, X. codiaei, X. cucurbitae, X. fragariae, X. hortorum, X. melonis, X. oryzae, X. pisi, X. populi, X. vasicola and X. vesicatoria. In spite of efforts to find alternative primers, attempts to amplify fyuA for X. cassavae ICMP 8666 and 8667 and Stenotrophomonas maltophilia ICMP 17033, and gyrB for X. melonis were unsuccessful. There was also a consistent failure to obtain

Discussion

In this study, relationships indicated by comparative analysis of 16S rDNA [13] were mimicked by the analyses of single and concatenated nucleotides and concatenated peptides. The MLSA was based on the four house-keeping genes, and did not include 16S rDNA sequences because degeneracy in codons means that house-keeping genes are subject to a lower level of selection constraint at the second and third codons compared with 16S rDNA, in which all bases are equally conserved and which directly

Acknowledgements

The New Zealand Foundation for Research, Science and Technology provided financial support. Robyn L. Howitt and Helen M. Harman, Landcare Research, are thanked for critically reading the manuscript.

References (41)

  • D.W. Dye

    The inadequacy of the usual determinative tests for the identification of Xanthomonas spp.

    N.Z. J. Sci.

    (1962)
  • E. Fargier, L’étude de la pathologie de Xanthomonas campestris et de la structure génétique de ses pathovars a permis...
  • E. Fargier, C. Manceau, Pathogenic and genetic characterization of Xanthomonas campestris by plant inoculation and...
  • L. Gardan et al.

    DNA relatedness among the pathovar strains of Pseudomonas syringae subsp. savastanoi Janse (1982) and proposal of Pseudomonas savastanoi sp. nov.

    Int. J. Syst. Bact.

    (1992)
  • L. Gardan et al.

    DNA relatedness among the pathovars of Pseudomonas syringae and descriptions of Pseudomonas tremae sp. nov. and Pseudomonas cannabina sp. nov. (ex Šutić and Dowson 1959)

    Int. J. Syst. Bact.

    (1999)
  • D. Gevers et al.

    Re-evaluating prokaryotic species

    Nat. Rev. Microbiol.

    (2005)
  • W.P. Hanage et al.

    Fuzzy species among recombinogenic bacteria

    BMC Biol.

    (2005)
  • H. Hauben et al.

    Comparison of 16S ribosomal DNA sequences of all Xanthomonas species

    Int. J. Syst. Bacteriol.

    (1997)
  • M.S.H. Hwang et al.

    Phylogenetic characterization of virulence and resistance phenotypes of Pseudomonas syringae

    Appl. Environ. Microbiol.

    (2005)
  • J.B. Jones et al.

    Validation list no. 109. List of new names and new combinations previously effectively, but not validly, published

    Int. J. Syst. Evol. Microbiol.

    (2006)
    J.B. Jones et al.

    Reclassification of the xanthomonads associated with bacterial spot disease of tomato and pepper

    Syst. Appl. Microbiol.

    (2004)
  • Cited by (260)

    View all citing articles on Scopus

    Nucleotide sequence data reported are available in the DDBJ/EMBL/GenBank databases under the accession numbers: dnaK, EU498747EU498848; fyuA, EU498849EU498947; gyrB, EU498948EU499066; rpoD, EU499067–EU499186.

    1

    Present address: Microbiology Department, Biomerit Research Centre, National University of Ireland, Cork, Ireland.

    View full text