- freely available
Pathogens 2014, 3(1), 211-237; doi:10.3390/pathogens3010211
Published: 18 March 2014
Abstract: Xanthomonas vasicola pathovar vasculorum (Xvv) is the bacterial agent causing gumming disease in sugarcane. Here, we compare complete genome sequences for five isolates of Xvv originating from sugarcane and one from maize. This identified two distinct types of lipopolysaccharide synthesis gene clusters among Xvv isolates: one is similar to that of Xanthomonas axonopodis pathovar citri (Xac) and is probably the ancestral type, while the other is similar to those of the sugarcane-inhabiting species, Xanthomonas sacchari. Four of six Xvv isolates harboured sequences similar to the Xac plasmid, pXAC47, and showed a distinct Type-IV pilus (T4P) sequence type, whereas the T4P locus of the other two isolates resembled that of the closely related banana pathogen, Xanthomonas campestris pathovar musacearum (Xcm). The Xvv isolate from maize has lost a gene encoding a homologue of the virulence effector, xopAF, which was present in all five of the sugarcane isolates, while xopL contained a premature stop codon in four out of six isolates. These findings shed new light on evolutionary events since the divergence of Xvv and Xcm, as well as further elucidating the relationships between the two closely related pathogens.
The bacterial genus, Xanthomonas, includes many economically important pathogens of crop plants . Genome sequencing of species and pathovars of Xanthomonas has led to insights into the evolution and mechanisms of virulence and their ability to overcome host defences . One pathovar whose study at the molecular level has been relatively limited is Xanthomonas vasicola pathovar vasculorum (Xvv).
Strains of Xvv are responsible for gumming disease in sugarcane and are also found naturally infecting some other monocotyledonous hosts, including maize and sorghum . In addition to the importance of gumming disease and Xvv per se, strains of this pathovar share a recent common ancestor with the causal agent of banana Xanthomonas wilt, namely Xanthomonas campestris pathovar musacearum (Xcm), with which they share an identical gyrB DNA sequence . Strains of Xvv do not cause disease in banana , but Xcm can infect maize and sugarcane.
The narrow genetic diversity in Xcm [5,6] suggests a population bottleneck, and Xcm may represent a single clonal lineage of the species, X. vasicola, or a closely related species, that has recently emerged to colonise bananas in east Africa. Whole-genome sequence data can provide greater phylogenetic resolution than analysis of a single gene fragment. Although all isolates shared identical gyrB amplicon sequences, analyses of genome-wide sequence data revealed that Xcm and Xvv comprise two distinct clades, although they are closely related to each other . This suggests that Xcm did not arise from Xvv, but rather from some currently unknown close relative of Xvv, possibly another pathovar belonging to the same species (X. vasicola).
Given the close phylogenetic relationship between Xcm and Xvv, molecular comparison between Xcm and Xvv is of importance for understanding the evolution of Xcm and the adaptation to banana. To better understand the range of diversity within Xvv, we analysed available genome sequence data for four isolates of Xvv (described in previous publications [7,8]) and new sequence data from a further two isolates of Xvv.
There has been some confusion in the literature regarding the taxonomy of the Xvv strains, whose genome sequences we analyse here. Lewis Ivey and colleagues  listed strains NCPPB (National Collection of Plant Pathogenic Bacteria) 702, NCPPB 1326 and NCPPB 206 as X. axonopodis pathovar vasculorum. In contrast, Qhobela and Claflin  classified strain NCPPB 1326 as X. campestris pathovar vasculorum. However, Vauterin and colleagues  proposed that these strains be included in the species, X. vasicola (as pathovar vasculorum), on the basis of genomic DNA hybridization studies. A subsequent analysis of cellular fatty acid profiles also confirmed a close affinity between these strains and strains of the species, X. vasicola, and only rather distant similarity to species X. axonopodis and X. campestris . Rademaker and colleagues  also list strain NCPPB 206 (synonymous with LMG 8284) as belonging to species X. vasicola (pathovar vasculorum, i.e., Xvv). Overall, the available evidence overwhelmingly supports the inclusion of these isolates in species X. vasicola, and therefore, in this manuscript, we refer to them as Xvv (rather than as X. campestris or X. axonopodis).
We previously generated draft genome sequences for one isolate of Xcm and one isolate of Xvv and identified several differences between them that might contribute to their distinct host ranges ; we found differences between Xcm NCPPB 4381 and Xvv NCPPB 702 in their repertoires of Type-III secretion system (T3SS) effectors, lipopolysaccharide (LPS) synthesis genes and Type-IV pilus (T4P) genes, but these differences were between just a pair of isolates, and so, it was unclear how generalisable these results were to other isolates of Xcm and Xvv. In a more recent publication, we reported sequencing the genomes of a further 13 isolates of Xcm and three isolates of Xvv . The main focus of that study  was genetic variation among isolates of Xcm, revealing two main phylogenetic groups (or sub-lineages) among Xcm. That previous study  reported a list of genes that were consistently conserved in Xcm and absent from the four Xvv genome sequences then available, and the study used data from the four Xvv genome sequences to generate a phylogenetic tree; however, no further analysis of the Xvv genome sequences was reported.
The current study extends the previous work by systematically searching for genetic differences among Xvv isolates, rather than focusing on variation between Xvv and Xcm or variation among isolates of Xcm. Furthermore, included in the current study are sequence data from two additional strains of Xvv, bringing the total number of sequenced Xvv isolates up to six. The sequence analyses presented here revealed large differences in gene-content among isolates of Xvv, both among isolates from sugarcane and also between isolates from sugarcane and a single isolate from maize. Some of these differences in gene content are ascribed to the gain and loss of plasmids, though many of the differences are likely to be chromosomally located.
We also report genetic differences implicated in important extracellular structures, such as lipopolysaccharide (LPS) synthesis, the T4P and candidate substrates (i.e., effectors) of the T3SS. Furthermore, further analysis of previously published data revealed some hitherto undetected differences in gene content among Xcm isolates, including a homologue of T3SS effector XopL and a plasmid.
2. Results and Discussion
By comparison of genome sequence data, we identified several likely important events in the evolutionary history of Xvv and Xcm; these are summarized in Figure 1 and include the acquisitions of plasmids and the exchange of genes encoding LPS biosynthesis and T4P, as well as the loss and gain of candidate T3SS effector genes. The evidence supporting the proposal of each of these events is presented in Subsections 2.2–2.9 and Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6. Further details of these findings are available in the Supplementary Material, which contains an additional 19 figures. The main findings were that: a plasmid similar to pXAC47 is found in a clade of four Xvv isolates from sugarcane (Section 2.4); the most ancestrally branching Xvv sugarcane isolate (NCPPB 895) may contain a plasmid similar to that from a cassava pathogen (Section 2.5); some (but not all) Xcm isolates harbour sequences similar to that from a plasmid in a cotton pathogen (Section 2.6); there is considerable variation in T4P genes among Xvv isolates (Section 2.7); there are two distinct sequence types of the LPS biosynthesis gene cluster among Xvv isolates (Section 2.8); and Xvv isolates vary with respect to their repertoires of putative T3SS effector genes (Section 2.9).
2.1. Overview of Sequence Data
Table 1 gives a brief description of the bacterial isolates from which the sequencing data used in this study originate. Table 2 lists summary statistics for the raw sequence data and Table 2 summarises the de novo assembly statistics. The sequence data from Xcm and for one Xvv isolate (Xvv 702) were described in previous publications [7,8]. The sequence data for four of the Xvv isolates were mentioned in a previous publication , but assembly statistics were not given. Therefore, we provide details of the assemblies here (Table 3). The contiguity of the assemblies for the two newly presented isolates (Xvv 890 and Xvv 895) are much lower than those of the previously presented assemblies (see N50 in Table 3). However, it should be noted that the incompleteness of de novo assemblies does not invalidate the results presented in the current study, since our inferences are based upon comparisons of alignments of raw reads rather than comparisons between assemblies; these alignments consist of unassembled read-pairs aligned against various reference genome sequences, and several examples of such alignments are illustrated in the Supplementary Material. The single exception is Figure 2, in which the comparison consists of alignments between de novo assemblies; it is possible that some gaps in the alignments in Figure 2 could arise through the incompleteness of the de novo assemblies. Figure 4 and Figure 5 are based on alignments between de novo assemblies, but the findings were also validated by the inspection of alignments of raw (unassembled) sequence reads (see Supplementary Material).
Five out of the six Xvv strains were originally isolated from sugarcane; the exception is 206, which was isolated from maize. Genomes of Xvv 890 and 895 were newly sequenced for this study. Genome sequences of Xvv 206, 1326 and 1381 were reported in a previous publication , but with only very limited analysis, since the focus of that study was on single-nucleotide polymorphism (SNP) in Xcm. The genome sequence of Xvv 702 was previously reported and compared against that of Xcm 4381 . We also included genome sequence data from several isolates of Xcm that have previously been published [7,8], because these are the sequenced genomes most closely related to Xvv and probably belong to the same species, X. vasicola .
|Table 1. Bacterial isolates and sequence datasets used in this study.|
|Isolate (NCPPB number)||Source and date of isolation||Source|
|Xvv 206 a||South Africa 1948||Zea mays|
|Xvv 702 b||Zimbabwe 1959||Saccharum officinarum|
|Xvv 890 c||South Africa 1960||S. officinarum|
|Xvv 895 c||Malagasy Republic 1960||S. officinarum|
|Xvv 1326 a||Zimbabwe 1962||S. officinarum|
|Xvv 1381 a||Zimbabwe 1962||S. officinarum|
|Xcm 2005 a||Ethiopia 1967||Ensete ventricosum|
|Xcm 2251 a||Ethiopia 1969||Musa sp.|
|Xcm 4379 a||Uganda 2007||Musa sp.|
|Xcm 4380 a||Uganda 2007||Musa sp.|
|Xcm 4381 b||Uganda 2007||Musa sp.|
|Xcm 4383 a||Uganda 2007||Musa sp.|
|Xcm 4384 a||Uganda 2007||Musa sp.|
|Xcm 4387 a||D. R. Congo 2007||Musa sp.|
|Xcm 4389 a||Rwanda 2007||Musa sp.|
|Xcm 4392 a||Tanzania 2007||Musa sp.|
|Xcm 4394 a||Tanzania 2007||Musa sp.|
|Xcm 4395 a||Tanzania 2007||Musa sp.|
|Xcm “Kenyan” d||Kenya (year not known)||Musa sp.|
a These sequence data were previously reported in . b These sequence data were previously reported in . c These sequences were newly generated for this study. d This sequence assembly was submitted to GenBank by the International Institute of Tropical Agriculture in 2011, but no accompanying manuscript has been published to the best of our knowledge. NCPPB, National Collection of Plant Pathogenic Bacteria.
|Table 2. Summary of sequence data.|
|Isolate (NCPPB number)||Number of read pairs||Read length||Coverage||SRA accession|
|Xvv 206 a||2,579,404||76||70 x||SRR494500.3|
|Xvv 702 b||2,913,785||36||35 x||SRR020202.3|
|Xvv 890 c||4,843,028||67||58 x||SRR1045340|
|Xvv 895 c||2,867,513||67||50 x||SRR1045341|
|Xvv 1326 a||2,365,912||76||63 x||SRR494491.5|
|Xvv 1381 a||2,450,234||76||66 x||SRR494499.3|
|Xcm 2005 a||2,536,030||76||72 x||SRR489154.7|
|Xcm 4379 a||3,652,875||76||102 x||SRR494484.2|
|Xcm 4380 a||4,069,509||76||113 x||SRR494485.2|
|Xcm 4381 b||5,052,905||36||56 x||SRR020203.3|
|Xcm 4384 a||1,976,797||76||55 x||SRR494488.2|
|Xcm 4392 a||2,554,161||76||72 x||SRR494498.3|
|Xcm 4394 a||3,295,956||76||92 x||SRR494489.1|
|Xcm 4395 a||4,117,662||76||117 x||SRR494490.2|
|Table 3. Summary statistics for de novo genome sequence assemblies.|
|Isolate (NCPPB number)||Number of scaffolds||Scaffolds N50||Total length (b.p.)||GenBank accession|
|Xvv 206 a||177||103,376||4,825,935||AKBM00000000.1|
|Xvv 702 b||97||205,000||5,478,002||ACHS00000000.1|
|Xvv 890 c||2,565||3,627||4,951,053||AKBN00000000.1|
|Xvv 895 c||1,229||8,362||4,803,807||AKBO00000000.1|
|Xvv 1326 a||253||74,455||4,951,570||AKBK00000000.1|
|Xvv 1381 a||137||108,088||4,958,101||AKBL00000000.1|
|Xcm 2005 a||156||61,233||4,692,764||AKBE00000000.1|
|Xcm 4379 a||95||147,554||4,758,198||AKBF00000000.1|
|Xcm 4380 a||87||149,435||4,751,644||AKBG00000000.1|
|Xcm 4381 b||115||143,688||4,793,900||ACHT00000000.1|
|Xcm 4384 a||84||157,780||4,741,777||AKBH00000000.1|
|Xcm 4392 a||162||90,974||4,728,564||AKBI01000000.1|
|Xcm 4394 a||85||151,917||4,792,617||AKBJ00000000.1|
|Xcm “Kenyan” d||510||36,401||4,907,936||AGFQ01000000.1|
a These sequence data were previously reported in . b These sequence data were previously reported in . c These sequences were newly generated for this study. d This sequence assembly was submitted to GenBank by the International Institute of Tropical Agriculture in 2011, but no accompanying manuscript has been published to the best of our knowledge.
2.2. Phylogenetic Relationships among Xvv Strains
We generated a phylogenetic reconstruction of the sequenced Xvv and Xcm strains based on single-nucleotide polymorphisms called against the reference sequence of the X. oryzae pathovar oryzae (Xoo) MAFF 311018 . The maximum parsimony phylogenetic tree is shown in Figure 1; the maximum likelihood method produced identical topology, and the topology is consistent with the tree that we previously presented . The Xvv strains form a monophyletic clade closely related to, but distinct from, Xcm and the genetic distances among these sequenced Xvv isolates are considerably larger than those among Xcm isolates.
Within the sequenced Xvv isolates, the single isolate from maize falls within the diversity of sugarcane isolates; in other words, there are not separate monophyletic groups for isolates from the two different hosts. Two of the sugarcane-derived isolates (Xvv 1326 and 1381) are indistinguishable on the basis of the SNPs used to generate the tree in Figure 1. They had both been collected from sugarcane in Zimbabwe in 1962 and may be essentially two isolates of the same bacterial population. However, these two isolates are genetically distinct from the isolate collected from sugarcane in Zimbabwe three years earlier (i.e., Xvv 702).
2.3. Global Genomic Comparison of Genomes of Xvv and Xcm Isolates
Alignment of our genome assemblies against a closely related reference chromosome sequence (Xoo MAFF 311018) suggested numerous differences in gene content both between Xvv and Xcm and also among Xvv isolates (see Figure 2). Particularly noticeable in Figure 2 is a region of the Xoo genome (centred at position 2,508 Mb) that is absent from Xvv 206, the isolate from maize. This consists of an 18-kb region of the genome, including the eight loci XOO2253–XOO2263 (GenBank accession numbers: BAE69008.1–BAE69018.1) that includes several predicted efflux proteins of unknown function. The absence of this region in Xvv 206 was confirmed by the alignment of sequence reads against the Xoo reference genome, independently of any de novo assembly.
We identified Xoo genes that are differentially present or missing in each of the Xvv and Xcm isolates (Figure 3). To do this, we aligned the Illumina sequence reads against the Xoo chromosome sequence using BWA . We then calculated the breadth of coverage of each Xoo gene in each Xvv and Xcm isolate using the coverageBed tools from the BEDtools suite . Note that this analysis involved the alignment of raw sequence reads against the reference; it was not dependent on any de novo assembly of our sequence data. We confirmed differential presence/absence by PCR amplification from genomic DNA (Figure 4). Breadths of coverage for each gene are indicated by the heatmap in Figure 3, which reveals numerous genes, whose presence distinguishes Xcm from Xvv and also several that distinguish among Xvv isolates. However, this approach is limited to the analysis of genes that are in the chromosome of Xoo. It excludes plasmids, as well as chromosomal genes in Xvv or Xcm that are not conserved in Xoo. Therefore, we also performed similar analyses based on using genomic assemblies of Xvv and Xcm genomes instead of the Xoo chromosome sequence. The results of these analyses are presented as heatmaps in the Supplementary Material. Some of these differences in gene content are discussed in more detail below.
2.4. Xvv Isolates NCPPB 890, 702, 1326 and 1381 Contain Sequences Similar to Plasmid pXAC47
We searched for evidence of plasmids in the Xvv and Xcm genomes by aligning the Illumina reads against all bacterial plasmid sequences in the RefSeq database as of the October 31, 2013. The Xvv Isolates 890, 702, 1326 and 1381 all yielded Illumina sequence reads that cover about 70% of the length of plasmid pXAC47 from Xanthomonas axonopodis pathovar citri (Xac) 29-1 (the RefSeq accession number is NC_020798.1). The genomes of Xvv Isolates 890, 702, 1326 and 1381 all contain sequences with extensive similarity to pXAC47, while the other isolates of Xvv (and Xcm) do not.
2.5. Xvv Isolate NCPPB 895 Contains Sequences Similar to a Plasmid from X. axonopodis pv. manihotis (Xam)
We found extensive sequence similarity between Xvv 895 and sequences from several recently published draft genome sequences  of Xam, a pathogen of cassava. These sequences are not found in the other sequenced Xvv isolates nor in the sequenced Xcm isolates. Specifically, a 34.7-kb Xvv 890 contig (GenBank: AKBO01000002.1) shared 99% nucleotide sequence identity with contigs from Xam strains NG1, UA306, ORST17 and ORST X27 originally isolated from Nigeria, Columbia, Congo and Togo (the GenBank accession numbers for these Xam sequences are, respectively: AKEG01000263.1, AKEN01000114.1, AKEH01000160.1 and AKEJ01000091.1).
Some other strains of Xam also contained some slightly less-similar sequences, sharing up to 89% nucleotide sequence identity, including strains IBSBF725, IBSBF436, IBSBF285, IBSBF2820 and UA324. Most of the 41.4-kb sequence from Xam ORST X27 is conserved in Xvv 895, but absent from the other sequenced Xvv and Xcm isolates. This sequence is annotated as a (partial) plasmid , presumably because it contains several conjugative transfer genes. Overall, these observations indicate that Xvv 895 has acquired a plasmid that is closely related to plasmids that are circulating in Xam strains, suggesting the inter-species exchange of plasmids over a wide geographic range.
2.6. Xcm Isolates NCPPB 4379, 4380, 4383, 4384, 4392 and 4395 Contain Sequences Similar to a Plasmid from X. citri pv. malvacearum Strain X20
Previously, we found no evidence for the presence of plasmids in Xcm 4381 . However, examination of data from Xcm Isolates 4379, 4380, 4383, 4384, 4392 and 4395 revealed extensive sequence similarity to a recently reported plasmid sequence from X. citri pv. malvacearum strain X20, a highly virulent pathogen of cotton isolated in Burkina Faso . This 39-kb plasmid sequence (GenBank: CM002030.1) was not present in any of the sequenced Xvv isolates nor in any of the Xcm isolates belonging to Sub-lineage I (as defined by ); it appears to be restricted to Xcm Sub-lineage II among the isolates of which it is present in all, except Xcm 4381. The most parsimonious explanation for this pattern of distribution is that this plasmid was acquired by the common ancestor of Xcm Sub-lineage II and subsequently lost in Xcm 4381. The GenBank accession number for the corresponding plasmid sequence in the Xcm 4384 de novo assembly is AKBH01000036.1. The average nucleotide sequence identity between the X. citri pv. Malvacearum plasmid and Xcm was 92%, somewhat lower than the 99% identity between Xvv 895 and Xam plasmid sequences.
2.7. Genetic Variation in the Type-IV Pilus (T4P) Among Isolates of Xvv
The global analyses of gene content (see Figure 3) revealed differences among the Xvv isolates with respect to their T4P genes (Figure 4 and Figure 5). For example, Figure 3 shows that Xvv Isolates 702, 890, 1326 and 1381 are distinguished from the other sequenced isolates by the presence of several T4P-related genes (including GenBank accessions: BAE68223, BAE6788, BAE6789, BAE69785, BAE69787 and BAE69790).
2.7.1. The pilVWXYE Gene Cluster
In Xanthomonas species, the T4P apparatus is encoded by clusters of genes scattered over several genomic locations, including two large clusters containing pilVWXYE and pilCABRS, respectively. The first of these two clusters falls between a gene encoding an excinuclease ABC subunit B and a gene encoding a decarboxylase-family protein. We found two distinct sequence types at this locus among Xvv isolates. The corresponding gene cluster in Xvv 702 encodes homologues of FimT, PilV, PilW, PilX, PilY and PilE and is highly conserved (at least 99% identical nucleotide sequence) in Xvv Isolates 890, 1326 and 1381 and Xoo (92% identity). However, the corresponding gene cluster is quite different in the other two sequenced Xvv isolates, namely 206 and 895. These two isolates share 99% nucleotide sequence identity with Xcm NCPPB 4381 and the other sequenced isolates of Xcm and 90% identity with X. vesicatoria ATCC 35937 .
In summary, there are two distinct sequence types at this pilVWXYE locus among Xvv isolates: (i) Xoo-like (in Xvv 890, 702, 1326 and 1381); and (ii) X. vesicatoria-like (Xvv 895, 206 and all Xcm); see Figure 4. The most parsimonious explanation for this pattern is horizontal transfer of the locus in a common ancestor of Xvv Isolates 890, 702, 1326 and 1381 after splitting from the Xvv 206 lineage (see Figure 1). A less parsimonious explanation would be multiple acquisitions of the same sequence.
2.7.2. The pilCABRS Gene Cluster
In addition to the pilVWXYE gene cluster described above, there is also variation at the pilCABRS gene cluster, with pilA (locus tag XOO1468 and accession number BAE68223 in Xoo) being particularly variable (Figure 5). The pattern of variation at this locus is similar to that at pilVWXYE, insofar as Xvv Isolates 890, 702, 1326 and 1381 have a Xoo-like sequence at this locus (94% nucleotide sequence identity between Xvv and Xoo), whereas Xvv Isolates 206 and 895 have a different sequence type.
The pilA genes are of different sequence types in Xvv 206, Xvv 895 and Xcm. The nucleotide sequence of Xvv 206 pilC and pilA is 87% identical to those of Xcm. This degree of sequence identity is significantly lower than for the core genome; most orthologous genes share at least 99% nucleotide sequence identity between Xvv and Xcm. Apart from Xcm, the next most similar sequence to Xvv 206 pilC and pilA comes from X. alfalfae subsp. alfalfae (Xaa) CFBP 3836 (79% identity). In Xvv 895, pilB is 95% identical to Xvv 206 and 94% identical to Xcm. However, at the pilA locus, there is further variation between Xvv 206, Xvv 895 and Xcm. The most similar sequence (as of November, 2013) in the public databases to the pilA gene of Xvv 895 is X. alfalfae subsp. alfalfae CFBP 3836 (90% identity). The pilA of Xcm shares 94% nucleotide sequence identity with X. gardneri ATCC 19865  and shows no detectable nucleotide sequence similarity with any other sequence in the public databases.
The pattern of sequence variation in the pilCABRS gene cluster indicates multiple superimposed horizontal transfer events, resulting in Xvv having three distinct sequence types of pilA: (i) the Xoo-type in Xvv 702, 890, 1326 and 1381; (ii) the Xvv 895 type that is 94% identical to Xaa; and (iii) the Xvv 206 type that is 79% identical to Xaa. The pilA of Xcm belongs to a fourth, X. gardneri-like type.
2.7.3. Concluding Remarks about Variation in T4P Genes
The T4P is a key virulence factor for phytopathogenic bacteria . It performs a range of functions, including twitching motility [20,21,22,23,24,25,26] and cell-to-cell adhesion [27,28], thereby playing a role in the formation of micro-colonies and biofilms [29,30,31,32], and PilA has been implicated in the transmission of the pathogen to seed . It is clear from our results that there have been several horizontal genetic transfers resulting in the replacement of T4P genes with alternative alleles in X. vasicola. It is not clear what functional significance, if any, arises from such allele exchanges, and there is no clear-cut correlation between T4P sequence type and host plant species. However, given the key role of the T4P in bacteria—plant interactions and the previously reported observation that some T4P genes are under selection in Xanthomonas species —this may warrant further investigation. It is also possible that a phage might exert a selective pressure on T4P; for example, some phages require the T4P to infect Pseudomonas aeruginosa .
2.8. There Are Two Distinct Sequence Types of LPS Biosynthesis Clusters in Xvv
The lipopolysaccharide (LPS) molecules that cover the outer membranes of Gram-negative bacteria [35,36] can be an important virulence factor in plant pathogens [37,38,39,40,41,42,43,44,45,46]. Horizontal transfer has led to hyper-variability among different strains within individual Xanthomonas species [47,48,49].
We previously reported  that the LPS locus in Xcm 4381 most closely matches that of X. axonopodis pv. citri (Xac) strain 306 , whereas half of the LPS locus in Xvv 702 was not detectably similar to Xcm 4381, but rather resembled that of X. albilineans strain GPE PC73 [8,51]. We  and subsequently others  pointed out that this pattern of sequence similarity is incongruent with the close phylogenetic relationship between Xcm 4381 and Xvv 702 and indicates recent horizontal transfer in one or both strains.
Sequencing of additional isolates indicated that the LPS biosynthesis gene cluster DNA sequence is highly conserved among Xcm isolates . However, in the present study, additional genome sequencing revealed variation in this locus among isolates of Xvv. As illustrated in Figure 6, Xvv 895 shares 99% nucleotide sequence identity with isolates of Xcm, which, in turn, share 97% identity with Xac 306. However, in Xvv 206, 890, 1326 and 1381, the LPS cluster shares at least 99% identity to that of Xvv 702. Approximately one half of the LPS cluster in these Xvv isolates (adjacent to etfA) is 99% identical to that of Xvv 895 and Xcm. However, the other half (adjacent to metB) shares no detectable sequence similarity with Xvv 895, but it does share 84% identity with X. sacchari  and 86% with X. albilineans [8,51].
As Pieretti and colleagues noted , it is interesting that the Xvv 702-type LPS cluster is common to three distinct Xanthomonas species that all inhabit the xylem of sugarcane, and thus, there might be opportunities for these species to come into contact with each other and exchange genetic material. However, it is unlikely that this type of LPS is uniquely adapted to evading recognition as a pathogen-associated molecular pattern [54,55] by sugarcane, since Xvv 895, isolated from sugarcane, has a Xac-type LPS cluster, as does Xcm, which can also infect sugarcane . Furthermore, Xvv 206 has the Xvv 702-type LPS cluster and was originally isolated from maize; this further emphasizes that there is not a clear correlation between LPS cluster type and host plant species. Drivers for variation in the LPS might be interactions with phages [36,37,56,57] or with insect vectors .
2.9. Xvv Isolates Differ in Their Repertoires of T3SS Effector Genes
The repertoire of T3SS effectors can significantly influence a bacterial phytopathogen’s host range [59,60,61,62]. Therefore, differences in T3SS repertoires between Xcm and Xvv are of great interest, as they might partly explain the ability of Xcm to cause disease in banana, whilst Xvv appears to be non-pathogenic in banana. We previously compared the set of T3SS effectors encoded by Xcm 4381 against the set encoded by Xvv 702 . In that previous study, we identified only a few differences. Specifically, Xcm encodes two homologues of XopJ that are absent from the genome of Xvv 702, and Xvv702 encodes a homologue of XopAF that is absent from Xcm 4381 . However, in the present study, utilizing additional sequence data, we found that XopAF is absent from the genome of Xvv 206, though it is present in Xvv 895, 890, 1326 and 1381, as well as in Xvv 702. It is absent from all sequenced isolates of Xcm.
2.9.1. Gene for XopAF is Absent from Xvv 206, Isolated from Maize
We previously reported that Xvv 702 encodes a homologue of XopAF (GenBank: ACHS01000051.1, bases 7184–7783; RefSeq: WP_010364039) that is not present in the genome of Xcm 4381 . This sequence shares 86% amino acid sequence identity with the XopAF (also known as AvrXv3) from X. euvesicatoria that was originally identified as an avirulence factor, inducing the hypersensitive response (HR) in resistant tomato and pepper plants . It is also identical at the amino acid sequence level to proteins encoded by X. translucens pv. translucens (Xtt) DSM 18974 (RefSeq: WP_003475568). In the present study, we found that this xopAF gene is present in Xvv 895, 890, 1326 and 1381, as well as in Xvv 702. It is not present in Xcm nor in Xvv 206. Thus, XopAF is encoded by Xvv isolates from sugarcane, but not by the Xvv isolate from maize and not by Xcm isolates from banana and enset.
This begs the question of whether XopAF1 contributes to the limitation of host range in Xvv; that is, one might hypothesise that XopAF confers avirulence in banana and that the absence of XopAF in Xcm enables its pathogenicity in banana. In a recent comparative study of the genomes of different pathotypes of X. citri pv. citri (Xca) , the authors noted that XopAF was encoded in the genomes of a narrow-host-range strain, but absent from a closely related broad-host-range strain. Therefore, they hypothesised that XopAF might confer avirulence and contribute to the limitation of the host range. However, their mutational analysis showed that xopAF did not affect host range, but it did contribute to the ability of X. citri pv. citri pathotype Aw (Xcaw) to grow in a Mexican lime host plant.
It should be noted that the predicted XopAF protein in Xcaw (RefSeq: WP_007652722.1) is much more divergent from the originally described sequence from X. euvesicatoria (WP_008577605.1), sharing only 31% amino acid sequence identity, whereas the Xvv protein shares 86% identity with X. euvesicatoria XopAF. Therefore, the Xvv XopAF protein is likely to interact with plants differently. Furthermore, the host plants in question are very different with X. euvesicatoria and Xcaw infecting dicots and Xvv and Xcm infecting monocots. However, it is reasonable to suppose that the Xvv protein is likely to be a T3SS effector and a potential avirulence factor, given the 86% identity between it and the experimentally characterised X. euvesicatoria XopAF . XopAF contains a DNA-binding domain at its C terminus and may allow the pathogen to manipulate its host by affecting the expression of plant genes. It would be interesting to test whether heterologous expression of xopAF in Xcm would cause avirulence in banana and whether deletion of xopAF in Xvv would have any impact on virulence in sugarcane or maize.
In Xvv 702, the xopAF gene falls between Positions 7181 and 7543, on the reverse strand in GenBank accession ACHS01000051.1. This resides within a genomic region that also encodes several phage-associated proteins, including phage-related lytic enzyme, phage-tail protein, baseplate assembly protein J, phage-tail fibres and phage-tail fibre protein, and may result from the integration of a pro-phage into the Xvv genome. In the wheat pathogen, Xtt DSM 18974, the xopAF gene encoding an identical protein sequence is located in a different genomic context; it resides at Positions 40277 to 40933 in GenBank accession CAPJ01000122.1 (locus tag: BN444_00905). This region of the Xtt genome does not contain any obvious phage-related genes, but does contain a predicted transposase for insertion sequence element IS629 (locus tag: BN444_00906), suggesting a mechanism for the mobility of this gene.
2.9.2. Gene Encoding Homologue of XopL Underwent Truncation in Common Ancestor of Xvv 890, 702, 1326 and 1381
We also found polymorphism with respect to a xopL-like gene in Xvv. The genomes of Xvv isolates 206 and 895 each encode a protein with 70% amino acid sequence identity to XopL from X. campestris pv. vesicatoria (Xcv) strain 85-10 (RefSeq: YP_364951.1) . The GenBank accession numbers for the Xvv 895 and 206 sequences are AKBO01000570.1 (Positions 1804–3768) and AKBM01000211.1 (Positions 29591–31555), where they encode a full-length protein. However, in Xvv Isolates 890, 702, 1326 and 1381, there is a single-nucleotide C→Asubstitution resulting in a TCA codon being transformed to a TAA stop codon. This substitution occurs at Position 14191 in GenBank accession ACHS01000315.1 and is predicted to result in the protein being truncated to 265 amino acids (compared to the full-length 654 amino acids).
This xopL-like sequence is completely absent from Xcm isolates belonging to Sub-lineage II (Xcm 4379, 4380, 4381, 4383, 4384, 4392 and 4395). However, it is present in Xcm isolates belonging to Sub-lineage I (Xcm 2005, 2251, 4387 and 4389), where each genome encodes a full-length protein. Thus, it appears that this xopL homologue may have been lost twice: (i) completely deleted in a common ancestor of Xcm Sub-lineage II; and (ii) truncated by a premature stop codon in a common ancestor of Xvv 890, 702, 1326 and 1381.
It has recently been demonstrated that XopL possesses E3 ubiquitin ligase activity, induces plant cell death and subverts plant immunity and that the ligase activity is associated with the C-terminal region of the protein . The premature stop codon found in xopL of some Xvv isolates has split the XopL-encoding open reading frame (ORF) into two.
Interestingly, there is a candidate plant inducible promoter (PIP) box [67,68,69,70] upstream of the second ORF, which corresponds to the C-terminal region of XopL, in which the E3 ubiquitin ligase resides. This suggests the hypothesis that this truncated ORF still has the potential to be expressed and be induced in planta and that it might still have biochemical activity, though it is unclear whether it would be a substrate of the T3SS. This potential PIP box (sequence TTCCGgcgaacatgcagcaaTTCGC) is located at Positions 14107 to 14137 in ACHS01000315.1, which is approximately 160 bp upstream of the C-terminal XopL ORF at 14296 to 15366. There is another PIP box (sequence TTCGCtacgataaagatgacTTCGC) located at 13300 to 13347, which is approximately 50 bp upstream of the ORF homologous to the XopL N terminus located at 13395 to 14192. The complete set of predicted PIP boxes in Xvv 702 and Xcm 4381 is tabulated in the Supplementary Material.
2.9.3. Absence of Genes Encoding Homologues of XopJ Distinguishes Xcm from Xvv
In addition to the homologues of XopAF and XopL, a further two potential T3SS show differential presence/absence among our sequenced strains. We previously reported that Xcm 4381 encodes two homologues of XopJ that are absent from Xvv 702 . Our subsequent analyses have confirmed that these are conserved in all the sequenced Xcm isolates and are absent from all the sequenced Xvv. Therefore, these predicted T3SS effectors remain as candidates for contributing to the differences in host range between Xvv and Xcm.
3. Experimental Section
3.1. Sources of Bacterial Strains
Bacterial strains were obtained from the National Collection of Plant Pathogenic Bacteria (NCPPB) at The Food and Environment Research Agency, UK (Fera). DNA library preparation and genome sequencing using the Illumina GA2x were performed using standard Illumina protocols, as previously described [7,8].
3.2. Preparation of Genomic DNA
For DNA preparation, bacterial strains were grown overnight at 28 °C in 10 mL King Broth shaken at 200 rpm. Bacterial cells were harvested by centrifugation and re-suspended in TE buffer (50 mM Tris-HCl, 40 mM EDTA, pH 8.0) containing 12 µL of 20 mg/mL lysozyme and 10 mg/mL RNase and incubated at 25 °C for 10 min with 17 µL 10% sodium dodecyl sulphate, then incubated on ice for 5 min. Proteins were dissolved with 170 µL of 8 M ammonium acetate, vortexed vigorously for 30 s centrifuged at 4 °C and for 15 min. DNA was precipitated with isopropanol and re-dissolved in 100 μL of 10 mM Tris, pH 8.0, and 1 mM Na2EDTA.
3.3. Genome Sequencing
We used the Illumina GA2x platform to sequence genomes of Xvv strains NCPPB 895 and 890, generating paired sequence reads of length 67 nucleotides, according to the manufacturer’s instructions.
3.4. Alignment of Sequence Reads Against Reference Genome Sequences
We used BWA  to align GA2x sequence reads against a reference genome sequence and used IGV  to visualize the alignments and SAMtools  to manipulate the alignments and convert between formats.
3.5. SNP Calling and Phylogenetic Analysis
We used a very conservative approach to infer SNPs from the alignments of Illumina reads against the previously published Xoo reference draft genome assembly. To avoid false positives and false negatives, we only used those regions of the Xoo genome with a coverage depth of 10 or more for every sequenced Xcm and Xvv genome and where there was at least 95% consensus among the sequence reads within each isolate. Just over 30% of the length (1,507,606 out of 4,940,217 nt) of the Xoo genome fulfilled these two criteria. In other words, for 30% of the Xoo chromosome, there was sufficient quantity and consistency in our data to be almost certain of the sequence in all of the eight isolates (six Xvv and two Xcm; see Figure 1); for the remaining 70% of the genome, there was some degree of ambiguity in the data for one or more of the isolates. The phylogenetic tree was inferred using the Maximum Parsimony method implemented in MEGA5  based on 39,665 single-nucleotide variants with respect to the chromosome of Xoo MAFF 311018. Bootstrap values were calculated as percentages of 500 trials.
3.6. Genome Assembly
De novo assembly of Illumina sequence reads was performed using Velvet 1.1.04 . We discarded any sequence reads that contained one or more “N” prior to assembly. It is difficult or impossible to predict the optimal parameter values for Velvet assembly. Therefore, we generated assemblies using a range of combinations of hash length and coverage cut-off and chose the assemblies giving the largest N50 values. For the newly presented assemblies (i.e., for Xvv 890 and 895), the parameter values were for Xvv 895: hash length = 25 and coverage cut-off = 2 and for Xvv 890: hash length = 29 and coverage cut-off = 3.
3.7. Identification of Presence and Absence of Genes
We used BEDtools  to infer the breadths of coverage for genomic features based on Binary Alignment Map (BAM) files from the Burrows-Wheeler Aligner (BWA) and General Feature Format (GFF) files from Rapid Annotation using Subsystem Technology (RAST) . We used the pheatmap package in R to generate heatmaps . However, it should be noted that incompleteness of de novo assemblies do not invalidate these, since our inferences are based upon comparisons of alignments of raw reads rather than comparisons between assemblies; these alignments consist of unassembled read-pairs aligned with BWA against various reference genome sequences, and several examples of such alignments are illustrated in the Supplementary Material. The single exception is Figure 2, in which the comparison consists of alignments between de novo assemblies; it is possible that some gaps in the alignments in Figure 2 could arise through the incompleteness of the de novo assemblies. Figure 4 and Figure 5 are based on BWA alignments between de novo assemblies, but the findings were also validated by the inspection of alignments of raw (unassembled) sequence reads (see the Supplementary Material).
3.8. Visualisation of Genome-Wide Patterns of Sequence Conservation
We used BLASTN  to align assembled sequences and visualized the alignments using the Artemis Comparison Tool  and BLAST Ring Image Generator (BRIG) , which is a wrapper for Circular Genome Viewer (CGView) .
3.9. Identification of Potential PIP Boxes
We built a profile hidden Markov model (HMM) based on a multiple sequence alignment of 22 known PIP boxes from X. vesicatoria (from Table 3 in ) using hmmb from the HMMER 1.8.5 package . The DNA sequence was scanned against this profile-HMM using hmmls from HMMER 1.8.5 with a bit-score cut-off of 10.0.
Here, we analyse draft genome sequences for two isolates of Xvv to augment the four previously published [7,8] draft genome sequences of Xvv. Comparative analyses of these genome sequences and previously published genome sequences of the closely related pathovar, Xcm, have revealed extensive differences in gene content among Xvv. This manuscript describes some of these differences in detail, including differences in plasmid content, LPS biosynthesis clusters, T4P and T3SS effectors. The main evolutionary events are summarized graphically in Figure 1. As well as providing some insight into evolutionary events within Xvv, these sequence analyses also further refine our understanding of the genomic differences between Xvv and the very closely related Xcm, which is a recently emerging pathogen in banana and enset; the availability of the sequence from multiple isolates allows us to distinguish between inherent variation within Xvv that might confound attempts to identify important genetic differences between the two pathovars and for functional analysis of important virulence factors.
It is clear that Xcm is genetically highly monomorphic [5,6,7]; here, we show that, apart from several phage-related genes and the SNPs in the core genome described previously , the few genetic differences among Xcm isolates can be explained by the acquisition of a plasmid in Xcm Sub-lineage II, which is not present in Sub-lineage I, nor in at least one isolate of Sub-lineage II (Xcm 4381). Additionally, the two Xcm sub-lineages differ in that members of Sub-lineage II have lost a gene encoding a homologue of XopL; interestingly, this gene has acquired a premature stop codon in four of the six Xvv isolates, suggesting that isolates of both Xcm and Xvv have independently converged on eliminating XopL.
In contrast to the limited genetic diversity within Xcm, there is considerable diversity within Xvv, both at the level of SNPs in the core genome and at the level of gene content. Some of the differences in gene content are ascribable to the acquisition of two different plasmids (one in Xvv 895 and one in Xvv 890, 702, 1326 and 1381), but there are also differences in chromosomally located gene clusters, such as those encoding LPS biosynthesis and T4P.
Overall, this work suggests hypotheses for future work towards understanding the molecular basis for the ability of Xcm to emerge as an important pathogen of banana and enset. For example, one consistent difference is that all sequenced Xcm isolates encode two homologues of XopJ that are absent from all sequenced isolates of Xvv. In X. campestris pv. vesicatoria, this T3SS effector has been shown to interfere with salicylic acid-dependent defence responses to attenuate the onset of necrosis and to alter host transcription ; it will be enlightening to test the contribution of the two XopJ homologues in Xcm interaction with banana and enset. Furthermore, understanding the emergence of Xcm will require the study of genome sequences of a wider range of strains within the species, X. vasicola, to which Xcm probably belongs ; several isolates are available in strain collections for X. vasicola pv. holcicola , but there is also a need to survey other as yet unknown members of the species that might inhabit the centre of origin, perhaps colonizing other monocot plants.
This study was supported in part by the National Agriculture Research Organisation, Uganda, under the MSI/World Bank grant 2009. The authors wish to thank Karen Moore and Alex Moorhouse for their invaluable technical assistance with sequencing and Marta De Torres Zabala for expert guidance to Arthur Wasukira and Max Coulter in the laboratory.
Conceived the study: David J. Studholme, Murray Grant, Julian Smith, Arthur Wasukira and Jerome Kubiriba. Wrote the manuscript and prepared figures: David J. Studholme, Julian Smith, Murray Grant and Arthur Wasukira. Performed laboratory-based work: Arthur Wasukira, Max Coulter and Richard Thwaites. Supervised DNA sequencing: Konrad Paszkiewicz. Bioinformatics analysis: David J. Studholme and Noorah Al-Sowayeh.
Conflicts of Interest
The authors declare no conflict of interest.
- Hayward, A.C. The hosts of Xanthomonas. In Xanthomonas; Springer: Dordrecht, The Netherlands, 1993; pp. 1–119.
- Ryan, R.P.; Vorhölter, F.-J.; Potnis, N.; Jones, J.B.; van Sluys, M.-A.; Bogdanove, A.J.; Dow, J.M. Pathogenomics of Xanthomonas: understanding bacterium-plant interactions. Nat. Rev. Microbiol. 2011, 9, 344–355, doi:10.1038/nrmicro2558.
- Dookun, A.; Stead, D.E.; Autrey, L.J. Variation among strains of Xanthomonas campestris pv. vasculorum from Mauritius and other countries based on fatty acid analysis. Syst. Appl. Microbiol. 2000, 23, 148–155, doi:10.1016/S0723-2020(00)80056-9.
- Aritua, V.; Parkinson, N.; Thwaites, R.; Heeney, J.V.; Jones, D.R.; Tushemereirwe, W.; Crozier, J.; Reeder, R.; Stead, D.E.; Smith, J. Characterization of the Xanthomonas sp causing wilt of enset and banana and its proposed reclassification as a strain of X. vasicola. Plant Pathol. 2008, 57, 170–177.
- Odipio, J.; Tusiime, G.; Tripathi, L. Genetic homogeneity among Ugandan isolates of Xanthomonas campestris pv. musacearum revealed by randomly amplified polymorphic DNA analysis. Afr. J. Biotechnol. 2009, doi:10.4314/ajb.v8i21.66028.
- Aritua, V.; Nanyonjo, A.; Kumakech, F.; Tushemereirwe, W. Rep-PCR reveals a high genetic homogeneity among Ugandan isolates of Xanthomonas campestris pv musacearum. Afr J Biotechnol. 2007, 6, 179–183.
- Wasukira, A.; Tayebwa, J.; Thwaites, R.; Paszkiewicz, K.; Aritua, V.; Kubiriba, J.; Smith, J.; Grant, M.; Studholme, D.J. Genome-wide sequencing reveals two major sub-lineages in the genetically monomorphic pathogen Xanthomonas campestris pathovar musacearum. Genes. 2012, 3, 361–377, doi:10.3390/genes3030361.
- Studholme, D.J.; Kemen, E.; MacLean, D.; Schornack, S.; Aritua, V.; Thwaites, R.; Grant, M.; Smith, J.; Jones, J.D.G. Genome-wide sequencing data reveals virulence factors implicated in banana Xanthomonas wilt. FEMS Microbiol. Lett. 2010, 310, 182–192, doi:10.1111/j.1574-6968.2010.02065.x.
- Ivey, M.L.L.; Tusiime, G.; Miller, S.A. A Polymerase Chain Reaction Assay for the Detection of Xanthomonas campestris pv. musacearum in Banana. Plant Dis. 2010, 94, 109–114, doi:10.1094/PDIS-94-1-0109.
- Qhobela, M.; Claflin, L.E. Eastern and southern African strains of Xanthomonas campestris pv. vasculorum are distinguishable by restriction fragment length polymorphism of DNA and polyacrylamide gel electrophoresis of membrane proteins. Plant Pathol. 1992, 41, 113–121, doi:10.1111/j.1365-3059.1992.tb02327.x.
- Vauterin, L.; Hoste, B.; Kersters, K.; Swings, J. Reclassification of Xanthomonas. Int. J. Syst. Bacteriol. 1995, 472–489.
- Rademaker, J.L.W.; Louws, F.J.; Schultz, M.H.; Rossbach, U.; Vauterin, L.; Swings, J.; de Bruijn, F.J. A comprehensive species to strain taxonomic framework for Xanthomonas. Phytopathology 2005, 95, 1098–1111, doi:10.1094/PHYTO-95-1098.
- Salzberg, S.L.; Sommer, D.D.; Schatz, M.C.; Phillippy, A.M.; Rabinowicz, P.D.; Tsuge, S.; Furutani, A.; Ochiai, H.; Delcher, A.L.; Kelley, D.; et al. Genome sequence and rapid evolution of the rice pathogen Xanthomonas oryzae pv. oryzae PXO99A. BMC Genom. 2008, 9, 204, doi:10.1186/1471-2164-9-204.
- Li, H.; Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 2009, 25, 1754–1760, doi:10.1093/bioinformatics/btp324.
- Quinlan, A.R.; Hall, I.M. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 2010, 26, 841–842, doi:10.1093/bioinformatics/btq033.
- Bart, R.; Cohn, M.; Kassen, A.; McCallum, E.J.; Shybut, M.; Petriello, A.; Krasileva, K.; Dahlbeck, D.; Medina, C.; Alicai, T.; et al. High-throughput genomic sequencing of cassava bacterial blight strains identifies conserved effectors to target for durable resistance. Proc. Natl. Acad. Sci. USA. 2012, 109, E1972–E1979, doi:10.1073/pnas.1208003109.
- Cunnac, S.; Bolot, S.; Serna, N.F.; Ortiz, E.; Szurek, B.; Noël, L.D.; Arlat, M.; Jacques, M.-A.; Gagnevin, L.; Carrere, S.; et al. High-quality draft genome sequences of two Xanthomonas citri pv. malvacearum strains. Genome Announc. 2013, doi:10.1128/genomeA.00674-13.
- Potnis, N.; Krasileva, K.; Chow, V.; Almeida, N.F.; Patil, P.B.; Ryan, R.P.; Sharlach, M.; Behlau, F.; Dow, J.M.; Momol, M.T.; et al. Comparative genomics reveals diversity among xanthomonads infecting tomato and pepper. BMC Genom. 2011, 12, 146, doi:10.1186/1471-2164-12-146.
- Taguchi, F.; Ichinose, Y. Role of type IV pili in virulence of Pseudomonas syringae pv. tabaci 6605: Correlation of motility, multidrug resistance, and HR-inducing activity on a nonhost plant. Mol. Plant. Microbe. Interact. 2011, 24, 1001–1011, doi:10.1094/MPMI-02-11-0026.
- Liu, H.; Kang, Y.; Genin, S.; Schell, M.A.; Denny, T.P. Twitching motility of Ralstonia solanacearum requires a type IV pilus system. Microbiology 2001, 147, 3215–3229.
- Mattick, J.S. Type IV pili and twitching motility. Annu. Rev. Microbiol. 2002, 56, 289–314, doi:10.1146/annurev.micro.56.012302.160938.
- De La Fuente, L.; Burr, T.J.; Hoch, H.C. Mutations in type I and type IV pilus biosynthetic genes affect twitching motility rates in Xylella fastidiosa. J. Bacteriol. 2007, 189, 7507–7510, doi:10.1128/JB.00934-07.
- Li, Y.; Hao, G.; Galvani, C.D.; Meng, Y.; De La Fuente, L.; Hoch, H.C.; Burr, T.J. Type I and type IV pili of Xylella fastidiosa affect twitching motility, biofilm formation and cell-cell aggregation. Microbiology 2007, 153, 719–726, doi:10.1099/mic.0.2006/002311-0.
- Li, Y.-Q.; Wan, D.-S.; Huang, S.-S.; Leng, F.-F.; Yan, L.; Ni, Y.-Q.; Li, H.-Y. Type IV pili of Acidithiobacillus ferrooxidans are necessary for sliding, twitching motility, and adherence. Curr. Microbiol. 2010, 60, 17–24, doi:10.1007/s00284-009-9494-8.
- Pelicic, V. Type IV pili: E pluribus unum? Mol. Microbiol. 2008, 68, 827–837, doi:10.1111/j.1365-2958.2008.06197.x.
- Burdman, S.; Bahar, O.; Parker, J.K.; De La Fuente, L. Involvement of type IV Pili in pathogenicity of plant pathogenic bacteria. Genes. 2011, 2, 706–735, doi:10.3390/genes2040706.
- Jenkins, A.T.A.; Buckling, A.; McGhee, M.; ffrench-Constant, R.H. Surface plasmon resonance shows that type IV pili are important in surface attachment by Pseudomonas aeruginosa. J. R. Soc. Interface 2005, 2, 255–259, doi:10.1098/rsif.2005.0030.
- Heijstra, B.D.; Pichler, F.B.; Liang, Q.; Blaza, R.G.; Turner, S.J. Extracellular DNA and Type IV pili mediate surface attachment by Acidovorax temperans. Antonie Van Leeuwenhoek 2009, 95, 343–349, doi:10.1007/s10482-009-9320-0.
- Roine, E.; Raineri, D.M.; Romantschuk, M.; Wilson, M.; Nunn, D.N. Characterization of type IV pilus genes in Pseudomonas syringae pv. tomato DC3000. Mol. Plant. Microbe. Interact. 1998, 11, 1048–1056, doi:10.1094/MPMI.1922.214.171.1248.
- Shime-Hattori, A.; Iida, T.; Arita, M.; Park, K.-S.; Kodama, T.; Honda, T. Two type IV pili of Vibrio parahaemolyticus play different roles in biofilm formation. FEMS Microbiol. Lett. 2006, 264, 89–97, doi:10.1111/j.1574-6968.2006.00438.x.
- Darsonval, A.; Darrasse, A.; Meyer, D.; Demarty, M.; Durand, K.; Bureau, C.; Manceau, C.; Jacques, M. The Type III secretion system of Xanthomonas fuscans subsp. fuscans is involved in the phyllosphere colonization process and in transmission to seeds of susceptible beans. Appl. Environ. Microbiol. 2008, 74, 2669–2678, doi:10.1128/AEM.02906-07.
- Varga, J.J.; Therit, B.; Melville, S.B. Type IV pili and the CcpA protein are needed for maximal biofilm formation by the gram-positive anaerobic pathogen Clostridium perfringens. Infect. Immun. 2008, 76, 4944–4951, doi:10.1128/IAI.00692-08.
- Mhedbi-Hajri, N.; Darrasse, A.; Pigné, S.; Durand, K.; Fouteau, S.; Barbe, V.; Manceau, C.; Lemaire, C.; Jacques, M.-A. Sensing and adhesion are adaptive functions in the plant pathogenic xanthomonads. BMC Evol. Biol. 2011, 11, 67, doi:10.1186/1471-2148-11-67.
- Heo, Y.-J.; Chung, I.-Y.; Choi, K.B.; Lau, G.W.; Cho, Y.-H. Genome sequence comparison and superinfection between two related Pseudomonas aeruginosa phages, D3112 and MP22. Microbiology 2007, 153, 2885–2895, doi:10.1099/mic.0.2007/007260-0.
- Lerouge, I.; Vanderleyden, J. O-antigen structural variation: Mechanisms and possible roles in animal/plant-microbe interactions. FEMS Microbiol. Rev. 2002, 26, 17–47, doi:10.1111/j.1574-6976.2002.tb00597.x.
- Raetz, C.R.H.; Whitfield, C. Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 2002, 71, 635–700, doi:10.1146/annurev.biochem.71.110601.135414.
- Hendrick, C.A.; Sequeira, L. Lipopolysaccharide-DEFECTIVE Mutants of the Wilt Pathogen Pseudomonas solanacearum. Appl. Environ. Microbiol. 1984, 48, 94–101.
- Drigues, P.; Demery-Lafforgue, D.; Trigalet, A.; Dupin, P.; Samain, D.; Asselineau, J. Comparative studies of lipopolysaccharide and exopolysaccharide from a virulent strain of Pseudomonas solanacearum and from three avirulent mutants. J. Bacteriol. 1985, 162, 504–509.
- Jayaswal, R.K.; Bressan, R.A.; Handa, A.K. Effects of a mutation that eliminates UDP glucose-pyrophosphorylase on the pathogenicity of Erwinia carotovora subsp. carotovora. J. Bacteriol. 1985, 164, 473–476.
- Schoonejans, E.; Expert, D.; Toussaint, A. Characterization and virulence properties of Erwinia chrysanthemi lipopolysaccharide-defective, phi EC2-resistant mutants. J. Bacteriol. 1987, 169, 4011–4017.
- Kingsley, M.T.; Gabriel, D.W.; Marlow, G.C.; Roberts, P.D. The opsX locus of Xanthomonas campestris affects host range and biosynthesis of lipopolysaccharide and extracellular polysaccharide. J. Bacteriol. 1993, 175, 5839–5850.
- Dow, J.M.; Osbourn, A.E.; Wilson, T.J.; Daniels, M.J. A locus determining pathogenicity of Xanthomonas campestris is involved in lipopolysaccharide biosynthesis. Mol. Plant. Microbe. Interact. 1995, 8, 768–777.
- Titarenko, E.; López-Solanilla, E.; García-Olmedo, F.; Rodríguez-Palenzuela, P. Mutants of Ralstonia (Pseudomonas) solanacearum sensitive to antimicrobial peptides are altered in their lipopolysaccharide structure and are avirulent in tobacco. J. Bacteriol. 1997, 179, 6699–6704.
- Li, J.; Wang, N. The gpsX gene encoding a glycosyltransferase is important for polysaccharide production and required for full virulence in Xanthomonas citri subsp. citri. BMC Microbiol. 2012, 12, 31, doi:10.1186/1471-2180-12-31.
- Yan, Q.; Hu, X.; Wang, N. The novel virulence-related gene nlxA in the lipopolysaccharide cluster of Xanthomonas citri ssp. citri is involved in the production of lipopolysaccharide and extracellular polysaccharide, motility, biofilm formation and stress resistance. Mol. Plant Pathol. 2012, 13, 923–934, doi:10.1111/j.1364-3703.2012.00800.x.
- Li, J.; Wang, N. The wxacO gene of Xanthomonas citri ssp. citri encodes a protein with a role in lipopolysaccharide biosynthesis, biofilm formation, stress tolerance and virulence. Mol. Plant Pathol. 2011, 12, 381–396, doi:10.1111/j.1364-3703.2010.00681.x.
- Patil, P.B.; Bogdanove, A.J.; Sonti, R.V. The role of horizontal transfer in the evolution of a highly variable lipopolysaccharide biosynthesis locus in xanthomonads that infect rice, citrus and crucifers. BMC Evol Biol 2007, 7, 243, doi:10.1186/1471-2148-7-243.
- Patil, P.; Sonti, R. Variation suggestive of horizontal gene transfer at a lipopolysaccharide (lps) biosynthetic locus in Xanthomonas oryzae pv. oryzae, the bacterial leaf blight pathogen of rice. BMC Microbiol. 2004, 4, 40, doi:10.1186/1471-2180-4-40.
- Lu, H.; Patil, P.; van Sluys, M.-A.; White, F.F.; Ryan, R.P.; Dow, J.M.; Rabinowicz, P.; Salzberg, S.L.; Leach, J.E.; Sonti, R.; et al. Acquisition and evolution of plant pathogenesis-associated gene clusters and candidate determinants of tissue-specificity in Xanthomonas. PLoS One 2008, 3, e3828, doi:10.1371/journal.pone.0003828.
- Da Silva, A.C.R.; Ferro, J.A.; Reinach, F.C.; Farah, C.S.; Furlan, L.R.; Quaggio, R.B.; Monteiro-Vitorello, C.B.; van Sluys, M.A.; Almeida, N.F.; Alves, L.M.C.; et al. Comparison of the genomes of two Xanthomonas pathogens with differing host specificities. Nature 2002, 417, 459–463, doi:10.1038/417459a.
- Pieretti, I.; Royer, M.; Barbe, V.; Carrere, S.; Koebnik, R.; Cociancich, S.; Couloux, A.; Darrasse, A.; Gouzy, J.; Jacques, M.-A. The complete genome sequence of Xanthomonas albilineans provides new insights into the reductive genome evolution of the xylem-limited Xanthomonadaceae. BMC Genom. 2009, 10, 616, doi:10.1186/1471-2164-10-616.
- Pieretti, I.; Royer, M.; Barbe, V.; Carrere, S.; Koebnik, R.; Couloux, A.; Darrasse, A.; Gouzy, J.; Jacques, M.-A.; Lauber, E.; et al. Genomic insights into strategies used by Xanthomonas albilineans with its reduced artillery to spread within sugarcane xylem vessels. BMC Genom. 2012, 13, 658, doi:10.1186/1471-2164-13-658.
- Studholme, D.J.; Wasukira, A.; Paszkiewicz, K.; Aritua, V.; Thwaites, R.; Smith, J.; Grant, M. Correction: Studholme et al. Draft Genome Sequences of Xanthomonas sacchari and Two Banana-Associated Xanthomonads Reveal Insights into the Xanthomonas Group 1 clade. Genes 2011, 2, 1050–1065, doi:10.3390/genes2041050.
- Dow, M.; Newman, M.-A.; von Roepenack, E. The induction and modulation of plant defense responses by bacterial lipopolysaccharides. Annu. Rev. Phytopathol. 2000, 38, 241–261, doi:10.1146/annurev.phyto.38.1.241.
- Meyer, A.; Pühler, A.; Niehaus, K. The lipopolysaccharides of the phytopathogen Xanthomonas campestris pv. campestris induce an oxidative burst reaction in cell cultures of Nicotiana tabacum. Planta 2001, 213, 214–222, doi:10.1007/s004250000493.
- Keshavarzi, M.; Soylu, S.; Brown, I.; Bonas, U.; Nicole, M.; Rossiter, J.; Mansfield, J. Basal defenses induced in pepper by lipopolysaccharides are suppressed by Xanthomonas campestris pv. vesicatoria. Mol. Plant. Microbe. Interact. 2004, 17, 805–815, doi:10.1094/MPMI.2004.17.7.805.
- Yang, Y.-C.; Chou, C.-P.; Kuo, T.-T.; Lin, S.-H.; Yang, M.-K. PilR enhances the sensitivity of Xanthomonas axonopodis pv. citri to the infection of filamentous bacteriophage Cf. Curr. Microbiol. 2004, 48, 251–261, doi:10.1007/s00284-003-4191-5.
- Pal, S.; Wu, L.P. Pattern recognition receptors in the fly: Lessons we can learn from the Drosophila melanogaster immune system. Fly 2009, 3, 121–129.
- Sarkar, S.F.; Gordon, J.S.; Martin, G.B.; Guttman, D.S. Comparative genomics of host-specific virulence in Pseudomonas syringae. Genetics 2006, 174, 1041–1056, doi:10.1534/genetics.106.060996.
- Hajri, A.; Brin, C.; Hunault, G.; Lardeux, F.; Lemaire, C.; Manceau, C.; Boureau, T.; Poussier, S. A “repertoire for repertoire” hypothesis: Repertoires of type three effectors are candidate determinants of host specificity in Xanthomonas. PLoS One 2009, 4, e6632.
- Hajri, A.; Pothier, J.F.; Fischer-Le Saux, M.; Bonneau, S.; Poussier, S.; Boureau, T.; Duffy, B.; Manceau, C. Type three effector genes distribution and sequence analysis provides new insights into pathogenicity of plant pathogenic Xanthomonas arboricola. Appl. Environ. Microbiol. 2012, 78, 371–384, doi:10.1128/AEM.06119-11.
- Hajri, A.; Brin, C.; Zhao, S.; David, P.; Feng, J.-X.; Koebnik, R.; Szurek, B.; Verdier, V.; Boureau, T.; Poussier, S. Multilocus sequence analysis and type III effector repertoire mining provide new insights into the evolutionary history and virulence of Xanthomonas oryzae. Mol. Plant Pathol. 2012, 13, 288–302, doi:10.1111/j.1364-3703.2011.00745.x.
- Astua-Monge, G.; Minsavage, G.V.; Stall, R.E.; Davis, M.J.; Bonas, U.; Jones, J.B. Resistance of tomato and pepper to T3 strains of Xanthomonas campestris pv. vesicatoria is specified by a plant-inducible avirulence gene. Mol. Plant. Microbe. Interact. 2000, 13, 911–921, doi:10.1094/MPMI.2000.13.9.911.
- Jalan, N.; Kumar, D.; Andrade, M.O.; Yu, F.; Jones, J.B.; Graham, J.H.; White, F.F.; Setubal, J.C.; Wang, N. Comparative genomic and transcriptome analyses of pathotypes of Xanthomonas citri subsp. citri provide insights into mechanisms of bacterial virulence and host range. BMC Genom. 2013, 14, 551, doi:10.1186/1471-2164-14-551.
- Thieme, F.; Koebnik, R.; Bekel, T.; Berger, C.; Boch, J.; Büttner, D.; Caldana, C.; Gaigalat, L.; Goesmann, A.; Kay, S.; et al. Insights into genome plasticity and pathogenicity of the plant pathogenic bacterium Xanthomonas campestris pv. vesicatoria revealed by the complete genome sequence. J. Bacteriol. 2005, 187, 7254–7266, doi:10.1128/JB.187.21.7254-7266.2005.
- Singer, A.U.; Schulze, S.; Skarina, T.; Xu, X.; Cui, H.; Eschen-Lippold, L.; Egler, M.; Srikumar, T.; Raught, B.; Lee, J.; Scheel, D.; Savchenko, A.; Bonas, U. A pathogen type III effector with a novel E3 ubiquitin ligase architecture. PLoS Pathog. 2013, 9, e1003121, doi:10.1371/journal.ppat.1003121.
- Jiang, W.; Jiang, B.-L.; Xu, R.-Q.; Huang, J.-D.; Wei, H.-Y.; Jiang, G.-F.; Cen, W.-J.; Liu, J.; Ge, Y.-Y.; Li, G.-H.; et al. Identification of six type III effector genes with the PIP box in Xanthomonas campestris pv. campestris and five of them contribute individually to full pathogenicity. Mol. Plant. Microbe. Interact. 2009, 22, 1401–1411, doi:10.1094/MPMI-22-11-1401.
- Büttner, D.; Bonas, U. Regulation and secretion of Xanthomonas virulence factors. FEMS Microbiol. Rev. 2010, 34, 107–133, doi:10.1111/j.1574-6976.2009.00192.x.
- Koebnik, R.; Krüger, A.; Thieme, F.; Urban, A.; Bonas, U. Specific binding of the Xanthomonas campestris pv. vesicatoria AraC-type transcriptional activator HrpX to plant-inducible promoter boxes. J. Bacteriol. 2006, 188, 7652–7660, doi:10.1128/JB.00795-06.
- Wengelnik, K.; Bonas, U. HrpXv, an AraC-type regulator, activates expression of five of the six loci in the hrp cluster of Xanthomonas campestris pv. vesicatoria. J. Bacteriol. 1996, 178, 3462–3469.
- Robinson, J.T.; Thorvaldsdóttir, H.; Winckler, W.; Guttman, M.; Lander, E.S.; Getz, G.; Mesirov, J.P. Integrative genomics viewer. Nat. Biotechnol. 2011, 29, 24–26, doi:10.1038/nbt.1754.
- Li, H.; Handsaker, B.; Wysoker, A.; Fennell, T.; Ruan, J.; Homer, N.; Marth, G.; Abecasis, G.; Durbin, R.; 1000 Genome Project Data Processing Subgroup. The Sequence Alignment/Map format and SAMtools. Bioinformatics 2009, 25, 2078–2079, doi:10.1093/bioinformatics/btp352.
- Tamura, K.; Peterson, D.; Peterson, N.; Stecher, G.; Nei, M.; Kumar, S. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 2011, 28, 2731–2739, doi:10.1093/molbev/msr121.
- Zerbino, D.R.; Birney, E. Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. 2008, 18, 821–829, doi:10.1101/gr.074492.107.
- Aziz, R.K.; Bartels, D.; Best, A.A.; DeJongh, M.; Disz, T.; Edwards, R.A.; Formsma, K.; Gerdes, S.; Glass, E.M.; Kubal, M.; et al. The RAST Server: rapid annotations using subsystems technology. BMC Genom. 2008, 9, 75, doi:10.1186/1471-2164-9-75.
- R Development Core Team. R: A Language and Environment for Statistical Computing. R Found. Stat. Comput. 2013, 1, 409.
- Altschul, S.F.; Gish, W.; Miller, W.; Myers, E.W.; Lipman, D.J. Basic local alignment search tool. J. Mol. Biol. 1990, 215, 403–410.
- Carver, T.J.; Rutherford, K.M.; Berriman, M.; Rajandream, M.-A.; Barrell, B.G.; Parkhill, J. ACT: The artemis comparison tool. Bioinformatics 2005, 21, 3422–3423, doi:10.1093/bioinformatics/bti553.
- Alikhan, N.-F.; Petty, N.K.; Ben Zakour, N.L.; Beatson, S.A. BLAST Ring Image Generator (BRIG): Simple prokaryote genome comparisons. BMC Genom. 2011, 12, 402, doi:10.1186/1471-2164-12-402.
- Stothard, P.; Wishart, D.S. Circular genome visualization and exploration using CGView. Bioinformatics 2005, 21, 537–539, doi:10.1093/bioinformatics/bti054.
- Eddy, S.R. Profile Hidden Markov Models. Bioinformatics 1998, 14, 755–763, doi:10.1093/bioinformatics/14.9.755.
- stün, S.; Bartetzko, V.; Börnke, F. The Xanthomonas campestris Type III Effector XopJ Targets the Host Cell Proteasome to Suppress Salicylic-Acid Mediated Plant Defence. PLoS Pathog. 2013, 9, e1003427, doi:10.1371/journal.ppat.1003427.
- Qhobela, M.; Leach, J.E. Characterisation of strains of Xanthomonas campestris pv. holcicola by PAGE of membrane proteins and by REA and RFLP analysis of genomic DNA. Plant Dis. 1991, 75, 32–36, doi:10.1094/PD-75-0032.
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).