ARTICLE IN PRESS
Systematic and Applied Microbiology 32 (2009) 549–557 www.elsevier.de/syapm
Proposal of Xanthomonas translucens pv. pistaciae pv. nov., pathogenic to pistachio (Pistacia vera) Danie`le Giblot-Ducraya,, Alireza Marefata,1, Michael R. Gillingsb, Neil M. Parkinsonc, John P. Bowmand, Kathy Ophel-Kellere, Cathy Taylorf, Evelina Facellia, Eileen S. Scotta a
The University of Adelaide, School of Agriculture, Food and Wine, Adelaide, South Australia 5005, Australia Macquarie University, Department of Biological Sciences, Sydney, New South Wales 2019, Australia c Central Science Laboratory, Sand Hutton, York YO41 1LZ, United Kingdom d University of Tasmania, School of Agricultural Science, Hobart, Tasmania 7001, Australia e South Australian Research and Development Institute, Adelaide, South Australia 5001, Australia f Department of Primary Industries, Mildura, Victoria 3502, Australia b
Received 9 March 2009
Abstract Strains of Xanthomonas translucens have caused dieback in the Australian pistachio industry for the last 15 years. Such pathogenicity to a dicotyledonous woody host contrasts with that of other pathovars of X. translucens, which are characterized by their pathogenicity to monocotyledonous plant families. Further investigations, using DNA-DNA hybridization, gyrB gene sequencing and integron screening, were conducted to confirm the taxonomic status of the X. translucens pathogenic to pistachio. DNA-DNA hybridization provided a clear classification, at the species level, of the pistachio pathogen as a X. translucens. In the gyrB-based phylogeny, strains of the pistachio pathogen clustered among the X. translucens pathovars as two distinct lineages. Integron screening revealed that the cassette arrays of strains of the pistachio pathogen were different from those of other Xanthomonas species, and again distinguished two groups. Together with previously reported pathogenicity data, these results confirm that the pistachio pathogen is a new pathovar of X. translucens and allow hypotheses about its origin. The proposed name is Xanthomonas translucens pv. pistaciae pv. nov. & 2009 Elsevier B.V. All rights reserved. Keywords: Pistachio; Phylogeny; Pathogenicity
Introduction Corresponding author at: The University of Adelaide, School of
Agriculture, Food and Wine, Waite Road, Urrbrae, SA 5064, Australia. Tel.: +61 8 83037495; fax: +61 8 83037109. E-mail address:
[email protected] (D. Giblot-Ducray). 1 Current address: Department of Plant Protection, Zanjan University, Zanjan, Iran. 0723-2020/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.syapm.2009.08.001
For the past 15 years, Australian pistachio (Pistacia vera) orchards have been affected by dieback, a disease characterized by trunk and limb lesions, discoloration of mature xylem and shoot dieback. Affected trees gradually decline, fail to produce marketable nuts and eventually die, making this disease of serious economic concern. A bacterium belonging to the genus
ARTICLE IN PRESS 550
D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
Xanthomonas was identified as the causal agent of the disease [4,5]. Subsequent phenotypic and genotypic characterisation, including biochemical and physiological tests, SDS-PAGE of whole-cell proteins, comparison of 16S ribosomal DNA sequences and rep-PCR, showed that the pathogen was related to X. translucens and also revealed genetic heterogeneity. Two distinct groups of strains, named A and B, were identified, with group A being further divided into 4 subgroups by rep-PCR [5,13]. Most species in the genus Xanthomonas are characterized at the infrasubspecific level into pathovars, based on distinctive pathogenicity. The species X. translucens consists of 10 recognized pathovars, namely X. translucens pv. translucens, 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 and X. translucens pv. undolosa, all originally described based on distinctive pathogenicity to plants in the family Poaceae [23]. Recently, Rademaker et al. [18] found X. translucens pv. undulosa in ornamental asparagus (Liliaceae), extending the host range of X. translucens to include another monocotyledonous family. In contrast, both A and B strains of the X. translucens isolated from pistachio naturally infect cultivated pistachio, a member of the family Anacardiaceae. Their artificial host range also includes several other species of Anacardiaceae, with some differences between the 2 groups, whereas, using the same conditions, X. translucens pv. translucens does not infect pistachio nor any of the other species of Anacardiaceae [13]. Following infiltration and moist incubation, the 2 groups of strains of X. translucens isolated from pistachio also induced lesions on several species of Poaceae, including the major hosts of the X. translucens pathovars. However, they have not been detected in grasses growing in pistachio orchards [5,13]. Several Xanthomonas species have been described as comprising distinct populations that are pathogenic to diverse and unrelated hosts. For example, Xanthomonas strains known to cause diseases of mango and cashew, 2 species of Anacardiaceae, were recently classified as pathovars of the species Xanthomonas citri [1]. Nevertheless, the identification of strains related to X. translucens that were pathogenic to a woody host and in a family outside the monocotyledons was novel and warranted further investigation to confirm their taxonomic status. The genus Xanthomonas has been the subject of numerous taxonomic and phylogenetic studies. Initially, classification was mainly based on a single phenotypic attribute, that of host specificity. A major reclassification was proposed by Vauterin et al. [23] using DNA–DNA hybridization as a criterion to delineate Xanthomonas species. From that study, the minimum
DNA–DNA homology value to assign strains to the same species is 70%, whereas strains assigned to different species show homology values averaging 40–50%. Despite its limitations, DNA–DNA hybridization is an essential component of bacterial species description and, according to Stackebrand et al. [20], remains the acknowledged standard for species delineation. Since then, other techniques have been tested to assess relationships between and/or within Xanthomonas species, including comparison of 16S ribosomal DNA sequences and rep-PCR [8,17]. Recently, the potential of sequence typing for classification of Xanthomonas was investigated. A multilocus sequence analysis (MLSA), based on dnaK, kyuA, gyrB and rpoD genes, allowed differentiation of most species previously established by DNA–DNA hybridization [25]. Furthermore, Parkinson et al. [14,15] showed that comparison of partial gyrB gene sequences alone provided an efficient identification tool, at both specific and subspecific levels, and in good agreement with DNA–DNA hybridization. Integrons are genetic features that allow their bacterial hosts to acquire and assemble new genes as gene cassettes. By mapping the integrons and gene cassettes of 32 Xanthomonas strains, encompassing 2 species and 12 pathovars, Gillings et al. [7] showed that integrons have contributed to the genome diversity among Xanthomonas. A direct link between integron diversity and the specialization of Xanthomonas into pathovars was not established, but that study showed that integron cassette arrays can be used to assess relationships among Xanthomonas taxa at species and subspecies levels. In this study, we conducted DNA-DNA hybridization analysis to definitely establish the species of the pistachio pathogen. Then, we sequenced the gyrB locus [14] from A and B strains to determine the relatedness between these 2 groups, the X. translucens pathovars and the other Xanthomonas species. Additionally, we used PCR and sequence analysis to screen A and B strains for integron content and compare them with other Xanthomonas. The results confirmed the identification of the pistachio pathogens as an X. translucens and, together with previously reported pathogenicity data [13], support the proposal for a new pathovar, Xanthomonas translucens pv. pistaciae pv. nov.
Material and methods Strains The reference strains of groups A (ICMP 16316) and B (ICMP 16317), previously characterized among X. translucens infecting pistachio [13], were used in all experiments. All other strains and their respective uses
ARTICLE IN PRESS D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
are listed in Table 1. Freeze-dried cultures were revived and subcultured at 28 1C on Nutrient Agar or Sucrose Peptone Agar [5] for subsequent use.
DNA–DNA hybridization DNA–DNA hybridization was conducted with the 2 reference strains of the pistachio pathogen, ICMP 16316 and ICMP 16317, and with the pathotype strains of 3 X. translucens pathovars, namely X. translucens pv. translucens LMG 876, X. translucens pv. poae LMG 728 and X. translucens pv. graminis LMG 726 (Table 1). The type strains of X. theicola LMG 8764 and X. hyacinthi LMG 739 were included as outgroups (Table 1). DNA was Table 1.
extracted as described previously [13], using the protocol of Rademaker and de Bruijn [16]. DNA concentration was determined spectrophotometrically and adjusted to 50 ng/ml before use. The percentage of DNA homology among the strains was determined using the spectrophotometric renaturation rate kinetic procedure [9]. Briefly, after shearing and denaturation of the DNA, hybridization was performed at the optimal temperature for renaturation. Decline in absorbance of DNA mixtures over a 40-min interval was used to calculate DNA hybridization value. Hybridization between the reference strains of the pistachio pathogen was performed 4 times, while hybridization with other reference and type strains was performed 2 or 3 times.
Xanthomonas strains used in this study. Placea and year of collection
Pistachio isolates Name
Group/subgroup
ICMPe 16316 DARf 75532 A20 A76 A5 A28 A34 A2 A3 A78 A15 A29 A77 A431 A451 A457 ICMPe 16317 VPRIh 21750a B54 B63
A1 A1 A1 A1 A2 A2 A2 A3 A3 A3 A4 A4 A4 Ag Ag Ag B B B B
Other Xanthomonas X. translucens pv. translucens DARf 35705 X. translucens pv. translucens LMGi 876 X. oryzae pv. oryzae DARf 61713 X. theicola LMGi 8684 X. hyacinthi LMGi 739 a
Kyalite, NSW, 2000 Kyalite, NSW, 2000 Kyalite, NSW, 2000 Kyalite, NSW, 2003 Kyalite, NSW, 2000 Renmark, SA, 2000 Renmark, SA, 2000 Kyalite, NSW, 2000 Kyalite, NSW, 2000 Kyalite, NSW, 2003 Red Cliffs, Vic., 2000 Renmark, SA, 2000 Kyalite, NSW, 2003 Robinvale, Vic., 2006 Robinvale, Vic., 2006 Robinvale, Vic., 2006 Robinvale, Vic., 2000 Robinvale, Vic., 2000 Robinvale, Vic., 2003 Robinvale, Vic., 2003
Used for DNAb
gyrBc
Integrond
x
x x
x
x
Tamworth, NSW, 1981 USA, 1979
x x
x x x x x x x x x x x x x x x x x
x
x
x
Queensland, 1987
x
Japan, 1974 Netherlands, 1958
x x
NSW: New South Wales; SA: South Australia; Vic.: Victoria. DNA–DNA hybridization. c gyrB sequence genbank accession numbers FJ794272–FJ794275 and GQ387665. d Integron sequence genbank accession numbers FJ797422–FJ797424. e Source ICMP: International Collection of Micro-organisms from Plants, Auckland, New Zealand. f Source ACPPB: Australian Collection of Plant Pathogenic Bacteria, Orange, Australia. g Strains identified to group but not to subgroup level. h Source VPRI: Victorian Plant Pathology Herbarium, Knoxfield, Australia. i Source BCCM: Belgium Coordinated Collection of Microorganisms, University of Ghent, Belgium. b
551
ARTICLE IN PRESS 552
D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
gyrB sequence and phylogenetic analysis The gyrB gene sequence was determined for the reference strains of the pistachio pathogen, ICMP 16316 and ICMP 16317, plus 1 other strain from each group. The gyrB gene was also sequenced for the isolate DAR 35705 of X. translucens pv. translucens, which was used as a reference in pathogenicity tests [13] and integron screening (Table 1). DNA extraction, PCR and sequence analysis of the gyrB gene were performed as described by Parkinson et al. [14]. Primers based on X. euvesicatoria gyrB gene and previously shown to amplify gyrB from a large number of Xanthomonas species and pathovars [14,15] were used to amplify the gyrB gene of the pistachio pathogen. Purity and yield of PCR were checked and PCR products were purified using a Wizard PCR Clean-up kit (Promega) prior to sequencing. Sequences were cropped to 530nt and aligned, using the program Clustal W [21] (http://www.ebi.ac.uk/ clustalw), with the sequences of 35 type and pathotype strains of Xanthomonas published by Parkinson et al. [15] (Table S1), which included 9 of the 10 pathovars of X. translucens. Sequence alignments are provided as Supplementary data. The phylogeny was established using the program TREECON [22] and rooted with the gyrB sequence of Stenotrophomonas sp. [14].
Screening for integron presence and cassette array diversity To examine further the pistachio pathogen, taking into account the genetic heterogeneity described previously [13], 18 strains were screened for integrons (Table 1). There were 15 strains from group A, comprising 3 strains from each subgroup defined by rep-PCR (namely A1, including ICMP 16316, A2, A3 and A4) plus 3 isolated in 2006 and identified to group but not to subgroup level. There were 3 strains from group B, including ICMP 16317. These 18 pistachio strains were compared with X. translucens pv. translucens (DAR 37705) and X. oryzae pv. oryzae (DAR 61713). DNA was extracted using an SDS/phenol extraction protocol [6]. The independence of each strain was first confirmed using rep-PCR as described by Gillings et al. [7]. Typical integrons comprise a gene for a DNA integrase (intI) and a specific integration site (attI) followed by an array of acquired genes, each separated by a conserved 59-base element (59-be) (Fig. S1). Based on this common organization, Gillings et al. [7] designed PCR protocols to study integrons of the genus Xanthomonas. These protocols were used to screen the pistachio strains. An initial PCR with primers AJH72 and MRG17 was conducted, followed by sequencing, to check for the presence of intI and attI and to ascertain the presence of an integron (Fig. S1).
Cassette arrays were then analyzed using primers MRG18 and AJH60, which target attI and 59-be, respectively (Fig. S1). Primer sequences and PCR conditions were as described by Gillings et al. [7].
Results DNA–DNA hybridization DNA–DNA hybridization between the reference strains of the pistachio pathogen was 84% (Table 2). The DNA homology of the pistachio strains with the pathotype strains of 3 X. translucens pathovars was also above the cut-off value of 70%. The strain from group A had the highest homology with X. translucens pv. poae, with 84% hybridization, whereas the strain from group B had the highest homology with X. translucens pv. graminis, with 90% hybridization (Table 2). In contrast, hybridization values of both pistachio strains with X. theicola and X. hyacinthi averaged no more than 54% (Table 2).
gyrB phylogeny A fragment corresponding to 530 nucleotides of the gyrB gene was amplified for the 4 strains of the pistachio pathogen and for X. translucens pv. translucens DAR 35705. In sequence alignments, X. translucens pv. translucens DAR 35705 revealed 100% similarity with the type strain of X. translucens pv. translucens, NCPPB 973. Strains from groups A and B of the pistachio Table 2. DNA–DNA hybridization analysis between the reference strains of X. translucens pathogen to pistachio and the type strains of selected Xanthomonas species and pathovars. Strain
DNA-binding valuea with pistachio pathogen ICMP 16316
X. translucens-pistachio pathogen group A (ICMP 16316) X. translucens-pistachio pathogen group B (ICMP 16317) X. translucens pv. translucens (LMG 876) X. translucens pv. poae (LMG 728) X. translucens pv. graminis (LMG 726) X. theicola (LMG 8684) X. hyacinthi (LMG 739)
ICMP 16317
100 8475
100
8177 84710 7874 4578 47
8675 7776 9078 54710 52
a Values are means of 2–4 hybridizations followed by standard deviation. For each group, only 1 hybridization was performed with X. hyacinthi DNA.
ARTICLE IN PRESS D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
553
X. perforans NCPPB 4321T (EU007533) X. euvesicatoria NCPPB 2968T (EU007517) X. alfalfae NCPPB 2062T (EU007542) X. melonis NCPPB 3434T (EU007531) X. axonopodis NCPPB 457T (EU007522) X. fuscans NCPPB 381T (EU007541) X. citri LMG 9322T (EU007540) X. bromi NCPPB 4343T (EU007523) X. vasicola NCPPB 2417T (DQ676938) X. oryzae NCPPB 3002T (EU007532) X. pisi NCPPB 762T (EU007520) X. vesicatoria NCPPB 422T (EU007519) X. arboricola NCPPB 411T (EU007516) X. fragariae NCPPB 1469T (EU007528) X. campestris NCPPB 528T (EU007524) X. populi NCPPB 2959T (EU007534) X. hortorum NCPPB 939T (EU007529) X. gardneri NCPPB 881T (EU007538) X. cynarae NCPPB 4356T (EU7527) X. cucurbitae NCPPB 2597T (EU007526) X. codiaei NCPPB 4350T (EU007537) X. cassavae NCPPB 101T (EU007525) X. sacchari NCPPB 4341T (EU007535) X. theicola NCPPB 4353T (EU007539) X. hyacinthi NCPPB 599T (EU007530) X.t. graminis NCPPB 2700 (EU285223) X.t. cerealis NCPPB 1944 (EU285238) X.t. pistaciae B ICMP 16317 (FJ794274) X.t. phlei NCPPB 3231 (EU285230) X.t. phleipratensis NCPPB 1837 (EU285221) X.t. arrhenatheri NCPPB 3229 (EU285240) X.t. poae NCPPB 3230 (EU285229) X.t. pistaciae A ICMP 16316 (FJ794272) X.t. translucens DAR 35705 (GQ387665) X.t. secalis NCPPB 2822 (EU285225) X.t. translucens NCPPB 973T (EU007536) X.t. undulosa NCPPB 2821 (EU285224) X. albilineans NCPPB 2969T (EU007521) Stenotrophomonas sp.
Fig. 1. Phylogenetic tree of the genus Xanthomonas based on the gyrB gene and highlighting the position of the 2 groups of X. translucens isolated from pistachio. The tree was constructed using the neighbor-joining method. The numbers are the percentages of occurrence in 500 bootstrapped values and only values 450% are shown. Strains are listed with their culture collection and GenBank accession numbers. Xanthomonas translucens pathovars are listed as X.t. followed by the pathovar names. T indicates type strains. Stenotrophomonas sp. was used as outgroup [14].
pathogen showed 96% homology, i.e. they shared 511 nucleotides of the 530 sequenced, whereas the sequences were identical within each group (Table S2). When compared with other Xanthomonas, the pistachio strains showed the highest similarity to pathovars of X. translucens, with percentages ranging from 95% to 99%, whereas the similarity to other Xanthomonas species and their pathovars ranged from 81% to 86% (Table S2). The 2 exceptions were X. theicola and X. hyacinthi, which showed similarity values of 92% and 93%, respectively, both with strains of groups A and B (Table S2). The phylogenetic tree constructed using the gyrB sequences of the pistachio pathogen, X. translucens pv.
translucens, DAR 35705, and 35 type and pathotype strains of Xanthomonas published in Parkinson et al. [14,15] showed that the strains of both groups A and B clustered in the X. translucens clade defined by Parkinson et al. [15] (Fig. 1). However, group A strains were closest to X. translucens pvs translucens, secalis and undulosa, whereas group B strains were closest to X. translucens pv. cerealis.
Integron screening When subjected to rep-PCR, all the pistachio strains generated patterns in accordance with their previous
ARTICLE IN PRESS 554
D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
intI
attI
AJH72 1 M
3 2
5 4
7 6
11
9 8
10
13 12
15 14
17 16
Xtt 18
MRG18
N
Xoo
M
All pvs in Gillings et al. [7] deletion
Pistachio group A Pistachio group B Xtt DAR 35705
IS1477 transposase
Fig. 2. Screening for integrons of the X. translucens isolated from pistachio. (A) Confirmation of the presence of the attI site and intI gene by PCR using primers AJH72 and MRG18. Lanes 1–15: X. translucens isolated from pistachio group A (1–3: subgroup A1, including ICMP 16316 in lane 2; 4–6: subgroup A2; 7–9: subgroup A3; 10–12: other group A isolates not characterized at the subgroup level). Lanes 16–18: X. translucens isolated from pistachio group B, including ICMP 16317 in lane 16. Lane Xtt: X. translucens pv. translucens DAR 35705. Lane Xoo: X. oryzae pv. oryzae DAR 61713. Lanes M: molecular weight marker. Lane N: negative control. (B) Schematic maps of the intI-attI region deduced from sequences of the PCR products obtained for the 2 groups of X. translucens isolated from pistachio (Pistachio group A and Pistachio group B) and for X. translucens pv. translucens (Xtt DAR 35705) and compared to the pathovars tested in Gillings et al. [7].
characterisation as group A or B [13] (data not shown). The PCR with primers AJH72 and MRG17 was expected to generate a 300 bp product [7]. A PCR product was observed for the 18 pistachio strains, as well as the X. translucens pv. translucens and X. oryzae pv. oryzae strains (Fig. 2a). For the pistachio group B strains and X. oryzae pv. oryzae, the product was of the expected size. In contrast, all group A strains yielded a 210 bp product and X. translucens pv. translucens yielded 2 bands, one of the expected size and the other about 1500 bp. The PCR products obtained from 1 strain of each group of the pistachio pathogen as well as the atypical amplicons of X. translucens pv. translucens were sequenced (Fig. 2b). For group A, the sequence showed a 95 bp deletion spanning the start of the integrase gene and a portion of the non-coding sequence between intI and attI. For group B, the PCR product had the sequence expected for this region. The dominant, larger product of X. translucens pv. translucens was shown to be due to the insertion of an IS1477 transposon into intI, whereas the fainter band at 300 bp appeared to be the typical intI/attI product caused by excision of the IS1477 transposon in a subset of the cells examined. The X. oryzae pv. oryzae product was sequenced in a previous study and shown to have the expected intI/attI sequence [7]. The PCR with primers MRG18 and AJH60 was expected to produce a complex pattern of bands varying with the proximal gene cassette [7]. In this PCR, the pistachio pathogen generated patterns different from X. translucens pv. translucens and X. oryzae pv. oryzae (Fig. S2). Strains belonging to group A shared a similar pattern and there was no difference between the
subgroups previously distinguished by rep-PCR [13]. Strains from group B shared a single pattern, which was different from that of group A strains (Fig. S2).
Discussion This research was conducted to clarify the taxonomic position of Xanthomonas strains pathogenic to pistachio, previously characterized as X. translucens [5,13]. In the International Standards for Naming Pathovars of Plant Pathogenic Bacteria, a pathovar is defined on the basis of distinctive pathogenicity to one or more plant hosts. To our knowledge, the natural host range of the X. translucens isolated from pistachio is limited to cultivated pistachio, P. vera, whereas other X. translucens pathovars were previously distinguished on the basis of their pathogenicity mainly among species of Poaceae [23]. When inoculated artificially, the pistachio strains were also able to infect several species of Anacardiaceae and Poaceae, including the main hosts of the different X. translucens pathovars [13]. However, it is not uncommon to find in the literature different pathovars of the same species showing overlapping host ranges [10], including among X. translucens pathovars [3]. Furthermore, the inability of a strain of X. translucens pv. translucens endemic to Australia to infect pistachio and other species of Anacardiaceae [13] suggested that the pathogenicity to these species was a distinctive trait of the strains isolated from pistachio. As containment requirements precluded the assessment of other X. translucens pathovars for their pathogenicity to
ARTICLE IN PRESS D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
pistachio, molecular characterisations were conducted here to confirm that the Xanthomonas strains pathogenic to pistachio represent a new pathovar of X. translucens, for which we propose the name Xanthomonas translucens pv. pistaciae pv. nov. For the past 30 years, DNA–DNA homology has formed the basis upon which to differentiate and classify bacteria at the species level [20]. It was used by Vauterin et al. [23] to establish a classification of the genus Xanthomonas, which remains a key reference to assess the efficiency of new genotyping approaches [2,8,14,17,25]. Previous genetic characterisation of X. translucens pv. pistaciae convincingly demonstrated its close relationship with X. translucens, but DNA–DNA homology values were lacking. DNA–DNA hybridization results clearly assigned X. translucens pv. pistaciae to the species X. translucens. DNA similarity between X. translucens pv. pistaciae and 3 of the X. translucens pathovars, including the type strain of the species, ranged from 77% to 90%, which is in agreement with Vauterin et al. [23], who described mean similarity values of 78% among X. translucens. Vauterin et al. [23] also identified X. theicola and X. hyacinthi as the closest species to X. translucens, with 48% and 51% homology, respectively. Likewise, the similarity values between X. translucens pv. pistaciae and these 2 species were approximately 50%. The 16S rDNA sequences also grouped X. translucens, X. theicola, and X. hyacinthi, along with X. albilineans, in a single cluster [8]. At the time, X. theicola, a pathogen of tea (Camellia sinensis), was the only species in that cluster to be pathogenic to a dicotyledonous host [8]. Interestingly, with homology of 84%, the 2 groups of strains of X. translucens pv. pistaciae did not show greater relatedness to one another than to the other 3 pathovars of X. translucens that were tested, in spite of their common pathogenicity to pistachio. The gyrB gene is universally distributed among species of bacteria and has proved useful for assessing relationships at the specific and infrasubspecific level in several species of bacteria [11,24]. For the genus Xanthomonas, gyrB-based phylogeny provides greater resolution than is possible based on the more conserved 16S rRNA or 16S–23S intergenic loci [14,15], the locus being discriminating enough to identify most of the pathovars within Xanthomonas species [15]. In accordance with previous phenotypic and genotypic characterisation [13], the representatives of the 2 groups of X. translucens pv. pistaciae were assigned to 2 distinct branches among X. translucens pathovars. Strains in group A were closest to X. translucens pv. secalis, X. translucens pv. undulosa and X. translucens pv. translucens, whereas strains in group B were closest to X. translucens pv. cerealis. This further confirms the close genetic relatedness between X. translucens pv. pistaciae strains in group B and X. translucens pv. cerealis previously reported by Marefat et al. [12], using
555
a PCR assay based on the 16S–23S intergenic spacer. In agreement with other analyses of the genus based on 16S rRNA [8] and on MLSA [25], X. translucens clustered in the most basal region of the tree along with X. hyacinthi, X. theicola, X. albilineans, and X. sacchari in the gyrBbased phylogeny [15]. As previously noted [8,15], with the exception of X. theicola, these species are pathogens of monocotyledonous plants. In this context, the placement of X. translucens pv. pistaciae in this cluster is particularly significant and raises questions about the processes that have allowed a Xanthomonas to adapt to pistachio, a dicotyledonous woody host. Among bacteria, adaptation can be achieved by conjugation and transformation, or by lateral gene transfer, which involves genetic elements such as plasmids, transposons and integrons [19]. Gillings et al. [7] showed that integrons have played a role in structuring the genetic diversity of Xanthomonas species. Accordingly, integron gene cassettes clearly distinguished X. translucens pv. pistaciae from X. oryzae pv. oryzae and from X. translucens pv. translucens. Furthermore, strains in groups A and B harbored different integrase genes and cassette arrays, which shows that their integrons have evolved separately and again supports the distinction of the 2 groups. However, the significance of this observation in terms of their adaptation to pistachio has not been investigated. In summary, we propose that the X. translucens infecting pistachio be classified as a novel pathovar, Xanthomonas translucens pv. pistaciae pv. nov. Based on several criteria, 2 groups of strains, X. translucens pv. pistaciae A and X. translucens pv. pistaciae B, were consistently identified. The pathotype strain is proposed to be X. translucens pv. pistaciae A strain ICMP 16316, as it is the most widespread in pistachio orchards. The overall genetic similarity between X. translucens pv. pistaciae and the other X. translucens pathovars shown by DNA–DNA hybridization is supported by gyrB sequences and previously published data [13]. Such a similarity in the genetic background suggests that X. translucens pv. pistaciae has evolved from another X. translucens by acquiring new pathogenicity genes. This may have occurred through recent adaptation, since the introduction of pistachio to Australia in the mid-1930s, or long-term co-evolution in association with pistachio in Iran, the country of origin. Further investigation is needed to understand the origin of X. translucens pv. pistaciae and to determine if integrons have played a role in its adaptation to pistachio. The gyrB phylogeny has clarified the relatedness of X. translucens pv. pistaciae to the other pathovars in the species. However, the consistent and strong discrimination of the pistachio pathogen and of its 2 genetic groups suggest that there is a deeper structure to its classification. The sequences of more housekeeping genes will be useful to address this issue.
ARTICLE IN PRESS 556
D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
Description of Xanthomonas translucens pv. pistaciae pv. nov. X. translucens pv. pistaciae was a motile Gram-negative rod-shaped bacterium of about 0.4 1.5 mm in size [5]. It produced typical pale yellow, mucoid and domed colonies on Sucrose Peptone Agar [5]. All strains had an oxidative metabolism of glucose, produced H2S, grew at 36 1C, were positive for starch, gelatin and esculin hydrolysis but were oxidase and urease negative and did not grow in the presence of 5% NaCl [13]. All strains metabolized a-keto glutaric acid, D, L-lactic acid, bromosuccinic acid, L-alaninamide, L-alanine, L-alanylglycine, L-glutamic acid and L-serine [13]. X. translucens pv. pistaciae A and X. translucens pv. pistaciae B differed in their ability to digest milk proteins, reaction in litmus milk, ice nucleation activity at 4 1C and ability to induce a hypersensitive reaction on tobacco leaves [13]. Natural hosts [5]: pistachio (P. vera). Hosts indicated by inoculation [5,13]: P. atlantica, P. terebinthus, P. sinensis, P. palaestina, Rhus tripartita, Schinus latifolius, S. lentiscifolius, S. polygamus, Triticum aestivum (bread wheat), T. durum (durum wheat), Hordeum vulgare (barley), H. leporinum (barley grass), Secale cereale (rye), Avena sativa (oat), Bromus inermis (smooth brome) and B. diandrus (brome grass), X Triticosecale (triticale), Phleum pratense (timothy), Dactylis glomerata (cocksfoot) and Lolium rigidum (annual rye grass). Rhus leptodictya and Avena fatua (common wild oat) were not hosts. X. translucens pv. pistaciae A also infected P. lentiscus and Lolium perenne (perennial rye grass). The pathotype strain is X. translucens pv. pistaciae A, ICMP 16316 (also referenced as NCPPB 4448). The reference strain for X. translucens pv. pistaciae B is ICMP 16317 (also referenced as NCPPB 4449).
Acknowledgements We thank Carolee Bull for critical comments on the manuscript, Margaret Sedgley and Bob Emmett for useful discussion, and Eric Cother for providing X. translucens pv. translucens DAR 35705. We are indebted to Chris Burrell and Brian Lewis from the Institute for Medical and Veterinary Science in Adelaide, for allowing access to PC3 containment facilities. This work was funded by the Pistachio Growers Association of Australia and Horticulture Australia Limited (Grant PS06002).
Appendix. Supporting Information Supplementary data associated with this article can be found in the online version at doi:10.1016/ j.syapm.2009.08.001.
References [1] N. Ah-You, L. Gagnevin, P.A.D. Grimont, S. Brisse, X. Nesme, F. Chiroleu, L. Bui Thi Ngoc, E. Jouen, P. Lefeuvre, C. Vernie`re, O. Pruvost, Polyphasic characterisation of xanthomonads pathogenic to Anacardiaceae and their relatedness to species of Xanthomonas, Int. J. Syst. Evol. Microbiol. 59 (2009) 306–318. [2] J. Cubero, H. Graham, The leucine-responsive regulatory protein (lrp) gene for characterization of the relationship among Xanthomonas species, Int. J. Syst. Evol. Microbiol. 54 (2004) 429–437. [3] B.M. Cunfer, B.L. Scolari, Xanthomonas campestris pv. translucens on triticale and other small grains, Phytopathology 72 (1982) 683–686. [4] E. Facelli, C. Taylor, E. Scott, R. Emmett, M. Fegan, M. Sedgley, Bacterial dieback of pistachio in Australia, Australas. Plant Pathol. 31 (2002) 95–96. [5] E. Facelli, C. Taylor, E. Scott, M. Fegan, G. Huys, R. Emmett, D. Noble, J. Swings, M. Sedgley, Identification of the causal agent of pistachio dieback in Australia, Eur. J. Plant Pathol. 112 (2005) 155–165. [6] M. Gillings, P. Fahy, Genetic diversity of Pseudomonas solanacearum biovars 2 and N2 assessed using restriction endonuclease analysis of total genomic DNA, Plant Pathol. 42 (1993) 744–753. [7] M.R. Gillings, M.P. Holley, H.W. Stokes, A.J. Holmes, Integrons in Xanthomonas: a source of species genome diversity, Proc. Natl. Acad. Sci. USA 102 (2005) 4419–4424. [8] L. Hauben, L. Vauterin, J. Swings, E. Moore, Comparison of 16S ribosomal DNA sequences of all Xanthomonas species, Int. J. Syst. Bacteriol. 47 (1997) 328–335. [9] V.A.R. Huss, H. Festl, K.H. Schleifer, Studies on the spectrophotometric determination of DNA hybridization from renaturation rates, Syst. Appl. Microbiol. 4 (1983) 184–192. [10] J.D. Janse, Pseudomonas syringae subsp. savastanoi (ex Smith) subsp. nov., nom. rev., the bacterium causing excrescences on Oleaceae and Nerium oleander, L. Int. J. Syst. Bacteriol. 32 (1982) 166–169. [11] M.T. La Duc, M. Satomi, N. Agata, K. Venkateswaran, gyrB as a phylogenetic discriminator for members of the Bacillus anthracis–cereus–thuringiensis group, J. Microbiol. Methods 56 (2004) 383–394. [12] A. Marefat, K. Ophel-Keller, E. Scott, M. Sedgley, The use of ARMs PCR in detection and identification of Xanthomonads associated with pistachio dieback in Australia, Eur. J. Plant Pathol. 116 (2006) 57–68. [13] A. Marefat, E. Scott, K. Ophel-Keller, M. Sedgley, Genetic, phenotypic and pathogenic diversity among xanthomonads isolated from pistachio (Pistacia vera) in Australia, Plant Pathol. 55 (2006) 639–649. [14] N. Parkinson, V. Aritua, J. Heeney, C. Cowie, J. Bew, D. Stead, Phylogenetic analysis of Xanthomonas species by comparison of partial gyrase B gene sequences, Int. J. Syst. Evol. Microbiol. 57 (2007) 2881–2887. [15] N.M. Parkinson, C. Cowie, J. Heeney, D. Stead, Phylogenetic structure of Xanthomonas determined by comparison of gyrB sequences, Int. J. Syst. Evol. Microbiol. 59 (2009) 264–274.
ARTICLE IN PRESS D. Giblot-Ducray et al. / Systematic and Applied Microbiology 32 (2009) 549–557
[16] J.L.W. Rademaker, F.J. de Bruijn, Characterization and classification of microbes by rep-PCR genomic fingerprinting and computer-assisted pattern analysis, in: G. Caetano-Anolles, P.M. Gresshoff (Eds.), DNA Markers: Protocols, Application and Overviews, Wiley, New York, USA, 1997, pp. 151–171. [17] J.L.W. Rademaker, F.J. Louws, M.H. Schultz, U. Rossbach, L. Vauterin, J. Swings, F.J. de Bruijn, A comprehensive species to strain taxonomic framework for Xanthomonas, Phytopathology 95 (2005) 1098–1111. [18] J.L.W. Rademaker, D.J. Norman, R.L. Foster, F.J. Louws, M.H. Shultz, F.J. de Bruijn, Classification and identification of Xanthomonas translucens isolates, including those pathogenic to ornamental asparagus, Phytopathology 96 (2006) 876–884. [19] M. Scortichini, The population structure of some plant pathogenic bacteria: an ecological and adaptive perspective, J. Plant Pathol. 87 (2005) 5–12. [20] E. Stackebrandt, W. Frederiksen, G.M. Garrity, P.A.D. Grimont, P. Kampfer, M.C.J. Maiden, X. Nesme, R. Rossello´-Mo´ra, J. Swings, H.G. Tru¨per, L. Vauterin, A.C. Ward, W.B. Whitman, Report of the ad hoc
[21]
[22]
[23]
[24]
[25]
557
committee for the re-evaluation of the species definition in bacteriology, Int. J. Syst. Evol. Microbiol. 52 (2002) 1043–1047. J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673–4680. Y. Van de Peer, R. De Watcher, TREECON: a software package for the construction and drawing of RT evolutionary trees, Comput. Appl. Biosci. 9 (1993) 177–182. L. Vauterin, B. Hoste, K. Kersters, J. Swings, Reclassification of Xanthomonas, Int. J. Syst. Bacteriol. 45 (1995) 472–489. S. Yamamoto, S. Harayama, PCR amplification and direct sequencing of gyrB genes with universal primers and their application to the detection and taxonomy analysis of Pseudomonas putida strains, Appl. Environ. Microbiol. 61 (1995) 1104–1109. J.M. Young, D.-C. Park, H.M. Shearman, E. Fargier, A multilocus sequence analysis of the genus Xanthomonas, Syst. Appl. Microbiol. 31 (2008) 366–377.
