ARTICLE IN PRESS

Systematic and Applied Microbiology 33 (2010) 105–115

Contents lists available at ScienceDirect

Systematic and Applied Microbiology

journal homepage: www.elsevier.de/syapm

Pseudomonas cannabina pv. cannabina pv. nov., and cannabina pv. alisalensis (Cintas Koike and Bull, 2000) comb. nov., are members of the emended species Pseudomonas cannabina (ex Suticˇ ˇ & Dowson 1959) Gardan, Shafik, Belouin, Brosch, Grimont & Grimont 1999$

Carolee T. Bull a,n, Charles Manceau b, John Lydon c, Hyesuk Kong c,1, Boris A. Vinatzer d, Marion Fischer-Le Saux b a United States Department of Agriculture, Agricultural Research Service (USDA/ARS), 1636 E. Alisal St., Salinas, CA 93905, United States b INRA UMR077 Pathologie Ve´ge´tale, F-49070 Beaucouze, France c USDA/ARS, Sustainable Agricultural Systems Laboratory, Belstville, MD, 20705-2350, United States d 551 Latham Hall, PPWS Department, Virginia Tech, Blacksburg, VA 24061, United States article info abstract

Article history: Sequence similarity in the 16S rDNA gene confirmed that crucifer pathogen pv. Received 29 September 2009 alisalensis belongs to P. syringae sensu lato. In reciprocal DNA/DNA hybridization experiments, DNA relatedness was high (69–100%) between P. syringae pv. alisalensis strains and the type strain of Keywords: P. cannabina (genomospecies 9). In contrast, DNA relatedness was low (below 48%) between P. syringae pv. alisalensis and reference strains from the remaining genomospecies of P. syringae including the type Hops strain of P. syringae and reference strain of genomospecies 3 (P. syringae pv. tomato) although the well- Marijuana known crucifer pathogen, P. syringae pv. maculicola, also belongs to genomospecies 3. Additional Host range evidence that P. syringae pv. alisalensis belongs to P. cannabina was sequence similarity in five gene Brassica rapa fragments used in multilocus sequence typing, as well as similar rep-PCR patterns when using the BOX- Broccoli raab A1R primers. The description of P. cannabina has been emended to include P. syringae pv. alisalensis. Broccoli Pseudomonas syringae pv. maculicola Host range testing demonstrated that P. syringae pv. alisalensis strains, originally isolated from broccoli, Pseudomonas syringae pv. tomato broccoli raab or arugula, were not pathogenic on Cannabis sativa (family Cannabinaceae). Additionally, P. cannabina strains, originally isolated from the C. sativa were not pathogenic on broccoli raab or oat while P. syringae pv. alisalensis strains were pathogenic on these hosts. Distinct host ranges for these two groups indicate that P. cannabina emend. consists of at least two distinct pathovars, P. cannabina pv. cannabina pv. nov., and P. cannabina pv. alisalensis comb. nov. Pseudomonas syringae pv. maculicola strain CFBP 1637 is a member of P. cannabina. Published by Elsevier GmbH.

Introduction syringae is further delineated into pathovars (an infrasubspecific designation for phytopathogenically distinct members of a Pseudomonas syringae is a heterogeneous species consisting of species; [18]). A comprehensive genetic analysis grouped plant pathogens and epiphytes with broad pathogenic capabilities many pathovars of P. syringae into nine genomospecies based on and taxonomic characteristics [11,19,22,35,44,55]. Pseudomonas DNA/DNA hybridization and ribotyping [19]. However, only two of the nine genomospecies, P. cannabina and P. tremae,were proposed as authentic species due to a lack of distinguishing Abbreviations: CFBP, Collection Franc-aise de Bacte´ries Phytopathogenes; G1–G9, phenotypic characteristics among the strains in the other genomospecies 1–9; KMB, King’s medium B; LMG, Laboratorium voor Microbiologie genomospecies. University of Gent; LOPAT, levan production, oxidase reaction, potato rot, arginine Pseudomonas syringae pv. maculicola has long been known as dihydrolase production, tobacco hypersensitivity $Nucleotide sequence data reported are available in the DDBJ/EMBL/GenBank an important pathogen of crucifers world-wide [30].Itisa databases under the accession number(s): GQ470207-GQ470215; GQ870338- heterogeneous taxon with strains identified as P. syringae GQ870341; GQ859258-GQ859264. pv. maculicola assigned to three different groups by multilocus n Corresponding author. Tel.: +1 831 755 2889; fax: +1 831 755 2814. sequence typing analysis (MLST [35]). The pathotype of P. syringae E-mail address: [email protected] (C.T. Bull). 1 Current address: U.S. Food and Drug Administration, Center for Biologics pv. maculicola was assigned to genomospecies 3 (the P. syringae Evaluation and Research, Rockville, MD 20852, United States pv. tomato group) in the analysis of Gardan et al. [19].

0723-2020/$ - see front matter Published by Elsevier GmbH. doi:10.1016/j.syapm.2010.02.001 ARTICLE IN PRESS

106 C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115

Subsequently a novel P. syringae (sensu lato) isolated from the Materials and methods cruciferous plant broccoli raab (Brassica rapa subsp. rapa) was designated as P. syrinage pv. alisalensis because it is genetically Bacterial strains and media and pathogenically distinct from P. syringae pv. maculicola [14]. Pseudomonas syringae pv. alisalensis has a unique and broad host All strains used in these studies came from the Collection Franc- range including crucifers and monocots, a distinctive rep-PCR aise de Bacte´ries Phytopathogenes (CFBP) or Laboratorium voor pattern, and is uniquely sensitive to a bacteriophage isolated from Microbiologie University of Gent (LMG) unless otherwise stated a diseased broccoli raab field. This pathogen has been isolated (Table 1). The pathotype strain of Pseudomonas syringae pv. from symptomatic plants from disease outbreaks in convention- alisalensis, CFBP 6866Pt (BS91Pt; from broccoli raab, Brassica rapa ally and organically managed crucifer production fields across subsp. rapa [14]), and additional P. syringae pv. alisalensis strains the US [9,10,12,13,25,26]. The host ranges for P. syringae pv. from broccoli (CFBP 6867 and 6873, Brassica olearacea var botrytis maculicola and P. syringae pv. alisalensis overlap [14] and disease [14]) and arugula (CFBP 6869 and 6875, Eruca sativa [9,10])in outbreaks caused by P. syringae pv. alisalensis have been California and New Jersey, USA, were used in these experiments. incorrectly attributed to P. syringae pv. maculicola (Bull et al., Additional strains from more recent disease outbreaks in rutabaga unpublished). (Brassica napus var. napobrassica; CFBP 7253 [26]), Brussels sprouts Although there are phenotypic and pathogenic similarities (Brassica oleracea L. var. gemmifera; CFBP 7254; Bull unpublished), between P. syringae pv. maculicola and P. syringae pv. alisalensis,it cauliflower transplants (Brassica oleracea var. botrytis; CFBP 7251; is not clear whether they are closely related. This study was Bull unpublished), and Romanesca (Brassica oleracea var. botrytis; undertaken to determine if P. syringae pv.maculicolaand P. syringae CFBP 7252 [25]) were included in some analyses. pv. alisalensis belong to the same or different genomospecies and Reference or other representative strains from each of the if not to determine the appropriate taxonomic placement of eight valid genomospecies (G1–G9) proposed by Gardan et al. [19] P. syringae pv. alisalensis. were used as controls in some experiments (G1, Pseudomonas

Table 1 Species and Pathovars used in this study.

Species and Pathovars Strain Genomospecies 16S rDNA accession Location of Host of origin Source no. isolation

Pseudomonas cannabina CFBP G9 AJ492827 Hungary , Cannabis sativa Klement 1957 2341T Pseudomonas cannabina CFBP G9 GQ870340 Hungary Hemp, Cannabis sativa Klement 1957 1619 Pseudomonas cannabina CFBP G9 GQ470211 Yugoslavia Hemp, Cannabis sativa Dowson 1968 1631 Pseudomonas cannabina LMG 5540 GQ870338 Yugoslavia Hemp, Cannabis sativa Sutic 1958 Pseudomonas cannabina LMG 5650 GQ870339 Hemp, Cannabis sativa Klement 1957 P. viridiflava CFBP G6 Z76671 Switzerland Phaseolus sp. 2107T P. syringae CFBP G1 Z76669 United Kingdom Syringa vulgaris 1392T P. syringae pv. phaseolicola CFBP G2 AB001448 Phaseolus vulgaris Pt Canada 1390 P. syringae pv. tomato CFBP G3 GQ470214 United Kingdom Lycopersicon esculentum 2212Pt Pseudomonas syringae pv. CFBP G3 GQ470210 New Zealand Brassica oleracea var. botrytis maculicola 1657Pt Pseudomonas syringae pv. CFBP Determined GQ470209 CA, USA Radish, Raphanus sativus [51] maculicola 1637 here P. syringae pv. porri CFBP G4 France Allium porrum 1908Pt P. syringae pv. helianthi CFBP G7 GQ870341 Mexico Helianthus annuus 2067Pt P. syringae pv. theae CFBP G8 AB001450 Japan Thea sinensis 2353Pt Pseudomonas syringae pv. CFBP Determined GQ470207 CA, USA Broccoli raab, Brassica rapa subsp. rapa [14] alisalensis 6866Pt here Pseudomonas syringae pv. CFBP Determined GQ470212 CA, USA Broccoli, Brassica olearacea var botrytis [14] alisalensis 6867 here Pseudomonas syringae pv. CFBP Determined GQ470215 CA, USA Arugula [9] alisalensis 6869 here Pseudomonas syringae pv. CFBP Determined GQ470208 NJ, USA Broccoli raab [14] alisalensis 6873 here Pseudomonas syringae pv. CFBP Determined GQ470213 NJ, USA Arugula [9] alisalensis 6875 here Pseudomonas syringae pv. CFBP CA, USA Cauliflower, Brassica oleracea var. botrytis Bull et al., alisalensis 7251 unpublished Pseudomonas syringae pv. CFBP CA, USA Romanesca, Brassica oleracea var. botrytis [24] alisalensis 7252 Pseudomonas syringae pv. CFBP CA, USA Rutabega, Brassica napus var. napobrassica [25] alisalensis 7253 Pseudomonas syringae pv. CFBP CA, USA Brussels sprouts, Brassica oleracea L. var. Bull et al., alisalensis 7254 gemmifera unpublished ARTICLE IN PRESS

C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 107 syringae CFBP 1392T; G2, Pseudomonas syringae pv. phaseolicola Multilocus sequence typing (MLST) CFBP 1390Pt; G3, Pseudomonas syringae pv. tomato CFBP 2212Pt; G4, Pseudomonas syringae pv. porri CFBP 1908Pt; G6, Pseudomonas Gene fragments of the four loci gap1, gltA, gyrB, and rpoD used viridiflava CFBP 2107Pt; G7, Pseudomonas syringae pv. helianthi in the MLST scheme developed by Hwang et al. [22] and a CFBP 2067Pt; G8, Pseudomonas syringae pv. theae CFBP 2353Pt; and fragment of the kup gene from the MLST scheme developed by G9, P. cannabina CFPB 2341T). The type strain from P. tremae (G5) Yan et al. [54] were amplified by PCR with primers described in was not used in this study, because we have evidence that it the respective papers and PCR products were sequenced as belongs to genomospecies 2 (Fischer-LeSaux et al., unpublished described in Yan et al. [54]. Sequences were aligned with the data). In some cases additional representatives from an individual allele sequences of P. syringae pv. tomato for all loci, cut to size, genomospecies were included. For example, P. cannabina strains and single nucleotide polymorphisms (SNPs) were identified CFBP 1619, CFBP 1631, LMG 5540, and LMG 5650 and Pseudomo- using Seqman (DNAstar, Madison, WI, USA). Sequence distances nas syringae pv. maculicola CFBP 1657Pt and CFBP 1637 (strain were determined in MegAlign (DNAstar, Madison). B-70 [51]) were included in some analyses. The were stored at À801C in a solution of 50% glycerol and 50% nutrient broth (NB) and were routinely cultured on King’s medium B agar (KMB [23]). Fatty acid profiles

Strains were grown and extracted for fatty acid methyl ester 16S rDNA analysis using a previously published method [14]. Fatty acid methyl esters were analyzed with the Sherlock Microbial For amplification of the 16S rDNA gene, genomic DNA was Identification System Version 6.1 (MIDI Inc., Newark, DE) using extracted using DNeasy Tissue Kits (Qiagen, Valencia, CA). An MJ an automated GC 6890 Hewlett-Packard gas chromatograph fitted Research DNA-Engine thermo-cycler (MJ Research, Waltham, MA) 2 with a 25 Â 0.2 mm phenyl methyl silicone-fused silica capillary with a heated lid in the ‘block’ mode was used for all polymerase column, an HP 7683 automatic sampler, and Agilent ChemStation chain reactions (PCR). Amplicons for the 16S rDNA were Software (Ver. B.03.02). The mean and standard deviation of the generated using universal primers 27F and 1492R [28] using area for each named peak from three independent replications published reaction conditions [50]. After visually checking for was reported as a percentage of the total area of all peaks in the amplification by gel electrophoresis, amplicons were purified chromatogram not including the solvent peak. using Ultra Clean PCR Clean-Up kits (MoBio, Carlsbad, CA). Amplicons were sequenced directly by an outside vendor (McLab, South San Francisco, CA). Geneious software (http://www. geneious.com) was used to align sequences from forward and Phenotypic characters reverse strands, generate sequence alignments and trees and conduct phylogenetic analysis using a neighbour-joining algo- The Biotype 100 (bioMerieux, Marcy l’Etoile, France) system rithm and bootstrapping (n=1000 simulations). was used to determine the ability of the strains to utilize 99 carbon sources. Biotype medium 1 was used to inoculate the DNA–DNA similarity cupules of the strips according to manufacturer’s recommenda- tions. The strips were then incubated at 28 1C and growth was Strains representing each of the nine genomospecies defined recorded after 2, 4 and 6 days for each cupule. Additional by Gardan et al. [19] were compared to P. syringae pv. alisalensis in phenotypic characterization included evaluation of fluorescence DNA–DNA hybridization experiments. Genomic DNA for DNA– on KMB, the KOH test to determine Gram character, levan DNA hybridization experiments was extracted and purified production, oxidase reaction, potato rot, arginine dihydrolase following published methods [7]. Native DNAs from P. syringae production, tobacco hypersensitivity (LOPAT), reduction of pv. alisalensis strains CFBP 6866Pt and CFBP 6867 and the type nitrate, hydrolysis of gelatine, DNA and Tween 80, pectinolytic strain of P. cannabina (CFBP 2341T) were labeled in vitro by nick activity at pH 5 and pH 8, acidification of sucrose, sorbitol, translation with 3H-labeled nucleotides (Amersham) using a erythritol and mannitol, alcalinisation of DL lactate and L(+) previously published S1 nuclease trichloroacetic acid procedure tartrate, sensitivity to bacteriophage PBS1, and ice nucleating [16]. The reassociation temperature was 70 1C. Each hybridization ability using published methods [1,14,29,37]. was conducted at least three times. DNA relatedness is reported To determine if coronatine biosynthesis genes were present in as the average percent reassociation relative to the reassociation the pathogens, primers to the cfl gene encoded within the of the probe DNA to itself. coronatine gene cluster were used to amplify fragments using published methods [57]. DNA from Pseudomonas syringae pv. maculicola (CFBP 1657Pt) and P. syringae pv. syringae (CFBP Rep-PCR 1392T) were used, respectively, as the positive and negative controls. To determine if ethylene biosynthesis genes were The BOX-A1R primer, designed to prime DNA synthesis from present in the pathogens, primers to the efe gene were used to the boxA subunit of the BOX element, was used in the PCR of amplify fragments using published methods [36] with minor repetitive bacterial sequences on purified genomic DNA using modifications. We used purified DNA and a touchdown PCR published methods [14,34]. Amplified DNA fragments were protocol in which annealing temperatures were 60, 58 and 56 1C examined by agarose gel electrophoresis in 1.5% agarose gels in for two cycles each followed by the standard reaction conditions. 0.5X Tris acetic acid EDTA buffer, TAE. Gels were stained with Three negative (Pseudomonas syringae pv. syringae CFBP 1392T; ethidium bromide or GelRed and photographed on a UV P. syrinage pv. maculicola CFBP 1657Pt; Pseudomonas syrinage transilluminator using a digital camera and Kodak Molecular pv. tomato CFBP 2212Pt) and one positive control (Pseudomonas Imaging software (v. 4.5.1, Carestream Health, Inc., Rochester, cannabina CFBP 2341T) were used in these experiments. Numer- NY). DNA fragment banding patterns generated from different ical analysis and identification of discriminative strains were compared visually. characters were performed as described by Achouack et al. [1]. ARTICLE IN PRESS

108 C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115

Pathogenicity P. syringae pv. maculicola from radish strain (CFBP 1637) was identical to strains of P. syringae pv. alisalensis from arugula, For pathogenicity tests on Cannabis sativa (L.) (accession having a cytosine residue at base 264. 910972 [32]), 24-day-old plants were inoculated. Cells from A basic local alignment search tool (BLAST [3]) comparison overnight cultures of bacteria grown in KMB broth were with sequences in public databases and phylogenetic analysis centrifuged, the broth decanted, and the resulting bacterial pellet including type strains of some of the main Pseudomonas resuspended in sterile, distilled H2O and adjusted to an OD600 nm phylogenetic clusters previously defined revealed unequivocally of 0.6. Three milliliters of the bacterial suspension were applied to that P. syringae pv. alisalensis strains are members of the leaves using a TLC sprayer (Sigma, St. Lewis, MO) at a pressure of P. syringae cluster (Fig. 1. [5,33]). The tree presented includes 138 Kpa. Plants sprayed with sterile distilled water served as representative strains from each of the genomospecies defined negative controls. Immediately after treatment, plants were within the P. syringae cluster. placed into plastic bags and held at 23 1C without light. After All P. syringae pv. alisalensis strains were nearly identical 48 h, plants were removed from the plastic bags and grown with (99.2%) to the type strains of either P. cannabina (genomo- light. Plants were evaluated for foliar symptoms at 11 and 21 days species 9) or P. syringae (CFBP 1392, genemospecies 1) and after treatment. (499.9 %) to the pathotype strains of P. syringae pv. tomato and Pathogenicity was also evaluated on broccoli (Brassica olear- P. syringae pv. maculicola, (genomospecies 3). Additionally, there acea var botrytis cv. ‘Greenbelt’), broccoli raab (Brassica rapa was only moderate or low statistical support (bootstrap values subsp. rapa cvs. ‘Spring’ and ‘Sorento’), and oat (Avena sativa cv. below 60) for separating P. syringae pv. alisalensis from either ‘Montezuma’) seedlings, using previously published methods P. cannabina or P. syringae type strains. Because of the high [14]. The pathogens were grown in nutrient broth with shaking similarity between all sequences from members of P. syringae at 200 rpm for 24 h at 27 1C. Nutrients were removed from cells sensu lato and low bootstrap values, we were unable to assign by centrifugation and cultures were prepared in sterile 0.01 M P. syringae pv. alisalensis to a genomospecies based on 16S rDNA phosphate buffer as described above except that Tween 20 (0.05%) sequence analysis. All sequences were submitted to Genbank was added to the suspensions. Each suspension was sprayed until and were given accession numbers GQ470207-GQ470215 and runoff using a hand mister. After inoculation, plants were placed GQ870338-GQ870341. in a humidity chamber for 48 h. Plants were then maintained at 20—25 1C and evaluated for symptoms after 14 days. For negative controls, plants were sprayed with sterile distilled water amended with Tween 20 (0.05%). An experimental unit was six plants in a DNA—DNA hybridizations six-pack container. For all pathogenicity tests, lesions or symptomatic tissues Representative strains of P. syringae pv. alisalensis isolated from from leaves were excised two or three weeks after treatment, different hosts and originating from distant geographic origins surface-disinfested by soaking the leaf tissue in 0.5% sodium (California and New Jersey) belonged to the same genomospecies, hypochlorite for 1 min (with or without a 1 min pretreatment with DNA relatedness values ranging from 90% to 100% in in 70% ethanol) and rinsed with sterile distilled water. The leaf experiments using the P. syringae pv. alisalensis pathotype (CFBP tissue was then macerated in sterile distilled water. The resulting 6866Pt) as the labeled probe (Table 2). tissue suspensions or dilutions were streaked on KMB supple- Hybridization experiments were conducted between selected mented with boric acid, cephalexin, and cycloheximide [37] P. syringae pv. alisalensis strains and representative strains of the 8 and incubated at 28 1C. After 4 to 5 days, single colonies were genomospecies defined within the P. syringae cluster. Hybridiza- purified, rep-PCR using the BOX-A1R primer performed, and DNA tion of probes made from P. syringae pv. alisalensis strains CFBP banding patterns were compared to the strains used to inoculate 6866Pt or CFBP 6867 to P. cannabina type strain CFBP 2341T (the plants. Pathogenicity experiments were conducted at least twice. reference strain for genomospecies 9) averaged 97% and 100%, The results for pathogenicity tests were recorded as positive for respectively, indicating that P. syringae pv. alisalensis strains are plants from which the appropriate pathogen (as determined members of P. cannabina (genomospecies 9). by rep-PCR) was isolated from the margins of typical leaf In reciprocal experiments using P. cannabina CFBP 2341T as the symptoms. labeled probe, the percent hybridization to CFBP 6866Pt and CFBP 6867 averaged 81% and 79%, respectively. The average hybridiza- tion of the P. cannabina probe to additional strains of P. syringae Results pv. alisalensis ranged from 69% to 84%. An additional P. syringae strain, CFBP 1637, isolated in the US in 1965 from radish 16S rDNA sequences (Raphanus sativus) and previously identified as P. syringae pv. maculicola [51], was evaluated. Strain CFBP 1637 had 83% DNA sequences of 16S rDNA of five P. syringae pv. alisalensis DNA relatedness to the probe of the type strain of P. cannabina. strains and of strain CFBP 1637 were obtained and compared to These data clearly indicated that P. syringae pv. alisalensis strains each other and to reference strains. The 1326 nucleotide bases and CFBP 1637 (formerly designated P. syringae pv. maculicola) are sequenced represent the majority of the primary structure of the members of genomospecies 9 and should be transferred to 16S rDNA from positions 97 to 1423 based on Escherichia coli P. cannabina. nomenclature [8]. Moreover, hybridization of CFBP 6867 probe DNA to The sequences of the 16S rDNA differed slightly among representative strains of the seven other genomospecies ranged P. syringae pv. alisalensis strains depending on the hosts (arugula from 34% to 48%. In particular, hybridization of P. syringae vs. other hosts) from which the strains were isolated but not by pv. alisalensis probes to the reference strain of genomospecies 3, region (New Jersey or California). At nucleotide 264 (according to P. syringae pv. tomato strain CFBP 2212, averaged 42—43%. E. coli numbering) the P. syringae pv. alisalensis strains from This value is far below the threshold value of 70% [47] and broccoli raab and broccoli possessed the nucleotide thymine, indicates that unlike P. syringae pv. maculicola and P. syringae while the strains from arugula had a cytosine residue at this pv. tomato, P. syringae pv. alisalensis is not a member of location, but were otherwise identical. The 16S rDNA sequence of genomospecies 3. ARTICLE IN PRESS

C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 109

Fig. 1. Phylogenetic analysis of the 16S rDNA gene of P. syringae pv. alisalensis, P. cannabina, and other P. syringae strains. The trees were constructed from evolutionary distance data generated by neighbour joining for 16S rDNA sequences. Numbers by the nodes are bootstrap frequencies expressed as percentages of 1000 replicates (o50% are not shown). Numbers in parentheses are accession numbers for the sequences used in the analyses. G# refers to the genomospecies published by Gardan et al. [19] to which the bacteria belong. In the cases where the sequence was available, no strain designation is given and all but Pseudomonas syringae pv. tomato DC3000 are from the type (T) or pathotype (Pt) strains.

Rep-PCR patterns and P. syringae pv. maculicola were significantly different from the patterns for P. cannabina and P. syringae pv. alisalensis strains. The rep-PCR DNA fragment banding patterns for P. cannabina and P. syringae pv. alisalensis strains were similar but formed three MLST distinct DNA fragment banding patterns based on the presence or absence of four significant bands (Fig. 2; Table 2). The rep-PCR The sequences for the loci gap1, gltA, gyrB, and rpoD were patterns for the five P. cannabina strains (CFBP 2341T, CFBP 1619, identical for all P. syringae pv. alisalensis strains (data not shown) CFBP 1631, LMG 5540, and LMG5650) were identical. Within the while one SNP was found in the kup locus between the arugula nine P. syringae pv. alisalensis strains there were two different strains and all the other strains tested (Table 2). Strain P. syringae banding patterns. The P. syringae pv. alisalensis strains from pv. maculicola CFBP 1637 was identical in all loci to the P. syringae arugula differ in that they are missing a 1284 kb band that is pv. alisalensis strains from arugula. The P. cannabina allele present in all of the other P. syringae pv. alisalensis and sequences were different for all five loci when compared to the P. cannabina strains and a 1092 kb band that is present in all P. syringae pv. alisalensis alleles. However, sequence identity was other P. syringae pv. alisalensis strains. Overall, the majority of the very high: acnB (98.8%), gap1 (99.0%), gyrB (99.9%), kup (95.9%), bands are shared between P. cannabina and P. syringae pgi (99.6%), rpoD (98.3%), and gltA (98.8%). Sequence identity was pv. alisalensis strains. However, there is a 799 kb band present significantly lower when comparing P. syringae pv. alisalensis in patterns from P. cannabina but not in patterns from P. syringae alleles to the corresponding alleles of the pathotypes of P. syringae pv. alisalensis and a 271 kb band that is present in P. syringae pv. tomato and of P. syringae pv. maculicola and to the alleles of pv. alisalensis but not in P. cannabina. Additionally, the P. syringae pv. syringae B728a (between 88.7% and 93.4%). All P. cannabina strains, like the P. syringae pv. alisalensis from allele sequences have been submitted to Genbank with accession arugula, do not have the 1092 kb band. The rep-PCR patterns for numbers GQ859258 to GQ859264 and have been deposited in the the pathotypes of P. syringae pv. syringae, P. syringae pv. tomato PAMDB database ([4], and www.pamdb.org). ARTICLE IN PRESS

110 C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115

Table 2 DNA–DNA homology, MLST for the kup locus and rep-PCR pattern types for Pseudomonas syringae strains and related species. species.

Target DNA or test organism Genomospecies CFBP number DNA–DNA Hybridization (percent) Source of probe DNA Additional genetic characterization

P. syringae pv. alisalensis P. cannabina MLST kup allelea Rep-PCR typeb

CFBP 6867 CFBP 6866 CFBP 2341

P. syringae pv. syringae G1c 1392Pt 41 NTd NT NT NT P. syringae pv. phaseolicola G2 1390Pt 48 NT NT NT NT P. syringae pv. tomato G3 2212Pt 43 42 NT NT NT P. syringae pv. porri G4 1908Pt 48 NT NT NT NT P. viridiflava G6 2107Pt 35 NT NT NT NT P. syringae pv. helianthi G7 2067Pt 43 NT NT NT NT P. syringae pv. theae G8 2353Pt 34 NT NT NT NT P. cannabina G9 2341T 97 100 100 17 3 P. syringae pv. alisalensis 6866Pt 100 100 81 18 1 P. syringae pv. alisalensis 6867 100 90 79 18 1 P. syringae pv. alisalensis 6869 NT 94 69 19 2 P. syringae pv. alisalensis 6873 NT 100 84 18 1 P. syringae pv. alisalensis 6875 NT 93 73 19 2 P. syringae pv. alisalensis 7251 NT NT NT 18 1 P. syringae pv. alisalensis 7252 NT NT NT 18 1 P. syringae pv. alisalensis 7253 NT NT NT 18 1 P. syringae pv. alisalensis 7254 NT NT NT 18 1 P. syringae pv. maculicola 1637 NT NT 83 19 2

a Allele sequences for the kup locus were deposited in the PAMDB database ([4], and www.pamdb.org). b Rep-PCR type refers to the corresponding rep-PCR fingerprint pattern (Fig. 2). c G1—9 designate reference strains or representative strains of genomonspecies delineated by Gardan et al. [19]. d NT=not tested.

Band sizes (kb) 517, 506 1018 1636 2036 3054 220 298 344 396

1 KB ladder P. cannabina LMG 5650 P. cannabina LMG 5540 P. cannabina CFBP 1619 P. cannabina CFBP 1631 P. cannabina CFBP 2341T P. syringae pv. alisalensis CFBP 6866Pt P. syringae pv. alisalensis CFBP 6867 P. syringae pv. alisalensis CFBP 6873 P. syringae pv. alisalensis CFBP 7251 P. syringae pv. alisalensis CFBP 7252 P. syringae pv. alisalensis CFBP 7253 P. syringae pv. alisalensis CFBP 7254 P. syringae pv. alisalensis CFBP 6869 P. syringae pv. alisalensis CFBP 6875 P. syringae pv. maculicola CFBP 1637 P. syringae pv. maculicola CFBP 1657Pt P. syringae pv. tomato CFBP 2212Pt P. syringae CFBP 1392T

Fig. 2. Rep-PCR fragment banding patterns obtained using the BOX-A1R primers.

Analysis of fatty acids Phenotypic analysis

All of the P. cannabina and P. syringae strains evaluated Phenotypic characters for the strains evaluated are given in produced 10:0 3OH, 12:0, 12:0 2OH, 12:3OH, 16:0, 16:1 o7c, Table 4, Supplemental Table 1 and in species and pathovar 18:0, and 18:1 o7c fatty acids (Table 3). All strains evaluated descriptions. All P. cannabina and P. syringae pv alisalensis strains except the two P. cannabina strains also produced 18:1 o7c were KOH positive, i.e. Gram negative, and produced a 11-methyl. Additionally, P. syringae pv. syringae, and P. syringae hypersensitive response on tobacco. The strains were oxidase pv. maculicola produced 17:0 iso which was not produced by and arginine dihydrolase negative and did not rot potato slices, P. cannabina or P. syringae pv. alisalensis. Trace amounts of 17:0 iso indicating that they all belonged to Lelliot’s LOPAT group 1 [29]. were produced in each of three independent analyses of Levan production was variable for individual P. syringae P. syringae pv. tomato strain CFBP 2212. pv. alisalensis strains, whereas levan was consistently positive ARTICLE IN PRESS

C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 111

Table 3 Fatty acids from P. cannabina, P. syringae pv. alisalensis, and related strains.

Feature Bacterial strains

Pseudomonas cannabina Pseudomonas syringae

pv. cannabina pv. alisalensis pv. tomato pv. syringae maculicola

CFBP 1631 CFBP 2341T CFBP 6866Pt CFBP 6867 CFBP 6869 CFBP 6873 CFBP 6875 CFBP 2212Pt CFBP 1392T CFBP 1657Pt

10:0 3OH 2.7370.66 2.5870.30 4.3670.23 4.3070.41 4.3370.50 4.1570.39 4.2870.65 4.0070.3 2.7570.29 4.1570.21 12:0 4.2370.27 4.3070.09 4.4570.27 4.3770.10 4.3070.03 4.2970.12 4.4470.06 4.2870.07 4.5870.13 4.5570.21 12:0 2OH 2.6470.15 2.7570.15 2.6670.12 2.6370.19 2.7770.11 2.6770.13 2.9170.09 2.3870.05 2.8870.10 2.5670.23 12:0 3OH 3.3670.53 3.5570.23 3.9470.34 3.8070.52 3.9870.32 3.6970.33 3.9370.25 3.6070.25 3.9570.35 3.9270.49 14:0 Tra Tr Tr Tr – 0.3270.02 Tr Tr Tr Tr 16:0 25.972.66 25.672.51 27.970.97 26.672.09 24.271.22 28.271.39 24.871.33 26.371.74 26.471.47 25.571.48 17:0 iso – – – – – – – Trb 0.4370.08 0.2670.05 18:0 0.8470.03 1.0070.00 1.0570.13 0.6570.23 1.0770.31 0.6970.20 0.9570.29 1.3570.23 1.5170.15 1.0770.42 18:1 o7c 11-methyl – – 1.6270.01 0.8270.29 0.5970.54 1.2370.22 0.9270.41 0.6470.09 1.1370.15 0.6570.11 18:1 o7c 21.2371.87 21.871.89 13.870.50 15.670.92 18.970.43 14.070.16 18.070.11 18.970.87 20.670.76 18.570.76 Summed Feature 3 38.870.60 38.470.07 39.871.17 41.070.79 39.4071.27 40.5370.90 39.7470.98 38.070.58 35.271.15 38.671.61 (16:1 o7c/16:1 o6c) Summed Feature 8 21.2371.87 21.871.89 13.870.50 15.670.92 18.970.43 14.070.16 18.070.11 18.970.87 20.670.76 18.570.76 (18:1 o7c/18:1 o6c)

a Tr, trace amounts detected in at least one of three assays. b For this strain, trace amounts of 17:0 iso were produced in each of three independent analyses of CFBP 2212.

Table 4 Characteristics relevant for distinguishing Pseudomonas cannabina emend. and its pathovars from closely related P. syringae pathovars.

Species P. cannabina emend. P. syringae

Genomospecies G9a G3 G1

Pathovars P. cannabina pv. cannabina P. cannabina pv. alisalensis pv. pv. pv. tomato maculicola syringae Phenotypic characters CFBP CFBP CFBP LMG LMG CFBP CFBP CFBP CFBP CFBP CFBP CFBP CFBP CFBP 2341T 1619 1631 5540 5650 6866Pt 6867 6869 6873 6875 1637 2212Pt 1657Pt 1392T

Ethylene production genes ++b +++++++++ +++++ÀÀ À Coronatine production genes + + + + + + ++++++ + À Brown pigment on KMB + + + + + À ÀÀÀÀÀÀ À À Sensitivity to PBS1 À ÀÀÀÀ+ ++++NTc ÀÀ À

Pathogenicity on Broccoli raaba, Brassica rapa subsp. À d À NT NT NT + e ++++NTÀÀ À rapa, cv. ‘Spring’ or ‘Sorento’ Oat, Avena sativa,cv. ‘Montezuma’ ÀÀNTNTNT+++++NTÀÀ À Broccoli, Brassica olearacea var NTNTNTNTNT+++++NTÀ + À botrytis, cv. ‘Greenbelt’ Cannnabis sativa, accession 91097 + + NT NT NT À ÀÀÀÀNT NT NT NT

a G1, G3, and G9 refer to genomonspecies as delineated by Gardan et al. [19]. b While the bands present for P. syringae pv. alisalensis strains (+) clearly indicated the presence of ethylene production genes as specified by the eth primers of Sato et al. [36], significantly more product was produced in reactions containing P. cannabina strains (++). No amplicons (À) were detected among the other strains. c NT=not tested. d The data represent uniform results from three replications in at least two independent experiments. e Identity of the pathogens associated with symptoms were verified after reisolation by rep-PCR. for P. cannabina strains as well as the P. syringae pv. syringae, pv. alisalensis) can be differentiated from P. syringae strain CFBP P. syringae pv. tomato and P. syringae pv. maculicola pathotypes. 1392T (G1) by the presence of coronatine production genes, Additional phenotypic traits common to both P. cannabina and hydrolysis of Tween 80, lack of gelatinase and lack of assimilation P. syringae pv. alisalensis are given in the emended description of of trans-aconitate, DL-glycerate, i-erythritol, DL-lactate and the P. cannabina species. Among them presence of ethylene DL-beta-hydroxybutyrate. Pseudomonas cannabina emend. can be production genes is a common trait among P. cannabina emend. differentiated from P. syringae pv. maculicola strain CFBP 1657Pt within the studied collection. Seventeen carbon sources evaluated and P. syringae pv. tomato strain CFBP 2212Pt (G3) by ice by the Biotype 100 system were assimilated by both P. cannabina nucleation ability and lack of assimilation of meso-tartrate, and P. syringae pv. alisalensis strains tested, whereas 50 other trans-aconitate, and DL-glycerate. carbon sources did not allow growth of these strains The P. cannabina strains tested (CFBP 2341T, CFBP 1619, CFBP (Supplemental Table 1). 1631, LMG 5540, and LMG 5650) reacted identically in 89.5% of Pseudomonas cannabina emend. (including pathogens of Can- the 124 phenotypic tests conducted. The type strain CFBP 2341T nabis sativa and crucifers, i.e. P. cannabina and P. syringae differed from at least one of the other P. cannabina strains for ARTICLE IN PRESS

112 C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 assimilation of esculine, L(+) arabinose, D(+) malate, D-galactur- P. syringae and rRNA I groups. However, the 16S rDNA analysis onate, D-alanine, D(+)-xylose, D-tagatose, betain, DL-alpha-amino- failed to clearly resolve the relationships between P. syringae pv. n-butyrate, succinate, fumarate, and malonate. Additionally, the alisalensis and other organisms in this group. Ribotyping or reactions varied for polypectate hydrolysis at pH 5. sequencing of the 16S—23S gene cluster may have been more Among the 99 carbon sources tested with the biotype 100, useful in resolving these relationships [19,27]. Strains within a Pseudomonas syringae pv. alisalensis CFBP 6866Pt was able to single species rarely differ in their 16S rDNA sequences by more assimilate 41 carbon sources whereas P. cannabina CFBP 2341T than 1.3%. The 16S rDNA sequences of P. syringae pv. alisalensis was only able to assimilate 22 carbon sources. Overall, these two and P. cannabina, differed by only 0.8%. Although these data would strains differed by 28 characters (23 biotype 100 tests and 5 other have been insufficient to propose the transfer of P. syringae tests). In contrast, P. syringae pv. alisalensis CFBP 6866Pt differed pv. alisalensis to P. cannabina, they do not contradict it. from P. syringae pv. maculicola CFBP 1657Pt in only 13 phenotypes The genetic data presented here clearly indicate that the (9 biotype 100 tests and 4 other tests). Differences in the crucifer pathogen P. syringae pv. alisalensis is a member of assimilation of numerous carbon sources can differentiate these P. cannabina. DNA/DNA hybridization levels were consistently two pathogens. greater than 70% in pair-wise and reciprocal comparisons Characters that differentiate P. cannabina and P. syringae pv. between P. syringae pv. alisalensis and P. cannabina. In all but alisalensis are given in pathovar descriptions and in Table 4. All one case, in which the hybridization level was 69%, the average five strains of P. cannabina produce a brown pigment on KMB hybridization levels were above the convention of 70% or greater which is not produced by strains of P. syringae pv. alisalensis or for delineating bacterial species [39,47]. The 70 % DNA/DNA other P. syringae strains evaluated. Of the organisms tested, only hybridization threshold was among the criteria used by Gardan strains of P. syringae pv. alisalensis (both broccoli raab and arugula et al. [19] to support the elevation of P. syringae pv. cannabina to strains) were sensitive to the bacteriophage PBS1 (Table 4). None P. cannabina. of the Pseudomonas cannabina strains evaluated, nor pathotype Additional genetic data support the transfer of P. syringae strains of P. syringae pv. syringae, P. syringae pv. tomato, and pv. alisalensis to P. cannabina. MLST is particularly useful in P. syringae pv. maculicola were sensitive to the PBS1. understanding phylogeny because genetic distance can be analyzed based on several independent loci. The sequences of Pathogenicity the genes used in MLST of P. syringae pv. alisalensis strains and P. cannabina were very similar ranging from 95.9% (kup) to 99.9 % The host range of Pseudomonas cannabina strain CFBP 2341T (gyrB) DNA identity. Specifically, the DNA identity of 99.9% in the was significantly different from the host range of P. syringae pv. gyrB fragment corresponds well to relationships determined by alisalensis (Table 4, Supplemental Fig. 1). Whereas strain CFBP DNA/DNA hybridization [45,52]. Among species in the Bacillus 2341T caused multiple 1–2 mm necrotic lesions with yellow halos subtilis group, gyrB sequences of 95.5–100% identity corresponded on all inoculated leaves of Cannabis sativa, as verified by rep-PCR, to DNA/DNA hybridization levels of 70–100% [45]. no disease was apparent on leaves inoculated with P. syringae pv. In addition to sequence data, relationships among rep-PCR alisalensis CFBP 6866Pt. Likewise P. syringae pv. alisalensis strains DNA fragment banding patterns for P. cannabina and P. syringae CFBP 6867, CFBP 6869, CFBP 6873, and CFBP 6875 were not pv. alisalensis indicated that these organisms are more similar to pathogenic to C. sativa in the two experiments conducted. each other than to other members of P. syringae. Three distinct but All P. syringae pv. alisalensis strains cause a bacterial blight on similar rep-PCR DNA fragment banding patterns were obtained broccoli raab (cv ‘Sorento’ and cv ‘Spring’) and individual small for P. cannabina strains and P. syringae pv. alisalensis strains. DNA lesions on oats (cv ‘Montezuma’; Table 4, Supplemental Fig. 1 banding patterns from P. cannabina and P. syringae pv. alisalensis [14]). Minor symptoms including diffuse yellowing and small strains had little similarity to those of P. syringae pv. syringae, lesions were occasionally detected on broccoli raab or oat plants P. syringae pv. tomato or P. syringae pv. maculicola. The inoculated with P. cannabina strains (CFBP 2341T and CFBP 1631), P. cannabina and P. syringae pv. alisalensis DNA banding patterns but P. cannabina was never isolated from these symptoms and corresponded to the MLST sequence types assigned to these was therefore not considered pathogenic on these hosts. None of strains. Correlation between rep-PCR and MLST are rarely the pathotype strains of P. syringae pv. syringae, P. syringae reported. However, similar clades have been reported from pv. tomato, and P. syringae pv. maculicola caused symptoms on analysis of rep-PCR and MLST data for P. syringae pv. avellanae broccoli raab or oats [14]. strains from Italy and Greece [38,46]. Moreover, the kup locus that distinguished strains within P. syringae pv. alisalensis and displayed the lowest percentage of DNA identity between Discussion P. cannabina and P. syringae pv. alisalensis is more diverse than other MLST loci [54] and thus gives an unusually high resolution Analysis of 16S rDNA sequences is useful for the development to differentiate between very similar strains. of evolutionary inferences at the genus level but has proven to be Additional genetic data support the transfer of P. syringae pv. less useful for the resolution of relationships at the species and alisalensis to P. cannabina. Pseudomonas syringae pv. alisalensis pathovar levels in the genus Pseudomonas [33,53]. Specifically, the grouped with P. cannabina when relationships were analyzed by limited sequence variation in the 16S rDNA gene among species AFLP (Manceau et al., unpublished). and pathovars within the P. syringae group of Anzai et al. [5] and The fatty acid profiles for all strains tested were indicative of rRNA group I of Palleroni [17]) is not particularly useful for fluorescent pseudomonads in subgroup 1a as described by Stead resolving the relationships among these taxa [53]. Reliance on a [40]. Subgroup 1a is characterized by profiles having 12:0 2-OH single locus or single method for deducing phylogeny is less and 12:0 3-OH, and less than 6% 10:0 3-OH. Similar to the findings sound than analysis of multiple types of data and proposed of others, in this study the fatty acid profiles were not phylogenetic relationships among pseudomonads can be signifi- discriminatory at subspecific ranks [40,43] and did not differ- cantly different depending on the genes or regions used to infer entiate P. cannabina and P. syringae pv. alisalensis from related the relationships [2,53]. Moreover, 16S rDNA has less resolving species and pathovars in this study. The fatty acid profiles of the power than either rpoDorgyrB [53]. In this study, the 16S rDNA two P. cannabina strains tested (CFBP 2341T and CFBP 1631) sequences of P. syringae pv. alisalensis clearly fell within the lacked 18:1 o7c 11-methyl, thus differed from P. syringae. ARTICLE IN PRESS

C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 113

Additionally, the absence of 17:0 iso in the profiles of P. cannabina range than P. cannabina we propose that the emended and P. syringae pv. alisalensis strains distinguished these from P. cannabina be split into the two pathovars P. cannabina P. syringae pv. syringae and P. syringae pv. maculicola which pv. cannabina pv. nov. and P. cannabina pv. alisalensis comb. nov produced 17:0 iso. Pseudomonas syringae pv. tomato also lacked (see descriptions and pathotype strains below). this fatty acid and could not be distinguished from the strains of Prior to the publication by Cintas et al. [14] describing interest by this criterion. P. syringae pv. alisalensis, all new bacterial diseases on crucifers Pathovars of P. syringae sensu lato are not yet identifiable by caused by fluorescent pseudomonads were reported to be caused means of routine biochemical tests [6,19] and it was suspected by P. syringae pv. maculicola. However, Pseudomonas syringae that many of the pathovars were synonyms [11,18,55]. Of the nine pv. alisalensis has been reported from crucifers across the US genomospecies designated from pathovars of P. syringae, only two [9,10,12,13,25,26]. Distinguishing between these two pathogens is genomospecies were elevated to species because they could be important because leaf symptoms caused by P. syringae clearly distinguished using routine biochemical tests. Pseudomo- pv. alisalensis aremoreseverethanthosecausedbyP. syringae pv. nas cannabina was elevated from P. syringae pv. cannabina because maculicola and the significant differences in the host ranges of the it differed phenotypically from the other genomospecies of pathogens affect disease management strategies involving crop P. syringae by means of not assimilating fumarate, D(+)-malate, rotation practices [14,15]. Additionally, P. syringae pv. alisalensis has succinate, and DL-lactate. In this study, we tested four additional not yet been reported from outside the USA, and accurate strains of P. cannabina and confirmed lack of assimilation of identification is important to plant protection agencies. Although DL-lactate although we found that assimilation of D(+)-malate, Pseudomonas syringae pv. alisalensis can be easily distinguished from fumarate, and succinate was variable among these strains. Pseudomonas syringae pv. maculicola in standard biochemical and However, P. syringae pv. alisalensis assimilated fumarate, succi- genetic bioassays [14], P. syringae pv. alisalensis is phenotypically nate, and D(+)-malate. Although the previous description of the more similar to P. syringae pv. maculicola than P. cannabina. species P. cannabina indicates otherwise, P. cannabina strains from Pseudomonas syringae pv. alisalensis differs from P. syringae pv. C. sativa assimilate L(-)malate and D-glucuronate and did not maculicola in only 10.5% of the characters tested whereas it differs assimilates L-histidine. Additionally, the inclusion of Pseudomonas from P. cannabina in 22.6 % of the phenotypic characters (Supple- syringae pv. alisalensis strains in P. cannabina will significantly mental Fig. 2). This helps to explain why P. syringae pv. alisalensis had expand the number of carbon sources utilized by at least some not previously been differentiated from P. syringae pv. maculicola strains of this pathogen because Pseudomonas syringae [14]. Consequently, some historical strains including CFBP 1637, pv. alisalensis assimilated 74% more of the carbon sources identified as P. syringae pv. maculicola in the literature, are identical evaluated than P. cannabina assimilated. to P. syringae pv. alisalensis (Bull et al., unpublished). Based on the genetic evidence discussed above we propose to The data from this manuscript support the hypothesis that transfer P. syringae pv. alisalensis from P. syringae to P. cannabina P. syringae pv. alisalensis is a member of P. cannabina. Previously, and emend the description of P. cannabina (below). only five strains of P. cannabina were accessible in international Pseudomonas cannabina (ex Suticˇ ˇ & Dowson 1959) Gardan, public culture collections and these strains appear to be identical Shafik, Belouin, Brosch, Grimont & Grimont 1999 (formerly according to rep-PCR analysis. There are currently 137 P. syringae P. syringae pv. cannabina), the causal agent of bacterial leaf spot pv. alisalensis strains from a variety of hosts and locations in the and ulcer stripe of hemp was originally isolated from hemp culture collection at the USDA/ARS in Salinas, CA and a (Cannabis sativa L.) in ex-Yugoslavia in 1955 [41] and subse- representative subset in international public culture collections. quently from Italy, Germany, Hungary, Bulgaria, Rumania and the This research significantly expands our understanding of the ex-USSR during the 1950s and 1960s [20,21,24,31]. There have species P. cannabina and evolution within this species. The data been no new reports of the disease or isolations of the pathogen also indicate the need for further research to elucidate the from this host since the 1960s. Thus, the diversity of strains of P. phylogenetic relationships between these organisms and geno- cannabina available for research is limited to the five strains mospecies 3 of P. syringae as defined by Gardan et al. [19]. available in publicly accessible culture collections. More recent research has not increased the number of strains available because the strains reported are not available, were misidentified Emended description of Pseudomonas cannabina or are one of the five strains available in public culture collections (Gardan et al., 1999). (although alternatively labeled [42,48,49]). Because P. cannabina and P. syringae pv. alisalensis have been Pseudomonas cannabina (can.na’bi.na. L. fem. Adj. cannabina isolated from different hosts, yet are similar based on genetic and pertaining to Cannabis, the generic name of the host plant, phenotypic data, it was important to investigate the host range of Cannabis sativa L.) these organisms in order to complete the analysis of their Cells are Gram-negative rods that are l.1–3.0 mm wide  3.0– pathovar status. Pathovars are ‘‘a set of strains with the same or 4.0 mm long and motile by means of one to four polar flagella. similar characteristics, differentiated at the infrasubspecific level Colonies have a grey color and are slightly convex on YPGA and from other strains of the same species or subspecies on the basis produces a fluorescent pigment on King’s B medium. Metabolism of distinctive pathogenicity to one or more plant hosts’’ according is respiratory. Results of LOPAT tests are +(v), –. –. – and + [29]. to The International Standards for Naming Pathovars of Phyto- Cell wall fatty acids include 12:0 2-OH and 12:0 3-OH, and less pathogenic Bacteria [11,18,56]. Here we demonstrated that than 6% 10:0 3-OH, but 17:0 iso is absent. Nitrate is not reduced. P. cannabina and P. syringae pv. alisalensis have significantly Hydrolysis of starch is negative. Arginine test (Thornley), indole different host ranges. Pseudomonas cannabina causes leaf spots on production, DNase activity, and gelatin hydrolysis are negative. C. sativa whereas P. syringae pv. alisalensis does not. Additionally, The species hydrolyses tween 80, nucleates ice and assimilates P. syringae pv. alisalensis causes disease on broccoli raab and oats, D-glucose, D-fructose, D-galactose, D-mannose, D-ribose, glycerol, whereas P. cannabina does not. Neither of the pathogens produced D-saccharate, mucate, L(-)malate, citrate, D-glucuronate, D-gluco- lesions on hops (Bull and Gent, unpublished), confirming the nate, L-aspartate, L-glutamate, L-proline, L-alanine and L-serine. findings of Suticˇ ˇ and Dowson [41] that P. cannabina is not a Assimilation varied or was negative, respectively, for 32 and 50 pathogen on hops or other members of the Cannabinaceae. of the remaining carbon sources of the Biotype 100 strips Because P. syringae pv. alisalensis has a significantly different host (bioMe´rieux). These organisms contain genes for the production ARTICLE IN PRESS

114 C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 of coronatine (clf) and ethylene (eth). The DNA G+C content is and analysis database and website for plant-associated microbes. Phyto- 60.2 mol%. The type strain is CFBP 2341T=ICMP 2823T =NCPPB pathology 100, 208–215. T [5] Anzai, Y., Kim, H., Park, J.Y., Wakabayashi, H., Oyaizu, H., 2000. Phylogenetic 1437 . affiliation of the pseudomonads based on 16S rRNA sequence. Int. J. Syst. Evol. Microbiol. 50, 1563–1589. Description of Pseudomonas cannabina pv. cannabina pv. nov. [6] Bradbury, J.F., 1986. Guide to Plant Pathogenic Bacteria. CAB International, Slough, UK (332 pp). [7] Brenner, D.J., McWhorter, A.C., Leete Knutson, J.K., Steigerwalt, A.G., 1982. Description is the same as is given for P. cannabina. This Escherichia vulneris: a species of Enterobacteriaceae associated with human pathogen produces necrotic halos on Cannabis sativa; does not wounds. J. Clin. Microbiol. 15, 1133–1140. [8] Brosius, J., Palmer, M.L., Kennedy, P.J., Noller, H.F., 1978. Complete nucleotide cause disease on broccoli raab (cv Sorento) or on oats (Avena sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc. Natl. Acad. sativa cv. Montezuma). All strains produce a brown pigment on Sci. USA 75, 4801–4805. King B medium and produce levan from sucrose. They do not [9] Bull, C.T., Goldman, P., Koike, S.T., 2004. Bacterial blight on arugula, a new disease caused by Pseudomonas syringae pv. alisalensis in California. Plant Dis. produce 18:1 o7c 11-methyl and are not lysed by bacteriophage 88, 1384. PBS1. Strains can be differentiated from other pathovars by rep- [10] Bull, C.T., Goldman, P.H., Morris, N.C., Koike, S.T., Kobayashi, D.Y., 2004. PCR using the BOX-A1R primers and by 16S rDNA sequences. The Expanded host and geographic range of Pseudomonas syringae pv alisalensis. pathotype strain is CFBP 2341Pt =ICMP 2823Pt =NCPPB 1437Pt. Phytopathology 94, S12. [11] Bull, C.T., De Boer, S.H., Denny, T.P., Firrao, G., Fischer-Le Saux, M., Saddler, G.S., Scortichini, M., Stead, D.E., Takikawa, Y., 2008. Demystifying nomen- Description of Pseudomonas cannabina pv. alisalensis comb. nov. clature of bacterial plant pathogens. J. Plant Pathol. 90, 403–417. [12] Bull, C.T., du Toit, L.J., 2009. First report of bacterial blight on conventionally and organically grown arugula in Nevada caused by Pseudomonas syringae pv. Description is the same as is given for P. cannabina. This alisalensis. Plant Dis. 93, 109. pathogen produces a severe blight on broccoli raab (cv ‘Sorento’ or [13] Cintas, N.A., Koike, S.T., Bull, C.T., 2001. Emerging bacterial pathogens of crucifers in the Salinas Valley of California. In: De Boer, S.H. (Ed.) Plant ‘Spring’) and minor spots on oats (Avena sativa cv. ‘Montezuma’) Pathogenic Bacteria. Kluewer Academic Publishers, Dordrecht, pp. and does not cause disease on Cannabis sativa. All strains tested 272–274. produce 18:1 o7c 11-methyl and are lysed by bacteriophage [14] Cintas, N.A., Koike, S.T., Bull, C.T., 2002. A new pathovar, Pseudomonas syringae pv. alisalensis pv. nov, proposed for the causal agent of bacterial PBS1. Strains assimilate sucrose, D(+) arabitol, myo-inositol, blight of broccoli and broccoli raab. Plant Dis. 86, 992–998. D-mannitol, D(-) tartrate, D(+) malate, D-galacturonate, protoca- [15] Cintas, N.A., Koike, S.T., Bunch, R.A., Bull, C.T., 2006. Holdover inoculum of techuate, (-) quinate, D-alanine, a-ketoglutarate. This pathovar Pseudomonas syringae pv. alisalensis from broccoli raab causes disease in can be differentiated from other pathovars by rep-PCR using the subsequent plantings. Plant Dis. 90, 1077–1084. [16] Crosa, J.H., Brenner, D.J., Falkow, S., 1973. Use of a single strand-specific BOX-A1R primers and by 16S rDNA sequences. The pathotype nuclease for analysis of bacterial and plasmid deoxyribonucleic acid homo- strain is ATCC BAA-566Pt= CFBP 6866Pt= ICMP 15200Pt (originally and heteroduplexes. J. Bacteriol. 115, 904–911. BS91Pt). [17] DeParasis, J., Roth, D.A., 1990. Nucleic acid probes for identification of phytobacteria: identification of genus-specific 16S rRNA sequences. Phyto- pathology 80, 618–621. [18] Dye, D.W., Bradbury, J.F., Goto, M., Hayward, A.C., Lelliot, R.A., Schroth, M.N., Acknowledgements 1980. International standards for naming pathovars of phytopathogenic bacteria and a list of pathovar names and pathotype strains. Rev. Plant Pathol. 59, 153–168. The authors are thankful to Drs. Pat Wechter and Yuichi [19] Gardan, L., Shafik, H., Belouin, S., Broch, R., Grimont, F., Grimont, P.A.D., 1999. Takemoto for their helpful comments in review of the manuscript. DNA relatedness among the pathovars of Pseudomonas syringae and The authors would like to thank Polly Goldman and Sophie description of Pseudomonas tremae sp. nov. and Pseudomonas cannabina sp. nov. (ex Suticˇ ˇ and Dowson 1959). Int. J. System. Bact. 49, 469–478. Bonneau for their technical support of this research and Nicole [20] Gitman, L.S., 1968. Little-known diseases of hemp. Zashch-Rast.—Mosk. 13, Morris, Freddy Rosales, Robert Lomeli, Stacy Mauzey and Teresa 44–45. Jardini for their labor. [21] Goidanich, G., Ferri, F., 1959. La batteriosi della canapa da Pseudomonas ˇ The mention of a trade name, proprietary product, or vendor does cannabina Suticˇ et Dowson var. italica Dowson. Phytopath. Z. 37, 21–32. [22] Hwang, M.S.H., Morgan, R.L., Sarkar, S.F., Wang, P.W., Guttman, D.S., 2005. not constitute an endorsement, guarantee or warranty by the United Phylogenetic characterization of virulence and resistance phenotypes of States Department of Agriculture and does not imply its approval or Pseudomonas syringae. Appl. Environ. Microbiol. 71, 5182–5191. the exclusion of these or other products that may be suitable. [23] King, E.O., Ward, M.K., Raney, D.E., 1954. Two simple media for the demonstration of pyocyanin and fluorescin. J. Lab. Clin. Med. 44, 301–307. [24] Klement, Z., Kaszonyi, S., 1960. A new bacterial disease of hemp in Hungary. Novenytermeles 9, 153–158. Appendix A. Supplementary materials [25] Koike, S.T., Kammeijer, K., Bull, C.T., O’Brien, D., 2006. First report of bacterial blight of romanesco cauliflower (Brassica oleracea var. botrytis) caused by Pseudomonas syringae pv. alisalensis in California. Plant Dis. 90, 1551. Supplementary data associated with this article can be found [26] Koike, S.T., Kammeijer, K., Bull, C.T., O’Brien, D., 2007. First report of bacterial in the online version at doi:10.1016/j.syapm.2010.02.001. blight of rutabaga (Brassica napus var. napobrassica) caused by Pseudomonas syringae pv. alisalensis in California. Plant Dis. 91, 112. [27] Kong, H., Blackwood, C., Buyer, J.S., Gulya Jr., T.J., Lydon, J., 2005. The genetic characterization of Pseudomonas syringae pv. tagetis based on the 16S–23S rDNA intergenic spacer regions. Biol. Cont. 32, 356–362. References [28] Lane, D.J., 1991. 16S/23S rRNA sequencing. In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley and [1] Achouak, W., Sutra, L., Heulin, T., Meyer, J.M., Fromin, N., Degraeve, S., Sons, New York, NY, pp. 115–175. Christen, R., Gardan, L., 2000. Pseudomonas brassicacearum sp. nov. and [29] Lelliott, R.A., Billing, E., Hayward, A.C., 1966. A determinative scheme for the Pseudomonas thivervalensis sp. nov., two root-associated bacteria isolated fluorescent plant pathogenic pseudomonads. J. Appl. Bacteriol. 29, 470–489. from Brassica napus and Arabidopsis thaliana. Int. J. Syst. Evol. Microbiol. 50, 9– [30] McCulloch, L., 1911. A spot disease of cauliflower. U.S. Dept. Agric. Bur. Plant 18. Ind. Bull. 225, 7–15. [2] Ait Tayeb, L., Ageron, E., Grimont, F., Grimont, P.A.D., 2005. Molecular [31] McEno, J. (Ed.) 1991) Cannabis Ecology: a Compendium of Diseases and Pests. phylogeny of the genus Pseudomonas based on rpoB sequences and Amrita Press, Middlebury, VT. application for the identification of isolates. Res. Microbiol. 156, 763–773. [32] de Meijer, E.P.M., van Soest, L.J.M., 1992. The CPRO Cannabis germplasm [3] Altschul, S.F., Madden, T.L., Schaffer,¨ A.A., Zhang, J., Zhang, Z., Miller, W., collection. Euphytica 62, 201–211. Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein [33] Moore, E.R.B., Mau, M., Arnscheidt, A., Bottger,¨ E.C., Hutson, R.A., Collins, M.D., database search programs. Nucl. Acids Res. 25, 3389–3402. van de Peer, Y., de Wachter, R., Timmis, K.N., 1996. The determination and [4] Almeida, N., Yan, S., Cai, R., Clarke, C.R., Morris, C.E., Schaad, N.W., Schuenzel, comparison of the 16S rRNA gene sequences of species of the genus E.L., Lacy, G.H., Sun, X., Jones, J.B., Castillo, J.A., Bull, C.T., Leman, S., Guttman, Pseudomonas (senus stricto) and estimation of the natural intragenic D.S., Setubal, J.C., Vinatzer, B.A., 2010. PAMDB, a multilocus sequence typing relationships. Syst. Appl. Microbiol. 19, 478–492. ARTICLE IN PRESS

C.T. Bull et al. / Systematic and Applied Microbiology 33 (2010) 105–115 115

[34] Rademaker, J.L.W., Louws, F.J., De Bruijn, F.J., 1998. Characterization of the [46] Wang, P.W., Morgan, R.L., Scortichini, M., Guttman, D.S., 2007. Convergent diversity of ecologically important microbes by rep-PCR genomic fingerprint- evolution of phytopathogenic pseudomonads onto hazelnut. Microbiology ing. Mol. Microb. Ecol. Man. 3.4.3, 1–27. 153, 2067–2073. [35] Sarkar, S.F., Guttman, D.S., 2004. Evolution of the core genome of [47] Wayne, L.G., Brenner, D.J., Colwell, R.R., Grimont, P.A.D., Kandler, O., Pseudomonas syringae, a highly clonal, endemic plant pathogen. Appl. Krichevsky, M.I., Moore, L.H., Moore, W.E.C., Murray, R.G.E., Stackebrandt, Environ. Microbiol. 70, 1999–2012. E., Starr, M.P., Truper,¨ H.G., 1987. Report of the ad hoc committee on [36] Sato, M., Watanabe, K., Yazawa, M., Takikawa, Y., Nishiyama, K., 1997. reconciliation of approaches to bacterial systematics. Int. J. Syst. Bacterol. 37, Detection of new ethylene producing bacteria, Pseudomonas syringae pvs. 463–464. cannabina and sesami, by PCR amplification of genes for the ethylene-forming [48] Weingart, H., Volksch,¨ B., 1997. Ethylene production by Pseudomonas syringae enzyme. Phytopathology 87, 1192–1196. pathovars in vitro and in planta. Appl. Environ. Microbiol. 63, 156–161. [37] Schaad, N.W., Jones, J.B., Chun, W. (Eds.) (2001) Laboratory Guide for [49] Weingart, H., Volksch,¨ B., Ullrich, M.S., 1999. Comparison of ethylene Identification of Plant Pathogenic Bacteria Third Edition APS Press, St. Paul, MN. production by Pseudomonas syringae and Ralstonia solanacearum. Phyto- [38] Scortichini, M., Dettori, M.T., Marchesi, U., Palombi, M.A., Rossi, M.P., 1998. pathology 89, 360–365. Differentiation of Pseudomonas avellanae strains from Greece and Italy by [50] Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S ribosomal rep-PCR genomic fingerprinting. J. Phytopathol. 146, 417–420. DNA amplification for phylogenetic study. J. Bacteriol. 173, 697–703. [39] Stackebrandt, E., Frederiksen, W., Garrity, G.M., Grimont, P.A.D., Kampfer,¨ P., [51] Williams, P.H., Keen, N.T., 1966. Bacterial blight of radish. Plant Dis. 50, Maiden, M.C.J., Nesme, X., Rossello´ -Mora, R., Swings, J., Truper,¨ H.G., Vauterin, 192–195. L., Ward, A.C., Whitman, W.B., 2002. Report of the ad hoc committee for the [52] Yamamoto, S., Bouvet, P.J.M., Harayama, S., 1999. Phylogenetic structures of re-evaluation of the species definition in bacteriology. Int. J. Syst. Evol. the genus Acinetobacter based on gyrB sequences: comparison with the Microbiol. 52, 1043–1047. grouping by DNA-DNA hybridization. Int. J. Syst. Bacteriol. 49, 87–95. [40] Stead, D.E., 1992. Grouping of plant-pathogenic and some other Pseudomonas [53] Yamamoto, S., Kasai, H., Arnold, D.L., Jackson, R.W., Vivian, A., Harayama, S., spp. by using cellular fatty acid profiles. Int. J. Syst. Bacteriol. 42, 281–295. 2000. Phylogeny of the genus Pseudomonas: intrageneric structure recon- [41] Suticˇ ˇ, D., Dowson, W.J., 1959. An investigation of a serious disease of Hemp structed from the nucleotide sequences of gyrB and rpoD genes. Microbiology (Cannabis sativa L.) in Jugoslavia. Phytopath. Z. 34, 307–314. 146, 2385–2394. [42] Terasawa, N., Sugiyama, A., Murata, M., Homma, S., 2002. Isolation of a [54] Yan, S., Liu, H., Mohr, T.J., Jenrette, J., Chiodini, R., Zaccardelli, M., Setubal, J.C., microorganism to oxidize 5-Hydroxymethylfurfural. Food Sci. Technol. Res. 8, Vinatzer, B.A., 2008. Role of recombination in the evolution of the model 28–31. plant pathogen Pseudomonas syringae pv. tomato DC3000, a very atypical [43] Vancanneyt, M., Witt, S., Abraham, W.R., Kersters, K., Frederickson, H.L., 1996. tomato strain. Appl. Environ. Microbiol. 74, 37–38. Fatty acid content in whole-cell hydrolysates and phospholipid fractions of [55] Young, J.M., Takikawa, Y., Gardan, L., Stead, D.E., 1992. Changing concepts in pseudomonads: a taxonomic evaluation, System. Appl. Microbiol. 19, the taxonomy of plant pathogenic bacteria. Annu. Rev. Phytopathol. 30, 528–540. 67–105. [44] Vinatzer, B.A., Bull, C.T., 2009. The impact of genomic approaches on our [56] Young J.M., Bull C.T., De Boer S.H., Firrao G., Gardan L., Saddler G.E., Stead D.E. understanding of diversity and taxonomy of plant pathogenic bacteria. In: and Takikawa Y., 2001. International standards for naming pathovars of Jackson, R.W. (Ed.) Plant Pathogenic Bacteria: Genomics and Molecular phytopathogenic bacteria. /http://www.isppweb.org/about_tppb_naming. Biology. Horizon Scientific Press Hethersett, Norwich, UK, pp. 37–61. aspS. [45] Wang, L.T., Lee, F.L., Tai, C.J., Kasai, H., 2007. Comparison of gyrB gene [57] Zhao, Y., Damicone, J.P., Bender, C.L., 2002. Detection, survival, and sources of sequences, 16S rRNA gene sequences and DNA–DNA hybridization in the inoculum for bacterial diseases of leafy crucifers in Oklahoma. Plant Dis. 86, Bacillus subtilis group. Int. J. Syst. Evol. Microbiol. 57, 1846–1850. 883–888.