PECTOBACTERIUM CAROTOVORUM Subsp

Home , Dickeya dadantii, Erwinia, Pectobacterium

Journal of Plant Pathology (2017), 99 (1), 149-160 Edizioni ETS Pisa, 2017 149

PECTOBACTERIUM CAROTOVORUM subsp. ODORIFERUM ON CABBAGE AND CHINESE CABBAGE: IDENTIFICATION, CHARACTERIZATION AND TAXONOMIC RELATEDNESS OF BACTERIAL SOFT ROT CAUSAL AGENTS*

M. Oskiera, M. Kałuz˙na, B. Kowalska and U. Smolin´ska

Research Institute of Horticulture, Konstytucji 3 Maja 1/3, 96-100, Skierniewice, Poland

SUMMARY INTRODUCTION

This study was aimed to isolate, identify and character- Cabbage (Brassica oleracea L. var. capitata L.) and Chi- ize Pectobacterium spp. causing soft rot disease of cabbage nese cabbage (Brassica rapa L. subsp. pekinensis L.) are im- and Chinese cabbage in Central Poland. Of fifty-two plant portant vegetable crops commonly cultivated in Poland. samples of cabbage and Chinese cabbage showing disease The main pathogens of Brassicaceae plants are Pectobacte- symptoms collected in Central Poland from 2007-2010, 542 rium carotovorum subsp. carotovorum (Pcc), Pseudomonas bacterial isolates were obtained. Of isolates 117 caused soft marginalis pv. marginalis, Pseudomonas syringae pv. maculi- rot on cabbage and Chinese cabbage leaves and potato cola and Xanthomonas campestris pv. campestris (Rimmer et slices and showed pectinolytic activity on crystal violet al., 2007). It is also known that Pseudomonas viridiflava can pectate medium. PCR using Y1/Y2 primers specific for occur on Brassicaceae plants, such as Chinese cabbage (Ma- Pectobacterium genus revealed that twenty-three of them ciel et al., 2010). The most common disease of Brassicaceae belonged to this genus. Phenotypic characterization in is soft rot mainly caused by highly pectinolytic bacteria combination with DNA-based typing methods (rep-PCRs) from the genus Pectobacterium (formerly Erwinia). Symp- and sequence analysis of 16S rRNA, and housekeeping toms of bacterial soft rot on cabbage include water-soaked genes gyrB, infB, rpoB, atpD, and rpoS (MultiLocus Se- lesions which later become rotted mass of macerated tis- quence Typing) done for 12 Polish representatives and ref- sue (Rimmer et al., 2007). P. carotovorum may infect the erence strains allowed for their identification and revealed plant through natural openings (in favorable conditions genetic diversity. Additional phylogeny analysis of the rpoS warm temperatures and high humidity), wounds caused by gene for Polish and worldwide well defined Pectobacte- insects or other diseases causal agent or damages caused rium spp. strains allowed to classify obtained strains into by abiotic factors such as low temperatures or mechanical Pectobacterium spp. phylogenetic clusters (PcI-PcV). The damages. Pectobacterium derives from the genus Erwinia presented studies showed that 8 of 12 isolated Pectobac- and was described as separate genus from Erwinia, Pan- terium carotovorum strains belong to the Pectobacterium toea and Brenneria on the basis of phylogenetic analyses carotovorum subsp. carotovorum (Pcc) cluster PcII and of the 16S rDNA sequences (Hauben et al., 1998). Later, other four strains belong to the subspecies Pectobacteri- Gardan et al. (2003) reclassified existed subspecies of Pec- um carotovorum subsp. odoriferum (Pco) cluster PcIII. To tobacterium carotovorum: i.e. atrosepticum, betavasculorum our knowledge, this is the first report that P. carotovorum and wasabiae to species: Pectobacterium atrosepticum, Pec- subsp. odoriferum causes soft rot of cabbage and Chinese tobacterium betavasculorum and Pectobacterium wasabiae cabbage in Central Poland. respectively. Duarte et al. (2004) described Pectobacterium carotovorum subsp. brasiliense as one of the Pectobacterium Keywords: rep-PCR, 16S rRNA, rpoS, MLST, pectino- carotovorum subspecies. Later, Nabhan et al. (2012a) and lytic bacteria. Waleron et al. (2014) described its appearance in Europe. Phylogenetic studies of the P. carotovorum pectinolytic strains performed by Nabhan et al. (2012b) reveals high heterogeneity of the species and divided it into five clusters (PcI to PcV) as follows: Pcc represented by two clus- teres (PcI and PcII), Pectobacterium carotovorum subsp. odoriferum (Pco) (PcIII), P. carotovorum subsp. brasiliense (PcIV) and P. atrosepticum, P. betavasculorum, P. wasabiae. Further, cluster PcV was described as Pectobacterium aroidearum by Nabhan et al. (2013). Corresponding author: M. Oskiera Fax: +48.46.833.31.86 * Supplementary materials can be downloaded at: E-mail: [email protected] http://dx.doi.org/10.4454/jpp.v99i1.3831.s11 150 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160

Several methods are commonly used for detection, were macerated in sterile distilled water. Serial dilutions identification and differentiation of plant pathogenic and of the mixture were plated on NA medium. The obtained pectinolytic bacteria. For many years identification of selected colonies were purified on NA medium, preserved bacteria was based on isolation of viable cells on selective in mixture of glycerol and nutrient broth (1:1) and stored culture agar media followed by serological and biochemi- in −80ºC until use. Reference strains of the Pectobacte- cal analyses, bioassays and microscopic observations in- rium carotovorum subsp. carotovorum (Pcc) (LMG 2431, cluding Gram staining (Schaad et al., 2001). Later, some LMG 2404T), Pectobacterium carotovorum subsp. odorif- improvements such as plating on CVP medium with im- erum (Pco) (LMG 17566T) and some pectinolytic strains of munomagnetic separation (IMS) step prior to plating (Van Pectobacterium atrosepticum (Eca59, Ec32), Pectobacterium Der Wolf and Perombelon, 2002), fatty acid methyl ester aroidearum (Ecc4M, Ea853), Pectobacterium sp. (Eca2M) analysis (FAME) (Dawyndt et al., 2006), immunofluores- and Dickeya dadantii (Ech0201) were included in all analy- cence staining (IF) (Van Vuurde and Van der Wolf, 2002) sis. The data on studied isolates and reference strains are and enzyme-linked immunosorbent assay (ELISA) (Lopez summarized in Table 1. et al., 2002) were developed for Pectobacterium detection. Molecular methods, such as PCR with primers specific for Selection of strains on the basis of their pectinolytic detection of Pectobacterium spp., multiplex-PCR, real-time activity on potato slices, cabbage leaves and CVP me- PCR, padlock probes and LAMP are currently widely dium. All the bacterial isolates were tested on potato tuber adapted and were reviewed by Czajkowski et al. (2015). slices, cabbage and Chinese cabbage leaves. Cabbage or Genus specific PCR primers, such as Y1/Y2 (Darrasse et Chinese cabbage leaves and potato tubers were washed al., 1994) for identification Pectobacterium spp., and species under tap water, disinfected in 70% ethanol for 1 min, specific primers, such as Eca1f/Eca2 (De Boer et al., 1995) then 5 min in 0.5% sodium hypochloride, followed by two and Y45/Y46 (Frenchon et al., 1995) for identification P. washes with sterile water. Fragments of the plants: cabbage atrosepticum are routinely used in many laboratories. As with a diameter of approx. 8 cm or 3 slices of potato 10- for many bacterial pathogens, also for soft rot causal agent, mm thick were placed on the prepared Petri dishes (9 cm fingerprinting methods through the analysis of repetitive in diameter) with wet Whatman filter paper. Each bacte- regions [rep-PCR i.e., Enterobacterial Repetitive Intergenic rial isolate grown 24 h on NA medium was placed with Consensus (ERIC), BOX and Repetitive Extragenic Pal- toothpicks on the injured tissues in 3-4 places (per plate). indromic Elements (REP)] (Weingart and Völksch, 1997) Three plates were assayed per bacterial isolate. The Petri have been successfully used for their identification and dif- dishes were then incubated at 28°C. Tissue maceration was ferentiation within the species and subspecies level (Rezaei determined by visual inspection (cabbage leaves) twice: and Taghavi, 2010). In case of Pectobacterium spp., also the after 48 h and after 5 days, while in the case of potato slices several primer pairs amplifying fragments of 16S rDNA with use of toothpicks after 24 and 48 h. Each assay was and housekeeping genes localized independently in bac- performed three times. Obtained bacterial isolates were al- terial core genome were already published and used for so tested on crystal violet pectate agar (CVP) (Perombelon sequencing (Waleron et al., 2002, 2014; Brady et al., 2008). and Burnett, 1991) with use of sodium polypectate (pectin The sequence analysis of housekeeping genes combined classic kat. CU902; Herbstreith & Fox Corporate Group, with bioinformatics tools (MultiLocus Sequence Typing) Neuenbürg/Württ, Germany), for checking their ability to allowing for identification and genotyping of the soft rot form of liquefied deep pits on the medium surface. bacteria. The objective of this study was the isolation, identification and differentiation of P. carotovorum spp. strains DNA isolation. Bacterial DNA was isolated using the obtained from diseased cabbage and Chinese cabbage in method described by Aljanabi and Martinez (1997). The Poland using phenotypic tests, repetitive-sequence PCRs purity and quantity of isolated DNA were evaluated by and MLST. visualization on 0.8% agarose gel after electrophoresis with sodium-borate (NaB) buffer (Brody et al., 2004) and determined spectrophotometrically by absorbance mea- MATERIALS AND METHODS surement of A260/A280 using a BioPhotometer UV/Vis Spectrophotometer (Eppendorf, Hamburg, Germany). Plant samples and isolation of bacteria. Fifty two Finally, DNA concentration was adjusted to 10 ng/μl by plant samples of the cabbage (25) and Chinese cabbage diluting samples with double distilled water, 2 μl of DNA (27) were collected from the Central Poland (Skierniewice was used as a template for PCR. area) in years 2007-2010. Samples of symptomatic tissue of different organs i.e., leaves with tiny necrotic spots with PCR with primers specific for Pectobacterium chlorotic halo, V-shaped chlorosis, leaves and stem with spp. Selected isolates with pectinolytic activity con- brown surface as soft rot symptoms, or macerated tissue, firmed on cabbage leaves, potato slices and CVP me- were collected. Small pieces from the margin between dium, were tested by PCR using primers Y1 (5’TTAC- healthy and diseased tissue after disinfection with ethanol CGGACGCCGAGCTGTGGCGT3’) and Y2 Journal of Plant Pathology (2017), 99 (1), 149-160 Oskiera et al. 151

Table 1. Origin and collection date of strains obtained and used in this study.

Strain name Host plant Geographic origin Year of isolation 1Kp9 Brassica rapa L. subsp. pekinensis (Lour.) Hanelt Skierniewice 2007 8Kp1 Skierniewice 2007 9Kp6 Łowicz 2007 19Kp1 Skierniewice 2008 30Kp Skierniewice 2010 6K1 Brassica oleracea L. var. capitata L Skierniewice 2007 8K6 Rawa Mazowiecka 2007 12K4 Skierniewice 2007 15K3 Skierniewice 2007 17K7b(2) Bednary 2007 27K1 Skierniewice 2009 Ky5 Skierniewice N/A Ecc4M Zantedeschia spp. Central Poland (A. Mikicin´ski) N/A Ea853 Central Poland (A. Mikicin´ski) N/A Eca2M Zantedeschia spp. Central Poland (A. Mikicin´ski) N/A Eca59 (Eca5/06) Solanum tuberosum N/A (IHAR Młochów) 2006 Ec32 N/A N/A N/A Ech0201 N/A N/A N/A LMG 2431 Allium cepa Japan (M. Goto) 1965 LMG 2404T Solanum tuberosum Denmark (E. Hellmers) 1952 LMG 17566T Cichorium intybus France (R. Samson) 1979

(5’CAGGAAGATGTCGTTATCGCGAGT3’) specific (Fermentas, Vilnius, Lithuania), and 20 ng of DNA tem- for the genus Pectobacterium spp. (Darrasse et al., 1994). plate. Following cycling parameters were used for ERIC- The PCR mixes and cycling conditions were carried out PCR: 4 min at 95°C followed by 35 cycles each of 1 min at according to those used in original paper. The amplified 94°C, 1 min and 30 s at 52°C, 8 min at 65°C and, finally, products were electrophoretically separated in 1.5% aga- 16 min at 65°C; BOX-PCR: 4 min at 94°C followed by 35 rose gel containing ethidium bromide with sodium-borate cycles each of 1 min at 94°C, 2 min at 40°C, 2 min at 72°C (NaB) buffer (Brody et al., 2004) and visualized under UV and, finally, 10 min at 72°C; REP-PCR: 3 min at 94°C fol- light. lowed by 35 cycles each of 30 s at 94°C, 1 min at 52°C, 8 min at 65°C and, finally, 16 min at 65°C. All PCRs (con- Phenotypic characterization. For all pectinolytic Pecto- ducted for each primer pairs separately) were carried out bacterium spp. strains physiological and biochemical test, in a Mastercycler EP gradient thermocycler (Eppendorf, including levan production from sucrose, presence of oxi- Hamburg, Germany). For separation of obtained patterns dase, catalase and arginine dihydrolase, starch hydrolysis, electrophoresis were conducted with 2% agarose gel in nitrate reduction and acid production from D-arabitol, 0.5 × Tris-borate-EDTA (TBE) containing ethidium bro- D-sorbitol, melibiose, D-glucose, lactose, D-mannitol, D- mide and visualized under UV light using the GelDoc- raffinose and D-xylose were carried out according to Lel- It System with VisionWorks LS 6.7.4 (UVP, Upland, CA, liott and Stead (1987) and Schaad et al. (2001). The tests USA). Each obtained band was scored as present (1) or were repeated for each isolate and reference strains at least absent (0) for all isolates to determine a distance matrix. twice. Only reproducible and unambiguous genomic fingerprints (tested by repeating the PCRs at least twice for each iso- Repetitive PCRs (rep-PCRs). For the evaluation of late) were taken into account. Cluster analysis of the binary genetic diversity of the 12 representative strains of previ- data was performed using the NTSYS-pc v.2.10q program ously identified Pectobacterium spp. and 3 reference strains (Exeter Software, Setauket, NY, USA). The SIMQUAL LMG 2404T, LMG 2431 and LMG 17566T three meth- program was used to generate similarity matrix using Jac- ods of repetitive PCR (rep-PCR) were used. Fingerprints card’s coefficient. Unweighted pair-group method using were performed using following primers: ERIC1R and arithmetic averages (UPGMA) within the SAHN mod- ERIC2, one primer BOX and two primers REP1R and ule was used to perform clustering analysis and generate REP2I (Louws et al., 1994; Versalovic et al., 1991, 1994). the dendrogram from rep-PCR similarity matrix. To de- PCR mixture (20 μl) contained: 1× Dream Taq Green buf- termine the goodness of fit of a cluster analysis to a da- fer, 0.25 mM of dNTPs, 0.5 μM of each primer, 4 μg bo- taset Mantel tests were used and the cophenetic correla- vine serum albumin (BSA), 1 U of Taq DNA Polymerase tion coefficient parameter (r) was calculated by compare 152 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160

Table 2. List of names, sequences, annealing temperatures and references of the primers used in this study for amplification and sequencing.

Primer Gene name Annealing temp [°C] Reference Name Sequence (5’-3’) 16S rRNA fD1 AGAGTTTGATCMTGGCTC 56 Weisburg et al., 1991 rP2 ACGGCTACCTTGTTACGACTT gyrB gyrB 01-F TAARTTYGAYGAYAACTCYTAYAAAGT 56 Brady et al., 2008 gyrB 02-R CMCCYTCCACCARGTAMAGTT infB infB 01-F ATYATGGGHCAYGTHGAYCA 56 infB 02-R ACKGAGTARTAACGCAGATCCA rpoB rpoB CM7-F AACCAGTTCCGCGTTGGCCTG 66 rpoB CM31b-R CCTGAACAACACGCTCGGA atpD atpD 01-F RTAATYGGMGCSGTRGTNGAYGT 66 atpD 02-R TCATCCGCMGGWACRTAWAYNGCCTG rpoS rpoS1 ATGAGCCAAAGTACGCTGAA 56 Waleron et al., 2008 rpoS2 ACCTGAATCTGACGAACACG

similarity matrix with cophenetic matix. Statistical signifi- Sequencing and Oligonucleotide Synthesis Lab (IBB PAS, cance was tested using 1000 random permutations within Warsaw, Poland). All amplicons were sequenced from both MXCOMP module. directions with primers used for amplification (Table 2).

Sequence analysis of 16S rDNA and housekeeping Bioinformatic analyses. All obtained sequences of genes. The sequences of primers for amplification and 16S rRNA, gyrB, infB, rpoB, atpD and rpoS were edited in sequencing of 16S rRNA and housekeeping genes frag- CLC GenomicWorkbench 4.0-8.0. Raw sequences were ments: gyrB, infB, rpoB, rpoS, atpD are listed in Table 2. trimmed, assembled from both directions and consensus Twelve representative stains of previously identified Pec- sequence was saved. For each strain consensus sequences tobacterium spp. and additionally 3 reference strains from were obtained individually. All similar raw sequences were international collection LMG 2404T, LMG 2431 and LMG compared, mismatches examined and resulted consensus 17566T and some pectinolytic strains Eca59, Ec32, Ecc4M, sequences were compared with Blastclust (www.toolkit. Ea853, Eca2M and Ech0201 (Table 1) were included into tuebingen.mpg.de/blastclust) and individual sequence the sequence analyses. haplotypes (ST) were selected with 100% identity crite- PCR mixture (20 μl) contained: 1× Taq buffer, 0.25 mM rion. All STs were identified by BLASTn (www.blast.ncbi. dNTPs, 0.5 μM of each primer, 1 U of Dream Taq Poly- nlm.nih.gov/Blast.cgi) cross checks with NCBI GenBank merase (Fermentas, Vilnius, Lithuania), and 20 ng of DNA nucleotide collection (nr/nt). Phylogenetic analyses, based template. 1.5 mM MgCl2 concentration was used for am- on the part of 16S rRNA, housekeeping genes gyrB, infB, plification of gene fragments infB, rpoS and 16S rRNA; rpoB, atpD, rpoS sequences and also based on concatenated 2.0 mM MgCl2 was used for amplification of gyrB and atpD data of four housekeeping genes fragments (gyrB, rpoB, gene fragments. The following conditions were applied: atpD, rpoS), were performed for 20 Pectobacterium spp. initial denaturation at 94°C for 3 min, followed by 35 cycles strains and one D. dadantii strain used as an outgroup of each: denaturation at 94°C for 30 s, annealing at 56°C (Table 1; Table 4). Phylogenetic analyses were performed, for 16S rRNA, gyrB, infB and rpoS and 66°C for rpoB and as described more detailed below, according to bayesian atpD for 30 s, extension at 72°C for 1 min and single final procedure with use of the linux standalone versions of the extension cycle 7 min at 72°C. PCRs (conducted for each following programs: ClustalX (Thompson et al., 1997), gene separately) were carried out in a Mastercycler EP gra- GBlocks (Castresana, 2000), jModelTest (Darriba et al., dient thermocycler (Eppendorf, Hamburg, Germany). To 2012) and MrBayes 3.2.2 x64 (Ronquist et al., 2012). Also, confirm DNA amplification 1 μl of each products obtained Compare2Trees: pairwise comparison of phylogenies (Nye for each gene and isolate were separated on 1.5% agarose et al., 2005) program was used to compare topologies of gel with sodium-borate (NaB) buffer (Brody et al., 2004). phylogenetic trees obtained with single locus sequences to Gels stained with ethidium bromide were visualized under trees obtained with concatenated data sequences. Over- UV light using the GelDoc-It System with VisionWorks LS all topological scores were noticed and the locus with the 6.7.4 (UVP, Upland, CA, USA). Prior to the sequencing, highest topological similarity tree (to the concatenated products were cleaned using enzymatic ExoSAP clean-up data tree) was chosen for further phylogenetic analysis kit according to the manufacturer’s instructions (Affym- with worldwide characterized strains. DNAsp 5.10.01 (Li- etrix, Santa Clara, CA, USA). PCR products were diluted brado and Rozas, 2009) was used to perform analysis of with H2O and mixed with 5 pmol of primer up to 6 μl, and sequence data. Diversity indices, such as the number of sequenced by Genomed S.A. (Warsaw, Poland) or by DNA alleles, G/C content, number of polymorphic sites, number Journal of Plant Pathology (2017), 99 (1), 149-160 Oskiera et al. 153 of parasimony informative sites, nucleotide diversity [π] average number of nucleotide differences nucleotide dif- ference per site, [θ] average number of nucleotide differences per site, [k] average number of nucleotide differences per sequence and Tajima’s D were calculated. All calculations were performed for all collected from Bras- sicaceae Pectobacterium carotovorum isolates and other Pectobacterium spp. pectinolytic strains (20 isolates listed in Tables 1 and 4), and separately for the isolates belonging to the Pcc (10 isolates) and Pco (5 isolates). Addition- ally, sequences of the rpoS gene locus were chosen to define phylogenetic relationship with other Pectobacterium spp. strains (known worldwide and classified in details). Dickeya dadantii sequence (Supplementary Table 1) was used as an outgroup, and also sequences used by Nah- ban et al. (2012a) and Waleron et al. (2014) were retrieved from NCBI (www.ncbi.nlm.nih.gov/genbank). Sequences were assembled with ClustalX (Thompson et al., 1997) and manually edited in CLC Genomic Workbench 8.0 (CLCBio, Aarhus, Denmark). Sequence sets of equal Fig. 1. Dendrogram generated using Jaccard’s coefficient and the UPGMA clustering method. Binary matrices for Pcc and length were processed with BLASTclust (toolkit.tuebin- Pco strains were constructed based on evaluation of rep-PCR gen.mpg.de/blastclust) with clustering of 100% iden- amplicons. Cluster analysis of the binary data was performed tity sequence. A single GenBank sequence was selected using NTSYS-pc 2.1 software. Similarity matrices were gener- for each sequence type (NCBI GenBank accessions are ated using Jaccard’s coefficient and an unweighted pair-group provided at Fig. 3). Final block alignment was prepared method using arithmetic averages (UPGMA) was used to with a less stringent selection option using GBlocks (Cas- generate the dendrogram. Pcc, P. carotovorum subsp. carotovo- tresana, 2000). Alignment length was 776 base pair and rum; Pco, P. carotovorum subsp. odoriferum. consist of 52 sequences. The substitution model, nucleotide frequencies and substitution values were estimated results in the PCR with primers Y1/Y2 primers (Darrasse with jModelTest 0.1.1 (Darriba et al., 2012), with the AICc et al., 1994) which allowed for their classification to Pecto- selection criterion (model GTR + I + G, -lnL = 2813.1496, bacterium genus. Identified isolates were obtained from 12 K = 112, freqA = 0.2531, freqC = 0.2339, freqG = 0.2744, different plant samples, due to that 12 Pectobacterium rep- freqT = 0.2386, ti/tv = 4.6967, gamma = 0.5032). Metropolis- resentative strains (1Kp9, 8Kp1, 9Kp6, 19Kp1, 30Kp, 6K1, coupled Markov chain Monte Carlo (MCMCMC) analy- 8K6, 15K3, 12K4, 17K7b2, 27K1, Ky5) were selected for sis was performed with 2 runs for 10 million generations further phenotypic and genetic analyzes (Table 1). with 4 chains with heating coefficient λ = 0.1 with MrBayes 3.2.2 ×64 (Ronquist et al., 2012). Phenotypic characterization. On NA medium Pectobac- terium colonies were irregular, shiny, transparent to white- creamy and on NA with 5% of sucrose were negative for RESULTS levan formation. Identified bacteria were Gram-negative rods, catalase, arginine dihydrolase and oxidase positive. Pectinolytic strains selection, bioassays. In total 542 Isolates were able to hydrolyze gelatin and aesculin, and bacterial isolates were obtained from 25 cabbage (273 iso- reduce nitrate. Bacteria were also able to produce acid lates) and 27 Chinese cabbage (269 isolates) plant samples. from lactose, melibiose, D-mannitol, D-raffinose and D- The pathogenicity tests performed on potato tubers, cab- xylose. Four strains (19Kp1, 12K4, 17K7b2, 30Kp) released bage and Chinese cabbage leaves gives similar results (se- acetone-citrus aroma during growth on the agar plates or lection of pectionolytic isolates) as on the CVP agar plates. on plant tissues (in bioassays) and were preliminary identi- Performed bioassays and tests on CVP agar plates revealed fied as Pco, and were different from other 8 collected Pcc 117 pectinolytic isolates: 57 isolates were obtained from strains by capability to produce acids from D-arabitol and Chinese cabbage (including 15 Gram-positive isolates) D-sorbitol. Results of phenotypic tests are listed in Table 3. and 60 isolates were obtained from cabbage (including 12 Gram-positive isolates). 90 Gram-negative pectinolytic Rep-PCR. Rep-PCR allowed differentiation of the isolates were selected for further identification. studied isolates. A higher variability was shown by ERIC- PCR and BOX-PCR than REP-PCR. All repetitive PCRs Molecular identification. Twenty three of 90 Gram-neg- (ERIC, BOX and REP) performed yielded from two to ative pectinolytic bacterial isolates analyzed gave positive thirteen products depending on the isolate, the size of 154 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160

Fig. 2. Bayesian phylogram of Pectobacterium spp. strains used in this study based on concatenated data of rpoS, rpoB, atpD, gyrB partial gene sequences alignment. D. dadantii Ech0201 sequence was used as an outgroup, numbers at nodes indicate the posterior probability coefficients obtained after ten million generations. Pcc, P. carotovorum subsp. carotovorum; Pco, P. carotovorum subsp. odoriferum; Pa, P. atrosepticum; Psp, P. sp.; Par, P. aroidearum, Dd, D. Dadantii. the DNA fragments ranged from about 200 to 3500 bp. used rep-PCR (ERIC, REP, BOX) generate 8 amplification The results were combined into a data matrix. Polymor- profiles and was not sufficient to discriminate three strains phic DNA patterns from all rep-PCRs gave 39 different (8K6, Ky5, 9Kp6). Strains from the different subspecies amplicons which were used to construct a binary matrix, grouped separately on the tree and there was no relation- and to generate a dendrogram presenting genetic diver- ship between strains and host-plant observed (Fig. 1). sity (Fig. 1). The cophenetic correlation coefficient (r) for the dendrogram was 0.97, indicating a good fit between Descriptive analysis of sequence data (MLST). One the unmolded data and the dendrogram. All tested Pol- hundred twenty three sequences of collected 20 Pectobac- ish P. carotovorum strains were divided into two separate terium spp. isolates: P. carotovorum newly isolated in this clusters and each contained respective type strains of Pcc study (12 isolates), 3 Pectobacterium carotovorum reference and Pco. The first group was represented by the Pcc type strains (2 strains of Pcc and 1 strain of Pco) and 5 isolates strain LMG 2404T and reference strain LMG 2431 and of others Pectobacterium species, and one Dickeya dadantii eight of the Polish strains identified as Pcc. The second strain (Table 1) were obtained in this study and deposited group contained LMG 17566T the reference Pco strain and in GenBank with accession numbers listed in Supplemen- the four of the tested strains of the Pco (Fig. 1). Generally tary Table 1. The analysis of obtained sequences of six loci: genome profiles obtained for Pco were less diverse than 16S rRNA, rpoS, rpoB, infB, atpD, gyrB with fragments of genome profiles obtained for Pcc. All used rep-PCR meth- 1384, 852, 597, 996, 711, 744 bp, respectively and 2904 bp ods (ERIC, REP, BOX) were insufficient to distinguish of concatenated data set (rpoS, rpoB, atpD, gyrB) were done two Pco strains (12K4, 17K7b2). Least discriminating used for all (20) Pectobacterium spp. strains analyzed within this technique was the REP-PCR, which in case of Pco generate study. Twelve of them were newly isolated from Brassica- one genomic profile with all (obtained from Brassicaceae) ceae plants in Poland and performed sequences analyses Pco strains (19Kp1, 12K4, 17K7b2, 30Kp) and reference allowed to clearly classify all of them into two genetically LMG 17566T. BOX-PCR gave three profile types, while distinct groups: one included eight isolates (1Kp9, 8Kp1, ERIC-PCR resulted in four profile types among five ana- 9Kp6, 8K6, Ky5, 6K1, 15K3, 27K1) identified as Pcc, while lyzed Pco strains. In case of 10 analyzed Pcc strains, all another four isolates identified as Pco (19Kp1, 30Kp, 12K4, Journal of Plant Pathology (2017), 99 (1), 149-160 Oskiera et al. 155

Fig. 3. Bayesian identification phylogram of Pectobacterium strains used in this study (bold) and used by Nabhan et al. (2012a) and Waleron et al. (2014) based on rpoS gene sequence alignments. D. dadantii Ech0201 sequence was used as an outgroup, numbers at nodes indicate the posterior probability coefficients obtained after ten million generations. PcI, PcII - Pcc; PcIII-Pco; PcIV-P. carotovorum subsp. brasiliensis; Pa-P. atrosepticum; Pb-P. betavasculorum; Pw-P. wasabiae; Par-P. aroidearum; Dd-D. dadantii.

17K7b2). Molecular analyses done on the basis of the ob- four Pco ST. Moderately diverse were rpoS and infB loci tained allelic profiles for 20 collected Pectobacterium spp. with six ST among ten Pcc strains and three ST among strains (Table 1; Table 4) resulted in the assignment of five Pco strains. Generally for all analyzed Pectobacterium 16 ST (sequence types) (Table 4). In fact, two Pco strains spp. strains the most diverse of all analyzed loci was gyrB, (12K4, 17K7b2) and three Pcc strains (8K6, 9Kp6, Ky5) which revealed 14 ST among 20 (all) analyzed Pectobacte- could not be differentiated in this study. Also, two P. atro- rium spp. isolates and reference strains, however within septicum strains (Ec32, Eca59) used for comparison were collected ten Pcc and five Pco strains, seven and three ST not distinguished. Five analyzed Pco strains (including were distinguished respectively (Table 4). reference strain LMG17566T) reveals four ST and ten Pcc strains reveals eight ST. The least diverse of all analyzed Phylogenetic analysis. Dendrograms constructed on loci was 16S rRNA (Table 4). Only six 16S rRNA ST were the basis of sequence data of each genes separately showed distinguished among 20 strains of Pectobacterium spp. slightly different phylogenetic relationships between Analysis of obtained 16S rRNA sequences were sufficient strains. However, both analysis done for each gene sepa- to identify each of all collected strains to the species and rately and concatenated data set allowed for identification subspecies, however sequences did not differ between and differentiation of the all Pectobacterium carotovorum strains of the same species, except of LMG 2404T, which Polish strains and references included. They have been differs from others Pcc strains sequences by one base pair. grouped into separate clusters according to the deter- Sequences of atpD and rpoB were the least diverse of all mined subspecies. Strains of Pcc and Pco were separated analyzed Pcc housekeeping genes fragments and allowed with high posterior probability value equal 1. Pco strains to differentiate only five ST among ten Pcc strains. In case grouped with type strain of this subspecies LMG 17566T. of five Pco strains only two atpD ST could be differenti- Pcc strains grouped with strain LMG 2431, but type strain ated, but rpoB locus turned to be the most diverse within of this subspecies LMG 2404T was diverse. Some divisions 156 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160

Table 3. Results of phenotypic assay performed for differentiation of Pectobacterium spp. T T LMG 17566 19Kp1 30Kp 12K4 17K7b2 LMG 2404 1Kp9 8Kp1 8K6 15K3 27K1 6K1 Ky5 LMG 2431 9Kp6 Eca 2M Eca 59 Ec32 Ea 853 Ecc 4M 0201 Ech

TEST

P. carotovorum subsp.

P. carotovorum subsp. carotovorum (Pcc) . sp.

odoriferum (Pco) P D. dadantii D. P. aroidearum P. P. atrosepticumP.

Pectolytic activity on potato + + + + + + + + + + + + + + + + + + + + + Pectolytic activity on CVP + + + + + + + + + + + + + + + + + + + + + Levan − − − − − − − − − − − − − − − − − − − − − Oxidase − − − − − − − − − − − − − − − − − − − − − Catalase + + + + + + + + + + + + + + + + + + + + + Arginine dihydrolase − − − − − − − − − − − − − − − − − − − − +/− Starch hydrolysis N/A − − − − N/A − − − − − − − − − − − − − N/A N/A Nitrate reduction N/A + + + + N/A + + + + + + + N/A − + N/A − N/A N/A + Acid production from D-Arabitol + + + + + − − − − − − − − − − − − − − − − D-Sorbitol + + + + + − − − − − − − − − − − − − − − − Melibiose + + + + + + + + + + + + + + + + + + + + + D-glucose + + + + + + + + + + + + + + + + + + + + + Lactose + + + + + + + + + + + + + + + + + + + + + D-mannitol + + + + + + + + + + + + + + + + + + + + + D-raffinose + + + + + + + + + + + + + + + + + + + + + D-xylose + + + + + + + + + + + + + + + + + + + + +

according to host-plant and year of isolation were noticed strains of this subspecies (Fig. 3). Based on this analysis in case of Pco strains. Among Pcc strains two clusters strains isolated from Zantedeschia spp. (Ecc4M and Ea853) were distinguished, however division of strains accord- used for comparison in this study previously described as ing to host plant and year of isolation were not observed Pcc (Mikicin´ski et al., 2010) were reclassified to P. aroi- (Fig. 2). Overall topological scores computed with Com- dearum (cluster PcV) according to Nabhan et al. (2013). In pare2Trees (Nye et al., 2005) program showed that con- general, relation to host among Pcc and Pco strains were catenated data phylogenetic tree was similar to the rpoS; not observed, both subspecies were isolated from cabbage infB; 16S rRNA; atpD; rpoB and gyrB partial gene locus and Chinese cabbage and grouped in the same clusters on tree in 91,7; 86; 84,8; 82,9; 78 and 77,8% respectively. As the phylogenetic tree (Fig. 3). topology of the bayesian phylogenetic tree obtained with rpoS gene fragment sequences was the most similar to the tree obtained with concatenated data set, therefore, ad- DISCUSSION ditionally, phylogenetic analysis of the rpoS gene fragment (Fig. 3) were done to better known phylogenetic relation- Presented study allowed for the identification, charac- ship of obtained Polish strains with other worldwide well terization and genetic diversity of the Pcc and Pco strains defined Pectobacterium spp. strains and clusters accord- obtained from Brassicaceae plants in Poland. This study ing to Nabhan et al. (2012a). Phylogenetic analysis of se- constitutes a significant contribution to the current state of quences obtained in this study and sequences of this gene knowledge of P. carotovorum occurrence on soft rot bacte- retrieved from NCBI showed that all Polish strains identi- ria occurring on cabbage and Chinese cabbage in Poland. fied as Pco clearly grouped with type strain of this subspe- Since now Pectobacterium spp. have been isolated from cies LMG 17566T sequence in cluster PcIII according to wide range of vegetables including Brassicaceae. In that Nabhan et al. (2012a). All strains classified as a Pcc were genus in some cases bacterial host specification or limita- grouped into two clusters. Cluster named PcII contained tions are observed (Ma et al., 2007). There are some cases the all Polish Pcc strains from cabbage and Chinese cab- where strains of D. chrysanthemi and P. atrosepticum were bage and cluster PcI contained type strain LMG 2404T isolated from cabbage or Chinese cabbage, but generally isolated from potato (Solanum tuberosum L.) and other these bacterial species are minor Brassicaceae pathogens Journal of Plant Pathology (2017), 99 (1), 149-160 Oskiera et al. 157

Table 4. Nucleotide diversity observed within Pectobacterium strains characterized in this study.

No. of Nucleotide Average number Average Locus and No. of No. of Fragment G+C No. of parasimony diversity of nucleotide number of Tajima’s concatenated sequence alleles polymorphic size [bp] content strains informative (per site) differences per nucleotide D test* for species and genus (ST) sites sites [Π] site [θ] differences [k] Pectobacterium carotovorum subsp. carotovorum (Pcc) 16SrRNA 1383 0.542 10 2 1 0 0.000150 0.000260 0.200000 −1.111730 rpoS 852 0.514 10 6 13 9 0.006020 0.005410 5.111000 0.510760 rpoB 597 0.532 10 5 11 1 0.004240 0.006510 2.533000 −1.560430 infB 942 0.519 9 6 54 12 0.014950 0.019950 14.889000 −1.280140 atpD 711 0.541 10 5 12 5 0.005440 0.005970 3.867000 −0.399470 gyrB 744 0.505 10 7 59 11 0.019830 0.028030 14.756000 −1.435430 concatenated 2904 0.522 10 8 95 26 0.00905 0.01158 26.267 −1.07841 Pectobacterium carotovorum subsp. odoriferum (Pco) 16SrRNA 1383 0.546 5 1 0 0 0.000000 0.000000 0.000000 n.a. rpoS 852 0.506 5 3 2 1 0.001170 0.001130 1.000000 0.243140 rpoB 597 0.536 5 4 6 4 0.005360 0.004820 3.200000 0.763690 infB 996 0.52 5 3 6 3 0.003010 0.002890 3.000000 0.286380 atpD 711 0.538 5 2 1 1 0.000840 0.000680 0.600000 1.224740 gyrB 744 0.52 5 3 9 9 0.007260 0.005810 5.400000 1.776610 concatenated 2904 0.524 5 4 18 15 0.01158 0.01158 10.200000 1.32981 Pectobacterium spp. (Pcc, Pco, P. atrosepticum, P. aroidearum, P. sp.) 16SrRNA 1384 0.544 20 6 54 51 0.01148 0.01107 15.779 −0.2073 rpoS 852 0.509 20 13 85 74 0.02882 0.02822 24.468 −0.05551 rpoB 597 0.528 20 12 49 37 0.02447 0.02321 14.558 −0.02731 infB 996 0.519 19 13 114 86 0.03801 0.03275 37.86 −0.11285 atpD 711 0.539 20 11 52 38 0.01871 0.02061 13.305 −0.50801 gyrB 744 0.512 20 14 127 109 0.05693 0.04811 42.358 0.40037 concatenated 2904 0.521 20 16 313 258 0.03266 0.03218 94.68947 0.06216 *Tajima’s D test were not significant, P > 0.10.

(http://www.cabi.org/isc/). The most polyphagous and di- 8K6, 27K1, Ky5 and three strains from Chinese cabbage: verse species of the genus Pectobacterium is P. carotovorum. 1Kp9, 8Kp1, 9Kp6. Of Pco strains two were isolated from Pcc is the main pathogen of the Brassicaceae plants, it may cabbage: 12K4, 17K7b2 and two from Chinese cabbage: cause soft rot of following vegetables: cabbage, Chinese 19Kp1, 30Kp. It is worth to mention that all pectinolytic cabbage, cauliflower (Brassica oleracea Botrytis Group), strains, which released intense characteristic acetone-citrus savoy cabbage (Brassica oleracea Savoy Cabbage Group), aroma were identified as Pco. That is in agreement with potato, onion (Allium cepa L.), sugar cane (Saccharum statement from original paper describing this subspecies officinarum L.) and others (Nabhan et al., 2012a, 2012b; (Gallois et al., 1992). Banana-like odor was observed on Waleron et al., 2014). Pco has been isolated mainly from rotting chicory and was helpful to distinguish Pco from the Chicory (Cichorium intybus L.) in France and Swit- other Pectobacterium spp., since gas chromatography per- zerland, but also from leeks (Allium porrum) and celery formed by Gallois et al. (1992) revealed that all analyzed (Apium graveolens) in France and Poland (Gallois et al., Pco strains differs from other Pectobacterium spp. in pro- 1992; Waleron et al., 2014), parsley (Petroselinum crispum duction of propyl acetate, isobutyl acetate, isoamyl acetate, (Mill.) Fuss) in Serbia, carrot (Daucus carota subsp. sati- and 2-actamyl acetate. vus (Hoffm.) Schübl. & G. Martens) and onion in Poland Waleron et al. (2013, 2014, 2015) investigated occur- (Waleron et al., 2014), potato, turnip (Brassica rapa var. rapa rence of the Pectobacterium spp. on different vegetables L.) and cabbage in Iran (Rezaei and Taghavi, 2010), and in Poland and apart P. atrosepticum and Pcc also P. wasa- Chinese cabbage and garlic (Allium fistulosum Linn.) in biae, Pco and P. carotovorum subsp. brasiliense in Poland. China (Zhu et al., 2010). Complete genome sequence of P. Mikiciński et al. (2010) analyzed pectinolytic bacteria from carotovorum subsp. odoriferum strain BC S7 (GenBank ac- Zantedeschia spp. and identified P. atrosepticum (strain cession No. CP009678.1) obtained from infected Chinese Eca2M) and Pcc (strain Ecc4M and Ea853). Majority of cabbage leaf was released in 2014 by Xie H. and Ma R. mentioned strains were included in this study and reiden- from Beijing Academy of Agriculture and Forestry Sci- tified on the basis of obtained sequences (Supplementary ences. However, in literature there is no report on pres- Table 1). Only strain Eca2M could not be classified as P. ence of Pco on cabbage and Chinese cabbage in Poland. atrosepticum or any other known Pectobacterium species In our study among twelve obtained P. carotovorum strains (data not shown). Thanks to the sequence analyses per- eight strains were identified as Pcc and the other four as formed another strain (Ecc4M) was identified as P. aroi- Pco. Of Pcc strains five were from cabbage: 6K1, 15K3, dearum, this species have not been described yet in 2010, 158 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160 but now is known as common pathogen of Zantedeschia and Dickeya species and subspecies. Later, when useful- spp. (Nabhan et al., 2013). Similar situation was observed ness for differentiation Pectobacterium spp. strains of recA when Zhu et al. (2010) isolated one Pectobacterium strain gene sequences were compared to rpoS gene sequences, it from Chinese cabbage which in phylogeny analysis clus- showed that grouping were almost identical (Waleron et tered with typical Pectobacterium strain PC1. Strain PC1 al., 2014). Ngadze et al. (2012) used gyrB and recA genes was further reclassified to the species P. aroidearum (Nab- sequences for identification of bacteria causing potato soft han et al., 2013). It means that also P. aroidearum was iso- rot and blackleg in South Africa and Zimbabwe. Performed lated from Chinese cabbage. identifications were reliable with use of sequences of both CVP medium and bioassays on cabbage and Chinese loci, but gyrB demonstrated a greater intraspecies diver- cabbage used in this study for pectinolytic strains selec- sity. That is in agreement with presented analyses where tion, connected with use of the genus specific primers Y1/ gyrB occurs the most diverse. Recently, in the main taxo- Y2 (Darrasse et al., 1994), proved to be very useful, cheap nomic studies of Pectobacterium spp. eight housekeeping and efficient method for identification of Pectobacterium genes (acnA, icdA, gapA, mdh, mtlD, pgi, proA and rpoS) spp., from cabbage what is in agreement with study on were used to define relations between Pectobacterium spe- strains causing soft rot of lettuce (Lactuca sativa var. ro- cies and subspecies (Ma et al., 2007; Nabhan et al., 2012a, mana) (Nazerian et al., 2013). However, for classification 2012b) and finally for new species discrimination, such as and differentiation, more complex study including bio- P. aroidearum (Nabhan et al., 2013). In presented study, of chemical tests, rep-PCR, MLST and phylogenetic analysis them only rpoS locus was investigated and phylogenetic tree had to be performed. Acids production from D-arabitol of this locus showed the most similar topology with concat- and D-sorbitol by Pco strains makes it possible to easily enated data set. Also, use of rpoS locus in many taxonomic distinguish those two closely related subspecies (Pco and studies (Nabhan et al., 2012a, 2012b, 2013; Waleron et al., Pcc), what is with accordance to Gallois et al. (1992) and 2008, 2014) makes this locus valuable to use as a reference Schaad et al. (2001). Also, partial genes sequences from in sequences analyses and due that, was recommended for all analyzed loci were able to distinguish both subspecies, phylogenetic analyses of Pectobacterium spp. as well as all rep-PCR methods used in this study. PCR Occurrence of Pco strains in Chinese cabbage and cab- amplification of repetitive bacterial DNA elements using bage confirms that two subspecies of P. carotovorum (Pcc the REP, ERIC and BOX primers discriminated the soft and Pco) cause diseases of those vegetables in the cultiva- rot bacteria isolated from cabbage and Chinese cabbage tion area in Poland. These results are in agreement with in Poland at the species and subspecies level. Since now, earlier studies of the Pco identification in Poland (Waleron several studies have shown the usefulness of the rep-PCR et al., 2014). This suggests that more intensive inspections technique for identification and differentiation of many of and monitoring should be conducted in different geo- bacteria into species, pathovar or strain level (Versalovic et graphical regions of our country. From the epidemiologi- al., 1991; Rademaker et al., 2004; Kałuz˙na et al., 2010). In cal point of view, this is a very important observation since our study, Pco strains were less diverse as compared to Pcc the bacterium was mainly considered to be pathogen of strains which is in agreement with results of Rezaei and chicory. To our knowledge this is the first report of Pco Taghavi (2010) and Ngadze and coworkers (2012). Similar- occurence on cabbage and Chinese cabbage in Poland. ly, different fingerprints were observed among Pcc strains isolated from Solanum tuberosum in Zimbabwe (Ngadze et al., 2012). ACKNOWLEDGEMENTS In literature it was reported that phylogenetic analyses with use of 16S rRNA sequences allowed to differentiate Authors are grateful to Dr Renata Lebecka (IHAR, such genus as Erwinia, Pectobacterium, Brenneria (Hauben Młochów, Poland) for providing to presented research et al., 1998) and Dickeya (Samson et al., 2005). In the pre- strain Eca59 (original name Eca 5/06) of P. atrosepticum, sented study 16S rRNA was a lowest variable nucleotide and Artur Mikicin´ski (Research Institute of Horticul- sequence of all analyzed loci, and although were suitable ture, Skierniewice, Poland) for providing strains (Ecc4M, to identify Pectobacterium spp. strains to the species level, Ea853, Eca2M) isolated from Zantedeschia spp. Presented did not differentiate strains of the same subspecies level. studies have been conducted in the implementation of the This statement is in agreement with study of Enterobacte- statutory tasks of the Research Institute of Horticulture in riaceae (Dauga, 2002). Dauga (2002) postulated that 16S Skierniewice (task 15.6 realized in years: 2010-2012). rRNA is more suitable to describe relationships between distantly related Enterobacteriaceae strains, than between strains within the same species. In the present study the use REFERENCES of sequences of housekeeping genes was more useful for intraspecies differentiation. Partial gene sequences of recA Aljanabi S.M., Martinez I., 1997. Universal and rapid salt-ex- was successfully used by Waleron et al. (2002) for identifi- traction of high quality genomic DNA for PCR-based tech- cation and differentiation of strains within Pectobacterium niques. Nucleic Acids Research 25: 4692-4693. Journal of Plant Pathology (2017), 99 (1), 149-160 Oskiera et al. 159

Brady C., Cleenwerck I., Venter S., Vancanneyt M., Swings J., syringae from stone fruits and hazelnut using repetitive-PCR Coutinho T., 2008. Phylogeny and identification of Pantoea and MLST. Journal of Plant Pathology 92: 781-787. species associated with plants, humans and the natural en- Lelliott R.A., Stead D.E., 1987. Methods for the diagnosis of vironment based on multilocus sequence analysis (MLSA). bacterial diseases of plants. Blackwell Scientific Publica- Systematic and Applied Microbiology 31: 447-460. tions, UK. Brody J.R., Calhoun E.S., Gallmeier E., Creavalle T.D., Kern Librado P., Rozas J., 2009. DnaSP v5: a software for compre- S.E., 2004. Ultra-fast high-resolution agarose electrophoresis hensive analysis of DNA polymorphism data. Bioinformatics of DNA and RNA using low-molarity conductive media. 25: 1451-1452. BioTechniques : 598-602. 37 Lopez M.M., Cambra M., Gorris M.T., 2002. Enrichment ELI- Castresana J., 2000. Selection of conserved blocks from multi- SA (E-ELISA). In: Perombelon M.C.M., Van Der Wolf J.M. ple alignments for their use in phylogenetic analysis. Mo- (eds). Methods for detection and quantification of Erwinia lecular Biology and Evolution 17: 540-552. carotovora subsp. atroseptica (Pectobacterium carotovorum Czajkowski R., Pérombelon M.C.M., Jafra S., Łojkowska E., subsp. atrosepticum) on potatoes: a laboratory manual, pp. Potrykus M., van der Wolf J.M., Śledź W., 2015. Detection, 34-43. Scottish Crop Research Institute, Scotland, UK. identification and differentiation of Pectobacterium and Louws F.J., Fulbright D.W., Stephens C.T., De Bruijn F.J. 1994. Dickeya species causing potato blackleg and tuber soft rot: Specific genomic fingerprints of phytopathogenic Xan- a review. Annals of Applied Biology 166: 18-38. thomonas and Pseudomonas pathovars and strains generated Darrasse A., Priou S., Kotoujansky A., Bertheau Y., 1994. PCR with repetitive sequences and PCR. Applied and Environ- and restriction fragment length polymorphism of pel gene mental Microbiology 60: 2286-2295. as a tool to identify Erwinia carotovora in relation to potato Ma B., Hibbing M.E., Kim H.S., Reedy R.M., Yedidia I., Breuer diseases. Applied Environmental Microbiology : 1437-1443. 60 J., Breuer J., Glasner J.D., Perna N.T., Kelman A., Charkows- Darriba D., Taboada G.L., Doallo R., Posada D., 2012. jModel- ki A.O., 2007. Host range and molecular phylogenies of the Test 2: more models, new heuristics and parallel computing. soft rot enterobacterial genera Pectobacterium and Dickeya. Nature Methods : 772-772. 9 Phytopathology 97: 1150-1163. Dauga C., 2002. Evolution of the gyrB gene and the molecular Maciel K.W., Almeida I.M.G., Silva H.S.A., Rodrigues L.M.S., phylogeny of Enterobacteriaceae: a model molecule for mo- Beriam L.O.S., 2010. Detection of Pseudomonas viridiflava lecular systematic studies. International Journal of System- in imported chinese cabbage (Brassica rapa var. pekinensis) atic and Evolutionary Microbiology 52: 531-547. seeds. Summa Phytopathologica 36: 83-84. Dawyndt P., Vancanneyt M., Snauwaert C., De Baets B., De Mikicin´ski A., Sobiczewski P., Sulikowska M., Puławska J., Meyer H., Swings J., 2006. Mining fatty acid databases for Treder J., 2010. Pectolytic bacteria associated with soft rot detection of novel compounds in aerobic bacteria. Journal of of calla lily (Zantedeschia spp.) tubers. Journal of Phytopathol- Microbiological Methods 66: 410-433. ogy 158: 201-210. De Boer S.H., Ward L.J., 1995. PCR detection of Erwinia caro- Nabhan S., Boer S.H., Maiss E., Wydra K., 2012a. Taxonomic tovora subsp. atroseptica associated with potato tissue. Phy- relatedness between Pectobacterium carotovorum subsp. topathology 85: 854-858. carotovorum, Pectobacterium carotovorum subsp. odoriferum Duarte V., De Boer S.H., Ward L.J., de Oliveira A.M.R., 2004. and Pectobacterium carotovorum subsp. brasiliense subsp. Characterization of atypical Erwinia carotovora strains caus- nov. Journal of Applied Microbiology 113: 904-913. ing blackleg of potato in Brazil. Journal of Applied Microbiol- Nabhan S., Wydra K., Linde M., Debener T., 2012b. The use of ogy 96: 535-545. two complementary DNA assays, AFLP and MLSA, for Frenchon D., Exbrayat P., Gallet O., Guillot E., LeClerc V., epidemic and phylogenetic studies of pectolytic enterobacte- Payet N., Bertheau Y., 1995. Sequences nucleotidiques pour rial strains with focus on the heterogeneous species Pecto- la detection des Erwinia carotovora subsp. atroseptica. Brevet bacterium carotovorum. Plant Pathology : 498-508. 95. 12-803 INRA INA-PG, SDP. 61 Nabhan S., De Boer S.H., Maiss E., Wydra K., 2013. Pectobac- Gallois A., Samson R., Ageron E., Grimont P.A., 1992. Erwinia terium aroidearum sp. nov., a soft rot pathogen with prefer- carotovora subsp. odorifera subsp. nov., associated with odor- ous soft rot of chicory (Cichorium intybus L.). International ence for monocotyledonous plants. International Journal of Systematic and Evolutionary microbiology : 2520-2525. Journal of Systematic Bacteriology 42: 582-588. 63 Gardan L., Gouy C., Christen R., Samson R., 2003. Elevation Nazerian E., Sijam K., Ahmad Z.A.M., Vadamalai G., 2013. of three subspecies of Pectobacterium carotovorum to species Characterization of Pectobacterium carotovorum subsp. caro- level: Pectobacterium atrosepticum sp. nov., Pectobacterium tovorum as a new disease on Lettuce in Malaysia. Austral- betavasculorum sp. nov., and Pectobacterium wasabiae sp. asian Plant Disease Notes 8: 105-107. nov. International Journal of Systematic and Evolutionary Mi- Ngadze E., Brady C.L., Coutinho T.A., Van der Waals J.E., crobiology 53: 381-391. 2012. Pectinolytic bacteria associated with potato soft rot Hauben L., Moore E.R., Vauterin L., Steenackers M., Mergaert and blackleg in South Africa and Zimbabwe. European Jour- J., Verdonck L., Swings J., 1998. Phylogenetic position of nal of Plant Pathology 134: 533-549. phytopathogens within the Enterobacteriaceae. Systematic Nye T.M., Lio P., Gilks W.R., 2006. A novel algorithm and web- and Applied Microbiology 21: 384-397. based tool for comparing two alternative phylogenetic trees. Kałuz˙na M., Ferrante P., Sobiczewski P., Scortichini M., 2010. Bioinformatics 22: 117-119. Characterization and genetic diversity of Pseudomonas Pérombelon M.C.M., Burnett E.M., 1991. Two modified 160 P. carotovorum identification in Poland Journal of Plant Pathology (2017), 99 (1), 149-160

crystal violet pectate (CVP) media for the detection, isola- on potatoes: a laboratory manual, pp. 17-24. Scottish Crop tion and enumeration of soft rot Erwinias. Potato Research Research Institute, Scotland, UK. 34: 79-85. Van Vuurde J.W.L., Van der Wolf J.M., 2002. Immunofluores- Rademaker J.L.W., Louws F.J., Versalovic J., De Bruijn F.J., cence colony staining (IFC). In: Perombelon M.C.M., Van Kowalchuk G.A., Head I.M., Akkermans A.D.L., van Elsas Der Wolf J.M. (eds). Methods for detection and quantifica- J.D., 2004. Characterization of the diversity of ecologically tion of Erwinia carotovora subsp. atroseptica (Pectobacterium important microbes by rep-PCR genomic fingerprinting. In: carotovorum subsp. atrosepticum) on potatoes: a laboratory Kowalchuk G.A., de Bruijn F., Head I.M., Van der Zijpp manual, pp. 25-33. Scottish Crop Research Institute, Scot- A.J., van Elsas J.D. (eds). Molecular Microbial Ecology Man- land, UK. ual. 2nd Ed., pp. 611-643. Springer, The Netherlands. Versalovic J., Schneider M., De Bruijn F.J., Lupski J.R., 1994. Rezaei R., Taghavi S.M., 2010. Phenotypic and genotypic char- Genomic fingerprinting of bacteria using repetitive se- acteristics of Iranian soft rot bacteria isolates from different quence-based polymerase chain reaction. Methods in Mo- hosts. Phytopathologia Mediterranea 49: 194-204. lecular and Cellular Biology 5: 25-40. Rimmer R., Shattuck V.,I., Buchwaldt L., 2007. Compendium Waleron M., Waleron K., Podhajska A.J., Łojkowska E., 2002. of Brassica Diseases, pp. 58-59. The American Phytopatho- Genotyping of bacteria belonging to the former Erwinia logical Society, APS Press, USA. genus by PCR-RFLP analysis of a recA gene fragment. Mi- Ronquist F., Teslenko M., van der Mark P., Ayres D.L., Darling crobiology 148: 583-595. A., Höhna S., Larget B., Liu L., Suchard M.A., Huelsenbeck Waleron M., Waleron K., Geider K., Łojkowska E., 2008. Ap- J.P., 2012. MrBayes 3.2: efficient Bayesian phylogenetic infer- plication of RFLP analysis of recA, gyrA and rpoS gene frag- ence and model choice across a large model space. System- ments for rapid differentiation of Erwinia amylovora from atic Biology 61: 539-542. Erwinia strains isolated in Korea and Japan. European Jour- Samson R., Legendre J.B., Christen R., Fischer-Le Saux M., Ach- nal of Plant Pathology 121: 161-172. ouak W., Gardan L., 2005. Transfer of Pectobacterium chry- Waleron M., Waleron K., Łojkowska E., 2013. Occurrence of santhemi (Burkholder et al., 1953) Brenner et al., 1973 and Pectobacterium wasabiae in potato field samples. European Brenneria paradisiaca to the genus Dickeya gen. nov. as Dickeya Journal of Plant Pathology 137: 149-158. chrysanthemi comb. nov. and Dickeya paradisiaca comb. nov. Waleron M., Waleron K., Łojkowska E., 2014. Characterization and delineation of four novel species, Dickeya dadantii sp. of Pectobacterium carotovorum subsp. odoriferum causing nov., Dickeya dianthicola sp. nov., Dickeya dieffenbachiae sp. soft rot of stored vegetables. European Journal of Plant Pa- nov. and Dickeya zeae sp. nov. International Journal of System- thology 139: 457-469. atic and Evolutionary Microbiology 55: 1415-1427. Waleron M., Waleron K., Łojkowska E., 2015. First report of Schaad N.W., Jones J.B., Chun W., 2001. Laboratory Guide for Pectobacterium carotovorum subsp. brasiliense causing soft Identification of Plant Pathogenic Bacteria. 3rd Ed. APS rot on potato and other vegetables in Poland. Plant Disease Press, USA. 99: 1271. Thompson J.D., Gibson T.J., Plewniak F., Jeanmougin F., Hig- Weingart H., Völksch B., 1997. Genetic Fingerprinting of Pseu- gins D.G., 1997. The CLUSTAL_X windows interface: flex- domonas syringae pathovars using ERIC, REP, and IS50 ible strategies for multiple sequence alignment aided by PCR. Journal of Phytopathology 145: 339-345. quality analysis tools. Nucleic Acids Research 25: 4876-4882. Weisburg W.G., Barns S.M., Pelletier D.A., Lane D.J., 1991. 16S Van Der Wolf J.M., Perombelon M.C.M., 2002. Immunomag- ribosomal DNA amplification for phylogenetic study. Journal netic separation-colony count on CVP medium (IMS-CVP). of Bacteriology 173: 697-703. In: Perombelon M.C.M., Van Der Wolf J.M. (eds). Methods Zhu L., Xie H., Chen S., Ma R., 2010. Rapid isolation, identifi- for detection and quantification of Erwinia carotovora subsp. cation and phylogenetic analysis of Pectobacterium carotovo- atroseptica (Pectobacterium carotovorum subsp. atrosepticum) rum ssp. Journal of Plant Pathology 92: 479-483.

Received July 22, 2016 Accepted November 9, 2016