Allelic profile of Serbian Xanthomonas campestris pv. campestris isolates from cabbage

1 for the P-X- dnaK, fyuA, gyrB and rpoD genes, respectively. This profile is assigned as sequence type 2 (ST2) and it coincides with a Portuguese B. oleracea Xcc strain, CPBF 213, originating from B. oleracea var. costata. No connection between sequence type (ST) and the host was detected.


INTRODUCTION
Yields and quality of cruciferous vegetables (fam. Brassicaceae), such as cabbage (Brassica oleracea var. capitata L.), cauliflower (B. oleracea var. botrytis L.), broccoli (B. oleracea var. italica L.), kale (B. oleracea var. sabellica L.), etc. are widely affected by various pathogens (Neik et al., 2017). Their high nutritional value, extensive use in human diet, and annual production rate of about 250.000 hg/ha (www.fao.org/ faostat/en) in Serbia strongly indicate the importance of these vegetables in agricultural production of this country. Cabbage is one of the most important crucifers worldwide, especially because of its richness in beneficial phytochemicals which are an important part of human nutrition (Singh et al., 2006).
Cabbage yields may be reduced by the activity of many plant pests and pathogens, and Xanthomonas campestris pv. campestris (Xcc), the causal agent of black rot disease, is one of the most important pathogens (Massomo et al., 2004;Popović et al., 2013Popović et al., , 2019Afrin et al., 2019;da Silva et al., 2020). This pathogen affects cabbage growth, leading to head weight and quality losses, and thereby drastically reducing its economic value. Under favorable weather conditions, losses caused by Xcc on cabbage can exceed 50% (Williams, 1980). The main source of infection is the infected seed, but the pathogen can also spread via raindrops, wind, irrigation, non-sterile tools, etc., thus entering the host plant through leaf openings such as stomata, hydathodes or some artificially created wounds (Hugouvieux et al., 1998;Afrin et al., 2019). Disease symptoms can be easily noticed in the field as the blackening of leaf veins, which is one of the first indicators of pathogen presence and multiplication inside plant vascular vessels. Also, symptoms caused by Xcc can be noticed in the form of V-shaped chlorotic to necrotic lesions on leaf margins, which further expand towards leaf base. The presence of these symptoms contributes to stopping growth, plant wilting, and finally rotting (Peňázová et al., 2015).
Considering the great yield loss rate and a resulting need for fast detection and suppression of Xcc, many molecular, PCR-based techniques for fast and sensitive detection of this pathogen in seeds and plants have been developed. Pathovar-specific primers are mainly designed to target genes involved in pathogen virulence, such as zinc uptake regulator (Zur) and genes belonging to the hypersensitive reaction and pathogenicity (hrpF) cluster (Eichmeier et al., 2019;Rubel et al., 2019). The multilocus sequence typing (MLST) approach, based on PCR amplifications of various housekeeping genes is becoming one of the main techniques for more precise identification and characterization (Young et al., 2008;Almeida et al., 2010). Such techniques provide useful information, especially now that globally accessible databases are available, such as the National Center for Biotechnology Information (NCBI), and Plant Associated and Environmental Microbes Database (PAMDB). Usage of these databases is of great importance because they allow fast comparison of test strains with all previously deposited strains of certain species, isolated from various hosts and regions/ countries. Multilocus sequence analysis (MLSA) is also widely used for establishing phylogenetic relatedness among test strains, allowing us to provide information about recombination events, the presence/absence of genetic diversity among species and their evolution (Fargier et al., 2011). Besides the nucleotide-based approach for analyzing MLST data, a new allelicbased approach has been recently developed (Feil et al., 2004). That approach is based on globally optimized implementation of the eBURST algorithm (goeBURST), which is applied for analyzing sequences of a set of typically seven selected housekeeping genes which are assumed to be under moderate to strong purifying selection (Feil et al., 2004).
This study aimed to apply a new technique for allelic profiles determination of Serbian cabbage Xcc strains and thus provide more information about the genetic structure of the cabbage Xcc population.

Bacterial strains
Three representative Xcc cabbage strains isolated in Serbia in 2014, coded as Xc40, Xc48, and Xc75, were selected for allelic profile determination out of a total of 44 strains previously characterized by Popović et al. (2019). The three strains were representative considering the high homology among Xcc populations from cabbage determined by the MLSA with ten housekeeping genes.

DNA extraction
Total genomic DNA of the test Xcc strains was extracted using a modified CTAB protocol proposed by Le Marrec et al. (2000). Single bacterial colonies of each strain, grown on YDC agar for 48 h, were resuspended in 500 μl of sterile distilled water and centrifuged at 13000 rpm for 5 min. The obtained pellet was resuspended in a mix of 567 μl of Tris-EDTA (TE) buffer (pH7, 10mM TRIS, 1mM EDTA), 30 μl of 10% (w/v) sodium dodecyl sulfate (SDS) and 20 mg ml -1 proteinase K, and incubated at 37 °C for 30 min. The samples were then treated with 100 μl of 5M NaCl and 250 μl of 3% hexadecyltrimethylammonium bromide (CTAB, pH 8.0). The mix was then heated at 65 °C for 20 min. After the incubation period, the DNA was purified with 750 μl of chloroform and centrifuged for 10 min at 13000 rpm. The obtained upper phase was transferred to clean tubes and 1/10 of 3M sodiumacetate (pH5.0) was added to precipitate DNA. The next step was the addition of ice-cold isopropanol and centrifugation at 13000 rpm for 15 min. The final phase consisted of washing the obtained pellet with 1 ml of 96% of ice-cold ethanol and centrifugation for 10 min at 13000 rpm. The obtained DNA was dried at room temperature for 30 min, dissolved in 50 μl TE buffer, and stored at -20 °C until use.
PCR amplification was performed in 25 μl mixture made with 12.5 μl of 2 × PCR TaqNova-RED Master Mix, 9.5 μl of PCR water, 1 μl of each primer and 1 μl of total sample DNA. PCR conditions were set according to a protocol given by Fargier et al., (2011) with an initial denaturation step at 94 °C for 3 min, followed by 35 cycles of denaturation (94 °C for 50 sec), hybridization (62 °C for 50 sec) and elongation (72 °C for 1 min). The final elongation step was performed at 72 °C for 7 min.
The obtained PCR products were visually checked for band presence on the searched position (540 bp) in comparison to 200-10000 bp SmartLadder MW-1700-10 DNA Ladder (Eugentec) on 1% agarose gel stained with ethidium bromide. Before sending them for sequencing in the Macrogen Sequence service (Amsterdam, the Nederlands), the PCR products were purified with QIAquick PCR Purification Kit (Qiagen) according to the protocol proposed by the manufacturer.
To confirm the identity of the strains, the obtained sequences were checked for their quality and compared with sequences deposited in the NCBI database using the nucleotide BLAST (BLASTn) option. The obtained sequences were deposited into the NCBI database to obtain accession numbers ( Table 1).
Sequences of the other three genes (fyuA, gyrB, and rpoD) were retrieved from the GenBank, where they had been deposited earlier by Popović et al. (2019) ( Table 1) and used for allelic profile determination.

Allelic-profile determination
Allelic profiles of the Serbian cabbage Xcc strains were determined based on sequences of four housekeeping genes (P-X-dnaK, fyuA, gyrB, and rpoD). The genetic structure and position of test strains were determined in relation to 33 Xcc strains selected to represent each of the 33 Xcc sequence types (STs) described so far by Cruz et al. (2018) and Bella et al. (2019). A list of strains used for comparison, hosts, and countries of their isolation are presented in Table 2. Sequences of all strains included in the comparison for each of the four genes were retrieved from the GenBank, aligned using the ClustalW segment of BioEdit v. 7.0.5 and trimmed to the same size (P-X-dnaK-470 nt, fyuA-538 nt, gyrB-418 nt, rpoD-422 nt). Allelic profiles for each gene were determined using a Non-redundant PubMLST database (https://pubmlst.org/analysis/ nrdb.shtml). Each DNA sequence that differed from all other sequences of the same gene in one or more nucleotides was marked with a different allele number. The combination of the obtained allelic profiles for each strain was considered as ST.
To represent graphically the position of Serbian cabbage Xcc strains in relation to the other 33 Xcc strains from the database used for comparison, a Minimum Spanning (MS) tree was constructed using the Phyloviz software (http://www.phyloviz.net/goeburst/) with implemented global optimal eBURST (goeBURST) algorithm (Feil et al., 2004). Each nodule on the MS tree represents one ST and its size is determined based on the number of isolates comprising it. The links connecting nodules represent single (SLVs), double (DLVs) and triple (TLVs) locus variants, meaning that the tested genotype differs from the founder genotype at only one, two or three out of four MLST loci, respectively. Different colors of the links connecting nodules (STs) represent SLVs (blue), DLVs (green), TLVs (light grey), and edges without ties (black).

RESULTS AND DISCUSSION
The allelic profiles and STs obtained for the tested Serbian cabbage Xcc strains (Xc40, Xc48, and Xc75) and comparative Xcc strains from diverse hosts are presented in Table 2. All three Serbian strains had the same allelic profile: 1, 3, 1, 1, for the genes P-X-dnaK, fyuA, gyrB, and rpoD, respectively, indicating their affiliation to the ST2. The same allelic profile combination was obtained for the strain CPBF 213 isolated from B. oleracea var. costata in Portugal ( Table  2). The obtained results indicate the absence of genetic diversity among the Serbian Xcc strains originating from cabbage, but also highlight ST2 as a new allelic profile found for the cabbage Xcc strains. Conversely, genetic heterogeneity was confirmed among eight strains of different cabbage types (B. oleracea var. capitata, B. oleracea convar. capitata var. alba and B. oleracea convar. capitata var. sabauda) used for comparison -CPBF 1135, CFBP 6865, CPBF 1136, CFBP 1712, CPBF 604, ISPaVe 1032, DAPP-PG 249, and LMG 8001, which were proved to belong to the ST5, ST9, ST10, ST20, ST29, ST31, ST32, and ST33, respectively.
The obtained MS tree shown in Figure 1 represents the distribution of STs among hosts. No strict connection between a host of isolation and ST affiliation was noticed, meaning that the Xcc isolates obtained from the same host can be genetically different. The central position of ST5 on the obtained tree indicates that it is the founder genotype, and diversification of the remaining 32 STs (ST1-ST4, ST6-ST33) included in comparison is shown. The ST2 differs from the founder genotype ST5 in one (rpoD) locus out of four tested.
A recent MLST-based study performed with the same four genes (P-X-dnaK, fyuA, gyrB, and rpoD) revealed the presence of as many as five (ST3, ST5, ST9, ST26, and ST47) different STs in the collection of fifteen Serbian Xcc strains obtained from winter oilseed rape (Brassica napus) over a nine-year period (2010-2018) (Jelušić et al., 2020). In another study on Xcc strains from Serbia, performed by Popović et al. (2019), which was based on the phylogenetic analysis of ten housekeeping genes (dnaK, fusA, fyuA, rpoD, gapA, gyrB1, gyrB2, gltA, lacF, and lepA), it was noticed that strains that originated from broccoli, cauliflower, collard greens, kohlrabi and kale, along with cabbage, were genetically homogenous. The absence of genetic diversity among the strains obtained in 2014 indicates that the long tradition of brassica crops cultivation in Serbia has not caused an evolution of different Xcc populations. Based on DNAfingerprinting profiles obtained in 2010 after performing repetitive element palindromic PCR (rep-PCR) with the primers GTG 5 and ERIC, Serbian cabbage Xcc strains were distinguished from Xcc strains isolated from kale and broccoli in the same year (Popović et al., 2013). An allelic-profile determination study performed with seven genes (atpD, dnaK, efp, fyuA, glnA, gyrB, and rpoD) for a collection of 31 Italian Xcc strains from various cruciferous crops (broccoli, cabbage, cauliflower, kale, kohlrabi, Savoy cabbage, seakale and rutabaga) showed the presence of 14 STs (Bella et al., 2019). In the same study, five strains from cabbage (ISPaVe 1032; PVCT 189.1.1; ISCI 88; OMP-BO 588/90; DAPP-PG 249) had four different allelic profiles (AP1=ST5, AP3=ST32, AP4=ST10, and AP8=ST31). In conclusion, the present study challenges our earlier knowledge about the genetic diversity of Xcc within specific plant species, and confirmed genetic homogeneity of cabbage Xcc strains, revealing the presence of only one ST (ST2), which was additionally found to be a new Xcc ST on cabbage.