Pseudomonas syringae pv. syringae causes bacterial canker on Japanese quince (Chaenomeles japonica)

Copyright: © 2022 T. Allahverdipour, M. Rastgou, H. Rahimian, E. Noris, S. Matić. This is an open access, peerreviewed article published by Firenze University Press (http://www.fupress. com/pm) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


INTRODUCTION
Japanese quince (Chaenomeles japonica (Thumb.) Lindley ex Spach) is one of the four species of Chaenomeles Lindl. (Maloidae, Rosaceae). This small tree produces white to pink flowers in late winter or early spring. The flowers have five petals and can grow up to 4.5 cm in diameter. The tree is spiny and bears simple alternately arranged leaves with serrated margins. Being native to Eastern Asia and naturally grown in central and Southern Japan, it is also cultivated in many countries including Iran (mostly together with forsythia), primarily as an ornamental plant for its showy flowers. Moreover, Japanese quince is also cultivated in Latvia, Lithuania, Russia, Poland, Belarus, Sweden, and Finland for its astringent apple-like pome fruits used in preserves and liqueurs, with potential as an alternate fruit crop (Jakobija and Bankina, 2018). Although Japanese quince is usually described as a healthy ornamental plant (Norin and Rumpunen, 2003), it can be attacked by a few pests and diseases, including aphids and scales, fungi and fire blight, reported as the main bacterial disease (Jewell, 1998).
Pseudomonas syringae is a heterogeneous group including more than 60 pathovars, belonging to the RNA homology group I of Pseudomonas, subclass γ from the Proteobacteria (Young, 2010). Its pathovars are host-specific on a range of plants, including beans, grasses and Prunus species (Little et al., 1998). Pseudomonas syringae pv. syringae (Pss) ( van Hall, 1902) causes bacterial canker and blast of stone fruit trees. It is one of the most important plant pathogens, causing foliar necrosis in host plants and hypersensitive responses (HR) in non-hosts (Bender et al., 1999;Collmer et al., 2000), leading to economic losses on more than 180 plant species (Bultreys and Kaluzna, 2010;Young, 2010). Pss is also an ice nucleation-active (INA + ) pathogen causing frost injury to plants (Hirano and Upper, 2000).
In Iran, Pss strains were isolated and characterized as causal agents of bacterial canker of stone fruit hosts (cherry, peach, apricot), almond, olive, rose, mallow, pelargonium, pear, and quince, red streak of sugarcane, blast of mandarin, and leaf blight of wheat, rice, oat, barley, and cabbage, from different areas (Bahar et al., 1982;Banapour et al., 1990;Rahimian, 1995;Najafi Pour and Taghavi, 2011;Aeini and Khodakaramian, 2018;Khezri and Mohammadi, 2018;Vasebi et al., 2019;Basavand et al., 2021). Japanese quince trees showing sunken brown cankers were observed in the Sari county during the 2016 growing season. As these trees are grown in parks, often close to stone fruit and pome fruit orchards, the aim of the present study was to determine the causal agent of canker symptoms on Japanese quince using pathogenicity, phenotypic, biochemical, and molecular analyses.

Sampling and isolation of bacteria
During the 2016 growing season, twigs of Japanese quince plants with canker symptoms and asymptomatic samples were collected from Sari in the Mazandaran province in the North of Iran. To isolate bacterial strains, plant tissues were surface sterilized, ground in 10 mL sterilized distilled water for 30 min, and a loopfull of the resulting suspension was streaked on nutrient agar with 0.5% sucrose (NAS). Resulting colonies were purified on nutrient agar, and selected according to morphology, color, and shape (Rademarker et al., 2000) for further biochemical, pathogenicity, and molecular tests.
Bacteria grown on NAS for 24 h at 28°C were inoculated into liquid culture, reaching 10 7 cfu·mL -1 . Ten µL of each suspension were inoculated into one-year-old Japanese quince and quince seedlings by subepidermal injections, infiltrating approximately 1 cm 2 of the bark tissue of each plant (Rahimian, 1995). For each bacterial isolate, ten plants were inoculated. Seedlings were incubated at 28°C in sterilized plastic boxes (≈80% moisture) for 1 month, and then observed for canker symptoms.
One-year-old seedlings inoculated with sterile water were used as negative controls (Jones, 1971). To fulfill Koch's postulates, after development of typical bacterial canker symptoms on inoculated seedlings, bacteria from each of the inoculated plants were re-isolated as above described.

Biochemical and physiological tests
All bacterial isolates were assessed for biochemical, physiological, and nutritional characteristics daily, for one month, as shown in Table 1 (Schaad et al., 2001). Carbohydrate utilization from glucose, fructose, sorbitol, melibiose, galactose, cellobiose, lactose, maltose or mannose (Fahy and Pearsly, 1983) was determined using the basal medium (Ayers et al., 1919). Nineteen isolates were selected for further analysis. The reference strain Pss ICMP 4916 was used in all tests as a positive control.

Isolation of total cell proteins and SDS-PAGE
Bacterial isolates were cultured on nutrient agar at 25-28°C for 24 h and single colonies were re-suspended in distilled water. Optical density of the suspensions was adjusted to 2.0 at 600 nm using a spectrophotometer. Samples were centrifuged at 2,576 g for 5 min, the pellet was re-suspended in 1 mL SDW and then dissolved in 0.1 volume of 10% SDS (Sodium Dodecyl Sulphate (SDS) and denatured by boiling for 5 min. Cells were then fractured by sonication (UP200, Hielscher) for 1 min at low amplification in an ice-cooling bath. The resulting cell suspensions were incubated for 5 min on ice and centrifuged at 10,000 rpm for 10 min. The supernatants representing total cell protein extracts were analyzed by SDS-PAGE, following addition of an equal volume of glycerol. SDS-PAGE analysis was carried out in 10% polyacrylamide gels using the Hoefer mini-electrophoresis system (Amersham Biosciences), according to Ahmadvand and Rahimian (2005). Electrophoresis was carried out at a constant voltage of 18 mA for approx. 2 h. Gels were then stained overnight in 50% (v/v) methanol, 10% (v/v) acetic acid, 0.1% Coomassie brilliant blue and then de-stained with in 50% (v/v) methanol and 10% (v/v) acetic acid under constant shaking. Protein profiles of Pss ICMP 4916 and P. viridiflava isolated from citrus in Mazandaran were used for comparison.

Genomic DNA fingerprinting using REP-PCR, BOX-PCR and IS50-PCR
Bacterial genomic DNA was extracted according to Bertheau et al. (1995) and Mahmoudi et al. (2007). Following quality checks on 1% agarose gel, DNA samples were kept at -20°C until use. Repetitive PCR (REP-PCR) using REP1R/REP2 primers (Versalovic et al. 1991) and BOXAIR primers (Louws et al., 1999), and Insertion sequence PCR (IS50-PCR) using 1S50 primers (Weingart and Völksch, 1997) were carried out for genomic DNA fingerprinting. Amplifications were performed in an Applied Biosystems 2720 thermal cycler (ThermoFisher) in 25 µL volumes, containing 200 mM of each dNTPs, 2 mM MgCl 2 , 1.5 pM primers, 1 U Taq polymerase, and 4 µL of DNA template. Thermal cycling was carried out with an initial denaturation cycle at 95°C for 5 min, followed by 35 cycles each consisting of denaturation at 94°C for 1 min, annealing at either 40°C (for REP primers), 52°C (for BOXAIR primers), or 44°C (for IS50 primers) for 1 min, extension cycle at 72°C for 3 min, with a final extension at 72°C for 7 min. Amplified products were separated by electrophoresis at 90 V in 2% agarose gels in Tris-Borate-EDTA (TBE) electrophoresis buffer (pH 8), and bands were visualized following ethidium bromide staining.

Data analysis
Data were statistically analyzed using a numerical taxonomy [amplified fragment of each strain scored as 1 (present) or 0 (absent)] and multivariate analysis system (NTSYS, version 2.1). Forty phenotypic characters were included in the analyses. Genetic relationships within and between bacterial strains were determined by cluster analysis, using the UPGMA method on distance calculated with the Jaccard coefficient matrices (Rohlf, 1990).

Partial ITS and rpoD gene amplification
The partial internal transcribed spacer (ITS) region and the RNA polymerase sigma factor D (rpoD)  gene were amplified using the primer pairs ITS-F (5' GGCTGGATCACCTCCTT 3') and ITS-R (5' TGC-CAAGGCATCCACC 3') and rpoD-F (5' ACCGCGCTG-GACACCGAAGGT 3') and rpoD-R (5' ACCACTTCT-GCTTGCTTGCGCTTGAGTG 3'). PCR amplification was carried out as described above. Thermal cycling was carried out with an initial denaturation cycle at 94°C for 5 min, 35 cycles each of denaturation at 94°C for 1 min, annealing at 58°C (ITS) and 54°C (rpoD) for 45 sec., extension at 72°C for 1 min, and a final extension at 72°C for 7 min (Versalovic et al., 1991). PCR products were purified and directly sequenced on both strands by Macrogen (South Korea). The sequences were edited using BioEdit v.7.0.5.2 and compared to the Pss sequences deposited in the GenBank database using BLASTn (https://www.ncbi.nlm.nih.gov). Nucleotide sequence similarity and multiple alignment and phylogenetic tree construction of partial ITS and rpoD gene sequences were carried out using the neighbor-joining method and bootstrap analysis replicated 1000 times, employing MEGA6 software (Tamura et al., 2013).

Ice nucleation activity
The ice nucleation activity of bacterial strains was conducted using the droplet-freezing method. Fifty µL droplets of bacterial suspension (each at 1 × 10 7 cfu·mL -1 ) were placed on paraffin soluble in xylenecoated aluminum foil and kept at a range of temperature between -5 and -10°C. Xanthomonas campestris and Pectobacterium carotovora were used as negative controls. Ice nucleation activity was observed visually after 30 sec (Schaad et al., 2001). The presence of the ice nucleation InaK gene was tested by PCR using InaK primers (forward 5´ GGGCCACAAAAGTATCCTGA 3´ and reverse 5´ CTGTGACAAGGTCGCTGTGT 3´) specific for Pss. Thermal cycling was carried out with an initial denaturation cycle at 94°C for 4 min, 30 cycles of denaturation each at 94°C for 1 min, annealing at 55°C for 1 min, extension cycle at 72°C for 45 sec., followed by a final extension cycle at 72°C for 6 min (Rashidaei et al., 2012).

Isolation and biochemical tests
The main symptoms of bacterial canker of Japanese quince in the surveyed area were blossom blight, twig wilting and dieback, blights around thorns, and sunken dark brown cankers, with no symptoms detected on leaves. Typical symptoms are shown in Figure 1.

Isolation and biochemical tests
In total, 33 bacterial colonies were obtained, and 19 of them (designated JQ1 to JQ19) were chosen as representative isolates, based on the colour and morphology of the colonies and on biochemical assays. All 19 isolates were Gram negative, produced fluorescent pigment on King's Medium B, were levan-positive, oxidase-negative, potato soft rot-negative, arginine dehydrolase-negative, induced hypersensitivity-reactions on tobacco and pelargonium, with characteristics of Pseudomonas LOPAT group Ia. These 19 isolates were selected for further biochemical and pathogenicity tests and for ice nucleation tests; their genomic DNA was tested for REP-, BOX-, and IS50-PCR, ITS and rpoD gene analysis. The results of biochemical and nutritional tests are reported in Table 1. According to the morphological and biochemical assays, and to the LOPAT and GATTa identification tests, all 19 representative isolates were identified as Pss.

Pathogenicity tests
All 19 strains inoculated were pathogenic to Japanese quince and quince seedlings. All isolates caused watersoaked lesions on twigs within 6 d after inoculation, and developed necrotic sunken cankers after 9 d (Figure 2). No symptoms occurred on twigs inoculated with sterile water as control. The symptoms observed on the inoculated seedlings were identical to those collected from the surveyed area. Koch's postulates were confirmed following re-isolation from healthy and young Japanese quince and quince seedlings; indeed, bacterial strains with the same characteristics as those inoculated were successfully re-isolated from artificially inoculated and diseased host plants.

Analysis of total bacterial cell proteins electrophoresis
Protein profiles of ten randomly selected isolates were compared visually to the protein profiles of the Pss reference strain ICMP 4916 and to P. viridiflava isolated from citrus in the Mazandaran province. The protein profiles of the Japanese quince isolates were similar to Pss ICMP 4916 reference strain and differed from P. viridiflava (Figure 3).

Genomic DNA fingerprinting using REP-PCR, BOX-PCR and IS50-PCR
The genomic DNA fingerprints of 13 randomly selected Pss strains isolated from Japanese quince and of the reference isolate of Pss were determined using BOX-, REP-and IS50-PCR products. Overall, the results showed approximately 721 fragments, with sizes ranging between 100 and 5000 bp (Figure 4). Five main clusters were obtained, with 67% similarity considering all the three PCR methods. In REP-PCR, JQ1 and JQ3 strains clustered together in cluster 1, JQ4 and JQ9   strains in cluster 2, JQ6 and JQ7 strains in cluster 3, and JQ2, JQ5, JQ6, JQ8, JQ10, JQ12 and JQ14 strains in cluster 4; the reference strain Pss made a separate cluster ( Figure 5A). According to BOX-PCR, the JQ1 strain grouped into cluster 1, JQ2 and JQ8 were grouped in cluster 2, JQ9, JQ4 and JQ10 grouped in cluster 4, and JQ6, JQ12, JQ14, JQ16 and Pss strains were grouped in cluster 5 ( Figure 5B). The strains were differentiated into five main clusters according to IS50-PCR. JQ1, JQ5 and JQ17 strains were grouped in cluster 1, JQ4, JQ12 and JQ16 in cluster 2, Pss strain in cluster 3, JQ3, JQ9 and JQ10 strains in cluster 4, and JQ6 and JQ14 strains in cluster 5 ( Figure 5C). The combined dendrogram from REP, BOX and IS50-PCR was also obtained for 13 strains, and obtaining seven main clusters with 67% similarity and four main groups with 56% similarity, including JQ1 and JQ16 as two separate clusters, JQ3, JQ4, JQ9 and JQ17 in cluster 3, and eight other strains in cluster 4 ( Figure 6).

ITS and rpoD gene amplification
PCR amplification with specific primers for ITS and the ropD gene correctly amplified specific bands of 600 bp and 850 bp in all the assessed strains ( Figure  7), while no amplification was obtained with negative control bacteria. The ITS sequences of two representative strains from Japanese quince, JQ4 and JQ9, had high similarity to each other (99%). The sequences of isolates JQ4 and JQ9 were deposited in the NCBI Gen-Bank database under the respective accession numbers KJ754379 and KJ754378 (for the ITS gene) and KJ769138 and KJ769139 (for the rpoD gene). The results confirmed that these isolates belong to Pss. A Neighbor-joining phylogenetic tree with 1000 bootstrap replicates showed that isolates JQ4 and JQ9 clustered together with the Pss AY342179, based on the ITS gene sequence (Figure 8), and grouped together with the Pss isolates JX867789 and KC852112 based on the rpoD gene sequence (Figure 9). The isolates and strains used for phylogenetic tree construction and their GenBank accession numbers are listed in Supplementary Table 1.

Ice Nucleation activity
Ice nucleation activity was positive for all the tested strains obtained from Japanese quince, showing freezing of water droplets in less than 30 sec. In contrast, isolates of X. campestris and P. carotovora did not induce freezing of water droplets in the same timeframe ( Figure  10A). Moreover, all PCR amplifications using Japanese  Figure 10B).

DISCUSSION
This study was carried out to identify the causal agent of canker of Japanese quince trees in the Sari County in the Mazandaran province located in the North of Iran. In total, 33 bacterial strains were isolated from canker lesions on plants located in different areas of the county. Characterization of the bacterial strains was achieved by biochemical, physiological and pathogenicity tests, indicating that the strains had phenotypic characteristics of Pss. Inoculations into quince and Japanese quince seedlings produced canker symptoms, consistent with previously described symptoms (Bahar et al., 1982;Banapour et al., 1990). Bacterial strains with the same features were also recovered from inoculated quince and Japanese quince seedlings, fulfilling Koch's postulates. However, as these tests do not precisely identify Pseudomonas strains (Peix et al., 2018), in order to perform a phylogenetic analysis, molecular tests based on DNA fingerprinting and partial gene sequence analysis using ITS and the housekeeping rpoD gene were also conducted. Thus, following BLAST analysis of the sequences of two representative bacterial strains JQ4 and JQ9 we could conclude that Pss was the causal agent of Japanese quince canker in the North of Iran.
All bacterial isolates belonged to Pss, previously reported as the causal agent of bacterial canker of many plants, including stone fruits, pear, quince, almond, olive, rose, malva and pelargonium, or causing red streak of sugarcane, blast of mandarin and leaf blight of wheat, rice, oat, barley and cabbage from different  areas of Iran (Bahar et al., 1982;Banapour et al., 1990;Rahimian, 1995;Najafi Pour and Taghavi, 2011;Aeini and Khodakaramian, 2018;Khezri and Mohammadi, 2018;Vasebi et al., 2019;Basavand et al., 2021). Genomic fingerprinting analyses, including REP-, BOX-and IS50-PCR showed some levels of variability among the Pseudomonas strains. The banding profiles for Pseudomonas isolates were also different and Pseudomonas isolates from the same region grouped into different clusters; therefore, genetic diversity among isolates was not related to their geographic origin. The methods here adopted are easy and rapid for distinguishing P. syringae at the pathovar level, with highly reproducible results. The present results agree with previous studies on genetic vari-ation of Pseudomonas strains (Weingart and Völksch, 1997;Little et al., 1998;Louws et al., 1999;Najafi Pour and Taghavi, 2011;Cetinkaya Yildiz et al., 2016;Abdellatif et al., 2020).
Noteworthy, this is the first study comparing these three genomic fingerprint analyses for the genetic characterization of Pss strains isolated from Japanese quince. Sequence analysis of the ITS and of the housekeeping rpoD gene, and the use of repetitive elements including REP-, BOX-and IS50-PCR have been described as useful methods for identification, classification and characterization of diversity of P. syringae isolates (Little et al., 1998;Najafipour and Taghavi, 2011). P. syringae can infect more than 200 plant species, due to a high degree of genetic variability (Ruinelli et al., 2019). The results of the present study show that a combination of different methods including biochemical, phenotypic, pathogenicity and molecular tests can be used for the identification of the causal agent of Japanese quince canker. Moreover, this is the first report of P. syringae as the causal agent of Japanese quince bacterial canker.
Bacterial canker of stone fruits, almond, olive, pear, and quince caused by Pss is a serious problem in all the areas where these crops are cultivated in Iran. Japanese quince, here reported as a new host for Pss in this country, can be grown within or near orchards and fields and can likely provide overwintering sites for Pss, serving as inoculum sources for disease development and outbreaks. The results of this study can facilitate further research on the ecology, epidemiology, diversity, and management of bacterial canker of Japanese quince and other host plants of Pss, not only in Iran but also in other countries. Further studies will allow to identify other infested areas and to identify the etiology of bacterial canker of Japanese quince in Iran.

AUTHOR CONTRIBUTIONS
TA designed and conducted experiments as a Masters student and analyzed the data. MR advised on experimental design, analyzed the data and wrote the manuscript. HR advised on experimental design, assisted with sample collection, and provided technical advice. EN and SM reviewed, critically revised and edited the manuscript LITERATURE CITED Abdellatif E., Kałużna M., Ferrante P., Scortichini M., Bahri B., … Rhouma A., 2020. Phylogenetic, genetic,