Peach leaves with bacterial shot hole symptoms were collected, and tissue sections from the lesions on the diseased leaves were disinfected with 70% ethanol, cut, crushed using a mortar and pestle in sterile distilled water, and serially diluted. Each diluted sample (100 μl) was spread onto nutrient-yeast-glycerol (NYG) agar (Tan et al., 2014) and incubated at 28°C for three days. On the plates containing the 10⁻³ diluted samples, 20-30 colonies appeared. The yellowish and transparent colonies were purified and identified. One positive strain was designated as Xap1, it was subcultured on NYG agar at 28°C and maintained in a glycerol suspension (20%, w/v) at −70°C.

The isolate was identified using 16S ribosomal RNA (Lane, 1991; Snaidr et al., 1997) and a putative ABC transporter ATP-binding protein gene sequence (Pagani, 2005; Pothier et al., 2011). The primers for the 16S rRNA gene sequence were 9F (5′-GAGTTTGATCCTGGCGGCTCAG-3′) and 1512R (5′-ACGGCTACCTTGTTACGACTT-3′), targeting all bacterial members. In recent years, several bacterial genomes have been sequenced, however, the genes sequences can be partial due to the whole genome shotgun sequencing method, which limiting the prediction of genes structural and functional properties. It is known that the putative ABC transporter ATP-binding protein gene sequence is a species-specific sequence for X. arboricola pv. pruni and has been widely used for its identification (Pagani, 2005; Pothier et al., 2011), furthermore, it was well studied and the gene sequences are available on National Center for Biotechnology Information (NCBI) website, therefore, the putative ABC transporter ATP-binding protein gene sequence was chosen as a target sequence for the identification of X. arboricola pv. pruni. Primers targeting the putative ABC transporter ATP-binding protein gene sequence were used for later confirmation, the forward and reverse primers used were Y17CoF (5′-GACGTGGTGATCAGCGAGTCATTC-3′) and Y17CoR (5′-GACGTGGTGATGATGATCTGC-3′), respectively. The Y17CoF/Y17CoR PCR primers are known for their detection of X. arboricola pv. pruni (Pagani, 2005). The LAMP primers were designed using PrimerExplorer V4 (http://primerexplorer.jp/e/), four primers were used in this study (Fig. 1): outer primers F3 (5′-AGACGTCTAGGGCAAGCC-3′) and B3 (5′-TGCTAGAACTGACGCTGAGA-3′) and inner primers FIP (5′-ATGGGGGCCCAATCTCTAGCAATTTTGTCGGGCGACAACTTCAAC-3′) and BIP (5′-TCCACACCGATTTGCAGAAGGATTTTCGTGAGTCAATCGCAGTGT-3′). A TTTT linker was added to the inner primer to form a loop structure.

Genomic DNA was extracted using the hexadecyl trimethyl ammonium bromide (Sigma-Aldrich, St. Louis, MO, USA) purification method, which has been described previously (Ausubel et al., 1992). To amplify the 16S rRNA gene sequence and the putative ABC transporter ATP-binding protein gene sequence, the PCR condition was conducted as following: initial denaturation at 95°C for 5 min followed by 35 cycles at 95°C for 30 s, 55°C or 60°C for 30 s, 72°C for 1.5 min, and a final extension step at 72°C for 7 min. The obtained sequences were compiled using Seq-Man software (DNASTAR, Madison, WI, USA). Pairwise 16S rRNA gene sequence similarities with closely related species were calculated using both the EzBioCloud server (Yoon et al., 2017) and the BLAST search program at the NCBI website (http://blast.ncbi.nlm.nih.gov/Blast.cgi). For species-level analysis, the specific sequence of the putative ABC transporter ATP-binding protein was analyzed using the BLAST search program on the NCBI website.

The LAMP reaction was contained 2 μl of 10× Bst reaction buffer, 0.5 μl of 10 mM dNTP, 0.5 μl of 10 pM F3 and B3, 2 μl of 10 pM FIP and BIP, and 1 μl of X. arboricola pv. pruni genomic DNA for a final volume of 19 μl (supplemented with double distilled water). To enhance the sensitivity of the LAMP assays, pre-heating took place at 60°C for 5 min. Following pre-heating, 1 μl of Bst DNA polymerase (8 units/μl, large fragment; New England Biolabs, Beverly, MA, USA) was added. To optimize the reaction temperature, the LAMP reaction was performed under various thermal conditions (50°C, 53°C, 55°C, 58°C, 60°C, 63°C, and 65°C). The resulting LAMP amplicons were visualized by adding 1 μl of 5-times diluted original SYBR Green I (Thermo Fisher Scientific, Waltham, MA, USA) to the reaction tube. A positive reaction was green, whereas the color remained orange to the naked eye for a negative reaction. An efficient reaction produced a white precipitate when visualized using a UV illuminator. Three microliters of LAMP products were electrophoresed using 2% agarose gels, followed by staining with ethidium bromide. And two microliters of LAMP products were digested by a restriction enzyme (HaeIII), followed by electrophoresis analysis.

The proposed LAMP method was applied for the diagnosis of field samples. Thirty-two peach varieties were used for the validation testing, 4 peach varieties were tested using diseased leaves, and 31 peach varieties were tested using healthy leaves. Diseased fruit collected from several peach orchards was also tested. All of the field samples were tested in triplicate, with peach leaf samples (diameter 1 cm) or fruit samples (0.05 g) ground in 300 μl of plant extraction buffer (200 mM Tris [pH 8.0]; 250 mM NaCl; 25 mM EDTA [pH 8.0]; 0.5% sodium dodecyl sulfate). The supernatant was diluted 10 times and then 1 μl of the diluted sample was used for the LAMP reaction. Two microliters of LAMP products were digested with the HaeIII restriction enzyme, with incubation at 37°C for 4 h. Electrophoresis was conducted using 2% agarose gels after enzyme digestion, followed by staining with ethidium bromide.

Colonies of the Gram-negative isolate Xap1 were grown on NYG agar at 28°C for 2 days. The cells were yellowish, glistening, circular, convex, mucous, slimy, and smooth. Comparative 16S rRNA gene sequence (1,398 bp) analysis revealed that strain Xap1 was phylogenetically affiliated with the members of the genus Xanthomonas, based on information from the EzBioCloud database. The 16S rRNA gene sequence similarity of Xap1 was investigated on the NCBI website, and it was found that it had a high similarity (100%) with various X. arboricola pv. pruni strains, such as 17-330, 17-328, 17-326, 17-324, 17-195, and a few X. campestris strains (MSSRF_2C28, YNCCT1, and YNCCT6). Due to the large number of closely related X. arboricola pv. pruni strains, it was impossible to identify isolate Xap1 at the species level using the 16S rRNA sequence alone. It is known that the putative ABC transporter ATP-binding protein gene sequence is unique to X. arboricola pv. pruni species and has been widely used for its identification. Therefore, the putative ABC transporter ATP-binding protein gene sequence (898 bp) obtained from Xap1 was analyzed using the BLAST search program on the NCBI website. According to the alignment, strain Xap1 exhibited 100% similarities with X. arboricola pv. pruni XAP HU1 and XAP HU2 (accession number: KY039173 and MG049921), and 99.78% similarities with X. arboricola pv. pruni BJ15 and CFBP 5530 (accession number: MF362224 and HQ896469). Moreover, the 16S rRNA gene and ABC transporter ATP-binding protein gene sequences of strain Xap1 were compared with the 13 X. arboricola pv. pruni strains (14-199, 15-096, 15-101, 15-106, 16-143, 16-153, 16-156, 16-244, 17-178, 17-195, 17-210, 17-213, and 17-215) which provided by Rural Development Administration, the 16S rRNA gene and ABC transporter ATP-binding protein gene sequences of strain Xap1 had a high similarity (100%) with the sequences derived from the thirteen X. arboricola pv. pruni strains. Based on this sequence data analysis, the isolate Xap1 was identified as X. arboricola pv. pruni. The 16S rRNA gene sequences of strains Xap1, 14-199, 15-096, 15-101, 15-106, 16-143, 16-153, 16-156, 16-244, 17-178, 17-195, 17-210, 17-213, and 17-215 were deposited in NCBI GenBank with following accession number: LC507165, LC507166, LC507167, LC507168, LC507169, LC507170, LC507171, LC507172, LC507173, LC507174, LC507175, LC507176, LC507177, and LC507178, respectively.

To optimize the reaction conditions, the LAMP reaction was tested at various temperatures (Fig. 2). High efficiency was achieved at temperatures of 65°C, 63°C, 60°C, and 58°C for 45 min, with 60°C identified as the optimal temperature. The positive reactions all turned green after the addition of SYBR Green I, while the negative reactions remained orange. To confirm specific amplification, the LAMP products were digested with the HaeIII restriction enzyme, which produced predicted fragment sizes of 88 bp and 114 bp. The difference in the LAMP products before and after enzyme treatment was analyzed using electrophoresis (Fig. 3).

To determine the analytical sensitivity of LAMP testing, the genomic DNA purified from Xap1 was diluted down to the following concentrations: 1 (180 ng/μl), 1/2, 1/4, 1/8, 1/10, 1/32, 1/50, 10⁻², and 10⁻³ (Fig. 4). It was found that 1.8 ng/μl of genomic DNA could be detected by the LAMP reaction (Fig. 4, lane 8). To evaluate the species specificity of the designed primers, 12 other Xanthomonas species strains and six strains from other genera were tested using the LAMP reaction with their genomic DNA (Table 1), and specific PCR was used to compare the sensitivity of the proposed LAMP assays. All of the Xanthomonas and non-Xanthomonas strains were tested with PCR using primers Y17CoF/Y17CoR. It was found that the LAMP assay was more accurate than the traditional PCR method, with only the strain Xap1 producing a positive reaction in the LAMP test, while two strains of X. campestris pv. hyacinthi and Acidovorax facilis had positive reactions in the PCR test using primers Y17CoF/Y17CoR. LAMP specificity was also tested using 1 μl of genomic DNA from 31 healthy peach varieties (Table 2). It was subsequently found that the LAMP assay was able to specifically detect X. arboricola pv. pruni.

In addition, the LAMP method was tested with 13 X. arboricola pv. pruni strains isolated from various regions in Korea (Table 3). Conforming to the results for regular PCR analysis using the primers Y17CoF/Y17CoR, all of the X. arboricola pv. pruni strains exhibited a positive reaction in the LAMP assays, indicating that the proposed LAMP primer system can specifically diagnose the presence of X. arboricola pv. pruni.

The proposed LAMP assay was validated using plant samples from the field. For this purpose, healthy leaves and leaves with shot hole disease from naturally infected plants were tested. Leaf samples from 32 peach varieties were analyzed with the proposed LAMP assay (Table 2, Fig. 5A), and it was found that only the diseased leaves had a positive reaction, in particular, the LAMP assay were succeed for leaves without invisible symptoms, with the healthy leaves producing a negative reaction (Fig. 5B). Confirming the specific amplification, all of the positive LAMP products were digested by the HaeIII restriction enzyme (Fig. 5C). This indicates that the proposed LAMP assay successfully detected the pathogen X. arboricola pv. pruni in the diseased peach leaves. The effectiveness of the LAMP method was also validated on peach fruit bearing the bacterial shot hole disease (Fig. 5A), all of the diseased fruit had a positive reaction in the LAMP assays (Fig. 5B and C).

In conclusion, the LAMP method was successful in the rapid and accurate detection of field samples, confirming that it is a reliable diagnostic method for the specific detection and identification of X. arboricola pv. pruni in peach orchards in Korea.