Morphology, DNA Phylogeny, and Pathogenicity of Wilsonomyces carpophilus Isolate Causing Shot ‐ Hole Disease of Prunus divaricata and Prunus armeniaca in Wild ‐ Fruit Forest of Western Tianshan Mountains, China

: Prunus divaricata and Prunus armeniaca are important wild fruit trees that grow in part of the Western Tianshan Mountains in Central Asia, and they have been listed as endangered species in China. Shot ‐ hole disease of stone fruits has become a major threat in the wild ‐ fruit forest of the Western Tianshan Mountains. Twenty ‐ five isolates were selected from diseased P. divaricata and P. armeniaca . According to the morphological characteristics of the culture, the 25 isolates were divided into eight morphological groups. Conidia were spindle ‐ shaped, with ovate apical cells and truncated basal cells, with the majority of conidia comprising 3–4 septa, and the conidia had the same shape and color in morphological groups. Based on morphological and cultural characteristics and multilocus analysis using the internal transcribed spacer (ITS) region, partial large subunit (LSU) nuclear ribosomal RNA (nrRNA) gene, and the translation elongation factor 1 ‐ alpha (tef1) gene, the fungus was identified as Wilsonomyces carpophilus . The 25 W. carpophilus isolates had high genetic diversity in phylogenetic analysis, and the morphological groups did not correspond to phylogenetic groups. The pathogenicity of all W. carpophilus isolates was confirmed by inoculating healthy P. divaricata and P. armeniaca leaves and fruits. The pathogen was re ‐ isolated from all inoculated tissues, thereby fulfilling Koch ʹ s postulates. There were no significant differences in the pathogenicity of different isolates inoculated on P. armeniaca and P. divaricata leaves ( p > 0.05). On fruit, G053 7m3 and G052 5m2 showed significant differences in inoculation on P. armeniaca , and G010 5m2 showed extremely significant differences with G004 7m2 and G004 5m2 on P. divaricata ( p < 0.05). This is the first report on shot ‐ hole disease of P. armeniaca (wild apricot) leaves and P. divaricata induced by W. carpophilus in China. Pathogens were re ‐ isolated from the resulting lesions and identified as described above [53]. Any changes in the tissues surrounding the inoculation sites on leaves and fruits were recorded over a period of 20 days. The length and width of lesions were measured using a vernier caliper (Mitutoyo 500 ‐ 196; Mitutoyo). Based on the length and width, the oval area of lesions was obtained to indicate lesion size [54]. The data were analyzed by Microsoft Excel 2010 and SPSS 20.0 software. One ‐ way ANOVA was used to determine significant differences in lesion size of different isolates and assess the pathogenicity of fungal isolates [55]. The treatment means were compared by Tukey’s honest significant difference (HSD) test at p = 0.05 [56]. pathogenicity tests also showed significant differences in some strains on inoculated fruits. This study has crucial implications for shot ‐ hole disease diagnosis and pathogen detection. We also suggest that the genetic diversity of W. carpophilus from the wild ‐ fruit forest needs further study.


Introduction
The Ili River Valley of the Western Tianshan Mountains is host to an important wild-fruit forest in Central Asia. The region has a great wealth of native plant species and has been considered an evolutionary center for fruit trees. Among them, Prunus divaricata and Prunus armeniaca are listed as endangered and protected plant species according to the regulations of the People's Republic of However, recent phylogenetic analyses based on large subunit (LSU) nuclear ribosomal RNA, internal transcribed spacer (ITS), and translation elongation factor 1-alpha (tef1) sequences supported Wilsonomyces as representing a distinct genus and its location in the Dothidotthiaceae [24]. The present study followed the classification of W. carpophilus as proposed by Marin-Felix et al. [24]. Conidia of W. carpophilus are initially fusiform, aseptate, and hyaline [26]. As conidia mature, they are delimited from conidiogenous cells by a single transverse septum, and multiple (3)(4)(5)(6)(7) transverse septa subsequently form as the conidia separate from the conidiogenous cells [26]. Shot-hole disease is most harmful under cool and wet spring conditions, although it can occur and cause damage at any time during the growing season following prolonged wet weather [27,28].
The aim of the present study was to investigate the cause of shot-hole disease associated with P. divaricata and P. armeniaca in different regions of the Ili wild-fruit forest in Xinjiang. Following surveys, morphological studies and DNA phylogenies were used to identify the disease causal agent. Moreover, pathogenicity studies were performed to determine the virulence of diverse fungal isolates in P. divaricata and P. armeniaca and confirm Koch's postulates.

Sampling and Isolation
Eight fruit samples from P. divaricata, and 14 fruit and 25 leaf samples from P. armeniaca were collected and brought to the laboratory. Dry specimens of infected leaves were deposited in the herbarium of the Forest Pathology Laboratory of Xinjiang Agricultural University. In order to isolate the disease causal agent, margins of infected fruit and leaf lesions were cut into small pieces (5 × 5 mm), the surface was disinfected by immersion in 75% ethanol for 30 s, followed by 3% NaClO for 5 min, then the samples were rinsed twice in sterile distilled water and dried in sterilized Petri dishes. Five pieces of diseased tissue from each symptomatic fruit and leaf sample were placed in a 90 mm Petri dish filled with full-strength potato dextrose agar (PDA) medium (39 g/L; Sangon Biotech). Plates were incubated at 25 °C under a 12 h photoperiod in a light incubator (GX-260A; Ningbo Southeast Instrument Co. Ltd.) for 6 days. Mycelial fragments taken from the growing colony margin were transferred to fresh PDA [31][32][33].
For single-spore isolation, conidia were scraped off with a sterile needle and suspended in 1 mL sterile distilled water with 0.1% Tween 20. An aliquot of 50 mL conidial suspension was spread on water agar in a Petri dish. After incubation at 25 °C for 24 h, single germinated conidia were transferred under a stereomicroscope to PDA plates and incubated at 25 °C for another 36 h for mycelium development [34]. Single-spore colonies were cultured on fresh PDA and pure cultures were stored in 15% glycerol at -80 °C [35,36]. Isolation plates were also checked regularly for the presence of the bacterial pathogen Xanthomonas arboricola pv. pruni.

Morphological Identification
Twenty-five isolates representing various morphological groups (three from P. divaricata fruits, 10 from P. armeniaca fruits, and 12 from P. armeniaca leaves) were used for morphological identification. Isolate morphology was assessed after 6 and 15 days of growth on PDA medium at 25 °C under the 12 h photoperiod. Anamorph characteristics, including the size and shape of conidiophores and conidia, were observed in water on a glass slide using an Olympus compound microscope (BX 53; Dongguan Yijiang Instrument Co. Ltd., Guangdong, China) [37]. Measurements of 50 conidia, including length and width, and 95% confidence intervals together with extreme values were determined [24].
The morphology of the 25 isolates growing on PDA medium was determined (Table 1). Mycelial plugs (5 mm in diameter) of the fungus growing on PDA were placed in the center of fresh PDA plates and incubated at 25 °C. Colony characteristics (color, concentric ring pattern, margin, presence of reproductive structures) were recorded. Mycelial growth rate was determined using three isolate replicates [38,39].
Conditions for PCR amplification were as follows: for ITS region: 94 °C for 2 min, followed by  [39]. The quality of the PCR products was checked by performing 1% agarose gel electrophoresis. DNA sequencing was performed using an ABI PRISM® 3730XL DNA analyzer with a BigDye®120 Terminator Kit v.3.1. The positive transformants were sequenced at Sangon Biological Engineering Co., Ltd. (Beijing, China) [46].

Phylogenetic Analysis
Sequences were assembled with the SeqMan program v.7.1.0 in DNASTAR Lasergene core suite software (DNASTAR Inc., Madison, WI, USA). Homologous sequences with high similarity from extype and non-type Wilsonomyces-like isolates were included to serve as phylogenetic references and obtained using the BLAST function in the National Center for Biotechnology (NCBI) database and extensive literature review. All sequences were subjected to Bayesian inference (BI) analysis using MrBayes 3.2.6 [47], maximum likelihood (ML) was performed using RAxML-HPC v.8 on XSEDE, and maximum parsimony (MP) analysis was performed using PAUP v. 4.0a150 [48][49][50]. Multiple sequences of concatenated ITS, LSU, and tef1 sequences were aligned using MAFFT v. 7 with default settings and edited manually using MEGA v.6.0. Stigmina platani was selected as an out-group in all analyses [50]. Phylogenetic trees were edited using FigTree v.1.4.0. The GenBank accession numbers for the sequences used in these analyses are given in Table 2 [24], and isolates collected for this study are reported in Table 1.
Bayesian inference (BI) analysis was performed using MrBayes v.3.2.6. Two independent analyses of two parallel runs and four chains were carried out for 5,000,000 generations and sampled every 5000 generations, resulting in 1000 trees in total. The first 25% of the resulting trees were eliminated as the burn-in phase of each analysis. Branches with significant BI posterior probability (BIPP) were estimated for the remaining 750 trees [51].
Maximum likelihood (ML) analysis was done using RAxML v.7.2.8 and a GTR (general-timereversible, one of the most popular models of nucleotide substitution because it constitutes a good trade-off between mathematical tractability and biological reality) model of site substitution including estimation of gamma-distributed rate heterogeneity and a proportion of invariant sites [48]. The branch support was evaluated with the bootstrapping method with 1000 bootstrap replicates [52].
A maximum parsimony (MP) analysis was conducted using the heuristic search option of 1000 random-addition sequences with tree bisection and reconnection (TBR) branch swapping algorithm. Clade stability was assessed with a bootstrap analysis of 1000 replicates [52]. Descriptive tree statistics for parsimony tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RC), and homoplasy index (HI) were calculated for the maximum parsimonious tree [51].

Pathogenicity Tests
Pathogenicity experiments were conducted in June 2018 to August 2019 in the Forest Pathology Laboratory of the College of Forestry and Horticulture, Xinjiang Agricultural University, Urumqi, China. In order to test for pathogenicity, young fruits and leaves were collected from healthy P. divaricata and P. armeniaca (wild apricots) from Xinyuan County of Ili, Xinjiang.
Leaves and fruits were inoculated using mycelial plugs (3 mm diameter). Inocula were prepared by growing individual fungal isolates on PDA at 25 °C in a 12 h photoperiod for 6 days. Healthy leaves and fruits of the same age and size were collected from trees, washed thoroughly with tap water, surface sterilized with 1% NaClO for 10 min, then washed with sterile distilled water three times, and placed on paper towels in the laminar flow cabinet for about 2 h. Inoculation was done following wounding with a sterile needle to create a single wound, which was then covered with a mycelial plug (3 mm) with the mycelium facing down, after being wrapped with parafilm. After 24 h, mycelial plugs were removed from the surface of the inoculated tissues. The control groups were mock inoculated with non-colonized PDA plugs. Inoculated leaves and fruits were placed on a grid tray inside a plastic crisper with water at the bottom and the lid closed to maintain high humidity. Crispers were incubated at room temperature at 25 ± 2 °C on a laboratory bench until development of symptoms. Each pathogenicity test used 10 replicates and all procedures were carried out under aseptic conditions. Pathogens were re-isolated from the resulting lesions and identified as described above [53]. Any changes in the tissues surrounding the inoculation sites on leaves and fruits were recorded over a period of 20 days. The length and width of lesions were measured using a vernier caliper (Mitutoyo 500-196; Mitutoyo). Based on the length and width, the oval area of lesions was obtained to indicate lesion size [54]. The data were analyzed by Microsoft Excel 2010 and SPSS 20.0 software. One-way ANOVA was used to determine significant differences in lesion size of different isolates and assess the pathogenicity of fungal isolates [55]. The treatment means were compared by Tukey's honest significant difference (HSD) test at p = 0.05 [56].

Field Symptoms and Isolating the Fungi
A total of 9600 leaves and 3200 fruits from 80 P. armeniaca trees were surveyed in Xinyuan, Yining, Huocheng, and Gongliu Counties; the average frequency of infection of leaves and fruits was 79.33% and 53.17%, respectively. Symptoms on leaves consisted of 1-2 mm circular spots, purplish in color with a yellow bordering halo, which eventually enlarged and became necrotic, causing abscission in the center of the lesion, giving the leaf the typical shot-hole appearance (Figure 1a). Symptoms on fruits appeared as raised brown sores or cracks, and the fruits were necrotic ( Figure  1f).
A total of 800 fruits from 20 P. divaricata trees were surveyed in Huocheng County, with an infection rate of 37.45%. Symptoms on P. divaricata fruits appeared as round spots, dark brown in the center and yellowish brown on the edges, with a hardened peel (Figure 1k).
Three types of fungi (Wilsonomyces-like, Didymella sp., and Alternaria sp.) were obtained from 705 tissue blocks of 47 samples. No bacterial colonies were isolated from any samples. Wilsonomyceslike isolations collected from P. armeniaca leaves accounted for 72.8% of the total isolates. All isolates collected from fruits of P. armeniaca and P. divaricata were Wilsonomyces-like. Occasionally isolates of Didymella sp. and Alternaria sp. were also collected from leaves, accounting for 9.6% and 17.6%, respectively, of the total fungi isolated. They were not included in morphological, phylogenetic, and pathogenicity studies because of their low frequency. Pure growths of 25 Wilsonomyces-like isolates (three from P. divaricata fruits, 10 from P. armeniaca fruits, and 12 from P. armeniaca leaves) were selected for single-spore purification for this study.

Cultural and Morphological Characteristics
Strong variability in cultural and morphological characteristics was recorded among the different isolates of Wilsonomyces-like on PDA medium. The 25 isolates of the pathogen were divided into eight morphological groups based on the shape of the colony margin, concentric growth patterns, and color in PDA medium. Groups I-V were characterized by a regular colony margin, nearly round, and an average growth rate of 0.47 cm/d. Group I, including nine isolates (G048 3m3, G048 5m2, G048 7m1, G049 7m1, G053 5m1, Y039 3m3, Y046 7m2, Y049 7m1, Y045 5m2-2), was characterized by dull white to pale yellow colonies with no concentric ring pattern (Figure 2a); Group II included four isolates (G052 5m2, G052 5m3, G053 7m3,Y057 7m3) with gray to light yellow colonies with concentric ring patterns (Figure 2 b); Group III included four isolates (G004 5m2, G004 7m2, G010 5m2, G059 5m2) with light-yellow to brown colonies with no concentric ring pattern (Figure 2c); Group IV included a single isolate (Y037 7m2) with dull olivaceous to brown colonies and a concentric ring pattern ( Figure 2d); Group V included three isolates (Y035 5m1, Y038 7m2, Y040 7m2) with brown to dark brown colonies with a concentric ring pattern (Figure 2e). Groups VI to VIII had irregular margins, no concentric ring pattern, and slow growth, averaging 0.21 cm/d. Group VI comprised two isolates (Y043 7m1, Y052 7m1) with dull white to dark gray colonies ( Figure 2f); Group VII had one isolate (Y048 5m2) with dark olivaceous to black colonies ( Figure 2g); Group VIII had one isolate (G048 5m3) showing dull white to pale yellow colonies (Figure 2h).
Three isolates obtained from P. divaricata fruits were classified into the morphological Group III. Twelve isolates from P. armeniaca leaves were classified into six morphological groups: I, II, IV, V, VI, and VII. Ten P. armeniaca fruit isolates were classified into four morphological groups: I, II, III, and VIII. Group I is the most common morphological type of P. armeniaca.
The shape and color of conidia were consistent among the isolate groups. Conidia were produced after 7 days in PDA medium and were most abundant at the edge and darkening part of the colony. Conidia were spindle-shaped, with ovate apical cells and truncate basal cells (2.11-4.

Phylogenetic Analysis
The ITS sequence of 25 Wilsonomyces-like isolates showed high similarity to Wilsonomyces carpophilus reference sequences available in GenBank. The ITS, LSU, and tef1 sequences obtained from the 25 isolates in this study were submitted to GenBank ( Table 1).
The ITS, LSU, and tef1 sequences from 44 taxa belonging to two families (genera of Dothidotthiaceae and Didymellaceae) were used in the phylogenetic analysis. The combined datasets comprised 2620 characters, of which 1724 characters were constant, 461 variable characters were parsimony uninformative, and 435 were parsimony informative. In the MP analysis, the reconstructed trees were 1367 steps long, with CI = 0.815, RI = 0.866, RC = 0.706, and HI = 0.185. BIPP, ML, and MP bootstrap support values above 50% are given above or below the branches (Figure 3). The three types of analysis trees resulting from the concatenated dataset showed the same relationships among Wilsonomyces-like isolates. The reference strains (CBS 159.51 and ex-epitype CBS 231.89) were clustered with the new isolates and had a higher support value (BI/ML/MP = 85/89/100) in Figure 3. The 25 W. carpophilus isolates also showed high genetic diversity in the phylogenetic analysis, and the phylogenetic groups did not correspond to the morphological groups. region, partial large subunit (LSU), and translation elongation factor 1-alpha (tef1) gene sequences. Backbone of the tree was constructed using Bayesian analysis. Stigmina platani was selected as an outgroup. Bootstrap percentages of Bayesian posterior probabilities, maximum-likelihood, and maximum parsimony from 1000 replicates are shown, from left to right, on the deep and major resolved branches. Morphological groups in this study were labeled after isolate numbers.

Pathogenicity Tests
The 25 isolates were used for pathogenicity tests. For inoculation of P. armeniaca, 12 isolates of W. carpophilus were inoculated on leaves, and 10 isolates of W. carpophilus were inoculated on fruits. All leaves inoculated with of W. carpophilus developed circular brown lesions with pale centers within 3-5 days after inoculation. Then, about 0.5 mm of the mesophyll tissue at the outer edge of the lesions became thinner, and the boundary between healthy and necrotic tissue became obvious. After 17 days post-inoculation, the whole lesion collapsed, and a small hole formed (Figure 1b-d). No significant difference in virulence was found among the 12 W. carpophilus isolates inoculated on leaves (p = 0.346) (Figure 4a), and the average lesion size was 5.38 mm 2 .
Inoculated P. armeniaca fruits developed sunken necrotic lesions within 3-12 days after inoculation (Figure 1g-i). There was a very significant difference between strains G053 7m3 and G052 5m2, with lesions averaging 10.75 and 5.57 mm 2 , respectively; the other eight inoculated strains showed no significant difference (p < 0.0001) (Figure 4b). Both leaves and fruits of P. armeniaca in the control group remained asymptomatic (Figure 1e,j). Lesions produced in the inoculated leaves and fruits were similar to symptoms observed in the field. W. carpophilus was re-isolated from symptomatic tissue in all inoculated leaves and fruits, thus fulfilling Koch's postulates.
For P. divaricata, three isolates of W. carpophilus from fruits were inoculated onto leaves and fruits, and lesions developed in all inoculated organs. Inoculated leaves developed circular brown lesions within 5-10 days after inoculation. Although the mesophyll tissue became thinner, the lesion did not detach to form a perforation (Figure 1n). There were no significant differences among the three isolates tested (p = 0.072) (Figure 4c). The average lesion size in leaves was 4.21 mm 2 .
Inoculated fruits of P. divaricata developed brown necrotic lesions within 3-6 days after inoculation with W. carpophilus isolates (Figure 1l). The isolate G010 5m2 showed extremely significant differences with G004 7m2 and G004 5m2. There was no significant difference between G004 7m2 and G004 5m2 (p = 0.014) (Figure 4d). The average lesion size for inoculated G004 7m2, G004 5m2, and G010 5m2 was 20.02, 19.95, and 13.65 mm 2 , respectively. Lesions produced in the inoculated fruits were similar to symptoms observed in the field. No lesions were found in the various control treatments (Figure 1m,o). W. carpophilus was also re-isolated from all symptomatic leaves and fruits, thus fulfilling Koch's postulates. .

Discussion
Shot-hole disease affecting stone fruits in China has been historically attributed to the bacterium Xanthomonas arboricola pv. pruni. Bacterial spot was shown in eight peach-producing areas in China [57,58]. The shot-hole disease affecting stone fruits in the Western Tianshan Mountains was generally thought to be caused by Xanthomonas arboricola pv. pruni, but P. armeniaca fruit spot caused by W. carpophilus in Gongliu County, Xinjiang, was reported in 2019 [1,59]. Our study indicates that W. carpophilus not only infected P. armeniaca fruits, but also caused shot-hole disease in P. divaricata fruits and P. armeniaca leaves [59].
The shot-hole disease of stone fruits caused by W. carpophilus can be confused with the bacterial spot disease caused by X. arboricola pv. pruni, due to overall similar symptoms [60]. It is likely that the two diseases were confused in previous efforts to document diseases affecting the wild-fruit forest of the Western Tianshan Mountains. Therefore, lesion characteristics alone should not be used to diagnose fruit and leaf spot diseases of stone fruits, and isolation procedures should be conducted systematically for accurate disease diagnosis.
In this study, the conidia characteristics of the isolates were similar to those of W. carpophilus: conidia ( (2)(3)(4)(5), colorless and transparent when immature, turning sub-hyaline, dark olivaceous to dark brown with age [24]. We observed large differences in culture morphology between the 25 isolates of W. carpophilus. According to morphological characteristics such as the shape of the colony margin, concentric growth patterns, and color, eight morphological groups were identified. The isolates obtained from P. divaricata fruits had consistent morphology and were classified into Group III. Strains derived from P. armeniaca leaves and fruits showed the highest variation in colony morphology. Strains derived from P. armeniaca fruit were divided into four groups, and strains derived from P. armeniaca leaves were divided into six groups, and the majority of P. armeniaca strains were included in Group I. However, the P. armeniaca strains could not be grouped according to either their geographical origin or the host tissue from which they were isolated. Although there were differences in culture morphology, no significant differences in color and size of the conidia were found among these isolates, indicating that these variations are more pronounced at the species level. These results agree with those reported by Nabi et al., who determined significant variations of morphology, particularly cultural characteristics, of W. carpophilus isolates from stone fruits in India, and suggested that the nonsignificant variation in conidial shape and color recorded in different isolates explained that these variations are much more pronounced at the species level and less pronounced at intra-species levels [61]. Studies of other pathogens (Magnaporthe grisea, Colletotrichum gloeosporioides, and Colletotrichum capsici) also indicate that conidial shape and color are significant only at the species level only [62,63].
Similarly, phylogenetic analysis shows an obvious clustering of W. carpophilus and the 25 isolates in this study into a single branch with higher support value. The W. carpophilus isolates also had high genetic diversity. However, morphological groups did not correspond to phylogenetic groups, despite some genetic variation among W. carpophilus isolates. Different W. carpophilus isolates have been previously reported to have high intraspecific genetic diversity. Ahmadpour et al. showed high genetic diversity in 28 isolates of W. carpophilus from different regions in Iran using DNA fingerprinting by random amplified polymorphic DNA polymerase chain reaction (RAPD-PCR) and four random primers [64]. A high level of polymorphism in different isolates of W. carpophilus in Kashmir using seven ISSR (inter-simple sequence repeat) markers also indicated that these markers were suitable for studying the genetic diversity in shot-hole pathogens [61]. The 25 isolate sequences in this study showed high similarity with W. carpophilus reference sequences available from GenBank. Wilsonomyces is a monotypic genus. Helminthosporium carpophilum was initially described and transferred to different genera until Adaskaveg et al. introduced Wilsonomyces to accommodate this species [21]. Sutton regarded W. carpophilus as a synonym of Thyrostroma [25]. However, Marin-Felix et al. separated these last species based on LSU, ITS, and tef1 sequence analysis and introduced Thyrostroma compactum, confirming that Wilsonomyces represents a distinct genus belonging to the Dothidotthiaceae [24]. In this study, the results of molecular phylogenetic analysis using multiple genes are consistent with those of Marin-Felix et al.
Isolates of W. carpophilus obtained from fruits of P. divaricata showed pathogenicity in fruits and leaves of P. divaricata. Isolates obtained from P. armeniaca leaves and fruits also were pathogenic to this host. Analysis of the morphological characteristics and gene sequences confirmed that strains reisolated from all lesions produced in the pathogenicity study were identical to the inoculated ones, thereby fulfilling Kochʹs postulates. The pathogenicity of P. divaricata and P. armeniaca determined that there were no significant differences among different isolates inoculated on P. divaricata and P. armeniaca leaves (p < 0.05). There were significant differences in inoculation on P. divaricata and P. armeniaca fruits (p > 0.05). The symptoms produced on the detached fruits were bigger than those on the fruits in the field, which may be due to the wound, the amount of fungus inoculated, or the influence of environmental factors. Nabi et al. observed that W. carpophilus produced more severe lesions on injured than uninjured tissues, and the isolates varied significantly with respect to the incubation period and the subsequent size (diameter) of the lesions produced [61].

Conclusions
This study presents the first research on the serious shot-hole disease on P. divaricata and P. armeniaca (wild apricot) leaves caused by W. carpophilus in the wild-fruit forest of the Western Tianshan Mountains. Twenty-five pure isolates of W. carpophilus were selected from P. armeniaca leaves and fruits and P. divaricata fruits. Morphological characteristics, phylogenetic analysis, and pathogenicity were recorded for all strains. All re-isolated strains from all lesions were identified, thereby fulfilling Kochʹs postulates. According to the morphological characteristics, the 25 isolates were divided into eight groups. The pathogenicity tests also showed significant differences in some strains on inoculated fruits. This study has crucial implications for shot-hole disease diagnosis and pathogen detection. We also suggest that the genetic diversity of W. carpophilus from the wild-fruit forest needs further study.