Apple (Malus domestica) is one of the most significant fruit crops in Korea, with a total production of approximately 566,000 tonnes on 34,600 ha in 2022 (FAOSTAT, 2022). However, apple production faces various constraints, including extreme weather conditions and various diseases caused by fungi, bacteria, and viruses. Among these, bitter rot caused by Colletotrichum species is a major factor limiting production, characterized by sunken lesions on the fruit that lead to decay and a significant reduction in quality and market value (Kim et al., 2020).
In Korea, species such as C. gloeosporioides, C. fructicola, C. siamense, C. fioriniae, C. nymphaeae, and C. aenigma have been reported to be associated with apple bitter rot over the last two decades (Kim et al., 2018; Oo et al., 2018). Currently, chemical control using fungicides like benzimidazoles and strobilurins is the most effective management strategy for this disease. Benzimidazoles, such as benomyl, target the β-tubulin gene to disrupt microtubule assembly and inhibit fungal growth (Zhou et al., 2016). Strobilurins, known as quinone outside inhibitors (QoIs), inhibit the mitochondrial respiratory chain by binding to the Qo site of cytochrome b (cyt b), thereby disrupting ATP production (Bartlett et al., 2002; Von and Link, 1986).
The consistent and intensive use of these site-specific fungicides has led to the emergence of resistant isolates among various fungal pathogens. Fungicide resistance is primarily driven by single nucleotide polymorphisms that result in specific amino acid substitutions within the target protein, commonly denoted by a notation indicating the original amino acid, its position, and the substituted amino acid. For benzimidazoles, resistance is linked to mutations at amino acid positions 198 and 200 of the β-tubulin (tub2) gene. For instance, E198A and E198K represent genotypes where the original glutamic acid (E) at the 198th position is replaced by alanine (A) or lysine (K), respectively (Albertini et al., 1999; Fujimura et al., 1992; Yarden and Katan, 1993). While multiple genotypes have been found in other hosts, only the E198A mutation has been reported in apple bitter rot in Korea to date (Cho, 2023).
Regarding strobilurins, resistance is associated with several key substitutions in the cyt b gene: F129L (phenylalanine [F] to leucine [L] at position 129), G137R (glycine [G] to arginine [R] at position 137), and G143A (glycine [G] to alanine [A] at position 143) (Ma and Michailides, 2005; Sierotzki et al., 2000). Among these, the G143A mutation confers the highest level of resistance and is the most common modification found in resistant Colletotrichum species causing apple bitter rot.
Although numerous studies have reported on the diversity and fungicide sensitivity of Colletotrichum species in Korea, comprehensive nationwide information on their long-term distribution and species-specific resistance remains inadequate. Therefore, this study aimed to evaluate the nationwide distribution of these species over a 10-year period (2015-2024) and monitor their resistance status against strobilurin and benzimidazole fungicides to provide a scientific basis for sustainable disease management in Korea.
Materials and Methods
Fungal isolation and culture conditions
Apples showing typical anthracnose symptoms were collected from five major apple-producing regions in South Korea (Gangwon, Gyeonggi, Chungcheong, Gyeongsang, and Jeolla), and the causal pathogens were isolated using the single-spore isolation method. A total of 319 Colletotrichum isolates were collected, including 66 in 2015 and 2016, 68 in 2022, 88 in 2023, and 97 in 2024. For isolates collected from 2022 to 2024, conidia were scraped from symptomatic tissues into sterile distilled water (SDW), adjusted to a concentration of 5 × 103 conidia/mL, and plated onto potato dextrose agar (PDA) by spreading 20 μL of the spore on a 90 mm Petri plate. After 48 hours of incubation at 25°C, single colonies were transferred to fresh PDA for further study. Isolates from 2015 and 2016 were retrieved from 20% glycerol stocks stored at −70°C and subcultured on PDA. All isolates were maintained at −70°C in 20% glycerol for long-term storage.
Genomic DNA extraction
Mycelial samples were harvested from 5-day-old cultures grown on PDA. The harvested mycelia were freeze-dried for 24 h using a freeze dryer (Ilshin BioBase, Dongducheon, Korea) and ground with glass beads in a bead beater (Next Advance Inc., Troy, NY, USA). Genomic DNA (gDNA) was extracted using the Exene™ Plant SV mini extraction kit (GeneAll Biotechnology, Seoul, Korea) following the manufacturer’s protocol. The extracted DNA was quantified using a NanoDrop spectrophotometer (K Lab Co. Ltd., Seoul, Korea), adjusted to a final concentration of 50 μg/mL, and stored at −20°C.
Molecular identification and phylogenetic analysis
For the 2015/2016 and 2022/2023 isolates, multi-locus phylogenetic analysis was performed using partial sequences of six genetic markers with the primer pairs listed in Supplementary Table 1: the internal transcribed spacer (ITS) region (White et al., 1990), actin (act) and chitin synthase 1 (chs-1) (Carbone and Kohn, 1999), glyceraldehyde-3-phosphate dehydrogenase (gapdh) (Guerber et al., 2003), β-tubulin (tub2) (Woudenberg et al., 2009), and calmodulin (cal), for which CL1 was adopted from O’Donnell et al. (2000) and CL2B was designed in this study. A 20 μL PCR mixture containing 12 μL SDW, 4 μL master mix (EzPCR 5X PCR master mix; Elpis Biotech, Daejeon, Korea), 1 μL each of the forward and reverse primers and 2 μL gDNA was prepared for each PCR reaction. The PCR reaction was performed in a thermal cycler (Bio-Rad 100TM; Bio-Rad Laboratories, Inc., Hercules, CA, USA) with the PCR conditions shown in Supplementary Table 1. The PCR products were separated by electrophoresis in 1% agarose gel in Tris-Borate-EDTA (TBE) buffer for 40 min at 100 V and viewed using a transilluminator. The PCR products were purified using the ExpinTM SV PCR mini (GeneAll Biotechnology) purification kit and sent for sequencing at Macrogen Inc. (Daejeon, Korea). Sequences were aligned using MAFFT v7 online (Katoh et al., 2019), and a phylogenetic tree was constructed using the maximum-likelihood method as implemented in MEGA11 software (Tamura et al., 2021). For the 2024 isolates, molecular identification was conducted via PCR using species-specific primers and PCR conditions (Supplementary Tables 2 and 3). The primers for C. fructicola, C. siamense, C. aenigma, and C. gloeosporioides were previously developed and validated for their specificity and sensitivity by Abdullahi and Kim (2024), whereas those for C. fioriniae and C. nymphaeae were adopted from Seo (2024). The primers targeting the C. gloeosporioides and C. acutatum species complexes were developed in this study. To ensure the accuracy of the 2024 identification, we validated the primers against the MLSA-confirmed reference isolates from the 2015-2023 collections. Each PCR assay was performed in triplicate to confirm reproducibility, and positive/negative controls were included in every run.
Morphological characterization and pathogenicity tests
Six representative isolates were selected for morphological and pathogenicity characterization based on phylogenetic analysis. Colony morphology and mycelial growth rates were assessed. To determine growth rates, a 5-mm diameter mycelial plug was taken from the edge of an actively growing colony and transferred onto PDA. The cultures were incubated at 25°C in the dark. After an initial 24 h of incubation, the colony diameter was recorded daily for 7 days. Colony color and texture were documented on the fifth day. The average growth rate (mm/day) was calculated based on the daily increase in diameter. Conidial and appressorial morphologies were examined using the slide culture technique. A 5 × 5 mm PDA block was placed aseptically on a glass slide within a moist chamber. The four sides of the agar block were inoculated with mycelial fragments of each representative isolate. A coverslip was placed over the agar, and the assembly was incubated at 25°C in the dark for 7 days. The shapes and sizes of conidia and appressoria formed on the underside of the coverslip were observed under a light microscope (Olympus Corp., Tokyo, Japan) at ×400 magnification. Mycelial growth rates and spore dimensions were subjected to analysis of variance (ANOVA), and means were compared using Duncan’s Multiple Range Test (DMRT) at P < 0.05.
Pathogenicity test
Healthy, unwounded mature apple fruits (cv. Hongro) were obtained from a commercial orchard. The fruits were washed with tap water, surface-disinfected in 1% sodium hypochlorite (NaOCl) for 3 min, rinsed three times with sterile distilled water, and air-dried in a laminar flow hood. The disinfected apples were then placed within plastic boxes lined with moist paper towels to maintain high humidity. For inoculation, a 3-mm diameter mycelial plug from each isolate was placed on the fruit surface, with the aerial mycelia facing the host tissue. Control fruits were inoculated with sterile agar plugs. Each treatment, including the control, was performed in triplicate. The inoculated fruits were initially incubated in sealed humid chambers at 25°C in the dark for 3 days. Subsequently, the covers were removed, and the fruits were kept on laboratory benches at room temperature until 10 days post-inoculation (dpi). Lesion diameters were measured at 13 dpi. Data were subjected to a one-way ANOVA using IBM SPSS Statistics (version 26; IBM Corporation, Armonk, NY, USA). Treatment means were separated using DMRT at P < 0.05. To fulfill Koch’s postulates, the causal agents were re-isolated from the lesions and re-identified. The entire experiment was conducted twice.
In vitro fungicide sensitivity assay
The sensitivity of 319 Colletotrichum spp. isolates to fungicides was evaluated using an in vitro agar dilution assay. Pyraclostrobin (a.i. 20%, WG) and benomyl (a.i. 50%, WP) were serially diluted in SDW to final concentrations of 20.0, 4.0, 0.8, 0.16, and 0.032 μg/mL, and 10.0, 1.0, 0.1, 0.01, and 0.001 μg/mL, respectively. These solutions were used to amend PDA at a 1% (v/v) ratio. A 3-mm diameter mycelial plug, taken from the margin of a 7-day-old colony, was transferred onto the fungicide-amended PDA plates. Non-amended PDA plates served as the control, and three replicates were maintained for each treatment. The plates were sealed with Parafilm (Parafilm M®; Bemis Company Inc., Neenah, WI, USA) and incubated at 25°C in the dark. After 5 days of incubation, the colony diameter was measured in two perpendicular directions. The final diameter was determined by averaging these measurements and subtracting the original 3-mm plug diameter. The mycelial growth inhibition rate (MGIR, %) was calculated using the formula: MGIR (%) = [(C − T) / C] × 100, where C is the radial growth of the control and T is the radial growth on the fungicide-amended medium. The median effective concentration (EC50) was determined by linear regression analysis of the probit-transformed inhibition rates against the log10-transformed fungicide concentrations using Microsoft Excel. Also, EC50 threshold values for classifying resistant and sensitive isolates were determined from the distribution of sensitivity data. They were confirmed by the presence of target-site mutations (G143A and E198A).
Detached fruit sensitivity assay
The in vivo sensitivity of Colletotrichum isolates to fungicides was evaluated using detached apple fruits. Two concentrations of pyraclostrobin and benomyl were prepared based on the manufacturers’ recommended field rates. For pyraclostrobin, final concentrations of 67 μg/mL (recommended concentration) and 13.4 μg/mL (1/5 of the recommended concentration for apple field treatment) were prepared. For benomyl, 325 μg/mL (recommended concentration) and 32.5 μg/mL (1/10 of the recommended concentration for apple field treatment) were used. Each fungicide solution was mixed with a spore suspension in a 1:9 (v/v) ratio. Both pyraclostrobin and benomyl were mixed with a 106 spores/mL suspension. Healthy apple fruits, prepared as described in the pathogenicity test, were wounded with a sterile needle. Subsequently, 10 μL of the spore-fungicide mixture was inoculated into the fruits at three equidistant points. The inoculated fruits were incubated at 25°C for 3 days. Seven representative isolates (C. fructicola GgGpCG23-3-1, GwPcCG23-1-1, and GwYgCG23-3-1; C. siamense GgHsCG23-1-1, GwYwCG23-1-1, and GgYcCG23-2-1; and C. gloeosporioides GgYcCG23-3-1) were used in this assay, with untreated fruits serving as the control. Lesion diameters were measured 13 days post-inoculation (dpi). Control efficacy (CE, %) was calculated using the following formula: CE (%) = [(DC − DT) / DC] × 100, where DC and DT represent the average lesion diameters of the untreated control and the fungicide-treated fruits, respectively.
Analysis of resistance-related mutations
Partial sequences of the cyt b and tub2 genes were amplified to identify point mutations associated with fungicide resistance. For the cyt b gene, a fragment flanking the 129th, 137th, and 143rd amino acid positions was amplified using the primer pair RF-CYT1A/RF-CYT2B from cDNA. For the tub2 gene, a fragment flanking the 198th and 200th amino acid positions was amplified using the primer pair TB2L/TB2R from genomic DNA (Supplementary Table 4). PCR amplifications were performed under the following conditions: initial denaturation at 95°C for 2 min; followed by 30 cycles of denaturation at 95°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min; with a final extension at 72°C for 5 min. The resulting PCR products were separated by electrophoresis on a 1% agarose gel in TBE buffer at 100 V for 40 min and visualized using a UV transilluminator. The PCR products were purified using the Expin™ SV PCR mini kit (GeneAll Biotechnology) and sequenced by Macrogen Inc. Nucleotide sequences from 222 isolates collected during the 2015/2016 and 2022/2023 seasons, along with reference sequences, were aligned and analyzed using MEGA 11 software (version 11.0.11). For the 97 isolates collected in 2024, resistance to pyraclostrobin and benomyl was determined using allele-specific PCR (AS-PCR) targeting the G143A and E198A mutations, respectively (Abdullahi et al., 2024).
Evaluation of field fitness of C. fructicola and C. siamense
To evaluate the field fitness of C. fructicola and C. siamense, the two primary pathogens causing apple bitter rot, five representative isolates were selected for each species: GwHcCG22-3-1, GwPcCG23-3-1, CcYsCG23-3-1, JlJsCG22-1-1, and CcCjCG22-2-2 for C. fructicola; and GwHcCG22-1-15, GwYwCG23-3-1, GgPcCG22-1-1, GsYcCG23-3-1, and CcCjCG22-4-2 for C. siamense. Daily mycelial growth rates and pathogenicity were assessed at 20, 25, 30, and 35°C, following the protocols described above. To determine conidial production on PDA, each isolate was initially incubated at 25°C for 10 days. After incubation, the aerial mycelia were removed using a sterile glass slide to stimulate sporulation. The cultures were then exposed to UV light for 3 days at 12-hour intervals at 25°C. Conidia were harvested by flooding the plates with sterile distilled water, and the spore concentration was quantified using a hemocytometer. All experiments were conducted in triplicate for each isolate. For the statistical comparison between the two species, each of the five isolates was treated as a biological replicate (n = 5). Data were subjected to t-test using the mean values of the five isolates per species to determine significant differences between C. fructicola and C. siamense at P < 0.05.
Results
Molecular identification of the Colletotrichum species
To determine the species identity of the Colletotrichum isolates, a multi-locus phylogenetic analysis was performed using the maximum likelihood (ML) method based on concatenated sequences of act, cal, chs-1, gapdh, ITS, and tub2 (Fig. 1, Supplementary Table 5). A total of 222 isolates collected in 2015/2016 (n = 66), 2022 (n = 68), and 2023 (n = 88) were analyzed. Among the 66 isolates from 2015/2016, which were previously identified only at the genus level, 41% were identified as C. fructicola, 27% as C. siamense, 11% each as C. fioriniae and C. nymphaeae, 7% as C. gloeosporioides, and 3% as C. aenigma. For the 2022 collection (n = 68), the predominant species were C. fructicola (51%) and C. siamense (47%), with a minor occurrence of C. aenigma (2%). Similarly, in 2023 (n = 88), C. fructicola remained the dominant species (57%), followed by C. siamense (42%) and C. gloeosporioides (1%). All nucleotide sequences obtained in this study were deposited in GenBank (accession numbers provided in Supplementary Table 5). For the 2024 isolates (n = 97), species-specific primers targeting the tub2 gene were employed for rapid identification. Gel electrophoresis results confirmed that 62% of the isolates were C. fructicola, 31% were C. siamense, 2% were C. gloeosporioides, and the remaining were C. nymphaeae (3%) and C. fioriniae (2%). Overall, these results demonstrate a consistent dominance of C. fructicola and C. siamense across the sampling regions over the years.
Morphological characteristics of the Colletotrichum species
The 6 selected isolates included one isolate each of the six Colletotrichum spp. (Table 1 and Fig. 2). The colonies of C. fructicola JlJsCG22-1-1 were off-white to slightly grey aerial mycelia to yellowish dark grey in reverse (Fig. 2A). The average mycelial growth was 14.2 ± 0.33 mm/day on PDA at 25°C. The conidia were cylindrical, with rounded ends, and averaged 15.25 ± 0.80 μm in length and 5.16 ± 0.20 μm wide. The appresoria were clavate or irregular. The colonies of C. siamense GwHcCG22-1-15 isolates on PDA were fast-growing with whitish to grey dense mycelia on 7-day-old culture (Fig. 2E). The average mycelial growth rate was 14.0 ± 0.26 mm/day at 25°C on PDA. The conidia were cylindrical, straight, non-septate, with obtuse or round ends. The average conidia size was 14.92 ± 0.75 μm long and 4.75 ± 0.55 μm wide. Appresoria were clavate or irregular and entire in outline. For C. aenigma JlJaCG22-2-1, the colonies were dark and fast growing, with growth rate of 13.5 ± 0.58 mm/day (Fig. 2I). The average conidia size were 14.46 ± 0.32 by 4.10 ± 0.40 μm, cylindrical with rounded ends and the appresoria were clavate or irregular in outline. The isolates of C. fioriniae 16MPNJ1 produced salmon-colored to red mycelia on 7-day-old cultures on PDA in both front and reverse sides and produced fusiform conidia after 7 days on PDA (Fig. 2Q). Isolate of C. nymphaeae 16MPBE1 formed white to grey-colored mycelia with orange-colored conidial masses on PDA medium (Fig. 2U). Conidia were mostly fusiform in shape, both ends pointed or acute. Appressoria were single, medium brown, clavate, or irregular in outline.
Pathogenicity of Colletotrichum spp. on apple fruits
Artificial inoculation of cv. Hongro apple fruits inoculated with representative Colletotrichum isolates developed typical bitter rot symptoms, characterized by light-to-dark brown, sunken, circular lesions (Fig. 2). No symptoms were observed on the control fruits inoculated with sterile agar plugs. To satisfy Koch’s postulates, the causal agents were successfully re-isolated from the symptomatic lesions and morphologically re-identified. Quantitative assessment of pathogenicity revealed significant variations in lesion diameters between the two species complexes as well as among species within the C. gloeosporioides species complex (CGSC) (Table 1). Isolates belonging to the C. acutatum species complex (CASC), namely C. fioriniae 16MPNJ1 and C. nymphaeae 16MPBE1, produced significantly smaller lesions (18.4 ± 2.72 mm and 16.9 ± 0.61 mm, respectively; P < 0.01) compared to those caused by members of the CGSC, including C. fructicola JlJsCG22-1-1, C. siamense GwHcCG22-1-15, C. aenigma JlJaCG22-2-1, and C. gloeosporioides GgYcCG23-3-1. These results indicate that members of the CGSC generally exhibit higher virulence on cv. Hongro apple fruits under the tested conditions.
Regional distribution of Colletotrichum species
The distribution of Colletotrichum species causing apple bitter rot was analyzed by collection region and year, as illustrated in the distribution map and pie charts (Fig. 3). In the 2015/2016 collection, six Colletotrichum species were isolated across all sampled regions except Gangwon-do. While Gyeonggi-do yielded only C. siamense, all six species were present in the Chungcheong and Gyeongsang regions, with C. fructicola being the most prevalent at 51% and 33%, respectively. By 2022, a distinct geographical pattern emerged: C. siamense predominated in the northern regions of South Korea, whereas C. fructicola was primarily isolated from the southern regions. Although the proportion of C. fructicola increased in Gyeonggi-do by 2023, C. siamense remained the dominant species in Gangwon-do. Other species identified in 2015/2016, such as C. aenigma and C. gloeosporioides, were rarely detected in 2022 and 2023, with only single isolates found in Jeolla and Gyeonggi, respectively. Among the 2024 isolates, C. siamense continued to lead in Gangwon-do (55%), followed by C. fructicola (40%) and members of the CASC (5%). In contrast, C. fructicola dominated the other four regions, accounting for 67%, 72%, 52%, and 79% of the isolates from Gyeonggi, Chungcheong, Gyeongsang, and Jeolla, respectively. Notably, CASC isolates were also recovered from Gyeongsang and Jeolla, as well as from Gangwon-do. Collectively, C. fructicola was the most frequent species identified in this study (54%), followed by C. siamense (37%) and other minor species (9%).
Fungicide sensitivity of Colletotrichum species to pyraclostrobin and benomyl
The in vitro sensitivity of Colletotrichum isolates to pyraclostrobin and benomyl was evaluated using mycelial growth inhibition assays on PDA amended with serial concentrations of each fungicide (Tables 2 and 3). Among the isolates identified as C. fructicola, a high frequency of resistance to pyraclostrobin (EC50 ≥2.0 μg/mL) was consistently observed across all collection years: 92.6% in 2015/2016, 97.1% in 2022, 74.0% in 2023, and 81.7% in 2024. In contrast, all other Colletotrichum species, including C. siamense, C. fioriniae, and C. nymphaeae, were sensitive to pyraclostrobin (EC50 <2.0 μg/mL), with the sole exception of a single C. aenigma isolate from 2022. Regarding benomyl sensitivity, C. fructicola also exhibited a high prevalence of resistance, with 96.3%, 91.4%, 74.0%, and 80.0% of isolates being resistant in the 2015/2016, 2022, 2023, and 2024 collections, respectively. Conversely, all 117 C. siamense isolates tested were sensitive to benomyl, being completely inhibited at a concentration of 1.0 μg/mL. Interestingly, all isolates of C. fioriniae and C. nymphaeae collected in 2015/2016 and 2024 were found to be less-sensitive to benomyl. These findings highlight that C. fructicola is the primary species harboring dual resistance to both pyraclostrobin and benomyl within the tested populations.
Molecular analysis of mutations in cyt b and tub2 genes
To identify the molecular basis of fungicide resistance, partial sequences of the cyt b and tub2 genes were analyzed. The primer pair RF-CYT1A/RF-CYT2B successfully amplified a 325-bp fragment of the cyt b gene from 222 isolates. Sequence alignment with a reference sequence (GenBank accession no. KM885304.1) revealed that 96 out of 112 C. fructicola isolates collected between 2015 and 2023 harbored the G143A mutation, which is known to confer high resistance to pyraclostrobin. This G143A mutation was also detected in a single resistant C. aenigma isolate (JlJaCG22-2-1), whereas no target-site modifications were found in C. siamense, C. gloeosporioides, C. fioriniae, or C. nymphaeae (Table 4). Regarding MBC resistance, the tub2 gene was amplified as a 474-bp fragment using the TB2L/TB2R primers. Alignment with the reference (P40904.2) showed that 96 isolates, comprising 95 C. fructicola isolates and one C. aenigma isolate, possessed the E198A substitution, which is associated with benomyl resistance. No other mutations, including those at position 200, were detected in any of the isolates (Table 5). For the 2024 isolates, AS-PCR targeting the G143A and E198A mutations correctly identified the resistant and sensitive genotypes, showing 100% consistency with the results of the in vitro sensitivity assays.
Control efficacy of pyraclostrobin and benomyl against Colletotrichum species on detached apple fruits
The in vivo sensitivity of seven representative Colletotrichum isolates to pyraclostrobin and benomyl was further evaluated using a detached apple fruit assay (Figs. 4 and 5). For pyraclostrobin, the control efficacy against resistant isolates was less than 20% at both the recommended field rate (67 μg/mL) and the 5-fold dilution (13.4 μg/mL). In contrast, pyraclostrobin was highly effective against sensitive isolates at the field rate (>80% efficacy), although its efficacy at the 5-fold dilution ranged from 40% to 65% depending on the species. Similar patterns were observed with benomyl treatment; the control efficacy against resistant isolates was negligible (<12%) at both the recommended rate (325 μg/mL) and the 10-fold dilution (32.5 μg/mL). While the recommended field rate of benomyl provided complete protection (100% efficacy) against all sensitive isolates, the 10-fold dilution showed reduced and variable efficacy, ranging from 17% to 63% among the tested Colletotrichum species. These results demonstrate that while recommended field rates remain effective against sensitive populations, they fail to suppress the infection of resistant isolates, and even slight dilutions significantly compromise the disease control potential.
Field Fitness of C. fructicola and C. siamense under different temperatures
The field fitness of C. fructicola and C. siamense, the two most prevalent species identified in this study, was evaluated by assessing mycelial growth and pathogenicity at 20, 25, 30, and 35°C, and sporulation at 25°C (Fig. 6). No significant differences in radial mycelial growth after 7 days of incubation as well as the average daily growth rates were observed between the two species at 20, 25, 30 and 35°C (P > 0.05, Fig. 6A). Pathogenicity assays revealed that the average lesion diameters produced by the two species were comparable at each temperature, showing no statistically significant difference (Fig. 6B). Although significant variations in conidia production were noted among individual isolates within each species (P < 0.05), the average conidial production showed no significant interspecific difference (P > 0.05; Fig. 6C).
Discussion
Apple bitter rot has been regarded as the most prevalent fruit disease in some provinces of Korea and the most challenging summer disease to control (Kim et al., 2020). Over the years, Colletotrichum species have been reported to be associated with this disease in many areas of the country, with species belonging to C. gloeosporioides and C. acutatum species complexes been the major pathogens (Kim et al., 2018; Oo et al., 2018). While C. gloeosporioides and C. acutatum were initially identified as the major pathogens (Lee et al., 2007), recent molecular studies have expanded the species diversity to include C. siamense, C. fructicola, C. fioriniae, C. nymphaeae, C. aenigma, C. grevilleae, and C. scovillei. In 2018, C. siamense and C. fructicola were added to the pathogens from isolates collected in 2013 from orchards in Andong, Gyeongsang region (Park et al., 2018) and in the same year, Oo et al. (2018) reported two additional species, namely C. fioriniae and C. nymphaeae from 2015 and 2016 isolates collected from the same region. Later in 2021, C. aenigma was reported as a pathogen of apple bitter rot by Lee et al. (2021), but according to the results of this study, this pathogen has been associated with bitter rot-infected apples since 2016 in Chungcheong and Gyeongsang regions, but was classified as C. gloeosporioides. In 2022, a new Colletotrichum species, C. grevilleae was isolated from bitter rot-infected apples collected from Gyeonggi-do (Nam et al., 2024). Each of the aforementioned studies were either on first reports or a region-based study, as information on the country-wide distribution of the species has been very limited.
In this study, 66 isolates collected from bitter rot-infected apple fruits and earlier identified as Colletotrichum spp. in 2015/2016 were re-identified. All six Colletotrichum spp. belonging to either of the two species complexes were found among the 2015/2016 isolates. However, about 156 isolates collected from bitter rot-infected apple fields across Korea in 2022 and 2023 were identified mainly as C. fructicola and C. siamense using morphological and molecular techniques. This indicates a complete dominance of C. gloeosporioides species complex as the major pathogens of apple bitter rot currently in Korea. In addition, earlier reports (Kim et al., 2018; Kim et al., 2020) on pathogens of apple bitter rot are mainly on samples collected from the southern region, including Gyeongsang, Chungcheong and Jeolla regions, with no reports from Gangwon or Gyeonggi, which are also apple-producing regions of Korea. Analysis of the distribution of the Colletotrichum species across these five regions in 2022 and 2023 showed that most of the species in Gangwon and Gyeonggi regions are C. siamense, while those of Gyeongsang, Chungcheong or Jeolla regions were mainly C. fructicola. However, in 2024, the frequency of isolation of C. fructicola increased in Gangwon and Gyeonggi regions, becoming the most dominant species in the latter. Though the frequency of isolation of each species was not reported, Oo et al. (2018) reported both C. fructicola and C. siamense from Gyeongsang and Chungcheong, while the only species isolated from Jeolla region was C. fructicola. If we consider this based on the re-identification of 2015/2016 isolates alongside the analysis of 2022-2024 collections, it was revealed a distinct species replacement process within the Colletotrichum community. The stable physiological fitness between C. fructicola and C. siamense strongly suggests that the population shift is not a result of natural competition, but rather a direct consequence of anthropogenic selection pressure.
A notable geographical pattern was observed: C. siamense predominated in the northern regions (Gangwon and Gyeonggi), whereas C. fructicola was the major pathogen in the southern regions (Gyeongsang, Chungcheong, and Jeolla). Interestingly, the isolation frequency of C. fructicola has steadily increased in the northern regions since 2024. Although we compared the fitness parameters, including mycelial growth, sporulation, and pathogenicity, of C. fructicola and C. siamense at various temperatures, no significant differences were found to explain this regional distribution. This suggests that fungicide resistance likely serves as a primary selective advantage for C. fructicola in commercial orchards. However, we recognize that our fitness evaluations were conducted under controlled laboratory conditions, which may not fully reflect the complex biotic and abiotic stresses in the field. Other ecological factors, such as differential overwintering capacity or varying responses to fluctuating microclimates, might also contribute to the population dynamics and should be explored in future studies.
Fungicide-driven selection pressure has served as a primary catalyst for the species replacement observed in Korean apple orchards. The widespread and intensive use of site-specific fungicides has created a selective environment favoring dual-resistant populations of C. fructicola. Since the introduction of site-specific fungicides, resistance in Colletotrichum has been linked to point mutations in target genes, specifically cyt b for QoIs and tub2 for MBCs, resulting in altered amino acid sequences at the fungicide-binding sites (Brent and Hollomon, 2007). In this study, isolates with EC50 values ≥2.0 μg/mL for pyraclostrobin and >100.0 μg/mL for benomyl were classified as resistant. The E198A mutation found in the tub2 gene of the resistant isolates in this study is consistent with the previous findings (Leroux et al., 2002; Ma and Michailides, 2005; Peres et al., 2004). Similarly, G143A mutation was found in the cyt b gene of all the resistant isolates in this study, which also confirmed earlier reports (An, 2023; Cho, 2023; Isa and Kim, 2022; Kim et al., 2019). With the exception of three isolates of C. fructicola, out of the 222 isolates used in this study, all the isolates that were resistant to MBC and possessed the E198A mutation were also resistant to QoI with the G143A mutation. Interestingly, resistant isolates to both fungicides were found mainly in the C. fructicola. All the isolates identified as C. siamense were sensitive to pyraclostrobin and benomyl. C. gloeosporioides and C. aenigma isolates obtained in 2015/2016 were all sensitive to both fungicides. C. fioriniae and C. nymphaeae were sensitive to pyraclostrobin. In their fungicide sensitivity assay with thiophanate-methyl, Kim et al. (2020) reported low and high EC50 values among C. siamense and C. fructicola isolates, respectively, but the report did not classify the isolates into sensitive and resistant. Contrary to our results, Chechi et al. (2019) and Martin et al. (2022) reported that C. siamense isolates obtained from bitter rot infected apples collected from Illinois and California, USA respectively were resistant to azoxystrobin/pyraclostrobin and thiophanate-methyl, while the C fructicola isolate obtained from South Carolina were sensitive to both fungicides. The insensitive response to benomyl observed among the C. fioriniae and C. nymphaeae isolates in this study is consistent with previous studies, that species within the C. acutatum species complex are less sensitive to benzimidazoles (Kim et al., 2007). As previously reported, almost all resistant C. fructicola isolates exhibited dual resistance to both pyraclostrobin and benomyl, whereas, in contrast, all C. siamense isolates retained susceptibility to both fungicides. This distinct difference in fungicide susceptibility provides a valid basis for explaining the phenomenon of C. fructicola becoming increasingly dominant over C. siamense in commercial orchards where these fungicides are frequently used.
The lack of QoI-resistant C. siamense in apple orchards contrasts with reports of high resistance in the same species on other hosts, such as strawberry crown rot (Abdullahi et al., 2024). This discrepancy highlights the potential role of the host plant and specific spray programs in shaping the fungicide sensitivity of Colletotrichum populations. Furthermore, the inherent insensitivity of C. fioriniae and C. nymphaeae (CASC) to benomyl observed in this study aligns with their known biological characteristics within the species complex.
Colletotrichum species are cosmopolitan, found on a variety of hosts, in different geographical regions, and under a range of ecological settings (Leonberger et al., 2019). The dominance of a species on a particular host is geographical location-specific (Babadoost, 2022; Leonberger, et al., 2019; Munir et al., 2016; Wu et al., 2017). In some reports on olive anthracnose caused by Colletotrichum species, high pathogenic specialization on the major cultivar, better adaptability to agroclimatic conditions, difference in the capacity of the species to produce inoculum, the order of species arrival on olive or a high ability to infect and remain on the host have been reported as probable ecological reasons for the dominance of the Colletotrichum species in a particular region (Garcia‐Lopez et al., 2023; Materatski et al., 2018; Moral et al., 2021; Talhinhas et al., 2011). The reasons for the dominance of C. fructicola in the southern region of Korea and the continued increase of its frequency of isolation in the northern region are not yet clear. However, fungicide selection pressure is likely a key driver behind the dominance of C. fructicola in Korean apple orchards, as the majority of these isolates exhibited dual resistance to both strobilurin and benzimidazole fungicides. Though C. siamense is the second most isolated species of apple bitter rot in Korea, no strobilurin or benzimidazole resistant isolates in this species were detected in this study, explaining a possible reason its continued decrease in isolation frequency in relation to C. fructicola. The lack of resistant isolates of C. siamense on apple bitter rot to QoI fungicides particularly in this study, is a clear contrast of the same species on strawberry crown rot which has been reported to be highly resistant to the fungicide (Abdullahi et al., 2024). This further highlights the possible role of plant host in the sensitivity of Colletotrichum spp. to fungicides.
In conclusion, the prevalence of C. fructicola in Korean apple orchards is strongly associated with its robust dual resistance to QoI and MBC fungicides. While basic physiological parameters appear comparable between the dominant species, the intensive selection pressure from site-specific fungicides likely acts as a key determinant in the shifting dominance toward resistant C. fructicola populations. As C. fructicola continues to spread northward, there is an urgent need to re-evaluate current chemical control strategies. Integrating fungicides with different modes of action and monitoring the rapid shift in species composition will be essential for the sustainable management of apple bitter rot in Korea.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Supplement1
Print
