Including isolates from a previous study (Lim et al., 2024), mono-conidial isolates of 329 Botryosphaeria spp. were derived from infected apple trees and fruits collected from 2005 to 2023 in Korea (Supplementary Table 1). Sampling covered various apple trial sites in Gyeongbuk, Gyeongnam, Chungbuk, Chungnam, Jeonbuk, Jeonnam, Gangwon, and Gyeonggi provinces. All isolates were cultured on potato dextrose agar (PDA; Difco, Detroit, MI, USA) at 25°C and stored at −80°C in a 25% glycerol solution.

Botryosphaeria strains were precultured on PDA media and incubated at 25°C for 5 days. To evaluate hyphal growth rate when challenged with pyraclostrobin, mycelial disks (4 mm in diameter) were cut from the edge of precultured fungal colonies and inoculated onto PDA media amended with pyraclostrobin WG (a.i. 20%) at 0.016, 0.08, 0.4, 2, or 10 μg/ml. All pyraclostrobin concentrations were tested in the presence of salicylhydroxamic acid at 100 μg/ml. The experiment was conducted in triplicate for each isolate. After incubation for 6 days at 25°C, the mean diameter minus the diameter of inoculation disk was measured at two perpendicular axes using a digital caliper (Mitutoyo, Kawasaki, Japan). The EC₅₀ (effective concentration for 50% inhibition) values for each isolate were determined using the “log/logit dose response” parameter of the GraphPadPrism software (version 10.2.3, GraphPad Software, La Jolla, CA, USA) (Pugliese et al., 2018). Twenty isolates with the lowest EC₅₀ values were designated as sensitive, and 20 isolates with the highest EC₅₀ values were designated as less-sensitive for, and their sensitivity to three additional QoI fungicides, azoxystrobin WP (a.i. 10%), kresoxim-methyl WG (a.i. 50%), and trifloxystrobin WG (a.i. 50%), was evaluated.

The total genomic DNA was extracted from mycelia cultured on PDA using HiGene Genomic DNA prep kits (BIOFACT, Daejeon, Korea) according to the manufacturer’s protocol. Molecular identification of the strains was conducted by phylogenetic analysis based on the concatenated nucleotide sequences of the internal transcribed spacer (ITS) regions and the translation elongation factor 1-alpha (tef1) and beta-tubulin (tub2) genes using the neighbor-joining method in the MEGA 11.0. program (Tamura et al., 2021). The ITS regions, tef1, and tub2 genes were amplified using the primer sets ITS1F/ITS4 (Gardes and Bruns, 1993; White et al., 1990), EF1-668F/EF1-1251R (Alves et al., 2008), and Bt2a/Bt2b (Glass and Donaldson, 1995). The amplified PCR products were purified with ExoSAP-IT (Thermo Fisher Scientific, Waltham, MA, USA) and sequenced by Macrogen Co. Ltd. (Daejeon, Korea).

Genomic DNA of the 40 less-sensitive and sensitive Botryosphaeria strains were used for the detection of F129L, G137R, and G143A mutations in the cyt b gene using the six different primers (Supplementary Table 2). In addition, the full-length cyt b gene of two less-sensitive and two sensitive strains was amplified using 24 primers for the detection of additional amino-acid substitutions and analysis of the cyt b gene structure (Supplementary Table 2). All primer pairs were designed based on the mitochondrion sequence of Botryosphaeria spp. (accession nos. KY801668, MG593780, and MG593782). The amplification was conducted using an initial denaturation at 94°C for 5 min, followed by 35 cycles of denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and extension at 72°C for 1 min and 30 s and a final extension at 72°C for 10 min. Purification and sequencing were performed as described above. The sequence data were analyzed using MEGA 11.0 software and compared with reference strains (Tamura et al., 2021).

Total RNA from the four less-sensitive and sensitive Botryosphaeria strains was extracted using column-based RNeasy Mini Kits (Qiagen, Mississauga, ON, Canada) following the instructions of the manufacturer. After measuring concentrations using a NanoDrop 2000 (Thermo Fisher Scientific), the RNA was subjected to a reverse transcription-PCR (RT-PCR) using the One-Step RT-PCR kit (Qiagen, Hilden, Germany), as described by the manufacturers, along with four additional primers annealing to the exon sequences of the cyt b gene (Supplementary Table 2). The reverse transcription and PCR conditions included an initial 30-minute step at 55°C, and then initial denaturation at 95°C for 10 min, followed by 35 cycles of denaturation at 95°C for 40 s, annealing at 58°C for 1 min, and extension at 72°C for 1 min and a final extension at 72°C for 10 min. All products were directly purified and sequenced as described above. The nucleotide sequences and amino acid sequences of each strain were compared using MEGA 11.0 (Tamura et al., 2021).

Apple fruits (cv. ‘picnic’) were sterilized with 70% ethanol for 1 min and washed using double-distilled water (ddH₂O). After drying to remove the remaining moisture, three consistent wounds were created using a sterile 4 mm-diameter cork borer. Subsequently, wounded fruits were dipped for 30 s in 500 ml of a 67 μg/ml (recommended concentration) pyraclostrobin solution or ddH₂O (control). After drying, the treated fruits were inoculated with 4 mm-diameter agar plugs taken from the five least sensitive and five most sensitive isolates. Subsequently, the inoculated fruits were incubated at 25°C in darkness for 7 days, and the lesion diameter was measured using a digital caliper. The control efficacy was calculated with the formula

, where Lt represents the mean lesion diameter when treated with pyraclostrobin and L0 represents the mean lesion diameter of the control group. This experimental procedure was performed in triplicate.

The normality of the data for the individual control efficacy of each Botryosphaeria strain was assessed using the D’Agostino-Pearson normality test in the GraphPadPrism software. After calculating control efficacy parameters, the Mann-Whitney test was performed to evaluate the significance of differences between the less-sensitive and sensitive strains. The significance level was set at α = 0.05.

A multilocus phylogenetic analysis was carried out based on the concatenated ITS, tef1, and tub2 sequences with Botryosphaeria parvum (ATCC 58191) as the outgroup taxon (Supplementary Fig. 1). The phylogenetic tree showed that 287 strains (approximately 87%) were identified as B. sinensis, a close relative of B. dothidea (Zhang et al., 2021), while the other 42 (approximately 13%) were clustered with B. kuwatsukai. Therefore, B. sinensis was the dominant species in Korean apple orchards, with some B. kuwatsukai also present in the white rot population, which is consistent with our previous study’s results (Lim et al., 2024).

The sensitivity of the Botryosphaeria strains against pyraclostrobin WG covered a broad mycelial inhibition range at each concentration. The EC₅₀ value of all 329 strains ranged from 0.01 to 34.16 μg/ml (average: 3.03 ± 3.62 μg/ml) (Table 1). Estimated EC₅₀ values were < 2.5 μg/ml in 59.3% of the isolates, 2.5 to 9.5 μg/ml in 37.1% of isolates, and > 10.0 μg/ml in 3.6% of isolates (Fig. 1). The frequency distribution of the EC₅₀ values of these Botryosphaeria isolates showed a unimodal curve with a long right-hand tail. The EC₅₀ values of the less-sensitive group ranged from 5.49 μg/ml (strain 20-GC72) to 34.16 μg/ml (strain GwYwBD22-2-1). For the sensitive group, EC₅₀ values ranged from 0.01 μg/ml (strain GsBhBD23-5-1) to 0.99 μg/ml (strain CcYsBD22-2-1) (Table 1).

Pyraclostrobin sensitivity was compared between various isolate populations using the EC₅₀ values. When comparing sensitivity of those isolated from 2005 to 2020 to those isolated from 2021 to 2023, the mean EC₅₀ value was 2.57 μg/ml in the former group and 3.17 μg/ml in the latter (Table 1), and 76% and 64% of the strains, respectively, were below the total mean EC₅₀ value of 3.03 μg/ml (Fig. 2A). Comparing the sensitivity of the B. sinensis and B. kuwatsukai strains based on the phylogenetic analysis, mean EC₅₀ values were similar, 3.04 μg/ml for B. sinensis and 3.27 μg/ml for B. kuwatsukai (Table 1). For both species, 67% and 65% of the isolates, respectively, presented EC₅₀ values below 3.03 μg/ml (Fig. 2B). When comparing the EC₅₀ values of isolates collected in the eight different regions in Korea, the lowest mean EC₅₀ value was 2.29 μg/ml, in Gyeongbuk province, and the highest was 4.62 μg/ml, in Chungnam province (Table 1, Fig. 2C), and no significant differences in sensitivity among regions were detected (Fig. 2C). When comparing strains by isolation source, trunk and fruit, most strains (82 and 55%, respectively) were distributed below 3.03 μg/ml (Fig. 2D), with mean EC₅₀ values of 2.22 μg/ml and 3.68 μg/ml, respectively, for the two groups (Table 1). All groups showed a similar EC₅₀ value range and a unimodal curve with a positive skew, indicating no significant differences in EC₅₀ distribution.

The sensitivity of the 40 less-sensitive and sensitive strains were evaluated against other QoI fungicides, including azoxystrobin WP (a.i. 10%), kresoxim-methyl WG (a.i. 50%), and trifloxystrobin WG (a.i. 50%), using the same methods as those for pyraclostrobin WG. Compared to the results for pyraclostrobin, all isolates showed relatively lower sensitivity to these fungicides. However, no significant differences in growth inhibition rates were observed between the less-sensitive and sensitive strains for all three fungicides at each concentration (Supplementary Fig. 2). The emergence of resistant strains in a pathogen population leads to a bimodal EC₅₀ value distribution and differences in mean EC₅₀ values when comparing isolate populations (Brent and Hollomon, 2007). For example, the EC₅₀ value distributions of Macrophomina phaseolina against fludioxonil changed from a unimodal distribution into a bimodal distribution between 2018 and 2020 (An et al., 2022). Oculimacula acuformis and O. yallundae showed changes in the distributions of their EC₅₀ values against prochloraz and carbendazim between 1984 and 2000 (Parnell et al., 2008). Similarly, the sensitivity of Zymoseptoria tritici to pyraclostrobin differed significantly between isolation years, with the EC₅₀ values in 2022 being 7 times higher than those in 2019 (Lavrukaitė et al., 2024). In the present study, Botryosphaeria populations did not show a bimodal distribution or significant differences in EC₅₀ values among isolate populations based on any categorization criteria, indicating consistent pyraclostrobin sensitivity across the entire Korean population and no significant resistance development. Similarly, a previous study found that 97 B. dothidea isolates from China indicated low EC₅₀ values (avg. 3.09 μg/ml) with a unimodal distribution, revealing that all isolates were sensitive to pyraclostrobin (Fan et al., 2019).

Sequencing of the partial cyt b gene from the 40 less-sensitive and sensitive isolates revealed a 1,700 bp sequence for B. sinensis using the BDcy687F/BDcy2467R primer pair and a 1,800 bp sequence for B. kuwatsukai using the BDcy687F/BDcy9347R primer pair. For B. kuwatsukai CcYsBD22-3-1, the primer pairs BDcy4986F/BDcy1542R and BDcy7586F/BDcy9347R were used to reveal partial cyt b gene sizes of 2,200 bp and 3,000 bp, respectively (Supplementary Table 2). The sequencing results covered the mutations at codons F129, G137, and G143, which are associated with resistance or reduced sensitivity to QoI fungicides. All 40 isolates had TTC (phenylalanine) in the 129th, GGT (glycine) in the 137th, and GGT (glycine) in the 143rd codon, which are all codons associated with sensitivity to QoI fungicides (Fig. 3). The primary mechanisms of fungicide resistance in phytopathogenic fungi are target gene mutations (He et al., 2021). Especially for QoI fungicides, single amino acid substitutions in the cyt b gene have been found in various plant pathogens, including Alternaria alternata, Pyrenophora teres, and Colletotrichum acutatum (Kim et al., 2019; Ma et al., 2003; Sierotzki et al., 2007). In most cases, resistance was induced by a single point mutation in the 143rd codon of the cyt b gene that led to the substitution of glycine by alanine (Banno et al., 2009). In some species, such as Pyricularia grisea and Colletotrichum spp., other amino acid substitutions (F129L and G137R) have also been associated with QoI resistance (Gisi et al., 2002; Shi et al., 2023). However, the 40 less-sensitive and sensitive isolates in this study showed no mutations at the 129th, 137th, or 143rd codon, indicating no resistant isolates based on these known QoI resistance mutation sites. These findings were consistent with the calculated EC₅₀ values against pyraclostrobin, which indicated no significant resistance against pyraclostrobin among the Botryosphaeria isolates evaluated in this study (Table 1).

To compare the cyt b exon/intron sequences and structures in Botryosphaeria spp., full-length gene sequences were determined through the primer walking strategy. The junctions between intron and exon sequences were identified by comparing their corresponding mRNA sequences to each other and to those of Botryosphaeria isolates from a previous study (Wang et al., 2021). All of the cyt b exon sequences from B. sinensis GsYjBD22-1-1 (EC₅₀: 0.25 μg/ml), B. sinensis GwYwBD22-2-1 (EC₅₀: 34.16 μg/ml), B. kuwatsukai CcYsBD22-3-1 (EC₅₀: 17.58 μg/ml), and B. kuwatsukai 22-MG32 (EC₅₀: 0.09 μg/ml) were 1,158 bp in length (Fig. 4A) and highly conserved (at least 98.5% identity) at the nucleotide level among all four isolates. Only 17 nucleotide substitutions were observed, all at different locations between B. sinensis and B. kuwatsukai, and these variations did not lead to any differences in amino acid sequences among the four isolates. Consequently, the primary mutations associated with QoI fungicide resistance (i.e., F129L, G137R, and G143A) were not observed in any isolates, nor were other mutations, including S34L, G37V, Y132C, D203V, and D230N (Fig. 4B and C), which are also known to potentially lead to reduced sensitivity or resistance against QoIs (Fontaine et al., 2019; Fouché et al., 2022; Oliver et al., 2024; Zhou et al., 2015). However, the structures of cyt b gene differed in other ways among four Botryosphaeria isolates.

The cyt b gene sequences of the two B. sinensis isolates were consistent in their exon/intron organization but differed from those of both B. kuwatsukai isolates. In brief, the four isolates revealed three distinct cyt b structures (Fig. 5). Although exons of the cyt b genes were relatively similar in sequence and structure, vast differences in their lengths, approximately 9.7 kb for both B. sinensis strains and 6.5 and 4.0 kb for the B. kuwatsukai strains, respectively, was mainly the result of presence/absence of introns. B. sinensis GsYjBD22-1-1 and GwYwBD22-2-1 revealed cyt b genes sized at 9,718 bp, containing five introns of 2,521, 2,100, 1,522, 1,356, and 1,061 bp in length. In contrast, B. kuwatsukai CcYsBD22-3-1 had an additional 2,402 bp long intron between the 229th/230th codons, and the introns between the 33rd and 34th, 51st and 52nd, and 143rd and 144th codons in the B. sinensis isolates were absent, and B. kuwatsukai 22-MG32 had an additional missing intron between the 229th and 230th codons that was present in B. kuwatsukai CcYsBD22-3-1. This presence or absence of introns in the four selected isolates were consistent with structures seen in the reference strains, including B. dothidea PGLWX, B. kuwatsukai PGYT19, and kuwatsukai PGZH18 (Wang et al., 2021). Based on these similarities, B. sinensis GwYwBD22-2-1 and GsYjBD22-1-1 were estimated to be constructed with four group I introns and one group II intron, while B. kuwatsukai CcYsBD22-3-1 was estimated to have two group I introns and one group II intron and B. kuwatsukai 22-MG32 to have two group I introns only.

Mitochondrial cyt b genes in fungi can contain group I and II introns, which encode catalytic RNA sequences that self-splice and join exons during transcription (Tourasse and Kolstø, 2008). Group I introns in particular are primarily reported in the mitochondria of plant pathogenic fungi, and the number and locations of group I introns in the cyt b gene are highly conserved within closely related species (Grasso et al., 2006; Malbert et al., 2023). However, the presence or absence of specific group I introns have been reported even within the same species. It has been suggested that the absence or presence of a group I intron immediately after codon 143 is associated with the recognition of splice sites, thereby effecting the occurrence of the G143A mutation and the development of qualitative resistance to QoI fungicides (Rosenzweig et al., 2008; Sierotzki et al., 2007). To date, the presence of this group I intron has been associated with fewer G143A mutations in Puccinia spp., B. cinerea, Alternaria solani, and Pyrenophora spp. (Banno et al., 2009; Grasso et al., 2006; Jiang et al., 2009; Sierotzki et al., 2007). Notably, QoI fungicide resistance due to the G143A mutation has been observed in B. cinerea strains in which this intron was absent (Banno et al., 2009; Yin et al., 2012). Intriguingly, while exons 5 and 6 were separated by the 143/144 group I intron in B. sinensis, this intron was absent in both B. kuwatsukai strains (Fig. 5). This intron was observed in B. dothidea PGLWX, PGLB9, and SLBM-2-2, which have cyt b structures similar to those of B. sinensis GsYjBD22-1-1 and GwYwBD22-2-1 (Wang et al., 2021). Additionally, as this intron was absent in both B. kuwatsukai strains from the present study, as well as those from previous studies, the possibility of QoI resistance emergence due to the G143A mutation is thought to be higher in B. kuwatsukai than in B. sinensis and B. dothidea. Notably, the mean EC₅₀ value of B. kuwatsukai strains increased from 2.34 to 3.40 μg/ml when comparing the 2005-2019 group and 2020-2023 group, while that of B. sinensis decreased from 3.13 to 2.98 μg/ml. In addition, the increase in the maximum EC₅₀ values between the 2005-2019 and 2020-2023 groups for B. sinensis and B. kuwatsukai were approximately 2.9- and 5.5-fold, respectively (Supplementary Fig. 3). Therefore, further studies are necessary to monitor changes in the pyraclostrobin sensitivity of B. kuwatsukai. Additionally, some B. cinerea and Monilinia fructicola isolates have reported G143A mutations despite the presence of the intron immediately after G143, suggesting an association with not only presence/absence patterns but also the mobility of intron (Cinget and Bélanger, 2020; Leroux et al., 2010; Luo et al., 2010). Therefore, the possibility of QoI resistance still exists in B. sinensis and B. dothidea. Clearly, further evaluation of the association between cyt b structure and QoI resistance is essential, and continuous resistance monitoring should be conducted for Botryosphaeria spp.

The effectiveness of pyraclostrobin fungicides in controlling Botryosphaeria infection in apple fruits of five sensitive and five less-sensitive strains and the phenotypic differences between them were examined. When treated with the recommended concentration of pyraclostrobin (67 μg/ml), a fungicide typically used in orchards, effective disease control was observed in both sensitive and less-sensitive strains seven days after inoculation (Fig. 6A). The control efficacy (%) of the less-sensitive strains ranged from 74.1% to 100.0% (Avg. 89.9%), and that of the sensitive strains ranged from 84.1% to 100.0% (Avg. 91.8%) (Fig. 6B). Since the normality tests showed that the control efficacy of all isolate groups did not follow a normal distribution, a statistical analysis was performed using the Mann-Whitney test to investigate the differences between less-sensitive and sensitive isolates, and no significant difference was observed S (P = 0.3346). Fungicide-resistant strains exhibited lower control efficacy than sensitive strains when inoculated on fungicide-treated fruits or leaves. For instance, in a previous study, benzimidazole-resistant B. cinerea strains showed a control efficacy of approximately 10% or less, revealing lower inhibition rates compared to sensitive strains when inoculated on the fruits and leaves of strawberry (Kim et al., 2023). Additionally, prothioconazole-resistant Leptosphaeria maculans and L. biglobosa also revealed larger lesion diameters compared to sensitive strains in in planta assays (Sewell et al., 2017). In the present study, there was no significant difference in control efficacy between the less-sensitive and sensitive strains, again indicating that no significant fungicide resistance currently exists in Korea. In addition, the mean lesion diameters of control fruits were similar, averaging 30.9 mm and 37.8 mm for less-sensitive and sensitive strains, respectively (Fig. 6B), which is consistent with a previous study indicating that resistant B. dothidea isolates and sensitive isolates have similar virulence (Fan et al., 2019).

In this study, 329 Botryosphaeria spp. were found to be sensitive to pyraclostrobin WG, providing a reference for future evaluations of pyraclostrobin sensitivity among Botryosphaeria populations in Korea and allowing for the identification of potential shifts toward resistant strains. While apparently not currently present, the risk of fungicide resistance emergence in Botryosphaeria spp. still exist, and resistance to thiophanate-methyl, benomyl, and tebuconazole has already been reported. In addition, multiple studies have reported QoI resistance associated with non-target-site mechanisms, such as drug efflux transporters, alternative respiration, and intron deletions (Dorigan et al., 2023; Sun et al., 2024). Therefore, examining both target- and non-target site resistance mechanisms will be necessary for the regular long-term monitoring of changes in fungicide resistance, and the association between various cyt b gene types in Botryosphaeria and QoI fungicide resistance should be demonstrated.