C. scovillei strain KC05 was used as the wild-type strain in this study (Han et al., 2016). The deletion mutant and complementation strain were selected on TB3 agar (200 g of sucrose, 3 g of yeast extract, 3 g of casamino acids, 10 g of glucose, and 8 g of agar per liter) supplemented with 200 μg/ml of hygromycin B (EMD Millipore, Billerica, MA, USA) and 400 μg/ml of G418 geneticin (Gibco, Carlsbad, CA, USA), respectively. CM agar (10 g of sucrose, 6 g of yeast extract, 6 g of casamino acids, and 15 g of agar per liter) supplemented with 0.4 M NaCl or KCl was used to test for osmotic stress tolerance. V8 agar (V8A, 80 ml of V8 juice, 310 μl 10 N NaOH, and 15 g of agar/l) was used to measure conidiation.

All DNA and protein sequences were obtained from the Comparative Fungal Genomics Platform (http://cfgp.snu.ac.kr) (Choi et al., 2013; Park et al., 2007), and the BLAST program in the National Center for Biotechnology Information (NCBI, http://www.ncbi.nlm.nih.gov). Protein sequence alignment and phylogenetic analysis were performed using the ClustalW program in MEGA 6.0 (Tamura et al., 2013; Thompson et al., 1994). Domain structure analysis was performed using InterPro Scan v83.0 (http://www.ebi.ac.uk/interpro/) (Mulder et al., 2005). Oligonucleotide primers used in this study were synthesized by Bioneer (Daejeon, Korea).

Total RNA was isolated from frozen fungal tissues using the Easy-Spin Total RNA Extraction Kit (Intron Biotechnology, Seoul, Korea) according to the manufacturer’s instructions. First-strand cDNA synthesis was performed from 5 μg total RNA using the oligo(dT) primer with the SuperScript III First-Strand Synthesis System Kit (Invitrogen Life Technologies, CA, USA). Detection of CsSTE50 and β-tubulin (CAP_007327) expression was performed using the primers CsSTE50_RTF/RTR as described by Shin et al. (2021). The primer sets used for reverse transcription polymerase chain reaction (RT-PCR) are listed in Supplementary Table 1. Experiments were conducted in triplicate and repeated three times.

Fungal genomic DNA was isolated according to a standard method or the quick method (Chi et al., 2009; Sambrook et al., 1989). Approximately 1.5 kb fragments of upstream and downstream of CsSTE50 were amplified from wild-type KC05 genomic DNA using primers CsSTE50_5F/5R and CsSTE50_3F/3R, respectively. The 1.5 kb hygromycin resistance gene (hyg) cassette was amplified from pBCATPH using primers HPH_F/HPH_R, and fused to the amplified upstream and downstream fragments via the double-joint polymerase chain reaction (PCR) method (Yu et al., 2004). The resulting products were finally amplified using primers CsSTE50_NF/NR and transformed into protoplasts of the wild-type strain by the polyethylene glycol-mediated transformation method (Shin et al., 2019b; Sweigard et al., 1992). Putative knockout mutants were selected by screening PCR using primers CsSTE50_SF/SR and confirmed by southern blot hybridization and RT-PCR (Sambrook et al., 1989). For southern blot hybridization, genomic DNA was digested with MseI restriction enzyme and blot was probed with 0.5 kb downstream cassette. Biotin-High Prime (Roche, Indianapolis, IN, USA) was used to label the probe. ChemiDoc XRS + system with Quantity One software (Bio-Rad Laboratories, Hercules, CA, USA) was used to detect chemiluminescent signal. To complement the mutant, the CsSTE50 gene including 1.7 kb upstream and 500 kb downstream was amplified from wild-type genomic DNA using the primers CsSTE50_CF/CR, and the amplified fragments were co-transformed into protoplasts of the mutant with pII99 that contains a geneticin resistance gene (gen) cassette.

To evaluate vegetative growth, fungal colonies were grown on CM agar and CM agar supplemented with an osmotic stress agent (0.4 M NaCl or KCl) for 5 days at 25°C in the dark. Conidiation was measured by counting the number of conidia harvested with 5 ml of sterile distilled water from 5-day-old V8 agar under continuous light, using a hemocytometer. Conidial morphology was observed under a light microscope, and conidial length was measured using the ZEN imaging software. To measure conidial germination and appressorium formation, conidial drops (5 × 10⁴ conidia/ml) were placed on hydrophobic coverslips and incubated in a moistened box. Exogenous cAMP (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in sterile distilled water to yield a 10 mM solution, and mixed with an equal volume of conidial drops. To induce appressorium formation from hyphal tips, mycelial agar plugs obtained from 5-day-old oatmeal agar were placed on slide glasses, covered with coverslips, and incubated in a moistened box. For conidial penetration and infection assays, conidial drops (5 × 10⁴ conidia/ml or 15 × 10⁴ conidia/ml, respectively) were inoculated onto the surface of pepper fruits and incubated in a moistened box. For mycelial infection assays, mycelial agar plugs were placed onto the surface of pepper fruits and incubated in a moistened box. All experiments were repeated three times with three replicates.

We identified CsSTE50 (CAP_001511), a homolog of M. oryzae MST50, via BLAST search of the Comparative Fungal Genomics Platform (Choi et al., 2013; Park et al., 2006). The CsSTE50 gene was predicted to encode a 486-amino-acid protein with SAM and RA domains (Fig. 1A and B). Phylogenetic analyses of Ste50 homologs in fungal species revealed that Ste50 homologs are conserved in the subphylum Pezizomycotina, distinct from yeast (Fig. 1A and B). The amino acids of the CsSTE50 and Ste50 homologs share less than 30% amino acid identity in yeast, compared with more than 72% in Pezizomycotina fungi. Ste50 homologs of M. oryzae, Neurospora crassa, Botrytis cinerea, C. gloeosporioides, C. higginsianum, Fusarium oxysporum, F. graminearum, Yarrowia lipolytica, S. cerevisiae, and Candida albicans showed 76.6%, 72.2%, 72.2%, 88.9%, 97.7%, 76.0%, 75.4%, 27.2%, 17.5%, and 20.9% amino acid identity, respectively, with CsSTE50.

To investigate the functional roles of CsSTE50, a gene replacement construct was generated using double-joint PCR (Fig. 2A). Approximately 1.5 kb upstream and downstream flanking sequences of the CsSTE50 gene were fused to the 1.5 kb hyg cassette via overlapping sequences (Supplementary Table 1, underlined), which generated a 4.5 kb construct. The construct was transformed into C. scovillei KC05 protoplasts. Southern blot hybridization was used to select a ΔCsste50 mutant exhibiting a 1.6-kb band instead of the 2.6-kb band of the wild-type (Fig. 2B), indicating that the CsSTE50 gene was replaced with the hyg cassette without ectopic insertion (Fig. 2B). Finally, RT-PCR confirmed that expression of the CsSTE50 gene was completely abolished in the ΔCsste50 mutant (Fig. 2C). All strains contained a band for the β-tubulin gene, as a positive control (Fig. 2C).

To investigate the role of CsSTE50 in the vegetative growth of C. scovillei, mycelial agar plugs of the ΔCsste50 mutant were inoculated on CM agar and incubated for 5 days. The growth of the ΔCsste50 mutant was slightly greater than that of the wild-type (Fig. 3A and B). We also evaluated the osmotic stress tolerance of the ΔCsste50 mutant by assessing the growth of the mutant on CM agar supplemented with an osmotic stress agent. The growth relative to that of the wild-type on CM agar without supplements was assessed. Compared with the wild-type, the ΔCsste50 mutant was highly sensitive to CM agar supplemented with 0.4 M NaCl or 0.4 M KCl (45% or 44% reduction, respectively) (Fig. 3A and B). These results suggest that CsSTE50 is involved in the vegetative growth and stress tolerance of C. scovillei.

Conidiation is an important developmental process involved in disease dissemination in plant pathogenic fungi (Dean et al., 2012). To assess the role of CsSTE50 in C. scovillei conidiation, the ΔCsste50 mutant was inoculated on V8 agar and incubated for 5 days under continuous light. Remarkably, the ΔCsste50 mutant produced approximately 1.4 times more conidia than did the wild-type (Fig. 4A). The ΔCsste50 mutant also produced much longer conidia (average length 13.5 μm) compared with those of the wild-type (average 10.6 μm) (Fig. 4B). These results indicate that CsSTE50 is involved in conidiation and conidial morphology in C. scovillei.

We evaluated the conidial germination and appressorium development of the ΔCsste50 mutant on an artificial hydrophobic surface. We placed 20 μl conidial drops (5 × 10⁴ conidia/ml) on coverslips. By 16 h post-inoculation, most of the wild-type conidia (>90%) formed a single germ tube and appressorium (Fig. 5A-C). However, the ΔCsste50 mutant showed significantly delayed germination and failed to form appressoria, indicating that the ΔCsste50 mutant is defective in intracellular signaling or hydrophobic surface recognition for germination and appressorium formation (Fig. 5A-C). We evaluated whether exogenous cAMP, a signaling molecule, could restore appressorium formation in the ΔCsste50 mutant. At 16 h post-inoculation, we found that treatment of conidia with exogenous cAMP did not restore appressorium formation in the mutant, suggesting that CsSTE50 functions in a cAMP-independent manner (Fig. 5D). To observe appressorium formation from hyphal tips, we inoculated mycelial agar plugs of the wild-type and mutant on coverslips. At 72 h post-inoculation, the wild-type formed melanized appressoria on coverslips, but the ΔCsste50 mutant failed to form appressoria (Fig. 5E). Collectively, these results suggest that CsSTE50 is indispensable for appressorium formation in C. scovillei.

To investigate whether the ΔCsste50 mutant can form appressoria on pepper fruit, we inoculated the fruit with conidial drops (5 × 10⁴ conidia/ml). At 3 days post-inoculation, the wild-type conidia formed appressoria and penetrated the plant cuticle (Fig. 6A and B), whereas the ΔCsste50 mutant failed to form appressoria and formed long germ tubes on the peppers (Fig. 6A and B). This indicates that CsSTE50 is important for appressorium formation on pepper fruits. To determine fungal pathogenicity, the surfaces of pepper fruit were inoculated with conidial drops (15 × 10⁴ conidia/ml) or mycelial agar plugs. At 8 days post-inoculation, the wild-type developed anthracnose lesions on both intact (left panel) and wounded (right panel) fruits (Fig. 6C). However, the ΔCsste50 mutant failed to develop disease lesions on intact peppers, and only small spots were observed on wounded fruit (Fig. 6C). Collectively, these results indicate that CsSTE50 is important for appressorium formation and invasive growth in C. scovillei.