C. rosea 67-1 (ACCC 39160) was originally isolated from a vegetable yard in Hainan Province, China, using the sclerotia-baiting method (Zhang et al., 2004). S. sclerotiorum Ss-H (ACCC 39161) was separated from sclerotia-infected soybean stems in a field in Heilongjiang Province, China. B. cinerea TC-B1 was isolated from infected tomato fruits in a greenhouse (Sun et al., 2015b). All strains were maintained at 4°C in the Biocontrol of Soilborne Diseases Lab of the Institute of Plant Protection, Chinese Academy of Agricultural Sciences.

The DNA sequence of CrSsd1 was obtained from the draft genome sequence of C. rosea 67-1. NCBI ¹ and UniProt ² were used for BLASTp analysis. Functional domains of CrSsd1 were predicted using SMART ³ . The Clustal X program was used for amino acid alignments. The phylogenetic tree was constructed by MEGA 7.0 using the maximum likelihood method with 1,000 bootstrap replicates.

Strain 67-1 genomic DNA was extracted using a Biospin Fungus Genomic DNA Extraction Kit (Bioer Technology Co. Ltd., Hangzhou, China) according to the manufacturer’s instructions. Plasmid DNA was isolated using a Plasmid Miniprep Purification Kit (BioDev Co., Beijing, China).

We analyzed the expression levels of CrSsd1 in strain 67-1 during different stages of mycoparasitizing sclerotia. Strain 67-1 was incubated on potato dextrose agar (PDA) at 26°C for 10 days, spores were washed with sterile water and adjusted to 1 × 10⁷ spores/ml, and spore suspensions were smeared evenly on a PDA plate and covered with cellophane. Uniformly sized sclerotia were placed onto the surface of 67-1 plates evenly after culturing for 48 h, and C. rosea 67-1 mycelia were collected at 8, 24, and 48 h and placed immediately in liquid nitrogen. Each treatment included five replicates. Total RNA was extracted using TRIzol reagent (Invitrogen, CA, USA) according to the manufacturer’s instructions. RNase-free DNase I (Invitrogen) was used to eliminate DNA contamination. Reverse transcription was performed using a cDNA FastQuant RT Kit (Tiangen, Beijing, China). Gene expression was analyzed by quantitative reverse transcription PCR (qRT-PCR) using a Bio-Rad IQ 5 Real-Time System (Bio-Rad, CA, USA) and SYBR Premix Ex Taq (Takara, Dalian, China) with primers listed in Table 1. Elongation factor gene EF1 (GenBank accession number: KP274074) was used as a reference gene to normalize gene expression in C. rosea 67-1 under sclerotia induction (Sun et al., 2015a,b), and mycelial samples without added sclerotia acted as a control. The relative expression levels of CrSsd1 were calculated using the 2^−∆∆Ct method, and three replicates were included for each sample.

The plasmid pKH-KO containing two uracil-specific excision reagent (USER) cloning sites (USC1 and USC2) on either side of the hygromycin resistance gene hph was used to construct a CrSsd1 disruption vector (Frandsen et al., 2008). Upstream and downstream flanking sequences of CrSsd1 were amplified using primer pairs CrSsd1-uF/CrSsd1-uR and CrSsd1-dF/CrSsd1-dR, respectively, and cloned into two USC sites using the USER-friendly cloning method to generate CrSsd1-deletion vector pKH-KO-CrSsd1.

For construction of the gene complementation vector, the full-length sequence of CrSsd1, including the promoter, protein-coding, and terminator regions was amplified from 67-1 genomic DNA and cloned into the pKN vector (carrying the G418 resistance gene neo; Kong et al., 2019). The resulting gene deletion and complementation vectors were transformed into protoplasts of 67-1 and ΔCrSsd1, respectively, to generate gene deletion and complementation mutants using protoplast formation and transformation of C. rosea (Sun et al., 2017). Primers (Table 1) were designed and mutants were verified by PCR and DNA sequencing. Furthermore, the expression levels of CrSsd1 in wild-type (WT), deletion and complementation strains were tested using reverse transcription PCR (RT-PCR) with primers CrSsd1-F and CrSsd1-R and reference gene EF-1 gene (Table 1; Sun et al., 2015a,b).

To analyze differences in vegetative growth among C. rosea 67-1, ΔCrSsd1, and ΔCrSsd1-C strains, agar blocks (3 mm) of strains were inoculated onto the center of a PDA plate and cultured at 26°C. The size and morphology of colonies were measured daily. After 15 days, fungal spores were collected by adding 5 ml sterile distilled water, and spores were counted under a BX41 microscope (Olympus, Tokyo, Japan). The conidial germination rates of all strains were determined. Spore suspensions of WT, ΔCrSsd1 and ΔCrSsd1-C strains with a concentration of 1 × 10⁷ spores/ml were prepared and inoculated in potato dextrose (PD) broth on a rotary shaker at a speed of 180 rpm. Samples were incubated at 26°C, and the germinated conidia were counted at 8 and 16 h post inoculation. To evaluate the stress response, cultures were grown on PDA plates amended with different stress agents [1 M NaCl, 1 M KCl, 1 M glycerin, 1 M sorbitol, 20 mM H₂O₂, 0.03% SDS, and 0.3 mg/ml Congo red (CR)] for 10 days. Diameters of colonies were counted, and microscopic observation of the hypha under different stress conditions was performed with a fluorescence microscope system (DM6 B, Leica, Germany). All assays were repeated three times.

Antagonistic activity of C. rosea 67-1, ΔCrSsd1, and ΔCrSsd1-C strains against B. cinerea was tested on 9 cm PDA plates. A 3-mm agar plug of strains was inoculated 2 cm from the edge of the plate and cultured at 26°C for 5 days. A plug of B. cinerea was then placed equidistant from the other side of the plate and cultured at 26°C for 20 days. The distance of hyphal extension for each strain was measured (Dubey et al., 2014; Tzelepis et al., 2015; Filizola et al., 2019).

Sclerotia of uniform size were surface-sterilized with 1% NaClO for 3 min, rinsed three times with sterile water, and then immersed in spore suspensions of wild-type (WT) 67-1, ΔCrSsd1, and ΔCrSsd1-C strains at a concentration of 1 × 10⁷ spores/ml for 10 min. Sclerotia were picked and placed onto a piece of wet sterile filter paper in a Petri dish (diameter 9 cm) and incubated at 26°C. Treatment with sterile water was used as a control. The number of sclerotia infected by the transformants was counted under a stereo microscope (SMZ-10, Nikon, Tokyo, Japan) at 8, 16 and 24 h. Sclerotia covered with C. roses mycelia were regarded as parasitized, and parasitic rates of all strains were calculated (Sun et al., 2017). After 7 days, we investigated parasitic severity of sclerotia using a BX41 inverted microscope (Olympus) based on a four-grade scale (0 = no C. rosea hyphae were detected on the surface of sclerotia; 1 = loose C. rosea hyphae extended to the sclerotia; 2 = sclerotia were covered with C. rosea hyphae but not softened; and 3 = sclerotia were covered with C. rosea hyphae and exhibited soft rot; Sun et al., 2019a). A total of 30 sclerotia were tested for each treatment, and three replicates were performed.

Pot experiments were carried out to test the ability of C. rosea 67-1, ΔCrSsd1, and ΔCrSsd1-C strains to control S. sclerotiorum on soybean in the greenhouse. Soybean seeds (Zhonghuang 13; Institute of Crop Sciences, CAAS, China) were sown in sterile soil in plastic pots (diameter 11 cm). When nine compound leaves had grown, seedlings were sprayed with 100 ml spore suspension (1 × 10⁷ spores/ml) from each strain. After drying for 2 h, an equivalent amount of S. sclerotiorum mycelial suspension was inoculated onto leaves. Plants treated with sterile water followed by the pathogen served as controls, and 12 pots were tested for each isolate. The greenhouse was maintained at 26–28°C and 60% relative humidity, and all pots were arranged randomly. After 7 days, disease severity of Sclerotinia rot was scored using grades 0–4 according to the percentage of lesions on soybean leaves (0 = no symptoms on soybean leaves; 1 = less than 10% lesions on soybean leaves, 2 = 10–30% lesions on soybean leaves; 3 = 30–50% lesions on soybean leaves; and 4 = over 50% lesions on soybean leaves). All unfolded compound leaves were checked and three replicates were performed for each treatment.

Statistical software SPSS 2.0 (Chicago, IL, USA) was used for ANOVA. Statistical tests were carried out using Tukey’s test for multiple comparisons and a p < 0.05 was considered statistically significant.

Gene cloning and bioinformatics analysis showed that CrSsd1 (GenBank accession number: MN816008) is 3,894 bp in length with no introns and encodes a 1,298-amino-acid polypeptide that contains the RNB domain (Figure 1A), which is the catalytic domain of ribonuclease II. CrSsd1 shares 45.8% identity with S. cerevisiae Ssd1, which is involved in a range of cellular processes, including cell wall integrity, signal transduction, and RNA deterioration. Phylogenetic analysis and sequence alignment of CrSsd1 with other fungal species revealed close homology with homologs of Ssd1 in Fusarium oxysporum and Trichoderma arundinaceum, and it is highly conserved among various fungi (Figure 1B).

The expression levels of CrSsd1 in 67-1 were also investigated during different stages of mycoparasitizing sclerotia by qRT-PCR. Analysis of gene expression indicated that CrSsd1 was upregulated in C. rosea throughout mycoparasitism, particularly at 24 h, and expression levels were more than four-fold higher than the control (Figure 2), which is consistent with the transcriptome data from C. rosea parasitizing S. sclorotiorum (Sun et al., 2015b).

To identify the role of CrSsd1 in C. rosea, single gene deletion mutants were generated using a homologous recombination strategy (Figure 3A). Among 187 hygromycin-resistant transformants, three ΔCrSsd1 strains with identical phenotypic characteristics were confirmed by PCR analysis with primers CrSsd1-in-F/R (inside of the target gene), CrSsd1-yz-F/R (outside of the homologous fragment), HPH-F/R (on both ends of the hph gene), and CrSsd1-yz-F/HPH-R (Figure 3C). Moreover, fragments amplified by primer pair CrSsd1-yz-F/R were sequenced, and the results showed that the CrSsd1 gene was successfully replaced with a hygromycin B resistance cassette as expected. For complementation of CrSsd1, the vector pKN-CrSsd1-C was transformed into the ΔCrSsd1 strain and 11 complementation strains were finally obtained. RT-PCR verification demonstrated a complete loss of CrSsd1 transcript in ΔCrSsd1 mutants, whereas specific products were detected in the WT and complementation strains. In addition, the expression of EF-1 gene was detected in all strains (Figure 3D).

Three ΔCrSsd1 and ΔCrSsd1-C mutants were selected to analyze the functions of CrSsd1 gene. The colony morphology showed that ΔCrSsd1 mutants had flatter and thinner mycelia than those of the WT 67-1 and the complemented transformant ΔCrSsd1-C (Figure 3B). Moreover, the mycelial growth rates of mutants were slightly slower than that of the WT and ΔCrSsd1-C strains. When grown on PDA for 9 days, the colony diameter of 67-1 reached 5.95 cm, while that of ΔCrSsd1 was 5.29 cm, and the difference was significant (p < 0.05; Figures 4A,B). Surprisingly, gene-deficient strains lost almost all ability to undergo conidiation. After incubation on PDA for 15 days, only 1 × 10⁶ spores/plate were harvested for ΔCrSsd1, compared with 4.9 × 10⁷ spores/plate for the WT strain (p < 0.01; Figures 4A,C). Conidial germination rate of the ΔCrSsd1 mutants was 46.9%, significantly lower than that of the WT strain (68.7%) at 8 h (p < 0.05); however, both strains increased to approximately 100% at 16 h (Supplementary Figure S1). The complemented transformants showed similar results with WT.

The sensitivity of mutants to a variety of environmental stresses, including osmotic stress, oxidative stress, cell membrane stress, and cell wall stress, was investigated. The results showed no significant differences among strains under treatment with NaCl (1 M), KCl (1 M), glycerin (1 M), H₂O₂ (20 mM), or SDS (0.03%). However, interestingly, ΔCrSsd1 grew much slower in media containing sorbitol (1 M) or CR (0.3 mg/ml) compared with WT and complemented strains, indicating that ΔCrSsd1 deletion mutants were more sensitive to osmotic and cell wall stresses (Figures 5A,B). To further investigate the stress sensitivity of ΔCrSsd1 and 67-1, the hyphal phenotypes under different stress conditions were observed. Our findings demonstrated that the loss of CrSsd1 impaired hyphae branching under NaCl, KCl, sorbitol, and CR, indicating that the CrSsd1 gene played an important role in C. rosea response to osmotic and cell wall stresses (Supplementary Figure S2).

In vitro antagonistic activity tests showed that C. rosea 67-1, ΔCrSsd1, and ΔCrSsd1-C strains could all overgrow a colony of B. cinerea after culturing for 20 days. However, the hyphal extension ability was decreased by 41.3% for ΔCrSsd1 mutants compared with the WT strain (p < 0.05), and the complemented strain ΔCrSsd1-C recovered this ability almost to the WT level (a decrease of only 3.6%; Figure 6).

No hyphae of ΔCrSsd1 mutants were detected on the surface of the sclerotia at 8 h after inoculation, and the parasitic rate was 20.5% at 16 h, which was remarkably lower than the WT (57.8%, p < 0.05). By 24 h, hypha of 67-1 and ΔCrSsd1-C covered the whole sclerotia surface, while only 48.3% were parasitized by the ΔCrSsd1 mutants (Table 2). After 7 days of cultivation in a moist environment, the mycoparasitism level of ΔCrSsd1 on sclerotia was markedly reduced compared with that of WT 67-1 and complemented strain ΔCrSsd1-C. From the external phenotype and the inner structure of the sclerotia, we could see that infected sclerotia were completely softened and rotten, resulting in high parasitic severity (grade 4), whereas those treated with the ΔCrSsd1 deletion mutant were covered only sparsely in hyphae and remained relatively firm, equating to mycoparasitism grade 1, indicating that deletion of the CrSsd1 gene substantially weakened the mycoparasitism of C. rosea. Additionally, mycoparasitic ability was recovered in the complemented strain (Figures 7A,B).

After inoculation with S. sclerotiorum for 7 days, severe leaf lesions were observed in control soybean seedlings. However, soybean seedlings treated with the biocontrol fungus 67-1 were much healthier and displayed less damage, consistent with excellent control efficacy against soybean Sclerotinia rot. Interestingly, when the CrSsd1 gene was deleted, the control efficacy of the mutant was markedly reduced, while the efficiency was regained in the complemented strain (Figure 8 and Table 3), demonstrating that CrSsd1 could dramatically affect the biocontrol efficacy of C. rosea.