Serine peptidases and increased amounts of soluble proteins contribute to heat priming of the plant pathogenic fungus Botrytis cinerea

ABSTRACT Botrytis cinerea causes gray mold disease in leading crop plants. The disease develops only at cool temperatures, but the fungus remains viable in warm climates and can survive periods of extreme heat. We discovered a strong heat priming effect in which the exposure of B. cinerea to moderately high temperatures greatly improves its ability to cope with subsequent, potentially lethal temperature conditions. We showed that priming promotes protein solubility during heat stress and discovered a group of priming-induced serine-type peptidases. Several lines of evidence, including transcriptomics, proteomics, pharmacology, and mutagenesis data, link these peptidases to the B. cinerea priming response, highlighting their important roles in regulating priming-mediated heat adaptation. By imposing a series of sub-lethal temperature pulses that subverted the priming effect, we managed to eliminate the fungus and prevent disease development, demonstrating the potential for developing temperature-based plant protection methods by targeting the fungal heat priming response. IMPORTANCE Priming is a general and important stress adaptation mechanism. Our work highlights the importance of priming in fungal heat adaptation, reveals novel regulators and aspects of heat adaptation mechanisms, and demonstrates the potential of affecting microorganisms, including pathogens through manipulations of the heat adaptation response.


Introduction 36
Botrytis cinerea is a notorious plant pathogen that causes grey mold disease, leading to massive crop losses 37 worldwide. Its optimum growth temperature is 18-22°C, and grey mold disease is widespread in relatively cool 38 environments. At temperatures just a few degrees higher than the optimum, the disease completely 39 disappears; however, the fungus remains viable in warm climate and resumes growth and infection when 40 temperatures drop to the optimal range. How B. cinerea copes with high temperatures and remains viable in 41 warm environments, including occasional extreme heat waves, is unclear. 42 Temperature shifts alter the expression of thousands of genes. In budding yeast, the most significantly 43 enriched gene ontology (GO) categories that are upregulated under heat stress include heat shock proteins 44 (HSPs), oligosaccharide metabolism, protein folding, and protein catabolic and recycling processes (1, 2). 45 Proteins in these categories, as well as the disaccharide trehalose, help maintain protein homeostasis 46 (proteostasis) (3,4). Downregulated genes are typically enriched in translation and ribosome biogenesis, which 47 correlates with growth arrest at high temperatures (1, 5). Apart from chaperones, the roles of most heat-48 induced genes and their protein products remain enigmatic. Most of these genes might help replenish 49 As several serine type peptidase-encoding genes were specifically upregulated under priming conditions, we 150 analyzed the transcript profiles of all 21 annotated B. cinerea serine-type peptidase genes by heatmap analysis. 151 All 21 genes were upregulated under MHT or SHT-P, whereas most were expressed at low levels under OT and 152 SHT (Fig. 3b). 153 Proteomic analysis. We grew fungi as described in Table S6, extracted soluble and aggregated proteins (Fig. 154 S4a), separated the proteins by SDS-PAGE and stained them with Coomassie brilliant blue. Significantly more 155 protein aggregates formed under SHT and SHT-P compared to OT and MHT, but minor differences were also 156 observed between MHT and OT (Fig. S4b). We then analyzed the soluble and aggregated proteins using liquid 157 chromatography with tandem mass spectrometry (LC-MS/MS). We detected 4,492 proteins (Table S7), with 158 good separation between soluble and aggregated proteins as well as the four heat treatments (Fig. S4c). The 159 aggregated proteins under SHT-P and SHT clustered together, whereas the soluble proteins at MHT and SHT-P 160 formed a subcluster. These results suggest that priming mainly affects the amount and nature of proteins that 161 remain soluble under heat stress. Therefore, we subsequently analyzed only the soluble proteins. 162 The levels of 1,374 soluble proteins (Table S8) were significantly altered (q-value < 0.05, lFCl ≥ 2) under the 163 three heat treatments compared to the control (Fig. S4d). The levels of 147 proteins were higher under SHT vs. 164 OT, a number significantly lower than those under MHT (429 proteins) and SHT-P (584 proteins). We set two 165 criteria for identifying priming-associated candidates among soluble proteins: (i) {log 2 (SHT-P/OT)} -{log2 166 (SHT/OT)} ≥ 1 for proteins more abundant at SHT-P than at SHT; and (ii) log 2 (SHT-P/OT) > 1 and log 2 (SHT/OT) < 167 1 for proteins more abundant at SHT-P but not SHT compared to OT. This analysis yielded 355 proteins (Fig. 3c 168 and Table S9), which we considered to be candidate regulators (MHT-specific) or executors (MHT and/or SHT-P) 169 of the priming response. 170 GO enrichment analysis of the 355 priming candidates (Table S10) highlighted several major categories,  171 including transferases (GO:0000030 and GO:0016758), transporters (GO:0005347 and GO:1901505), and 172 serine-type peptidases (GO:0008236). Serine-type peptidases (STPs) were of particular interest since this group 173 was also revealed as priming-related by RNA-seq (Fig. 3b). Except for bcin_16g02790, all 9 STPs identified 174 among the 1,374 soluble proteins were more abundant at MHT and SHT-P, but not at SHT vs. OT (Fig. 3d). The 175 six serine-type peptidases identified by RNA-seq were included in this set. 176 Collectively, the transcriptomics and proteomics analyses revealed subsets of priming-related GO categories 177 that only partly overlapped. We also noticed a significant difference between changes in genes and protein 178 levels during priming: compared with control, the number of DGEs was lowest in MHT (Fig. S3d and S3e)., 179 whereas the changes in soluble proteins were lower at SHT (Fig. S4d). 180 181 Changes in soluble protein abundance are only partly dependent on gene expression levels 182 To extract more information from our data, we compared the gene expression and protein abundance data. 183 For each treatment, we selected all shared genes and soluble proteins with pFDR < 0.05 and evaluated the 184 correlation between relative gene expression level and protein abundance. The correlations were similar under 185 MHT (R = 0.631) (Fig. 4a) and SHT (R = 0.594) (Fig. 4b) and somewhat smaller under SHT-P (R = 0.477) (Fig. 4c). 186 We also noticed a relatively high number of opposite correlations at SHT-P, namely increased protein and 187 reduced transcript abundance. To explore the nature of these proteins, for each treatment, we selected all 188 genes with log 2 (treatment/OT) < 0 and all proteins with log 2 (treatment/OT) ≥ 1 and calculated their 189 proportions. Under SHT-P, 11.56% of all gene-protein pairs had negative correlations (Fig. 4d), a much larger 190 proportion than under MHT (4.62%) and SHT (1.7%). These results suggest that the levels of many soluble 191 proteins are regulated in a different manner under SHT-P vs. SHT. 192 We performed a similar analysis of the 355 priming candidate proteins, using only the data from MHT and  P, since only 20 gene-protein pairs were found at SHT. At MHT, the correlation between transcript levels and 194 protein abundance was positive, but it dropped from 0.631 for the entire set to 0.33 (Fig. 4e). Unexpectedly, at 195 SHT-P, the correlation between shared genes and proteins dropped from 0.477 for the entire set to nearly zero 196 (Fig. 4f, R = 0.04) for the 355 priming-specific candidate proteins. In 54.6% (Fig. 4d To determine whether STPs are required for heat priming, we tested the effects on priming of PMSF, which 206 specifically inhibits STPs (32, 33). We produced GTs at 29°C, replaced the medium with fresh medium 207 containing 0.5 or 2 mM PMSF, transferred the samples to 37°C for 4 h and stained them with PI. Treatment 208 with 2 mM PMSF resulted in 85% cell death vs. no cell death in untreated control GTs and <5% cell death in 209 samples treated with 0.5mM PMSF or a 2× concentration of a general protease inhibitor cocktail (Fig. 5a and 210 S5a). Treatment with 2 mM PMSF had almost no effect on GTs that were kept at 29°C, indicating that the drug 211 suppressed the priming response but did not kill the fungus. 212 To test the possible roles of the six identified STPs in priming, we generated deletion strains for each of the six 213 genes and tested their priming responses. All deletion strains had normal colony morphology, hyphal growth 214 and germination rates at 22°C and 29°C (not shown). To evaluate priming responses, we produced GTs at 29°C, 215 transferred them to 37°C and stained them with PI. Compared to the wild type, strain ∆bcin_08g02390 showed 216 markedly increased cell death, whereas the five other deletion strains showed normal levels of cell death (Fig. 217 5b and S5b). Similarly, the membrane potential of strain ∆bcin_08g02390 was compromised following transfer 218 from 29°C to 37°C (Fig. 5c and S5c). 219 Four of the STPs belong to peptidase group S53 (BCIN_06g00620, BCIN_06g00330, BCIN_15g04670 and 220 BCIN_15g03150). To determine whether the lack of a priming phenotype in a single deletion strain of these 221 genes resulted from functional redundancy, we generated the strain Δ4stp, with deletions of all four S53 genes. 222 Similar to the single deletions trains, the Δ4stp strain did not show developmental defects and had a normal 223 priming response (not shown). Hence, the role of these four S53 genes in priming remains unverified. 224 Because the deletion of bcin_08g02390 compromised the priming response of the fungus, we named this gene 225 bcprm1 (B. cinerea priming 1). To further examine the physiological relevance of the priming defects, we 226 examined the effect of priming on biomass production of the ∆bcprm1 strain. We produced uniform wild-type 227 and ∆bcprm1 cultures in malt medium at 22°C, transferred them to 29°C for 12 h to induce priming, and 228 measured the biomass. We then incubated cultures for 12 h at 37°C followed by 22°C for 24 h, and measured 229 their biomass once again (Fig. S5d). There were no differences in biomass between the wild-type and ∆bcprm1 230 strains after incubation at 29°C. However, the recovery of ∆bcprm1 after incubation at 37°C was compromised, 231 as it produced ~50% less biomass than the wild type (Fig. 5d and S5e). To examine the effect of the deletion of 232 bcprm1 on pathogenicity, we inoculated the leaves of P. vulgaris plants with a spore suspension and incubated 233 them at 29°C for 8 h, then at 37°C for 4 h and then at 22°C for 3 d. Understanding the mechanisms of heat adaptation in fungi is essential for risk evaluation in preparation for 240 temperature-driven disease outbreaks (34, 35) and for the rational design of temperature-driven disease 241 control methods. To help achieve these goals, we studied the mechanisms of heat adaptation in B. cinerea, a 242 cosmopolitan, devastating plant pathogen (36, 37). 243 We found that heat stress adaptation via priming is a powerful mechanism that enables B. cinerea to cope with 244 potentially lethal temperature conditions. Comparative analysis revealed poor correlations between changes in 245 gene expression and protein abundance under priming conditions, which were most prominent within a set of 246 355 priming-related soluble proteins (Fig. 3c). Muhlhofer et al. (2019) (2) reported similar results in yeast and 247 proposed that the massive upregulation of gene expression under moderate heat stress is required to 248 counterbalance increased protein turnover and to maintain metabolism under temperature stress. Accordingly, 249 we propose that the main function of priming is to maintain protein solubility under SHTs by activating the 250 compensation system, which stimulates gene expression and protein synthesis (Fig. 6). This assumption is 251 supported by the relatively good correlation between upregulated genes and proteins under MHT (Fig. 4a). 252 Under MHT, cellular damage is initially low, and high protein levels lead to excess cellular production and 253 hence accelerated GT growth (Fig. S1c and S1d), as also demonstrated in yeast (2). During longer stress periods, 254 damage slowly accumulates, cellular programs deviate from their optimal functions and mycelial growth 255 decreases over time (Fig. S1b). The high levels of priming-induced soluble proteins might serve as a buffer that 256 mitigates the detrimental effects of exposure to SHTs. Apart from their buffering capacity, some proteins, such 257 as the priming-induced STPs, likely have more specific roles. 258 Several lines of evidence, including transcriptomics, proteomics, pharmacological and mutagenesis data, link 259 STPs to the B. cinerea priming response, suggesting that STPs are an important group of heat adaptation 260 proteins. STPs are highly abundant in all organisms (38) and are involved in proteostasis, thereby contributing 261 to cell fitness and survival (9, 39). In Arabidopsis thaliana, the extracellular subtilase SBT3.3 is required for the 262 activation of immune priming, which is mediated by a chromatin-remodeling-and salicylic-acid-dependent 263 mechanism (40). Specific serine peptidases, such as the HTRA (41, 42), Clp (43, 44) and Lon1 (8,9)  supplemented with near UV light. Additional media that were used for specific experiments included malt 281 medium (5 g glucose, 15 g malt extract, 1 g peptone, 1 g casamino acids) and GB5-Glc medium (Gamborg B5 282 with vitamins and 2% glucose, Duchefa Biochemie). 283

Germination, colony growth, and germ tube elongation 284
Germination. Fungi were cultured on PDA for 7 d. Spores were collected by washing with PDB and filtered 285 through two layers of Miracloth, and the spore density was adjusted to 5 × 10 5 /ml. A 20-μl droplet of spore 286 suspension was placed on a coverslip and incubated under continuous light at the specified temperature and 287 time. After incubation, the slides were visualized under a light microscope and germination rates were scored. 288 Each experiment was repeated at least four times, with >200 randomly selected spores each time. 289 Colony growth. Cultures were initiated from 4-mm plugs that were cut from the edge of a 2-day-old colony. The 290 plugs were placed in the center of a Petri dish containing PDA and incubated at the indicated temperatures 291 under continuous light. Colony diameter was recorded after 3 d, the diameter of the initial inoculation plug (4 292 mm) was subtracted, and radial growth was calculated. Each experiment was performed with three replications 293 (3 plates per treatment) and was repeated three times. 294 Germ tube elongation. A 20-μl droplet of spore suspension (5 × 10 4 /ml) was placed on a coverslip and 295 incubated under the indicated conditions. After incubation, the slides were visualized under a light microscope 296 and GT length was recorded. Each experiment was repeated three times, with more than 120 randomly 297 selected GTs each time. 298 Germination after heat stress. Fungi were cultured on PDA for 7 d to allow mycelia and spores to develop. The 299 plates were transferred to 37°C or 42°C for the indicated time. After heat treatment, spores were collected into 300 PDB, and spore density was adjusted to 200 spores/ml. To determine germination rates, a 5-μl spore 301 suspension was placed on a small (1 cm 2 ) PDA cube and incubated at 22°C for 72 h. The number of germinated 302 and un-germinated spores was scored, and germination rate was calculated. Each experiment was performed 303 at least three times with >200 spores per treatment. 304

Priming-related assays 305
Cell death. Spores (2 × 10 5 /ml PBD) were placed on a coverslip, incubated at 22°C or 29°C for 12 h under 306 continuous light and transferred to 37°C. Following incubation at 37°C, the GTs were stained with 10 µg/ml PI 307 for 15 min. Samples were visualized under a Zeiss Axio imager M1 fluorescence microscope using a rhodamine 308 filter. The number of dead cells (PI-positive) was counted, and the proportion of dead cells was calculated. Each 309 experiment was repeated at least three times with more than 120 randomly selected GTs each time. 310 Membrane potential. GTs were suspended in PBS containing 100 µM DiBAC4(5) (Interchim), incubated for 15 311 min at room temperature and visualized under a fluorescence microscope using a rhodamine filter. Images 312 were captured using Zeiss AxioCam MRm camera. The fluorescence intensity of each germ tube was quantified 313 using ImageJ, and the mean signal intensity of each treatment was calculated. The experiment was repeated 314 four times with at least 50 randomly selected GTs each time. 315 Recovery of colony growth after heat shock. Colonies were initiated by placing a 5-μl droplet of spore 316 suspension (5 × 10 5 /ml) in the center of a Petri dish containing PDA. The plates were incubated under 317 continuous light at 22°C or 29°C for 12 h and then at 37°C for 6 or 9 h. After heat treatment, the cultures were 318 incubated at 22°C for 3 d and the colony diameter measured. Each experiment was repeated six times with five 319 replications per treatment. 320 High/low temperature cycles. Spores (2 × 10 5 /ml) were incubated in liquid GB5-Glc on a coverslip at 22°C for 6 321 h, transferred to 42°C for 2 h (heat treatment) and incubated at 22°C for 22 h (recovery). The cycle was 322 repeated three times, and the levels of cell death were determined after each cycle by PI staining. The 323 experiments were repeated at least three times with >300 randomly selected GTs each time. 324 Infection assays. Pathogenicity assays were performed using the first two leaves of 8-day-old French bean 325 (Phaseolus vulgaris L. genotype N9059) plants as described previously (45). The leaves were inoculated with 326 7.5-μl droplets of a spore suspension containing 2 × 10 5 spores/ml in GB5-Glc.

DNA and RNA extraction and analysis 338
Genomic DNA extraction was performed using Extract-N-Amp™ Tissue PCR Kits (Sigma/Aldrich). For cDNA 339 synthesis, total RNA was treated with DNase I (Thermo Scientific) and first-strand cDNA was synthesized from 340 1μg of DNA-free RNA using a RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). qRT-PCR was 341 performed with SYBR Premix Ex Taq II (Takara, Dalian, China) using a StepOne (Applied Biosystems) Real-time 342 PCR instrument (Applied Biosystems). Relative fold changes of mRNA levels determined by RNA-seq were 343 normalized to the expression of the ribosomal protein gene bcrsm18, which has relatively stable mRNA 344 expression. 345

RNA-seq and analysis 346
Biological materials. B. cinerea cultures were produced by spreading 500 μl of spore suspension (10 7 /ml) on a 347 cellophane-covered PDA plate. Treatments (Table S1) included incubation at 22°C or 29°C for 10 h (OT and 348 MHT, respectively), and 22°C or 29°C for 8 h followed by 37°C for 2 h (SHT and SHT-P, respectively). GTs were 349 collected from the plates by washing with water, centrifuged, separated into two 1.5-ml Eppendorf tubes, 350 flash-frozen in liquid nitrogen and stored at -80°C. RNA was extracted from one tube per sample, with four 351 biological replications per treatment and a total of 16 samples. Samples were ground in liquid nitrogen with a 352 mortar and pestle, and RNA was extracted with TRIzol reagent (Sigma-Aldrich) according to the manufacturer's 353 instructions. 354 RNA quality, library preparation, and RNA-seq. cDNA libraries were prepared using a NEBNext Ultra II RNA 355 Library Prep Kit (NEB). RNA and cDNA quality and quantity were evaluated by performing a Qubit assay with 356 the TapeStation  with high quality (Phred score > 34). The number of reads after QC and trimming ranged from 19,604,233 to 363 29,043,828 (Table S2). The clean reads were mapped to the B. cinerea reference genome (assembly 364 ASM14353v4, https://www.ncbi.nlm.nih.gov/genome/494) using STAR-2.7.3a. Over 97% of the reads in each 365 sample were mapped to the B. cinerea genome (Table S2), and 11,219 of the 11,700 annotated B. cinerea 366 strain B05.10 genes were identified. Principal component analysis (PCA) separated the samples into four 367 distinct clusters based on treatment (Fig. S3a). RT-PCR analysis of eight genes validated the RNA-seq results, 368 with highly similar expression patterns, confirming the accuracy of the RNA-seq data (Fig. S3b and S3c). 369 Quantification and normalization of gene expression were performed using the Partek E/M algorithm and 370 DESeq2, respectively. Gene expression values were calculated as fragments per kilobase per million (FRKM). 371 Principal coordinate analysis and heatmap analysis were performed using the Partek Flow tool or R software. 372 Identification of DEGs was performed with the DESeq2 package with cutoffs of pFDR < 0.05 and |fold change| 373 (FC) ≥ 2 or ≥ 5. GO enrichment analysis of the DEGs was performed using the FungiDB platform 374 (https://fungidb.org/fungidb/app) (46). 375

Isolation of soluble and aggregated proteins 376
To isolate soluble and aggregated proteins, fungal material was produced as described for the RNA-seq 377 experiment, except that the initial incubation at 22°C or 29°C was performed for 14 h instead 8 h to allow the 378 accumulation of sufficient biomass. Soluble and aggregated proteins were extracted from the samples 379 according to Koplin and Brennan (47,48) with some modifications. At the end of the heat treatments, the 380 fungal biomass was collected (0.8-1.6 g), ground to a fine powder in liquid nitrogen and suspended in 1.2 ml of 381 freshly prepared protein extraction buffer (20 mM Na-phosphate, pH 6.8, 10 mM DTT, 1 mM EDTA, 0.1% 382 Tween, 1 mM PMSF, 1× Mini Protease Inhibitor Cocktail, Roche). The samples were centrifuged at 3000g for 6 383 min at 4°C using a tabletop centrifuge, and each supernatant was transferred to a new 1.5-ml Eppendorf tube. 384 The samples were centrifuged at 5,000g for 5 min at 4°C to precipitate remnants of cellular debris, and the 385 supernatant was transferred to a new tube. A sample of the protein extract was measured in a 386 spectrophotometer, and the amount of total protein in each sample was calculated using a 1 mg/ml standard 387 sample of bovine serum albumin (Sigma-Aldrich). To separate the soluble and aggregated proteins, extracted 388 proteins were centrifuged at 16,000g for 20 min at 4°C, and the supernatant was carefully removed, 389 transferred to a new tube and stored at -80°C as the soluble protein fraction. The pellet was washed twice 390 with 300 μl sodium phosphate buffer (20 mM Na-phosphate pH 6.8, 1 mM PMSF, 1× Mini Protease Inhibitor 391 Cocktail, Roche) with centrifugation at 16,000g for 20 min between washes. The supernatant was removed and 392 the clean pellet stored at -80°C as the protein aggregate fraction. 393

SDS-PAGE and proteomic analyses 394
For soluble proteins, samples were mixed with protein loading buffer and boiled for 5 min. The samples were then mixed with 2 M urea, 80 mM ammonium bicarbonate solution and digested with 402 modified trypsin (Promega) by overnight incubation at a 1:50 enzyme-to-substrate ratio at 37°C. The resulting 403 tryptic peptides were desalted using C 18 tips (Harvard), dried and resuspended in 0.1% formic acid. The samples 404 were analyzed by LC-MS/MS using a Q Exactive Plus mass spectrometer (Thermo) fitted with a capillary high-405 performance liquid chromatograph (HPLC; easy nLC 1000; Thermo). The peptides were loaded onto a 406 homemade capillary column (25-cm, 75-μm internal diameter) packed with Reprosil C 18 Aqua (Dr. Maisch 407 GmbH, Ammerbuch, Germany) in solvent A (0.1% formic acid in water). The peptide mixture was resolved with 408 a linear gradient (5%-28%) of solvent B (95% acetonitrile with 0.1% formic acid) for 105 min, followed by a 15-409 min gradient of 28%-95% and 15 min at 95% acetonitrile with 0.1% formic acid in water at a flow rate of 0.15 410 μl/min. Mass spectrometry was performed in positive mode (m/z 350 to 1,800; resolution 70,000) using a 411 repetitively full MS scan, followed by collision-induced dissociation (high-energy collision dissociation [HCD], at 412 a normalized collision energy of 35) of the 10 most dominant ions (>1 charges) selected from the first MS scan. 413 The AGC settings were 3 × 10 6 for the full MS scan and 1 × 10 5 for the MS/MS scans. The intensity threshold for 414 triggering MS/MS analysis was 1 × 10 4 . A dynamic exclusion list was enabled with an exclusion duration of 20 s. 415 The mass spectrometry data from all three replications were analyzed using MaxQuant software v.1.5.2.8. (50)  416 for peak identification and quantitation using the Andromeda search engine, which searches for tryptic 417 peptides against the B05.10 UniProt database (51), with a mass tolerance of 20 ppm for both the precursor 418 masses and fragment ions. Oxidation on methionine and protein N-terminus acetylation were accepted as 419 variable modifications, and carbamidomethyl on cysteine was accepted as a static modification, as the 420 percentage of carbamylation was low. Minimal peptide length was set to 6 amino acids, and a maximum of two 421 missed cleavages were allowed. Peptide-and protein-level false discovery rates (FDRs) were filtered to 1% 422 using the target decoy strategy. Protein tables were filtered to eliminate identifications from the reverse 423 database and common contaminants and single peptide identifications. The data were quantified by SILAC 424 analysis using the same software. H/L ratios for all peptides belonging to a particular protein species were 425 pooled, providing a ratio for each protein. 426

Protease inhibitor assay 427
A 20-μl droplet of spore suspension (10 5 /ml) was placed on a coverslip and incubated at 29°C for 8 h. The 428 suspension medium was carefully removed and replaced with fresh PDB supplemented with 2× protease 429 inhibitor (Mini Protease Inhibitor Cocktail, Roche), 0.5 mM or 2 mM PMSF (Sigma-Aldrich), or an equal volume 430 of PDB (control treatment). The samples were incubated at 29°C or 37°C for 4 h, the GTs were stained with PI, 431 and the death rate was calculated. Each experiment was repeated at least three times with more than 150 432 randomly selected GTs per sample. 433

Generation of serine protease deletion strains 434
Deletion strains were generated using a marker-free CRISPR-Cas9 genome editing method (52) with two 435 sgRNAs per gene. All CRISPR-Cas9 reagents were purchased from Integrated DNA Technology (IDT). Selection 436 and design of sgRNAs were conducted with a sgRNA design web platform (http://grna.ctegd.uga.edu/). To 437 assemble a single sgRNA duplex (33µM), 0.2 nm each of Alt-R CRISPR-Cas9 crRNA (2µl) and Alt-R CRISPR-Cas9 438 tracrRNA (2µl) were mixed with Nuclease-Free Duplex Buffer (2 µl). The mixture was incubated for 5 min at 439 95°C and allowed to cool at room temperature. To assemble a Cas9/sgRNA ribonucleoprotein (RNP) complex, 3 440 μg Cas9 (3 μl), 33 µM sgRNA duplex (2 µl) and 5 µl of Cas9 working buffer were mixed and incubated at 37°C for 441 30 min. 442 Transformation of B. cinerea was performed according to Leisen et al. (53). Briefly, protoplasts were incubated 443 with 2 µg pTEL-Fen telomeric plasmid (52) and two RNP complexes per gene, each containing 3 µg Cas9 and 1 444 µg sgRNA. The protoplasts were mixed with 50 ml of liquified SH agar medium supplemented with 30 mg/l 445 fenhexamid (Fen; Teldor), and the mixture was dispensed into 90-mm Petri dishes. The plates were incubated 446 at 22°C for 3 d, and Fen-resistant colonies were collected. The colonies were transferred to PDA plates without 447 selection to allow rapid growth and the loss of pTEL-Fen selection. Hyphal tips from fast-growing isolates were 448 transferred to fresh PDA plates, allowed to grow for 4 d and subjected to DNA extraction and PCR analysis to 449 verify the deletion. A single round of single spore isolation was performed, followed by DNA extraction and PCR 450 analysis to verify that the strains were homokaryotic for the deletion. To generate strains with deletions of 451 multiple genes, purified strains were selected and transformed with RNP complexes for additional genes. For 452 each strain, three to four individual isolates were selected and subjected to initial phenotyping by examining 453 growth morphology, sporulation, spore germination, and the priming responses of GTs. 454

Biomass production 455
Liquid cultures were initiated by inoculating 50 ml malt medium with 10 6 spores/ml. The cultures were 456 incubated for 12 h at 22°C with shaking at 150 rpm under continuous light and transferred to 29°C. After 12 h 457 incubation with shaking at 29°C, two samples of 10 ml each were removed, placed in 15-ml tubes and 458 centrifuged at 4000g for 15min. The supernatant was discarded and the fresh weight was measured. The tubes 459 with pellets were incubated in an oven at 55°C for 24 h and the dry weight measured. After removing 20-ml 460 aliquots, the remaining cultures were transferred to 37°C and incubated 12 h, followed by 22°C for 24 h. A 20-461 ml aliquot was removed from each sample, and the fresh and dry weights were measured. Each experiment 462 was repeated three times with two replications per treatment. 463

Statistical analysis 464
The statistical significance between means of treatments was evaluated by Student's t-test (two-tailed t-test). declare that they have no competing interests. Data and materials availability: All data needed to evaluate the 474 conclusions in the paper are present in the paper and/or the supplementary materials. RNA sequencing data 475 have been uploaded to NCBI and will be available once granted a project accession number.