Effects of Carbon, Nitrogen, Ambient pH and Light on Mycelial Growth, Sporulation, Sorbicillinoid Biosynthesis and Related Gene Expression in Ustilaginoidea virens

Sorbicillinoids are a class of hexaketide metabolites produced by Ustilaginoidea virens (teleomorph: Villosiclava virens), an important fungal pathogen that causes a devastating rice disease. In this study, we investigated the effects of environmental factors, including carbon and nitrogen sources, ambient pH and light exposure, on mycelial growth, sporulation, as well as the accumulation of sorbicillinoids, and the expression of related genes involved in sorbicillinoid biosynthesis. It was found that the environmental factors had great influences on mycelial growth and sporulation of U. virens. Fructose and glucose, complex nitrogen sources, acidic conditions and light exposure were favorable for sorbicillinoid production. The relative transcript levels of sorbicillinoid biosynthesis genes were up-regulated when U. virens was separately treated with those environmental factors that favored sorbicillinoid production, indicating that sorbicillinoid biosynthesis was mainly regulated at the transcriptional level by different environmental factors. Two pathway-specific transcription factor genes, UvSorR1 and UvSorR2, were found to participate in the regulation of sorbicillinoid biosynthesis. These results will provide useful information to better understand the regulation mechanisms of sorbicillinoid biosynthesis, and be conducive to develop effective means for controlling sorbicillinoid production in U. virens.

Different regulation levels of secondary metabolism include environmental stimuli, global regulators, specific regulators, signal transduction pathways, epigenetic regulations, and post-translational modifications, as well as the combination of those regulations; they form a complex and multi-level regulatory network in fungi [23][24][25][26].
For the carbon source assessment, the PB medium was supplemented with different carbon sources (i.e., soluble starch, apple pectin, fructose, glucose, sucrose, maltose, mannose, stachyose, dextrin, lactose, galactose, and xylose) at the final concentration of 10 g/L. The different basal media were employed for the determination of the effects of the following other environmental factors. They were used in independent experiments for the determination of the effects of different environmental factors. Only one variable was changed in each independent experiment.
For nitrogen source assessment, the PB medium was supplemented with glucose (10 g/L) as basal medium, with different nitrogen sources (i.e., yeast extract paste, yeast extract powder, beef extract, malt extract, maize powder, soybean powder, nutrient broth, peptone, polypeptone, tryptone, KNO 3 and urea), at the final concentration of 10 g/L.
For the assessment of carbon and nitrogen sources, PB medium was used as the basal medium. In order to better evaluate the influences of environmental factors, we carried out the following pH and light experiments, based on the assessment results of the carbon and nitrogen sources.
To test the effects of ambient pH on mycelial growth, sporulation and sorbicillinoid production in U. virens, the pH values of PBGY medium (PB medium, supplemented with 10 g/L glucose and 10 g/L yeast extract paste) were amended to between 3.0 and 8.0, with either acid or alkali solution (1 M HCl and 1 M NaOH).
To determine U. virens's' response to light exposure, U. virens was grown on PBGB medium (PB medium, supplemented with 10 g/L glucose and 10 g/L beef extract) and was cultured in an LED light incubator at 28 • C, with different light treatments (DD, 24 h dark/day; LL, 24 h light/day; DL, 12 h dark-12 h light/day). Light was provided by conventional LED bulbs with 4000 Lux light intensity, and the color temperature (Kelvin) of the LED bulbs was 5000 K.

Mycelial Growth and Sporulation Assessments
To evaluate the effects of different carbon and nitrogen sources, ambient pH, and light exposure on mycelial growth and sporulation, the plates with different environmental factors were inoculated with 2 µL conidia suspension (1 × 10 6 conidia/mL) of U. virens, and were cultivated at 28 • C in darkness. For growth assessment, the plates with mycelia were photographed on day 18 of cultivation, then the colony extension diameters were measured. For mycelia biomass measurement, 2 µL of conidia suspension (1 × 10 6 conidia/mL) was cultivated on different medium plates, covered with a sheet of sterile cellophane at 28 • C, in darkness, for 18 days. Then, the mycelia were collected from the medium through cellophane separation, dried in an oven at 55 • C to a constant weight, and the dry weight was then determined. During the experiment, we also directly inoculated U. virens onto the agar plates supplemented with different carbon sources, without cellophane covering the agar medium. It was found that the growing status of the fungus was almost the same as that grown with cellophane. For sporulation assessment, a 100 µL conidia suspension (1 × 10 6 conidia/mL) of U. virens was coated onto a 60 mm-diameter plate with different environmental factors, and cultured at 28 • C in darkness for 18 days. Two mycelial plugs with diameters of 10 mm were taken from each plate, then homogenized in 5 mL water containing 0.05% Tween 80, and filtered with Miracloth. The spore suspension was diluted to a suitable concentration. The conidia quantity of each plate was determined using a hemocytometer. The sporulation is presented as conidia number per square millimeter (conidia/mm 2 ) of agar surface [51,52].

Sorbicillinoid Production Determination
The 28 day-old mycelia on the cellophane were separated from different media in plates, and dried at 55 • C in an oven to obtain the dry weight. Then, the dried mycelia and their medium were extracted with 10 mL analytical reagent-grade methanol in an ultrasonic bath three times (35 kHz, 30 min). After removal of the solid, by filtration, the filtrate was evaporated to dryness and re-dissolved in 1 mL of chromatographic grade methanol. Three main sorbicillinoids, including trichotetronine (also called bislongiquinolide), demethyltrichodimerol, and trichodimerol, were detected and quantified with the standards obtained from our previous study, which were identified according to their molecular weight, retention times and spectrometric data, with their structures shown in Figure S1A [11]. In terms of analyzing the contents and yields of major sorbicillinoids, HPLC analysis was performed by injecting 10 µL of the filtrate extract, with a microporous filter (pore size, 0.22 µm), into a Shimadzu instrument consisting of an SPD-M20A photodiode array detector (LC-20A, Shimadzu Corp., Tokyo, Japan), using an analytic C18 column (250 mm × 4.6 mm i.d., 5 µm; Phenomenex Inc., Torrance, CA, USA). The column temperature was set to 30 • C. The mobile phase was composed of methanol (mobile phase B), and water, containing 0.02% TFA (mobile phase A). A gradient elution program, eluting from 60% to 100% methanol over 40 min, was used; the flow rate was 1.0 mL/min and detection wavelength was 370 nm. Three main sorbicillinoids were quantified with the standard regression equations; the regression equations were Y = 1,138,395.3477 X − 3022.8909 (R = 0.9982) for trichotetronine, Y = 2,254,860.6801 X − 18,274.5152 (R = 0.9995) for demethyltrichodimerol, and Y = 4,854,544.1694 X + 29618.4711 (R = 0.9992) for trichodimerol, where Y is the peak area, X is the quality (µg) of the sample injected each time, and R is the correlation coefficient. Sorbicillinoid production was assessed by content (mg/g) and yield (mg/L). The sum of trichotetronine, demethyltrichodimerol and trichodimerol contents was considered to be the major sorbicillinoid content. The sum of three main sorbicillinoid yields was considered to be the major sorbicillinoid yield. Content (mg/g) was equal to total sorbicillinoids (mg), divided by mycelia dry weight (g), and yield (mg/L) was the total sorbicillinoids (mg) of per liter of culture broth, according to the method of Zong et al. [50].

RNA Preparation, RT-qPCR and Sorbicillinoid BGC Genes Analyses
A total of 2 µL of spore suspension (1 × 10 6 conidia/mL) of U. virens was inoculated onto each plate, and covered with sterile cellophane at 28 • C for 10 days. Mycelia were harvested and stored at −80 • C for RNA extraction. All tips and mortars were autoclaved twice to prevent RNase contamination, and the mycelia were ground into powder, with liquid nitrogen, in a mortar, manually. According to the operation manual, total RNA was extracted from 100 mg of mycelial sample using Trizol reagent (TransGen Biotech, Beijing, China), in order to lyse cells and inhibit nuclease release. The concentration and purity were measured with an ultraviolet spectrophotometer, and RNA integrity was checked by agarose gel electrophoresis. The reverse transcription of RNA into cDNA was performed using the Fast Quant RT Kit (TransGen Biotech, Beijing, China). The specific primers were designed and chosen to analyze the transcript level of genes (Table S2). The reverse transcription-quantitative PCR (RT-qPCR) assays were performed using the UltraSYBR One Step RT-qPCR Kit (CWBIO, Beijing, China) and the QuantStudio ® 6 Flex System (BioRad, Hercules, CA, USA). RT-qPCR was carried out in a total volume of 20 µL, containing 10 µL of SYBR Premix Ex Taq™, 0.4 µL of each primer (10 µM), 0.4 µL of Lox ROX Reference Dye, 1 µL of cDNA, and 7.8 µL of RNase free-water. The transcript levels from different samples were normalized to that of β-tubulin gene to compensate for variations in the amount of cDNA input. The β-Tubulin gene, used as the reference gene in this study, was considered to be stable [53][54][55]. Three biological and three technical replicates performed for each sample. Each relative transcript level was calculated by using the 2 −∆∆CT method, with 100% amplification efficiency and using suitable internal reference gene [56]. To normalize the date and aesthetics of the heatmap, the UvSorA gene, showing moderate transcriptional levels in our previous research, was selected as the control [22]. Additionally, one culture condition, with moderate transcriptional levels of sor BGC genes, was selected as the control group in each independent environmental factor experiment. The UvSorA gene was used as the control in carbon and nitrogen assays, respectively, with the cultures grown in the media supplemented with maltose and malt extract. In the ambient pH assay, the control was the UvSorA gene, with the cultures grown in pH 5.5 medium. In the light exposure assay, the control was the UvSorA gene, with the cultures treated with DL light exposure. The relative transcript level was determined by log2FC (fold change) between the normalized value of the target gene in each sample and the control sample gene. Then, the package pheatmap in the R programming language (accessed on 20 July 2022) was used to visualize the gene expression data (https://cran.microsoft.com/snapshot/ 2018-06-22/web/packages/pheatmap/pheatmap.pdf).

Gene Deletion and Phenotyping
All primers used to construct UvSorR1 and UvSorR2 knockout mutants are listed in Table S2. The gene deletion via the CRISPR/Cas9 system was performed according to the previous method [57,58]. The gene deletion cassette was constructed by upstream and downstream flanking sequences of the target gene, fused with the geneticin-resistance gene (NeoR). The gRNA spacers were cloned into the CRISPR-Cas9 vector (pCas9-tRp-gRNA). U. virens protoplasts were transformed by PEG-mediated transformation, and the mutants were selected from the medium containing 700 µg/mL of G418 (Sigma-Aldrich, St. Louis, MO, USA). Genomic DNA was extracted from the mycelia of transformants, by using a regular phenol-chloroform method, and correct gene replacements for upstream, downstream and target genes were confirmed by diagnostic PCR.
For mycelial growth and sporulation assessments, 2 µL of spore suspensions (1 × 10 6 conidia /mL) from the WT strain and deletion mutants were inoculated on PDA medium (potato, 200 g/L; glucose, 20 g/L; agar, 20 g/L) at 28 • C for 18 days. Colony extension diameter, mycelia dry weight and spore concentration were measured. For chemical analyses, the WT strain and deletion mutants were grown on GYES medium (yeast extract, 10 g/L; glucose, 10 g/L; starch 10 g/L; NaCl, 5 g/L, CaCO 3 , 3 g/L) at 28 • C for 28 days. The MeOH extracts of the plates with hyphae were analyzed by HPLC.

Statistical Analysis
All experiments were designed with three independent biological replicates. Five replicates were performed for each treatment. Similar results were garnered for each biological experiment. Statistical analysis was carried out using SPSS version 17.0 (SPSS, Inc., Chicago, IL, USA), with a one-way analysis of variance (ANOVA) and Duncan's multiple range test. Data are expressed as standard error (SE) of the mean, and differences at p < 0.05 were considered statistically significant.

Effects of Carbon Sources on Mycelial Growth, Sporulation and Sorbicillinoid Production
Twelve different carbon sources were tested for their effects on mycelial growth, sporulation and sorbicillinoid production in U. virens. Except for the fact that U. virens could not grow on medium with xylose as the carbon source, the colony morphology and pigment accumulation varied when U. virens was cultured with other carbon sources ( Figure 1A). Colony diameter and mycelial biomass were used to evaluate mycelial growth. An increase in mycelial radial expansion was observed when the medium was separately supplemented with dextrin, soluble starch, stachyose and lactose, but these carbon sources had less mycelial biomass accumulation than glucose, sucrose, maltose, and mannose. For sporulation, it was observed that the maximum sporulation (9.3 × 10 3 conidia/mm 2 ) of U. virens was observed when adding fructose, while minimum sporulation (2.0 × 10 2 conidia/ mm 2 ) was recorded when adding lactose ( Figure 1B).
Sorbicillinoid production was also obviously influenced by the carbon source. Fructose and glucose significantly promoted sorbicillinoid production, followed by maltose, sucrose, apple pectin, mannose, and galactose. In contrast, low sorbicillinoid production was detected in the medium supplemented with dextrin, stachyose, lactose and soluble starch, respectively ( Figure 1C). Except for galactose, the content and yield of trichotetronine were the highest, while the content and yield of demethyltrichodimerol were the lowest among the three tested sorbicillinoids, regarding the media supplemented with all tested carbon sources (Table S3).
sucrose, apple pectin, mannose, and galactose. In contrast, low sorbicillinoid production was detected in the medium supplemented with dextrin, stachyose, lactose and soluble starch, respectively ( Figure 1C). Except for galactose, the content and yield of trichotetronine were the highest, while the content and yield of demethyltrichodimerol were the lowest among the three tested sorbicillinoids, regarding the media supplemented with all tested carbon sources (Table S3).

Effects of Nitrogen Sources on Mycelial Growth, Sporulation and Sorbicillinoid Production
To investigate the effects of nitrogen sources on mycelial growth, sporulation and sorbicillinoid production, twelve different nitrogen sources, including one inorganic nitrogen source, and eleven organic nitrogen sources were tested in this study. The colony morphology and pigment accumulation were markedly influenced by the nitrogen source supplemented in the media, but U. virens could not grow in the media supplemented with urea ( Figure 2A). Nitrogen sources affected the mycelial growth and sporulation of U. virens. Overall, organic nitrogen sources were more conducive to mycelial growth than potassium nitrate (KNO 3 ). Peptone nitrogen sources (i.e., polypeptone, peptone and tryptone) promoted mycelial biomass accumulation. The maximum sporulation (3.32 × 10 4 conidia/mm 2 ) of U. virens was recorded in media supplemented with beef extract, followed by yeast extract paste, peptone, nutrient broth, malt extract and tryptone. Other nitrogen sources resulted in relatively less sporulation ( Figure 2B).
Organic nitrogen sources resulted in increased sorbicillinoid production when compared to KNO3, except for yeast extract powder. Among ten complex nitrogen sources, yeast extract paste, peptone and beef extract were the most favorable nitrogen sources for sorbicillinoid production in U. virens. Tryptone was better for sorbicillinoid production, compared with polypeptone, maize powder, nutrient broth, malt extract, and soybean powder ( Figure 2C). In addition, yeast extract paste and beef extract were more favorable for trichotetronine and demethylthyltrichodimerol production, while peptone was more favorable for trichodimerol production (Table S4).

Effects of Ambient pH on Mycelial Growth, Sporulation and Sorbicillinoid Production
The effects of ambient pH on mycelial growth, sporulation and sorbicillinoid production were established by culturing U. virens in PBGY medium, with pH values ranged from 3.0 to 8.0. U. virens had a broad pH range for mycelial growth and sporulation, but could not grow on the medium with pH values of 3.0 and 8.0 ( Figure 3A). Mycelia grew well at pH 4.5 to 7.0, whereas mycelial growth was significantly inhibited at pH ≤ 4.0 and pH ≥ 7.5. The optimum pH value for mycelial growth and biomass accumulation in U. virens was pH 6.0. The highest sporulation (5.59 × 10 4 conidia/mm 2 ) was obtained at pH 5.0 ( Figure 3B).
Ambient pH had a significant effect on sorbicillinoid production. The highest yield (8.3 mg/L) of sorbicillinoids was observed at pH 4.5, which was on par with that observed at pH 5.0. An acidic environment was more beneficial for sorbicillinoid accumulation in mycelia. The highest content (2.0 mg/g) of sorbicillinoids was observed at pH 3.5, and the content of sorbicillinoids was gradually decreased when pH was increased ( Figure 3C). In addition, ambient pH also affected the composition of sorbicillinoids (Table S5). Trichotetronine was the major component in the sorbicillinoids when pH was 3.5, and distinctly Organic nitrogen sources resulted in increased sorbicillinoid production when compared to KNO 3 , except for yeast extract powder. Among ten complex nitrogen sources, yeast extract paste, peptone and beef extract were the most favorable nitrogen sources for sorbicillinoid production in U. virens. Tryptone was better for sorbicillinoid production, compared with polypeptone, maize powder, nutrient broth, malt extract, and soybean powder ( Figure 2C). In addition, yeast extract paste and beef extract were more favorable for trichotetronine and demethylthyltrichodimerol production, while peptone was more favorable for trichodimerol production (Table S4).

Effects of Ambient pH on Mycelial Growth, Sporulation and Sorbicillinoid Production
The effects of ambient pH on mycelial growth, sporulation and sorbicillinoid production were established by culturing U. virens in PBGY medium, with pH values ranged from 3.0 to 8.0. U. virens had a broad pH range for mycelial growth and sporulation, but could not grow on the medium with pH values of 3.0 and 8.0 ( Figure 3A). Mycelia grew well at pH 4.5 to 7.0, whereas mycelial growth was significantly inhibited at pH ≤ 4.0 and pH ≥ 7.5. The optimum pH value for mycelial growth and biomass accumulation in U. virens was pH 6.0. The highest sporulation (5.59 × 10 4 conidia/mm 2 ) was obtained at pH 5.0 ( Figure 3B).

Effects of Light Exposure on Mycelial Growth, Sporulation and Sorbicillinoid Production
To investigate the effects of light exposure on mycelial growth, sporulation and sorbicillinoid production in U. virens, 24 h dark/day (DD), 24 h light/day (LL), and 12 h dark-12 h light/day (DL) periods were used for different light treatments. Light exposure influenced colony morphology and pigment accumulation ( Figure 4A). When compared to darkness, the mycelial growth was reduced with light treatment, whereas sporulation was promoted by light exposure (Figure 4B).  Ambient pH had a significant effect on sorbicillinoid production. The highest yield (8.3 mg/L) of sorbicillinoids was observed at pH 4.5, which was on par with that observed at pH 5.0. An acidic environment was more beneficial for sorbicillinoid accumulation in mycelia. The highest content (2.0 mg/g) of sorbicillinoids was observed at pH 3.5, and the content of sorbicillinoids was gradually decreased when pH was increased ( Figure 3C). In addition, ambient pH also affected the composition of sorbicillinoids (Table S5). Trichotetronine was the major component in the sorbicillinoids when pH was 3.5, and distinctly reduced when the pH value was higher than 5.0. Trichodimerol was rarely detectable at pH 3.5, but its yield (2.2 mg/L) and content (0.4 mg/g) reached the maximum when the pH increased to 4.5 ( Figure S1).

Effects of Light Exposure on Mycelial Growth, Sporulation and Sorbicillinoid Production
To investigate the effects of light exposure on mycelial growth, sporulation and sorbicillinoid production in U. virens, 24 h dark/day (DD), 24 h light/day (LL), and 12 h dark-12 h light/day (DL) periods were used for different light treatments. Light exposure influenced colony morphology and pigment accumulation ( Figure 4A). When compared to darkness, the mycelial growth was reduced with light treatment, whereas sporulation was promoted by light exposure ( Figure 4B). Light exposure enhanced sorbicillinoid production, particularly regarding the con tent of sorbicillinoids in mycelia. Sorbicillinoid content was gradually increased with ligh exposure, and reached the maximum (5.12 mg/g) when the mycelia were treated wit constant light (LL) ( Figure 4C). The content of trichotetronine, demethylthyltrichodimero and trichodimerol all displayed similar trends with light exposure. While the yields o trichotetronine (5.8 mg/L) and demethylthyltrichodimerol (1.1 mg/L) were highest wit the DL treatment, the yield (1.4 mg/L) of trichodimerol was highest with the DD treatmen (Table S6).  Light exposure enhanced sorbicillinoid production, particularly regarding the content of sorbicillinoids in mycelia. Sorbicillinoid content was gradually increased with light exposure, and reached the maximum (5.12 mg/g) when the mycelia were treated with constant light (LL) ( Figure 4C). The content of trichotetronine, demethylthyltrichodimerol and trichodimerol all displayed similar trends with light exposure. While the yields of trichotetronine (5.8 mg/L) and demethylthyltrichodimerol (1.1 mg/L) were highest with the DL treatment, the yield (1.4 mg/L) of trichodimerol was highest with the DD treatment (Table S6).

Relative Transcripts Levels of the Genes Involved in Sorbicillinoid Biosynthesis in Response to the Environmental Factors
Seven conserved genes, UvSorA, UvSorB, UvSorR1, UvSorR2, UvSorC, UvSorT and UvSorD, participated in sorbicillinoid biosynthesis in U. virens [22] ( Figure 5A). The RT-qPCR method was used to explore the transcript level patterns of these seven genes in response to different environmental factors. Overall, most of the transcript levels of these genes under various environmental factors were consistent with sorbicillinoid production in U. virens ( Figure 5B-E).
When U. virens was cultured in the media with fructose, glucose, sucrose, and maltose as carbon sources, respectively for conducive sorbicillinoid production, the transcript levels of sorbicillinoid biosynthesis-related genes were up-regulated significantly when compared with low-sorbicillinoid-production carbon sources, such as soluble starch, stachyose, dextrin and lactose. The correlation heatmap of sorbicillinoid biosynthesis genes under different carbon sources shows that the high-transcript-level carbon sources fructose and glucose are clustered in one group. Two low-transcript-level carbon sources, stachyose and lactose, are clustered in another group. From gene cluster analysis, UvSorB, UvSorC and UvSorD, with the similar transcript patterns, are clustered in one group, while UvSorR1, UvSorR2 and UvSorA are clustered in another group, and UvSorT is in a separated group ( Figure 5B).
The transcript levels of sorbicillinoid biosynthesis genes were up-regulated significantly with the use of favorable nitrogen sources, including yeast extract paste, beef extract and peptone, and these nitrogen sources are clustered in one group. Malt extract, maize powder and soybean powder are clustered together, and the sorbicillinoid biosynthesis genes for this cluster present a moderate transcript level. Yeast extract powder, KNO 3 , tryptone, nutrient broth and polypeptone, as low-transcript-level nitrogen sources, belong to the same group ( Figure 5C).
An acidic environment was more beneficial for high transcript levels of sorbicillinoid biosynthesis genes in U. virens, according to correlation heatmap of ambient pH. Interestingly, the relative transcript levels for most of the genes displayed a similar trend to sorbicillinoid content. Cluster analysis based on ambient pH shows that pH values 3.5-5.0 are grouped into one cluster, and pH values 5.5-7.5 are grouped into another cluster. By utilizing gene transcript profile cluster analysis, UvSorT and UvSorR2 are in the same cluster, while the other genes are in another cluster ( Figure 5D). Light exposure with high expression level of sorbicillinoid biosynthesis genes enhanced sorbicillinoid content in U. virens. In particular, the expression level of the transcription factor gene UvSorR2 in constant light (LL) was up-regulated 86-fold, relative to that observed in constant dark (DD). The heatmap cluster analysis shows that light exposure treatment clustered in one group, and sorbicillinoid biosynthesis genes are perfectly clustered according to their function. Two backbone genes (UvSorA and UvSorB), two tailoring genes (UvSorC and UvSorD), two regulatory genes (UvSorR1 and UvSorR2) and one transporter gene (UvSorT) are clustered in different groups ( Figure 5E).
On the whole, the expressions of these seven genes were increased in response to favorable carbon and nitrogen sources, an acidic pH and light exposure. However, the expressions of some genes did not change significantly in response to the environmental factors, or were even down-regulated (i.e., ambient pH). Therefore, the genes for which the fold changes were significant should be further analyzed.

Deletion of Transcription Factor Genes UvSorR1 and UvSorR2
Two transcription factors, UvSorR1 and UvSorR2, in sor BGC were deleted with a geneticin-resistance (NeoR) cassette, in order to confirm their function in sorbicillinoid biosynthesis regulation. The deletion fragments, flanking approximately 1.0 kb upstream and downstream of UvSorR1 or UvSorR2 ORF regions, were fused partially with the geneticin-resistance gene. The gRNA spacers of UvSorR1 and UvSorR2 were cloned into the CRISPR-Cas9 vector (pCas9-tRp-gRNA). Both ∆uvSorR1 and ∆uvSorR2 mutants were generated by replacing the endogenous UvSorR1 or UvSorR2 ORF with the deletion cassette, via protoplast transformation with linear donor DNA fragments and the CRISPR construct. Genomic DNA was extracted from the transformants, and correct gene replacements were confirmed by diagnostic PCR (Figure 6). PCR assays using specific pairs of primers (Table S6) for the deletion of UvSorR1 or UvSorR2 were conducted, and then we further verified homologous recombination in both upstream and downstream flanking sequences in transformants.
(C) The correlation heatmap between nitrogen sources and relative transcript level of sorbicillinoid biosynthesis genes; (D) The correlation heatmap between ambient pH and relative transcript level of sorbicillinoid biosynthesis genes; (E) The correlation heatmap between different light exposure treatments and relative transcript level of sorbicillinoid biosynthesis genes.

Deletion of Transcription Factor Genes UvSorR1 and UvSorR2
Two transcription factors, UvSorR1 and UvSorR2, in sor BGC were deleted with a geneticin-resistance (NeoR) cassette, in order to confirm their function in sorbicillinoid biosynthesis regulation. The deletion fragments, flanking approximately 1.0 kb upstream and downstream of UvSorR1 or UvSorR2 ORF regions, were fused partially with the geneticin-resistance gene. The gRNA spacers of UvSorR1 and UvSorR2 were cloned into the CRISPR-Cas9 vector (pCas9-tRp-gRNA). Both ΔUvSorR1 and ΔUvSorR2 mutants were generated by replacing the endogenous UvSorR1 or UvSorR2 ORF with the deletion cassette, via protoplast transformation with linear donor DNA fragments and the CRISPR construct. Genomic DNA was extracted from the transformants, and correct gene replacements were confirmed by diagnostic PCR (Figure 6). PCR assays using specific pairs of primers (Table S6) for the deletion of UvSorR1 or UvSorR2 were conducted, and then we further verified homologous recombination in both upstream and downstream flanking sequences in transformants.

Regulation of UvSorR1 and UvSorR2 Involved in Sorbicillinoid Biosynthesis
There was no difference in mycelial growth and sporulation among ∆UvSorR1 and ∆UvSorR2 mutants and WT strain ( Figure S2). In terms of sorbicillinoid synthesis of the wild-type (WT) strain, and the ∆UvSorR1 and ∆UvSorR2 mutants, the sorbicillinoids trichotetronine (1), demethylthyltrichodimerol (2) and trichodimerol (3) could be detected in the WT strain, but there was no dimeric sorbicillinoid detected in the ∆UvSorR1 mutant; only a small amount of sorbicillin (4) could be detected when comparing it with the authentic compound. The sorbicillinoid production was drastically decreased in the ∆UvSorR2 mutant. From the back views of the colonies, there was less pigment accumulation in ∆UvSorR1 and ∆UvSorR2 when compared with the WT strain (Figure 7). The gene deletion results showed that UvSorR1 and UvSorR2, as pathway-specific transcription factors, indeed participate in the regulation of sorbicillinoid in U. virens.
There was no difference in mycelial growth and sporulation among ∆UvSorR1 ∆UvSorR2 mutants and WT strain ( Figure S2). In terms of sorbicillinoid synthesis o wild-type (WT) strain, and the ∆UvSorR1 and ∆UvSorR2 mutants, the sorbicillin trichotetronine (1), demethylthyltrichodimerol (2) and trichodimerol (3) could be dete in the WT strain, but there was no dimeric sorbicillinoid detected in the ∆UvSorR1 mu only a small amount of sorbicillin (4) could be detected when comparing it with the thentic compound. The sorbicillinoid production was drastically decreased in ∆UvSorR2 mutant. From the back views of the colonies, there was less pigment accu lation in ∆UvSorR1 and ∆UvSorR2 when compared with the WT strain (Figure 7). The deletion results showed that UvSorR1 and UvSorR2, as pathway-specific transcription tors, indeed participate in the regulation of sorbicillinoid in U. virens.

UvSorR1 and UvSorR2 Regulated Sorbicillinoid Biosynthesis at Transcriptiona Level
The relative transcript levels of seven sorbicillinoid biosynthetic genes in the strain, as well as in the ΔUvSorR1 and ΔUvSorR2 mutants, were analyzed by qRT-P

UvSorR1 and UvSorR2 Regulated Sorbicillinoid Biosynthesis at Transcriptional Level
The relative transcript levels of seven sorbicillinoid biosynthetic genes in the WT strain, as well as in the ∆UvSorR1 and ∆UvSorR2 mutants, were analyzed by qRT-PCR. The results reconfirm that UvSorR1 and ∆UvSorR2 had been successfully deleted according to the absence of fluorescent signal. When compared with the WT strain, the transcript levels of sor BGC genes were markedly down-regulated in ∆UvSorR1 and ∆UvSorR2, except UvSorD, indicating that two pathway-specific regulators affected sorbicillinoid production by positively regulating the expression of sor BGC genes at the transcriptional level. Moreover, the expressions of UvSorA and UvSorB, as two PKS backbone genes, were still detected, but the relative expression level of the monooxygenase gene UvSorC was extremely low; a small amount of the backbone product sorbicillin was accumulated in ∆UvSorR1 mutant (Figure 8).
cept UvSorD, indicating that two pathway-specific regulators affected sorbicillinoid production by positively regulating the expression of sor BGC genes at the transcriptional level. Moreover, the expressions of UvSorA and UvSorB, as two PKS backbone genes, were still detected, but the relative expression level of the monooxygenase gene UvSorC was extremely low; a small amount of the backbone product sorbicillin was accumulated in ∆UvSorR1 mutant (Figure 8).

Discussion
Environmental factors could affect mycotoxin biosynthesis [23,[27][28][29]50]. Our results demonstrated that carbon and nitrogen sources, ambient pH and light exposure strongly influenced mycelial growth, sporulation and sorbicillinoid production in U. virens. A previous study reported that stachyose was a preferential carbon source utilized by U. virens [59]. However, our study showed that stachyose, soluble starch and dextrin only strongly increased colony diameter expansion, while their biomass accumulations were significantly less than sucrose, glucose, sucrose, maltose, and mannose, because of mycelium thinness. It was probable that, with the use of PB medium, rather than Czapek medium, as a basal medium, fructose could be utilized as a carbon source for U. virens mycelial growth, while xylose could not, which was inconsistent with previous reports [60]. Glucose was also a favorable carbon source for sorbicillinoid synthesis in U. virens, which was also observed in T. reesei [45,46]. Among the nitrogen sources, polypeptone, peptone and tryptone promoted mycelial biomass accumulation, while urea inhibited the mycelial growth of U. virens, which is consistent with the previous observations [60]. Yeast extract paste, peptone and beef extract could increase sporulation, and enhance sorbicillinoid production. U. virens adapted to pH values from 3.0 to 8.0, and the optimum pH value for mycelial growth and biomass accumulation was pH 6.0, which was the same as previously reports [60,61]. An acidic environment was more favorable for sorbicillinoid production than alkaline conditions, which was similar to the reports on the production of other mycotoxins, such as patulin [50], aflatoxin [62], and trichothecenes [63]. Darkness was more conducive for mycelial growth, but light enhanced the sporulation of U. virens, which also observed in other fungal species [23,60]. Light exposure considerably increased the sorbicillinoid content in mycelia, which was different from the study reporting the constitutive hyper production of sorbicillinoids in T. reesei ZC121 [64].

Discussion
Environmental factors could affect mycotoxin biosynthesis [23,[27][28][29]50]. Our results demonstrated that carbon and nitrogen sources, ambient pH and light exposure strongly influenced mycelial growth, sporulation and sorbicillinoid production in U. virens. A previous study reported that stachyose was a preferential carbon source utilized by U. virens [59]. However, our study showed that stachyose, soluble starch and dextrin only strongly increased colony diameter expansion, while their biomass accumulations were significantly less than sucrose, glucose, sucrose, maltose, and mannose, because of mycelium thinness. It was probable that, with the use of PB medium, rather than Czapek medium, as a basal medium, fructose could be utilized as a carbon source for U. virens mycelial growth, while xylose could not, which was inconsistent with previous reports [60]. Glucose was also a favorable carbon source for sorbicillinoid synthesis in U. virens, which was also observed in T. reesei [45,46]. Among the nitrogen sources, polypeptone, peptone and tryptone promoted mycelial biomass accumulation, while urea inhibited the mycelial growth of U. virens, which is consistent with the previous observations [60]. Yeast extract paste, peptone and beef extract could increase sporulation, and enhance sorbicillinoid production. U. virens adapted to pH values from 3.0 to 8.0, and the optimum pH value for mycelial growth and biomass accumulation was pH 6.0, which was the same as previously reports [60,61]. An acidic environment was more favorable for sorbicillinoid production than alkaline conditions, which was similar to the reports on the production of other mycotoxins, such as patulin [50], aflatoxin [62], and trichothecenes [63]. Darkness was more conducive for mycelial growth, but light enhanced the sporulation of U. virens, which also observed in other fungal species [23,60]. Light exposure considerably increased the sorbicillinoid content in mycelia, which was different from the study reporting the constitutive hyper production of sorbicillinoids in T. reesei ZC121 [64].
The expression of genes participating in secondary metabolism is usually strictly controlled by environmental factors through global regulators, and is regulated at the transcriptional level [24,26]. Recently, we identified the gene cluster responsible for sorbicillinoid biosynthesis; six conserved genes in sor BGC and one gene outside the cluster were involved in sorbicillinoid biosynthesis pathway in U. virens [22]. The expression levels of most sorbicillinoid-biosynthesis-related genes demonstrated a positive association with sorbicillinoid production with different carbon and nitrogen sources, an ambient pH and light exposure. In particular, UvSorT, an MFS transporter, is responsible for the export and efflux of sorbicillinoids, and its transcriptional levels were up-regulated significantly when using the culture conditions that resulted in the high production of sorbicillinoids. The correlation between UvSorT transcript levels and sorbicillinoid production should be studied in detail. However, it is difficult to reveal the correlation between carbon sources and sorbicillinoid biosynthesis. The possible reason is that there is a correlation between a certain carbon source and other genes' expression (not sor BGC expression). This might explain why the addition of lactose, stachyose, and dextrin in the medium resulted in an increased mycelial radial expansion, as well as why these carbon sources were clustered in the same clade. Conducive carbon and nitrogen sources, an acidic environment and light exposure induced a remarkable up-regulation of sor BGC genes and enhanced sorbicillinoid production, in U. virens. Expression levels of CreA, AreA, PacC and velvet complex (VeA/VelB/LaeA) orthologue genes under different environmental factors in U. virens were also investigated ( Figure S3). Both CreA and LaeA had been confirmed to be involved in sorbicillinoid regulation in T. reesei, in previous studies. [42,44]. CreA-mediated carbon catabolite repression (CCR) participates in morphology, pathogenicity and SM production [65,66]. Glucose can stimulate high transcript levels of CreA in CCR, which was also observed in our study [67,68]. The relative expression levels of UvAreA in various nitrogen sources were different, supporting the function of AreA as a nitrogen metabolism regulator, used to confer the ability to use a wide variety of nitrogen sources in fungi [34]. PacC is a pH factor involved in regulation of SM biosynthetic pathways in Aspergillus or Fusarium species [63]. In U. virens, the expression levels of UvPacC were higher under alkaline environments, with similar reports for PePacC, an activator under acidic conditions, of patulin biosynthesis in P. expansum [50]. The velvet complex (VeA/VelB/LaeA) has been implicated as a light-sensitive regulator of SMs in fungi [32]. The function of UvVeA in regulating the development and virulence of U. virens was revealed recently [69]. UvLaeA and UvVeA expression levels decreased slightly in the presence of light, revealing that they might mediate light regulation of sorbicillinoids. Considering that these global regulators mediate the regulation of carbon and nitrogen sources, ambient pH and light on sorbicillinoid biosynthesis in U. virens, other strategies, such as gene deletion, overexpression and RNA-seq are worth studying in future investigations.
Both SorR1 and SorR2, as the pathway-specific regulator genes in sor BGC, have been reported in P. chrysogenum and T. reesei. SorR1 acted as a transcriptional activator, while SorR2 controlled the expression of SorR1, and they balanced carbon and secondary metabolism in T. reesei [39,44,46,70]. Gene deletion and RT-qPCR confirmed that UvSorR1 and UvSorR2 positively participate in the sorbicillinoid biosynthesis regulation, at transcriptional levels. As two backbone genes, UvSorA and UvSorB were detected with low expression levels, and small amounts of the backbone product sorbicillin still accumulated in the medium with culture of the ∆UvSorR1 mutant, which was different from P. chrysogenum and T. reesei [39,70]. Additionally, UvSorR2 under constant light exposure (LL) was remarkably up-regulated by 86-fold when compared to the constant dark (DD), which supported previous results, in which SorR2 was considered as a regulator in response to light, and subjected to regulation by photoreceptors [21]. More details are needed to reveal the connection between the sorbicillinoid biosynthesis and environmental factors, global transcription factors and pathway-specific regulators. Our results may provide valuable information for understanding the complex regulation mechanism and network of sorbicillinoid biosynthesis in U. virens.

Conclusions
In summary, this study revealed the effects of carbon and nitrogen sources, ambient pH, and light exposure on mycelial growth, sporulation and sorbicillinoid production in U. virens. Among these environmental factors, glucose and sucrose, peptone and tryptone, an ambient pH of 6.0 and darkness were more conducive to mycelial growth. Fructose, beef extract, a pH of 5.0 and light exposure stimulated sporulation. Moreover, glucose and fructose, yeast extract paste and beef extract, acidic conditions and light exposure were favorable for sorbicillinoid production. Sorbicillinoid biosynthesis was mainly regulated at the transcriptional level by different environmental factors, and environmental factors also modulated the expression of global transcription factors. Two pathway-specific regula-tion genes, UvSorR1 and UvSorR2, as positive regulators, directly regulated sorbicillinoid biosynthesis at the transcriptional level. These findings will provide useful information, in order to better understand the complex and multi-level regulatory network regarding fungal development and sorbicillinoid biosynthesis in U. virens, as well as be advantageous in developing effective strategies for controlling fungal diseases and mycotoxin contamination. Other environmental factors, such as the incubation temperature, shaking, aeration, redox status, and metal ions, along with their regulation mechanisms, should be further studied. Investigating gene deletions and overexpressions by phenotypic analysis of relevant regulators to reveal the mechanisms of regulation, also merit further investigation.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/jof9040390/s1, Figure S1: Effects of ambient pH on the production of main sorbicillinoids in U. virens; Figure S2: Comparisons of the colony extension, mycelial growth and sporulation of wild-type strain P1 and gene deletion mutants of U. virens cultured on PDA medium at 28 • C for 18 days; Figure S3. Expression of global regulator genes in U. virens grown on different carbon and nitrogen sources, ambient pH, and light treatments; Table S1: Fungal strains and plasmids used in this study; Table S2: Primers used in this study; Table S3: Effects of carbon sources on the contents and yields of main sorbicillinoids in U. virens; Table S4: Effects of nitrogen sources on the contents and yields of main sorbicillinoids in U. virens; Table S5: Effects of ambient pH on contents and yields of main sorbicillinoids in U. virens; Table S6: Effects of light exposure on contents and yields of main sorbicillinoids in U. virens. The references [57,58,71] cited in Table S1