A. alternata P3 and the mutant strains were cultured on potato dextrose agar (PDA) at 25°C. The spores were then washed with 0.1% Tween-80 water, collected from the culture of A. alternata, and diluted to 1 × 10⁵ spores/ml (Jiang et al., 2019; Wang et al., 2019). The spore suspension was cultured on PDA at 25°C for 10 days to monitor hyphal extension, colony morphology, and conidial formation (Wang et al., 2019). The spores were inoculated and incubated on potato dextrose broth (PDB) under dark, white, blue, and red-light conditions at 25°C for 10 days for the following analysis on mycotoxin determination and gene expression. For the continuous light experiment, white, blue, or red LED lamps (18 W/lamp) were used, and the distance between the light source and the medium was 30 cm. The average illumination intensities of white, blue, and red light were separately measured at 5200, 4350, and 2,270 lux, respectively.

The deletion of AaVeA was used to investigate the functional roles for the growth, development, and secondary metabolite biosynthesis. AaVeA was disrupted by homologous recombination using the hygromycin resistance gene as a marker gene for antibiotic resistance screening (Supplementary Figure S1). Flanking sequences of 1,458-bp upstream and 1,485-bp downstream were separately amplified by polymerase chain reaction (PCR) using genomic DNA of A. alternata P3 as a template with the primer pairs AaVeA-up-F/R and AaVeA-down-F/R (Supplementary Table S1). The hygromycin resistance gene encoding hygromycin phosphotransferase (hph) was amplified using the primer pair hph-F/R. The reaction mixture (50 μl) consisted of 2 μl of the template DNA, 1 μl of each forward and backward primer (10 μM), 25 μl of GoTaq Master Mix (Promega, Madison, WI, United States), and 21 μl of nuclease-free water. The PCR program was set as follows: initial denaturation at 95°C for 5 min, 35 cycles of the amplification at 95°C for 30 s, 54°C for 30 s and 72°C for 90 s, and finally extension for 5 min. The PCR product was extracted using a Gel Extraction Kit (Sangon Biotech, Shanghai, China). Three fragments were assembled by overlapping extension PCR in combination with nested PCR using the primer pair AaVeA-knock-F/R. The assembled fragment was also purified for transformation. The protoplast of A. alternata was obtained by enzymatic hydrolysis of fungal cell wall with 30 mg/ml of snailase, 5 mg/ml of driselase, 2 mg/ml of lyticase, 8 mg/ml of β-D-dextranase, and 20 mg/ml of cellulase. Then, the PCR product was transformed into the protoplast by polyethylene glycol (PEG)-CaCl₂ mediated genetic transformation. The transformants were screened by 100 μg/ml hygromycin B (Roche Diagnostics, Mannheim, Germany) and identified by PCR and quantitative PCR according to the method of Zhang et al. (2016). The mutant of AaVeA deletion was finally obtained for the following study (Supplementary Figure S1).

The spore suspension of A. alternata P3 and the mutant was acquired (1 × 10⁵ spores/ml). A 5-μl aliquot of the suspension was inoculated in a PDA medium and grown for 10 days at 25°C. The characteristics of the colony were observed, and then the conidiation was compared between the wild type and ΔAaVeA mutant under white, red, blue, and dark conditions. The spore counts were determined using the hemocytometer after they were washed completely from the culture using 0.1% Tween-80.

The microscopical structure was evaluated by scanning electron microscopy (SEM). The samples were harvested and fixed in 2.5% glutaraldehyde. Then, the samples were washed, dehydrated, and vacuum-dried following the processes by Wang et al. (2019).

The 1%(v/v) spore suspension of A. alternata P3 and the AaVeA null mutant were inoculated and cultured in PDB. The strains were incubated for 10 days at 25°C with or without various light treatments. The supernatant was centrifuged and used to determine the concentration of mycotoxin. The determination of mycotoxin production was assayed under the direction of Jiang et al. (2019) using LC–MS/MS.

The mycelial samples were collected after culture under various light conditions and ground with liquid nitrogen. Total RNA was extracted from the mycelia with three biological replications cultivated in PDB medium with or without light using TRIzol reagent according to the Cold Spring Harbor Laboratory (Rio et al., 2010). The quality and purity of the extracted RNA were evaluated using the Bioanalyzer 2,100 and RNA 6000 Nano LabChip Kit (Agilent, CA, United States) with RIN number > 7.0. cDNA library construction and RNA-seq were carried out by Hangzhou Lianchuan Biotechnology Co., Ltd. (Hangzhou, China). Using the Illumina paired-end RNA-seq approach, the transcriptome was sequenced, and millions of paired-end reads were generated. After removing the low-quality reads, we mapped the reads to the reference genome of A. alternata using the HISAT package (Kim et al., 2015). HISAT allows multiple alignments per read (up to 20 by default) and a maximum of two mismatches when mapping the reads to the reference. The remaining reads, after trimmed from the raw reads, were then bioinformatically analyzed based on the genome annotation of A. alternata (genome assembly: GCF_001642055.1; Dang et al., 2015).

Differentially expressed genes (DEGs) were selected with a |log2 (fold change) | > 1 and statistical significance of value of p < 0.05 by the R package. Gene Ontology (GO) functional annotation (Young et al., 2010) and Kyoto Encyclopedia of Genes and Genomes (KEGG; Kanehisa et al., 2008) pathway enrichment analysis were conducted with the DEGs to further understand their functions.

The relative expression levels of TeA biosynthetic genes and velvet genes were further confirmed and quantified by reverse transcription quantitative real-time PCR (RT-qPCR). Total RNA was separately extracted by the EasyPure Plant RNA Kit (TransGen Biotech, Beijing, China) containing DNase I digestion for the genomic DNA residue. The RNA samples were diluted to an equal concentration, and then cDNA was reverse transcribed by EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China). Quantitative PCR was carried out by TransStart Top Green qPCR SuperMix (+Dye I), using cDNA as the template. The 20-μL reaction mixture comprised the following: 2 μl of cDNA, 0.4 μl of forward and backward primer (10 μM; Supplementary Table S2), 10 of 2 × TransStart Top Green qPCR SuperMix (+Dye I), and 7.2 μl of nuclease-free water. The transcriptomic level of clustered biosynthetic and regulatory genes was separately normalized to the expression of β-tubulin and evaluated by the 2^-ΔΔCt calculation from the results of StepOnePlus Real-Time PCR Systems (Applied Biosystems, Foster City, CA, United States).

To evaluate the pathogenicity and mycelial extension in vivo for the wild and VeA-disrupted strains of A. alternata, an equal volume (5 μl) of conidial suspension (10⁵/ml) was separately inoculated into the man-made wound on the surface of tomatoes after they were sterilized with 10% sodium hypochlorite and 75% alcohol. Tomatoes were treated with or without light at 25°C for 10 days.

Fungal growth was measured, and the infected area on tomatoes was used to extract and determine mycotoxin production. The samples were first dried by vacuum freezing and drying technology. Then, they were weighed and ground into powder, of which 0.2 g was used to determine the mycotoxin content. The final mycotoxin concentration was expressed as mycotoxin content per unit area.

All the experiments were carried out with at least three replicates. The data of fungal extension diameters, spore counts, and mycotoxins concentrations were analyzed by IBM SPSS statistics 23.0 (IBM Inc., Armonk, NY, United States) and compared by the one-way analysis of variance (ANOVA) using Tukey’s test. Significance was considered as p < 0.05. The figures were generated by Graphpad Prism 8.02 (GraphPad Software Inc., San Diego, CA, United States).

To understand the effects of the AaVeA gene on fungal growth and conidiation, the wild type and AaVeA knockout strains were used to characterize the properties with white, red, or blue light or without light. The wild type and the ΔAaVeA mutant strain were separately grown on PDA for 10 days. The colonies and the microscopic morphology are shown in Figure 1. Light seemed to hardly influence the mycelial extension of A. alternata from the colonies with light illumination (Figure 2A). However, lacking the gene AaVeA resulted in the inhibition of hyphal growth by 24.2% in the ΔAaVeA mutant under dark conditions (p < 0.001). Similarly, the mycelia of the mutant extended 25.5%, 23.0%, and 18.9% more slowly than the wild type with white, red, or blue light, respectively. There was no significant difference in the inhibition rate under the different light conditions due to the gene deletion (p > 0.05).

Although no obvious difference and a similar trend was found on the hyphal extension, the conidiation of A. alternata was markedly changed under different conditions (Figures 1B, 2B). White light illumination caused a 97.2% decrease in spores (p < 0.001). Red light stimulated conidiation, while blue light had a negative effect on conidial formation. Compared to the dark condition, red light recovered about 68.7% of conidial formation, but under blue light, the formation of conidia was repressed by 99.8%. Different from mycelial extension, the disruption of AaVeA led to a significant reduction in conidia under both dark and light conditions. Interestingly, unlike the wild strain, red light hardly stimulated asexual sporulation when the AaVeA gene was inactivated. Blue light also resulted in the lack of sporulation in both wild and ΔAaVeA strains.

In general, light hardly affected fungal growth, while red light favored spore formation and blue light did the opposite.

To evaluate the AaVeA roles in TeA biosynthesis, mycotoxin extraction and quantitation were performed, and the changes in mycotoxin production were compared between the wild-type and ΔAaVeA strains under various light sources. The TeA concentration reached up to 48.8 μg/ml in darkness, and no significant change was observed with white light illumination (Figure 3A). Furthermore, mycotoxin production also seemed no more marked fluctuation under red and blue light conditions. TeA production became much weaker after the gene, AaVeA, was disrupted under dark or light conditions. Similarly, no obviously positive or negative effect on TeA biosynthesis was detected in the mutant strain in response to various light sources (p > 0.05).

In a previous study, light promoted AOH and AME production (Pruß et al., 2014; Igbalajobi et al., 2019), which was also confirmed in this study (Figures 3B,C), although it could not result in more accumulation in TeA biosynthesis. AOH and AME concentrations were extremely stimulated by light illumination, especially blue light. AOH and AME contents were separately promoted 2.08- and 1.26-times with white light higher than its production in the dark (p < 0.001). Furthermore, blue light exposure caused much more AOH and AME accumulation while red light had no positive stimulation. When AaVeA was deleted, AOH and AME production was significantly repressed by 96.4% and 100% in the dark (p < 0.001). Light could no longer stimulate more abundant AOH and AME biosynthesis, even blue light.

To uncover the regulatory roles of AaVeA, the detailed transcriptomic profile of A. alternata with or without light illumination was separately analyzed and shown in Supplementary Tables S3 and S4. In total, gene expression at the transcriptional level exhibited the most dramatic changes from the number of DEGs in wild type strain compared to the ΔAaVeA mutant grown in the dark (Supplementary Figure S2). There were 1,496 DEGs, including 614 upregulated and 882 downregulated DEGs. Nevertheless, the comparative transcriptomic profile of the mutant had fewer fluctuations between the dark and white light conditions, as there were 660 DEGs in total (350 upregulated and 310 downregulated genes). This demonstrated that AaVeA is a critically important regulator for the physiological-biochemical processes in A. alternata.

From the analysis of GO enrichment, some biological processes, including conidium formation, pathogenicity, and secondary metabolite biosynthetic process, were distinctly enriched from the comparatively expressed DEGs (Figure 4). The biological process related to conidial formation and development was enriched from the DEGs between the wild and AaVeA null strains under white light and dark conditions (Figure 5). The secondary metabolite biosynthetic process was markedly influenced by AaVeA regulation. The biosynthesis of mycotoxins, including TeA, AOH, and AME, were affected to various degrees under light regulation via the regulator.

Many DEGs partly enriched in the process of hyphal growth (Figure 5). This explains why the mycelial extension became slower when AaVeA was disrupted in A. alternata. In addition, the DEGs responsible for pathogenesis were enriched and hence AaVeA was probaly involved in the pathogenicity of fungal infection. To investigate whether AaVeA was responsible for pathogenicity, the trials were carried out in tomato infected by A. alternata with or without AaVeA.

From KEGG pathway enrichment analysis, the metabolism, biosynthesis, and degradation of amino acids, including valine, leucine, and isoleucine were found to be enriched (Supplementary Figure S3). The substrate of TeA biosynthesis comprises isoleucine and acetoacetyl-CoA. The fluctuation of gene expression related to amino acid metabolism could lead to the alteration of the supply of isoleucine. Acetoacetyl-CoA could be formed by carbohydrate metabolism, including fatty acid biosynthesis and degradation, which were also enriched in KEGG pathway enrichment analysis.

From the prediction of the secondary metabolite biosynthesis gene clusters, the TeA biosynthetic gene cluster contains two genes encoding a biosynthetic enzyme and a transporter. The enzymatic protein has a NRPS portion and a ketosynthase (KS) domain for the condensation and cyclization of the mycotoxin, which is similar to the homologous protein encoded by the TAS1 (MGG_07803) gene in Magnaporthe oryzae (Yun et al., 2015). Additionally, another TAS2 gene encoding a Zn(II)2-Cys6-type transcription factor harbored the neighboring TAS1 and was identified to be responsible for the regulation of TAS1 expression and mycotoxin biosynthesis (Yun et al., 2017). However, we searched for the homologous gene throughout the genome of A. alternata, but no matched protein was found. The transporter gene may be responsible for the export of TeA from the cytoplasm to the extracellular matrix. Generally, the trend of these two genes expression was similar to the production of the mycotoxin (Figure 6A). Accordingly, AOH/AME biosynthetic gene cluster has been identified, which consists of four biosynthetic enzymatic genes (pksI, omtI, moxI, sdrI, and doxI) and a transcriptional factor gene (aohR; Chooi et al., 2015; Wenderoth et al., 2019). Among them, pksI is responsible for AOH formation. AOH is then converted to AME by an O-methyltransferase encoded by omtI. Both the transcription of pksI and omtI became lower but the AaLaeA gene transcript had no downtrend in the AaVeA knockout strain (Figure 6B). This demonstrates that AaVeA could positively affect TeA biosynthesis by regulating the transcription of the clustered biosynthetic genes, but it did not modulate AaLaeA expression.

To validate the data of RNA-seq, RT-qPCR was carried out for the analysis of fungal gene expression (Figure 6C). The trend of the clustered genes and LaeA transcription was similar to the analysis from RNA-seq. In addition, the red and blue light treatments had similar effects in the wild type and ΔAaVeA strains.

The wild and ΔAaVeA mutant strains of A. alternata were separately used to inoculate and colonize the surface of sterilized tomatoes. For 10 days culture, the disruption of AaVeA resulted in a reduction of 34.6% or 24.7% in lesion size under dark or light conditions, respectively (Figure 7). The hyphae of the mutant had weakness to be extending. Light seemed to have no significant influence on hyphal growth in the wild strain, which was consistent with the results of medium culture.

Light had no obvious influence as an activator or inhibitor for TeA biosynthesis. Accordingly, TeA production was obviously decreased by 98.3% in the ΔAaVeA strain grown in the dark (p < 0.001). A similar reduction occurred under white light illumination. However, light stimulated the accumulation of AOH and AME by 2.81- and 1.24-times more than that of mycotoxin production in the dark (p < 0.001). Moreover, AOH and AME contents were separately suppressed by 45.9% and 53.7%, respectively, due to the deletion of AaVeA under dark condition. Similarly, the loss of AaVeA led to a decrease in the AOH and AME concentration by 84.3% and 74.7% in ΔAaVeA with light illumination. This showed that AOH and AME biosynthesis was regulated by light via AaVeA.

In total, these results from the pathogenic infection on tomato in vivo were similar to the outcome in vitro, in which AaVeA positively played vital roles in hyphal growth, pathogenicity, and mycotoxin biosynthesis.