Analysis of TabZIP15 transcription factor from Trichoderma asperellum ACCC30536 and its function under pathogenic toxin stress

The TabZIP15 gene encoding a 396 amino acid (aa) polypeptide in the fungus Trichoderma asperellum ACCC30536 was cloned and characterised. The protein includes a basic region motif (NR-x2-QR-x2-R) and has a pillar-like structure. The 25 basic region/leucine zipper transcription factors (TFs) identified in the T. asperellum genome were divided into YAP (14 TFs), ATF2 (5), GCN4 (2), Zip1 (2), BRLZ (1) and u1 (1) subfamilies based on conserved domains. T. asperellum was cultured in minimal media (MM) control, C-Hungry and N-Hungry medium (to simulate nutrient competition and interaction with pathogens, respectively), and differential expression analysis showed that 14 TabZIP genes (including TabZIP15) were significantly altered under both conditions; TabZIP23 responded strongly to N-Hungry media and TabZIP24 responded strongly to C-Hungry media. However, only YAP genes TabZIP15, TabZIP12 and TabZIP2 were significantly upregulated under both conditions, and expression levels of TabZIP15 were highest. T. asperellum was also cultured in the presence of five fungal pathogenic toxins, and RT-qPCR results showed that TabZIP15 was significantly upregulated in four of the five toxin stress conditions (MM + Rhizoctonia solani, MM + Fusarium oxysporum, MM + Alternaria alternata and MM + Cytospora chrysosperma).

Scientific RepoRtS | (2020) 10:15084 | https://doi.org/10.1038/s41598-020-72226-w www.nature.com/scientificreports/ fungi, and a similar regulatory function can be inferred in other fungal species. However, most of the research on fungal bZIP proteins has focused on pathogens, which bZIP proteins in biological control agents such as Trichoderma spp. have received little attention. Phytopathogen toxins are metabolites in pathogens that can harm plants and other fungi. To better understand the detoxification mechanism of Trichoderma that is triggered in response to phytopathogen toxins, the roles of bZIP TFs should be investigated under different stress conditions. Herein, the TabZIP15 TF gene of Trichoderma asperellum ACCC30536 was cloned, and characterised alongside another 24 similar bZIP TFs in the Joint Genome Institute (JGI) database. The basic biochemical characteristics of these bZIP TFs were investigated, and expression levels were analysed based on RT-PCR data obtained for T. asperellum under three conditions at 72 h. Furthermore, transcription levels of TabZIP15 were measured by RT-qPCR following exposure to toxins produced by five pathogenic fungi (Rhizoctonia solani, Fusarium oxysporum, Sclerotinia sclerotiorum, Alternaria alternata, and Cytospora chrysosperma). The results provide theoretical support for the analysis and development of TFs from T. asperellum.

Materials and methods
Strains and materials. T. asperellum ACCC30536 was obtained from the Agricultural Culture Collection of China. The five phytopathogens were R. solani, F. oxysporum, S. sclerotiorum, A. alternata and C. chrysosperma, and they were stored at the Laboratory of Forestry Protection, Northeast Forestry University, Harbin, China. T. asperellum was cultured on potato dextrose agar (PDA) slant culture medium at 28 °C for 7 days and stored at 4 °C. A. alternata, S. sclerotiorum, R. solani, C. chrysosperma and F. oxysporum were inoculated in 200 mL of 1/4, 1, 1/2, 1 and 1/2 strength PD medium and cultured in shake flasks for 10 days at 28 °C with shaking at 200 rpm. The medium was filtered using 0.2 μm filters (Pall Corporation, MI, USA) to remove spores, and the filtered fermentation liquid from each pathogen was combined, mixed, and stored in tubes at − 20 °C.

Cloning and analysis of TabZIP15 transcription factor from T. asperellum. Primers for cloning
TabZIP15 (TabZIP15-1, 5′-GCG AAT CCG GAT GAG TTC AC-3′; TabZIP15-2, 5′-TCG CCG CAC CCT ATA CTT TT-3′) were designed using Primer Premier 6.0 software (PREMIER Biosoft, Vancouver, Canada) 16 . Extraction of T. asperellum DNA was performed as described previously 17 , and thermal cycling for PCR was performed at 94 °C for 5 min, followed by 35 cycles at 94 °C for 25 s, 56 °C for 30 s, and 72 °C for 30 s, and followed by 72 °C for 7 min. The PCR product was purified and ligated into the pGEM-T vector (A3600; Promega, Madison, USA) and sequenced (Shanghai Sangon Co., Shanghai, China). Conserved domain prediction and identification of other homologous proteins was performed using the Blastp tool 18 from the National Center for Biotechnology Information (NCBI). Multiple sequence alignment was conducted using the Clustal X program (https ://www. ebi.ac.uk/Tools /clust alw2/) 19 , and three-dimensional structure prediction was carried out by SWISS-MODEL ( https ://swiss model .expas y.org/) 20 .
Characteristic analysis and construction of phylogenetic tree of 25 bZIPs in the T. asperellum genome. Full-length T. asperellum bZIP protein (TabZIP protein) sequences were obtained from the T. asperellum CBS 433.97 v1.0 JGI database (https ://genom e.jgi.doe.gov/) 21 , and conserved domains were predicted using the NCBI tool Blastp. Proteins without conserved bZIP domains were omitted. The molecular weight (MW) and isoelectric point (pI) of each bZIP protein was calculated using ExPASy (https ://www.expas y.org/) 22 , and gene structural information for exons and introns was obtained from the JGI database (https ://genom e.jgi. doe.gov/) 21 and analysed by the Gene Structure Display Server 2.0 website (https ://gsds.cbi.pku.edu.cn/index .php) 23 . A phylogenetic tree was constructed via the maximum likelihood method with 1,000 bootstrap replicates using the MEGA 7.0 program 24 . Motifs were searched using the MEME suite (http://meme-suite .org/tools /meme) 25 with the site distribution set as 'zoops' , and the number of motifs set as '3' . Conserved domains and motifs were analysed by the MEME suite and TBtools 26 . TabZIP15  MM + F comprised MM plus 5% (v/v) fermentation liquid from F. oxysporum, C-Hungry was MM without a carbon source, and N-hungry was MM without a nitrogen source. Differential expression of 25 bZIP family genes was analysed using T. asperellum ACCC30536 transcriptome data obtained previously 27 , and the heatmap was drawn by TBtools 26 . Expression of TabZIP15 genes was analysed by RT-qPCR under five pathogenic toxin stress conditions. Spores from T. asperellum were inoculated into 200 mL 1/4 strength PD broth medium at a final concentration 1 × 10 4 spores/mL and cultured at 28 °C with continuous shaking at 200 rpm for 48 h. The resulting mycelia were filtered, washed, transferred to MM for 2 h, then transferred into medium containing different pathogenic toxins for 72 h, three replicates were set. Mycelia were collected at 0, 4, 8, 12, 24, 48, and 72 h. At each time point, the biomass was calculated (averaged from three replicates), and 25 mL mycelium cultures were collected and stored at − 80 °C. Total RNA was extracted from mycelia 28 using TRIzol reagent (Invitrogen, Carlsbad, USA), digested with DNaseI (Promega, Madison, USA), and reverse-transcribed into cDNA using a PrimeScript RT Kit (TaKaRa, Dalian, China) according to the manufacturer's instructions.

Differential expression of bZIP family genes and expression of
Expression levels of TabZIP15 in T. asperellum were measured by RT-qPCR, and calculated according to the 2 −ΔΔCt method 29 using cDNA as template, with actin, α-tubulin and β-tubulin as reference genes. Three RT-qPCR replicates were performed per cDNA sample. Primers for RT-qPCR (Table 1)  Statistical analysis. All experimental data were subjected to analysis of variance (ANOVA) and independent sample t tests using SPSS19.0 31 . Statistical significance of differences between groups was determined by Turkey's multiple range tests (p < 0.05).

Results
Characteristics of 25 TabZIP TFs in the T. asperellum genome. The 25 TabZIP genes encode proteins ranging from 177 to 753 aa in length, with an average length of 408 aa. The molecular weight of TabZIP proteins ranges from 19.63 to 80.93 kDa, with an average of 44.7 kDa. The pI ranges from 4.66 to 10.05, and the average pI is 6.65. TabZIP TF genes in T. asperellum are distributed on scaffold 1 to scaffold 41, and some chromosomes have more than one TabZIP TF gene (four, three and three TFs are in scaffold 5, scaffold 2 and scaffold 9, respectively; Table 2). The number of exons in the 25 TabZIP genes ranges from one to six, with most having two or three exons. Eight genes possess untranslated exon regions (Fig. 1). Table 1. Primers for RT-qPCR.

Gene name Primers Sequences (5′-3′) Tm (°C) Product size (bp)
TabZIP15 Ta-bZIP 15-L TGA CCC TGA CGC TCC TTT ACCT  59.1  230  Ta-bZIP 15-R GAG GCA TGA GAC GCT CCA TACG  between the 25 TabZIP proteins were explored, and sequences were divided into seven clades (Fig. 2a). The number of conserved domains in TabZIP TFs ranges from one to four (Fig. 2b) HMN]). We found that 10 bZIP TFs have two motifs, but interestingly, none of the bZIP TFs have three motifs and bZIP TFs in the same clade have similar numbers and types of motifs (Fig. 2).

Differential expression of 25 TabZIPs in the three transcriptomes. Expression of the 25 TabZIP
genes in 3 transcriptomes was measured ( Fig. 3a; Supplementary Table S1). Interestingly, 13, 14 and 11 TabZIP genes were highly expressed (Log 2 RPKM > 6, Reads Per Kilobase per Million mapped reads) in MM, C-Hungry and N-Hungry media, respectively. Genes were clustered based on their expression, and TabZIP15, TabZIP13 and TabZIP23 were highly expressed in all three conditions, while TabZIP18, TabZIP24 and TabZIP 25 were highly expressed in MM and N-Hungry media, and TabZIP6 and TabZIP20 were highly expressed in MM and C-Hungry media.
In the differential expression analysis, a two-fold (Log 2 value > 1) change in expression level relative to MM was considered significant. The results showed that six TabZIP genes were significantly upregulated and six were significantly downregulated in C-Hungry media (Fig. 3b). Meanwhile, four TabZIP genes were significantly upregulated and 11 were significantly downregulated in N-Hungry media (Fig. 3b). Additionally, 14 genes were significantly altered in N-Hungry and C-Hungry media compared with MM media, but only three genes (TabZIP15, TabZIP12 and TabZIP2) were significantly upregulated in both N-Hungry and C-Hungry media, while TabZIP6, TabZIP11, TabZIP20 and TabZIP25 were the only genes significantly downregulated in these two conditions. During the stress response process, upregulated genes may act as positive response factors. Thus, we focused on TabZIP15, TabZIP12 and TabZIP2, and TabZIP15 displayed the highest expression, hence this gene was subjected to further analysis.
Characteristics of TabZip15 TF in T. asperellum. The TabZIP15 DNA sequence (2062 bp) includes six exons and five introns. The TabZIP15 cDNA (1,188 bp) encodes a 396 aa polypeptide with a calculated molecular weight of 45.05 kDa, and a pI of 5.54. TabZIP15 is a member of the bZIP-YAP family of TFs (Fig. 4a). Multiple sequence alignment revealed a highly conserved basic region motif (NR-x2-QR-x2-R), and three conserved hydrophobic amino acids in the sequence following the basic region (Fig. 4b). Three-dimensional structure prediction showed that the TabZIP15 protein is a cylindrical bundle of α-helices (Fig. 4c).   -fold), respectively. Analysis of differential expression (relative to controls) showed that TabZIP15 was highly expressed when T. asperellum was exposed to toxins from A. alternata, C. chrysosperma, F. oxysporum and R. solani, but expression in MM + S media did not differ significantly from that in MM media.

Discussion
Members of the Trichoderma genus have been widely used as biological control agents, and the biocontrol mechanism has been investigated 3 . bZIP proteins regulate various biological processes [10][11][12][13] and are present in all eukaryotes 5 . Thus, bZIP TFs may regulate biological control processes in T. asperellum. Herein, 25 TabZIP TFs were identified in T. asperellum, and multiple sequence alignment revealed relatively low sequence similarity (data not shown). Seventeen sequences sharing homology with TabZIP15 were identified using Blastp (Fig. 4), and multiple sequence alignment showed that these sequences share short conserved regions (~ 30 aa), as expected for bZIP proteins that typically include 18 conserved residues followed by a leucine zipper 32,33 . Thus, the sequences of bZIP TFs are diverse, and this may explain why they perform diverse functions 33 . The bZIP TFs in T. asperellum were divided into seven clades, with TabZIP15, TabZIP9 and TabZIP10 in the same clade (Fig. 2a). The close phylogenetic relationship indicates that these three proteins may perform similar functions and share similar expression profiles. Consistently, motif prediction showed that TabZIP15, TabZIP19 and TabZIP21 are www.nature.com/scientificreports/ the only proteins possessing motif 3, and they share a similar motif arrangement (Fig. 2c). This further suggests that TabZIP15, TabZIP19 and TabZIP21 may perform similar functions or share similar expression profiles. Next, we explored the expression of these 25 TabZIP genes in C-Hungry and N-Hungry media using an RNA sequencing (RNA-Seq), which simulated Trichoderma nutrient competition and interaction with the pathogens 27 , TabZIP19 and TabZIP21 have similar motif arrangements to TabZIP15, hence they may be expected to share similar expression modes, but the RNA-Seq data showed that they did not have similar expression modes. Similarly, TabZIP9 and TabZIP10 are in the same clade as TabZIP15, and they share similar phylogenetic relationships, hence they may be expected to share similar expression modes, but again, RNA-Seq data showed that they did not have similar expression modes (Fig. 3).
The YAP protein and its homologs have been analysed in several fungal pathogens, and they all perform similar functions in stress response 9,15,34 . In T. asperellum, 14 of the 25 bZIP TFs have a conserved bZIP-YAP domain (Fig. 2b). The presence of a large number of YAP bZIPs may explain the robust ability of T. asperellum to adapt to environmental changes, but their exact functions need to be further investigated. In Claviceps purpurea, the transcription factor CPTF1, a homolog of ATF in mammals, is related to oxygen stress responses 35 . In T. asperellum, five bZIP proteins (TabZIP4, TabZIP7, TabZIP8, TabZIP13 and TabZIP18) have a conserved bZIP-ATF2 domain, and they may perform similar functions, but this should be further investigated (Fig. 2b). The TabZIP18 protein includes Aft1-HRR, Aft1-HRA and Aft1-OSA domains, and is a homolog of UvbZIP8 9 , AtfB 36 and Aft1 37 . Aft1 proteins are reportedly involved in the regulation of sexual development and various stress responses, and they activate meiotic recombination, which suggests that TabZIP18 may also play a key role in these processes.
Our transcriptome data showed that 16 of the 25 bZIP TF genes were significantly differentially expressed in C-Hungry medium (simulating competition with pathogens) or N-Hungry medium (simulating interaction with pathogens; Fig. 2b). Of these, 15 genes appear to be involved in nutrient competition and 15 appear to be related to interactions with pathogens. The 15 nutrient competition-related genes include nine YAP, four ATF2, one Zip1 and one u1 genes, while the 15 pathogen interaction-related genes include eight YAP, four ATF2, one Zip1, one u1 and one BRLZ genes. These results suggest that YAP and ATF2 proteins in T. asperellum may be involved in various stress responses, similar to other homologs 9,15,34-37 . One BRLZ gene was significantly downregulated in N-Hungry media (interaction with pathogens), and a homolog of this protein is reportedly involved in fungal development and differentiation 38 . Meanwhile one u1 and one Zip1 were significantly downregulated in both N-Hungry and C-Hungry media. A homolog of Zip1 was reportedly induced under stress conditions 39 , but u1 proteins have not been investigated, and further studies are needed. Thirteen genes were highly expressed in MM media (Log 2 RPKM > 6), and they may be involved in normal physiological regulation (Fig. 2a). Three YAP genes (TabZIP15, TabZIP12 and TabZIP2) were significantly upregulated in both C-Hungry and N-Hungry media (simulating nutrient competition and interactions with pathogens, respectively), which suggests that they may be important for resistance against pathogens. TabZIP12 and TabZIP2 were both significantly upregulated. TabZIP12 expression levels were upregulated 2 2.31 -and 2 1.85 -fold in C-Hungry and N-Hungry media respectively, while TabZIP2 expression levels were upregulated 2 2.14 -and 2 1.58 -fold in C-Hungry and N-Hungry media respectively. However, the RPKM value for TabZIP12 in MM and C-Hungry media was 45 and 224, respectively; compared with 22 and 97 for TabZIP2. The RPKM value for TabZIP12 in MM and N-Hungry media was 45 and 163, respectively, compared with 22 and 66 for TabZIP2. Thus, the RPKM values for both genes were not high, and can be easily influenced by experimental error 40 , but fold-changes in expression were relatively large, hence further investigation is clearly needed.
The most highly expressed gene (TabZIP15) was chosen for further analysis, and its expression was measured in the presence of five pathogenic toxins by RT-qPCR. The results confirmed our hypothesis (Fig. 5); TabZIP15 was strongly upregulated in four of the five pathogenic toxin conditions, indicating an important function in resistance to pathogens. Interestingly, TabZIP15 was also highly expressed in MM media, which suggests that it may be involved in other functions beyond resistance to pathogens.
Herein, we identified 25 TabZIP genes in T. asperellum, which were divided into seven clades based on phylogenetic analysis, and six conserved bZIP domains were detected in the aa sequences. Analysis of the transcriptome data showed that expression of 16 TabZIP genes was significantly altered under C-Hungry and N-Hungry media conditions. The TabZIP15 gene was identified as a potential biocontrol factor, and its characteristics were further analysed. The RT-qPCR results showed that the YAP gene TabZIP15 was significantly upregulated in the presence of four of the five pathogenic toxins (those from A. alternata, C. chrysosperma, F. oxysporum and R. solani). Our results reveal some of the properties of bZIP TFs in T. asperellum, identify potential biological control-related bZIP TFs, and will guide further research in this area.

Data availability
All data generated or analysed during this study are included in this published article (and its Supplementary  Information files), if there is any other request, the datasets are available from the corresponding author on reasonable request.