Identification of a novel fungus, Trichoderma asperellum GDFS1009, and comprehensive evaluation of its biocontrol efficacy

Due to its efficient broad-spectrum antimicrobial activity, Trichoderma has been established as an internationally recognized biocontrol fungus. In this study, we found and identified a novel strain of Trichoderma asperellum, named GDFS1009. The mycelium of T. asperellum GDFS1009 exhibits a high growth rate, high sporulation capacity, and strong inhibitory effects against pathogens that cause cucumber fusarium wilt and corn stalk rot. T. asperellum GDFS1009 secretes chitinase, glucanase, and protease, which can degrade the cell walls of fungi and contribute to mycoparasitism. The secreted xylanases are good candidates for inducing plant resistance and enhancing plant immunity against pathogens. RNA sequencing (RNA-seq) and gas chromatography-mass spectrometry (GC-MS) showed that T. asperellum GDFS1009 produces primary metabolites that are precursors of antimicrobial compounds; it also produces a variety of antimicrobial secondary metabolites, including polyketides and alkanes. In addition, this study speculated the presence of six antimicrobial peptides via ultra-performance liquid chromatography quadrupole time of flight mass spectrometry (UPLC-QTOF-MS/MS). Future studies should focus on these antimicrobial metabolites for facilitating widespread application in the field of agricultural bio-control.


Introduction
Trichoderma spp. exhibits antagonistic effects against at least 18 genera and 29 species of pathogenic fungi, as well as a variety of pathogenic bacteria. The biocontrol mechanisms of Trichoderma spp. primarily include competition and mycoparasitism, followed by the stimulation of plant resistance and immunity [1].
Trichoderma spp. is highly adaptive to the environment, and their growth rates are generally much faster than those of plant pathogens. Trichoderma spp. could compete with nearby a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 T. asperellum GDFS1009 was cultured in potato-dextrose (PD) broth in a rotary shaker for 2 d at 28˚C and 180 rpm until its mycelium was well developed. The supernatant was removed by vacuum filtration, the mycelium was collected as the pellet, and the genome was extracted using the hexadecyl trimethyl ammonium bromide (CTAB) method [20]. Using universal ITS primers (ITS1-F and ITS4) [21], ITS sequences were amplified by PCR using the T. asperellum GDFS1009 genome as the template. The PCR product was ligated into the pMD19-T vector (Takara, Shiga, Japan) and was then sequenced by the Shanghai Biosune Biotechnology Co. (Shanghai, China). The sequencing results were aligned with the ITS sequences of known strains using NCBI BLAST (http://www.ncbi.nlm.nih.gov/BLAST). MEGA 4.0 software was used to build a phylogenetic tree for ITS [22].

Detection of biocontrol efficacy
Discs were obtained from 4-day-old cultures of pathogens (Fusarium oxysporum f. sp. cucumerinum Owen and Fusarium graminearum) and T. asperellum GDFS1009 using sterile 7 mm hole-punchers. Each disc was inverted and placed on one side of a 9-cm Petri dish with PDA medium, with 4 cm between pathogen and T. asperellum GDFS1009 discs. The control was a dish containing only the pathogen disc. Dishes were incubated at 28˚C for 5 d [23].
Then spore suspension (10 6 cfu/mL) of T. asperellum GDFS1009 was cultured in PD broth in a rotary shaker at 28˚C and 180 rpm for 6 d. A total of 50 mL of filtered fermentation liquor from T. asperellum GDFS1009 or ddH 2 O (control) were mixed into 100mL PDA medium, and the plate was made in 9-cm Petri dish. 7 mm discs of 4-day-old cultures of pathogens (Fusarium oxysporum f. sp. cucumerinum Owen and Fusarium graminearum) were inverted and placed on the center of each plate, then were incubated at 28˚C for 4 d. Additionally, related micro well dilution assay of pathogens inhibition was also done with filtered fermentation liquor from T. asperellum GDFS1009.
Cucumber (Jinyan no. 4) plants were grown to the 3-5-leaf stage in soil under a 10-h light/ 14-h dark cycle in a 28˚C phytotron. A 20-mL spore suspension of F. oxysporum f. sp. cucumerinum Owen (10 6 cfu/mL) was irrigated onto the roots of cucumber seedlings, followed by a 20-mL spore suspension of T. asperellum GDFS1009 (10 7 cfu/mL) in the treatment group. The control group was irrigated with sterile water instead of T. asperellum GDFS1009. The inoculated plants were placed in the 28˚C phytotron for at least 7 d, and biocontrol effects were observed and analyzed [24].
Assay of enzymes related to mycoparasitism and induced resistance T. asperellum GDFS1009 was cultured in PD broth in a rotary shaker at 28˚C and 180 rpm for 2 d until the mycelium was well developed. The supernatant was removed by vacuum filtration, and 1 g of mycelium pellet was transferred to xylanase induction medium includes 2% corncob powder, 0.4% NH4NO3, 0.05% K2HPO4, 0.05% MgSO4Á7H2O, 0.05% KCl, and 0.001% FeSO4Á7H2O. The activities of xylanase were measured using 3, 5-dinitrosalicylic acid (DNS). At 50˚C and pH 5.0, 1 U of xylanase activity was defined as the amount of enzyme required to release 1 μmol xylose per minute. Xylose was used as standard for xylanase analysis [25].
Then spore suspension (10 4 cfu/mL) of T. asperellum GDFS1009 was cultured in protease induction medium (1% casein, 0.05% K2HPO4, 0.05% MgSO4Á7H2O, 0.05% KCl, and 0.001% FeSO4Á7H2O) in a rotary shaker at 28˚C and 180 rpm for 5 d. The activities of protease were measured every day using foline-phenol reagent. At 50˚C and pH 7.5, 1 U of protease activity was defined as the amount of enzyme required to release 1 μg L-tyrosine per minute. L-tyrosine was used as standard for protease analysis. Enzyme activity was statistically analyzed via SPSS software. Each measurement was performed in triplicate. Furthermore, discs were obtained from 2-day-old cultures of T. asperellum GDFS1009 using sterile 3 mm hole-punchers. Each disc was inverted and placed on the center of a 9-cm Petri dish with 1.5% skim milk and 2.5% agar. After incubation at 28˚C for 36h, proteolytic circle was observed [26][27].
Tobacco (Nicotiana benthamiana) plants and cucumber plants were grown to the 6-8-leaf stage in soil under a 10-h light/14-h dark cycle in a 24˚C phytotron. A total of 20 μL of filtered xylanase preparation from T. asperellum GDFS1009 or blank broth (control) were injected into tobacco or cucumber leaves using 1 mL syringes. Injected leaves in each group were of the same size and position in the same tobacco or cucumber plant. The inoculated plant was placed in darkness in a 24˚C phytotron, and tissue necrosis was observed every 12 h. For oxidative burst assay, injected leaves were stained in DAB solution (pH 3.8) for 8 h in darkness, boiled in ethanol, glycerin, and acetic acid (1:1:1, v/v) for 10-15 min, and washed twice with 60% ethanol [28].

Global analysis of gene expression based on RNA-seq
The spore suspension of T. asperellum GDFS1009 was transferred into PD medium with a final spore concentration of 10 6 cfu/mL. After culturing at 28˚C and 180 rpm for 24 h, the mycelium was collected by vacuum filtration. Single-end (1×50) Solexa sequencing was performed, producing 20 million reads per sample. Additionally, 48-h samples were similarly sequenced, producing 10 million reads.
The reference genome for transcriptome analysis was acquired from http://genome.jgi.doe. gov/Trias1/Trias1.download.html. Conventional analysis was performed on the transcriptome data, including data pre-processing, genomic mapping, gene expression analysis, transcript expression analysis, alternative splicing analysis, analysis of differentially expressed genes, and GO/KEGG enrichment analysis of differentially expressed genes and non-differentially expressed genes [29][30][31]. On the basis of these results, genes related to mycoparasitism, induced resistance, and antibiosis in other Trichoderma strains were downloaded, and then local BLAST was used to compare them to the public T. asperellum genomic data mentioned above. Then all homologous genes were assembled and matched to the RNA-seq data of T. asperellum GDFS1009 to determine their expression levels. When the FPKM value exceeded 100, the expression level was defined as "high", while an FPKM value of less than 10 was defined as "low", and a value less than 1 was considered "ultra-low" expression. A heat map was constructed in the R language [32].

Assay of primary metabolites based on GC-MS
The mycelia of T. asperellum GDFS1009 were inoculated into a PDA plate at 28˚C for 4 d. Then, spores were collected in sterile water and diluted to 10 8 cfu/mL. Spores (1 mL) were added to 100 mL PD broth and incubated on a 180-rpm rotary shaker at 28˚C for 1 d. Next, 50 mL of this sample was dropped directly into 200 mL 60% (v/v) methanol solution (taken from a -40˚C deep freezer). The mixture was allowed to cool for 10 min to allow the temperature to return to -40˚C. The mixture was then centrifuged at 5000 rpm for 5 min at -10˚C, and the mycelium pellet was collected. Following this procedure, the mycelium pellet was freeze-dried at -40˚C.
Subsequently, 500 μL 100% methanol at -80˚C was added to 100 mg of sample, vortexed for 30 s, and placed in liquid nitrogen for 15 min. After thawing at 4˚C for 15 min and vortexing for 1 min, the mixture was centrifuged at 13000 rpm for 5 min at 4˚C, and the resulting supernatant was transferred to tube A. Milli-Q water (250 mL) was used to resuspend the pellet after centrifugation. The resuspension was then vortexed for 30 s and placed in liquid nitrogen for 15 min. After thawing at 4˚C for 15 min, glass beads (G8772-100G, Sigma, St. Louis, MO, USA) were added and vortexed for 10 min, and then the mixture was centrifuged at 13000 rpm for 5 min at 4˚C. The supernatant was also transferred to tube A. After drying with nitrogen, 60 μL methoxyl amine hydrochloride (15 mg/mL) was added to tube A and vortexed for 1 min, and then the mixture was allowed to react at 25˚C for 12 h. After adding 60 μL MSTFA with 1% TMCS and allowing the reaction occur at room temperature for 1 h, the mixture was centrifuged at 13000 rpm for 5 min at 4˚C.
The supernatant was then used for gas chromatography-mass spectrometry (GC-MS; Agilent 7890A/5975C, Agilent, Santa Clara, CA, USA). The derivatized samples (1 μL) were injected at a 1:20 split ratio onto a HP-5MS column (5% phenyl methyl silox; 30 m × 250 μm i. d., 0.25-μm; Agilent J&W Scientific, Folsom, CA, USA). During GC, helium was supplied at a constant rate of 1 mL/min. The temperature of the injection port was 280˚C, the ion source temperature was 250˚C, and the interface temperature was 150˚C. The temperature program started at 40˚C for 5 min and increased at a rate of 10˚C/min until it reached 300˚C, where it was held constant for 5 min. MS was determined using the full-scan method with a range from 35 to 780 m/z. Identification of metabolites was performed in six biological replicates. Bioinformatics analyses included data pre-processing (XCMS, www.bioconductor.org) and compound identification (NIST 2008, Wiley 9) [33][34][35].

Assay of secondary metabolites based on GC-MS
The mycelia of T. asperellum GDFS1009 were inoculated into PDA plate at 28˚C for 6 d. After conidia fully covered the plate, they were carefully scraped from the plate surface with a sterile spoon. Spores (6 g) were weighed and extracted with 50 volumes of dichloromethane at 4˚C for 2 d. The extract was exposed to 5% activated carbon under oscillation for 2 h and filtered with four layers of sterilized gauze. The supernatant was washed twice with an equal volume of 3% sodium carbonate solution. The resulting dichloromethane fraction was dehydrated with anhydrous sodium sulfate and then filtered with fast filter paper. Finally, the dehydrated liquid was vacuum-evaporated at 40˚C to obtain 1 mL of viscous crude extract [36]. GC-MS analysis was carried out using an Auto System XL GC/Turbo Mass MSUSA (Perkin Elmer, USA, DB-5MS column: 30 m×0.25 mm, 0.25 μm). The GC conditions were as follows: after injection of 1 μL, the initial column temperature was held at 50˚C for 5 min and then raised to 300˚C at 5˚C/min, and the vaporizing chamber temperature was then maintained at 300˚C. The GC was operated with a split ratio of 10:1 using helium as the carrier gas. The electron-ionization (EI) source was held at 230˚C and the ionization voltage was 70 eV. The quadrupole mass filter was run at an effective scanning range of 290-500 amu. Metabolites were qualitatively analyzed by GC-MS. The peak area of each metabolite, acquired in SIM mode (selective ion monitoring; signal/ noise > 5.0), was normalized to the peak area of 2-heptone prior to further data processing.
Then spore suspension (10 6 cfu/mL) of T. asperellum GDFS1009 was cultured in PD broth in a rotary shaker at 28˚C and 180 rpm for 6 d. A total of 20 mL of filtered fermentation liquor from T. asperellum GDFS1009 (PD broth as control) was enriched to 5 mL, extracted three times with 5mL dichloromethane, and was washed twice with an equal volume of 3% sodium carbonate solution. The resulting dichloromethane fraction was vacuum-evaporated at 40˚C to obtain 1 mL of viscous crude extract, and then was dehydrated with anhydrous sodium sulfate. Gas chromatography was performed on a HP-5MS capillary column (5% phenyl/95% methylpolysiloxane (30 m × 250 μm i.d., 0.25 μm film thickness, Agilent J & W Scientific, Folsom, CA, USA) to separate the derivatives at a constant flow of 1 mL/min helium. 1 μL of sample was injected in split mode in a 20:1 split ratio by the auto-sampler. Injection temperature was 280˚C, the interface set to 150˚C and the ion source adjusted to 230˚C. The programs of temperature-rise was followed by initial temperature of 80˚C for 5 min, 20˚C/min rate up to 300˚C and staying at 300˚C for 6 min. Mass spectrometry was determined by full-scan method with range from 35 to 500 (m/z). Identification of metabolites using the Automatic Mass Spectral Deconvolution and Identification System (AMIDS) was searched against commercial available databases such as National Institute of Standards and Technology (NIST) and Wiley libraries [33][34][35].

Assay of antimicrobial peptides based on UPLC-QTOF-MS/MS
Mycelia of T. asperellum GDFS1009 were inoculated onto a PDA plate at 28˚C for 4 d. Then, spores were collected with sterile water to 10 8 cfu/mL. Spores (1 mL) were added to 100 mL antimicrobial peptide media [0.5% glucose, 0.08% KH 2 PO 4 , 0.07% KNO 3 , 0.02% Ca(H 2 PO 4 ) 2 , 0.05% MgSO 4 .7H2O, 0.001% MnSO 4 .5H 2 O, 0.0005% CuSO 4 .5H 2 O, and 0.0001% FeSO 4 .7H2O], and the mixture was incubated on a 180-rpm rotary shaker at 28˚C for 20 d. The broth was filtered using two layers of qualitative filter paper, extracted with a filtrate:butanol ratio of 3:1, and the lower layer liquid was transferred to a vacuum distillation flask. After distillation, the dry powder was completely dissolved in 80 mL of methanol:dichloromethane at a ratio of 1:1, and then the dissolved liquid was filtered with a 0.45-μm PTFE membrane. After evaporating to dryness in a vacuum distillation flask, the solution was completely dissolved in 5 mL of dichloromethane:methanol at a ratio of 85:15 in a column that had been previously washed with 85:15 dichloromethane:methanol to remove acetone. This was followed by the addition of 100 mL of 85:15 dichloromethane:methanol. The collected filtrate was again placed in a vacuum distillation flask and evaporated to dryness. The dry powder was completely dissolved in 5 mL of methanol: H 2 O at a ratio of 85:15, transferred to a 1.5-mL centrifuge tube, and centrifuged at 13000 rpm for 10 min. After centrifugation, the supernatant was stored in a 2.0-mL glass vial at 4˚C [37].
UPLC-QTOF-MS/MS was carried out using an ACQUITY UPLC & SCIEX SelexION Triple Quad TM 5500 System (Waters, USA) equipped with an Acquity BEH C18 column (2.1 × 100 mm, 1.7 μm). The UPLC conditions were as follows: the mobile phase consisted of water (eluent A) and 0.1% acetonitrile (eluent B), the column temperature was 45˚C, the injection volume was 3.00 μL, and the running time was 26.50 min. MS parameters were as follows: the spectrometer was operated in the positive ion mode using an electrospray-ionization source, the capillary voltage was 3.0 kV, the sampling cone voltage was 4.0 V, the extraction cone voltage was 3.0 V, the ion source temperature was 105˚C, the atomization air temperature was 350˚C, the atomizing gas flow was 700.0 L/h, the collision voltage was 3.0 eV, the scan time was 0.300 s, the scan range was 200-2200 m/z, and LockSpray online correction was performed using 200 ng/mL leucine enkephalin (556.2771 m/z). MassLynx software was used for the alignment, fitting, identification, and analysis of different antimicrobial peptides.

Colony morphology, ITS identification, and phylogenetic tree
Morphological observations showed that after five days of culture, the front of the novel isolate's colony was dark green (spore color) and the back was white (mycelium color). The morphology of the mycelium was coarse, and dark green spores began to form in the center of the colony at high rates ( Fig 1A). Based on preliminary morphological observations, the strain was confirmed to be Trichoderma.
The result of the PCR amplification of the ITS (KF367522) is showed in Fig 1B, with a product size of 642 bp. NCBI BLAST followed by phylogenetic analysis of ITS sequences revealed that this novel strain shares the highest homology with T. asperellum G, followed by T. harzianum, T. longibrachiatum, T. hamatum, T. reesei, and T. virens, with Protocrea pallida used as an outgroup (Fig 1C). Together with the morphological identification, we determined the strain to be T. asperellum and named it GDFS1009.

Biocontrol effects under in vitro and greenhouse conditions
In vitro dual-culture analysis showed that the inhibition rate of T. asperellum GDFS1009 against F. oxysporum f. sp. cucumerinum Owen was 80.82% (Fig 2A). The inhibition rate of filtered fermentation liquor from T. asperellum GDFS1009 against F. oxysporum f. sp. cucumerinum Owen was 67.59% (Fig 2B). Micro well dilution assay on 2 days showed consistent results, up to 76.28% (Fig 2C). Under greenhouse conditions, T. asperellum GDFS1009 exhibited 86.34% growth inhibition (Fig 2D). Based on the in vitro antagonism assay and greenhouse study, T. asperellum GDFS1009 exhibits significant biocontrol effects on F. oxysporum f. sp. cucumerinum Owen. Additionally, except in the case of a low incidence of fusarium wilt, after treatment with Trichoderma spore suspension, cucumber seedling growth was more robust, with bright green leaves and strong roots (Fig 2D).
Comparing with F. oxysporum f. sp. cucumerinum Owen, in vitro dual-culture analysis showed that the inhibition rate of T. asperellum GDFS1009 against F. graminearum was not so high. However, the inhibition rate of filtered fermentation liquor from T. asperellum GDFS1009 against F. graminearum was 100%. Micro well dilution assay on 2 days showed consistent results, nearly 100% (S1 Fig). Comprehensive evaluation of biocontrol efficacy of T. asperellum

Mycoparasitism-related enzyme gene expression and activity
The relative expression levels of mycoparasitism genes were analyzed using RNA-seq analysis and illustrated with a heat map. Local BLAST identified 16 chitinase genes in the T. asperellum GDFS1009 genome and eight protease genes, with two being highly expressed after 24 h and one after 48 h. A total of 11 glucanase genes were discovered, with only one of these being highly expressed after 24 h and 48 h. Most of the non-highly expressed genes mentioned above were expressed at low levels, and a few did not appear to be expressed under the experimental conditions (Fig 3A and S1 Table).
After incubation at 28˚C for 36h, proteolytic circle in skim milk plate was 1.656 ± 0.030 cm (Fig 3B). Activity of mycoparasitism-related protease was assessed in corresponding induction broth. And protease activity was highest after fermentation for 2 d, reaching 1.418 ± 0.023 U/ mL at 50˚C and pH 7.5 (Fig 3C). In addition, chitinase and glucanase were also discovered with high level expression in induction broth (Data not shown).

Induced resistance-related gene expression and induction of tissue necrosis and resistance in plants
In the T. asperellum GDFS1009 genome, 12 elicitor genes with different functions were found, including two endopolygalacturonase genes, two Epl protein genes, two hydrophobin genes, one polygalacturonase gene, one swollenin gene, and four xylanase genes. Most of the elicitor genes were expressed at very low levels or not at all, with only three being expressed highly in PD broth after 24 h and 48 h (Fig 4A and S2 Table). The xylanase activity of T. asperellum GDFS1009 was assessed in induction broth, and exhibiting a maximum activity of 11.477 ± 0.613 U/mL at 50˚C and pH 5.0 after fermentation for 3 d (Fig 4B).
At 12 h after inoculation, necrotic lesions were seen on tobacco and cucumber leaves injected with a xylanase solution from T. asperellum GDFS1009, but no necrotic lesions were seen on leaves treated with control broth. This result was most apparent at 24 h. In addition, sites both near and far from the inoculation sites of tobacco and cucumber leaves injected with xylanase solution were stained by DAB, while those in tobacco and cucumber leaves injected with control broth exhibited nearly no staining (Fig 4C and 4D).

Biocontrol-related primary metabolites determined by GC-MS analysis
Using GC-MS analysis, nine were related to biocontrol as fungicides or a precursor/ intermediate of a fungicide, insecticide, or herbicide. Acetamide and diethylamine are fungicide precursors. Ethylene Glycol, ethanolamine, and o-toluic acid are fungicide intermediates. Citric acid is a fungicide. Ethylamine is a precursor of triazine herbicides, and glycine is a precursor of fungicides, insecticides, and herbicides (Table 1).
Local BLAST revealed 16 gene clusters of polyketide synthases (PKSs) in the T. asperellum GDFS1009 genome, among which 12 clusters were fully reducing PKSs and four were nonreducing PKSs. Based on the analysis of RNA-seq data, after 24 h and 48 h in PD broth, almost all of the gene clusters exhibited low levels of expression or were not expressed ( Table 2).

Antimicrobial peptides determined by UPLC-QTOF-MS/MS analysis and gene cluster prediction
Through UPLC-QTOF-MS/MS analysis, a total of six types of antimicrobial peptides were speculated from T. asperellum GDFS1009 with molecular weights ranging from 1092.6450 to 1734.0618. These were TBV (11 aa), trichotoxin_A-50_F (18 aa), hypomurocin_B_IIIa (18 aa), trichotoxin_A-50_I (18 aa), hypomurocin_B_I (18 aa), and hypomurocin_B_II (18 aa; Table 3). These antimicrobial peptides are formed through the catalysis of non-ribosomal peptide synthases (NRPS). Local BLAST revealed 16 NRPS gene clusters in the T. asperellum GDFS1009 genome, with two PKS/NRPS hetero-gene clusters. Based on the analysis of RNAseq data, after 24 h and 48 h in PD broth, almost all of the gene clusters were lowly expressed or not expressed (Table 2).

Discussion
In recent years, the United States, Canada, Australia, Poland, and many other countries have participated in collecting Trichoderma resources. At present, 141 Trichoderma species have been discovered in the world [38][39][40][41]. Early studies on Trichoderma mostly focused on aspects of soil microbiology and general mycology. As our understanding of Trichoderma has deepened, research into the mechanism of biocontrol and application has gradually increased.
This study began with the identification and biocontrol assessment of a novel Trichoderma, named T. asperellum GDFS1009, which exhibits strong control effects on F. graminearum and F. oxysporum (Figs 1 and 2; S1 Fig). As is typical of soil-borne diseases, F. graminearum and F. oxysporum seriously harm vegetables and food crops. The biocontrol data from in vitro and greenhouse experiments provide an experimental basis for further investigations into spore granule production for use in agricultural settings. Comprehensive evaluation of biocontrol efficacy of T. asperellum Previously, research into the resistance mechanisms of Trichoderma spp. have been fairly broad, with most concentrated on a single or a few biocontrol factors in a single strain. Our study reveals the global resistance mechanisms of T. asperellum GDFS1009 at the molecular level, including almost all biocontrol factors related to mycoparasitism, induced resistance, and antibiosis using genomics, transcriptomics, and metabolomics analyses as well as a series of physiological and biochemical analyses.
Chitinases, involved in mycoparasitism, are diverse, inducible secretory proteins. Characteristics of chitinases differ in different Trichoderma species. Trichoderma koningii inhibits the growth of Sclerotium cepivorum in onion root via the production of exo-and endo-chitinases, which digest the cell walls of pathogens [42]. β-1, 3-glucanase, also associated with mycoparasitism, induces strong hydrolysis of pathogen cell walls, thereby inhibiting germination of pathogen spores and digestion of the mycelium [43]. In addition, proteases play an important role in mycoparasitism, as they are necessary for the complete hydrolysis of pathogen cell walls [44]. In our study, 16 chitinase genes, eight protease genes, and 11 glucanase genes were discovered in T. asperellum GDFS1009, only a few of which were highly expressed in PD broth without induction medium (Fig 3 and S1 Table). In future experiments, the specific functions of these enzymes, such their tolerances to heat, cold, or metal ions, could be predicted with bioinformatics analysis or via expression in E. coli or Saccharomyces cerevisiae in order to more fully understand their enzymatic properties. In the induction broths, the three types of enzymes were all detected and exhibited high activities, providing an experimental basis for subsequent development of enzyme preparations. Induction broths should be optimized for greater enzyme activity, and single or mixed enzyme preparations may be applied in agricultural production.
In recent years, it was discovered that Trichoderma and its formulations could elicit pathogen-related gene expression in host plants and induce systematic acquired resistance (SAR). In 1999, Yedidia found that when the Trichoderma mycelium crosses the epidermis layer of cucumber root, it induces a cascade of pathogen-related protein expression, similar to that caused by the chemical mutagen 2, 6-dichlorphenoxyacetic acid [45,46]. Similarly, after treatment of the cucumber root with T. harzianum T-203, the JA/SA signal pathway was induced, and Pseudomonas lachrymans was inhibited [47,48]. Numerous studies have shown that Trichoderma produces several types of inducible factors or elicitors, such as sm1 and xylanase. In this study, xylanase activity was also detected in induction broth. This may induce plant resistance due to the induction of tissue necrosis. Additionally, in the T. asperellum GDFS1009 genome, 12 elicitor genes with different functions were found, including two endopolygalacturonase genes, two Epl protein genes, two hydrophobin genes, one polygalacturonase gene, one swollenin gene, and four xylanase genes. The expression levels of most of these elicitor genes were very low or not detected, with only a few being highly expressed in PD broth after 24 h and 48 h (Fig 4 and S2 Table). In future experiments, elicitors which mechanisms are unknown could be expressed in E. coli or Saccharomyces and be purified for further investigation into the interactions with plants and pathogens.
Nine primary metabolites believed to be fungicides or precursors/intermediates of fungicides, insecticides, or herbicides were identified in T. asperellum GDFS1009 (Table 1). Acetamide is widely used as a pesticide. Dibromocyanoacetamide (DBNPA) is a novel potent sanitizer and excellent water treatment agent [49]. Ethylamine can be used as a raw material to Comprehensive evaluation of biocontrol efficacy of T. asperellum produce triazine herbicides [50], while ethylene glycol can be used as an intermediate for pesticide production [51]. Glycine can be used to synthesize fungicides, and research has also shown that copper glycine inhibits B. cinerea and F. oxysporum [52]. Ethanolamine is an intermediate in the synthesis of pesticides. The combined application of Schiff bases of salicylaldehyde ethanolamine with copper acetate or zinc sulfate exhibited significant inhibition of Aspergillus niger and F. oxysporum [53]. As an important organic synthesis intermediate, otoluic acid is also widely used in the production of pesticides [54]. L-malic acid can be used for pharmaceutical formulations and also for the synthesis of insect repellents [55]. Citric acid strongly inhibits Aeromonas hydrophila, Pseudomonas fluorescens, and other aquatic pathogens [56].
Antimicrobial peptides exhibit broad antimicrobial abilities and strong non-specific resistance to pathogens. In 2005, Nielsen isolated an antimicrobial peptide polymer with 20 residues from T. brevicompactum [18]. In 2006, Kada isolated a new, short antimicrobial peptide from Trichoderma [66]. In 2007, Ruiz identified 11 residual fragments from antimicrobial peptides in T. longibrachiatum [19]. In this study, six types of antimicrobial peptides were detected and speculated from T. asperellum GDFS1009: TBV, trichotoxin_A-50_F, hypomurocin_ B_IIIa, trichotoxin_A-50_I, hypomurocin_B_I, and hypomurocin_B_II ( Table 3). As antimicrobial peptides are formed through the catalysis of NRPS, related gene clusters were analyzed and 16 NRPSs were found in T. asperellum GDFS1009, with two PKS/NRPS hetero-gene clusters. After 24 h and 48 h in PD broth, almost all of these gene clusters were lowly expressed or not detected (Table 2).
Based on these results, we infer that there are many mycoparasitism-related enzymes, elicitors, primary and secondary metabolites in Trichoderma that have not yet been discovered, including chitinase, protease, polyketides, antimicrobial peptides and so on. Future experiments should focus on optimizing the ingredients of the induction broth and the incubation conditions during the fermentation process. In addition, the silent genes and gene clusters should be expressed in Saccharomycetes. In conclusion, Trichoderma spp. offer great promise for ongoing agricultural development, expressing many promising enzymes, elicitors, and secondary metabolites.

Conclusion
Due to its efficient broad-spectrum antimicrobial activity, Trichoderma has been established as an internationally recognized biocontrol fungus. Though researches about Trichoderma spp. were very broad, most just concentrated in one or several biocontrol factors in one strain. Our study reveals resistant molecular mechanisms of T. asperellum in global aspects, including almost all biocontrol factors related to mycoparasitism, induced resistance and antibiosis, based on genomics, transcriptomics, metabolomics and a series of physiological and biochemical analyses. Firstly, we found and identified a novel strain of T. asperellum, named GDFS1009. The mycelium of T. asperellum GDFS1009 has high growth rate, high sporulation capacity, and very strong inhibitory effects on pathogens that cause cucumber fusarium wilt and corn stalk rot. T. asperellum GDFS1009 secretes chitinase, glucanase and protease, which can degrade the cell walls of fungi and participate in the mycoparasitism. The secreted xylanase are good candidates in inducing plant resistance and enhancing plant immunity. Then, RNA-seq and GC-MS analysis showed that T. asperellum GDFS1009 can produce nine primary metabolites which are precursors of antimicrobial compounds, and a variety of antimicrobial secondary metabolites, including polyketides, alkanes and so on. In addition, this study speculated six antimicrobial peptides via UPLC-QTOF-MS/MS. However, through synthetic analysis we can infer that there are some enzymes and elicitors with specific function and lots of secondary metabolites in Trichoderma have not been discovered. Subsequently, these related genes or gene clusters could be predicted and expressed via E. coli and Saccharomyces systems for wider application in the field of agricultural biocontrol.