Genome Mining of the Biocontrol Agent Trichoderma afroharzianum Unearths a Key Gene in the Biosynthesis of Anti-Fungal Volatile Sesquiterpenoids

Abstract

The volatile organic compounds (VOCs) in Trichoderma afroharzianum ACCC 33109 have the biological activities of both hydrolytic enzymes and antimicrobial peptides to mitigate attack by phytopathogens and spread over long distances in soil. However, the biosynthesis pathway of anti-fungal VOCs has not been elucidated yet. In this study, we identified 15 genes (TaTS1–15) coding for putative terpene synthase with low identities (<79.54%) to functionally characterized homologs through genome mining. Upon Fusarium induction, the relative expression levels of nine TaTS genes were up-regulated by up to 2793-fold (TaTS9). To verify the contribution of TaTS9 to the synthesis of anti-fungal VOCs, the TaTS9 knockout mutant strain was constructed and characterized by its antagonistic activities, transcript profiles, and VOC metabolomes. Heterologous expression of TaTS9 in Escherichia coli produced the target gene product, which converted the precursor farnesyl pyrophosphate (FPP) into β-cubenene (>90%) and γ-amorphene. Thus, TaTS9 was confirmed as the first β-cubenene synthase of Trichoderma, which catalyzes the biosynthesis of various sesquiterpenes with anti-fungal activities. This study provides insight into the key terpene synthase gene in the biosynthesis of anti-fungal sesquiterpenoids for potential applications in the agriculture and food industries.

1. Introduction

Trichoderma (phylum Ascomycota, class Sordariomycetes, order Hypocreales, and family Hipocreaceae) are ubiquitous on Earth and prominent in studies due to their ability to grow fast, spread and colonize easily, and produce a vast array of metabolites, including extracellular enzymes, antibiotics, polyketides, as well as terpenoids [1]. Different bioactive compounds vary in the mechanism of action. Some Trichoderma strains exert biocontrol action over a wide range of phytopathogens of agro-industrial importance through competition, mycoparasitism, and antagonism, and they confer beneficial effects on crops through symbiotic interaction as well [2,3], thus representing a safe and environmentally friendly biocontrol agent and biofertilizer as an alternative to chemical pesticides and fertilizers.

Upon recognition of phytopathogens, antagonistic Trichoderma strains initiate the synthesis of bioactive compounds to mitigate attack by pathogens. For examples, the cell-wall-degrading enzymes, such glucanases, chitinases, and proteases, are able to disintegrate the conidia and mycelia of phytopathogens [4,5]; the peptaibols can alter the osmotic pressure of pathogen cells and inhibit the β-1,3-glucan synthetase activities [6]; antibiotics such trichodermin, viridin, and harzianolide counteract pathogens by inhibiting their ribosomal activities [7]; and bisvertinolone inhibits the biosynthesis of β-1,6-glucan of the pathogen cell wall [8]. In comparison to these soluble compounds that play roles through close contact, the volatile organic compounds (VOCs) produced by Trichoderma, such as pyrones and terpenoids, are a large group of carbon-based chemicals with low molecular weights, low polarities, low boiling points, and high vapor pressure, and they can break through physical barriers and spread over long distances in the air, water, and soil, providing greater advantages over other secondary metabolites of a microbial nature in preventing and controlling soil-borne diseases and food contamination [9,10]. Moreover, the Trichoderma VOCs have the biological activities of both hydrolytic enzymes and antimicrobial peptides, and they can promote plant growth, biomass production, and root development [11,12]. For example, the Trichoderma terpenoids have antibacterial, anti-fungal, inhibitory, cytotoxic, and anti-virus activities, and the 6-pentyl-α-pyrone has anti-fungal activities against a wide range of phytopathogens of Fusarium, Phytophthora, Rhizoctonia, Macrophomina, Sclerotium, Colletotrichum, and Cylindrocarpon, as well as the stored grain pathogen Aspergillus flavus, by deconstructing their cell wall structure, altering the amino acid metabolism, causing downregulation of the ECHS1 protein, and inducing autophagy [13,14,15,16]. Thus, application of the Trichoderma VOCs is considered a promising biocontrol strategy for the management of plant diseases in agriculture and food contamination during storage.

Thus far, a total of 253 terpenoids, including 202 sesquiterpenes, 48 diterpenes, two monoterpenes and one meroterpenoid, have been isolated from Trichoderma spp. since 1948, and the sesquiterpenes can be divided into nine categories based on the skeleton: cyclonerane-type, cadinane-type, carotane-type, acorane-type, trichothecene-type, bisabolane-type, botryane-type, drimane-type and other sesquiterpenes [11]. The abundance of the bioactivities and skeletons of terpenoids presents complex biosynthetic pathways involving successive head-to-tail condensation, cyclization, and side-chain decoration of the linear chain of prenyl diphosphate of varying lengths (C5 and C10), which are catalyzed by terpenoid synthases/cyclases (TSs/TCs), polyketide synthases (PKSs), non-ribosomal peptide synthases (NRPSs), and PKS-NRPS hybrids [17,18]. Thanks to sequencing technology developments, the genomes of 158 Trichoderma strains have been published in the NCBI genome database (https://www.ncbi.nlm.nih.gov/datasets/genome/?taxon=5543; accessed on 18 March 2025), and a large number of TSs have been annotated in 13 Trichoderma spp. [11,19]. However, only a few TSs from Trichoderma have been functionally characterized and structurally resolved. For example, the terpene cyclase from Trichoderma atroviride is able to convert farnesyl diphosphate (FPP) into a mixture of sesquiterpene hydrocarbons and alcohols, including the main product trichobrasilenol and minor side products such alpha-humulene, caryophyllene, 2-epi-caryophyllene, african-3-ene, african-1-ene, isoafricanol, and pristinol [20]; the gene cluster of UbiA terpene synthase from Trichoderma erinaceum F1-1 was validated to encode the biosynthesis of (−)-α-trans-bergamotene and eight novel bergamotene-derived sesquiterpenoids [21]; and other Trichoderma TS studies have focused on the trichodiene synthases or squalene synthases (https://www.uniprot.org/uniprotkb?query=*trichoderma+terpene+synthase&facets=proteins_with%3A13; accessed on 18 March 2025). Thus, genome mining and enzymology exploration are promising in terms of the discovery of novel sesquiterpenes derived from Trichoderma in the areas of chemistry, agriculture, microbiology, and synthetic biology.

In a previous study, we found that the VOCs produced by Trichoderma afroharzianum ACCC 33109 exert strong inhibitory effects on the growth of Fusarium oxysporum f. sp. cucumerinum (Foc) [22]. However, which types of volatile sesquiterpenes were released and functioned in the antagonism remained unclear. The aims of this study were to identify the key TS gene associated with the biosynthesis of anti-fungal sesquiterpenes against Foc, to clarify the type of TSs and sesquiterpenes produced, and to provide reference data on anti-fungal sesquiterpenes as novel chemical weapons for confronting other pathogens in a highly efficient and environmentally friendly way.

2. Results and Discussion

2.1. Genome Mining and Bioinformatic Analysis

The genus Trichoderma is of great importance to agriculture and industry, and it has been a hot issue in research and development. Of the 158 Trichoderma genomes published in the NCBI database, six sequencing projects have targeted T. afroharzianum strains through Illumina NovaSeq or Sanger technologies since 2021. In this study, Illumina NovaSeq/PacBio Sequel platforms were used to sequence the whole genome of ACCC 33109, which was fine-mapped and submitted to GenBank under BioProject PRJNA1206798. Similar to the genomes of the T. afroharzianum strains that are 38.4–44.1 Mb in size, with the G+C contents of 46.5–50.5% (Supplementary Table S1), ACCC 33109 is 40.7 Mb in size, with the G+C content of 47.0%, and encodes 11,022 genes, including 9827 genes involved in the KEGG metabolic pathway, 54 secondary metabolic gene clusters (645 genes), and 10 families of cytochrome P450 (176 genes).

Genome mining is an efficient and reliable strategy for the discovery of novel terpenoids from filamentous fungi. Using this strategy, Yang et al. [21] identified a biosynthetic gene cluster of terpene synthase BgtA in the genome of Trichoderma erinaceum F1-1 and discovered eight new bergamotene-derived sesquiterpenoids. As for ACCC 33109, 35 genes were mapped to the putative KEGG biosynthesis pathway of the terpenoid skeleton (

Table 1. Putative genes involved in the biosynthesis of terpenoid.

No.	Gene ID	EC	Enzyme	Product
1	A03930	2.3.1.9	Acetyl-CoA acetyltransferase	MVA pathway to produce IPP
2	A05584	2.3.1.9	Acetyl-CoA acetyltransferase
3	A02030	2.3.3.10	Hydroxymethylglutaryl-CoA synthase
4	A07965	1.1.1.34	Hydroxymethylglutaryl-CoA reductase
5	A02844	2.7.1.36	Mevalonate kinase
6	A03938	2.7.4.2	Phosphomevalonate kinase
7	A10372	4.1.1.33	Diphosphomevalonate decarboxylase
8	A03978	5.3.3.2	Isopentenyl diphosphate delta-isomerase	DMAPP (C5)
9	A03044	2.5.1.1	Geranyl diphosphate synthase	GPP (C10)
10	A03290	2.5.1.10	Farnesyl diphosphate synthase	FPP (C15)
11	A01536	2.5.1.29	Geranylgeranyl diphosphate synthase	GGPP (C20)
12	A05498	2.5.1.29	Geranylgeranyl diphosphate synthase	GGPP (C20)
13	A06847	2.5.1.83	Hexaprenyl pyrophosphate synthase	HexPP (C20)
14	A05610	2.5.1.21	Farnesyl-diphosphate farnesyltransferase	FFPP (C30)
15	A01534	4.2.3.-	Cycloaraneosene synthase	Terpenoid skeleton
16	A01643	4.2.3.-	Delta(6)-protoilludene synthase
17	A03322	4.2.3.-	Silphinene synthase
18	A04687	4.2.3.-	Sesquiterpene cyclase
19	A04858	4.2.3.-	Longiborneol synthase
20	A06313	4.2.3.-	Fusicoccadiene synthase
21	A09488	4.2.3.-	Terpene cyclase
22	A09872	4.2.3.-	Sesquiterpene cyclase
23	A10648	4.2.3.-	(+)-Eremophilene synthase
24	A01608	2.5.1.87	Polyprenyl diphosphate synthase	Side-chain-decorated terpenoids
25	A02633	2.5.1.87	Dehydrodolichyl diphosphate synthase
26	A02814	2.5.1.75	tRNA dimethylallyl transferase
27	A02831	3.4.24.84	STE24 endopeptidase
28	A02842	2.5.1.58/59	Farnesyl/geranylgeranyl transferase
29	A05170	2.1.1.100	S-isoprenylcysteine O-methyltransferase
30	A06568	1.8.3.5/6	Prenylcysteine oxidase/farnesylcysteine lyase
31	A06978	2.5.1.58	Farnesyl transferase
32	A07199	1.13.11.59	Torulene dioxygenase
33	A08871	3.4.22.-	Prenyl protein peptidase
34	A09359	1.2.1.82	Beta-apo-4′-carotenal oxygenase
35	A03646	1.14.14.17	Squalene monooxygenase

- means bigger and unidentified family.

). Acetyl CoA generated through glycolysis, the tricarboxylic acid cycle, and other pathways is catalyzed by seven putative enzymes (A03930, A05584, A02030, A07965, A02844, A03938, and A10372) of the mevalonic acid pathway (MVA) to form the terpene precursor isopentenyl pyrophosphate (IPP; C5), which is further catalytically isomerized by A03978 to produce dimethylallyl pyrophosphate (DMAPP; C5) or condensed by six putative isopentenyl transferases (PTs; A01536, A03044, A03290, A05498, A05610, and A06847) to produce geranyl/farnesyl/geranylgeranyl/farnesylfarnesy/hexaprenyl pyrophosphate (GPP, FPP, GGPP, FFPP, and HexPP; C10–C30). These long carbon skeletons are further cyclized by nine putative terpene cyclases (TCs; A01534, A01643, A09488, A04687, A10648, A06313, A03322, A04858, and A09872) to produce the terpenoid skeleton. Through secondary enzyme reactions such as dehydrogenation, hydroxylation, methylation, and acylation catalyzed by, i.e., A01608, A02633, A02814, A02831, A02842, A05170, A06568, A06978, A07199, A08871, A09359, A03646, etc., the side-chains of the terpenoid backbones are decorated to generate various terpenoid compounds with diverse bioactivities. Moreover, gene clusters of ten TaTS genes involving 57 genes were also identified (Supplementary Table S2), which also contribute to the diversification of terpenoid compounds.

Phylogeny analysis of the deduced proteins indicated that the 15 putative TaTS genes were mainly confined to Class I TC and Class I PT, and they shared 25.82–79.54% identities with the Swiss-Prot homologs (Figure 1). Of them, A05610 and A04858 represent the ancestors of ACCC 33109 TSs and TCs, respectively, and A03044 and A03290 are bifunctional PT-TC. Further phylogeny analysis divided the 133 TCs from Trichoderma spp. into seven clades, and nine TCs from ACCC 33109 are evenly distributed (Supplementary Figure S1). This suggests that the evolution of ACCC 33109 TSs is complex, contributing to the multiple enzyme functions and terpenoid diversity.

2.2. Expression Levels of the TaTS Genes upon Foc Induction

It is common practice to gain insights into the expression level of TS genes by RT-qPCR with the beta-tubulin (housekeeping) gene as the internal reference [23]. Considering the contribution of TaTS transcriptional profiles to the biosynthesis of diverse and distinct terpenoids, the mRNA levels of 10 Tats genes upon Foc induction were analyzed over 5 days using the RT-qPCR approach. As shown in Figure 2, the TaTS genes were differentially expressed in the presence of Foc and over the days of Foc induction. Similar up–down transcriptional profiles were observed for the TaTS1, TaTS7, TaTS8, and TaTS15, with the highest levels (2.03–6.16-fold) on day 3. The TaTS2 and TaTS4 showed similar transcriptional profiles, peaking on day 1 (8.06- and 4.03-fold) and declining afterwards. The TaTS9 showed the most significant increasing trend of transcription from 3.14-fold on day 1 to 2793.12-fold on day 4, followed by decreased transcription to 1103.94-fold on day 5. The TaTS3, TaTS5, and TaTS6 were suppressed or not expressed in ACCC 33109 upon Foc induction. The results indicated that the transcriptional patterns of seven TaTS genes are complementary in functions (isozymes) and in chronological order, which contributed to the biosynthesis of diverse anti-fungal terpenes upon Foc induction. Since none of the TaTSs from ACCC 33109 have been experimentally characterized, it will be informative to understand the function of these genes via either the heterologous expression or the gene deletions. Therefore, the TaTS3, TaTS4, TaTS8, and TaTS9 genes showing representative suppressed or highly expressed transcription profiles were selected for function verification through the deletion of target genes.

2.3. Creation and Verification of the TaTS Mutants

Recent advances in genome-mining-assisted synthetic biology and metabolic engineering have reinforced filamentous fungi as promising chassis cells to produce bioactive natural products [24]. In comparison to well-known engineering strains such E. coli, Saccharomyces cerevisiae, Pichia pastoris, and Aspergillus spp., the development of Trichoderma as a chassis strain has been limited by the scarcity of efficient genetic manipulation tools and application potentials. In the present study, hygromycin B and G418 were selected as the antibiotics for fungal sensitivity. It was found that 100 μg/mL hygromycin B and 400 μg/mL G418 completely inhibit the mycelial growth of ACCC 33109, respectively (Supplementary Figure S2). To prepare the protoplasts, fresh mycelia treated with lysing enzymes for 8 h yielded approximately 8.9 × 10⁷ mL⁻¹ protoplasts with a recovery rate of 20% on TB3 agar plates (Supplementary Figure S3). To generate the chassis strain ΔTalig4::neo, the recombinant fragments of neo and Talig4 upstream (neo-Lig-up) and Talig4 downstream and neo (Lig-down-neo) were obtained by two rounds of PCR (Figure 3a) and transformed into the protoplasts of ACCC 33109 according to the homologous recombination strategy [25]. As expected, the transformants of the Talig4 deletion strain were highly resistant to G418 and grew well on selective TB3 agar plates supplemented with 400 µg/mL of G418 (Figure 3b). To verify these mutants, genomic DNA was extracted and diagnostic PCR was performed using the designated primers. The correct mutants carrying the gene fragment of neo showed a band of 1 kb and missed the gene fragment of Talig4 (Figure 3c,d). The results demonstrated that the mutant with the deletion of Talig4 in ACCC 33109 was successfully created.

Mutant strains of four TaTS gene deletions were then created using ΔTalig4::neo as the chassis strain following the protocol described above. As shown in Supplementary Figure S4, the recombinant fragments of TaTS3/4/8/9 upstream/downstream and hph (hph-TaTS-up and TaTS-down-hph) obtained by the fusion PCR reaction were successfully transformed into the protoplasts of ΔTalig4::neo, which grew well on selective TB3 agar plates supplemented with 100  µg/mL hygromycin B and showed a hph fragment of 600 bp on agarose gel. The knockout mutant strains were verified with the absence of TaTS3/4/8/9 on gel (Figure 4). In comparison to the very low recombination efficiency of ACCC 33109 (one out of three hundred transformants for Talig4 replacement with the neo gene), the recombination efficiency of ΔTalig4::neo as the chassis strain increased a lot, which was 12–162 out of 300 transformants for TaTS replacement with the hph gene. These results indicate that the deletion of the Talig4 gene in ACCC 33109 significantly increased the frequency of the gene replacement, as reported in T. hypoxylon [25].

2.4. Characterization of the TaST Mutants

Through integrative analysis of the fungal genomics and phenotype characters, the roles of the four TaTS genes in the terpene biosynthesis of ACCC 33109 were identified. The TaTS gene disruption mutants ΔTalig4::neoΔTaTS::hph and chassis strain ΔTalig4::neo were cultured on sandwiched PDA plates with Foc for 5 days at 28 °C for the antagonism assay and qPCR analysis. Considering the antagonism of VOCs is ascribed to the combined action of multiple compounds, the absence of a certain gene cannot cause the complete loss of anti-fungal activity of the VOCs in Trichoderma [9,26]. As shown in Supplementary Figure S5 and

Table 2. Inhibitory rates of the VOCs of different mutants of ACCC 33109 against Foc ^a.

Time	Inhibitory Rate (%)
	ΔTalig4::neo	ΔTalig4::neoΔTaTS3::hph	ΔTalig4::neoΔTaTS4::hph	ΔTalig4::neoΔTaTS8::hph-5	ΔTalig4::neoΔTaTS9::hph-3	ΔTalig4::neoΔTaTS9::hph-130
Day 3	18.56	14.90 **	17.21	14.70 **	14.63 **	17.83
Day 4	17.16	14.45 **	17.96	10.60 ***	13.78 ***	18.17
Day 5	22.37	15.33 **	18.88 *	13.14 ***	19.40 *	19.08 **

^a *, **, and *** indicate statistically significant differences of the same row between ΔTalig4::neo and ΔTalig4::neoΔTaTS3/4/8/9::hph at p < 0.05, p < 0.01, and p < 0.001, respectively.

, the VOCs produced by the chassis strain significantly inhibited the Foc growth by 18.56%, 17.16%, and 22.37% at days 3, 4, and 5 respectively. Compared with the chassis strain, deletion of the TaTS3/4/8/9 showed significant effects on the biosynthesis of terpenoids, which further changed the composition and contents of the anti-fungal VOCs. Significant inhibition reduction (up to 9.23%) of Foc growth was detected in the ΔTalig4::neoΔTaTS9::hph-3-, ΔTalig4::neoΔTaTS9::hph-130-, and ΔTalig4::neoΔTaTS8::hph-5-treated plates on days 3–5 and ΔTalig4::neoΔTaTS3::hph- and ΔTalig4::neoΔTaTS4::hph-treated plates on day 5 (p < 0.05). The results indicated that the four terpene-synthase-encoding genes are crucial to the biosynthesis of antagonistic VOCs.

Deletion of TaTS3/4/8/9 also affected the expression levels of other TaTS genes (Figure 5). In comparison to the chassis strain ΔTalig4::neo, some genes were depressed in the mutant strains, and some genes were overexpressed. In the mutant strains ΔTalig4::neoΔTaTS3::hph and ΔTalig4::neoΔTaTS4::hph, the mRNA transcription levels of the TaTS genes were unchanged or depressed significantly (0.05–0.36-fold; p < 0.05). In the ΔTalig4::neoΔTaTS8::hph, the mRNA transcription levels of TaTS3 and TaTS6 increased (2.42- and 2.03-fold; p < 0.05), while the other TaTS genes had no changes or decreased in terms of the mRNA expression levels (0.09–0.39-fold; p < 0.05). Only one gene, TaTS5, was overexpressed in the ΔTalig4::neoΔTaTS9::hph, with the highest level of 32.5-fold (p < 0.001), and the other TaTS genes had no changes or decreased in the mRNA expression levels (0.05–0.38-fold; p < 0.05). The results indicate that the TaTS genes of ACCC 33019 can be classified into the constitutive and inducible types, which form a complex network and function in independent (i.e., TaTS1 and TaTS4), synchronous (i.e., TaTS2 and TaTS7) as well as complementary (i.e., TaTS5 and TaTS9; TaTS3, TaTS6 and TaTS8) ways in the biosynthesis of terpenoids. Moreover, no TaTS genes were detected in the corresponding deletion mutants by qPCR, suggesting the mutant strains lacked genome complementarity to homologs and retained recombination stability.

In comparison to previous studies that focused on a TS gene or gene cluster [21,23], we assessed the whole TaTS gene family of ACCC 33109 to find the key gene in the biosynthesis of anti-fungal terpenes. Combined analysis of the antagonism and TaTS gene transcripts of the wild-type strain, chassis strain, and four mutant strains revealed that TaTS8 and TaTS9 are the key genes by intriguing shifts of more TaTS genes or overexpression upon Foc induction.

2.5. VOC Metabolomes of the Chassis and Mutant Strains

The VOCs released by ΔTalig4::neo, ΔTalig4::neoΔTaTS8::hph, and ΔTalig4::neoΔTaTS9::hph were collected at day 5 by the SPEM and identified by the GC-MS against the NIST database. As shown in

Table 3. Identities and contents of the VOCs detected by SPEM-GC-MS ^a.

No.	Compound	Retention Time (min)	Formula	Molecular Weight	Absolute Amount
					ΔTalig4::neo	ΔTalig4::neoΔTaTS8::hph	ΔTalig4::neoΔTaTS9::hph
1	1-Pentanol	3.146	C5H12O	88.15	1.07 × 108	2.68 × 107	2.68 × 107
2	2-Ethyl-1-hexanol	10.781	C8H18O	130.23	3.35 × 107	2.07 × 106	2.59 × 106
3	Phenylethyl alcohol	12.723	C8H10O	122.16	3.47 × 107	8.68 × 106	1.09 × 107
4	3-Octanol, acetate	12.950	C10H20O2	172.26	2.01 × 105	*	1.01 × 105
5	2-Heptenoic acid, 4-nitrophenyl ester	13.159	C13H15NO4	249.26	7.80 × 105	5.04 × 104	2.08 × 105
6	β-Cubenene	18.511	C15H24	204.35	6.58 × 106	1.95 × 105	*
7	β-Cubenene	18.620	C15H24	204.35	7.54 × 106	1.64 × 106	*
8	cis-α-Bergamotene	18.741	C15H24	204.35	3.70 × 105	*	*
9	Bicyclo[5.2.0]nonane, 4-methylene-2,8,8-trimethyl-2-vinyl-	18.849	C15H24	204.35	7.57 × 105	*	*
10	Cyclohexane, 1,2-diethenyl-4-(1-methylethylidene)-, cis-	18.958	C13H20	176.33	*	1.88 × 106	1.69 × 106
11	1,3a-Ethano-3aH-indene, 1,2,3,6,7,7a-hexahydro-2,2,4,7a-tetramethyl-, [1R-(1.alpha.,3a.alpha.,7a.alpha.)]-	18.969	C15H24	204.35	1.13 × 106	*	2.30 × 106
12	(1S,4aR,7R)-1,4a-Dimethyl-7-(prop-1-en-2-yl)-1,2,3,4,4a,5,6,7-octahydronaphthalene	19.041	C15H24	204.35	7.92 × 105	9.25 × 104	*
13	1H-Cyclopenta[1,3]cyclopropa[1,2]benzene, octahydro-7-methyl-3-methylene-4- (1-methylethyl)-, 3aS-(3a.alpha.,3b.beta.,4.beta.,7	19.370	C15H24	204.35	*	1.89 × 105	*
14	γ-Amorphene	19.380	C15H24	204.35	2.72 × 105	*	*
15	(1R,4R,5S)-1,8-Dimethyl-4-(prop-1-en-2-yl)spiro[4.5]dec-7-ene	19.434	C15H24	204.35	1.03 × 106	6.84 × 101	*
16	(1R,5S)-1,8-Dimethyl-4-(propan-2-ylidene)spiro[4.5]dec-7-ene	20.017	C15H24	204.35	7.97 × 105	2.84 × 105	*
17	Bicyclo[2.2.1]heptane, 2-(1,1-dimethyl-2-propenyl)-	20.320	C12H20	164.29	2.02 × 105	1.98 × 105	*
18	cis-Calamenene	20.340	C15H22	202.33	3.84 × 105	7.89 × 101	*
19	Tricyclo[6.3.0.0(1,5)]undec-2-en-4-one, 5,9-dimethyl-	20.590	C13H18O	190.28	1.29 × 105	*	1.23 × 105
20	Biphenylene, 1,2,3,6,7,8,8a,8b-octahydro-4,5-dimethyl-	21.290	C14H20	188.31	1.13 × 105	6.81 × 104	2.52 × 105
21	1H-Benzocyclohepten-7-ol, 2,3,4,4a,5,6,7,8-octahydro-1,1,4a,7-tetramethyl-, cis-	22.305	C15H26O	222.37	5.09 × 105	2.56 × 105	6.28 × 105
22	β-Ocimene	22.421	C10H16	136.26	3.31 × 105	1.99 × 105
23	α-Bisabolol	22.454	C15H26O	222.37	1.67 × 105	5.05 × 104	*
24	Spiro[4.5]dec-6-en-8-one, 1,7-dimethyl-4-(1-methylethyl)-	22.635	C15H24O	220.35	6.10 × 105	9.60 × 104	*
25	Di-epi-1,10-cubenol	23.477	C15H26O	222.37	4.96 × 105	3.23 × 104	*
26	Tricyclo[5.4.0.0(2,8)]undec-9-ene, 2,6,6,9-tetramethyl-, (1R,2S,7R,8R)-	23.670	C15H24	204.35	1.03 × 105	2.84 × 104	*
27	Hinesol	24.433	C15H26O	222.37	2.68 × 105	1.27 × 105	*
28	Hexadecanoic acid, methyl ester	25.256	C17H34O2	270.50	2.97 × 105	*	*
29	n-Hexadecanoic acid	25.621	C16H32O2	256.42	3.87 × 106	*	*
30	Hexadecanoic acid, ethyl ester	25.974	C18H36O2	284.50	1.16 × 105	*	*
31	Bicyclo[9.3.1]pentadeca-3,7-dien-12-ol, 4,8,12,15,15-pentamethyl-, [1R-(1R*,3E,7E,11R*,12R*)]-	26.150	C20H34O	290.50	2.86 × 106	8.27 × 104	3.78 × 105
32	Bicyclo[9.3.1]pentadeca-3,7-dien-12-ol, 4,8,12,15,15-pentamethyl-, [1R-(1R*,3E,7E,11R*,12R*)]-	26.353	C20H34O	290.50	2.93 × 105	4.19 × 104	2.21 × 105
33	(S,E)-8,12,15,15-Tetramethyl-4-methylenebicyclo[9.3.1]pentadeca-7,11-diene	26.555	C20H32	272.50	4.86 × 105	1.52 × 105	*
34	1-Octadecanol	26.879	C18H38O	270.50	2.76 × 105	*	*
35	Panaxene	26.935	C15H24	204.35	1.24 × 105	1.24 × 105	*
36	Methyl stearate	27.293	C19H38O2	298.50	1.69 × 105	*	*
37	Bicyclo[9.3.1]pentadeca-3,7-dien-12-ol, 4,8,12,15,15-pentamethyl-, [1R-(1R*,3E,7E,11R*,12R*)]-	29.062	C20H34O	290.50	2.12 × 106	2.58 × 104	*
38	Hexanedioic acid, bis(2-ethylhexyl) ester	30.678	C22H42O4	370.60	6.34 × 105	6.71 × 104	6.31 × 104
39	Cholesteryl formate	36.302	C28H46O2	414.70	5.56 × 105	*	*
40	Cholesta-2,4-diene	37.471	C27H44	368.60	5.17 × 105	*	*
41	Cholest-5-en-3-ol (3.beta.)-, nonanoate	37.909	C36H62O2	526.90	1.74 × 106	*	*
Total					1.88 × 108	4.34 × 107	5.29 × 107

^a *, not detected.

, a total of 39 compounds were detected in ΔTalig4::neo, including seven alcohols (81.89%), 17 sesquiterpenes (11.71%), four diterpenes (3.08%), six esters (1.17%), one each of fatty acid (1.72%), monoterpene (0.18%), alkane (0.11%), ketone (0.07%), and benzene (0.06%). In comparison to the high content and diversity of the VOCs produced by ΔTalig4::neo, less volatile compounds (27 and 13, respectively) were produced in ΔTalig4::neoΔTaTS8::hph and ΔTalig4::neoΔTaTS9::hph (Figure 6). Notable differences were detected among the diversity and contents of the sesquiterpenes (RT at 18–24 min, Supplementary Figure S6). ΔTalig4::neoΔTaTS8::hph produced 14 sesquiterpenes with less relative content of 11.22%, and ΔTalig4::neoΔTaTS9::hph only produced two sesquiterpenes, i.e., 1H-benzocyclohepten-7-ol,2,3,4,4a,5,6,7,8-octahydro-1,1,4a,7-tetramethyl-cis-19 and cyclohexane (undetected in ΔTalig4::neo), with the low content of 0.35%. Similar results have been reported in the mutant strain of Trichoderma virines with the deletion of the sesquiterpene synthase gene vir4, which produced 24 less sesquiterpenes [27]. In addition, β-cubenene, as the predominant sesquiterpene of ΔTalig4::neo (7.54%), was completely absent in the VOCs of ΔTalig4::neoΔTaTS9::hph. This indicated that TaTS9 is the most critical gene for the biosynthesis of sesquiterpenes, especially β-cubenene. A sesquiterpene synthase from filamentous Acremonium chrysogenum (AcTPS3; KFH47841.1) has been identified through genome mining, and its product was functionally characterized to be (+)-cubenene [28]. However, TaTS9 shared very low similarity (<30%) to AcTPS3, except for the conserved motifs DDXXD and NSE of the Class I terpene synthases. Therefore, it is inaccurate to predict the functions of TaTS9 through sequence similarity and homology modeling. Thus, biochemical characterization of TaTS9 through heterologous expression is crucial and necessary.

2.6. Production and Purification of the Recombinant TaTS9

The deduced TaTS9 contains 375 amino acid residues with a pI value of 6.12 and a molecular weight of 42.04 kDa, and it shares 34.06% similarity with the functionally characterized and structurally resolved longiborenol synthase of Fusarium graminearum (I1S104.1). The recombinant plasmid pET28-TaTS9 was transformed into E. coli BL21 (DE3) and cultured overnight, followed by 20 h of induction with 0.5 mM IPTG. The cells were harvested, resuspended, and disrupted, and the crude enzyme was subjected to Ni-NTA affinity purification and ultraconcentration. As shown in Figure 7a, the purified recombinant TaTS9 showed a single band of 44 kDa on the SDS-PAGE, which was similar to the theoretical weight of TaTS9 plus His₆-TEV. Although the recombinant TaTS9 was successfully produced in the E. coli, the yield was low, which might be due to the prokaryotic incompatibility. However, the heterologous expression system of E. coli is always preferred for the functional characterization and product identification of TSs due to the absence of the mevalonate (MVA) pathway and the simpleness of the reaction system consisting of the precursor FPP and recombinant TS. For instance, the terpenoid cyclase TaTC6 from T. atroviride FKI-3849 produced in E. coli BL21 (DE3) could convert FPP into trichobrasilenol with some sesquiterpenes by-products [20].

2.7. Identification of the Sesquiterpene Products Catalyzed by TaTS9

The purified TaTS9 (100 μg/mL) and FPP (100 μm) were incubated at 50 rpm and 28 °C overnight, and the sesquiterpene products were extracted by hexane and identified by the GC-MS. As shown in Figure 7b, three peaks with the retention time at 18.462 min, 18.569 min, and 19.394 min were detected, corresponding to β-cubenene, β-cubenene, and γ-amorphene, respectively. The chromatogram revealed the presence of cyclic sesquiterpenes with typical mass fragmentation patterns at 105, 119, 161, and 204 m/z, without and with the hydroxyl group (Supplementary Figure S7). The major peaks were predicted to be the large amounts of β-cubenene isomers (>90%) and tiny γ-amorphene, respectively. As a negative control, no terpenoids were detected in the reaction system containing only FPP. Therefore, TaTS9 was then classified as a sesquiterpene synthase that catalyzes the cyclization of FPP to produce β-cubenene.

Cubenene synthase is common in plants like Pyrus spp. that catalyze the biosynthesis of volatile cubenene [29]. Its existence in fungi was reported by Chen et al. [29] for the first time, in which a (+)-cubenene synthase was identified in A. chrysogenum. Due to the very low sequence similarity of fungal sesquiterpene synthases, such as the <30% similarity between TaTS9 and AcTPS3 and 25.82–47.71% similarity of seven TaTSs to known Swiss-Prot homologues (Figure 1), it is inaccurate to predict the functions of new sesquiterpene synthases [20,30]. Moreover, the low yield of sesquiterpenes obtained through direct extraction from the native host, especially the volatile sesquiterpenes, hinders the pace of biochemical characterization. Synthesis of new products through biological instead of chemical approaches is more environmentally friendly and sustainable. Heterologous expression of terpene synthases in engineering strains is more feasible to increase the yields of terpenoids. In the case of AcTPS3, cubenene achieved the highest titer of 597.3 mg/L in the optimized S. cerevisiae so far [28]. However, the application potential of cubenene has not been assayed yet. In contrast, we identified TaTS9 as the key gene in the biosynthesis of anti-fungal sesquiterpenoids through genome mining, antagonism assay, transcript quantification, and VOC metabolome analysis, and we determined its products to be mainly β-cubenene. However, due to the low yield in the present biosynthesis system, the biological activities of β-cubenene against plant pathogens were unassessed. Besides the MVA pathway of S. cerevisiae, an isopentenol utilization IU pathway-dependent (IUPD) strain of S. cerevisiae has been developed, which surpassed the MVA pathway in the growth-coupled production of complex terpenoids [31]. Thus, our future studies would focus on the improvement and optimization of the β-cubenene biosynthesis system for application in the agriculture industry.

3. Materials and Methods

3.1. Strains, Plasmids, and Culture Conditions

Trichoderma afroharzianum ACCC 33109 and Fusarium oxysporum f. sp. cucumerinum (Foc) ACCC 37438 were obtained from the Agriculture Culture Collection of China (ACCC). The chassis strain and mutant strains of ACCC 33109 constructed in this study were conserved in the laboratory stock. The fungal culture were positioned onto Petri dishes with potato dextrose agar (PDA; Oxoid, UK) medium or into potato dextrose broth (PDB; Oxoid) and thereafter subjected to incubation at 28 °C for 5 days. The plasmids, pKS666 and pKH-KO, donated by Drs. Rongjun Guo and Xiliang Jiang from the Institute of Plant Protection, Chinese Academy of Agricultural Sciences, were used in the construction of the gene knockout mutant strains. The vectors pEASY-T3 and pET28a (TransGen, Beijing, China) were used to construct the recombinant plasmids for DNA manipulation and gene cloning, respectively. The Escherichia coli strains Trans1-T1 and BL21 (DE3) were propagated in Luria–Bertani (LB) medium with appropriate antibiotics for plasmid DNA isolation and gene expression, respectively.

3.2. Genome Sequencing and Mining

The genomic DNA of T. afroharzianum ACCC 33109 was extracted using the CTAB method. Two libraries with an insert size of 20 kb and 350 bp were constructed using the SMRTbell^TM Template Kit (Pacific Biosciences, Menlo Park, CA, USA) and Next^® Ultra™ DNA Library Prep Kit (New England Biolabs, Ipswich, MA, USA) following the manufacturers’ instructions, respectively. The PacBio Sequel and Illumina NovaSeq platforms were used to sequence the whole genome at Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). Clean reads (>6000 bp) were preliminarily assembled using the SMRT Link v5.0.1 and corrected using the variant Caller module for fine mapping. Genes were annotated using Genewise, AUGUSTUS v.3, and MAKER v.2 with the default parameters, which were further integrated by GLEAN (https://sourceforge.net/projects/glean-gene/, accessed on 18 March 2025). The gene models were functionally annotated against the NR, GO, KEGG, KOG, Pfam, Swiss-Prot, CAZy, Secondary Metabolism, Secretory, DFVF, P450, PHI, and TCDB databases, respectively. Genes and gene clusters related to terpene biosynthesis were predicted using antiSMASH v.4.0.2 and mapped to the KEGG pathway of terpenoid synthesis. Core genes and specific genes were analyzed by the CD-HIT rapid clustering of similar proteins software with a threshold of 50% pairwise identity and 0.7 length difference cutoff in amino acid. Homologue searches for putative functions and multiple sequence alignments of the 15 TaTS genes from ACCC 33109 were performed using the NCBI blastX and ClustalW, respectively. The unrooted phylogenetic trees were constructed using the maximum likelihood method in MEGA X. The bootstrap value was set as 1000 replicates.

3.3. Expression Induction of the TaTS Genes

The sandwiched method described by Li et al. [32] was employed to induce the TaTSs expression of ACCC 33109 by Foc. Each Petri dish contained 20 mL of PDA medium, and a 6 mm diameter plug of each strain was placed on the center of one of the PDA plates. ACCC 33109 was inoculated 1 day earlier and placed on top of the Foc plate. The sandwiched plates were separated with cellophane, sealed with three layers of Parafilm, and incubated at 28 °C. Uninoculated plates were sandwiched with Foc or ACCC 33109 plates as the control. The colony diameters of Foc were measured 3 days later, and the inhibition rate was calculated using the following formula: inhibition rate (%) = [(C_d−6) – (T_d−6)] × 100%/(C_d−6), where C_d and T_d are the colony diameters on the control Foc plates and on the treated Foc plates. The experiment was repeated three times, and each treatment had ten replicates.

3.4. RT-qPCR Analysis of the TaTS Transcript Profiles

The total RNAs were extracted using the FastPure Plant Total RNA Isolation Kit (Vazyme, Nanjing, China) from the cultures of ACCC 33109 treated with or without Foc. The experiment was performed once for 5 days with three biological replicates. Each biological replicate consisted of two ACCC 33109 cultures independently treated with Foc or blank PDA. The RNAs were treated by DNase I (Ambion, Foster City, CA, USA), followed by purification using the RNeasy Kit (Qiagen, Valencia, CA, USA). The concentration and quality of the purified RNAs were assessed using the A₂₆₀/A₂₈₀ and A₂₆₀/A₂₃₀ absorbance ratios and gel electrophoresis. The RNAs (1 µg) were reverse transcribed to cDNAs in a 20 µL reaction using the HiScript II 1st Strand cDNA Synthesis Kit (Vazyme).

The qPCR primers for the 15 Tasts genes and the β-tubulin gene as the reference were designed using Primer3web v.4.1.0 (http://primer3.ut.ee/, accessed on 18 March 2025) and are listed in Supplementary Table S3. The primer specificity was tested using the genomic DNA and the cDNAs of 3-day-old ACCC 33109 cultures with Foc induction as templates. Each PCR reaction with three technical replicates was performed using the ChamQ SYBR qPCR kit (Vazyme) consisting of 12.5 µL 2× PCR master mix, 1 µL each of forward and reverse primers (10 µM), 2 µL diluted DNA/cDNAs (1:5–10 dilution), and 8.5 µL UltraPure^TM DNase/RNase-free distilled water. The qPCR conditions were as follows: one cycle of 10 min at 95 °C, 40 cycles of 30 s at 95 °C, 30 s at 60 °C, and 45 s at 72 °C, one cycle of 1 min at 95 °C and 30 s at 55 °C, and a final ramp to 95 °C to check the primer dimerization and nonspecific amplification. RNase-free distilled water was used as blank controls. The relative expression levels of the TaTS and β-tubulin genes were calculated using the ViiA^TM7 software (Applied Biosystems, Foster City, CA, USA) and the Comparative ΔΔC_t method [33].

3.5. Construction of the TaST Knockout Strains

Using the highly efficient genetic modification strategy of Trichoderma hypoxylon [25], a chassis strain of ACCC 33109, ΔTalig4::neo, was constructed and used for the creation of other TaTS knockout strains by the double homologous gene replacement approach. Hygromycin B (K547; Amresco, Solon, OH, USA) and G418 (FG401-01; TransGen) were used to evaluate the fungal sensitivity. PDA plates supplemented with 100–800 µg/mL G418 and 10–100 µg/mL hygromycin B were inoculated with a 6 mm diameter culture plug of ACCC 33109 in the center and incubated for 5 days at 28 °C. To yield the maximum amount of protoplasts, fresh mycelia of ACCC 33109 cultured for 24 h in PDB were disrupted with sonication (50 Hz and 60 s) and digested by 20 mg/mL lysing enzyme (L1412; Sigma-Aldrich, St. Louis, MO, USA) for 2–10 h. Protoplasts were suspended in STC solution (1.2 M sorbitol, 10 mM Tris-HCl, pH = 7.5, 50 mM CaCl₂) and enumerated under a microscope. The recovery rate of the protoplasts was determined by serial dilution and inoculation on TB3 agar plates (200 g/L sucrose, 5 g/L yeast extract, 3 g/L casein, and 15 g/L agar).

The oligonucleotide primers used for the construction of the ΔTalig4::neo and ΔTalig4::neoΔTaTS::hph strains are given in Supplementary Table S4. In the first round of the PCR, approximately 800 bp fragments upstream and downstream of the target genes, 1300 bp of the neomycin phosphotransferase gene (neo), and 800 bp of the hygromycin phosphotransferase gene (hph) were amplified from the genomic DNA of ACCC 33109 and plasmids pKS666 and pKH-KO, respectively. The 50 µL PCR reaction systems consisted of 25 µL of 2× PCR buffer, 1 µL each of forward and reverse primers (10 µM), 1 µL of genomic DNA or plasmids, 1 µL of TksGflex DNA Polymerase (TaKaRa, Dalian, China), and 21 µL of distilled water. The PCR conditions were as follows: 5 min at 95 °C, 34 cycles of 10 s at 94 °C, 45 s at 60 °C, and 30 s at 72 °C, and 10 min at 72 °C. The PCR fragments were purified with an EasyPure Quick Gel Extraction Kit (Transgene) and quantified. In the second round of the PCR, the upstream and downstream of the Talig4 gene were ligated with the forward 2/3 and backward 1/3 fragments of neo, respectively, as well as the upstream and downstream of the TaTS gene with the forward 2/3 and backward 1/3 fragments of hph. The 50 µL PCR reaction systems consisted of 25 µL of 2× PCR buffer, 1 µL each of forward and reverse primers (10 µM), 1 µL of Talig4 or TaTS PCR products (100 ng/µL), 1 µL of neo or hph PCR products (100 ng/µL), 1 µL of TksGflex DNA polymerase, and 20 µL of distilled water. The PCR conditions were as follows: 5 min at 95 °C, 34 cycles of 10 s at 94 °C, 90 s at 60 °C, and 30 s at 72 °C, and 10 min at 72 °C. The fused fragments were purified and quantified as described above.

The fused fragments of Talig4::neo and TaTS::hph were mixed at a molar ratio of 1:1, and 10 µL of the mixture were added into 250 µL of ACCC 33109 or ΔTalig4::neo protoplasts (10^7–8 cfu/mL). After 30 min storage on ice, the mixture was added with 1.4 mL of PEG, placed on ice for another 20 min, and transferred into 5 mL of TB3 medium supplemented with 100 µg/mL ampicillin for overnight incubation at 28 °C with a constant agitation of 80 rpm. The liquid cultures were mixed with 50 °C preheated TB3 agar medium containing 200 µg/mL G418, poured into 9 cm in diameter Petri dishes, and covered with 400 µg/mL G418. After incubation at 28 °C for 3 days, the colonies were picked up and subcultured on TB3 agar plates containing 400 µg/mL G418 for another 3 days. Successful deletion of the target genes were confirmed by diagnostic PCR with primers inside and outside the corresponding gene, as shown in Supplementary Table S4. The 50 µL PCR reaction systems consisted of 25 µL of 2× PCR buffer, 1 µL each of forward and reverse primers (10 µM), 1 µL of DNA template (100 ng/µL), 1 µL of TksGflex DNA Polymerase, and 21 µL of distilled water. The PCR conditions were as follows: 5 min at 95 °C, 34 cycles of 10 s at 94 °C, 40 s at 60 °C, and 30 s at 72 °C, and 10 min at 72 °C. Sandwiched cultures with Foc and RT-qPCR were also used to evaluate the effects of gene deletion on the antagonism and TaTS transcript levels of the mutant strains, as described above.

3.6. GC-MS Analysis of the VOC Metabolomes

Fifty microliters of fresh spore suspension (1 × 10⁶ cfu/mL) of the chassis strain ΔTalig4::neo and TaTS knockout strain ΔTalig4::neoΔTaTS::hph were inoculated in 20 mL of headspace bottles containing 3 mL PDA medium and cultured at 28° C for 5 days. Solid-phase microextraction (SPME) was used to collect the volatiles as described previously [22]. The 65 µm PDMS/DVB fiber tip was aged at 250 °C for 30 min, inserted at 1/2 height, and adsorbed volatiles at 28 °C for 20 min. The samples were then analyzed by GC-MS (Agilent QP2010; Santa Clara, CA, USA). The GC-MS conditions were as follows: column, HP-5MS; sample volume, 1 µL; carrier gas, helium; inlet temperature, 280 °C; process, initial temperature of 50 °C for 2 min, increased at 8 °C/min to 180 °C, increased at 10 °C/min to 240 °C, and maintained at 240 °C for 6 min; the ion source, EI; spectrum, 30–450 m/z. The search was conducted using NIST14 and the NIST14 library.

3.7. Protein Expression and Purification

For the prokaryotic expression in E. coli, the mRNA sequence of TaTS9 was codon-optimized by using the ExpOptimizer and synthesized by Shenggong (Shanghai, China). The recombinant plasmid pET28-TaTS9 containing four restriction sites (EcoRI, SacI, HindIII, and NotI), an N-terminal His₆ tag, and a TEV cleavage site was transformed into E. coli BL21 (DE3) and cultured at 37 °C and 220 rpm overnight in 200 mL LB medium supplemented with 100 μg/mL ampicillin. When the OD₆₀₀ of the culture reached 0.6, the protein expression was induced by adding 0.5 mM isopropyl β-D-thiogalactopyranoside (IPTG), followed by further incubation at 160 rpm and 16 °C for 20 h. The cells were harvested, resuspended in 25 mL of buffer A (50 mM NaH₂PO₄, 300 mM NaCl, and 20 mM imidazole; pH 8.0), and disrupted by ultrasonication on ice. The cell debris was removed by centrifugation, and the soluble protein in the supernatant was purified by an Ni-NTA affinity chromatography column containing 100–300 mM imidazole and concentrated by an ultrafiltration centrifugal tube (10 kDa; Millipore, Bedford, MA, USA). SDS-PAGE of 5% stacking gel and 12% separation gel was used to analyze the purity and apparent molecular masses of the purified protein. The protein concentration was measured using the Protein Assay Kit (Bio-Rad) in triplicate.

3.8. In Vitro Enzyme Assay

Farnesyl diphosphate (FPP, Sigma-Aldrich) was used as the terpenoid precursor [34]. The in vitro enzyme assay was performed in a 1 mL reaction system (25 mM Tris-HCl, 5 mM MgCl₂, 100 μg of purified TaTS9 and 100 μM FPP). The assay was incubated at 28 °C and 50 rpm overnight, and the products were extracted twice with 200 μL of hexane. The hexane layers were subjected for GC-MS analysis as described above for the volatolome analysis. The assay reaction without purified TaTS9 was included as a negative control and was also analyzed using GC-MS.

3.9. Statistical Analysis

The data were generated by analysis of variance and Duncan’s multiple comparison using SPSS v.22.0 or the independent samples t-test to determine significant differences (p < 0.05). The graphs were drawn using GraphPad Prism v.8.0.

4. Conclusions

In this study, we identified a key terpene synthase gene, TaTS9, in the biocontrol agent T. afroharzianum ACCC 33109 through genome mining, qPCR analysis, gene knockout, and differential volatolome comparison. The deduced TaTS9 shares a low identity of 34.06% with the functionally characterized longiborneol synthase, and catalyzes the cyclization of FPP to produce β-cubenene. Furthermore, its knockout mutant strain is unable to synthesize 22 kinds of terpenoids (19 sesquiterpene, two diterpene, and one monoterpene) and shows significantly decreased anti-fungal activity against Foc. The results indicate that TaTS9 is crucial to the biosynthesis of anti-fungal sesquiterpenes. This reveals the effectiveness of genome mining and omics in the identification of the key gene involved in the enzyme-catalyzed biosynthesis of bioactive natural products.