VdEGe1, a cell wall-degrading enzyme gene from Verticillium dahliae required for cotton pathogenesis

Abstract

Background

Verticillium wilt, caused by Verticillium dahliae, is a destructive disease threatening global cotton production, leading to severe yield and fiber quality losses. Cell wall-degrading enzymes (CWDEs) are known to be crucial for fungal pathogenicity, but the specific roles of individual endoglucanases V. dahliae virulence remain poorly understood. This study characterized an endoglucanase gene VdEGe1 to elucidate its function and mechanistic contribution to V. dahliae pathogenicity.

Methods

Bioinformatic analyses were performed to identify endoglucanase genes in the V. dahliae genome, followed by phylogenetic clustering. Host-Induced Gene Silencing (HIGS) was employed to specifically inhibit VdEGe1 expression to evaluate the cotton resistance to Verticillium wilt. Knockout mutagenesis in V. dahliae was used to assess the roles of VdEGe1 in fungal growth, development, and pathogenicity. Secretory activity of VdEGe1 was validated using a yeast signal peptide trap system, and Agrobacterium-mediated transient expression in Nicotiana benthamiana were conducted to evaluate cell death induction. Secretome profiling was analyzed to decipher the pathogenic mechanisms associated with VdEGe1.

Results

VdEGe1 was classified into the glycoside hydrolase family 45 (GH45). HIGS of VdEGe1 significantly reduced Verticillium wilt symptoms, disease index, and fungal biomass in infected cotton. The ΔVdEGe1 knockout mutant exhibited impaired colony expansion, hyphal growth, sporulation, and spore germination, accompanied by severely attenuated virulence. VdEGe1 was confirmed as a secreted protein but did not trigger or suppress cell death in N. benthamiana. Secretome analysis identified 311 differentially expressed secretory proteins, with 156 being down-regulated. The VdEGe1 deletion mutant exhibited a significant down-regulation of mitochondrial-related proteins and several carbohydrate-active enzymes, accompanied by a decrease in intracellular ATP levels and a slower growth rate on media containing various carbon sources.

Conclusion

VdEGe1 deletion leads to mitochondrial dysfunction and impaired carbon utilization, which induces energy starvation in V. dahliae, thereby compromising its growth, development and virulence. These findings provide novel molecular insights into V. dahliae-host interactions and lay a theoretical foundation for developing targeted control strategies against cotton Verticillium wilt.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-026-04977-x.

Introduction

Cotton (Gossypium hirsutum L.), the world’s most important natural fiber crop, accounts for about 35% of global fiber supply [1]. Its production is severely threatened by Verticillium wilt, a destructive vascular disease primarily caused by the soil‑borne fungus V. dahliae. This disease is often described as the ‘cancer’ of cotton, leading to significant yield and quality losses in major cotton‑growing regions [2]. The pathogen persists in soil as microsclerotia, which can survive for many years even without a host plant [3]. Infection is initiated when root exudates stimulate the germination of microsclerotia. Following root surface colonization, the fungus typically enters through wounds or natural openings, penetrates the root cortex, and invades the xylem vessels. It then spreads systemically through the vascular system via conidia and mycelial growth, causing vessel blockage and characteristic symptoms including wilting, chlorosis, necrosis, defoliation and, in severe cases, plant death [4, 5]. The durable survival structures and vascular nature of the infection render Verticillium wilt particularly challenging to control, highlighting the importance of elucidating the molecular mechanisms underlying V. dahliae pathogenicity.

The plant cell wall, composed of cellulose, hemicellulose, pectin, lignin and structural proteins, serves as the primary structural barrier against pathogen invasion [6]. To overcome this barrier, fungal pathogens secrete a suite of cell wall‑degrading enzymes (CWDEs) during infection, including cutinases, pectinases, hemicellulases and cellulases, which act synergistically to break down host cell wall components and promote fungal penetration and colonization. Transcriptomic analyses consistently reveal the up‑regulation of numerous CWDE genes during infection, and functional studies have demonstrated their key roles in pathogenicity. For instance, knockout mutants of pectate lyase genes (VdPEL1, VdPL3.1, VdPL3.3) [7, 8], a cutinase gene (VdCUT11) [9], a xylan‑degrading enzyme gene (VdXyn4) [10], and cellulose‑degrading enzyme genes (VdEg‑1, VdEG1, VdEG3) all exhibit significantly reduced virulence [11, 12]. Comparative genomic studies further indicate that V. dahliae harbors an expanded repertoire of carbohydrate‑active enzymes (CAZymes) compared with many other pathogenic fungi [13]; however, only a limited subset of these enzymes has been functionally characterized to date.

Endoglucanases are key components of the cellulose‑degrading enzyme system, catalyzing the cleavage of β‑1,4‑glycosidic bonds within amorphous cellulose regions to produce oligosaccharides and generate new non‑reducing ends. Based on the CAZy database, endoglucanases are distributed into multiple glycoside hydrolase families, including GH3, GH5, GH6, GH7, GH8, GH9, GH10, GH12, GH44, GH45, GH48, GH61, GH74 [14]. Functional studies in various phytopathogens have demonstrated that endoglucanases contribute to virulence in diverse ways. For instance, in Magnaporthe oryzae, the GH12 family endoglucanases MoCel12A and MoCel12B specifically hydrolyze β-1,3 − 1,4-linked glucans; Unexpectedly, knockout of these two genes enhances fungal virulence, suggesting a negative regulatory role in pathogenicity [15]. In Rhizoctonia solani AG1 IA, the GH16 family protein RsEG146 exhibits both canonical enzymatic structural features and effector functions, facilitating cell wall degradation while modulating host immune responses, thereby highlighting the dual role of glycoside hydrolases [16]. In Ustilago esculenta, the GH45 family endoglucanase UeEgl1 is secreted during infection and promotes hyphal proliferation within Zizania latifolia tissues; this gene is significantly upregulated in highly virulent strains, and its overexpression enhances fungal colonization and spread within the host [17]. Beyond their primary role in cell wall degradation, certain cell wall‑degrading enzymes (CWDEs) can also act as pathogen‑associated molecular patterns (PAMPs) that are recognized by plant pattern recognition receptors (PRRs), thereby triggering pattern‑triggered immunity (PTI) [18]. Although this dual functionality of CWDEs is increasingly recognized, the specific roles and mechanisms of endoglucanases in the pathogenesis of V. dahliae remain largely unexplored.

Transcriptomic data from our laboratory revealed that a candidate cell wall‑degrading enzyme (CWDE) gene, designated as VDAG_00931, was significantly up‑regulated during cotton infection by a highly aggressive strain of V. dahliae. This gene, here named VdEGe1, encodes an endoglucanase. In the present study, we first performed a comprehensive phylogenetic analysis of endoglucanase genes in V. dahliae. Subsequently, a combination of experimental approaches—including host‑induced gene silencing (HIGS), targeted gene knockout and complementation, secretory activity validation, pathogen‑associated molecular pattern (PAMP) function assays, and secretome and transcriptome analyses—was employed to systematically investigate the role of VdEGe1 in fungal growth, development, and pathogenicity. Our findings elucidate the molecular mechanisms by which VdEGe1 contributes to fungal virulence and identify potential targets for novel Verticillium wilt control strategies, thereby providing a foundation for future breeding of resistant cotton varieties.

Results

Identifying and phylogenetic analysis of the VdEGe1 gene

Based on previous transcriptome data, an endoglucanase-encoding gene (VDAG_00931) was found to be significantly up-regulated during cotton root infection by highly aggressive V. dahliae and named VdEGe1 [19]. The ORF of VdEGe1 encodes a poly-peptide of 287 amino acids with a predicted molecular weight of 29.71 kDa and a predicted isoelectric point of 9.33. To further elucidate the evolutionary characteristics of VdEGe1, endoglucanase protein sequences from V. dahliae and other pathogenic fungi were used for phylogenetic tree construction. The phylogenetic analysis revealed that these endoglucanase proteins can be classified into six distinct clades: GH7, GH45, GH12, GH5, GH6, and GH61 (Fig. 1). VdEGe1 clusters within the GH45 family, indicating that VdEGe1 is a GH45 family gene.

Fig. 1.

Phylogenetic analysis of endoglucanase genes in V. dahliae

Phylogenetic tree of endoglucanase genes constructed using the Neighbor-Joining method. The tree illustrates the evolutionary relationships among endoglucanase genes from V. dahliae and other fungal species, including Trichoderma reesei QM6a, Trichoderma longibrachiatum, Fusarium xylarioides, Fusarium sp. VM40, Fusarium oxysporum, Fusarium austroafricanum, Fusarium agapanthi, Colletotrichum tofieldiae, Colletotrichum spaethianum, Colletotrichum sojae, Colletotrichum shisoi, Colletotrichum liriopes, Colletotrichum incanum, and Colletotrichum graminicola. The tree was constructed using the Maximum Likelihood method in MEGA 11.0 software with 1000 bootstrap replicates. The six subfamilies of endoglucanases are denoted by six distinct colors. The red star marks the gene VdEGe1 from V.dahliae studied here.

Host-induced gene silencing of VdEGe1 attenuates disease symptoms in cotton

The TRV-mediated HIGS approach was employed to silence the VdEGe1 gene in V. dahliae. When HIGS treated cotton seedlings reached two-true-leaf stage, they were inoculated with Vd991 spore suspension using the root wounding method. Disease symptoms were assessed at 14 and 21 days post-inoculation (dpi). At 14 dpi, the pTRV2:00 treated plants (CK) exhibited pronounced leaf yellowing and wilting, while pTRV2:VdEGe1 treated plants only had mild yellowing (Fig. 2A). At 21 dpi, the CK showed severe defoliation, whereas pTRV2:VdEGe1 treated plants displayed noticeable yellowing, wilting, and limited defoliation (Fig. 2B). Stem dissection at 21 dpi revealed less vascular browning in pTRV2:VdEGe1 plants than in the CK (Fig. 2C). Fungal re-isolation and biomass quantification demonstrated a substantial reduction in fungal load within the pTRV2:VdEGe1 plants (Fig. 2D and G). At both 14 and 21 dpi, the Disease Index of pTRV2-VdEGe1 plants (34.2 and 46.3) were significantly lower than that of CK (39.5 and 63.4) (Fig. 2E). To evaluate the silencing efficiency of the TRV-mediated HIGS system, the expression level of VdEGe1 in the roots, stems, and leaves from pTRV2:VdEGe1 plants at 21 dpi were analyzed by using qRT-PCR. It was found that the expression level of VdEGe1 was significantly lower in pTRV2:VdEGe1 plants compared to the CK (Fig. 2F). In summary, the TRV-mediated HIGS targeting the VdEGe1 gene in V. dahliae effectively restricted fungal biomass accumulation in cotton plants and mitigated disease severity, suggesting that VdEGe1 may play a key role in the pathogenicity of V. dahliae.

Fig. 2.

Host-induced gene silencing of VdEGe1 attenuates Verticillium wilt of cotton. A Disease symptoms in VdEGe1-silenced (pTRV2:VdEGe1) and control (pTRV2:00) cotton plants at 14 dpi. B Disease symptoms at 21 dpi. C Vascular browning in longitudinal stem sections of HIGS-treated plants at 14 dpi (scale bar = 1 mm). D Fungal re-isolation from stem segments of HIGS-treated plants cultured on PDA for 7 days at 25°C. E Disease index in VdEGe1-silenced plants at 14 and 21 dpi. Disease index was calculated as: DI = [∑(disease rating × number of plants at that rating) / (total number of plants × 4)] × 100. F Relative expression of VdEGe1 in roots, stems and leaves of HIGS-treated plants analyzed by qRT-PCR using V. dahliae β-tubulin as the reference gene. G Fungal biomass in different tissues of HIGS-treated plants at 21 dpi detected by qRT-PCR with GhUBQ7 as the reference gene. All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (Mean ± SD). Statistical analysis was performed using IBM SPSS Statistics 26.0, with significance assessed via one-way ANOVA followed by Duncan’s multiple range test. Different lowercase letters on the bars indicate significant differences between treatments at the p<0.05 level

VdEGe1 deletion impairs growth and development of V. dahliae

To elucidate the function of VdEGe1, two knockout mutants (ΔVdEGe1-4, ΔVdEGe1-6) and a complemented strain (ΔVdEGe1-C) were generated in the Vd991 background via Agrobacterium tumefaciens-mediated transformation method and verified (Supplementary Fig. 1). On both CM and PDA media, the knockout mutants exhibited reduced colony diameters compared to the wild-type (WT) and complemented strains (Fig. 3A, B). Further phenotypic analysis revealed severe defects in sporulation and hyphal development in the ΔVdEGe1 mutants: the conidial germination rates were lower (56.7% and 59.67%, respectively) compared to the WT (86.3%) and the complemented strain ΔVdEGe1-C (89%), conidial production in liquid culture was decreased, and hyphal growth was retarded relative to the WT and complemented strains (Fig. 3C-E). Collectively, these results suggested that VdEGe1 is essential for normal colonial expansion, hyphal growth and sporulation in V. dahliae.

Fig. 3.

A Colony morphology of different strains on PDA, CM and Czapek medium after 15 d of growth; B Colony diameters of different strains on PDA, CM and Czapek medium after 15 d of growth; C Spore germination rates of different strains on sterile glass slides after 12 h of culture; D Sporulation quantities of different strains in Czapek liquid medium after 1, 2, 3, 4, and 5 d of culture; E Mycelial growth morphology of different strains on PDA medium after 24 h of culture, Bar=200 μm. Significance analysis was performed using one-way ANOVA followed by Duncan's multiple range test, Different letters indicate significant differences among groups (p < 0.05)

VdEGe1 deletion impairs pathogenicity of V. dahliae on cotton

Pathogenicity assays were performed to characterize the role of VdEGe1 during V. dahliae infection. Cotton plants inoculated with the ΔVdEGe1-4 and ΔVdEGe1-6 knockout mutants displayed significantly milder symptoms compared to plants infected with the wild-type (WT) or the complemented strain ΔVdEGe1-C at 18 and 30 dpi (days post-inoculation) (Fig. 4A). Consistently, the Disease Index (DI) of plants infected with ΔVdEGe1-4 (19.38 and 37.45) and ΔVdEGe1-6 (18.89 and 37.85) mutants were significantly lower than that of CK (36.02 and 58.4) at both time points (Fig. 4G, H). Stem vascular browning was also noticeably attenuated in plants infected with the two knockdown mutants (Fig. 4B). Fungal re-isolation assays and biomass quantification further confirmed a significant reduction in fungal colonization in plants inoculated with ΔVdEGe1 mutants (Fig. 4C–F). Collectively, these results demonstrated that VdEGe1 is required for pathogenicity of V. dahliae.

Fig. 4.

VdEGe1 deletion attenuates the Pathogenicity of V. dahliae on Cotton. A Disease symptoms on cotton seedlings inoculated with wild-type (Vd991), VdEGe1 knockout mutants (ΔVdEGe1-4, ΔVdEGe1-6), and complemented strain (ΔVdEGe1-C) at 18 and 30 dpi. B Vascular browning in longitudinal stem sections at 18 dpi. C Re-isolation of fungi from infected stem segments cultured on PDA for 5 days. D-F Fungal biomass in roots (D), stems (E), and leaves (F) at 30 dpi detected by qRT-PCR with GhUBQ7 as the reference gene. G, H Disease index at 18 dpi (G) and 30 dpi (H). For each treatment, 45 plants were evaluated across three independent replicates.he numbers 0, 1, 2, 3, and 4 represent the disease severity rating scale for Verticillium wilt used in this study, which is defined as follows: Grade 0: no symptoms; Grade 1: 1–25% of leaves wilted; Grade 2: 26–50% of leaves wilted; Grade 3: 51–75% of leaves wilted

Grade 4: 76–100% of leaves wilted or plant dead. Disease index was calculated as: DI = [∑(plants per rating × rating number) / (total plants × 4)] × 100. All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (Mean ± SD). Statistical analysis was performed using IBM SPSS Statistics 26.0, with significance assessed via one-way ANOVA followed by Duncan’s multiple range test. Different lowercase letters on the bars indicate significant differences between treatments at the p < 0.05 level.

VdEGe1 is a secreted protein that does not trigger cell death in tobacco

Bioinformatic prediction using SignalP 6.0 identified a 23-amino acid signal peptide at the N-terminus of VdEGe1 (Fig. 5A), indicating its potential as a secreted protein. Secretory function was experimentally validated via a yeast signal sequence trap system. Yeast transformants carrying pSUC2-VdEGe1^sp grew normally on CMD-W medium and successfully utilized raffinose as the sole carbon source on YPRAA medium. Consistently, positive red precipitation was detected in 2,3,5-triphenyltetrazolium chloride (TTC) assays, confirming the functional secretory function of the VdEGe1 signal peptide (Fig. 5B). To explore the potential role of VdEGe1 in modulating plant immunity, VdEGe1 was transiently expressed in tobacco leaves either individually or co-expressed with the cell death inducer BAX. Phenotypic observation found that VdEGe1 neither triggered cell death when expressed alone nor suppressed BAX-triggered cell death compared to control treatments (Fig. 5C). These findings indicate that VdEGe1 is a secreted protein during V. dahliae infection, but does not act as a cell death elicitor or suppressor in plant immunity.

Fig. 5.

VdEGe1 is a secreted protein that does not trigger cell death in tobacco. A Signal peptide prediction for VdEGe1. Colored curves represent different prediction categories: Sec/SPI signal peptide (pink/orange/yellow) and cleavage site (green). B Validation of secretory function using the yeast signal sequence trap system. Yeast strains carrying pSUC2-VdEGe1^sp, empty pSUC2 (negative control), and pSUC2-Avr1b^sp (positive control) were assessed for growth and invertase secretion. C Cell death induction assay in N. benthamiana leaves. Agrobacterium-mediated transient expression was performed with pGR107:VdEGe1, pGR107:GFP (negative control), and pGR107:BAX (positive control). Phenotypes were recorded at 7 days post-infiltration

VdEGe1 deletion attenuates plant immunity by remodeling the fungal secretome

To investigate the impact of VdEGe1 deletion on secretion of V. dahliae, we conducted a comparative secretome analysis between the wild-type strain Vd991 (WT) and the ΔVdEGe1-6 mutant. A total of 311 differentially expressed secreted proteins (DEPs) were identified, with 156 down-regulated and 155 up-regulated DEPs (Fig. 6A). GO enrichment analysis indicated that down-regulated proteins were significantly enriched in ‘translation’ and ‘organelle organization’ terms in BP category, ‘mitochondrial part’ and ‘mitochondrial inner membrane’ terms in CC category, and ‘RNA binding’ and ‘ribosome binding’ terms in BP category (Fig. 6B).

Fig. 6.

VdEGe1 Deletion Remodels the Fungal Secretome and Attenuates Secretome-induced Plant Immunity. A Volcano plot of differentially expressed secretory proteins. B GO enrichment analysis of down-regulated proteins in the mutant. C Phenotype of tobacco leaves at 24 h post-infiltration (hpi) with secretory proteins from different strains. D H₂O₂ accumulation detected by DAB staining at 24 hpi. E Callose deposition visualized by aniline blue staining. Scale = 200 μm. F Dynamic changes in electrolyte leakage in leaves infiltrated with secretory proteins from different strains. G Relative expression levels of defense-related genes at 12 h post-inoculation. All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (Mean ± SD). Statistical analysis was performed using IBM SPSS Statistics 26.0, with significance assessed via one-way ANOVA followed by Duncan’s multiple range test. Different lowercase letters above the bars indicate statistically significant differences (p < 0.05) between the ΔVdEGe1-6 and Vd991 treatment groups

To determine whether these secretome alterations affect fungal pathogenicity, we compared immune responses triggered by secretory proteins from both strains. Leaves treated with ΔVdEGe1-6 secretory proteins exhibited significantly reduced necrosis, weaker H₂O₂ accumulation (by DAB staining), and less callose deposition compared to those treated with Vd991 proteins (Fig. 6C-E). Electrolyte leakage, an early PTI marker, was also consistently lower in leaves infiltrated with the ΔVdEGe1-6 secretory proteins over 36 h (Fig. 6F). However, the expression levels of defense-related genes (PR5, PAL, NPR1, COI1) in leaves treated with ΔVdEGe1-6 secretory proteins were higher than those in leaves treated with the wild-type Vd991 secretory proteins (Fig. 6G). Together, these results indicate that deletion of VdEGe1 alters the pathogen’s secretome and attenuates the ability of V. dahliae to induce typical PTI-associated responses, which is consistent with the reduced pathogenicity observed in the ΔVdEGe1-6 mutant.

VdEGe1 deletion impairs mitochondrial function

GO enrichment analysis revealed that significant enrichment of down-regulated of proteins associated with cellular component terms ‘mitochondrial part’ and ‘mitochondrial inner membrane’ (Fig. 6B), suggesting possible mitochondrial impairment in the ΔVdEGe1 mutant. These two terms encompassed eight proteins, all showing significantly lower FPKM values in the ΔVdEGe1 mutant than in the wild-type strain Vd991 (Fig. 7A). Notably, 6 of these proteins (cytochrome c domain-containing protein, single-stranded DNA-binding protein, mitochondrial acidic protein MAM33, Mitochondrial phosphate carrier protein, mitochondrial outer membrane protein porin, mitochondrial phosphate carrier protein) are known to be closely associated with mitochondrial structure and core functions, leading us to hypothesize that mitochondrial dysfunction occurs in the ΔVdEGe1 mutant. To test this, we measured intracellular ATP levels. It was found that the ΔVdEGe1 mutants (ΔVdEGe1-4 and ΔVdEGe1-6) exhibited significantly reduced ATP content relative to the wild type, whereas the complemented strain (ΔVdEGe1-C) restored ATP production to wild-type levels (Fig. 7B). qRT-PCR analysis further confirmed that transcript levels of the corresponding genes for these six key proteins were significantly down-regulated in the ΔVdEGe1 mutant (Fig. 7C-H). Collectively, these results demonstrate that deletion of VdEGe1 disrupts mitochondrial function by suppressing the expression of mitochondria-associated genes, resulting in decreased ATP synthesis and thereby impairing fungal growth and pathogenicity.

Fig. 7.

VdEGe1 deletion impairs mitochondrial function. A Heatmap showing the expression levels (FPKM values) and fold-change (FC) of mitochondrial-related proteins in the ΔVdEGe1-6 mutant versus wild-type Vd991. B Intracellular ATP content in the wild-type Vd991, knockout mutants (ΔVdEGe1-4 and ΔVdEGe1-6) and complemented strain (ΔVdEGe1-C). C–H Relative expression levels of six mitochondrial-related proteins in different strains, detected by qRT-PCR. All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (Mean ± SD). Statistical analysis was performed using IBM SPSS Statistics 26.0, with significance assessed via one-way ANOVA followed by Duncan’s multiple range test. Different lowercase letters on the bars indicate significant differences between treatments at the p < 0.05 level

VdEGe1 deletion leads to the down-regulation of CAZymes and Small Cysteine-Rich Proteins (SCRPs)

The down-regulated secreted proteins in the ΔVdEGe1-6 mutant included 88 classical secreted proteins (CSPs) and 68 non-classical secreted proteins (NCSPs). Among the CSPs, we identified 14 CAZymes and 27 SCRPs (SCRPs, < 400 amino acids with ≥ 4 cysteine residues) (Fig. 8A). The 14 CAZymes were further classified into 2 glycoside hydrolases, 9 auxiliary activities, 2 carbohydrate esterases, and 1 polysaccharide lyase (Fig. 8B). Functional annotation indicated that 5 of these proteins are involved in plant cell wall degradation, including pectinases, cellulases, hemicellulases (Fig. 8C). This down-regulation of CWDEs suggests a reduced capacity for carbon source utilization of the ΔVdEGe1-6 strain. To test this, we cultured the ΔVdEGe1-6 mutant on media with different sole carbon sources. Compared to the wild-type and complemented strains, the ΔVdEGe1-6 mutant showed significantly impaired growth on methylcellulose, glucose, sucrose, fructose, xylose, and pectin (Fig. 9A, B). These results confirm that VdEGe1 is required for carbon source utilization in V. dahliae, and the down-regulation of cell wall-degrading CAZymes likely contributes to the attenuated pathogenicity of the mutant.

Fig. 8.

VdEGe1 deletion results in the down-regulated of CAZymes and SCRPs. A Number and functional classification of down-regulated differentially expressed proteinsin the ΔVdEGe1-6 mutant. SPs:total secreted proteins; CSPs: classical secreted proteins; N-CSPs: non-classical secreted proteins; SCRPs: small cysteine-rich proteins; CAZymes: carbohydrate-active enzymes; PHI: pathogen-host interaction-related proteins. B Pie chart showing the family distribution of CAZymes. GHs: glycoside hydrolases; PLs: polysaccharide lyases; AAs: auxiliary activities; CEs: carbohydrate esterases. C Heatmap showing the FPKM and fold change of CAZymes in the ΔVdEGe1-6 mutant versus wild-type Vd991. Only genes with a fold change ≥ 1.2 are displayed. D Heatmap showing the FPKM and fold change of SCRPs in the ΔVdEGe1-6 mutant versus wild-type Vd991. Blue represents known functional proteins; orange represents uncharacterized proteins

Fig. 9.

VdEGe1 is required for efficient carbon source utilization in V. dahliae. A Colony morphology of the wild-type Vd991, VdEGe1 knockout mutants (ΔVdEGe1-4, ΔVdEGe1-6), and complemented strain (ΔVdEGe1-C) on Czapek-Dox medium containing different carbon sources at 15 days post incubation. “MC” denotes Methyl Cellulose. B Colony diameters under each carbon condition. All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (Mean ± SD). Statistical analysis was performed using IBM SPSS Statistics 26.0, with significance assessed via one-way ANOVA followed by Duncan’s multiple range test. Different lowercase letters on the bars indicate significant differences between treatments at the p < 0.05 level

To strengthen the causal relationship between CWDEs down-regulation and impaired carbon source utilization, we further quantified the correlation between changes in protein expression and the growth phenotypes of ΔVdEGe1 mutants. For the pectin degradation pathway, three pectinase enzymes were down-regulated by more than 2.33-fold (Fig. 8C). The colony diameter of ΔVdEGe1 mutants was reduced by 0.28 cm on pectin-supplemented media, directly reflecting compromised pectin utilization due to down-regulation of pectinase enzymes. For the cellulose degradation pathway, one cellulase enzyme showed a significantly reduction of 2.2-fold (Fig. 8C), and this was associated with a pronounced 0.67 cm reduction in colony diameter on methylcellulose-supplemented media. Collectively, these results establish a direct link between the down-regulation of CWDEs and the reduced carbon source utilization capacity of ΔVdEGe1 mutants. Additionally, 27 DEPs were annotated as SCRPs. These include known effectors such as CFEM domain-containing extracellular membrane protein, AAI-like domain-containing proteins, hydrophobin and chitin-binding protein. However, the majority represent novel effector candidates that have not yet been functionally characterized (Fig. 8D).

Discussion

The plant cell wall is the primary barrier for pathogenic fungi, and successful colonization by these microbes riles heavily on CWDEs [20]. In this study, we functionally characterized VdEGe1, a GH45 family endoglucanase from V. dahliae, by using knockout analysis. In contrast to classic GH45 members such as Uwtilago esculenta UeEgl1, which solely mediate cell wall degradation [17], VdEGe1 acts as a regulatory CWDE coordinating cell wall degradation, mitochondrial function and other physiological processes in V. dahliae. Comparisons with reported V. dahliae endoglucanases VdEG1/VdEG3 from GH12 family, further highlight the unique nature of VdEGe1. Either VdEG1 or VdEG3 deletion resulted in enhanced fungal pathogenicity [12], while VdEGe1 deletion led to comprehensive virulence attenuation; VdEG1/VdEG3 act as PAMPs to directly induce PTI, whereas VdEGe1 does not induce or suppress plant cell death. Instead, VdEGe1 modulates plant immunity indirectly via secretome remodeling. Collectively, our studies on VdEGe1 knockout mutants establish its critical role in fungal growth, development and pathogenicity. Additionally, we performed HIGS assays to further evaluate its potential application value in plant disease resistance and provide additional evidence fro its function. HIGS, an RNA interference (RNAi)-based technology, has been widely applied in crop protection [21–23]. This technology enhances disease resistance by expressing pathogen-specific double-stranded RNA or hairpin in plants, which is subsequently taken up by invading pathogen to trigger RNAi, inhibit homologous gene expression, impair pathogen virulence and thus enhance plant desease resistance. Consistent with previous reports [24–26], our results showed that VdEGe1 silencing via HIGS obviously alleviated cotton Verticillium wilt symptoms (Fig. 2), highlighting its potential as a target for disease control.

The significant down-regulation of key mitochondrial proteins in the ΔVdEGe1 mutant, including cytochrome c domain-containing protein involved in electron transport [27], single-stranded DNA-binding protein essential for mtDNA replication and repair [28], mitochondrial acidic protein MAM33 associated with matrix organization and metabolic homeostasis [29], mitochondrial phosphate carrier protein critical for ATP/ADP and phosphate exchange [30], and mitochondrial outer membrane porin regulating metabolite flux [31], collectively points to profound mitochondrial dysfunction. These proteins play integral roles in oxidative phosphorylation, organellar genome maintenance, ion homeostasis, and metabolite transport [32, 33]. Their coordinated suppression likely disrupts the electron transport chain, impairs ATP synthesis, and compromises mitochondrial integrity. Consistent with this model, direct measurement of intracellular ATP content confirmed a sharp decline in the ΔVdEGe1 mutant, while complemented strain restored ATP to wild-type levels. This systemic energy deficit provides a mechanistic explanation for the observed pleiotropic defects, ultimately underpinning the impaired vegetative growth and reduced pathogenicity in the ΔVdEGe1 mutant.

Emerging evidence suggests that impaired cell wall‑degrading enzyme (CWDE) function can disrupt mitochondrial performance through metabolic or signaling networks, a conserved phenomenon in fungi. In Trichoderma reesei, deletion of a β‑glucanase GLU1 gene alters cell wall β‑glucan content, which activates the cell wall integrity pathway and represses mitochondrial ATP synthase expression [34]. Similarly, In Aspergillus nidulans, deletion of an endoglucanase EngA gene resulted in abnormal cell wall structure, increased β‑1,3‑glucan content, reduced intracellular ATP levels, and decreased mitochondrial respiration rate [35] (Fig. 7). Transcriptomic analyses across plant‑pathogenic fungi further confirm that loss of key CWDEs down-regulates energy metabolism and oxidative phosphorylation pathways, highlighting the tight functional coupling between cell wall degradation and mitochondrial energy supply [36–38]. These studies collectively indicate that efficient cell wall degradation is energy‑intensive and relies on coordinated mitochondrial function, governed by conserved regulatory pathways. In this context, the ΔVdEGe1 mutant, which exhibits coordinated down-regulation of mitochondrial proteins and a sharp drop in ATP levels, provides direct evidence that the endoglucanase gene VdEGe1 contributes to mitochondrial function.

The ΔVdEGe1 mutant exhibited significant down-regulation of five key plant cell wall-degrading proteins and impaired growth on multiple carbon sources, directly linking this gene to carbon utilization via the CWDE system (Figs. 8 and 9). This finding underscores the central role of CWDEs, which degrade complex plant polysaccharides and remodel fungal cell walls, in nutrient acquisition and colonization [39]. The coordinated change in multiple CWDE genes upon deletion of a single gene reflects an evolutionarily conserved regulatory mechanism in fungi. For instance, in V. dahliae, deletion of a xylosidase-encoding gene VdxyL3 up-regulated a suite of CWDE genes, enhancing carbon utilization, growth, and pathogenicity [19]. In Trichoderma reesei, deletion of an extracellular major β-glucosidase-encoding gene bgl1 broadly affects the coordinated regulation of the entire cellulase system at the transcriptional level [40]. In Penicillium decumbens, deletion of an intracellular β-glucosidase gene bgl2 not only leads to a significant decrease in intracellular β-glucosidase activity but also causes a coordinated and marked increase in the production of multiple extracellular lignocellulose-degrading enzymes [41]. The multi-CWDE downregulation and carbon metabolism defects in ΔVdEGe1 align with these cross-species patterns, supporting a pivotal role for VdEGe1 in co-regulating cell wall degradation, carbon metabolism, and pathogenicity.

Our study uncovered an intriguing phenomenon: while secretory proteins from the ΔVdEGe1 mutant triggered attenuated early PTI responses—including reduced H₂O₂ accumulation, callose deposition, and electrolyte leakage—they notably induced upregulation of the defense-related genes PR5, PAL, NPR1, and COI1 in tobacco leaves (Fig. 6G). This indicates that deletion of VdEGe1 may remodel the fungal secretome in a way that weakens canonical PAMP-triggered immunity while simultaneously activating alternative defense-signaling pathways. The up-regulation of PR5 and PAL suggests engagement of the salicylic acid (SA) pathway, whereas induction of NPR1 and COI1 points to involvement of both SA and jasmonic acid (JA) signaling. Such a response could be interpreted in two ways: (i) as a compensatory mechanism whereby the plant, perceiving a weakened pathogen threat, reallocates resources from energetically costly early PTI toward sustained systemic defense via SA/JA-mediated gene expression; or (ii) as a consequence of reduced secretion of fungal suppressors that normally repress host defense transcription. Similar immune rewiring has been reported in other pathosystems, where impaired pathogen virulence enhances SA-dependent gene expression without strong concurrent activation of early PTI markers [42, 43]. Future studies measuring SA/JA levels in cotton upon ΔVdEGe1 infection, or employing SA/JA signaling mutants, could help elucidate the regulatory logic underlying this transcriptomic response.

Collectively, Our study reveals that deletion of the VdEGe1 gene in V. dahliae triggers secretome reprogramming, which disrupts the pathogen’s physiological functions through two key pathways: by down-regulating mitochondrial-related proteins, leading to mitochondrial dysfunction and insufficient energy supply; and by suppressing the expression of CWDEs, markedly reducing the pathogen’s carbon utilization capacity (Fig. 10). These dual impairments collectively inhibit the growth, development, and pathogenicity of V. dahliae on cotton, thereby elucidating the role of VdEGe1 as a critical regulatory factor in the pathogenicity of this fungus. Limitations of this study include unclear signaling mechanisms between VdEGe1 and mitochondria, as well as uncharacterized interactions with cotton immune components. Further investigations will be conducted in future work to address these issues.

Fig. 10.

A proposed model depicting how VdEGe1 deletion affects the growth, development, and pathogenicity of V. dahliae

Conclusion

Collectively, our findings demonstrate that VdEGe1, a GH45 family endoglucanase gene, plays a multifaceted role in the colony expansion, hyphal growth, sporulation, conidial germination and virulence of V. dahliae. Although secreted, VdEGe1 does not trigger or suppress cell death in N. benthamiana. Instead, it modulates host immunity by remodeling the fungal secretome. VdEGe1 acts as a GH45 family endoglucanase directly involved in extracellular plant cell wall degradation, and also function as a key regulatory factor: its dedeletion down-regulates mitochondrial-related proteins and CWDEs, leading to mitochondrial dysfunction and impaired carbon utilization, thus disrupting fungal growth and virulence. These findings offer novel molecular insights into the interactions between V. dahliae and its host and provide a theoretical basis for developing targeted strategies to control cotton Verticillium wilt.

Materials and methods

Vector, plant and microbial materials

The V. dahliae strain Vd991, Agrobacterium tumefaciens strains GV3101, LBA4404, and EHA105, the upland cotton (Gossypium hirsutum) cultivar ‘Xinluzao 7’ (susceptible to Verticillium wilt), and N. benthamiana were all provided by the Cotton Research Institute, College of Agriculture, Shihezi University, Xinjiang, China. The VIGS vectors pTRV1, pTRV2, and pTRV2:GhCHLI were kindly provided by Prof. Longfu Zhu (Huazhong Agricultural University). The plant expression vector pGR107, knockout vector pGKO-HPT, and complemented vector pCAMBIA1302-neo were provided by Res. Hongjie Feng (Institute of Cotton Research, Chinese Academy of Agricultural Sciences).

Cotton cultivation and fungal suspension preparation

Soil-based cultivation. Seeds of the upland cotton cultivar ‘Xinluzao 7’ were germinated on four layers of moist gauze in boxes at 25 °C for 36–48 h. Seedlings with hypocotyls of 1–2 cm length were used for subsequent cultivation. Seedlings were then planted into plastic pots (10 cm diameter) and maintained in a growth chamber under a 16 h/8 h (light/dark) photoperiod, at 25 ± 1 °C, and 60–70% relative humidity.

Hydroponic cultivation. Seeds were surface‑sterilized with 75% ethanol for 1 min, followed by 5% sodium hypochlorite for 15 min, and then thoroughly rinsed with sterile water before germination. Germinated seedlings were transferred to a hydroponic setup and secured with sterile sponges. Roots were immersed in half‑strength Hoagland nutrient solution (pH 5.8–6.0). Plants were grown under the same environmental conditions as described for soil‑based cultivation. The nutrient solution was replaced every 5 days.

To prepare the V. dahliae spore suspension, strain Vd991 was cultured in 200 mL of Czapek-Dox liquid medium at 25 °C with shaking at 150 rpm/min for 5–7 days. The mycelia were then removed by filtration through four layers of sterile gauze. The conidial concentration was determined using a hemocytometer and adjusted to 1 × 10⁷ conidia/mL. The resulting suspension was stored at 4 °C for subsequent inoculations.

Nucleic acid extraction and cDNA synthesis

Total DNA from V. dahliae was extracted using the Fungal DNA Kit (Omega Bio-tek, USA). Total RNA was isolated from V. dahliae and cotton tissues with the Fungal RNA Kit (Omega Bio-tek, USA) and EASYspin Plus Plant RNA Kit (Aidlab, China), respectively. Genomic DNA contamination was removed from all RNA samples, followed by reverse and reverse transcription using the One-step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, China).

Bioinformatics analysis

Transmembrane domains of the VdEGe1 protein were predicted using TMHMM-2.0 (https://services.healthtech.dtu.dk/services/TMHMM-2.0/), and its signal peptide was predicted using SignalP-5.0 (https://services.healthtech.dtu.dk/services/SignalP-5.0/). Basic properties of protein, including amino acid length, molecular weight, and isoelectric point, were predicted using the Sequence Manipulation Suite. The genomic, gene annotations (GFF3 format), and protein sequences of V. dahliae (assembly ASM15067v2) were obtained from the Ensembl Fungi database. All putative endoglucanase genes were identified by screening the GFF3 annotations with TBtools. Multiple sequence alignment was performed with Clustal W program in MEGA11. A phylogenetic tree was then constructed using the Neighbor‑Joining method with 1000 bootstrap replicates and visualized using EvolView (https://evolgenius.info/).

Host-induced gene silencing (HIGS)

The coding sequence (CDS) of VdEGe1 was retrieved from the NCBI database. A specific interfering fragment was amplified from V. dahliae strain Vd991 cDNA using primers HIGS-VdEGe1-F/R (Supplementary Table 1), which were designed with Primer3 Plus and contained 15–20 bp homologous arms matching the termini of BamH I- and EcoR I-linearized pTRV2 vector. The purified PCR product was cloned into the linearized pTRV2 vector via homologous recombination using the ClonExpress II One Step Cloning Kit-C112 (Vazyme, Nanjing). The recombinant plasmid was verified and introduced into A. tumefaciens strain GV3101 by electroporation. Agrobacterium cultures harboring pTRV1, pTRV2 (empty vector), pTRV2:VdEGe1, or pTRV2:GhCHLI were prepared, mixed at a 1:1 ratio (pTRV1 with each pTRV2-derived vector), and infiltrated into cotton leaves as described previously [44]. Two weeks post-infiltration, the efficiency of VIGS was confirmed by observing the photobleaching phenotype in pTRV2:GhCHLI-infiltrated plants. For pathogen challenge, cotton plants at two-true-leaf stage were inoculated with prepared conidial suspension (1 × 10⁷ conidia/mL) using a root-wounding method [45].

For detection of gene silencing efficiency, total DNA was extracted from root, stem, and leaf tissues at 21 days post-inoculation (dpi), and subjected to quantitative real‑time PCR (qRT‑PCR) using the perfectStart™ Green qPCR SuperMix on a Roche LightCycler 480 II system. The relative expression level of VdEGe1 was determined using the V. dahliae β‑tubulin gene (VDAG_10074) as an internal reference, with gene-specific primers qRT‑PCR‑VdEGe1‑F/R (Supplementary Table 1). Thermal cycling conditions were: 95 °C for 10 s; 45 cycles of 95 °C for 10 s, 60 °C for 15 s, 72 °C for 20 s. Relative expression levels were calculated using the 2^⁻ΔΔCT method [46].

Disease resistance evaluation

Disease symptoms were assessed, and the disease index (DI) was recorded at 14 and 21 dpi using a 0–4 rating scale [47]. The DI was calculated as follows: DI = [Σ (number of plants per grade × grade number) / (total number of plants × 4)] × 100% [48]. At 14 dpi, stem segments (⁓3 cm) above the roots were collected, longitudinally split, and vascular browning was examined and photographed under a stereomicroscope.

For pathogen re‑isolation, stems from symptomatic plants at 21 dpi were surface‑sterilized with 5% NaOCl for three 10-min intervals, thoroughly rinsed with sterile water, trimmed at both ends, and cut into 0.5 cm segments. These segments were plated on potato dextrose agar (PDA) medium and incubated at 25 °C for 3–7 days to observe fungal growth. Fungal biomass was quantified using ITS1‑F/ST‑Ve1‑R as primers and the cotton GhUBQ7 gene (DQ116441.1) as internal reference. The performance and data processing of qRT-PCR were as described above.

Construction of VdEGe1 knockout and complemented strains

For the knockout construct, approximately 1 kb upstream and downstream flanking sequences of VdEGe1 were amplified from Vd991 genomic DNA (Supplementary Table 1) and inserted into the Pac I-linearized vector PGKO-HPT using the ClonExpress II One Step Cloning Kit-C115 (Vazyme, Nanjing). The resulting plasmid was introduced into A. tumefaciens strain EHA105 and subsequently transformed into V. dahliae. Positive transformants were selected on PDA plates containing hygromycin and verified by PCR with primers Hyg‑F/R. For complementation, a fragment containing the native upstream region and the full-length VdEGe1 coding sequence was cloned into pCAMBIA1302, yielding the plasmid Up‑VdEGe1‑Down‑pCAMBIA1302. This construct was introduced into A. tumefaciens and then transformed into the corresponding knockout mutants to generate complemented strains. Both the knockout mutants and complemented strains were generated via Agrobacterium tumefaciens-mediated transformation (ATMT) as previously described [49, 50].

Phenotypic characterization of VdEGe1 deletion mutants

Phenotypic assays were performed using the wild‑type strain Vd991 as the control, and the deletion mutants (ΔVdEGe1‑4 and ΔVdEGe1‑6) and the complemented strain (ΔVdEGe1‑C) as test strains. All strains were cultured in 50 mL Czapek‑Dox liquid medium at 25 °C with shaking (200 rpm) for 5–7 days. Conidia were collected by filtration through sterile gauze and adjusted to 1 × 10⁷ or 1 × 10⁵ conidia/mL with sterile water for subsequent assays.

Colony morphology and growth. A 10 µL droplet of the 1 × 10⁷ conidia/mL suspension was spot‑inoculated onto the center of PDA, CM (complete medium), and Czapek‑Dox solid plates. Plates were incubated at 25 °C in the dark for 15 d, after which colony morphology was recorded and colony diameter measured.

Hyphal growth. A 10 µL aliquot of the 1 × 10⁵ conidia/mL suspension was centrally inoculated onto PDA plates. Hyphal growth at the colony edge was examined microscopically after 24 h incubation.

Spore germination. A 10 µL droplet of the 1 × 10⁵ conidia/mL suspension was placed on a sterile glass slide and incubated at 25 °C under high humidity for 12 h. Conidial germination was observed using an optical microscope (Leica DM750), and the germination rate was calculated.

Sporulation dynamics. A 500 µL aliquot of the 1 × 10⁷ conidia/mL suspension was inoculated into 50 mL Czapek‑Dox liquid medium and incubated at 25 °C with shaking (220 rpm) in the dark. From day 1 to day 5 post‑inoculation, spore concentrations were quantified daily using a hemocytometer.

All experiments were conducted with three independent biological replicates.

Pathogenicity assays of VdEGe1 deletion mutants

Pathogenicity of the wild‑type strain Vd991, deletion mutants (ΔVdEGe1‑4 and ΔVdEGe1‑6), and complemented strain (ΔVdEGe1‑C) was evaluated on hydroponically grown cotton plants (cultivar ‘Xinluzao 7’) at the two‑true‑leaf stage. The roots were immersed in conidial suspension (1 × 10⁷ conidia/mL) of each strain for 1 h. Mock‑inoculated controls were treated with sterile water. The inoculated seedlings were transplanted into hydroponic boxes, with 12 plants per box and maintained under standard growth conditions. Disease symptoms were recorded at 18 and 30 days post‑inoculation (dpi), and the disease index (DI) was calculated using a 0–4 rating scale as described above.

Transient expression in N. benthamiana

The coding sequence of VdEGe1 was cloned into the plant expression vector pGR107 for transient expression assays. The full‑length CDS was amplified from V. dahliae strain Vd991 cDNA using primers pGR107‑VdEGe1‑F/R (Supplementary Table 1). These primers were designed with Primer3 Plus and extended with 15–20 bp homology arms matching the Xba I‑ and BamH I‑linearized vector ends. The amplified fragment was inserted into linearized pGR107 by using ClonExpress II One Step Cloning Kit (Vazyme, Nanjing). The resulting plasmid (pGR107:VdEGe1) was verified by colony PCR and sequencing, then introduced into Agrobacterium tumefaciens strain GV3101 by electroporation. For agroinfiltration assays, Agrobacterium strains carrying pGR107:VdEGe1, pGR107:GFP (negative control), or pGR107:BAX (positive control) were grown in LB medium with appropriate antibiotics at 28 °C to the late‑logarithmic phase. The bacterial cells were harvested, resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES, 150 µM acetosyringone), and adjusted to an OD₆₀₀_nm of 0.8–1.0. The resuspended cells were incubated at 28 °C in the dark for 3 h prior to infiltration. The abaxial surface of leaves from 4‑week‑old N. benthamiana plants was infiltrated using a needle‑less syringe. Necrosis symptoms on the leaves was monitored and recorded at 5–9 days post‑infiltration.

Yeast signal sequence trap system assay

To assess the secretory function of the VdEGe1 signal peptide, the full‑length sequence of VdEGe1 was amplified using primers pSUC2‑VdEGe1^SP‑F/R (Supplementary Table 1) with 5′‑terminal homology arms complementary to the EcoR I- and Xho I‑linearized pSUC2 vector. The amplified fragment was inserted into pSUC2 via homologous recombination to generate the recombinant plasmid pSUC2‑VdEGe1^sp. The construct was transformed into E. coli DH5α, and positive clones were verified by colony PCR and sequencing. The verified plasmid was then introduced into the yeast strain YTK12. Positive transformants were selected on CMD‑W plates and cultured in liquid CMD‑W medium at 30 °C with shaking. Yeast cultures of the transformant (pSUC2‑VdEGe1^sp), empty‑vector control (pSUC2), and positive control (pSUC2‑Avr1b^sp) were spot-inoculated onto CMD‑W (tryptophan‑deficient) and YPRAA (raffinose‑based) plates. Growth was evaluated after incubation at 30 °C for 1–2 days. Growth on YPRAA indicates functional signal peptide-mediated invertase secretion, which enables enables yeast to utilize sucrose as the carbon source [51]. Secretory activity was further evaluated using 2,3,5‑triphenyltetrazolium chloride (TTC) staining. Yeast cells were harvested, washed, and incubated in a reaction mixture containing 5% sucrose and 1% TTC. A functional signal peptide facilitates SUC2 secretion, which reduces colorless TTC to a red, insoluble formazan precipitate [52, 53].

Secreted protein profiling

V. dahliae strains Vd991 and ΔVdEGe1‑6 were individually inoculated into 200 mL of Czapek‑Dox liquid medium and cultured at 25 °C with shaking at 220 rpm for 7 days. The culture broth was centrifuged (8,000 × g, 10 min, 4 °C), and the supernatant was collected, vacuum‑filtered through a 0.22 μm membrane to remove mycelial debris and other impurities, and subsequently concentrated using a 3 kDa molecular weight cutoff ultrafiltration device. The secreted proteins were then identified by label‑free quantitative proteomics. This approach is based on MS1‑level quantification by integrating the chromatographic peak areas of peptide signals in LC‑MS runs. For statistical analysis of quantitative results, only proteins with at least two valid values across three biological replicates were retained. Proteins meeting the criteria of |fold change| > 1.2 and P < 0.05 were considered differentially expressed [54].

Gene Ontology (GO) enrichment analysis was performed using the online Microbiology Letter Mapping tool. In addition, secreted proteins were predicted based on SignalP 6.0, TMHMM 2.0, and SecretomeP 2.0 [55–57]. The prediction of PHI homologs was performed based on the PHI database [58]. Annotation of putative CAZymes was performed using a Hidden Markov Model (HMM) routine based on the Carbohydrate-Activity-enzymes database [59].

Protein extraction and plant immune response assays

Secreted proteins were precipitated from 100 mL supernatants using 56.8 g ammonium sulfate. The pellet was resuspended in 500 µL PBS, and protein concentration was determined with the Easy II Protein Quantitative Kit (TransGen Biotech; BCA method). All protein samples were adjusted to 0.3 µM for subsequent assays .

ROS burst detection: Reactive oxygen species (ROS) accumulation was detected using 3,3′‑diaminobenzidine (DAB) staining, which produces a brown precipitate in the presence of H₂O₂ [60]. Leaves of 4‑week‑old N. benthamiana plants were infiltrated with 50 µL of 0.3 µM protein solutions from Vd991 or ΔVdEGe1‑6. After 3 h of incubation, leaf discs were vacuum‑infiltrated with 1 mg/mL DAB (pH 3.8), stained in the dark for 1 h, and then decolorized in ethanol for observation.

Electrolyte leakage measurement: Leaves of 4-week-old N. benthamiana plants were infiltrated with 50 µL of 0.3 µM protein solutions from either Vd991 or ΔVdEGe1-6 using a 1 mL needleless syringe, then returned to the greenhouse. Leaf discs were collected at 0, 1, 3, 6, 12, 24, and 36 h post-infiltration, immersed in deionized water, and electrolyte leakage was assessed by measuring conductivity as previously described [61].

Analysis of defense‑related gene expression: Leaves infiltrated with the protein solutions (as described above) were harvested 12 h post-treatment, flash-frozen in liquid nitrogen, and stored at − 80 °C. Total RNA was extracted using the EasyPure Plant RNA Kit (TransGen Biotech). The expression levels of defense-related genes (Actin, PR5, PAL, NPR1, and COI1) were analyzed by quantitative real-time PCR (qRT-PCR) [62].

Observation of callose deposition: Following infiltration with the protein solutions, leaf samples were immersed in 0.1% aniline blue staining solution (150 mM K₂HPO₄, pH 9.5) and stained in the dark for 3 h. Callose deposits were then visualized and examined under a fluorescence microscope [63].

qRT‑PCR analysis

To validate the secretome sequencing results at the transcriptional level, the wild-type Vd991 and the ΔVdEGe1 mutant cultured under identical conditions (medium, temperature, shaking speed, and duration) were collected. Total RNA was extracted and reverse-transcribed into cDNA. Using the V. dahliae β-tubulin gene as an internal reference, the relative expression levels of six mitochondrial-related proteins in each strain were analyzed by quantitative real-time PCR (qRT‑PCR). The qRT‑PCR experimental procedure and data analysis followed the methods described in Sect.  5.5 of this paper, and the primer sequences used are listed in Supplementary Table 1.All primers were designed using Primer Premier 5.0 software following these principles: length 18–24 bp, annealing temperature (Tm) 58–62℃, GC content 40–60%, and amplicon length 80–200 bp. All primers were designed to span exon–exon junctions to avoid genomic DNA amplification. Primer specificity was verified using the NCBI BLAST tool to ensure exclusive matching with the target gene sequences, thereby guaranteeing the accuracy and reproducibility of qRT-PCR results.

ATP content assay

To measure ATP content, spore suspensions (1 × 10⁷ conidia/mL) of the wild‑type strain Vd991, deletion mutants ΔVdEGe1‑4 and ΔVdEGe1‑6, and the complemented strain ΔVdEGe1‑C were separately cultured in 200 mL Czapek‑Dox liquid medium at 25 °C with shaking (150 rpm) for 7 days. Mycelia were harvested by centrifugation (4,000 rpm, 5 min). Precisely 0.1 g of fresh mycelium was homogenized in 1 mL ice‑cold distilled water using a tissue homogenizer on ice. The homogenate was heat‑treated at 100 °C for 5 min, then centrifuged at 8,000 × g for 15 min at 4 °C. The supernatant was collected for ATP quantification.

ATP content was determined by a creatine kinase‑coupled enzymatic assay, in which creatine kinase catalyzes the conversion of ATP and creatine to phosphocreatine [64]. The reaction product was quantified using phosphomolybdic acid colorimetry; absorbance was measured at 700 nm after wavelength scanning to confirm the peak absorption.

Carbon source utilization analysis

To examine the carbon utilization capacity of different strains, glucose (50 g/L), sucrose (50 g/L), xylose (50 g/L), fructose (10 g/L), pectin (50 g/L) or methylcellulose (50 g/L). was added individually into Czapek Dox media without carbon sources [26]. Carbon-free Czapek Dox media were used as controls. Drops of conidial suspension (1 × 10⁷ conidia/mL) were inoculated on Czapek Dox media with or without carbon sources and incubated at 25◦C.The colony diameters of all strains were measured after 15 days of inoculation. Each strain was replicated at least three times.

Data statistical analysis

All data were analyzed using IBM SPSS Statistics software (version 26.0). Results are presented as the mean ± standard error (SE). For comparisons involving two or more groups, one-way analysis of variance (ANOVA) was performed. Where ANOVA indicated significant differences (P < 0.05), Duncan’s multiple range test was used for post‑hoc comparisons. In the figures, different lowercase letters (e.g., a, b, c) indicate statistically significant differences (P < 0.05) between groups, while groups sharing the same letter are not significantly different. A P‑value < 0.05 was considered statistically significant.

Author’s contributions

Yanjun Li and Xinyu Zhang designed the research and modified the manuscript. Yuanjing Li and Ruixiang Yuan performed the bioinformatics analysis and wrote the original manuscript. Jie Sun, Feng Liu revised the manuscript. Yongtai Li, Tiange Sun, Yating Wei, and Qingwen Yang conducted the HIGS experiments, gene knockout, yeast signal peptide trap assay, transient expression assay, secretome sequencing, RNA extraction, reverse transcription, and qPCR analysis. All authors reviewed the manuscript.

Funding

This work was supported by the Science and Technology Major Program of BINGTUAN (No. 2023AA008), the Key technology research and development projects of BINGTUAN (No. 2024AB001), the National Natural Science Foundation of China (No. 32360494), the Xinjiang Tianshan Talents Program (No. SN-SHZU-202402), and the Key Research and Development Program of Shihezi City (No. 2021NY01).

Data availability

The proteomics data in this article can be accessed in the iProX proteomics database with the accession number: PXD073085. Access link: http://proteomecentral.proteomexchange.org/cgi/GetDataset? ID=PXD073085.

Declarations

Ethics approval and consent to participate

The plant materials used in this study were grown in an artificial climate chamber at Shihezi University. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.