Multi-omics profiling reveals antagonistic strategies of endophytic Streptomyces sp. MEPP0209 against Lasiodiplodia theobromae causing Dendrobium officinale stem rot

Abstract

Stem rot disease frequently occurs during the artificial cultivation of Dendrobium officinale Kimura et Migo, which has adverse effects on its yield and quality. Endophytic bacteria present in plants represented potential sources of biocontrol agents. This study isolated 43 distinct microbial strains from wild Dendrobium moniliforme (L.) Sw., among which strain MEPP0209 exhibited significant inhibition of Lasiodiplodia theobromae hyphal growth. Based on the 16S rDNA sequences, the strain was identified as Streptomyces sp. MEPP0209. The cell-free supernatant (CFS) of MEPP0209 significantly inhibited L. theobromae growth, and effectively reduced the disease index of stem rot in D. officinale tissue-cultured seedlings. Transcriptome analysis showed that differentially expressed genes (DEGs) related to cell division and cell cycle were significantly downregulated in L. theobromae hyphae after MEPP0209 CFS treatment, such as MCM2/3/4/5/6/7, NDT80, POLA1, and MBP1, while DEGs related to energy metabolism and oxidative stress were significantly upregulated. Metabolomics analysis and in vitro antagonistic assays indicated that the candidate antagonistic substances, 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid, can effectively inhibit the growth of L. theobromae. And qRT-PCR results obtained from the pathogen treated with these two metabolites further indicated the downregulation of genes implicated in cell division and cell cycle. Overall, our findings suggested that the inhibition of fungal growth by MEPP0209 may be associated with the downregulation of genes related to cell division and proliferation in L. theobromae. The metabolites produced by MEPP0209, including 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid, likely contributed to this inhibitory effect. This study provides a theoretical basis for future investigations into the role of MEPP0209 in biological control of D. officinale stem rot.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00253-026-13950-7.

Introduction

Dendrobium officinale Kimura et Migo, a medicinal and edible epiphytic plant from the Orchidaceae family (He et al. 2022a, b), is mainly distributed in Southern China (Chen et al. 2021). D. officinale contains various bioactive components, including polysaccharides, flavonoids, alkaloids, and dibenzyl compounds (Yang et al. 2023). Modern pharmacological studies have demonstrated that D. officinale exhibits various therapeutic properties, including anti-cancer, anti-diabetic, cardiovascular protective, gastrointestinal regulatory, hepatoprotective, pulmonoprotective, and neuroprotective effects (Xu et al. 2022). However, due to overharvesting, low seed germination, slow growth, and environmental degradation, wild D. officinale resources have become scarce and are nearing extinction (Wang et al. 2021a, b; Mu et al. 2023). Consequently, artificial cultivation has emerged as the primary method to meet market demand (Jia et al. 2022). Nevertheless, high-density cultivation, coupled with high humidity and poor ventilation in greenhouse environments, frequently result in plant diseases and pest infestations (Mirghasempour et al. 2022). Stem rot is among the most prevalent diseases affecting the artificial cultivation of D. officinale, posing a severe threat to its yield and causing substantial economic losses (Cao et al. 2022).

Lasiodiplodia theobromae is a major pathogenic fungus responsible for stem rot in D. officinale (Zhao et al. 2016). Beyond D. officinale, L. theobromae infects other plants, including Adenia globosa, Eriobotrya japonica, and Mangifera indica (Hu et al. 2023; Kong et al. 2023; Ma et al. 2021). Symptoms of infection include light brown, water-stained circular lesions on the stem that gradually expand, leading to leaf and stem death, leaf shedding, stem wrinkling, and eventually the formation of a black mold layer. The black mold layer consists of the mature conidiophores and conidial structures of L. theobromae (Zhao et al. 2016). To meet production demands, chemical fungicides are predominantly used to prevent yield losses caused by fungal infections in plants (Ons et al. 2020). However, their use not only pollutes the environment but also promotes the development of fungicide resistance in pathogenic fungi (Han et al. 2022). Consequently, substantial research efforts are focused on reducing chemical fungicide use in plant disease management, with the goal of identifying safe, environmentally friendly, and effective alternatives, such as specific microorganisms or biological agents, to control plant diseases (Thambugala et al. 2020).

Endophytic bacteria are capable of colonizing living plant tissues without eliciting disease symptoms (Taulé et al. 2021). They contribute to the prevention and treatment of plant diseases through multiple mechanisms (Ghadamgahi et al. 2022). They directly interact with pathogens by secreting antibacterial compounds that destroy the pathogens and reduce their virulence. Endophytic bacteria also indirectly inhibit pathogens by competing for nutrients and space (Oleńska et al. 2020) or by modulating the plant's immune system and hormone levels (Bonaterra et al. 2022). For instance, the protease produced by Bacillus amyloliquefaciens SP1 degrades the cell wall of the pathogenic fungus Fusarium oxysporum, thereby inhibiting mycelial growth (Guleria et al. 2016). Bacillus subtilis effectively prevents and controls diseases by producing lipopeptides that disrupt the integrity of pathogenic fungal cell membranes (Yaraguppi et al. 2023). Streptomyces species produce secondary metabolites, such as antibiotics, that protect plants from pathogens (Procópio et al.2012). Studies have demonstrated that when Pseudomonas pseudoalcaligenes AVO110 is co-inoculated with Rosellinia necatrix, it invades plants more rapidly, outcompeting R. necatrix for nutrients and space, thereby inhibiting R. necatrix colonization and growth (Pliego et al. 2019). Strains of Agrobacterium and Bacillus protected lettuce against Sclerotinia sclerotiorum and Rhizoctonia solani by inducing salicylic acid (SA)- and ethylene/jasmonic acid (ET/JA)-dependent defenses during early infection stages and ET/JA-dependent defenses during later disease stages (Aggeli et al. 2020). Streptomyces sp. strain AcH 505 induced systemic resistance in oak by activating jasmonic acid/ethylene (JA/ET)- and salicylic acid (SA)-dependent pathways, which were further enhanced upon pathogen challenge (Kurth et al. 2014).

Currently, some studies have investigated the use of endophytic bacteria derived from Dendrobium species to manage plant diseases. A Streptomyces strain isolated from wild Dendrobium plants demonstrated effective biocontrol activity against southern blight in D. officinale caused by the fungus Sclerotium delphinii (Chen et al. 2024). Yi et al. isolated Paenibacillus polymyxa Y-1 from Dendrobium nobile. Eight metabolites extracted from the fermentation broth of this strain exhibited strong antibacterial activity against Xanthomonas oryzae pv. oryzicola (Xoo) and Xanthomonas oryzae pv. oryzae (Xoc) (Yi et al. 2022). However, these limited studies have primarily focused on the transcriptional changes in pathogen in response to antagonistic bacterium, or on the extraction, isolation, and antibacterial activity assessment of active metabolites from biocontrol agent. Within the genus Dendrobium, there remains a significant gap in research that integrates multi-omics approaches to comprehensively elucidate the mechanisms by which endophytic antagonistic microorganism inhibit pathogens. Additionally, few studies have explored endophytic biocontrol agents to combat stem rot in D. officinale. Therefore, the novelty of this study lies in employing an integrated approach that combines pathogen inhibition assays with multi-omics techniques to explore the potential antifungal mechanisms of endophytic bacterium against L. theobromae. Specifically, this study was carried out from the following aspects: (1) isolate and screen for the endophytic antagonistic bacterium MEPP0209 by evaluating its antifungal effects; (2) assess the inhibitory effects of MEPP0209, its volatile organic compounds (VOCs), and cell-free supernatant (CFS) on L. theobromae growth; (3) determine whether MEPP0209 can protect D. officinale against stem rot; and (4) explore the potential mechanisms underlying MEPP0209's inhibition of L. theobromae through a combination of multi-omics techniques and in vitro validation experiments of potential antifungal substances.

Materials and methods

Isolation and identification of the pathogen from D. officinale and verification of its pathogenicity

The diseased D. officinale plants were collected from a cultivation site in Shicha Village, Shishan Town, Haikou City, Hainan Province, China. These samples were used for pathogen isolation. The surface sterilization of samples was carried out by rinsing them with 75% ethanol first and then rinsed twice with sterile water. Subsequently, they were treated with 5% sodium hypochlorite and rinsed three times with sterile water. Subsequently, the lesion margins (disease-healthy tissue junctions) were excised and transferred to 9 cm Petri dishes containing potato dextrose agar (PDA), and incubated at 28 ± 2 °C for 5 days. Identical fungal colonies growing from multiple lesion margins were considered as the potential pathogen.

Molecular identification of the isolated pathogen was performed through ITS sequencing. The genomic DNA extraction was performed with the FastPure Plant DNA Isolation Mini Kit (Nanjing Vazyme Biotech Co., Ltd., China), followed by storage at – 20 °C for subsequent analysis. For ITS region amplification, PCR was conducted employing universal primer pairs: ITS1 (5'-TCCGTAGGTGAACCTGCGG-3') and ITS4 (5'-TCCTCCGCTTATTGATATGC-3') (Jian et al. 2021). The PCR reaction mixture (25 μL) contained 2.5 μL DNA template, 12.5 μL 2 × Taq master mix, 1 μL each primer, and 8 μL sterilized water. The PCR program was executed as follows: the initial denaturation at 94 °C for 4 min, followed by 32 cycles of denaturation at 94 °C for 30 s, annealing at 54 °C for 30 s, and extension at 72 °C for 1 min, with a final extension at 72 °C for 10 min. PCR products were subjected to electrophoretic separation on 1% agarose gels, followed by nucleotide sequencing performed at Sangon Biotech Co., Ltd. (Shanghai, China).The acquired sequences were subjected to analysis utilizing the Basic Local Alignment Search Tool (BLAST; https://blast.ncbi.nlm.nih.gov/Blast.cgi), which is provided by the National Center for Biotechnology Information (NCBI). For the phylogenetic examination, the ITS sequences were analyzed using MEGA 11 software (Tamura et al. 2021), employing the maximum likelihood approach.

The isolated fungus was inoculated into sterile seedlings of D. officinale to evaluate pathogenicity. The D. officinale seedlings were transferred to 350 mL tissue culture bottles containing Murashige and Skoog (MS) medium (Murashige and Skoog 1962) supplemented with 30 g/L sucrose, 60 g/L potato extract, and 7.5 g/L agar, with three plants per bottle. Fungus was cultured on PDA medium for 4 days, and mycelial plugs (approximately 5 mm in diameter) were excised from the edge of the culture plate. Each plant stem was inoculated with 2 mycelial plugs, while control plants received PDA plug without mycelium to serve as negative controls. Three biological replicates were set up for each treatment. The inoculated seedlings were incubated in an artificial light chamber maintained at 26 ± 2 °C for 10 days. The seedlings were inspected daily for disease symptoms to evaluate the pathogenicity of the isolated fungus. Based on Koch's postulates, if the seedlings showed typical disease symptoms, and identical isolate and ITS sequence were retrieved from symptomatic tissues, the tested isolate was verified as the causal pathogen.

Isolation of endophytic bacteria from wild Dendrobium moniliforme

The wild samples of Dendrobium moniliforme (L.) Sw. were collected from Jinxiu Yao Autonomous County, Guangxi Province, China. The collected samples were transferred to the laboratory for the isolation of their endophytic bacteria, and the isolation method for endophytic bacteria followed Mahlangu and Serepa-Dlamini (2018) with modifications. The plant roots, stems, and leaves were separated, and 1 g of each tissue type was weighed. Initially, the samples were rinsed under running tap water for 30 min, followed by immersion in 75% ethanol for 5 min, and then rinsed three times with sterile water. Subsequently, they were soaked in a 5% sodium hypochlorite (NaClO) solution for 4 min, with an additional three rinses with sterile water. The effectiveness of the sterilization procedure was confirmed by plating the final rinse water onto LB agar plates. Under sterile conditions, the surface-sterilized root, stem, and leaf materials were individually homogenized with 9 mL of sterile water using a sterile mortar and pestle. The homogenates from the roots, stems, and leaves were then subjected to serial dilution with sterile water to achieve dilution factors of undiluted (10⁰), 10⁻¹, and 10⁻². Different concentrations of the homogenates (10⁰, 10⁻¹, and 10⁻²) from the roots, stems, and leaves were inoculated onto LB, ISP2, and HV plates (Hamaki et al. 2005; Golinska et al. 2015), with three replicates set up for each medium. LB plates were incubated at 37 ± 2 °C, while ISP2 and HV plates were maintained at 28 ± 2 °C. Distinct colonies were subcultured, and pure cultures were preserved at – 80 °C for antifungal activity screening. The endophytic bacterium with the highest inhibition rate was subjected to further investigation.

Identification and physiological and biochemical characterization of the endophytic bacteria MEPP0209

Genomic DNA of the endophytic bacteria was extracted using the FastPure Bacterial DNA Isolation Mini Kit-BOX2 (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China) and stored at – 20 °C. The 16S rDNA sequence of MEPP0209 was amplified using primers PA(5'-CAGAGTTTGATCCTGGCT-3') and PB (5'-AGGAGGTGATCCAGCCGCA-3'), which are more specific to actinobacteria (Qin et al. 2009), while the 16S rDNA amplification of other endophytic bacteria was performed using universal primers 27 F (5'-AGAGTTTGATCCTGGCTCA-3') and 1492R (5'-GGTTACCTTGTTACGACTT-3'). The PCR reaction mixture contained 2.5 μL DNA template, 12.5 μL 2 × Taq master mix, 1 μL each primer, and 8 μL sterile water. The PCR program consisted of initial denaturation at 95 °C for 3 min, followed by 30 cycles of denaturation at 95 °C for 15 s, annealing at 60 °C for 15 s, and extension at 72 °C for 2.5 min, with a final extension at 72 °C for 5 min. The amplified products were analyzed using 1% agarose gel electrophoresis and sequenced by Sangon Biotech Co., Ltd. (Shanghai, China). The BLAST program was used to analyze the 16S rDNA sequences against the NCBI GenBank database. The morphological characteristics of the endophytic bacterial strain MEPF0209 were observed using scanning electron microscopy (SEM; Hitachi S-3400N, Hitachi High-Technologies Corporation, Tokyo, Japan) at 30 kV and × 30,000 magnification. A phylogenetic tree of strain MEPF0209 based on the 16S rDNA sequences was constructed using the MEGA 11 program with the maximum likelihood method. Concurrently, the physiological and biochemical characteristics of MEPP0209 were determined, with the formulas and methods formulations provided in Supplementary Table S1.

In vitro dual culture assays and hyphal morphology observation of L. theobromae

Dual culture experiments were performed using PDA medium in 9 cm diameter culture dishes (Passari et al. 2015). The inoculation method of pathogenic fungus and endophytic strain MEPF0209 on the culture dishes referred to Chen et al. (Chen et al. 2024). A pathogenic fungus culture dish without endophytic strain served as the control. The culture dishes were incubated at 28 ± 2 °C for 5 days in a DNP-9052 incubator (Jinghong, China). Fungal colony diameters were measured daily, and the growth inhibition rates were calculated according to the following formula:

$$\text{Inhibition rate (\%)}=\frac{(\text{diameter of the control fungal colony}-\text{diameter of fungal colony in endophytic strain presence})}{\text{diameter of the control fungal colony}}\times 100$$ 1

. A DM5000B fluorescence microscope (Leica, Wetzlar, Germany) equipped with Las X (Leica, Wetzlar, Germany) was used to observe the morphological characteristics of the antagonistically treated and control mycelia. Three biological replicates were set for each experiment and repeated three times (n = 9).

Inhibitory effects of the MEPP0209 volatile organic compounds and cell-free supernatant on the growth of L. theobromae

The inhibitory impact of VOCs was assessed using a double-plate assay (Win et al. 2022). L. theobromae plugs (5 mm) were placed on PDA, while MEPP0209 was grown on ISP2. Plates were sealed face-to-face with parafilm, preventing the diffusion of VOCs and exposing the mycelia to VOCs emitted by MEPP0209, and incubated at 28 ± 2 °C for 5 days. The medium without MEPP0209 was used as the control group. Three biological replicates were set for each experiment and repeated three times (n = 9). The impact of VOCs was assessed by evaluating the reduction in fungal colony size. Following the incubation period, the colony diameters were measured along two perpendicular axes, and the inhibition efficacy was calculated using Eq. (1).

The CFS of strain MEPP0209 was prepared as previously described by Liao et al. (Liao et al. 2023), with minor modifications. MEPP0209 was cultured in 200 mL ISP2 medium at 28 ± 2 °C, 180 rpm for 7 days. The fermentation broth was centrifuged at 12,000 × g for 10 min, and supernatants were filtered using the 0.22 μm microporous membrane to generate CFS. The CFS was incorporated into PDA medium at 40 °C to achieve ultimate concentrations of 10%, 20%, 30%, 40%, and 50%, with PDA medium lacking CFS used as the control group. Agar plates were centrally inoculated with L. theobromae mycelial plugs and maintained at 28 ± 2 °C for 5 days incubation period. After incubation, the final diameters of the fungal colonies were measured. The growth inhibition rate was calculated using formula (1) as described above. Fluorescent dye DCFH-DA (2',7'-Dichlorodihydrofluorescein diacetate, Beyotime Biotechnology, Shanghai, China) was used to detect reactive oxygen species (ROS) levels in both the control group (PDA) and the treated group (PDA with 30% MEPP0209 CFS). A DM5000B fluorescence microscope (Leica, Germany) equipped with Las X was used to observe the ROS levels of treated and untreated mycelia. Three biological replicates were set for each experiment and repeated three times (n = 9).

Evaluation of MEPP0209 in controlling stem rot of D. officinale

Healthy, similarly sized, and sterile D. officinale seedlings were used to investigate whether MEPP0209 can effectively control stem rot. The CFS was prepared following the procedure outlined in the preceding section. Bacterial suspension (BS) was prepared by lightly scraping off bacterial biomass from ISP2 plates cultured for 7 days, suspending it in sterile water, and adjusting the cell density to 10⁸–10⁹ CFU/mL using a spectrophotometer (OD₆₀₀ = 1.0) and plate count (He et al. 2022a, b). D. officinale seedlings were transferred to 350 mL flasks with 30 mL of sterile water in each flask, and five plant seedlings were placed in each flask. The pathogen inoculation method referred to the pathogenicity determination section mentioned earlier. Each plant stem was inoculated with 2 mycelial plugs, while control plants received PDA plug without mycelium to serve as negative controls. Six treatment groups were established as follows: (1) treated with sterile water 6 h prior to pathogen inoculation (SW + LT); (2) treated with CFS 6 h prior to pathogen inoculation (CFS + LT); (3) treated with BS 6 h prior to pathogen inoculation (BS + LT); (4) treated with sterile water 12 h post-inoculation (LT + SW); (5) treated with CFS 12 h post-inoculation (LT + CFS); and (6) treated with BS 12 h post-inoculation (LT + BS). Groups 1 and 4 served as controls. The seedlings were maintained at 26 ± 2 °C with a 12-h light/dark cycle for 7 days. L. theobromae infection was monitored over 7 days by observing necrotic lesions on the infected plants. The disease severity was assessed on a 0–4 scale: 0 (asymptomatic), 1 (1–2 leaves with minor lesions), 2 (1–2 leaves with extensive lesions and stem with disease spots), 3 (more than 3 leaves with necrosis, partial defoliation, and stem was rotted and softened) and 4 (the whole plant decay or death). The disease index (%) was calculated according to formula (2) (Fan et al. 2021). Five biological replicates were set for each experiment and repeated three times (n = 15).

$$\text{Disease index =}\frac{\sum \text{disease grade}\times \text{number of infected plantlets}}{\text{total number of plantlets}\times \text{4}}\times \text{100}$$ 2

Transcriptome analysis of L. theobromae treated with MEPP0209 CFS

Transcriptome samples were prepared as described (Lou et al. 2018), with modifications. Three L. theobromae mycelial plugs were inoculated into 200 mL of potato glucose broth (PDB) as the control group. Three L. theobromae mycelial plugs were also inoculated into conical flasks containing 140 mL of PDB and 60 mL of MEPP0209 CFS as the treatment group and cultured for 3 days at 26 ± 2 °C and 180 rpm in a constant temperature shaker (Ruihua, Changzhou, China). Four replicates were set for each group of treatments. The mycelia were filtered through nine layers of sterile gauze, collected, and prepared as transcriptome sequencing (RNA-seq) samples. Prior to RNA isolation, the harvested mycelial samples were rapidly cryopreserved using liquid nitrogen.

Total RNA isolation was performed with the MJZol Total RNA Extraction Kit (Shanghai Majorbio Bio-Pharm Biotechnology Co., Ltd., Shanghai, China) following the manufacturer's protocol. The quality and integrity of the RNA samples were assessed using 1% agarose gel electrophoresis, the 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and the NanoDrop-2000 spectrophotometer (NanoDrop Technologies, Waltham, MA, USA). The transcriptome library was prepared with 1 μg of total RNA using the TruSeq™ RNA Sample Preparation Kit (Illumina, San Diego, CA, USA). The library was quantified using TBS380 and sequenced on the Illumina NovaSeq 6000 system (150 bp paired-end reads) at Shanghai Majorbio Bio-Pharm Biotechnology Co., Ltd. (Shanghai, China). The RNA sequencing (RNA-seq) datasets have been deposited in the NCBI database under accession number PRJNA1166406.

Raw paired-end reads were trimmed and quality-screened using the default parameters of fastp (Chen 2023) (https://github.com/OpenGene/fastp). Clean reads were then aligned to the L. theobromae reference genome (GCF_012971845.1) using HISAT2 (Kim et al. 2019) (https://ccb.jhu.edu/software/hisat2/index.shtml). Differentially expressed genes (DEGs) between the two groups were identified by measuring gene expression levels using RSEM (Djebali et al. 2017) (https://deweylab.biostat.wisc.edu/rsem/). Gene expression levels were normalized to transcripts per million reads, and DEGs analysis was performed using DESeq2 (Liu et al. 2021). Significant DEGs were identified using the criteria |log2(fold-change) |≥ 1 and P < 0.05. Furthermore, Gene Ontology (GO; https://www.geneontology.org) and Kyoto Encyclopedia of Genes and Genomes database (KEGG; https://www.genome.jp/kegg/) were utilized for functional annotation analysis, revealing statistically significant GO terms and metabolic pathways associated with DEGs (P < 0.05).

Quantitative real-time polymerase chain reaction analysis

To validate the RNA-seq results, quantitative real-time PCR (qRT-PCR) was conducted on the selected genes mainly related to cell division, cell cycle and oxidative stress. The samples from the control group and the treatment group were the same as those used in transcriptome sequencing. First-strand cDNA is synthesized with HisyGo qRT Red SuperMix for qPCR (+ gDNA wiper) (Nanjing Vazyme Biotech Co., Ltd., China). The qRT-PCR analysis was performed using the ChamQ Blue Universal SYBR qPCR Master Mix (Nanjing Vazyme Biotech Co., Ltd., China), with Actin serving as the internal reference (F: 5'-CTTCGAGACCTTCAACGCC-3'; R: 5'-TAGATGGGGACGACGTGAGT-3'). The gene-specific primers were presented in Supplementary Table S2, and relative expression levels of genes were determined based on the 2^⁻ΔΔCt method (Zhao et al. 2021). Three biological replicates were set for each treatment group, and the experiment was repeated three times (n = 9).

Non-targeted metabolome analysis of MEPP0209 cell-free supernatant

The preparation of non-target metabolome samples was based on previous instructions with appropriate modifications (Cong et al. 2023). CFS of MEPP0209 was prepared as described in the preceding section and served as the treatment group. ISP2 liquid culture medium without MEPP0209 inoculation served as the control group. The liquid culture medium and CFS were freeze-dried using a lyophilizer (Christ Beta 1–8 LD, Osterode, Germany). For each group, 25 mg of sample was mixed with 400 µL of methanol/water (4:1, v/v) solution containing 0.02 mg/mL L-2-chloroalanine as the internal standard. The mixture was precipitated at – 10 °C, ground for 10 min, sonicated at 5 °C for 45 min, and frozen at – 20 °C for 20 min. The sample was centrifuged at 13,000 × g for 15 min at 4 °C, and the supernatant containing metabolites was collected. The samples were sent to Shanghai Majorbio Bio-Pharm Biotechnology Co., Ltd. (Shanghai, China) for liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis. Each group consisted of six independent biological replicates.

LC–MS analysis was conducted using a UHPLC-Q Exactive HF-X system (Thermo Fisher, Waltham, MA, USA) with chromatographic conditions set to an ACQUITY UPLC HSS T3 column (100 mm × 2.1 mm, 1.8 μm; Waters, Milford, MA, USA). Mobile phase A consisted of 95% water (7732–18-5, Fisher Chemical, Waltham, MA, USA) and 5% acetonitrile (75–05–8, Fisher Chemical, Waltham, MA, USA) with 0.1% formic acid, while mobile phase B consisted of 47.5% acetonitrile, 47.5% isopropanol (67–63-0, ANPEL Laboratory Technologies Inc., Shanghai, China), and 5% water with 0.1% formic acid (64–18-6, ANPEL Laboratory Technologies Inc., Shanghai, China). The injection volume was 2 μL, and the column temperature was maintained at 40 °C. Samples were ionized using electrospray, and mass spectral signals were collected in both positive and negative ion scanning modes. Mixed quality control (QC) samples were prepared by combining equal volumes of all samples and processed in the same manner as the experimental samples. Potential differentially accumulated metabolites (PDAMs) were identified based on variable importance in projection (VIP) values > 1 and P-values < 0.05. VIP values and P-values were derived from the orthogonal partial least squares discriminant analysis (OPLS-DA) model and Student's t-test, respectively. Identified PDAMs were mapped to their respective biochemical pathways through metabolic enrichment and pathway analysis using the KEGG database (KEGG; http://www.genome.jp/kegg/). The non-targeted metabolomic data from this study were deposited in the OMIX database under accession ID OMIX007605 (https://ngdc.cncb.ac.cn/omix/view/OMIX007605).

In vitro antagonistic assay of metabolites

Based on the combined analysis of transcriptomics and metabolomics, as well as literature review (Nguyen et al. 2015; Aron et al. 2021; Ortiz-Ramírez et al. 2022; Zhao et al. 2024; Legendre et al. 2020; Lu et al. 2024), six potential antagonistic compounds were screened for in vitro antagonistic assays, such as L-glutamine, 5,6-dimethylbenzimidazole, D-glutamine, 2-ketoglutaric acid, pyrrole-2-carboxylic acid and pimelic acid (Shanghai Macklin Biochemical Co., Ltd, Shanghai, China). Following established methodologies in previous studies with minor modifications (Nguyen et al. 2015; Zhao et al. 2024; Yang et al. 2021), the final concentrations of 125, 250, 500, and 1000 mg/L were achieved for each compound in PDA medium. PDA medium without compounds was used as a control. A mycelial plug of L. theobromae was inoculated at the center of each plate, followed by incubation at 28 ± 2 °C for 5 days. The final fungal colony diameters were recorded, and the inhibition rates were calculated using Eq. (1), screening for antagonistic metabolites. Perform qRT-PCR analysis on L. theobromae treated with screened antagonistic metabolites, and qRT-PCR procedure followed the protocol outlined in the preceding section. Genes primarily related to cell division, cell cycle and oxidative stress in L. theobromae were selected for qRT-PCR analysis, and gene-specific primers were listed in Supplementary Table S2. Three biological replicates were set for each experiment, and the experiment was repeated three times (n = 9).

Statistical analyses

All statistical analyses and graphical representations were conducted using GraphPad Prism (version 8.0.2; GraphPad Software Inc., San Diego, CA, USA). Error bars represent standard deviation. Statistical significance among mean values was assessed using Student’s t-test for comparisons between two groups or one-way ANOVA with Tukey’s post hoc test for multiple comparisons, with significance set at P < 0.05.

Results

Isolation, identification, and pathogenicity assessment of the pathogen

On PDA medium, the hyphae of the isolated pathogen initially appeared white with abundant aerial growth, expanding radially and gradually transitioning to light gray, mouse gray, and eventually black villous (Supplementary Fig. S1a). During the pathogenicity assessment, water-stained circular lesions appeared on leaves in the early stages of pathogen infection, progressively expanding over time. Black lesions appeared on the stems, followed by leaf shedding during the middle stage. In the later stages, the leaves completely fell off, and the plants rotted and died, consistent with stem rot symptoms (Supplementary Fig. S1b). Sequence alignment with the NCBI database revealed that the ITS sequence (PQ459379) of the isolated pathogen showed 100% similarity to L. theobromae. In the phylogenetic tree, the isolated pathogen clustered with L. theobromae in the same clade (Supplementary Fig. S2). The mycelial morphological characteristics and ITS sequence of the strain re-isolated from diseased D. officinale seedlings were identical to those of the initially isolated isolate. Therefore, the isolate tested was confirmed as the causal agent of stem rot in D. officinale and was identified as L. theobromae. The pathogen L. theobromae was deposited in the China Center for Type Culture Collection [CCTCC M2025201].

Identification, and physiological and biochemical characteristics of MEPP0209

Forty-three bacterial strains were isolated from wild D. moniliforme (Supplementary Table S3) and subjected to in vitro dual culture experiments. The results showed varying levels of antagonistic activity against L. theobromae, with strain MEPP0209 exhibiting the strongest antagonistic effect (Supplementary Fig. S3). On solid ISP2 medium, the colony of MEPP0209 initially appeared white and circular, gradually transitioning to a mouse gray color with black spots over time (Supplementary Fig. S4a). Scanning electron microscopy (SEM) images showed that MEPP0209 exhibited slender cellular morphology with characteristic spiral spore chains (Supplementary Fig. S4b). The phylogentic tree based on 16S rDNA sequences indicated that strain MEPP0209 was a member of the Streptomyces genus. It formed a subclade with Streptomyces platensis (EU502864) at a bootstrap value of 93% (Fig. 1). Based on morphological characteristics and phylogenetic analysis, the strain was referred to as Streptomyces sp. MEPP0209, and stored as a patent strain in the China General Microbiological Culture Collection Center [CGMCC 32171], Beijing, China. The 16S rDNA sequences of all 43 endophytic bacteria isolated in this study have been deposited in the NCBI database, and Supplementary Table S3 lists their GenBank accession numbers. Physiological and biochemical assays indicated that MEPP0209 could dissolve phosphorus, fix nitrogen, produce ACC deaminase, and siderophores, and hydrolyze starch. However, it did not produce protease or oxidase, nor could it hydrolyze cellulose (Supplementary Fig. S5).

Fig. 1.

Phylogenetic tree of MEPP0209 constructed based on 16S rDNA sequences. Bootstrap values > 50% are shown on the branches. The reference sequence contains 18 other strains of Streptomyces (GenBank accession number is shown in the branch annotation)

In vitro antifungal efficacy of MEPP0209 on L. theobromae

The in vitro antifungal activity of MEPP0209 against L. theobromae was evaluated by measuring the diameter of fungal colonies treated with MEPP0209. After incubation at 28 ± 2 °C for 4 days, the diameter of fungal colonies treated with MEPP0209 was 0.68 ± 0.09 cm, compared to 8.55 ± 0.005 cm for the control colonies (Fig. 2). The inhibition rate of mycelial growth was calculated to be 92.09%, demonstrating the strong antagonistic effect of MEPP0209 on L. theobromae. To further investigate the morphological changes in L. theobromae mycelia after antagonistic treatment, the control and treated L. theobromae mycelia were comparatively observed using a fluorescence microscope. The control L. theobromae mycelia appeared smooth and structurally normal (Figs. 3a-c). In contrast, mycelia treated with MEPP0209 exhibited morphological abnormalities, including swelling, vacuolization, deformation, and increased branching at the ends (Figs. 3d-f).

Fig. 2.

Dual-culture assay of MEPP0209 antagonistic effects on L. theobromae mycelia. (a) Control (left) and MEPP0209-treated (right) L. theobromae cultures on PDA medium. Scale bar = 2 cm. (b) Hyphal growth inhibition of L. theobromae by MEPP0209. The X-axis is different treatment groups, and the Y-axis is the diameter of fungal colony. An asterisk indicates a significant difference between treatments according to the student’s t-test (**** P < 0.0001)

Fig. 3.

The effects of MEPP0209 on the mycelial morphology of L. theobromae. (a-c) Microscopic features of control L. theobromae mycelia: (a) Minimal terminal branching, (b) Smooth structure, and (c) Absence of deformation. (d-f) Microscopic features of MEPP0209-treated L. theobromae mycelia: (d) Enhanced terminal branching, (e) Mycelial dissolution, and (f) Swelling or vacuolization. The red arrow indicates the abnormal mycelial morphology

Effects of the MEPP0209 VOCs and CFS on the growth of L. theobromae

To characterize the antifungal activity of MEPP0209, the effects of VOCs and different concentrations of CFS were analyzed. Exposure to VOCs produced by MEPP0209 showed weak inhibitory effect on the growth of L. theobromae (Fig. 4a). The inhibition rate of L. theobromae mycelial growth after VOCs treatment was only 4.44% (Fig. 4b). In contrast, the inhibitory effect of CFS on L. theobromae increased with rising CFS concentrations (Figs. 4b and 4c). The antifungal assay revealed concentration-dependent inhibition, with mycelial growth suppression reaching 6.67% at 10% CFS and 73.52% at 50% CFS concentration. (Fig. 4b). These results indicated that CFS exhibit a significantly stronger inhibitory effect on L. theobromae than VOCs (Fig. 4). ROS levels were also observed in the control and treatment groups. In the control group, almost no green fluorescence was detected in L. theobromae hyphae, indicating very low ROS content under normal conditions. However, the hyphae treated with MEPP0209 CFS displayed green fluorescence (Supplementary Fig. S6), suggesting the accumulation of ROS in the treated group.

Fig. 4.

Impacts of MEPP0209 VOCs and CFS on L. theobromae growth. (a) Influence of MEPP0209 VOCs on L. theobromae growth. Scale bar = 2 cm. (b) Inhibition rates of L. theobromae growth by MEPP0209 VOCs and CFS at different concentrations. The X-axis is different treatment groups, and the Y-axis is the Inhibition rate. Different letters indicate significant differences between treatments according to one-way ANOVA (P < 0.05). (c) Growth responses of L. theobromae to CFS concentrations (10%, 20%, 30%, 40%, and 50%). Scale bar = 2 cm

Biological control efficiency of MEPP0209 against stem rot of D. officinale

To evaluate the ability of MEPP0209 to control stem rot disease in D. officinale, sterile D. officinale seedlings were treated with MEPP0209 CFS and BS before and after inoculation with the pathogen. Following inoculation, the seedlings were monitored daily for symptoms of stem rot. After 7 days, in the SW + LT and LT + SW groups (control groups), the leaves were decayed, softened, and had mostly fallen off. Varying degrees of disease symptoms were also observed on the stems (Fig. 5a), with incidence rates reaching 100% (Supplementary Fig. S7). In contrast, in the other four treatment groups, some plants exhibited small lesions on leaves and stems, but most showed no obvious symptoms (Fig. 5a). The disease indices for the LT + CFS and LT + BS groups were 16.0% and 12.0%, respectively, while the indices for the CFS + LT and BS + LT groups were 16.33% and 14.33%, respectively (Fig. 5b). These results demonstrated that both BS and CFS of MEPP0209 exhibited significant biocontrol effects against stem rot in D. officinale.

Fig. 5.

Biocontrol efficacy of strain MEPP0209 on stem rot in D. officinale. (a) Disease symptoms in six groups of D. officinale seedlings. Scale bar = 2 cm. (b) Disease indices across the six groups of D. officinale seedlings. The X-axis is different treatment groups, and the Y-axis is the disease index. Different letters indicate significant differences between treatments according to one-way ANOVA (P < 0.05)

Transcriptome analysis

A total of 60.69 Gb of clean sequencing data were obtained for the control and MEPP0209-treated samples, and Q20 and Q30 values for each sample exceeded 98.80% and 96.12%, respectively (Supplementary Table S4), reflecting the high quality of the sequencing data used for subsequent bioinformatic analyses. Using thresholds of |log₂(fold-change) |≥ 1 and adjusted P < 0.05, 4,407 significant DEGs were identified between MEPP0209-treated and control L. theobromae mycelia. Of these, 2,111 (47.90%) were upregulated, and 2,296 (52.09%) were downregulated (Supplementary Fig. S8). GO enrichment analysis showed that upregulated DEGs were primarily associated with catalytic activity and multiple metabolism-related terms, including carbohydrate, organic acid, carboxylic acid and oxoacid metabolic processes (Fig. 6a; Supplementary Table S5). In contrast, downregulated DEGs were mainly enriched for organelle-related categories (e.g., organelle, intracellular organelle, non-membrane-bounded and intracellular non-membrane-bounded), cellular nitrogen compound metabolic process and RNA processing (Fig. 6b; Supplementary Table S6).

Fig. 6.

Functional enrichment analysis of differentially expressed genes (DEGs) in GO terms and KEGG pathways. (a) Enriched GO terms for the upregulated DEGs. (b) Enriched GO terms for the downregulated DEGs. The vertical axis represents GO term, and the horizontal axis represents rich factor. The larger the rich factor, the greater the degree of enrichment. The size of the dots represents the number of genes/transcripts in this GO term, and the color of the dots corresponds to different P-adjust ranges. (c) Enriched KEGG pathways for the upregulated DEGs. (d) Enriched KEGG pathways for the downregulated DEGs. The vertical axis represents the pathway name, and the horizontal axis represents the rich factor. The larger the rich factor, the greater the degree of enrichment. The size of the dots indicates the number of genes in this pathway, while the color of the dots corresponds to different P-adjust ranges

The upregulated KEGG pathway analysis further supported a survival-oriented metabolic reprogramming. Upregulated DEGs were primarily enriched in central carbon and energy metabolism, including starch and sucrose metabolism, glycolysis/gluconeogenesis, the pentose phosphate pathway (PPP), and galactose metabolism (Fig. 6c; Supplementary Table S7). Consistent with these enrichments, genes encoding key glycolytic enzymes were strongly induced, including pyruvate kinase (PK/pyk), glyceraldehyde-3-phosphate dehydrogenase (GAPDH/gapA) and triosephosphate isomerase (TPI/tpiA) (Fig. 7a). In parallel, induction of PPP-associated 6-phosphogluconate dehydrogenase (PGD/gnd/gntZ) supports the production of NADPH. Multiple carbohydrate-active enzymes (e.g., glycoside hydrolases bglX and SGA1), together with glycogen synthase (GYS), were upregulated, consistent with remodeling of carbohydrate utilization and storage (Fig. 7a). In addition to activating energy metabolism, MEPP0209 CFS elicited a coordinated detoxification/oxidative stress-linked response of pathogen. Upregulated genes mainly included oxidase (iscS/NFS1, DAO/aao), glutathione-associated dehydrogenase (frmA/ADH5/adhC) and oxidative metabolism component cytochrome P450 (PDC/pdc) (Fig. 7a).

Fig. 7.

Representative DEGs in the top five upregulated and downregulated KEGG pathways. (a) Energy metabolism and detoxification/oxidative stress-linked genes were significantly upregulated. (b) Fungal growth and proliferation related genes were broadly downregulated. (a–b) Dot–heatmaps display log2 fold changes (Log2FC) of selected genes across the four treatment group samples (A7_1, A7_10, A7_13, A7_14) relative to CK_mean. Each gene is shown as a fixed grey square frame, within which a filled square is overlaid. The left side of the square frames is gene name, and the right side is gene description. Color encodes Log2FC (red, upregulated; blue, downregulated) and square size encodes the adjusted P-value (Padjust), with smaller Padjust shown as larger squares

Conversely, the downregulated DEG set was mainly enriched in pathways related to growth and proliferation, particularly ribosome, ribosome biogenesis, cell cycle–yeast, and DNA replication (Fig. 6d; Supplementary Table S8). The heatmap representation of genes drawn from the top five enriched downregulated KEGG pathways highlighted broad repression of core replication and cell-cycle machinery (Fig. 7b). This included a significant decrease in the expression of replication initiation and licensing factors (ORC2, ORC4, MCM2/3/4/5/6/7), DNA replication process-related genes (CDC45, POLA1, MRC1), and checkpoint/cell division regulators (TTK/MPS1, BUB1, MAD1, CDC28/CDC2, MBP1, NDT80). Consistently, ribosome-linked genes (e.g., RP-S16e/RPS16, EMG1/NEP1) were also significantly downregulated. Collectively, these transcriptional patterns were consistent with the inhibited state of pathogen growth, providing a plausible molecular basis for reduced fungal growth and pathogenicity under MEPP0209 CFS treatment.

Validation of the RNA-seq results through qRT-PCR analysis

The qRT-PCR was performed to validate the expression of 15 selected DEGs, the majority of which are associated with oxidative stress and cell proliferation. The results demonstrated that the expression patterns of these genes were consistent with the RNA-seq analysis (P < 0.05), confirming the reliability of the RNA-seq data (Supplementary Fig. S9).

Non-targeted metabolome analysis

Metabolomic database analysis identified several potential metabolites from ion peaks detected via mass spectrometry in both the treatment (CFS) and control (ISP2) groups. In POS (Positive Ion Mode) mode, a total of 1,168 potential metabolites were detected across both samples, with 69 exclusives to the treatment group and 23 specifics to the control group (Fig. 8a). Similarly, NEG (Negative Ion Mode) mode detected 865 potential metabolites, including 48 exclusives to the treatment group and 15 specifics to the control group. Comparing both modes between the treatment and control groups revealed 664 potential differentially accumulated metabolites (PDAMs), of which 365 were upregulated and 299 were downregulated (Fig. 8b). The primary metabolic pathways associated with the upregulated PDAMs included biosynthesis of cofactors (map01240), D-amino acid metabolism (map00470), tryptophan metabolism (map00380), and ascorbate and aldarate metabolism (map00053) (Fig. 8c, Supplementary Table S9).

Fig. 8.

Analysis of PDAM quantity and composition in the treatment and control groups. (a) Venn diagram of PDAMs detected in POS and NEG modes for the two groups. (b) Volcano plot of PDAMs between the treatment and control groups. Nosig was a non-significantly different metabolite. (c) Enriched KEGG pathways for the upregulated PDAMs. (d) Classification of PDAMs based on KEGG pathways

A total of 517 PDAMs were categorized into 11 groups based on the HMDB database (Supplementary Table S10), while 62 PDAMs were further classified into 18 categories using the KEGG database (Fig. 8d). Among these metabolites, one PDAM was classified as a carbohydrate, one as a vitamin or cofactor, two as antibiotics, two as steroids, five as lipids, eight as organic acids, eight as nucleic acids, eleven as hormones or transmitters, and twenty-four as peptides (Fig. 8d). Additionally, the representative PDAMs in the top two upregulated KEGG enrichment pathways (Supplementary Table S9) were displayed in Supplementary Table S11.

In vitro antagonistic assays of 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid

Based on literature review and transcriptomic and metabolomic analysis, six metabolites were selected for in vitro antagonistic assays in the upregulated top two KEGG-enriched pathways, and it was found that 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid could significantly inhibit the mycelial growth of L. theobromae (Fig. 9a). At final concentrations of 125, 250, 500 and 1000 mg/L in the medium, the inhibition rates of 5,6-dimethylbenzimidazole against L. theobromae were 15.56%, 38.17%, 62.49%, and 100%, respectively (Fig. 9b). And pyrrole-2-carboxylic acid exhibited the inhibition rates of 0%, 7.12%, 23.01%, and 49.68%, respectively (Fig. 9c). Besides, the colony morphology of L. theobromae exhibited obvious changes after treatment with pyrrole-2-carboxylic acid, and the colony became loose (Fig. 9a). Further biomass quantification showed that the mycelial biomass of L. theobromae was significantly reduced after treatment with 500 mg/L pyrrole-2-carboxylic acid (Fig. 9d). Additionally, the qRT-PCR results suggested that the expression trends of genes related to cell proliferation and oxidative stress in L. theobromae treated with 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid were consistent with transcriptome analysis (Figs. 9e and 9f).

Fig. 9.

Inhibitory effects of L. theobromae mycelial growth with 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid. (a) The inhibitory effects of 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid at different concentrations (0, 125, 250, 500, and 1000 mg/L) on the growth of L. theobromae. Scale bar = 2 cm. (b) The inhibition rates of L. theobromae with different concentrations of 5,6-dimethylbenzimidazole. (c) The inhibition rates of L. theobromae with different concentrations of pyrrole-2-carboxylic acid. (d) Biomass statistics of L. theobromae with 500 mg/L pyrrole-2-carboxylic acid. (e) The qRT-PCR analysis of L. theobromae after 5,6-dimethylbenzimidazole treatment. (f) The qRT-PCR analysis of L. theobromae after pyrrole-2-carboxylic acid treatment. Different lowercase letters indicate significant differences between groups (P < 0.05), ****P < 0.0001

Discussion

Among biocontrol agents, Streptomyces sp. stands out for suppressing plant diseases through unique traits and metabolite-mediated antagonism of phytopathogens (Wang et al. 2024). In this study, strain MEPP0209 exhibited strong antagonistic activity against L. theobromae. We further explored the potential molecular mechanisms of MEPP0209 in antagonizing L. theobromae using transcriptome and non-targeted metabolome analyses.

Microscopic observation revealed abnormal hyphal morphology in L. theobromae following MEPP0209 treatment, including deformation, increased terminal branching, swelling, and vacuolization (Figs. 3d-f). Consistent with these findings, previous studies have reported that pathogenic fungi often exhibit abnormal hyphal morphology when antagonized (Naing et al. 2014; Asril et al. 2022). Two interconnected mechanisms likely explain these effects: One of the mechanisms is resource competition, under conditions of limited nutrients and space, actinomycetes like Streptomyces outcompete pathogenic fungi for survival, disrupting their normal growth (Ebrahimi-Zarandi et al. 2022). Pathogens may adapt by increasing hyphal branching to expand surface area for nutrient acquisition (Baranger et al. 2020). Another is metabolite-mediated antifungal activity, which may induce abnormal growth of mycelium (Guo 2010; Kavitha et al. 2005). In this study, CFS had a significantly stronger inhibitory effect on L. theobromae than VOCs (Fig. 4), and spraying D. officinale seedlings with CFS notably reduced the incidence of stem rot (Fig. 5, Supplementary Fig. S7), suggested that the primary antagonistic substances in MEPP0209 were concentrated in the CFS. By analyzing the non-targeted metabolomics of MEPP0209 CFS, it was found that differential metabolites such as pyrrole-2-carboxylic acid and 5,6-dimethylbenzimidazole were significantly upregulated (Supplementary Table S11). Among which, Nguyen et al. (Nguyen et al. 2015) demonstrated that pyrrole-2-carboxylic acid, a secondary metabolite produced by Streptomyces, exhibits antifungal activity and significantly inhibits the growth of Phytophthora capsici hyphae. Gálvez-Iriqui et al. (Gálvez-Iriqui et al. 2019) found that CS-PCA (chitosan-1H-pyrrole-2-carboxylic acid) biocomposite materials alter fungal morphology, leading to abnormal hyphal structures. Consistently, pyrrole-2-carboxylic acid inhibited the growth of L. theobromae and altered its colony morphology in this study (Fig. 9a). Compared with the control group, the hyphae of L. theobromae became loose and their biomass significantly decreased (Figs. 9a and 9 d). Benzimidazole compounds, known to inhibit fungal growth (Yin et al. 2023), exhibit antibacterial activity due to their structural similarity to purines (Bansal et al. 2019), which are essential for nucleic acid and protein biosynthesis (Shi et al. 2014). As competitive inhibitors, benzimidazole derivatives can substitute for purines, blocking fungal nucleotide biosynthesis and interfering with spindle formation during mitosis. This disruption affects cell division and inhibits fungal reproduction (Bai et al. 2024). In this study, the application of 5,6-dimethylbenzimidazole significantly inhibited the growth of L. theobromae hyphae (Figs. 9a and 9b), consistent with previous research.

Correspondingly, transcriptome analysis of pathogen responding to MEPP0209 CFS revealed that downregulated genes were mainly associated with fungal growth and proliferation. The broad repression of DNA replication/cell-cycle machinery (e.g., MCM2/3/4/5/6/7, ORC2, ORC4, CDC45, POLA1, MRC1, MBP1, NDT80) indicated that cell division and proliferation processes were constrained. Specifically, the microchromosome maintenance (MCM) complex is crucial for initiating DNA replication, which provides DNA unwinding functionality. It can maintain correct DNA replication during normal cell division (Dang et al. 2011; Shastri et al. 2021). In our transcriptome analysis, the genes MCM2/3/4/5/6/7 showed comprehensive downregulation (Fig. 7b). POLA1 encodes the catalytic subunit of DNA polymerase alpha, which, together with the primase complex, initiates the DNA replication process. The complete absence of this gene is considered fatal (Starokadomskyy et al. 2021). In fungi, NDT80 encodes a protein specifically expressed during meiosis, serving as a critical transcription factor for cells entering the first chromosome segregation stage. Phosphorylation of the Ndt80 protein is essential for the normal progression of the meiotic cycle (Chu and Herskowitz 1998; Shubassi et al. 2003). In budding yeast, Mbp1 is a core component of G1/S transcriptional regulation, responsible for coordinating the expression of genes related to DNA synthesis and S-phase entry (Harris et al. 2013). In Fusarium verticillioides, Mbp1 acts as an APSES-type transcription factor that regulates cell cycle progression, and FvMbp1-Swi6 complex controls vegetative growth, stress tolerance, and virulence (Huang et al. 2024). Beyond cell-cycle transcription, mbp1 also contributes to hyphal fitness via cell-envelope programs: in Pleurotus ostreatus, disruption of mbp1 impairs normal mycelial growth and perturbs β-glucan/chitin synthesis, supporting a role for mbp1 in sustaining cell wall integrity during hyphal expansion (Kojima et al. 2024). These genes were significantly downregulated in our transcriptome data (Fig. 7b), and were also validated by qRT-PCR results of pathogen treated with metabolites 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid (Figs. 9e and 9f). The suppressed expression of genes related to cell division and cell cycle was consistent with the inhibition of fungal growth, offering a crucial molecular foundation for the reduction of fungal growth observed following treatment with MEPP0209 CFS. Based on a comprehensive analysis of non-targeted metabolomics, transcriptomics, and in vitro antagonistic assays, the inhibition of fungal growth by MEPP0209 may be associated with the disruption of expression of genes related to cell division and cell cycle in L. theobromae. The two candidate antagonistic compounds derived from MEPP0209, 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid, may be involved in this inhibitory effect.

ROS assume a critical function in regulating cell viability, with their accumulation serving as a key biochemical indicator of cellular dysfunction. This buildup precipitates oxidative injury to cellular constituents, induces enzyme deactivation, and results in cellular swelling (Kobayashi et al. 2002). In this study, ROS accumulation was observed in L. theobromae hyphae treated with MEPP0209 CFS (Supplementary Fig. S6), suggesting that metabolites of strain MEPP0209 induce ROS accumulation in L. theobromae, resulting in cellular damage and abnormal hyphal morphology (Figs. 3d-f and 9a). Combined with transcriptome analysis and qRT-PCR results, MEPP0209 treatment significantly enhanced pathways associated with detoxification/oxidative stress response (Fig. 6c) and upregulated related genes, such as COX17 and PEX11B (Figs. 9e and 9f). COX17 has been shown to influence fungal superoxide accumulation, regulate ROS clearance by pathogenic fungi, and enhance fungal survival and pathogenicity (Ding et al. 2022). The PEX11B gene encodes a peroxidase that synthesizes peroxisomes, which play a key role in maintaining the intracellular balance of ROS. This regulation is essential for plant pathogenic fungi to clear ROS and successfully establish infection (Gao et al. 2017). This suggested that L. theobromae activates oxidative stress defenses to counteract MEPP0209 metabolites. In response to stress, besides the activation of detoxification capabilities, energy metabolism also showed an enhancement, as evidenced by the enrichment of corresponding metabolic pathways in transcriptomic analysis, such as starch and sucrose metabolism, glycolysis/gluconeogenesis, and pentose phosphate pathway (Fig. 6c). Together, the L. theobromae RNA-seq data suggested a classic stress–survival reallocation under MEPP0209 CFS: energy-generating carbon metabolism and detoxification/antioxidant defenses functions were transcriptionally reinforced, while DNA replication and cell-cycle progression were suppressed, ultimately resulting in growth inhibition rather than growth promotion.

In addition to direct antifungal activity, MEPP0209 may indirectly enhance disease resistance in D. officinale by promoting plant growth. Endophytic bacteria with biological control activity commonly exhibit growth-promoting traits, including the production of siderophores and ACC deaminase, phosphorus solubilization, and nitrogen fixation (Bhattacharyya and Jha 2012; Wang et al. 2021a, b). Siderophores play a vital role in plant growth by providing iron to promote development, enhancing plant resistance to iron deficiency stress, and inhibiting pathogen growth through competition for iron (Cui et al. 2022; Li et al. 2023). Streptomyces sp. have been shown to convert insoluble phosphorus into a bioavailable form, significantly increasing soil phosphorus content throughout the plant growth and development cycle (Bautista-Cruz et al. 2019). Nitrogen-fixing bacteria not only promote plant growth (Onofre-Lemus et al. 2009; Guo et al. 2020) but may also enhance plant defense mechanisms against pathogen colonization (Basak and Paramanik 2023). Additionally, ACC deaminase-producing bacteria can reduce the ethylene content, alleviate the obstacles to plant growth and improve the survival rate of plants. In this study, MEPP0209 demonstrated phosphorus solubilization, nitrogen fixation, and the production of siderophores, ACC deaminase, and urease (Supplementary Fig. S5), suggesting its potential to enhance D. officinale resistance to L. theobromae through growth promotion. However, some limitations of the host plant assays in this study need to be acknowledged. First, host physiological or defense-related indicators were not measured, so the physiological responses of the host plant induced by MEPP0209 against stem rot remain unclear. Second, no commercial fungicide treatment was included as a positive control, so the relative efficacy of MEPP0209 compared with current chemical control measures cannot yet be evaluated. Third, the colonization of MEPP0209 in the host was not examined, and its long-term biocontrol efficacy remains to be further verified. These host plant-related experiments will be further investigated at D. officinale planting bases subsequently to provide more evidence for elucidating the mechanism by which MEPP0209 controls stem rot in D. officinale.

Conclusion

In our study, strain MEPP0209 was able to inhibit the growth of L. theobromae and reduce the disease severity of stem rot in D. officinale tissue-cultured seedlings. The primary antagonistic substances of MEPP0209 were concentrated in CFS. By integrating multi-omics analysis with in vitro antagonistic assays, this study elucidated the potential transcriptional pathways disrupted by MEPP0209 in L. theobromae and identified the potential antagonistic metabolites produced by MEPP0209, including 5,6-dimethylbenzimidazole and pyrrole-2-carboxylic acid. Our study findings indicate that MEPP0209 may serve as a potential biocontrol agent for managing D. officinale stem rot, and provide a basis for further validation in planting bases.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contribution

LDL, ZLJ, HLF and JHJ conceived and designed research. ZLJ, HLF, YJC and YJZ conducted experiments. ZLJ, JS, JD and CYQ analyzed the data. ZLJ, HLF, JG, XYC and LDL drafted the manuscript. All authors have read and agreed to the final manuscript.

Funding

This study was supported by National Natural Science Foundation of China (grant number 32101519); the Key Research and Development Projects in Hainan Province (grant number ZDYF2024XDNY168); the Haikou Science and Technology Planning Project (grant number 2022–008, 2022–011); the Postgraduate Research & Practice Innovation Program of Jiangsu Province (grant number KYCX24_3020).

Data availability

The RNA sequencing (RNA-seq) datasets in this study were deposited in the NCBI database under accession number PRJNA1166406. The non-targeted metabolomic data were deposited in the OMIX database under accession ID OMIX007605. The 16S rDNA sequences of all 43 endophytic bacteria isolated in this study were deposited in the NCBI database, with their GenBank accession numbers listed in Supplementary Table S3. The ITS sequence of the isolated pathogen (L. theobromae) was deposited in the NCBI database, with the GenBank accession number PQ459379.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.