Application of streptomyces maleus WHL7 in prevention and treatment of banana wilt

Abstract

The invention provides application of streptomyces maleus WHL7 in prevention and control of banana wilt, and belongs to the technical field of fungus damage prevention and control. The invention relates to an application of streptomyces maleus WHL7 in inhibition and prevention of banana wilt, which is characterized in that the classification name of the streptomyces maleus WHL7 is Streptomyces maleus, the preservation unit is Guangdong Microbial Culture Collection Center, the preservation date is June 18, 2025, and the preservation number is GDMCC No: 66544. The WHL7 metabolite provided by the invention induces banana G3P to synthesize beneficial microorganisms such as bacillus recruitment and pseudomonas, further inhibits colonization of banana fusarium oxysporum (Foc TR4) in soil, reduces the morbidity of fusarium wilt, and provides a new way for preventing and treating banana fusarium wilt.

Description

Technical Field

[0001] This invention relates to the field of fungal pest control technology, specifically to the application of Streptomyces malayi WHL7 in the control of banana wilt disease. Background Technology

[0002] Bananas and plantains are important fruits and staple crops globally, widely cultivated in tropical and subtropical regions. In 2023, global banana production exceeded 183.67 million tons, with a planted area of ​​12.93 million hectares, supporting the livelihoods of millions (FAO Statistical Database, 2025). However, most cultivated bananas are triploid and reproduce asexually, resulting in a narrow genetic background and increased susceptibility to pests and diseases. Fusarium wilt, caused by *Fusarium oxysporum* specializing in tropical physiological race 4 (Foc TR4), is particularly devastating. This pathogen can infect over 80% of banana varieties, leading to severe yield losses. Its chlamydospores can survive in the soil for over 30 years without a host, making disease management extremely challenging. While chemical fungicides and cultivation practices have been used for control, their effectiveness remains limited. These fungicides also pose significant environmental pollution risks. Biological control is a promising and sustainable alternative for controlling this devastating soil-borne disease.

[0003] Biocontrol microorganisms are valuable biological resources that can be used to control plant diseases and reduce agriculture's reliance on chemical pesticides. Commonly used strains include Bacillus, Pseudomonas, Trichoderma, and Actinomycetes. Among them, Streptomyces are particularly noteworthy because they can produce a variety of bioactive compounds, showing great potential in the biocontrol of plant diseases. Our recent research has identified several Streptomyces strains with antifungal activity against Foc TR4, such as Streptomyces CB-75, SCA3-4, and YYS-7. Despite extensive research, few biocontrol agents have been successfully implemented on a large scale in agriculture. Their main limitations include poor environmental stability, inconsistent field efficacy, narrow host specificity, and insufficient soil persistence. A key knowledge gap lies in understanding the interaction between biocontrol agents and the soil microbial community, which profoundly affects the efficacy of biocontrol agents. These unresolved challenges hinder the application of biocontrol microorganisms in sustainable agriculture.

[0004] The rhizosphere is a crucial interface for plant health and soil ecosystem function. This dynamic microenvironment fosters a rich microbial community. These microbes interact with plant roots through complex signal transduction and metabolic exchange, forming symbiotic or antagonistic relationships. Rhizosphere microbes make significant contributions to plant disease resistance. Beneficial microbes can directly inhibit pathogens by competing for nutrients and niches, producing antimicrobial compounds, or interfering with pathogenic signal transduction. Furthermore, they indirectly enhance plant immunity by activating systemic resistance mechanisms, such as induced systemic resistance (ISR) and acquired systemic resistance (SAR). Moreover, the diversity and stability of the rhizosphere microbial community are closely related to plant disease resistance. A robust and diverse rhizosphere microbiome can serve as a "microbial barrier" against pathogen infection. However, the complex interactions between biocontrol agents and host plants remain unclear.

[0005] The aggregation of rhizosphere microbial communities is controlled by the dynamic interactions of plant genotype, soil properties, and environmental factors. Root exudates, such as organic acids, sugars, amino acids, flavonoids, and plant hormones, are the primary mediators of communication between plants and microorganisms. They recruit microorganisms as nutrient substrates and molecular signals. Compared to non-rhizosphere soils, rhizosphere microbiomes exhibit higher diversity and metabolic activity. Their composition is selectively regulated by plant-derived biochemical signals. Notably, the dialogue between plants and microorganisms is bidirectional. For example, beneficial microorganisms such as *Pseudomonas* can alter host amino acid cycling pathways, thereby changing exudate profiles, while *Streptomyces* significantly modulate plant root metabolic patterns and promote the accumulation of immune-related metabolites such as flavonoids, organic acids, and phenylalanine. However, our understanding of *Streptomyces*-plant metabolite-microbial community interactions remains incomplete, particularly regarding metabolite function, molecular mechanisms, and in-situ ecological performance. Summary of the Invention

[0006] The purpose of this invention is to propose the application of Streptomyces maleicus WHL7 in the control of banana wilt disease, to inhibit the colonization of banana wilt pathogen (Foc TR4) in the soil, reduce the incidence of wilt disease, and provide a new approach for the control of banana wilt disease.

[0007] The technical solution of this invention is implemented as follows: This invention provides the application of *Streptomyces malayi* WHL7 in the control of banana wilt disease, wherein *Streptomyces malayi* WHL7 is classified as... Streptomyces malaysiensis The depositary institution is Guangdong Provincial Center for Microbial Culture Collection, the deposit date is June 18, 2025, and the accession number is GDMCC No: 66544.

[0008] As a further improvement of the present invention, the banana wilt disease is a disease caused by the fungus *Fusarium wiltii*.

[0009] As a further improvement of the present invention, the banana wilt pathogen is tropical race 4 of the banana wilt pathogen.

[0010] As a further improvement of the present invention, the application method is to use the crude extract of fermentation broth of Streptomyces malayi WHL7 to prepare bacterial agents or drug compositions for the prevention and control of diseases.

[0011] As a further improvement of the present invention, the method for preparing the crude extract of the fermentation broth of Streptomyces malayi WHL7 is as follows: S1: Activation of Streptomyces malayi WHL7 strain; S2: Inoculate the activated Streptomyces maleicus WHL7 into the soybean flour fermentation medium, culture, add anhydrous ethanol, continue culture, and filter out the residual bacterial cells to obtain the fermentation broth. S3: Concentrate and collect the fermentation broth to obtain the crude extract of the fermentation broth.

[0012] As a further improvement of the present invention, the soybean powder fermentation culture medium formula in step S2 is as follows: 15-25g soluble starch, 10-20g soybean powder, 3-8g yeast powder, 1-3g peptone, 3-6g CaCO3, 3-7g NaCl, to be brought to a final volume of 1L, and the pH value is adjusted to 6.8-7.2.

[0013] As a further improvement of the present invention, the culture conditions described in step S2 are 100-200 rpm and 25-30°C for 5-10 days.

[0014] As a further improvement of the present invention, the culture time in step S2 is 0.5-1.5 days.

[0015] As a further improvement of the present invention, the amount of anhydrous ethanol added in step S2 is the same as the volume of the culture medium.

[0016] The present invention has the following beneficial effects: The present invention proposes that WHL7 metabolites recruit beneficial microorganisms such as Bacillus and Pseudomonas by inducing the synthesis of banana G3P, thereby inhibiting the colonization of banana wilt pathogen (Foc TR4) in the soil, reducing the incidence of wilt disease, and providing a new approach for the prevention and control of banana wilt disease. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 Antifungal activity and genomic characteristics of *Streptomyces* WHL7: (a) Assessment of the antifungal activity of volatile organic compounds (VOCs) in strain WHL7 against Foc TR4 growth. (b) Circular genome map of *Streptomyces* WHL7. Loops 1–6 represent (from inside to outside): genome-scale markers, coding sequences on the forward and reverse strands, rRNA and tRNA sites, GC content, and GC tilt. (c) KEGG annotation of the *Streptomyces* WHL7 genome. (d) Prediction of antifungal metabolites in strain WHL7. (e) Taxonomic identification of *Streptomyces* WHL7. Average nucleotide identity (ANI) values ​​are shown in the last lane. (f) Number of single-copy orthologs and unique paralogs in 12 selected *Streptomyces* genomes. (g) KEGG enrichment analysis of unique paralogs in the *Streptomyces* WHL7 genome. (h) Broad-spectrum antifungal analysis of Streptomyces WHL7.

[0019] Figure 2 Characteristics of banana bulbs infected with Foc TR4 at 14 dpi in sterilized soil.

[0020] Figure 3 For the analysis of bacterial community diversity and composition; (a) disease index, chlorophyll content, plant height and stem diameter of bananas under different treatments. An asterisk indicates a significant difference ("*" p<0.01, "*" p<0.01, "***" p<0.001, "****" p<0.0001). (b) Relative abundance of bacterial phyla in different samples. (c) Abundance of Streptomyces in different treatment groups. (d) Bacterial richness assessed using the Chao1 index. (e) Bacterial diversity assessed using the Shannon index. (f) β-diversity of bacterial communities between treatment groups visualized using PCoA. (g) Differential abundance of ASVs between the Foc TR4 and WHL7+Foc TR4 treatments. Blue and red dots represent downregulated and upregulated ASVs in the WHL7+Foc TR4 group, respectively. The size of the dot represents the abundance of the ASV. (h) Differential abundance of ASVs at the class level between the Foc TR4 and WHL7+Foc TR4 treatments.

[0021] Figure 4 Clustering of microbiome WGCNA modules. (a) Samples of microbiome clustering analysis. (b) Module correlation clustering and heatmap of correlation between modules. (c) Proportion of different bacteria in gray modules.

[0022] Figure 5To identify disease-associated ASVs using WGCNA analysis; (a) Cluster dendrograms and correlation heatmaps of metabolite co-expression modules show the relationships between modules. Each color represents a different module. (b) Heatmaps show the correlation between modules and incidence / chlorophyll. Blue and red indicate negative and positive correlations, respectively. Numbers in parentheses represent p-values. (c) Scatter plots show the correlation between ASV importance (GS) and module membership (MM). (d) The network of gray modules is negatively correlated with incidence. Node size reflects the relative abundance of ASVs, and colors represent different phyla. (e) Co-expression network of central ASVs. (f) Overlap results of WGCNA and DAA in Venn diagrams. (g) Antimicrobial evaluation of isolated strains against Foc TR4.

[0023] Figure 6 Analysis of banana root metabolites at different treatments and specified time points: (a) The correlation between the ability of Streptomyces WHL7 to directly inhibit pathogens (i.e., its antibacterial effect against Foc TR4 on YE solid medium) and the disease index in natural soil was analyzed. (b) PCA analysis of root metabolites in the CK, WHL7, Foc TR4, and WHL7+Foc TR4 groups at 0, 1, and 3 dpi. (c) Volcano plot showing the difference in median elevation between WHL7+Foc TR4 and Foc TR4 treatments at 1 dpi and 3 dpi. Blue and red dots represent upregulated and downregulated metabolites after WHL7+Foc TR4 treatment, respectively. (d) Correlation heatmap illustrating the relationship between metabolite modules and disease incidence / chlorophyll content. (e) Scatter plot showing the correlation between metabolite importance (GS) and module membership (MM). (f) Hub metabolite co-expression network constructed using 12 methods. (g) Venn diagram highlights the overlap between DAMs and WGCNA in the salmon module. (h) Comprehensive analysis of key metabolites and beneficial ASVs. Colors indicate correlation coefficients (red: positive correlation; blue: negative correlation).

[0024] Figure 7 KEGG enrichment analysis for differentially expressed metabolites. (a) Venn diagrams of upregulated DAMs identified at 1 dpi and 3 dpi after WF treatment. (b) KEGG enrichment of 42 common DAMs was analyzed.

[0025] Figure 8Clustering of WGCNA modules for the metabolome. (a) Module correlation clustering and correlation heatmap within each module. (b) Cluster dendrogram and correlation heatmap of metabolite co-expression modules showing the relationships between modules. Each color represents a different module. (c) Visualization of important salmon modules using the network. Nodes and size represent metabolites and connectivity, respectively. Red and blue nodes represent upregulated and downregulated metabolites. Green nodes represent metabolites with no significant changes.

[0026] Figure 9 To validate the recruitment of beneficial bacteria for G3P; (a) the growth response trend of beneficial bacteria to G3P; (b) the growth rate of beneficial bacteria after 24 hours of G3P treatment.

[0027] Figure 10 (a) Experimental protocol for G3P treatment of banana root Foc TR4 infection. (b) Correlation between G3P and disease incidence. (c) GFP-Foc TR4 infection in banana corms and roots after 14 dpi. White arrows indicate green fluorescence of Foc TR4. Horizontal axis = 150 μm. (d) Abundance of Foc TR4 at different G3P concentrations detected by qPCR. (e) Microbial species accumulation curve. (f) Relative abundance of bacterial phyla in different samples. (g) Assessment of bacterial richness using the Chao1 index. (h) Assessment of bacterial diversity using the Shannon index. (i) Visualization of β-diversity of bacterial communities among treatment groups using PCoA. (j) Bar graph of LEfSe analysis results of rhizosphere microorganisms in G3P-treated and control groups. This graph shows the log-transformed LDA scores of bacterial taxa identified by LEfSe analysis, with a threshold of 2.0 for the log-transformed LDA score. (k) The overlapping results of beneficial microorganisms were enriched after treatment with Streptomyces WHL7 and G3P.

[0028] Figure 11 To construct a standard curve for Foc TR4 using qPCR.

[0029] Figure 12 This indicates the antifungal activity of G3P against Foc TR4. Detailed Implementation

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] The classification of Streptomyces malayi WHL7 is as follows: Streptomyces malaysiensis The depositary institution is Guangdong Provincial Center for Microbial Culture Collection, located at 5th Floor, Building 59, No. 100 Xianlie Middle Road, Guangzhou. The deposit date is June 18, 2025, and the accession number is GDMCC No: 66544. Example 1

[0032] 1. Genome annotation and classification of Streptomyces WHL7 Genomic DNA was extracted from *Streptomyces* WHL7 using the TIANGEN bacterial DNA kit (Tiangen Biotech Co., Ltd., Beijing, China), strictly following the manufacturer's instructions. DNA integrity was assessed by 1% (w / v) agarose gel electrophoresis, and purity was determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA), with an OD260 / 280 ratio between 1.8 and 2.0 considered acceptable. DNA concentration was further quantified using a Qubit quantitative PCR instrument (Invitrogen, USA). Whole-genome sequencing was performed using the Illumina HiSeq × 10 platform (Hangzhou Majorbio Co., Ltd., Shanghai, China), with paired-end sequencing reads of 150 bp in length. Raw data were filtered using FastP v1.99.2 software, with parameters including removal of adapter sequences, bases with a quality score <20, and filtering of reads shorter than 50 bp. Clean reads were assembled de novo using Spades v4.0.0 with default parameters, and the assembly results were evaluated using QUAST v5.2.0, outputting metrics including N50, L50, GC content, and total assembly length. Protein coding sequences (CDs) were predicted using Prodigal v2.6.3 software. Based on the results, structurally incomplete gene sequences were removed, retaining complete genes containing both start and stop codons. The predicted protein coding sequences were annotated using the emapper v2.1.12 program with the EggNOG database (v5.0) as a reference, with an e-value less than 1e-5 as a significance threshold. Functional enrichment analysis of the annotated genes was performed using the clusterProfiler v4.12.6 package in R v4.2.1, with a false positive rate (FDR) <0.05 as a significance criterion for KEGG pathway enrichment analysis. Secondary metabolite biosynthetic gene clusters (BGCs) were predicted using antiSMASH v7.1, enabling comprehensive detection and comparison with known biosynthetic pathways.

[0033] Using WHL7 genomic DNA as a template and universal primers, the introduced sequences are shown in Table 1 below.

[0034] Table 1

[0035] The PCR system is shown in Table 2 below.

[0036] Table 2

[0037] 2. Comparative genomic analysis of Streptomyces strains in different ecological niches Twelve *Streptomyces* genomes from different ecological niches were downloaded from the NCBI database (www.ncbi.nlm.nih.gov / datasets / genome) for comparative genomic analysis. Unique paralogous genes between different strains were identified using the same method with OrthoFinder2 software. Specific paralogous genes of strain WHL7 were extracted using seqkit software and annotated using emapper v2.1.12 with reference to the EggNOG database (v5.0), with default parameters set (e-value = 1e-5). KEGG pathway enrichment analysis was performed on the specific paralogous genes of strain WHL7 using ClusterProfiler v4.12.6.

[0038] Sequencing was performed by Sangon Biotech Co., Ltd. The obtained 16S rRNA sequence (SEQ ID NO. 3, Accession: MW429336) was then BLAST-aligned in the NCBI database (https: / / www.ncbi.nlm.nih.gov / ). Based on the homology analysis results, the genome sequences of 15 closely related strains were obtained. Single-copy orthologous genes among these strains were identified using OrthoFinder2 software, and all single-copy orthologous gene sequences were extracted using seqkit software. Multiple sequence alignment of the single-copy orthologous genes was performed using ParaAT software, and the concat module of seqkit software was used to merge the alignment results in the same order. The average nucleotide identity (ANI) between *Streptococcus pyogenes* WHL7 and the selected strains was calculated using FastANI v1.33. Example 2

[0039] 1. Determination of antifungal activity of volatile organic compounds (VOCs) in Streptomyces WHL7 The inhibitory activity of volatile organic compounds (VOCs) produced by *Streptomyces* WHL7 against *Foc* TR4 was evaluated using a two-plate assay, a minor modification of the previously described method. Briefly, strain WHL7 was inoculated onto yeast extract (YE) basal medium agar plates (10 g / L yeast extract, 20 g / L agar) and incubated at 28°C for 48 hours. Simultaneously, 5 mm diameter *Foc* TR4 mycelial discs (taken from the edge of 5-day-old colonies grown on potato dextrose agar [PDA; 200 g / L potato extract, 20 g / L dextrose, 20 g / L agar]) were transferred to the center of a fresh PDA plate. In the two-culture protocol, a YE plate containing WHL7 and a PDA plate containing *Foc* TR4 were sealed together in an inverted, face-to-face configuration and sealed with a triple paraffin film to allow VOC exchange while preventing physical contact. The plates were incubated at 28°C for 6 days. The control group consisted of PDA culture plates inoculated with Foc TR4 and cultured on sterile YE medium. Each treatment group was tested in triplicate for each assay, with the experiment being independently repeated twice (a total of 6 assays per treatment group).

[0040] 2. Broad-spectrum antifungal activity of Streptomyces WHL7 Detection of Streptomyces WHL7 against Alternaria leaf spot disease in mango ( A. tenuissima ), gray mold ( B. cinerea Banana long spot disease C. fallax Banana gray spot disease C. lunata ),anthrax( C. gloeosporioides Strawberry anthracnose ( C. fragariae Anthracnose of peppers ( C. acutatum ), Foc TR1 ( F. oxysporum ), wheat scab ( F. graminearum ) and cucumber wilt ( F. oxysporum Antifungal activity against pathogens (f. sp.). All fungal strains were provided by the Institute of Tropical Biotechnology, Chinese Academy of Tropical Agricultural Sciences. The antagonistic activity of WHL7 against these plant pathogens was determined using a slightly modified plate antagonism test. Briefly, a 5 mm mycelial disc was cut from the active growth edge of a 5-day-old fungal colony and placed in the center of a PDA plate. Strain WHL7 was inoculated at four equidistant points around the mycelial disc, each 2.5 cm from the pathogen inoculation point. The control plate contained only the pathogen mycelial disc and was not inoculated with strain WHL7. All plates were incubated at 28°C in the dark for 7 days. Antifungal activity was assessed by measuring the radial growth of the fungal colonies, and the inhibition rate was calculated.

[0041] To further clarify the taxonomic position of strain WHL7, we constructed a phylogenetic tree using 16 Streptomyces genomes. Based on 1,471 single-copy orthologous genes, phylogenetic analysis using maximum likelihood alignment showed that strain WHL7 formed a stable cluster with Streptomyces malaysia, with a bootstrap value of 100% and an ANI of 98.71%. Figure 1 e, Table S9). These results definitively identified the strain as *Streptomyces maculata* WHL7. To explore its potential as an antifungal agent, we used the antiSMASH software to predict the biosynthetic gene clusters (BGCs) in the *Streptomyces maculata* WHL7 genome. Analysis identified several BGCs encoding polyketide compounds and other secondary metabolites with presumed antifungal properties. Figure 1 d), including geldmycin (100% similarity), kulamycin (100% similarity), nigramycin (94% similarity), hygromycin (96% similarity), and azaclomycin F3a (91% similarity). In addition, two siderophore synthesis gene clusters were detected ( Figure 2 This suggests that it may play a role in microbial competition.

[0042] Considering the genetic potential of antifungal metabolite production, we evaluated the antifungal activity of Salmonella Malaysia WHL7 against 10 plant pathogenic fungi. Figure 1 The strain exhibited broad-spectrum inhibition rates, ranging from 45.40% to 77.67%. It showed the strongest inhibition against Foc TR4 (77.67 ± 0.55%), followed by *Alternaria tenuissima* (76.09 ± 1.16%), *Colletotrichum gloeosporioides* (74.66 ± 1.13%), and *Curvularia lunata* (72.28 ± 2.03%). In contrast, it showed the weakest inhibition against *Colletotrichum fallax* (45.40 ± 1.18%). Example 3

[0043] 1. Collection of rhizosphere soil from banana plants inoculated with Streptomyces WHL7 To investigate the effects of Streptomyces WHL7 on the rhizosphere microbial community structure, four experimental treatments were set up: (i) inoculation with Foc TR4 alone (10 6 CFU g -1 Soil; Foc TR4); (ii) Inoculate WHL7 alone (10 6 CFU g -1Soil; WHL7); (iii) Co-inoculation of WHL7 and Foc TR4 (10 each) 6 CFU g -1 Soil; WHL7 + Foc TR4); (iv) Watering-only control (CK). Each treatment used 40 healthy, uniformly shaped banana plants ( Musa acuminata Cavendish cv. 'Brazil' seedlings were transplanted into soil. In the Foc TR4 and WHL7 + Foc TR4 groups, the soil was mixed with a concentration of 10 6 CFU g -1 The Foc TR4 spore suspension was thoroughly mixed and inoculated. After culturing at room temperature for 24 hours, the WHL7 fermentation broth was added to the soil of the WHL7 + Foc TR4 and WHL7 groups to achieve a final density of 10. 6 CFU g -1 All plants were cultured in a greenhouse at 28°C, 70% relative humidity, and natural light for 30 days. Rhizosphere soil samples were collected on day 30 post-treatment. Procedure: Banana seedlings were carefully uprooted, gently shaken to remove loose soil, and the soil tightly adhering to the roots was brushed off. Samples were collected in 50 mL sterile tubes. Samples were immediately transported on dry ice and stored at -80°C until DNA extraction and 16S amplicon sequencing were performed. Disease index, plant height, stem diameter, and chlorophyll content (SPAD values ​​measured using a Konica Minolta SPAD-502 Plus chlorophyll meter) were recorded for each treatment group.

[0044] 2. Effects of Streptomyces WHL7 on the rhizosphere microbial community of banana The bacterial community composition of banana rhizosphere soil was assessed by high-throughput sequencing of the hypervariable V3-V4 region of the 16S rRNA gene. Genomic DNA was extracted from rhizosphere soil samples using the FastDNA Soil Centrifugation Kit (MP Biomedicals, USA) according to the manufacturer's instructions. The V3-V4 region was amplified using universal primers, and the sequences are shown in Table 3.

[0045] Table 3

[0046] After purification and quantification, the PCR products were sequenced on the Illumina HiSeq platform (Magigene, Guangzhou, China) to obtain 2×250 bp paired-end reads. Raw reads were processed using fastp v1.99.2 to obtain high-quality sequences, including adapter removal, removal of reads shorter than 50 bp, and trimming of bases with a Q-score <20. Chimeric sequences were identified and removed using VSEARCH v2.29.2. Clean reads were merged with FLASH v1.2.11 (minimum overlap 10 bp, mismatch ≤ 2 bp) and de-noiseed using the deblur plugin in QIIME2 v2023.5 to obtain amplicon sequence variants (ASVs) with 100% similarity. Representative ASVs were classified according to the SILVA 138 database (version 138.1) with a confidence threshold of 80%. Alpha diversity indices, including Chao1 (species richness) and Shannon (community diversity), were calculated in QIIME2. Statistical comparisons of α diversity between treatments were performed using the Wilcoxon rank-sum test in R (v4.2.1). β diversity was assessed based on a weighted UniFrac distance matrix and visualized using principal coordinate analysis (PCoA). Statistical significance of differences in β diversity between groups was determined using permuted multivariate analysis of variance (PERMANOVA). Differential abundance ASVs (DAAs) were identified using the ANCOM method in the Composition module of QIIME2, with thresholds set at fold change (FC) ≥ 2 and false discovery rate (FDR) ≤ 0.01.

[0047] 3. Effects of Streptomyces WHL7 on banana root metabolomics Banana root samples were collected at 0, 1, and 3 days (dpi) after inoculation with Streptomyces WHL7 for non-targeted metabolomics analysis. Sample preparation and metabolite extraction were performed according to the standard protocol provided by Beijing Biomark Technology Co., Ltd. Metabolite detection was performed using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS / MS). Raw data were processed to remove metabolites with missing values ​​from more than 50% of the samples. Remaining features were normalized using the total ion intensity of each sample, followed by log2 transformation to improve data distribution. Principal component analysis (PCA) was used to visualize overall metabolic changes across different treatment groups and time points. Differentially accumulating metabolites (DAMs) were identified using the limma software package (v3.60.6, R Bioconductor), with a threshold set at |log2 (fold change)| ≥ 1 and p < 0.05. DAMs were functionally enriched using the KEGG database via MetaboAnalyst 6.0 software, and FDR-corrected p < 0.05 was used to determine statistical significance of pathway enrichment.

[0048] Our previous research showed that the Streptomyces WHL7 strain can inhibit Foc TR4 infection in banana roots. Furthermore, compared with Foc TR4 alone, WHL7 treatment significantly reduced disease incidence, significantly increased chlorophyll content, and promoted plant height and stem diameter development. Figure 3 a). However, in sterilized soil, the resistance of the WHL7 strain did not differ significantly ( Figure 3 This indicates that Streptomyces WHL7 primarily inhibits FocTR4 through modulating the rhizosphere microbial community, rather than through direct antagonism. To validate this hypothesis, we evaluated plant health and soil microbiome under four treatments: control (CK), Foc TR4 inoculation (F), WHL7 inoculation (W), and co-inoculation of WHL7+Foc TR4 (WF), and monitored rhizosphere soil bacterial community composition using Illumina MiSeq sequencing of 16S rRNA genes. Bacterial community composition analysis showed that, in the presence of Foc TR4, WHL7 application altered the relative abundance of major bacterial phyla, significantly increasing Proteobacteria and decreasing Acidobacteria (…). Figure 1 b S2). There was no significant difference in the relative abundance of Streptomyces intercalans among the treatments ( Figure 1 c). Alpha diversity index, including Chao1 richness ( Figure 1 d) and Shannon diversity ( Figure 1e), indicating that WHL7 mitigated the decline in microbial diversity caused by Foc TR4 infection. Beta diversity analysis based on principal coordinates showed that the microbial communities under all four treatments (P=0.004) exhibited significant clustering, with WHL7+Foc TR4 clustering together with Foc TR4 (e). Figure 1 f). The abundant differential amplicon sequence variations (DAA) between the Foc TR4 and WHL7+Foc TR4 treatment groups indicated that the WHL7+Foc TR4 group had 302 significantly enriched taxa and 222 missing taxa. Figure 1 g). Taxonomic classification shows that the diverse ASVs (DAA) mostly belong to the phylum Gammaproteobacteria, among which Pseudomonas genus ( Pseudomonas ) is the dominant genus ( Figure 1 In summary, these results indicate that WHL7 inhibits banana wilt in non-sterile soils by reorganizing the rhizosphere microbial community, which favors beneficial communities such as Proteobacteria and reduces potentially harmful communities such as Acidobacterium, thereby enhancing plant health and resistance to Foc TR4.

[0049] To identify key microorganisms inhibiting Foc TR4 infection in banana roots, WGCNA analysis based on rhizosphere microbial ASV abundance identified unique modules highly correlated with disease inhibition. Treatment with strain WHL7 significantly reduced the correlation between sample and disease incidence. Figure 4 a). The clustering tree divides ASV into 12 co-expression modules, each represented by a unique color ( Figure 5 (a, 4b). Correlation analysis of the characteristic genes of the gray module and the phenotypic traits showed that the gray module was significantly negatively correlated with the incidence of disease (r = -0.76, p = 0.004) and moderately positively correlated with chlorophyll content. Figure 5 b). In the gray module, MM was significantly associated with GS (r = 0.58, p < 1 × 10-200), indicating that highly connected ASVs were also most associated with disease suppression ( Figure 5 c). Network visualization of the gray modules shows that Proteobacteria (41.22%), Acidobacteria (19.52%), and Bacteroidetes (12.41%) dominate the composition of the hub ASVs, and the node size is proportional to their relative abundance and color, indicating the gating affiliation ( Figure 5 d, 4c). Detailed co-expression networks further highlight the major pivotal taxa, including Bacillus, Pseudomonas, and Streptomyces, which form tightly interconnected subnetworks ( Figure 5 e). Integrating the WGCNA results with differential abundance analysis (DAA), 143 overlapping ASVs were identified from the gray module, all of which were central taxonomic units ( Figure 5 f). Therefore, these groups may be key contributors to WHL7-mediated natural soil-borne disease suppression. To verify their functional relevance, in vitro antibacterial activity against Foc TR4 was tested on representative strains corresponding to the central ASV. Subsequently, 20 bacterial strains, mainly Bacillus and Pseudomonas, were isolated from rhizosphere soil. Several Bacillus isolates (5-36, 5-88, 5-136) and one Pseudomonas isolate (6-34) showed strong inhibitory effects on pathogen growth in PDA plate assays. Figure 5 (g), which is consistent with their potential role in suppressing banana wilt disease in vivo. These findings provide strong evidence that the suppression of WHL7-related diseases is mediated by the enrichment of specific beneficial taxa in the rhizosphere microbiome. Example 4

[0050] Identification of key ASVs and metabolites associated with wilt disease incidence To reveal differentially abundant ASVs (DAA) and differentially accumulated metabolites (DAM) associated with Foc TR4 infection, we integrated amplicon sequencing and metabolomics datasets and analyzed them using weighted gene co-expression network analysis (WGCNA) in the R package (v1.73). ASVs and metabolites were clustered using an optimized soft-threshold power based on a topological overlap differential measure (1-TOM). Modules with fewer than 30 members were excluded from downstream analysis. Subsequently, we correlated representative ASVs and metabolites in each module with disease incidence. By integrating DAA, DAM, and WGCNA results, we identified key pivotal ASVs and metabolites closely associated with disease progression.

[0051] 1. Isolation of beneficial rhizosphere microorganisms A fresh soil sample (5 g) was suspended in 45 mL of sterile distilled water and stirred at 180 rpm for 1 h at 28 °C to prepare a homogeneous soil suspension. The suspension was serially diluted to 10⁻¹, 10⁻², and 10⁻³, with 100 μL of the suspension transferred to 900 μL of sterile water each time. 100 μL of the diluted solution was evenly spread onto an isolation agar plate. The plates were inverted and incubated at 28 °C for 7 days to allow colony development. Three independent biological replicates were performed to ensure reproducibility. After incubation, morphologically distinct colonies were picked, streaked onto fresh LB agar plates, and repeatedly passaged until pure isolates were obtained. The purified isolates were short-term stored on LB agar slants at 4 °C. For molecular identification, genomic DNA was extracted from each isolate, and the 16S rRNA gene was amplified using universal primers (27F and 1493R). PCR products were validated by gel electrophoresis and sequenced by Shanghai Sangon Biotech Co., Ltd. The sequencing results were compared with the EZBioCloud database (https: / / www.ezbiocloud.net / ) to determine the taxonomic classification at the genus and species levels.

[0052] 2. Comprehensive multi-omics network analysis of metabolites and microbial communities Pearson correlation analysis was used to assess the association between rhizosphere microbial community and banana root metabolites after treatment with strain WHL7. Significant ASVs and metabolites were identified based on correlation coefficients and p-values. Furthermore, key metabolites that may contribute to changes in rhizosphere microbial community structure were identified, providing insights into the metabolic drivers behind microbe-metabolite interactions.

[0053] 3. Evaluation of the biocontrol efficacy of key metabolites against Foc TR4 The biocontrol efficacy of glycerol-3-phosphate (G3P) was evaluated through in vitro and pot experiments. In the in vitro experiment, G3P was incorporated into PDA medium at final concentrations of 0.75 nM, 1.5 nM, and 3 nM, with PDA without metabolites serving as a control. A 5 mm disc of Foc TR4 mycelium was placed in the center of each petri dish. After culturing at 28°C for 7 days, colony diameter was measured and inhibition rate was calculated. In the pot experiment, banana seedlings were divided into three treatment groups: (i) water control (CK), (ii) inoculated with Foc TR4 only, and (iii) challenged with Foc TR4 after G3P pretreatment. G3P treatment group: Banana seedlings were pretreated with different concentrations of G3P for two weeks, and then the roots were inoculated with GFP-labeled Foc TR4 (106 CFU g⁻¹ soil). The control group received only water, while the Foc TR4 group was inoculated with the pathogen without G3P pretreatment. All plants were placed in a greenhouse at 28°C and 60% relative humidity, and cultured under natural light. The colonization of GFP-Foc TR4 in the roots was observed using a confocal microscope (FV3000, Olympus, Tokyo, Japan), and rhizosphere soil was collected for further analysis. The GFP-Foc TR4 strain was kindly provided by the Institute of Tropical Biotechnology, Chinese Academy of Tropical Agricultural Sciences (Haikou, China).

[0054] 4. The impact of key metabolites on the rhizosphere microbial community Rhizosphere soil samples were collected from banana plants treated with different concentrations of G3P, with Foc TR4 and water treatment (CK) serving as controls. Bacterial 16S rDNA genes (V3-V4 region) were amplified and sequenced using the Illumina HiSeq platform (Magigene, Guangzhou). A consistent workflow was used to analyze the microbial community composition and diversity of the different treatment groups, and the results were compared with those of the Streptomyces WHL7 treatment group.

[0055] 5. Detection of Foc TR4 abundance in banana rhizosphere soil Rhizosphere soil samples were collected from banana plants in different treatment groups, and fungal abundance (especially Foc TR4) was quantitatively analyzed by real-time quantitative PCR (qPCR). Foc TR4 genomic DNA was serially diluted 10-fold (range from 10...). 8 Up to 10 4A standard curve was constructed using a copy. Soil DNA was extracted using the EZNA® Soil DNA Kit (OmegaBio-Tek, Norcross, Georgia, USA). qPCR assays were performed on a LightCycler® 96 system (Roche, Basel, Switzerland), with five technical replicates per sample. Foc TR4 abundance is expressed as log(s) per gram of soil. 10 The number of copies after conversion. Primer sequences are shown in Table 4: Table 4

[0056] Correlation analysis between the in vitro anti-Foc TR4 activity of Streptomyces WHL7 and its biocontrol effect in natural soil (R = -0.34, p = 0.15, Figure 6 a) showed no significant correlation between the two, indicating that the antibacterial ability of WHL7 was not the direct cause of its in vivo disease suppression. To explore its potential mechanism, principal component analysis (PCA) was used to analyze the root metabolite profiles of banana plants under different treatments (CK, WHL7, Foc TR4, WHL7+Foc TR4) at 0, 1, and 3 days (dpi) after inoculation. The results showed that the metabolite profiles of different treatment groups were significantly separated, especially at 1 dpi, where the plants treated with WHL7 showed significant differences from the pathogen-only group and the control group. Figure 6 b). Differential metabolite analysis revealed 249 significantly altered metabolites between WHL7 + Foc TR4 and Foc TR4-only treatments, of which 98 were downregulated at 1 dpi, 151 were upregulated at 1 dpi, 156 were downregulated at 3 dpi, and 132 were upregulated at 3 dpi. Figure 6 c). Among them, 42 differentially accumulated metabolites (DAMs) showed commonalities at different time points ( Figure 7 a). KEGG enrichment analysis showed that these DAMs were associated with flavonoid biosynthesis, phenylalanine / tyrosine metabolism, α-linolenic acid metabolism, and tryptophan biosynthesis. Figure 7 b). Example 5

[0057] The recruitment mechanism of key metabolites for the enrichment of beneficial microorganisms in banana rhizosphere To investigate how banana root metabolites recruit beneficial microorganisms in the rhizosphere, we conducted bacterial growth and chemotaxis experiments. In the growth experiment, G3P was added to LB liquid medium at a specified concentration. Beneficial strains were inoculated into the medium and cultured with shaking at 28°C for 24 hours. The optical density (OD) at 600 nm was measured using a spectrophotometer. 600To assess bacterial growth, a syringe containing LB medium supplemented with G3P was used as the attractant chamber in the chemotaxis assay. The syringe was immersed in the beneficial bacterial cell suspension and incubated at 28°C for 30 minutes. After incubation, residual liquid at the syringe needle was carefully removed, and the syringe was rinsed three times with sterile water to remove external contaminants. The remaining solution in the syringe barrel was then spread onto LB agar plates and incubated at 28°C in the dark for 24 hours. Colony forming units (CFU) were then counted to quantify the chemotactic response of beneficial bacteria to the tested metabolites.

[0058] Similar to the microbiome results, treatment with the WHL7 strain significantly reduced the correlation between the sample and the incidence rate. Figure 8 To further identify key metabolites associated with incidence, WGCNA analysis of metabolomics data revealed a total of nine significantly differentially expressed metabolic co-expression modules. Figure 8 (b) and (c). The salmon module showed a strong negative correlation with disease incidence (r = -0.77, p = 0.003) and chlorophyll content (r = 0.82, p < 0.001). Within this module, metabolite significance (GS) was positively correlated with module members (MM; r = 0.82, p = 1.2 × 10⁻³⁹), indicating that pivotal metabolites are key contributors to disease suppression. Further pivotal metabolites were detected in the salmon module. Based on network construction using 12 algorithms in the Cytohubba plugin, central metabolites such as 3-phosphoglycerate, 17-phenyltriene prostaglandin, and F2a serine amide were identified as potential regulatory pivots. Integrating WGCNA and DAMs, 30 overlapping metabolites were found in the salmon module. Correlation analysis of pivotal metabolites of WHL7-related bacterial communities with beneficial ASVs showed strong correlations, with 3-phosphoglycerate showing a significant positive correlation with multiple beneficial bacterial communities, and this correlation was much higher than that of other metabolites. This suggests that G3P may be a key metabolite driving the enrichment of beneficial microorganisms.

[0059] This study used G3P to verify whether the enrichment of beneficial rhizosphere microorganisms is mediated by host-derived metabolites. The experimental design involved inoculating banana seedlings treated with different concentration gradients of G3P with Foc TR4 (Foc TR4 + G3P), with control groups consisting of banana seedlings inoculated only with Foc TR4 (Foc TR4) and untreated seedlings (CK). Figure 10a). All Foc TR4 pathogens used in the experiment were labeled with green fluorescent protein (GFP) to facilitate subsequent observation of hyphal infection using confocal microscopy. Correlation analysis showed a significant negative correlation between G3P accumulation and disease incidence (R = -0.99, p = 0.01), indicating that G3P levels are closely related to resistance to Foc TR4. Figure 10 b). After 14 days of infection, microscopic observation of GFP-labeled Foc TR4 revealed that numerous hyphae colonized the corms and root tissues of untreated plants, while G3P treatment significantly reduced pathogen invasion in a concentration-dependent manner. Figure 10 c). To further investigate the effect of G3P on Foc TR4 colonization in soil, real-time quantitative PCR (qPCR) was used to detect Foc TR in soils of different treatment groups. A standard curve was constructed using different concentration gradients of Foc TR4 spores (R = 0.9906). Figure 11 qPCR quantification results consistently confirmed that the abundance of Foc TR4 was significantly reduced under G3P treatment, with the lowest abundance at a G3P concentration of 3.0 nM. Figure 10 d). However, in vitro experiments showed that G3P had no direct inhibitory effect on the growth of Foc TR4 ( Figure 12 This indicates that G3P inhibits FocTR4 colonization in the soil and infection of banana plants by regulating the rhizosphere microbial community.

[0060] Microbial community analysis further revealed that G3P treatment altered the composition of the rhizosphere microbiota. Dilution curves showed that all samples had sufficient sequencing depth ( Figure 10 e), Phylum-level taxonomic analysis showed significant changes in bacterial community structure among different treatments ( Figure 10 f). α-diversity analysis showed that G3P application restored the microbial richness and diversity (Chao1 and Shannon indices) suppressed by Foc TR4 infection. Figure 10 gh). β-diversity analysis using robust principal component analysis (RPCA) highlighted significant aggregation of the microbial community, with clear separation between the G3P-treated group and the group with only pathogens added. LEfSe analysis identified specific bacterial communities enriched in the G3P-treated samples, including *Rhodotorula* spp. (gh). Rhodomicrobia ), Pseudomonas spp. Pseudomonas ) and Mesophytic rhizobia ( Mesorhizobium These microbial communities have been reported as beneficial bacteria or bacteria that promote plant growth. Figure 10 j). Overlap analysis of the microbial communities enriched in WHL7 and G3P treatments revealed that 75 beneficial microbial communities (52.45%) were shared, indicating that WHL7 and G3P regulate similar rhizosphere microbial communities. Figure 10 These results indicate that exogenous application of G3P significantly inhibited Foc TR4 infection in banana roots, which not only reduced pathogen colonization but also remodeled the rhizosphere microbial community. The enrichment of beneficial bacteria strongly supports the hypothesis that G3P, as a key host metabolite, mediates WHL7-induced natural soil microbiome remodeling and disease suppression.

[0061] To elucidate the mechanism by which G3P mediates the enrichment of beneficial rhizosphere microorganisms, growth-promoting and chemotactic assays were conducted using Bacillus and Pseudomonas strains as representatives. Growth dynamics results showed that, compared to the control (CK), the addition of G3P to LB medium significantly promoted bacterial proliferation. Figure 9 a). After 24 h of culture, the optical density (OD600) at 600 nm significantly increased in Bacillus 5-36, Bacillus 5-88, Bacillus 5-136, and Pseudomonas 6-34 under the action of G3P, indicating that G3P has a positive regulatory effect on microbial growth. This stimulating effect was consistent in all tested strains, with Pseudomonas 6-34 showing the strongest growth response. Figure 9 a).

[0062] To further verify whether G3P can directly promote microbial proliferation by inducing chemotaxis, beneficial bacteria were pretreated with G3P, rinsed with sterile water, and then cultured for 24 hours. Figure 9 (b) The results showed that bacteria treated with G3P formed denser colonies than those treated with water, confirming that G3P not only stimulated bacterial chemotaxis but also significantly accelerated colony expansion. All tested strains showed similar trends, with Bacillus 5-88 and Pseudomonas 6-34 exhibiting particularly strong colony formation under G3P treatment. In summary, these results demonstrate the role of G3P as a microbial growth-promoting metabolite, directly enhancing the proliferation of beneficial bacteria and increasing their colonization potential. The dual role of G3P in stimulating bacterial growth and promoting chemotactic recruitment provides mechanistic evidence for its involvement in reshaping the rhizosphere microbiome and promoting the enrichment of beneficial microorganisms in host-pathogen interactions. This mechanistic insight supports the hypothesis that host-derived G3P plays a central role in establishing a protective rhizosphere environment against soil-borne pathogens.

[0063] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. The application of Streptomyces malayi WHL7 in the control of banana wilt disease, characterized in that, The classification name of the Streptomyces malayi WHL7 is... Streptomyces malaysiensis The depositary institution is Guangdong Provincial Center for Microbial Culture Collection, the deposit date is June 18, 2025, and the deposit number is GDMCC No: 66544. The pathogen of banana wilt disease is tropical race 4 of Fusarium wilt.

2. The application according to claim 1, characterized in that, The application method involves using crude extracts of Streptomyces malayi WHL7 fermentation broth to prepare bacterial agents or drug compositions for the prevention and control of diseases.

3. The application according to claim 2, characterized in that, The method for preparing the crude extract of the fermentation broth of Streptomyces maleicus WHL7 is as follows: S1: Activation of Streptomyces malayi WHL7 strain; S2: Inoculate the activated Streptomyces maleicus WHL7 into the soybean flour fermentation medium, culture, add anhydrous ethanol, continue culture, and filter out the residual bacterial cells to obtain the fermentation broth. S3: Concentrate and collect the fermentation broth to obtain the crude extract of the fermentation broth.

4. The application according to claim 3, characterized in that, The soybean flour fermentation culture medium formula mentioned in step S2 is as follows: 15-25g soluble starch, 10-20g soybean flour, 3-8g yeast powder, 1-3g peptone, 3-6g CaCO3, 3-7g NaCl, to a final volume of 1L, and the pH value is adjusted to 6.8-7.

2.

5. The application according to claim 3, characterized in that, The culture conditions described in step S2 are 100-200 rpm and 25-30℃ for 5-10 days.

6. The application according to claim 3, characterized in that, The culturing time described in step S2 is 0.5-1.5 days.

7. The application according to claim 3, characterized in that, The amount of anhydrous ethanol added in step S2 is the same as the volume of the culture medium.

Images