Production‐optimized fermentation of antifungal compounds by bacillus velezensis LZN01 and transcriptome analysis

Abstract

Fusarium wilt is one of the major constraints on global watermelon production, and Fusarium oxysporum f. sp. niveum (Fon) is the causative agent of Fusarium wilt in watermelon and results in severe yield and quality losses worldwide. The enhancement of antifungal activity from antagonistic bacteria against Fon is highly practical for managing Fusarium wilt in watermelon. The aim of this study was to maximize the antifungal activity of Bacillus velezensis LZN01 by optimizing fermentation conditions and analysing its regulatory mechanism via transcriptome sequencing. The culture and fermentation conditions for strain LZN01 were optimized by single‐factor and response surface experiments. The optimum culture conditions for this strain were as follows: the addition of D‐fructose at 35 g/L and NH₄Cl at 5 g/L in LB medium, pH 7, and incubation at 30°C for 72 h. The fungal inhibition rate for strain LZN01 reached 71.1%. The improvement of inhibition rate for strain LZN01 in optimization fermentation was supported by transcriptomic analysis; a total of 491 genes were upregulated, while 736 genes were downregulated. Transcriptome analysis revealed that some differentially expressed genes involved in carbon and nitrogen metabolism, oxidation–reduction, fatty acid and secondary metabolism; This optimization process could potentially lead to significant alterations in the production levels and types of antimicrobial compounds by the strain. Metabolomics and UPLC/Q‐Exactive Orbitrap MS analysis revealed that the production yields of antimicrobial compounds, such as surfactin, fengycin, shikimic acid, and myriocin, increased or were detected in the cell‐free supernatant (CFS) after the fermentation optimization process. Our results indicate that fermentation optimization enhances the antifungal activity of the LZN01 strain by influencing the expression of genes responsible for the synthesis of antimicrobial compounds.

The yields of antimicrobial compounds, such as surfactin, fengycin, shikimic acid, and myriocin, increased or were observed in the optimized fermentation CFS. In conclusion, fermentation conditions improved the antifungal activity of the LZN01 strain and influenced the expression of genes involved in the synthesis of antimicrobial compounds.

INTRODUCTION

Watermelon (Citrullus lanatus L.) is a popular fruit crop that is cultivated worldwide due to its high nutritional and economic value (Al Mutar, Alzawar, et al., 2023). Fusarium wilt, caused by Fusarium oxysporum f. sp. niveum (Fon), is one of the most destructive soil‐borne diseases in watermelon cultivation (Lv et al., 2018; Xu, Wang, et al., 2019). The long‐term survival of Fon in soil and ineffective control measures have led to significant economic losses and food safety threats (Keinath et al., 2019). Therefore, research is urgently needed to develop effective, sustainable, and eco‐safe green strategies to manage and control Fusarium wilt in watermelon (Al Mutar, Alzawar, et al., 2023).

Biological control methods are effective, environmentally friendly and sustainable for controlling Fusarium wilt in watermelon (Fenta et al., 2023). Antagonistic microorganisms, such as Bacillus amyloliquefaciens, Bacillus velezensis, Penicillium oxalicum and Streptomyces goshikiensis, have been found to be effective in controlling Fusarium wilt in watermelon (Al Mutar, Noman, et al., 2023; De Cal et al., 2009; Faheem et al., 2015; Wang et al., 2022). The cell‐free supernatant (CFS) of B. velezensis LZN01 exhibits antifungal activity by deforming the conidial structures and damaging the membranes of Fon (Xu, Zhang, et al., 2019).

Biocontrol agents can suppress plant diseases by producing antimicrobial compounds and different enzymes to inhibit the growth of plant pathogens, competing for niches/nutrients, inducing systemic resistance in host plants, changing the structure of the microbial community in the rhizosphere and secreting specific compounds or enzymes to interfere with quorum sensing (QS) signals produced by pathogens (Dimkić et al., 2022; Xu et al., 2020; Zhang et al., 2023). Bacillus spp. has been reported to secrete a wide range of antimicrobial compounds, including nonribosomally encoded lipopeptides (LPs), such as fengycin, surfactin, iturin and polyketides (e.g. difficidin, macrolactin, butirosin and bacillaene) (Cao et al., 2021; Fazle Rabbee & Baek, 2020). Antifungal compounds from Bacillus spp. are considered major contributors to controlling Fon. Iturin, surfactin, bacillomycin, syringfactin, and pumilacidin from B. amyloliquefaciens DHA6 exhibit significant antifungal activity against Fon by inducing oxidative stress, disrupting structural integrity, and inhibiting mycelial growth and spore germination (Al Mutar, Noman, et al., 2023). Iturin A, fengycin, surfactin and bacitracin from B. velezensis WB are responsible for inhibiting the growth of Fon (Wang et al., 2022). The ability of biocontrol agents to inhibit plant pathogens is associated with their capacity to produce a variety of antimicrobial metabolites (Chowdhury et al., 2015); the more antimicrobial metabolites that the strain produces, the stronger the biocontrol efficiency (Nguyen et al., 2017).

The large‐scale production of antimicrobial metabolites from microorganisms is a crucial step for their commercial use. The production of antimicrobial metabolites secreted by biocontrol agents can be significantly impacted by their culture conditions (Kubiak et al., 2019). Cell growth and metabolite concentrations are strongly affected by carbon and nitrogen sources, inorganic salts, aeration, agitation, temperature and pH (Pan et al., 2019; Romero Rodríguez et al., 2018; Ruiz Villafán et al., 2021; Singh et al., 2017). Among them, the carbon source, nitrogen source and initial pH in the fermentation medium significantly affected the target substances (Pan et al., 2019; Wang et al., 2017; Wang et al., 2024). Therefore, fermentation optimization is important for the production of antimicrobial active substances and for reducing commercial costs (Wang et al., 2018). Genomics and transcriptomics provide powerful tools for uncovering the vast untapped biosynthetic potential of numerous microorganisms, potentially leading to the discovery of novel biosynthetic pathways that produce antimicrobial compounds (Bremer et al., 2023; Lu et al., 2022; Nanjani et al., 2022). Metabolomics, as a useful tool, enables researchers to predict the presence, absence, or alterations of metabolites in biosynthesis systems, facilitating the understanding of metabolic process (Yang et al., 2023; Zeng & Chen, 2023; Zhang et al., 2011).

Previously, we demonstrated that the Bacillus velezensis strain LZN01 showed antagonistic activity against Fon and produced myriocin, which is involved in inducing membrane disruption in Fon (Wang, Wang, Liu, et al., 2021; Xu, Wang, et al., 2019). The objective of this work was to optimize the fermentation conditions for B. velezensis LZN01 culture by combining single factor and response surface methodology (RSM) tests. Therefore, the aim of the present study was to determine the optimal fermentation conditions for the LZN01 strain for improving its antagonistic activity against Fon. Additionally, the gene expression (mRNA levels) of strain LZN01 during fermentation optimization was analysed. Moreover, the antimicrobial metabolites were analysed by metabolomics and UPLC/Q‐Exactive Orbitrap MS to explore the mechanism of the improvement in antagonistic activity. This study provides a good foundation for exploring the commercial potential of B. velezensis LZN01.

EXPERIMENTAL PROCEDURES

Experimental strain

The B. velezensis LZN01 and Fon strains used in this study were provided by the Laboratory of Microbial Ecology, Qiqihar University in China. Fon mycelia were incubated in potato dextrose agar (PDA) in the dark at 28°C for 7 days. The LZN01 strain was cultured in Luria–Bertani (LB) medium according to the methods described by (Xu, Wang, et al., 2019).

The antifungal activity assay of CFS from the LZN01 strain

B. velezensis LZN01 was cultured at 30°C with shaking at 150 rpm for 72 h in LB medium. The culture was centrifuged at 9820 g for 10 min. The supernatants were filtered through a syringe filter (0.22 μm) and collected, which was used as the CFS at 72 h (Xu, et al., 2019).

The CFS (5 mL) at 72 h and autoclaved PDA medium (5 mL) were mixed and added to a petri dish (60 mm diameter). The supernatant of the LB medium was used as a control. After solidification of the medium, Petri dishes were inoculated with 4 mm diameter fungal mycelial sections cut from a 1‐week‐old PDA‐grown Fon plate. The inoculated plates were sealed with sealing film and cultured in a growth chamber at 30°C. The colony diameters were measured in two directions on each plate after incubation for 2 days, and the inhibition rate was calculated according to the following formula (Xu, Wang, et al., 2019):

$$\mathrm{R}\left(\%\right)=\left(\mathrm{C}-\mathrm{T}\right)/\mathrm{C}\times 100$$

where C is the colony diameter of the control plate and T is the colony diameter of the treatment plate after culturing for 2 days. Each treatment involved three replicates.

Optimization of the conditions for improving the antifungal activity of B. Velezensis LZN01 fermentation broth

The Luria‐Bertani (LB) is a basic medium for B. velezensis LZN01 cultivation at 30°C for 72 h. Under the same culture conditions, 5 g/L D‐fructose, D‐glucose, D‐mannose, or L‐arabinose was added to the basic fermentation culture (LB medium). Then, 5 g/L NH₄Cl, beef extract, urea, and soya peptone were added to the basic fermentation culture. The optimal carbon and nitrogen sources were ascertained through triplicate evaluations following the methods described in Section ‘The antifungal activity assay of CFS from the LZN01 strain’. Then, the concentrations of carbon (15, 25, 35, 45 and 55 g/L) and nitrogen (3, 5, 7, 9, and 11 g/L) sources were set. The optimal concentrations of carbon and nitrogen sources were ascertained through triplicate evaluations according to the methods described above. The initial pH of the culture was set at 3, 5, 7, 9 and 11, the temperature was set at 30°C, and the time was set at 72 h. The determination of the optimum pH was conducted three times, adhering to the aforementioned methods.

On the basis of the single‐factor experiment, D‐fructose, NH₄Cl, and the fermentation pH were selected as the investigation factors, and RSM I‐based optimal design was used to determine the optimum conditions for improving the antifungal activity of strain LZN01. Statistical significance was determined at the 95% confidence level (p ≤ 0.05), and the interaction effects of each variable and the predicted value were determined via the following equation:

$$\begin{array}{c}Y={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+{\beta }_{12}{X}_1{X}_2+{\beta }_{13}{X}_1{X}_3\hfill \\ +{\beta }_{23}{X}_2{X}_3+{\beta }_{11}{X_1}^2+{\beta }_{22}{X_2}^2+{\beta }_{33}{X_3}^2\hfill \end{array}$$

Y is the predicted response; β ₀ is a constant; β ₁, β ₂, and β ₃ are linear coefficients; β ₁₁, β ₂₂, and β ₃₃ are squared coefficients; β ₁₂, β ₁₃, and β ₂₃ are interaction coefficients; and X ₁, X ₂ and X ₃ are the coded levels of the independent variables. Regression analysis and response surface plots were obtained using Design‐Expert version 13.0.

RNA extraction and RNA‐seq

B. velezensis LZN01 was cultured at 30°C with shaking at 150 rpm for 12 h in LB medium. Next, 1 mL cultures were inoculated into the basic fermentation culture medium and optimized fermentation culture medium. Then, the cultures were incubated at 30°C and shaken at 150 rpm for 9 h. The culture was centrifuged at 9820 g for 10 min at 4°C. Then, the precipitated bacterial cells were immediately frozen in liquid nitrogen for RNA extraction and RNA sequencing (RNA‐Seq). Three replicates were prepared for the treatment (optimized fermentation culture medium) and the control (basic fermentation culture) conditions.

Total RNA was extracted from the samples using TRIzol^® Reagent according to the manufacturer's instructions (Invitrogen, USA), and genomic DNA was removed using DNase I (TaKaRa, Kofu, Japan). The purity and concentration of the RNA were measured using a 2100 Bioanalyzer (Agilent, Palo Alto, CA, USA) and a NanoDrop 2000 (ND‐2000, Thermo, USA), respectively. High‐quality RNA samples (OD_260/280 = 1.8–2.2, >2 μg, ≥100 ng/μL) were used to construct the sequencing libraries. The RNA‐Seq library was prepared using the TruSeq™ RNA Sample Preparation Kit from Illumina (San Diego, CA, USA). First, the mRNA was enriched and then fragmented with fragmentation buffer. Second, first‐strand cDNA was synthesized using random hexamer primers, and double‐stranded cDNA was synthesized using a SuperScript Double‐Strand cDNA Synthesis Kit (Invitrogen, CA, USA). Phusion DNA polymerase (NEB) was used to perform 15 cycles of PCR amplification. The quantification of the PCR amplification products was carried out using TBS380 (Pico Green). An Illumina HiSeq X Ten/NovaSeq 6000 sequencer (2 × 150 bp read length) was used to sequence the paired‐end RNA sequencing library.

The raw paired‐end reads were trimmed and controlled for quality using SeqPrep (https://github.com/jstjohn/SeqPrep) and Sickle (https://github.com/najoshi/sickle) with the default parameters. The clean reads were separately aligned to the reference genome in orientation mode using HISAT2 (http://ccb.jhu.edu/software/hisat2/index.shtml) software (Kim et al., 2015). The expression levels of differentially expressed genes (DEGs) between two different samples were calculated according to the transcripts per million reads (TPM) method (Wang et al., 2022).

RSEM (http://deweylab.github.io/RSEM/) was used to quantify gene abundance. Differential expression analysis was performed using DESeq2 (http://bioconductor.org/packages/release/bioc/html/DESeq2.html)/EdgeR (http://www.bioconductor.org/packages/2.12/bioc/html/edgeR.html) with p ≤ 0.05, and the significant DEGs were screened according to the criteria for a fold change ≥2 or ≤0.5 and a false discovery rate (FDR) < 0.001. Functional enrichment analyses of DEGs, including gene ontology (GO) functional enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses, were performed using Goatools (https://github.com/tanghaibao/GOatools) and KOBAS 2.0 (http://kobas.cbi.pku.edu.cn) software (Xie et al., 2011).

qRT–PCR

To further identify the expression of the function genes involved in the synthesis of antifungal metabolites (Lu et al., 2022; Tan et al., 2022; Wu, Zhi, & Xu, 2019), six DEGs from the RNA‐Seq analysis were selected, and qRT–PCR was used to verify the changes in the expression of these six DEGs. The RNA samples used for RNA‐Seq were subjected to qRT–PCR. The qRT–PCR primers used are listed in Table 1, and the 16S rRNA gene was used as an internal control. Each qRT–PCR system (20 μL) included 10 μL of 2X ChamQ SYBR Colour qPCR Master Mix, 0.8 μL of forward primer (5 μmol/L), 0.8 μL of reverse primer (5 μmol/L), 0.4 μL of 50 X ROX Reference Dye 1, 2 μL of template (cDNA), and 6.0 μL of ddH₂O. PCR was carried out as follows: 5 min at 95°C, 40 cycles of 5 s at 95°C and 30 s at 55°C, and then 40 s at 72°C. Melting curves were generated to detect amplification specificity. Relative quantitative analysis was performed according to the 2−^ΔΔCT method. Each measurement was carried out three times (Wang, Wang, Liu, et al., 2021).

TABLE 1.

Primer sequences for the 16S rRNA and the strain LZN01 genes.

Gene name	Forward primer (5′ to 3′)—reverse primer (5' to 3′)	Gene description
16S	ACTCCTACGGGAGGCAGCAG‐ GGACTACHVGGGTWTCTAAT
serA	TGATGCCCTGTTAGTCCG‐ TGCCGTTGAAATGGTGTT	Phosphoglycerate dehydrogenase
serC	ATGCTGAGCCTCGTCCTT‐ TGACGGCGTTGTAGATAGA	3‐phosphoserine/phosphohydroxythreonine transaminase
fbaA	CTGTCGCCATTCACTTAG‐ CTTTAGGGTCAGCATAGATT	Fructose bisphosphate aldolase
gap	CAAATACGATGCGGCTAA‐ CAGGAAGAACAAGGGAAA	Glyceraldehyde‐3‐phosphate dehydrogenase type I
accA	CGTATTCCGTTATTTCTCCT‐ ACTTTCCCGATTGCTTTAT	Acetyl‐CoA carboxylase, alpha subunit
accD	CAGCATGGGTTCGGTTGT‐ CTGTCGTCGGATGGGTCA	Acetyl‐CoA carboxylase carboxyl transferase, beta subunit

Differentially expressed metabolites identified by untargeted metabolomics

To identify differentially abundant metabolites produced by strain LZN01 in the basic fermentation culture and optimized fermentation culture medium, B. velezensis LZN01 was cultured at 30°C with shaking at 150 rpm for 72 h in LB medium or optimized fermentation culture medium; B. velezensis LZN01 cultures were centrifuged at 9820 g for 10 min at 4°C. Then, the supernatants were filtered through 0.22 μm filters and collected. The supernatant of LB medium cultured for 72 h (without strain LZN01) was used as a control. Each treatment was repeated six times.

Metabolites extracted from B. velezensis LZN01 cells and their growth media were analysed on a UHPLC‐Q‐Exactive HF‐X system (Thermo Fisher Scientific Inc., USA). The samples were injected into an ACQUITY UPLC HSS T3 (100 mm × 2.1 mm i.d. 1.8 μm; Waters, Milford, USA) equilibrated with water containing 0.1% formic acid and eluted in a gradient with acetonitrile containing 0.1% formic acid at a flow rate of 0.4 mL/min for 16 min. The eluted metabolites were analysed using a Thermo UHPLC‐Q‐Exactive Mass Spectrometer equipped with an electrospray ionization (ESI) source operating in either positive or negative ion mode. The capillary voltages, sampling cones and collision voltages were set at 3.5 kV, 50 V and 20–40–60 eV, respectively. The capillary and source temperatures were set at 325°C, and the desolvation gas flow rate was 50 L/h. Centroid data were collected in the m/z 70–1050 range with a resolution of 60,000. The quality control (QC) sample was prepared by mixing aliquots of all samples as a pooled sample, which was analysed and tested via the same method as the analytic samples. The QCs were inserted at regular intervals (every 5–15 samples) throughout the analytical run to provide a set of data from which reproducibility can be evaluated. The positive and negative data were combined to acquire a combined dataset that was imported into Progenesis QI software (Waters Corporation, Milford, USA). Then, baseline filtration, peak recognition, retention time correction, and peak alignment were performed. Finally, a data matrix containing the retention time, mass–charge ratio and peak intensity was obtained. The characteristic peak was identified using Progenesis QI software, the mass spectrum information was matched with the metabolic database, and the metabolites were identified according to the matching score.

Assessment and detection of antifungal compounds from strain LZN01

The antifungal activity of the differentially abundant metabolites against Fon was estimated by the disk diffusion method (Kacem et al., 2016). Stock solutions of the differentially abundant metabolites were prepared by dissolving them in methanol. The working solutions were generated by serial dilution of the respective stock solutions with methanol. The 100 μL working solution was diffused on PDA medium (10 mL), and Fon colonies with a diameter of 4 mm were inoculated in the center of a Petri dish. The petri dishes with PDA media were incubated at 30°C for 48 h. Petri dishes supplemented with 100 μL of methanol were used as controls. Tests were performed in triplicate. The inhibition rate was calculated using the aforementioned methods. The LZN01 CFS at 72 h was prepared as described above, and the supernatant of LB medium cultured for 72 h was used as a control. The antifungal compounds were identified by a UHPLC system (Thermo Scientific™ Ultimate 3000 UHPLC, USA) coupled with a high‐resolution mass spectrometer (Thermo Scientific™ Q‐Exactive™ Orbitrap MS, USA) according to the method described by Wang et al. (Wang et al., 2022). Chromatographic separation was performed on a Hypersil GOLD C₁₈ column (50 mm × 2.1 mm, 1.9 μm). The mobile phase included water containing 0.1% formic acid and was eluted in a gradient with methanol containing 0.1% formic acid. The flow rate was set to 300 μL/min. The analysis employed standards of surfactin, fengycin (Sigma–Aldrich, USA) and shikimic acid (Aladdin, China), which were fully dissolved in methanol and then diluted to 0.1 mg/mL. Surfactin, fengycin and shikimic acid were detected by plotting the peaks of multiple reaction monitoring transitions of standards. Triplicates were performed.

Statistical analysis

SPSS 25.0 software was used for the statistical analysis. The data are expressed as the mean with standard deviation from three biological replicates (mean ± SD). Tukey's test and t‐tests were used for the statistical analyses, and the significance level was set at p < 0.05.

RESULTS

Optimization of the fermentation conditions for B. Velezensis LZN01 fungistasis

The carbon source was optimized on the basis of LB culture. D‐fructose, D‐glucose, D‐mannose, and L‐arabinose (5 g/L) were added to the LB media. As shown in Figure 1A,D fructose as a carbon source had the highest inhibition rate. D‐fructose was the most favourable carbon source for the strain LZN01 to produce fungistatic substances. By adjusting the D‐fructose concentration, the optimal carbon source concentration was achieved. As shown in Figure. 1B, when the D‐fructose concentration was 35 g/L, the fermentation CFS had a fungistatic effect, and the inhibition rate reached 50.3 ± 1.2%.

FIGURE. 1.

Antagonistic Fon experiments of the cell‐free supernatant (CFS). (A) Inhibitory effect of different carbon sources (p < 0.05). (B) Effect of different concentrations of D‐fructose on the growth of Fon mycelia. (C) Inhibitory effect of different nitrogen sources on the growth of Fon mycelia (p < 0.05). (D) Inhibitory effect of different concentrations of NH₄Cl on the growth of Fon mycelia. (E) Inhibitory effects of different pH values on the growth of Fon mycelia (p < 0.05). The bar chart represents the Fon colony diameter, and the line chart represents the inhibition rate (%).

The nitrogen source was optimized for the basal culture (LB medium). As shown in Figure. 1C, the culture with NH₄Cl was the best nitrogen source, with a Fon colony diameter of 12.50 ± 0.41 mm. When the NH₄Cl concentration was 5 g/L, the fermentation CFS had the greatest inhibition rate (62.3 ± 0.7%, Figure. 1D).

An appropriate pH contributed to the production of antifungal substances by B. velezensis LZN01. Figure. 1E shows that the colony diameter of Fon varied at pH 3 to 11. At pH 7, the inhibition rate of the fermentation filtrates reached 32.8 ± 3.6%.

Analysis and optimization of the response surface

Based on the single‐factor experiments, the pH, D‐fructose concentration and NH₄Cl concentration were selected as the independent variables in the response surface optimization experiments. The results were matched via multivariate regression. The quadratic regression equation between the inhibition rate and D‐fructose concentration (A), NH₄Cl concentration (B), and pH (C) was as follows:

$$Y=70.20+1.90\mathrm{A}₋.20\mathrm{B}+2.15\mathrm{C}+1.23\mathrm{AB}₋0.6250\mathrm{AC}+3.28\mathrm{BC}₋2.66{\mathrm{A}}^2₋1.66{\mathrm{B}}^2₋4.96{\mathrm{C}}^2$$

The ANOVA results for the regression equation model are shown in Table 2.

TABLE 2.

Regression analysis of the experimental results based on the Box–Behnken design.

Source	Sum of squares	Df	Mean square	F value	p value	Significance
Model	313.26	9	34.81	14.39	0.0010	Significant
A‐D‐fructose	28.88	1	28.88	11.94	0.0106
B‐NH4Cl	38.72	1	38.72	16.00	0.0052
C‐pH	36.98	1	36.98	15.29	0.0058
AB	6.00	1	6.00	2.48	0.1592
AC	1.56	1	1.56	0.6459	0.4480
BC	42.90	1	42.90	17.73	0.0040
A2	29.85	1	29.85	12.34	0.0098
B2	11.64	1	11.64	4.81	0.0644
C2	103.69	1	103.69	42.86	0.0003
Residual	16.93	7	2.42
Lack of Fit	13.86	3	4.62	6.00	0.0581	not significant
Pure Error	3.08	4	0.7700
Cor Total	330.20	16
R2	0.9487

The plot of the response surface analysis shows the influence of the variables and their interactions on the response value (Figure. 2). Table 2 suggests that the regression equation model was significant, indicating that the model was reliable. The correction coefficient of the model was 0.9487, showing that the fit of the model was good. The influence on the inhibition rate was in the following order: NH₄Cl > pH > D‐fructose.

FIGURE. 2.

The inhibitory effect of D‐fructose, NH₄Cl and pH on the antifungal activity of LZN01 against Fon, as shown by contour plots and 3D response surface representations (A–C).

The predicted conditions were obtained by Design‐Expert V13.0 software: 37.26 g/L D‐fructose, 3.88 g/L NH₄Cl, and pH 7.03. Under these conditions, the theoretical inhibition rate was 71.05%. A validation experiment was performed under the predicted conditions, and the results showed that the inhibition rate was 67 ± 0.93%. The conditions were as follows: 35 g/L D‐fructose, 5 g/L NH₄Cl, and pH 7. The inhibition rate was 71.1 ± 1.0%, which was close to the predicted theoretical value.

Functional annotation, pathway enrichment and expression analysis of DEGs

In the transcriptome analysis, 1227 DEGs were identified (|log2FC| ≥ 1, p‐adjusted <0.05). Among them, 491 and 736 DEGs were significantly upregulated and downregulated, respectively, under the optimized fermentation culture conditions for LZN01 (Figure. 3A). KEGG annotation revealed that the DEGs could be classified into 6 categories (Figure. 3B). Most of the DEGs were enriched in metabolism categories, and the number of enriched DEGs in carbohydrate metabolism was greater than that in other pathways with the same function (Figure. 3B). KEGG pathway enrichment analysis of the 155 upregulated DEGs was performed, as shown in Figure. 3C. The top 10 significantly enriched pathways (p < 0.05) were mainly associated with metabolism, including cysteine and methionine metabolism, methane metabolism, histidine metabolism, and valine, leucine and isoleucine biosynthesis. These results indicate that some genes related to metabolism are affected by the optimized fermentation conditions for LZN01.

FIGURE. 3.

Expression analysis of DEGs (p < 0.05). (A) Number of DEGs. (B) KEGG annotation analysis. The vertical axis represents the number of genes, and the horizontal axis represents the pathway name. (C) KEGG enrichment analysis of upregulated DEGs (pathway enrichment is shown based on p < 0.05). The vertical axis represents the pathway name, and the horizontal axis represents the ratio of the number of genes enriched in this pathway to the background number of annotated genes (the larger the enrichment factor, the greater the enrichment degree); the size of the dot indicates the number of genes in the pathway, and the colour of the dot corresponds to different p value ranges. (D) GO enrichment analysis of upregulated DEGs (p < 0.05). The ordinate represents the GO term, the left nether abscissa represents the significance level of enrichment, and the right upper abscissa represents the number of DEGs.

To better understand the functions of the upregulated DEGs, a GO functional classification analysis of the upregulated DEGs was performed (Figure. 3D). The DEGs were divided into three categories based on the GO enrichment analysis: biological process, molecular function and cellular component. The top 10 significantly enriched genes are shown in Figure. 3D and were associated with carbohydrate, amino acid, organic acid, carboxylic and small molecule metabolites, including organic acid metabolic process (GO:0006082), oxoacid metabolic process (GO:0043436), small molecule biosynthetic process (GO:0044283), small molecule metabolic process (GO:0044281), cellular amino acid metabolic process (GO:0006520), carboxylic acid biosynthetic process (GO:0046394), serine family amino acid metabolic process (GO:1902494), carboxylic acid metabolic process (GO:0009069), histidine biosynthetic process (GO:0019752), and organic acid biosynthetic process (GO:0000105). In addition, Figure. 3D shows that the enriched DEGs were related to oxidation–reduction reactions, such as oxidoreductase complex (GO:1990204), oxidoreductase activity (GO:0016491) and ferrochelatase activity (GO:0004325). These results indicate that the expression levels of some genes associated with carbon and nitrogen metabolism and oxidation–reduction reactions were influenced by the optimized fermentation conditions.

The expression levels of genes encoding enzymes related to the Embden‐Meyerhof pathway (EMP), which is involved in carbon and nitrogen metabolism, were significantly upregulated (Figure. 4A), including ptsG (glucokinase), pfkA (fructose phosphate isomerase), fbaA (aldolase), gap (glyceraldehyde 3‐phosphate dehydrogenase), pgk (phosphoglycerate kinase), gpmI (phosphoglycerate mutase) and eno (phosphopyruvate hydratase). Figure. 4A shows that the expression levels of csn (chitosanase), ccpA and tnrA (carbon and nitrogen metabolism) increased in the optimized fermentation culture. However, ccpC and codY are also related to carbon and nitrogen metabolism, and their expression levels were decreased in the optimized fermentation culture.

FIGURE. 4.

Expression analysis of DEGs. (A) Expression analysis of 12 DEGs. The abscissa represents log2 (fold change), and the ordinate represents the gene name. (B) KEGG enrichment analysis of downregulated DEGs (pathway enrichment is shown based on p < 0.05). The vertical axis represents the pathway name, and the horizontal axis represents the ratio of the number of genes enriched in this pathway to the background number of annotated genes (the larger the enrichment factor, the greater the enrichment degree); the size of the dot indicates the number of genes in the pathway, and the colour of the dot corresponds to different P value ranges. (C) DEGs involved in the secondary metabolic synthesis pathway. log2 (fold change) represents the multiple of differential expression.

To better understand the functions of the downregulated DEGs, the top 10 significantly enriched KEGG pathways (p < 0.05) of the downregulated DEGs were identified (Figure. 4B). We found that DEGs related to the biosynthesis of secondary metabolites were also significantly enriched in the optimized culture group (Figure. 4C). Figure. 4C shows that genes related to the biosynthesis of surfactin, iturin A and fengycin, including K4120_09460 (ituB), K4120_09455 (ituC), K4120_09465 (ituA), K4120_01790 (srfAB), K4120_01795 (srfAC), K4120_01785 (srfAA), K4120_09585 (fenB), K4120_09590 (fenA), K4120_09600 (fenD), and K4120_09605 (fenC), were significantly downregulated in the optimized fermentation group. However, some genes related to the biosynthesis of terpenoids and other secondary metabolite pathways were significantly upregulated in the optimized fermentation group at the mRNA level (Figure. 4C). These results suggest that some genes related to secondary metabolite pathways were altered in response to the optimized fermentation culture conditions.

We also found that the expression levels of genes involved in amino acid synthesis and fatty acid synthesis, including alsS, ilvBCDN, leuABCD, accABD and lipA, were significantly upregulated in the optimized fermentation group (Figure. 5A). The expression levels of genes related to NRPS pathway regulation, including K4120_01825 (phosphopantetheinyl transferase), degQ and degU (two‐component system), were also upregulated. In addition, the expression levels of genes encoding enzymes related to the biosynthesis of shikimic acid and sphingosine were significantly upregulated in the fermentation optimization group, including K4120_13830 (bifunctional 3‐deoxy‐7‐phosphoheptulonate synthase/chorismate mutase), K4120_01665 (shikimate kinase), K4120_04005 (shikimate dehydrogenase) and K4120_10745 (3‐phosphoshikimate 1‐carboxyvinyltransferase); serA (phosphoglycerate dehydrogenase), serC (3‐phosphoserine/phosphohydroxythreonine transaminase), K4120_07650 and K4120_01805 (pyridoxal phosphate‐dependent aminotransferase), K4120_15570, K4120_15845, K4120_13595 and K4120_17885 (SDR family oxidoreductase). However, the expression levels of most genes related to sporulation, including spoVS, spoIIAB, and spoIIAA, were downregulated (Figure. 5A). Moreover, upregulated expression of genes involved in quorum quenching was identified, particularly those annotated as “quorum‐quenching lactonase YtnP” (K4120_13895) and “Metallo‐beta‐lactamase superfamily” (K4120_05490, K4120_12505, K4120_11945 and K4120_08865) (Figure. 5A).

FIGURE. 5.

Expression analysis of DEGs and a schematic view of the metabolic pathways related to fengycin and surfactin. (A) Gene expression change heatmap, LB: Gene expression of LZN01 cultured in LB media, XYM: Gene expression of LZN01 cultured in the optimized fermentation media. (B) A schematic view and of the metabolic pathways for fengycin and surfactin. Red represents upregulation, and blue represents downregulation. (C) The expression of six typical genes determined by qRT–PCR. The horizontal axis represents the gene names in LZN01, and the left panel represents the mean values of log2 (fold change).

Carbon and nitrogen metabolism (EMP pathway and amino acid metabolism), fatty acid metabolism, sporulation and NRPS pathway regulation may play important roles in the synthesis of LPs (Lu et al., 2022; Tan et al., 2022; Wu, Zhi, & Xu, 2019). As shown in Figure. 5B, changes in multiple metabolic pathways affect the synthesis of fengycin and surfactin.

To confirm the RNA‐Seq data, the expression of six typical genes was analysed via qRT–PCR (Figure. 5C). The expression levels of serA, serC, fbaA, gap, accA and accD were significantly greater in the optimized fermentation culture than in the basal culture. These results are consistent with those of the transcriptome analysis.

Increased antifungal compound production in the optimized fermentation system

We detected and analysed the differentially abundant metabolites between the basal culture and optimized fermentation culture conditions using metabolomics and multivariate statistical analysis. In addition, differentially abundant metabolites with antimicrobial functions, such as shikimic acid, myriocin, neomycin, leucomycin a5, penicillin K, curvacin A and myristoleic acid, were found according to the characteristics of publicly available databases (Figure. 6).

FIGURE. 6.

Specialized metabolites produced by LZN01. The levels of several differentially abundant metabolites were significantly increased in the LZN01 optimized fermentation broth. ****p < 0.0001. CK indicates LB medium, LB indicates the fermentation supernatant of strain LZN01 cultured in basic LB medium, and XYM indicates the fermentation supernatant of strain LZN01 cultured in the optimized medium.

Biosynthesis‐related gene clusters (BGCs) encoding fengycin and surfactin were found in the LZN01 genome (Hu et al., 2023). To further verify the increase in antifungal compounds in the optimized fermentation culture, we tested the CFS by UPLC/Q‐Exactive Orbitrap MS, and fengycin and surfactin were detected (Figure. 7A,C). Ions with m/z values of 1461.7887 and 1036.6904 were detected in the CFS from LZN01 and assigned to fengycin [M‐H]⁻ and surfactin [M + H]⁺, respectively (Figure. 7A,C). We found that the peak areas of fengycin and surfactin were significantly greater in the optimized fermentation culture than in the basal fermentation culture (Figure. 7B,D). These results indicate that the LZN01 strain produced more fengycin and surfactin in the optimized fermentation medium.

FIGURE. 7.

Identification of fengycin and surfactin in the CFS from LZN01. (A) MS analysis of fengycin in the CFS from LZN01. (B) Fengycin peak area analysis by UPLC/Q‐Exactive Orbitrap MS. (C) MS analysis of surfactin in the CFS from LZN01. (D) Surfactin peak area analysis by UPLC/Q‐Exactive Orbitrap MS. CK, the supernatant of LB medium cultured for 72 h; LB, the cell‐free supernatant (CFS) from LZN01 cultured in LB medium for 72 h; XYM, the CFS from LZN01 cultured in the optimized medium for 72 h. ***p < 0.001.

Shikimic acid from the CFS of strain LZN01 was tested by UPLC/Q‐Exactive Orbitrap MS. Ions with m/z values of 173.0443 were detected in the CFS from strain LZN01 and assigned to shikimic acid [M‐H]^₋ (Figure. 8A). In addition, the peak area of shikimic acid in the optimized fermentation CFS was significantly greater than that in the basic fermentation CFS (Figure. 8B). Shikimic acid showed antibacterial activity by causing cell membrane dysfunction and bacterial damage (Bai et al., 2015). We evaluated the activity of shikimic acid against Fon by the disk diffusion method. The results showed that different concentrations of shikimic acid had antagonistic effects on Fon (Figure. 8C). These results suggested that one of the reasons for the increase in antifungal activity in the optimized fermentation culture was related to the increase in shikimic acid.

FIGURE. 8.

Identification of shikimic acid in the CFS from LZN01. (A) MS analysis of shikimic acid in the CFS from LZN01. (B) Shikimic acid peak area analysis by UPLC/Q‐Exactive Orbitrap MS. CK, the supernatant of LB medium cultured for 72 h; LB, the cell‐free supernatant (CFS) from LZN01 cultured in LB medium for 72 h; XYM, the CFS from LZN01 cultured in the optimized medium for 72 h. ****p < 0.0001. (C) Antagonistic experiments on the inhibition of Fon by different concentrations of shikimic acid solutions.

In our previous study, we confirmed that myriocin was one of the major functional components of the CFS from strain LZN01 and that myriocin exhibited antifungal activity by inducing membrane damage in Fon (Wang, Wang, Liu, et al., 2021; Xu, Wang, et al., 2019). Myriocin is a sphingosine analog (Craveri et al., 1972). The RAW intensities of sphingosine and myriocin in the optimized fermentation culture were greater than those in the basal fermentation culture (Figure. 6).

DISCUSSION

The antifungal activity of B. Velezensis LZN01 was improved by single factor and response surface optimization

Fusarium wilt in watermelon plants is a devastating disease. From the viewpoint of effective, sustainable and ecologically safe green strategies to control Fusarium wilt, the growth and reproduction of pathogenic fungi can be effectively controlled by improving the biological control capability of antagonistic microorganisms in soil (Al Mutar, Alzawar, et al., 2023; Yánez Mendizábal et al., 2023; Zalila Kolsi et al., 2022). Bacillus sp. are important biocontrol agents because they produce antimicrobial substances to inhibit the growth of plant pathogens (Bonaterra et al., 2022; Luo et al., 2022). The strain LZN01, isolated from the wheat rhizosphere in the watermelon/wheat companion system, exhibits antifungal activity through its CFS (Xu, Zhang, et al., 2019). The strain was named Bacillus velezensis LZN01 based on phylogenomic analysis (Hu et al., 2023).

The yield of antimicrobial active substances determines the actual effect of bacteriostasis (Duan et al., 2020; Pageni et al., 2014; Roy et al., 2006). The yield and type of antimicrobial active substances can be improved by adjusting the nitrogen and carbon sources in the culture and regulating the pH of the medium (Pan et al., 2019; Wang et al., 2024). In this study, the fermentation conditions of B. velezensis LZN01 were optimized by single factor and response surface tests. The optimized fermentation conditions were as follows: 35 g/L D‐fructose, 5 g/L NH₄Cl, and pH 7. The antifungal activity of the optimized fermentation culture was significantly greater (p < 0.05) than that of the basic fermentation culture. The maximum inhibition rate of the optimized fermentation culture for B. velezensis LZN01 was 71.1%, whereas that of the basic fermentation culture was 32.8%, which was an increase of 38.3%. Similarly, the activity of B. subtilis BS501a against Magnaporthe grisea DWBJ329 was increased by 2.4 times under the optimized fermentation conditions (Li & Xu, 2011). The optimization of incubation temperatures, stirring speeds, and other parameters need to further be explored and improved. In this study, the state of culture and pH before fermentation were considered, and the expression of genes at the mRNA level was monitored in the optimized fermentation culture.

B. Velezensis LZN01 responses to the optimized fermentation culture conditions

To explain the gene expression pattern and compare gene transcription under the basic fermentation culture and optimized fermentation culture conditions, we screened 1227 significantly DEGs related to carbon and nitrogen metabolism, oxidation–reduction and secondary metabolites (Figure. 3).

Enrichment of carbon and nitrogen metabolism pathways indicated that the optimized fermentation culture might promote B. velezensis LZN01 to accelerate the synthesis and transformation of carbon and nitrogen sources, efficiently synthesize functional molecules to meet the requirements for metabolism and growth and ensure more efficient life activities (Figure. S1F). In addition, carbon and nitrogen metabolism play a critical role in antibiotic synthesis (Cai et al., 2019). The ccpA and ccpC genes encode the main catabolite control protein A and transcriptional regulator ccpC (Bremer et al., 2023; Sonenshein, 2007). The tnrA and codY genes encode the main MerR family transcriptional regulator tnrA and GTP‐sensing pleiotropic transcriptional regulator codY in Bacillus (Mirouze et al., 2015; Zhu et al., 2017). Simultaneously deleting ccpC and overexpressing ccpA and tnrA in strain DW2 improves the bacitracin yield (Cai et al., 2019). The yield of lipopeptide antibiotic lichenysin is enhanced by 31% in the codY null strain of B. licheniformis, with additional precursor amino acids (Zhu et al., 2017). The results showed that the genes ccpA and tnrA were upregulated, while the genes codY and ccpC were downregulated in the optimized fermentation culture (Figure. 4A). The expression levels of several genes involved in EMP in the optimized fermentation culture were significantly upregulated (Figure. 4A). It has been reported that overexpressing EMP genes can slightly increase the production of surfactin (Wu, Zhi, & Xu, 2019). This finding implies that B. velezensis LZN01 produces more antimicrobial substances in the optimized fermentation culture.

Oxidation–reduction reactions, which involve secondary metabolite synthesis and intracellular energy and material metabolism, play important roles in life activities (Ma et al., 2021, Wang, Ma, et al., 2021). The results showed that some of the upregulated genes were enriched in GO terms, including oxidoreductase activity (GO:0016491), oxidoreductase complex (GO:1990204) and ferrochelatase activity (GO:0004325) (Figure. 3D), which indicated that the optimized fermentation culture might improve the oxidoreduction metabolism of B. velezensis LZN01 to satisfy self‐growth and synthesize antimicrobial substances. Oxidoreductase can cause the biosynthetic antibiotics in bacterial cells to be exported outside of the cells to become mature compounds (Song et al., 2017). The improvement of bacteria in regard to carbon and nitrogen metabolism and oxidation–reduction reactions is beneficial to bacterial growth and antibiotic synthesis (Guo et al., 2020; Song et al., 2017). Therefore, we infer that the optimized fermentation culture could enhance the metabolic capacity of B. velezensis LZN01 to accelerate material transformation and the biosynthesis of antimicrobial compounds, thereby improving its growth (Figure. S1F).

We found that the number of downregulated genes related to the biosynthesis of secondary metabolites was greater than that of upregulated genes in the optimized fermentation culture group, especially those involved in nonribosomal peptide synthetase (ituA‐C, srfAA‐AC and fenA‐C) and polyketide synthesis (Figure. 4C). However, genes related to the metabolism of terpenoids and other pathways were significantly enriched in the optimized fermentation group, which contained many upregulated genes. These genes were mainly involved in antifungal, antibacterial and plant growth‐promoting activities (Dimkić et al., 2022; Rabbee et al., 2019).

In addition, the upregulation of genes associated with quorum quenching (QQ) has been noted, particularly those annotated as “quorum‐quenching lactonase YtnP” and classified within the “Metallo‐beta‐lactamase superfamily.” This mechanism, whereby enzymes degrade N‐acyl homoserine lactones (AHLs), a key QS signal molecule, has been instrumental in controlling the virulence of various plant pathogens (Roca et al., 2024; Zhang et al., 2023). In this study, we have identified four genes in the mRNA of LZN01 that are associated with enzymes displaying AHLs degrading activity, specifically annotated as “Quorum‐quenching lactonase YtnP” (K4120_13895), and belonging to the “Metallo‐beta‐lactamase superfamily” (K4120_05490, K4120_12505, K4120_11945 and K4120_08865) (Figure 5A). The up‐regulated expression of these genes, encoding homologues to QQ putative proteins (Roca et al., 2024), may potentially enhance the antimicrobial capability of strain LZN01.

Analysis of antagonistic Fon compounds

Fengycin and surfactin are lipopeptide molecules with circular structures that contain amino acid residues and exhibit antifungal activity (Liao et al., 2016). The lipopeptide synthesis process starts with connecting amino acids in a circular polypeptide structure under the action of NRPS synthetase and then combining with fatty acids to form an amphiphilic molecule (Biniarz et al., 2017). Our UPLC/Q‐Exactive Orbitrap MS identification and analysis results confirmed that yields of fengycin and surfactin were significantly increased in the optimized fermentation culture group (Figure. 7). These results may be related to amino acid and fatty acid synthesis, the NRPS regulation pathway and sporulation (Figure. 5A,B).

The synthesis of fengycin requires multiple amino acids as structural precursors, including valine, isoleucine, tyrosine, alanine, glutamine, threonine, ornithine, and glutamic acid (Chavarria Quicaño et al., 2023). Amino acids are essential precursors for surfactin biosynthesis (Wang et al., 2019). In this study, a large number of genes related to amino acid biosynthesis were significantly upregulated in the optimized fermentation culture group (Figure. 3B,C), and the upregulated genes were enriched in the valine, leucine and isoleucine synthesis pathways (Figure. 3B). The synthesis and metabolism of amino acids play important roles in the synthesis of fengycin and surfactin (Gao et al., 2023; Liu et al., 2012). Therefore, increasing the amino acid supply might have positive effects on surfactin and fengycin production (Na et al., 2013). The expression levels of genes involved in amino acid synthesis, including alsS, ilvBCDN and leuABCD, were upregulated, and surfactin production was significantly improved (Wu, Zhi, & Xu, 2019). Fengycin and surfactin levels were elevated by upregulating the expression of relevant genes involved in the fatty acid pathway (Gao et al., 2022; Jing et al., 2022; Wu, Zhou, et al., 2019). The overexpression of accABCD, fabD, and lipA increases the production of surfactin and fengycin (Rasetto et al., 2019; Tan et al., 2022; Wu, Zhou, et al., 2019). The expression levels of the above genes related to amino acid synthesis and the fatty acid pathway were upregulated in the optimized fermentation culture group (Figure. 5A). Therefore, we speculate that the inhibition rate was enhanced in the optimized fermentation culture group, which may be related to the upregulated expression of the above genes.

The genes sfp, degU and degQ play crucial roles in the NRPS pathway (Lu et al., 2022; Tan et al., 2022). When the gene sfp (4 phosphopantetheinyl transferase) in the NRPS pathway is deleted, LPs cannot be synthesized (Nakano et al., 1992). The degQ gene is a pleiotropic factor that can regulate the production of fengycin (Tan et al., 2022), and the expression of degQ is regulated by degU (Lu et al., 2022). Knockout of genes related to spore formation also effectively improve fengycin and surfactin production (Gao et al., 2022; Wang, Wang, Liu, et al., 2021). The reason for these results may be that blocking the sporulation pathway can upregulate the expression of genes related to the glycolysis pathway and increase the utilization of metabolic substrates (Wang, Yu, et al., 2020). In this study, KEGG enrichment analysis revealed that the expression of most of the genes mentioned above was downregulated in the optimized fermentation culture group (Figure. 5A). These results imply that the downregulation of genes related to sporulation and the upregulation of genes related to the NRPS regulation pathway contribute to the synthesis of fengycin and surfactin.

The strain LZN01 exhibited no discernible presence of fengycin when cultured in the LB medium (Figure. 7A). Similarly, strain 168DS (sfp and degQ overexpression strain based on B. subtilis168) did not produce fengycin in LB culture (Tan et al., 2022). The reason for this result may be attributed to the production levels of fengycin not reaching the detection threshold (Tan et al., 2022). In other Bacillus species, the production of fengycin when cultured in LB medium was also observed to be very low, similar to the findings in strains LZN01 and 168DS (Fan et al., 2017; Yaseen et al., 2016).

Shikimic acid has been reported to possess antibacterial activity against Staphylococcus aureus (Bai et al., 2015). Some strains can produce shikimic acid via microbial fermentation, such as Escherichia coli, Corynebacterium glutamicum, Bacillus megaterium, B. subtilis, and Saccharomyces cerevisiae (Sheng et al., 2023). In this study, the shikimic acid yield in the optimized fermentation culture was greater than that in the basic fermentation culture, which may be related to the upregulated expression of genes involved in the shikimic acid biosynthesis pathway (Figure. 5A) (Sheng et al., 2023). We demonstrated that shikimic acid possesses antifungal activity against Fon (Figure. 8C), which may be another important reason for the increase in the inhibition rate.

In our previous study, we confirmed that strain LZN01 produces myriocin (Hu et al., 2023), which inhibits the growth of Fon by inducing membrane damage and targeting intracellular molecules (Wang, Wang, Liu, et al., 2021; Wang, Wang, Xu, & Wang, 2021). Here, the expression of putative genes involved in the myriocin biosynthesis pathway were upregulated according to the transcriptomic analysis, and the peak area of myriocin increased according to the metabolomic analysis (Figure. 5A and Figure. 6). We further found that compounds involved in antimicrobial activity, including neomycin (Lee et al., 2005), leucomycin a5 (Sakakibara et al., 1981), penicillin K (Eagle, 1947), myristoleic acid (Park et al., 2022), and curvacin A (Ahmadova et al., 2013) (Figure. 6), were present in the CFS from the optimized fermentation system.

CONCLUSION

In the present study, the optimized fermentation parameters were obtained, and the inhibition rate of CFS was found to be 71.1%. Transcriptome analysis revealed significant alterations in gene expression related to carbon and nitrogen metabolism, oxidation–reduction, amino acid metabolism, fatty acid metabolism, and secondary metabolism under the optimized fermentation conditions. The optimized fermentation system led to an increase or presence of antimicrobial compounds in the CFS, including surfactin, fengycin, shikimic acid, and myriocin. Our study provides valuable insights into optimized fermentation for strain LZN01, with a focus on improving its inhibition rate.

AUTHOR CONTRIBUTIONS

Jiale Hu: Data curation; formal analysis; writing – original draft. Zhigang Wang: Methodology; supervision. Weihui Xu: Conceptualization; funding acquisition; supervision; writing – original draft.

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no potential conflict of interest.

Supporting information

Figure S1.

Hu, J. , Wang, Z. & Xu, W. (2024) Production‐optimized fermentation of antifungal compounds by bacillus velezensis LZN01 and transcriptome analysis. Microbial Biotechnology, 17, e70026. Available from: 10.1111/1751-7915.70026

DATA AVAILABILITY STATEMENT

All data generated during this study are presented in this article or supplementary material. Further inquiries can be directed to the corresponding author.