Bacillus subtilis 1JN2 was isolated from ginger stem in our previous study and identified as an effective biocontrol agent (GeneBank accession number GU549436，China General Microbiology Culture Collection Center, CGMCC accession number 9759). The strain was grown on LB medium (10 g of tryptone, 5 g of yeast extract, and 5 g of NaCl per liter).

Fusarium oxysporum (Schl.) F.sp. cucumerinum Owen strain was provided by Kai Chen from Ecology Institute of Shandong Academy of Sciences. The strain was grown on PDA medium (filtrate of 200 g boiled potatoes and 20 g glucose per liter).

Cucumber seeds (Jindian 303, purchased from Academy of Agricultural Sciences of Huai’an, Jiangsu Province, China) were grown in a nursery site, transplanted to a new pot (10 cm height x 10 cm diameter of bottom), and filled with 0.5 kg of nutrient soil composed of vermiculite, peat, and perlite (purchased from Huai’an Chaimihe Agricultural Technology Co., Ltd) when it had 3–4 true leaves.

Two treatment conditions were tested in the first round greenhouse experiment: A: Blank control, B: B. subtilis 1JN2. Each group has three repetitions, and each repetition contains 20 cucumber seedlings. After transplantation of the cucumber plants, suspension of B. subtilis 1JN2 was adjusted to 10⁷ CFU/mL, and 20 mL of the suspension was irrigated around the base of the stem for each plant in the group B. After 1 week, 20 mL of spore suspension (10⁵ CFU/mL) of F. oxysporum was irrigated for each plant in the two groups. The temperature of the greenhouse was maintained at 30°C with 16/8 h of light/dark light cycle. The disease index was recorded 30 days after transplanting.

For colonization analysis, a chloramphenicol-resistant mutant 1JN2-chl was screened using methods from the study by Xue (2008) and prepared in LB medium with 20 μg/mL of chloramphenicol. As described above, a second-round greenhouse experiment was performed with two groups: A: cucumber, B: nutrient soil without plant. Each group has three repetitions, and each repetition contains 10 cucumber seedlings. After transplantation of the cucumber plants, suspension of B. subtilis 1JN2-chl was adjusted to 10⁸ CFU/mL, and 20 mL of the suspension was irrigated for each plant. In total, 1 g of rhizosphere soil from each group was collected on 1, 3, 7, 14, and 21 days after treatment. The colonization ability of the strain was measured by dilution spreading on solid medium with 20 μg/mL of chloramphenicol.

For scanning electron microscopy analysis, 20 cucumber seedlings with 4–5 true leaves were selected and divided into two groups. First, the roots of the plants were washed with sterile distilled water thrice and placed into a 250 mL flask containing 100 mL of 1JN2 suspension with a concentration of 10⁷ CFU/mL for 2 days. After incubation, the roots were washed with sterile distilled water thrice and cut into 0.5 cm pieces and then fixed in 2.5% glutaraldehyde for 2 h. The root pieces were washed thrice for 10 min in 25 mM sodium phosphate buffer, and then, they were serially dehydrated in 50, 70, 80, and 90% acetone solutions for 15 min and, finally, in 100% acetone solution for 90 min. During this process, the acetone was changed every 30 min. Cucumber root pieces were transferred into tertbutyl alcohol thrice to displace acetone for 30 min each time and then freeze-dried in a freeze dryer (Thermo Scientific, Waltham, MA, USA). Finally, the root samples were observed by scanning electron microscopy (SU8010 SEM, Hitachi, Japan).

Cucumber root exudates were collected as followed: cucumber seedlings with four leaves were isolated from the soil, the roots of the cucumber were gently washed with sterilized water three times, and then transferred into Hoagland’s solution for 3 days. Then roots were washed again, and then, each root was divided into two parts, and each part was immersed into a 50 mL centrifuge tube containing 40 mL of sterilized water. Overall, 5 mL of B. subtilis 1JN2 (1*10⁷ CFU/mL) was added to tube A and designated as the treated group, and 5 mL of sterilized water was added to tube B and designated as the induced group. Cucumber plants without root separation were set as the control group. Each treatment contained 10 cucumber plants, and 3 independent repeats were conducted in this procedure. After 3 days, the solutions from tubes A and B in each treatment were collected together and filtered with 30-μm filter paper, then, the filtrates were freeze-dried.

For component analysis, 50 mg powder of each sample was transferred to 5 mL of EP tube and extracted with 3,000 μL of 75% methanol with 1% acetic acid. After vortexing for 30 s, the samples were homogenized at 40 Hz for 4 min and sonicated for 30 min in an ice-water bath. All the samples were centrifuged at 12,000 rpm (RCF = 13,800 (×g), R = 8.6 cm) for 15 min at 4°C. In total, 2,500 μL of the supernatant was dried under gentle nitrogen flow. The residues of each sample were reconstituted in 250 μL of 50% methanol with 0.1% formic acid containing internal standard. The samples were vortexed for 30 s and ultra-sonicated for 15 min in an ice bath. Then, all the samples were centrifuged at 12,000 rpm (RCF = 13,800 (×g), R = 8.6 cm) for 15 min at 4°C. The resulting supernatants were filtered through the 0.22-μm filter membrane. Then, they were transferred to 2 mL glass vials and stored at −80°C until the UHPLC–MS/MS analysis. The quality control (QC) sample was prepared by mixing an equal aliquot of the supernatants from all of the samples.

Two rounds of field experiment were conducted from September to November 2022 and 2023 in the vegetable experimental base of Huai’an Agricultural Technology Extension Center (E119.020292，N33.484059) to determine the diversity of the rhizo-microorganisms in cucumbers treated with B. subtilis 1JN2 and the exogenous combinations of amino acid induced from cucumber roots. The same variety of cucumber (Jindian 303) was used in the experiments. In the first round, suspension of B. subtilis 1JN2 was adjusted to 10⁷ CFU/mL and irrigated to 20 mL for each plant after transplantation of the cucumber plants. Three sites, which were placed 5 m apart, were selected for irrigation and each site consist of three cucumber plants. Rhizosphere soil samples were collected 1 day before treatment and 1 day, 3 days, 7 days, and 14 days after treatment. All the soil samples at five time points were stored at −80°C before sequencing and further carried out microbiota analysis by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

A second round of experiment was conducted at the same site on 2023, the experimental materials and methods were described above, but the strain treatment was replaced with an exogenous amino acid solution treatment. According to the results of cucumber root exudate analysis, four amino acids that were induced by the biocontrol strain were selected. Amino acid solutions in water were prepared such that they contained each of the selected compounds in equal amounts (2.5 mM glutamine, 2.5 mM tryptophan, 2.5 mM glycine, and 2.5 mM glutamate) and at a final concentration of 10 mM. In total, 10 mL of the amino acid mixture was irrigated to each plant, and the soil samples were collected at the same time as described above.

Total microbial genomic DNA was extracted from soil samples using the E.Z.N.A.^® Soil DNA Kit (Omega Bio-tek, Norcross, GA, USA), according to the manufacturer’s instructions. The quality and concentration of DNA were determined by 1.0% agarose gel electrophoresis and a NanoDrop2000 spectrophotometer (Thermo Scientific, United States) and kept at −80°C prior to further use. The hypervariable region V3-V4 of the bacterial 16S rRNA gene was amplified with primer pairs 338F (5′-ACT CCT ACG GGA GGC AGC AG-3′) and 806R (5′-GGA CTA CHV GGG TWT CTA AT-3′) (Liu et al., 2016) by a T100 Thermal Cycler PCR (BIO-RAD, USA). The PCR reaction mixture consists of 4 μL of 5 × Fast Pfu buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of each primer (5 μM), 0.4 μL of Fast Pfu polymerase, 10 ng of template DNA, and ddH2O to a final volume of 20 μL. PCR amplification cycling conditions were as follows: initial denaturation at 95°C for 3 min, followed by 27 cycles of denaturing at 95°C for 30 s, annealing at 55°C for 30 s, extension at 72°Cfor 45 s, single extension at 72°C for 10 min, and a final extension at 4°C. The PCR product was extracted from 2% agarose gel and purified using the PCR Clean-Up Kit (YuHua, Shanghai, China), according to the manufacturer’s instructions and quantified using Qubit 4.0 (Thermo Fisher Scientific, USA).

Purified amplicons were pooled in equimolar amounts and paired-end sequenced on an Illumina PE300 platform (Illumina, San Diego, USA), according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The raw sequencing reads were deposited into the NCBI Sequence Read Archive (SRA) database (Accession Number: PRJNA1019504).

Raw FASTQ files were de-multiplexed using an in-house perl script, quality-filtered by fastp version 0.19.6 (Chen et al., 2018), and then merged by FLASH version 1.2.7 (Magoc and Salzberg, 2011) with the following criteria: (i) the reads were truncated at any site receiving an average quality score of <20 over a 50-bp sliding window, and the truncated reads shorter than 50 bp were discarded; reads containing ambiguous characteristics were also discarded; (ii) only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. The maximum mismatch ratio of overlap region is 0.2, and reads that could not be assembled were discarded; and (iii) samples were distinguished according to the barcode and primers, and the sequence direction was adjusted. Then, the optimized sequences were clustered into operational taxonomic units (OTUs) using UPARSE 7.1 (Stackebrandt and Goebel, 1994; Edgar, 2013) with 97% sequence similarity level. The most abundant sequence for each OTU was selected as a representative sequence. The OTU table was manually filtered, i.e., chloroplast sequences in all samples were removed. To minimize the effects of sequencing depth on alpha and beta diversity measure, the number of 16S rRNA gene sequences from each sample was rarefied to 20,000, which still yielded an average Good’s coverage of 99.09%, respectively.

The taxonomy of each OTU representative sequence was analyzed by RDP Classifier version 2.2 (Wang et al., 2007) against the 16S rRNA gene database (e.g., Silva v138) using the confidence threshold of 0.7. The metagenomic function was predicted by Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) (Douglas et al., 2020) based on OTU representative sequences. PICRUSt2 is software containing a series of tools as follows: HMMER was used to align OTU representative sequences with reference sequences. EPA-NG and Gappa were used to put OTU representative sequences into a reference tree. The castor was used to normalize the 16S gene copies. MinPath was used to predict gene family profiles and locate into the gene pathways. The entire analysis process was performed according to the protocols of PICRUSt2.

Statistical analysis was performed using a t-test, to compare the tested samples and control in the greenhouse experiment (SPSS 16.0). Statistical significance was determined at a p-value of <0.05.

Bioinformatic analysis of the soil microbiota was carried out using the Majorbio Cloud platform. ¹ Based on the OTU information, rarefaction curves and alpha diversity indices including observed OTUs, Chao1 richness, Shannon index, and Good’s coverage were calculated with Mothur v1.30.1 (Schloss et al., 2009). The similarity among the microbial communities in different samples was determined by principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity using the Vegan v2.5–3 package. The PERMANOVA test was used to assess the percentage of variation explained by the treatment along with its statistical significance using the Vegan v2.5–3 package. The linear discriminant analysis (LDA) effect size (LEfSe) (Segata et al., 2011) ² was performed to identify the significantly abundant taxa (phylum to genera) of bacteria among the different groups (LDA score > 2, p < 0.05). Since there is a multicollinearity problem among the 12 soil physicochemical properties/clinical parameters, the variance inflation factor (VIF) for each variable was estimated using the VIF function in the car package. ³ The distance-based redundancy analysis (db-RDA) was performed using the Vegan v2.5-3 package to investigate the effect of soil physicochemical properties/clinical parameters on soil/gut bacterial community structure. Forward selection was based on Monte Carlo permutation tests (permutations = 9,999). Values of the x-axis and y-axis and the length of the corresponding arrows represented the importance of each soil physicochemical properties/clinical parameters in explaining the distribution of taxa across communities. Linear regression analysis was applied to determine the association between major physicochemical properties/clinical parameters identified by db-RDA analysis and microbial alpha diversity indices. The co-occurrence networks were constructed to explore the internal community relationships between the samples (Barberan et al., 2012). A correlation between two nodes was considered to be statistically robust if Spearman’s correlation coefficient was over 0.6 or less than −0.6 and the p-value was less than 0.01.

For the composition analysis of the root exudates of cucumber, the SIMCA16.0.2 software package (Sartorius Stedim Data Analytics AB, Umea, Sweden) was used for multivariate analysis. Data were scaled and logarithmically transformed to minimize the impact of both noise and high variance of the variables. After the transformation, principal component analysis (PCA), an unsupervised analysis that reduces the dimension of the data, was carried out to visualize the distribution and the grouping of the samples. Overall, 95% confidence interval in the PCA score plot was used as the threshold to identify potential outliers in the dataset.

To visualize group separation and find significantly changed metabolites, supervised orthogonal projections to latent structures-discriminate analysis (OPLS-DA) was carried out. Then, a 7-fold cross validation was performed to calculate the value of R2 and Q2. R2 indicates how well the variation of a variable is explained, and Q2 means how well a variable could be predicted. To check the robustness and predictive ability of the OPLS-DA model, permutation was further conducted 200 times. Afterward, the R2 and Q2 intercept values were obtained. In this study, the intercept value of Q2 indicates the robustness of the model, the risk of overfitting, and the reliability of the model, with smaller values indicating better performance.

Furthermore, the value of variable importance in the projection (VIP) of the first principal component in OPLS-DA analysis was obtained. It summarizes the contribution of each variable to the model. The metabolites with VIP > 1 and p < 0.05 (Student’s t-test) were considered as significantly changed metabolites. In addition, commercial databases including KEGG ⁴ and MetaboAnalyst ⁵ were used for pathway enrichment analysis.

Bacillus subtilis isolate 1JN2 was proven to be an efficient biocontrol agent against Ralstonia wilt; the strain was isolated from the stem of a healthy ginger plant, where the Ralstonia wilt was spread severely (Yang et al., 2012). In this study, the plate antagonistic activity of the strain 1JN2 was tested first against Fusarium oxyporum which caused serious wilt on cucumber. According to the results, the strain can inhibit the growth of F. oxyporum, but the inhibition effect was not significant. The antagonistic radius reached 1–3 mm on LB plate (Figure 1A). The biocontrol efficacy against Fusarium wilt of the strain in greenhouse experiment was detected. Interestingly, compared with the plate antagonistic activity, the strain significantly reduced the occurrence of cucumber Fusarium wilt in greenhouse experiment. Compared with the blank control group, the strain could significantly reduce the occurrence of Fusarium wilt disease in artificially controlled greenhouse conditions (Figure 1B). The disease severity of the blank control group reached 58%, while after the pretreatment of biocontrol bacteria, the disease severity of the treatment group was only 22.7%. All the results above indicated that the strain could effectively decrease cucumber Fusarium wilt indirectly.

Colonization ability is generally considered as the first but crucial step to achieve the effect of biocontrol strains. Whether direct or indirect actions, biocontrol strains must accumulate enough population density to achieve disease control or microecological reset. Moreover, the colonization activity of the isolate 1JN2 on cucumber root was detected through plate spreading and SEM analysis. First, the survival of the strain in the presence or absence of cucumber plants was detected. Cucumber plants are promotive for the survival of the strain 1JN2 in soil. The population density of the strain 1JN2 in soil increased slightly from 1 to 3 days after inoculation but decreased significantly from the third day. The decline trend of the control group without cucumber plants was more obvious, which was only 10⁴ CFU/g soil on the 14th day after inoculation. In contrast, the density of the strain in the cucumber plant treatment group was close to 10⁶ CFU /g soil on the 14th day after inoculation and began to rise slowly and reached 10⁶ CFU/g soil 21 days after inoculation (Figure 2B). The colonization sites of the strain on cucumber root were detected by SEM. Compared with the taproot, the strain preferred to colonize on the root tip of cucumber (Figure 2A). It may be because the root tip provides more nutrition and colonization sites.

First, non-target detection for the root exudates from cucumber was performed with LC–MS/MS in order to identify the types of components that changed in cucumber root exudates after treatment with the biocontrol strain 1JN2 (Supplementary Figure S1). Compared with the blank control, the main components of cucumber root exudates in the biocontrol bacteria treatment group included lipids and lipid-like molecules (30.79%), organoheterocyclic compounds (9.49%), phenylpropanoids and polyketides (9.49%), benzenoids (6.94%), organic oxygen compounds (5.32%), and organic acids and derivatives (4.4%), while in the biocontrol bacterial-induced group, similar results were obtained, except for that the content of organic acids and derivatives (6.59%) was higher than that of organic oxygen compounds (5.27%). When we focus on the differences between the 1JN2-treated group and induced group, we found that sterebin G, choline, ginsenoside Rh1, and sulfoquinovosyl diacylglycerol were higher in the 1JN2-treated group, and S−adenosylmethioninamine, agnuside, glycine, and triethanolamine were higher in the 1JN2-induced group. From the above results, it can be inferred that alkaloids and their derivatives are the representative components of the treatment group of the biocontrol strain, which may come from the metabolites of the biocontrol strain 1JN2. The amino acids in the induction group may come from the changes in cucumber root exudates after the treatment of biocontrol strain. Because the strain 1JN2 showed strong control effect on Fusarium wilt in greenhouse experiment, the direct inhibition effect on Fusarium wilt pathogen in plate antagonistic experiment was not significant. Moreover, we speculated that the strain could control Fusarium wilt mainly by an indirect method, changing the microecology of rhizosphere. To identify the specific components that may play a role in the microecological reset, we analyzed the amino acid composition of cucumber root exudates treated with the biocontrol strain 1JN2.

According to the results, each group has its own unique amino acids compared with the control group. In the biocontrol strain-inoculated group (group A1), 5-hydroxylysine, L-methionine, 1-methyl-L-histidine, 4-aminobutyric acid, and L-aspartic acid are the top five amino acids that significantly upregulated compared with the blank control. While in the biocontrol strain-induced group (group A2), four out of the top five upregulated amino acids are the same as the strain inoculated group include 5-hydroxylysine, L-methionine, 1-methyl-L-histidine, and 4-aminobutyric acid. The only amino acid different from the A1 group is glycine rather than L-aspartic acid. When we focus on the differences between the A1 and A2 groups, the content of glutamine, L-tryptophan, glycine, and L-glutamic acid in A2 was significantly higher than those in the blank control group and also higher than those in the A1 group (Figure 3). Therefore, this group of amino acids is considered to be induced by biocontrol strain in cucumber roots and may play a key role in the assembly of cucumber rhizosphere microecology.

Bacterial communities in the rhizosphere of cucumber treated by biocontrol strain and mixture of amino acids were characterized with Illumina HiSeq sequencing. In total, 1,537,449 high-quality sequences were obtained, and each sample contained 27,341 to 90,019 sequences. All the samples were analyzed by PLS-DA on the genus level first, regardless of the biocontrol bacteria treatment group or the amino acids combination treatment group, and it was observed that the samples at different time points before and after treatment were significantly different (Supplementary Figure S2), which means that treatments with biocontrol strain or mixture of amino acids significantly changed the structure of rhizobacterial communities of cucumber. On the phylum level, whether in the treatment group or the control group, the dominant populations of cucumber rhizosphere bacteria are mainly concentrated in six phyla, namely, Firmicutes, Actinobacteriota, Proteobacteria, Chloroflexi, Gemmatimonadota, and Acidobacteriota (Figure 4). Among them, the bacterial population density of Firmicutes in soil samples at the last four time points in the biocontrol bacteria treatment group decreased compared with that before treatment. However, there was no significant change in Firmicutes bacteria in the amino acid mixture treatment group. The bacterial populations of Actinobacteriota, Proteobacteria, Chloroflexi, and Gemmatimonadota had no significant changes in either the biocontrol bacteria treatment group or the amino acid mixture treatment group. The most significant change occurred in the bacterial population of Acidobacteriota. After the treatment of biocontrol strain or amino acid mixture, the bacterial population of Acidobacteriota showed a significant upward trend, that is, from 1.88 and 1.3% before treatment to 11% after treatment (Supplementary Figure S3).

Focusing on the genus level, although the total number of bacteria from Firmicutes in the biocontrol strain treatment group was decreasing, the bacterial diversity of Firmicutes was increasing, and among them, bacteria of the genus Hydrogenispora increased significantly after treatment. The same trend also appeared in the amino acid mixture treatment group. Although there was no significant change in the number of bacteria of Firmicutes in the treatment group, bacteria belong to the genus Hydrogenispora increased significantly after treatment. Among the Acidobacteria bacteria with the largest changes in the biocontrol strain treatment group and the amino acid mixture treatment group, bacteria of the genus Vicinamibacterales were the dominant population. Among the phyla Actinobacteria, Proteobacteria, Chloroflexi, and Gemmatimonadota without obvious changes, the dominant populations mainly include the genera of Marmoricola, Nocardioides, Streptomyces, Sphingomonas, and Gemmatimonadaceae (Supplementary Figure S4). The number of Bacillus was the largest in all dominant populations, exceeding 20% in all treatment groups (Figure 5).

Based on the above results, we can speculate that the bacteria of Hydrogenispora and Vicinamibacteriales may play a key role in the process of the biocontrol strain 1JN2, resetting the bacterial community structure of cucumber rhizosphere, which may come from the recruitment of significantly increased amino acids in cucumber root exudates.