The biocontrol strains, isolated from pig biogas slurry, were obtained from the Institute of Plant Protection, Liaoning Academy of Agricultural Sciences. S. rolfsii was provided by the College of Plant Protection, Shenyang Agricultural University. PHT0-P43GFPmut3a, a plasmid carrying chloramphenicol-and green fluorescent protein-encoding genes, was provided by the Rice Disease Research Laboratory, Institute of Plant Protection, Liaoning Academy of Agricultural Sciences. A peanut rot susceptible variety of the Fuyu four red peanut (Fan, 2021) was used for the analysis (Registration Number: GPD [2018]220156).

Plate confrontation method (Shan et al., 2013) was employed to determine the inhibitory effect of antagonistic bacteria against S. rolfsii. The S. rolfsii isolates were grown on a potato dextrose agar (PDA) medium at 35°C for 5 days. Approximately 5-mm disks of S. rolfsii were placed at the center of a 90-mm diameter PDA plate, and the bacteria were inoculated 25 mm away from the center. In the controls, 5-mm disks of S. rolfsii were placed at the center of a 90-mm diameter PDA plate. The plates were cultured at 28°C until the fungus in the control plate covered the full plate. A light microscope was used to examine the morphology of the hyphae present at the edge of the colony. Antagonistic activity of endophytes was evaluated by the inhibition rate (IR), which was calculated using the following formula:

where D is the diameter of S. rolfsii disks in control treatment, d is the diameter of S. rolfsii with the antagonistic bacteria, and 5 is the diameter of the disks inoculated with S. rolfsii. The strain showing the strongest antagonistic activity against S. rolfsii was selected for further studies.

The edges of S. rolfsii mycelia cultured with the antagonistic bacteria were fixed in 3% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) at 4°C for 24 h. The samples were treated as described by Jia et al. (2020). The samples were observed using a transmission electron microscope (Hitachi H-7650; Hitachi Ltd., Tokyo, Japan) at 115 kV.

The biocontrol strain CB13 was observed with an optical microscope. The physiological and biochemical tests, including Gram staining, methyl red test, starch hydrolysis, peptone hydrolysis, gelatin liquefaction test, VP test, catalase test, and salt tolerance test, were performed with reference to previous literature (Zhu, 2011). The CB13 strain was identified using 16S rDNA, gyrA, gyrB, and rpoB gene sequences. The genomic DNA of CB13 was extracted using a bacterial genomic DNA FastPrep Extraction Kit (Sangon Biotech, Shanghai, China). DNA was quantified using a NanoDrop ND100 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States) and stored at −20°C. Partial sequences of 16S rDNA, gyrA, gyrB, and rpoB were amplified using specific primers (Supplementary Table S1; Ki et al., 2009; Cao et al., 2014; Jing et al., 2019). The 50 μL polymerase chain reaction (PCR)-system contained 5 μL of 10× PCR buffer, 2 μL dNTPs, 1 μL Pfu DNA polymerase (Takara Bio Inc., Beijing, China), 3 μL genomic DNA template (25 ng/μL DNA), 2 μL of each primers (0.1 μM), and 35 μL deionized distilled water. The PCR program included a denaturation step at 94°C for 5 min, followed by 30 cycles of denaturation at 95°C for 1 min, annealing at 50–63°C for 30 s, and elongation at 72\u00B0C for 1 min, followed by a final elongation cycle at 72°C for 5 min, after which the PCR product was kept at 4°C. The PCR products were purified using a DNA gel extraction kit (Qiagen, Hilden North Rhine-Westphalia, Germany) and sent to Sangon Biotech (China) for sequencing. DNA sequence homology analysis was performed using the BLAST tool of the GenBank nucleotide database of the National Center for Biotechnology Information (Bethesda, MD, United States).¹ Multiple alignments with sequences of closely related Bacillus strains and sequence similarity calculations were performed using CLUSTAL X (Thompson et al., 1997). Phylogenetic trees were constructed with Mega 6.0 (Kim et al., 1993) using the neighbor-joining method with 1,000 bootstrap resampling replications (Felsenstein, 1985).

A bacterial suspension of the CB13 bacterial inoculum was prepared at a concentration of 10⁷ CFU/mL. The bacteria were seeded at a density of 1% into 100 mL lysogeny broth (LB) liquid medium and shaken at 150 rpm at 35°C for 24 h to obtain a bacterial inoculum for plant treatment.

The phytopathogen inoculum was prepared by autoclaving wet sorghum seeds in a 150 mL Erlenmeyer flask three times and then infecting them with mycelium plugs of S. rolfsii (6 mm diameter). The Erlenmeyer flask was maintained at 28°C until abundant mycelial growth was observed (approximately 7–10 days).

The pot experiment was conducted four times, in June and August 2021 and 2022. The peanut seeds were surface disinfected as described by Lu et al. (2015). The surface-sterilized seeds were sown in plastic pots filled with nutrient soil, with each pot containing 4–5 seeds. Three peanut plants were maintained after seedling growth (at temperature 28 ± 5°C and relative humidity 65 ± 5%) for 30 days. The peanut seedlings were divided into four treatment groups, with 20 pots in each group, and the experiments were performed in triplicate. Treatment groups were as follows: (A) CK—infected peanut plants without treatment; (B) THI—infected peanut plants sprayed with 20 mL of 240 g/L thifluzamide (diluted 500×; Zhejiang Hangzhou Yulong Chemical Co. LTD; thifluzamide treatment group); (C) RI—infected peanut plants watered with 20 mL bacterial inoculum (diluted 10×); (D) ST—seeds were soaked in bacterial inoculum (diluted 10×) for 24 h before sowing. The plants were treated thrice with CB13 and chemical agents once every 7 days. Infection with S. rolfsii was performed 1 day after the third treatment. Approximately 3 g of infected sorghum grains was used to inoculate the root of the plant, and the plants were examined after 5 days. The extent of the stem rot disease was scored as 0 for no stem lesion, 1 for small stem lesion, and 2, 3, and 4 for the wilting and death of ≤25, 26% ~ 50, and > 50% of whole plants in each group, respectively. The disease incidence rate (DIR), disease index (DI), and control efficacy (CE) were calculated as follows (Li et al., 2017):

where n is the number of infected plants, N is the total number of investigated plants, Ni is the number of infected plants at a certain severity level, i is certain severity level, DI_CK is the disease index in control, and DI_T is the disease index in treatment.

The natural transformation method (Li, 2011) was used to obtain GFP-labeled strains. Peanut seedlings were watered with 10 mL of CB13-GFP strain seed liquid at a concentration of 1 × 10⁷ CFU/mL, and flower roots and root soil were taken on the 3rd, 6th, 11th, 16th, 23rd, 30th, 40th, and 50th days after treatment. The roots in large granular soil were carefully removed, placed in a bottle containing 50 mL sterile water, and oscillated for 30 min. The soil suspension was shaken; 20 mL of the soil diluent was placed in a bottle, and the dry weight of soil in the soil suspension was determined; the residual soil was diluted to form a slurry gradient, and a coated chloramphenicol-resistant plate count was determined. The roots were dried with absorbent paper, weighed, cut into small segments of approximately 0.5 mm, ground fully with sterile water, diluted in gradient, and coated with a chloramphenicol screening medium (10 μg/mL concentration). The number of CB13-GFP colonies was determined after 2 days.

The seed disinfection and potted plant-handling methods were conducted as described above. After 30 days of seedling growth (at a temperature of 28 ± 5°C and relative humidity of 65 ± 5%), one plant per pot was maintained for inoculation. There were four treatments, as follows, with 20 pots in each treatment: (A) Uninoculated—uninfected and untreated plants; (B) Uninfected CB13—seeds soaked in bacterial inoculum (diluted 10×) for 24 h before sowing; (C) Inoculated—plants infected with S. rolfsii alone; (D) CB13 + infection—seeds soaked in bacterial inoculum (diluted 10×) for 24 h before sowing and then infected with S. rolfsii. Leaves from the same part of the plant were sampled at five time points (0, 12, 24, 48, and 72 h) post-infection, immediately wrapped in a tin foil, and frozen in liquid nitrogen. Enzyme activity was measured according to the kit manufacturer’s protocols (Sangon Biotech (Shanghai) Co., LTD, Shanghai, China).

Seed disinfection and potted plant-handling were performed as described above. Peanut seedlings were divided into four treatments, with three pots per treatment, repeated in triplicate. Treatment groups were as follows: (A) untreated peanut plants (without pathogen or bioagent), labeled as “NS group;” (B) peanut plants watered with 20 mL of bacterial inoculum alone, labeled as “CS group;” (C) peanut plants infected with S. rolfsii alone, labeled as “SS group;” (D) peanut plants watered with 20 mL of bacterial inoculum and then infected with S. rolfsii 24 h later, labeled as “CSS group.” CB13 treatment was performed once every 7 days a total of three times; infection with S. rolfsii was performed 1 day after the third treatment, and soil samples were collected 10 days later. The root soils of the peanut plants were collected according to Courchesne’s root-shaking method (Courchesne and Gobran, 1997), and three soils were mixed together to form one soil sample. Microbial DNA was extracted from soil samples using the E.Z.N.A.^® soil DNA Kit (Omega Bio-Tek, Norcross, GA, United States) according to the manufacturer’s protocols. The final DNA concentration and purification were determined using a NanoDrop 2000 UV–vis spectrophotometer (Thermo Scientific, Wilmington, DE, United States), and DNA quality was checked using 1% agarose gel electrophoresis. The internal transcribed spacer (ITS) regions were amplified with primers ITSI and ITS2R, and V3-V4 hypervariable regions of the bacterial 16S rDNA gene were amplified with primers 338F and 806R (Supplementary Table S1; Xu et al., 2016) using a thermocycler PCR system (GeneAmp 9,700, ABI; Thermo Fisher Scientific, Waltham, MA, United States). PCR reactions were conducted using the following program: 3 min of denaturation at 95°C; 35 cycles (16S rDNA; 27 cycles) of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, and elongation at 72°C for 45 s; and a final extension at 72°C for 10 min. PCR was performed in triplicate using 20 μL reaction mixtures containing 2 μL (16S rDNA, 4 μL) of 5× FastPfu Buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of each primer (5 μM), 0.4 μL FastPfu Polymerase, and 10 ng template DNA. The resulting PCR products were extracted from a 2% agarose gel, further purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, United States), and quantified using QuantiFluor™-ST (Promega, Madison, WI, United States).

Raw FASTQ files were de-multiplexed using an in-house perl script, and then quality-filtered by fastp (version 0.19.6; Chen S. F. et al., 2018) and merged by FLASH (version 1.2.7; Tanja and Steven, 2011) with the following criteria: (i) the 300 bp 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 characters 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. Reads that could not be assembled were discarded; (iii) Samples were distinguished according to the barcode and primers, and the sequence direction was adjusted, exact barcode matching, 2 nucleotide mismatch in primer matching. Then, the optimized sequences were clustered into operational taxonomic units (OTUs) using UPARSE (version 7.1; Edgar, 2013) with 97% sequence similarity level. The most abundant sequence for each OTU was selected as a representative sequence. The taxonomy of each 16S rDNA was analyzed by RDP Classifier (version 2.11)² against the Silva (version 138)³ 16S rDNA database using confidence threshold of 70% (Cole et al., 2009). The taxonomy of each ITS gene sequence was analyzed by RDP Classifier against the UNITE (version 8.0)⁴ using confidence threshold of 70% (Cole et al., 2009).

To determine the content in peanut rhizosphere soil, the 16S rDNA/ITS copies of Bacillus spp./S. rolfsii were determined using real-time quantitative PCR (qPCR) by designing specific primers for Bacillus spp.: B1F/B1R and S. rolfsii: SRITSF/SRITSR (Supplementary Table S1; Pravi et al., 2014). The qPCR was performed in a 20-μL reaction mixture that consisted of 2 μL of template DNA, 10 μL of 2 × SYBR Mix, 0.25 μL of each primer, and 7.5 μL of ddH₂O. The PCR protocol included a 5-min initial denaturation at 95°C, 40 cycles of 95°C for 30 s, 56°C for 30 s, and 72°C for 40 s. A standard curve that shows the relationship between copy numbers logarithm of Bacillus spp./S. rolfsii and Ct values was constructed, as described by Jing et al. (2019). The copies of Bacillus spp. and S. rolfsii were determined by using LineGene 9600 Plus (FQD-96A) Fluorescence Quantitative Thermal Cycler (Bioer Technology, China). The content was expressed by log copies DNA·g⁻¹ soil.

Data were expressed as mean ± SD, and a suite of statistical analysis (SPSS 19 and MS Excel 2007) was used to find the mean and standard error. Statistical differences between treatments were analyzed using one-way ANOVA test at a 5% significance level. The software Mothur (Schloss et al., 2009) was used to calculate the alpha diversity index under random sampling, and alpha diversity and richness were determined at the OTU level using the Shannon index and Chao1 index, respectively. The Wilcoxon rank-sum test was used to compare statistical differences between groups. Principal coordinates analysis (PCoA) was conducted based on a Bray–Curtis dissimilarity matrix computed from the samples. The one-way analysis of similarity (ANOSIM) test (Clarke and Warwick, 2001) with 999 permutations was conducted to determine statistical significance. The functions of bacteria and fungi in peanut rhizosphere soil were analyzed using PICRUSt2 (version 2.2.0).⁵ The p value was corrected using a multiple test with false discovery rate (Fdr) and results with p < 0.05 were considered statistically significant. Differential abundance analysis for sequence count data was performed using DESeq2 (version 1.14.1; Love et al., 2014), with significance defined as p < 0.05, and Benjamini-Hochberg-adjusted p < 0.05. Unless otherwise stated, statistical analyses and plotting were carried out using R software (version 3.3.1).

The antagonistic bacteria effectively inhibited the radial growth of S. rolfsii (Figures 1A,B). The CB13 strain exhibited the highest inhibition zone and IR against the radial growth of S. rolfsii, at 12.00 mm and 72.56%, respectively. Thus, the CB13 strain was selected for further studies.

Under the light microscope, the hyphae of S. rolfsii with normal growth were smooth and uniform (Figure 1C), while some hyphae of the inhibited S. rolfsii appeared darker, ablated, and ruptured, with vesicles present at the top (Figure 1D). Based on the transmission electron micrographs, normal S. rolfsii mycelial cells appeared full, with intact mitochondria, cell and nuclear membranes, and gathered nucleoli (Figure 1E), whereas the mycelial cells of S. rolfsii co-cultured with CB13 appeared pyknotic, with disintegrated mitochondria, cleaved nuclear membranes, and separated nucleoli (Figure 1F).

On LB medium, the colonies of CB13 appeared nearly round and milky-white–opaque, with dry folds on the surface and irregular edges. Optical microscopy identified the CB13 strain as rod-shaped with a size of 0.5 μm × 1.5–3.5 μm. The results of physiological and biochemical tests performed on strain CB13 are shown in Supplementary Table S2. Based on these results, along with morphological examinations, strain CB13 was preliminarily classified as a strain of the Bacillus spp. Multigene 16S rDNA, gyrA, gyrB, and rpoB phylogenetic analysis revealed 99% consistency between strain CB13 and Bacillus velezensis (Figure 2). Accordingly, strain CB13 was identified as B. velezensis and named B. velezensis CB13 based on the multigene sequence analysis and morphological, physiological, and biochemical characteristics.

Bacillus velezensis CB13 treatment reduced the incidence and disease index of peanut stem rot in pot assays (Figure 3). Thifluzamide, RI, and ST significantly reduced stem rot incidence and disease index (p < 0.05). There were no significant differences in stem rot incidence and disease index between the chemical control and ST (p > 0.05); however, the disease index (May 2020 and July and May 2021) and disease incidence (May 2021) were significantly higher than those in RI (p < 0.05). The control efficiencies of ST in four pot experiments were 65.44, 73.33, 85.13, and 94.92%. The levels of control effect achieved with thifluzamide or ST had no significant differences (p > 0.05) but were significantly higher than those achieved with RI (p < 0.05).

The antagonistic strain B. velezensis CB13 was labeled with GFP using a natural transformation method (Figure 4C). The ability of B. velezensis CB13 to colonize peanut roots and rhizosphere soil was evaluated over 50 days (Figures 4A,B). CB13-GFP reached its highest density in peanut root after 23 days (51.44 × 10⁴ CFU/g of peanut root), which decreased thereafter to a final density of 18.89 × 10⁴ CFU/g of peanut root after 50 days. Furthermore, CB13-GFP reached its highest density in rhizosphere soil after 30 days (12.10 × 10⁸ CFU/g of rhizosphere soil), which decreased thereafter to a final density of 2.62 × 10⁸ CFU/g of rhizosphere soil after 50 days. The root colonization of CB13-GFP was observed under a fluorescence microscope (Figure 4D).

The activities of defense enzymes increased to varying degrees in infected leaves after B. velezensis CB13 treatment compared to those in infected leaves without treatment (Figure 5). At 12 h, the activities of superoxide dismutase (SOD) and β-1,3-glucanase (GLU) in the CB13 + inoculated group increased significantly by 1.32 and 1.14 folds, respectively, compared to those in the inoculated group, and then decreased. The activities of polyphenol oxidase (PPO), ascorbate peroxidase (APX), catalase (CAT), phenylalanine ammonia-lyase (PAL), peroxidase (POD), and chitinase (CHI) in the CB13 + inoculated group increased significantly and were higher than those in the positive control group at all time points examined after inoculation. The activities of POD, CAT, and APX increased significantly by 1.72, 1.57, and 3.72 folds, respectively, compared to those in the inoculated group.

After the removal of low-quality reads, 420,323 valid bacterial reads and 785,669 valid fungal reads were obtained from 12 samples. The number of valid bacterial reads ranged from 26,649 to 44,340 among the 12 samples, and the number of valid fungal reads ranged from 57,323 to 83,454. The sequences representing OTUs were identified with 97% similarity (Supplementary Tables S3, S4). The rarefaction curves for individual samples were close to the platform period, indicating that sequencing depth reflected the microflora richness (Supplementary Figure S1).

The Shannon index for the soil microflora revealed that bacterial diversity in the four treatment groups decreased in the order CS > CSS > SS > NS, while fungal diversity decreased in the order NS > CSS > SS > CS; these results were consistent with the Simpson’s index values (Figure 6). The Shannon index of species diversity indicated that the bacterial diversity of the CS group was significantly higher than that of the other treatment groups (p < 0.01). The results showed that B. velezensis CB13 treatment significantly increased the diversity of soil bacteria. The fungal diversity of the soil after B. velezensis CB13 treatment was not significantly different from that of the control (p > 0.05), whereas the fungal diversity of soil inoculated with S. rolfsii decreased significantly compared to that of the control (p < 0.05).

At the phylum level, the composition and abundance of soil bacterial and fungal flora in the peanut root system were analyzed (Figures 6E,F; Supplementary Tables S5, S6). The predominant bacterial phyla (>1%) were Actinobacteriota, Proteobacteria, Chloroflexi, Acidobacteriota, Bacteroidota, Firmicutes, Patescibacteria, Gemmatimonadota, and Myxococcota. Inoculation with B. velezensis CB13 significantly decreased the abundance of Actinobacteriota and significantly increased the abundance of Proteobacteria, Acidobacteriota, Firmicutes, Gemmatimonadota, and Myxococcota. The predominant fungal phyla (>1%) were Ascomycota, Basidiomycota, unclassified_k_Fungi, Mortierellomycota, and Olpidiomycota. Inoculation with B. velezensis CB13 significantly decreased the abundance of Olpidiomycota (p < 0.05). Differential abundance analysis by using DESeq2, we confirmed that B. velezensis CB13 significantly increased the abundance of Proteobacteria, Firmicutes, Gemmatimonadota, and Myxococcota and significantly decreased the abundance of Actinobacteriota and Olpidiomycota (p < 0.05; Supplementary Table S7).

Venn analysis revealed that the soil samples from different groups had 530 bacterial genera and 165 fungal genera in common (Figures 7A,B). The 10 dominant genera identified in the soil samples were selected for relative proportion variance (Supplementary Figure S2). Compared to that in the uninoculated soil, the proportion of the 10 dominant bacterial genera did not change significantly after B. velezensis CB13 treatment (p > 0.05). Regarding fungi, after B. velezensis CB13 treatment, the proportion of Pseudogymnoascus increased compared with that in the uninoculated soil (p < 0.05). Differential abundance analysis showed that there were no significant differences in the proportion of the 10 dominant bacterial and fungal genera between the uninoculated and inoculation with B. velezensis CB13 soil (p > 0.05). PCoA confirmed that the samples from soil inoculated with S. rolfsii clustered together, indicating that S. rolfsii influences rhizosphere microbiome community structure (Figures 7C,D). Contrastingly, the results for other samples that were separately clustered suggested that B. velezensis CB13 could alter the structure and maintain the stability of rhizosphere microbiome community structure.

The functional composition and proportion of the microbial diversity in the peanut rhizosphere soil were analyzed in different samples. According to the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, the primary functional layer consists of cellular processes, environmental and genetic information processing, human diseases, metabolism, and organism system (Figure 7E). The different sample groups significantly differed in terms of human diseases, cellular processes, and metabolism. Further functional analysis showed that the secondary functional layer of genes predominantly comprised 46 subfunctions, including global and overview maps, carbohydrate 337 metabolism, amino acid metabolism, and energy metabolism, etc. (Supplementary Table S8). Additionally, we focused on the pathways related to disease resistance in the third functional layer (Figure 7F). Compared with that in the uninoculated soil, the functional proportion of genes involved in lipopolysaccharide biosynthesis, biosynthesis of siderophore group non-ribosomal peptides, and tryptophan metabolism in the soil inoculated with B. velezensis CB13 and S. rolfsii increased significantly (p < 0.05).

Based on the MetaCyc pathway, the gene functions of fungi comprised 74 pathways, including aerobic respiration, the glyoxylate cycle, the pentose phosphate pathway (non-oxidative branch), and guanosine nucleotide degradation (Supplementary Table S9). After B. velezensis CB13 treatment, the proportion of the functional pathways of NAD/NADH phosphorylation and dephosphorylation and 4-amino-2-methyl-5-phosphomethylpyrimidine biosynthesis (yeast) increased significantly (p < 0.05), and the proportion of functional pathways of GDP-mannose biosynthesis, adenine and adenosine salvage III, phospholipid remodeling (phosphatidylethanolamine, yeast), and phospholipases decreased significantly (p < 0.05). Compared with that in the uninoculated soil, the proportion of 11 pathways in the soil inoculated with S. rolfsii was significantly higher (p < 0.05), including glyoxylate cycle, fatty acid, and beta-oxidation (peroxisome, yeast), etc. In the soil inoculated with S. rolfsii, the proportion of 16 pathways, including pentose phosphate pathway (non-oxidative branch), D-myo-inositol (1,4,5)-trisphosphate biosynthesis, etc. were significantly lower compared to the uninoculated soil (p < 0.05).

In order to determine the content of Bacillus spp. and S. rolfsii, we designed the specific primers and confirmed the specificity by electrophoresis analysis. The results indicated that there was a single and clear band amplified by B1F/B1R in four treatment groups, and there was no visible band observed amplified by SRITSF/SRITSR in uninoculated S. rolfsii groups (Supplementary Figure S3). Thus, these primers could be accepted and used for the quantification of Bacillus spp. and S. rolfsii in peanut rhizosphere soil. The qPCR results of Bacillus spp. and S. rolfsii (Figures 4E,F) show that the population of Bacillus spp. in the CS and CSS groups were significantly higher than those in the NS and SS groups, respectively (p < 0.05), indicating that B. velezensis CB13 stably colonized or increased the content of Bacillus spp. in the soil. The population of S. rolfsii in SS group was significantly higher than that in the CSS group (p < 0.05), indicating that the biocontrol bacterium B. velezensis CB13 significantly inhibited the reproduction of S. rolfsii in the soil.