Culture of microbes and plant material

The antagonistic bacteria KRS010 was isolated from the seed of Verticillium wilt-resistant G. hirsutum cultivar “Zhongzhimian No. 2” and incubated on Luria-Bertani (LB, tryptone 10 g, NaCl 10 g, yeast extract 5 g, double-distilled H2O 1000 mL) plates at 28 °C. The pathogenic fungi were cultured on potato dextrose agar (PDA, potato 200 g, glucose 20 g, agar 15 g, ddH2O 1000 mL) plates at 25 °C. G. hirsutum cultivar Junmian No. 1 cotton seedlings and NahG transgenic tobacco plants were cultured in 25 °C greenhouse with a 16-h light and 8-h darkness photoperiod. The NahG transgenic tobacco plants expressing a SA hydroxylase gene from Pseudomonas putida are able to metabolize SA to catechol, leading to a dramatic decrease in the plant SA content. Throughout the growing period, these cotton plants were cultivated in a fixed position in the greenhouse.

Identification of antagonistic bacterium KRS010Morphological identification

The strain KRS010 was streaked on an LB plate and cultured at 28 °C for 12 h, in which the morphology, color, wet or dry, smooth or rough of a single colony was observed. Gram staining was conducted by a series of steps including initial dyeing, mordant dyeing, decolorization, and redyeing by Gram Staining kit (Coolaber, SL7040) [60]. Escherichia coli DH5α was used as negative control. In order to observe the biofilm formation, a single colony of the KRS010 was inoculated in 50 mL of LB broth with 200 rpm at 28 °C overnight, followed by stationary incubation at 28 °C for 7 days for the formation of biofilm.

Molecular identification

A phylogenetic analysis of KRS010 and other Bacillus species strains was conducted with Phylophlan v3.0.2 based on the complete genome sequence. Phylogenetic tree was established by TVBOT (https://www.chiplot.online/circleTree.html).

Physiological and biochemical characteristics analysis

Physiological and biochemical characteristics were assessed according to the methods described in Bergey’s Manual of Systematic Bacteriology and the Common Bacterial Identification Manual [61, 62]. The experiment includes the characterization of capability for inorganic phosphorus utilization (Bacterial Inorganic Medium, Hopebio, HB8670), organic phosphorus utilization (Bacterial organophosphorus Medium, Hopebio, HB8673), potassium utilization (Silicate bacteria medium, Coolaber, MM6021), nitrogen fixation (Ashby’s Medium, COOLABER, MM5041), siderophore production (modified CAS agar medium kit, Coolaber, PM0821), IAA synthesis, oxidase activity, amylase activity, hydrogen sulfide production (Triple sugar iron agar, COOLABER, MM3011), gelatin liquefaction, phenylalanine deaminase test (Phenylalanine medium, Solarbio, LA1720), protease activity, methyl red test, gluconate production, nitrate reduction, citrate production, and anaerobic determination. All the above tests were performed twice, and each experiment was performed in triplicate.

Safety evaluation

The roots of cotton seedlings were soaked in KRS010 cell suspension (OD600 = 1.0) for 1 h, followed by staining with trypan blue for 3 h, and the roots were observed with a microscope. Roots soaking with LB broth were severed as control. Mueller Hinton (MH) plates and Columbia blood plates were used to detect the hemolysis of KRS010. The E. coli DH5α has no hemolysis [63] that was used as a negative control. All of these assays were repeated twice, and each treatment was performed in triplicate.

Identification for antifungal activity of KRS010 in vitro

The activity of KRS010 against the colony growth of seven pathogenic fungi was evaluated through confrontation tests and the treatments of volatile organic compounds (VOCs). Firstly, seven pathogenic fungi were firstly cultured on the center of the PDA medium for 4 days, and the hyphal bulks with a diameter of 6 mm were inoculated to the center of the PDA plates.

For confrontation assays, the 10 μL KRS010 (OD600 = 1.0) cell suspension was dropped at 20 mm from the top, bottom, left, and right of the fungal bulks, respectively. The 10-μL LB broth at the same position of the hyphal bulks was severed as controls. All plates were incubated at 25 °C for 3~7 days, and the semidiameter of the pathogen mycelium disc grown on the treated plates and the control plates were measured by cross-measurement method. The semidiameter of the colony was recorded, of which the colony semi-diameters of the control and treatment group were marked as “A” and “B,” respectively. The inhibition rate was calculated by the following formula: inhibition rate (%) = [(A − B)/(A − 6)] × 100.

For the assays of VOCs released from KRS010 on mycelial growth, LB plates covering 100 μL KRS010 cell suspension were prepared, and these treated LB plates were sealed with the prepared plates with fungal bulks by parafilm. The LB plates without KRS010 cell suspension were used as control. All plates were incubated at 25 °C for 3~7 days, and the diameter of the colony was measured, of which the colony diameters of control and treatment were marked as “a” and “b,” respectively. The inhibition rate was calculated using the formula: inhibition rate (%) = [(a − b)/(a − 6)] × 100.

The cell-free culture filtrate was obtained as described in a previous study with some modifications [64]. KRS010 fermentation broth was centrifuged at 8000 rpm for 20 min, and the supernatant was collected and filtrated through a 0.22-µm microporous membrane and stored at 4 °C. The cell-free supernatant was mixed into PDA plates at the proportion of 5%, 10%, and 20% (v/v), respectively, and inoculated with fungal bulks of V. dahliae. The hyphal growth and development were observed after 5 days.

All the above experiments were performed twice, and each treatment/group in every experiment was conducted at least three plates.

Biocontrol effect of KRS010 against cotton Verticillium wilt

Three-week-old cotton seedlings were used to evaluate the biocontrol activity of isolate KRS010 against Verticillium wilt caused by V. dahliae. The KRS010 cell suspension with OD600 = 1.0 and conidial suspension of V. dahliae as 1 × 107 conidia/mL were prepared. This assay included four groups: cotton seedlings were treated with KRS010 cells suspension individually (marked with “KRS010”), conidial suspension of V. dahliae individually (marked with “Vd”), KRS010 cells suspension with conidial suspension of V. dahliae (marked with “KRS010+Vd”), and water only treatment (marked with “CK”). The amount of inoculation was 20 mL, and 20 mL of corresponding solvent was used for other treatments each time. All of the treatments were incubated at the same conditions as the previous cotton seedlings culture. Disease symptoms were observed and quantified at 2 days post-treatment following the method in our previous study [65]. Symptoms of the seedlings at 30 days were surveyed and classified into 5 levels: level 0, healthy seedlings; level 1, ≤ 25% of leaves showing chlorosis or wilt; level 2, ≤ 50% of leaves showing chlorosis or wilt; level 3, ≤ 75% of leaves showing chlorosis or wilt; and level 4, more than 75% of leaves showing chlorosis or wilt [66]. At least 16 cotton seedlings were used for each group in 4 pots, and the experiment was conducted 3 times. The disease index (DI) and the control efficacy were calculated with the following formulas [67, 68]:

$$\mathrm=\;\left[\sum\left(\mathrm\;\mathrm\;\mathrm\;\mathrm\;\mathrm\;\times\;\mathrm\;\mathrm\right)\;/\left(\mathrm\;\mathrm\;\times\;\mathrm\;\mathrm\;\mathrm\;\mathrm\right)\;\right]\;\times\;100\;$$

$$\mathrm\;\mathrm\;(\%)\;=\left[\left(\mathrm\;\mathrm\;\mathrm\;\mathrm-\;\mathrm\;\mathrm\;\mathrm\;\mathrm\;\mathrm)/\mathrm\;\mathrm\;\mathrm\;\mathrm\right)\right]\;\times\;100$$

Subsequently, the stems of the seedlings were slit longitudinally to check the damage caused by V. dahliae, and the roots of the seedlings were collected for V. dahliae biomass analysis.

The genomic DNA (gDNA) of cotton was extracted according to a DNAsecure plant kit (Tiangen, Beijing, China) for the detection of relative fungal biomass. SYBR green-based qPCR was used to detect the biomass of V. dahliae with an initial 95 °C denaturation step for 3 min, followed by denaturation for 15 s at 95 °C, annealing for 20 s at 60 °C, and extension for 20 s at 72 °C for 34 cycles. The V. dahliae EF1α gene was used to quantify fungal colonization, and the Ubiquitin gene served as an endogenous control. These primer pairs are listed in Additional file 1: Table S1. Moreover, the plant height, stem thickness, dry weight, fresh weight, chlorophyll content, and nitrogen content of cotton in four treatments were also compared to estimate the growth-promoting effect of KRS010. Every experiment was performed with the same batch of cotton seedlings. At least 20 cotton seedlings were treated for each treatment, and the experiment was conducted three times.

RT-qPCR for the relative gene expression levels in cotton or V. dahliae

The expression levels of defense-related genes in cotton roots and leaves were detected by reverse transcription quantitative PCR (RT-qPCR). The KRS010 fermentation broth (OD600 = 1.0), the solution of 10 mmol/L SA, and the solution of 100 μmol/L JA was sprayed on the 3-week-old cotton leaves for 12 h before collecting leaves [69]. The KRS010 fermentation broth (OD600 = 1.0) was inoculated on the 3-week-old cotton roots for 48 h before collecting the roots. The water treatment group was severed as a control. For the detection of defense-related gene expression levels in SA-deficient tobacco, the water, the KRS010 fermentation broth (OD600 = 1.0), and the solution of 10 μmol/L SA were sprayed on the 3-week-old tobacco leaves for 6 h before collecting leaves [70].

Total RNA extraction and first-strand cDNA synthesis were performed using the EASYspin Plus RNA speed extract kit (Aidlab, Beijing, China) and a cDNA synthesis supermix kit (TransGen, Beijing, China) according to the manufacturer’s instructions (both kits included a gDNA-removal procedure). For cotton, gene expression levels of the salicylic acid (SA) pathway-related genes (GhEDS1, GhICS, GhNPR1, GhNDR1, WRKY7, GhPR1, and GhPR5) and jasmonic acid (JA) pathway-related genes (GhAOS, GhJAZ, GhAOCS, GhLOX1, and GhOPR3) were normalized to the GhUBQ7 gene by RT-qPCR. For tobacco, gene expression levels of the salicylic acid (SA) pathway-related genes (NbPR1, NbPR2, NbPR5, and NbPAL) were determined by RT-qPCR after normalizing to the NbEF-1α gene.

For the detection of hypha melanin biosynthesis-related gene expression levels in V. dahliae, the 100-mL cell suspension of KRS010 (OD600 = 1.0) was coated on an LB plate, and 100 mL conidial suspension of V. dahliae was inoculated on an organic filter membrane on top of a PDA plate. Then, the two treated plates were sealed together with parafilm and incubated at 25 °C for 6 days. The two sealed base plates, involving a LB plate with 100 mL LB broth and a PDA plate with V. dahliae, served as control. All treatments and controls were performed with five replicates. The fungal hypha was collected from the organic filter followed by total RNA extraction and synthesis of first-strand cDNA. The melanin biosynthesis and autophagy processes genes (VdVayg1, VdAG_00183, VdAG_03393, VdCmr1, VdLac1, and VdH1) were detected and normalized to the V. dahliae EF1α gene by RT-qPCR.

Reverse transcription-quantitative PCR (RT-qPCR) was carried out using TransStart Top Green qPCR SuperMix (+DyeII) (TransGen, Beijing, China) following the manufacturer’s instructions. Relative transcript levels of different genes among various samples were evaluated using the 2−∆∆CT method as described previously [71]. The RT-qPCR experiment was repeated twice, and each contained three technical replicates. The RT-qPCR primer pairs are listed in Additional file 1: Table S1.

Microscopic observations

After 4 days of treatment of VOCs or confrontation culture of KRS010, the hyphae at the edge area of V. dahliae were observed via a stereomicroscope (Nikon SMZ18). The grayscale area is used to evaluate the production of melanin using ImageJ calculation.

Strain KRS010 was incubated in LB at 28 °C for 96 h with shaking (180 rpm) and the cultural liquid was centrifuged to separate a supernatant and precipitate at 4 °C and 8000 rpm for 15 min. The supernatant was filtrated through a bacterial filter (0.22 μm), and the aseptic filtrate was obtained. After the conidial suspension of V. dahliae was incubated with cell-free culture filtrate of KRS010, conidial of V. dahliae was observed via a differential interference contrast microscope (DIC) at 12 h, 24 h, and 48 h. After 4 days of treatment with VOCs and a 20% cell-free supernatant of KRS010, the hyphae at the edge of the bacteriostatic area were picked with an inoculation blade and placed in a sterile centrifuge tube that contained 2.5% glutaraldehyde fixative for 48 h. These samples were observed with a scanning electron microscope (SEM). Before observations, the pretreated samples were dehydrated using graded ethanol at 30%, 50%, 70%, 80%, 90%, 95%, and 100%, respectively. The samples were dried using a CO2 critical point dryer.

Statistical analysis

The standard errors in all involved figures were calculated for each treatment or group with at least three replicates. Unpaired Student’s t test was performed to determine statistical significance. One-way analysis of variance (ANOVA) and least significant difference (LSD) were used to analyze the statistical significance of multiple groups.