Role of Biofilm Formation by Bacillus pumilus HR10 in Biocontrol against Pine Seedling Damping-Off Disease Caused by Rhizoctonia solani

: The biocontrol process mediated by plant growth-promoting rhizobacteria (PGPR) relies on multiple mechanisms. Bioﬁlm formation plays an important role in the ability of PGPR to control plant diseases. Bacillus pumilus HR10, one such PGPR, promotes the growth of Pinus thunbergii . This study showed that the wild-type strain B. pumilus HR10 produces a stable and mature bioﬁlm in vitro. Bioﬁlm-deﬁcient mutants of B. pumilus HR10 with di ﬀ erent phenotypes were screened by mutagenesis. The contents of extracellular polysaccharides (EPS) and proteins produced by the mutant strains were signiﬁcantly reduced, and the bioﬁlms of the mutants were weakened to varying degrees. The swarming abilities of the wild-type and mutant strains were positively correlated with bioﬁlm formation. A colonization assay demonstrated that B. pumilus HR10 could colonize the roots of Pinus massoniana seedlings in a large population and persist, while bioﬁlm-deﬁcient mutants showed weak colonization ability. Furthermore, a biocontrol assay showed that biocontrol e ﬃ cacy of the mutants was reduced to a certain degree. We determined the inhibitory activity of B. pumilus HR10 and its ability to induce systemic resistance against Rhizoctonia solani of plants. The synthesis of lipopeptide antibiotics is probably involved in bioﬁlm formation by B. pumilus HR10. These observations not only provide a reference for further research about the coordinated action between bioﬁlm formation and the multiple biocontrol mechanisms of B. pumilus HR10 but also improve the understanding of the regulatory pathway of bioﬁlm formation by B. pumilus HR10. strains induce plant systemic resistance, the transcription levels of PR2 , PAL , COI , and GPX were measured and evaluated. The seedlings not inoculated with bacteria served as control (CK) plants. The seedlings inoculated with WT B. pumilus HR10 or bioﬁlm-deﬁcient mutants exhibited high expression levels of PR2 (4.2-fold, 1.8-fold, 2.5-fold and 1.9-fold compared to the expression of PR2 in the CK, respectively), while the expression of PAL , COI , and GPX showed no signiﬁcant di ﬀ erences (Figure 6). The results indicate that inoculation with WT B. pumilus HR10 or mutants contributes to inducing P. massoniana resistance to some extent.


Introduction
Plant growth-promoting rhizobacteria (PGPR) have been reported to exhibit biocontrol processes including competition, antibiosis, growth promotion, and the induction of systemic resistance in plants [1][2][3][4][5][6]. The Bacillus genus is one of the most frequently reported and promising PGPR with various biocontrol mechanisms. The nematicidal activity of the guanidine compound from B. pumilus LYMC-3 and its high potential in the biocontrol of Bursaphelenchus xylophilus was demonstrated [7]. Bacillus velezensis FZB42, a model strain for excellent plant growth promoting and biocontrol rhizobacteria, has been studied on its genes and expression involved in biocontrol process and the bacteria-plant interaction [8][9][10].
Due to colonizing the root of plants persistently, PGPR can effectively promote the growth of plants and inhibit the pathogens [10][11][12]. Persistent colonization of PGPR on roots is essential for suppressing different plant diseases, and biofilm formation is considered a key factor for efficient root

Quantitative Estimation of Biofilm Polysaccharides
Standard curve of the glucose: the glucose (1 mL) concentrations were set as 0, 40, 80, 120, 160 and 200 µg/mL. Anthrone (0.2 g) was dissolved in 100 mL of sulfuric acid, and then 2 mL of liquid was slowly added into the glucose solutions. The tubes were incubated in boiling water for 10 min and then in ice-cold water for 10 min. Subsequently, the solutions were placed at room temperature for 2 min, and the absorbances were measured at 620 nm with a Multiskan Spectrum microplate reader.
Biofilm polysaccharide extraction and measurement: the cultures of B. pumilus HR10 wild strain and biofilm-deficient mutants were harvested at OD 600 = 1.0. For quantitative analysis of biofilm polysaccharides, 10 µL of liquid culture of each strain was used to inoculate into the well with 1 mL LB of 24-well microliter plates and was incubated without agitation at 30 • C for 48 h. The medium was carefully removed, and the wells were washed with sterilized water. Three milliliters of ethanol were added into each well. The biofilm polysaccharides were precipitated at 4 • C overnight. Subsequently, biofilm polysaccharides were collected by centrifugation (10,000 rpm, 4 • C, 10 min). The polysaccharide pellet was washed three times with ice-cold 75% ethanol and centrifuged. Similarly, two washing steps with distilled water were performed. The biofilm polysaccharides of the wild-type strain and mutants were dissolved, and the absorbances were measured as described above [8,31].

Quantitative Estimation of Biofilm Proteins
Standard curve of bovine serum albumin (BSA): the BSA solution (1 mL) concentrations were set as 0, 20, 40, 60, 80 and 100 µg/mL. Coomassie brilliant blue G250 (0.1 g) was dissolved in 1000 mL of distilled water containing 50 mL of 95% ethanol and 100 mL of 85% phosphoric acid. The BSA solution was stained with the dye liquid (5 mL) respectively for 5 min and then the absorbance at 595 nm was measured with a Multiskan Spectrum microplate reader.
Biofilm protein extraction and measurement: the biofilms of wild-type (WT) B. pumilus HR10 and biofilm-deficient mutants were harvested as described above. Ammonium sulfate was added to the supernatant liquid to a final concentration of 70%. The biofilm proteins were precipitated at Forests 2020, 11, 652 4 of 17 4 • C overnight. Subsequently, extracellular proteins were collected by centrifugation (10,000 rpm, 4 • C, 15 min). The protein pellet was washed three times with ice-cold 50% ethanol and centrifuged. Similarly, two washing steps with distilled water were performed. The extracellular proteins obtained from WT and mutants were dissolved, and the absorbances were measured as described above.

qRT-PCR Assay of Biofilm-Related Gene Expression
The biofilms of WT and biofilm mutants were collected after 12 h, 24 h, 48 h and 72 h of incubation. The biofilms were collected by centrifugation and ground into powder with liquid nitrogen. Then, 1 mL of TRIzol (Invitrogen, Waltham, MA, USA) was added to the powder (100 mg) of each sample. The 1.5 mL tubes with powder were shaken and the samples were mixed and then incubated for 5 min at 25 • C. The solution was extracted with 200 µL of trichloromethane (99%), mixed and centrifugated. The supernatant was collected and washed 2 times with 75% ethanol. The extracted RNA was dried and dissolved with RNase-Free H 2 O. The quantity and purity of isolated RNA were measured by ultraviolet absorbance NanoDrop 2000C at A260/280 (Thermo Scientific, Waltham, MA, USA) and assessed by electrophoresis on a 1% agarose gel. The first-strand cDNA was synthesized from 1 µg total RNA using the HiScript IIqRT SuperMix for qPCR (+gDNA wiper) (Vazyme, Nanjing, China), which could completely remove the residual genomic DNA in the RNA template.qRT-PCR was performed using a HiScript IIqRT reagent kit (Vazyme, Nanjing, China). The cDNA samples were diluted to 90 ng/µL. A solution of cDNA was prepared with 2 × ChamQ SYBR qPCR Master Mix. The reaction system (20 µL) included 10 µL of SYBR qPCR Master Mix (2×), 0.4 µL of forward primer (10 µM), 0.4 µL of reverse primer (10 µM), 2 µL of cDNA sample, and 7.2 µL of ddH 2 O. Thermal conditions were as follows: 30 s at 95 • C for initial denaturation and 40 cycles of 10 s at 95 • C, followed by 34 s at 60 • C. The relative expression levels of epsA (extracellular polysaccharide-related gene), tasA (extracellular protein-related gene), fliG (flagellum motility-related gene), and sfp (surfactin synthesis-related gene) were measured. The B. pumilus gyrB gene was used as an internal reference. Experiments were performed in triplicate with three biological replicates on an ABI Prism 7500 (Applied Biosystems, Foster City, CA, USA). The relative gene expression levels were analyzed using ABI Prism 7500 software and the 2 −∆∆Ct method. The primers used to amplify these genes are listed in Table 1.

Antifungal Activity Assay
R. solani plate was grown on PDA plates for the assay. The antifungal activity of the fermentation liquid of B. pumilus HR10 culture against R. solani was also measured. The cultures of B. pumilus HR10 WT and mutants were harvested to OD 600 = 1.0 in 20 mL LB respectively. Subsequently, fermentation liquid was collected by centrifugation (10,000 rpm, 10 min) and the further removing cells with a 0.22 µm filter. One milliliter of the fermentation culture of WT B. pumilus HR10 and mutants was added into a melted PDA plate respectively. The plates were incubated and observed, and the colony diameter for each treatment was recorded [6].

Cultivation of P. massoniana Seedlings
The 300 seeds of P. massoniana were surface-sterilized with 30% H 2 O 2 , after which they were placed in water agar medium with 1% agar. After growing to the cotyledon periods, 100 seedlings were grown in sterilized soil matrix for the root colonization assay. The remaining 200 seeds of P. massoniana were for reduced-plant-resistance assay and biocontrol assay against pine seedling damping-off disease.

Root Colonization Assay
The assay of root colonization was carried out following a described method with few modifications [32]. The overnight-grown bacterial cultures (OD 600 = 1.0) of wild-type B. pumilus HR10 and biofilm mutants were collected by centrifugation at 6000 rpm for 5 min and diluted to 1 × 10 8 CFU/mL with PBS for use. P. massoniana seeds were soaked in water for 18 h. The seeds were disinfected with 30% H 2 O 2 , washed with sterilized water and sown on medium with 1% agar. The 10-day-old seedlings were then transplanted into sterilized bottles containing sterile soil matrix. Forty-day-old seedlings grown in sterilized bottles were inoculated with five milliliters of diluted culture. Subsequently, 1 cm of root below the base of the stem was collected at 2, 4, 6, 8, 10 and 12 dpi. The roots were gently shaken in sterilized water. Crystal violet solution (1%) was used to stain the roots for 15 min, and the excess stain was rinsed off with sterilized water. The 5 mL of pure ethanol was used to extract the dye bound to the roots, and the absorbance at 570 nm was visualized with a Multiskan Spectrum microplate reader. Simultaneously, other gently shaken roots were placed into the tube with 5 mL sterilized water. Afterwards, the tubes were prepared with shaking at 200 rpm for 20 min, and 100 µL liquid was collected, diluted, and coated on an LB plate. The number of colonies was counted two days later. Subsequently, the roots with a length of 1 cm were taken at the time with most colonies. After drying with a critical point dryer (K850, Laughton, East Sussex, UK), colonization of the HR10 WT strain and mutants on the surface of pine roots was observed under a scanning electron microscope (Quanta 200, Hillsboro, OR, USA). Each treatment contains 3 seedlings in the assays of colony counting, crystal violet staining, and SEM observation.

Assay of Plant Defense Gene Expression
P. massoniana seeds were pregerminated as described above and sown in a sterilized soil matrix. Fifteen of the fifty-day-old P. massoniana seedlings grown in plastic holes containing sterilized soil matrix were inoculated with two milliliters of B. pumilus HR10 and mutant cultures diluted to 10 8 CFU/mL with sterilized water. After 4 days, the seedlings were carefully removed, flash-frozen in liquid nitrogen separately and ground. RNA extraction was performed using the Fast Plant RNA Kit for Polysaccharides and Polyphenolics-Rlch (Zomanbio, Beijing, China). The measurement of RNA quantity, removing gDNA, synthesis of cDNA, and qRT-PCR were conducted as described above (2.7). The relative expression levels of PR2, PAL (regulation of salicylic acid-related gene), COI (cytochrome oxidase-related gene), and GPX (stress tolerance-related gene) were measured. The P. massoniana splicing factor U2af gene was used as an internal actin reference. Experiments were performed in triplicate with three biological replicates on an ABI Prism 7500 (Applied Biosystems, Foster City, CA, USA). The relative gene expression levels were analyzed using ABI Prism 7500 software and the 2 −∆∆Ct method. The primers used to amplify these genes are listed in Table 2. Table 2. The qRT-PCR primer pairs of P. massoniana defense resistance-related genes.

Biocontrol Assay
P. massoniana seedlings were cultured as described in Section 2.11 and inoculated with two milliliters of solution (10 8 CFU/mL) of B. pumilus HR10 and mutants dissolved with sterilized water. Four days later, each seedling was inoculated with R. solani in the soil at the same time of sampling in Section 2.11 [5]. One week after inoculation with the pathogen, disease severity and biocontrol efficacy were recorded and calculated according to the following formula: Disease incidence (%) = (the number of damping-off diseased seedlings in each treatment/ total number of seedlings in each treatment) × 100.

Statistical Analyses
One-way analysis of variance (ANOVA) was carried out with SPSS software, and different letters indicate significant differences (p < 0.05).

Colony Morphology and Growth Curve of Cells and Spores
By observing the colony morphology, three biofilm-deficient mutants of B. pumilus HR10 wild-type were obtained. The results are shown in Figure 1A. The three-dimensional structure of the wild-type strain was the most obvious and ductile. The structure of the MA15 mutant was relatively the simplest. The structure of the MA1330 mutant had a slight depression in the middle and poor ductility. The structure of the MA4828 mutant was simpler than that of the wild-type strain and more complicated than that of the other two mutants. These results showed that the mutants had different degrees of deficiency in colony structure.
As shown in Figure 1B, the growth in broth cultures of the mutant cell culture showed no significant changes compared with that of wild-type strain. To better understand the growth curves of B. pumilus HR10 and biofilm mutants, we detected and compared the quantity of cells and spores. The test results indicated that the mutations had no significant effects on the growth curves of cells and spores of mutants ( Figure 1C). Figure 1B, the growth in broth cultures of the mutant cell culture showed no significant changes compared with that of wild-type strain. To better understand the growth curves of B. pumilus HR10 and biofilm mutants, we detected and compared the quantity of cells and spores. The test results indicated that the mutations had no significant effects on the growth curves of cells and spores of mutants ( Figure 1C).

The Biofilm Substance Contents and Swarming Motility of the Biofilm-Deficient Mutants Declined
After incubation for 12 h, 24 h, 48 h, and 72 h, B. pumilus HR10 formed wrinkly floating pellicles at the liquid-air interface of liquid cultures. In contrast, the mutants formed thinner pellicles in LB ( Figure 2A). All of the strains reached peak biofilm formation at 48 h, and MA15 formed the weakest biofilm. The biofilm of MA1330 degraded at the fastest rate in the later stages. Among the mutants, MA4828 had a relatively stable ability to form biofilm, and the degradation rate of MA4828 was relatively slow (Figure 2A). After staining the biofilm formed by B. pumilus HR10 and mutants with crystal violet dye, the absorbances at 570 nm were quantitatively measured. These results indicate that the abilities of B. pumilus HR10 and the biofilm mutants to form biofilms at the liquid-gas

The Biofilm Substance Contents and Swarming Motility of the Biofilm-Deficient Mutants Declined
After incubation for 12 h, 24 h, 48 h, and 72 h, B. pumilus HR10 formed wrinkly floating pellicles at the liquid-air interface of liquid cultures. In contrast, the mutants formed thinner pellicles in LB ( Figure 2A). All of the strains reached peak biofilm formation at 48 h, and MA15 formed the weakest biofilm. The biofilm of MA1330 degraded at the fastest rate in the later stages. Among the mutants, MA4828 had a relatively stable ability to form biofilm, and the degradation rate of MA4828 was relatively slow (Figure 2A). After staining the biofilm formed by B. pumilus HR10 and mutants with crystal violet dye, the absorbances at 570 nm were quantitatively measured. These results indicate that the abilities of B. pumilus HR10 and the biofilm mutants to form biofilms at the liquid-gas interface are consistent with the changes in colony morphology at the solid-gas surface. The more complex the fold on the solid-gas surface is, the denser and more complete the biofilm structure on the liquid-gas surface becomes.
Biofilm polysaccharides are the key components of the three-dimensional structure of biofilms. Extracellular proteins help to maintain the hydrophobicity and surface tension of biofilms [20]. The polysaccharide contents of the mutants were significantly less than that of the wild-type strain, indicating that polysaccharides are an indispensable component for biofilm formation by B. pumilus HR10 ( Figure 2B). The biofilm protein content of MA15 was significantly less than that of the wild-type strain, while the contents of MA1330 and MA4828 were almost the same as that of the wild-type strain.
Swarming motility has a certain effect on the social movement of bacteria and the speed of biofilm formation [15,37]. To analyze swarming motility, cell suspensions were spotted at the center of the swarming plates, and the expansion of cells was photographed ( Figure 2C). As shown, the wild-type strain showed excellent swarming motility and the plate was nearly fully covered with the swarming cells. The biofilm mutants exhibited a markedly weakened swarming motility compared to the wild-type strain. Based on the above results, it is indicated that the weakened swarming motility of the mutants restricted the adhesion of bacteria and biofilm formation. biofilm formation [15,37]. To analyze swarming motility, cell suspensions were spotted at the center of the swarming plates, and the expansion of cells was photographed ( Figure 2C). As shown, the wild-type strain showed excellent swarming motility and the plate was nearly fully covered with the swarming cells. The biofilm mutants exhibited a markedly weakened swarming motility compared to the wild-type strain. Based on the above results, it is indicated that the weakened swarming motility of the mutants restricted the adhesion of bacteria and biofilm formation.

Biofilm-Related Genes of HR10 Wild-Type and Mutants Showed Different Expression Levels
To further examine the effects of extracellular substances and swarming motility on biofilm formation, the transcription levels of biofilm-related genes were evaluated after incubation for 12 h, 24 h, 48 h, and 72 h. The epsA gene, which is a transcriptional regulator of EPS synthesis, of the four strains increased to the highest value at 48 h post-incubation ( Figure 3A). The expression of the epsA gene in MA4828 was maintained at a high level compared to that in MA15 and MA1330. No significant changes in the tasA expression in the mutants were measured at different times, except for the difference in MA15 at 12 h and 24 h post-incubation ( Figure 3B). The expression levels of fliG (movement-related gene) and sfp (surfactin synthesis-related gene) were maintained at the highest levels at 24 h, 48 h, and 72 h in WT, respectively, but were the lowest levels in MA15 ( Figure 3C,D). strains increased to the highest value at 48 h post-incubation ( Figure 3A). The expression of the epsA gene in MA4828 was maintained at a high level compared to that in MA15 and MA1330. No significant changes in the tasA expression in the mutants were measured at different times, except for the difference in MA15 at 12 h and 24 h post-incubation ( Figure 3B). The expression levels of fliG (movement-related gene) and sfp (surfactin synthesis-related gene) were maintained at the highest levels at 24 h, 48 h, and 72 h in WT, respectively, but were the lowest levels in MA15 ( Figure 3C,D).

Deficiency in Biofilm Formation Reduced Root Colonization on P. massoniana Seedlings
To investigate the root colonization on P. massoniana seedlings of strains, the population dynamics of WT and mutants on roots were evaluated over time. The results showed that the colonization trends of strains were consistent with their biofilm formation. The population of WT was higher at each testing time point, where the highest level was reached at 10 days post inoculation (DPI). In contrast, the populations of the mutants, especially MA15 and MA1330 at 6-12 dpi, were obviously lower than that of WT. The quantitative results of crystal violet dye staining ( Figure 4B,C) also showed that the absorbances of WT treatment were the highest at 4-12 dpi, while those of MA15 treatment were significantly reduced. Similar results were obtained in the SEM observations. In comparison with WT, the mutants showed different levels of reduced colonization on the roots of P. massoniana. In general, the deficient biofilm formation by B. pumilus HR10 leads to decreased colonization of the roots of P. massoniana seedlings.
obviously lower than that of WT. The quantitative results of crystal violet dye staining ( Figure 4B,C) also showed that the absorbances of WT treatment were the highest at 4-12 dpi, while those of MA15 treatment were significantly reduced. Similar results were obtained in the SEM observations. In comparison with WT, the mutants showed different levels of reduced colonization on the roots of P. massoniana. In general, the deficient biofilm formation by B. pumilus HR10 leads to decreased colonization of the roots of P. massoniana seedlings.

Mutants Exhibited Different Inhibitory Activities against R. Solani
A test of the inhibition activity of the fermentation broth of the strains against R. solani was performed. The treatments with additional fermentation broth significantly inhibited the growth of R. solani in comparison with the control group. Moreover, the MA15 treatment exhibited weaker inhibitory activity than the other treatments ( Figure 5A,B). A test of the inhibition activity of the fermentation broth of the strains against R. solani was performed. The treatments with additional fermentation broth significantly inhibited the growth of R. solani in comparison with the control group. Moreover, the MA15 treatment exhibited weaker inhibitory activity than the other treatments ( Figure 5A,B).

Biofilm Formation Contributes to Inducing the Systemic Resistance of Plants
To investigate whether the strains induce plant systemic resistance, the transcription levels of PR2, PAL, COI, and GPX were measured and evaluated. The seedlings not inoculated with bacteria served as control (CK) plants. The seedlings inoculated with WT B. pumilus HR10 or biofilm-deficient mutants exhibited high expression levels of PR2 (4.2-fold, 1.8-fold, 2.5-fold and 1.9-fold compared to the expression of PR2 in the CK, respectively), while the expression of PAL, COI, and GPX showed no significant differences ( Figure 6). The results indicate that inoculation with WT B. pumilus HR10 or mutants contributes to inducing P. massoniana resistance to some extent.

Biofilm Formation Contributes to Inducing the Systemic Resistance of Plants
To investigate whether the strains induce plant systemic resistance, the transcription levels of PR2, PAL, COI, and GPX were measured and evaluated. The seedlings not inoculated with bacteria served as control (CK) plants. The seedlings inoculated with WT B. pumilus HR10 or biofilm-deficient mutants exhibited high expression levels of PR2 (4.2-fold, 1.8-fold, 2.5-fold and 1.9-fold compared to the expression of PR2 in the CK, respectively), while the expression of PAL, COI, and GPX showed no significant differences ( Figure 6). The results indicate that inoculation with WT B. pumilus HR10 or mutants contributes to inducing P. massoniana resistance to some extent.

Biofilm Formation and Inhibitory Activity are Both Required for Biocontrol against Pine Seedling Damping-off Disease
We observed that the disease incidence of seedlings in the CK with R. solani reached 81.3%, and the treatment exhibited a decreasing trend. The treatment with the wild-type strain decreased disease incidence to 18.8% with excellent biocontrol efficacy. The disease incidence following MA1330 and MA4828 inoculation was 31.25%, while that of MA15 inoculation was lower (Table 3). In brief, the biofilm-deficient mutants are inefficient in the biocontrol of pine seedling damping-off disease.

Biofilm Formation and Inhibitory Activity Are Both Required for Biocontrol against Pine Seedling Damping-Off Disease
We observed that the disease incidence of seedlings in the CK with R. solani reached 81.3%, and the treatment exhibited a decreasing trend. The treatment with the wild-type strain decreased disease incidence to 18.8% with excellent biocontrol efficacy. The disease incidence following MA1330 and MA4828 inoculation was 31.25%, while that of MA15 inoculation was lower (Table 3). In brief, the biofilm-deficient mutants are inefficient in the biocontrol of pine seedling damping-off disease.

Discussion
Bacillus spp. have been identified as biological control agents against fungal and bacterial diseases. Bacillus subtilis and Bacillus amyloliquefaciens exert growth-promoting and biocontrol functions by competition, antibiosis, and the induction of plant systemic resistance [38][39][40]. Successful colonization of PGPR is considered a crucial step for biocontrol [41,42]. Swarming motility and biofilm formation are major adaptive strategies for Bacillus in colonization [43,44]. The results of this study demonstrated that B. pumilus HR10 produces a mature biofilm structure, which contributes to its colonization on the roots of pine seedlings. The biocontrol efficacy of B. pumilus HR10 was proposed to be the result of multiple causes. This study explains the mechanisms of the biocontrol efficacy from many aspects.
The biofilm-deficient mutants of B. pumilus HR10 were screened by colony phenotype. The aggregate structure of three mutants, such as the middle or colony edge, exhibited different degrees of defects. The WT B. pumilus HR10 strain was found to be most complicated. The defects of MA15 and MA1330 were more significant than those of MA4828. The mutants formed fewer floating pellicles at the liquid air interface of liquid cultures compared to WT, consistent with their simpler colony morphology than that of WT on the solid-air interface. MA15 appeared the sparsest, and MA1330 degraded the quickest. The results suggested that the defectiveness of biofilm substances and swarming motility result in defective pellicle formation and rapid degradation.
Surfactin, a typical lipopeptide that antagonizes fungal and bacterial pathogens, is also known to be involved in swarming motility and biofilm formation [20,44]. For example, the inability of B. subtilis 9407 to produce surfactin led to a reduction in antagonistic activity, swarming motility and biofilm formation [4]. In accordance with qRT-PCR measurements, we found that the expression levels of sfp at 48 h in MA15 and MA1330 were lower than that of WT, indicating that surfactin may play an important role in the swarming motility and biofilm formation of B.pumilus HR10 (Figure 3).
Biofilm formation facilitates the colonization of Bacillus on the roots or leaves of plants [45]. Similarly, in our study, WT B. pumilus HR10 effectively colonized on the roots of Pinus massoniana seedlings, while the biofilm-deficient mutants showed significantly decreased in the colonization on the roots (Figure 4). These results demonstrated that biofilm formation by B. pumilus HR10 contributes to efficient colonization.
The biological control of Bacillus depends on its inhibitory activity against R. solani to a large extent [3]. The inhibitory activities of MA1330 and MA4828 showed no significant differences compared with that of WT, and MA15 exhibited the weakest activity. Surfactin is a major lipopeptide compound that suppresses pathogens [4,20]. According to the results of Figures 3D and 5, we proposed that sfp expression of B. pumilus HR10 is related to biocontrol against R. solani.
Previous studies have reported that biofilm formation is associated with rhizobacteria-induced systemic resistance [29,46]. However, the function of biofilm formation in the induction of systemic resistance is not very evident. In this study, we conducted qRT-PCR measurements of defense-related genes in P. massoniana seedlings in response to inoculation with WT or mutants. We found that the expression levels of PR2 exhibited obvious differences, whereas the expression levels of other genes were not significant. The typical gene of the salicylic acid (SA) signaling pathway of seedlings, PR2, exhibited high levels when seedlings were inoculated with WT HR10 or mutants. The PR2 transcription level increased approximately fivefold after inoculation with WT. These results demonstrated that biofilm formation by B. pumilus HR10 contributes its ability to induce seedling resistance by promoting the SA signaling pathway of P. massoniana seedlings. Deficiency in biofilm formation of the mutants significantly weakened the induced seedling resistance.
There were obvious differences in the effects of WT B. pumilus HR10 and mutants on the biocontrol against pine seedling damping-off disease. The biocontrol results suggested that biofilm formation plays a significant role during the biocontrol process mediated by the WT and mutants. This finding was consistent with a previous study by Luo et al., who reported that biofilm formation further strengthens the colonization of B. subtilis 916 and biocontrol efficacy against rice sheath blight [20]. Indeed, chemotaxis of PGPR and subsequent biofilm formation are very important for root colonization and for providing beneficial functions for plants and PGPR [47,48]. Interestingly, the biocontrol effect of MA1330, which produced defective biofilm, was equal to that of MA4828. This may be due to MA1330 triggering higher expression of SA signaling pathway-related genes, which is necessary for the resistance of plants and biocontrol of PGPR.
Overall, the results reported here indicate that the antagonistic activity of B. pumilus HR10 suppresses R. solani, and biofilm formation also contributes to the promotion of colonization, induction of plant resistance and biocontrol against seedling damping-off disease. The results of our study may provide a theoretical basis for the field application of B. pumilus HR10, which is beneficial for promoting the growth of pine trees, symbiosis of mycorrhizae and biocontrol against pine diseases. A major challenge for the future will be to explore the regulatory pathway of biofilm formation of B. pumilus HR10 and the role of lipopeptides in the formation process. It will be interesting and necessary to determine how multiple mechanisms are synergistically involved in regulating biocontrol.

Conclusions
There are few studies on the relationship between biofilm formation of Bacillus pumilus and its biocontrol process. B. pumilus HR10 has high-efficiency functions such as promoting the growth of plants, assisting mycorrhizal-formation and inhibiting pathogens. In this research, biofilm related features of the HR10 strain were analyzed qualitatively or quantitatively. It was also discussed in terms of colonization on the root of P. massoniana seedlings and the induction of plant system resistance, which were more in line with the status of PGPR in the natural environment. The results showed that biofilm formation of B. pumilus HR10 plays an important role in the prevention of pine seedlings damping-off disease, and it affects biocontrol with induced resistance synergistically. It provides a basis for better exerting the application potential of B. pumilus HR10 in the field to promote growth and disease resistance.

Conflicts of Interest:
The authors declare no conflict of interest.