Perception of Biocontrol Potential of Bacillus inaquosorum KR2-7 against Tomato Fusarium Wilt through Merging Genome Mining with Chemical Analysis

Simple Summary Bacillus is a bacterial genus that is widely used as a promising alternative to chemical pesticides due to its protective activity toward economically important plant pathogens. Fusarium wilt of tomato is a serious fungal disease limiting tomato production worldwide. Recently, the newly isolated B. inaquosorum strain KR2-7 considerably suppressed Fusarium wilt of tomato plants. The present study was performed to perceive potential direct and indirect biocontrol mechanisms implemented by KR2-7 against this disease through genome and chemical analysis. The potential direct biocontrol mechanisms of KR2-7 were determined through the identification of genes involved in the synthesis of antibiotically active compounds suppressing tomato Fusarium wilt. Furthermore, the indirect mechanisms of this bacterium were perceived through recognizing genes that contributed to the resource acquisition or modulation of plant hormone levels. This is the first study that aimed at the modes of actions of B. inaquosorum against Fusarium wilt of tomatoes and the results strongly indicate that strain KR2-7 could be a good candidate for microbial biopesticide formulations to be used for biological control of plant diseases and plant growth promotion. Abstract Tomato Fusarium wilt, caused by Fusarium oxysporum f. sp. lycopersici (Fol), is a destructive disease that threatens the agricultural production of tomatoes. In the present study, the biocontrol potential of strain KR2-7 against Fol was investigated through integrated genome mining and chemical analysis. Strain KR2-7 was identified as B. inaquosorum based on phylogenetic analysis. Through the genome mining of strain KR2-7, we identified nine antifungal and antibacterial compound biosynthetic gene clusters (BGCs) including fengycin, surfactin and Bacillomycin F, bacillaene, macrolactin, sporulation killing factor (skf), subtilosin A, bacilysin, and bacillibactin. The corresponding compounds were confirmed through MALDI-TOF-MS chemical analysis. The gene/gene clusters involved in plant colonization, plant growth promotion, and induced systemic resistance were also identified in the KR2-7 genome, and their related secondary metabolites were detected. In light of these results, the biocontrol potential of strain KR2-7 against tomato Fusarium wilt was identified. This study highlights the potential to use strain KR2-7 as a plant-growth promotion agent.


Introduction
Tomato Fusarium wilt, caused by Fusarium oxysporum f. sp. lycopersici (Fol), is one of the most destructive diseases, causing a considerable loss in the production of both field and greenhouse tomatoes worldwide [1]. Since Fusarium wilt is a difficult disease to control [2][3][4], control strategies including physical and cultural methods, chemical fungicides treatment, and the cultivation of resistance tomato cultivars [5] achieved limited efficacy [6]. In addition, excessive usage of agrochemicals imposed serious negative impacts on the environment, causing the pollution of soil and groundwater reservoirs, an accumulation of chemical residues in the food chain, an emergence of pesticide-resistance pathogens, and health hazards [7]. As a result, biocontrol microbes have been suggested as a promising alternative to agrochemicals in plant disease control. Numerous biocontrol microbes, especially Bacillus strains, have been commercially developed as biopesticides and biofertilizers worldwide [7].
Biocontrol microbes protect the crops from an invasion of phytopathogens via (1) direct modes of action, e.g., the antibiosis and production of antimicrobial secondary metabolites [8,9]; and (2), the indirect modes of action, including induced systemic resistance (ISR) and the competition for nutrients and space [10,11]. The investigation of biocontrol microbes through conventional genetic and biochemical approaches could not unveil the full potential of these microbes due to the absence of appropriate natural triggers or stress signals under laboratory conditions [12]. With the development of high-throughput DNA sequencing technologies and genome mining, along with MS-based analytical methods (e.g., GC/LC-MS, LC-ESI-MS, and MALDI-TOF-MS), more potential biocontrol microbes can be revealed. For instance, the Bacillus amyloliquefaciens FZB42 genome contains nine giant gene clusters synthesizing secondary metabolites which are involved in the suppression of soil-born plant pathogens. Several gene/gene clusters are implicated in swarming motility, plant colonization, biofilm formation, and the synthesis of plant growth-promoting volatile compounds and hormones [13]. A wide range of extracellular proteins and phytase in the FZB42 secretome were detected through two-dimensional electrophoresis, MALDI-TOF-MS, and the proteomics approach, indicating this strain can grow on the plant's surface and supply phosphorus for the plant under phosphorus starvation. Additionally, four members of the macrolactin family were identified in an FZB42 culture filtrate by combining mass spectrometric and ultraviolet-visible data which perfectly agree with the overall structure of the macrolactin gene cluster found in the FZB42 genome [13]. Recently, the genome analysis of plant-protecting bacterium B. velezensis 9D-6 demonstrated that this strain can synthesize 13 secondary metabolites, of which surfacin B and surfactin C were detected as antimicrobial compounds against Clavibacter michiganensis through LC-MS/MS [14]. Furthermore, the genome mining of B. inaquosorum strain HU Biol-II revealed that this bacterial genome contains eight bioactive metabolite clusters and the production of seven metabolites was confirmed through HPLC MS/MS [15].
In our previous study, the B. inaquosorum strain KR2-7 was isolated from the rhizosphere soil of the tomato (Solanum lycopersicum) and was introduced as a highly effective biocontrol agent against Fol with a biocontrol efficiency of 80% under greenhouse conditions [16]. To better understand the biocontrol mechanisms of strain KR2-7 against Fol, whole-genome sequencing was conducted to identify putative gene clusters for secondary metabolites biosynthesis and to characterize gene/gene clusters involved in plant colonization, plant growth promotion, and induced systemic resistance (ISR). Moreover, secondary metabolites and other compounds related to identified BGCs and gene/gene clusters were detected using MALDI-TOF-MS analysis to confirm the results of genome mining.

Strains and Culture Conditions
The fungal pathogen Fol strain Fo-To-S-V-1 used in this study was obtained from the culture collection from the Iranian Research Institute of Plant Protection. The fungus Biology 2022, 11, 137 3 of 26 was maintained on a potato dextrose agar (PDA, Merck, Germany) slant at 4 • C and was sub-cultured onto a fresh PDA plate at 27 • C for 7 days for further tests.
Strain KR2-7 was maintained on nutrient agar (NA, Merck, Germany; with a 0.3% beef extract, 0.5% peptone, and 1.5% agar) plate with a periodic transfer to a fresh medium. For long-term storage, it was kept at −80 • C in lysogeny broth (LB, Merck, Germany) with 20% glycerol (v/v).

Dual Culture Assay
In order to investigate the antagonism efficiency of strain KR2-7 against various tomato pathogens, five destructive fungal pathogens, including Alternaria alternata f. sp. lycopersici, Athelia rolfsii, Botrytis cinerea, Rhizoctonia solani, and Verticillium albo-atrum, were selected. The antifungal activity of strain KR2-7 against each pathogen was evaluated through a dual culture assay in three replications. In the dual culture assay, strain KR2-7 was simultaneously cultured 3cm apart from the 5-mm plug of a pathogen in a 9 cm PDA plate. The control plate was inoculated only with the pathogen. Plates were incubated at 27 • C. The fungal growth was checked daily by measuring the diameter of the colony for a period of three days. The percentage of fungal growth inhibition (PFGI) was calculated by the formula (1) developed by Skidmore and Dickinson [17], where R1 is the maximum radius of the growing fungal colony in the control plate, and R2 is the radius of the fungal colony that grew in the presence of strain KR2-7:

MALDI-TOF-MS Analysis of KR2-7 Secondary Metabolites
The secondary metabolite analysis was performed from the whole-cell surface extract of bacterium obtained during the dual culture of KR2-7 and Fol. The bacterial surface extract was prepared according to the methodology described by Vater et al. [18]. The Dual culture was done on potato dextrose agar (PDA) instead of Landy agar. Strain KR2-7 was streaked on one side of the plate and a 5 mm plug of Fol was placed on the opposite side simultaneously and incubated at 27 • C. After 24 h, two loops of bacterial cells from the interface of the bacterium-fungus in the inhibition zone were suspended in 500 µL of 70% acetonitrile with 0.1% trifluoroacetic acid for 2 min. The suspension was gently vortexed to produce a homogenized suspension. The bacterial cells were pelleted by centrifuging at 5000 rpm for 10 min. The cell-free supernatant was transferred to a new microcentrifuge tube and stored at 4 • C for further analysis. One microliter of supernatant liquid was spotted onto the target of the mass spectrometer with an equal volume of α-cyano-4hydroxycinnamic acid (CHCA) matrix and was air-dried. The sample mass fingerprints were obtained using an ultrafleXtreme MALDI-TOF/TOF-MS (Bruker, Billerica, MA, USA) within a mass range of 100-3000 Da. The MALDI-TOF-MS analysis was performed at the school of life sciences, Chinese University of Hong Kong (CUHK), Hong Kong. The whole-cell surface extract of strain KR2-7 grown on a potato dextrose agar was used as a control.

Genome Sequencing, Assembly, and Annotation
The genomic DNA of strain KR2-7 was extracted using a commercial DNA extraction kit (Thermo Fisher Scientific, Waltham, MA, USA). The whole-genome sequencing was performed using the Illumina HiSeq 4000 and PacBio RSII platforms (BGI, Shenzhen, China). The quality control of raw sequences was performed by FastQC v0.11.9 and the de novo assembly was done using SPAdes v3.14.1. The genome was annotated using the NCBI Prokaryotic Genomes Automatic Annotation Pipeline (PGAAP), and Bacterial Annotation System (BASys) webserver. The proteome of KR2-7 was subjected to BLASTP against the Cluster of Orthologous Group (COGs) database at E-value < 1 × 10 −5 to identify the Cluster of Orthologous Groups (COGs) [19].

Genome Phylogeny
In this study, 32 Bacillus strains belonging to various species were selected among those recorded in the NCBI GenBank database. For all the selected strains, the nucleotide and the corresponding amino acid sequences were retrieved from the GenBank database. Whole-genome alignments were performed using REALPHY (http://realphy.unibas.ch (accessed on 22 December 2021); [20]) and the phylogenetic tree was constructed using the MEGA v. 7 [21] by maximum likelihood method [22], with evolutionary distances computed using the general time-reversible model [23]. Branch validity was evaluated by the bootstrap test with 1000 replications. The average nucleotide identity (ANI) values of selected Bacillus strains were calculated using the server EzBioCloud (http://www. ezbiocloud.net/tools/ani (accessed on 21 August 2021); [24]). According to the algorithm developed by Goris et al. [25], 95∼96% cut-off value was used for the species boundary [26]. The web-based DSMZ service (http://ggdc.dsmz.de (accessed on 7 January 2022); [27]) with 70% species and sub-species cut-off was used to estimate the in silico genome-togenome distance values for the selected strains.

Pathway Analysis
The annotated genome was analyzed using KEGG (Kyoto Encyclopedia of Genes and Genomes) to determine the existing pathways, which were then manually validated through matching the assigned gene functions to the corresponding KEGG pathway.

Genome-Wide Identification of Secondary Metabolite Biosynthesis Gene Clusters
The antibiotics and secondary metabolite analysis shell (antiSMASH) is a comprehensive resource that allows the automatic genome-wide identification and analysis of secondary metabolite biosynthesis gene clusters in bacterial and fungal genomes [28,29]. Thereby, the KR2-7 genome was submitted to the antiSMASH web server (https://antismash. secondarymetabolites.org) (accessed on 22 December 2021) to detect the putative BGCs for secondary metabolites. Each identified BGC in the KR2-7 genome was aligned against corresponding BGC in B. subtilis strain 168 and B. amyloliquefacience stain FZB42 using Geneious Prime v.2021.2.2. to find out the BGCs similarity between KR2-7, 168 and, FZB42.

General Genomic Features of Strain KR2-7
The assembled genome of B. inaquosorum KR2-7 contained 4 contigs, with an N50 of 2,144,057 bp and 700X sequence coverage. The KR2-7 genome was obtained with a length of 4,248,657 bp, the G+C content of 43.1%, and 4265 predicted genes consisting of 4017 protein-coding genes, 50 rRNA genes, and 83 tRNA genes. Interestingly, strain KR2-7 possesses the larger number of genes contributing to amino acid transport and metabolism (322 genes), carbohydrate transport and metabolism (278 genes), inorganic ion transport and metabolism (200 genes), and secondary metabolites biosynthesis, transport and catabolism (76) compared to the reputable biocontrol agent B. velezensis strain FZB42 ( Figure S1; Table S1). Therefore, the genome content of KR2-7 indicates that the strain has considerable potential as a biocontrol agent. The genome sequence of B. inaquosorum KR2-7 was deposited in NCBI GenBank under the accession number QZDE00000000.2.

Genome Phylogeny
The genomes of 31 Bacillus strains were selected for aligning with the KR2-7 genome and phylogenomic analysis. The selected strains and their corresponding genome sequence accession numbers were presented in Table 1. Selected Bacillus strains were accurately distributed on branches of the maximum likelihood phylogenomic tree ( Figure 1).   Moreover, closely related Bacillus species such as B. amyloliquefaciens and B. velezensis were distributed on the same branch ( Figure 1). The genome-based phylogeny approaches well recognized B. methylotrophicus, B. amyloliquefaciens subsp. plantarum, and B. oryzicola as the heterotypic synonyms of B. velezensis [30]. Recently, three subspecies of Bacillus subtilis, including B. subtilis subsp. inaquosorum, B. subtilis subsp. Spizizenii, and B. subtilis subsp. stercoris were promoted to species status through comparative genomics. Each subspecies encompasses unique bioactive secondary metabolite genes which cause the unique phenotypes [31]. According to REALPHY results, strain KR2-7 was identified as B. inaquosorum, owing to being placed within the B. subtilis branch close to B. subtilis subsp. inaquosorum strain KCTC 13429 and strain DE111 in the phylogenomic tree ( Figure 1). Notably, the results of ANI and GGDC analysis were consistent with REALPHY results as the KR2-7 genome displayed the highest ANI values (99.26%) and the lowest GGDC values (0.0075) with respect to the genome of strain KCTC 13429 (Table 1). Interestingly, the phylogeny analysis of several B. amyloliquefaciens strains based on core-genome was consistent with the ANI and GGDC values [32]. Altogether, the aforesaid genome-based phylogeny approaches identified strain KR2-7 as Bacillus inaquosorum.

Secondary Metabolites Biosynthetic Gene Clusters in the KR2-7 Genome
Genome mining of the strain KR2-7 revealed that more than 700 kb (i.e., nearly 17% of the genome) is devoted to 13 putative BGCs. Of the 13 found BGCs, nine were identified to contain one polyketide synthase (PKS) for macrolactin; five non-ribosomal peptide synthetases (NRPSs) for bacillibactin, bacillomycin F, bacilysin, fengycin, and surfactin; one PKS-NRPS hybrid synthetases (PKS-NRPS) for bacillaene; one thiopeptide synthase for subtilosin A, and one head-to-tail cyclised peptide for the sporulation killing factor. The nine annotated BGCs encode secondary metabolites which contribute to plant growth promotion through the fungal/bacterial pathogen suppression, ISR, nutrient uptake, and plant colonization ( Table 2) [33][34][35][36]. The distribution of identified BGCs within the KR2-7 genome underlies its vigorous potential in plant disease biocontrol application [16]. The coding genes of secondary metabolites in KR2-7 were different from those in B. velezensis FZB42, while these genes showed more similarity with those in B. subtilis 168 (Table 2). Interestingly, the BGC of bacillomycin F in KR2-7 was absent in B. subtilis 168 and B. velezensis FZB42 (Table 2) as this gene cluster conserved in B. inaquosorum [31]. Moreover, the KR2-7 genome contains four unannotated BGCs (data not shown) which showed less similarity to compounds listed in the MIBiG database.

Antifungal Secondary Metabolites Production in Strain KR2-7
The KR2-7 genome mining showed that this strain harbors three BGCs with antifungal function, including fengycin, surfacing, and bacillomycin F (a variant of iturin) belonging to Bacillus cyclic-lipopeptides (CLPs). Bacillus CLPs represented the powerful fungitoxicity properties by interfering with cell membrane integrity, permeabilizing the cell membrane, and perturbing membrane osmotic balance due to the formation of ion-conducting pores [37].
Strain KR2-7 not only suppressed the Fusarium wilt of tomato caused by Fusarium oxysporum f. sp. lycopersici [16] but also showed a broad-spectrum antifungal activity towards various phytopathogenic fungi including Alternaria alternata f. sp. lycopersici, Athelia rolfsii, Botrytis cinerea, Rhizoctonia solani, and Verticillium albo-atrum ( Figure 2).  Fengycin (plipastatin), the powerful fungitoxic compound-especially against filamentous fungi [37]-is synthesized by NRPS and encoded by a 39302 bp gene cluster with five genes including ppsA-E in KR2-7, which showed a 92% and 72.05% similarity to the fengycin gene cluster of B. subtilis 168 and B. amyloliquefacience FZB42, respectively ( Table 2). The first three genes (ppsABC) each encode two amino acid modules. The fourth gene (ppsD) encodes three amino acid modules, and the last gene (ppsE) encodes one amino acid module ( Figure 3). Ions of m/z values 1471.8, 1485.7, 1487.9, 1499.9, 1501.9, 1513.8, 1515.9, 1527.8 1529.9 and, 1543.8 were observed in a whole-cell surface extract of KR2-7 grown on a dual culture plate (thereafter, dual culture cell extract) and assigned to C15 to C18 fengycin homologues while in the whole-cell surface extract of KR2-7 grown on the control plate (thereafter, control cell extract) only four aforesaid peaks (m/z 1501.9, 1515.9, 1529.9 and, 1543.8) were detected (Table 3, Figure S2). The result indicated that the KR2-7 strain secreted various fengycin homologues to inhibit the growth of Fol.    More strikingly, a 37074 bp gene cluster encoding bacillomycin F was also identified immediately downstream of the fengycin gene cluster of KR2-7 ( Figure 4). The bacillomycin F is one of seven main variants within the iturin family [40] encoded by a gene cluster consisting of four genes designated ituD, ituA, ituB and, ituC. The gene cluster code a cyclic heptapeptide in which the first three amino acids are shared among iturin family members, whereas the remaining four amino acids are conserved in B. inaquosorum [31]. Furthermore, iturins are characterized by a heptapeptide of α-amino acids attached to a β-amino fatty acid chain with a length of 14 to 17 carbons [37]. They possess potent antifungal activity against a wide variety of fungi and yeast, but bounded antibacterial and no antiviral actions [41][42][43]. Furthermore, these molecules also showed strong haemolytic activity, which limits their clinical use [44]. The antifungal mechanism of iturins launches by their interaction with the target cell membrane and osmotic perturbation of the membrane, owing to the formation of ion-conducting pores. Subsequently, the change in the permeability of a membrane is conducive to the release of biomolecules, such as proteins, nucleotides, and lipids from cells, which ultimately causes cell death [44,45]. In the dual culture cell extract of KR2-7, six mass peaks assigned to C16, C18 and C19 forms of iturin were observed while they were absent in the control cell extract of KR2-7 (Table 4, Figure S3). This result indicated that strain KR2-7 produced different variants of iturins to limit the growth of Fol hyphae.      [18] Similar to fengycin, surfactin was synthesized by NRPS and encoded by a srf gene cluster that spans 26073 bp in the KR2-7 genome. The gene cluster harbors four genes (srf AA-AD) and showed a 92.20% and 74.65% similarity to those of B. subtilis 168 and B. amyloliquefacience FZB42, respectively. The product of the srf gene cluster is a linear array of seven modules, six of which are encoded by srf AA and srf AB genes and the last module is encoded by a srf AC gene ( Figure 5). The fourth gene (srf AD) encodes thioesterase/acyltransferase (Te/At-domain) which triggers surfactin biosynthesis [37]. Hence, the sfp gene encodes an essential enzyme (phosphopantetheinyl transferase) for the non-ribosomal synthesis of lipopeptides and the synthesis of polyketides. The regulatory gene yczE encoding an integral membrane protein was detected within the KR2-7 genome ( Figure 5). Surfactin enables bacteria cells to interact with plant cells as a bacterial elicitor for stimulating ISR [37], especially through the activation of jasmonate-and salicylic acid-dependent signaling pathways [46]. Several studies indicated the ISR-elicitor role of surfactin against phytopathogens in various host plants, e.g., tomato [47], wheat [48], citrus fruit [49], lettuce [50], and grapevine [51]. Comparing the MALDI-TOF mass spectra of KR2-7 grown on a PDA control and dual culture revealed that surfactin contributed to the suppression of Fol as eight mass peaks assigned to C13, C14 and C15 surfactin homologs were detected in dual culture cell extracts, while only four of which were observed in the control cell extract (Table 5, Figure S3). Furthermore, the bacterium produced more surfactin to suppress Fol. Additionally, C14 and C15 surfactin tend to stimulate stronger ISR rather than those with shorter chain lengths [47]. Moreover, suppression of taxonomically diverse fungal pathogens including Fusarium oxysporum, F. moniliforme, F. solani, F. verticillioides, Magnaporthe grisea, Saccharicola bicolor, Cochliobolus hawaiiensis, and Alternaria alternata by the surfactin family demonstrated that surfactins are strong fungitoxic compounds [52][53][54][55].

Antibacterial Secondary Metabolites Production in Strain KR2-7
The KR2-7 genome contained six BGCs coding for antibacterial compounds including bacillaene and macrolactin, sporulation killing factor (skf), subtilosin A, bacilysin, and surfactin. Several studies on surfactin and its isoforms proved that these metabolites played a major role in combating bacterial plant diseases, such as fruit bloch caused by Acidovorax citrulli in melon [56], tomato wilt caused by Ralstonia solanacearum [57], and root infection by Pseudomonas syringae in Arabidopsis [58]. Moreover, surfactin produced by B. subtilis R14 exhibited pronounced antagonistic efficacy against several multidrug-resistance bacterial strains of Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Enteroccoccus faecalis [59].
Bacillaene is a polyketide known as a selective bacteriostatic agent that inhibits prokaryotic, not eukaryotic growth by disrupting protein synthesis [60]. Its antimicrobial efficacy against various bacteria (Myxococcus xanthus and Staphylococcus aureus) and fungi (Fusarium spp) have been reported [60][61][62]. In the KR2-7 genome, bacillaene was synthesized by a PKS/NRPS hybrid pathway and encoded by a giant pks gene cluster (76.355 Kbp) containing 16 genes (pksA-S and acpK) showing an 89.63% and 75.25% similarity to those of B. subtlis 168 and B. velezensis FZB42, respectively (Table 2, Figure 6B). Another polyketide, macrolactin, can be encoded by a 54.225 kbp gene cluster in strain KR2-7 and showed a 74.12% similarity to the mln cluster of B. velezensis FZB42 (Table 2, Figure 6A). Macrolactins are a large class of macrolide antibiotics that inhibited the growth of several bacteria, including Ralstonia solanacearum, Staphylococcus aureus, and Burkholderia epacian [63,64]  Bacilysin (also known as tetaine) is a dipeptide suppressing a wide variety of destructive phytopathogenic bacteria, e.g., Erwinia amylovora, Xanthomonas oryzae pv. oryzae, X. oryzae pv. Oryzicola, and Clavibacter michiganense subsp. sepedonicum [65][66][67]. This bactericidal property is due to the inhibition of glucosamine-6-phosphate synthase by the anticapsin moiety of bacilysin. Its inhibition represses the biosynthesis of peptidoglycans, the essential constituents of the bacterial cell wall [68,69]. In the KR2-7 genome, bacilysin can be encoded by a 7128 bp bac gene cluster consisting of seven genes (bacA-E, ywf AG), and display high gene similarity to those of B. subtilis 168 (Table 2, Figure 7). This metabolite and its derivatives were detected neither in the KR2-7 control cell extract nor dual culture cell extract, likely due to culture conditions or the assay method. Furthermore, the KR2-7 genome encompassed two distinct gene clusters encoding bacteriocins, including subtilosin A and sporulation killing factors (SKFs). Subtilosin A is a macrocyclic anionic antimicrobial peptide originally obtained from wild-type strain B. subtilis 168 [70] but is also produced by B. amyloliquefaciens and B. atrophaeus [71,72]. This bacteriocin displayed a bactericidal effect on a broad spectrum of bacteria, including Grampositive and Gram-negative bacteria and both aerobes and anaerobes [73], possibly through an interaction with membrane-associated receptors, or binding to the outer cell membrane, and is conducive to membrane permeabilization [73][74][75]. Subtilosin A is ribosomally synthesized by an alb gene cluster containing eight genes (albA-G, sboA) spanning 6.8 kbp in the KR2-7 genome (Figure 8). The sboA gene encodes presubtilosin, and albA-G genes encode proteins whose functions are presubtilosin processing and subtilosin export [76].
The mass peaks corresponding to subtilosin A and its homologs appeared neither in the KR2-7control cell extract nor the dual culture cell extract. These peaks are detectable by altering the culture condition and/or evaluating the procedure. The KR2-7 genome also harbored a 5976 bp skf gene cluster encompassing skf ABCEFGH, and involves the production and release of killing factors during sporulation (Figure 9). During the early stages of sporulation, sporulating cells of B. subtilis exude extracellular killing factors to kill the nonsporulating sister cells whose immunity to these toxins was not developed. As a result, the nutrient from the dead cells are released and then used by the sporulating cells to resume their growth. This phenomenon is termed "cannibalism" and causes a delay in sporulation [77,78]. The SKF bacteriocin produced by the sporulating cells can destroy other soil-inhabiting bacteria. Similarly, the expression of skf genes in B. subtilis inhibits the growth of X. orzae pv. oryzae, the causative agent of rice bacterial blight [79].
1 Figure 9. The biosynthetic gene cluster of sporulation killing factor antibacterial metabolite in strain KR2-7.

Plant Colonization by Strain KR2-7
The most crucial step for a PGPR (Plant Growth Promoting Rhizobacteria) agent to survive, enhance plant growth, and suppress plant disease is the efficient colonization of plant tissues. The plant colonization process comprises two steps. In the first step, PGPR agents reach the surface of plant tissue either by passive movement in water flow or by flagellar movement. The second step is to establish the plant-bacterium interaction reliant on bacterial biofilm formation [36,80].
The KR2-7 genome harbored the gene clusters for flagellar assembly (flg cluster, flh cluster and, fli cluster) and bacterial chemotaxis (che cluster) together with other genes known to be necessary for swarming motility, including hag, two stator elements (motAB), as well as regulatory genes swrAA, swrAB, swrB and, swrC ( Table 6). In the step of efficient colonization, the PGPR agent forms bacterial biofilm and not only strengthens the plantbacterium interaction but protects the plant root system as a bio-barrier against pathogen attacks [80]. The main component of bacterial biofilm is the extracellular polymeric substances (EPS) with a chemical composition including proteins, neutral polysaccharides, charged polymers, and amphiphilic molecules [80]. The eps cluster (epsC-O) encoding exopolysaccharide of biofilm and its regulatory genes sinR and arbA (repressors) and, sinI (antirepressor), the yqxM-sipW-tasA gene cluster encoding amyloid fiber (TasA protein of biofilm) and pgcA encoding phosphoglucomutase were found in the KR2-7 genome (Table 6). Moreover, the involvement of surfactin in cell adhesion and biofilm formation due to its 3D topology and amphiphilic nature has been illustrated [81,82]. Baise et al. [58] declared that deleting surfactin gene expression in B. subtilis strain 6051 led to disability to form robust biofilm on Arabidopsis root surface, and reduced the suppression of disease caused by Pseudomonas syringae. Besides, the deficiency in surfactin production in B. subtilis, strain UMAF6614 resulted in a biofilm formation defect on melon phylloplane and partially reduced the suppression of bacterial soft root rot, bacterial leaf spot, and cucurbit powdery-mildew by the biocontrol stain [83].

Genes Involved in Bacterium-Plant Interactions
Quite apart from the antagonistic mechanisms of bacterial biocontrol strains, these bacterial microorganisms are also involved in plant growth augmentation through making nutrients available for host plants, production of plant growth-promotion hormones, and the induction of systemic resistance within the plant by specific metabolite secretion [96,97]. Similar to other biocontrol microorganisms, the KR2-7 genome contains the genes/gene clusters related to plant growth promotion ( Table 6).
The KR2-7 genome contained moaA-E genes encoding molybdenum cofactor and may be a relic of a nitrogen-fixing gene cluster or a cofactor for nitrogen assimilation [80]. Moreover, the genes for nitrate reduction (narG-J), nitrate transport (narK), probable transcription regulator genes (arf M), regulatory protein (fnr), an ammonium acid transporter (nrgA), and its regulator gene (nrgB), along with the nas gene cluster (nasA-F), were also identified in the KR2-7 genome. The nas gene cluster is involved in nitrite transport and reduction (Table 6).
Furthermore, ions of m/z values 883.4 and 905.2 were detected in the KR2-7dual culture cell extract and were identified as bacillibactin [M + H] + and bacillibactin [M + Na] + ( Figure 10) by comparison with previously reported data [99,100]. Notably, the molecular ion peaks corresponding to bacillibactin were not observed in the control cell extract of KR2-7. Siderophores are low-molecular-weight molecules with a high affinity for ferric iron that solubilize iron from minerals and organic compounds under iron limitation conditions [101]. Siderophore-producing bacterial strains impact plant health by supplying iron to the host plants [102,103], depriving fungal pathogens of iron [104], and suppressing fungal phytopathogens, including F. oxysporum f. sp. capsici [105] and Phytophthora capsici [106]. Furthermore, it was reported that siderophores mitigate heavy metal contamination of soil through the formation of a stable complex with environmental toxic metals such as Al, Cd, Cu, Ga, In, Pb, and Zn [101]. Volatile organic compounds (VOCs) produced by PGPR agents play a significant role in promoting plant growth through the regulation of synthesis or metabolism of phytohormones [107], the induction of systemic disease resistance [108,109], and the control of plant pathogens [110]. A 2,3-butanediol and 3-hydroxy-2-butanone (acetoin) are the best-known growth-promoting VOCs that produced B. subtilis and B. amyloliquefaciens. The genome of the KR2-7 harbored als gene cluster (alsR, alsS, alsD), along with the bdhA gene is together required for the biosynthesis pathway of 2,3-butanediol from pyruvate. In this pathway, alsS encodes the acetolactate synthase enzyme, which catalyzes the condensation of two pyruvate molecules into acetolactate. Then, acetolactate decarboxylase, encoded by alsD, converts decarboxylate acetolactate into acetoin. The alsR regulates two aforesaid steps. Finally, the bdhA encoded (R, R)-butanediol dehydrogenase enzyme catalyzes 3-hydroxy-2butanone (acetoin) to 2,3-butanediol [111]. In addition, the KR2-7 genome contained ilvH, ilvB, ilvC genes and a leu gene cluster (leuABCD) which are required for the biosynthesis pathway of three branched-chain amino acids (BCAA), including leucine, isoleucine, and valine. Acetolactate is a central metabolite between 2,3-butanediol and BCAA biosynthesis and can involve in both anabolism and catabolism by acetolactate decarboxylase. It was reported that acetolactate decarboxylase is an enzyme with a dual role that can direct acetolactate flux to catabolism in favour of valine and leucine biosynthesis or can catalyze the second step of the 2, 3-butanediol anabolism pathway [112]. Bacillus spp. can enhance plant growth through the synthesis of plant growth-promoting hormones, such as auxin, indole-3-acetic acid (IAA), and gibberellic acid. The genome of KR2-7 may encompass genes/gene clusters responsible for the biosynthesis of indole acetic acid, phytase, and trehalose (Table 6). Moreover, a large variety of PGPRs produce polyamines, such as putrescine, spermine, spermidine, and cadaverine, and are known to be involved in promoting plant growth and improving abiotic stress tolerance in plants [113]. The genes coding for arginine decarboxylase (SpeA), agmatinase (SpeB), and spermidine synthase (SpeE), which direct polyamines biosynthesis, were also found in the KR2-7 genome (Table 6).

Discussion
Previously, the Bacillus subtilis species complex was composed of four close subspecies, i.e., subspecies subtilis, spizizenii, inaquosorum, and stercoris, which were differentiated through a phylogenetic analysis of multiple protein-coding genes and genome-based comparative analysis [114,115]. B. subtilis subsp. Inaquosorum was deemed as a distinctive taxon encompassing strains KCTC 13429 and NRRL B-14697 [116]. Recent phylogenomic studies clearly distinguished subspecies inaquosorum from subspecies spizizenii, as the estimated ANI among them was smaller than the defined ANI for species delineation (95% ANI) [117]. In addition to a low ANI value (<95%), the BGC of subtilin exclusively presents in the genomes of subspecies spizizenii, but was not characterized in subspecies inaquosorum genome [115]. Accordingly, B. inaquosorum KR2-7 was clearly differentiated from B. subtilis subsp. spizizenii W23 because of the low ANI value among them (94.18%) and the lack of subtilin gene cluster in the genome content of strain KR2-7. In addition, it was reported that B. inaquosorum is the only species to produce bacillomycin F. It was approved by detecting a unique MALDI-TOF-MS biomarker at m/z 1120 in the MALDI-TOF-MS spectra of B. inaquosorum that cannot be produced by other species [114]. Since this unique biomarker (m/z value 1120.6) was observed in the MALDI-TOF-MS spectra of strain KR2-7, it can be concluded that this strain is a B. inaquosorum. Recently, the ability of B. inaquosorum strain HU Biol-II in producing bacillomycin F was confirmed through HPLC MS/MS [15]. Most recently, subspecies spizizenii, inaquosorum, and stercoris were promoted to species status through a comparative genome study [118]. This study determined that each subspecies had unique secondary metabolite genes encoding unique phenotypes, thereby each subspecies can be promoted to species. According to the aforesaid results, strain KR2-7 was identified as a B. inaquosorum.
The genome-driven data highlighted the plant-beneficial functions of strain KR2-7. This strain can efficiently colonize the plant root surface, relying on its swarming motility and biofilm formation abilities. Efficient root colonization of biocontrol bacteria is necessary for suppressing phytopathogens, and biofilm formation is an essential prerequisite for persistent root colonization [119,120]. The biofilm-deficient mutant of B. pumilus HR10 produced weakened biofilms with reduced contents of extracellular polysaccharides and proteins, and thereby could not efficiently control pine seedling damping-off disease [121]. Hence, the suppression of tomato Fusarium wilt by strain KR2-7 [16] may contribute to efficient tomato root colonization of this strain.
In addition to efficient root colonization, strain KR2-7 is able to directly suppress soildwelling phytopathogens through producing eight antimicrobial secondary metabolites, e.g., fengycin, surfactin, bacillomycin F, macrolactin, bacillaene, bacilysin, subtilosin A, and sporulation killing factor. The combination of obtained data via MALDI-TOF-MS with our previous observations [16] confirmed that strain KR2-7 produced at least four bioactive metabolites (including fengycin, surfactin, macrolactin, and bacillaene) to directly protect the tomato plant from the invasion and penetration of Fol. The cyclic lipodecapeptide fengycin exhibits strong fungitoxic properties by inhibiting phospholipase A2 and aromatase functions [122], disruption of biological membrane integrity [123], deformation and permeabilization of hyphae [124,125], and induction of ISR [126]. In this context, the strong antifungal activity of B. inaquosorum strain HU Biol-II against a diverse group of fungi highly pertained to the fengycin produced by this strain. Interestingly, 97.47% of the ppsA-E gene cluster in KR2-7 was similar to the fengycin gene cluster in strain HU Biol-II [15]. Fengycin produced by B. subtilis SQR9 and B. amyloliquefaciens NJN-6 significantly inhibited the growth of F. oxysporum [127,128]. Moreover, fengycin BS155 isolated from B. subtilis BS155 destroyed Magnaporthe grisea through damaging the plasma membrane and cell wall, disruption of mitochondrial membrane potential (MMP), chromatin condensation, and the induction of reactive oxygen species (ROS) [129]. In addition to fengycin, the contribution of other secondary metabolites in the biocontrol of various pathogens has been reported. The supernatant of B. subtilis GLB191, consisting of surfactin and fengycin, highly controlled grapevine downy mildew caused by Plasmopara viticola by means of direct antagonistic activity and the stimulation of plant defence [51]. Furthermore, the strong antifungal effect of B. velezensis strains Y6 and F7 against Ralstonia solanacearum and F. oxysporum was attributed to the production of fengycin, iturin, and surfactin, among which iturin played a key role in the suppression of F. oxysporum [130]. The biocontrol mechanism of B. amyloliquefaciens DH-4 against Penicillium digitatum, the causal agent of citrus green mold, was secreting a cocktail of antimicrobial compounds consisting of macrolactin, bacillaene, iturins, fengycin, and surfactin [100].
Additionally, strain KR2-7 can exert hormones, such as IAA, phytase, and trehalose for root uptake and rebalance hormones in the host plant to boost growth and stress response. Phytate (inositol hexa-and penta-phosphates) is the predominant form of soil organic phosphorus, which is unavailable for plant uptake due to the rapid immobilization of phosphorus and the lack of adequate phytase levels in plants [131]. Phytase is a phosphatase enzyme responsible for the transformation of organic phytate to inorganic phosphate, which is acquirable for plant roots. Similarly, phytase-producing Bacillus strains can effectively enhance plant growth through the liberation of reactive phosphorus from phytate and make this element available for plant uptake. In the presence of phytate, the comparison of the culture filtrate of B. amyloliquefaciens strain FZB45 with those of a phytase-deficient mutant provided evidence that the phytase activity of strain FZB45 enhanced the growth of corn seedling [132]. The bacterization of Brassica juncea with Bacillus sp. PB-13 considerably boost phosphorus content and growth parameters in 30-day-old seedlings [133]. More recently, the soil inoculation of Bacillus strain SD01N-014 resulted in the enhancement of soil phosphorus content and the promotion of maize seedling growth [134]. Accordingly, extracellular phytase activity of strain KR2-7 mediated with phy gene can be expectable. In addition, the presence of genes involved in the biosynthesis of IAA and trehalose in the KR2-7 genome (Table 6) is an indication of this strain's potential in the mitigation of salt toxic stress on plants. Inoculation of tomato plants subjected to salt stress with OxtreS (trehalose over-expressing strain) mutant of Pseudomonas sp. UW4 considerably boosted the dry weight, root and shoot length, and chlorophyll content of the tomato plant [135]. Moreover, canola seedlings treated with over-expressed IAA transformant of UW4 represented longer primary root with an increased number of root hairs than seedlings treated with wild-type UW4 [136]. The growth promotion of root hairs by IAA improves the assimilation of water and nutrients from the soil, which in turn raises plant biomass [136]. Similarly, Japanese cypress seedlings inoculated with B. velezensis CE 100 showed significant increases in growth parameters and biomass due to the production of indole-3-acetic acid (IAA) by CE 100 strain [137].

Conclusions
According to genome-driven data, along with chemical analysis results, strain KR2-7 most likely exploits four possible modes of action to control tomato Fusarium wilt, as shown in Figure 11: (1) Inhibition of the pathogen growth through the diffusion of antifungal and antibacterial secondary metabolites and biofilm formation; (2) Stimulation of ISR in tomato via the production of surfactin and volatile organic compounds; (3) Promotion of plant health and growth by producing plant growth promotion hormones and polyamines, supplying iron for tomato, depriving the pathogen of iron, and relieving heavy metal stress in the soil as a result of siderophore bacillibactin activity; (4) Efficient colonization of plant roots.
The described modes of action were highly based on the identified gene clusters encoding secondary metabolites and characterized gene/gene clusters involved in plant colonization, plant growth promotion, and ISR. Furthermore, future studies using integrated omics approaches and the mutagenesis of strain KR2-7 are required to approve the aforesaid possible modes of action of strain KR2-7 and exact functions of the putative genes and gene clusters in the suppression of fungal pathogen Fol. Figure 11. Schematic presentation of putative biocontrol mechanism of strain KR2-7 against Fol. (A) An untreated tomato plant in which Fol (yellow 16-point star) penetrated root tissue, colonized and blocked vascular system to prevent water and nutrients from being transferred to plant organs. It caused yellowing began with bottom leaves, followed by wilting, browning, and defoliation. Growth is typically stunted, and little or no fruit develops. (B) Strain KR2-7 (blue rod) reaches to tomato root and colonizes on the root surface through its motility potential and biofilm formation. As a result of root colonization, strain KR2-7 diffuses a wide variety of antifungal and antibacterial secondary metabolites to establish a protective zone (green dash line semicircular) in tomato rhizosphere. Strain KR2-7 directly limits the invasion of Fol fungal pathogen through diffused antifungal secondary metabolites and also control the bacterial pathogens of tomato by means of produced antibacterial secondary metabolites. Meanwhile, volatile organic compounds and surfactin stimulate tomato systemic resistance to provide ISR-mediated protection (yellow dash line arrow) against phytopathogens. Moreover, the tomato growth is enhanced assisted by growth-promoting hormones, polyamines, and siderophore bacillibactin. Figure 11. Schematic presentation of putative biocontrol mechanism of strain KR2-7 against Fol. (A) An untreated tomato plant in which Fol (yellow 16-point star) penetrated root tissue, colonized and blocked the vascular system to prevent water and nutrients from being transferred to plant organs. It caused yellowing began with bottom leaves, followed by wilting, browning, and defoliation. Growth is typically stunted, and little or no fruit develops. (B) Strain KR2-7 (blue rod) reaches to tomato root and colonizes on the root surface through its motility potential and biofilm formation. As a result of root colonization, strain KR2-7 diffuses a wide variety of antifungal and antibacterial secondary metabolites to establish a protective zone (green dash line semicircular) in the tomato rhizosphere. Strain KR2-7 directly limits the invasion of Fol fungal pathogen through diffused antifungal secondary metabolites and also control the bacterial pathogens of tomato by means of produced antibacterial secondary metabolites. Meanwhile, volatile organic compounds and surfactin stimulate tomato systemic resistance to provide ISR-mediated protection (yellow dash line arrow) against phytopathogens. Moreover, tomato growth is enhanced assisted by growth-promoting hormones, polyamines, and siderophore bacillibactin.