Genome and Transcriptome Analysis to Elucidate the Biocontrol Mechanism of Bacillus amyloliquefaciens XJ5 against Alternaria solani

Early blight, caused by Alternaria solani, is an important disease affecting tomatoes. Biological control offers an environmentally friendly approach to controlling pathogens. Herein, we identified a B. amyloliquefaciens strain XJ5 and investigated its biocontrol mechanism against A. solani. A. solani growth was significantly inhibited by XJ5, with the inhibition rate of cell-free culture supernatants reaching 82.3%. Furthermore, XJ5 crude protein extracts inhibited conidia germination and altered the mycelial morphology of A. solani. To uncover the potential biocontrol mechanism of XJ5, we analyzed its genome sequence and transcriptome. The genome of XJ5 comprised a 4.16 Mb circular chromosome and two circular plasmids. A total of 13 biosynthetic gene clusters and 127 genes encoding hydrolases were identified, suggestive of the ability of XJ5 to secrete antagonistic secondary metabolites and hydrolases. Transcript analysis revealed 174 differentially expressed genes on exposing A. solani to XJ5 crude protein extracts. The expression of genes related to chitin and mannose synthesis was downregulated, indicating that XJ5 metabolites may impact chitin and mannose synthesis in A. solani. Overall, these findings enhance our understanding of the interactions between B. amyloliquefaciens and phytopathogens and pave the way for the agricultural application of this promising biocontrol agent.


Introduction
Early blight of tomato is one of the most devastating diseases caused by the necrotrophic pathogen Alternaria solani, which is distributed worldwide [1]. The pathogen can infect leaves, stems, petioles, twigs, and fruits under favorable conditions, causing approximately 50-86% loss in fruit yield [2]. Fungicide treatments are the most efficient approach for controlling early blight; however, the frequent and incorrect application of agrochemicals can lead to the emergence of resistant pathogens and adversely impact the environment [3][4][5]. Biological control of early blight has received increased attention, and various microorganisms and their metabolites with antifungal activity against A. solani have been reported [2,6].

In Vitro Antifungal Activity Assays
Dual culture assays were performed to evaluate the in vitro antifungal activity of B. amyloliquefaciens against A. solani. A. solani was cultivated on PDA for 5 days at 25 • C; subsequently, fresh hyphae discs from A. solani colony margins, measuring 6 mm in diameter, were placed on a new PDA plate, and cell suspension of XJ5 was spread in a straight line on both sides of the disc [35]. A fresh disc from A. solani colony margins (diameter: 6 mm) was inoculated on PDA as the control. All PDA plates were inoculated at 25 • C for 5 days. There were at least four biological repeats for each treatment, and the assay was replicated twice. The inhibition rate (IR) was calculated according to a previously Microorganisms 2023, 11, 2055 3 of 16 described method [35]: IR (%) = (colony diameter of control A. solani − colony diameter of A. solani treated with XJ5) / colony diameter of control A. solani × 100%.

Antifungal Activity Assessment of Crude Protein Extracts from XJ5
The protein component was harvested using a previously reported method with some modifications [36]. XJ5 cells were streaked onto BPY agar plates, and a single colony was inoculated into 100 mL BPY broth and cultured at 25 • C for 72 h with shaking at 200 rpm. This entire culture was then centrifuged at 4 • C, and 12,000 rpm for 10 min. Subsequently, it was filtered through a 0.22-µm filter to obtain cell-free culture supernatant. The supernatant was cooled on ice, followed by the addition of saturated ammonium sulfate solution to bring the ammonium sulfate concentration to 70% saturation. This mixture was then incubated at 4 • C for 2 days. The precipitation products were collected by centrifugation at 4 • C and 10,000 rpm for 5 min, dissolved in Tris-HCL buffer (10 mM, pH 7.4), and dialyzed using dialysis bags (Molecular Weight Cut-Off 8000-14,000, Solarbio, Beijing, China) at 4 • C for 12 h. Precipitates were freeze-dried, dissolved in distilled water, and filtered through a 0.22-µm filter to obtain crude protein extracts. Then, 50 µL spore suspensions of A. solani (2 × 10 5 ) were inoculated on a PDA plate; using a sterilized punch, two wells (6 mm in diameter) were then created on the plate 2 cm from the dried spore suspension, crude protein extracts (100 µL) were pipetted into one of the wells, and ddH 2 O (100 µL) without crude protein extracts was pipetted into the other well as a control.

Light Microscopy and Scanning Electron Microscopy (SEM) of A. solani
A. solani conidia were observed under a light microscope 48 h after incubation with crude protein extracts from XJ5. A. solani hyphae disc (diameter: 6 mm) was inoculated into 100 mL potato dextrose broth, which contained crude protein extracts from XJ5 at a concentration of 1% and cultivated with constant shaking at 25 • C and 120 rpm for 6 h, 12 h, 24 h, and 48 h, respectively. Afterward, hyphae were collected and fixed with a solution containing 2.5% glutaraldehyde, followed by dehydration using progressively increasing concentrations of ethanol (30%, 50%, 70%, 80%, 90%, 95%, and 100%). The samples were dried with a critical point drier using the critical-point drying method and observed under a scanning electron microscope (JSM-6490LV, JEOL Ltd., Tokyo, Japan). A. solani hyphae grown on a PDA plate served as a control.

Total RNA Extraction, RNA Sequencing, and Transcriptomic Analysis
A. solani hyphae treated with crude protein extracts from XJ5 for 6 h, 12 h, 24 h, and 48 h were collected. A. solani hyphae grown on PDA served as a control. Each treatment consisted of three biological replicates, and fifteen samples were collected. Hyphae were ground in liquid nitrogen with a mortar and pestled to a fine powder. Total RNA was extracted using a TRIzol RNA Extraction Kit (Takara Bio, Inc., Kusatsu, Japan) as recommended by the manufacturer. Nanodrop 2000 (Thermo Scientific™, Waltham, MA, USA) and agarose gel electrophoresis were used to check RNA concentrations. mRNA enrichment and library preparation were performed, and the libraries were sequenced on the Illumina NovaSeq 6000 platform (Majorbio Bio-Pharm Technology Corporation, Shanghai, China), with each library generating >6 Gb data. The raw data were deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) Database under the accession number PRJNA989022. The raw reads from 15 libraries were processed to remove adaptor sequences, unqualified reads with a high content of unknown base (N) reads (content > 10%), and those with a low-quality score (<10). HISAT2 (v2.1.0) was used to map the clean reads to the reference genome of A. alternata (accession number: GCF_001642055) [44], and bam files were then obtained. StringTie2 was utilized to assemble the bam files with default parameters [45,46]. Gene expression levels were extracted using the prepDE.py script provided by StringTie2. Differentially expressed genes (DEGs) were analyzed using the DESeq2 package in R with the following filtering criteria: adjusted p-value < 0.05 and |log2 (fold change) | > 1 [47]. DEGs were annotated in six different databases (NR, Swiss-Prot, Pfam, eggNOG, GO, and KEGG). GO functional annotation was performed using Blast2GO (version:5.2, https://www.blast2go.com/). KEGG annotation was conducted using KOBAS 3.0 [48]. GO and KEGG pathway enrichment analyses were performed using the clusterProfiler (v4.8.1) package in R with default parameters. Correlations between samples were calculated using Pearson's correlation coefficient and principal component analysis (PCA).

Real-Time Quantitative Reverse Transcription PCR (qRT-PCR) Analysis
For qRT-PCR, A. solani cDNA was synthesized with an oligo d(T) primer using a PrimeScript™ RT Reagent Kit with gDNA Eraser (Perfect Real Time) (Takara Bio, Inc., Kusatsu, Japan) according to the manufacturer's instructions. Beacon Designer v7.92 was used to design gene-specific primers for selected DEG sequences (Supplementary Table S1). qRT-PCR was performed on the CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA) with TB Green ® Premix Ex Taq™ II (Takara Bio, Inc., Kusatsu, Japan). The PCR reaction condition is as follows: pre-denaturation at 95 • C for 30 s, followed by 40 cycles of denaturation at 95 • C for 5 s, and extension at 60 • C for 30 s. A. solani actin gene (GenBank: MK388241.1) was used as an internal reference gene. The relative expression levels of each DEG were calculated using the 2 −∆∆CT method [49], and qRT-PCR results were then compared with RNA-seq data.

Morphological Observation and Antifungal Activity of B. amyloliquefaciens XJ5
XJ5 was observed to grow as white, smooth-faced, nontransparent colonies after 24 h of cultivation on BPY agar plate ( Figure 1A). The dual culture assays indicated that XJ5 inhibited the mycelial growth of A. solani on PDA, resulting in a clear inhibition zone ( Figure 1B,C). Further, the cell-free culture supernatants of XJ5 were found to suppress the mycelial growth of A. solani, and A. solani colony growth diameters decreased by 82.3% ( Figure 1C,D). To further investigate the antifungal constituents of XJ5, crude protein was extracted from the fermentation supernatant broth of XJ5, and its antifungal activity was analyzed. After 3 days of treatment with crude protein extracts from XJ5, A. solani growth was found to be inhibited, with an inhibition zone diameter of 48 mm ( Figure 1E); a significant antifungal effect was observed even 5 days after treatment ( Figure 1E). These results indicated that XJ5 inhibits the mycelial growth of A. solani by secreting antifungal proteins. analyzed. After 3 days of treatment with crude protein extracts from XJ5, A. solani growth was found to be inhibited, with an inhibition zone diameter of 48 mm ( Figure 1E); a sig nificant antifungal effect was observed even 5 days after treatment ( Figure 1E). These re sults indicated that XJ5 inhibits the mycelial growth of A. solani by secreting antifunga proteins.

XJ5 Inhibits Conidia Germination and Disturbs the Hyphal Structure of A. solani
A. solani was treated with crude protein extracts from XJ5, and its morphology was observed by light microscopy and SEM. Microscopic observation revealed that the germination and germ tube elongation of A. solani conidia were inhibited on treatment with crude protein extracts from XJ5 ( Figure 2A). The SEM observation showed that A. solani hyphae in the control group appeared plump and smooth and showed clear outlines ( Figure 2F), while A. solani hyphae treated with crude protein extracts from XJ5 appeared swollen and enlarged ( Figure 2B-E). After 6 h of treatment with crude protein extracts from XJ5, A. solani mycelium was swollen and showed spherical vesicles with surface wrinkles ( Figure 2B); after 12 h of treatment, spherical vesicles were enlarged, and their abundance was higher ( Figure 2C). Further, after 24 h of treatment, spherical vesicles were expanded and hyphal cell walls appeared rough and shriveled ( Figure 2D), and after 48 h of treatment, damage and distortion of hyphae were apparent, and degradation of some mycelia was evident ( Figure 2E). These results suggested that crude protein extracts from XJ5 target the cell wall of A. solani to exert their inhibitory effect against this tomato early blight fungus.

XJ5 Inhibits Conidia Germination and Disturbs the Hyphal Structure of A. solani
A. solani was treated with crude protein extracts from XJ5, and its morphology was observed by light microscopy and SEM. Microscopic observation revealed that the germination and germ tube elongation of A. solani conidia were inhibited on treatment with crude protein extracts from XJ5 ( Figure 2A). The SEM observation showed that A. solani hyphae in the control group appeared plump and smooth and showed clear outlines (Figure 2F), while A. solani hyphae treated with crude protein extracts from XJ5 appeared swollen and enlarged ( Figure 2B-E). After 6 h of treatment with crude protein extracts from XJ5, A. solani mycelium was swollen and showed spherical vesicles with surface wrinkles ( Figure 2B); after 12 h of treatment, spherical vesicles were enlarged, and their abundance was higher ( Figure 2C). Further, after 24 h of treatment, spherical vesicles were expanded and hyphal cell walls appeared rough and shriveled ( Figure 2D), and after 48 h of treatment, damage and distortion of hyphae were apparent, and degradation of some mycelia was evident ( Figure 2E). These results suggested that crude protein extracts from XJ5 target the cell wall of A. solani to exert their inhibitory effect against this tomato early blight fungus.

Genome Features of B. amyloliquefaciens XJ5
The B. amyloliquefaciens XJ5 genome was sequenced to elucidate its biocontrol mechanism. The genome features of XJ5 are summarized in Figure 3 and Supplementary Table  S2. The genome of B. amyloliquefaciens XJ5 showed a circular chromosome (4,160,003 bp) and two plasmids (215,881 and 63,621 bp); the GC content was 46.09%, 37.28%, and 40.93%, respectively (GenBank: CP071970.1, CP071971.1, and CP071972.1, respectively). There was a total of 4199 predicted protein-coding genes in the chromosome; 310 in the first plasmid, and 77 in the second plasmid. The total length of protein-coding genes was 3,939,753 bp, accounting for 88.74% of the genome sequence. A total of 43 tandem repeat sequences, 88 tRNA-coding genes, and 27 rRNA genes were predicted in the chromosome sequence ( Figure 3A). The number of genes annotated in NR, Swiss-Prot, Pfam, COG, GO, and KEGG databases was 4586, 3695, 3460, 3111, 2464, and 2241, respectively ( Figure 3B).

Genome Features of B. amyloliquefaciens XJ5
The B. amyloliquefaciens XJ5 genome was sequenced to elucidate its biocontrol mechanism. The genome features of XJ5 are summarized in Figure 3 and Supplementary Table S2. The genome of B. amyloliquefaciens XJ5 showed a circular chromosome (4,160,003 bp) and two plasmids (215,881 and 63,621 bp); the GC content was 46.09%, 37.28%, and 40.93%, respectively (GenBank: CP071970.1, CP071971.1, and CP071972.1, respectively). There was a total of 4199 predicted protein-coding genes in the chromosome; 310 in the first plasmid, and 77 in the second plasmid. The total length of protein-coding genes was 3,939,753 bp, accounting for 88.74% of the genome sequence. A total of 43 tandem repeat sequences, 88 tRNA-coding genes, and 27 rRNA genes were predicted in the chromosome sequence ( Figure 3A). The number of genes annotated in NR, Swiss-Prot, Pfam, COG, GO, and KEGG databases was 4586, 3695, 3460, 3111, 2464, and 2241, respectively ( Figure 3B).
Thirteen secondary metabolite biosynthetic gene clusters (BCGs) were found in the genome of XJ5 (Table 1), including five NRPs (locillonycin, surfactin, fengycin, bacillibactin, and a gene cluster that is not matched to a known NRPS), two polyketide synthetases (butirosin A/butirosin B), three trans-acyltransferase polyketide synthetases (macrolactin H, bacillaene, and difficidin), two terpenes, and one other BCGs (bacilysin). These BCGs reportedly exhibit high efficacy against bacteria, fungi, and viruses. The production of antimicrobial peptides through the nonribosomal synthesis pathway is an important mechanism employed by biocontrol bacteria to suppress phytopathogens. These findings indicated that the antagonistic activity of B. amyloliquefaciens XJ5 against A. solani may be related to the synthesis of these biocontrol agents. Furthermore, four functional unknown gene clusters (clusters 4,8,9,12) were found, including two terpenes, one type III polyketide synthase (T3PKS), and one NRPS, indicating that the presence of additional gene clusters in XJ5 for the synthesis of potential novel antifungal substances. Consequently, XJ5 may have significant application potential in agriculture.

Transcriptomic Changes in A. solani Treated with B. amyloliquefaciens XJ5
Upon sequencing, each sample generated an average of 6.5 Gb raw reads; 745,252,900 clean reads were eventually obtained, with Q30 value > 96%. The genome mapping rates for reads from different samples ranged between 82.27% and 86.79% (Supplementary Table S3). PCA indicated a relatively high correlation between sequenced duplicate samples ( Figure 4A). After 6 h, 12 h, 24 h, and 48 h of treatment with crude protein extracts from XJ5, 2502, 1791, 1787, and 3157 DEGs were identified, respectively, and they were grouped into three clusters ( Figure 4B Figure 4B). Cluster 3 was mainly enriched in the metabolic process (GO:0008152), carbohydrate metabolic process (GO:0005975), and secondary metabolic process (GO:0019748). Overall, 174 genes were differentially expressed in total, of which the expression of 60 genes was upregulated and that of 114 genes was downregulated ( Figure 4D,E).
After 6 h and 12 h of treatment with crude protein extracts from XJ5, a higher number of upregulated genes was identified, which were primarily enriched in terms such as ribosome and secondary metabolite synthesis ( Figure 5A). KEGG pathway analysis indicated enrichment in ribosome pathways, structural constituents of ribosomes, catalytic activity, and others ( Figure 5C). After 48 h of treatment, there were more downregulated genes, mainly enriched in pathways such as biosynthesis of amino acids and ATPase-coupled transmembrane transporter activity ( Figure 5B,D). In addition, the expression of genes involved in cell wall synthesis, such as chitin synthase and mannose synthase-related genes, including CC77DRAFT_1027903, CC77DRAFT_1062346, CC77DRAFT_1021349, CC77DRAFT_987140, CC77DRAFT_1031975, CC77DRAFT_1020620, CC77DRAFT_1022845, and CC77DRAFT_868963, exhibited significant downregulation across all four stages of treatment, potentially impacting chitin and mannose synthesis in A. solani ( Figure 6). The expression of genes related to glutathione S-transferases (GSTs), O-mannosylation, and the N glycan-processing pathway, including CC77DRAFT_1026794, CC77DRAFT_950947, CC77DRAFT_1015189, and CC77DRAFT_687848, was upregulated at 6 h and 12 h and downregulated at 24 h and 48 h of treatment (Figure 6), suggesting a potential role in fungal cell wall integrity. These results demonstrated that crude protein extracts from XJ5 significantly impact the transcriptional profile of A. solani, especially the genes associated with cell wall synthesis and cell wall integrity. qRT-PCR results of nine randomly selected DEGs correlated well with the transcriptome data (Figure 7). The Pearson correlation coefficient was 0.7465 ( Figure 7J), indicating a strong correlation between qRT-PCR results and the transcriptome data.  After 6 h and 12 h of treatment with crude protein extracts from XJ5, a higher number of upregulated genes was identified, which were primarily enriched in terms such as ribosome and secondary metabolite synthesis ( Figure 5A). KEGG pathway analysis indicated enrichment in ribosome pathways, structural constituents of ribosomes, catalytic activity, and others ( Figure 5C). After 48 h of treatment, there were more downregulated transcriptional profile of A. solani, especially the genes associated with cell wall synthesis and cell wall integrity. qRT-PCR results of nine randomly selected DEGs correlated well with the transcriptome data ( Figure 7). The Pearson correlation coefficient was 0.7465 (Figure 7J), indicating a strong correlation between qRT-PCR results and the transcriptome data.   transcriptional profile of A. solani, especially the genes associated with cell wall synthesis and cell wall integrity. qRT-PCR results of nine randomly selected DEGs correlated well with the transcriptome data ( Figure 7). The Pearson correlation coefficient was 0.7465 (Figure 7J), indicating a strong correlation between qRT-PCR results and the transcriptome data.

Discussion
B. amyloliquefaciens is a nonpathogenic bacterium known for its biological control characteristics, including colonization ability, the inhibition of pathogens, and the induction of systemic resistance in plants [33]. While many reports exist on the biocontrol potential of B. amyloliquefaciens against diverse phytopathogens [50][51][52][53], there have been limited studies on its effectiveness in reducing the growth of A. solani. Herein, we isolated and identified B. amyloliquefaciens strain XJ5. In vitro experiments indicated that XJ5 exhibits potent growth inhibition effects on A. solani, the causal agent of early blight of tomato, making it a promising biocontrol agent. B. amyloliquefaciens strains evidently secrete some antifungal proteins or lipopeptides [15,54]. Similarly, crude protein extracts from XJ5 are known to inhibit mycelial growth and conidia germination and disrupt the hyphal structure of A. solani ( Figures 1E and 2B-E). These results demonstrated that crude protein extracts of XJ5 have significant antifungal activity; however, the characteristics and mechanisms of crude protein extracts require further exploration.
The complete genome of B. amyloliquefaciens offers new insights pertaining to metabolites with potential biocontrol activity [33,55]. Strain FZB42, for example, has an impressive capability to synthesize various secondary metabolites, with approximately 8% of its genome dedicated to antimicrobial metabolite synthesis, while genes related to antimicrobial syntheses in B. subtilis account for approximately 4-5% of its genome on average [7,33,56]. Moreover, gene clusters associated with antibacterial substance synthesis have been identified in the genomes of B. amyloliquefaciens AS43.3 and DSM7 T , providing valuable information to unveil the molecular mechanisms of antimicrobial activity in B. amyloliquefaciens [28,34]. To elucidate the potential biocontrol mechanism of XJ5, we annotated and analyzed its whole genome sequence. Thirteen BCGs were consequently identified, which included seven gene clusters coding for proteins with amino acid similarities of >90% to seven known classes of antibiotics (surfactin, macrolactin, bacillaene, fengycin, difficidin, bacillibactin, and bacilysin), two synthesis gene clusters with amino acid similarities of <30% to known classes (locillomycin and butirosin A/B), and four functional unknown gene clusters. These well-characterized antibiotics possess broad-spectrum antifungal or antibacterial activity [57,58]. Furthermore, the genome of XJ5 was found to contain genes encoding hydrolases, such as chitinase, xylanase, phosphatase, and protease, which play a vital role in degrading the main components of fungal cell walls (e.g., chitin and chitosan). Our results suggested that the genome of XJ5 encodes several secondary metabolites (lipopeptides and antifungal compounds) and antifungal proteins such as hydrolytic enzymes, which may endow XJ5 with the ability to produce different antifungal substances for biocontrol. Indeed, our biocontrol experiment results indicated that XJ5 can inhibit the growth of A. solani, and SEM observations demonstrated that the hyphal cell walls of A. solani were deformed and degraded after treatment with XJ5 crude protein extracts ( Figure 2). These findings are likely related to the production of different secondary metabolites and cell wall degradation enzymes.
The transcriptome analysis results indicated that many DEGs were enriched in all four stages of treatment; rude protein extracts from XJ5 were observed to significantly impact the transcriptional profile of A. solani. KEGG and GO enrichment analyses revealed more upregulated GO terms and KEGG pathways at the early stages (6 and 12 h) of treatment. At this time, the mycelium of A. solani was swollen, which coincided with the appearance of spherical vesicles ( Figure 2B,C). After 48 h of treatment, the expression of most DEGs was downregulated and hyphal cell walls appeared rough and shriveled, indicating that hyphal cells may be in a stressed physiological metabolic state.
Filamentous fungal cell walls comprise galactomannans, chitin, and β-1,3-glucans, which play an important role in fungal viability and pathogenicity [59]. Bacillus spp. can produce various antifungal substances that target fungal cell walls and membranes and then inhibit fungal growth [15,[60][61][62]. The expression of genes involved in cell wall synthesis, such as chitin synthase and glycosylation modification-related genes, was inhibited by XJ5 crude protein extracts. The SEM observation indicated that the crude protein extracts from XJ5 can cause hyphal damage and even degradation, these results demonstrated that the hyphal damage may be associated with the downregulation of chitin and mannose synthase-related genes. Furthermore, genes involved in the O-mannosylation (Afpmt1) and N glycan-processing pathways (Afcwh41 and Afams1) can cause deficiencies in cell wall integrity in Aspergillus fumigatus [63][64][65][66]. Glutathione S-transferases are apparently involved in protecting cells against damage caused by oxidative stress in Schizosaccharomyces pombe and Saccharomyces cerevisiae [67,68]. The expression of related genes was upregulated at 6 h and 12 h and downregulated at 24 h and 48 h of treatment. We believe that A. solani upregulates the expression of related genes to alleviate stress when exposed to antifungal substances, but the accumulation of these substances ultimately inhibits its growth. XJ5 could disrupt cell wall synthesis and affect the cell wall integrity of A. solani. A. solani also responds to stress by regulating multiple pathways and gene expression to mitigate the pressure of antifungal substances. To summarize, an antagonistic B. amyloliquefaciens strain XJ5 was identified and its biocontrol mechanism against the phytopathogen A. solani was investigated by genome and transcriptome sequencing. Our findings should further our understanding of the interactions between B. amyloliquefaciens and phytopathogens.