Identification and functional analysis of non-coding regulatory small RNA FenSr3 in Bacillus amyloliquefaciens LPB-18

Microorganisms and cultivation conditions

The experimental strains and plasmids used in this study are shown in Table 1. B. amyloliquefaciens LPB-18 (can produce a series of isoforms of fengycin, fengycin A and fengycin B), and Escherichia coli DH5 α were cultivated in Luria-Bertani (LB) medium at 33 °C. The strain LPB-18 was grown in LB medium (beef extract 5 g/L peptone 10 g/L, yeast extract 5 g/L, NaCl 5 g/L, and glucose 10 g/L) at 33 °Cfor seed culture. The modified Landy medium was used for Fengycin fermentation. The medium components were: fructose 20 g/L, yeast extract 1 g/L, L-glutamic acid 5 g/L, KCl 0.5 g/L, MgSO₄ 0.5 g/L, KH₂PO₄ 1 g/L, L-phenylalanine 2 mg/L, MnSO₄ 5 mg/L, FeSO₄ 0.15 mg/L, and CuSO₄ 0.16 mg/L. The fermentation was carried out in a 500 mL shake flask containing 200 mL medium at 33 °C with 180 rpm for 72 h. Antibiotics (1 mg/L erythromycin or 5 mg/L chloromycetin for the strain LPB-18, 1 mg/L ampicillin for E. coli) were added to the medium appropriately.

Construction of plasmids and mutation strain

The thermo-sensitive knockout vectors pCBS were obtained from the Nanjing Agricultural University (Sun et al., 2021). All primers used for this study are summarised in Table 2. The mutation construction was performed in the same manner as described previously (Sun et al., 2021). Briefly, the upstream and downstream regions of fenSr3 were amplified with specific primers. The two fragments were linked by Seamless Cloning/In-Fusion Cloning polymerase chain reaction (PCR) with the primers up-fenSr3-F and down-fenSr3-R. The integration fragment was inserted into the corresponding restriction sites (EcoRI/BamHI) of pCBS to yield B. amyloliquefaciens LPB-18N via electrical transformation technology, and the knockout strain of B. amyloliquefaciens LPB-18N was validated by PCR amplification using primers up- fenSr3-F and down-fenSr3-R.

The expression vector was constructed as previously described(Zhang et al., 2017). First, the fenSr3 was amplified from B. amyloliquefaciens LPB-18 with the primers fenSr3-F/R and the Pgrac promoter was amplified from pHT43 using the primers Pgrac-F/R. The fusion fragment of fenSr3 and Pgrac were linked by Seamless Cloning/In-Fusion Cloning PCR. After the KpnI and BamHI function, it resulted in pHT-fenSr3. Then, the fenSr3 expression plasmid was transferred into B. amyloliquefaciens LPB-18 and the recombinant strain B. amyloliquefaciens LPB-18P were validated by PCR.

Phenotypic assays of wild-type and mutant strains

The dry cell weight was performed to evaluate the effect of non-coding small RNA FenSr3 on cell growth. Different types of strains were cultured, and aliquots were sampled into a pre-weighed tube. After centrifuging at 8,000 rpm for 5 min, the supernatant was discarded, and the cell pellet was collected. The cell pellet was washed three times with normal saline and dried for 5 h to a constant weight at 90 °C. The crude extract was filtered through a 0.22 µm pore filter, and 30 µL of aliquot was injected into an Eclipse XDB-C18 column (5 µm, 250 × 4.6 mm; Agilent Technologies, Santa Clara, CA, USA) in the RT-HPLC system (Shimadzu, Japan) for further identification and purification. The mobile phase was deionized water containing 0.1% trifluoroacetic acid, and mobile phase B was acetonitrile containing 0.1% trifluoroacetic acid. The flow rate was maintained to be six mL/min under the following conditions: 0–15 min, 30% A to 45% A, 70% B to 55% B; 15–55 min, 45% A-55% A, 55% B-45% B. Monitoring was conducted at 205 nm UV detector (Wang et al., 2015). The crude extracts were dried for 4 h to a constant weight at 65 °C. The plate diffusion assays were used to detect the in vitro antagonistic activity of wild-type and mutant strains against Rhizopus and incubated at 28 °C for 48 h in PDA. The biofilm formation assay was carried out in 96 deep well plates incubating in LB medium at 33 °C for 24 h.

RNA extraction and transcriptome sequencing

B. amyloliquefaciens LPB-18 and B. amyloliquefaciens LPB-18N were cultured for 12 h before their RNA was extracted. NCBI_GCF_000262385.1 was used as the reference genome. Total RNA from these strains was extracted using the RNeasy Extraction kit (Invitrogen Life Technologies, Gaithersburg, MD, USA) according to its manufacturer’s instructions. RNA quality was assessed using Nanodrop 2000 (Thermo Scientific, Waltham, MA, USA). Before sequencing analysis, the ribosomal RNAs were wiped using the RiboZero rRNA Removal Kit (Epicentre) for a gram-positive bacterium. The extracted total RNA was used as a template to synthesize the cDNA first strand using HiScript First Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). Subsequently, the cDNA was sequenced by Genedenovo Biotechnology Co., Ltd (Guangzhou, China) using the Illumina sequencing platform (http://www.omicshare.com). EdgeR 4 software was used to analyze the quantity of gene expression differences between groups using FDR and log2FC genetic variations to filter, filter condition for FDR <0.05 and — log2FC —>1. The functional classification analysis was conducted by Gene Ontology (GO) enrichment analysis (http://www.omicsmart.com/RNAseq/home.html), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis (https://www.omicsmart.com/). Data were collected as previously described in Sun (2021). Specifically, RT-qPCR was used to determine the results of RNA-seq, and the genes of 16s rRNA were used as the reference to normalize the expression of examined genes. The specific procedure of RT-qPCR was performed according to Lu et al. (2022). The relative fold change of the target gene expression was evaluated by the calculation of 2^−ΔΔCt. The threshold cycle (Ct) values were obtained by the real-time PCR System software.

Analysis of biofilm formation and spore count

The wild strain LPB-18 and mutation strain Δ fenSr3 were inoculated into sterile MRS for 5 days at 33 °C. After that, the plates were washed with ddH₂O, dried at 40 °C for 20 min, and strained with five mL of 0.5% (v/v) crystal violet solution for 3 min. The biofilm on the wells was dissolved with one mL of 95% (v/v) ethanol, and the OD 590 value was measured as their biofilm formation ability (Caro-Astorga et al., 2020). All the samples were analyzed in triplicate. The wild strain LPB-18 and mutation strain Δ fenSr3 were inoculated in LB medium at 33 °C for 12 h. The spore counting was conducted by plate counting method. Briefly, 100 uL nutrient medium was heated at 65 °C for 30 min. After that, the samples were incubated in culture dish with LB for 24 h, and colonies were counted. Each sample was analyzed in triplicate.

Statistical analysis

One-way analysis of variance (ANOVA) in SPSS software (SPSS, ver. 17.0, IBM, Armonk, NY) was used for statistical analysis. P values were given with the results of ANOVA. The test of Duncan was performed to detect the significance below 0.05. All results were conducted with a triplicate of experiment and mean standard deviation (SD).

The characteristic of sRNA FenSr3 and its role in biomass and fengycin production in B. amyloliquefaciens

FenSr3 was identified in B. amyloliquefaciens as a ∼355 nt sRNA that is considered an important regulatory factor in fengycin biosynthesis. The sRNA is transcribed from an intragenic region of fenC, which was the biosynthesis gene of fengycin. The genomic context of this sRNA gene is highly conserved in the family of B. subtilis (Fig. S1).

The strains were cultivated and regularly sampled to investigate the effect of FenSr3 on the phenotype of B. amyloliquefaciens. Fig. 1B displays the result of the determination of dry cell weight. The strain LPB-18 exhibited a standard S-shaped growth curve, which reached a stable phase after about 20 h. After 50 h, the biomass of the strain LPB-18N increased persistently, reaching a maximum biomass value of 6.8 g/L. Additionally, regarding the LPB-18N culture was worth noting that the biomass increased slowly at approximately 12 h and exhibited a more extended logarithmic growth than strain LPB-18 and LPB-18P. As shown in Fig. 1A, fengycin production was significantly increased in B. amyloliquefaciens LPB-18N, from 190.464 to 327.5 ± 2.6 mg/L. The strain LPB-18P showed a decreased fengycin concentration of 38.6 ± 9.8 mg/L. In order to compare biofilm development between wild-type and mutant strains, which were consistent with the antagonistic activities of B. amyloliquefaciens against F. oxysporum in PDA (Figs. 1G–1I). It has already been reported that B. amyloliquefaciens could suppress F. oxysporum and other filamentous fungi (Wang et al. 2022). The wild-type strain LPB-18 and the mutant exhibited less antagonism against F. oxysporum than the fenSr3-deficient strain B. amyloliquefaciens LPB-18N, as indicated by clear zones in the inoculated bacteria culture. To compare biofilm development between wild-type and mutant strains, the crystal violet assay was performed to evaluate biofilm formation. The wild-type strain and the strain LPB-18P formed moderate biofilms with OD590 nm equivalent to 0.81 and 0.52 in TSB. The strain LPB-18N formed robust biofilms and recorded OD590 nm equivalent to 2.09 (Fig. 1C). Under high magnification, B. amyloliquefaciens LPB-18 demonstrated the 3D structure of fold cell surface morphology (Fig. 1D). The micrographs of cell surface also revealed the presence of appendages which were conducive to make the interconnection of biofilm cells and physical surface (Fig. 1E). On the other hand, the strain LPB-18P exhibited less complex and smooth cellular surface (Fig. 1F).

Transcriptome analysis of B. amyloliquefaciens LPB-18N in response to fengycin biosynthesis

A comparative transcriptome analysis was performed between strain LPB-18 and strain LPB-18N to comprehensively research the gene expression global changes. As shown in Fig. 2A, the correction between the samples was greater than 0.96 in B. amyloliquefaciens LPB-18 and B. amyloliquefaciens LPB-18N, which were analyzed by the algorithm of Pearson Correlation Coefficient. This finding demonstrated high repeatability of sample Parallelism and reliability of the transcriptome data. The correction of strain LPB-18 vs. strain LPB-18N was 0.65, indicating that sRNA FenSr3 significantly affects the gene expression of B. amyloliquefaciens LPB-18. The statistical power of this experimental design, calculated in RNASeqPower, is 0.999882 (https://rodrigo-arcoverde.shinyapps.io/rnaseq_power_calc/). Subsequently, DgeR 4 software was used to summarize difference expression genes (DEGs). According to the criteria of FDR<0.05 and —log₂FC—>1, 1037 DEGs were screened in B. amyloliquefaciens LPB-18N, which included 548 genes were significantly upregulated, and 489 genes were remarkably downregulated. As shown in Fig. 2B, the volcano plots were created to exhibit the different expression genes in wild-type strain LPB-18 and strain LPB-18N, with upregulated genes in red and downregulated genes in green.

For a comprehensive describing gene function, P < 0.05 was considered the threshold for significant enrichment, and DEGs were enriched in 20 pathways involving 267 genes: 17 pathways for cell metabolic processes, 2 pathway refer to cell processes and 1 pathway participates in environmental signal sensing. (As shown in Fig. 3). In a nutshell, the DEGs enrichment in the biological process includes arginine biosynthesis (ko00220), valine, leucine, and isoleucine biosynthesis (ko00290), flagellar assembly (ko02040), fatty acid biosynthesis (ko02040), biosynthesis of antibiotics (ko01130), secondary metabolites (ko01110) and carbon metabolism (ko01200). The molecular functions enriched for identified DEGs include DNA replication (ko03030), RNA synthesis and degradation (ko03018 and ko03020), and aminoacyl-tRNA biosynthesis (ko00970), and base excision repair (ko03410). The main cellular components enriched for the identified DEGs include cellular protein-containing complexes (ko02040), ribosomes(ko00750), translocator (ko02010 and ko03060), non-ribosomal peptide structures (ko01054), and cell growth and death (ko04112).

Metabolic pathway analysis of DEGs in the biosynthesis of fengycin

A global fengycin biosynthesis pathway was constructed in B. amyloliquefaciens LPB-18N, as shown in Fig. 4. Changes in individual metabolic processes were analyzed and shown with a metabolic pathway map. A unitive theme of the metabolic processes during fengycin synthesis development was an upregulation of energy-generating and precursor biosynthetic pathways in B. amyloliquefaciens LPB-18N. The glycolytic and the tricarboxylic acid (TCA) cycle provide multiple precursors for cellar metabolism and generate a large amount of NADH/NADPH. These positive alterations in metabolite levels are primarily correlated with carbon metabolism and fatty acid synthesis. Specifically, the results revealed the acoA-acoB-acoC-citZ operon (encoding pyruvate dehydrogenase system) exhibited an increase in transcript abundance that providing enough precursor in the TCA cycle and branched fatty acid biosynthesis. It is important that pyruvate is a pivotal metabolite linking amino acids and branched fatty acid biosynthesis (Xu et al., 2020). Upregulation of the TCA cycle was consistent with increasing levels of de novo nucleotide biosynthesis, including ATP, CTP, and GTP, which could increase the concentration of extracellular DNA (eDNA), a key component of biofilm. The effects of the TCA cycle on promoting biofilm formation had detailed in B. subtilis (Pisithkul et al., 2019), and it is consistent with transcript sequencing and biofilm phenotype in B. amyloliquefaciens LPB-18N. Besides, a few intracellular branched amino acids also displayed an increasing trend in the transcript, including Orn, Thr, Asp, Glu, Ser, Leu, and Val. The ¹³C-tracer experiments demonstrated the carbon could incorporate into various amino acids and acetylated amino acids at 16 to 24 h in biofilm formation in B. subtilis (Pisithkul et al., 2019). The phenotypic experiments and transcript sequencing reveal that carbon cycle may be critical for providing precursors and energy to the cellular process involved in fengycin and biofilm biosynthesis.

Fatty acids are another natural constituent of fengycin, making up the hydrophobic part of fengcyin. A higher expression level of acyl-carrier protein synthase provides enough acyl-carrier protein (ACP), which is dominant in the biosynthesis of fatty acids. Acetyl-CoA is an essential initiator and precursor in the biosynthesis of fatty acids. Interestingly, the genes that participate in the conversion of pyruvate with acetyl-CoA and generation were up-regulated in B. amyloliquefaciens LPB-18N, which would be essential for biosynthesizing and assembling fengycin. The deletion of fenSr3 could affect the synthesis of non-ribosomal peptides, in which fengycin synthesis genes of fenA, fenC, and fenD upregulate 8.91 folds, 7.03 folds, and 8.82 folds, respectively. In addition, the knockout of fenSr3 could affect carbohydrate metabolism, amino acid, and lipid biosynthesis, which was manifested in better fengycin production and higher expression of genes encoding transport protein, such as ABC transporters (znuB, yhcB, opuBA, and tcyA) and lipid transport protein (estA, lcfB, plsC). And the global regulation factor abrB (−4.7 folds) was significantly downregulated. According to transcriptome results, we also found that the comX, phrA, and phrC genes were remarkably upregulated, participating in environmental responses in quorum sensing systems. The deletion of non-coding RNA FenSr3 also enhanced the DNA replication and transcription, manifested in responding gene-upregulated including dnaG, dnaE, rnhB, ygzB, and hupA. After the absence of components that promote the adaptation of bacteria to the challenging environment, bacteria could survive in harsh conditions by enhancing their metabolism and self-repair, which may be a self-defense behavior of bacteria.

Differentially expressed genes associated with a knockout of fenSr3 in B. amyloliquefaciens LPB-18N

The huge distinction in gene expression between strains LPB-18 and LPB-18N was analyzed by metabolic pathway analysis, including spore, biofilm, and fengycin biosynthesis (Fig. 5). According to transcriptome data, the coding gene abrB was down-regulation (−2.1 folds) in B. amyloliquefaciens LPB-18N. The opposite of fengycin biosynthesis genes fenA, fenC, and fenD exhibited a remarkable increase in deletion strain of fenSr3, which encoded fengycin synthetases FenA, FenC, and FenD, respectively. These results were consistent with qPCR, as shown in Fig. 6. Fengycin has an amphipathic structure comprised of a cycle peptide and a hydrophobic fatty acid chain. In addition to the non-ribosomal peptide synthetases (NRPSs), it is also essential to synthesize amino acid and fatty acid residues. In the transcriptome data, nine genes involved in carbon metabolism, including three rate-limiting enzyme genes of glycolysis and tricarboxylic acid cycle (TCA): pdhD, citA, and mdh, which encodes dihydrolipoamide dehydrogenase E3, citroyl synthetase and malic dehydrogenase, respectively, were significantly upregulated. These metabolic processes promote energy supplements and gain.

Also, gene cluster fabG/D/Lmore precursors for the synthesis of amino acid and fatty acid. It was remarkably increased, which encodes ACP S–malonyltransferase and fatty acids synthesis, and there was a great consistency with RT-qPCR (Fig. 6). It indicates good fatty acid synthetic ability in B. amyloliquefaciens LPB-18N. The biofilm formation and sporulation were also evaluated (Figs. 5B and 5C). It shows a significant difference between wild-type LPB-18 and strain LPB-18N, which indicates that non-coding small RNA fenSr3 is essential for biofilm and sporulation.