Arabinose Plays an Important Role in Regulating the Growth and Sporulation of Bacillus subtilis NCD-2

A microbial fungicide developed from Bacillus subtilis NCD-2 has been registered for suppressing verticillium wilt in crops in China. Spores are the main ingredient of this fungicide and play a crucial role in suppressing plant disease. Therefore, increasing the number of spores of strain NCD-2 during fermentation is important for reducing the cost of the fungicide. In this study, five kinds of carbon sources were found to promote the metabolism of strain NCD-2 revealed via Biolog Phenotype MicroArray (PM) technology. L-arabinose showed the strongest ability to promote the growth and sporulation of strain NCD-2. L-arabinose increased the bacterial concentration and the sporulation efficiency of strain NCD-2 by 2.04 times and 1.99 times compared with D-glucose, respectively. Moreover, L-arabinose significantly decreased the autolysis of strain NCD-2. Genes associated with arabinose metabolism, sporulation, spore resistance to heat, and spore coat formation were significantly up-regulated, and genes associated with sporulation-delaying protein were significantly down-regulated under L-arabinose treatment. The deletion of msmX, which is involved in arabinose transport in the Bacillus genus, decreased growth and sporulation by 53.71% and 86.46% compared with wild-type strain NCD-2, respectively. Complementing the mutant strain by importing an intact msmX gene restored the strain’s growth and sporulation.


Introduction
Plant soil-borne diseases such as verticillium wilt and fusarium wilt cause significant losses in plant production and are very difficult to control.Microbial fungicides using living microorganisms as active compounds are effective and environmentally friendly methods to suppress plant soil-borne diseases and reduce the amount of chemical fungicides needed [1,2].The control capabilities of biocontrol agents are influenced by their concentration in the plant rhizosphere.Therefore, it is important to increase the application dose of biocontrol agents.
B. subtilis is an important resource for developing microbial fungicides due to its ability to produce a variety of antibiotics and form highly resistant spores [3].Wettable powders and dry powder seed-coating agents are the main formulations of microbial fungicides for suppressing plant soil-borne diseases.However, processing these two formulations involves an instantaneous high temperature of as high as 170 • C to dry the bacteria, and only the spores can survive under such high temperatures.Therefore, microbial fermentation should consider the bacterial concentration as well as the sporulation [4].An ideal fermentation system first increases the fermentation level of the bacterium and, subsequently, carries out the maximum possible conversion of the bacteria into spores via nutrient regulation and other methods [5].
The carbon and nitrogen sources in a medium are the main factors affecting bacterial growth and sporulation [6][7][8].Suitable and sufficient carbon and nitrogen sources can promote the growth of the bacterium, but sporulation generally occurs in unfavorable environments, such as nutrient starvation [9].During the stable phase of bacterial growth, residual carbon and nitrogen sources in the medium might inhibit the sporulation of Bacillus spp.[10].In microbes, carbohydrates are catabolized into pyruvate, which enters the tricarboxylic acid cycle, mainly through the glycolytic pathway (EMP) and pentose phosphate pathway (PPP).By comparing the effects of the intermediate products in the EMP and PPP pathways on the sporulation of B. subtilis, it was found that the sugars existing in the PPP pathway but not in the EMP pathway could increase the sporulation of Bacillus.Therefore, it was concluded that the PPP pathway is an important carbohydrate catabolic pathway that affects sporulation [11].In addition, metal ions such as Ca 2+ , Mg 2+ , Zn 2+ , and Mn 2+ also affect the sporulation of B. subtilis, which is promoted via the addition of appropriate concentrations of the metal ions in the medium [12][13][14].Different strains have different nutrient requirements suitable for their growth and sporulation; therefore, specific nutrients must be explored for a specific strain.
Bacillus subtilis NCD-2 showed a promising biocontrol effect against plant soil-borne diseases and was developed as a commercial microbial fungicide against cotton verticillium wilt in China [15][16][17].This study aimed to screen the nutrients suitable for the growth and sporulation of strain NCD-2 and then explore the mechanism for regulating sporulation via these nutrients.The results of this study will provide important information for the large-scale and efficient fermentation of strain NCD-2.

Screening of Nutrients That Facilitate the Metabolism of Strain NCD-2
The metabolic activities of strain NCD-2 for carbon, nitrogen, phosphorus, sulfur, and trace elements were determined using Biolog Phenotype MicroArray (PM) technology (Figure S1).Regarding carbon source utilization, strain NCD-2 had higher metabolic activity under L-arabinose, D-arabinose, D-xylose, D-ribose, and D-glucosamine treatment.Regarding nitrogen source utilization, strain NCD-2 had higher metabolic activity under cysteine treatment.Regarding phosphorus and sulfur source utilization, strain NCD-2 had low metabolic activities in 59 phosphorus sources and 35 sulfur sources.

Effects of L-Arabinose, D-Ribose, and D-Xylose on Growth and Sporulation Efficiency of Strain NCD-2
The effects of different carbohydrates on the growth of strain NCD-2 were evaluated (Figure 1a).When D-glucose was used as a carbon source, the bacterial concentration of strain NCD-2 reached a maximum of 2.55 × 10 7 CFU/mL, while that using L-arabinose was 2.04 times higher, at 5.20 × 10 7 CFU/mL.When D-ribose and D-xylose were used as sole carbon sources, the bacterial concentrations of strain NCD-2 reached maximums of 3.90 × 10 7 CFU/mL and 3.10 × 10 7 CFU/mL, respectively.It was revealed that L-arabinose was the most suitable carbon source to promote the growth of strain NCD-2.
The effects of different carbohydrates on the sporulation of strain NCD-2 were evaluated (Figure 1b).Using D-glucose, L-arabinose, D-ribose, and D-xylose as sole carbon sources, the sporulation efficiencies of strain NCD-2 were 2.33%, 27.12%, 11.40%, and 15.39% after 24 h of inoculation, respectively.The sporulation efficiencies of strain NCD-2 were 41.35%, 82.18%, 20.26%, and 20.45% after 48 h of inoculation when using D-glucose, L-arabinose, D-ribose, and D-xylose as carbon sources, respectively.The sporulation efficiencies of strain NCD-2 increased from 24 h to 48 h of inoculation.The increase in sporulation with L-arabinose was also confirmed using microscopy (Figure S2).The results indicate that L-arabinose increased the sporulation of strain NCD-2 during the early and later growth stages, but D-ribose and D-xylose only increased the sporulation of strain NCD-2 in the early growth stage compared with D-glucose.The effects of different carbohydrates on the sporulation of strain NCD-2 were evaluated (Figure 1b).Using D-glucose, L-arabinose, D-ribose, and D-xylose as sole carbon sources, the sporulation efficiencies of strain NCD-2 were 2.33%, 27.12%, 11.40%, and 15.39% after 24 h of inoculation, respectively.The sporulation efficiencies of strain NCD-2 were 41.35%, 82.18%, 20.26%, and 20.45% after 48 h of inoculation when using D-glucose, L-arabinose, D-ribose, and D-xylose as carbon sources, respectively.The sporulation efficiencies of strain NCD-2 increased from 24 h to 48 h of inoculation.The increase in sporulation with L-arabinose was also confirmed using microscopy (Figure S2).The results indicate that L-arabinose increased the sporulation of strain NCD-2 during the early and later growth stages, but D-ribose and D-xylose only increased the sporulation of strain NCD-2 in the early growth stage compared with D-glucose.

Effects of Different Proportions of L-Arabinose and D-Glucose on Sporulation of Strain NCD-2
The effects of different proportions of L-arabinose and D-glucose on the sporulation of strain NCD-2 were evaluated (Figure 2).Forty-eight hours after inoculation, the sporulation efficiencies of strain NCD-2 were 41.35% and 82.43% when using D-glucose and Larabinose as sole carbon sources, respectively.The sporulation efficiency of strain NCD-2 gradually decreased with the decrease in the L-arabinose proportion in a mixture of Larabinose and D-glucose.The sporulation efficiencies of strain NCD-2 were 68.82%, 56.62%, and 32.79% when L-arabinose and D-glucose were at proportions of 2:1, 1:1, and 1:2, respectively.

Effects of Different Proportions of L-Arabinose and D-Glucose on Sporulation of Strain NCD-2
The effects of different proportions of L-arabinose and D-glucose on the sporulation of strain NCD-2 were evaluated (Figure 2).Forty-eight hours after inoculation, the sporulation efficiencies of strain NCD-2 were 41.35% and 82.43% when using D-glucose and L-arabinose as sole carbon sources, respectively.The sporulation efficiency of strain NCD-2 gradually decreased with the decrease in the L-arabinose proportion in a mixture of L-arabinose and D-glucose.The sporulation efficiencies of strain NCD-2 were 68.82%, 56.62%, and 32.79% when L-arabinose and D-glucose were at proportions of 2:1, 1:1, and 1:2, respectively.

Transcriptome Analysis
The effects of L-arabinose and D-glucose on the gene expression of strain NCD-2 were compared using transcriptome sequencing.Compared with D-glucose, a total of 1483, 1773, and 2271 differential expression genes (DEGs) were identified under L-arabinose treatment at 8 h, 12 h, and 16 h after inoculation, respectively (Figure 3).GO annotations revealed that the DEGs associated with sporulation (GO:0043934), spore walls

Confirmation of Transcriptional Results with qRT-PCR
To verify the transcriptome results, 14 DEGs associated with arabinose transportation, sporulation, spore resistance to heat, etc., were selected to analyze the expression under L-arabinose treatment with qRT-PCR.The results reveal that all 14 genes showed consistent expression trends with transcriptome analysis, indicating the transcriptome results were reliable and could be used for further experimental analysis (Table S1).

Analysis of Genes Associated with Sporulation in Strain NCD-2
The Venn plot shows that 717 genes were significantly differentially expressed at all three time points (Figure S3), and most of the up-regulated genes were involved in sporulation (Table 1).Among them, sigK was a transcriptional regulator-encoding spore formation.cotE, cotF, cotG, cotS, cotT, cotV, cotW, cotX, and yheD were responsible for encoding spore coat proteins.dpaA and dpaB were responsible for synthetases-encoding pyridine dicarboxylic acid (DPA), a substance within the spore core.gerBA, gerE, gerQ, and gerT were responsible for encoding spore germination proteins.spoIIIAH and spoIIQ encoded polymeric complexes that connected the forespore and mother cell [18].spoIVB and

Confirmation of Transcriptional Results with qRT-PCR
To verify the transcriptome results, 14 DEGs associated with arabinose transportation, sporulation, spore resistance to heat, etc., were selected to analyze the expression under L-arabinose treatment with qRT-PCR.The results reveal that all 14 genes showed consistent expression trends with transcriptome analysis, indicating the transcriptome results were reliable and could be used for further experimental analysis (Table S1).

Analysis of Genes Associated with Sporulation in Strain NCD-2
The Venn plot shows that 717 genes were significantly differentially expressed at all three time points (Figure S3), and most of the up-regulated genes were involved in sporulation (Table 1).Among them, sigK was a transcriptional regulator-encoding spore formation.cotE, cotF, cotG, cotS, cotT, cotV, cotW, cotX, and yheD were responsible for encoding spore coat proteins.dpaA and dpaB were responsible for synthetases-encoding pyridine dicarboxylic acid (DPA), a substance within the spore core.gerBA, gerE, gerQ, and gerT were responsible for encoding spore germination proteins.spoIIIAH and spoIIQ encoded polymeric complexes that connected the forespore and mother cell [18].spoIVB and spoIVFB encoded the activating proteins of protease and Sig-K, which catalyzed the formation of the spore cortex, respectively.sspA, sspB, sspD, and sspE encoded small acid-soluble proteins (SASPs) associated with spore resistance to heat.sdpC encoded a cannibalism factor that delayed sporulation in B. subtilis and was significantly downregulated at all three time points.

Discussion
Plant soil-borne diseases such as verticillium wilt and fusarium wilt are difficult to suppress with chemical fungicides, rotation, and resistant varieties, mainly due to phytopathogens that can survive for decades in soil [19].Recent studies have revealed that applying microbial fungicides can successfully suppress plant soil-borne diseases, and the genus Bacillus is a major resource for developing microbial fungicides [20].Spores of Bacillus species are resistant to stresses and are used as the key ingredient in the formulation of microbial fungicides.Generally speaking, the control effect of plant soil-borne diseases is positively correlated with the population of bacteria colonized in the plant rhizosphere [21].To obtain an ideal biocontrol effect, the amount of spores applied to the soil should be increased as much as possible.Therefore, promoting the yield of spores during the fermentation process of Bacillus is important for reducing the cost and ensuring the wide application of micro-fungicides.Promoting the growth of bacteria and the yield of spores during the fermentation process is one of the key factors in reducing the cost of bio-fungicides.It is known that the sporulation process and final spore yield depend on carbohydrates and amino acids [22].The combined effects of yeast extract, peptone, and glucose enhanced the spore yield of B. megaterium [23].Likewise, the addition of glucose and ribose into the sporulation medium increased the spore yields of B. subtilis and B. cereus [6,11,24].In this study, we focused on nutrients that promoted strain NCD-2's growth and spore formation, which required screening for a large number of nutrients due to different strains having different nutrient requirements.Phenotype MicroArrays (Biolog) are commercially available microplate assays that can be used to test more than 1000 phenotypic traits simultaneously by recording the microorganism's respiration over time on many distinct substrates [25,26].Therefore, PMs can be used to screen nutrients suitable for the catabolism of a specific organism quickly and in high throughput, which has the advantages of producing a large amount of information and saving time [27].In this way, the catabolic capability of strain NCD-2 in approximately 200 carbon sources, 400 nitrogen sources, and 100 phosphorous and sulfur sources was determined using PMs.The results show that strain NCD-2 had a higher metabolic capacity with L-arabinose, D-xylose, and D-ribose as sole carbon sources, among which L-arabinose significantly increased the bacterial concentration and sporulation efficiency of strain NCD-2.
B. subtilis could grow on a medium with L-arabinose as the sole carbon and energy source.Without L-arabinose, the AraR protein bounded to a site within the araABDLMNPQ-abfA operon promoter region, preventing transcription.With L-arabinose, a conformational change was induced in AraR, such that recognition and binding to DNA were no longer possible, and the operon could be expressed [28].After entering the cell, L-arabinose was sequentially converted to L-ribulose, L-ribulose 5-phosphate, and D-xylulose 5-phosphate via the action of L-arabinose isomerase (encoded by araA), L-ribulokinase (encoded by araB), and L-ribulose-5-phosphate 4-epimerase (encoded by araD), respectively.D-xylulose 5-phosphate was further catabolized through the pentose phosphate pathway [29,30].Transcriptome analysis showed that L-arabinose strongly up-regulated araABDLMNPQ-abfA operon expression in strain NCD-2 (accession number: SUB12858722), and genes associated with sporulation were also strongly up-regulated (Table 1).Moreover, compared with using L-arabinose as the sole carbon source, the sporulation efficiency of strain NCD-2 significantly decreased when both glucose and arabinose were present in the M9 medium (Figure 2), which might be because glucose repressed the expression of both araE, a gene for the L-arabinose transporter, and the ara operon at the transcriptional level [31,32].Thus, L-arabinose might be involved in regulating the expression of genes related to sporulation in strain NCD-2 by regulating the ara operon.
Previous studies found that the AraNPQ-MsmX system was involved in the transport of arabinans, and knocking out araNPQ reduced the growth rate of B. subtilis [28,33].Therefore, deleting the msmX gene, which encodes ATPase to provide energy to the AraNPQ transporter, inevitably led to a decrease in the growth rate of B. subtilis.In this study, msmX was deleted from wild-type strain NCD-2, and the mutant reduced the bacterial growth and sporulation efficiency of strain NCD-2 with L-arabinose as the sole carbon source (Figure 6)-which is consistent with the previous study-but not with D-glucose as the sole carbon source.L-arabinose entered the metabolic process of strain NCD-2 via the AraNPQ-MsmX system, affecting its growth and sporulation.These results provide knowledge for effectively improving growth and spore production during the fermentation of strain NCD-2.
Autolysis of B. subtilis was observed during fermentation, resulting in a large number of cell deaths and reducing the bacterial fermentation concentration [34].Many factors led to the autolysis of the bacterium [35][36][37][38].Among them, a phenomenon of "cannibalism" was described [39,40], in which the master regulator of sporulation Spo0A was active and released two toxins, Skf and SdpC, to kill Spo0A-inactive sister cells.The nutrients released by the dead cells were used for the growth of cells that were not yet committed to sporulating.In this study, it was observed that cell autolysis produced a large amount of cell debris in the medium with D-glucose as the sole carbon resource, but not in the medium with L-arabinose.In the transcriptome analysis, L-arabinose significantly down-regulated the transcription of sdpC compared with D-glucose at 8 h, 12 h, and 16 h post-inoculation.Therefore, we speculated that L-arabinose increased the bacterial concentration of strain NCD-2 by inhibiting the process of "cannibalism" in strain NCD-2.

Bacterial Strains and Growth Conditions
The strains used in this study are listed in Table 2. B. subtilis strains were stored at −80 • C in LB medium containing 30% glycerol.Strain CPmsmX was inoculated in M9 medium containing chloramphenicol (Sangon Biotech, Shanghai, China) at a final concentration of 5 µg•mL −1 and cultured at 30 • C under continuous agitation (180 rpm).

Biolog Phenotype MicroArray Analysis
The metabolic phenotype of B. subtilis strain NCD-2 in 755 nutrients was evaluated using the Biolog Phenotype MicroArray system (Biolog, Hayward, CA, USA).Ninetysix-well PM1-8 MicroPlates TM (Biolog, Hayward, CA, USA), including a carbon source (PM1 or PM2A), a nitrogen source (PM3B or PM6-8), and a phosphorus source and sulfur source (PM5), were assayed.The names of nutrients are described in the literature of Bochner et al. [25].The experiment was conducted according to the procedures developed by the manufacturer [27].Briefly, strain NCD-2 was grown overnight at 33 • C on BUG + B plates (Biolog, Hayward, CA, USA).A single colony was selected, grown on BUG + B plates again, under the same conditions.Cells were picked up from the plates with sterile cotton swabs and transferred into a sterile, capped tube containing 20 mL of inoculation fluid (IF-0a, Biolog, Hayward, CA, USA).Cell density was adjusted to 81% transmittance on the Biolog turbidimeter.The PM1-8 MicroPlates TM were inoculated with the cell suspension (100 µL/well), and then incubated at 30 • C for 48 h in the OmniLog incubator (Biolog, Hayward, CA, USA).The plates were scanned every 15 min, and the results were analyzed and plotted using the OmniLog software (OL_FM 12 analysis package, v1.2) at the end of the incubation.

Determination of Cell Concentration and Sporulation Efficiency
Strain NCD-2 was inoculated in Luria-Bertani (LB) broth and cultured at 30 • C and 180 rpm for 12 h.Cells were collected via centrifugation at 10,000 rpm for 2 min and adjusted to OD 600 = 1.0 with sterile water.The cell suspension was inoculated in 100 mL M9 medium (12.8 To evaluate the effects of different carbohydrates on the sporulation of strain NCD-2, the D-glucose was replaced with the same concentration of L-arabinose, D-xylose, and D-ribose (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China).Strain NCD-2 was continuously cultured for 24, 36, and 48 h at 30 • C, and the sporulation efficiency was then determined following the protocol described by Eswaramoorthy et al. [41].

RNA Extraction and RNA Sequencing
Strain NCD-2 was inoculated in 100 mL of M9 medium with D-glucose or L-arabinose as the sole carbon source and cultured at 30 • C and 180 rpm for 8 h, 12 h, and 16 h.Cells were harvested via centrifugation at 4 • C and 10,000 rpm for 5 min, and three biological replicates were included.The bacterium was rapidly frozen with liquid nitrogen and stored at −80 • C. The total RNA of the collected bacteria was extracted according to the instructions of the RNAprep Pure Cell/Bactria Kit (TianGen Biotech, Beijing, China), and the quality and concentration of total RNA were measured with the NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA).The cDNA library construction and RNA sequencing (RNA-seq) were performed with the Illumina platform at Majorbio Co., Ltd.(Shanghai, China).

Confirmation of Transcriptome Analysis Results
To validate the transcriptome results, 14 genes (Supplementary Table S2) were selected for the expression analysis with qRT-PCR.Primers were designed using Primer Premier 5.0 software (Applied Biosystems, Waltham, MA, USA).B. subtilis strain NCD-2 was cultured in M9 medium with L-arabinose or D-glucose as the sole carbon source and cultured at 30 • C under continuous agitation (180 rpm).Cells were collected via centrifugation at 10,000 rpm for 1 min after 8 h, 12 h, and 16 h of incubation, respectively.Total RNA was extracted as described above and adjusted to 50 ng•µL −1 .The extracted total RNA was used as a template to synthesize the first-strand cDNA using TransScript ® One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China), and all the cDNA sample concentrations were diluted 3-fold with double distilled water.qRT-PCR assays using SYBR Green as a detector were performed in a StepOne™ Real-Time PCR system, conditions for amplification were according to the instructions of TransStart ® Top Green qPCR SuperMix (TransGen Biotech, Beijing, China).Three replicates were set up for each gene detection.The gyrB gene was used as the internal reference gene, and the relative change in target gene expression was calculated using the formula 2 −∆∆Ct [42].

Function Analysis of msmX Gene
To delete the msmX gene from strain NCD-2, msmX upstream fragments were amplified using primers msmX-P1 (CGAGCTCTTTCAGCGGTTCGGGTG) and msmX-P2 (GGGGTACCGATCAAAAAAACCGGACATGGGG), and msmX downstream fragments were amplified using primers msmX-P3 (GGGGTACCACCCAGCCATCTAACATCCCCC) and msmX-P4 (GCTCTAGATCCCGGTTCGATTGTGTCTG).The upstream and downstream amplification fragments were digested with Kpn I restriction enzyme, and then the two fragments were ligated with T4 DNA ligase.Using the ligation product as a template, PCR amplified with the msmX-P1 and msmX-P4 primers.The amplificon was digested with Sac I and Xba I restriction enzymes and then attached to the corresponding digestion site of the pKSV7 plasmid [43].The recombinant plasmid was transformed into strain NCD-2 via electroporation, and msmX knockout was conducted via in-frame deletion, as described by Arnaud et al. [44].The deletion of msmX (∆msmX) was confirmed with PCR and sequencing with primers msmX-P1 and msmX-P4.To complement the ∆msmX mutant, intact msmX was amplified from strain NCD-2 with primers CPmsmX-F (GGGGTACCTTATCGAATTCT-CATTTCTG) and CPmsmX-R (GCAGGTCGACATTGGAAATATGCACGAAAA), which included the Kpn I and Sal I restriction sites, respectively.The amplicon was digested with Kpn I and Sal I and inserted into pHY300PLK, which is a shuttle vector for E. coli and B. subtilis [45].The recombinant plasmid was transformed into mutant strain ∆msmX via electroporation to obtain the complemented strain (CPmsmX).All strains were cultured in M9 medium with L-arabinose as the sole carbon source, after which the bacterial concentration and sporulation efficiencies were calculated via plate-counting as described above.

Statistical Analyses
Statistically significant differences (p < 0.05) in NCD-2 CFU, sporulation efficiency, as well as the CFU and sporulation efficiency, between wild-type and mutant strains were evaluated with ANOVA using SPSS 18.0 software (SPSS, Chicago, IL, USA) followed by Tukey's post hoc test.Figures were prepared with Origin Pro 8.6 software (OriginLab Corporation, Hampton, MA, USA).

Conclusions
In this study, we used PMs technology to screen several nutrients with high metabolic activity in strain NCD-2, among which L-arabinose can significantly increase the bacterial concentration and sporulation efficiency of strain NCD-2 and repress cell autolysis.The transcriptome results show that L-arabinose up-regulated the expression of sporulationrelated genes and down-regulated the expression of cannibalism-related genes.Knocking out msmX, which is responsible for transporting arabinose, significantly decreased the bacterial concentration and sporulation efficiency of strain NCD-2 in the medium with Larabinose as the carbon source.These results will assist in the study of directed fermentation and the mechanism of regulating the sporulation of strain NCD-2.

Figure 1 .
Figure 1.Effects of different carbon sources on the bacterial concentration (a) and sporulation efficiency (b) of strain NCD-2.Columns represent the averages of four replicates; error bars show standard deviations; and different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 1 .
Figure 1.Effects of different carbon sources on the bacterial concentration (a) and sporulation efficiency (b) of strain NCD-2.Columns represent the averages of four replicates; error bars show standard deviations; and different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 2 .
Figure 2. The effects of different proportions of L-arabinose and D-glucose on the sporulation of strain NCD-2.Strain NCD-2 was cultured in M9 medium containing 4 g•L −1 of carbohydrates, and the sporulation efficiency was determined 48 h after inoculation.Columns represent the averages of four replicates; error bars show standard deviations; and different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 2 .
Figure 2. The effects of different proportions of L-arabinose and D-glucose on the sporulation of strain NCD-2.Strain NCD-2 was cultured in M9 medium containing 4 g•L −1 of carbohydrates, and the sporulation efficiency was determined 48 h after inoculation.Columns represent the averages of four replicates; error bars show standard deviations; and different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 2 .
Figure 2. The effects of different proportions of L-arabinose and D-glucose on the sporulation of strain NCD-2.Strain NCD-2 was cultured in M9 medium containing 4 g•L −1 of carbohydrates, and the sporulation efficiency was determined 48 h after inoculation.Columns represent the averages of four replicates; error bars show standard deviations; and different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 3 . 1 Figure 3 .
Figure 3. Volcano plots of transcriptomics comparison of strain NCD-2.X-axis indicates the average of log2 fold change from the replicates.Negative values indicate down-regulation, and positive values indicate up-regulation.Y-axis is -log10 padj.Dots in blue or red indicate differentially expressed genes.Dots in grey indicate proteins that are not significantly changed in gene expression.(a) Eight hours post-inoculation.(b) Twelve hours post-inoculation.(c) Sixteen hours post-inoculation.

Figure 4 .
Figure 4. GO enrichment analysis based on the differential expression genes between L-arabinose and D-glucose-cultured strain NCD-2.The X-axis indicates the number of genes classified into reg ulatory or functional categories, as depicted on the Y-axis.Columns in blue indicate down-regula tion and those in red indicate up-regulation.(a) Eight hours post-inoculation.(b) Twelve hours pos inoculation.(c) Sixteen hours post-inoculation.

Figure 4 .
Figure 4. GO enrichment analysis based on the differential expression genes between L-arabinoseand D-glucose-cultured strain NCD-2.The X-axis indicates the number of genes classified into regulatory or functional categories, as depicted on the Y-axis.Columns in blue indicate downregulation and those in red indicate up-regulation.(a) Eight hours post-inoculation.(b) Twelve hours post-inoculation.(c) Sixteen hours post-inoculation.

Figure 4 .
Figure 4. GO enrichment analysis based on the differential expression genes between L-arabinoseand D-glucose-cultured strain NCD-2.The X-axis indicates the number of genes classified into regulatory or functional categories, as depicted on the Y-axis.Columns in blue indicate down-regulation and those in red indicate up-regulation.(a) Eight hours post-inoculation.(b) Twelve hours postinoculation.(c) Sixteen hours post-inoculation.

Figure 5 .
Figure 5. KEGG enrichment analysis based on the differential expression genes between L-arabinose and D-glucose cultured strain NCD-2.The X-axis indicates the number of genes classified into regulatory or functional categories, as depicted on the Y-axis.Columns in blue indicate down-regulation and those in red indicate up-regulation.(a) Eight hours post-inoculation.(b) Twelve hours postinoculation.(c) Sixteen hours post-inoculation.

Figure 5 .
Figure 5. KEGG enrichment analysis based on the differential expression genes between L-arabinose and D-glucose cultured strain NCD-2.The X-axis indicates the number of genes classified into regulatory or functional categories, as depicted on the Y-axis.Columns in blue indicate downregulation and those in red indicate up-regulation.(a) Eight hours post-inoculation.(b) Twelve hours post-inoculation.(c) Sixteen hours post-inoculation.

Figure 6 .
Figure 6.The bacterial concentration (a) and sporulation efficiency (b) of wild-type strain NCD-2 (WT), msmX-null mutant (ΔmsmX), and its complemented strain (CPmsmX) in M9 medium with Larabinose as carbon source at 24, 36, and 48 h post-inoculation.Columns represent the averages of three replicates; error bars show standard deviations; different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Figure 6 .
Figure 6.The bacterial concentration (a) and sporulation efficiency (b) of wild-type strain NCD-2 (WT), msmX-null mutant (∆msmX), and its complemented strain (CPmsmX) in M9 medium with L-arabinose as carbon source at 24, 36, and 48 h post-inoculation.Columns represent the averages of three replicates; error bars show standard deviations; different letters indicate significant (p < 0.05) differences according to ANOVA with Tukey's post hoc test.

Table 2 .
Strains used in this study.