Metabolome Analysis of Constituents in Membrane Vesicles for Clostridium thermocellum Growth Stimulation

The cultivation of the cellulolytic bacterium, Clostridium thermocellum, can have cost-effective cellulosic biomass utilizations, such as consolidated bioprocessing, simultaneous biological enzyme production and saccharification. However, these processes require a longer cultivation term of approximately 1 week. We demonstrate that constituents of the C. thermocellum membrane vesicle fraction significantly promoted the growth rate of C. thermocellum. Similarly, cell-free Bacillus subtilis broth was able to increase C. thermocellum growth rate, while several B. subtilis single-gene deletion mutants, e.g., yxeJ, yxeH, ahpC, yxdK, iolF, decreased the growth stimulation ability. Metabolome analysis revealed signal compounds for cell–cell communication in the C. thermocellum membrane vesicle fraction (ethyl 2-decenoate, ethyl 4-decenoate, and 2-dodecenoic acid) and B. subtilis broth (nicotinamide, indole-3-carboxaldehyde, urocanic acid, nopaline, and 6-paradol). These findings suggest that the constituents in membrane vesicles from C. thermocellum and B. subtilis could promote C. thermocellum growth, leading to improved efficiency of cellulosic biomass utilization.


Introduction
Cellulose is one of the most abundant organic materials on Earth. Bacteria that can grow on cellulose have been isolated from many environments that include soil, hot springs, cow rumen, termite gut, and the human intestinal tract [1]. Clostridium thermocellum (Acetivibrio thermocellus) [2], a Gram-positive thermophilic anaerobic soil bacterium, is a candidate for cellulosic biomass utilization. C. thermocellum completely degrades 4.4 g/L purified cellulose in one day [3]. It also degrades 65% of 5 g/L switchgrass in five days and 70% of 10 g/L corn hull in seven days [4,5].
One hundred microliters of C. thermocellum DSM 1313 culture was inoculated in 5 mL of CTFUD medium containing 0.5% cellobiose with the supplementation of the collected MV fraction. C. thermocellum was cultured at 60 • C under anaerobic conditions with nitrogen gas. The C. thermocellum growth was evaluated with optical density of the broth at 600 nm.

Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) Analysis of C. thermocellum MV and B. subtilis Broth
The C. thermocellum MV fraction was treated with 10 mg/L surfactin, and the filtrate obtained after ultrafiltration with Vivaspin 2-100 K (Cytiva, Marlborough, MA, USA) was used to obtain the constituents in MVs. Cell-free supernatants of B. subtilis trpC2 and trpC2 yxeJ broth were prepared by centrifugation and filtration with a 0.22-µm syringe filter. These specimens were homogenized with zirconia beads in 75% methanol, and the supernatants were collected after centrifugation at 15,000× g rpm for 10 min. The supernatants were applied to a MonoSpin C18 column (GL Science, Tokyo, Japan) and were filtered through a 0.22-µm syringe filter.

MV Constituents Promote C. thermocellum Growth
A previous study reported that the co-culture of the engineered C. thermocellum and T. saccharolyticum strains produced 3.8% ethanol from cellulose for 6 days [9]. C. thermocellum cultivation with BGL supplementation for 10 days reportedly produced 76.7 g/L glucose from alkali pretreated rice straw [18]. It seems that the growth rate of C. thermocellum is an important factor in improving the efficiency of CBP and BSES. In this study, we collected MVs from C. thermocellum broth ( Figure S1). MVs contain various compounds, such as DNA and RNA, which function in cell-cell communication. When C. thermocellum was grown in the presence of the MV fraction, the growth rate did not change.
However, when the MVs were lysed using the lipopeptide surfactin [41] the cell density of C. thermocellum had significantly increased at 24 h after the inoculation (Figure 1). The surfaction supplementation alone did not affect the C. thermocellum growth rate. The final growth yield in each sample had not changed significantly. These results suggest that the constituents in the MV fraction could promote the growth rate of C. thermocellum.
Mukamolova et al. purified the resuscitation promoting factor (Rpf) from the broth of the Gram-positive bacterium, Micrococcus luteus. The purified Rpf promoted the growth of this bacterium as well as Mycobacterium avium, M. bovis, M. kansasii, M. smegmatis, and M. tuberculosis [42]. Genes homologous to the rpf gene were found to be widespread in a number of Mycobacterium species, as well as in Gram-positive bacteria with a high GC content, such as Corynebacterium gultamicum and Streptomyces rimosus. The Rpf protein shows peptidoglycan degradation activity [43]. Shah et al. reported that muropeptide fragments released from the peptidoglycan of the Gram-positive bacterium, B. subtilis, stimulate the germination of bacterial spores. Staurosporine, which inhibits related eukaryotic kinases in bacteria, blocks muropeptide-dependent bacterial spore germination [44]. We evaluated the effect of staurosporine on C. thermocellum growth with cell-free B. subtilis broth, however no significant inhibition was observed.

B. subtilis Broth Promotes C. thermocellum Growth Rate
Cell-free B. subtilis broth containing MVs also promoted the C. thermocellum growth rate, similar to the C. thermocellum MV fraction ( Figure S1 and Figure 2a). Again, the surfaction supplementation alone did not affect the C. thermocellum growth rate (Figure 2a [42]. Genes homologous to the rpf gene were found to be widespread in a number of Mycobacterium species, as well as in Gram-positive bacteria with a high GC content, such as Corynebacterium gultamicum and Streptomyces rimosus. The Rpf protein shows peptidoglycan degradation activity [43]. Shah et al. reported that muropeptide fragments released from the peptidoglycan of the Gram-positive bacterium, B. subtilis, stimulate the germination of bacterial spores. Staurosporine, which inhibits related eukaryotic kinases in bacteria, blocks muropeptide-dependent bacterial spore germination [44]. We evaluated the effect of staurosporine on C. thermocellum growth with cell-free B. subtilis broth, however no significant inhibition was observed. We further evaluated the C. thermocellum growth promotion effect of the broth of B. subtilis genome deletion mutants [38]. All the mutants, especially six mutants in which the pdp-rocR genomic region, were deleted (MGB723, MGB773, MGB822, MGB834, MGB860, MGB874) promoted C. thermocellum growth by accelerating the growth rate ( Figure 2b, Table S1). Subsequently, we evaluated the C. thermocellum growth promotion effect of 100 B. subtilis mutants in which single genes within the pdp-rocR genomic region were deleted under a trpC2 gene deletion background (Table S2) [39]. We did not find B. subtilis mutants that promoted C. thermocellum growth more than trpC2 strain as the parent strain. Contrary to our expectation, the effect of 23 B. subtilis mutants was significantly lower than that of the parent strain ( Figure 2c). Microorganisms 2021, 9, x FOR PEER REVIEW 5 of 14  Table S1. The experiment was duplicated. (c). Evaluation of the C. thermocellum growth promotion effect of the broth of B. subtilis single-gene deletion mutants. Dark and light blue indicate significant differences compared with the effect of the parent strain (trpC2) with Student's t-test at p < 0.01 and < 0.05, respectively. The experiment was duplicated. Error bars indicate the standard error.
We further evaluated the C. thermocellum growth promotion effect of the broth of B. subtilis genome deletion mutants [38]. All the mutants, especially six mutants in which the pdp-rocR genomic region, were deleted (MGB723, MGB773, MGB822, MGB834, MGB860, MGB874) promoted C. thermocellum growth by accelerating the growth rate ( Figure 2b, Table S1). Subsequently, we evaluated the C. thermocellum growth promotion effect of 100 B. subtilis mutants in which single genes within the pdp-rocR genomic region were deleted under a trpC2 gene deletion background (Table S2) [39]. We did not find B. subtilis mutants that promoted C. thermocellum growth more than trpC2 strain as the parent strain. Contrary to our expectation, the effect of 23 B. subtilis mutants was significantly lower than that of the parent strain (Figure 2c).
Among these 23 genes, the functions of several genes have been experimentally evaluated. The asnH operon, which comprises yxbB, yxbA, yxnB, asnH, and yxaM, might be involved in the biosynthesis of asparagine [45]. The iolJ, iolG, iolF, iolE, iolC, iolB, and iolR genes in the iolABCDEFGHIJ and iolRS operon are responsible for myo-inositol catabolism involving multiple and stepwise reactions [46][47][48]. We observed a slight growth inhibition of C. thermocellum in the presence of myo-inositol, however this required a high concentration (1 mg/mL) of myo-inositol ( Figure S2). YydF is predicted to be an exported and modified peptide that has antimicrobial and/or signaling properties [49,50]. YxaL, which contains a repeated pyrrolo-quinoline quinone (PQQ) domain that forms a beta-propeller structure, interacts with the DNA helicase PcrA in B. subtilis [51]. Kim et al. reported that treatment of Arabidopsis thaliana and Oryza sativa L. seeds with 1 mg/L Among these 23 genes, the functions of several genes have been experimentally evaluated. The asnH operon, which comprises yxbB, yxbA, yxnB, asnH, and yxaM, might be involved in the biosynthesis of asparagine [45]. The iolJ, iolG, iolF, iolE, iolC, iolB, and iolR genes in the iolABCDEFGHIJ and iolRS operon are responsible for myo-inositol catabolism involving multiple and stepwise reactions [46][47][48]. We observed a slight growth inhibition of C. thermocellum in the presence of myo-inositol, however this required a high concentration (1 mg/mL) of myo-inositol ( Figure S2). YydF is predicted to be an exported and modified peptide that has antimicrobial and/or signaling properties [49,50]. YxaL, which contains a repeated pyrrolo-quinoline quinone (PQQ) domain that forms a beta-propeller structure, interacts with the DNA helicase PcrA in B. subtilis [51]. Kim et al. reported that treatment of Arabidopsis thaliana and Oryza sativa L. seeds with 1 mg/L purified YxaL was effective in improving root growth [52]. PQQ, which was first recognized as an enzyme cofactor in bacteria, displays bioactivities for various eukaryotes and prokaryotes. For many bacterial species, PQQ has growth stimulation effect and serves as a cofactor for a special class of dehydrogenases/oxidoreductases [53]. PQQ has been described as an essential growth factor for various microbes [54][55][56]. We observed a slight C. thermocellum growth promotion effect by PQQ. This effect was not enough to explain the effect of B. subtilis broth ( Figure S3). More than 50 proteins are involved in B. subtilis spore coat assembly. Of these, YxeE is an inner spore coat protein [57,58]. ahpC encodes thiol-specific peroxidase that plays a role in protecting cells against oxidative stress by detoxifying peroxides [59]. Utilization of a hydroxamate siderophore, ferrioxamine, requires the FhuBGC ABC transporter together with a ferrioxamine-binding protein, YxeB [60]. A range of siderophores can act as growth factors for various previously uncultured bacteria [61]. YxdK is assumed to be a subunit of the two-component sensor histidine kinase, with its potential cognate response regulator, YxdJ [62]. Co-cultivation with B. subtilis allows the growth of Synechococcus leopoliensis CCAP1405/1 on solid media. However, the yxdK deletion mutant reportedly loses this ability [63]. The yxeK gene, which encodes FAD-dependent monooxygenase, contributes to the metabolism of S-(2-succino)cysteine to cysteine [64].

Metabolome Analysis of the Constituents in C. thermocellum MV and B. subtilis Broth
We collected the constituents in C. thermocellum MVs and analyzed them using LC-MS/MS. Among the 534 detected peaks, the intensities of seven peaks were significantly higher in the fraction where MVs had been disrupted by surfactin compared to MVs not disrupted using surfactin (Table S3). The structure of five significantly detected compounds in surfactin-treated C. thermocellum MVs specimen can be estimated by MS/MS analysis ( Table 1 and Table S5). An aliphatic compound with the chemical formula C 12 H 22 O 2 was specifically detected in surfactin-treated C. thermocellum MVs (Table 1). Cis-2-decenoic acid was reported to decrease persister formation and revert dormant cells to a metabolically active state. Wang et al. demonstrated that three medium-chain unsaturated fatty acid ethyl esters (ethyl trans-2-decenoate, ethyl trans-2-octenoate, and ethyl cis-4-decenoate) decreased persister formation in Escherichia coli, P. aeruginosa, and Serratia marcescens, suggesting that fatty acid ethyl esters disrupt bacterial dormancy [65].
Some aliphatic acids function as diffusible signal factors (DSFs). These include cis-11-methyl-2-dodecenoic acid from Xanthomonas campestris and cis-2-dodecenoic acid from Burkholderia cenocepacia, among others [66]. DSFs are synthesized by and interact with a diverse group of microbes, including fungi, suggesting a broad conservation of cell-cell communication among these organisms [67][68][69][70]. Mutation of the DSF biosynthesis gene in B. cenocepacia results in substantially impaired growth in minimal medium [71]. Dean et al. demonstrated that Burkholderia DSF inhibits the formation and disperses Francisella biofilms. Furthermore, Burkholderia DSF was reported to upregulate the genes involved in iron acquisition in F. novicida, which increased siderophore production [72].
Subsequently, we compared the metabolites in the broth of B. subtilis trpC2 and trpC2 yxeJ (Figure 2). Among the 3150 detected peaks, the intensities of 40 peaks were significantly higher in the broth of B. subtilis trpC2 compared to that of trpC2 yxeJ (Table S4). The structures of 32 significantly detected compounds in B. subtilis trpC2 broth were estimated by MS/MS analysis ( Table 2 and Table S5). Diverse peptides were detected in B. subtilis trpC2 broth. Nicotinamide reportedly enhances growth of both Gram-negative and Gram-positive bacteria, such as M. avium, Propionibacterium acnes, S. aureus, and B. macerans [73][74][75][76]. Indole-3-carboxaldehyde was shown to efficiently inhibit biofilm formation by Vibrio cholerae O1 [77]. The utilization of urocanic acid by Pseudomonas and Aeromonas strains has been reported [78,79]. Nopaline is a carbon and nitrogen source metabolized by Agrobacterium. 6-Paradol was reported to have significant anti-adhesive activity against S. aureus [80].   In this study, we demonstrated that constituents in membrane vesicles significantly promoted the growth rate of C. thermocellum. Additionally, the MV constituents with growth stimulation were described by LC-MS/MS analysis. These findings suggest that the constituents in membrane vesicles could promote C. thermocellum growth, leading to improved efficiency of cellulosic biomass utilization.
Supplementary Materials: The following are available online at https://www.mdpi.com/2076-2 607/9/3/593/s1, Figure S1: MVs from C. thermocellum and B. subtilis. Figure S2: Effect of myoinositol on C. thermocellum growth. Figure S3: Effect of pyrrolo-quinoline quinone on C. thermocellum growth. Table S1: Genotypes of B. subtilis genome deletion mutants. Table S2: B. subtilis single gene deletion mutants used in this study. Table S3: Intensities of detected peaks in the MV fraction of C. thermocellum by LC-MS/MS. Table S4: Intensities of the detected peaks in cell-free B. subtilis trpC2 broth by LC-MS/MS. Table S5: Structures of constituents detected by LC-MS/MS in this study.
Author Contributions: Conceptualization, investigation, methodology, writing, review, editing, project administration, funding acquisition, S.I.; investigation, methodology, review, Y.T.; conceptualization, review. S.K. All authors have read and agreed to the published version of the manuscript.