Microbial interaction between the succinate‐utilizing bacterium Phascolarctobacterium faecium and the gut commensal Bacteroides thetaiotaomicron

Abstract A large variety of microbes are present in the human gut, some of which are considered to interact with each other. Most of these interactions involve bacterial metabolites. Phascolarctobacterium faecium hardly uses carbohydrates for growth and instead uses succinate as a substrate. This study investigated the growth behavior of the co‐culture of the succinate‐specific utilizer P. faecium and the succinogenic gut commensal Bacteroides thetaiotaomicron. Succinate production by B. thetaiotaomicron supported the growth of P. faecium and concomitant propionate production via the succinate pathway. The succinate produced was completely converted to propionate. This result was comparable with the monoculture of P. faecium in the medium supplemented with 1% (w/v) succinate. We analyzed the transcriptional response (RNA‐Seq) between the mono‐ and co‐culture of P. faecium and B. thetaiotaomicron. Comparison of the expression levels of genes of P. faecium between the mono‐ and co‐cultured conditions highlighted that the genes putatively involved in the transportation of succinate were notably expressed under the co‐cultured conditions. Differential expression analysis showed that the presence of P. faecium induced changes in the B. thetaiotaomicron transcriptional pattern, for example, expression changes in the genes for vitamin B12 transporters and reduced expression of glutamate‐dependent acid resistance system‐related genes. Also, transcriptome analysis of P. faecium suggested that glutamate and succinate might be used as sources of succinyl‐CoA, an intermediate in the succinate pathway. This study revealed some survival strategies of asaccharolytic bacteria, such as Phascolarctobacterium spp., in the human gut.


| INTRODUC TI ON
Microbe-microbe interactions in the human gut have been increasingly recognized and analyzed in multidisciplinary fields. However, key factors in the interactions remain incompletely understood.
The microbiota of the human gut consists of a variety of microorganisms. Many are unclassified or uncultured anaerobic bacteria.
During our attempts to recover new microbes from human feces, we observed bacteria that barely use carbohydrates for growth and instead use succinate as a substrate.
Phascolarctobacterium faecium is an obligately anaerobic and Gram-negative bacterium that was first isolated from koala feces (Del Dot, Osawa, & Stackebrandt, 1993). Recently, it was reported that P. faecium abundantly colonizes the human gut (Wu et al., 2017). The functional role of P. faecium in the human gut is unknown. P. faecium utilizes succinate. It grows poorly on common blood agar, but adding succinate to the medium improves growth. In a previous study, we found that the P. faecium JCM 30894 genome lacked fumarate reductase, which is an enzyme that is necessary for the production of succinate (Ogata et al., 2019).
To determine microbial interactions in the human gut, we used P. faecium and B. thetaiotaomicron as the model organisms.
The co-culture of P. faecium and B. thetaiotaomicron has not been previously studied. The trophic interaction between the mucin-degrading bacterium Akkermansia muciniphila and the butyrate-producing bacterium Anaerostipes caccae has been described (Chia et al., 2018). The authors demonstrated the use of metatranscriptomics (RNA-Seq) as an explorative approach to study the expressional changes of A. muciniphila in response to A. caccae. We also used metatranscriptomics to explore the interaction of succinate-producing and succinate-utilizing bacteria from the human gut.

| Growth stimulation of P. faecium by B. thetaiotaomicron
Bacteroides thetaiotaomicron JCM 5827 T was streaked on one-half of an EG plate using an inoculating loop. An inoculum of P. faecium JCM 30894 was similarly streaked on another half of the same EG plate.
The monoculture of each strain was also performed as a control.

| Growth stimulation of P. faecium by the addition of succinate
Phascolarctobacterium faecium JCM 30894 from a 6-day plate culture were suspended in phosphate-buffered saline (PBS). A 1% (v/v) suspension (MacFarland standard 3 turbidity) was inoculated into Gifu Anaerobic Medium Broth (GAM Broth, Nissui Pharmaceutical Co., Tokyo, Japan) that was not supplemented or supplemented with 1% (w/v) succinate (adjusted to pH 7.0). The broth was cultured experiments that were performed in anaerobic serum bottles sealed with butyl-rubber stoppers at 37°C in an atmosphere of CO 2 and N 2 (1:9, v/v). Cultures were sampled at 0,18,20,22,24,42,44,46, and 48 hr for analysis of metabolites and measurements of optical density at 660 nm (OD 660 ). OD 660 was measured using an Ultrospec 2100 pro spectrophotometer (Amersham Biosciences, Piscataway, NJ, USA).
The pH of the medium was measured using a Twin pH compact pH meter (HORIBA, Kyoto, Japan).

| Co-culture
Co-culture experiments were performed in GAM broth using anaerobic serum bottles sealed with butyl-rubber stoppers at 37°C at the aforementioned culture conditions. B. thetaiotaomicron JCM 5827 T cells from a 2-day plate culture were suspended in PBS. Suspensions (1% v/v; MacFarland standard 3) were added to GAM broth followed by 5 hr of incubation to allow accumulation of metabolites. A 1% (v/v) suspension (MacFarland standard 3) of P. faecium JCM 30894 was then added to the B. thetaiotaomicron cultures. Cultures were sampled at 0, 5, 23, 29, 47, and 53 hr for analysis of metabolites and measurements of OD 660 . For transcriptomic analysis, bacterial pellets received after 2 days of incubation were suspended in TRIzol Reagent (Life Technologies, Carlsbad, CA, USA) and stored at −20°C until used for RNA purification. A pure culture of each strain was also incubated for 2 days.

| Analysis of metabolites
One milliliter of bacterial culture was centrifuged, and the supernatant was used for high-performance liquid chromatography (HPLC) analysis. Metabolites were quantified using an HPLC system equipped with a model SPD-M20A diode array detector model (Shimadzu, Kyoto, Japan) and a Rezex ROA-Organic acid H + (8%) column (Phenomenex, Torrance, CA, USA). The analytical conditions were as follows: eluent, 0.0025 N sulfuric acid; flow rate, 0.5 ml/min; detection, ultraviolet (UV) 210 nm; and column temperature 55°C. Succinate, propionate, and acetate were used as standards.

| RNA purification
Total RNA was isolated by using the TRIzol Max Bacterial RNA Isolation Kit (Life Technologies) and the RNeasy Mini Kit (QIAGEN, Valencia, CA, USA) as described previously (Chia et al., 2018).

| Transcriptome analysis
Illumina adapter sequences and low-quality bases were trimmed from raw fastq reads with fastp v0.20 (Chen, Zhou, Chen, & Gu, 2018). Forward and reverse reads were independently quality filtered, and qualified reads were combined into one fastq file.

| Co-culture of P. faecium and B. thetaiotaomicron
In a preliminary experiment, the growth of B. thetaiotaomicron JCM 5827 T was investigated ( Figure 3a). Based on these results, P. faecium

| Transcriptomes of co-culture of P. faecium and B. thetaiotaomicron
On average, 20 million reads were generated per sample, which is above the recommended sequence depth of 5-10 million reads for a single bacterial transcriptome (Haas, Chin, Nusbaum, Birren, & Livny, 2012). The Nx kit enabled the removal of more rRNA. The number of reads mapped to the CDSs was higher. In the co-culture sample, most of the reads were derived from B. thetaiotaomicron. An average of 9.78% (Ex kit) and 56.7% (Nx kit) of reads were mapped on the CDSs of F I G U R E 1 Growth stimulation of Phascolarctobacterium faecium by Bacteroides thetaiotaomicron. Strains were cultured on EG medium for 4 days at 37°C in an atmosphere of 1:1:8 H 2 /CO 2 /N 2 B. thetaiotaomicron, while the percentages of P. faecium CDS-mapped read count were average 0.04% (Ex kit) and 0.3% (Nx kit; Table A1).

| Highly expressed genes of P. faecium cocultured with B. thetaiotaomicron
The genes involved in the succinate pathway were expressed in mono-and co-culture. However, due to the insufficient number of reads, it was difficult to determine whether there were significant changes in the expression levels. In the co-culture, two genes en- (2-oxoacid:acceptor oxidoreductase subunit alpha), PFJ30894_ RS01160 (2-oxoacid:ferredoxin oxidoreductase subunit beta), and PFJ30894_RS01165 (pyruvate/ketoisovalerate oxidoreductase gamma subunit) were highly expressed in the co-culture (Table A2).
Moreover, genes encoding chaperones and stress response factors also exhibited larger TPM values in the co-culture than in the monoculture (Table 1).

| Differential expression between B. thetaiotaomicron in monoculture and co-culture with P. faecium
The genome of B. thetaiotaomicron possesses 4794 CDSs, of which 4786 (99.8%) were expressed in at least one sample. We used an F I G U R E 2 Growth stimulation of Phascolarctobacterium faecium by succinate. (a) P. faecium was inoculated into GAM broth that was not supplemented or supplemented with 1% (w/v) succinate. Experiments were performed in triplicate, and error bars represent the standard deviation between each biological replicate. (b) pH of the medium was measured at the time of sampling. Significant differences were determined by Student's t test. Significance was set at p < 0.05 (two-tailed). * indicates significant differences, p < 0.01. (c) 5% (w/v) succinate solution (a final concentration of 1%) was added to P. faecium cultures at 120 hr. (d) The metabolite profile was analyzed by using HPLC  was downregulated in the co-culture (Table 2). Besides, a cluster of genes encoding subunits of ATP synthase was upregulated in the monoculture (Table 3).

F I G U R E 3
Co-culture of Phascolarctobacterium faecium and Bacteroides thetaiotaomicron. (a) The pH and growth profile of B. thetaiotaomicron were investigated in a preliminary experiment. Results of (b) monoculture and (c) co-culture. B. thetaiotaomicron was inoculated at 0 h followed by P. faecium at 5 hr. (d) The pH of monoculture and co-culture was measured at the time of sampling. **significant differences p < 0.05

| D ISCUSS I ON
In this study, we investigated the growth behavior of the co-cultured P. faecium and B. thetaiotaomicron. The pK a values for succinate and propionate are 4.16 and 4.87, respectively. The pH of the culture supplemented with succinate increased from 6.6 to 7.4 after 42 hr.
As inferred from the above-mentioned pK a values for each organic acid, this is consistent with a decrease in succinate and increase in propionate. The pH of the co-culture also increased slightly from 5.0 to 5.3 after 29 hr (Figure 3d). The collective results support the view that the pH change is associated with the conversion of succinate to propionate.
Phascolarctobacterium wakonense was isolated from common marmoset feces (Shigeno, Kitahara, Shime, & Benno, 2019). These two species, as well as P. faecium, grew well in medium supplemented with succinate. This finding may be one of the characteristics of this genus. In the human gut, Phascolarctobacterium spp. convert succinate to propionate, which is a health-promoting microbial metabolite (Hosseini et al., 2011).
thetaiotaomicron changes the expression patterns of these transporter genes in response to the corrinoids provided by P. faecium. Previous experiments observed in the presence of vitamin B 12 suggested that propionate production was associated with the conservation of biologically useful energy (Strobel, 1992). P. faecium may exist in an energy-limited environment, and maximizing energy conservation during the production of propionate may be one strategy this bacterium uses to survive in the human gut. The conversion of the energy of the decarboxylation reaction into sodium ion (Na + ) gradients by methylmalonyl-CoA decarboxylase is the biological use of decarboxylation energy (Dimroth, 1987;Hilpert & Dimroth, 1982). The central energy conservation step in Propionigenium modestum (Schink & Pfennig, 1982) is the conversion of the energy of methylmalonyl-CoA decarboxylation into a Na + gradient, which in turn drives ATP synthesis via Na + -activated ATPase (Dimroth, 1997;Dimroth & Schink, 1998;Hilpert, Schink, & Dimroth, 1984). Note: Negative values indicate upregulation in the monoculture, and the positive value indicates upregulation in the co-culture.
In this study, although the difference in the expression of the genes involved in the succinate pathway of P. faecium was not clear, high expressions of two genes encoding SLC13/DASS family transporter were observed in the co-culture. One of the genes (PFJ30894_RS03075) is encoded consecutively with the gene cluster of the succinate pathway. The SLC13 transporter is part of the divalent anion:Na + symporter (DASS) family (Mulligan, Fitzgerald, Wang, & Mindell, 2014). VcINDY, an SLC13 homologue from Vibrio cholerae, couples a Na + gradient to the transport of succinate, a C 4 -dicarboxylate (Mulligan et al., 2014). Therefore, it seems that P.
faecium upregulated the SLC13 transporter to transport succinate produced by B. thetaiotaomicron into the cell.
It has been suggested that lower pH achieved in the co-culture of B. thetaiotaomicron and Bifidobacterium adolescentis could potentially slow the growth and metabolism of B. thetaiotaomicron (Das et al., 2018). Furthermore, it has been reported that B. thetaiotaomicron DSM 2079 T (=JCM 5827 T ) showed a growth rate at pH 5.5 of approximately 40% of the growth at pH 6.7 (Duncan, Louis, Thomson, & Flint, 2009). As mentioned above, the pH of the co-culture of P. faecium and B. thetaiotaomicron decreased from 7.0 to 5.0, but then increased to 5.3. Increased pH in co-culture would improve the growth of B. thetaiotaomicron. Consequently, the co-existence of these two species seems to be beneficial for each species. In the presence of P. faecium, B. thetaiotaomicron downregulated glutamate-dependent acid resistance system-related genes involved in glutaminase A and glutamate decarboxylase activity and the antiporter GadC. Strategies adopted to face acid encounters include amino acid-dependent systems (Lu et al., 2013;Pennacchietti, D'Alonzo, Freddi, Occhialini, & De Biase, 2018). In particular, the glutamate-dependent acid resistance system is extremely powerful. B. thetaiotaomicron possesses this system (Pennacchietti et al., 2018). The difference in pH values between the monoculture (pH 4.9) and co-culture (pH 5.3) resulted in the downregulation of the glutamate-dependent acid resistance system-related genes in the co-culture. On the other hand, co-cultured P. faecium highly expressed genes for chaperones and stress response factors. P. faecium did not grow in the monoculture but could grow in the co-culture. Active transcription of chaperone genes could be associated with acid stress caused by succinate and acetate produced in the co-culture. The molecular mechanisms adopted by Gram-positive and Gram-negative bacteria for coping with acid stress have been reviewed (Lund, Tramonti, & De Biase, 2014 (Foster & Hall, 1990, 1991, and Streptococcus faecalis (Kobayashi, Suzuki, & Umemoto, 1986).
In this study, B. thetaiotaomicron upregulated ATP synthase genes in the monoculture. This result is consistent with that of the aforementioned glutamate-dependent acid resistance system, which is one of the mechanisms of protection against acid stress.
B. thetaiotaomicron may release compounds including glutamate and ammonium into the medium in response to acidic stress. Therefore, it is conceivable that P. faecium also transports glutamate into the cell using have also been reported (Watanabe et al., 2012). In the co-culture, the numbers of P. succinatutens cells increase and succinate is converted to propionate. These findings may indicate one of the survival strategies of asaccharolytic Phascolarctobacterium spp. in the human gut. This idea is supported by the greater abundance of Phascolarctobacterium along with the increased abundance of Bacteroides in rats fed a high-fat diet (Lecomte et al., 2015). The abundance of B. fragilis and Bacteroides vulgatus was positively correlated with both changes in body weight and fat mass. A previous study demonstrated that P. faecium colonizes the human gut in early life and develops to a high level in healthy adults, followed by a decrease in elderly individuals (Wu et al., 2017). The authors described that elderly individuals and those <1 year of age consumed relatively less fat and had a relatively low body weight. As inferred from rat experiments described above, this may result in a decrease in Bacteroides and the decrease in the available succinate for P. faecium.
In conclusion, we reveal some survival strategies of asaccharolytic bacteria, such as Phascolarctobacterium spp., in the human gut. The encounter between P. faecium and B. thetaiotaomicron in the human gut may result in a beneficial conversion of succinate to propionate. An overview of the microbial interaction between the succinate-utilizing bacterium P. faecium and the gut commensal B.
thetaiotaomicron is shown in Figure 5.

ACK N OWLED G M ENTS
This work was supported by PRIME, the Japan Agency for

E TH I C S S TATEM ENT
None required.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used and analyzed during the current study are included in this published article.