Fecal Short-Chain Fatty Acid Concentrations Increase in Newly Paired Male Marmosets (Callithrix jacchus)

This study addressed a knowledge gap about longitudinal changes in the gut microbiome metabolites during animal pairing. This research in the laboratory common marmoset can control for the confounding factors such as diet and other environmental conditions. Phascolarctobacterium showed the highest contribution to the sex-biased transmission of the female to the male after pairing. Here, we observed the sex difference in the increase in short-chain fatty acid concentration in the feces of newly paired marmosets, which may be caused by the sex-biased gut microbiome transmission after pairing.

and female feces after pairing (1). This finding may be associated with their social contact and affiliative behavior (e.g., social grooming, genital investigation, social approach, and mounts and copulations) (2)(3)(4). In addition, considering the pairing experiment performed at the same time, the unknown factors other than pairing could also have affected gut microbiome changes.
Phascolarctobacterium showed the highest contribution in the gut microbiome transmission in the newly paired marmosets, and this microbiome group exhibited the highest contribution to the sex-biased transmission pattern of the female to the male (1). Some studies have suggested that the transmitted gut microbiome within primate social groups may be beneficial to the host health, but with no direct evidence (5,6).
The gut microbiome plays an important role in host immune development and health (7,8). Short-chain fatty acids (SCFAs) are the end products of fermentation of dietary fibers and protein by the gut microbiome (9)(10)(11)(12) ) have a key role in the maintenance of intestinal homeostasis and help maintain energy equilibrium, immune system function, and health (9)(10)(11)(12). For example, PA can increase host serotonin (5-hydroxytryptamine) biosynthesis in the intestine and affect gut motility and hemostasis (10). PA also reduces host inflammation, enhances tissue insulin sensitivity, maintains homeostasis of glucose, and induces hormone release (e.g., of leptin) (11,12).
Phascolarctobacterium can produce short-chain fatty acids, such as propionate (PA) (13). Based on the significant elevation in Phascolarctobacterium following social pairing in marmosets, we hypothesized that these changes in gut microbiome community following pairing would result in significant increases in production of SCFAs, especially PA.
We 0.41 Ϯ 0.341 mol/g). We compared the SCFA concentration from prepairing (PRE) to postpairing (POST) for each sex. In males, we observed that the concentrations of ACE, PA, ISOB, and ISOV significantly increased in the POST stage compared to the PRE stage. No significant changes were found in female fecal samples after pairing, but the changes in PA after pairing were at the marginally significant level (Table 1, Benjamini-Hochberg [B-H] correction). Although there were significant increases in the abundance of five genera (Phascolarctobacterium, Alloprevotella, Anaerobiospirillum, Sutterella, and Coprobacter) in fecal samples from both males and females, the PA concentration was significantly associated only with the abundance of Phascolarctobacterium in fecal samples from male marmosets (Fig. 1, B-H correction). Therefore, we speculated that the highest contribution of Phascolarctobacterium to the sex-biased transmission pattern of the female to the male after pairing might result in the sex differences in the abundance changes of some SCFA concentrations.
Phascolarctobacterium bacteria are Gram-negative, obligately anaerobic, and nonspore-forming environmental microorganisms and have been found in soil, water, and some mammal feces (e.g., koala, human, and nonhuman primate) (1,(13)(14)(15). One of the main end products of the fermentation in Phascolarctobacterium is PA (13), which may have beneficial effects on the host (10-12, [16][17][18]. For example, Phascolarctobacterium is positively correlated with the human positive mood (17), and Phascolarctobacterium has beneficial effects on host health and putatively decreases susceptibility to hepatic steatosis based on a rat model of nonalcoholic fatty liver (18). Moreover, PA is the essential mediator in the link between host nutrition, symbiotic gut microbiomes, and host physiology (19,20). The purpose of physiological homeostasis is to neutralize or repair disturbance and to maintain stability (e.g., blood glucose level and blood pressure) (21). In addition, several diseases in humans (e.g., diabetes, obesity, and autoimmune disorders) are associated with reduced SCFA production in the gut (22)(23)(24)(25).
Interestingly, the establishment of social bonds within pairmates (e.g., pair bonding) is important for maintaining physiological homeostasis (26). The data presented here extend these findings to the sex difference in changes in the fermentation metabolites  of the gut microbiome during pair formation. Although the fecal SCFAs do not entirely reflect the total intestinal production (19,20), examining SCFAs in common marmoset feces is a possible way (noninvasive method) to show SCFA metabolism in the intestine indirectly. We suggest that the gut microbiome metabolites in marmosets may have effects on the host physiology after pairing. Therefore, this study provides the new insight that one of the putative mechanisms between social bonding and physiological homeostasis in humans and nonhuman primates might involve microbial transmission in the gut and the associated metabolites under the condition of sex difference. Newly paired marmoset experiment. We examined fecal short-chain fatty acid concentrations in eight adult common marmoset pairs (eight females and eight males, aged 1.5 to 7.5 years). Baseline fecal samples (PRE; 59 fecal samples) were collected during a 2-week period prior to pairing (PRE), during which marmosets resided with an opposite-sex partner or in a family group (1). Marmosets were then rehoused in a new enclosure with a new opposite-sex partner who was a previously unfamiliar and unrelated marmoset. Fecal samples were collected in the postpairing phase (POST; 169 fecal samples) for an 8-week period following pairing. During both phases, marmosets were fed a consistent diet (commercial marmoset diet [Zupreem; Science Diet], Tenebrio larvae, scrambled eggs, fruits [red apple and cantaloupe], and gum arabic [Mazuri]). Fresh fecal samples were collected from marmosets in sterilized aluminum pans immediately after the light-on phase of the photoperiod. Samples were snap-frozen in liquid nitrogen and stored at Ϫ80°C. Most of these fecal samples were used in the previous gut microbiome study (1). This study was performed following the guidelines of the University of Nebraska Medical Center and the University of Nebraska at Omaha Institutional Animal Care and Use Committee. The protocol was approved by the University of Nebraska Medical Center/University of Nebraska at Omaha Institutional Animal Care and Use Committee .
SCFA measurement. Fecal samples were taken from the Ϫ80°C freezer and thawed at room temperature. Fecal samples (0.21 g) were homogenized with 1 ml sterile phosphate-buffered saline and then centrifuged (10,000 ϫ g, 5 min). Next, 0.4 ml of supernatant from each sample was mixed with 0.1 ml of internal standard (7 mM 2-ethylbutyrate) followed by acidification, extraction into diethyl ether, and quantification by gas chromatography (Clarus 580; PerkinElmer, Waltham, MA, USA) based on the standard protocol (27). For each pair, we estimated two single average points, one from the PRE stage, the other from the POST stage (days after pairing: from 29 to 55). Wilcoxon paired tests were used to calculate the P value from these two groups (PRE versus POST) within each sex. The relationships between the abundance of the five gut microbiome genera (significantly increased in both females and males after pairing) and the concentration of PA were tested via linear and quadratic models. The better model was selected according to the lower value of Akaike's information criterion (28). False discovery rates were addressed by the Benjamini-Hochberg procedure (B-H correction, ␣ ϭ 0.05) for multiple-comparison testing within each sex (29). The above statistical analysis and plotting were conducted in R software (30).

ACKNOWLEDGMENTS
We are grateful to members of the French lab for helping with sample collection and providing feedback during manuscript preparation. We are grateful to Qibin Gao and Wenqian Zhao for helping with the statistics analysis during manuscript revision.
The The authors declared no conflicts of interest relevant to this paper.