Sex-specific effects of voluntary wheel running on behavior and the gut microbiota-immune-brain axis in mice

Abstract

Physical exercise has been positioned as a promising strategy to prevent and/or alleviate anxiety and depression, but the biological processes associated with its effects on mental health have yet to be entirely determined. Although the prevalence of depression and anxiety in women is about twice that of men, very few studies have examined whether physical exercise could affect mental health differently according to sex. This study examined, in singly-housed mice, the sex-specific effects of voluntary exercise on depressive- and anxiety-like behaviors as well as on different markers along the gut microbiota-immune-brain axis. Male and female C57BL/6N mice had voluntary access to running wheels in their home-cages for 24 days or were left undisturbed in identical home-cages without running wheels. Behaviors were then examined in the open field, splash, elevated plus maze, and tail suspension tests. Gene expression of pro-inflammatory cytokines, microglia activation-related genes, and tight junction proteins was determined in the jejunum and the hippocampus, while microbiota composition and predicted function were verified in cecum contents. Voluntary exercise reduced anxiety-like behaviors and altered grooming patterns in males exclusively. Although the exercise intervention resulted in changes to brain inflammatory activity and to cecal microbiota composition and inferred function in both sexes, reductions in the jejunal expression of pro-inflammatory markers were observed in females only. These findings support the view that voluntary exercise, even when performed during a short period, is beneficial for mental and intestinal health and that its sex-specific effects on behavior could be, at least in part, related to some components of the gut microbiota-immune-brain axis.

Fig. 1

Summary of experimental procedures. Female (n = 19) and male (n = 20) C57BL/6N mice were given voluntary access to running wheels in their home-cages (Exercise condition) or were housed in identical cages without running wheels (Control condition) from Days 1–24. Behavior was assessed on Days 22 (open field and splash tests) and 23 (elevated plus maze and tail suspension tests), and the hippocampus, jejunum, and cecum contents were collected approximately 16 h after on Day 24 for the determination of inflammatory and tight junction markers (RT-qPCR) and of microbiota diversity, composition, and inferred function (16S rRNA, PICRUSt2).

Fig. 2

Female and male mice did the same amount of exercise during access to running wheels. The total distance run in the wheels (in km) from Days 1–22, computed form the total number of revolutions made during this period, was equivalent in female and male C57BL/6N mice with voluntary access to running wheels. Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. (Females: n = 9; males: n = 10).

Fig. 3

Access to running wheels changed grooming patterns in the splash test in males only. (A) Total time spent grooming during the 10-min test session (in seconds [s]) was comparable among groups. (B) Males with exercise initiated more grooming sessions than those without exercise. (C) Males with exercise spent less time on average during these grooming sessions (in seconds [s]) than those without exercise. Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. Females without exercise (Females CON: n = 7); females with exercise (Females EXE: n = 7); males without exercise (Males CON: n = 9); males with exercise (Males EXE: n = 8). **p < .01 and ***p < .005 relative to Males CON.

Fig. 4

Access to running wheels reduced the fear of open arms in the elevated plus maze in males only. Males with exercise spent (A) more time in the open arms of the elevated plus maze (in seconds [s]) and (B) less time in the closed arms of the maze (in seconds [s]) than those without exercise. Males with exercise made (C) more entries into the open arms of the maze but (D) comparable entries into the closed arms than those without exercise. (E) The distance traveled in the apparatus (in cm) was comparable among groups. Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. Females without exercise (Females CON: n = 10); females with exercise (Females EXE: n = 9); males without exercise (Males CON: n = 10); males with exercise (Males EXE: n = 10). **p < .01 and ***p < .005 relative to Males CON.

Fig. 5

Gene expression of inflammatory and tight junction markers in the hippocampus and jejunum of female and male mice with or without access to running wheels. (A) Hippocampal expression of Cx3cr1 was reduced in both females and males with exercise whereas (B) the apparent hippocampal increases in BDNF expression in females and males with exercise did not reach significance. Females with exercise had lower jejunal expression of (C) IL-6 and of (D) TNF-α than their counterparts without exercise whereas (E) IL-1β was unchanged by any of the manipulations. The modest changes in (F) Cldn3 and (G) Ocln expression in the jejunum after exercise did not reach significance. (H) Jejunal Cldn2 expression was comparable among groups. Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. Females without exercise (Females CON: n = 7–10); females with exercise (Females EXE: n = 8–9); males without exercise (Males CON: n = 6–10); males with exercise (Males EXE: n = 7–10). *p < .05 and ***p < .005 relative to sex-matched CON.

Fig. 6

Alpha- and beta-diversity metrics in contents from the cecum of female and male mice with or without access to running wheels. (A) The Chao1 index of bacterial species richness was reduced in males with exercise compared to males without exercise whereas (B) the Shannon diversity index was comparable among groups. (C) Beta-diversity as illustrated by Bray-Curtis dissimilarity was comparable among all groups. Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. Females without exercise (Females CON: n = 10); females with exercise (Females EXE: n = 9); males without exercise (Males CON: n = 10); males with exercise (Males EXE: n = 10). *p < .05 relative to Males CON.

Fig. 7

Relative abundance of families and genera in contents from the cecum of female and male mice with or without access to running wheels. Increased abundance of (A) Muribaculaceae and (B) Delftia and reduced abundance of (C) Roseburiain both females and males with exercise. Access to exercise promoted sex-specific changes in particular taxa, with (D) Lachnoclostridium being increased in males, and (E) Clostridia_UCG-014, (F) Erysipelatoclostridiaceae UCG-004, (G) Ruminococcaceae uncultured, and (H) Pygmaiobacter, being decreased in females. Sex-dependent effects of exercise on KEGG pathways related to inflammation, with the peptidoglycan biosynthesis pathway (ko00550) being decreased in females (I). Dots and triangles represent individual mice and bar plots and error bars represent group means ± S.E.M. Females without exercise (Females CON: n = 10); females with exercise (Females EXE: n = 9); males without exercise (Males CON: n = 10); males with exercise (Males EXE: n = 10). *p < .05, **p < .01, and ***p < .005 relative to sex-matched CON.

Fig. 8

Relationships between microbiota-related outcomes and behavioral parameters affected by Exercise and/or Sex among female and male mice. In females (left panel), enrichment of the peptidoglycan synthesis pathway positively correlated with the mean time spent in grooming bouts (Splash Mean T) in the splash test but negatively correlated with the number of grooming bouts (Splash # bouts) in this test as well as with the number of entries into the corners (OF Corners) in the open field. The Prevotellaceae_NK3B31_group in females also negatively correlated with the number of entries in the corners in the open field (OF Corners). **p < .01, ***p < .005, and ****p < .001.