The Response of the Rodent Gut Microbiome to Broad-Spectrum Antibiotics Is Different in Males and Females

Abstract

Gut microbiome composition is different in males and females, but sex is rarely considered when prescribing antibiotics, and sex-based differences in gut microbiome recovery following antibiotic treatment are poorly understood. Here, we compared the effects of broad-spectrum antibiotics on both the stool and small bowel microbiomes in male and female rats. Adult male and female Sprague Dawley rats were exposed to a multi-drug antibiotic cocktail for 8 days, or remained unexposed as controls. Following cessation of antibiotics, rats were monitored for an additional 13-day recovery period prior to euthanasia. Baseline stool microbiome composition was similar in males and females. By antibiotic exposure day 8 (AbxD8), exposed male rats exhibited greater loss of stool microbial diversity compared to exposed females, and the relative abundance (RA) of numerous taxa were significantly different in exposed males vs. exposed females. Specifically, RA of phylum Proteobacteria and genera Lactobacillus, Sutterella, Akkermansia, and Serratia were higher in exposed males vs. exposed females, whereas RA of phyla Firmicutes and Actinobacteria and genera Turicibacter and Enterococcus were lower. By 13 days post antibiotics cessation (PAbxD13), the stool RA of these and other taxa remained significantly different from baseline, and also remained significantly different between exposed males and exposed females. RA of phyla Firmicutes and Actinobacteria and genus Enterococcus remained lower in exposed males vs. exposed females, and genus Sutterella remained higher. However, RA of phylum Proteobacteria and genus Akkermansia were now also lower in exposed males vs. females, whereas RA of phylum Bacteroidetes and genus Turicibacter were now higher in exposed males. Further, the small bowel microbiome of exposed rats on PAbxD13 was also significantly different from unexposed controls, with higher RA of Firmicutes, Turicibacter and Parabacteroides in exposed males vs. females, and lower RA of Bacteroidetes, Proteobacteria, Actinobacteria, Oscillospira, Sutterella, and Akkermansia in exposed males vs. females. These findings indicate that broad-spectrum antibiotics have significant and sex-specific effects on gut microbial populations in both stool and the small bowel, and that the recovery of gut microbial populations following exposure to broad-spectrum antibiotics also differs between sexes. These findings may have clinical implications for the way antibiotics are prescribed.

Keywords: antibiotics recovery; antibiotics response; female; male; sex; small bowel microbiome; stool microbiome.

Figure 1

Average weights in male and female control and antibiotic-exposed rats at baseline, during the 8-day antibiotic exposure period, and during the 13-day recovery period following cessation of antibiotics. P values denote differences between antibiotic-exposed rats and controls of the same sex during the antibiotic-exposure and recovery periods.

Figure 2

Microbial alpha and beta diversity in stool samples from male and female control and exposed rats. (A) Stool alpha diversity in male and female control and exposed rats at baseline, on AbxD8, and on PAbxD13, as measured by Shannon’s and Simpson’s indices. (B) Stool beta diversity in male and female exposed rats at baseline and all controls (males—red and females—blue), on AbxD8 (exposed males—orange and exposed females—light blue), and on PAbxD13 (exposed males yellow and exposed females purple).

Figure 3

Heat trees of the stool microbiomes in male vs. female antibiotic-exposed rats. (A) Exposed males vs. exposed females on AbxD8. (B) Exposed males vs. exposed females on PAbxD13. Taxa with lower RA in males vs. females are shown in red. Taxa with higher RA in males vs. females are shown in blue. Nodes represent taxonomic levels. Greater line thicknesses and font sizes denote higher RA.

Figure 4

Differences in predicted stool microbial functional metabolic pathways in exposed male and female rats on PAbxD13. Pathways shown in red are predicted to be enriched in exposed males, and pathways shown in green are predicted to be enriched in exposed females. Only significant differences are shown.

Figure 5

Microbial alpha and beta diversity in small bowel samples from male and female control and exposed rats. (A) Small bowel alpha diversity in male and female control and exposed rats on PAbxD13, as measured by Simpson’s and Shannon’s indices. Exposed males vs. control males: Simpson’s index p = 0.0006, Shannon’s index p = 0.0002. All other comparisons p = NS. (B) Small bowel beta diversity in male and female rats on PAbxD13 (control males—red, control females—blue, exposed males—purple, and exposed females—green). Male vs. female controls: PERMANOVA p = 1. Exposed females vs. exposed males PERMANOVA p = 0.01026. Exposed females vs. female controls: PERMANOVA p = 0.0022. Exposed males vs. male controls: PERMANOVA p = 0.00006.

Figure 6

Heat trees of the small bowel microbiomes of male vs. female rats on PAbxD13. (A) Control males vs. control females. (B) Exposed males vs. exposed females. Taxa with lower RA in males vs. females are shown in red. Taxa with higher RA in males vs. females are shown in blue. Nodes represent taxonomic levels, and greater line thickness denotes higher RA.

Figure 7

Differences in predicted small bowel microbial functional metabolic pathways in exposed male and female rats on PAbxD13. Pathways shown in red are predicted to be enriched in exposed males, and pathways shown in green are predicted to be enriched in exposed females. Only significant differences are shown.

Figure 8

Heat trees of the stool vs. small bowel microbiomes in control and exposed female rats on PAbxD13. (A) Stool vs. small bowel microbiomes in control females. (B) Stool vs. small bowel microbiomes in exposed females. Taxa with lower RA in the small bowel vs. stool are shown in red. Taxa with higher RA in the small bowel vs. stool are shown in blue. Nodes represent taxonomic levels. Greater line thicknesses and greater font sizes denote higher RA.