Sex-specific post-inflammatory dysbiosis mediates chronic visceral pain in colitis

Abstract

Background: Despite achieving endoscopic remission, over 20% of inflammatory bowel disease (IBD) patients experience chronic abdominal pain. Visceral pain and the microbiome exhibit sex-dependent interactions, while visceral pain in IBD shows a sex bias. Our aim was to evaluate whether post-inflammatory microbial perturbations contribute to visceral hypersensitivity in a sex-dependent manner.

Methods: Males, cycling females, ovariectomized, and sham-operated females were given dextran sodium sulfate to induce colitis and allowed to recover. Germ-free recipients received sex-appropriate and cross-sex fecal microbial transplants (FMT) from post-inflammatory donor mice. Visceral sensitivity was assessed by recording visceromotor responses to colorectal distention. The composition of the microbiota was evaluated via 16S rRNA gene V4 amplicon sequencing, while the metabolome was assessed using targeted (short chain fatty acids - SCFA) and semi-targeted mass spectrometry.

Results: Post-inflammatory cycling females developed visceral hyperalgesia when compared to males. This effect was reversed by ovariectomy. Both post-inflammatory males and females exhibited increased SCFA-producing species, but only males had elevated fecal SCFA content. FMT from post-inflammatory females transferred visceral hyperalgesia to both males and females, while FMT from post-inflammatory males could only transfer visceral hyperalgesia to males.

Conclusions: Female sex, hormonal status as well as the gut microbiota play a role in pain modulation. Our data highlight the importance of considering biological sex in the evaluation of visceral pain.

Keywords: DSS colitis; Microbiome; colorectal distention; fecal microbiota transplant; sex-differences; short-chain fatty acids; visceral pain.

Figure 1.

Assessment of DSS colitis severity in males and females. (a). Experimental timeline. Male and female mice were given 5 days of DSS or water and allowed to recover for 5 weeks. At day 31, a vasectomized male was introduced to synchronize females. Somatic pain assessment was performed at day 35. Electrodes were implanted at day 40 and visceral pain assessment was performed at day 42 followed by euthanasia of animals. Stool collection time points are indicated by arrows. (b). Weight changes in post-inflammatory mice. Two-way ANOVA, Tukey post hoc test, ***p≤0.001, comparing male DSS vs female DSS (from day 9–15), male control vs male DSS (from days 7–20), female control vs female DSS group (from day 8–13), n=12/group. (c). Disease activity index in DSS mice. Two-way ANOVA, Tukey post hoc test ***p≤0.001 from day 7–12, n=12/group. (d-f). Macroscopic and microscopic inflammation in DSS male and cycling females. (d). Colonic length changes in post-inflammatory mice. One-way ANOVA, Bonferroni post hoc test, *p=0.035, comparing male DSS vs male control (n=11-13/group). No significant differences were seen between male DSS vs female DSS (p=0.14) or between female control and female DSS (p>0.99). No changes in thickness (e; n=4/group), or tissue myeloperoxidase levels (f; n=4-5/group) were seen. (g). Representative photomicrographs of colonic cross sections. No microscopic evidence of inflammation was seen. Scale bar, 100 µm.

Figure 2.

Somatic and visceral sensitivity in post-inflammatory males and females. (a). Mechanical sensitivity assessed using the automated Von Frey test. One-way ANOVA, Bonferroni post-hoc test *p=0.02 control vs DSS males and control vs DSS females. N=8-9/group. (b). Thermal sensitivity was evaluated using the hot plate test. One-way ANOVA, Bonferroni post-hoc test *p=0.048 control vs. DSS females. No differences were seen when comparing control vs DSS males, p=0.18. N=8-9/group. (c). Visceral sensitivity was assessed using the visceromotor response to colorectal distention. 45mmHg distention: control males vs DSS males, *p≤0.05; control females vs DSS females, ** p≤0.01. 60mmHg distention: control males vs DSS males, *p≤0.05; control females vs DSS females: **p≤0.01, DSS males vs DSS females, ^##p=0.003. Two-way ANOVA, Tukey post-hoc test, n=11-12/group.

Figure 3.

Assessment of DSS colitis severity in sham-operated and OVX-females. (a). Experimental timeline. Ovariectomy or sham surgery was performed one week prior to DSS administration. Mice were then given 2.5% DSS and allowed to recover as outlined above. Stool collection time points are indicated by arrows. (b). Weight changes in sham and OVX females. Two-way ANOVA, Tukey post hoc test, *p≤0.05, comparing sham DSS vs OVX DSS on day 11, ***p≤0.001 comparing sham control vs sham DSS (from day 8–12), and ***p≤0.001 OVX control vs OVX DSS (from day 9–11), n=13/group. (c). Disease severity in sham and OVX females. Two-way ANOVA, Tukey post hoc test, **p=0.005 at day 11, n=13/group. (d). Colonic length. One-way ANOVA, Bonferroni post-hoc test, *p=0.02, sham control vs OVX control, ***p=0.0002, sham control vs DSS control, n=11-13/group. No significant differences were seen between DSS sham vs DSS OVX (p>0.99). (e). Colonic thickness. One-way ANOVA, Bonferroni post-hoc test, *p≤0.05, sham control vs OVX control, n=7-8/group. (f). MPO activity, n=4-5/group.

Figure 4.

Somatic and visceral sensitivity in post-inflammatory sham-operated and OVX females (a). Mechanical sensitivity in sham vs OVX females. N=11-12/group. One-way ANOVA, Bonferroni post-hoc test. (b). Thermal sensitivity in sham vs OVX females N=8-9/group. One-way ANOVA, Bonferroni post-hoc test. (c). Visceral sensitivity in sham vs OVX mice. OVX DSS females displayed a significant decrease in visceral hypersensitivity compared to sham DSS females. 45mmHg distension: control sham vs DSS sham group, *p≤0.05. 60mmHg distension: control sham vs DSS sham, *p≤0.05; control OVX vs DSS OVX *p≤0.05, DSS sham vs DSS OVX p≤0.05. Two-way ANOVA, Tukey post hoc test. n=12-13/group.

Figure 5.

Changes in gut microbiota composition in post-inflammatory males and females. (a). Shannon diversity index. There was a significant decrease in the alpha diversity [p<0.0001, two-way ANOVA, Tukey post-hoc test] in male mice 1 week after DSS treatment compared to baseline, which recovered to baseline levels by day 35. Female post-inflammatory mice displayed a significant increase in their microbial diversity compared to baseline [p=0.008, two-way ANOVA, Tukey post-hoc test] at day 35, which recovered to baseline levels by day 42 [p=0.001, two-way ANOVA, Tukey post-hoc test compared to day 35]. No change was seen in alpha diversity between control male or female mice at any point of the experiment. (b-c). Beta diversity (Bray-Curtis index, NMDS) in male (b) and female (c) post-inflammatory mice. Circles indicate 80% confidence interval. Males and females treated with DSS showed a significant shift in diversity after DSS administration, which persisted until day 42. All timepoints compared to control with PERMANOVA. (d). Log2 fold change of the ASVs differentially increased in males (right) and females (left) at day 0. ASVs with an adjusted p-value below p<0.01 were considered differentially abundant. DESeq2, Wald test. (e). Log2 fold change of the ASVs differentially increased in males (right) and females (left) at day 42. ASVs with an adjusted p-value below p<0.01 were considered differentially abundant. DESeq2, Wald test.

Figure 6.

Changes in fecal short chain fatty acids in post-inflammatory male vs female mice. (a). Fecal acetate levels; p<0.001 DSS males vs male controls and p=0.0002 DSS males vs DSS females. (b). Fecal butyrate levels; p=0.0123 DSS males vs male controls, p=0.0040 DSS males vs DSS females. (c). Fecal propionate levels. (d). Fecal isobutyrate levels; p= 0.0167 DSS males vs control males, p=0.0064 DSS males vs DSS females. (e). Fecal valerate levels. (f). Fecal isovalerate levels. p=0.0167 DSS males vs DSS females. One-way ANOVA, Bonferroni post-hoc test, n=3-5/group.

Figure 7.

Fecal metabolites known to participate in visceral hyperalgesia. (a). Adenine signal intensity across groups. (b). Tryptophan signal intensity across groups. The box plot represents the median value ± the interquartile range. One-way ANOVA, Tukey’s post-hoc test; FDR <0.05. **p<0.01, *p<0.05. n=4-9 samples per group.

Figure 8.

Changes in gut microbiota composition in post-inflammatory sham-operated and OVX females (a). Shannon diversity index. Post-inflammatory sham females showed a significant increase in alpha diversity at day 42 compared to day 0 (p=0.0461), while post-inflammatory OVX females displayed a significant increase in diversity at days 35 (p=0.0067) and 42 (p=0.0234) when compared to baseline. No differences were seen between OVX or sham control mice. Two-way ANOVA, Tukey post-hoc test. (b-c). Beta diversity (Bray-Curtis index, NMDS) in sham-operated (b) and OVX (c) post-inflammatory mice. Circles indicate 80% confidence interval. Sham [p<0.03, PERMANOVA] and OVX [p<0.02, PERMANOVA] surgery resulted in a significant shift in the gut microbiota. Sham and OVX females treated with DSS showed a significant shift in diversity after DSS administration [p<0.01 PERMANOVA] which persisted until day 42. (d). Log2 fold change of the ASVs differentially increased in OVX (right) and sham-operated females (left) at day 0. ASVs with an adjusted p-value below p<0.01 were considered differentially abundant. DESeq2, Wald test. (e). Log2 fold change of the ASVs differentially increased in OVX (right) and sham-operated females (left) at day 42. ASVs with an adjusted p-value below p<0.01 were considered differentially abundant. DESeq2, Wald test.

Figure 9.

Evaluation of visceral sensitivity following sex-appropriate or cross-sex FMT. (a). Experimental timeline. Germ-free male and female recipients received sex-appropriate or cross-sex FMT at days 1 and 5. At day 19, a vasectomized male mouse was introduced to female recipients to synchronize the estrous cycle. Electrodes were implanted at day 23 and visceral sensitivity assessment was performed at day 25, 3 weeks post-FMT. Stool collection time point is indicated. (b). Visceromotor response to colorectal distention in female recipients who received stool from female donors. 60mmHg distention: p=0.01. Two-way ANOVA, Tukey post-hoc test, n=10. (c). Visceromotor response to colorectal distention in male recipients who received stool from female donors. 45mmHg distention: p=0.03 and 60mmHg distention: p=0.008. Two-way ANOVA, Tukey post-hoc test, n=8-11. (d). Visceromotor response to colorectal distention in female recipients who received stool from male donors, n=8-9. (e). Visceromotor response to colorectal distention in male recipients who received stool from male donors. 45mmHg distention: p=0.004 and 60mmHg distention: p=0.006. Two-way ANOVA, Tukey post-hoc test, n=9.