Sex hormone deprivation abolishes sex-specific differences in murine colon inflammation and related lipid mediator production

Abstract

Unresolved inflammation, due to unfavorable imbalances between pro-inflammatory and pro-resolving mediators, leads to chronic inflammatory pathologies that are often sex-biased and regulated by sex hormones, including inflammatory bowel disease. Lipid mediators (LM) produced from polyunsaturated fatty acids by various lipoxygenases (LOX) and cyclooxygenases govern all stages of inflammation, i.e., the initiation and progression by pro-inflammatory eicosanoids and its resolution by specialized pro-resolving mediators (SPM). Here, we reveal sex-specific differences in murine experimental colitis with male preponderance, which was abolished by sex hormone deprivation using gonadectomy, and this correlated to the levels of inflammation-relevant mediators in the colon. Oral dextran sodium sulfate administration caused more severe colon inflammation in male CD-1 mice than in female counterparts during the acute phase. Colitis in males yielded higher colonic cytokine/chemokine levels but lower 12-/15-LOX-derived LM including SPM compared to female animals in the resolving phase. Sex hormone deprivation in male mice by orchidectomy ameliorated colitis and impaired pro-inflammatory cytokine/chemokine levels but elevated 12-/15-LOX products including SPM, thus abolishing the observed sex differences. Conversely, ovariectomy impaired the levels of those LM that dominated in females and that were increased in males after gonadectomy. Our findings suggest that male sex hormones promote the development of colitis connected to the biosynthesis of inflammatory cytokines, chemokines, and certain LM, especially pro-resolving 12-/15-LOX products that appear to be suppressed in the male colon due to androgens.

Keywords: colitis; inflammation; lipid mediators; lipoxygenases; sex differences; specialized pro‐resolving mediators.

Figure 1.. Sex dimorphism in murine DSS-induced colitis.

A) Experimental timeline. Male and female mice had access to drinking water containing 3% dextran sulfate sodium (DSS) or pure drinking water (control groups) for 5 days. Then, the mice had access to pure drinking water until sacrifice on day 8 or day 14, corresponding to the acute or resolving phase of inflammation, respectively. Control (ctrl) mice were sacrificed at day 8 and day 14. B) Changes in body weight of male and female mice. Data are expressed as difference between the day of sacrifice (day 8 or day 14) and the day of onset of the experiment (day 0). Male ctrl. n = 16, female ctrl. n = 16, male DSS day 8 n = 15, female DSS day 8 n = 16, male DSS day 14 n = 6, female DSS day 14 n = 8. C) Daily score of stool consistency (diarrhea score) over 14 days (0 = normal faeces, 1 = loose stool, 2 = watery diarrhea, 3 = severe watery diarrhea) of DSS-treated mice; male n = 16, female n = 16. D) Daily score of rectal bleeding over 14 days (0 = normal faeces, 1 = visible blood, 2 = severe bleeding) of male and female animals during DSS-induced colitis, male n = 16, female n = 16. E) Colon weight/length ratio from male and female animals in control groups and on day 8 and day 14. Data are given as percentage of control; male ctrl. n = 16, female ctrl. n = 16, male DSS day 8 n = 7, female DSS day 8 n = 8, male DSS day 14 n = 6, female DSS day 14 n = 8. F) Intestinal MPO activity of control and DSS-treated (day 8 and day 14) male and female mice. Data are expressed as units (U) per mg of tissue; male ctrl. n = 12, female ctrl. n = 12, male DSS day 8 n = 6, female DSS day 8 n = 6, male DSS day 14 n = 6, female DSS day 14 n = 6. G) Levels of IL-1β in the colons was measured by ELISA. Data are expressed as ng/mL colon lysate, corresponding to 20 mg organ. Male ctrl. n = 5, female ctrl. n = 5 male DSS day 8 n = 5, female DSS day 8 n = 5, male DSS day 14 n = 5, female DSS day 14 n = 5. Statistics: Data are shown as mean ± SEM. p-Values were calculated by unpaired two-tailed Student’s t-test with or without Welch’s correction. Statistical analysis was performed after identification of outliers by ROUT (Q = 0.1%) and transformation to logarithmic values depending on results for Shapiro-Wilk normality tests for normal (Gaussian) and lognormal distribution. *p <0.05, ****p <0.0001; ns, not significant.

Figure 2.. Lipid mediator and cytokine levels in the colons of naïve male and female mice.

A) Principal component analysis (PCA) of the LM profiles in colons of male and female mice. B) LM and PUFA levels in colons of male and female mice. Data are expressed as pg/20 mg colon; male n = 7, female n = 7. C) PUFA metabolomes of colons from male and female animals. Metabolomes are given in ng/20 mg as sum of all LM derived from AA (t-LTB₄, et-LTB₄, LTB₄, 5S,12S-diHETE, 5S,15S-diHETE, 5S,6R-diHETE, 15-HETE, 11-HETE, 12-HETE, 8-HETE, 5-HETE, PGE₂, PGD₂, 6-keto-PGF_1α, 15-keto-PGE₂, 20-OH-PGE₂, PGF_2α, PGF_2β, PGJ₂, TXB₂, 15-deoxy-12,14-PGJ₂), LA (9-HODE, 13-HODE), DHA (MaR2, PDX, RvD5, 17-HDHA, 14-HDHA, 13-HDHA, 10-HDHA, 7-HDHA, 4-HDHA) or EPA (18-HEPE, 15-HEPE, 12-HEPE, 11-HEPE, 5-HEPE, PGF_3α, TXB₃, PGB₃, PGE₃/PGD₃); male n = 7, female n = 7. D) Levels of IL-10, IL-1β, CXCL-1/KC, and MCP-1 in colon lysates from male and female mice were measured by ELISA. Data are given in ng/g colon; male n = 7, female n = 7. Statistics: Data are shown as mean ± SEM. p-Values were calculated by unpaired two-tailed Student’s t-test with or without Welch’s correction. Values below the LOD were set to LOD and values below LLOQ were replaced by ½ LLOQ. Statistical analysis was performed after identification of outliers by ROUT (Q = 0.1%) and transformation to logarithmic values depending on results for Shapiro-Wilk normality tests for normal (Gaussian) and lognormal distribution. *p <0.05, **p <0.01; ns, not significant.

Figure 3.. Gonadectomy abolishes sex differences in the development of murine colitis.

A) Experimental setup and timeline. Male and female mice were orchidectomized (orch.) and ovariectomized (ovx.), respectively, or sham-operated, followed by a recovery period of four to five weeks. Then, animals had access to drinking water with 3% DSS for 5 days. Afterwards, mice were provided access to normal drinking water (without DSS) until sacrifice on day 8 or day 14, corresponding to acute or resolving phases of inflammation, respectively. B) Changes in body weight of male and female mice expressed as difference between the day of sacrifice (day 8 or day 14) and the day of onset of the experiment (day 0). Male sham day 8 n = 16, female sham day 8 n = 15, male orch. day 8 n = 16, female ovx. day 8 n = 16, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. C) Daily score of stool consistency over 14 days (0 = normal faeces, 1 = loose stool, 2 = watery diarrhea, 3 = severe watery diarrhea) of gonadectomized and sham-operated mice; male sham day 1–8 n = 16, male sham day 9–14 n = 7, female sham day 1–8 n = 15, female sham day 9–14 n = 7, male orch. day 1–8 n = 16, male orch. day 9–14 n = 8, female ovx. day 1–8 n = 8, female ovx. day 9–14 n = 8. D) Daily score of rectal bleeding over 14 days (0 = normal faeces, 1 = visible blood, 2 = severe bleeding) of gonadectomized and sham-operated mice; male sham day 1–8 n = 16, male sham day 9–14 n = 7, female sham day 1–8 n = 15, female sham day 9–14 n = 7, male orch. day 1–8 n = 16, male orch. day 9–14 n = 8, female ovx. day 1–8 n = 8, female ovx. day 9–14 n = 8. E) Weight to length ratio of colons from male and female animals after sham operation or gonadectomy at day 8 and day 14; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. F) Intestinal MPO activity of DSS-treated male and female mice (day 8) that were sham-operated or gonadectomized. Data are expressed as units (U) per mg of tissue; male sham n = 8, female sham n = 7, male orch. n = 8, female ovx. n = 8. G) Spleen weight of male and female animals after sham surgery or gonadectomy on days 8 and day 14; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. H) Levels of IL-10, IL-1β, CXCL-1/KC, and MCP-1 in colon lysates from gonadectomized and sham-operated mice on day 8 and day 14. Data are given in ng/g colon; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. I) Representative images (five samples for each group) of H&E-stained colon sections from male and female mice at day 8 and at day 14 under normal gonadal condition as well as after gonadectomies; original magnification 20 X. Statistics: Data are shown as mean ± SEM. p-Values were calculated by Brown-Forsythe and Welch ANOVA with Dunnett’s T3 or Šídák’s multiple comparisons test. Statistical analysis was performed after identification of outliers by ROUT (Q = 0.1%) and transformation to logarithmic values depending on results for Shapiro-Wilk normality tests for normal (Gaussian) and lognormal distribution. *p <0.05, **p <0.01, ***p <0.001, ****p <0.0001; ns, not significant.

Figure 4.. Impact of gonadectomy on lipid mediator profiles during DSS-induced colitis in mice.

A) PCA of colonic LM profiles of male and female sham-operated and gonadectomized mice 8 days and 14 days after the onset of DSS-treatment. B) Levels of selected LM in colons of male and female sham-operated animals on day 8 and 14. Data are expressed as pg/20 mg colon; male sham day 8 n = 8, female sham day 8 n = 7, male sham day 14 n = 7, female sham day 14 n = 8. C) Heatmap showing changes in LM profiles due to gonadectomy versus sham operation in colons of male and female animals after 8 days and 14 days. The fold change is implicated by color scale. For individual LM only detectable or quantifiable in one group, the fold change was calculated to corresponding LOD or ½ LLOQ; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. D) LM levels in colons of male and female sham-operated and gonadectomized mice at 8 days and 14 days after onset of DSS-treatment. COX products: PGB₃, PGD₁, PGD₂, PGD₃/PGE₃, PGE₁, 6-keto-PGE₁, PGE₂, PGF_1α, 6-keto-PGF_1α, PGF_2α, PGF_2β, PGF_3α, PGJ₂, 15-d-PGJ₂; 5-LOX products – LTB₄, t-LTB₄, epi-t-LTB₄, 5-HEPE, 5-HETE, 5S,12S-diHETE, 7-HDHA; 12-/15-LOX products – 14-HDHA, 12-HEPE, 12-HETE, 17-HDHA, 15-HEPE, 15-HETE, 5S,15S-diHETE, PDX, MaR2, RvD5, and PDX; DHA metabolome – PDX, 17-HDHA, 13-HDHA, 14-HDHA, 10-HDHA, 7-HDHA, 4-HDHA; EPA metabolome – 18-HEPE, 11-HEPE, 15-HEPE, 12-HEPE, 5-HEPE, PGF_3α, PGE₃/PGD₃. LM product groups and metabolomes are given in ng/20 mg; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8. E) Levels of selected LM in colons of sham-operated and gonadectomized male and female mice at day 14 after the onset of DSS-treatment. LM are given in pg/20 mg; male sham day 8 n = 8, female sham day 8 n = 7, male orch. day 8 n = 8, female ovx. day 8 n = 8, male sham day 14 n = 7, female sham day 14 n = 8, male orch. day 14 n = 8, female ovx. day 14 n = 8. Statistics: Data are shown as (B, D, E) mean + SEM, and p-values (B) were calculated by unpaired two-tailed Student’s t-test with or without Welch’s correction, or (D,E) by Brown-Forsythe and Welch ANOVA with Dunnett’s T3 or Šídák’s multiple comparisons test. Values that were below the LOD were set to LOD and values below LLOQ were replaced by ½ LLOQ. Statistical analysis was performed after identification of outliers by ROUT (Q = 0.1 %) and transformation to logarithmic values depending on results for Shapiro-Wilk normality tests for normal (Gaussian) and lognormal distribution. *p <0.05, **p <0.01, ***p <0.001, ****p <0.0001, ns, not significant.