Long-Term Administration of Conjugated Estrogen and Bazedoxifene Decreased Murine Fecal β-Glucuronidase Activity Without Impacting Overall Microbiome Community

Abstract

Conjugated estrogens (CE) and Bazedoxifene (BZA) combination is used to alleviate menopause-associated symptoms in women. CE+BZA undergo first-pass-metabolism in the liver and deconjugation by gut microbiome via β-glucuronidase (GUS) enzyme inside the distal gut. To date, the impact of long-term exposure to CE+BZA on the gut microbiome or GUS activity has not been examined. Our study using an ovariectomized mouse model showed that CE+BZA administration did not affect the overall cecal or fecal microbiome community except that it decreased the abundance of Akkermansia, which was identified as a fecal biomarker correlated with weight gain. The fecal GUS activity was reduced significantly and was positively correlated with the abundance of Lactobacillaceae in the fecal microbiome. We further confirmed in Escherichia coli K12 and Lactobacillus gasseri ADH that Tamoxifen-, 4-hydroxy-Tamoxifen- and Estradiol-Glucuronides competed for GUS activity. Our study for the first time demonstrated that long-term estrogen supplementation directly modulated gut microbial GUS activity. Our findings implicate that long-term estrogen supplementation impacts composition of gut microbiota and microbial activity, which affects estrogen metabolism in the gut. Thus, it is possible to manipulate such activity to improve the efficacy and safety of long-term administered estrogens for postmenopausal women or breast cancer patients.

Figure 1

(A) Venn diagram comparing the observed bacterial genera in the cecal and fecal microbiome of mice after Veh or CE+BZA treatment. (B) Clustering analysis based on weighted Unifrac distances of cecal and fecal microbiome of mice treated with: Veh - vehicle (n = 8); E2 - estradiol (n = 5); CE - conjugated estrogen (n = 7); BZA - Bazedoxifene (n = 8); CE+BZA - conjugated estrogen combined with Bazedoxifene (n = 6).

Figure 2

Top 10 most abundant bacterial families (percentage total) in the cecal and fecal microbiome of each mouse treated with: Veh - vehicle (n = 8); E2 - estradiol (n = 5); CE - conjugated estrogen (n = 7); BZA - Bazedoxifene (n = 8); CE+BZA - conjugated estrogen combined with Bazedoxifene (n = 6).

Figure 3

(A) Identification of important features including fecal bacterial taxa and plasma metabolites that correlate with lowest% weight gain. Detailed association data for samples (n = 4) were shown for several features’ association against (B) weight gain and Akkermansia. (C) Correlation between the abundance of fecal Akkermansia, Anaerotruncus and blood glucuronic acid levels, liver weight and percent weight change. Two-tailed pairwise t-tests were performed to identify correlation between indicated factors, metabolites or bacteria level. *P < 0.05, **P < 0.01, ***P < 0.001. (D) The levels of Akkermansia and glucuronic acid following each treatment. Statistical significance was established at α = 0.05. One-way Anova with a Dunn’s correction were used to identify treatments that were significantly different from Veh. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4

(A) Total GUS activity of fecal microbiota of mice (n = 8 per group) treated with: Veh - vehicle; E2 - estradiol; CE - conjugated estrogen; BZA - Bazedoxifene; CB - conjugated estrogen plus Bazedoxifene. *P < 0.05 compared to the activity of the control group at 12 h. Results are mean ± SEM. (B) Abundance of bacterial families in the fecal microbiome highly correlated (P < 0.05) with fecal GUS activity.

Figure 5

(A) S9 fraction glucuronidation metabolites of BZA, E2 and TAM. Competition between Tamoxifen-Glucuronide (TAM-Glc), 4-hydroxy-Tamoxifen-Glucuronide (4OH-TAM-Glc) or Estradiol-Glucuronide (E2-Glc) with GUS activity in (B) Escherichia coli K12 and (C) Lactobacillus gasseri ADH. Data showing min to max including three biological replicates and two technical duplicates in each treatment group (background corrected - ave. 1.9 nmol p-nitrophenol per 10⁹ CFU cells). Results are min to max. ****P < 0.0001 compared to the activity of the control group at 12 h.