Glycodeoxycholic acid alleviates central precocious puberty by modulating gut microbiota and metabolites in high-fat diet-fed female rats

Abstract

Objective: Central precocious puberty (CPP) is a common pediatric endocrine disorder and a significant global public health concern. Emerging evidence suggests an association between bile acids (BAs) and CPP, although their regulatory roles and underlying mechanisms remain poorly understood.

Methods: We conducted untargeted metabolomics and targeted BA analysis on serum samples from female rats with high-fat diet-induced CPP to identify metabolites potentially involved in regulating puberty through modulation of Sirt1 and Kiss1 expression in the hypothalamus. Identified BAs were then administered via gavage to female rats with CPP to assess their effects. To explore the mechanisms by which these BAs affect the development of CPP, gut microbiota and their metabolites were analyzed using 16S rRNA sequencing and untargeted metabolomics.

Results: Our findings revealed significant reductions in glycodeoxycholic acid (GDCA) and glycoursodeoxycholic acid (GUDCA) levels in female rats with CPP. GDCA treatment delayed the onset of puberty, accompanied by alterations in the gut microbiota functions and metabolic pathways related to oxidative stress (OS) and fatty acid metabolism. Mediation analysis suggested that OS-related metabolites, including gamma-glutamylcysteine and malonic acid, which increased with the abundance of Lachnospiraceae UCG-001, facilitated the reduction of Sirt1 expression. Additionally, pregnenolone appeared to suppress the beneficial effect of Parasutterella in enhancing Sirt1 expression.

Conclusion: This study demonstrates that GDCA exhibits a potential therapeutic effect on CPP through a unique mechanism that involves gut microbiota modulation, alterations in serum metabolites, and changes in the expression of key regulatory factors Sirt1.

Keywords: Sirt1 regulation; Central precocious puberty; Glycodeoxycholic acid; Gut microbiota; Gut-brain axis.

Fig. 1

Significantly changed metabolic profiles in HFD group. A. OPLS-DA score plots of all peak features in positive (ES+) and negative (ES−) ion modes from the untargeted metabolomics data of serum samples from CPP rats (HFD, n = 10) and normal rats (NN, n = 12). B. Volcano plot showing the up-regulated (represented by red dots) and down-regulated metabolites (represented by blue dots) in the HFD group compared to the NN group. Significantly changed metabolites related to BAs and derivatives were highlighted. C. Heatmap displaying distinct serum BAs in HFD (n = 7) and NN (n = 10). D. Variable importance plot of 10 differential serum bile acids (y-axis) ranked by contribution to mean decrease accuracy of coefficient (x-axis) in the random forest model for discerning group difference. E. Spearman correlations between Sirt1 and Kiss1 expression levels and GDCA and GUDCA in the HFD (n = 5) and NN groups (n = 7). F. Comparison of serum levels of GDCA and GUDCA in the NN and HFD groups. Two-tailed Wilcoxon rank-sum test (B-C, F) and Spearman rank correlation (E) were used for statistical analysis with Benjamini-Hochberg adjustment. r: Spearman correlation coefficient; adj.p: p value corrected by the Benjamini-Hochberg method; *adj.p < 0.05

Fig. 2

GDCA supplementation has therapeutic effects in CPP. A. The experimental protocol for assessing the effects of GDCA and GUDCA on HFD-induced CPP in female rats. B. Evolution of BW. Cumulative percentages of the NS and GDCA groups at VO (C) and FE (D). E. Histological score of follicular development and ovulation. F. Serum LH levels among the groups. G. The mRNA expression of GnRH, Kiss1, and Sirt1 in the hypothalamus at PND 34. LME (B), Fisher’s exact test (C-D) and two-tailed Wilcoxon rank-sum test (E-G) were used for statistical analysis. *p < 0.05; **p < 0.01; ****p < 0.0001. Total group sizes were: NS = 12 and GDCA = 10; while phenotypic and hormonal parameters were assayed in the whole groups, hypothalamic RNA (Fig. 2G) analyses were conducted in a representative subset of randomly assigned samples from each group, with the following distribution: NS = 8; GDCA = 7

Fig. 3

GDCA modulates the composition of gut microbiota in CPP female rats. A. Principal coordinate analysis (PCoA) of beta diversity based on the Bray-Curtis distance in the NS (n = 11) and GDCA (n = 9) groups. B. LEfSe analysis showing the differential abundance of species and genera. The differential species (C) and genera (D) influencing the increase in mean squared error (%IncMSE) measure of the random forest regression of Sirt1 gene expression. E. Top 20 significant differential KEGG pathways by PICRUSt2 among the groups

Fig. 4

Relationships among gut microbiota, metabolites and hypothalamic Sirt1 in GDCA-induced changes. A. Volcano plots showing the up-regulated (represented by blue dots) and down-regulated metabolites (represented by yellow dots) in GDCA (n = 5) compared to NS (n = 6) groups. Significantly changed metabolites from different origins were labeled with different colors. B. The enrichment analysis of KEGG metabolic pathways according to differential metabolites related to Sirt1 expression in ES−. C. Spearman correlations between gut microbiota and metabolites related to Sirt1 expression levels in GDCA and NS groups. Only the correlation coefficients of|r| ≥ 0.7, along with adj.p < 0.05 were displayed. Red and blue represent positive and negative correlations, respectively. The black border highlights the top 4 bacteria or metabolites with the highest number of edges. D. Mediation analysis for the role of metabolites in the association between gut microbes and hypothalamic Sirt1 expression. Different types of causal effects are colored by different colors (red: positive, blue: negative) and the strength of effect is represented by line thickness. *p < 0.05; **p < 0.01