Characteristics of the gut microbiota and serum metabolites in postmenopausal women with reduced bone mineral density

Abstract

Introduction: Emerging evidence suggests that the gut microbiota is closely associated with bone homeostasis. However, little is known about the relationships among the bone mineral density (BMD) index, bone turnover markers, and the gut microbiota and its metabolites in postmenopausal women.

Methods: In this study, to understand gut microbiota signatures and serum metabolite changes in postmenopausal women with reduced BMD, postmenopausal individuals with normal or reduced BMD were recruited and divided into normal and OS groups. Feces and serum samples were collected for 16S rRNA gene sequencing, liquid chromatography coupled with mass spectrometry (LC-MS)-based metabolomics and integrated analysis.

Results: The results demonstrated that bacterial richness and diversity were greater in the OS group than in the normal group. Additionally, distinguishing bacteria were found among the two groups and were closely associated with the BMD index and bone turnover markers. Metabolomic analysis revealed that the expression of serum metabolites, such as etiocholanolone, testosterone sulfate, and indole-3-pyruvic acid, and the corresponding signaling pathways, especially those involved in tryptophan metabolism, fatty acid degradation and steroid hormone biosynthesis, also changed significantly. Correlation analysis revealed positive associations between normal group-enriched Bacteroides abundance and normal group-enriched etiocholanolone and testosterone sulfate abundances; in particular, Bacteroides correlated positively with BMD. Importantly, the tryptophan-indole metabolism pathway was uniquely metabolized by the gut bacteria-derived tnaA gene, the predicted abundance of which was significantly greater in the normal group than in the control group, and the abundance of Bacteroides was strongly correlated with the tnaA gene.

Discussion: Our results indicated a clear difference in the gut microbiota and serum metabolites of postmenopausal women. Specifically altered bacteria and derived metabolites were closely associated with the BMD index and bone turnover markers, indicating the potential of the gut microbiota and serum metabolites as modifiable factors and therapeutic targets for preventing osteoporosis.

Keywords: gut microbiota; osteoporosis; postmenopausal women; serum metabolites; tryptophan-indole metabolism.

Figure 1

Characteristic the gut microbiota composition. (A, B) The alpha diversity comparison of microbiota between two groups. (C) PCoA analysis of gut microbes at genus level. (D, E) The gut microbiota composition in normal and OS cases at phylum and genus levels, respectively. (F–L) The significant difference of gut bacteria at genus level. *: p<0.05, **: p<0.01.

Figure 2

Characteristic the function of gut microbiota. (A) LDA analysis of gut microbiota between two groups at genus level. (B) Spearman’s correlation between significantly altered bacteria and clinical index. (C) The remarkably difference predicted function of bacteria between two groups. (D) The correlation analysis of predicted function and bacteria abundance. *: p<0.05.

Figure 3

The evaluation of gut microbiota interaction and community stability. (A, B) The correlation ship among normal and OS cases, respectively. The orange and green dote indicated bacterial which dominated in OS and normal cases, respectively. (C, D) The bacterial co-occurrence network in two groups at ASVs level. (E) The vulnerability of co-occurrence network between two groups. (F) The comparison of average variation degree between two groups. **: p<0.01.

Figure 4

Characteristic of the serum metabolites. (A, B) The OPLS-DA analysis of serum metabolites between two groups in PIM and NIM, respectively. (C, D) The VIP score of metabolites in PIM and NIM, respectively. (E, F) Volcano plots indicated the different metabolites in PIM and NIM, respectively. (G, H) The metabolites’ sets and KEGG pathway enrichment based on differ metabolites.

Figure 5

The origin analysis of serum metabolites. (A) The source of differ serum metabolites between two groups. (B) Venn plot indicated the number of metabolites origin from host and gut microbiota. (C) The pathway enrichment based on host and gut microbiota derived metabolites. (D–F) The illustration of significantly enriched pathways.

Figure 6

The interaction of gut microbiota and serum metabolites. (A) The altered predicted function of gut microbiota and significantly enriched pathway based on metabolites. (B) The predicted gut microbiota tnaA gene abundance based on PICRUTS2. (C) The correlation between tnaA gene abundance and gut bacteria. (D) The comparison of Bacteroides at species level. (E) The correlation between tnaA gene abundance and Bacteroides species. (F) Mantel’s analysis the interaction of differ gut microbiota, serum metabolites and clinical index. *: p<0.05, **: p<0.01.

Figure 7

Multiple markers for diagnosis of reduced BMD. (A–C) The separated gut microbiota, serum metabolites of PIM and NIM diagnose OS from normal subjects. (D) The accuracy of combined marker panels for diagnose OS from normal subjects.