Effects of traditional Chinese medicine Zuo-Gui-Wan on gut microbiota in an osteoporotic mouse model

Abstract

Background

The target and mechanism of oral traditional Chinese medicine (TCM) have been important research directions for a long time. The close relationship between osteoporosis and gut microbiota (GM) has been confirmed. However, the relevance of oral TCM and the “Gut-Bone Axis” is still poorly understood.

Methods

Twenty-one SPF C57BL/6J female mice were divided into sham (Sham), ovariectomized (OVX), and Zuo-Gui-Wan-treated (ZGW, 1.4 g/kg) groups. The osteoporosis mouse model was established through ovariectomy. After eight weeks of Zuo-Gui-Wan treatment via gavage, serum calcium, phosphorus, ALT, AST, CREA, and other biochemical indicators were measured. Subsequently, Micro-CT, HE staining, and analysis of gut microbiota were conducted to further explore the potential mechanism.

Results

The anti-osteoporotic effects of ZGW were confirmed through micro-CT, histological, and biochemical tests in an OVX-induced osteoporosis mouse model. ZGW treatment also alters the diversity and composition of the gut microbiota and altered the Firmicutes/Bacteroidetes ratio. Further analysis reveals a correlation between specific bacterial groups and serum indicators. Mfuzz clustering analysis and metagenomeSeq analysis identified important microbiota species that were rescued or modulated by ZGW treatment.

Conclusion

These findings suggest that changes in gut microbiota abundance may be linked to ZGW’s ability to improve osteoporosis. This study provides new insights into how ZGW treats osteoporosis, though further research is needed to clarify the mechanisms by which specific gut microbiota influence bone health.

Introduction

Osteoporosis is a chronic skeletal disorder characterized by low bone mass and microarchitectural deterioration of bone tissue, leading to increased bone fragility and fracture risk [1]. It is a major public health problem that affects millions of people worldwide, especially postmenopausal women [2, 3]. Osteoporotic fractures can cause severe pain, disability, and even death, thus impairing the quality of life and increasing the health care costs [4, 5]. While current treatments can partially mitigate the symptoms of osteoporosis, there remains a pressing demand for more efficacious and safe therapeutic options for patients [6]. Therefore, it is imperative to investigate and develop effective strategies for the prevention and management of osteoporosis.

Traditional Chinese medicine (TCM) has a long-standing history of treating various bone diseases, including osteoporosis [7,8,9]. Unlike single-agent drugs, traditional Chinese medicine prescriptions for osteoporosis offer potential advantages due to their ability to target multiple signaling pathways and exhibit diverse biological functions [10]. Zuo-Gui-Wan (ZGW), traditionally described as having the effects of “nourishing the kidney and replenishing the essence,” contains a variety of active components, including loganin, quercetin, oleanolic acid, and isorhamnetin [11]. Research indicates that various compounds in ZGW are closely associated with its pharmacological effects in preventing osteoporosis [11,12,13,14,15]. However, the precise mechanisms underlying its therapeutic effects remain insufficiently understood.

On the one hand, numerous studies have established a significant connection between gut microbiota and osteoporosis, providing evidence for the existence of a “Gut-Bone Axis” [16,17,18]. Gut microorganisms influence bone metabolism through various mechanisms, including enhancing nutrient absorption [19, 20], regulating host endocrine [21], immune system [22], and other pathways [23]. The “Gut-Bone Axis” is consistent with the traditional Chinese medicine theory. From the perspective of traditional Chinese medicine, because the kidney is the congenital foundation, and the spleen is the acquired foundation, they depend on each other and serve each other [24]. Therefore, the intestine functions of intestinal microbiota are closely related to the basic pathological changes of “kidney yin deficiency” in osteoporosis patients [25]. Studies have reported that traditional Chinese medicine can regulate bone metabolism through multiple pathways, such as modulating host immunity [26,27,28] and estrogen metabolism [29].

On the other hand, traditional Chinese medicine (TCM) components are often not directly absorbed by the host after oral administration. Instead, they enter the gut, where they are transformed by the gut microbiota. Ingesting different TCMs can affect the abundance of potentially beneficial (e.g., anti-inflammatory or short-chain fatty acids (SCFAs)-producing) and harmful (e.g., pro-inflammatory and pathogenic) bacteria in varying ways. Thus, changes in gut microbiota composition are closely related to the development of different immune and metabolic activities within the host [30]. Moreover, the poor oral bioavailability of most active TCM components cannot account for their significant biological effects. The unabsorbed TCM components can be metabolized by the gut microbiota, with some metabolites being absorbed into the bloodstream and subsequently metabolized by the liver, where their products can exert biological activity [31]. Consequently, research on TCM and its regulation of bone metabolism through the gut microbiota (GM) has garnered widespread attention.

Recent studies have shown that TCM can positively influence bone metabolism by modulating gut microbiota composition. Traditional Chinese medicine compound and its effective component have been reported to prevent bone loss and improve bone structure in osteoporotic models by altering gut microbiota [32, 33]. While there is evidence supporting the beneficial effects of TCM on bone health through gut microbiota modulation, the exact mechanisms and interactions are not fully understood. More studies are needed to elucidate these complex relationships and confirm causality. Treating osteoporosis from the perspective of GM may overcome the shortcomings of existing clinical treatment, and could be a new target for preventing and treating postmenopausal osteoporosis.

In this study, high-throughput sequencing of 16 S rRNA from mouse intestinal contents was performed, to explore the therapeutic effect of ZGW on osteoporotic mice models and investigate its possible involvement in GM.

Materials and methods

ZGW preparation

The ingredients of ZGW include eight medicinal herbs: Rehmanniae Radix Praeparata (Shudihuang in Chinese, 24 g), Dioscoreae Rhizome (Shanyao in Chinese, 12 g), Corni Fructus (Shanzhuyu in Chinese, 12 g), Lych Fructus (Gouqizi in Chinese, 12 g), Cyathulae Radix (Chuanniuxi in Chinese, 9 g), Cuscutae Semen (Tusizi in Chinese, 12 g), Cervi Cornus Colla (Lujiaojiao in Chinese, 12 g), Testudinis Tarapacis Et Plastri Colla (Guibanjiao in Chinese, 12 g). The raw herbs of Zuo-Gui-Wan were obtained from the Sichuan Provincial Orthopedic Hospital and verified by the research team. The herbs were decocted in distilled water and the decoction was filtered. The filtrate was frozen at -80 ℃ for 2 h, then transferred to a designated facility and vacuumed to 10 Pa. The lyophilization process lasted for 72 h, resulting in a freeze-dried powder. The powder was stored at -20 ℃ until further use.

Animals and drug administration

The animal protocol was reviewed and approved by the Animal Ethics Committee of Chengdu University of Traditional Chinese Medicine (Record number: 2020-34). All procedures involving animals were conducted in accordance with the ethical standards of the institution and the national guidelines for the care and use of laboratory animals. Female C57BL/6J mice (n = 21, 12 weeks old) were obtained from SPF (Beijing) Biotechnology (permit No. SCXK(Jing) 2019-0010) and housed at Chengdu University of Traditional Chinese Medicine (permit No. SYXK (chuan) 2019-049). The mice were kept at 22–26 °C with 45–55% humidity and provided free access to water and diet. One week was allowed to adapt to their new environment. Following one week of acclimation, mice were randomly assigned to three groups of seven animals each: sham-operated (Sham), ovariectomized (OVX) and Zuo-Gui-Wan-treated. Prior to any invasive procedures, the animals were anesthetized using intraperitoneal injection of 1.25% Avertin (Tribromoethanol in tert-amyl alcohol) at a dosage of 0.02 mL/g. We have detailed the animal modeling process in our previous work [34]. In brief, we performed bilateral ovariectomy on the OVX and ZGW groups, while the Sham group had an equivalent amount of fat removed as the ovaries. Ovariectomized mice are a well-established animal model of osteoporosis [35]. After a one-week postoperative recovery, the mice received daily gavage administration of 10 ml/kg of either distilled water (OVX and Sham groups) or ZGW solutions (1.4 g/kg, ZGW group) for 8 weeks. The dose of ZGW was based on the human equivalent dose (18 g/kg) calculated by body surface area. The animals were sacrificed using decapitation. This method was chosen to ensure a quick and humane end to the animals’ lives. To minimize suffering, all procedures were performed by trained personnel under strict adherence to ethical guidelines. Additionally, animals were closely monitored for any signs of distress or discomfort. Appropriate measures were taken to alleviate any observed suffering, including providing a warm and quiet environment to reduce stress, ensuring easy access to food and water to promote recovery, and administering additional analgesics if signs of unmanageable pain were observed.

Specimen preparation

Blood samples were collected from the orbits before euthanasia. To avoid post-mortem bacterial changes, we collected the full intestinal contents as soon as possible and immediately froze them in liquid nitrogen. The intestinal contents were stored at -80 °C until DNA extraction. The femurs of mouse were isolated and carefully removed of surrounding soft tissues for micro-CT scanning. The bones were fixed in 4% paraformaldehyde after being scanned with micro-CT.

Micro-CT analysis

The microstructure of the fresh femurs were scanned by a Quantum GX (PerkinElmer, Waltham, MA) instrument with a pixel size of 10 μm. The region of interest (ROI) was manually selected as the area about 540 μm above the distal femur growth plate (layer thickness of 1.5 mm), and the bone volume fraction (BV/TV, %), trabecular separation (Tb.Sp, mm), and trabecular thickness (Tb.Th, mm) were calculated. Data were calculated and exported using PerkinElmer Analyze 12.0 software (PerkinElmer Waltham, MA) with its Bone Microarchitecture Analysis (BMA) Add-On. The trabecular 3D structure was also reconstructed for visual comparison with 3D Slicer program (Slicer 5.2.2; slicer.org). The imaging parameters were set as follows: 18*18 mm FOV, 90 kV voltage, 88 µA current, and 14-min high resolution scan time.

Histopathological analysis

After fixation in 4% paraformaldehyde solution for 48 h, the femurs of all groups were decalcified for about three to four weeks using 0.5 M EDTA. Samples were dehydrated with graded ethanol, cleared with xylene twice, and embedded in paraffin. Sections were visualized and digitally recorded after staining with hematoxylin and eosin. The percentage of trabecular bone area (%) was calculated as trabecular bone area/analysis area*100%.

Serum biochemical analysis

To evaluate the effects of ZGW on liver and kidney function and bone metabolism, we tested multiple indicators in the serum collected from each group, including serum alanine aminotransferase (ALT), serum aspartate aminotransferase (AST) and serum creatinine (CREA) to verify whether ZGW had obvious liver and kidney toxicity at this dose. To preliminarily assess the bone metabolism, we measured the serum concentrations of calcium (Ca) and phosphorus (P) using an automatic biochemistry analyzer (BS360S, Mindray). According to the manufacturer’s instructions, the ELISA kit (E-EL-M0200c, Elabscience) was utilized to measure the levels of bone-specific alkaline phosphatase (BALP), a biomarker for bone formation in serum [36].

Gut microbiota analysis

Referring to the report of other study [37], we determined the following steps for sequencing process. Firstly, sample collection and DNA purification: genomic DNA (gDNA) was extracted from the samples using the Zymo Research BIOMICS DNA Microprep Kit (Cat# D4301), following the manufacturer’s instructions. Second, target amplification: we specifically amplified the V4 region of the 16S rDNA using custom-designed primers. Primer sequences: forward primer (5’-3’): 515 F (5’-GTGYCAGCMGCCGCGGTAA-3’), reverse primer (5’-3’): 806R (5’-GGACTACHVGGGTWTCTAAT-3’). Third, PCR replicates and library construction: PCR reactions were performed using TOYOBO KOD-Plus-Neo DNA Polymerase (KOD-401B) in an Applied Biosystems PCR System 9700. Cycling conditions:1. Initial denaturation at 94 °C for 1 min (one cycle). 2. Denaturation at 94 °C for 20 s, annealing at 54 °C for 30 s, and extension at 72 °C for 30 s (25–30 cycles). 3. Final extension at 72 °C for 5 min (one cycle). 4. Hold at 4 °C. Three technical replicates of PCR were performed for each sample. PCR products within the linear range were pooled equally for subsequent library preparation. Fourth, fragment detection and quantification: Target fragments were detected by electrophoresis on a 2% agarose gel. DNA quantification was performed using the Qubit 2.0 Fluorometer (Thermo Scientific). Fifth, library preparation: The library was constructed using the NEBNext Ultra II DNA Library Prep Kit for Illumina (NEB#E7645L) from NEW ENGLAND BioLabs. Sixth, Sequencing: PE250 sequencing mode was employed. The sequencing reagent kit used was NovaSeq 6000 SP Reagent Kit V1.5 from Illumina. After that, the sequencing data are analyzed as follows. With FLASh v1.2.11 and default parameters, paired-end sequences were assembled, and sequences were demultiplexed using Sabre v1.000, allowing zero base mismatches between the sequences. Denoising sequences was performed using the QIIME2 DADA2 plugin with default parameters and a 200-bp trim length to remove those with a base quality below 30. Vegan’s vegdist function was used to calculate the Bray-Curtis distance matrix. Then, the PCoA analysis was performed using the pcoa function in the ape package in R (https://www.r-project.org/). To obtain metabolic information from KEGG, we used Picrust2 to predict the microbial functions of the community. As part of the LEfSe analysis, we used the LEfSe software (https://huttenhower.sph.harvard.edu/lefse/). The Mfuzz analysis was carried out using the Mfuzz package in R.

Statistical analysis

Data were described as mean ± SD. The Shapiro-Wilk test was first performed on all data to determine its normality. Two-tailed single-sample t-tests or factor variance analysis (ANOVA) was used to determine whether the difference in the mean was significant, and post-hoc Bonferroni tests were used to compare each group. The test level was α = 0.05. Statistical tests were conducted with IBM SPSS 26.0 (IBM UK Ltd.; Portsmouth, United Kingdom) to detect statistically significant differences. Processing and analysis of all digital images were performed using FIJI/ImageJ (https://fiji.sc/) software.

Results

In vivo experiments on ZGW show good safety

All postoperative mice survived, with no wound infections, and showed normal activity throughout the experiment. To investigate the safety of equivalent ZGW in mice, we monitored their body weight during the entire experiment (Fig. 1A). A loss of body weight was observed in each group after surgery, which might be related to surgical trauma. Across all time points, no significant difference in body weight was observed between the ZGW and OVX groups. We tested serum indicators of mice in each group to assess the pharmacological effects of ZGW on bone, liver, and kidney metabolism (Fig. 1B). Neither serum calcium nor serum phosphate levels were significantly different between groups. The serum analysis of the mice in each group indicated that oral administration of ZGW for 8 weeks did not cause significant hepatorenal toxicity at the current equivalent dose. CERA, ALT and AST serum levels were within the normal reference range, and no statistically significant differences were found between the groups.

Fig. 1

Effects of Zuo-Gui-Wan on osteoporosis. (A) Mouse weight changes over time before and after surgery. The line graph shows the mean ± SD of mouse weight measured weekly for 9 weeks. The surgery was performed at week 1. (n = 7). (B) Serum levels of phosphorus (P), calcium (Ca), alanine transaminase (ALT), aspartate transaminase (AST), creatinine (CREA), and bone alkaline phosphatase (BALP). (n = 7). (C) Representative micro-CT images of the distal femoral trabeculae. The red dashed area represents the region used for three-dimensional reconstruction. (D) Microstructural parameters of ROI in the distal femoral trabeculae for each group. (n = 5). (E) Hematoxylin and eosin (H&E) staining of distal femur sections. (n = 7). (F) Quantitative analysis of the trabecular bone area of the total tissue volume in the region of interest of HE images. (n = 7). (“ns”: no significant difference; ∗: P < 0.05; ∗∗: P < 0.01)

ZGW can partially reverse OVX-induced osteoporosis

After 9 weeks of ovariectomy, the successful establishment of an osteoporotic mouse model was confirmed by Micro-CT. To be specific, as compared with the Sham group, the OVX group had abnormal bone trabecular structure at the distal femur, manifested as osteoporosis (Fig. 1C). The microstructure of bone trabeculae also changed: the volume fraction of bone trabeculae (BV/TV) decreased, the trabecular thickness (Tb.Th) decreased, and trabecular separation (Tb.Sp) with increased. The differences between the two groups in each index were statistically significant (Fig. 1D). It indicates that a successful mouse model of osteoporosis after menopause has been established, and significant changes have occurred in the microstructure of bone trabeculae compared with the Sham group. Although there was no statistically significant difference, ZGW also decreased Tb.Sp compared with the OVX group. Subsequently, histopathological examination further confirmed the changes observed in Micro-CT analysis (Fig. 1E). Compared with the Sham group, the OVX mice exhibited a significant reduction in trabecular area and number, as well as the presence of numerous fat-like granules. After 8 weeks of ZGW intervention, the osteoblasts around the trabeculae increased, and the trabecular microstructure improved markedly. Improvement in bone trabeculae was also confirmed by statistical analysis (Fig. 1F). In conclusion, OVX-induced bone loss and associated changes can be partially reversed with an 8-week ZGW gavage intervention.

ZGW affects the bacterial diversity and abundance of the GM

OVX-induced estrogen deficiency significantly affects the composition of the GM. Principal coordinate analysis (PCoA) (Fig. 2A) and Anosim analysis (Supplementary Material) showed that the GM structure of the OVX group was significantly different from the Sham group. The GM composition of the Sham and ZGW groups were partially similar (Fig. 2A). After OVX, the gut community structure changed significantly, while the dominant phyla were still Bacteroidetes, Firmicutes, Proteobacteria, and Epsilonbacteraeota (Fig. 2B). To ensure clarity, the relative abundance of the top genera has been visually represented in the provided boxplot (Fig. 2C). The genera Lachnospiraceae NK4A136 group, Alloprevotella, and Bacteroides are highlighted as the most dominant genera across the groups, with noticeable differences in their abundance patterns. Alpha diversity showed that the gut microbial richness and evenness of the OVX group were slightly higher than those of the Sham and ZGW groups. The OVX group differed significantly from the sham group in the Chao1 and Simpson indices, while ZGW partially reversed this difference (Fig. 2D). To identify unique and shared bacterial operational taxonomic units (OTUs) among different groups, a Venn diagram analysis was performed. The analysis revealed that there were 260, 713 and 388 unique OTUs in Sham, OVX and ZGW groups, respectively (Fig S1).

Fig. 2

Microbial composition and diversity across Sham, OVX, and ZGW groups. (A) Principal Co-ordinates Analysis (PCoA) based on Weighted UniFrac distance. Each point represents an individual sample, and ellipses indicate the 95% confidence intervals for each group. (B) Stacked bar plot of the top 10 species with abundance at the phylum level across the three groups. (C) The boxplot shows the relative abundance of microbial genera across the three groups. (D) Alpha diversity indices, including Chao1, Faith’s Phylogenetic Diversity (PD), Shannon, and Simpson indices, are shown for each group. (For each group n = 6)

We analyzed the correlation between GM and serum calcium, phosphorus and BALP in OVX mice. Regression analysis was used to assess relationship between serum biomarkers and Shannon, ACE index, Observed species, and PCoA axis 1 and 2 scores as a representative of bacterial community structure (Fig. 3A). The analysis showed that the bacterial diversity of GM was not strongly correlated with serum bone metabolism related biomarkers. Also, we performed dbRDA-bray-curtis analysis to examine the relationship between GM and serum biomarkers. As shown in Fig. 3B, serum phosphorus and serum BALP levels correlate with Alitipes and Bacteroides (at the genus level).

Fig. 3

(A) Bivariate correlations among important variables and serum biomarkers. The color indicates the direction of correlation. ∗∗: P < 0.01. (B) Distance-based redundancy analysis (db-RDA) between serum biomarkers and bacteria phyla

ZGW affects the difference in bacterial abundance of gut microbiota

We further performed LEfSe analysis on the samples to examine which taxa had the largest differences among different groups. Figure 4 A and 4B show that there are many taxa that are common among different groups (yellow circles), but there are also some that are specific to each group. The OVX group has a higher relative abundance of Firmicutes, whereas the Sham and ZGW groups have increased relative abundances of Bacteroidetes and Epsilonbacteraeota. Additionally, a higher Firmicutes/Bacteroidetes (F/B) ratio was observed in the OVX group than in the Sham group, and ZGW intervention reduced the F/B ratio (Fig. 4C). To compare the GM between the ZGW group and the OVX group, we performed metagenomeSeq analysis. Results revealed that some bacteria had significant discriminability between the two groups (Fig. 4D). ZGW intervention reduced the relative abundance of Epsilonbacteraeota, Proteobacteria and Cyanobacteria, and regulated the relative abundance of Firmicutes and Bacteroidetes (Fig. 4D).

Fig. 4

Species difference analysis. (A) LDA score of Lefse. LDA scores > 3 and P < 0.05 are shown. (B) Cladogram of Lefse. (C) Ratio of Firmicutes/Bacteroides. (D) MetagenomeSeq analysis. Log2 (fold change) and difference significance of gene expression in ZGW-OVX

As for KEGG, 3 pathways showed significant differences in the secondary pathway (LDA > 2.5, P < 0.05, Fig S2), among which there were significant differences in pathways (Lipopolysaccharide biosynthesis, biosynthesis and metabolism and Glycan biosynthesis and metabolism) in Sham, 9 pathways in sham such as “ABC transporters”, “ABC transporters Membrane transport”, “Membrane transport Environmental Information Processing”, 9 pathways such as “Two component system”, “Signal transduction”, “Oxidative phosphorylation” in ZGW.

In Mfuzz clustering analysis of GM, we divided them into four groups based on changing trends of differential flora (Fig. S3). Among 605 genera detected, Cluster 1 had 56 genera, Cluster 2 had 127 genera, Cluster 3 had 49 genera, and Cluster 4 had 30 genera. Specifically, relative abundance of shared GM in Cluster 1 and Cluster 2 were higher in the OVX group, but lower in the ZGW group. Therefore, the altered GM in Cluster 1 and Cluster 2 were considered as important microbiota that were rescued or modulated by ZGW treatment. The important microbiota obtained by cluster analysis and the differential microbiota obtained by metagenomeSeq analysis shared the followings: Rikenella, Bacteroides, Helicobacter, Allobaculum, Family XIII AD3011 group, ASF356, Ruminococcaceae UCG-013, Ruminiclostridium 6 and Ruminococcaceae UCG-010. It may help us address some unresolved questions about the interaction between gut microbiota and osteoporosis in the future.

Discussion

It is estimated that the age-standardized prevalence of osteoporosis among middle-aged and elderly Chinese residents is 33.49% [38, 39], and it is expected to continue to rise in the future with the aggravation of population aging [40, 41]. As a common metabolic bone disease in the elderly, osteoporosis can cause low back pain, general weakness, and even lead to fractures and disabilities, seriously affecting the quality of life of patients [42]. Therefore, it is imperative to have reasonable and effective prevention and treatment management for osteoporosis. The main types of clinical treatment for osteoporosis are bone resorption inhibitors, bone formation stimulators, and dual-action drugs. However, these treatments have some limitations, such as causing adverse reactions in the digestive tract and nervous system, and being expensive [43, 44]. Traditional Chinese medicine formulas consist of complex active ingredients that target multiple sites for the treatment of osteoporosis, achieving systemic regulation and exhibiting advantages such as good efficacy and low toxicity and side effects [45]. In summary, our experiments show the pharmacological effect of ZGW in preventing OVX-induced osteoporosis.

ZGW is a classic traditional Chinese medicine formula with significant effects in improving osteoporosis. Network pharmacology analysis has identified many active components in ZGW, among which key constituents such as quercetin and kaempferol play critical roles in regulating bone metabolism and mitigating oxidative stress [46]. These findings highlight the molecular foundation of ZGW’s therapeutic potential. However, in oral administration, ZGW’s active ingredients aren’t directly absorbed. For instance, polysaccharides and saponins are minimally absorbed and depend on gut microbiota to convert them into bioactive metabolites [47]. Therefore, investigating whether ZGW exerts its therapeutic effects through modulation of the gut microbiota is particularly important.

The effect of GM diversity on host osteoporosis is currently unclear. Several studies have suggested that individuals with osteoporosis exhibit a higher gut microbiome diversity compared to healthy individuals, with a negative correlation observed between the Alpha diversity index and bone mineral density (BMD) [18, 48]. Furthermore, animal studies have shown an increase in gut microbiome Alpha diversity in rats undergoing ovariectomy [49], while other osteoporosis model rats demonstrate lower Alpha diversity [50, 51]. In this study, GM compositions were different between sham and OVX groups, and also between ZGW and OVX groups. The Chao1 and Shannon indices of GM were significantly higher in osteoporosis model mice than in sham mice, which was opposite to the changes reported in osteoporosis model rats [50, 51]. ZGW appeared to diminish the alpha diversity of the mouse microbiota, a finding that aligns with prior research demonstrating that a decrease in gut microbiome diversity can potentially mitigate bone loss resulting from ovariectomy [52]. Insufficient sample size might explain this phenomenon, which could be investigated further in the future.

Another potential mechanism by which GM could affect bone health is by modulating mineral absorption. Previous studies have shown that increasing the concentration of specific microbiota, such as Lactobacillus and Bifidobacterium, can enhance the mineral absorption and increase the bone mineral density [53]. In our study, we did not find any significant correlation between the serum levels of calcium, phosphorus and BALP and the diversity indices of species. On the other hand, we found a correlation between Alistipes and Bacteroides and serum biomarkers. Previous studies have reported that Bacteroides, which are involved in vitamin K metabolism [54], were more abundant in osteoporosis patients than in health controls. Vitamin K is essential for bone homeostasis, as it activates osteocalcin, a protein that regulates bone mineralization [55]. Moreover, study have also shown that reducing the abundance of Alistipes in the gut seems to be associated with reduced inflammation and improved osteoporosis [50]. Therefore, the relationship between specific bacteria and serum markers related to osteoporosis needs to be further explored.

Firmicutes and Bacteroidetes are two major phyla of bacteria in the human gut microbiota. When studying the gut microbiome, the abundance ratio of Firmicutes to Bacteroidetes (F/B ratio) is often used to investigate differences between various health conditions and their correlations with conditions such as osteoporosis [56]. The Firmicutes and Bacteroidetes in the gut ferment dietary fibers, producing SCFAs such as butyrate, propionate, and acetate. These SCFAs impact host metabolism in multiple ways by interacting with G-protein coupled receptors expressed in enteroendocrine cells [57]. As a result of ZGW administration, Bacteroidetes were significantly more abundant and Firmicutes were significantly less abundant, leading to a significant reduction in the F/B ratio. By influencing the production of SCFAs by the GM, the F/B ratio plays an important role in osteoporosis. Li [28] and Zhang [58] reported that traditional Chinese medicine can reduce the Firmicutes/Bacteroidetes ratio and exert a protective effect against osteoporosis. In our study, we found that ovariectomy intervention increased the F/B ratio, while ZGW treatment reversed this change. These indicates that ZGW intervention may could reduce the proportion of pro-inflammatory microbes.

Notably, Lachnospiraceae NK4A136 group was the most abundant genus in the ZGW group, with a significant increase in abundance compared to both OVX and Sham groups, indicating a potential role in ZGW’s therapeutic effects. The Lachnospiraceae NK4A136 group is a specific genus within the family Lachnospiraceae, known to play a pivotal role in the gut microbiota, particularly in the production of SCFAs. SCFAs, the primary metabolites produced by gut microbial fermentation of dietary fibers, are crucial for maintaining gut health and regulating host metabolism, immunity, and inflammatory responses. Previous Mendelian randomization studies have demonstrated that the Lachnospiraceae NK4A136 group exhibits protective effects against osteoporosis [59]. This aligns with our findings, where the abundance of the Lachnospiraceae NK4A136 group was significantly reduced in the OVX group but markedly increased in the ZGW-treated group. These results suggest that the Lachnospiraceae NK4A136 group may play a protective role in maintaining gut barrier integrity and immune function, thereby influencing the onset and progression of postmenopausal osteoporosis [60, 61]. However, the precise functional roles and mechanisms of specific genera, such as the Lachnospiraceae NK4A136 group, in postmenopausal osteoporosis remain unclear and warrant further investigation.

The functional prediction results of the gut microbiota (GM) suggest that ZGW’s potential to improve osteoporosis may involve regulation across diverse KEGG pathways. KEGG enrichment analysis serves as a valuable tool to infer the roles of gut microbiota in metabolic pathways. Monitoring changes in these pathways is expected to shed light on the underlying mechanisms and therapeutic targets of osteoporosis, providing a robust reference for future studies. While these KEGG pathways are not directly linked to osteoporosis, they are integral to various cellular processes such as signal transduction, energy metabolism, and cell motility. By integrating KEGG analyses of gut microbiota with ZGW’s network pharmacology [11], we hypothesize that the gut microbiota in the ZGW treatment group may participate in pathways such as oxidative phosphorylation. This pathway likely contributes to regulating gut microbiota balance, enhancing the production of metabolites like SCFAs, and subsequently activating critical pathways such as PI3K/AKT signaling. These actions collectively promote bone formation and gut health. Future research should combine microbiomics and metabolomics to further elucidate these mechanisms.

Furthermore, we used Mfuzz analysis to observe the changes of intestinal microbiota under different interventions. The significance of Mfuzz analysis is to cluster the species with the same distribution patterns by clustering [62]. The species with the same distribution patterns indicate that these species have the same co-occurrence patterns in microbial ecology, which can help to elucidate the community assembly mechanisms. Mfuzz analysis showed that many osteoporosis-related microbiota in the Sham group and ZGW group had similar changes, such as Bacteroides, Prevotellaceae, etc. Our study results indicate that ZGW can protect mice from ovariectomy-induced osteoporosis by potentially modulating gut microbiota diversity, changing the Firmicutes to Bacteroidetes ratio and altering the relative abundance of specific bacterial genera, including Alistipes and Lachnospiraceae NK4A136 group.

Conclusions

Our study demonstrated that ZGW improves osteoporosis by modulating gut microbiota composition at both phylum and genus levels. Key bacterial taxa, such as Bacteroides and Lachnospiraceae NK4A136, were identified as potential mediators of ZGW’s therapeutic effects. These findings provide new insights into the “Gut-Bone Axis” and its role in bone metabolism, highlighting the potential of TCM in osteoporosis prevention and treatment. Further research integrating microbiomics and metabolomics is necessary to clarify the mechanisms by which gut microbiota influence bone health and to develop targeted interventions for osteoporosis.