Methods for the joint construction and evaluation of microbial-metabolomics in different types of osteoporosis animal models
By constructing a combined microbiome-metabolome evaluation method, the problem of lack of microecological and metabolic characteristics in the evaluation of osteoporosis models was solved. This method enables multi-indicator joint evaluation, screening out model-specific microbiota and metabolic biomarkers, and supporting drug screening and intervention research.
Patent Information
- Application Number
- CN202611291843.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-25
- Publication Date
- 2026-11-17
AI Technical Summary
Current evaluation of osteoporosis animal models mainly relies on a single bone phenotype, lacking comparable microecological and metabolic characteristics. There is a lack of joint evaluation between the microbiota and metabolites, resulting in an imperfect classification evaluation system and a shortage of early diagnostic biomarkers.
We employed parallel control, postmenopausal osteoporosis model, disuse osteoporosis model, and glucocorticoid-induced osteoporosis model groups to perform bone phenotype verification, gut microbiota analysis, fecal metabolomics analysis, and microbiota-metabolite association analysis. We constructed a combined microbiome-metabolome evaluation method, screened differentially expressed bacterial genera using LEfSe and Wilcoxon rank-sum tests, screened differentially expressed metabolites using PCA and OPLS-DA models, and constructed a microbiota-metabolite association network.
It enabled the comparability evaluation of different types of osteoporosis models, screened model-specific microbial biomarkers and common metabolic biomarkers, and formed a non-invasive, multi-indicator combined model evaluation system to support the screening of anti-osteoporosis drugs and microecological intervention research.
Smart Images

Figure CN123100562A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of biomedical model construction technology, specifically to a method for the combined construction and evaluation of microbial-metabolomics in different types of osteoporosis animal models. Background Technology
[0002] Osteoporosis (OP) is a systemic metabolic bone disease characterized by decreased bone mineral density, deterioration of bone microstructure, and increased susceptibility to fractures. The original text states that there are approximately 200 million OP patients worldwide, with about 9 million fragility fractures occurring annually. Osteoporosis and related fractures impose a heavy burden on society and families. Therefore, establishing stable, reproducible animal models that reflect different pathogenesis mechanisms is a crucial foundation for conducting pathogenesis research, drug screening, and intervention evaluation.
[0003] Based on etiology, osteoporosis can be divided into primary and secondary types. Postmenopausal osteoporosis is a typical type of primary osteoporosis, while disuse osteoporosis and glucocorticoid-induced osteoporosis are typical types of secondary osteoporosis. The main pathological bases of these three types are not the same: postmenopausal osteoporosis is mainly related to high turnover bone loss due to estrogen deficiency; disuse osteoporosis is mainly related to bone formation inhibition due to lack of mechanical stimulation; and glucocorticoid-induced osteoporosis is related to glucocorticoids interfering with osteogenic differentiation and promoting osteoclast formation. Due to differences in pathogenesis and bone turnover status, different types of models may also respond differently to intervention factors and drugs.
[0004] Current animal model evaluations primarily focus on host bone phenotypes such as bone mineral density, bone microstructure, bone biomechanics, serum bone turnover markers, and histological morphology. While these indicators can confirm changes in bone mass and quality, they remain insufficient for differentiating between different model types, identifying model characteristics from non-invasive samples, and revealing the link between gut microbiota and host metabolism. In particular, there is a relative lack of early diagnostic biomarkers for osteoporosis, an incomplete classification and evaluation system, and insufficient cross-omics analysis methods.
[0005] Changes in the gut microbiota can interact with the host through metabolites, and fecal samples are convenient, non-invasive, and suitable for continuous observation. Therefore, joint analysis of gut microbiota and fecal metabolites, and cross-validation with classic bone phenotypes, can provide a more comprehensive evaluation dimension for different types of osteoporosis animal models. Based on this, it is necessary to establish a method for the joint construction and evaluation of the microbiome and metabolome in different types of osteoporosis animal models. Summary of the Invention
[0006] This invention aims to address the problems of existing osteoporosis animal model evaluation mainly relying on a single bone phenotype, lack of comparable microecological and metabolic characteristics between different model types, and lack of joint evaluation between microbiota and metabolites, and provides a method for the joint construction and evaluation of microbial-metabolome in different types of osteoporosis animal models.
[0007] To solve the above-mentioned technical problems, this invention adopts the following technical solution: parallel control group, postmenopausal osteoporosis model group, disuse osteoporosis model group, and glucocorticoid-induced osteoporosis model group are set up; model establishment and sample collection are completed under the same feeding conditions and the same observation period; bone phenotype verification, gut microbiota analysis, fecal metabolomics analysis, and microbiota-metabolite association analysis are performed respectively. Specifically, the following steps are included: (1) Model construction: The OVX model was constructed by bilateral oophorectomy, the DIO model was constructed by right sciatic nerve resection, and the GIO model was constructed by intramuscular injection of 1 mg / kg dexamethasone every other day for 10 weeks. At the same time, corresponding control groups were set up. (2) Bone phenotype verification: Femoral and blood samples were collected 10 weeks after modeling. Bone mineral density, bone microstructure, bone biomechanics, bone turnover status and histological changes were confirmed by bone Micro-CT, three-point bending test, serum PINP and CTX-I detection and H&E histological staining. (3) Microbial community detection: Fecal samples were collected, total DNA was extracted, the V3-V4 region of the 16S rRNA gene was amplified and sequenced, and after quality control, noise reduction, chimerism removal, ASV construction and species annotation, Alpha diversity, Beta diversity and differential microbial community analysis were performed. (4) Metabolomics detection: Fecal samples were extracted and non-targeted detection in positive and negative ion modes was carried out using ultra-high performance liquid chromatography-high resolution mass spectrometry. Peak extraction, retention time alignment, blank subtraction and metabolite identification were completed. (5) Differential feature screening: LEfSe and Wilcoxon rank-sum tests were used to screen differential bacterial genera; PCA was used to evaluate the overall distribution of the samples; OPLS-DA model was used to screen differential metabolites with VIP>1, P<0.05 and |FC|≥2; and KEGG pathway enrichment was performed on differential metabolites. (6) Construction of association network: Z-score standardization was performed on the relative abundance data of differential bacterial genera and the normalized peak area data of differential metabolites. Spearman rank correlation analysis was used to screen bacterial genera-metabolite pairs with |r|>0.6 and P<0.05, and correlation heatmap and co-expression network were constructed. (7) Joint evaluation: The classic bone phenotype results were used to determine whether the model successfully formed osteoporotic changes. The different model types were characterized by the model-specific differential bacterial genera. The 4-(β-acetaminoethyl)imidazole, which was upregulated in all three models, and its association with key bacterial genera were used as common fecal metabolism evaluation indicators.
[0008] In the extended cross-omics analysis, host transcriptomics data can be integrated based on the aforementioned microbiome and metabolome data. A cross-omics association network can be constructed using bioinformatics methods, and Lasso regression can be used to screen for biomarker combinations with early warning value. This extended analysis does not change the basic technical approach of this invention, which involves joint evaluation based on bone phenotype, microbiome characteristics, and fecal metabolites.
[0009] III. Evaluation Indicators and Judgment Relationships The evaluation of this invention is not based on a single indicator, but consists of three interconnected levels. The first level is bone phenotype confirmation, used to confirm whether animals exhibit osteoporotic changes in bone mass, bone microstructure, bone mechanics, and bone turnover. The second level is model classification, identifying microecological differences through characteristic bacterial genera in different models. The third level is common metabolic characteristics, forming a common evaluation indicator across models through the consistent upregulation of 4-(β-acetaminoethyl)imidazole in the three models and its significant correlation with key bacterial genera.
[0010] The key differentially expressed genera in the OVX model were *g_Bacteroidetes_unclassified*, *g_Firmicutes_unclassified*, and *g_Eggerthella*; the key differentially expressed genera in the DIO model was *g_Akkermansia*; and the key differentially expressed genera in the GIO model was *g_Rothia*. 4-(β-acetaminoethyl)imidazole was negatively correlated with *g_Bacteroidetes_unclassified*, *g_Akkermansia*, and *g_Rothia*, and positively correlated with *g_Firmicutes_unclassified* and *g_Eggerthella*.
[0011] Beneficial effects 1. This invention constructs and compares three different types of osteoporosis models—OVX, DIO, and GIO—in parallel within a unified experimental period, making the models comparable across different etiologies and bone turnover states; 2. This invention uses bone Micro-CT, bone biomechanics, serum bone metabolism indicators and histological results for bone phenotype confirmation, avoiding the evaluation of the model based on only a single indicator; 3. This invention systematically compares the gut microbiota and fecal metabolome characteristics of three models and screens out model-specific microbiota markers such as Eggerthella in the OVX model, Akkermansia in the DIO model, and Rothia in the GIO model. 4. This invention found that 4-(β-acetaminoethyl)imidazole was significantly upregulated in three osteoporosis models and was significantly associated with multiple key bacterial genera, and can be used as a common fecal metabolic marker. 5. This invention uses fecal samples as the target for microbial and metabolomics detection, forming a non-invasive, multi-indicator combined model evaluation system, which can provide an animal experimental basis for the screening of anti-osteoporosis drugs and microecological intervention research. Attached Figure Description
[0012] Figure 1 Micro-CT three-dimensional reconstruction images of the femur of rats in each group and bone microstructure parameters; Figure 2 Results of H&E staining, serum bone metabolism markers, and three-point bending test; Figure 3 ASVVenn plot, Alpha / Beta diversity analysis, and PCoA plot for each group of gut microbiota; Figure 4 Stacked bar charts, heatmaps, and differential bacterial community analysis at the phylum, genus, and species levels; Figure 5 For fecal metabolites PCA, heat map, volcano map and KEGG enrichment bubble map; Figure 6 A heatmap showing the correlation between differentially expressed bacterial genera and metabolites, and a co-expression network diagram. Detailed Implementation
[0013] The present invention will be further illustrated below with specific experimental examples and embodiments. The following embodiments are used to illustrate the technical approach, detection method, and evaluation relationship of the present invention, and are not intended to limit the scope of protection of the present invention. Unless otherwise specified, all groups of samples are processed and detected under the same conditions to ensure comparability between different models.
[0014] Example 1: Construction and Bone Phenotypic Validation of Three Types of Osteoporosis Rats 1. Laboratory animals and grouping Twenty-four 12-week-old SPF-grade female SD rats, weighing 220-250g, were selected. The rats were acclimatized for 7 days in an environment with a temperature of 22±2℃ and humidity of 50±10%, using a 12-hour light / 12-hour dark cycle, and had free access to sterile feed and water. The rats were then randomly divided into four groups of six each.
[0015] (1) CON group, i.e. sham surgery control group: the skin on the back was cut and sutured, but the ovary was not removed; the right thigh was cut and the sciatic nerve was exposed, but not severed; an equal volume of normal saline was injected intramuscularly every other day for 10 weeks. (2) OVX group, namely the postmenopausal osteoporosis model group: anesthesia was performed by intraperitoneal injection of sodium pentobarbital 40mg / kg, bilateral back incisions were made to expose the ovaries, the fallopian tubes were ligated and the bilateral ovaries were completely removed and sutured layer by layer; penicillin 80,000 U / animal was injected intramuscularly for 3 consecutive days after the operation to prevent infection. (3) DIO group, i.e. disuse osteoporosis model group: After intraperitoneal anesthesia, a longitudinal incision was made on the back of the right thigh, the muscles were bluntly dissected and the sciatic nerve was exposed, a nerve segment of about 5 mm was removed to avoid nerve regeneration, and the incision was sutured to make the right lower limb completely disused; 80,000 U of penicillin per animal was injected intramuscularly for 3 consecutive days after the operation to prevent infection. (4) GIO group, namely the glucocorticoid-induced osteoporosis model group: no surgical treatment, dexamethasone sodium phosphate injection 1mg / kg was injected into the thigh muscle every other day, and the body weight was measured weekly and the dosage was adjusted accordingly, for a total of 10 weeks.
[0016] 2. Sample Collection Ten weeks after modeling, rats were sacrificed by CO2 inhalation, and abdominal aortic blood, bilateral femur and fecal samples were collected for serum bone metabolism indicators, bone micro-CT, bone biomechanics, histology, 16S rDNA sequencing and non-targeted metabolomics detection.
[0017] 3. Bone Micro-CT Examination The right femur was harvested, and after removing the attached muscle, it was fixed in 4% paraformaldehyde for 24 hours. Scanning was performed using a vivaCT80 Micro-CT scanner with parameters of 70 kVp, 114 μA, and a resolution of 10.5 μm. The region of interest was selected as the cancellous bone 1 mm distal to the growth plate. The accompanying software was used to calculate bone mineral density (BMD), bone volume fraction (BV / TV), trabecular bone number (Tb.N), trabecular bone thickness (Tb.Th), trabecular bone spacing (Tb.Sp), and structural model index (SMI).
[0018] The results showed that compared with the CON group, the BMD of the OVX, DIO, and GIO groups decreased by 18.7%, 15.2%, and 20.4%, respectively (P < 0.01); BV / TV decreased by 32%, 28%, and 35%, respectively; Tb.N decreased significantly (P < 0.05); and Tb.Sp and SMI increased significantly (P < 0.05). These results indicate that all three models exhibited typical osteoporotic bone microstructural destruction, with trabecular bone degeneration from plate-like to rod-like structures.
[0019] 4. Three-point bending test Three-point bending tests were conducted using an ElectroForce 3200 mechanical testing machine with a span of 20 mm and a loading speed of 2 mm / min. Load-displacement curves were recorded. Compared with the CON group, the femoral fracture load in the OVX, DIO, and GIO groups decreased by 25.6%, 22.3%, and 28.1%, respectively (P < 0.05); the stiffness decreased by 22.1%, 19.8%, and 26.4%, respectively (P < 0.05), indicating that the bone biomechanical properties of all three models deteriorated significantly.
[0020] 5. Serum bone metabolism marker detection Abdominal aortic blood was centrifuged at 3000 rpm for 15 minutes, and the supernatant was collected. Serum levels of type I procollagen N-terminal propeptide (PINP) and type I collagen C-terminal peptide (CTX-I) were detected using an ELISA kit from Xiamen Lunchangshuo Biotechnology Co., Ltd. Results showed that PINP levels were compensatorily increased in the OVX group (P<0.05), indicating high-turnover bone loss; PINP levels were inhibited in both the DIO and GIO groups (P<0.05), indicating low-turnover or mixed-type bone loss; CTX-I levels in all three groups were significantly higher than in the CON group (P<0.01), indicating increased bone resorption in all three groups.
[0021] 6. H&E histological observation After decalcification of the distal femur, sections were prepared and stained with H&E. The morphology of trabeculae and medullary cavity was observed under a light microscope. Compared with the CON group, the number of trabeculae in all three models was significantly reduced, and the trabeculae changed from plate-like to rod-like. The number of fat vacuoles in the medullary cavity was significantly increased, which was consistent with the SMI results of Micro-CT.
[0022] 7. Conclusion of this embodiment Bone micro-CT, three-point bending test, serum bone metabolism indicators and histological results corroborate each other, indicating that the three osteoporosis models, OVX, DIO and GIO, were successfully constructed. At the same time, each model exhibits different bone turnover states, which can be used for subsequent microbiome and metabolomics comparative studies.
[0023] Example 2: 16S rDNA sequencing and gut microbiota composition analysis of fecal samples 1. DNA extraction, amplification, and sequencing Approximately 200 mg of fecal sample stored at -80℃ was collected, and total DNA was extracted using the CTAB method. DNA concentration and purity were detected using a NanoDrop 2000. The 16S rRNA gene V3-V4 region was amplified using primers 341F (5'-CCTACGGGNGGCWGCAG-3') and 805R (5'-GACTACHVGGGTATCTAATCC-3'). The PCR reaction volume was 25 μL, and the reaction program was as follows: 98℃ pre-denaturation for 30 seconds; 32 cycles, each cycle consisting of 98℃ denaturation for 10 seconds, 54℃ annealing for 30 seconds, and 72℃ extension for 45 seconds; and a final extension at 72℃ for 10 minutes. After confirmation by 2% agarose gel electrophoresis, the amplified products were purified using AMPureXT magnetic beads, quantified using Qubit, and sequenced using the NovaSeq 6000 platform.
[0024] 2. Sequencing data processing and diversity analysis After FASTQ quality control, the raw sequencing data were denoised and chimeras removed using the DADA2 algorithm in the QIIME2 software to generate an ASV feature table and representative sequences. Species annotation was performed based on the Greengenes database v13.8. Alpha diversity was evaluated using the Shannon, Simpson, and Chao1 indices, while Beta diversity PCoA analysis was performed based on weighted Unifrac distance.
[0025] The alpha diversity sparse curves all tended to flatten, indicating sufficient sequencing depth. Beta diversity PCoA analysis showed that the OVX, DIO, and GIO groups clustered into three independent clusters, all far from the CON group, indicating that different types of osteoporosis models have different gut microbiota structures. Venn plots showed a total of 553 ASVs across the four groups; the CON, OVX, DIO, and GIO groups had 2,449, 2,689, 2,683, and 1,176 unique ASVs, respectively.
[0026] 3. Screening for differentially expressed microbiota LEfSe multilevel species difference discriminant analysis and Wilcoxon rank-sum test were used to screen for significantly different bacterial genera among the groups. The main changes in bacterial communities in each model are as follows.
[0027] OVX model: At the gate level, the relative abundance of p_Firmicutes increases, log2FC=0.22, P=0.02; the relative abundance of p_Bacteroidetes decreases, log2FC=-1.02, P=0.02.
[0028] At the genus level, g_Bacteroidetes_unclassified was significantly downregulated (log2FC = -1.65); g_Firmicutes_unclassified was significantly upregulated (log2FC = 0.60); and g_Eggerthella was significantly upregulated (log2FC = 1.18). The abnormally elevated levels of Eggerthella, in relation to Th17 lymphocyte activation, increased IL-17 secretion, and osteoclastogenesis, could be used to explain the pro-inflammatory bacterial flora characteristics of this model.
[0029] In the DIO model: at the phylum level, p_Verrucomicrobia was significantly reduced (log2FC = -6.97); at the genus level, g_Akkermansia was significantly downregulated (log2FC = -7.09). Akkermansia is associated with intestinal mucus layer integrity and intestinal barrier function. Its reduction may be accompanied by increased intestinal permeability, metabolic endotoxin entry into the bloodstream, and systemic low-grade inflammation, and is also associated with inhibited bone formation.
[0030] GIO model: At the genus level, g_Rothia was significantly downregulated (log2FC = -2.37). Rothia has the ability to convert xylose, galactose, raffinose, and glucose into butyrate; butyrate, as a short-chain fatty acid, can promote regulatory T cell differentiation and inhibit osteoclast differentiation. Therefore, reduced Rothia may weaken the protective effect of intestinal short-chain fatty acids on bone and could be considered a characteristic microbiota change in glucocorticoid-induced osteoporosis.
[0031] The results above show that the three osteoporosis models have distinguishable microbial community characteristics, which provides a basis for model classification and microbial community-metabolite association analysis.
[0032] Example 3: Fecal Non-Targeted Metabolomics Detection and Differential Metabolite Identification 1. Sample pretreatment Accurately weigh 50 mg of fecal sample and place it in a 1.5 mL centrifuge tube. Add 200 μL of 50% methanol-acetonitrile extraction solution containing 2 μg / mL L-2-chlorophenylalanine as an internal standard. Vortex to mix for 1 minute and centrifuge at 4℃ and 4000×g for 20 minutes. Transfer the supernatant to a vial.
[0033] 2. Chromatographic and Mass Spectrometric Conditions Detection was performed using a Thermo Scientific UltiMate 3000 ultra-high performance liquid chromatography system coupled with a Q-Exactive high-resolution mass spectrometer. The chromatographic column was an ACQUITY ULC HSS ST3 column, 2.1 × 100 mm, 1.8 μm, with a column temperature of 40 °C. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was 0.1% formic acid acetonitrile solution.
[0034] The gradient elution program was as follows: 0–1 min, 2% B; 1–9 min, 2%–50% B; 9–12 min, 50%–98% B; 12–13.5 min, 98% B; 13.5–14 min, 98%–2% B; 14–15 min, 2% B. The flow rate was 0.3 mL / min, and the injection volume was 5 μL.
[0035] Mass spectrometry was performed using an electrospray ionization source with spray voltages of 3.5 kV (positive ion mode) and 2.8 kV (negative ion mode), a capillary temperature of 320 °C, a sheath gas flow rate of 35 arb, and an auxiliary gas flow rate of 10 arb. The scanning mode was Full MS / dd-MS. 2 The primary scan resolution is 70,000, and the secondary scan resolution is 17,500.
[0036] 3. Data processing and metabolite identification Peak extraction, retention time alignment, and blank subtraction were performed using CompoundDiscoverer 3.1 software. Metabolite identification was accomplished by matching precise molecular weights and secondary fragment spectra with the KEGG and HMDB databases, with mass deviation controlled to <5 ppm. Unsupervised PCA was used to perform unsupervised analysis on the preprocessed data, and the OPLS-DA model was used to screen for differentially expressed metabolites with a screening threshold of VIP>1, P<0.05, and |FC|≥2.
[0037] 4. Results of differential metabolite and pathway enrichment In both positive and negative ion modes, the PCA score plots showed significant separation between the OVX, DIO, and GIO groups and the CON group, indicating that the fecal metabolic profiles of all three models underwent significant remodeling. In positive ion mode, 4,942, 4,579, and 5,041 differentially expressed metabolites were identified in the OVX, DIO, and GIO groups, respectively; while in negative ion mode, 2,190, 2,110, and 2,398 differentially expressed metabolites were identified, respectively.
[0038] KEGG pathway enrichment analysis of differentially metabolites yielded the following main results: (1) Lipid metabolism pathways: Steroid hormone biosynthesis map00140, secondary bile acid biosynthesis map00121, and primary bile acid biosynthesis map00120 were significantly enriched in all three models, with P values ranging from 3.60 × 10⁻⁶. -11 Up to 5.61×10 -7 This indicates that lipid metabolism disorder is a common metabolic feature of the three models; (2) Amino acid metabolic pathways: Histidine metabolism map00340 was significantly enriched in all three models, with P values ranging from 5.61 × 10⁻⁶. -7Up to 8.81×10 -8 Tryptophan metabolism map00380 was significantly enriched in all three models, with P values ranging from 1.09 × 10⁻⁶. -9 Up to 7.45×10 -7 Arginine and proline metabolism map00330 was significantly enriched in the DIO and GIO models, with P values of 1.27 × 10⁻⁶. -7 and 2.56×10 -9 .
[0039] 5. Common and Differential Metabolites The KEGGID of 4-(β-acetamidoethyl)imidazole, an intermediate in the histidine metabolic pathway, is C05135. This metabolite was consistently and significantly upregulated in the OVX, DIO, and GIO models (VIP>1, P<0.05). In positive ion mode, its mass-to-charge ratio (m / z) was 153.08976, and its retention time was 0.746 min. This metabolite is a key product of histidine degradation via decarboxylation and acetylation; its sustained elevation reflects a common histidine metabolic pathway disorder in all three models.
[0040] In addition to the common differential metabolites mentioned above, metabolites in the tryptophan metabolic pathway, such as kynurenic acid, indole-3-acetic acid, and 5-hydroxyindole-3-acetic acid, also showed significant changes in all three models; secondary bile acids in the lipid metabolic pathway, such as deoxycholic acid, chenodeoxycholic acid, and lithocholic acid, were all significantly upregulated in all three models. These results collectively indicate that lipid metabolism and amino acid metabolism are shared metabolic changes in all three models.
[0041] Example 4: Construction and Joint Evaluation of Microbial Community-Metabolite Association Network 1. Data integration and standardization The relative abundance data of differentially abundant bacterial genera related to bone metabolism, screened by LEfSe and Wilcoxon rank-sum tests, were integrated with the normalized peak area data of differentially abundant metabolites related to lipid and amino acid metabolism, screened by OPLS-DA. Before the association analysis, all variables were Z-score standardized to ensure that the bacterial community data and metabolite data were on the same scale.
[0042] 2. Spearman correlation analysis The Spearman rank correlation coefficient was used to evaluate the association strength between differentially expressed bacterial genera and differentially expressed metabolites. This method is a non-parametric test, does not require the data to follow a normal distribution, and is applicable to microbiome and metabolome data. Specifically, the Pearson correlation coefficient was calculated after sorting the two sets of data for each genus-metabolite pair. The cor.test function in R language v4.0.3 was used for calculation, with |r|>0.6 and P<0.05 as the screening criteria for significant correlation.
[0043] 3. Correlation heatmap and co-expression network The correlation heatmap was plotted using the pheatmap package v1.0.12 in R. The X-axis represents differentially expressed metabolites, and the Y-axis represents differentially expressed bacterial genera. The cell color indicates the magnitude of the correlation coefficient r, with red indicating a positive correlation and blue indicating a negative correlation. P < 0.05 is marked with "*", and P < 0.01 is marked with "**".
[0044] Differentially expressed bacterial species and metabolites significantly associated with 4-(β-acetaminoethyl)imidazole were introduced into Cytoscape v3.8.0 to construct a co-expression network. The network construction rules are as follows: (1) Nodes represent fungal genera or metabolites; (2) The connection indicates that the Spearman correlation reaches |r|>0.6 and P<0.05; (3) The thickness of the connecting edge indicates the magnitude of the absolute value of the correlation coefficient; (4) Node size represents the node connectivity, that is, the number of other nodes that are significantly related to the node.
[0045] Use the NetworkAnalyzer tool built into Cytoscape to calculate network topology parameters.
[0046] 4. Key Relationship Results Correlation heatmaps were plotted using differentially expressed bacterial genera specific to each model as rows and differentially expressed metabolites in the top 10 lipid and amino acid metabolic pathways identified by KEGG enrichment analysis as columns. The model-specific bacterial genera used were: g_Bacteroidetes_unclassified, g_Firmicutes_unclassified, and g_Eggerthella in the OVX model; g_Akkermansia in the DIO model; and g_Rothia in the GIO model.
[0047] 4-(β-acetaminoethyl)imidazole was significantly correlated with all five key bacterial genera mentioned above, as detailed below: (1) The correlation coefficient with g_Bacteroidetes_unclassified was r=-0.783, P=0.004, indicating a negative correlation; (2) The correlation coefficient with g_Akkermansia was r=-0.832, P=0.001, indicating a negative correlation; (3) The correlation coefficient with g_Rothia was r=-0.636, P=0.030, indicating a negative correlation; (4) The correlation coefficient with g_Firmicutes_unclassified was r=0.727, P=0.001, indicating a positive correlation; (5) The correlation coefficient with g_Eggerthella was r=0.748, P=0.007, indicating a positive correlation.
[0048] Except for g_Rothia, whose correlation coefficient was 0.636, the absolute values of the correlation coefficients between the other key bacterial genera and this metabolite were all between 0.727 and 0.832. Other differentially metabolites in the histidine metabolic pathway, such as hercynine, 1-histidinol, and thiourocanicacid, also showed varying degrees of correlation with the above bacterial genera, but their correlation coefficients were generally below 0.6 or their p-values were greater than 0.05, failing to meet the established significance criteria. 4-(β-acetaminoethyl)imidazole was the only metabolite in the analysis that met the criteria of |r|>0.6 and p<0.05 with all five key bacterial genera.
[0049] 5. Network topology characteristics The co-expression network constructed with 4-(β-acetaminoethyl)imidazole as the central node has the following characteristics: the metabolite is located at the core of the network with a node degree of 5, which is the node with the highest connectivity in the network; five key bacterial genera are connected to the central node in a radial pattern, forming a star-shaped topology; network edges represent the direct association between bacterial genera and metabolites, and do not include indirect connections between bacterial genus nodes; g_Bacteroidetes_unclassified, g_Akkermansia, and g_Rothia, which are downregulated in the model, are negatively correlated with the metabolite, while g_Firmicutes_unclassified and g_Eggerthella, which are upregulated in the model, are positively correlated with the metabolite, and the direction of change in bacterial genera corresponds to the upregulation trend of the metabolite.
[0050] 6. Joint evaluation method According to the above embodiments, animal models can be jointly evaluated at the following levels: First, the osteoporosis phenotype is confirmed based on BMD, BV / TV, Tb.N, Tb.Th, Tb.Sp, SMI, femoral fracture load, stiffness, PINP, CTX-I, and H&E histological results; second, the microecological characteristics of different models are identified based on the key differentially expressed bacterial genera of the OVX, DIO, and GIO models; third, whether 4-(β-acetaminoethyl)imidazole is upregulated in all three models is detected, and whether it satisfies |r|>0.6 and P<0.05 with key bacterial genera is analyzed; finally, a comprehensive evaluation is made by combining bone phenotype, model-specific flora, and common metabolic markers. This evaluation method retains the classic bone phenotype as basic evidence of model success while adding fecal flora and metabolites as non-invasive joint indicators.
[0051] Summary of Implementation Examples The above results indicate that the OVX, DIO, and GIO models all exhibit osteoporotic changes in bone mass, bone microstructure, bone biomechanics, and bone turnover, and possess different gut microbiota characteristics. Simultaneously, the three models share common changes in lipid and amino acid metabolism; 4-(β-acetaminoethyl)imidazole is significantly upregulated in all three models and reaches the set significant correlation criteria with five key differentially expressed bacterial genera. Therefore, this invention enables the joint construction and evaluation of different types of osteoporosis animal models through a technical chain of "bone phenotype confirmation—model-specific microbiota identification—common metabolic marker evaluation—microbiota and metabolite association verification."
[0052] The above description is merely a specific embodiment of the present invention, intended to enable those skilled in the art to understand and implement the present invention. Those skilled in the art can make corresponding adjustments or substitutions to the specific embodiments without departing from the basic technical concept of the present invention; all such adjustments or substitutions should fall within the scope defined by the claims of the present invention.
Claims
1. A method for the joint construction and evaluation of microbial-metabolomes in different types of osteoporosis animal models, characterized in that, Includes the following steps: S1. Three animal models of postmenopausal osteoporosis, disuse osteoporosis, and glucocorticoid-induced osteoporosis were established, along with a control group. S2. After the modeling process, bone tissue, blood and fecal samples were collected. The osteoporosis phenotype was verified by bone Micro-CT, three-point bending test, serum bone metabolism biochemical indicators and histological observation. S3. Perform 16S rDNA sequencing on fecal samples to obtain gut microbiota composition data; S4. Perform non-targeted metabolomics detection on fecal samples to obtain metabolite expression profiles; S5. Screen for differentially expressed bacterial genera and differentially expressed metabolites, and screen for differentially expressed metabolites that show significant changes in common in different animal models; S6. Calculate the correlation between differentially expressed bacterial genera and differentially expressed metabolites, and construct a "gut microbiota-metabolite" association network; S7. Based on the bone phenotype validation results, model-specific differential bacterial genera, common differential metabolites and their correlations, the construction effects and model types of the three animal models are jointly evaluated.
2. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The three animal models were constructed as follows: the postmenopausal osteoporosis model was constructed by bilateral oophorectomy, the disuse osteoporosis model was constructed by right sciatic nerve resection, and the glucocorticoid-induced osteoporosis model was constructed by intramuscular injection of 1 mg / kg dexamethasone every other day for 10 weeks.
3. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The bone phenotype verification included detecting bone mineral density (BMD), bone volume fraction (BV / TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular spacing (Tb.Sp), and structural model index (SMI); detecting femoral fracture load and stiffness; detecting serum PINP and CTX-I levels; and observing histological changes in distal femoral trabeculae and medullary cavity.
4. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The 16S rDNA sequencing includes amplifying the V3-V4 region of the 16S rRNA gene, performing noise reduction, chimera removal, ASV feature table generation, species annotation, and diversity analysis on the sequencing data, and screening for differentially expressed bacterial genera through LEfSe multi-level species difference discrimination analysis and Wilcoxon rank-sum test.
5. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The non-targeted metabolomics detection was performed using ultra-high performance liquid chromatography-high resolution mass spectrometry; PCA was used for unsupervised analysis; the OPLS-DA model was used to screen differential metabolites with VIP>1, P<0.05 and |FC|≥2 as thresholds; and KEGG pathway enrichment analysis was performed on the differential metabolites.
6. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The common differential metabolite includes 4-(β-acetaminoethyl)imidazole, an intermediate in the histidine metabolic pathway, with KEGGID C05135; this metabolite was significantly upregulated in three models: postmenopausal osteoporosis, disuse osteoporosis, and glucocorticoid-induced osteoporosis.
7. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The model-specific differentially expressed bacterial genera include: g_Bacteroidetes_unclassified, g_Firmicutes_unclassified, and g_Eggerthella in a postmenopausal osteoporosis model; g_Akkermansia in the disuse osteoporosis model; g_Rothia in the glucocorticoid-induced osteoporosis model.
8. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The 4-(β-acetaminoethyl)imidazole was negatively correlated with g_Bacteroidetes_unclassified, g_Akkermansia, and g_Rothia, and positively correlated with g_Firmicutes_unclassified and g_Eggerthella.
9. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The correlation was evaluated using the Spearman rank correlation coefficient, with |r|>0.6 and P<0.05 as the screening criteria for significant correlation. Based on the screening results, an association network was constructed with 4-(β-acetaminoethyl)imidazole as the central node.
10. The method for constructing and evaluating the combined microbial-metabolome of different types of osteoporosis animal models according to claim 1, characterized in that: The joint evaluation includes: using bone phenotypic indicators to confirm whether the model has developed osteoporotic changes, using the model-specific differential bacterial genera to characterize different model types, and using the co-upregulation of 4-(β-acetaminoethyl)imidazole and its correlation with key bacterial genera as joint evaluation indicators of fecal metabolomics.
