Revealing the core active pharmaceutical ingredients and targets of Jie-gu capsules for fracture treatment through network pharmacology and mendelian randomization Authors: Ying Wang, Shuang Ding, Feng Gao, Yuyan Jia, Xukai Wang Published: 2024 ABSTRACT Jie-gu capsules are widely used for the treatment of fractures in China. However, the core active pharmaceutical ingredients of Jie-gu capsules and the potential mechanisms for treating fractures remain unclear. This study aims to preliminarily elucidate the potential mechanisms of Jie-gu capsules in the treatment of fractures through network pharmacology and mendelian randomization methods. Data of fracture patients were obtained from the GEO database (GSE93215), and the active pharmaceutical ingredients and therapeutic targets of Jie-gu capsules were retrieved from the TCMSP and TCMID databases to identify the intersection genes. Subsequently, a protein-protein interaction network of the intersection genes was constructed using the STRING database. Then, GO and KEGG analyses were conducted on the intersection genes. In addition, mendelian randomization was employed to identify core targets. Finally, molecular docking techniques were used to perform molecular docking of the core active pharmaceutical ingredients and core targets for Jie-gu capsules in the treatment of fractures. In this study, a total of 65 intersection genes involved in Jie-gu capsule treatment of fractures were identified. GO and KEGG results indicated that these 65 intersection genes were primarily associated with biological processes such as response to tumor necrosis factor and are involved in signaling pathways, especially the regulation of the MAPK signaling pathway. We identified 5 core active ingredients of Jie-gu capsules (quercetin, baicalein, kaempferol, luteolin, and succinic acid). Mendelian randomization confirmed 2 core targets (ALOX12 and EGF). Molecular docking results demonstrated that the core active pharmaceutical ingredients (quercetin, baicalein, kaempferol, luteolin, and succinic acid) exhibit high affinities with the core targets (ALOX12 and EGF). This study has unveiled the core active pharmaceutical ingredients and potential action targets of the Jie-gu capsules in treating fractures, offering valuable insights for subsequent foundational research and the development of new medications. 1. Introduction Fractures are among the most common major injuries in the human body, typically caused by external trauma or skeletal system disorders. The pathogenesis of fractures involves the biomechanical properties and physiological processes of bone tissue.[1] Bone is a dynamic tissue undergoing constant remodeling and repair. In certain circumstances, external forces or abnormal conditions of the skeletal system may lead to fractures. Fractures can be categorized into various types based on their characteristics and location, such as closed fractures, open fractures, pathological fractures, and more.[2,3] Each type of fracture necessitates specific treatment approaches and considerations. Different types of fractures may exhibit varying symptoms, including pain, swelling, deformity, and functional impairment.[4] Research indicated that pro-inflammatory cytokines, such as Interleukin-1 (IL-1), Interleukin-6 (IL-6), and Tumor Necrosis Factor-α (TNF-α), played roles in initiating repair cascade reactions.[5,6] Transforming Growth Factor-β (TGF-β) superfamily and other growth factors exerted strong mitogenic and chemotactic effects on osteogenic cells and chemotactic effects on macrophages and other inflammatory cells.[7] Vascular endothelial growth factors promoted new blood vessel formation, metalloproteinases were involved in extracellular matrix remodeling,[8] and Bone Morphogenetic Proteins (BMPs) were distributed in the extracellular collagen matrix, playing a vital role in bone healing.[9] These research findings collectively demonstrated that fracture repair involved a multitude of intricate mechanisms. With the rising trend of an aging population and changes in modern lifestyles, the incidence of fractures is gradually increasing. This not only is having a significant impact on patients’ health and quality of life but also is placing a substantial burden on the socio-economic sector. Therefore, the further development of effective medications for treating fractures remains of paramount significance. Jie-gu capsules were an in-house formulation actively promoted by the Affiliated Hospital of Changchun University of Traditional Chinese Medicine. They were widely utilized in the treatment of fracture patients, offering a unique therapeutic approach and support for their recovery. This traditional Chinese medical method integrated ancient herbal traditions with modern medical knowledge, bringing new insights into the rehabilitation of fracture patients. The preparation of Jie-gu capsules adhered to a millennium-old tradition of traditional Chinese medicine, utilizing 18 types of Chinese medicinal herbs, including huang gua zi, hong hua, ru xiang, mo yao, long gu, chen pi, san qi, bu gu zhi, hu po, tu bie chong, fang hai, gu sui bu, xu duan, zi ran tong, bai ji, peng sha, er cha, and bing pian. Through a strict formulation and preparation process, a unique herbal capsule was formed, possessing the functions of promoting blood circulation, resolving stasis, facilitating fracture healing, and relieving pain. These capsules played a positive role in the treatment of fractures. Pharmacology networks, a multidisciplinary field, delve into the interactions between drugs and biological molecules. It leverages bioinformatics, systems biology, and data mining techniques to construct protein-protein interaction networks and biological pathway models, revealing drug targets, drug mechanisms of action, and potential applications in disease treatment. Pharmacology networks can expedite drug discovery, reduce drug development costs, and contribute to personalized medicine by precisely matching a patient’s genotype with drug therapy, providing crucial tools for medical research and clinical practice, driving innovation in the field of pharmaceuticals.[10,11] Mendelian randomization (MR) is a widely used causal inference method in epidemiology. It employs genetic information as instrumental variables (IVs) to assess the causal effects of exposure factors on outcomes. MR is based on 3 primary assumptions. First, there is an association between genetic variations and exposure. Second, genetic variations are not associated with confounding factors. Third, genetic variations can only influence the risk of outcomes through exposure and not through other pathways. MR finds extensive application in scientific research and clinical trials, offering the advantage of reducing bias and the impact of confounding factors.[12,13] Although Jie-gu capsules have been widely utilized in the treatment of fracture patients, their exact molecular mechanism remains unclear. Therefore, our research aims to explore the molecular-level mechanisms of their actions. By gaining a deeper understanding of the molecular mechanisms underlying Jie-gu capsules treatment for fractures, we hope to simplify their composition and enhance their therapeutic efficacy. This, in turn, will provide fracture patients with a more cost-effective and efficient treatment choice, making it easier for them to access the necessary medical care and promoting the development of fracture recovery therapies. 2. Materials and methods 2.1. Data source and preprocessing We obtained the GSE93215 dataset (Platform: GPL6244)[14] from the Gene Expression Omnibus database,[15] which consists of 19 normal samples and 9 fracture patients. Differential gene expression analysis between the normal samples and fracture samples in GSE93215 was conducted using the “limma” package in R 4.2.2, with a significance threshold of P < .05 to identify differentially expressed genes (DEGs). The active pharmaceutical ingredients contained in the 18 traditional Chinese medicines in the Jie-gu capsules, along with their corresponding targets, were obtained from the TCMSP and TCMID databases.[16,17] We applied the screening criteria of oral bioavailability (OB) ≥ 30% and drug likeness (DL) ≥ 0.18 to select effective active pharmaceutical ingredients that met both of these conditions, removing duplicates and ineffective components. Finally, the “Venn” package in R 4.2.2 was used to identify the intersection of DEGs from the GSE93215 dataset and all gene targets in the Jie-gu capsules. This resulted in a set of intersection genes, which are likely to play a crucial role in the process of Jie-gu capsule treatment for fractures. 2.2. Building protein-protein interaction network and active pharmaceutical ingredient-target network We used the STRING database[18] to import the intersection genes, specifying the species as “Homo sapiens” and keeping the other options at their default settings. This allowed us to retrieve the protein-protein interaction relationships corresponding to the intersection genes. In Cytoscape 3.7.2, we constructed the “Active Pharmaceutical Ingredient-Target” network for the Jie-gu capsules. In this network graph, “nodes” represented the active ingredients contained in the bone cement capsules and their respective targets, while “edges” represented the relationships between the active ingredients and targets. 2.3. Gene ontology and Kyoto encyclopedia of genes and genomes enrichment analysis Gene ontology (GO) is a universal method for defining the biological processes (BP), cellular components (CC), and molecular functions (MF) of many genes. Kyoto Encyclopedia of Genes and Genomes (KEGG) encompasses various biological pathways for multiple organisms. We performed GO and KEGG enrichment analysis on the intersection genes using the “clusterProfiler” package and the “KEGGgraph” package in R 4.2.2. This analysis aimed to reveal the relevant biological functions or pathways in which these intersection genes may be involved. 2.4. Mendelian randomization analysis For data collection, we utilized the Open GWAS database to gather data on intersection genes and fractures.[19] The GWAS ID for fractures is ukb-b-13979, which included 461,055 normal samples and 1878 fracture patients. To determine the single nucleotide polymorphisms (SNPs) of the intersection genes for use in MR analysis, we selected SNPs associated with the exposure and set a genome-wide significance threshold of P < 5 × 10−8 as IVs. Subsequently, we applied a linkage disequilibrium region range (kb) > 10,000 and an r2 < 0.001 to ensure the independence of the chosen SNPs. We then calculated an F-statistic for each IV to assess the strength of the association with the exposure factor, excluding IVs with an F-statistic less than 10. This process ensured that only SNPs significantly associated with the exposure factor and mutually independent were used as IVs in the MR analysis. To assess the causal effects of the exposure factor on the outcome, we employed various MR analysis methods to ensure the reliability and accuracy of the results. These methods included Inverse Variance Weighting (IVW), Weighted Median, MR-Egger, Simple mode, and Weighted mode. Each method has its unique characteristics and assumptions, providing a multifaceted estimation of causality. Furthermore, to evaluate the potential pleiotropy of the IVs, we conducted a leave-one-out analysis, sequentially excluding each SNP and then performing MR analysis on the remaining SNPs to detect potential outlier IVs. All MR analyses were conducted using the “TwoSampleMR” and “MR-PRESSO” packages in R 4.2.2. The statistically significant intersection genes obtained through MR analysis serve as marker genes for the treatment of fractures with Jie-gu capsules. Since all data used in this study originated from public databases, there was no need for approval from an ethics committee. This data source ensured the transparency and reliability of the data and allowed our research results to be shared and discussed more widely within the scientific community. 2.5. Constructing the ceRNA network for marker genes We used the miRcode database to match lncRNAs with miRNAs.[20] Subsequently, we employed 3 databases, miRTarBase, miRDB, and TargetScan, to identify miRNAs that match the marker genes.[21–23] Afterward, we integrated the interactions between miRNAs, lncRNAs, and marker genes to build a ceRNA regulatory network. Finally, we visualized the ceRNA network using Cytoscape 3.7.2. This network provides insights into the regulatory relationships and interactions between miRNAs, lncRNAs, and marker genes, shedding light on potential mechanisms and pathways involved in the context of Jie-gu capsules treatment for fractures. 2.6. Molecular docking Molecular docking is a common method in drug research and a pharmaceutical design technique for studying the interaction and recognition between receptors and ligands. It is a theoretical simulation method used to investigate molecular interactions and predict their binding modes and affinities.[24] In this study, we retrieved core proteins from the PDB database[25] and the core active pharmaceutical ingredients from the PubChem database.[26] CB-Dock2 is an online molecular docking platform that we used to determine how the core active pharmaceutical ingredients of Jie-gu capsules interact with the core proteins.[27] We evaluated the binding ability between molecular ligands (small molecular compounds) and protein targets (marker genes) using Vina scores.[28] The absolute value of the Vina score is indicative of the strength of the attraction between the ligand and the protein, with larger absolute values suggesting a more stable binding between them. 3. Results 3.1. Data acquisition and identification of intersection genes Using a significance threshold of P < .05, differential gene expression analysis of the GSE93215 dataset was performed with the “limma” package in R 4.2.2, resulting in a total of 4931 DEGs. This included 2081 upregulated DEGs and 2850 downregulated DEGs (Fig. 1A) (File S1, Supplemental Digital Content, http://links.lww.com/MD/O93). From the TCMSP and TCMID databases, using filtering criteria of OB ≥ 30% and DL ≥ 0.18, ineffective components were removed. This led to the identification of a total of 76 active pharmaceutical ingredients in Jie-gu capsules and 266 gene targets (File S2, Supplemental Digital Content, http://links.lww.com/MD/O93). Ineffective ingredients and drugs removed were described in File S3, Supplemental Digital Content, http://links.lww.com/MD/O93. In addition, detailed information on the 18 herbal medicines was provided in Table S1, Supplemental Digital Content, http://links.lww.com/MD/O94. The intersection genes were obtained by taking the intersection between DEGs in fractures and the gene targets in Jie-gu capsules, resulting in 65 intersection genes (Fig. 1B). These 65 intersection genes were suggested to play a crucial role in the process of Jie-gu capsule treatment for fractures. Intersection genes between bone-healing capsules and fractures (GSE93215 Dataset). (A) Volcano plot of differentially expressed genes in the GSE93215 dataset. Red represents up-regulated genes and green represents down-regulated genes. (B) Venn diagram illustrating the intersection genes between Bone-Healing Capsules and fractures. (C) Protein-protein interaction network of the 65 intersection genes. Each node represents an independent protein. Edges (connections) represent interactions between proteins. 3.2. Building protein-protein interaction network and active pharmaceutical ingredient-target network Using the STRING database, a protein-protein interaction network for the 65 intersection genes was generated. The network revealed that core proteins such as Epidermal Growth Factor (EGF), MMP1, MAPK8, SOD1, and others play central roles (Fig. 1C). In Cytoscape 3.7.2, a network was established to represent the relationships between the 65 intersection genes and the active pharmaceutical ingredients in Jie-gu capsules (File S4, Supplemental Digital Content, http://links.lww.com/MD/O93). This network illustrated that the compounds such as MOL000098 (quercetin), MOL002714 (baicalein), MOL000422 (kaempferol), MOL000006 (luteolin), and MOL000346 (succinic acid) acted on a significant portion of the intersection genes, making them central components among the 76 active pharmaceutical ingredients in Jie-gu capsules (Fig. 2). Interaction network between active ingredients of bone-healing capsules and gene targets. Blue nodes represent the active ingredients of Bone-Healing Capsules in relation to fractures. Red nodes represent the 65 gene targets of Bone-Healing Capsules in relation to fractures (intersection genes). 3.3. Gene ontology and Kyoto encyclopedia of genes and genomes enrichment analysis We conducted GO and KEGG enrichment analysis on 65 intersection genes. The GO analysis results revealed that these 65 genes were primarily enriched in BP including “response to tumor necrosis factor,” “cellular response to tumor necrosis factor,” “cellular response to chemical stress,” “response to lipopolysaccharide,” and “response to oxidative stress.” In terms of CC, the main enrichments were observed in “serine/threonine protein kinase complex,” “transcription regulator complex,” “protein kinase complex,” “spindle,” and “mitochondrial matrix,” among others. As for MF, the significant enrichments included “RNA polymerase II-specific DNA-binding transcription factor binding,” “DNA-binding transcription factor binding,” “fatty acid binding,” “carboxylic acid binding,” and “monocarboxylic acid binding,” among others (Fig. 3). Histogram of GO enrichment analysis results. GO = gene ontology. The KEGG analysis results demonstrated that the 65 intersection genes were predominantly enriched in pathways such as “Hepatitis B,” “Lipid and atherosclerosis,” “MAPK signaling pathway,” “TNF signaling pathway,” and “PD-L1 expression and PD-1 checkpoint pathway in cancer” (Fig. 4A). We extracted the top 20 signaling pathways along with their corresponding intersection genes for visualization using Cytoscape 3.7.2. This visualization revealed that these 20 signaling pathways collectively involved 36 intersection genes, which we subsequently employed for further analysis (Fig 4B, Table S2, Supplemental Digital Content, http://links.lww.com/MD/O95). KEGG enrichment analysis. (A) Histogram of the first 20 KEGG enrichment analysis results. (B) Network diagram of the top 20 KEGG enrichment analysis results. Orange represents genes, and green represents KEGG pathways. KEGG = Kyoto Encyclopedia of Genes and Genomes. 3.4. Mendelian randomization analysis We considered the 36 intersection genes as exposure factors and fractures (GWAS ID: ukb-b-13979) as the outcome in MR analysis. The results suggested that ALOX12 and EGF may serve as protective factors for fractures (Fig. 5A). ALOX12 used 9 SNPs strongly associated with fractures, while EGF used 4 SNPs (Table S3, Supplemental Digital Content, http://links.lww.com/MD/O96). Among the 5 methods employed, ALOX12 showed potential as a protective factor for fractures in the IVW method (OR: 0.9994, 95% CI: 0.9989–0.9999, P = .019), Weighted mode method (OR: 0.9992, 95% CI: 0.9986–0.9998, P = .035), and Weighted median method (OR: 0.9993, 95% CI: 0.9987–0.9998, P = .012). The results of the other 2 methods also indicated ALOX12’s potential protective role in fractures, although they did not reach statistical significance. For EGF, the IVW method (OR: 0.9993, 95% CI: 0.9986–1.0000, P = .045) suggested that EGF might act as a protective factor for fractures. While the results of the other 4 methods did not reach statistical significance, they also pointed to EGF’s potential as a protective factor (Fig. 5B). Figure 6A and B illustrated the influence of each SNP of ALOX12 on fractures. To ensure the robustness of the results and identify outliers, we conducted a leave-one-out analysis, which involves sequentially excluding each SNP and then repeating the MR analysis using all remaining SNPs. The results showed that the estimated causal effect did not change significantly (Fig. 6C). Figure 7A and B showed the impact of each SNP of EGF on fractures. Similarly, a leave-one-out analysis was conducted to validate the robustness of the results, and the estimated causal effects remained consistent (Fig. 7C). Therefore, based on MR analysis, it was determined that ALOX12 and EGF may be considered marker genes for the treatment of fractures with Jie-gu capsules. Mendelian randomization analysis. (A) Volcano plot of the results of the Mendelian randomization analysis. (B) Forest plot of the results of the Mendelian randomization analysis. A mendelian randomization analysis of the causal relationship between ALOX12 and fractures (GWAS ID: ukb-b-13979). (A) Scatterplots for the causal association between AOLX12 and fractures. The slope of a straight line indicates the magnitude of causality. Black dots represent genetic instruments included in the main mendelian randomization analysis. (B) Forest map visualization of the causal impact of each SNP on fractures. (C) “Leave-one-out” plots for the causal association between ALOX12 and fractures. SNPs = single nucleotide polymorphisms. A mendelian randomization analysis of the causal relationship between EGF and fractures (GWAS ID: ukb-b-13979). (A) Scatterplots for the causal association between EGF and fractures. The slope of a straight line indicates the magnitude of causality. Black dots represent genetic instruments included in the main mendelian randomization analysis. (B) Forest map visualization of the causal impact of each SNP on fractures. (C) “Leave-one-out” plots for the causal association between coffee EGF and fractures. EGF = Epidermal Growth Factor, SNPs = single nucleotide polymorphisms. 3.5. Building ceRNA networks with marker genes We have separately constructed ceRNA networks for ALOX12 and EGF. For ALOX12, there are 3 lncRNAs paired with 2 miRNAs, forming 3 lncRNA-miRNA interactions. Additionally, 2 miRNAs are matched with ALOX12, resulting in 2 miRNA-mRNA interactions (Fig. 8A). As for EGF, there are 20 lncRNAs paired with 6 miRNAs, forming 20 lncRNA-miRNA interactions. Furthermore, 6 miRNAs are matched with EGF, resulting in 6 miRNA-mRNA interactions (Fig. 8B). ceRNA network diagram. (A) The ceRNA network relationship diagram of ALOX12. (B) Diagram of the ceRNA network relationship of EGF. Red represents mRNA, green represents miRNA, and blue represents lncRNA. EGF = Epidermal Growth Factor. 3.6. Molecular docking To further explore the potential interactions between the core active pharmaceutical ingredients MOL000098 (quercetin), MOL002714 (baicalein), MOL000422 (kaempferol), MOL000006 (luteolin), and MOL000346 (succinic acid) with ALOX12 and EGF, molecular docking was employed. The protein’s three-dimensional structure for ALOX12 was obtained from the PDB database, with code 8GHB, while the three-dimensional structure for EGF was obtained from the PDB database with code 1JL9. Specific results indicate that MOL000098 (quercetin) achieved a vina score of −7.1 when docking with ALOX12, MOL002714 (baicalein) scored −8.0, MOL000422 (kaempferol) scored −6.6, MOL000006 (luteolin) scored −7.0, and MOL000346 (succinic acid) scored −4.8 (Fig. 9). In the case of docking with EGF, MOL000098 (quercetin) achieved a vina score of −6.5, MOL002714 (baicalein) scored −6.5, MOL000422 (kaempferol) scored −6.5, MOL000006 (luteolin) scored −6.6, and MOL000346 (succinic acid) scored −4.4 (Fig. 10). Finally, we add all the original figures to Table S4, Supplemental Digital Content, http://links.lww.com/MD/O97. Molecular docking results of core active ingredients of bone grafting capsules and ALOX12. (A) Molecular docking results of quercetin and ALOX12. (B) Molecular docking results of baicalein and ALOX12. (C) Molecular docking results of kaempferol and ALOX12. (D) Molecular docking results of luteolin and ALOX12. (E) Molecular docking results of Succinic Acid and ALOX12. Molecular docking results of core active ingredients of bone grafting capsules and EGF. (A) Molecular docking results of quercetin and EGF. (B) Molecular docking results of baicalein and EGF. (C) Molecular docking results of kaempferol and EGF. (D) Molecular docking results for luteolin and EGF. (E) Molecular docking results for Succinic Acid and EGF. EGF = Epidermal Growth Factor. 4. Discussion Fractures are a common form of traumatic injury, typically caused by external forces, factors such as osteoporosis, tumors, medications, and other variables.[29] In recent years, the incidence of fractures has been steadily rising, with a significant contributing factor being the increasing global trend toward an aging population. Firstly, in the elderly, a decline in bone density occurs, thereby increasing the risk of fractures.[30] Secondly, older individuals often face challenges related to decreased balance and coordination, which in turn elevate the likelihood of falls and fractures.[31] Lastly, older individuals may concurrently suffer from other health issues, such as chronic diseases or cognitive impairments, which can complicate the treatment and recovery process following a fracture when compared to younger individuals.[32] Therefore, as the global aging population trend intensifies and fracture incidence continues to rise, research and the development of new pharmaceutical treatments for fractures become particularly vital. In that study, we initially identified 65 intersection genes through differential gene expression analysis in the GSE93215 dataset (normal samples: 19, fracture patients: 9) and the therapeutic targets of Jie-gu capsules. These 65 intersection genes suggested their critical role in the process of Jie-gu capsule treatment. Subsequently, we constructed a network of drug active ingredients and their targets. The results indicate that MOL000098 (quercetin), MOL002714 (baicalein), MOL000422 (kaempferol), MOL000006 (luteolin), and MOL000346 (succinic acid) primarily target the majority of these intersection genes among the 76 active ingredients in Jie-gu capsules, thus playing a central role. Therefore, we consider these 5 core active pharmaceutical ingredients as core constituents of Jie-gu capsules. Furthermore, the GO and KEGG enrichment analysis of the 65 intersection genes showed their involvement in various biological processes, including “response to tumor necrosis factor,” “cellular response to tumor necrosis factor,” and “response to oxidative stress.” They were also associated with molecular functions such as “RNA polymerase II-specific DNA-binding transcription factor binding,” “DNA-binding transcription factor binding,” and “fatty acid binding.” These genes participated in the regulation of signaling pathways like the “MAPK signaling pathway” and “TNF signaling pathway.” In the pathology of fractures, the MAPK signaling pathway and the TNF signaling pathway play critical roles in bone repair and inflammation. The MAPK signaling pathway, through its 3 main cascades – ERK, JNK, and p38 – regulates the proliferation, differentiation, and inflammatory responses of bone cells, promoting bone regeneration and remodeling during fracture healing.[33] Meanwhile, the TNF signaling pathway, through the release of TNF-α, activates downstream inflammatory factors and induces osteoclast activity, playing a significant role in the early inflammatory phase of fracture healing and creating a favorable microenvironment for bone repair.[34] The coordinated action of these signaling pathways is essential for managing inflammation and tissue repair during the fracture healing process. In addition, we performed MR analysis using 36 intersection genes from the top 20 signaling pathways and fracture data (GWAS ID: ukb-b-13979). The results indicated that ALOX12 and EGF were protective factors for fractures. Therefore, we believed that ALOX12 and EGF were marker genes for the treatment of fractures with Jie-gu capsule. Subsequently, we constructed ceRNA networks for ALOX12 and EGF separately. The results showed that in ALOX12’s ceRNA network, ALOX12 could competitively bind with hsa-miR-15a-5p and hsa-miR-561-3p. These 2 miRNAs typically played essential roles in post-transcriptional gene regulation, and their binding could impact ALOX12’s translation or stability. Additionally, several lncRNAs, such as RP11-231G3.1, RP11-483P21.6, and RP11-34P13.7, competitively bound to the mentioned miRNAs. These lncRNAs could act as “sponges” for miRNAs, absorbing them and potentially reducing the influence of miRNAs on ALOX12 mRNA, which may have helped maintain ALOX12’s normal expression levels. In the ceRNA network of EGF, EGF competitively bound with multiple miRNAs, including miR-450b-5p, miR-539-5p, miR-570-3p, miR-576-5p, miR-198, and miR-335-3p. These miRNAs typically regulated gene expression within cells, and their binding could have influenced the stability or translation of EGF mRNA, thereby affecting EGF’s expression levels. Additionally, a series of lncRNAs, such as RP11-10J21.4, MUC19, RP11-830F9.6, CTC-459F4.1, etc., competitively bound to these miRNAs. These lncRNAs could have served as “sponges” for miRNAs, adsorbing them and potentially reducing the impact of miRNAs on EGF mRNA, which may have helped maintain EGF’s normal expression levels.[35] In the final part of this study, we conducted molecular docking between the core active pharmaceutical ingredients of the Jie-gu capsules (quercetin, baicalein, kaempferol, luteolin, and succinic acid) and the target proteins for treating fractures (ALOX12 and EGF). The results indicated that quercetin achieved a vina score of −7.1 when docking with ALOX12, suggesting that quercetin had a certain degree of affinity for ALOX12, indicating a potential interaction. Baicalein scored even lower with a vina score of −8.0, implying a stronger binding affinity between baicalein and ALOX12. Kaempferol and luteolin obtained vina scores of −6.6 and −7.0 when docking with ALOX12, respectively, also demonstrating their relatively high affinity for ALOX12. Succinic acid, on the other hand, achieved a vina score of −4.8 when docking with ALOX12, indicating a relatively weaker binding affinity. All of these active components (quercetin, baicalein, kaempferol, luteolin, and succinic acid) achieved vina scores centered around −6.5 when docking with EGF, signifying a similar level of affinity for EGF, with minimal differences. In comparison to ALOX12, succinic acid obtained a lower vina score of −4.4 when docking with EGF, possibly indicating weaker binding affinity between succinic acid and EGF. In summary, based on the vina scores, baicalein demonstrated the strongest binding affinity with ALOX12, while quercetin, kaempferol, and luteolin also exhibited relatively high affinities. For EGF, these active components displayed similar affinities when interacting with it, with only succinic acid showing weaker binding affinity in comparison to the others. Quercetin is a natural flavonoid compound found in various plant foods known for its antioxidant, anti-inflammatory, and immune-boosting properties. Research suggested that quercetin could promote osteogenic processes in bone cells, increase the expression of bone formation markers, and enhance the osteogenic differentiation of stem cells.[36,37] Another study indicated that the combination therapy of quercetin and Dasatinib could clear senescent cells in the body and accelerate the healing process of fractures. This combined therapy reduced the number of senescent cells, lowered inflammation, and promoted the proliferation and differentiation of bone cells.[38] Based on the TCMSP data results and existing related research, quercetin may interact with osteopontin – a critical glycoprotein involved in bone remodeling – through various mechanisms to enhance fracture healing. These mechanisms may include reducing oxidative stress to create a favorable environment for osteopontin function, modulating inflammation to optimize osteopontin activity, and affecting the activities of osteoblasts and osteoclasts, which are crucial for bone repair.[39–41] Furthermore, quercetin may influence key cell signaling pathways related to cellular activities, such as PI3K/Akt and NF-κB, and promote angiogenesis, thereby synergizing with osteopontin’s roles in vascularization and bone healing processes.[42,43] Baicalein, another natural flavonoid compound, also possesses antioxidant and anti-inflammatory properties.[44] Research showed that baicalein’s antioxidant properties helped reduce oxidative stress following fractures, thereby promoting fracture healing. This study further suggested that baicalein could promote fracture recovery by inhibiting the activity of 5-lipoxygenase, reducing inflammation, and bone degradation.[45] Kaempferol, belonging to the flavonoid class of compounds, has antioxidant, anti-inflammatory, cardiovascular protective, anti-cancer, and antibacterial effects.[46] Studies indicated that kaempferol protected the skeleton in various models, including neonatal rats, glucocorticoid-induced and ovariectomy-induced osteoporosis models, as well as fracture models. It achieved this by inhibiting fat production, inflammation, oxidative stress, osteoclast autophagy, and osteoblast apoptosis, while simultaneously activating osteoblast autophagy.[47,48] Kaempferol’s anti-osteoporotic effects were mediated through the regulation of the estrogen receptor, bone morphogenetic protein-2 (BMP-2), nuclear factor-kappa B (NF-κB), mitogen-activated protein kinase (MAPK), and mammalian target of rapamycin (mTOR) signaling pathways. Kaempferol appeared to have a significant impact on the treatment of osteoporosis.[49] Luteolin is a natural flavonoid compound with various biological activities, including antioxidant, anti-inflammatory, anti-cancer, and antibacterial properties.[50] Preliminary research suggested that luteolin could mitigate glucocorticoid-induced osteoporosis by modulating the ERK/Lrp-5/GSK-3β signaling pathway. This modulation might aid in promoting the proliferation and differentiation of bone cells, contributing to bone tissue repair and regeneration.[51] Succinic acid is a naturally occurring organic acid found abundantly in amber, among other sources. It possesses antioxidant properties that are beneficial for protecting cells from oxidative stress.[52] Succinic acid also participates in the tricarboxylic acid cycle (TCA cycle) within the body, an important metabolic process related to energy production.[53] However, there is currently no direct evidence to suggest that succinic acid has a direct role in bone fracture repair. ALOX12 (Arachidonate 12-Lipoxygenase, 12S Type) is a lipoxygenase enzyme that plays a crucial role in various biological processes, including the regulation of platelet aggregation, cell migration, and tumor cell proliferation. Aberrant DNA methylation and genetic variations in ALOX12 are associated with various human diseases and pathological phenotypes, such as cardiovascular diseases, diabetes, neurodegenerative diseases, respiratory system diseases, cancer, infections, and more.[54] For instance, high methylation of ALOX12 is correlated with poor overall survival in acute myeloid leukemia patients.[55] ALOX12 hypermethylation is also linked to chronic kidney disease-related atherosclerosis in dialysis patients.[56] ALOX12 rs434473 is associated with negative symptoms in schizophrenia,[57] while ALOX12 rs1126667 and rs434473 are significantly related to early peripheral neuropathy during treatment in multiple myeloma patients.[58] Moreover, ALOX12 may serve as a potential therapeutic target for various human diseases, including cancer. Downregulating ALOX12 can inhibit lipid synthesis, making breast cancer cells more sensitive to chemotherapy.[59] Some research suggested that lipids could influence the phenotypes of osteoclasts and osteoblasts in pathological conditions,[60] and ALOX12 might impact bone mineral density and fracture risk.[61] Thus, ALOX12 might influence fracture repair through lipid metabolism pathways. EGF (Epidermal Growth Factor) is a member of the epidermal growth factor superfamily encoded by this gene. The encoded precursor undergoes proteolytic processing to generate the 53-amino acid epidermal growth factor peptide. This protein is an effective mitogen and plays a significant role in the growth, proliferation, and differentiation of many cell types.[62] A study noted that EGF is one of the molecules that promote osteogenic differentiation and may contribute to bone regeneration.[63] Hematoma, or the blood clot formed after a fracture, is considered the foundation of bone healing and a critical factor in the entire healing process.[64] It contains all the signals required to initiate and actively promote new bone formation. Hematoma is a rich medium, characterized by the presence of various cell types, cytokines, and chemokines. It regulates the generation of osteoclasts, induces angiogenesis and extracellular matrix production, and contains a multitude of anti-inflammatory molecules.[65,66] According to a study, EGF was found to be upregulated 35-fold in the hematoma of a fracture (P < .01), indicating that EGF may play a crucial role in the fracture hematoma.[62] Another study suggested that local tissues at the fracture site released multiple growth factors shortly after injury, and bone and cartilage cells continued to produce growth factors for repair in the later stages. In this research, serum levels of EGF were significantly higher in the pure fracture group than in the healthy control group. Furthermore, EGF mRNA and protein levels gradually decreased with the duration of injury, suggesting that EGF might be involved in the fracture healing process.[67] While this study utilized network pharmacology to integrate a vast array of molecular data, including gene expression data, protein-protein interaction data, and metabolic pathway data, to provide a comprehensive and in-depth analysis, it does have its limitations. Firstly, there was limited research on the active pharmaceutical ingredients of Jie-gu capsules (quercetin, baicalein, kaempferol, luteolin, and succinic acid) and their role in fracture repair at the time of this study. Consequently, this study did not delve deeply into the potential mechanisms through which the core active pharmaceutical ingredients of Jie-gu capsules might promote fracture repair. Secondly, this study would have benefited from extensive in vivo and in vitro experiments to validate its conclusions. Further experiments and analysis would have been needed to confirm the regulatory mechanisms between the core active pharmaceutical ingredients (quercetin, baicalein, kaempferol, luteolin, and succinic-acid) and ALOX12 and EGF. Lastly, it is essential to consider other factors, such as the toxicity and potential side effects of these compounds, to determine whether they are suitable for further drug development or for application in fracture treatment. Additionally, although this study primarily focuses on the association between SNPs and fracture risk, it must be acknowledged that fractures are influenced by various factors, such as genetic predispositions, age, and lifestyle, among others. By concentrating solely on SNPs, there may be a bias in the data outcomes and the identified genes. 5. Conclusions This study, based on network pharmacology and MR methods, conducted preliminary investigations into the core active pharmaceutical ingredients and potential mechanisms of Jie-gu capsules for the treatment of fractures. The study indicated that quercetin, baicalein, kaempferol, luteolin, and succinic-acid might have been the core active pharmaceutical ingredients of Jie-gu capsules, while ALOX12 and EGF could have been potential therapeutic targets for the treatment of fractures with these capsules. Finally, this research provided a reference for further exploring the mechanisms of Jie-gu capsules in the treatment of fractures and the development of new fracture treatment drugs. Author contributions Conceptualization: Ying Wang, Shuang Ding, Feng Gao, Yuyan Jia, Xukai Wang. Data curation: Ying Wang, Shuang Ding, Xukai Wang. Formal analysis: Ying Wang, Yuyan Jia, Xukai Wang. Funding acquisition: Xukai Wang. Investigation: Feng Gao. Methodology: Feng Gao. Resources: Xukai Wang. Software: Xukai Wang. Supervision: Shuang Ding, Xukai Wang. Visualization: Ying Wang. Writing – original draft: Ying Wang, Shuang Ding, Xukai Wang. Writing – review & editing: Ying Wang, Feng Gao, Yuyan Jia, Xukai Wang. Supplementary Material