Analysis of the Potential Angiogenic Mechanisms of BuShenHuoXue Decoction against Osteonecrosis of the Femoral Head Based on Network Pharmacology and Experimental Validation

Abstract

Objective

Osteonecrosis of the femoral head (ONFH) is a common orthopedic disease with a high disability rate. The clinical effect of BuShenHuoXue decoction (BSHX) for ONFH is satisfactory. We aimed to elucidate the potential angiogenic mechanisms of BSHX in a rat femoral osteonecrosis model and bone marrow mesenchymal stem cells (BMSCs).

Methods

With in vivo experiments, we established the steroid‐induced osteonecrosis of the femoral head (SONFH) model using Sprague–Dawley (SD) rats (8‐week‐old). The rats were randomly divided into five group of 12 rats each and given the corresponding interventions: control, model (gavaged with 0.9% saline), BSHX low‐, medium‐ and high‐dose groups (0.132 3, 0.264 6, and 0.529 2 g/mL BSHX solution by gavage). After 12 weeks, haematoxylin and eosin (H&E) staining was preformed to evaluate rat osteonecrosis. the expression of angiogenic factors (CD31, VEGFA, KDR, VWF) in rat femoral head was detected by immunohistochemistry, qPCR and western blotting. In cell experiment, BMSCs were isolated and cultured in the femoral bone marrow cavity of 4‐week‐old SD rats. BMSCs were randomly divided into eight groups and intervened with different doses of BSHX‐containing serum and glucocorticoids: control group (CG); BSHX low‐, medium‐, and high‐dose groups (CG + 0.661 5, 1.323, and 2.646 g/kg BSHX gavage rat serum); dexamethasone (Dex) group; and Dex + BSHX low‐, medium‐, and high‐dose groups (Dex + 0.661 5, 1.323, and 2.646 g/kg BSHX gavaged rat serum), the effects of BSHX‐containing serum on the angiogenic capacity of BMSCs were examined by qPCR and Western blotting. A co‐culture system of rat aortic endothelial cells (RAOECs) and BMSCs was then established. Migration and angiogenesis of RAOECs were observed using angiogenesis and transwell assay. Identification of potential targets of BSHX against ONFH was obtained using network pharmacology.

Results

BSHX upregulated the expression of CD31, VEGFA, KDR, and VWF in rat femoral head samples and BMSCs (p < 0.05, vs. control group or model group). Different concentrations of BSHX‐containing serum significantly ameliorated the inhibition of CD31, VEGFA, KDR and VWF expression by high concentrations of Dex. BSHX‐containing serum‐induced BMSCs promoted the migration and angiogenesis of RAOECs, reversed to some extent the adverse effect of Dex on microangiogenesis in RAOECs, and increased the number of microangiogenic vessels. Furthermore, we identified VEGFA, COL1A1, COL3A1, and SPP1 as important targets of BSHX against ONFH.

Conclusion

BSHX upregulated the expression of angiogenic factors in the femoral head tissue of ONFH model rats and promoted the angiogenic capacity of rat RAOECs and BMSCs. This study provides an important basis for the use of BSHX for ONFH prevention and treatment.

BSHX ameliorated glucocorticoid‐induced decrease in angiogenesis in the rat femoral head. BSHX‐containing serum promoted the migratory proliferative activity of rat aortic endothelial cells (RAOECs) and the angiogenic differentiation ability of bone marrow mesenchymal stem cells (BMSCs). VEGFA, COL1A1, COL3A1, and SPP1 were critical targets of BSHX in the prevention and treatment of osteonecrosis of the femoral head (ONFH).

Introduction

Osteonecrosis of the femoral head (ONFH) is a pathological process that occurs in the femoral head and results from an interruption of the blood supply following an ischemic injury and is an orthopedic disease with a very high disability rate. ¹ , ² In China, the overall prevalence of non‐traumatic ONFH is 0.725%, with a cumulative total of about 8.12 million cases of non‐traumatic ONFH patients over 15 years of age, of which the prevalence is significantly higher in men than in women (1.02% vs. 0.51%). ¹ , ² , ³ Over 80% of non‐traumatic osteonecrosis cases are caused by corticosteroids or alcohol, decompression sickness, systemic lupus erythematosus, sickle cell anemia, and so forth, are also risk factors for ONFH that cannot be ignored. ⁴ , ⁵ To date, there is no specific drug that can effectively prevent the progression of ONFH. Total hip arthroplasty (THA) is the final choice for patients with advanced ONFH, however, this faces problems such as distant revision, making it crucial to prevent and control ONFH in the early and mid‐term stages. ⁴

In recent years, the “vascular hypothesis” has been gradually recognized in the pathogenesis of ONFH. An imbalance in the vascular homeostasis of the femoral head can cause impaired blood supply to the femoral head, which can affect the delivery of oxygen and nutrients as well as the removal of metabolites, thereby causing or aggravating osteonecrosis. ⁶ , ⁷ , ⁸ Glucocorticosteroid use and excessive alcohol consumption reduce vascular perfusion in the femoral head through mechanisms such as endothelial injury and microvascular thrombosis, and restoration of vascular perfusion is an important factor in determining the progression of ONFH. ⁴ , ⁹ Consequently, exploring therapeutic approaches to promote femoral angiogenesis would facilitate the design of more effective ONFH therapeutic strategies.

In China, traditional Chinese medicine (TCM) is an integral part of the medical system and is widely used in clinical practice due to its unique system of diagnosis and treatment and holistic concept. Numerous studies have shown that Chinese medicine monomers and compound can promote the reconstruction and regeneration of femoral head blood vessels as well as the repair of femoral head osteonecrosis through modulating the “osteogenic‐angiogenic coupling.” ¹⁰ , ¹¹ BSHX has the pharmacological effects of tonifying the kidney and strengthening the bones, activating blood circulation and removing blood stasis, and has been proved to have good efficacy in the early and middle stage of ONFH patients with “deficiency of kidney qi, stagnation of qi and blood stasis” in numerous clinical practices. ¹² BSHX consist of 10 herbs, comprising Epimedium sagittatum (Siebold & Zucc.) Maxim. (Chinese name: Yinyanghuo, YYH), Salvia miltiorrhiza Bunge (Chinese name: Danshen, DS), Eucommia ulmoides Oliv. (Chinese name: Duzhong, DZ), Conioselinum anthriscoides “Chuanxiong” (Chinese name: Chuanxiong, CX), Paeonia lactiflora Pall. (Chinese name: Baishao, BS), Achyranthes bidentata Blume (Chinese name: Niuxi, NX), Smilax glabra Roxb. (Fuling, FL), buckhorn gelatin (Lujiaojiao, LJJ), Cyperus rotundus L. (Xiangfu, XF), and Glycyrrhiza glabra L. (Gancao, GC). The English name of gelatin of buckhorn was verified using the Batman‐TCM database (http://bionet.ncpsb.org.cn/batman-tcm/), and the remaining nine herbs were fully verified using the “Medicinal Plant Name Services” key search tool (http://mpns.kew.org). Accumulating evidence has demonstrated that kidney tonifying and blood activating herbs can promote angiogenesis by mediating signaling pathways such as VEGFR2/MEK/ERK, Src/FAK and PI3K/AKT/Ras/MAPK, and play a critical role in angiogenesis and vascular repair. ¹³ , ¹⁴ , ¹⁵ Nevertheless, given the complexity of the active components in TCM compounds and the diversity of potential regulatory targets in humans, the potential angiogenic mechanism of BSHX for the treatment of ONFH remains to be systematically investigated.

The aim of this study was to elucidate the angiogenic mechanism underlying the prevention and treatment of ONFH by BSHX through promoting the angiogenic capacity of steroid‐induced osteonecrosis of the femoral head (SONFH) rat model and BMSCs. Meanwhile, the systemic network pharmacological study of BSHX could contribute to the clarification of potential angiogenesis‐related targets or pathways of BSHX, thus further confirming the potential angiogenic role of BSHX. Consequently, in this study, we hypothesized that: (i) BSHX could upregulate the expression of angiogenic factors in femoral head samples from SONFH rats; (ii) BSHX‐containing serum could ameliorate the inhibitory effect of high concentrations of Dex on the expression of angiogenic factors; and (iii) BMSCs induced by BSHX‐containing serum could promote the migratory‐proliferative activity and angiogenic‐differentiation capacity of RAOECs. The experimental flow was shown in Figure 1.

FIGURE 1.

Experimental procedure. In vivo experiments, the rat steroid‐induced osteonecrosis of the femoral head (SONFH) model was established and different interventions were given. After 12 weeks, the expression of angiogenic factors in the femoral head of rats was detected. In vitro experiments, BuShenHuoXue decoction (BSHX)‐containing serum was extracted, and rat bone marrow mesenchymal stem cells (BMSCs) were isolated and given different interventions. After 48h of intervention, the effect of BSHX‐containing serum on the angiogenic capacity of BMSCs was detected.

Materials and Methods

Animals and Drugs

Sixty healthy SPF‐grade male Sprague Dawley (SD) rats at 8 weeks of age weighing 200 ± 20 g were used for in vivo experiments; 10 male SD rats (4‐week‐old) weighing approximately 175 ± 25 g were used to extract bone marrow mesenchymal stem cells; and 40 male SD rats (8‐week‐old) weighing 275 ± 25 g were used to extract serum containing BSHX. Experimental animal license number: SYXK (Lu) 2018‐0015. All experimental operations and animal disposal procedures were approved by the Animal Research Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine (AWE‐2019‐107), and the relevant provisions of Health Guidelines for the Care and Use of Laboratory Animals ¹⁶ and the Guidelines for Reporting Animal Research: In vivo Experiments ¹⁷ were strictly implemented.

The gelatin of buckhorn and the other nine herbs used to prepare BSHX were purchased from ShangPharma Holdings Shandong Co. (Jinan City, China) and Anhui He Lin Chinese Medicine Tablet Technology Co. (Bozhou, China) respectively. The batch numbers of the 10 herbs were as follows: YYH (NO.170401), DZ (NO.170501), DS (NO.171201), CX (NO.171101), BS (NO.180201), FL (NO.170402), NX (NO.180301), XF (NO.170601), GC (NO.170502), LJJ (NO.2017030401). The above herbs were purchased and supplied by the Pharmacy Department of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, and were stored in a dry and light‐proof environment. The mass (g) ratios of the 10 herbs in BSHX are shown in Figure 2. With the in vivo experiments, the high, medium and low dose concentrations of the gavage solution for rats were 0.529 2, 0.264 6, and 0.132 3 g/mL, respectively. In order to ensure that the concentration of the BSHX drug in the blood of rats could reach the desired steady state, we prepared different concentrations of BSHX (about 0.264 6 g/mL for the high dose, 0.132 3 g/mL for the medium dose, and 0.066 15 g/mL for the low dose), and increased the number of times of gavage (twice/d), to extract stable and reliable BSHX‐containing serum at the low, medium and high concentrations. It was stored at 4°C for backup, heated in a water bath and stirred evenly for gavage.

FIGURE 2.

The mass (g) ratio of 10 herbs in BuShenHuoXue decoction (BSHX).

Detection of Angiogenic Indicators in Rat Femoral Head

Animal Grouping and Treatments

Sixty male rats (8‐weeks‐old) were randomly divided into five groups (n = 12): control, model, low‐dose, medium‐dose and high‐dose groups. The rats in the control group were injected with 0.9% saline in the gluteal muscle as a control, and the rats in the remaining groups were injected with methylprednisolone (MPS) in the gluteal muscle to make animal models of steroid‐induced ONFH. Meanwhile, the control and model groups were gavaged with 0.9% saline, while the low, medium and high dose groups were gavaged with low, medium and high dose BSHX solution. The subgroups and specific interventions were shown in Table 1.

TABLE 1.

The number of rats in different groups and the intervention program.

Group	Quantity	Intervention program
		Intramuscular injection	Gavage
Control group (CG)	12	Inject 0.9% saline into the bilateral gluteus muscles 35 mg/kg/d alternately, first 3 days of continuous weekly injections, 4 weeks in total	Drinking 0.9% saline gavage, 1 mL/100 g, once a day, for 8 weeks
Model group (MG)	12	Inject MPS into the bilateral gluteus muscles 35 mg/kg/d alternately, first 3 days of continuous weekly injections, 4 weeks in total	Drinking 0.9% saline gavage, 1 mL/100 g, once a day, for 8 weeks
Low dose group (LDG)	12	Inject MPS into the bilateral gluteus muscles 35 mg/kg/d alternately, first 3 days of continuous weekly injections, 4 weeks in total	Low dose BSHX gavage (0.132 3 g/mL), 1 mL/100 g, once a day, for 8 weeks
Middle dose group (MDG)	12	Inject MPS into the bilateral gluteus muscles 35 mg/kg/d alternately, first 3 days of continuous weekly injections, 4 weeks in total	Middle dose BSHX gavage (0.264 6 g/mL), 1 mL/100 g, once a day, for 8 weeks
High dose group (HDG)	12	Inject MPS into the bilateral gluteus muscles 35 mg/kg/d alternately, first 3 days of continuous weekly injections, 4 weeks in total	High dose BSHX gavage (0.529 2 g/mL), 1 mL/100 g, once a day, for 8 weeks

Acquisition of Femoral Bone Tissue

After the rats were anesthetized with 3% pentobarbital 45 mg/kg intraperitoneally, the hip joint was routinely exposed, the left femur was cut from the middle section and the femoral head was removed and fixed in 4% paraformaldehyde fixative, the right femoral head was separated at the femoral neck, snap frozen in liquid nitrogen and then transferred to −80°C refrigerator for freezing and storage. Mark the group and time of extraction. All animal carcasses were disposed of safely.

Hematoxylin and Eosin (H&E) Staining and Immunohistochemical Staining

Samples of 4% paraformaldehyde‐fixed femoral heads were taken, decalcified and paraffin‐embedded. Sections with a thickness of 5 μm were then cut on the coronal plane. The sections were stained with hematoxylin and eosin (H&E). Paraffin sections of femoral head samples were taken, and after paraffin rehydration and antigen repair, they were immersed in 3% hydrogen peroxide solution for incubation, phosphate buffered saline washing, blocking, dropwise addition of primary and secondary antibodies for immunoreactivity, and then dropwise addition of DAB chromogenic solution and hematoxylin re‐staining, dehydration and blocking to complete immunohistochemical staining, and semi‐quantitative analysis was performed using Image Pro Plus 6.0 software to observe and evaluate the expression intensity and distribution of angiogenesis‐related factors VEGFA, CD31, KDR, and VWF in the femoral head.

Real‐time Quantitative PCR Assay

Remove the femoral bone sample from the −80°C refrigerator, grind the bone tissue into powder using a special bone tissue grinder, and extract the total RNA from the bone tissue by referring to the instructions of the Bone Tissue RNA Rapid Extraction Kit (Centrifugal Column Type AC1301, Sparkjade, Shandong, China), and determine the concentration and purity. Subsequently, the mRNA expression levels of VEGFA, CD31, and KDR were detected by the qPCR kit (TaKaRa, Kyoto, Japan), and the experiments were repeated three times. The primer sequences were shown in Table 2.

TABLE 2.

PCR primers used in this study.

Gene	Primer name	Sequence	Product size/bp
Kdr	Kdr forward	CGCTGTGAACGCTTGCCTTAT	192
	Kdr reverse	GTGCTCGCTGTGTGTTGCTCC
Pecam1	Pecam1 forward	TCACAGACAAGCCCACCAGAG	186
	Pecam1 reverse	TCACAGAGCACCGAAGCAC
Vegfa	Vegfa forward	CCACGACAGAAGGGGAGC	163
	Vegfa reverse	CACCGCATTAGGGGCACA

Western Blotting Assay

The femoral head sample was removed from the −80°C refrigerator, and the bone tissue was ground into subdivision using a special bone tissue grinder, and then added to a 1.5 m LEP tube containing 1% PMSF in RIPA in the ratio of femoral head sample mass (mg): lysis fluid volume (μL) = 1:4, vortexed and shaken, then placed in 4°C refrigerator for 1 h, 12,000 rpm, centrifuged for 20 min, and the supernatant was aspirated, which was the total protein of each group of femoral bone samples. The protein concentration was measured according to the operating instructions of the BCA protein quantification kit, and the expression of vascular‐associated core proteins (VEGFA, CD31, KDR, and VWF) was detected by Western Blot.

Detection of Angiogenic Capacity of Rats RAOECs and Angiogenic Indicators of BMSCs

BSHX‐Containing Drug Serum Preparation

Forty male rats (8‐weeks‐old) were randomly divided into blank group and BSHX containing serum low, medium and high dose groups (n = 10). Rats in the blank control group were gavaged with 0.9% saline, and SD rats in the low, medium and high BSHX groups were gavaged with the low (0.066 15 g/mL), medium (0.132 3 g/mL) and high dose (0.264 6 g/mL) BSHX solution respectively, at a ratio of 1 mL/100 g, twice/d for 7 days. Weigh accurately daily to determine gavage dose.

Rats were injected intraperitoneally with 3% sodium pentobarbital 45 mg/kg 1 h after the completion of the last gavage, blood was taken from the abdominal aorta, labeling was completed and put into 4°C refrigerated for 2 h, centrifuged at 3000 rpm for 20 min, the upper serum was aspirated, and the serum of rats in the same group was combined and mixed, and after inactivation in a water bath at 56°C for 30 min, the blood was removed by filtration through a 0.22 μm needle‐tipped microporous filter and placed at −20°C for long‐term storage.

Dexamethasone Sodium Phosphate Preparation Solution

One hundred milligram Dex powder (Solarbio, Beijing, China) was dissolved in 1 mL dimethyl sulfoxide, and mixed well, and 19.62 μL of Dex solution was aspirated and added to 10% v/v sodium pentobarbital (FBS) low sugar dulbecco's modified eagle's medium (DMEM) culture solution to fix the volume to 50 mL to obtain a concentration of 3.924 6 × 10⁻⁴ g/L of dexamethasone sodium phosphate preparation solution. Short‐term refrigerated storage at 4°C, long‐term frozen storage at −20°C.

Extraction of Bone Marrow Mesenchymal Stem Cells

In male SD rats (4‐week‐old), 3% pentobarbital sodium 45 mg/kg was injected intraperitoneally and then executed to reveal the femoral bone marrow cavity. A syringe with 1 mL of low‐sugar DMEM culture solution with 10% v/v FBS was inserted into the bone marrow cavity of the femur to aspirate the bone marrow from the proximal femur and the femoral stem. Bone marrow suspension was centrifuged at 800 rmp for 5 min at 25°C, resuspended as cell suspension with 6 mL of low sugar DMEM culture medium containing 10% v/v FBS, inoculated in cell culture flasks, and cultured for 48 h, followed by half volume exchange, and then full volume exchange every 2–3 days. Cells were fused to 70%–90% and passed to the next generation by trypsin digestion at a ratio of 1:2 for 3 generations.

The 3rd generation rat BMSCs were taken for passaging and inoculated with 5 × 10⁵ cells in new culture flasks, cultured with 10% v/v FBS low sugar DMEM complete medium, and when the cells were fused to 60%–70%, the cells were randomly divided into 8 groups of control group (CG), dexamethasone (Dex) group (DG), low dose group (LG), middle dose group (MG), high dose group (HG), LG + Dex group (LDG), MG + Dex group (MDG), HG + Dex group (HDG), and replaced with different groups of BMSCs cell treatment culture medium for intervention culture. The cell treatment culture medium configurations for different groups of BMSCs were shown in Table 3.

TABLE 3.

Preparation of cell treatment cultures for different groups of BMSCs.

Group	Method
Control group, CG	CG serum 5 mL + low sugar DMEM 45 mL
Low dose group, LG	BSHX LG serum 5 mL + low sugar DMEM 45 mL
Middle dose group, MG	BSHX MG serum 5 mL+ low sugar DMEM 45 mL
High dose group, HG	BSHX HG serum 5 mL+ low sugar DMEM 45 mL
Dexamethasone (Dex) Group, DG	CG serum 5mL + low sugar DMEM 44.5 mL + Dex preparation 0.5 mL
LG + Dex group, LDG	BSHX LG serum 5 mL+ low sugar DMEM 44.5 mL+ Dex preparation 0.5 mL
MG + Dex group, MDG	BSHX MG serum 5 mL+ low sugar DMEM 44.5 mL+ Dex preparation 0.5 mL
HG + Dex group, HDG	BSHX HG serum 5 mL+ low sugar DMEM 44.5 mL + Dex preparation 0.5 mL

Abbreviations: BMSCs, bone marrow mesenchymal stem cells; BSHX, BuShenHuoXue decoction; CG, control group; MG, model group.

Establishment of Co‐culture System between RAOECs and BMSCs

To better simulate the environment of cells in vivo, we established a co‐culture system of rat aortic endothelial cells (RAOECs, Shanghai Bioleaf Biotech, Shanghai, China) with BMSCs. The 3rd generation SD rat BMSCs were inoculated in cell culture flasks and adherent growth. When the cells were fused to 60%–70%, the BMSCs cell treatment medium of CG, DG, LDG, MDG, and HDG groups were replaced and cultured for 48 h. BMSCs and RAOECs were resuspended as single cell suspensions with DMEM complete medium containing 10% v/v FBS high sugar, respectively. Subsequently, each group of BMSCs was added to the upper chamber of the 24‐well plate at a concentration of 1 × 10⁵ cells/cm² for 200 μL, while RAOECs (2 × 10⁴ cells/cm²) were inoculated in the lower chamber of the 24‐well plate for a total of 500 μL of single cell suspension. The upper chamber was embedded in the lower chamber lined with RAOECs and incubated for 48 h. After co‐culture, the RAOECs in each group in the lower chamber were made into single cell suspensions with high sugar DMEM complete medium without FBS and counted.

RAOECs Angiogenesis Assay

Refrigerate Matrigel matrix gel, experimental sterile gun cassettes, and 96‐well plates at 4°C overnight. The 96‐well plate was placed on ice, 60 μL Matrigel matrix gel was added to each well, incubated at 37°C, 5% CO₂, and saturated humidity for 30 min, then single cell suspension (5 × 10³ cells/well) of each group of RAOECs was placed in the 96‐well plate and incubated in the cell incubator for 4 h, observed under an inverted microscope and photographed.

RAOECs Transwell Assay

Transwell assay were performed using Transwell plates (24 wells) according to the manufacturer's instructions. Briefly, RAOECs (2 × 10⁴ cells/well) were grown in the upper chamber of Transwell plates while the lower chamber was immersed in 500 μL of high sugar DMEM complete medium containing 10% v/v FBS. After 24 h of incubation at 37°C, 5% CO₂, and saturated humidity, the culture fluid was aspirated and the cells remaining on the surface of the upper chamber were gently wiped. Five hundred microliter 4% paraformaldehyde was fixed for 30 min and then stained with hematoxylin staining solution for 20 min. After washing, the cells were re‐stained with eosin staining solution for 20 min. Observed under the microscope and photographed; multiple fields of view under the microscope were randomly selected for each group and the number of cells crossing the membrane was counted.

Real‐Time Quantitative PCR Assay

Eight groups of cells (CG, DG, LG, MG, HG, LDG, MDG, HDG) were extracted from total RNA after completing 48 h intervention culture of the corresponding BMSCs cell treatment medium. The mRNA expression levels of VEGFA, CD31, and KDR were detected according to the instructions of OMEGA RNA extraction kit, and the experiments were repeated three times.

Western Blotting Assay

After mixing the PMSF and RIPA in a volume fraction of 1:100, the proteins of BMSCs were extracted. Following centrifugation at 12000 RPM for 15 min at 4°C, the supernatant was aspirated to obtain the total protein of each group of BMSCs. Protein concentration was determined according to the operating instructions of the BCA protein quantification kit, and the expression of angiogenic‐associated core proteins (VEGFA, CD31, KDR, and VWF) were detected by Western Blot.

Data Sources and Processing

Through the Gene Expression Omnibus database (GEO, https://www.ncbi.nlm.nih.gov/geo/), we screened to obtain an ONFH dataset GSE74089 from the GPL13497 platform (Agilent‐026652 Whole Human Genome Microarray 4 × 44K v2). The dataset contained a total of eight hip cartilage samples, four from ONFH patients and four from healthy controls. Subsequently, the probe level data in the txt file were processed using the limma package in R (version: 4.1.2), using the normexp method for background correction, followed by quantile normalization and probe summarization to extract gene matrix related data.

Identification and Enrichment Analysis of Hub Genes for ONFH

Differentially expressed genes (DEGs) between ONFH and normal samples were identified by the R package “limma.” The significance criteria for DEGs were as follows: p < 0.05, |log2FC| > 2. After importing the DEGs into the STRING database ¹⁸ to construct the protein–protein interaction (PPI) network, we used the six common algorithms of the cytoHubba plugin in Cytoscape software to calculate the scores of DEGs. Subsequently, we screened genes in each algorithm that scored greater than the average of the DEGs scores of that algorithm and used the intersection of genes obtained from each algorithm as the hub genes. Using the STRING database, we constructed PPI network of hub genes. Additionally, through the ClueGO plugin in Cytoscape software, we performed biofunctional enrichment analysis of hub genes.

Screening of BSHX for Active Ingredients and Relevant Targets

To systematically elucidate the pharmacological mechanism of BSHX in ONFH, we used the Traditional Chinese Medicine Systems Pharmacology database ¹⁹ (TCMSP, http://lsp.nwu.edu.cn/tcmsp.php) to set the parameters related to absorption and distribution and metabolism and excretion (ADME) to oral bioavailability (OB) ≥ 30% and drug‐likeness (DL) ≥ 0.18, and the nine herbal active ingredients and their targets in BSHX were screened. Meanwhile, the effective active ingredients and targets of gelatin of buckhorn were collected through ETCM (The Encyclopedia of Traditional Chinese Medicine, http://www.tcmip.cn/ETCM/index.php/Home/Index/), BATMAN (http://bionet.ncpsb.org.cn/batman-tcm/) and Drugbank (https://go.drugbank.com/) databases.

Identification of BSHX Targets to Prevent and Control ONFH and Construction of a Hub Gene‐target‐active Component Network

The BSHX action target was mapped to the ONFH key gene, and the resulting gene is the action target of BSHX to prevent and treat ONFH. With Cytoscape software, we constructed the BSHX active ingredient‐target network, and subsequently, merge this network with the ONFH hub genes PPI network using the merge function to finally obtain the hub gene‐acting target‐active ingredient network. Furthermore, we further predicted the BSHX‐related targets using the STRING database, and constructed a PPI network using Cytoscape software to analyze target‐related parameters and perform topological analysis based on degrees (DC) to obtain the core targets.

Statistical Analysis

All data were presented as mean ± SD and statistically analyzed using SPSS23.0 software. Comparisons of means between multiple groups were analyzed by one‐way analysis of variance, and multiple comparisons were performed by the Bonferroni method. p < 0.05, the difference was of statistical significance.

Results

Oral Administration of BSHX Significantly Improved the Expression of Angiogenic Indicators in the Rat Femoral Head

H&E staining can clearly observe the effect of MPS on the bone loss of the femoral head of rats. The cartilage surface of the femoral head samples of the CG group was relatively smooth, with occasional clefts, rich in the number of chondrocytes and regular arrangement; the bone marrow cavity was rich in the content of hematopoietic cells, densely arranged, few fat cells and rich in the number of blood vessels. In the MG group, the cartilage surface of the femoral head was not smooth, with rich in the number of chondrocytes and regular arrangement; the marrow cavity of the femoral neck could be seen to have few missing hematopoietic cells and more blood vessels. A small amount of hematopoietic cells could be seen missing, more fat cells could be seen, and more blood vessels could be seen (Figure 3).

FIGURE 3.

Hematoxylin and Eosin (H&E) staining of femoral head sample. Brown arrows, cartilage clefts; yellow arrows, blood vessels; blue arrows, hematopoietic cell loss; green arrows, adipocytes.

To further evaluate the effect of BSHX on the angiogenic ability of rat femoral head tissue, we selected four factors involved in angiogenesis to detect their genes and proteins expression levels by immunohistochemical analysis, qPCR and Western blot analysis. Immunohistochemical staining for CD31, VEGFA, KDR, and VWF demonstrated a significant decrease in angiogenic activity in MG femoral samples, while the inhibitory angiogenic activity induced by MPS was reversed by co‐treatment with BSHX. qPCR and Western blot further verified that the expression of CD31, VEGFA, KDR and VWF, as angiogenesis‐related markers, was significantly decreased in response to MPS, however, BSHX could rescue this suppressive effect to a varying extent and upregulate the expression of genes (Figure 4).

FIGURE 4.

Immunohistochemical analysis, qPCR and Western blotting analysis of angiogenic‐related indicators in rat femoral head samples. (A) Immunohistochemical staining of CD31, VEGFA, KDR, and VWF in coronal sections of representative femoral head samples from each group; (B–E) Semi‐quantitative analysis of immunohistochemistry for each group; (F–H) qPCR analysis to detect the mRNA expression of CD31, VEGFA, and KDR; (I) Western blotting bands of angiogenic indicators; (J–M) Results of semi‐quantitative analysis of CD31, VEGFA, KDR, and VWF Western blotting, respectively. (black *p < 0.05 with CG group; red *p < 0.05 with model group [MG] group).

BSHX‐Containing Serum‐induced BMSCs Promoted the Migration and Angiogenesis of RAOECs

To determine the regulatory potential of BSHX‐containing serum‐induced BMSCs on the angiogenesis of RAOECs, we performed RAOECs angiogenesis assays. As shown in the figure, the indicators of the number of vascular branches in the CG group and the LDG, MDG and HDG groups were significantly higher than those in the DG group (p < 0.05); on the other hand, in terms of the total length of vascular branches, the CG group, MDG and HDG were significantly higher than those in the DG group (p < 0.05). The above experimental results confirmed that BMSCs treated with high concentration of Dex‐containing medium could significantly reduce the number of microangiogenesis in RAOECs through their secreted cytokines during co‐culture; whereas different doses of BSHX‐containing serum reversed the adverse effect of Dex on microangiogenesis in RAOECs to some extent and increased the number of microangiogenesis (Figure 5A–C).

FIGURE 5.

Analysis of the effect of co‐culture with bone marrow mesenchymal stem cells (BMSCs) on the angiogenic capacity and migration ability of rat aortic endothelial cells (RAOECs). (A) Results of angiogenic ability of RAOECs in each group after co‐culture with BMSCs under different interventions; (B) Number of branches generated by each group of RAOECs; (C) Total length of branches generated by each group of RAOECs; (D) Results of Transwell assay under different interventions; (E) Results of the number of migrated cells in each group of RAOECs. black *p < 0.05 with control group (CG) group and red, *p < 0.05 with DG group. (Scale bar: the first row in A is 200 μm, the second row is 100 μm; the first row in D is 200 μm, the second row is 100 μm, the third row is 50 μm).

To further investigate the changes in angiogenic capacity and cellular activity under BSHX‐containing serum and Dex intervention in culture, the Transwell system was used to observe its intervention on the migration of RAOECs. The results of Transwell assay revealed that the migratory ability of RAOECs in the DG group was significantly lower than that in the CG group, and the migrated cell number in the MDG group was significantly more than that in the DG group (p < 0.05); whereas the migrated cell number in the LDG and HDG groups was not significantly different than that in the DG group (p > 0.05). The above results indicated that BMSCs treated with medium containing high concentration of Dex still significantly reduced the migratory motility of RAOECs when co‐cultured; while BMSCs induced by medium dose of BSHX‐containing serum significantly promoted the migratory motility of RAOECs (Figure 5D,E).

BSHX‐containing Serum Significantly Improved the Angiogenic Indicators of BMSCs

To further clarify the interventional effect of BSHX‐containing serum on the angiogenesis of BMSCs, we verified the effect of BSHX on the expression levels of angiogenic indicators at the protein and mRNA levels by qPCR as well as Western blot assays. As shown in Figure 6, the expression levels of CD31, VEGFA, KDR and VWF were significantly inhibited by high concentrations of Dex, while the combination treatment with different concentrations of BSHX‐containing serum significantly ameliorated this adverse effect. Furthermore, the expression of angiogenic factors was also differentially upregulated in the LG, MG, and HG groups compared with the CG group, confirming that BSHX‐containing serum significantly enhanced the angiogenic ability of BMSCs.

FIGURE 6.

Expression of mRNA and protein of angiogenesis‐related indicators in rat bone marrow mesenchymal stem cells (BMSCs) in each group. (A–C) qPCR analysis to detect the mRNA expression of CD31, VEGFA and KDR; (D) Western blotting bands of angiogenic factors; (E–H) Western blotting semi‐quantitative analysis results of CD31, VEGFA, KDR, and VWF in each group. black, *p < 0.05 with control group (CG) group and red, *p < 0.05 with DG group.

Identification and Enrichment Analysis of ONFH Hub Genes

ONFH differential expression analysis identified 515 differential genes: 207 up‐regulated genes and 308 down‐regulated genes in cartilage tissue of ONFH patients as compared to healthy controls (Figure 7A,B). Subsequently, using the six common algorithms in the cytoHubba plugin, we screened 23 ONFH hub genes (Table 4). Through the STRING database, we constructed the PPI network of hub genes, which consists of 23 nodes and 164 edges. The nodes with the highest DC values were COL1A1 and COL1A2 (DC = 22) (Figure 8A,B). Enrichment analysis revealed that key genes were mainly enriched in biological processes such as collagen binding, catabolic processes of aminoglycan and cartilage development involved in morphogenesis of cartilage endosteum (Figure 8C).

FIGURE 7.

Osteonecrosis of the femoral head (ONFH) genes differential expression analysis. (A) Volcano map; (B) Heat map.

TABLE 4.

Results of differential expression analysis of 23 hub genes.

Gene symbol	Full name	Log2FC	p‐value
COL1A1	Collagen type I alpha 1 chain	5.314261989	7.68676E‐08
COL1A2	Collagen type I alpha 2 chain	2.606711488	5.00726E‐07
COL3A1	Collagen type III alpha 1 chain	3.6496041	8.50667E‐09
POSTN	Periostin	4.132726152	4.60196E‐11
LOX	Lysyl oxidase	2.599793808	7.80781E‐08
COL5A1	Collagen type V alpha 1 chain	2.326036254	3.3984E‐10
COL5A2	Collagen type V alpha 2 chain	2.99300997	4.24664E‐09
COL6A1	Collagen type VI alpha 1 chain	3.237473003	8.22832E‐07
SERPINH1	Serpin family H member 1	2.024753115	8.24018E‐08
LUM	Lumican	2.282302245	7.50079E‐06
LOXL2	Lysyl oxidase like 2	4.4292764	1.21625E‐08
SPP1	Secreted phosphoprotein 1	3.716699359	2.91431E‐08
FGF2	Fibroblast growth factor 2	2.600311089	7.58853E‐06
VCAN	Versican	2.26450908	6.26933E‐06
MMP13	Matrix metallopeptidase 13	3.597764712	8.1732E‐08
COL15A1	Collagen type XV alpha 1 chain	4.40047675	1.15937E‐07
PLOD2	Procollagen‐Lysine, 2‐Oxoglutarate 5‐Dioxygenase 2	2.899817319	5.15868E‐08
FSTL1	Follistatin like 1	2.494759025	6.12235E‐09
S100A4	S100 calcium binding protein A4	3.099484593	9.38273E‐09
VEGFA	Vascular endothelial growth factor A	2.117321447	2.28668E‐09
LTBP1	Latent transforming growth factor beta binding protein 1	2.411055657	5.66087E‐08
THY1	Thy‐1 cell surface antigen	3.394298371	2.5892E‐09
TGFBI	Transforming growth factor beta induced	7.323263575	7.78206E‐09

FIGURE 8.

Identification and enrichment analysis of osteonecrosis of the femoral head (ONFH) hub genes. (A) Venn diagram showing 23 ONFH hub genes obtained by screening with 6 cytoHubba algorithms; (B) Hub genes protein–protein interaction (PPI) network; (C) Hub genes enrichment analysis.

Candidate Active Ingredients and Potential Drug Targets of BSHX

For the 10 herbs contained in BSHX, a total of 222 active ingredients were identified. Subsequently, we further explored to obtain the potential drug targets of 335 candidate active ingredients. The candidate ingredients had many overlapping targets, suggesting that these ingredients might play critical synergistic roles.

Identification of BSHX Action Targets and Construction of ONFH Hub Genes‐BSHX Action Targets‐active Components Network

The Venn diagram indicated that a total of four key targets of action of BSHX to prevent and treat ONFH were obtained, including: VEGFA, SPP1, COL1A1, and COL3A1 (Figure 9A). Using Cytoscape software, we constructed an ONFH hub genes‐BSHX action targets‐active ingredients network to visualize the interaction between the drug action targets and ONFH hub genes (Figure 9B). Using the STRING database to mine 70 proteins interacting with the above four action targets, Cytoscpe was used to construct a PPI network consisting of 74 proteins and 936 interrelationships (Figure 9C). The top 10 genes of degrees were the core genes of this network, including: VEGFA, FN1, ITGB1, PTK2, EGFR, CTNNB1, KDR, HIF1A, CD44, and PECAM1. (Table 5).

FIGURE 9.

Identification and analysis of pivotal targets of action of BuShenHuoXue decoction (BSHX). (A) Screening of 4 pivotal action targets of BSHX for the prevention and treatment of osteonecrosis of the femoral head (ONFH); (B) ONFH hub genes‐acting targets‐active ingredients network. The circle was the hub genes of ONFH, the square was the action targets of BSHX to prevent and treat ONFH, and the V was the active ingredients of BSHX; (C) protein–protein interaction (PPI) network of predicted relevant targets of action. Red nodes were BSHX pivotal action targets, green nodes were STRING database predicted targets.

TABLE 5.

Degree of 10 core genes.

Gene symbol	Full name	Degree
VEGFA	Vascular endothelial growth factor A	58
FN1	Fibronectin 1	53
ITGB1	Integrin subunit beta 1	45
PTK2	Protein tyrosine kinase 2	44
EGFR	Epidermal growth factor receptor	44
CTNNB1	Catenin beta 1	43
KDR	Kinase insert domain receptor	43
HIF1A	Hypoxia inducible factor 1 subunit alpha	42
CD44	CD44 molecule (Indian blood group)	40
PECAM1	Platelet and endothelial cell adhesion molecule 1	39

Discussion

Based on normal rats, SONFH rat model and rat BMSCs, we performed a series of in vivo and in vitro experiments, which indicated that BSHX upregulated the expression of angiogenic factors in the femoral head tissues of SONFH rats, and BSHX‐containing serum reversed the inhibitory effect of high concentrations of Dex on the expression of angiogenic factors; furthermore, BSHX‐containing serum‐induced BMSCs promoted the angiogenic capacity of co‐cultured RAOECs. The above results confirmed that BSHX could prevent and treat ONFH by promoting the repair of vascular injury.

BSHX Ameliorated Femoral Head Angiogenesis in ONFH Rats

Although the exact pathogenesis of ONFH remains unknown, most of the risk factors are directly or indirectly associated with damage to the vascular supply of the femoral head. Based on the theory of intravascular coagulation, the pathogenic factors leading to endothelial cell damage or apoptosis and thus microcirculatory thrombosis are important mechanisms for the development of ONFH. ²⁰ Normal angiogenesis is central to tissue repair, and VEGFA, KDR, CD31, and VWF are important regulators of angiogenesis that help maintain and restore vascular integrity and are essential for vascular endothelial cell development, migration, proliferation and new vessel formation, and are often used as indicators of angiogenic capacity. ²¹ , ²² , ²³ , ²⁴ , ²⁵ VEGF can promote the coupling of angiogenesis and osteogenesis during bone repair, and the femoral head upregulates VEGF expression levels by stabilizing HIF‐1α expression in an ischemic–hypoxic environment. Then VEGF promotes endothelial cell migration by protease hydrolysis of the basement membrane, and endothelial cells migrate down the concentration gradient of VEGF and other growth factors to the area of necrotic tissue, initiating revascularization of the necrotic area of the femoral head, which in turn promotes neoangiogenesis. ²¹ , ²⁶ , ²⁷ , ²⁸ KDR, also known as Vascular Endothelial Growth Factor Receptor 2 (VEGFR2), can mediate endothelial cell migration through binding to the bridging protein Shb and activation of PI3K, which is essential for the revascularization of necrotic areas. ²¹ , ²⁹ In our study, the relatively lower expression level of angiogenic indicators in the model group is consistent with the hypothesis that glucocorticoids cause ONFH by triggering vascular damage in the femoral head, that is, glucocorticoids interfere with the differentiation direction of BMSCs by triggering apoptosis of osteoblasts and endothelial cells and increasing the activity of osteoclasts, thus reducing the blood supply in the femoral head, causing ischemia and hypoxia and decreasing the bone repair capacity, and ultimately leading to the development of ONFH. ³⁰ , ³¹ , ³² In vivo experiments in rats demonstrated that BSHX significantly upregulated the expression of VEGFA, KDR, CD31 and VWF in ONFH rats, which improved the imbalance of glucocorticoid‐induced reduction of angiogenic capacity in the femoral head of rats to some extent. Consequently, BSHX may prevent and treat ONFH by improving vascular injury and promoting revascularization of necrotic areas.

BSHX‐containing Serum Enhanced Angiogenesis in Rat BMSCs

The drug‐containing serum of traditional Chinese medicine has good controllability and reproducibility, which is closer to the real process of drug action in vivo while reducing external interfering factors, and has been widely used in the basic research on the effective substances of traditional Chinese medicine. ³³ As an important source of osteoblasts, vascular endothelial cells and chondrogenic cells, BMSCs play a vital role in the process of bone formation and bone repair. In vitro experiments revealed that BMSCs treated with high concentrations of Dex‐containing medium effectively inhibited the angiogenic ability of co‐cultured RAOECs, while BSHX‐containing serum significantly improved the migration and angiogenic ability of RAOECs, as well as the angiogenic indicators of BMSCs intervened by BSHX‐containing serum. It could be seen that the effect of BSHX on the prevention and treatment of ONFH might be based on the protection and promotion of cell proliferation, differentiation, and activity of RAOECs, which in turn promoted the expression levels of angiogenesis‐related indicators CD31, VEGFA, KDR, and VWF at the mRNA and protein levels, and finally achieved the purpose of alleviating and even treating ONFH.

BSHX Active Ingredients Had the Potential to Promote Angiogenesis and Improve Bone Health

In recent years, network pharmacology has provided new approaches to research in the field of TCM, especially in the development of TCM compounding. In order to further explore the mechanism of BSHX against ONFH and to excavate the relevant control targets and pathways, we performed a network pharmacology analysis. The analysis identified VEGFA as a differentially expressed hubgene of ONFH, as well as a therapeutic target for the joint action of quercetin, baicalein, β‐carotene and luteolin, the active components of BSHX. Hence, VEGFA may play a critical role in the underlying angiogenic mechanism of BSHX against ONFH. Additionally, the screened active ingredients have been documented to have the potential to promote angiogenesis and improve bone health, for example, quercetin from YYH, DZ, NX, and GC inhibited receptor activator of nuclear factor kappa‐Β ligand (RANKL)‐mediated apoptosis of osteoblasts and promoted osteogenesis, apoptosis of adipocytes, apoptosis of osteoblasts, and angiogenic factor expression in rat BMSCs; ³⁴ , ³⁵ Baicalein, an active constituent of NX, improves subchondral bone remodeling and has the potential to inhibit osteoclast differentiation and induce apoptosis of mature osteoclasts ³⁶ , ³⁷ ; β‐carotene in DZ inhibits the formation of osteoclasts and resorption pits via the NF‐ĸB pathway. ³⁸ A study reported that serum levels of β‐carotene were significantly lower in ONFH patients than in healthy controls, suggesting that β‐carotene may be involved in the pathogenesis of ONFH. ³⁹ Luteolin present in YYH, DS, and XF can increase collagen synthesis, alkaline phosphatase (ALP) activity, and osteocalcin secretion. ⁴⁰ Emerging evidence suggests that luteolin significantly promoted the angiogenic capacity of rat umbilical vein endothelial cells, as well as the osteogenic differentiation of BMSCs and the formation of calcified nodules. ⁴¹ Combined with previous studies, the promotion of angiogenic factor expression, inhibition of osteoblast apoptosis and promotion of osteoclast apoptosis may be the main mechanisms of BSHX against ONFH, providing evidence for the BSHX prevention and treatment of ONFH.

Strengths and Limitations

In this study, the interventional effect of BSHX on ONFH angiogenesis was comprehensively investigated by in vivo assay, in vitro assay and network pharmacology. The application of Chinese herbal medicine‐containing serum also avoided the various drawbacks of using compound crude extracts to directly undergo in vitro experiments, providing new ideas and evidence for the treatment of related diseases with Chinese medicine. The present study has some limitations, first, this study was conducted with the BSHX compound as a whole, but due to the objective conditions, the active ingredients of the drugs in the formula and their interactions still need to be further revealed. The network pharmacology analysis provided new targets of action and ideas for future in‐depth studies. Second, although network pharmacology can reflect the drug‐disease interrelationship at multiple levels, it cannot yet fully reflect all the real cellular network characteristics in living organisms, and the critical links for the mechanism of BSHX prevention and treatment remain to be explored by in‐depth studies. Additionally, the role of important monomeric components and specific targets in BSHX in the pathological process has not been further investigated, for example, experiments are needed to study the effect of quercetin on angiogenesis in ONFH rats. Finally, the results of animal and cellular experiments remain to be observed, and we will conduct further high‐quality, multicenter clinical trials.

Conclusion

Our study demonstrated that BSHX improved the imbalance of glucocorticoid‐induced reduction of angiogenic capacity in the rat femoral head by regulating the expression of angiogenic indicators VEGFA, CD31, KDR, and VWF. The BSHX‐containing serum enhanced the migratory proliferative activity of RAOECs and the angiogenic differentiation ability of BMSCs, which in turn served the purpose of delaying and treating ONFH. This study confirmed that VEGFA, COL1A1, COL3A1, and SPP1 were critical targets of BSHX in the prevention and treatment of ONFH, and that BSHX might reduce bone loss and promote vascular regeneration in patients with ONFH by modulating these targets to improve the local microenvironment in a multi‐component, multi‐target and multi‐linked manner to delay or reverse the process of ONFH. In conclusion, our study employed an integrated strategy combining in vivo and in vitro experiments with network pharmacology, thus providing a scientific basis for elucidating the mechanism of BSHX against ONFH.

Author Contributions

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, Di Luo and Hao Liu; methodology, Di Luo and Hao Liu; investigation, Di Luo, Cheng‐bo Hu, Chou Ding, and De‐zhi Yan; writing—original draft, Hao Liu and Di Luo; writing—review and editing, Di Luo, Wei Yan, Ji‐biao Wu, and Jin‐song Li; visualization, Hao Liu; supervision, Di Luo, Xue‐zhen Liang, Ji‐biao Wu, and Wei Yan.

Funding Information

This work was supported by the National Natural Science Foundation of China (NSFC) (grant number 82074453); the National Natural Science Foundation of China Youth Science Foundation Project (grant number 82205154); the Shandong Provincial Natural Science Foundation Joint Special Fund Project (grant number ZR2021LZY002); the Shandong Provincial Natural Science Foundation (grant number ZR2020KH012); the Shandong Province Natural Science Foundation Youth Branch (grant number ZR2021QH004); the Shandong Provincial Medical and Health Science and Technology Development Plan (grant number 202104070600); the Shandong Provincial Health and Wellness Commission project (Qilu Shao's special technique of Chinese medicine for bone and joint diseases 2021.01) and the Shandong Provincial TCM Science and Technology Development Plan Project (grant number 2019‐0148).

Conflict of Interest Statement

The authors declare that this study was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.

Ethics Statement

All experimental operations and animal disposal procedures were approved by the Animal Research Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine (AWE‐2019‐107), and the relevant provisions of Health Guidelines for the Care and Use of Laboratory Animals and the Guidelines for Reporting Animal Research: In vivo Experiments were strictly implemented.

Data Availability Statement

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.