A total of 60 male C57BL/6J mice (6–8 weeks old; weighing 30–35 g) were purchased from Shanghai SLAC Laboratory Animals Co., Ltd., (Shanghai, China). All mice were housed in a temperature-controlled facility at 23±1°C with a relative humidity of 50±5% and a 12-h light/dark cycle. All mice were identified by ear punching and were used to establish a glucocorticoid-induced (100 mg/kg) femoral head mouse model. The induction of glucocorticoid was achieved as described previously (20) via the subcutaneous injection of 100 mg/kg glucocorticoid (methylprednisolone, ModiQuest Research, Oss, The Netherlands) per mouse. The mice were divided into three groups and given a normal schedule and free access to a standard diet and water (n=20 in each group). On day 7 after model establishment (visible dead in cartilage area and partial necrosis and collapse of the cartilage surface), mice received treatment with PRP (10 mg/kg; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), dexamethasone (DEX; 10 mg/kg, Sigma-Aldrich; Merck KGaA) or the same volume of PBS by tail vein intravenous injection. Body weights of experimental mice were measured prior to each injection. Treatments were conducted seven times, once every 3 days. All mice were decapitated under 50 mg/kg of intraperitoneal injection of sodium pentobarbital (3%, Sigma-Aldrich; Merck KGaA) on day 32 for histological analysis and measurement of spleen weight loss. Clinical femoral head osteonecrosis scores were evaluated using a scale of 0–2, as previously described (0, health; 0.5, one leg disabled in action; 1.0, two legs disabled in action; 1.5, three legs disabled in action; and 2.0, four legs disabled in action) (21). The present study was performed in strict accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the Shanghai Jiaotong University (Shanghai, China). All animal procedures were reviewed and approved by the Shanghai Jiaotong University Animal Care Committee. All surgeries and euthanasia were performed using sodium pentobarbital anesthesia and efforts were made to minimize the suffering of experimental mice.

Interferon (IFN)-γ, interleukin (IL)-6, IL-17A, tumor necrosis factor (TNF)-α, IL-1β and receptor activator of nuclear factor κ-B ligand (RANKL) concentrations in the spleen of experimental mice were analyzed using ELISA. On day 32, spleens were harvested and homogenized in a commercialized buffer (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Subsequently, splenocytes were purified by centrifugation (4×1,000 × g) at 4°C for 15 min and the homogenates were quantified using a cell counter (Vi-CELL XR; Beckman Coulter, Inc., Brea, CA, USA) according to previous study (20). IFN-γ (ERIFNGALPHA), IL-6 (29-8061-65), IL-17A (EMIL17A), TNF-α (CMC3013), IL-1β (ERIL1B) and RANKL (EMTNFSF11) concentrations were measured using ELISA kits (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. In addition, glucocorticoid serum concentration level was also analyzed using an ELISA kit (EIACORT; Bio-Rad Laboratories, Inc.) according to the manufacturer's instructions.

Total RNA from murine spleens and synovial cells was extracted using an RNA isolation kit (AM1912M; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. All gene primers were synthesized and listed in Table I. The concentration of RNA was quantified by spectrophotometry at 260 nm (Smart Spec TM 3000; Bio-Rad Laboratories, Inc.). Quantified RNA was subjected to reverse transcription into cDNA using a reverse transcription kit (Qiagen Sciences, Inc., Gaithersburg, MD, USA) and quality was confirmed by electrophoresis. cDNA was diluted 1:10 with distilled water and 10 µl was used for amplification. Specific primer sets for IFN-γ, IL-6, IL-17A, VEGF, PDGF, IGF-1, TGF-β, TNF-α, IL-1β and RANKL, HGF, ICAM-1, OPN and PD-ECGF were conserved in our lab. RT-qPCR was performed using a real-time quantitative PCR kit (Roche Molecular Diagnostics, Pleasanton, CA, USA) with a SYBR Green Master Mix system according to the protocol provided with the parameter-specific kits (K0241; Roche Diagnostics, Basel, Switzerland). A total of 45 amplification cycles were performed, including 94°C for 30 sec, denaturation at 96°C for 5 sec, primer annealing at 68°C for 5 sec with touchdown to 58°C for 15 sec and applicant extension at 72°C for 10 sec. Relative gene expression levels were calculated by 2^−ΔΔCq method (22). The results were presented as the n-fold way compared with β-actin. All experiments were repeated three times.

Mice with osteonecrosis of the femoral head were sacrificed using pentobarbital anesthesia (50 mg/kg) on day 32 and the hips joints were separated, and 4-µm thick sections were fixed in 10% formalin for 12 h at 25°C. The joints were subsequently decalcified and embedded in paraffin. The hips joints from experimental mice were stained with hematoxylin and eosin. The therapeutic effects of PRP, DEX and PBS were analyzed in mice with glucocorticoid-induced osteonecrosis of the femoral head. The severity of osteonecrosis of the femoral head in the joints was scored on a scale of 0–5 (0, no joint inflammation arbitrary unit; 1, 1 joint inflammation arbitrary unit; 2, 2 joint inflammation in arbitrary unit; 3, 3 joint inflammation arbitrary unit; 4, 4 joint inflammation arbitrary unit; and 5, 5 joint inflammation arbitrary unit) as previously described (23).

All data were presented as mean ± standard deviation of triplicate results. Unpaired data were analyzed using Student's t test and comparisons of data between multiple groups were conducted using analysis of one-way variance and a Tukey Kramer test. A Kaplan-Meier curve was used to estimate the survival rate during the 120-day long-term observation to analyze the survival rate. P<0.05 was considered to indicate a statistically significant difference.

To investigate the efficacy of PRP in mice with glucocorticoid-induced osteonecrosis of the femoral head, the concentration of inflammatory cytokines in murine synovial cells and spleens were measured. It was determined that the levels of mRNA and concentrations of IL-17A, TNF-α, IL-1β and RANKL were significantly downregulated in the DEX group compared with the PBS control group in synovial cells (P<0.01; Fig. 1A) and the spleen (P<0.01; Fig. 1B) respectively. PRP treatment significantly reduced levels of mRNA and concentration of IL-17A, IL-1β, TNF-α and RANKL in the PRP group compared with the DEX group in murine synovial cells and spleens, respectively (P<0.01; Fig. 1A and B), contributing to the remission of osteonecrosis of the femoral head. Furthermore, the mRNA expression levels and concentrations of IFN-γ and IL-6 were significantly upregulated in the DEX group compared with the PBS group in murine synovial cells and spleens, respectively (P<0.05; Fig. 1C and D, respectively); moreover, PRP treatment significantly upregulated the mRNA expression levels and concentrations of IFN-γ and IL-6 in the PRP group compared with the DEX group in murine synovial cells and spleens, respectively (P<0.01; Fig. 1C and D, respectively). These results suggested that PRP treatment effectively suppressed inflammatory cytokines IL-17A, IL-1β, TNF-α and RANKL and promoted advantageous inflammatory cytokines IFN-γ and IL-6 for mice with osteonecrosis of the femoral head.

In order to evaluate the efficacy of PRP on the state of illness development, the femoral head osteonecrosis mouse model was established. Experimental mice were injected with PRP, DEX or PBS. The arthritic score of mice in each group was recorded every 3 days. PRP treatment significantly ameliorated both the clinical symptoms and hind paw swelling, thus reducing arthritic scores of PRP-treated mice compared with those of the DEX-treated mice (P<0.01; Fig. 2A). In addition, the therapeutic effects of PRP were evaluated according to the humoral glucocorticoid concentration. Serological levels of anti-glucocorticoid immunoglobulin (Ig)G prior to and following treatment were measured in experimental mice. The results demonstrated that a significant decrease in anti-glucocorticoid IgG serum levels occurred in PRP-treated mice compared with the DEX-treated group (P<0.01; Fig. 2B). Vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), insulin-like growth factor-1 (IGF-1) and transforming growth factor (TGF)-β mRNA expression levels were significantly increased in the DEX group compared with the PBS group in the synovial cells (P<0.01; Fig. 2C-F, respectively). Furthermore, PRP treatment significantly upregulated the mRNA expression levels of VEGF, PDGF, IGF-1 and TGF-β compared with the DEX group (P<0.01; Fig. 2C-F, respectively). These data suggest that PRP may induce the production of angiogenic factors, which may contribute to angiogenesis in femoral head osteonecrosis.

Body and spleen weight changes were evaluated following treatment with PRP DEX or PBS. The body weights of mice in the DEX group were significantly increased compared with the PBS group (P<0.01; Fig. 3A) and PRP-treated mice exhibited a significantly increased body weight compared with the DEX group (P<0.01; Fig. 3A). It was determined that spleen weight loss in mice from the DEX group was significantly decreased compared with that of the PBS group (P<0.01) and that the spleen weight loss of mice in the PRP group was significantly decreased compared with that of the DEX group (P<0.01; Fig. 3B). The DEX group exhibited significantly increased mRNA expression levels of hepatocyte growth factor (HGF; P<0.01), intercellular adhesion molecule-1 (ICAM-1; P<0.01), osteopontin (OPN; P<0.05) and platelet-derived endothelial cell growth factor (PD-ECGF; P<0.05) compared with the PBS group (Fig. 3C-F, respectively). Furthermore, PRP treatment significantly increased the mRNA expression levels of HGF, ICAM-1, OPN and PD-ECGF compared with the DEX group (P<0.01; Fig. 3C-F). The data suggest that PRP treatment significantly increased body weight and promoted the mRNA expression levels of angiogenesis-related factors.

All synovial specimens from mice with femoral head osteonecrosis that were treated with PRP, DEX or PBS were investigated. Data indicated that hypertrophy and leukocyte infiltration was significantly reduced following 4 weeks treatment with PRP compared with the DEX group (P<0.01; Fig. 4A and B). In addition, the therapeutic mechanism of PRP in synovial cells from mice with osteonecrosis of the femoral head was also analyzed. Histological staining indicated that PRP-treated mice with osteonecrosis of the femoral head exhibited markedly improved synovial hyperplasia, inflammatory cell influx, destruction of the cartilage and angiogenesis compared with the DEX group (Fig. 4C and D). Taken together, these preclinical data suggest that PRP may be an efficient agent for treatment of femoral head osteonecrosis.