BMSCs were isolated from bone marrow that was extracted from two male Sprague-Dawley rats (weight, 8–10 g; age, 3 days), which were purchased from the Animal Experimental Center of Guangxi Medical University (Nanning, China). The rats were housed at a constant temperature (26°C) and relative humidity (60%) under a reversed 12-h dark/light cycle (lights on until 8:00 p.m.), with free access to a standard diet and water. All experiments were conducted in accordance with the standard guidelines approved by the Animal Care and Experiment Committee of Guangxi Medical University (Guangxi, China; protocol number: 2014-12-3), and the present study was approved by the ethics committee of Guangxi Medical University. Briefly, the rats were sacrificed with an injected overdose of pentobarbital, and bone marrow was collected from the bilateral femur by flushing with α-modified Eagle's medium (α-MEM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 1% penicillin/streptomycin (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China). Following density gradient centrifugation (1,100 × g at 25°C for 30 min), the isolated BMSCs were cultured in α-MEM supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin. The culture medium was changed every 2 days. Characterization of isolated BMSCs was performed as described previously (5).

Two sources of NGF were added to the cell cultures. mNGF was purchased from Peprotech, Inc. (Rocky Hill, NJ, USA), and cvNGF was extracted and purified from the venom of Chinese cobra (Naja naja atra) as described in our previous study (8). Briefly, BMSCs were cultured in chondrogenic medium including 50 µg/ml ascorbic acid (A7506; Sigma-Aldrich; USA), 100 nM dexamethasone (D1756; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), 1% insulin-transferrin-selenium solution (41400045; Gibco; Thermo Fisher Scientific, Inc.) and 0.06 µg/ml mNGF or 6 µg/ml vNGF. For each experiment, cells at passage 2 were used and the seeding density was 2×10⁴ cells/cm². Experiments were performed at 7, 14 and 21 days.

NGF toxicity on BMSCs was detected using the MTT assay (Gibco; Thermo Fisher Scientific, Inc.). BMSCs were seeded onto 96-well plates and treated with various concentrations of mNGF (0, 0.0075, 0.015, 0.03, 0.06, 0.12, 0.24, 0.48 and 0.96 µg/ml) and cvNGF (0, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24 and 32) and incubated at 37°C. After 3 days, 20 µl MTT solution (5 mg/ml) was added and incubated at 37°C for 4 h in the dark. After removing the medium, the formazan crystals in each well were dissolved with 200 µl dimethyl sulfoxide (Gibco; Thermo Fisher Scientific, Inc.). The absorbance was measured at 570 nm using a microplate reader (Thermo Fisher Scientific, Inc.).

Cell viability was determined by fluorescein diacetate (FDA; GenWay Biotech, Inc., San Diego, CA, USA) and propidium iodide (PI; Sigma-Aldrich; Merck KGaA) staining on days 7, 14 and 21. Briefly, after rinsing with PBS for three times, 0.5 µM FDA and 2 µM PI in 1 ml PBS was added to the cells and incubated in the dark for 5 min. Images were captured under a laser scanning confocal microscope (Olympus Corporation, Tokyo, Japan).

The BMSCs were seeded onto 24-well-plates and were cultured with mNGF (0.06 µg/ml) or cvNGF (6 µg/ml) for 7, 14 and 21 days. The cells were then washed three times with PBS and fixed with 95% alcohol for 30 min at room temperature. Subsequently, the cells were washed with PBS and stained with 10% hematoxylin and 2% eosin (H&E; Nanjing Jiancheng Bioengineering Institute, Nanjing, China) at room temperature for 5 min and 30 sec, respectively. Finally, cells were observed and images were captured under an inverted phase contrast microscope (Olympus Corporation).

After being cultured for 7, 14 or 21 days, cells were washed with PBS and then digested with 0.25% trypsin/ETDA. Cells were centrifuged at 200 × g for 5 min at room temperature, and digested with proteinase K (20 mg/ml; Sigma-Aldrich; Merck KGaA) for 16 h at 56°C. After staining with 10 µg/ml Hoechst 33258 dye at room temperature for 10 min (Sigma-Aldrich; Merck KGaA), the DNA content was determined using a fluorescence microplate reader (BioTek Instruments, Inc., Winooski, VT, USA) at 460 nm; calf thymus DNA (Sigma-Aldrich; Merck KGaA) was used as a standard and absorption of Hoechst 33258 dye was considered the baseline. The total production of GAG was quantified by detecting absorbance at 525 nm, following treatment with 16 mg/l 1, 9-dimeth-ylmethylene blue (DMMB; Sigma-Aldrich; Merck KGaA) at room temperature for 10 min; chondroitin sulfate (Sigma-Aldrich; Merck KGaA) was used as a standard. The GAG content in each cell was normalized to the total DNA content of all cells, which represented its biosynthetic activity under various culture conditions.

Safranin O staining was performed to assess the synthesis of GAG. After being fixed with 95% alcohol for 30 min, at room temperature the cells were stained with 0.1% safranin O (Sigma-Aldrich; Merck KGaA) for 10 min. Subsequently, the cells were washed with water and sealed with neutral gum. Eventually, cells were observed and images were captured under an inverted phase contrast microscope (Olympus Corporation).

Immunohistochemistry was performed to analyze the expression levels of collagen type I α1 chain (COL1A1) and collagen type II α1 chain (COL2A1). After 7, 14 and 21 days, cells were washed three times with PBS, fixed with 95% alcohol at room temperature for 30 min and treated with Triton X-100 (Sigma-Aldrich; Merck KGaA) at room temperature for 10 min. To eliminate endogenous peroxidase activity, cells were incubated with 3% H₂O₂ at room temperature for 15 min and blocked with goat serum (Beijing Solarbio Science & Technology Co., Ltd.) for 15 min at room temperature. After incubating with COL1A1 (1:200; Wuhan Boster Biological Technology, Ltd., Wuhan, China) and COL2A1 (BA0533; 1:200; Wuhan Boster Biological Technology Ltd.) primary antibodies overnight at 4°C, cells were incubated with the secondary antibody (G1080, 1:200; Beijing Solarbio Science & Technology Co., Ltd.) and followed by biotin-labeled horseradish peroxidase (Zhongshanjinqiao Biotechnology Inc., Wuhan, China) at room temperature for 15 min. A DAB kit (Wuhan Boster Biological Technology, Ltd.) was used to visualize antibody binding. Finally, images of the cells were captured under an inverted phase contrast microscope (Olympus Corporation).

RT-qPCR was conducted to analyze the expression levels of aggrecan (Acan), SRY-box 9 (Sox9), Col2a1, Col1a1, runt-related transcription factor 2 (RUNX2) and enolase 2 (ENO2). The sequences of primers used for RT-qPCR are presented in Table I. Total RNA was extracted using a RNA isolation kit (Wuhan Megentec Biological Technology, Ltd., Wuhan, China), according to the manufacturer's protocol. RT of RNA was performed using a RT kit (Fermentas; Thermo Fisher Scientific, Inc.) and was carried out at 25°C for 5 min, 42°C for 60 min and 72°C for 5 min. RT-qPCR was performed on a qPCR Detection system with Fast Start Universal SYBR Green Master Mix (Roche Diagnostics, Basel, Switzerland) under the following conditions: 10 min at 95°C for initial denaturation, 15 sec at 95°C and 1 min at 60°C for final extension (40 cycles). The melting curve data were collected to verify PCR specificity; each gene was analyzed in triplicate. The relative mRNA expression levels were calculated using the 2^−ΔΔCq (17) method with β-actin as a reliable internal control.

Data are presented as the means ± standard deviation from 4 repeated experiments. Data were analyzed using one-way analysis of variance followed by Dunnett's post hoc test (SPSS version 21; IBM Corp., Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

An MTT assay was performed to detect the cytotoxicity of NGF on BMSCs. Cells were treated with increasing concentrations of mNGF (0–0.96 µg/ml) or cvNGF (0–32 µg/ml). As shown in Fig. 1, mNGF at concentrations ranging between 0.03 and 0.12 µg/ml and cvNGF at 1.5–16 µg/ml significantly accelerated cell growth (P<0.05). On the basis of these results, 0.06 µg/ml mNGF (Fig. 1A) and 6 µg/ml cvNGF (Fig. 1B), which exhibited the most obvious effect on BMSC proliferation, were chosen for further investigation.

To evaluate the effects of NGF on cell viability, a FDA/PI staining kit (Fig. 2) was used. The majority of the cells in all groups were stained green, which indicated a greater amount of viable cells. More viable cells and fewer apoptotic cells (stained red) were detected in the cvNGF group compared with the other groups.

Morphological alterations in BMSCs were detected following treatment with NGF for 7, 14 and 21 days using H&E staining. As shown in Fig. 3, more cells displayed chondrocyte-like morphology in the cvNGF group at the same culture period compared with the control and mNGF groups. In addition, more cell colonies were observed in the cv-NGF group.

The proliferation of NGF-treated BMSCs was analyzed according to DNA content. The DNA content in all groups was increased in a time-dependent manner (Fig. 4A). The growth rate of BMSCs was 126 and 59% higher in the cvNGF and mNGF groups at day 21, respectively, compared with in the control group, as evidenced by the significantly higher DNA content. In particular, cvNGF promoted cell growth compared with the other two groups; these findings were consistent with the results of the cell viability assay (Fig. 2) and H&E staining (Fig. 3).

GAG deposition in BMSCs treated with NGFs for 7, 14 and 21 days was measured by DMMB assay and Safranin O staining. As shown in Fig. 4B, GAG production was gradually increased in a time-dependent manner. GAG synthesis in NGF-treated BMSCs was significantly higher compared with in the control group at the same timepoint. Compared with the control group, cvNGF promoted GAG synthesis most effectively among all groups with an increase of 60% at day 21. Safranin O staining also detected intense staining in the NGF groups compared with in the control group (Fig. 5). In addition, more abundant and homogeneously distributed GAG was detected in BMSCs in the cvNGF group.

Immunohistochemical staining was used to detect the expression of type I and type II collagen in BMSCs following treatment with NGFs in vitro (Fig. 6). Strong type II collagen-positive staining (Fig. 6B) and weaker type I collagen staining (Fig. 6A) were detected in the NGF groups after 7, 14 and 21 days of culturing, particularly in the cvNGF group.

The mRNA expression levels of Acan, Sox9, Col2a1, Col1a1, RUNX2 and ENO2 were detected by RT-qPCR. As shown in Fig. 7, cartilage-specific genes, Acan, Sox9 and Col2a1, were significantly upregulated by cvNGF compared with the control and mNGF groups. Conversely, Col1a1 was significantly downregulated in the cvNGF group. The expression levels of RUNX2, a specific gene associated with hypertrophy and osteogenic differentiation, were similar to those of Col1a1. Furthermore, ENO2, a specific gene marker for neural differentiation, was not induced by NGF, as evidenced by reduced expression compared with in the control group.