A total of two male and two female New Zealand white rabbits (weight, 400–500 g; age, 4 weeks) were obtained from the Animal Experimentation Center of Qingdao University (Qingdao, China). Rabbits were housed individually in standard cages and maintained under standard laboratory conditions (relative humidity, 50±10%; temperature 25±1°C; 12-h light/dark cycles), with access to food twice a day and free access to water. Rabbits were sacrificed by peritoneal injection with 10 ml/kg of 10% chloral hydrate.

All experiments were conducted in accordance with the Guidance Suggestions for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of China (30). Animal procedures were approved by the Medical Ethics Committee of Yantai Yuhuangding Hospital (Yantai, China).

BMSCs were isolated from the tibial and femoral shafts of the rabbits. The ends of the femora were cut off at the epiphysis, and the marrow was flushed out using 20 ml α-minimum essential medium (α-MEM; Hyclone; GE Healthcare Life Sciences, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 100 U/ml of penicillin and 100 µg/ml of streptomycin (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) with a 20-gauge needle. The collected cells were collected into 25-cm² cell culture flasks (Nalge Nunc International, Penfield, NY, USA) containing 5 ml of the aforementioned medium. The medium was changed after 48 h to remove non-adherent cells and then renewed every day. Cultures were maintained in a humidified atmosphere of 5% CO₂ at 37°C. Following reaching 70–80% confluence (after ~1 week), the cells were harvested using 0.25% trypsin (Hyclone; GE Healthcare Life Sciences) and cell concentration was adjusted to 1×10⁶ cells/l. Following passage, the cells were plated in flasks and cultured until third-passage rBMSCs (P3) were obtained.

The DNA fragment that encoded human BMP-2 or TGF-β3 that had been cloned into the pIRES vector, was provided by the Central Laboratory at the Medical School of Qingdao University (Qingdao, China). Corresponding lentivirus packaging plasmids were produced by Shanghai GenePharma Co., Ltd. (Shanghai, China).

P3 rBMSCs were divided into four groups as follows: i) Group I, negative controls, consisting of untransfected rBMSCs or rBMSCs transfected with an empty vector (vehicle); ii) group II, rBMSCs transfected with lentivirus carrying green fluorescent protein (GFP)/BMP-2; iii) group III, rBMSCs transfected with lentivirus carrying TGF-β3; iv) group IV, rBMSCs co-transfected with BMP-2 and TGF-β3. The procedure of transfection was performed as previously described (30). Briefly, rBMSCs were seeded in 6-well culture plates and sequentially infected with lentivirus (Shanghai GenePharma Co., Ltd.) encompassing the indicated genes [multiplicity of infection (MOI) =20, 30, 40, 50, 55 and 60] or negative short hairpin RNA (Lenti-shcontrol) at 80% confluency (~500,000 cells/well) using Polybrene (8 µg/ml culture medium; Sigma-Aldrich; Merck Millipore). The efficiency of transfection was estimated by detecting the proportion of GFP-positive rBMSCs under a fluorescence microscope.

Following transfection with corresponding plasmids, after one week, total RNA was isolated from cells using TRIzol (Takara Bio, Inc., Otsu, Japan) according to the manufacturer's protocols, and subsequently digested with RNase-free DNase I. The concentration and quality of extracted RNA was evaluated by calculating the absorbance at 260 nm (A₂₆₀) and the A_260/280 ratio, respectively, using a spectrophotometer. cDNA was generated by reverse transcription using 1 µg RNA as a template, and RT-PCR was subsequently conducted. To evaluate BMP-2 and TGF-β3 expression, quantities of target genes were normalized to that of the housekeeping gene GAPDH, which served as the internal control. The sequences of forward and reverse primers (synthesized at Sangon Biotech Co., Ltd., Shanghai, China) used in the present study were as follows: Forward, 5′-CCAACCATGGATTCGTGGTG-3′ and reverse, 5′-GGTACAGCATCGAGATAGCA-3′ for BMP-2; forward, 5′-TGGCTGTTGAGAAGAGAGTCC-3′ and reverse, 5′-TGCTTCAGGGTTCAGAGTGTT-3′ for TGF-β3; and forward, 5′-GCCTGGAGAAAGCTGCTAAGTA-3′ and reverse, 5′-CGTTGTCATACCAGGAAATGAG-3′ for GAPDH. The amplification profile was 95°C for 5 min, followed by 38 cycles (36 cycles for GAPDH) of denaturation at 98°C for 10 sec, hybridization annealing at 62°C (60°C for GAPDH) for 30 sec, and extension at 72°C for 45 sec, followed by an extension cycle for 10 min at 72°C. PCR products were visualized on 1.0% (w/v) agarose gels stained with ethidium bromide. The band densities were quantified by detecting absorbance values and analyzed using Quantity One software (version 4.6; Bio-Rad Laboratories, Inc., Hercules, CA, USA) to measure mRNA levels. The signals were normalized to GAPDH expression. Experiments were performed in triplicate.

To induce osteogenic differentiation, the rBMSCs were cultured in α-MEM containing 10% FBS, 1% penicillin and streptomycin, 50 µg/ml ascorbic acid (Sigma-Aldrich; Merck Millipore), 10 mM β-glycerophosphate (Sigma-ldrich; Merck Millipore), and 10 nM dexamethasone (Sigma Aldrich; Merck Millipore). The culture medium was exchanged every 3 days.

Staining was performed as described in our previous study (30). Briefly, on day 21, cells were fixed and stained with Alizarin Red S staining solution. The stained monolayers were then washed 3 times with phosphate-buffered saline (PBS) and visualized using phase microscopy with an inverted microscope (DMI4000B, Leica Microsystems GmbH, Wetzlar, Germany).

The activity of ALP and total protein quantity were assessed on days 0, 3, 7, 14 and 21. The lysates was determined by LabAssay ALP colorimetric assay kit (Wako Pure Chemical Industries, Ltd., Osaka, Japan). Total proteins were determined by BCA Protein assay kit (Beyotime Institute of Biotechnology, Haimen, China) following the standard protocol. The activity of ALP was calculated as phosphorylated nitrophenol release in n/mol/h and was further normalized to the cell protein input. Each sample was assessed in triplicate.

Cultures were washed three times with PBS, and sequentially harvested cells were pelleted by centrifugation at 8,000 × g for 15 min at room temperature. Cell pellets were then resuspended in radioimmunoprecipitation assay lysis buffer containing 1% phenylmethylsulfonyl fluoride protease inhibitor, before samples were incubated on ice for 1 h. Lysates were subjected to ultrasonication on ice for further lysing and cell debris was removed by centrifugation at 16,000 × g for 10 min at 4°C. Following centrifugation, protein concentration was determined using a QuantiPro BCA assay kit (Sigma-Aldrich; Merck Millipore) and the protein supernatant was kept at −80°C for future analysis.

For immunoblotting, proteins (~40 µg) were separated on 8% SDS-PAGE and transferred to polyvinylidene difluoride membranes at 60 V for 1 h at 4°C. The membranes were blocked with milk and then incubated overnight at 4°C with primary antibodies against mouse monoclonal BMP-2 (dilution, 1:1,000; Abcam, Cambridge, MA, USA; cat no. ab6285), rabbit polyclonal TGF-β3 (dilution, 1:1,000; Abcam; cat no. ab15537), Runx2-C-terminal region (dilution, 1:1,000; Aviva Systems Biology Corporation, San Diego, CA, USA; cat no. ARP38453_P050), Osx (dilution, 1:200; Beijing Biosynthesis Biotechnology Co., Ltd., Beijing, China; cat no. bs-1110R), or mouse monoclonal β-actin (1:2,000; Abcam; cat no. ab6276,), followed by rinsing 3 times with PBS with Tween 20 for 30 min, and subsequently incubated with secondary antibodies at room temperature for 1 h. Secondary antibodies were horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (dilution, 1:2,000; Sigma-Aldrich; Merck Millipore; cat no. A0168) or HRP-conjugated goat anti-rabbit IgG (dilution, 1:3,000; Sigma-Aldrich; Merck Millipore; cat no. A0545). The membrane was washed with PBS containing 0.05% Tween 20 three times, for 10 min each time, prior to being developed using the Immobilon™ Western Chemiluminescent HRP Substrate (Merck Millipore; cat. no. WBKLS0500).

Data are presented as the mean ± standard deviation. Significance between various treatment samples was calculated using the Student's t-test. All statistical analyses were conducted with SPSS 19.0 software (IBM SPSS, Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

To evaluate the efficiency of lentivirus-mediated transfection, expression of a vector encoding GFP in rBMSCs was visualized using fluorescence microscopy. GFP was expressed in rBMSCs with high intensity and lasted stably, gradually reaching a peak value at 72 h after transfection. Images from three random fields were captured for each well and GFP-positive cells per microscope field were counted. The ratio of GFP-positive cells compared with total cells was defined as the transfection efficiency. As presented in Fig. 1A, a robust transfection efficiency of >90% was observed in each experimental group when cells were transfected with lentiviral-mediated BMP-2, TGF-β3, or BMP-2/TGF-β3 genes at an MOI of 40, 40 and 55, respectively. Furthermore, the incidence of suspended cells increased in the group undergoing co-transfection compared with single-gene transfected counterparts, indicating decreased proliferation of rBMSCs when co-transfected. However, it did not affect the transfection rate. This demonstrated that the present study efficiently delivered BMP-2 and TGF-β3 into rBMSCs together, which has been technically challenging to this point.

The expression status of corresponding exogenous genes was detected based on RT-PCR analysis, 5 days after transfection. The results indicated that expression levels of BMP-2 and TGF-β3 in lentivirus-treated rBMSCs were markedly increased compared with those of untreated cells. Notably, expression of BMP-2 in rBMSCs was significantly increased when co-expressed with TGF-β3, in comparison to that in rBMSCs transfected with BMP-2 (P=0.019). Similarly, increased TGF-β3 mRNA level was exacerbated by BMP-2 (P=0.021; Fig. 1B), indicating that BMP-2 and TGF-β3 were expressed in rBMSCs when cultured in vitro.

This RT-PCR result was consistent with results obtained by assessing BMP-2 and TGF-β3 protein expression levels using Western blotting. It was demonstrated that BMP-2 or TGF-β3 transfection significantly facilitated upregulation of the corresponding gene, and higher expression levels of BMP-2 and TGF-β3 proteins were observed in co-transfected rBMSCs than in cells transfected with a single gene (Fig. 1C), consistent with the PCR results. The data collectively suggested that BMP-2 and TGF-β3 synergistically induced expression of the other, with a possible association indicated by the mutual role they play in the bone-forming process. Following this, the present study next assessed the alteration of osteogenesis in post-co-expressed rBMSCs.

To further investigate the osteogenic function of rBMSCs undergoing BMP-2 and/or TGF-β3 delivery, expression levels of Runx2, and Osx, the representative early osteogenic-specific markers, were estimated. As expected, BMP-2 and TGF-β3 overexpression markedly upregulated Runx-2 and Osx expression levels (Fig. 2A). Notably, rBMSCs demonstrated increased expression of Runx2 and Osx when co-transfected with BMP-2 and TGF-β3, compared with those transfected with BMP-2 alone, which indicated that TGF-β3 enhanced osteogenic differentiation capacity for BMP-2 in rBMSCs.

Following lentivirus infection, ALP activities were measured to examine the mechanism by which BMP-2 and/or TGF-β3 overexpression affects the osteogenic differentiation process. Compared with the negative control, ALP activities in rBMSCs transfected with BMP-2 and TGF-β3 gradually increased with time. As presented in Fig. 2B, ALP activity in the BMP-2-transfected rBMSCs was higher than that in TGF-β3 transfected stem cells (P=0.0353 and P=0.023 at days 3 and 7, respectively; P<0.01 at days 14 and 21) possibly due to a more robust osteogenic activity of BMP-2. However, when co-transfected with BMP-2 and TGF-β3, rBMSCs presented significantly increased ALP activities at all time points compared with rBMSCs transfected with BMP2. (P=0.0187 at day 3; P<0.01 at days 7, 14 and 21). The capacity of TGF-β3 in osteogenic differentiation may be elevated by BMP-2, and BMP-2 mediated ossification was in turn enhanced by TGF-β3 delivery.

In addition, osteogenic capabilities were characterized by examining the mineralization of the extracellular matrix using Alizarin Red S staining after 21 days of culture. As hypothesized, marginal mineralized nodules were observed in negative control groups with or without osteogenic introduction (Fig. 2C). However, in agreement with data from the Western blotting, although there was no marked difference in the density of mineralized nodule areas between BMP-2 and TGF-β3 overexpressed rBMSCs, the proportion of mineralized nodules was notably increased in rBMSCs incubated with the BMP-2 and TGF-β3 encapsulated lentivirus.

The results of the present study demonstrated that when acting together, BMP-2 and TGF-β3 increased promotion of osteogenic differentiation compared with when functioning individually.