The present study was performed in accordance with the Declaration of Helsinki for investigations involving human subjects, and was approved by the Ethics Committee of Shanghai Ninth People’s Hospital Affiliated to Shanghai Second Medical University (Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China). Dermal fibroblasts were isolated from residual skin during circumcision surgery in 5 older children (age, 6-9 years) between June and July 2009. All patients provided written informed consent. The hDDFCs were isolated as described previously (13). Immunomagnetic isolation of the CD105⁺ cells was performed as described previously (14). Briefly, suspensions of hDDFCs obtained from human dermis were washed once with 1X PBS and resuspended with magnetic cell sorting (MACS) buffer (1X PBS containing 0.5% fetal bovine serum (FBS; cat. no. SH30087.01; HyClone; GE Healthcare Life Sciences, Logan, UT, USA) and 2 mM ethylenediamine tetraacetic acid, pH 7.2). A nylon mesh was used to filter cell suspensions (30-µm pore). The cells were resuspended in MACS buffer at 10⁷ cells per 80 µl, mixed with 20 µl microbeads of directly conjugated mouse anti-human CD105 antibody (1:200; cat. no. MCA1557; Bio-Rad Laboratories, Inc., Hercules, CA, USA), and incubated at 4°C for 15 min on a rotator in the dark. Following washing in 1X PBS, the DDFCs-CD105 cells were resuspended in MACS buffer and processed in an LS⁺/VS⁺ separation column. The column was removed from the magnetic device, and the cells were flushed out with MACS buffer. The CD105⁻ and CD105⁺ cells were recovered by centrifugation at 300 × g for 10 min at 4°C for future use.

To determine the purity of the CD105⁺ cells, the cells were analyzed using the FACSCalibur device (BD Biosciences, San Jose, CA). Aliquots containing 0.5×10⁶ CD105⁺ cells were incubated with phycoerythrin (PE)-conjugated anti-CD105 antibody (1:100; cat. no. FAB10971B, R&D Systems, Inc., Minneapolis, MN, USA) on ice for 30 min, washed three times, and resuspended in wash buffer. IgG-PE (1;200; cat. no. SC-3756; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) was used as the isotype control. The hDDFCs were suspended basal media, which consisted of DMEM (DMEM-HG; Invitrogen, Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% FBS (HyClone; GE Healthcare Life Sciences) and maintained in the original culture medium at 37°C with 5% CO₂.

The AdCMV-hBMP7 and AdCMV-LacZ viruses were generated by Shanghai Key Laboratory of Tissue Engineering (Shanghai, China) using the BD Adeno-Xä expression system (BD Biosciences). In vitro ligation was used to incorporate a mammalian expression cassette into a replication-incompetent (∆E1/∆E3) human adenoviral type 5 (Ad5) genome. Full-length human BMP7 (hBMP7) cDNA and pBlue-BMP7 (4.5 kb), obtained from the American Tissue Culture Collection (Manassas, VA, USA) were cloned into a shuttle vector via the KpnI and NotI sites. The expression cassette was excised from the recombinant pShuttle2 plasmid using the PI-SceI and I-CeuI restriction endonucleases and ligated into the BD Adeno-X viral DNA. The recombinant linearized pAdeno-X DNA (PacI digestion) was transfected into 293 cells (Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China) using FuGENE 6 (Roche Diagnostics, Basel, Switzerland) and the virus was purified by cesium chloride gradient ultracentrifugation at 100,000 × g for 2 h at 4°C and stored in 10% glycerol in PBS. Titers were determined by end-point dilution, yielding ~1.2×10¹⁰ plaque-forming units (pfu)/ml. Recombinant adenoviruses expressing β-galactosidase DNA (AdCMV-LacZ) were generated as controls.

The cultured CD105⁺ hDDFCs were harvested and treated with virus at a multiplicity of infection of 200 pfu/cell in 1 ml DMEM-HG medium for 4 h at 37°C. Following culture in DMEM with 10% FBS in a humidified atmosphere with 5% CO₂, the cells were harvested, counted, and resuspended in an alginate hydrogel at the indicated concentrations.

Small interfering RNAs (siRNAs) against Smad4 (target sequence: 5′-GCC ATA GTG AAG GAC TGT T-3′) were synthesized by GeneChem (Shanghai, China). The hDDFCs (1×10⁵ cells) infected with recombinant adenovirus were transfected with Smad4 siRNA (2.5 µg) for 48 h using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific, Inc.). For inhibitor treatment, the recombinant adenovirus-infected CD105⁺ hDDFCs were incubated with osteogenic medium (OM) consisting of DMEM, 10% FBS, 1% antibiotics, 0.01 µM 1,25-dihydroxyvitamin D3, 50 µM ascorbate-2-phosphate, and 10 mM β-glycerophosphate (Sigma, EMD Millipore, Billerica, MA, USA) in the presence or absence of the p38 inhibitor (SB203580; 10 µM; cat. no. tlrl-sb20; InvivoGen, San Diego, CA, USA) at 37°C for 24 h for western blot detection of p38 MAPK, 7 days for osteogenesis-associated gene expression, alkaline phosphatase (ALP) staining and ALP activity assay, and 21 days for Alizarin Red S staining.

Adipogenic differentiation was performed as previously described (13). Briefly, the cells were cultured at 37°C on cover slips at a density of 3,000 cells/cm² in adipogenic differentiation medium consisting of low-glucose DMEM, 10% FBS, 1% antibiotics, 0.5 mM isobutylmethylxanthine, 1 µM dexamethasone, 10 µM insulin, and 200 µM indomethacin (Sigma, EMD Millipore) for 3 weeks. To confirm differentiation, the cells were stained with Oil Red O (Sigma, EMD Millipore) for 5 min following induction, and photographed under an Axiovert inverted microscope (Zeiss AX10; Carl Zeiss AG, Oberkochen, Germany).

Osteogenic differentiation was examined as previously described (13). Briefly, the cells were seeded at a density of 3,000 cells/cm² and cultured at 37°C in OM consisting of DMEM, 10% FBS, 1% antibiotics, 0.01 µM 1,25-dihydroxyvitamin D3, 50 µM ascorbate-2-phosphate, and 10 mM β-glycerophosphate. Alizarin Red S staining was performed 4 weeks following induction to confirm calcium deposition.

ALP staining was performed using the BCIP/NBT Alkaline Phosphatase Color Development kit (Beyotime Institute of Biotechnology, Jiangsu, China) following 7 days of osteogenic induction. ALP activity was assessed using an Alkaline Phosphatase Yellow (pNpp) Liquid Substrate system for ELISA (Sigma, EMD Millipore) following 7 days of osteogenic induction.

Total RNA was extracted from the hDDFCs using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Inc.). For the synthesis of first strand cDNA, AMV reverse transcriptase (Promega, Madison, WI, USA) was used. qPCR was then performed using the SYBR Premix Ex Taq kit (Takara Biotechnology Co., Ltd., Dalian, China), in a volume of 25 µl containing 2 µl cDNA, 12.5 µl SYBR Premix Ex Taq, 0.5 µl each of the forward and reverse primers, and 9.5 µl RNase-free water, in an ABI 7900 sequence detection system under the following conditions: Initial denaturation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 30 sec, and extension at 72°C for 30 sec. The samples were then normalized to the expression of GAPDH using the 2^−ΔΔCq method (15). The following primers were used for qPCR: BMP7, sense 5′-GGG CTT CTC CTA CCC CTA CA-3′ and antisense 5′-ACG TCT CAT TGT CGA AGC GT-3′; osteopontin (OPN), sense 5′-AGG CCA AAA TAG AGC TGC CT-3′ and antisense 5′-GTG GTC ATG GCT TTC GTT GG-3′; osteocalcin (OCN), sense 5′-CGT AGA AGC GCC GAT AGG C-3′ and antisense 5′-ATG AGA GCC CTC ACA CTC CTC-3′; osterix (OSX), sense 5′-CTC TGC GGG ACT CAA CAA CT-3′ and antisense 5′-ATG GAT GCC TGC CTT GTA CC-3′; runt related transcription factor 2 (RUNX2), sense 5′-TCT GGC CTT CCA CTC TCA GT-3′ and antisense 5′-GTC CAC TCT GGC TTT GGG AA-3′; ALP, sense 5′-ACC GCT TCC CAT ATG TGG CT-3′ and antisense 5′-GGT CTG GAA GTT GCC CTT GA-3′; GAPDH, sense 5′-ACC ATC TTC CAG GAG CGA GA-3′ and antisense 5′-TGG TTC ACA CCC ATG ACG AA-3′.

Protein was extracted from the cells using RIPA buffer and protein concentration was quantified with a BCA kit (Bio-Rad Laboratories, Inc., Richmond, CA, USA). Aliquots of cell lysates containing equal quantities of protein (30 µg) were separated by 12% SDS-PAGE and transferred onto nitrocellulose membranes (Amersham, GE Healthcare Life Sciences, Chalfont, UK). The membranes were blocked with 5% nonfat milk in TBST overnight, and incubated with primary antibodies against p-Smad 1/5/8 (1:1,000; cat. no. 9511), Smad1 (1:1,000; cat. no. 9743), extracellular signal-regulated kinase (ERK; 1:2,000; cat. no. 4696), phosphorylated (p)-ERK (1:1,000; cat. no. 9101), c-Jun N-terminal kinase (JNK; 1:1,000; cat. no. 9252), and p-JNK (1:1,000; cat. no. 9251) from Cell Signaling Technology, Inc. (Beverly, MA, USA); p-p38 (1:500; cat. no. sc-7973), p38 (1:1,000; cat. no. sc-7972), BMP7 (1:200; cat. no. sc-9305), runt-related transcription factor 2 (RUNX2; 1:1,000; cat. no. sc-12488), and ALP (1:200; cat. no. sc-15065) from Santa Cruz Biotechnology, Inc.); and OSX (1:500; cat. no. ab94744), OPN (1:1,000; cat. no. ab8448), OCN (1:500; cat. no. ab93876) and GAPDH (1:2,000; cat. no. ab22555) from Abcam (Cambridge, UK) at 4°C overnight. The membranes were then washed three times in TBST, and incubated with the corresponding horseradish peroxidase-conjugated secondary antibodies (1:5,000; cat. no. 31462 and 32230; both from Thermo Fisher Scientific, Inc.) at 4°C for 1 h. Signals were detected using the Pierce ECL western blotting substrate. The quantification of western blot bands was conducted by comparison against the GAPDH bands using ImageJ software (version 1.48; National Institutes of Health, Bethesda, MD, USA).

A total of 10 immunodeficient C57BL/6 male mice aged 4-5 weeks (15±0.30 g) were obtained from Shanghai Second Medical University Center of Laboratory Animals (Shanghai, China). All experimental protocols were approved by the Animal Experiment Committee of Shanghai Second Medical University. Mice had ad libitum access to food and water and were maintained in a 12-h light/dark cycle at 21±2°C with a relative humidity of 45±10%. The mice were injected with CD105⁺ hDDFCs transfected with Ad-LacZ (negative control) or Ad-BMP7. Alginate (Sigma, EMD Millipore) was dissolved in PBS to a concentration of 2% (w/v), and the adenovirus-infected CD105⁺ hDDFCs were added to the alginate solution at a density of 2.5×10⁷ cells/ml. The cell-alginate preparation was mixed with excessive aqueous slurries of 2M calcium chloride to produce hydrogels and washed with PBS. The mice were anesthetized and injected subcutaneously with two 400 µl injections of cell-hydrogel mixture using an 18-gauge needle into the dorsal panniculus carnosus. The Ad-BMP7-transduced cells were injected into the right side and Ad-LacZ-transduced cells were injected into the left side of mice. At 12 weeks post-implantation, the mice were sacrificed by CO₂ asphyxiation. Bone constructs were dissected and surrounding soft tissue was removed. X-ray images were acquired and the bone constructs were fixed in 4% paraformaldehyde for the histologic and immunohistochemical analyses of bone formation.

Bone formation was monitored in vivo by fixing bone constructs with 4% paraformaldehyde, followed by decalcifi-cation in 10% formic acid. The samples were dehydrated in a graded alcohol series, diaphonized in xylene, and embedded in paraffin. The paraffin blocks were sectioned into 5-µm-thick slides, deparaffinized, hydrated, stained with hematoxylin and eosin (H&E) and photographed under an Axiovert inverted microscope (Zeiss AX10).

The results are expressed as the mean ± standard deviation. Student’s t-test was used for comparisons of two groups and one-way analysis of variance was used for multiple comparisons. Data were analyzed using SPSS software (version 14.0; SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Primary CD105^- hDDFCs exhibited an extended and long narrow shape, whereas CD105⁺ hDDFCs exhibited a spindle-like morphology following 12 h in culture (Fig. 1A). Flow cytometric analysis confirmed the expression of the surface marker CD105 in isolated cells and the isolated CD105⁺ cells had a purity of 95±2.5% (Fig. 1B). The differentiation of hDDFCs into osteogenic and adipogenic lineages was confirmed by Alizarin red staining and Oil-red-O staining 3-4 weeks following induction. Representative images of Alizarin red- and Oil-red-O-stained CD105^- and CD105⁺ cells are shown in Fig. 1C and D.

To examine the effect of BMP7 on the osteogenic differentiation potential of CD105⁺ hDDFCs, the cells were infected with recombinant adenovirus expressing hBMP7 and cultured in BM or OM. The expression of BMP7 and osteogenesis-associated genes was assessed by RT-qPCR and western blot analyses. The results showed that the expression levels of RUNX2, OSX, OCN, OPN, and ALP were significantly upregulated by OM, and BMP7 significantly enhanced this effect (Fig. 2A and B). Alizarin red staining at 21 days post-osteogenic induction showed that BMP7 enhanced the intensity of staining, indicating increased calcium deposition (Fig. 2C). BMP7 significantly increased the expression and activity of ALP by ~1.5-fold in the cells cultured in OM (Fig. 2D and E). Taken together, these results indicated that BMP7 has a potent effect on enhancing the osteogenic differentiation of CD105⁺ hDDFCs.

The present study further examined the mechanisms underlying the effect of BMP7 on the osteogenic differentiation of CD105⁺ hDDFCs by assessing the expression of proteins associated with the BMP canonical Smad-dependent and non-canonical Smad-independent pathways. The results of the western blot analysis and densitometric quantification showed that BMP7-expressing adenovirus infection significantly enhanced the OM-induced upregulation of p-Smad1/5/8, indicating the activation of Smad signaling (Fig. 3A). OM treatment induced the phosphorylation of ERK, p38 and JNK, and the overexpression of BMP7 further upregulated the OM-induced phosphorylated form of p38 (Fig. 3B). These results indicated the activation of p38/MAPK signaling by BMP7 in the CD105⁺ hDDFCs.

To further elucidate the involvement of the Smad pathway in the effect of BMP7, the present study examined the effect of Smad4 knockdown on the enhancing effect of BMP7 on the OM-induced osteogenic differentiation of CD105⁺ hDDFCs. The silencing of Smad eliminated the BMP7-induced upregulation of RUNX2, OSX, OCN, OPN and ALP in the presence of OM, compared with that in the si-Control transfected cells (Fig. 4A). Consistent with these results, Smad knockdown reversed the BMP7-induced increase in Alizarin red and ALP staining (Fig. 4B) and ALP activity (Fig. 4C), indicating that the effect of BMP7 on promoting the osteogenic differentiation of CD105⁺ hDDFCs occurred via a Smad-dependent pathway. The expression of p38 MAPK (p38/p-p38) was also measured and the results showed that it was not altered by siR-Smad4 (data not shown).

To determine the role of the p38 MAPK pathway, the cells were treated with the p38 inhibitor SB203580 and incubated in OM in the presence or absence of BMP7. The results of the western blot analysis showed that inhibition of p38 attenuated the BMP7-induced upregulation of OSX, OCN, OPN, and ALP, whereas the expression of RUNX2 was not affected by the inhibition of p38 (Fig. 5A). The Alizarin red and ALP staining showed that the inhibition of p38 attenuated the effect of BMP7 on enhancing the OM-induced osteogenic differentiation of CD105⁺ hDDFCs (Fig. 5B). In addition, treatment with SB203580 reversed the BMP7-induced increase in ALP activity (Fig. 5C). Taken together, these results indicated that the p38 MAPK signaling pathway was involved in the enhancement of the OM-induced osteogenic differentiation of CD105⁺ hDDFCs by BMP7. The results also showed that p38 inhibitor SB203580 suppressed BMP7-induced Smad1/5/8 phosphorylation (data not shown).

The effects of BMP7 on osteogenic differentiation were examined in nude mice injected intramuscularly with adenovirus infected CD105⁺ hDDFCs. Representative X-ray images and bone volumes from mice are shown in Fig. 6A, indicating enhanced bone formation by BMP7 in vivo. The H&E staining of tissue sections showed enhanced trabecular bone formation in Ad-BMP7 sections compared with Ad-LacZ sections (Fig. 6B). These results confirmed the effect of BMP7 on promoting the osteogenic differentiation of CD105⁺ hDDFCs in an in vivo model.