C3H10T1/2 and C2C12 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in complete Dulbecco's modified Eagle's medium (DMEM) (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY, USA), 100 U/ml of penicillin and 100 μg/ml of streptomycin. Cells were incubated at 37°C in 5% CO₂. Mouse embryo fibroblasts (MEFs) were isolated from post coitus day 12.5 mice (5 female and 5 male mice, purchased from Beijing Institute of Chinese Medicine), as described previously (39).

p-Smad1/5/8 (cat. no. 9516) was purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA) and Smad1/5/8 (cat. no. sc-6031-R) was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; cat. no. TA-08) and β-actin (cat. no. TA-09) were purchased from Zhongshan Golden Bridge Biotechnology (Beijing, China). γ-secretase inhibitor DAPT was purchased from Sigma-Aldrich (St. Louis, MO, USA), dissolved in dimethyl sulfoxide (DMSO) and stored at −80°C. The recombinant adenoviruses including Ad-BMP9, Ad-DLL1, Ad-dnNotch1, Ad-dnALK1, Ad-dnALK2, Ad-GFP and Ad-RFP were kindly provided by Dr Tong-Chuan He (University of Chicago Medical Center, Chicago, IL, USA).

Total RNA was extracted from the cells with TRIzol reagents (Takara, Otsu, Japan), and was reverse transcribed to cDNA using the Takara PrimeScript RT reagent kit. Semi-quantitative RT-PCR was performed as described previously (40,41). The cDNA products were further diluted 5- to 10-fold and used in the successive experiments. A touchdown cycling program was used as follows: 94°C for 5 min, 94°C for 30 sec, 68°C for 30 sec, and 72°C for 12 cycles with a decrease in 1°C/cycle; then, 94°C for 30 sec, 55°C for 30 sec, and 72°C for 30 sec for 18–27 cycles depending on the abundance of the target genes. The PCR products were resolved on 2% agarose gels. All samples were normalized with the expression level of mouse GAPDH.

qPCR analysis was performed using a SYBR Premix Ex Taq kit (Takara). The cycling program consisted of 94°C for 2 min and 30 cycles at 92°C for 20 sec, 57°C for 30 sec, and 72°C for 20 sec, followed by a plate read at 78°C for each cycle. All samples were evaluated in triplicate and normalized to GAPDH. The primer pairs of target genes are presented in Table I.

BMP9-conditioned medium (BMP9-CM) was prepared as described previously (42). Briefly, subconfluent HCT116 cells (in a 75 cm² flask) were infected with an optimal titer of Ad-BMP9 or Ad-GFP control. At 4 h after treatment, the medium was replaced with serum-free DMEM. The conditioned medium was collected at 48 h after infection and used as soon as possible.

ALP activity was analyzed by a modified Great EscAPe SEAP chemiluminescence kit (Takara). Each assay condition was performed in triplicate, and the results were repeated in three independent experiments. ALP activity was normalized by the total cellular protein concentrations among the samples.

Cultured cells were seeded in 24-well cell culture plates and infected with Ad-DLL1 followed by treatment with BMP9-CM, ascorbic acid (50 mg/ml) and β-glycerophosphate (10 mM). On day 11, the mineralized matrix nodules were stained for calcium precipitation by means of Alizarin Red S staining, as described previously (41). Briefly, the cells were fixed with 0.05% (v/v) glutaraldehyde at room temperature for 10 min. After being washed with double-distilled water, the fixed cells were incubated with 0.4% Alizarin Red S (Sigma-Aldrich) for 5 min, followed by extensive washing with double-distilled water. The staining of calcium mineral deposits was recorded under a microscope.

The cells were seeded in 75-cm² cell culture dishes and subjected to the indicated treatments. At the indicated time-points, the cells were harvested and washed with cold phosphate-buffered saline (PBS) and extracted in protein buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1% P40, and 1 mM EDTA) in the presence of protease and phosphatase inhibitors. Proteins were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Following electrophoretic separation, the proteins were transferred onto a PVDF membrane. The membrane was blocked with 5% bovine serum albumin (BSA) for 2 h at 37°C and probed with the primary antibody (diluted 1:1,000) overnight at 4°C, then washed three times with Tris-buffered saline contained 0.1% Tween-20 (TBST). The membrane was then incubated with horseradish peroxidase-conjugated goat anti-rabbit (cat. no. ZB-2301) or goat anti-mouse (cat. no. ZB-2305) secondary antibodies (diluted 1:5,000; Zhongshan Golden Bridge Biotechnology). Finally, the membrane was exposed with ECL (Thermo Fisher Scientific, Inc., Rockford, IL, USA).

The cells were transfected with 2 μg per flask of BMP R-Smad-binding element luciferase reporter (p12xSBE-Luc) using Lipofectamine 2000 (Invitrogen Corp., Carlsbad, CA, USA). At 24 h after transfection, the cells were replated in 24-well plates and treated with BMP9-CM and/or Ad-DLL1, Ad-dnNotch1, DAPT. At 36 h after treatment, the cells were lysed and harvested for luciferase assays using the Luciferase Assay kit (Promega Corp., Madison, WI, USA). Each assay condition was performed in triplicate. The results were repeated in at least three independent experiments. Luciferase activity was normalized with the total cellular protein concentrations among the samples.

MEFs were co-infected with Ad-BMP9 and/or Ad-DLL1, Ad-dnNotch1 for 24 h, and harvested for subcutaneous injection (5×10⁶ cells per injection) into the flanks of athymic nude (nu/nu) mice (4- to 6-week-old male Sprague-Dawley). At 4 week(s) after treatment, the animals were euthanized, and the bony masses were collected for micro-CT imaging and histologic evaluation. All animal experiments were approved by the Ethics Committee of Chongqing Medical University.

Animals were euthanized at 4 week(s) and bony masses were imaged using high-performance micro-CT imager component of a GE Triumph (GE Healthcare, Piscataway, NJ, USA) trimodality preclinical imaging system. All image data analysis was performed using Amira 5.3 (Visage Imaging, Inc., San Diego, CA, USA).

Retrieved tissues were fixed in 10% neutral-buffered formalin overnight and processed for paraffin embedding. Serial sections of the embedded tissues were stained with H&E and Masson's trichrome staining as previously described (43).

Cells were seeded in 6-well plates and harvested after a 72-h treatment, washed with PBS three times and fixed with 75% iced-ethanol at 4°C. The fixed cells were washed with PBS and stained with propidium iodide-containing RNase followed by fluorescence-activated cell sorting for cell cycle analysis.

Data are reported as the means ± SD. Statistical analysis was conducted using SPSS software version 14 (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered as statistically significant.

We sought to explore whether or not Notch signaling has any effect on a BMP9-induced early osteogenic marker. We first adopted DAPT, an inhibitor of the γ-secretase complex, to inhibit the Notch activity of MSCs. We also confirmed that Ad-DLL1 upregulated the level of DLL1 in MEFs, and the dominant-negative mutant of Notch1 (dnNotch1) which contains extracellular and transmembrane domains but lacks cytoplasmic domains was highly expressed in the Ad-dnNotch1-infected cells (data not shown). We used Ad-dnNotch1 and Ad-DLL1 to downregulate and upregulate Notch signaling, respectively (data not shown). Moreover, by using ALP staining and activity assay, we found that BMP9-induced ALP activity was significantly inhibited by DAPT and Ad-dnNotch1 in a concentration-dependent manner (Fig. 1A, B and D). Conversely, Ad-DLL1 enhanced BMP9-induced ALP activity (Fig. 1C). These data suggested that Notch signaling enhances the BMP9-induced early osteogenic differentiation of MSCs.

Although ALP is a well-established early osteogenic marker, it may not be an accurate predictor of the late stage of osteogenic differentiation and bone formation (41). Thus, we aimed to determine whether Notch signaling has an effect on the expression of BMP9-induced late osteogenic markers, such as osteopontin (OPN) and osteocalcin (OCN). By RT-PCR or Alizarin Red S staining, we found that the combination of Ad-BMP9 and DAPT treatment significantly decreased the expression of OCN in a concentration-dependent manner (Fig. 2A); The expression of OCN and OPN decreased when treated with Ad-dnNotch1 (Fig. 2B). Conversely, Ad-DLL1 treatment was found to enhance the matrix mineralization induced by BMP9 (Fig. 2C). These data suggested that Notch signaling facilitates BMP9-induced late osteogenic differentiation in MSCs.

Our in vitro data revealed that Notch signaling plays an important role in BMP9-induced osteogenic differentiation of MSCs. Thus, we determined the effect of Notch on BMP9-induced ectopic bone formation. The infected MEFs were collected and injected subcutaneously into athymic nude mice. After 4 week(s), the animals were sacrificed and the bony masses were retrieved (Fig. 3A). The overall sizes of bony masses from the Ad-BMP9 combined with Ad-DLL1 group were apparently larger than that from the Ad-BMP9 group, and the bony masses from the group treated with Ad-BMP9 and Ad-dnNotch1 were significantly smaller than that of the Ad-BMP9 group. Histologic analysis indicated that, compared with the Ad-BMP9 group, the trabecular bone and osteoid matrix area was obviously increased in the group of Ad-BMP9 combined with Ad-DLL1, and was significantly decreased in the Ad-BMP9 combined with Ad-dnNotch1 group (Fig. 3B). Micro-CT scanning analysis showed the same results (Fig. 3C). These data demonstrated that Notch signaling enhances BMP9-induced ectopic bone formation.

We then explored the possible mechanism behind the effect of Notch signaling on the BMP9-induced osteogenic differentiation of MSCs. BMP/Smad signaling is the classical pathway for BMP9 to induced osteogenic differentiation in MSCs. Thus, we aimed to ascertain whether or not Notch regulates this signaling. In the presence of BMP9-CM, we treated the cells for 36 h with DAPT, Ad-dnNotch1 or Ad-DLL1, respectively. Western blot analysis showed that DAPT had no obvious effects on total protein level of Smad1/5/8, but decreased the phosphorylation level of Smad1/5/8 in the C2C12 cells (Fig. 4A). Similar results were found in MEFs treated with Ad-BMP9 and Ad-dnNotch1 (Fig. 4A). Conversely, Ad-DLL1 was found to enhance the phosphorylation level of Smad1/5/8 induced by BMP9 in the C3H10T1/2 and C2C12 cells and MEFs (Fig. 4B). Using the BMP responsive Smad1/5/8 reporter, p12xSBE-luc, we found that Ad-DLL1 promoted BMP9-induced reporter activities prominently in the C3H10T1/2 and C2C12 cells and MEFs, and Ad-dnNotch1 impaired the BMP9-induced reporter act ivities (Fig. 4C). Collectively, these results suggested that Notch signaling can enhance the BMP/Smad signaling transduction induced by BMP9 in MSCs.

It has been demonstrated in our previous study that runt-related transcription factor 2 (Runx2), inhibitor of differentiation (Id)1, 2 and 3 are targets of BMP9, and are critical to BMP9-induced osteogenic differentiation in MSCs (40). Type I collagen (Colla1) is the special collagen secreted by osteoblast cells. With RT-PCR analysis, we found that the expression of Runx2 and Colla1 induced by BMP9 was decreased by DAPT in a concentration-dependent manner in C2C12 cells, and similar results were found in MEFs when treated with Ad-dnNotch1 combined with BMP9-CM. Yet, the BMP9-induced expression levels of Runx2 and Col1a1 were increased when treated with Ad-DLL1 combined with BMP9-CM (Fig. 5A). We also found that the expression levels of Id1, Id2 and Id3 were decreased by DAPT in the presence of BMP9, but only Id1 was decreased by Ad-dnNotch1 combined with BMP9-CM treatment in MEFs (Fig. 5B). These results indicated that Notch signaling was able to regulate BMP9-induced essential osteogenic factors, but there may be some differences between different ligands/receptors.

ALK1 and ALK2 are functional receptors essential for BMP9 osteogenic activity (42). BMP9 can increase the expression of ALK1 and ALK2 in MSCs, which is likely to be a novel clue to demonstrate the molecular mechanism underlying BMP9-induced osteogenic differentiation of MSCs. Therefore, we aimed to ascertain whether or not Notch can affect ALK1 and ALK2. With qPCR, we found that it was ALK2 but not ALK1 that was significantly downregulated by Ad-dnNotch1 combined with BMP9-CM in MEFs (Fig. 6A). Ad-DLL1 promoted the gene expression of ALK2 induced by BMP9 in MEFs, but had no apparent effect on ALK1 gene expression (Fig. 6A). In order to ascertain whether Notch signaling augments BMP9-induced osteogenic differentiation by increasing the gene expression of ALK2, we treated cells with Ad-dnALK1 and Ad-dnALK2, respectively, in the presence of BMP9-CM. We noted that BMP9-induced osteogenic differentiation was markedly impaired by Ad-dnALK1 and Ad-dnALK2, and Ad-DLL1 rescued the inhibitory effect of dnALK2, but had no significant effect on dnALK1 (Fig. 6B and C). These results suggested that DLL1/Notch signaling may regulate BMP9-induced osteogenic differentiation of MSCs by partly increasing BMP9-dependent upregulation of ALK2.

Notch signaling is involved in regulating the balance between cell differentiation and stem cell proliferation during the development of numerous tissues. Notch signaling may affect BMP9-induced osteogenic differentiation by regulating the proliferation of MEFs. Using FCM analysis, we found that following treatment withAd-DLL1 combined with BMP9-CM the percentage of MEFs in the S phase was increased compared to the percentage of cells treated with Ad-RFP combined with BMP9-CM. In addition, the percentage of cells in the G0/G1 phase decreased significantly. We also noted that Ad-dnNotch1 decreased the S phase cell percentage induced by BMP9-CM, but apparently increased the cell percentage in G0/G1 (Fig. 7A and B). Based on these data, Notch may promote the proliferation of MSCs in the presence of BMP9-CM.