Small molecule inhibitor of TGF-β signaling enables robust osteogenesis of autologous GMSCs to successfully repair minipig severe maxillofacial bone defects

Abstract

Background: Clinically, for stem cell-based therapy (SCBT), autologous stem cells are considered better than allogenic stem cells because of little immune rejection and no risk of communicable disease infection. However, severe maxillofacial bone defects restoration needs sufficient autologous stem cells, and this remains a challenge worldwide. Human gingival mesenchymal stem cells (hGMSCs) derived from clinically discarded, easily obtainable, and self-healing autologous gingival tissues, have higher proliferation rate compared with autologous bone marrow mesenchymal stem cells (hBMSCs). But for clinical bone regeneration purpose, GMSCs have inferior osteogenic differentiation capability. In this study, a TGF-β signaling inhibitor SB431542 was used to enhance GMSCs osteogenesis in vitro and to repair minipig severe maxillofacial bone defects.

Methods: hGMSCs were isolated and cultured from clinically discarded gingival tissues. The effects of SB431542 on proliferation, apoptosis, and osteogenic differentiation of hGMSCs were analyzed in vitro, and then, SB431542-treated hGMSCs composited with Bio-Oss® were transplanted into immunocompromised mice subcutaneously to explore osteogenic differentiation in vivo. After that, SB431542-treated autologous pig GMSCs (pGMSCs) composited with Bio-Oss® were transplanted into circular confined defects (5 mm × 12 mm) in minipigs maxillary to investigate severe bone defect regeneration. Minipigs were sacrificed at 2 months and nude mice at 8 weeks to retrieve specimens for histological or micro-CT or CBCT analysis. Effects of SB431542 on TGF-β and BMP signaling in hGMSCs were investigated by Western Blot or qRT-PCR.

Results: One micromolar of SB431542 treatment induced a robust osteogenesis of hGMSCs in vitro, without adverse effect on apoptosis and growth. In vivo, 1 μM SB431542 treatment also enabled striking osteogenesis of hGMSCs subcutaneously in nude mice and advanced new bone formation of pGMSCs in minipig maxillary bone defect model. In addition, SB431542-treated hGMSCs markedly increased bone-related proteins expression, and BMP2 and BMP4 gene expression. Conversely, SMAD3 protein-dependent TGF-β signal pathway phosphorylation was decreased.

Conclusions: Our study show that osteogenic differentiation of GMSCs treated with TGF-β signaling inhibitor SB431542 was increased, and SB431542-treated autologous pig GMSCs could successfully repair minipig severe maxillofacial bone defects. This preclinical study brings about a promising large bone regeneration therapeutic potential of autologous GMSCs induced by SB431542 in clinic settings.

Keywords: BMP; Bone defect; GMSCs; Osteogenic differentiation; SB431542; TGF-β signaling.

Fig. 1

Isolation and identification of human GMSCs. a Typical morphology of fibroblast-like cells growing out from a gingival tissue fragment after culturing 5 days. When the cells reached 80% confluence from the passage 0 for about 10 days, they were subcultured and proliferated. The fibroblast-like cells were digested by trypsin and had high purity with a homogeneous spindle shape at passage 2 for 2 weeks. Scale bar, 500 μm. b and c Results of flow cytometry analysis were showed of the expression of cell surface markers related to mesenchymal stem cells (CD44, CD73, CD90, and CD105), hematopoietic stem cells (CD34 and CD45), or macrophage and B-cells (CD19, CD11b, HLA-DR, and HLA-DQ). CD, cluster of differentiation. n = 3. Error bars represent the SD. d The colonies were formed by human gingival mesenchymal stem cells (hGMSCs) at low seeding density for 2 weeks culture. Scale bar, 500 μm

Fig. 2

Effects of SB431542 on Growing status of hGMSCs. a and b Concentration-dependent apoptosis assay of hGMSCs with different concentration of SB4315342(0, 0.1, 1, and10μM) cultured for 24 h or 48 h. n = 3. c Cell growth curve of hGMSCs with different concentrations of SB4315342 (0, 0.1, 1, and10μM) by cell growth assay. Cell growth assay showed that the growth of group with 10 μM SB431542 was worse than the other groups at day7. n = 4. The results represent mean ± SD, ns, no statistically significant difference, p > 0.05, *p < 0.05, **p < 0.01, vs. 0 μM, one-way ANOVA with Tukey’s multiple comparison test

Fig. 3

SB431542 treatment promotes osteogenic differentiation potential of hGMSCs. a hGMSCs formed mineralized nodules that stained positive for Alizarin Red following 21 days of osteogenic induction with various concentrations of SB4315342(0, 0.1, and1μM). hGMSCs showed improvement the capability of osteogenic differentiation as the concentration of SB4315342 increased. Scale bars, 500 μm. b Quantification of positive alizarin red staining showed that hGMSCs stimulated with SB431542 had significantly increased osteogenic differentiation capability compared with the other. n = 3. The results represent mean ± SD, ***p < 0.001, one-way ANOVA with Tukey’s multiple comparison test. c Western blotting analysis of osteoblastic differentiation-related protein in hGMSCs during osteogenic differentiation in the presence or absence of SB431542 (1 μM) at the indicated time points (days 0, 1, 3, 5, 7, 9, 11). GAPDH was used as a protein loading control. n = 3. OI, osteogenic induction medium; SB, 1 μM SB431542; D, day; ALP, alkaline phosphatase; COL-1, collagen type I; RUNX2, runt-related transcription factor 2; OPN, osteopontin

Fig. 4

In vivo bone regeneration of hGMSCs in immunocompromised mice subcutaneously. a Three-dimensional (3D) visualization based on the CBCT images of mice.3D image creation of soft tissue (a1), 3D rendering of materials at mouse bone density (a2), and maximum intensity projection (MIP) full range visualization (a3) were created using the Software NNT viewer. CBCT, cone beam computed tomography. b Macroscopic view of allograft subcutaneous transplanted in immunocompromised mice skin at 12 weeks. The graft Bio-Oss® at upper left (b1), Bio-Oss®/hGMSCs at lower left (b2), and Bio-Oss®/hGMSCs/SB431542 at lower right (b3). c Representative images of specimen stained with H&E and Masson staining retrieved from immunocompromised mice after implantation for 12 weeks. CT, connective tissues; NB, new bone. Scale bars, 200 μm. H&E, Hematoxylin-Eosin; hGMSCs, human gingival mesenchymal stem cells; SB,1 μM SB431542. d Quantification of the new bone formation with H&E and Masson staining. n = 4 mice/group, 5 views each. n = 4 mice/group. Data represent mean ± SEM. *p < 0.05, ***p < 0.001, one-way ANOVA with Tukey’s multiple comparison test

Fig. 5

In vivo bone regeneration of autogenous pGMSCs in minipigs maxillary bone defects model. a Macroscopic view of surgical procedure of bone defect model in Minipigs. Defects measuring 12 mm in diameter and 5 mm in depth were prepared on each side of the maxillary. b CBCT images of grafts in maxillary bone defects at 2 months post-implantation. Maximum intensity projection full range visualization (Ac), were created using the Software NNT viewer. Images of 0.5-mm-thick free-cut specimens were generated across the four implants (yellow rings) individually. b1, autogenous bone blocks; b2, Bio-Oss®/pGMSCs/1 μM SB431542; b3, Bio-Oss®/pGMSCs; b4, Bio-Oss®. pGMSCs, pig gingival mesenchymal stem cells; CBCT, cone beam computed tomography. c Coronal images of Micro-CT show the different reparation effects of grafts. c1, autogenous bone blocks; c2, Bio-Oss®/pGMSCs/1 μM SB431542; c3, Bio-Oss®/pGMSCs; c4, Bio-Oss®. d The average CT value and bone density (BD) of new bone in the bone defects varied between the different groups analyzed by micro-CT. BD, bone density. n = 3, in each group. e Representative images of sections stained with Masson and H&E retrieved from minipigs after implantation for 2 months. B, mature bone; CT, connective tissues; NB, new bone. Scale bars, 200 μm. H&E, Hematoxylin-Eosin; f Quantification of the new bone formation by H&E and Masson staining. n = 3 pigs/group, 5 views each. n = 3 pigs/group. Data represent mean ± SEM. ns, no statistically significant difference p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA with Tukey’s multiple comparison test

Fig. 6

Effects of SB431542 on the TGF-β and BMP signaling in hGMSCs. a Western blotting analysis of phosphorylation protein levels of SMAD2, SMAD3, ERK1/2, JNK, and P38 with or without SB431542 (1 μM) at the indicated time points (0, 15, 30, 60, 120, 180, 300 min). GAPDH was used as a protein loading control. b mRNA expression of BMP4 and BMP6 were analyzed by qRT-PCR during osteogenic differentiation of hGMSCs treatment with or without SB431542 (1 μM) at a serious time points (0, 3, 6, 9, 12 day). The relative mRNA expression is normalized to β-ACTIN. Student’s t test. c, d Alizarin Red staining of hGMSCs (c) with quantification (d) after 21 days in osteogenic induction medium with both SB and noggin or SB alone. Osteogenic induction medium treatment as control. One-way ANOVA with Tukey’s multiple comparison test. n = 3. Data are means ± SEM. ns, no statistically significant difference p > 0.05, *p < 0.05, **p < 0.001, ***p < 0.001. Scale bars, 500 μm. SB, 1 μM SB431542; N, Noggin; OI, osteogenic induction medium; P, phosphorylation; min, minutes