Periostin accelerates bone healing mediated by human mesenchymal stem cell-embedded hydroxyapatite/tricalcium phosphate scaffold

Abstract

Background: Periostin, an extracellular matrix protein, is expressed in bone, more specifically, the periosteum and periodontal ligaments, and plays a key role in formation and metabolism of bone tissues. Human adipose tissue-derived mesenchymal stem cells (hASCs) have been reported to differentiate into osteoblasts and stimulate bone repair. However, the role of periostin in hASC-mediated bone healing has not been clarified. In the current study, we examined the effect of periostin on bone healing capacity of hASCs in a critical size calvarial defect model.

Methods and results: Recombinant periostin protein stimulated migration, adhesion, and proliferation of hASCs in vitro. Implantation of either hASCs or periostin resulted in slight, but not significant, stimulation of bone healing, whereas co-implantation of hASCs together with periostin further potentiated bone healing. In addition, the number of Ki67-positive proliferating cells was significantly increased in calvarial defects by co-implantation of both hASCs and periostin. Consistently, proliferation of administered hASCs was stimulated by co-implantation with periostin in vivo. In addition, co-delivery of hASCs with periostin resulted in markedly increased numbers of CD31-positive endothelial cells and α-SMA-positive arterioles in calvarial defects.

Conclusions: These results suggest that recombinant periostin potentiates hASC-mediated bone healing by stimulating proliferation of transplanted hASCs and angiogenesis in calvarial defects.

Fig 1. Periostin stimulates chemotaxis migration, adhesion, and proliferation activity of hASCs in vitro.

(A) Dose-dependent effect of periostin on hASC migration. hASCs were loaded in the upper chamber of a chemotaxis chamber apparatus and recombinant periostin with the indicated concentration was placed in the lower chamber, followed by measurement of the number of migrated cells after 12 h incubation. (B) Checkerboard analysis of the periostin-induced migration of hASCs was measured using a chemotaxis chamber. Periostin was placed in either the bottom, top, or both chambers of the chemotaxis system and hASCs were loaded into the upper chamber. The number of migrated cells was quantified after incubation of the cells for 12 h. Data represent mean ± S.D.; *, p < 0.05. (C) 96-well plates were coated with indicated concentration of periostin and adhesion of hASCs onto the plates was determined. (D) Dose dependence of periostin-stimulated proliferation. hASCs were treated with the increasing concentrations of periostin for 3 days and proliferation of cells was determined by MTT assay. Data represent mean ± S.D. *, p < 0.05 vs control.

Fig 2. Micro-CT analysis of bone regeneration following implantation of hASCs and/or periostin into calvarial defects.

Calvarial defects were implanted with HA/TCP scaffold bearing hASCs and/or periostin, or mock-treated (control). (A) Posterior view of a coronal sliced micro-CT images of the calvaria was captured at week 0 and 8. (B) The distance between the advancing edges was quantified from coronal-section view images of calvarial defects by Micro-CT. Data represent mean ± S.D. (n = 8). *, p < 0.05 vs control.

Fig 3. Effects of hASCs and/or periostin on regeneration of calvarial bone defects.

(A) Masson’s trichrome staining of calvarial defects (blue color: mineralized bone tissue; red color: non-mineralized bone tissue). Higher magnification images of the regions highlighted by the black box in left panels are shown in right panels. (B) The gap of calvarial defects in the Massons’s trichrome staining images was quantified and shown as the distance between the advancing edges. Data represent mean ± S.D. (n = 8). *, p < 0.05 vs control.

Fig 4. Histological analysis of newly regenerated bone after implantation of hASCs and/or periostin.

(A) H&E staining of calvarial defects eight weeks after implantation of hASCs and/or periostin. Higher magnification images of the regions highlighted by the black box in left panels are shown in right panels. Scale bar = 100 μm. (B) Area of newly regenerated bone tissues was determined in the defected bone and the percentage of bone area per healed area was calculated. Data represent mean ± S.D. (n = 8). *, p < 0.05 vs control.

Fig 5. Effects of hASCs and periostin implantation on cell proliferation within defective calvarial bone.

(A) Proliferating cells within defective calvaria were detected by immunostaining with anti-Ki67 antibody at week 8 after implantation of hASCs and/or periostin. Scale bar = 50 μm. (B) Numbers of Ki67-positive cells and total nuclei per field were counted and expressed as the relative percentage of Ki67-positive cells per total nuclei. Data represent mean ± S.D. (n = 8). #, p < 0.05; *, p < 0.05 vs control.

Fig 6. Effects of periostin on proliferation of implanted hASCs in vivo.

(A) hASCs were labeled with CM-DiI and loaded into HA/TCP scaffold with or without periostin, followed by implantation of the scaffold into the defective calvaria. Two weeks after implantation, tissue specimens were immunostained with anti-PCNA antibody. Overlaid images of CM-DiI-positive hASCs (red color), nuclei (blue color), and PCNA (green color) are shown. Scale bar = 20 μm. (B) The numbers of PCNA- and CM-DiI-double positive cells, which indicate proliferating hASCs, were counted and the percentage of PCNA-positive cells per CM-DiI-positive cells was determined. Data represent mean ± S.D. (n = 8). *, p < 0.05 vs control.

Fig 7. Effects of hASCs and periostin on angiogenesis in calvarial defects.

(A) Immunostaining of endothelial cells and smooth muscle cells in calvarial defects implanted with hASCs and/or periostin. The specimens were immunostained with anti-CD31 (red color) or anti-α-SMA (green color) antibodies, and overlaid images with nuclei (blue color) are shown. Scale bar = 50 μm. The numbers of CD31-positive capillaries (B) and α-SMA-positive blood vessels (C) per high power field were counted. Data represent mean ± S.D. (n = 8). #, p < 0.05; *, p < 0.05 vs control.