De Novo Reconstruction of a Functional Bone by Tissue Engineering

Introduction Large bone defects are a challenge to the orthopaedic surgeon. In the present study, we explored the possibility of preparing tissue engineered bone in an autologous manner avoiding the use of either allogenic or xenogenic products in the mesenchymal stem cells (MSCs) cultures. For this purpose, we amplified MSCs from sheep in autologous serum (AS) and prepared constructs by loading MSCs on either a natural coral scaffold or a coralline-derived porous hydroxyapatite (HAC) scaffolds. The osteogenic potential of the constructs was then assessed in criticalsize defects in sheep. Materials and Methods MSCs amplification MSCs are expanded in 10% AS with osteogenic supplements. Experimental groups Coral-MSC2D constructs are composed of MSCs and a natural coral scaffold. Cells are loaded onto the scaffold the day prior implantation Coral/MSC3D constructs are composed of MSCs and a natural coral scaffold (n=7). Cells are loaded onto the scaffold 15 days prior implantation and cultured in static conditions. HAC/MSC3D constructs are composed of MSCs and a HAC scaffold(n=7). Cells are loaded onto the scaffold 15 days prior implantation and cultured in static conditions. Control scaffolds: Defects left empty (n=5) or filled either with coral (n=6) or HAC scaffolds (n=7) or autografts (n=7)harvested at the iliac crest, were also implanted as controls. Surgical model: A 2.5 cm long mid-diaphyseal resection was created on the metatarsus of adult sheep and filled with a construct, an autograft or a control material. Animals were X-rays each month and sacrified at 4 months except one which was kept for 14 months. Bone formation was determined by image analysis on undecalcified sections using Leica Qwin image analysis system. Results In vitro study of MSC distribution in constructs: The number of MSCs per scaffold was similar in all the constructs (≈1.8x107 cells). Confocal imaging of constructs before their implantation showed that in Coral/MSC3D or HAC/MSC3D constructs, MSCs were observed almost solely at the periphery, forming a pseudoperiosteal layer around the biomaterial whereas in Coral/MSC2D constructs, MSCs were present throughout the implants. Bone healing in defects filled with coral constructs at 16 weeks: Defects filled with coral scaffolds(fig.1A) or coral/MSC2D constructs (fig 1 B, C) showed no bridging of the cortices in all animals. In these groups, the mean bone surface areas (BSA) was less than 8 % of the total defect surface areas and these values did not differ from one to another. In contrast, defects filled with coral/MSC3D constructs exhibited a significant increase in the rate of bone formation when compared to bone defects filled with coral or coral /MSC 2D. However, clinical bone healing actually occurred in only one sheep.