(19c) An in Vitro Chondro-Osteo-Vascular Triphasic Model of the Osteochondral Complex

Introduction: The development of therapies against osteoarthritis (OA), a main cause of disability in the US, is hindered by the lack of veritable in vitro osteochondral (OC) models (1). In this work, we have developed a chondro-osteo-vascular (COV) triphasic model of the OC complex within a recently developed 3D printed microphysiological tissue system (MPS) bioreactor (2) that allows the separate flow of specific media to the chondral and osseous components while maintain them in contact and allowing tissue-tissue communication (3). The engineered osteochondral construct is based on wet spinning of poly(ε-caprolactone) (PCL) subsequently combined with photocrosslinkable methacrylated gelatin (gelMA)

Materials and Methods: PCL (CAPA 6500, 50000 g/mol) fibrous scaffolds manufactured by computer-aided wet-spinning (CAWS) (4) for bone were combined with gelMA scaffolds for cartilage (3). Human bone marrow mesenchymal stem cells (hMSCs) were seeded on PCL scaffolds at 8×10⁴ cells/construct, expanded for 10 days, then placed in the bottom chamber of the bioreactor (Fig. 1A), or suspended in 10% gelMA/0.15% LAP (w/v)/PBS at 1×10⁷ cells/ml, poured in the upper part of the insert and photopolymerized in situ. Chondrogenic/osteogenic medium (3) were supplied through the upper/lower conduits, at 1 ml/day. After 2 weeks, PCL scaffold pores were filled with a hMSCs:GFP-HUVECs 1:4 suspension (1×106 cell/ml) in 5% gelMA. Osseous constructs were incubated for 2 more weeks with 1:1 OM:endothelial growth medium.

Results and Discussion: At 4 weeks, histology showed chondrogenic (Alcian blue) and osteogenic differentiation (Alizarin red) in gelMA and PCL components, respectively. Similarly, RT-PCR of individual OC components showed upregulation of chondral (COL2, ACAN, SOX9) and osseous genes (RUNX2, BSPII, OPN). HUVECs formed interconnected capillary-like networks, induced stronger Alizarin red staining and enhanced osseous and chondral gene expression vs. no-HUVECs control. These results suggest that the engineered COV-OC construct mimics native OC tissue in terms of structural architecture and gene expression profile and highlights the crosstalk occurring between the vascularized bone compartment and the chondral compartment.

Conclusion: The COV-OC model recapitulating three different tissue types of the osteochondral unit could represent a key step towards an effective in vitro analog of the OC unit to understand the biology of cartilage and bone development and regeneration, and to develop high throughput screening approaches for drug development.

References: (1) Alexander P.G. et al. Exp Biol Med 2014, 239:1080-1095. (2) Lozito T.P. et al. Stem Cell Res & Ther 2013, 4(Suppl 1):S6. (3) Lin H. et al. Mol Pharm 2014, 11(7):2203-12. (4) Puppi D. et al. Biom Microdev 2012, 14(6):1115-1127.

Acknowledgements: Commonwealth of Pennsylvania, NIH (1UG3 TR002136-01), Ri.MED Foundation.