Engineering of Signaling Microenvironments with Fibronectin Domains to Enhance Tissue Regeneration

Conducting tissue healing and regeneration through biomaterials and morphogens is still an unrealized goal. Understand the multiple roles of the extracellular matrix (ECM) is indeed essential for the design of successful regenerative medicine strategies. During tissue repair and healing, cells receive numerous signals from their immediate ECM microenvironment and adhere by receptor-mediated interactions with ECM components by specialized adhesion receptors, such as integrins. As such, design and modulation of ECM analogs to ligate specific integrins is a promising approach to control cellular processes. Through production of variants of the 9th to 10th type III repeat of fibronectin (FN, FN III9-10) with variable stabilities, we engineered ligands that present different specificities for the integrin α5β1. Furthermore, the FN fragments have been engineered in order to be covalently incorporated into fibrin, a clinical relevant matrix for regenerative medicine. We demonstrated the capacity of α5β1 integrin-specific engagement to influence human mesenchymal stem cell behavior, showing that α5β1 integrin has an important role in the control of their osteogenic differentiation. Specifically, compared to FN, FN fragments with increased specificity for α5β1 versus αvβ3 integrins (FN III9*-10) results in significantly enhanced osteogenic differentiation in 2D and in a clinically relevant fibrin matrix system, while cell attachment/spreading and proliferation were comparable. On the other hand, growth factors (GFs) are key molecules for tissue morphogenesis and healing. However, while they are really promising molecules for a use in regenerative medicine applications, they often fail to prove cost-effective or even clinically efficacious during clinical trials. One of the reasons for this poor translation may lie in the rapid clearance of GFs from tissue sites in vivo, leading to the development of strategies controlling their release. Since FN has been shown to bind GFs from very different families, we first explored the possibility of FN to bind GFs much more broadly, and secondly the possibility of using FN fragments as an anchor for GF retention into fibrin matrix. We found that the 12th to 14th type III repeats of FN (FN III12-14) promiscuously bind GFs from the platelet-derived GF, fibroblast GF, transforming GF-β and neurotrophin families. Overall, 25 new binding interactions were demonstrated, supporting that GF binding may be one of FN's main physiological functions. However, the reasons for such promiscuous binding capacity were still unclear, while evidences from the literature suggested that the close proximity of the major integrin-binding domain, FN III9-10, allows joint integrin/GF-receptor signaling triggered by a complex FN/GFs. Accordingly, we found that FN fragments containing both the integrin- and GF-binding domains (FN III9(*)-10/12-14) could drastically enhance GF activities in vitro. In addition, testing which integrins were invol