A computational modeling approach for enhancing self-assembly and biofunctionalisation of collagen biomimetic peptides

Abstract

Collagen fibers are essential components of tissues, which are highly conserved across the animal kingdom and could be extremely useful in tissue engineering. The formation of these macromolecular fibers depends on molecular interactions-based self-assembly of the basic building blocks of collagen called tropocollagens. Several attempts to produce biomimetic collagen have been described, however the best method to achieve the optimal material for tissue engineering has not been established. Here, we describe a bottom-up approach to design two computationally mutated molecular models that use non-covalent interactions to cross-link triple helices of tropocollagen molecules and thus promote self-association. Implementing a graph theory approach in the software FIRST reveals the hotspots that are crucial for the overall rigidity of the supramolecular helical structures and the remaining non-hotspots available for mutations. The mutated models were further decorated with GFOGER, a known collagen cell binding motif, to depict a biofunctional model. In addition to their recognized role of cell binding, the charged residues of the binding motif appeared to enhance further the supramolecular helical association. These findings could help to produce biomimetic collagen for biomedical applications.