Molecular crowding of collagen: a pathway to produce highly-organized collagenous structures.

Collagen in vertebrate animals is often arranged in alternating lamellae or in bundles of aligned fibrils which are designed to withstand in vivo mechanical loads. The formation of these organized structures is thought to result from a complex, large-area integration of individual cell motion and locally-controlled synthesis of fibrillar arrays via cell-surface fibripositors (direct matrix printing). The difficulty of reproducing such a process in vitro has prevented tissue engineers from constructing clinically useful load-bearing connective tissue directly from collagen. However, we and others have taken the view that long-range organizational information is potentially encoded into the structure of the collagen molecule itself, allowing the control of fibril organization to extend far from cell (or bounding) surfaces. We here demonstrate a simple, fast, cell-free method capable of producing highly-organized, anistropic collagen fibrillar lamellae de novo which persist over relatively long-distances (tens to hundreds of microns). Our approach to nanoscale organizational control takes advantage of the intrinsic physiochemical properties of collagen molecules by inducing collagen association through molecular crowding and geometric confinement. To mimic biological tissues which comprise planar, aligned collagen lamellae (e.g. cornea, lamellar bone or annulus fibrosus), type I collagen was confined to a thin, planar geometry, concentrated through molecular crowding and polymerized. The resulting fibrillar lamellae show a striking resemblance to native load-bearing lamellae in that the fibrils are small, generally aligned in the plane of the confining space and change direction en masse throughout the thickness of the construct. The process of organizational control is consistent with embryonic development where the bounded planar cell sheets produced by fibroblasts suggest a similar confinement/concentration strategy. Such a simple approach to nanoscale organizational control of structure not only makes de novo tissue engineering a possibility, but also suggests a clearer pathway to organization for fibroblasts than direct matrix printing.