Where elastin is intrinsically random, collagen is rigidly ordered. Human collagen is defined by the tripeptide repetitive motif Gly-X-Y, dominated by Gly-Pro-Hyp sequences. it exists in a variety of types and undergoes a number of post-translational modifications, including the prolific but specific hydroxylation of proline. This makes it quite difficult to produce recombinantly, particularly in prokaryotic model organisms such as Escherichia coli. However, there is a large class of proteins defined by the Gly-X-Y motif with a similar supersecondary structure which does not take advantage of hydroxyproline; these proteins are sometimes referred to as bacterial collagens. Despite this name, these bacterial proteins come from a single cell and are not used in tissue-scale structures. A method to produce arbitrarily long bacterial collagen triple helices is thus desirable.
By combining the work of Yoshizumi et al. in 2011 in using coiled coil trimerization domains to promote folding of bacterial collagen with the much earlier work of Nautiyal et al. in 1995 designing an obligate heterotrimeric coiled coil, it becomes possible to imagine directed formation of a bacterial collagen trimer with sticky ends. This construct could then assemble into potentially infinitely long fibers. Variations on this concept have been designed in the past, with Tanrikulu et al. in 2016 being a recent example, but they have all relied on chemical peptide synthesis. This coiled coil directed method would be the first based on a bacterially expressed monomer without chemical modifications. The modular structure of this construct also opens up the possibility of including cross-linking functionality, which is being explored through co-opting exons 21 and 23 of human tropoelastin representing a simple lysine-based cross-linking domain.