Diffusion and Molecular Association in Artificial Protein Hydrogels

Citation

Rapp, Peter Butterweck (2017) Diffusion and Molecular Association in Artificial Protein Hydrogels. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/Z9CV4FSF. https://resolver.caltech.edu/CaltechTHESIS:06012017-074605131

Abstract

Artificial proteins may be programmed to reversibly self-assemble into water-soluble networks, or “hydrogels”, by encoding them with terminal coiled-coil forming domains. Such networks are model viscoelastic materials. The well-defined molecular structures adopted by proteins, combined with their facile preparation by recombinant synthesis, invite a careful exploration of the relationship between protein sequence and the resulting network properties.

This work explores the relationship between network reorganization and diffusion from the perspective of single chains, using artificial elastin-like proteins as a model system. We make use of fluorescence recovery after photobleaching (FRAP), a classic biophysical technique, to measure chain mobilities as a function of network structure and probe architecture. Reversible network association is demonstrated to control the effective diffusivity of network-bound chains, and a novel mechanism of chain transport is proposed: the chains naturally partition into various bound states, and move by “hopping” from site to site in between binding events.

A careful analysis of the equilibrium constants that control this partioning leads to the conclusion that the sequential binding of identical chain ends to the network is inherently asymmetric: the first association is always stronger than the second. This binding asymmetry is shown to arise from a strong entropic penalty for chain entry into the fully bound state due to local network structure. We derive a simple equation predicting the degree of binding asymmetry as a function of network geometry from equilibrium statistical mechanics. A large set of self-diffusivity measurements on a series of model telechelic proteins finds good agreement with this new theory. Generalized binding asymmetry for chains with many associative domains also holds.

Finally, the inherent viscoelasticity of the elastin-like network is found to couple with an entropically driven phase separation above a critical temperature set point. Relaxation of the viscoelastic stress throughout the process of phase domain segregation is found to induce highly dynamic phase patterns. The time evolution of these patterns illustrates that a delicate balance of surface tension and viscoelastic stress controls pattern formation in viscoelastic materials.

Thesis Files