De novo design of protein oligomers for formation of hydrogels with programmable viscoelasticity

Introduction:: As we continue to appreciate the many ways that material properties, from viscoelasticity to stress relaxation, affect cell fate, it is becoming ever more important to find materials systems that allow us to connect design features at the microscopic scale to these macroscopic properties. While synthetic polymers and naturally derived biomolecules have been a workhorse of the biomaterials space for decades, their intrinsic polydispersity and batch-to-batch variation add complications in rationally designing materials with desired mechanical properties. The recent rise of recombinant protein-based biomaterials provides a new avenue for creating monodisperse precursors, resulting in networks that are perfectly defined in a way that synthetic polymers are not.¹ The additional explosion of de novo protein design simultaneously provides us with new tools to precisely decipher the connections between hydrogel network design and material mechanics.² Here, we aim to develop a new class of modular hydrogel precursors that can be used to generate a materials with a variety of material properties in a plug-and-play fashion.

Materials and Methods:: Previously de novo designed protein oligomers with valencies of 5, 12, and 60 were utilized in combination with a divalent crosslinker to form hydrogel networks.^3,4 These proteins were modified in a variety of ways, including the crosslinking mechanism (SpyCatcher/Tag⁵ – covalent, LHD101A/B⁶ – noncovalent), rigid arm length, and flexible linker length. All proteins were expressed in E. coli, purified by Ni-NTA affinity chromatography, and spin concentrated to enable the formation of 10% w/v gels with a 1:1 molar ration of crosslinking moieties. Oscillatory rheology was used to characterize the storage moduli, frequency crossover points, and to determine self-healing properties for noncovalent gels. Gels were also formed intracellularly through transfection of HeLa cells with plasmids coding for GFP-tagged precursor proteins. FRAP was used to confirm the nature of the GFP labeled structures within the cells.

Results, Conclusions, and Discussions:: We first characterized the effect of each protein structure variant on macroscopic gel properties. Here, we identified that increased flexible linker length preceding the SpyTag and SpyCatcher groups on the pentavalent and divalent oligomers, respectively, leads to moderate increases in the resultant gel’s storage modulus. Similarly, increasing the rigid arm length on the pentavalent oligomer also leads to increased material stiffness over multiple orders of magnitude. These effects are thought to be a result of minimizing steric hinderance for crosslink formation, by increasing flexibility and space between crosslinks, leading to a greater percentage of total possible crosslinks forming within the material. Conversely, while valencies of 12 or 60 did not form gels with moduli unattainable in the pentavalent system, these materials see a general decrease in stiffness with increased flexible linker length. In these systems, we believe that crosslinking efficiency is no longer limited by the length of the oligomers and instead, the addition of longer linkers simply results in increased flexibility and softer materials. The replacement of covalent crosslinking interactions with noncovalent ones resulted in a network displaying non-Newtonian shear thickening, where the material behaves as a liquid until a force is applied. These gels also display self-healing properties, typical of noncovalently crosslinked materials. When these noncovalent proteins are expressed intracellularly in HeLa, droplet-like structures form. These coacervates show fluid-like properties in response to FRAP, with recovery times longer than intracellular un-crosslinked pentavalent proteins.

This work begins to uncover the relationship between the microscopic deign of hydrogel networks and their macroscopic mechanical properties, taking us one step closer to designing materials with pre-defined mechanics. Furthermore, we have identified a non-Newtonian hydrogel system, which to our knowledge is rare. These protein gel precursors also allow for the unique opportunity to form networks intracellularly, which could provide additional applications of this work in the field of complex coacervates. Overall, we see this work as a gateway to realizing the immense, untapped design space available to us when we utilize de novo protein design as a tool to create never before seen hydrogel systems.

Acknowledgements (Optional): :

References (Optional): :

1. Li, H., Kong, N., Laver, B. & Liu, J. Hydrogels Constructed from Engineered Proteins. Small 12, 973–987 (2016).


2. Huang, P.-S., Boyken, S. E. & Baker, D. The coming of age of de novo protein design. Nature 537, 320–327 (2016).


3. Hsia, Y. et al. Design of multi-scale protein complexes by hierarchical building block fusion. Nat. Commun. 12, 2294 (2021).


4. Ueda, G. et al. Tailored design of protein nanoparticle scaffolds for multivalent presentation of viral glycoprotein antigens. eLife 9, e57659 (2020).


5. Zakeri, B. et al. Peptide tag forming a rapid covalent bond to a protein, through engineering a bacterial adhesin. Proc. Natl. Acad. Sci. 109, E690–E697 (2012).


6. Sahtoe, D. D. et al. Reconfigurable asymmetric protein assemblies through implicit negative design. Science 375, eabj7662 (2022).