Seeing Double for Mechanobiology: Discrete biomaterial softening of IPNs through orthogonal enzymatic degradation

Introduction:: The interactions between cells and their extracellular environment are complex and dynamic. During development, normal physiological processes and disease progression, the extracellular matrix (ECM) undergoes a multitude of changes in biochemical composition, biomechanical properties, and topography. In particular, the mechanical contributions of the ECM are known to play a significant role in regulating signal transduction, cell motility, proliferation, differentiation and other decisions of cellular fate. Hydrogel biomaterials with spatiotemporal control of matrix stiffness have proven extremely useful for studying these mechanobiological phenomena in a physiologically relevant system. As such, we sought to design a hydrogel model that can be softened in situ with the goal of studying how softening processes impact cellular function and phenotype.

Materials and Methods:: Our design is based on two interpenetrating poly(ethylene glycol) (PEG) networks that can be independently formed and sequentially degraded upon treatment with sortases—bacterial transpeptidases—that act orthogonally to the mammalian proteome. Wild-type SrtA binds to the sorting motif “LPXTG” and cleaves between the Thr-Gly peptide bond and installs a polyglycine motif. Wild-type SrtA has been evolved to yield two orthogonal sortases—eSrtA(2A-9) and eSrtA(4S-9)—that recognize with high specificity the sequences LAETG and LPESG, respectively; these peptide sequences can be included in the crosslinking moieties for our two networks. The double networks (DNs) are based on two popular click chemistries employed in the biomaterials space: radical thiol-ene polymerization and the spontaneous azide-alkyne cycloaddition. Given that the second network is light-initiated, we can pattern stiffnesses in preformed substrates, and subsequently remove these patterns upon treatment with the eSrt(4S-9) enzyme. To fully degrade our hydrogels and recapture cells for downstream analysis, we degraded our remaining network with the eSrt(2A-9) enzyme. The gentle and bioorthogonal nature of the gelation as well as softening reactions allowed for cells to be encapsulated within the dynamic hydrogels for weeklong experiments with no adverse effects.

Results, Conclusions, and Discussions:: Atomic force microscopy of swollen DN gels showed the ability to modulate Young’s moduli from ~1kPa (single network) to 5.5kPa (DN); softening the gels led to a significant decrease in modulus, returning to baseline single network (SN) stiffness. Patterned hydrogels exhibited similar differences in moduli as bulk DN or SN gels. 10T1/2 fibroblasts displayed excellent viability after 7 days in all gel formulations, including dynamic gels that were softened on day 3. RT-qPCR revealed elevated expression of downstream Hippo pathway genes involved in mechanotransduction in 10T1/2 cells, such as CTGF and Cyr61, in stiffer double network gels, as compared to soft, single network gels; dynamic gels (softened on day 3) displayed signatures similar to those encapsulate for the whole culture duration in SNs. Human mesenchymal stem cells (hMSCs) were encapsulated in patterned hydrogels for 7 days and their spreading was analyzed via a Python pipeline. As expected, hMSCs exhibited greater spreading in the softer regions as opposed to the stiffer regions; additionally, we observed cells aligning perpendicularly to the striped pattern in the soft regions.

We have developed a dynamically softening hydrogel system based on the sensitivity of two polymer networks to orthogonal sortase variants. The mechanical properties of our model are highly tunable by varying network composition and exposure to light and can be dynamically altered to capture the mechanics of healthy and diseased tissue. Moreover, our design allows for spatial control over gelation; these patterns can be subsequently degraded, with no effect on the underlying substrate. We have demonstrated that our approach is cytocompatible and allows for terminal release of cells for further pooled and/or single-cell analysis. While we did not see significant differences in gene expression in the dynamic gels, we expect that including other time points would capture distinct signatures. We expect that our platform could be used to interrogate mechano-memory, or the ability for cells to remember their past mechanical environment and how this influences cell fate in normal physiological processes, or disease progression in fibrosis or cancer, in 3D, which has previously not been reported in similarly suited materials.


Acknowledgements (Optional): :

References (Optional): :