(586b) Poly(ethylene glycol) and Elastin-like Protein Double-Network Hydrogels As Central Nervous System Extracellular Matrix Models

Central nervous system (CNS) tissue does not substantially regenerate from injury and disease because the extracellular environment in the damaged lesion is not conducive to neuronal regeneration. Most current in vitro models of the CNS extracellular matrix (ECM) utilize tissue harvested from rat or mouse brain and spinal cord making it difficult to isolate the effects that each individual biochemical cue has on cells. The challenges with using native tissue arise from the complexity of the CNS ECM with multiple biochemical cue expression changes occurring simultaneously during injury and disease progression, the presence of redundant effects to cells from separate ECM biochemical cues, and changes to the native tissue structural stiffness and topography from scar tissue formation.

We have developed a poly(ethylene glycol) (PEG) hydrogel with elastin-like protein (ELP) microdomains by mixing the two polymers and maintaining separate crosslinking mechanisms. PEG is crosslinked through the free radical reaction of methacrylate end-groups in the presence of the photoinitiator lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) and ultraviolet light. ELP is crosslinked by mixing with tetrakis(hydroxymethyl) phosphonium chloride (THPC); primary amines on lysine residues of ELP undergo Mannich-type condensation reactions with THPC. ELP displays a lower critical solution temperature (LCST) transition where it forms coacervates at physiological temperature which causes the two polymers to phase separate when mixed. Our crosslinking procedure fixes the polymers to form regions with high ELP concentrations and ones with high PEG concentrations. These PEG and ELP hydrogels (PEG-ELP) mimic the basic orientation of the neural interstitial matrix, which is the part of the CNS ECM found in deep tissue. The neural interstitial matrix contains a network of proteoglycans, hyaluronan, and tenascins with small amounts of fibrous proteins such as elastin and collagen. Our PEG-ELP hydrogels contain large areas of the bio-inert PEG with smaller regions of fibrous ELP, which contains RGD, a cell integrin binding motif, in the protein backbone sequence.

Preliminary data shows that PEG-ELP double-network hydrogels are more compliant than PEG hydrogels. The addition of 1 wt% ELP to a 6 wt% PEG hydrogel reduces the unswelled storage modulus from 253 Pa to 127 Pa. Oligodendrocyte precursor cells (OPC) encapsulated at 1e6 cells/ml survive well in 3 wt% PEG hydrogels with 3 wt% ELP (3/3 PEG-ELP) but do not survive in 6 wt% PEG hydrogels, as determined by live/dead imaging. Increasing the OPC concentration by an order of magnitude, 1e7 cells/ml, causes them to survive in a 6 wt% PEG hydrogel but they do not extend processes. OPCs extend processes in 3/3 PEG-ELP hydrogels over the course of 7 days. This work demonstrates that PEG-ELP hydrogels are the first step towards mimicking the neural interstitial matrix from a bottom-up approach, making them a suitable mimetic for studying the fundamental effect that ECM remodeling during CNS injury or disease has on cell recovery.