(571e) Biomolecular Constructs for Localized Retention of Therapeutic Enzymes

Enzymes are promising drug candidates due to their ability to rapidly convert many copies of substrate molecules into products with high specificity and selectivity. Yet, only about 10% of all proteins currently approved for therapeutic use by the FDA are enzymes, in spite of their continued industrial evaluation since the 1960s. A key challenge hindering the successful translation of therapeutic enzymes is poor accumulation within the target tissue, which can lead to deleterious off-site effects, partial or total catalytic activity loss, and rapid clearance from the body. Here I will present two new strategies to prolong enzyme activity within specific tissue sites following minimally-invasive injection.

In the first strategy, enzymes are recombinantly fused to a protein domain that binds to carbohydrates that are abundant on the cell surface and within the extracellular matrix. Binding of the fusion protein to carbohydrates hinders enzyme diffusion away from the injection site by anchoring the enzyme to the tissue. To demonstrate feasibility of this approach, we first fused the carbohydrate-binding protein, galectin-3 (“G3”), to enzymes. Variants of this fusion construct are retained within different tissues for up to 7 days. Fusion of galectin-3 to the immunoregulatory enzyme, indoleamine-2,3-dioxygenase, provides robust suppression of inflammation in various preclinical models, including lipopolysaccharide challenge, periodontal disease, and osteoarthritis. Inspired by observations that the carbohydrate-binding affinity of galectin-3 is naturally increased through its self-association into multivalent quaternary structures, we further engineered G3-enzyme fusion constructs to self-assemble oligomeric “nanoassemblies”. An oligomer having three G3 and three enzyme domains (i.e., a “trimeric nanoassembly”) was retained in various tissues for up to 14 days. We have further developed this platform to create a library of nanoassemblies having 1-6 G3 domains, which provides a tunable range of enzyme binding affinities for cell surface and tissue glycans.

In the second strategy, we fuse enzymes to a tag that mediates their integration into peptide nanofibers. The nanofibers are assembled from a pair of charge-complementary peptides, known as CATCH, which resist self-association due to electrostatic repulsion, yet efficiently co-assemble into an elongated beta-sheet structure when combined in solution. An enzyme fused to either CATCH peptide can be mixed with the like-charged CATCH peptide at a desired molar ratio prior to the triggering of nanofiber formation by the addition of the complementary CATCH peptide. The dose of enzyme can be tuned by simply varying its amount in the pre-fibrillization mixture. Likewise, multiple enzymes can be co-immobilized by introducing each into the pre-fibrillization mixture. At sufficiently high concentrations, CATCH nanofibers physically crosslink into hydrogels that demonstrate shear thinning and recovery, which enables direct injection into tissues through a small-diameter needle. CATCH hydrogels persist at subcutaneous injection sites for up to two months, while eliciting only minimal inflammation and no significant adaptive immunity. Collectively, these strategies demonstrate the potential of biomolecular engineering to provide localized retention of therapeutic enzymes at specific tissue sites from days to weeks.