(504a) Engineering and Profiling the Substrate Specificity of Human Proteases for Therapeutic Applications

Autoimmune diseases such as inflammatory bowel disease (IBD) are characterized in part by an overproduction of regulatory proteins, such as tumor necrosis factor α (TNF-α). Antibodies that target TNF-α have been at the forefront of treatments for IBD. While antibody therapies can be effective, they are often limited by a loss of efficacy over time and can lead to an increased risk of opportunistic infections. To address these issues, an alternative approach is to degrade target proteins using proteases, some of which have been used as therapeutic agents for several decades. However, in order to expand the pharmacopeia of many human proteases, their substrate specificity needs to be fine-tuned to enhance their efficacy and avoid adverse side effects, as well as to exhibit resistance to in vivo inhibitors.

Recently, we reported a high-throughput yeast platform named Yeast Endoplasmic Reticulum (ER) Sequestration Screening (YESS). In YESS, a protease substrate display cassette, which is composed of selection and counter-selection substrates flanked by epitope tags, is co-expressed in the yeast ER along with a protease. Peptide display is detected by fluorescent labeling and cleavage is screened by fluorescence-activated cell sorting (FACS). Human neutrophil elastase (HNE), an extracellular serine protease with broad specificity, may modulate immune inflammatory responses through the degradation of both membrane-bound and soluble TNF-α. In order to develop a protease-based drug to treat IBD, we challenged a mutagenized library of HNE through YESS for increased specificity towards TNF-α. After four rounds of FACS, we isolated HNE variants that specifically degrade peptide substrates found in soluble TNF-α with minimal activity towards membrane-shedding TNF-α sequence preferred by the WT enzyme. Fortuitously, the HNE variants no longer cleaved peptides derived from serpin B1, a potent HNE in vivo inhibitor. Similarly, in an effort to develop a treatment for SLE, we repurposed the specificity of human matrilysin (hMMP7), a metalloprotease involved in the breakdown of extracellular matrix, to specifically cleave the IgG hinge region which leads to IgG deactivation. To characterize these protease variants biochemically, we are concurrently developing methods for human protease expression and purification. In the future, we will characterize these variants in vivo using mice models of IBD and SLE.

In addition, we coupled YESS with Next-Gen sequencing (YESS-NGS) to profile the substrate specificity of wild-type and our engineered HNEs. In this respect, YESS offers several advantages over exiting protease profiling methods. YESS circumvents the need to have purified proteins, enables access a large sample of sequences (10⁷) to determine fine patterns of reactivity not previously discovered. We have applied YESS-NGS to profile several other proteases, including TEV proteases and their variants, human kallikreins, neutrophil elastase and their variants, neprilysin, cathepsin E, among others.