Engineering Substrate Specificity of Mammalian Proteases
MetadataShow full item record
Proteases, enzymes that catalyze the hydrolysis of amide bonds in peptides and proteins, are ubiquitous across all forms of life and affect every protein in its natural life cycle. Engineering programmed specificity onto stable protease scaffolds holds promise as a route to a new generation of useful enzymes with requisite clinical and biotechnological properties. Combinatorial approach of screening targeted protein libraries (size > 10 million) has been successfully applied for engineering the substrate specificity of disulfide-free microbial proteases. To date however, the largest size of the library screened for any mammalian protease system is 150. This is primarily due to paucity of high-throughput screening platforms that are compatible with post-translational modifications such as zymogen activation and disulfide bond formation typically required for functional mammalian proteases. In the present study, we have established a comprehensive methodology that allows efficient FACS screening of large libraries of recombinant proteases in E.coli by developing and integrating three different components: (i) surface display system compatible with mammalian proteases such as chymotrypsin and caspase-3, (ii) protease activity assay with single cell resolution, and (iii) genotype recovery method to facilitate iterative enrichment of desired phenotype. Characterizing post-translational modifications (PTMs) has been of increasing interest due to their biological functions and this knowledge is used for novel target discovery and drug development. Since PTMs can be labile, proteomic techniques relying on tandem MS have difficulty in achieving comprehensive coverage of PTM sites within proteins. Proteases which selectively recognize PTMs may improve their identification in complex protein mixtures and are easily adaptable to current proteomic methods. Using our FACS-based methodology, we have screened large (> 10 million) libraries of chymotrypsin, and engineered its substrate binding pocket to cleave after Asn residues for mapping N-linked glycosylation sites. In comparison to the wild-type chymotrypsin, the engineered variant displayed a 20,000-fold reversal in catalytic selectivity for asparagine at P1 position over tyrosine. Unlike engineered microbial proteases, we demonstrate that the ability of the engineered chymotrypsin variant to cleave after Asn in peptide substrates translates to full length proteins, an essential attribute for application to proteomics. Our results show that comprehensive engineering of the substrate specificity of mammalian proteases is feasible and when executed in the context of chymotrypsin, can enable the detection of various kinds of PTMs.