(534g) Deciphering and Engineering the Substrate Specificity of Proteases (Invited Speaker)

Proteases are ubiquitous in biology, playing significant roles in initiating, regulating, and terminating cellular processes. The diversity and breadth of protease substrate preference have been harnessed for a variety of applications, including protease therapeutics, protein purification, and mass spectrometry-based proteomics. Furthermore, with advances in protein engineering and synthetic biology, proteases are now being utilized for studying protein-protein interactions, imaging newly synthesized proteins, and performing logic operations inside cells. To successfully repurpose proteases towards these applications, it is often necessary to engineer their catalytic properties using Directed Evolution. In this vein, to effectively isolate protease variants with the desired phenotypes from a large pool, the high-throughput enzyme screening or selection system should ideally exhibit a broad operational range (dose-response or sensitivity over a large variation in input), and a high dynamic range (signal-to-noise ratio). Unfortunately, current protease engineering platforms often require laborious design-build-test cycles to determine, optimize, and operate within these ranges. Failure at any or all three steps prevents one from assessing the functionality of variant libraries in between evolution cycles and lowers the probability of isolating protease variants with the desired phenotypes.

Here we describe YESS 2.0, a highly modular and customizable yeast endoplasmic sequestration screening system suitable for engineering and profiling the specificity of protein-modifying enzymes. By incorporating features to modulate gene transcription, as well as substrate and enzyme spatial sequestration within a versatile and seamless assembly method, YESS 2.0 achieves broad operational and dynamic range. To showcase YESS 2.0, we evolve a TEV protease variant (eTEV) with an 8-fold higher catalytic efficiency to obtain the fastest TEV protease variant to date. Due to the unique features of our system, this phenotype is strictly attributable to an increase in turnover number (k_cat). Second, we used YESS 2.0 coupled with NextGen Sequencing to profile the substrate specificity of insulin-degrading enzyme (IDE) and discover IDE substrates with increased activity compared to the prototypical insulin peptide. Lastly, we show for the first time that YESS 2.0 supports calcium-independent sortase-mediated ligations and confirm that residues directly C-terminal of the pentapeptide motif heavily influence the rate of SML. This state-of-the-art platform offers unmatched versatility in profiling and engineering the specificity of protein-modifying enzymes and should enable even more ambitious future undertakings.