Clustering protease domains for use in directed evolution of specificity (967.6)

The objective of this study is to determine the most successful protease domains for use in directed evolution experiments against substrates with novel amino acid sequences. Building on our previous work that created an in vivo expressible fluorescence resonance energy transfer (FRET) system for protease specificity determination, we have initiated a study of protease domain clustering for in vivo protease evolution against a single recognition sequence. Using backbone root‐mean‐square deviation (RMSD), buried active‐site surface area, secondary structure analysis, and hydrophobicity scores, we have clustered protease domains into groups with a likelihood of potential success for in vivo evolution against specific amino acid sequences of maximum length. Our results demonstrate that many protease domains are not amenable to evolution studies due to energetic concerns related to instability of secondary structure interactions around the active site. Additionally, many proteases exhibit active site recognition regions that are relatively small, which eliminate the potential to target these domains against longer, and more specific, amino acid sequences. Encouraging results were obtained for two types of protease domains that have the maximum likelihood for successful evolution studies. These protease domains will be used in fluorescence‐activated cell sorting (FACS) experiments and evolved against loop regions of signal transduction receptors using our FRET protease specificity substrates. Antibody conjugates of these domains will then be constructed to target these evolved proteases to specific cell types based on antigen availability on the cell surface. Grant Funding Source : Supported by NIH GM080691