Summary: The ability to design stimulus-responsiveness into any enzyme of choice would aid in our ability to interrogate and intervene in biological processes with exquisitely high spatial and temporal precision. By constructing stimulus-responsiveness into enzymes from the bottom up, insights into fundamental biophysical principles underlying allosteric effects in natural systems will also be achieved (learning by building). The focus of this proposal is the bottom-up development of stimulus-responsive prodrug-activating enzymes by synergistically combining protein design through computational approaches (Khare) with protein engineering through unnatural amino acid (UAA) mutagenesis (Deiters). Genetic incorporation of unnatural amino acids allows introducing bio-orthogonal switches in a site-specific manner, and computational modeling enables rational rewiring of the structure and conformational landscape of the protein active site and UAA microenvironment with atomic resolution. By combining these state-of-the-art techniques with well-established protocols for detailed kinetic, structural, and biophysical characterization, we hypothesize that a bottom up framework for introducing stimulus-responsiveness in proteins will be deciphered. As our model systems, we will use the carboxypeptidase G2/nitrogen mustard prodrug and cytosine deaminase/5-fluorouracil enzyme/prodrug pairs, both of which have been extensively investigated in a chemotherapy application called directed enzyme prodrug therapy (DEPT). Their utility in a therapeutic setting for DEPT will be enhanced by rendering them conditionally activatable. In Aim 1, we will use azobenzene-containing photo-responsive UAAs for rendering the designed enzymes photocontrollable. In Aim 2, we will develop methodology for designing zymogenized versions of enzymes that can be activated by tissue-specific proteolytic enzymes, such as matrix metalloproteases. In Aim 3, we will use a variety of structural and biophysical techniques to validate designed enzymes and provide feedback for further design iterations and modeling methodology improvement. While our focus is on the two enzymes mentioned, the methods we are developing, however, will be transferable to the control of activity and delivery of a variety of other enzymes, making this a general approach to design new stimulus-responsive enzymes and potentially also allowing unprecedented selectivity and optimal delivery of chemotherapies.