An Iterative Approach to de novo Computational Enzyme Design and the Successful Application to the Kemp Elimination
<p>The development of reliable methods for the 'on demand" de novo design of an enzymatic catalyst for an arbitrary chemical reaction has been an elusive goal of the computational protein design community. Recent successful results of de novo computational enzyme design have been en...
| Summary: | <p>The development of reliable methods for the 'on demand" de novo design of an enzymatic catalyst for an arbitrary chemical reaction has been an elusive goal of the computational protein design community. Recent successful results of de novo computational enzyme design have been encouraging, but the activity of the enzymes produced so far is still well below that of natural enzymes and the generalizability of these methods has yet to be established.</p>
<p>Presented in this thesis are methods that we have developed for the computational design of enzyme active sites as well as results from the evaluation of these methods through a test case, the Kemp elimination. Initial Kemp elimination designs were shown to be inactive. However, in the course of refining these design procedures, we carried out extensive theoretical and experimental evaluation of several of these inactive designs, which allowed us to identify the causes of the inactivity and led to adjustments of our design procedure. These modified methods were then successfully used to design four distinct enzymes for this reaction in three inert scaffolds including the scaffold that housed the previously inactive designs. In addition, we demonstrate that molecular dynamics simulations can accurately predict the activity of designed Kemp elimination enzymes and can be used as a reliable prescreening step, allowing us to focus our experimental efforts on designs that are most likely to be active.</p>
<p>The work presented here demonstrates that the cyclic evaluation and redesign of both active and inactive enzymes was instrumental in the identification and resolution of deficiencies in our computational methods and directly resulted in de novo designed enzymes with novel and increased activity.</p>
|
|---|