Multiparameter Optimization in Directed Evolution: Engineering Thermostability, Enantioselectivity, and Activity of an Epoxide Hydrolase

The challenge of optimizing several parameters in the directed evolution of enzymes remains a central issue. In this study we address the thermostability, enantioselectivity, and activity of limonene epoxide hydrolase (LEH) as the catalyst in the hydrolytic desymmetrization of cyclohexene oxide with formation of (R,R)- and (S,S)-cyclohexane-1,2-diol. Wild type LEH shows a thermostability of T-50(30) = 41 C and an enanioselectivity of 2% ee (S,S). Two approaches are described herein. In one strategy, the mutations generated previously by Janssen, Baker, and co-workers for notably increased thermostability are combined with mutations evolved earlier for enhanced enantioselectivity. Although highly enantioselective R,R and S,S variants (92-93% ee) with increases in T-50(30) by 10-11 C were obtained, relative to wild type LEH the tradeoff in activity was significant. The second strategy based on the simultaneous optimization of both parameters using iterative saturation mutagenesis (ISM) with minimized tradeoff in activity proved to be superior. Several notably improved variants were observed, a reasonable compromise being R,R- and S,S-selective LEH variants (80-94% ee) showing enhanced thermostability by 5-10 degrees C and still reasonable levels of activity. Analysis of the X-ray structure of the S,S variant (94% ee) with and without diol product sheds light on the origin of altered stereoselectivity.