Manipulating regioselectivity of an epoxide hydrolase for single enzymatic synthesis of (R)-1,2-diols from racemic epoxides.

Both the activity and regioselectivity of Phaseolus vulgaris epoxide hydrolase were remarkably improved via reshaping two substrate tunnels based on rational design. The elegant one-step enantioconvergent hydrolysis of seven rac-epoxides was achieved by single mutants, allowing green and efficient access to valuable (R)-1,2 diols with high eep (90.1-98.3%) and yields.

Substantially improving the enantioconvergence of PvEH1, a Phaseolus vulgaris epoxide hydrolase, towards m-chlorostyrene oxide by laboratory evolution.

BACKGROUND: Epoxide hydrolase can regioselectively catalyze the oxirane ring-opening hydrolysis of rac-epoxides producing the corresponding chiral diols. In our laboratory, a gene named pveh1 encoding an EH from Phaseolus vulgaris was cloned. Although the directed modification of PvEH1 was carried out, the mutant PvEH1Y3 showed a limited degree of enantioconvergence towards racemic (rac-) m-chlorostyrene oxide (mCSO). RESULTS: PvEH1 and PvEH1Y3 were combinatively subjected to laboratory evolution to further enhance the enantioconvergence of PvEH1Y3 towards rac-mCSO. Firstly, the substrate-binding pocket of PvEH1 was identified using a CAVER 3.0 software, and divided into three zones. After all residues in zones 1 and 3 were subjected to leucine scanning, two E. coli transformants, E. coli/pveh1Y149L and /pveh1P184L, were selected, by which rac-mCSO was transformed into (R)-m-chlorophenyl-1,2-ethanediol (mCPED) having 55.1% and 27.2% eep. Secondly, two saturation mutagenesis libraries, E. coli/pveh1Y149X and /pveh1P184X (X: any one of 20 residues) were created at sites Y149 and P184 of PvEH1. Among all transformants, both E. coli/pveh1Y149L (65.8% αS and 55.1% eep) and /pveh1P184W (66.6% αS and 59.8% eep) possessed the highest enantioconvergences. Finally, the combinatorial mutagenesis was conducted by replacements of both Y149L and P184W in PvEH1Y3, constructing E. coli/pveh1Y3Z2, whose αS reached 97.5%, higher than that (75.3%) of E. coli/pveh1Y3. In addition, the enantioconvergent hydrolysis of 20 mM rac-mCSO was performed by E. coli/pveh1Y3Z2, giving (R)-mCPED with 95.2% eep and 97.2% yield. CONCLUSIONS: In summary, the enantioconvergence of PvEH1Y3Z2 was successfully improved by laboratory evolution, which was based on the study of substrate-binding pocket by leucine scanning. Our present work introduced an effective strategy for the directed modification of enantioconvergence of PvEH1.