Cofactor engineering of Lactobacillus brevis alcohol dehydrogenase by computational design

The R-specific alcohol dehydrogenase from Lactobacillus brevis (Lb-ADH) catalyzes the enantioselective reduction of prochiral ketones to the corresponding secondary alcohols. It is stable and has broad substrate specificity. These features make this enzyme an attractive candidate for biotechnological applications. A drawback is its preference for NADP(H) as a cofactor, which is more expensive and labile than NAD(H). Structure-based computational protein engineering was used to predict mutations to alter the cofactor specificity of Lb-ADH. Mutations were introduced into Lb-ADH and tested against the substrate acetophenone, with either NAD(H) or NADP(H) as cofactor. The mutant Arg38Pro showed fourfold increased activity with acetophenone and NAD(H) relative to the wild type. Both Arg38Pro and wild type exhibit a pH optimum of 5.5 with NAD(H) as cofactor, significantly more acidic than with NADP(H). These and related Lb-ADH mutants may prove useful for the green synthesis of pharmaceutical precursors.