Ser67Asp and His68Asp substitutions in candida parapsilosis carbonyl reductase alter the coenzyme specificity and enantioselectivity of ketone reduction

Abstract

A short-chain carbonyl reductase (SCR) from Candida parapsilosis catalyzes an anti-Prelog reduction of 2-hydroxyacetophenone to (S)-1-phenyl-1,2-ethanediol (PED) and exhibits coenzyme specificity for NADPH over NADH. By using site-directed mutagenesis, the mutants were designed with different combinations of Ser67Asp, His68Asp, and Pro69Asp substitutions inside or adjacent to the coenzyme binding pocket. All mutations caused a significant shift of enantioselectivity toward the (R)-configuration during 2-hydroxyacetophenone reduction. The S67D/H68D mutant produced (R)-PED with high optical purity and yield in the NADH-linked reaction. By kinetic analysis, the S67D/H68D mutant resulted in a nearly 10-fold increase and a 20-fold decrease in the k(cat)/K(m) value when NADH and NADPH were used as the cofactors, respectively, but maintaining a k(cat) value essentially the same with respect to wild-type SCR. The ratio of K(d) (dissociation constant) values between NADH and NADPH for the S67D/H68D mutant and SCR were 0.28 and 1.9 respectively, which indicates that the S67D/H68D mutant has a stronger preference for NADH and weaker binding for NADPH. Moreover, the S67D/H68D enzyme exhibited a secondary structure and melting temperature similar to the wild-type form. It was also found that NADH provided maximal protection against thermal and urea denaturation for S67D/H68D, in contrast to the effective protection by NADP(H) for the wild-type enzyme. Thus, the double point mutation S67D/H68D successfully converted the coenzyme specificity of SCR from NADP(H) to NAD(H) as well as the product enantioselectivity without disturbing enzyme stability. This work provides a protein engineering approach to modify the coenzyme specificity and enantioselectivity of ketone reduction for short-chain reductases.

FIG. 1.

Partial structural alignment of the SCR with selected members of the SDR family. Left columns contain the Protein Data Bank accession codes of the structures. 3CTM, (S)-1-phenyl-1,2-ethanediol dehydrogenase from Candida parapsilosis; 1H5Q, mannitol-2-dehydrogenase from Agaricus bisporus; 1ZZE, carbonyl reductase from Sporobolomyces salmonicolor; 1CYD, mouse lung carbonyl reductase; and 1A4U, Drosophila lebanonensis alcohol dehydrogenase. Conserved residues are boxed with blue lines. Selected residue numbers of the SCR are labeled above the sequence. Secondary structure elements of the SCR are marked at the top of the alignment, and the glycine-rich consensus sequence is indicated. The residues (Arg43 and Ser44 in 1H5Q and Arg44 and Ser45 in 1ZZE) in contact with the pyrophosphate bridge of NADPH-dependent SDRs is boxed in black. The Thr38 residue in 1CYD corresponding to the hydrogen bonds of the 2′ phosphate of NADPH is highlighted in green, and the Thr38Asp substitution alters the 1CYD preference from NAD(P)H to NAD(H). The Asp37 in 1A4U responsible for NAD(H) specificity is marked with a filled circle. The three residues for mutations to convert coenzyme specificity in 3CTM are marked with solid triangles. This figure was prepared with the program Espript (http://espript.ibcp.fr/ESPript/cgi-bin/ESPript.cgi/).

FIG. 2.

Asymmetric reduction of 2-hydroxyacetophenone using purified SCR and S67D/H68D proteins. (A) Retention times of standard samples are as follows: (R)-PED, 15.0 min; (S)-PED, 18.3 min; 2-hydroxyacetophenone, 27.1 min. (B) Reaction products catalyzed by the SCR in the presence of NADPH. (C) Reaction products catalyzed by the S67D/H68D protein in the presence of NADH. AU, arbitrary units.

FIG. 3.

CD spectra (A) and thermal denaturation (B) of the SCR and the S67D/H68D variant. The CD spectra were recorded by measuring the ellipticity as a function of wavelength at 0.1-nm increments between 195 and 260 nm at 20°C. The thermal denaturation was determined by measuring the ellipticity at 209 nm as a function of temperature (T) at increments of 2 or 5°C between 20 and 70°C.

FIG. 4.

Effects of coenzymes on thermal inactivation of the WT (A) and the S67D/H68D mutant (B). The WT and S67D/H68D mutant enzymes were incubated at 40°C in the presence or absence of NAD(P)H.

FIG. 5.

Effects of coenzymes on urea denaturation of the WT (A) or the S67D/H68D mutant (B) at pH 8.0. The WT and S67D/H68D mutant enzymes were incubated at 25°C in the presence or absence of NAD(P)H.