Molecular basis for enantioselectivity in the (R)- and (S)-hydroxypropylthioethanesulfonate dehydrogenases, a unique pair of stereoselective short-chain dehydrogenases/reductases involved in aliphatic epoxide carboxylation

Abstract

(R)- and (S)-2-hydroxypropyl-CoM (R-HPC and S-HPC) are produced as intermediates in bacterial propylene metabolism from the nucleophilic addition of coenzyme M to (R)- and (S)-epoxypropane, respectively. Two highly enantioselective dehydrogenases (R-HPCDH and S-HPCDH) belonging to the short-chain dehydrogenase/reductase family catalyze the conversion of R-HPC and S-HPC to 2-ketopropyl-CoM (2-KPC), which undergoes reductive cleavage and carboxylation to produce acetoacetate. In the present study, one of three copies of S-HPCDH enzymes present on a linear megaplasmid in Xanthobacter autotrophicus strain Py2 has been cloned and overexpressed, allowing the first detailed side by side characterization of the R-HPCDH and S-HPCDH enzymes. The catalytic triad of S-HPCDH was found to consist of Y156, K160, and S143. R211 and K214 were identified as the amino acid residues coordinating the sulfonate of CoM in S-HPC. R211A and K214A mutants were severely impaired in the oxidation of S-HPC or reduction of 2-KPC but were largely unaffected in the oxidation and reduction of aliphatic alcohols and ketones. Kinetic analyses using R- and S-HPC as substrates revealed that enantioselectivity in R-HPCDH (value, 944) was dictated largely by differences in k(cat) while enantioselectivity for S-HPCDH (value, 1315) was dictated largely by changes in K(m). S-HPCDH had an inherent high enantioselectivity for producing (S)-2-butanol from 2-butanone that was unaffected by modulators that interact with the sulfonate binding site. The tertiary alcohol 2-methyl-2-hydroxypropyl-CoM (M-HPC) was a competitive inhibitor of R-HPCDH-catalyzed R-HPC oxidation, with a K(is) similar to the K(m) for R-HPC, but was not an inhibitor of S-HPCDH. The primary alcohol 2-hydroxyethyl-CoM was a substrate for both R-HPCDH and S-HPCDH with identical K(m) values. The pH dependence of kinetic parameters suggests that the hydroxyl group is a larger contributor to S-HPC binding to S-HPCDH than for R-HPC binding to R-HPCDH. It is proposed that active site constraints within the S-HPCDH prevent proper binding of R-HPC and M-HPC due to steric clashes with the improperly aligned methyl group on the C2 carbon, resulting in a different mechanism for controlling substrate specificity and enantioselectivity than present in the R-HPCDH.

Figure 1

Pathway of propylene oxidation in X. autotrophicus Py2. The reactions catalyzed by R-HPCDH and S-HPCDH are shown in the boxed region.

Figure 2

Multiple-sequence alignment of S-HPCDH1, S-HPCDH3 and R-HPCDH1 enzymes from X. autotrophicus Py2. Abbreviations: Cons.S1&3, consensus amino acid alignment for S-HPCDH1 and S-HPCDH3; Cons.all, consensus of S-HPCDH1, S-HPCDH3 and R-HPCDH1 Letter designations: a, Classic GXXXGXG glycine-rich NAD⁺ binding motif. b, Catalytic tetrad residues of Asn, Ser, Tyr and Lys. c, Positively charged residues that have been shown (R-HPCDH1) or are proposed (S-HPCDH) to interact with the sulfonate group of CoM in the substrate. The alignment was generated using MULTALIN with default parameters, while the consensus was derived using ClustalW2. The following symbols mean that the residues are: (*) identical; (:) conserved; (.) semi-conserved.

Figure 3

Changes of kinetic parameters with pH for rS-HPCDH3 catalyzed oxidation of S-HPC. (A) k_cat vs. pH, (B) k_cat/K_mS-HPC vs. pH, (C) K_mS-HPC vs. pH, represented in log scale. The plots in (A) and (C) are shown as simple line plots. The line in plot (B) was generated by a fit of the experimental data to equation 1.

Figure 4

Superimposed active sites of R-HPCDH1 and S-HPCDH3 based on the crystal structure of R-HPCDH1 and a homology model of S-HPCDH3. The cartoon structures and carbon atoms of amino acid residues of R-HPCDH1 (pdb ID 2cfc) and the homology model for S-HPCDH3 are colored grey and green, respectively. NAD⁺ is shown in magenta. R-HPC and S-HPC were modeled using the crystal structure for S-HPC bound at the active site of R-HPCDH1 as described previously (16). In both views, R-HPC (grey carbon atoms) and S-HPC (green carbon atoms) are modeled at the active sites such that the positions of the hydroxyl group and hydrogen atom occupy the same positions. The methyl groups and the methylene groups linking the hydroxypropyl groups to CoM are overlayed on top of each other to highlight the different spatial orientations of these groups in R- and S-HPC. Panel A, Superimposed structures highlighting the interactions of substrates with the catalytic triads. Panel B, Superimposed structures highlighting the interactions of substrates with the amino acids that coordinate the sulfonate of CoM.

Figure 5

Effects of 2-(2-methyl-2-hydroxypropylthio)ethanesulfonate (M-HPC) on R- and S-HPC oxidation by R-HPCDH1, S-HPCDH3, and S-HPCDH1. Panel A, Competitive inhibition of R-HPCDH1-catalyzed R-HPC oxidation by M-HPC. The double reciprocal plots for assays performed in the presence of different concentrations of M-HPC are shown in the main diagram. Data points represent the average of triplicate experiments. The solid lines were generated by nonlinear least-square fits of the ν vs. S data, shown in the inset, to the equation for a rectangular hyperbola using Sigmaplot. M-HPC concentrations: (●) 0 mM, (○) 0.2 mM, (▼) 0.4 mM, (△) 0.8 mM, (■) 1.6 mM. Panels B and C, ν vs. S plots for S-HPC oxidation by S-HPCDH3 and S-HPCDH1, respectively, in the presence of different concentrations of M-HPC. The lines were generated by fitting the data to the standard form of the Michaelis-Menten equation. M-HPC concentrations: (●) 0 mM, (○) 1.2 mM, (▼) 4.9 mM.

Scheme 1

Structures