Enhancing the biocatalytic manufacture of the key intermediate of atorvastatin by focused directed evolution of halohydrin dehalogenase

摘要

Halohydrin dehalogenases (HHDHs, EC 4.5.1.X) are enzymes that naturally catalyze dehalogenation of halohydrins to produce the corresponding epoxides, often with high stereoselectivity. In the reverse reaction, different nucleophiles than halides (i.e., N3-, CN-, and NO2-) are accepted to open the epoxide ring, thus making possible C-N, C-C, or C-O bonds (1). Consequently, HHDHs have been used for the stereoselective preparation of a number of β-substituted alcohols. The most important application involves the production of ethyl (R)-4-cyano-3-hydroxybutyrate from ethyl (S)-4-chloro-3-hydroxybutyrate via the corresponding (S)-epoxide; since ethyl (R)-4-cyano-3-hydroxybutyrate is the side-chain precursor in the synthesis of atorvastatin and other statins (2). HHDHscan be employed both in the so-called KRED route, where the two stereocenters are created by asymmetric induction of ketoreductases able to catalyze the hydrogenation of two carbonyls (Figure 1). An halohydrin dehalogenase from Agrobacterium radiobacter AD1 (named HheC) proved the most suited for the preparation of ethyl (R)-4-cyano- 3-hydroxybutanoate and evolved (re-modelled) for improving cyanolytic activity until getting a mutant (HheC236) with robustness (3) and efficiency appropriate for industrial purpose. The KRED route encompasses a subsequent Claisen condensation, which needs tough conditions. Therefore, another chemoenzymatic strategy was envisaged, the DERA route, where the overall stereochemistry derives from the action of an aldolase able to simultaneously form two C-C bonds and cyanolysis occurs on tert-butyl (3R,5S)-6-chloro-3,5-dihydroxyhexanoate; the enzymatic reaction was needed since under standard alkaline conditions used for chemical cyanolysis, ester hydrolysis is a major side-reaction (Figure 2). In this paper, HheC was modified for preparing a biocatalyst able to perform totally chemoselective cyanolysis of tert-butyl (3R,5S)-6-chloro-3,5-dihydroxyhexanoate under mild condition of pH. To have a solid model accounting for enzyme-ligands interactions, HheC was crystallyzed in the presence of tert-butyl (3R,5R)-6-cyano-3,5-dihydroxyhexanoate, the product of the desired cyanolysis; six residues were identified based on the potential effects on enzymatic activity. Combinational active-site saturation testing (CASTing) was then applied (4) for enhancing the performance of the biocatalyst; a major limitation in the development of directed evolution techniques, even if oriented like CASTing, is the number of variants to be screened, in this case the authors used a high throughput screening based on a colorimetric reaction among the chloride ions, formed after dehalogenation, and (iron(III)/thiocyanate). Variant V84G/W86F, gave 15-fold higher activity than the original protein, making this mutant an efficient biocatalyst for the practical production of atorvastatin intermediate.