Semi-rational Directed Evolution Of P450 BM-3 For The Biosynthesis Of Indirubin

Engineering enzymes to meet special requirements of diverse industrial applications is an important task in protein engineering, because not all wild-type enzymes have the desired qualities for specific applications. In this paper, in order to create biocatalysts suitable for the biosynthesis of indirubin (a drug for chronic myelogenous leukemia as well as a novel anticancer agent), cytochrome P450 BM-3, a promising monooxygenase for biotechnological applications, was engineered by semi-rational directed evolution. One mutant primarily produced indirubin was created, whereas the parent enzyme preferentially formed indigo. The mutant producing indirubin with purity higher than any other reported enzymes may be a good starting point for the biosynthesis of indirubin. The reaction process and the molecular mechanism of indirubin formation were investigated. With molecular docking, the effects of the amino acid substitution on the structure and function of P450 BM-3 related to indole hydroxylation was discussed.On the ground of analysis of existing directed evolution researches about P450 BM-3, a semi-rational directed evolution method was designed to engineer the enzyme. That is, site-directed saturation mutagenesis was carried out at position D168, a non-active site but possibly relative to indirubin production. Screening of the resulting library was carried out in liquid micro-cultures of 96-deep-well plates. With the difference in color and composition of products, one mutant, D168W, was found. It predominantly produced indirubin and resulted in a red product.Characterization of the mutant was performed and its kinetic parameters and effects of pH as well as temperature on its activity and stability were determined. The catalytic efficiency of D168W was decreased by nearly fivefold compared with that of the parent, mainly because of its lowered affinity towards indole. So the mutation of D168W may have an adverse effect on the binding of indole into the enzyme, and decrease the catalytic efficiency. Furthermore, the coupling efficiency of the mutant was reduced.The reaction process of the mutant hydroxylated indole was investigated to acertain the pathway and mechanism of indirubin formation. It was found that the C-2 of indole was the dominant hydroxylation site and the resulting product was the overwhelming hydroxyindole intermediate. This suggested that the mutant may preferentially hydroxylate indole at the C-2. As to the formation of indirubin, it may come from the combination of indoxyl and isatin (or oxindole). But the substitution of D168W was adverse to the transformation of indole because only a part of indole was transformed.Molecular simulation and molecular docking were applied to probe the molecular mechanisms of the alteration of the regioselectivity of P450 BM-3 to indole. The molecular docking clearly demonstrated that the mutation changed the model of indole binding and orienting towards the active site of P450 BM-3, making the C-2 of the indole pyrrole ring closer to the heme iron of P450 BM-3 than the C-3. Thus, the mutation possibly shifted the hydroxylation preference of P450 BM-3 for indole from the C-3 to C-2, which may be responsible for the reversal of distribution of the product yield. Because D168 is far from the active site or substrate access channel, it is unlikely to directly act on indole binding. As it is located at a loop connecting two a-helices and one of them is a substrate recognition sequence and involved in defining the substrate channel, D168W substitution may alter the spatial orientation of the a-helix related to the substrate access channel, and therefore impact the recognition and orientation of indole.With the aim of biosynthesis of indirubin, a novel P450 BM-3 mutant capable of transforming indole principally into indirubin rather than indigo was created by saturation mutagenesis at the non-active site D168. This research not only provided a novel promising biocatalyst for the preparation of indirubin, but shed some light on the regioselectivity and the corresponding structure-function relationships of P450 BM-3.