Evolution of efficient p450 biocatalysts for human drug metabolite production

Human cytochromes P450 (P450s) are essential enzymes in metabolism of drugs/xenobiotics entering the body, and for generation of numerous steroids and lipid mediators. P450s catalyse oxidative metabolism of their substrates, often with exquisite regio- and stereoselectivity of oxidation. There is great interest in application of P450s for production of oxyfunctionalized molecules as e.g. intermediates in chemical synthesis or high-value lipid mediators with specific biological functions (e.g. derivatives of arachidonic acid), and these molecules are often extremely difficult, inefficient and non-cost-effective to produce by organic synthesis. Of particular value in the human P450 field would be a ready source of key metabolites produced from a cluster of the P450s known to be most important for metabolism of prescribed drugs and pharmaceuticals (specifically CYP3A43, CYP2D6, CYP2C9, CYP2C19 and CYP1A2, which collectively oxidise and metabolise >90 % of administered drugs). These metabolites would be of great value in diagnostics for human P450 function (i.e. as standards for human P450 and hepatic metabolism) and as reagents that will e.g. help define variant/polymorphic P450 activity (and hence likelihood of adverse drug reactions) and enable quantification of key metabolite production. The problem is that the relevant human P450s and their diflavin P450 reductase partner are integral membrane proteins, are unstable and have low activities. Hence yields of relevant products are very small and a commercial process that exploits these enzymes is not financially viable. Instead, we will use the highest activity P450 in the entire P450 superfamily (P450 BM3 from Bacillus megaterium) for this purpose, and evolve BM3 to catalyse specific functions by a combination of rational mutagenesis and directed evolution, exploiting robotic facilities for screening provided already by BBSRC funding. The major reasons for exploiting BM3 are that a) it is a natural and soluble fusion of a P450 to its P450 reductase, with catalytic activity typically ~1000-fold that of the multi-component human enzymes, and b) it has already been shown (through work in our own labs and elsewhere) that the active site of the enzyme is pliable and can be engineered to bind and oxidise molecules quite different from its 'natural' fatty acid substrates (e.g. aromatics and polycyclic aromatics, short chain alkenes and alkanes). The student will undertake a combination of rational mutagenesis and directed evolution (with rational targets including Phe87, Val78, Ala264 and Leu181, which we have shown critical to control of substrate selectivity) in order to diversify reactivity of BM3 such that it will oxidise prototypical substrates of the aforementioned human P450s: testosterone (3A4), dextramethorphan (2D6), diclofenac (2C9), diazepam (2C19) and phenacetin (1A2). Our work has already shown that BM3 oxidises testosterone (but not at correct carbon for wild-type BM3), and all other human P450 substrates targeted have molecular dimensions and chemical characteristics similar to other compounds oxidised by various BM3 mutants. We will screen for activities towards new substrates using robotics and E. coli cell extracts expressing BM3 mutants, and oxo-plate technology whereby dye-linked oxygen consumption is measured. Point mutants that enhance activity with key substrates will be combined with variants randomly evolved in regions known to control substrate selectivity (in I helix, F/G and B/C loop regions) and improved mutants identified by further screening. Organic products will be characterized by GCMS/HPLC methods, and optimised enzymes purified and characterized enzymatically and structurally (in ligand-free and substrate-bound forms). The student will be supported by a PDRA working on a parallel P450 evolution project, and will develop novel catalysts for generation of important molecules to be exploited as reagents and diagnostics for commercial exploitation.