The crystal structures of the haem domains of Ala330Pro and Ile401Pro, two single-site proline variants of CYP102A1 (P450(BM3)) from Bacillus megaterium, have been solved. In the A330P structure, the active site is constricted by the relocation of the Pro329 side chain into the substrate access channel, providing a basis for the distinctive C-H bond oxidation profiles given by the variant and the enhanced activity with small molecules. I401P, which is exceptionally active towards non-natural substrates, displays a number of structural similarities to substrate-bound forms of the wild-type enzyme, notably an off-axial water ligand, a drop in the proximal loop, and the positioning of two I-helix residues, Gly265 and His266, the reorientation of which prevents the formation of several intrahelical hydrogen bonds. Second-generation I401P variants gave high in vitro oxidation rates with non-natural substrates as varied as fluorene and propane, towards which the wild-type enzyme is essentially inactive. The substrate-free I401P haem domain had a reduction potential slightly more oxidising than the palmitate-bound wild-type haem domain, and a first electron transfer rate that was about 10 % faster. The electronic properties of A330P were, by contrast, similar to those of the substrate-free wild-type enzyme.

There is intense interest in late-stage catalytic C-H bond functionalization as an integral part of synthesis. Effective catalysts must have a broad substrate range and tolerate diverse functional groups. Drug molecules provide a good test of these attributes of a catalyst. A library of P450BM3 mutants developed from four base mutants with high activity for hydrocarbon oxidation produced human metabolites of a panel of drugs that included neutral (chlorzoxazone, testosterone), cationic (amitriptyline, lidocaine) and anionic (diclofenac, naproxen) compounds. No single mutant was active for all the tested drugs but multiple variants in the library showed high activity with each compound. The high conversions enabled full product characterization that led to the discovery of the new P450 reaction type of oxidative decarboxylation of an α-hydroxy carboxylic acid and the formation a protected imine from an amine, offering a novel route to α-functionalization of amines. The substrate range and varied product profiles suggest that this library of enzymes is a good basis for developing late-stage C-H activation catalysts.

The power of proline: Bold amino acid substitutions in sensitive protein regions are frequently unproductive, while more subtle mutations can be sufficient to bring about dramatic changes. But introducing proline at the residue next to the sulfur ligand in P450(BM3) (CYP102A1) has the unexpected and desirable effect of enhancing the activity of this fatty acid hydroxylase with a broad range of non-natural substrates, as illustrated by the figure.

The substrate-free crystal structure of a five-mutation directed evolution variant of CYP102A1 (P450(BM3)) with generic activity-enhancing properties ("KT2") has been determined to 1.9-Å resolution. There is a close resemblance to substrate-bound structures of the wild-type enzyme (WT). The disruption of two salt bridges that link the G- and I-helices in WT causes conformational changes that break several hydrogen bonds and reduce the angle of the kink in the I-helix where dioxygen activation is thought to take place. The side-chain of a key active site residue, Phe87, is rotated in one molecule of the asymmetric unit, and the side-chains of Phe158 and Phe261 cascade into the orientations found in fatty-acid-bound forms of the enzyme. The iron is out of the porphyrin plane, towards the proximal cysteine. Unusually, the axial water ligand to the haem iron is not hydrogen-bonded to Ala264. The first electron transfer from the reductase domain to the haem domain of substrate-free KT2 is almost as fast as in palmitate-bound WT even though the reduction potential of the haem domain is only slightly more oxidising than that of substrate-free WT. However, NADPH is turned over slowly in the absence of substrate, so the catalytic cycle is gated by a step subsequent to the first electron transfer-a contrast to WT. Propylbenzene binding slightly raises the first electron transfer rate in WT but not in KT2. It is proposed that the generic rate accelerating properties of KT2 arise from the substrate-free form being in a catalytically ready conformation, such that substrate-induced changes to the structure play a less significant role in promoting the first electron transfer than in WT.

The new agreement specifically addresses what authors can do with different versions of their manuscripte.g. use in theses and collections, teaching and training, conference presentations, sharing with colleagues, and posting on websites and repositories. The terms under which these uses can occur are clearly identified to prevent misunderstandings that could jeopardize final publication of a manuscript (Section II, Permitted Uses by Authors).