This review summarizes how ultrahigh-throughput screening methods employ cells and biomimetic compartments to access the vast, unexplored diversity of biocatalysts with novel functions derived from directed evolution and metagenomics libraries.

The self-sufficient cytochrome P450 BM-3 enzyme from Bacillus megaterium catalyzes subterminal hydroxylation of saturated long-chain fatty acids and structurally related compounds. Since the primary structure of P450 BM-3 is homologous to that of mammalian P450 type II, it represents an excellent model for this family of enzymes. During studies on the directed evolution of P450 BM-3 into a medium-chain fatty-acid hydroxylase, several mutants, in particular the triple mutant Phe87Val, Leu188Gln, Ala74Gly, were observed to hydroxylate indole, producing indigo and indirubin at a catalytic efficiency of 1365 M(-1)s(-1) (kcat=2.73 s(-1) and Km=2.0 mM). Both products were unequivocally characterized by NMR and MS analysis. Wild-type P450 BM-3 is incapable to hydroxylate indole. These results demonstrate that an enzyme can be engineered to catalyze the transformation of substrates with structures widely divergent from those of its native substrate.