Engineered biosynthesis of novel polyketides: regiospecific methylation of an unnatural substrate by the tcmO O-methyltransferase

Abstract

TcmO is an O-methyltransferase that methylates the C-8 hydroxyl to Tcm B3, a four-ring aromatic intermediate in the tetracenomycin biosynthetic pathway of Streptomyces glaucescens. The gene encoding this enzyme was expressed in Streptomyces coelicolor CH999 together with the actinorhodin polyketide synthase (PKS) gene cluster, which is responsible for the biosynthesis of 3,8-dihydroxy-methylanthraquinone carboxylic acid (DMAC) and its decarboxylated analog, aloesaponarin. The resulting recombinant strain produced approximately equal quantities of aloesaponarin and a new product but no DMAC. Spectroscopic analysis revealed that the novel polyketide was the 3-O-methylated analog of DMAC. An in vitro radioisotopic assay was developed for tcmO. The enzyme requires S-adenosylmethionine as a co-substrate. It has a Km of 3 microM and a kcat of 2.7 min-1 for DMAC. A series of monocyclic, bicyclic, and tricyclic aromatic compounds were also tested as candidate substrates in vitro. Remarkably, none was modified by tcmO within detectable limits of the assay. Together, these results highlight the interesting molecular recognition features of this enzyme. On one hand, there appears to be some flexibility in the number and structures of unreactive rings, since both Tcm and B3 and DMAC are good substrates. However, 6-methylsalicylic acid, a monocyclic analog of the reactive ring, is not recognized by the enzyme. Likewise, neither aloesaponarin (which only differs from DMAC in the reactive ring) nor carminic acid (which only differs in the distal nonreactive ring) is modified. Thus, the binding energy for the tcmO-catalyzed methyl transfer appears to involve significant contributions from both the aromaticity and the functionality of polycyclic substrates.