Enzymatic and structural study of domains and modules from erythromycin and picromycin polyketide synthases

Complex polyketide synthases (PKSs) are becoming attractive systems for combinatorial biosynthesis of complex molecules, not only because of numerous pharmacologically important polyketides, such as the broad-spectrum antibiotic erythromycin A and the immunosuppressant rapamycin, that are biosynthesized by this family of proteins, but also because of the linearity between the catalytic domains of PKSs and structure of polyketide products. Although combinatorial biosynthesis has yielded promising results, the further use of this approach is limited by the problem of low or no polyketide production from the recombinant PKSs, due to the fact that the downstream domains and modules are not able to process readily the unnatural polyketides. The work described herein attempts to solve such problem by studying the substrate specificities and catalytic mechanisms of the domains and modules from both erythromycin PKS and picromycin PKS.; The KR domain has been deleted from module 3 of erythromycin PKS and the resulting mutant is as active as the wild-type enzyme indicating that the protein structural integrity is generally preserved, thereby supporting the two PKS structural models that individual domains fold autonomously and the positions of core domains do not change in the absence of modification domains. Detailed kinetic and structural studies on the thioesterase domain from picromycin PKS have revealed a broad substrate tolerance by this enzyme, which can be used as a potential releasing enzyme in combinatorial biosynthesis, and also exposed a structural fold that is possibly also present in other PKS thioesterases. Experiments on the modules from picromycin PKS have shown interesting substrate specificities that are different from the substrate specificities of erythromycin PKS modules, thereby extending our understanding of PKS specificity and function.