In vitro studies of the enzymes involved in fluorometabolite biosynthesis in Streptomyces cattleya

Enzymatic fluorination of natural products is extremely rare. Of the 4000 halogenated natural products identified, only 13 possess a fluorine atom. The C-F bond forming enzyme from the soil bacterium, Streptomyces cattleya, remains the only native enzyme to be identified that is capable of such biochemistry. It generates 5’-fluoro-5-deoxyadenosine (5‘-FDA) from S-adenosyl-L-methionine (SAM) and F-. The “fluorinase” is the first committed step toward the biosynthesis of the two fluorometabolites, 4-fluorothreonine and fluoroacetate, via the common intermediate, fluoroacetaldehyde (FAld). The enzymatic steps responsible for the conversion of 5’-FDA to the fluorometabolites remained to be fully characterised when this project began. Previously, a purine nucleoside phosphorylase was identified that was capable of generating 5-fluorodeoxyribose-1-phosphate (5-FDRP) from 5’-FDA. 5-FDRP is subsequently isomerised to 5-fluorodeoxyribulose-1-phosphate (5-FDRulP) by an aldose-ketose isomerase enzyme. Chapter 2 describes the identification of the isomerase gene from the genomic DNA of S. cattleya and the corresponding protein product was capable of generating 5-FDRulP from 5-FDRP. The next intermediate, FAld, is generated from 5-FDRulP by a fuculose aldolase. Attempts to identify the aldolase gene from S. cattleya were unsuccessful, however a putative fuculose aldolase from Streptomyces coelicolor was isolated that could generate FAld from 5-FDRulP, which is described in Chapter 3. Following the identification and over expression of a PLP-dependant transaldolase, which generates 4-fluorothreonine (4-FT) from FAld and L-threonine in S. cattleya, Chapter 4 details the successful in vitro reconstitution of fluorometabolite biosynthesis using five over- expressed enzymes. In Chapter 5, attempts to develop a novel assay for fluorinase activity was explored. The colorimetric detection of L-methionine produced by the fluorinase in a coupled L-amino acid oxidase and horseradish peroxidase assay, leading to the oxidation of a dye substance. This was carried out with interest in developing a high-throughput assay for fluorinase mutants, generated by random mutagenesis, in order to identify those with increased activity. In the event, it proved unsuccessful