UC Berkeley

Fluorination has become a very useful tool in the design and optimization of bioactive small molecules ranging from pesticides to pharmaceuticals. Its small size allows a sterically conservative substitution for a hydrogen or hydroxyl, thus maintaining the overall size and shape of a molecule. However, the extreme electronegativity of fluorine can dramatically alter other properties of the molecule. As a result, the development of new methods for fluorine incorporation is currently a major focus in synthetic chemistry. It is our goal to use a complementary biosynthetic approach to use enzymes for the regio-selective incorporation of fluorine into complex natural product scaffolds through the fluoroacetate building block.

Towards this goal, we initiated a study of the only known genetic host of a carbon-fluorine bond forming enzyme, the fluoroacetate- and fluorothreonine-producing bacterium Streptomyces cattleya. Sequencing and analysis of the genome identified several paralogs of enzymes predicted to be on the fluoroacetate and fluorothreonine biosynthetic pathway and whose function were probed by in vitro biochemistry and genetic knock-out studies. We sought to further explore how S. cattleya manages fluoroacetate, which is a potential biosynthetic building block but is also a potent toxin that operates by shutting down the tricarboxylic acid (TCA) cycle. Coordinated transcriptional changes of genes involved in central metabolism and organofluorine metabolism suggest transcriptional control may serve as the major mechanism for management of fluoroacetate toxicity. This hypothesis is further supported by biochemical analysis of S. cattleya enzymes involved in fluoroacetate toxicity, which showed that they were no more selective against fluorine than orthologs from non-producing bacteria.

To explore the possibility of incorporating fluorinated building blocks into more complex natural products, we turned to type I polyketide synthases as their modular assembly line nature would make them ideal candidates for the engineering of site-selective incorporation of fluorinated subunits. We first developed methods to enzymatically synthesize fluoromalonyl-CoA, which is a fluorinated congener of the malonyl-CoA extender unit. With this building block in hand, we were able to observe incorporation of fluorine into a triketide polyketide. We then showed that it is possible to site-selectively incorporate fluorine into tetraketide products. These results suggest that the production of complex fluorinated natural products is possible and may allow us to explore the medicinal chemistry of these compounds using site-selective fluorination.