Structural and Functional Analysis of the Thioesterase Domain of Human Fatty Acid Synthase

Abstract

Human fatty acid synthase (FAS) has recently become one of the most therapeutically relevant targets in the cell. Because of the strong correlation between the elevated expression levels of FAS and most aggressive carcinomas, FAS harbors extreme potential for the treatment of a broad range of human cancers. Recent advances with pharmacological inhibition of FAS have shown promising anti-tumor activity. The FDA-approved drug Orlistat was discovered as a potent inhibitor of the thioesterase activity, which is the seventh functional domain of the multienzyme FAS assembly. Responsible for the highly selective hydrolysis of palmitate (C16) and stearate (C18), thus termination of the fatty acid biosynthetic cycle, inhibition of thioesterase by Orlistat induces tumor cell-specific apoptosis, inhibits the growth of prostate tumors and prevents angiogenesis. This prompted the recent crystal structure of the thioesterase domain inhibited by Orlistat. These complexes confirmed the mechanism of Orlistat mediated inhibition and characterized the active site ‘molecular landscape.’ Thus, providing a foundation for the development of novel anti-tumor agents with the potential to translate into the clinical oncology setting. Futhermore, the binding modes of the Orlistat molecules have enabled a model to be proposed for chain-length selectivity of the thioesterase during the FAS catalytic cycle. These data have driven the design of additional experiments to test hypotheses concerning the mechanism of chain-length selectivity and the biochemistry of ACP mediated substrate delivery and potential domain interactions.

In this regard, considerable effort has been dedicated to understanding the highly selective mechanism of C16 and C18 chain-length hydrolysis catalyzed by the thioesterase domain. We have solved to high-resolution three crystal structures of the thioesterase domain in complex with the intact palmitoyl-coenzyme A (CoA) substrate, the double products (CoASH and palmitate), or the single palmitoyl-product. The phosphopantetheinyl arm interacts with a prominent channel suggesting a mechanism for the efficient delivery of acyl-substrates. Moreover, the binding mode of the C16 acyl-chains bound to the hydrophobic ‘specificity channel’ and not the ‘interface cavity’ supports a revised model for the discrimination of acyl-chain length during the FAS catalytic cycle. Mutagenesis designed to fill the interface cavity retained chain-length selectivity nearly identical to wild-type, whereas truncating or lengthening the specificity channel either completely abolished specificity toward C16-18 substrates shifting selectivity toward short- to medium-chain fatty acids (C8-12) or resulted in a notable trend toward longer-chain (≥C18) specificity, respectively. Our observations also provide a foundation for understanding the potential domain interactions through molecular modeling of the docked ACP onto the thioesterase.