Sialyltransferases (STs), which catalyze the sialylation reaction by adding sialic acids to the terminal positions of oligosaccharide of glycoproteins and glycolipids, are over-expressed in cancer cells and associated with cancer metastasis. Until now, ST inhibitors are not applicable for clinical use because of poor cell permeability, although showing potent effect in vitro. In this study, we synthesize a lithocholic acid-based ST inhibitor AL10 and test its anti-metastatic effect. Overexpression of £\-2,3-ST is found in highly metastatic A549 and CL1-5 lung cancer cells. Confocal microscopy demonstrates that AL10 is cell permeable and may attenuate total sialylation on cell surface. AL10 has no cytotoxicity but inhibits adhesion, migration, actin polymerization and invasion of A549 and CL1-5 cells in vitro. Inhibition of adhesion and migration by AL10 is associated with reduced sialylation of beta1 integrin. In addition, activation of the beta1 integrin downstream signaling molecule focal adhesion kinase is also attenuated. More importantly, AL10 suppresses lung metastasis in vivo and this effect may be linked with reduced sialylation of the chemokine receptor CXCR4 which has been found to play a critical role in organ-specific metastasis. Serum biochemical assay indicates that AL10 does not affect liver and kidney functions of experimental animals. Taken together, we conclude that AL10 is an effective sialyltransferase inhibitor and exerts anti-metastatic effect in vivo via suppression of sialylation of beta1 integrin and CXCR4.

Sialylation is an important carbohydrate modification of glycoconjugates, which introduces sialic acids (SA). The relatively large nine-carbon, negatively charged sugars are typically located at the termini of carbohydrate chains. SA's are often required for functionally important molecular and cellular interactions including virus-host interactions, tumor progression and malignancy, immune system development and function, and nervous system development and function. However, the study of sialylation in vertebrates, including man, encounters serious obstacles associated with the complexity of vertebrates' biology and limitations of available experimental approaches. Drosophila is a useful model system with many advantages including quick generation time, a large number of progeny, simplified glycosylation and neurophysiology, and ease of genetic manipulations. The primary focus of this thesis is on the functions of Drosophila melanogaster CMP sialic acid synthetase (DmCSAS) and sialyltransferase (DSiaT) in the central nervous system (CNS). A combination of genetic, immunostaining, and neurobiology approaches were used to characterize the functions of DmCSAS and DSiaT in Drosophila. This investigation revealed the expression of DmCSAS and suggested that it plays an important role in a specialized and developmentally regulated process in the nervous system of Drosophila. Further experiments examined sub-cellular localization of DmCSAS revealing that this protein has a complex mostly Golgi-associated distribution within the cell in vivo. I discovered a novel link between Drosophila sialylation and circadian rhythm regulation. I also characterized the electrophysiological phenotypes of DmCSAS mutants and compared them to the corresponding defects associated with DSiaT mutations. My experiments also revealed that the relationship between DmCSAS and DSiaT are more complex than originally thought; these genes may have independent functions while also participating in the same pathway. Taken together, these results elucidate the sialylation pathway in Drosophila and shed more light on the role of sialylation in the nervous system. My experiments provide a unique evolutionary perspective on the sialylation pathway in animals and suggest that the neural function of SA in Drosophila can be conserved in vertebrates, including humans.