Annabelle Gillig
Cancer cells exhibit different properties that make them dangerous to the host, such as the ability to invade other tissues and the abnormal response to the control mechanisms that regulate the division of normal cells, which allow them to divide in a relatively uncontrolled fashion. Moreover, cancer cells require fewer protein growth factors than normal cells and can go on dividing indefinitely, in contrast with normal. Thus, malignant tumors cells defy not only the normal controls on their proliferation, but also the normal controls on their position. The more widely such cells spread in the body, the harder they become to eradicate surgically. Therefore, cancer research has the difficult task of analyzing malignant transformation and metastatic diffusion in order to develop new therapies. In this context, we report in the present work the development of new chemical tools related to cancer research. The first part relies on the salvage pathway of nicotinamide adenine dinucleotide NAD+ by the enzyme nicotinamide phosphoribosyl transferase. The latter enzyme is over-expressed in cancer cells and a synthetic inhibitor, called FK866, shows efficient anti-cancer activity while having little toxicity to normal cells. However, FK866 has a poor bioavailability, which highlights the need to develop new drugs. In order to optimize the bioavailability and the interactions in the active site, we report in the present work the strategy for the design of a new class of potential nicotinamide phosphoribosyl transferase inhibitors. An efficient synthetic pathway involving the preparation of an iminoribitol, its substitution on the position 1 through a three stepone pot reaction, a cross-metathesis and an ester-amide exchange has been set up. A library of new glycomimetics analogous to FK866 has been prepared and their encouraging biological evaluation on cancer cells led to the selection of eight candidates for further biological studies. The second part of the project is based on the development of cancer vaccines. Carcinomas express a multitude of different carbohydrate antigens, such as the Thomsen-Friedenreich antigen, on plasma membrane. Interestingly, the changes of the glycopatterns during the malignant transformation generally result in shorter carbohydrate chains. In usual tissue, the Thomsen-Friedenreich antigen is masked and inaccessible to the immune system, but in various tumors, it is exposed and immunoreactive. A cluster of Thomsen-Friedenreich antigen administered as vaccines has proven to be a target for the immune system. Nevertheless, due to hydrolysis catalysed by glycosidases in vivo, these disaccharides conjugates are relatively short-lived in the blood stream. Our strategy to avoid hydrolytic degradation is the use of C-linked- glycosides. Thus, we report the synthesis of non-hydrolysable clusters of C-linked disaccharides analogous to the TF-antigen through the coupling of two glycosides, the construction of disaccharides analogous to the Thomsen-Friedenreich epitope and their clustering through a peptide synthesis strategy. Conjugation of these clusters to a carrier protein such as KLH will provide the expected cancer vaccines. The last part refers to an outstanding strategy in human therapy, which use oligonucleotide analogues to inhibit the genes expression at the level of translation. This strategy, called the antisense approach, is based on the introduction of an antisense oligonucleotide, complementary to the mRNA of interest, into the cell. Base-pairing with the targeted mRNA thus inhibits its translation into a protein. RNAs tend to fold into a variety of secondary structure such as hairpin loops. Therefore, the antisense oligonucleotide should have a restricted conformation in order to bind efficiently, and with high specificity, to the targeted nucleic acid structure. Thus, we report the synthesis of a conformationally restricted dinucleoside through the application of cross-metathesis in nucleic acid chemistry.
EPFL2010
