Cancer Epigenetics: New Drugs and Paradigms
Conference Recording Feb 20, 2013
About the Speaker
Frank Lyko is a Professor of Epigenetics at the University of Heidelberg and heads the Division of Epigenetics at the German Cancer Research Center in Heidelberg, Germany. He received his PhD from the University of Heidelberg in 1998. After postdoctoral work with Rudolf Jaenisch at the Whitehead Institute for Biomedical Research in Cambridge (USA) he joined the German Cancer Research Center as a group leader in 2001. His current research activities focus on the role of DNA methylation in phenotypic plasticity, the analysis of RNA methylation and the development of epigenetic cancer therapies.AbstractHypermethylation of tumor suppressor genes has been described in many tumors and represents an attractive target for cancer therapy. The DNA methyltransferase inhibitor 5-azacytidine represents an archetypal epigenetic drug that has found increasing use in the treatment of myeloid leukemias. However, clinical resistance to 5-azacytidine is common and has been linked to several pathways. Using cellular transport assays, we have identified the human equilibrative nucleoside transporter (hENT) system as a major factor for 5-azacytidine uptake in human cancer cells. Reduction of hENT transport activity induced pronounced resistance to 5-azacytidine and strongly reduced the DNA demethylating activity of the drug. This resistance could be overcome by CP-4200, a novel hENT-independent derivative of 5-azacytidine. Importantly, epigenetically silenced tumor suppressor genes that could not be reactivated by 5-azacytidine under hENT-deficient conditions, showed robust gene activation with CP-4200. Our findings define an important paradigm for 5-azacytidine tolerance and establish CP-4200 as a valuable tool for the modulation of epigenetic drug resistances. Genetic mutations in the DNA methyltransferase 3A (DNMT3A) gene have been frequently described in myeloid leukemias, but their functional significance has remained unclear. In the mouse, loss of DNMT3A has been shown to promote lung tumorigenesis, which provided an important model for further analysis. We have now used whole-genome bisulfite sequencing to characterize the genome-wide methylation patterns from various tumors of this model. The results show distinct similarities between human and mouse tumor methylomes and also demonstrate that the deletion of DNMT3A causes a pronounced spreading of tumor-specific hypomethylated domains. A detailed overview of our observations will be provided and implications for epigenetic drug dicovery will be discussed.