DNMTs et cancer : conception d'inhibiteurs sélectifs comme outils chimiques moléculaires en pharmacologie épigénétique

DNA methylation, primarily regulated by the maintenance methyltransferase DNMT1 and the de novo methyltransferases DNMT3A and DNMT3B, represents a critical epigenetic modification that governs gene expression and is frequently dysregulated in cancer. Widespread promoter hypermethylation and aberrant overexpression of DNA methyltransferases (DNMTs) are commonly observed in various malignancies, including pancreatic, breast, colorectal, and lung cancers, as well as in leukemias. Moreover, accumulating evidence suggests that DNMT1, traditionally regarded as a maintenance methyltransferase, can also contribute to de novo methylation events. This thesis is part of a medicinal chemistry program devoted to the rational design, synthesis and biological evaluation of DNMT inhibitors, with the twin aims of delivering pharmacological tools and uncovering new therapeutic leads. Chapter 1 reviews the fundamentals of DNA methylation, outlines the catalytic machinery of DNA methyltransferases and surveys the current landscape of isoform-selective DNMT1 and DNMT3 inhibitors. Chapter 2 details our structure-guided design strategy. We have designed DNMTs inhibitors using a structure-based approach. This strategy exploits the shared catalytic mechanism of DNMTs - methyl transfer from S-adenosyl-L-methionine (SAM) to cytosine - while targeting structural differences between DNMT1 and DNMT3. A key distinction is the valine-to-tryptophan substitution (DNMT1 Val1580 / DNMT3A Trp893) located between the substrate and cofactor-binding sites, which offers a unique opportunity for selective targeting. For DNMTs recognition, the pharmacophoric motif of the natural ligand cytosine is retained and coupled to a polycyclic aromatic heterocycle, thereby departing chemically from a DNA base while optimizing active-site interactions and potentially conferring selectivity. Docking-based virtual screening identified two privileged heterocyclic scaffolds - nitrogen-containing series A and oxygen-containing series B - and delineated their optimal substitution patterns. Chapters 3 and 4 detail the synthetic efforts. Streamlined four to five steps synthetic sequences were developed for both heterocyclic scaffolds, each initiated by a one-pot, three-component condensation of readily available starting materials. While only a single analogue was obtained from series A, approximately 16 derivatives were synthesized within series B. Chapter 5 is devoted to the biological evaluation of all synthesized compounds. An initial enzymatic screen - conducted in collaboration with Dr P. Arimondo (UMR 3523, Institut Pasteur) - quantified inhibition of DNMT1 and DNMT3 and revealed two molecules with pronounced selectivity for DNMT3. These lead compounds subsequently underwent cellular assays, and molecular modeling coupled with dynamic docking was performed to gain insights into the structural basis of their selectivity. The thesis concludes with an overall synthesis of the findings and outlines future research directions, followed by the full experimental section.