The role of Epigenetics in Epithelial Mesenchymal Transition (EMT) has recently emerged. Two epigenetic enzymes with paradoxical roles have previously been associated to EMT, EZH2 (Enhancer of Zeste 2 Polycomb Repressive Complex 2 (PRC2) Subunit), a lysine methyltranserase able to add the H3K27me3 mark, and the histone demethylase KDM6B (Lysine Demethylase 6B), which can remove the H3K27me3 mark. Nevertheless, it still remains unclear how these enzymes, with apparent opposite activities, could both promote EMT. In this study, we evaluated the function of these two enzymes using an EMT-inducible model, the lung cancer A549 cell line. ChIP-seq coupled with transcriptomic analysis showed that EZH2 and KDM6B were able to target and modulate the expression of different genes during EMT. Based on this analysis, we described INHBB, WTN5B, and ADAMTS6 as new EMT markers regulated by epigenetic modifications and directly implicated in EMT induction.

EMT (epithelial to mesenchymal transition) is a plastic phenomenon involved in metastasis formation. Its plasticity is conferred in a great part by its epigenetic regulation. It has been reported that the trimethylation of lysine 27 histone H3 (H3K27me3) was a master regulator of EMT through two antagonist enzymes that regulate this mark, the methyltransferase EZH2 (enhancer of zeste homolog 2) and the lysine demethylase KDM6B (lysine femethylase 6B). Here we report that EZH2 and KDM6B are overexpressed in numerous cancers and involved in the aggressive phenotype and EMT in various cell lines by regulating a specific subset of genes. The first paradoxical role of these enzymes is that they are antagonistic, but both involved in cancer aggressiveness and EMT. The second paradoxical role of EZH2 and KDM6B during EMT and cancer aggressiveness is that they are also inactivated or under-expressed in some cancer types and linked to epithelial phenotypes in other cancer cell lines. We also report that new cancer therapeutic strategies are targeting KDM6B and EZH2, but the specificity of these treatments may be increased by learning more about the mechanisms of action of these enzymes and their specific partners or target genes in different cancer types.

EZH2 et KDM6B catalysent respectivement la méthylation et la déméthylation de la marque épigénétique H3K27me3 (triméthylation de la lysine 27 de l’histone 3). Elles sont dérégulées dans de nombreux types de cancers. Nos travaux montrent que l’expression de KDM6B et EZH2 sont corrélées dans les cancers du poumon non à petites cellules et sont corrélées à l’expression du marqueur mésenchymateux VIMENTINE. Dans la lignée cellulaire de cancer du poumon A549, la surexpression de EZH2 ou de KDM6B entraine chacune les cellules vers un phénotype mésenchymateux. De plus, l’inhibition chimique de EZH2, mais plus significativement celle de KDM6B entraine également un phénotype mésenchymateux, au niveau moléculaire comme fonctionnel. Certains gènes sont fortement sur- ou sous-exprimés au cours de la TEM et leurs promoteurs pourraient être des cibles de EZH2 ou KDM6B d’après nos données de ChIP-seq. Nous avons alors émis l’hypothèse que KDM6B et EZH2 ciblent des gènes différents au cours de la TEM. Nous avons montré que la sous-expression de INHBB et la surexpression de WNT5B et ADAMTS6 au cours de la TEM est régulée au moins en partie par H3K27me3. La modulation de l’expression de ces gènes est suffisante pour impacter les capacités d’invasion des cellules A549. Pour conclure, EZH2 et KDM6B, malgré leur activité antagoniste, sont importantes pour la régulation de la TEM dans les cellules A549.