Suppression of EZH2 Accelerates MYC-Driven Lymphomagenesis By Inhibition of Apoptosis

EZH2 is the catalytic component of the polycomb repressive complex 2 (PRC2), which also contains the non-catalytic subunits suppressor of zeste 12 (SUZ12) and embryonic ectoderm development (EED). This complex methylates histone H3 at lysine 27 (H3K27) which, together with H3K4 methylation, generates a bivalent “code” that primes genes for either expression or silencing. EZH2 is highly expressed in stem cells and many proliferating cells, downregulated in differentiated cells and frequently altered in cancer in ways that point to a context dependent role for this gene. For instance, overexpression of EZH2 has been described in prostate and breast cancer, where this overexpression is associated with invasive growth metastatic potential and poor clinical outcome.

More recently, both gain and loss of function mutations of EZH2 were identified in human cancer. In particular, activating mutations of EZH2 affecting a tyrosine at position 641 located within the SET domain (Y641) were observed in lymphoma. In striking contrast to wild type EZH2 which catalyzes the monomethylation of H3K27 very efficiently and shows less efficient catalytic activity in the subsequent di- and trimethylation reactions, the mutation generates a neomorphic protein with enhanced catalytic activity and efficiency for di- and tri-methylation, hinting toward a “cooperation” of both wild type and mutant EZH2 to increase total H3K27me3 levels and representing a functional equivalent of EZH2 overexpression in human lymphoma.

Gain of function mutations in EZH2 are frequent in Diffuse Large B-cell Lymphoma (DLBCL; 22%) and in Follicular Lymphoma (FL; 7%) and recent data implicate EZH2 as an oncogene in DLBCL and FL lymphomas.

In contrast to DLBCL and FL, EZH2 mutations have so far never been identified in MYC-driven B-cell Non-Hodgkin Lymphoma (B-NHL), and EZH2 expression is suppressed in Burkitt’s Lymphoma (BL), a lymphoma for which a MYC translocation is pathognomonic, suggesting that in this context EZH2 could have a role different from its role in DLBCL and FL. We probed the role of EZH2 in MYC-driven lymphomagenesis by mimicking the loss of function mutations by RNAi mediated suppression of EZH2 and the gain of function mutations by over-expression of the Y641 mutant.

Our results show that suppression of EZH2, but not overexpression of the EZH2 Y641 mutant, accelerates MYC-driven lymphomagenesis by attenuating apoptosis. Our model recapitulates the transcriptional signature of a subset of B-NHLs driven by MYC overactivation and EZH2 suppression. Taken together, our data imply EZH2 as a tumor suppressor in the context of MYC activation and thus raise a warning for the use of EZH2-targeted therapies in some B-NHLs subtypes.