Lymphoma-Associated Mutations of EZH2 Result In a Change-of-Function

Abstract 707

EZH2, the catalytic subunit of the polycomb repressive complex 2 (PRC2), catalyzes the mono- through tri-methylation of lysine 27 on histone H3 (H3K27). While overexpression of EZH2 and increased H3K27 methylation have generally been associated with both hematologic malignancies and solid tumors, inactivating somatic mutations of Tyr641 (Y641F, Y641N, Y641S and Y641H) of EZH2 were recently reported to be associated with follicular lymphoma (FL) and the GCB subtype of diffuse large B-cell lymphoma (DLBCL) (Morin, Nat Genet 2010; 42: 181). In all cases, occurrence of the mutant EZH2 gene was heterozygous, and expression of both wild type and mutant alleles was detected in the mutant samples profiled by transcriptome sequencing. Further, the mutant forms of EZH2 could be incorporated into the multi-protein PRC2 complex, but the resulting complexes lacked the ability to catalyze trimethylation of an unmethylated H3K27 peptide substrate. To explore further the role of EZH2 in lymphomagenesis, we have evaluated the catalytic activity of the mutant EZH2 proteins in greater detail. Recombinant PRC2 complexes were prepared with wild type and Tyr641 mutant EZH2 forms. As previously reported, the wild type enzyme demonstrated robust activity but none of the mutant enzymes displayed significant methyltransferase activity on an unmodified H3K27 peptide. We next evaluated the activity of the enzymes using native avian erythrocyte olignucleosomes as the substrate in the reaction. In contrast to the peptide result, we found that the wild type and all of the mutant enzymes were active methyltransferases against the native nucleosome substrate. Since native nucleosome represents an admixture of the unmodified and mono-, di- and tri-methylated H3K27 we next evaluated the activity of the wild type and mutant enzymes on unmodified, and mono- and di-methylated H3K27 peptide. We demonstrate that the wild type enzyme displays greatest catalytic efficiency (k_cat/K) for the zero to mono-methylation reaction of H3K27, and diminished efficiency for subsequent (mono- to di- and di- to tri-methylation) reactions. In stark contrast, the disease-associated Y641 mutants display very limited ability to perform the first methylation reaction, but have enhanced catalytic efficiency for the subsequent reactions, relative to WT-enzyme. Catalytic coupling between the mutant EZH2 species and PRC2 complexes containing either wild type EZH2 or wild type EZH1 are predicted to augment H3K27 trimethylation and thus produce the malignant phenotype associated with mutant heterozygosity. To test this prediction, the level of H3K27 methylation was evaluated in lymphoma cell lines harboring only wild type EZH2 (OCI-LY-19) or heterozygous for EZH2 Y641N (DB, KARPAS and SU-DHL-6) or EZH2 Y641F (WSU-DLCL2) by immunoblotting. As predicted by simulations, the level of H3K27 trimethylation was elevated in all of the lymphoma cell lines harboring the mutant EZH2 relative the wild type. Additionally, we observe decreased H3K27 dimethylation and monomethylation in the cells harboring the mutated EZH2 relative to wild type enzyme; these reductions in di- and monomethylation are likewise consistent with expectations based on steady state kinetic simulations. The present results imply that the malignant phenotype of follicular lymphoma and diffuse large B cell lymphoma of the GCB subtype, associated with expression of mutant forms of EZH2, results from of an overall gain-of-function with respect to formation of the trimethylated form of H3K27. These data suggest that selective, small molecule inhibitors of EZH2 enzymatic function may form a rational underpinning for molecularly targeted therapeutics against mutant-harboring lymphomas and other malignancies in which EZH2 gain-of-function is pathogenic.