Ezh2 is the enzymatic component of Polycomb Repressive Complex 2, and it catalyzes the methylation of lysine 27 (K27) of histone H3, a histone post-translational modification associated with repressed genes. Many cancer types are characterized by gene inactivation. Ezh2 mRNA levels are increased in a variety of tumors, consistent with a role in repressing critical genes. Recently, however, a specific mutation in the catalytic site of Ezh2 (Y641) that hyperactivates its H3K27 tri-methyltransferase activity while eliminating its ability to mono- methylate H3K27 been detected with high frequency in one specific subtype of cancer: the germinal center (GC) B cell subtype of diffuse large B cell lymphoma (GCB-DLBCL). This mutation occurs only in the tumor cell, not in the germline DNA, and therefore is a somatically acquired and positively selected mutation. Such a unique gain of function mutation is unprecedented. Ezh2 is highly expressed in normal GC B cells, but is low in na?ve B cells. We hypothesize that Ezh2 expression is increased in GC B cells to repress a specific subset of genes that must be downregulated to allow the differentiation of a na?ve B cell into a GC B cell, and/or of a GC B cell into a post-GC B cell. We further hypothesize that the Y461F mutation will result in over-repression of a subset of genes that normally are activated or re-activated in order for a B cell to exit the GC compartment. If this hypothesis is correct, then we predict that this over-repression might retain the GC B cells in the mutagenic environment of the GC longer than usual. The direct effect to B cells of this Ezh2 mutation cannot be directly addressed in patients'lymphoma samples, since the genome of each individual is unique, and because the lymphomas may have translocations or a variety of other genetic or epigenetic changes. Here we will directly test the effect of this hyperactive Ezh2 in an inbred strain of mouse, so that we can unambiguously determine the downstream consequences of this lymphoma-associated mutation. We will make a mouse with the Ezh2 Y641F targeted into the ROSA26 locus. The targeting construct has a floxed transcriptional stop between the promoter and Ezh2 Y641F. When crossed to C?1-Cre mice, the mutant Ezh2 will be expressed in GC B cells. We will determine if Ezh2 Y641F, with all of its downstream epigenetic, transcriptional and biological consequences, will be sufficient to lead to lymphoma. We will determine if it results in delayed exit from the GC, which may provide additional time in the mutagenic environment of the GC. We will perform gene expression profiling of GC B cells from Ezh2 Y641F and WT mice, and also ChIP-seq for H3K27me3. We will determine the downstream consequences of the altered enzymatic activity of Ezh2 Y641 by ChIP-seq for PRC1 components and ubiquitinated H2AK119, which is catalyzed by PRC1. The proposed studies will provide insight into the downstream effects of this mutation on gene expression, epigenetic profile, B cell differentiation after antigen stimulation, and lymphomagenesis. If this Ezh2 Y641 is sufficient in itself to result in lymphomagenesis, then specific inhibitors of the mutant Ezh2 could potentially be designed as a therapeutic. )
Diffuse large B cell lymphoma (DLBCL) is an aggressive disease that accounts for 40% of all non- Hodgkin's lymphomas. A substantial fraction of the GCB subset of DLBCL have a unique mutation in the catalytic domain of the enzyme Ezh2 which marks genes for repression. Here we will test whether this hyperactive Ezh2 by itself is sufficient to cause lymphoma in a mouse model;if so, then allele- specific inhibitors of the mutant Ezh2 could well be a future therapeutic target for GCB-DLBCL patients with this mutation.)
