Abstract A22: Ezh2 haploinsufficiency and full insufficiency in KRAS/p53-null lung tumors drive distinct cellular phenotypes

EZH2, the functional enzymatic component of the Polycomb Repressive Complex 2 (PRC2), is a histone methyltransferase that trimethylates histone H3 at lysine 27 (H3K27me), and may act as an oncogene or tumor suppressor in different cellular contexts. EZH2 overexpression correlates with worse prognosis in many cancer types including breast, lung, and prostate cancer. However, other data suggest EZH2 activation results in better outcomes in colon cancer, and several publications show that inactivating mutations in EZH2 and its related proteins are common in T-cell acute lymphoblastic leukemia (T-ALL). Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related death worldwide. Because EZH2 inhibitors are currently in clinical trials for many cancer types, understanding the roles of EZH2 in lung tumor initiation and progression is important. We generated the LSL:Kras G12D/+ ; p53 flox/flox (LSL: lox-stop-lox) mice ( Kras/p53 mice) with either zero, one or two floxed alleles of Ezh2 and induced lung tumors with intranasal adeno-Cre virus administration. Similar to findings by other groups, mice with two floxed alleles of Ezh2 ( Ezh2 null) had higher tumor burden (57%), higher tumor grade, and lower survival (110 days) than Ezh2 wild-type Kras/p53 mice (45% and 116 days, respectively). Ezh2 null tumors also had a higher propensity to metastasize than Ezh2 wild-type tumors at time-points greater than 110 days after adeno-Cre virus administration (75% vs. 16%, respectively). However, we also observed that mice with only one floxed allele of Ezh2 ( Ezh2 heterozygous) had lower tumor burden (27%) and grade and increased survival (127 days) compared to Ezh2 wild-type mice. These data suggest that full and partial insufficiency for Ezh2 have distinct phenotypic outcomes. We generated two-dimensional tumor cell lines, from which the preliminary RNA-seq showed that 108 genes were significantly upregulated and 45 genes significantly downregulated in Ezh2 flox/flox cells relative to Ezh2 +/+ Kras/p53 cells (FDR 1.3). GSEA results of Ezh2 null vs. wild-type positively correlated with epithelial-mesenchymal transition (EMT), KRAS-driven tumorigenesis, metastasis, Hippo signaling, and imprinted genes. In contrast, Ezh2 heterozygous vs. wild-type showed enrichment of non-metastasis signatures and only partial enrichment of PRC2-target and imprinted genes. Consistent with RNA-seq results, Ezh2 null cell lines showed obvious EMT morphology compared to Ezh2 wild-type cells. ChIP-seq studies to learn how H3K27me3 correlates with gene expression are ongoing. In conclusion, Ezh2 haploinsufficiency (heterozygous) leads to better prognosis while full insufficiency (null) causes aggressive metastatic lung tumors in the Kras/p53 mouse model. Given that loss of the H3K27me3 methyltransferase Ezh2 leads to more aggressive KRAS-driven tumors, we are now exploring the efficacy of the H3K27me3 demethylase inhibitors in this genotype. Three-dimensional cultures of murine tumor cells have demonstrated sensitivity to the H3K27me3 demethylase inhibitors GSK-J4, which may be a method to epigenetically target KRAS+ tumors. Citation Format: Fan Chen, Aria L. Byrd, Yanming Zhao, Carla F. Kim, Christine Fillmore Brainson. Ezh2 haploinsufficiency and full insufficiency in KRAS/p53-null lung tumors drive distinct cellular phenotypes [abstract]. In: Proceedings of the AACR Special Conference: Advances in Modeling Cancer in Mice: Technology, Biology, and Beyond; 2017 Sep 24-27; Orlando, Florida. Philadelphia (PA): AACR; Cancer Res 2018;78(10 Suppl):Abstract nr A22.