Funktionelle und mechanistische Charakterisierung von EZH2 in Atm-defizienten Subtypen des Pankreaskarzinoms

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is predicted to be the second cancer associated cause of death after lung cancer by 2030. The dismal prognosis of PDAC is mainly driven by its resistance towards common cancer therapies which is predominantly caused by the intra- and interindividual molecular PDAC heterogeneity. Consequently, novel, stratification-based therapeutic strategies are urgently required to improve therapeutic responses and PDAC prognosis. One genetic alteration with therapy-predictive implications is the loss of the deoxyribonucleic acid damage gene ataxia teleangiectatica mutated (ATM) which occurs in around 12% of all PDAC specimen. In a transgenic mouse model, Atm loss in combination with oncogenic kirsten rat sarcoma viral oncogene homolog (KRAS) activation promoted tumor progression and shortened survival. Moreover, preliminary data links ATM-deficient PDAC subtypes with over-expression of the tumor promoting histone-methyltransferase enhancer of zeste homolog 2 (EZH2). The aim was to characterize the functional implications of EZH2 expression in ATM-deficient PDAC subtypes. In order to examine oncogenic EZH2-functions, genetic (siRNA) and pharmacological (EPZ-6438) approaches were used to interfere with EZH2 activity in murine PDAC cells with constitutively active KRAS and an ATM deficiency (KATMC cell lines). Annexin V-propodium iodide stainings and flow cytometry were used to compare apoptosis of Atm-deficient PDAC cells in the presence and absence of EZH2. Furthermore, MTT Assays were used to analyze cell viability. Moreover, a transgenic mouse model with conditional activation of KRAS and Atm- and Ezh2-deficiency was generated (KATMEC mice). Mice were evaluated for phenotypic differences, survival, tumor-incidence and metastasis in comparison to EZH2 expressing control littermates (KATMC mice). Histological analysis was utilized to characterize the impact of EZH2 expression on the formation of PDAC precursor lesions and PDAC development. Tumor baring KATMEC animals were further subjected to primary PDAC cell isolation. Characterization of the model showed that KATMEC mice have a significantly longer overall and tumor survival than KATMC control animals expressing EZH2. Despite similar growth during the first twelve weeks, KATMC mice showed a significantly reduced weight compared to KATMEC mice at time of death. During carcinogenesis higher proliferation rate and higher E-cadherin expression could be observed in KATMEC mice compared to KATMC mice. Furthermore, E-cadherin expression was higher in KATMEC pancreatic tumors. In vitro data showed increased apoptosis rates after Ezh2-knockdown in the context of Atm-deficiency. Nevertheless, after inhibiting EZH2 enzymatic activity by EPZ-6438 in the context of Atm-deficiency no induction of apoptosis was detectable. In addition, three stable tumor cell lines could be isolated out of the KATMEC transgenic mouse model to compare with KATMC cell lines in terms of functional parameters. When comparing primary isolated KATMC and KATMEC cell lines no differences in cell viability and apoptosis rate were perceived. Overall, our findings support the hypothesis of tumor progression-promoting EZH2 functions in ATM-deficient PDAC. Therefore, the inhibition of EZH2 might represent a promising therapeutic option in ATM-deficient PDAC subtypes. Together, these data argue for the relevance of molecular stratification regarding the ATM status prior to pharmacologically interfering with EZH2 activity and expression in PDAC therapy.