Targeting apurinic/apyrimidinic endonuclease 1 and 8-oxoguanine DNA glycosylase as a therapeutic strategy in acute myeloid leukaemia

Abstract

Acute myeloid leukaemia (AML) is an extremely heterogeneous disease characterised by genomic instability, epigenetic changes and a high oxidative stress burden that drives disease progression and plays an essential role in prognosis and treatment response. AML treatment is challenging and the majority of AML patients suffer relapse, particularly elderly patients. Targeting DNA repair mechanisms in cancer is a proven approach that can potentiate the anticancer activity of chemotherapeutic agents to yield better outcomes. The evidence suggests that the APE1 and OGG1 components of the base excision DNA repair pathway may be essential to cancer cell survival, and based on these reports it is hypothesised that targeting APE1 and OGG1 will have therapeutic value in AML. In order to test this hypothesis, APE1 and OGG1 gene expression was silenced in several AML cell lines using small hairpin RNA (shRNA) interference and cells were investigated for effects on proliferation, cloning efficiency, cell cycle, apurinic/apyrimidinic (AP) site accumulation and sensitivity to anti-leukaemic chemotherapy. Moreover, wildtype AML cells were treated with APE1 inhibitors methoxyamine (MX), E3330 and APE1 inhibitor III (APE1-III), and the effect on cell proliferation and cell cycle profile was investigated as single agents and in combination with anti-leukaemic chemotherapy. APE1 and OGG1 proteins were expressed in all AML cell lines investigated, but protein levels were not correlated with mRNA gene expression, suggesting that post-translational modification may regulate both proteins. APE1 shRNA knockdown slowed cellular proliferation and reduced cloning efficiency of AML cell lines HL-60, AML3 and U937. APE1 knockdown did not potentiate the sensitivity of AML cell lines to chemotherapeutic agents, including temozolomide, Ara-C, daunorubicin, clofarabine, fludarabine and etoposide. In contrast, chemotherapy-induced cell killing induced was antagonised by APE1 knockdown in AML cells. APE1 inhibition using MX, E3330 and APE1-III showed some single agent activity, with evidence of reduced proliferation and clonogenicity, but potentiation of chemotherapy-induced cytotoxicity was not evident. APE1 knockdown had no discernible effect on cell cycle kinetics. In contrast, APE1 inhibition using methoxyamine significantly induced cell cycle blockade in S phase, but no alteration in cell cycle profile was evident following APE1 inhibition with E3330 or APE1- III. ii RNA sequencing of APE1 knockdown cells identified several genes that were significantly upregulated, including many involved in cell cycle regulation and genes that may contribute to leukaemogenesis, including PAX5, CDKN1A and FOXO1. Targeting OGG1 using shRNA had no effect on proliferation of HL-60 and U937 AML cell lines, and only a very modest effect on colony formation in semi-solid soft agar. Furthermore, OGG1 silencing had no effect on cellular sensitivity to anti-leukaemic chemotherapy agents, and also did not affect cell cycle kinetics. In conclusion, the data presented in this thesis indicate that APE1, but not OGG1, may have potential therapeutic value as a single agent, but not in combination with established antileukaemic drugs. Additionally, the extensive genetic heterogeneity of AML suggests that targeting APE1 may have a utility in some but not all subtypes of AML. OGG1 may provide prognostic value in AML, but appears not to be a suitable therapeutic target.