Cause and therapeutic impact of DNA-protein crosslink repair defect in myeloid leukemias

ABSTRACT Acute myeloid leukemias (AMLs) are a genetically heterogeneous group of clonal hematopoietic disorders characterized by accumulation of immature non-lymphoid marrow progenitors. While there have been notable therapeutic advances over the past 5 years, many AML subtypes continue to have case fatality rates of >50%. Despite the introduction of a number of targeted therapies, conventional cytotoxic drugs – alone or in combination with the targeted agents – remain the mainstay of AML therapy. Among the conventional cytotoxic drugs used to treat AML, several act by increasing unique types of DNA lesions known as DNA-protein crosslinks (DPCs). In particular, topoisomerase poisons increase the number of DPCs containing TOP2 (daunorubicin, mitoxantrone, etoposide) or TOP1 (topotecan) covalently bound to DNA. In addition, the hypomethylating agents decitabine and azacitidine increase the number of DPCs containing DNA methyltransferases covalently bound to DNA. The mechanisms involved in DPC repair are at present incompletely understood. To facilitate the further development of topotecan and other TOP1 poisons, as well contribute to the study of TOP1-containing DPCs, we have generated an antibody that specifically recognizes TOP1-DNA covalent complexes (TOP1ccs). Using this antibody, we have shown that the nuclear metalloproteinase SPARTAN and the serine protease FAM111A, acting upstream of the phosphodiesterase TDP1, play important roles in the repair of TOP1ccs in some tissues. Importantly, loss of SPARTAN, FAM111A or TDP1 leads to accumulation of TOP1ccs in the absence of drug treatment as well as enhanced sensitivity to the prototypic TOP1 poison camptothecin. More recently, we have also observed that a variety of malignant myeloid cells, including AML cell lines and primary AML specimens, contain readily detectable TOP1ccs in the absence of drug treatment and are slow to repair TOP1ccs upon exposure to the TOP1 poison topotecan. In contrast, the vast majority of tissues, including normal and malignant lymphoid cells as well as normal marrow, contain very few TOP1ccs in the absence of drug treatment. These results lead to the hypothesis that many myeloid neoplasms have previously unsuspected defects in TOP1cc repair that might affect their therapeutic sensitivity. To test this hypothesis, we now propose to define the biochemical basis for the constitutive increase in TOP1ccs in myeloid neoplasms (Aim 1), examine the impact of low TOP1cc repair on leukemia cell sensitivity to agents that stabilize DPCs (Aim 2), and assess the relationship between TOP1cc levels (constitutive and drug-induced) and clinical response of myeloid neoplasms to a topotecan-containing regimen currently undergoing NCI-sponsored phase II clinical testing in high risk AML (Aim 3). Collectively, these studies will provide important new insight into a previously unsuspected DPC repair defect in myeloid malignancies and begin to determine whether this repair defect has therapeutic implications that can be used to guide improvements in AML therapy.