Targeting ATR/CHK1 pathway in acute myeloid leukemia to overcome chemoresistance

ABSTRACT

Resistance of acute myeloid leukemia to current therapies leads to frequent relapses. Identification of molecular mechanisms involved in chemoresistance constitutes a key challenge to define new therapeutic concepts. Here, we show that the ATR/CHK1 pathway, essential in maintaining genomic stability, is involved in resistance and proliferation characteristics of leukemic cells.

Genomic instability is a prominent source of genetic diversity within tumors, generating cell populations that can be selected in modified micro-environmental or therapeutic contexts. Genetic instability may arise from deregulation of the DNA replication program, a process known as DNA replicative stress (RS), which is observed at the earliest stages of cancer initiation, and constitutes a driving force for tumor development and a source of clinical biomarkers.¹ Oncogenes as well as certain chemotherapeutic agents are potent inducers of RS. The replication stress response (RSR), which leads to an adaptive response to overcome RS, is coordinated by the checkpoint kinase 1 (CHK1) and by the ataxia telangiectasia and Rad3-related kinase (ATR), which phosphorylates and activates CHK1. Studies with a mouse model expressing an extra allele of CHEK1 showed that supraphysiological abundance of CHK1 protects cells from oncogenic RS.² Consistently, increased expression and activation of ATR and CHK1 kinases were reported in various cancers. As a consequence, targeting ATR or CHK1 is preferentially toxic for tumors harboring high levels of RS .^3,4 Since, cancer cells with high RS may become highly reliant on ATR and CHK1 kinases for their survival upon treatment,⁵ therapeutic strategies have been consequently developed to exploit this dependency and are currently evaluated in clinical trials in association with DNA-damaging agents.

Genome-wide replication analyses detected widespread deregulation of replication timing in both acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) cells, suggesting common defects during early leukemogenesis,⁶ and supporting the notion of replication defects as a source of new therapeutic strategies. This aspect is clinically relevant, because patients with AML are currently treated with the so-called “3+7” regimen, combining daily three doses of anthracycline and seven doses of cytarabine, two DNA-damaging agents that induce RS. AML occurs due to excessively proliferating myeloid cells blocked in their differentiation, and is characterized by frequent relapse following initial remission. Since it is admitted that relapses initiate from a subpopulation of leukemic cells refractory to treatment, the identification of resistance mechanisms to these therapeutic drugs constitutes a key challenge for improving AML treatments. Indeed, AML is one of the hematological malignancies where no key specific treatment has been developed compared to the recent clinical progress made with multiple myeloma, B-cell lymphoma, chronic myelocytic leukemia, or ALL.

To investigate this issue, CHEK1 expression was assessed by real-time quantitative Polymerase Chain Reaction (qPCR) analysis in a cohort of 198 AML patients. By using multivariable models adjusted for several parameters (age, sex, etc.), we found that high CHEK1 expression is the most powerful independent predictor of shorter survival among the expression of 72 RSR genes tested. Moreover, our results determined that CHEK1 transcript levels correlate with CHK1 protein abundance, and that high levels of CHK1 in primary cells at diagnosis are not associated with constitutive replication checkpoint activation. However, high CHK1 abundance favored cellular resistance to cytarabine, as assessed by colony-forming assay, and CHK1 inhibitor (SCH900776) restored sensitivity of high CHK1 leukemic cells to clinically relevant cytarabine concentrations. A single molecule DNA fiber-spreading assay adapted to primary samples revealed that inhibition of fork progression by cytarabine was more pronounced in AML cells with low-abundance CHK1, supporting the hypothesis that high CHK1 levels could facilitate fork progression or stalled fork restart upon treatment. These results also suggest that a subpopulation of AML cells with increased abundance of CHK1 may survive the selective pressure of cytarabine during the treatment, becoming enriched and forming a residual aggressive tumor burden that is the source of the relapse.

In parallel with these comprehensive results in terms of resistance potential, we also showed that high CHK1 abundance is associated with higher proliferation capacities of AML cells. Patient samples with high CHK1 levels have a more pronounced clonogenic potential, characterized by the formation of a higher number of colonies of larger size in semisolid medium cultures. This may be relevant with recent studies highlighting important functions of CHK1 during unperturbed cell cycling. CHK1 has roles in replication fork activity during S phase,⁷ as well as for entry into mitosis, through transcriptional regulation of mitotic components such as cyclin-dependent kinase 1 (CDK1).⁸ We accordingly established that the abundance of CDK1 and CHK1 is correlated at mRNA and protein levels in AML samples, providing mechanistic clues to explain the proliferative ability of high CHK1 abundance in leukemic cells, and in a wider sense some rational explanation to understand the source of relapse. These data open the interesting possibility that high CHK1 expression is associated with a “mitotic signature” that may explain higher proliferation capacities of these cells. It is important to note in this respect that we also observed a frequent association of high CHK1 levels with high CDC25A expression, a phosphatase involved in the activation of CDK1 during mitosis, and recently identified as a biomarker for ATR inhibitor sensitivity.⁹ In view of these data, the expression of other important actors of proliferation (cyclin B, Wee1, PLK1) could be addressed in AML with high CHK1 abundance.

While we clearly documented the heterogeneity of CHK1 abundance in primary AML,¹⁰ the mechanisms by which CHK1 mRNA and protein expressions are controlled in these leukemic cells remain unclear. Although we found a correlation between the mRNA and the protein expression of CHK1 in primary samples from patients, the respective importance of transcriptional, post-transcriptional, and post-translational pathways governing CHK1 expression in AML remains to be investigated.

Overall, these different works stress the urgent need for establishing a more integrated view of CHK1 functions in AML biology, including transformation and progression. We propose in particular that leukemic cells with high levels of CHK1 expression are naturally selected because of their ability to proliferate, to prevent replication fork collapse, and/or to activate homologous DNA repair and to cope with increased RS. We propose that under cytarabine treatment, a subpopulation expressing high CHK1 level is selected and may be responsible for frequent relapses in these malignancies. Since high-dose cytarabine used as a single agent remains the backbone of post-remission therapy, our results pave the way to assess CHK1 inhibition strategies in combination with cytarabine or as maintenance therapy to reduce the risk of relapse (Fig. 1). In consequence, an immediate step could be to initiate clinical trial to assess the therapeutic benefit of CHK1 inhibitor (SCH900776) in AML patients with high CHK1 abundance specifically associated with RS markers.

Figure 1.

ATR or CHK1 inhibition enhances chemosensitivity in acute myeloid leukemia. Cancer cells are characterized by a range of molecular defects (oncogene expression) leading to hyperproliferative capacities associated with increase in replicative stress resulting from fork stalling, collapse, and genomic instability. Beside this, DNA antimetabolite drugs extensively used in clinical therapies are also strong replicative stress inducers. (A) Cancer cells, including AML cells, have been selected or have adapted to replication stress through increased expression and activation of key mediators of DNA replicative stress response: Ataxia Telangiectasia-Mutated and Rad3-related kinase (ATR) and checkpoint kinase 1 (CHK1), essential for the maintenance of cell viability by controlling DNA repair, cell cycle checkpoint, DNA replication, apoptotic process, and mitotic entry. (B) Alternatives therapies could be developed in AML to bypass the selected advantage of cancer cells by inhibiting the activity of these kinases. Targeting the replicative stress response could render cells more vulnerable to conventional therapy as it decreases the strength of cell cycle checkpoints and perturbs DNA replication process. Thus cells are more prone to enter into mitosis with massive DNA damage, under-replicated DNA and undergo a process called “mitotic catastrophy” that leads to apoptosis.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.