Inhibition of ATR Potentiates Acute Myeloid Leukemia Cells to the RNA Pol I Transcription Inhibitor CX-5461

Leukemia is the most common type of childhood cancer. It accounts for about 30% of all malignancies in children and adolescents less than 14 years of age. Acute myeloid leukemia (AML) accounts for one-fourth of acute childhood leukemias and is considered a leading cause of death in the pediatric population with an overall 5-year survival of 65% and relapse rates approaching 40-50%. AML stem cells are believed to be the source of AML relapse, thus new molecular therapies targeting AML stem cells are urgently needed to lower risk of relapse and improve clinical outcomes.

The rate of ribosome production controls cell growth and proliferation. RNA polymerase (Pol I), which is upregulated in acute myeloid leukemia (AML), transcribes ribosomal DNA, regulating ribosome production. CX-5461 is a novel inhibitor of Pol I transcription and has shown therapeutic efficacy in hematological cancers with minimal effect on normal cells. Recent evidence suggests that CX-5461 exerts its antitumor activity by blocking DNA replication forks and inducing DNA single strand breaks, leading to ATR activation and G2/M cell cycle arrest (Xu H, et al. Nat Commun. 2017;8:14432), representing a novel mechanism of resistance to this promising antitumor agent. Therefore, we hypothesized that targeting ATR using its selective inhibitor AZD-6738 would abolish CX-5461-induced G2 cell cycle checkpoint activation, preventing repair of damaged DNA, leading to more DNA damage and cell death.

To begin to determine if AZD-6738 enhances CX-5461-induced AML cell death, we treated a panel of AML cell lines with variable concentrations of CX-5461 alone or in combination with AZD-6738 for 48 hours. CX-5461 treatment induced cell death independent of the status of TP53gene, as determined by Annexin V/Propidium Iodide (PI) staining and flow cytometry analysis, and was accompanied by increased cleavage of caspase 3 and PARP, demonstrating that CX-5461 induces apoptosis. Lentiviral shRNA knockdown revealed that CX-5461-induced cell death was partly dependent on both Bax and Bak, demonstrating that CX-5461 induces apoptosis, at least partially through the intrinsic apoptotic pathway. When combined, CX-5461 and AZD-6738 synergistically induced apoptosis in AML cell lines (n=6) and primary patient samples (n=3), determined by Annexin V/PI staining, flow cytometry analyses, and calculation of combination index values using the CalcuSyn software. Synergy was also confirmed in an additional 10 primary AML patient samples using MTT assays and standard isobologram analyses. Colony formation assays using primary AML patient samples showed that both CX-5461 and AZD-6738 individual treatments significantly reduced colony formation capacity compared to vehicle control, and was further reduced by combination treatment. Treatment of AML cells with CX-5461 induced G2/M cell cycle arrest, which was abolished by AZD-6738. This was accompanied by significantly increased cell death. CHK-1-selective inhibitor LY2603618 also abolished CX-5461-induced G2/M cell cycle arrest and synergistically induced cell death when combined with CX-5461 in AML cell lines. Protein levels of γH2AX increased with single drug treatment, which was enhanced in combination treated AML cells, indicating that the two drugs cooperatively induced DNA damage, which was confirmed by comet assays. Further, chromatin fractionation results showed that AZD-6738 enhanced binding of RPA32 to chromatin induced by CX-5461 treatment. Moreover, CX-5461 treatment substantially increased protein levels for ribonucleotide reductase subunit M2, which was abolished by the addition of AZD-6738. Taken together, inhibition of ATR, using AZD-6738, can synergistically enhance cell death induced by the Pol I transcription inhibitor CX-5461 in both AML cell lines and primary patient samples in vitro. Additional studies in both AML cell line- and patient-derived xenograft mouse models are warranted to confirm the efficacy of the combinationin vivo. Leukemia is the most common type of childhood cancer. It accounts for about 30% of all malignancies in children and adolescents less than 14 years of age. Acute myeloid leukemia (AML) accounts for one-fourth of acute childhood leukemias and is considered a leading cause of death in the pediatric population with an overall 5-year survival of 65% and relapse rates approaching 40-50%. AML stem cells are believed to be the source of AML relapse, thus new molecular therapies targeting AML stem cells are urgently needed to lower risk of relapse and improve clinical outcomes.

The rate of ribosome production controls cell growth and proliferation. RNA polymerase (Pol I), which is upregulated in acute myeloid leukemia (AML), transcribes ribosomal DNA, regulating ribosome production. CX-5461 is a novel inhibitor of Pol I transcription and has shown therapeutic efficacy in hematological cancers with minimal effect on normal cells. Recent evidence suggests that CX-5461 exerts its antitumor activity by blocking DNA replication forks and inducing DNA single strand breaks, leading to ATR activation and G2/M cell cycle arrest (Xu H, et al. Nat Commun. 2017;8:14432), representing a novel mechanism of resistance to this promising antitumor agent. Therefore, we hypothesized that targeting ATR using its selective inhibitor AZD-6738 would abolish CX-5461-induced G2 cell cycle checkpoint activation, preventing repair of damaged DNA, leading to more DNA damage and cell death.

To begin to determine if AZD-6738 enhances CX-5461-induced AML cell death, we treated a panel of AML cell lines with variable concentrations of CX-5461 alone or in combination with AZD-6738 for 48 hours. CX-5461 treatment induced cell death independent of the status of TP53gene, as determined by Annexin V/Propidium Iodide (PI) staining and flow cytometry analysis, and was accompanied by increased cleavage of caspase 3 and PARP, demonstrating that CX-5461 induces apoptosis. Lentiviral shRNA knockdown revealed that CX-5461-induced cell death was partly dependent on both Bax and Bak, demonstrating that CX-5461 induces apoptosis, at least partially through the intrinsic apoptotic pathway. When combined, CX-5461 and AZD-6738 synergistically induced apoptosis in AML cell lines (n=6) and primary patient samples (n=3), determined by Annexin V/PI staining, flow cytometry analyses, and calculation of combination index values using the CalcuSyn software. Synergy was also confirmed in an additional 10 primary AML patient samples using MTT assays and standard isobologram analyses. Colony formation assays using primary AML patient samples showed that both CX-5461 and AZD-6738 individual treatments significantly reduced colony formation capacity compared to vehicle control, and was further reduced by combination treatment. Treatment of AML cells with CX-5461 induced G2/M cell cycle arrest, which was abolished by AZD-6738. This was accompanied by significantly increased cell death. CHK-1-selective inhibitor LY2603618 also abolished CX-5461-induced G2/M cell cycle arrest and synergistically induced cell death when combined with CX-5461 in AML cell lines. Protein levels of γH2AX increased with single drug treatment, which was enhanced in combination treated AML cells, indicating that the two drugs cooperatively induced DNA damage, which was confirmed by comet assays. Further, chromatin fractionation results showed that AZD-6738 enhanced binding of RPA32 to chromatin induced by CX-5461 treatment. Moreover, CX-5461 treatment substantially increased protein levels for ribonucleotide reductase subunit M2, which was abolished by the addition of AZD-6738. Taken together, inhibition of ATR, using AZD-6738, can synergistically enhance cell death induced by the Pol I transcription inhibitor CX-5461 in both AML cell lines and primary patient samples in vitro. Additional studies in both AML cell line- and patient-derived xenograft mouse models are warranted to confirm the efficacy of the combinationin vivo.

Keywords:CX-5461; AZD-6738; ATR; Pol I; acute myeloid leukemia