Replication-dependent irreversible topoisomerase 1 poisoning is responsible for FdUMP[10] anti-leukemic activity

Abstract

Previous studies have indicated that 5-Fluoro-2'-deoxyuridine-5'-O-monophosphate 10mer (FdUMP[10]) displays strong antileukemic activity through the dual targeting of thymidylate synthase (TS) and DNA topoisomerase 1 (Top1). The present studies were undertaken to clarify the relationship between the induction of a thymineless state and the formation of Top1 cleavage complexes (Top1CC) for inducing cell death and to clarify the role of DNA replication for induction of lethal DNA double-strand breaks (DSBs) in FdUMP[10]-treated acute myeloid leukemia (AML) cells. Human promyelocytic (HL60) and AML (KG1a, Molm13, THP-1) cells were synchronized by serum starvation and treated with FdUMP[10] with thymidine (Thy) rescue. Cells were assayed for TS inhibition, DNA DSBs, Top1CC, and apoptosis to clarify the interrelationship of TS inhibition and Top1CC for cell death. FdUMP[10] induced a thymineless state in AML cells and exogenous Thy administered within the first 18 hours of treatment rescued FdUMP[10]-induced Top1CC formation, γH2AX phosphorylation, and apoptosis induction. Exogenous Thy was not effective after cells had committed to mitosis and undergone cell division in the presence of FdUMP[10]. FdUMP[10] treatment resulted in Chk1 activation, and Chk1 inhibition enhanced FdUMP[10]-induced DNA damage and apoptosis. Jnk-signaling was required for FdUMP[10]-induced apoptosis in promyelocytic HL60 cells and in THP1 cells, but was antiapoptotic in Molm13 and to a lesser extent KG1a AML cells. The results are consistent with FdUMP[10] inducing a thymineless state, leading to misincorporation of FdU into genomic DNA of proliferating cells. Top1CC form in cells upon re-entry into S-phase, resulting in DNA double-strand breaks, and initiating apoptotic signaling that can be either muted or enhanced by Jnk-signaling depending on cell type.

Figure 1

FdUMP[10] is a potent inducer of thymineless death via induction of apoptosis in HL60 cells. (A) Time and concentration dependence of viability for HL60 cells following FdUMP[10] treatment for the indicated times. Experiments were based on six observations and FdUMP[10] cytotoxicity at 10⁻¹⁰ mol/L was significantly enhanced relative to control at 72 hours (p < 0.001) and at 10⁻⁹ mol/L at 24 hours (p < 0.05) and 48 hours (p < 0.009). (B) Luminescent readout of caspase activation indicates FdUMP[10] induces apoptosis. Data were based on at least 12 observations per condition with all pairwise comparisons versus control significant (all p < 0.0001). (C) Flow cytometry histogram for cleaved caspase-3 demonstrating mainly apoptotic cells 24 hours (center) and 48 hours (right) after FdUMP[10] (10⁻⁷ mol/L) treatment relative to control (left). (D) Western blot demonstrating cleavage of caspase-3 in HL60 cells treated with FdUMP[10] at 10⁻⁸ mol/L for the indicated times. (E) Thymidylate synthase (TS) inhibition in THP-1 (p < 0.008) and KG1a (p < 0.005) cells is significantly reduced relative to control (based on at least five observations) following FdUMP[10] (10⁻⁸ mol/L) treatment.

Figure 2

The consequences of thymineless death following FdUMP[10] treatment are effectively rescued with exogenous thymidine only when administered during the first replicative cycle (<18 hours). (A) Depiction of the experimental scheme used for Thy rescue experiments and the relative timing of induction of a thymineless state, Top1CC formation, and apoptotic cell death. (B) Cell viability assays showing the time dependence of Thy rescue following FdUMP[10] treatment (10⁻⁸ mol/L). (C) Induction of apoptotic cell death and time dependence of rescue by exogenous Thy (10⁻⁸ mol/L). Data are a summary of two to six experiments performed in triplicate with significant differences between adjacent time points indicated (time points being farther apart was also significant). (D) In vivo complex of enzyme (ICE) bioassays showing the induction of Top1CC formation following FdUMP[10] treatment (10⁻⁸ mol/L) and the time dependence of Thy rescue.

Figure 3

FdUMP[10] treatment results in cell-cycle arrest and induces DNA double-strand breaks (DSBs). (A) FdUMP[10]-induced DNA DSBs are not a consequence of apoptotic cell death. γH2AX phosphorylation following FdUMP[10] treatment (10⁻⁷ mol/L) is not rescued by Z-VAD (20 μmol/L). (B) FdUMP[10]-induced DNA DSBs (evaluated with γH2AX) can be rescued with exogenous Thy only when administered during the first replicative cycle (<18 hours). (C) Summary of tail moments (arbitrary units) from neutral Comet assays for HL60 cells treated with FdUMP [10] with or without 2′-deoxythymidine (dT) rescue. Values were based on 20 captured images, and mean value differences for 24- and 48-hour dT rescue are significant (p < 0.0006). (D) Summary of DNA histograms from HL60 cells treated with FdUMP[10] with Thy rescue as indicated. FdUMP[10] treatment results in apparent G1-arrest although EdU-incorporation indicates cells have entered S-phase (Supplemental Figure E8, online only, available at www.exphem.org). Thy rescue permits cells to progress further into S-phase before apoptosis induction.

Figure 4

Chk1 inhibition enhances FdUMP[10]-induced DNA damage. DNA histograms and Comet assays for HL60 cells treated with FdUMP [10] (Fd) for the indicated times. (A) Chk1 inhibition (SB218078 5 γmol/L) reduces viability of FUMP[10]-treated cells. The combination (FdUMP[10]/Chk1i) was significantly more cytotoxic than either single agent at both 24 (p < 0.0004) and 48 hours (p < 0.0001) based on six experiments in triplicate. (B) Effects of Chk1 inhibition in combination with FdUMP[10] (10 nmol/L, 48 hours) on cell-cycle progression. FdUMP[10] treatment results in S-phase arrest after one replicative cycle (<18 hours). The combination of FdUMP[10] and the Chk1 inhibitor SB218078 for 48 hours results in fewer arrested cells and greater sub-G₀ fraction. (C) Comet assays under neutral conditions indicated that FdUMP[10] induced DNA strand breaks and that Chk1 inhibition substantially enhances FdUMP[10]-induced DNA damage. Tail moments (TM = Tail length × % of DNA in the tail) were calculated from 20 individually imaged cells from parameters measured by computer software (Loats Associates). FdUMP[10] + Chk1i vs. FdUMP[10] p < 0.000004.

Figure 5

Selective activation of the Jnk pathway is necessary for FdUMP [10]-induced apoptosis, but not cytotoxicity, in HL60 cells. (A) Western blots demonstratingthatFdUMP[10]selectively activatesJnk butnotother mitogen-activated protein kinase pathways such as Erk and p38. (B) Caspase-glo apoptosis assay results demonstrating that Jnk inhibition (SP600125, 50 μmol/L) abrogated FdUMP[10]-induced apoptosis. Mean apoptosis values for FdUMP[10] + Jnki were significantly different from FdUMP[10] at both 24 and 48 hours (p < 0.03). Data were based on six experiments in triplicate. (C) Cell titer-glo assays demonstrating that Jnk inhibition does not significantly rescue FdUMP[10] cytotoxicity. (D) Western blots demonstrating that activation of Jnk following FdUMP[10] treatment may be prevented by exogenous Thy only if administered during the first replicative cycle.