Combined Inhibition of ATR and WEE1 as a Novel Therapeutic Strategy in Triple-Negative Breast Cancer

Abstract

Triple negative breast cancer (TNBC) is a highly aggressive subtype of breast cancer that poses a clinical challenge. Thus, new therapy strategies are urgently needed. The selective WEE1 inhibitor, AZD1775, has shown strong anti-proliferative effects on a variety of tumors. Here, we first demonstrate that inhibition of ATR by selective inhibitor AZD6738 can enhance AZD1775-caused growth inhibition in TNBC. Our results show that the enhanced cell death is attributed to repressed DNA damage repair and excessive replication stress, thereby causing increased DNA damage reflected by accumulation of the DNA double-strand-break marker γH2AX. On the other hand, combined treatment with AZD6738 and AZD1775 forces mitotic entry of cells with DNA damages by activating CDK1 activity, inducing severely aberrant mitosis and mitotic catastrophe, ultimately resulting in cell death. Dual inhibition of WEE1 and ATR also inactivated RAD51-mediated homologous recombination, which sensitized TNBC cells to cisplatin and PARP inhibitor. Here, based on the preclinical results that ATR inhibition synergizes with WEE1 inhibition in TNBC, we propose that this combination therapy alone, or in parallel with chemotherapy, represents an innovative and potent targeted therapy in TNBC.

Figure 1

ATR inhibitor AZD6738 potentiates anti-proliferation of WEE1 inhibitor AZD1775 in TNBC cells.A, MDA-231 and Hs578t cells were treated with 0.1-0.5 μM of AZD1775 for 1 and 3 h. Western blot showing activation of WEE1 targets. B, MDA-231 and Hs578t cells were treated with 0.05-50 μM of AZD1775 in combination with 0.1-10 μM of ATRi, ATMi or DNA-PKi for 72 h. Cell viability was detected by CCK8 (n = 3, mean). C, Proliferation curve of MDA-231 and Hs578t cells treated by 0.05-50 of μM AZD1775 with or without a constant dose of AZD6738 for 72 h (n = 3, mean ± SD) CI vs. Fa plot (combination index vs. fraction affected) for cell viability data presented in upper panel. The CI values were calculated by CompuSyn software. CI values below 1 are considered to have a synergistic interaction (n = 3, mean). D, Respective images of colony-forming results in MDA-231 and Hs578t treated by indicated concentrations of AZD1775 and AZD6738 (either single or both). Samples treated by inhibitors were compared to those of samples containing DMSO. In the upper small insets, the same cells were treated by the indicated concentrations of AZD6738. In the lower small insets, the CI values were calculated based on data in the main panel and upper inset (n = 3, mean ± SD). E, MDA-231 and Hs578t cells were treated for 24 hours with 100 nM of AZD1775, 500 nM of AZD6738 or both. Apoptotic cells were defined by flow cytometry with dual staining of annexin V (AV) and propidium iodide (PI) after treatment. Percentages of early (AV+PI−) and late (AV+PI+) apoptotic cells were calculated. Quantitative data of the apoptotic cells are shown (n = 3, mean ± SD) (**P < .01 and ***P < .001). F, Cells were treated with 0.1-50 μM of AZD1775 with AZD6738 for 72 h, and the dose ratio of AZD1775/AZD6738 was 1:5 or 5:1. Cell viability was investigated by CCK8 assay. The CI values at different inhibition levels (50%,75%,90% and 95%) analyzed by CompuSyn software are shown.

Figure 2

AZD6738 targets DDR activated by AZD1775 and enhances AZD1775-induced replication stress in TNBC cancer cells.A, MDA-231 or Hs578t cells were treated with 50 nM or 100 nM of AZD1775 with or without 500 nM of AZD6738 for 24 h. After treatment, western blot was performed using anti-γH2AX, anti-pCHK1 (S345), anti-CHK1, anti-pCDC25c (S216), anti-CDC25c, anti-pHH3, anti-RAD51 anti-pRPA32 (S4/S8), anti-cleaved-caspase 3, and anti-GAPDH antibodies. B, MDA-231 and Hs578t cells were treated for 24 hours with 100 nM of AZD1775, 500 nM of AZD6738 or both. Flow cytometry was used to identify the population of cells positive for γH2AX after treatment. Quantitative data of γH2AX-positive cell population are shown (n = 3, mean ± SD) (*P < .05, **P < .01, and ***P < .001). C, MDA-231 cells were treated with 50 nM or 100 nM of AZD1775 with or without 500 nM of AZD6738 for 24 h. Cells were probed with anti-γH2AX and anti-RAD51 antibodies. Scale bar: 5 μm. D, Quantitative data of the γH2AX-positive (five or more foci per cell) cells and pan-nuclear γH2AX signal after indicated treatments in the MDA-231 and Hs578t cells are shown (n = 3, mean ± SD). E, MDA-231 cells were treated with AZD1775 or AZD6738 alone or in combination for 24 h. Nuclear expression of pRPA32 (S4/S8) was determined by immunofluorescent imaging. Scale bar: 10 μm. F, Quantitative data of pRPA32 (S4/S8) positive cells after the indicated treatments are shown in MDA-231 and Hs578t cells (n = 3, mean ± SD) (**P < .01 and ***P < .001).

Figure 3

Combined WEE1 and ATR inhibition forces mitotic entry of cells with DNA damage. A, MDA-231 and Hs578t cells were treated with the indicated drugs for 48 h. Cells were probed with anti-pHH3 (Ser10) or γH2AX antibodies by flow cytometry. Quantitative data of flow cytometry in upper panel are shown (n = 2, mean ± SD). B, MDA-231 was co-treated with AZD1775 and AZD6738 for 24 h, followed by double staining for p-HH3 (red), or α-tubulin (green) together with DAPI (blue) by immunofluorescence. Scale bar: 10 μm. C, Respective images of typical mitotic phenotypes and several abnormal mitotic phenotypes in pHH3-positive MDA-231 cells are shown. Red, pHH3; green, α-tubulin. D, Quantification of pHH3-positive MDA-231 cells and abnormal mitotic cells in pHH3-positive MDA-231 cells after treatment for 24 h (n = 2, mean ± SD). E, Respective images of normal nuclei and micronuclei are shown. Blue, DAPI. F, Quantification of micronuclei numbers and total nucleuses in MDA-231 and Hs578t cells after treatment for 24 h (n = 2, mean ± SD).

Figure 4

CDK activities are required for the synergistic cytotoxic effect of combination treatment of AZD1775 and AZD6738 in MDA-231 and Hs578t cells.A, MDA-231 and Hs578t cells were treated by DMSO, roscovitine (CDK1/2i), or palbociclib (CDK4/6i) for 24 h, after which DMSO and CDK inhibitors were washed out, and cells were treated with the indicated concentrations of AZD1775 with 100 nM of AZD6738 for another 72 h. Cell viability was detected by CCK8 (n = 3, mean ± SD). B, MDA-231 cells were pretreated with 2 μM of CDK1/2 inhibitor roscovitine, or 2 μM of CDK4/6 inhibitor palbociclib for 6 h before the medium was changed with that containing 100 nM of AZD1775 and 500 nM of AZD6738 for 24 h. Cells after treatment were probed with anti-γH2AX and anti-RAD51 antibodies and the respective images are shown. Scale bar: 10 μm. C, Quantitative data of the pan-nuclear γH2AX after treatment in Figure 4B are presented (n = 2, mean ± SD) (**P < .01 and ***P < .001).

Figure 5

Combined WEE1 and ATR inhibition further sensitizes TNBC cancer cells with defective HRR to cisplatin.A, MDA-231 and Hs578t cells were treated for 72 h by the indicated concentrations of cisplatin with 100 nM of AZD1775, or 500 nM of AZD6738 or in combination. Cell viability was detected by CCK8 assay (n = 3, mean ± SD). B, Expression of RAD51, γH2AX and pHH3 of MDA-231 and Hs578t exposed to cisplatin alone, or in combination with AZD1775 or AZD6738, or both, were determined by immunoblot. C, An immunofluorescence assay was used to detect γH2AX and RAD51 expression in the nucleus in Hs578t following exposure to 3 μM of cisplatin with AZD1775 or AZD6738, or both. Scale bar: 2 μm. D, Percentage of cells with γH2AX foci (five or more foci per cell), pan-nuclear γH2AX signal or RAD51 foci (five or more foci per cell) after the indicated treatments in MDA-231 and Hs578t cells (n = 3, mean ± SD). E, Pearson’s coefficient shown as the quantification of RAD51 and γH2AX co-localization of three separate experiments in Figure 5C calculated by Image-Pro Plus software (n = 3, mean ± SD). F, After 48 h of treatment with vehicle or BRCA1 siRNA, MDA-231 cells were treated with the indicated concentrations of AZD1775 or AZD67398 for 72 h. Cell viability was analyzed by the CCK8 assay (n = 3, mean ± SD). G, Two days after the MDA-231 cells were transfected with vehicle or BRCA1 siRNA, cells were exposed to the indicated concentrations of cisplatin with or without the combination of 100 nM of AZD1775 and 500 nM of AZD6738 for 72 h, and cell viability was determined by CCK8 assay (n = 3, mean ± SD).

Figure 6

In Vivo efficacy of AZD1775 and AZD6738. Mice bearing MDA-231 xenografts with a tumor volume of 100±50mm³ (6 for each group) were dosed with AZD1775 (WEE1i; 30mg/kg/d, p.o.) or AZD6738 (ATRi; 60 mg/kg/d, p.o.) or both. Vehicle and drugs were administered on days 0-4 and 6-10. Tumors were taken 8 hours after last drug administration for western blot and immunofluorescence assay. A, Tumor volumes were evaluated every 3 days and calculated by the formula:(S² × L) /2. S: short diameter; L: long diameter (*P < .05). B, Body weight data for MDA-231 xenograft mouse. C, Apoptosis of the tumor tissues in different groups are shown by TUNEL assays. Scale bar: 20 μm. D, Quantification of apoptosis cells in different groups (n = 3, mean ± SD) (**P < .01, and ***P < .001). E, Immunofluorescence assay was utilized to analyze γH2AX, pRPA32(S4/S8) and pHH3 expression of tumor tissues from control group and combinational treatment group. Scale bar: 20 μm. F, Proposed mechanism for AZD1775/AZD6738 synthetic lethality in TNBC. WEE1 inhibition increases CDK1/2 activity, leading to replication stress and abrogation of G2/M checkpoint, which results in accumulated DNA damage including ssDNAs and DSBs. On the other hand, highly activated CDK1 activity forces cells with DNA damage into mitosis, triggering mitosis catastrophe. Replication stress and DNA damages activate ATR-CHK1 pathway, which activates WEE1 and inhibits CDC25c to inhibit CDK activity. Activated ATR also can directly regulate replication stress and DDR, which limits cytotoxic effect of AZD1775. AZD6738 inhibits ATR and thus enhances the forced mitosis and DNA damage, eventually leading to cell death. In almost all TNBC cells, the functions of p53/p21 are lost, resulting in unscheduled CDK activity that is more dependent on ATR-WEE1 pathway.

Figure S1.

Inhibitory activity of AZD1775 and AZD6738, Related toFigure 1. Different TNBC cell lines were treated with 0.05-50 μM of AZD1775 (A) and AZD6738 (B) for 72 h, and the anti-proliferative effects were determined by CCK8 (n = 3, mean ± SD). C, Proliferation curves of MDA-157, HCC1937 and BT549 cells treated by 0.05-50 μM of AZD1775 with or without a constant dose of AZD6738 for 72 h are shown (n = 3, mean ± SD). CI vs. Fa plot (combination index vs. fraction affected) for the cell viability data presented in upper panel. The CI values were calculated by CompuSyn software. CI values below 1 are considered to have a synergistic interaction (n = 3, mean). D, MDA-231 resistant cell lines were treated with 0.05-50 μM of AZD1775, AZD6738 and AZD1775 with 500 nM of AZD6738 for 72 h, and the cell viability was determined by CCK8 (n = 3, mean ± SD). E, MDA-231, Hs578t, MCF7 and MCF10A cells were treated for 24 hours with 100 nM of AZD1775, 500 nM of AZD6738 or both. Propidium iodide (PI) staining was carried out to detect the cell-cycle profile using flow cytometry. Quantitative cell cycle data was obtained using the Modfit software (n = 3, mean ± SD).

Figure S2.

Treatment with both AZD1775 and AZD6738 triggers DNA damage and replication stress in Hs578t. Related toFigure 2. Hs578t cells were treated with 50 nM or 100 nM of AZD1775 with or without 500 nM AZD6738 for 24 h. A, Cells were probed with anti-γH2AX and anti-RAD51 antibodies. Scale bar: 5 μm. B, Immunofluorescent images of the representative nuclei of Hs578t treated with AZD1775 or AZD6738 alone or in combination for 24 h. Scale bar: 2um. C, Cells were probed with anti-pRP32A (S4/S8) antibody. Scale bar: 10 μm.

Figure S3.

Dual inhibition of WEE1 and ATR further sensitizes TNBC cells with defective BRCA1 to cisplatin and veliparib. Related toFigure 5. A, MDA-231 and Hs578t cells were treated for 72 h with the indicated concentration of veliparib with 100 nM of AZD1775, or 500 nM of AZD6738 or a combination. Cell viability was detected by CCK8 assay (n = 3, mean ± SD). B, After treatment of 1 μM of cisplatin with 100 nM of AZD1775 or 500nM of AZD6738, or both, an immunofluorescence assay was utilized to analyze γH2AX and RAD51 expression in the nucleus. Scale bar: 2 um. C, After 48 h of treatment with vehicle or BRCA1 siRNA in MDA-231, cell lysates were used to analyze the BRCA1 expression. D, After MDA-231 cells were transfected with vehicle or BRCA1 siRNA for 48 h, cells were exposed to indicated concentration of veliparib with or without the combination of 100 nM of AZD1775 and 500 nM of AZD6738 for 72 h, and the cell viability was determined by CCK8 assay (n = 3, mean ± SD). E, MDA-231 was transfected with lentivirus encoding mutational BRCA1 for 72 h, after which cells were treated with the indicated concentration of cisplatin or veliparib in combination with 100 nM of AZD1775 and 500 nM of AZD6738 for 72 h. Cell viability was determined by CCK8 assay. F, Expression of pCHK1 and pHH3 of xenografted tumor exposed to AZD1775, or AZD6738 or both, were determined by immunoblot.