Targeting ErbB and tankyrase1/2 prevent the emergence of drug-tolerant persister cells in ALK-positive lung cancer

Abstract

Targeting the drug tolerant persister (DTP) state in cancer cells should prevent further development of resistance mechanisms. This study explored combination therapies to inhibit alectinib-induced DTP cell formation from anaplastic lymphoma kinase-positive non-small cell lung cancer (ALK + NSCLC) patient-derived cells. After drug-screening 3114 compounds, pan-HER inhibitors (ErbB pathway) and tankyrase1/2 inhibitors (Wnt/β-catenin signaling) emerged as top candidates to inhibit alectinib-induced DTP cells growth. We confirmed knockdown of both TNKS1/2 in DTP cells recovered the sensitivity to alectinib. Further, our study suggested knockdown of TNKS1/2 increased stability of Axin1/2, which induced β-catenin degradation and decreased its nuclear translocation, thereby suppressing transcription of antiapoptotic and proliferation-related genes (survivin, c-MYC). Targeting both pathways with alectinib+pan-HER inhibitor and alectinib+TNKS1/2 inhibitor suppressed alectinib-induced DTP cells, and the triple combination almost completely prevented the appearance of DTP cells. In conclusion, combination with ALK-TKI, pan-HER and TNKS1/2 inhibitors has the potential to prevent the emergence of DTP in ALK + NSCLC.

Fig. 1. Preparation and characterization of ALK1510-c4 DTP cells from ALK1510-c4 cells.

a ALK1510-c4 DTP cells were generated from ALK1510-c4 cells after treatment with 1000 nM alectinib for 9 days, ALK1510-c4 regrown cells were generated from ALK1510-c4 DTP cells cultured in alectinib-free cell culture for 37 days, and comparison with ALK1510-c4 cell controls were assessed in a cell proliferation assay (mean [SD] of n = 3 experiments, *P < 0.05 between ALK1510-c4 and ALK1510-c4 DTP and ALK1510-c4 regrown; Tukey’s HSD test). b Immunoblots of cell lysates showing changes in phosphorylated and/or total ALK, AKT, ERK, STAT3, and cleaved PARP in ALK1510-c4 cells treated with 1000 nM alectinib for 1, 3, 24, and 48 h and 9 days. c Immunoblots of cell lysates showing changes in BIM, stem cell marker CD133, vimentin, and E-cadherin protein levels in ALK1510-c4 cells and ALK1510-c4 DTP cells. d Immunoblots of cell lysates showing change in phosphorylated and/or total ALK, AKT, ERK and STAT3 protein levels in ALK1510-c4 cells and ALK1510-c4 DTP cells after treatment with 1000 nM alectinib for 1 h. Microscopic examination of ALK1510-c4 parental and DTP cells generated from ALK1510-c4 by treatment with 1000 nM alectinib for 9 days with phase contrast (e), and crystal violet staining (f). g Immunoblots of cell lysates in ALK1510-c4 DTP cells after 13 and 26 days culture in the alectinib-free medium.

Fig. 2. Screening the compound library to identify drugs with specific growth rate (ALK1510-c4 DTP cells/ALK1510-c4 cells).

a ALK1510-c4 cells and ALK1510-c4 DTP cells were cultured with each of the 3114 compounds in an anticancer drug compound library (n = 1). Antiproliferative effects on ALK1510-c4 DTP cells relative to ALK1510-c4 cells are plotted as ratios in an ascending order for each compound by evaluating cell viability of the respective cells. b Top ten compounds with antiproliferative effects on ALK1510-c4 DTP cells relative to ALK1510-c4 cells are tabulated. c Cell proliferation assay/cytostatic studies showing relative cell viability/difference in drug sensitivity of ALK1510-c4 cells and ALK1510-c4 DTP cells after treatment with dacomitinib, neratinib, AZ6102, G007-LK, LY2090314, and omecamtiv mecarbil; mean [SD] of n = 3 experiments, *P < 0.05 between ALK1510-c4 and ALK1510-c4 DTP ; Student’s t test).

Fig. 3. Evaluation of ErbB signaling in ALK1510-c4 cells using alectinib and ALK1510-c4 DTP cells treated with pan-HER inhibitor dacomitinib.

a Immunoblots of cell lysates evaluating ErbB signaling molecules from ALK1510-c4 cells treated with 1000 nM alectinib for 1, 3, 24, and 48 h and 9 days. b Relative HER2 phosphorylation levels in ALK1510-c4 cells to nontreated controls described in Fig. 3a were measured by ELISA (mean [SD] of n = 3 experiments). c Relative EGFR phosphorylation levels of ALK1510-c4 cells to total EGFR protein levels using cells described in Fig. 3a quantified by densitometric analysis with Sally Sue™. d Immunoblots of cell lysates from ALK1510-c4 DTP cells evaluating ErbB signaling and ALK signaling after treatment with 10, 100, and 1000 nM dacomitinib for 3 h. e Relative HER2 phosphorylation levels of ALK1510-c4 DTP cells described in Fig. 3d were measured by ELISA (mean [SD] of n = 3 experiments).

Fig. 4. Efficacy and characterization of the combination of alectinib with pan-HER inhibitors on ALK1510-c4 cells.

a Cell proliferation assay of ALK1510-c4 cells cultured with alectinib, lorlatinib, dacomitinib, and neratinib; and alectinib or lorlatinib in combination with dacomitinib or neratinib (100 nM and 300 nM) (mean [SD] of n = 3 experiments, *P < 0.05 between ALK-TKI single treatment and combination treatment; Dunnett’s test). b ALK1510-c4 cells cultured with alectinib, dacomitinib, or alectinib + dacomitinib (100 nM and 300 nM) were evaluated for caspase-3/7 activity relative to vehicle-treated cells (mean [SD] of n = 3 experiments). c Immunoblots of cell lysates evaluating ErbB and ALK downstream signaling in ALK1510-c4 cells treated with 1000 nM alectinib, 100 nM dacomitinib, and their combination for 24 h. d Cell proliferation assay of ALK1510-c4 cells transfected with three siRNAs against EGFR, HER2, and HER3, or a nontarget control siRNA and treated with alectinib or lorlatinib (mean [SD] of n = 3 experiments, *P < 0.05 between ALK-TKI + siCtrl and ALK-TKI + siPan-HER; Student’s t test). e Immunoblots of cell lysates from ALK1510-c4 cells described in Fig. 4d treated with 1000 nM alectinib for 48 h. f Mean (SD) change in tumor volume in mice bearing ALK1510-c4 cells xenograft tumors treated with vehicle, 2 mg/kg alectinib, 10 mg/kg dacomitinib, or their combination for 14 days (P < 0.05 versus vehicle (a), alectinib (b) and dacomitinib (c); Wilcoxon rank sum test by the Holm–Bonferroni method; n = 8 mice per group). g Immunoblots of tumor lysates from ALK1510-c4 cells xenograft tumors in mice treated with vehicle, alectinib (2 mg/kg), or alectinib (2 mg/kg) + dacomitinib (10 mg/kg) combination for 2 days. h Body weight change in mice relative to the start of the treatment in Fig. 4f.

Fig. 5. Characterization of TNKS1/2 pathway in ALK1510-c4 cells and ALK1510-c4 DTP cells.

a The relative mRNA expression of TNKS1 and TNKS2 (TNKS1/2) in ALK1510-c4 cells and ALK1510-c4 DTP cells determined by RT-PCR (using the TaqMan probes) was calculated as the ratio of the normalized value (with GAPDH mRNA expression) (mean [SD] of n = 3 experiments; *P < 0.05 versus ALK1510-c4 cells; Dunnett’s test). b Immunoblots of cell lysates evaluating Wnt/β-catenin signaling in ALK1510-c4 cells treated with 1000 nM alectinib for 1, 3, 24, and 48 h and 9 days. c Immunoblots of cytosolic or nuclear lysates from ALK1510-c4 cells treated with 1000 nM alectinib for 24 h and 9 days.

Fig. 6. Efficacy in cell-based assay and characterization of TNKS1/2 inhibitors in ALK1510-c4 cells.

a Cell proliferation assay of ALK1510-c4 cells cultured with alectinib, lorlatinib, AZ6102, or G007-LK singly, or alectinib or lorlatinib in combination with AZ6102 or G007-LK (100 nM or 300 nM) (mean [SD] of n = 3 experiments, *P < 0.05 between ALK-TKI single treatment and combination treatment; Dunnett’s test). b The relative mRNA expression of TNKS1/2 in ALK1510-c4 cells treated with 1000 nM alectinib was determined by RT-PCR (using the TaqMan probes) and calculated as the ratio of the normalized value with GAPDH mRNA expression (mean [SD] of n = 3 experiments, *P < 0.05 versus siCtrl; Dunnett’s test). c Cell proliferation assay of ALK1510-c4 cells transfected with two siRNAs against TNKS1, TNKS2, or a nontarget control siRNA and cultured with alectinib or lorlatinib (mean [SD] of n = 3 experiments, *P < 0.05 among siCtrl, siTNKS1, siTNKS2 and siTNKS1/2; Tukey’s HSD test). d Immunoblots assessing Wnt/β-catenin signaling molecules from cell lysates of ALK1510-c4 cells described in Fig. 6b treated with 1000 nM alectinib for 48 h.

Fig. 7. Efficacy in animal model and characterization of TNKS1/2 inhibitors in ALK1510-c4 cells.

a Assessment of caspase-3/7 activity in ALK1510-c4 cells cultured with alectinib, AZ6102, and the combination of alectinib and AZ6102 (100 nM or 300 nM) relative to caspase-3/7 activity in vehicle-treated cells (mean [SD] of n = 3 experiments). b Immunoblots assessing Wnt/β-catenin signaling molecules from cell lysates of ALK1510-c4 cells treated with alectinib (1000 nM), AZ6102 (100 nM), or their combination for 48 h. c Mean (SD) change in tumor volume in mice bearing xenograft tumors with ALK1510-c4 cells and treated with vehicle, 2 mg/kg alectinib, 10 mg/kg or 50 mg/kg of G007-LK, or their combinations for 14 days (P < 0.05 versus vehicle (a), alectinib (b) and corresponding dose of G007-LK (c); Wilcoxon rank sum test by the Holm–Bonferroni method; n = 6 mice per group). d Change in body weight from baseline of mice treated with vehicle, alectinib (2 mg/kg), G007-LK (10 mg/kg and 50 mg/kg), and a combination of alectinib (2 mg/kg) and G007-LK (10 mg/kg or 50 mg/kg). e Immunoblots of tumor lysates from mice bearing xenograft tumors with ALK1510-c4 cells and treated with alectinib (2 mg/kg), or a combination of alectinib (2 mg/kg) and G007-LK (10 mg/kg) for 2 days.

Fig. 8. Efficacy and characterization of triple combination therapy in ALK1510-c4 cells and animal models.

a Relative cell viability of ALK1510-c4 cells cultured with alectinib, alectinib with dacomitinib (100 nM), alectinib with AZ6102 (100 nM), and a triple combination of alectinib + dacomitinib + AZ6102 (mean [SD] of n = 3 experiments, *P < 0.05 between doublet treatment and alectinib + dacomitinib + AZ6102 treatment; Dunnett’s test). b Caspase-3/7 activity relative to vehicle treatment was assessed in ALK1510-c4 cells treatment with the drugs described in Fig. 8a (mean [SD] of n = 3 experiments). c Immunoblots of cell lysates from ALK1510-c4 cells described in Fig. 8a after treatment for 24 h. d Mean (SD) change in tumor volume of mice bearing xenograft tumors with ALK1510-c4 cells and treated with vehicle, alectinib (2 mg/kg) + dacomitinib (10 mg/kg), alectinib (2 mg/kg) + G007-LK (10 mg/kg), and their triple combination for 14 days (P < 0.05 versus alectinib + dacomitinib (a) or versus alectinib + G007-LK (b) ; Wilcoxon rank sum test by the Holm–Bonferroni method; n = 8 mice per group). e Immunoblots of tumor lysates evaluating ALK-signaling and Wnt/β-catenin-signaling from xenograft tumors with ALK1510-c4 cells in mice treated with alectinib, alectinib + dacomitinib, alectinib + G007-LK, and their triple combination for 2 days. f Mean (SD) change from baseline in body weight of mice described in Fig. 8d. g ALK1510-c4 cells were treated with vehicle, alectinib (1000 nM), alectinib in combination with dacomitinib (100 nM) or AZ6102 (100 nM); and a triple combination of alectinib + dacomitinib + AZ6102 for 5 weeks, and then washed after 5 weeks. Cell numbers were measured using a cell counter at weeks 1, 2, 3, 4, 5, and 11. If the cell number was 1 × 10⁴ or less, it was indicated and considered a limitation of the cell counter. The cell counts were terminated at the point (†) when ALK1510-c4 cells detached from the culture flask due to over-confluence (n = 1). h ALK1510-c4 cells were sequentially treated with alectinib (1000 nM) in combination with dacomitinib (100 nM) or AZ6102 (100 nM), and a triple combination of alectinib + dacomitinib + AZ6102 for 5 weeks (left). Cell numbers were measured using a cell counter at weeks 1, 2, 3, 4, and 5 (right). If the cell number was 1 × 10⁴ or less, it was indicated and considered a limitation of the cell counter (n = 1).