A low microRNA-630 expression confers resistance to tyrosine kinase inhibitors in EGFR-mutated lung adenocarcinomas via miR-630/YAP1/ERK feedback loop

Abstract

Purpose: MicroRNA-630 plays dual roles in apoptosis and drug resistance in human cancers. However, the role of miR-630 in resistance to tyrosine kinase inhibitors (TKIs) in lung adenocarcinoma remains to be elucidated. Methods: Manipulation of miR-630 and its targeted gene YAP1 and/or combination of inhibitor treatments was performed to explore whether low miR-630 could confer TKI resistance due to de-targeting YAP1, and this could decrease proapoptotic protein Bad expression through the miR-630/YAP1/ERK feedback loop. A retrospective study was conducted to examine whether the expression of miR-630 and YAP1 could be associated with TKI therapeutic response in patients with lung adenocarcinoma. Results: Low miR-630 expression may confer TKI resistance via increased SP1 binding to the miR-630 promoter due to ERK activation by YAP1 de-targeting. Persistent activation of ERK signaling via the miR-630/YAP1/ERK feedback loop may be responsible for TKI resistance in EGFR-mutated cells. Moreover, a decrease in Bad expression by its phosphorylation at Serine 75 through ERK activation conferred low miR-630-mediated TKI resistance by modulating the apoptotic pathway. Xenographic tumors induced by miR-630-knockdown PC9 and PC9GR cells in nude mice were nearly suppressed by the combination of gefitinib with the YAP1 inhibitor verteporfin or an MEK/ERK inhibitor AZD6244. Patients with low miR-630 and high YAP1 expressing tumors had a higher prevalence of unfavorable responses to TKI therapy and poorer outcomes when compared with their counterparts. Conclusion: MiR-630 may be a potential biomarker for the prediction of TKI therapeutic response and outcome in patients with lung adenocarcinoma.

Keywords: MiR-630; TKI resistance; YAP1; and lung adenocarcinoma.

Figure 1

MiR-630 expression levels are associated with gefitinib resistance and upregulation of miR-630 expression by gefitinib may occur by decreased SP1 binding to the miR-630 promoter due to ERK inactivation in PC9 cells, but not in PC9GR cells. (A) Six lung adenocarcinoma cells were treated with four concentrations of gefitinib. After 24 h, the dose-response curves, determined by the MTT assay, were used to calculate the IC50 values of these cells. MiR-630 expression of these cells was evaluated by real-time PCR. The expression of p-AKT, total AKT, p-ERK, total ERK and β-actin was evaluated by western blotting. (B) MiR-630 inhibitors were transfected into PC9 cells. MiR-630 mimic was transfected into low miR-630 expressing PC9GR cells. After 24 h, the cells were treated with four concentrations of gefitinib to calculate the IC50 values. NC: nonspecific shRNA control. VC: Vector control. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to NC (*P < 0.05). (C) PC9 and PC9GR cells were transfected with miR-630 inhibitors and mimic for 24 h. These cells were treated with 0.1% DMSO or 10 μM gefitinib for 24 h and then subjected to annexin-V and PI staining, followed by flow cytometry analysis. The percentages of apoptotic cells in the annexin V+/PI- population plus annexin-V+/PI+ are summarized. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to NC (*P < 0.05) for different treatments. (D) PC9 and PC9GR cells were treated with two concentrations of gefitinib for 5 h and then the cell lysates were evaluated for expression of p-EGFR, p-AKT, total AKT, p-ERK, total ERK, EGFR, and β-actin by western blotting. MiR-630 expression of these cells was evaluated by real-time PCR. The DNA binding activity of SP1 onto the miR-630 promoter was evaluated by ChIP. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to vehicle treatment (*P < 0.05). (E) PC9GR and PC9 cells were treated with or without ERK (10 μM; U0126 (U0)) and PI3K inhibitor (10 μM; LY294002 (LY)) for 5 h. MiR-630 expression of these cells was evaluated by real-time PCR. The DNA binding activity of SP1 onto the miR-630 promoter was evaluated by ChIP. (F) PC9GR and PC9 cells were transfected with shSP1 for 24 h, followed by treatment with or without ERK (10 μM; U0126 (U0)) for 5 h. The DNA binding activity of SP1 onto the miR-630 promoter was evaluated by ChIP. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to vehicle treatment (*P < 0.05).

Figure 2

The miR-630/YAP1/ERK feedback loop may be responsible for gefitinib resistance in EGFR-mutated lung adenocarcinoma cells. (A) The cell lysates of PC9 and PC9GR were evaluated for the expression of YAP1, Bcl-2, Slug, and IGF1R by real-time PCR. (B) MiR-630 inhibitor were transfected into PC9 cells. MiR-630 mimic was transfected into low PC9GR cells. After 48 h, the cells lysates were evaluated for the expression of YAP1, Bcl-2, Slug, and IGF1R by real-time PCR. (C) PC9 cells were transfected with the indicated combination of miR-630 inhibitor, shYAP1, shBcl-2, and shSlug for 24 h. PC9GR cells were transfected with the indicated combination of miR-630 mimic, YAP1, Bcl-2, and Slug overexpression plasmids for 24 h. These cells were treated with 0.1% DMSO or 10 μM of gefitinib for 24 h and were subjected to annexin-V and PI staining, followed by a flow cytometry analysis. The percentages of apoptotic cells in the annexin V+/PI- population plus annexin-V+/PI+ are summarized. (D) PC9 cells were transfected with the indicated combination of miR-630 inhibitor and shYAP1 for 48 h. PC9GR cells were transfected with the indicated combination of miR-630 mimic and YAP1 overexpression plasmids for 48 h. The cells lysates were evaluated for expression of p-AKT, total AKT, p-ERK, total ERK and β-actin by western blotting. (E) YAP1-overexpressing PC9 and PC9GR cells were treated with 10 μM AZD6244 for 5 h. The cell lysates were evaluated for the expression of YAP1, p-ERK, total ERK and β-actin by western blotting. MiR-630 expression of these cells were evaluated by real-time PCR. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to vehicle or indicated treatment (*P < 0.05). N.s.: Non-significant.

Figure 3

Decreased Bad expression by its phosphorylation at Serine 75 may be responsible for gefitinib resistance via the miR-630/YAP1/ERK feedback loop. (A) MiR-630 mimic was transfected into PC9GR cells. After 48 h, the cell lysates were evaluated for the expression of apoptotic protein using an apoptotic protein chip analysis. (B) PC9 cells were transfected with the indicated combination of miR-630 inhibitor and Bad overexpression plasmid for 24 h. PC9GR cells were transfected with the indicated combination of miR-630 mimic and shBad for 24 h. These cells were treated with 0.1% DMSO or 10 μM of gefitinib for 24 h and were subjected to annexin-V and PI staining, followed by a flow cytometry analysis. The percentage of apoptotic cells included with the annexin V+/PI- population plus annexin-V+/PI+ is summarized. (C) PC9 cells were transfected with the indicated combination of miR-630 inhibitor and shYAP1 for 48h. PC9GR cells were transfected with the indicated combination of miR-630 mimic and YAP1 overexpression plasmid for 48 h. The cells lysates were evaluated for expression of Bad, p-ERK, total ERK and β-actin by western blotting. (D) PC9 cells transfected with miR-630 inhibitors were treated with 5 μM verteporfin (VP) or 10 μM AZD6244 for 5 h. MiR-630-overexpressing PC9GR cells were treated with 10 μM AZD6244 for 5 h or transfected with YAP1 overexpression plasmid for 48 h. The cell lysates were evaluated for expression of Bad, pS75-Bad, pS99-Bad, p-ERK, total ERK and β-actin by western blotting. The Bad mRNA expression was evaluated by real-time PCR. (E) PC9 cells transfected with miR-630 inhibitors were treated with 5 μM MG132 for 5 h. MiR-630-overexpressing PC9GR cells were transfected with YAP1 overexpression plasmid for 43 h, followed by treatment with 5 μM MG132 for 5 h. The cell lysates were evaluated for expression of Bad, pS75-Bad, pS99-Bad, and β-actin by western blotting. (F) PC9 cells were transfected with indicated combinations of miR-630 inhibitor, wildtype (WT), and S75A mutant (Mut) Bad overexpression plasmids for 24 h. PC9GR cells were transfected with the indicated combination of wildtype (WT) and S75A mutant (Mut) Bad overexpression plasmids for 24 h. The cell lysates were evaluated for expression of Bad, pS75-Bad, p-ERK, total ERK and β-actin by western blotting. (G) PC9 cells were transfected with indicated combinations of miR-630 inhibitor, wildtype (WT), and S75A mutant (Mut) Bad overexpression plasmids for 24 h. PC9GR cells were transfected with the indicated combination of miR-630 mimic, wildtype (WT), and S75A mutant (Mut) Bad overexpression plasmids for 24 h. These cells were treated with 0.1% DMSO or 10 μM of gefitinib for 24 h and were subjected to annexin-V and PI staining, followed by flow cytometry analysis. The percentage of apoptotic cells included with the annexin V+/PI- population plus annexin-V+/PI+ is summarized. P value was calculated by the Student's t-test. The significant differences in experimental groups were compared to vehicle or indicated treatment (*P < 0.05). N.s.: Non-significant.

Figure 4

The formation of subcutaneous tumors induced by miR-630-knockdown PC9 and PC9GR cells in nude mice was nearly completely suppressed by VP+gefitinib or AZD6244+gefitinib. (A) The PC9, PC9GR, and miR-630-overexpressing PC9GR xenografts were treated with vehicle and gefitinib (5 mg/kg). The PC9GR and miR-630-inhibiting PC9 xenografts were treated with vehicle, gefitinib (5 mg/kg), verteporfin (5 mg/kg), AZD6244 (5 mg/kg), or a combination. Representative tumor burdens in the ten groups are illustrated. The tumor volumes in the nude mice of the 8 groups were measured at 3-day intervals from Day 9 to Day 27. Mean ± SD values (mm³) were calculated from the tumor volume of five nude mice in each group. (B) The miR-630 expression in tumors of each group of nude mice were evaluated by real-time PCR. (C) Representative immunostaining results of YAP1, p-ERK, Bad, and cleavage caspase-3 in tumors of each group of nude mice. Scale bar is 50 µm (D) The possible model of miR-630 influence on gefitinib resistance in lung adenocarcinoma cells with EGFR mutation. P values were calculated by the Student's t-test. The significant differences in experimental groups were compared to vehicle or indicated treatment (*P < 0.05).