Fig 1.
Reporter gene assay for transcription factors influenced by EGFR-TKIs.
(a) Luciferase reporter assay. The pGL4 vectors, which encode response elements for transcription factors (indicated) or a minimal promoter (minP; a negative control) linked with the Photinus luciferase gene, were transfected together with phRL-TK carrying the Renilla luciferase gene (as a reference) into PC-9 cells. After 24h-incubation, the cells were exposed to 10μM gefitinib or DMSO (a vehicle) for 6h and subjected to a dual-luciferase assay. The activity of the Photinus luciferase (P luc.) was normalized to that of the Renilla luciferase (R luc.). Data are shown as mean ± standard deviation (SD) (n = 4). Significant difference between gefitinib and DMSO was examined by means of two-tailed Student’s t-test (*<0.05, **p<0.01, ***p<0.001). (b, c) Effects of gefitinib or erlotinib on transcription factors. The GL4-plasmid carrying the NF-κB response element (b) or the serum response element (SRE) (c) was co-transfected with phRL-TK into PC-9 cells, and after 24h-incubation, the cells were treated with 0, 0.1, 1.0 and 10μM of gefitinib or erlotinib for 6h. Dual-luciferase assay followed by normalization were carried out as in a. Normalized data were further normalized to the data obtained from cells treated with no EGFR-TKIs (0 μM) as 1. Data are shown as mean ± standard deviation (SD) (n = 3).
Fig 2.
Effects of extracellular FGF2 on cell survival and transcriptional activation.
(a) ELISA. PC-9 cells were treated with various concentrations of gefitinib (indicated) for indicated days, and conditioned media were collected. The FGF2 level in the media was examined by an ELISA analysis. Data are shown as mean ± SD (n = 3). (b) Effects of PD173074 (PD), an FGFR inhibitor, on cell viability under the presence of gefitinib. PC-9 cells were treated with 0.5 and 1.0μM of PD173074 or DMSO (a vehicle) for 1h prior to exposure to 0, 0.1, 1.0 and 10μM of gefitinib. After 3day-incubation, cell viability was examined. Data are shown as mean ± SD (n = 3). The data of DMSO represent the results treated only by gefitinib. Significant difference between PD173074 and DMSO, i.e., between the presence and absence of PD173074, was examined by one-way analysis of variance (ANOVA) (Dunnett’s test; ***p<0.001). (c) Effects of extracellular FGF2 on cell viability under the presence of gefitinib. PC-9 cells were treated with FGF2 or DMSO together with gefitinib, and cell viability was examined and analyzed as in b (n = 3). (d) Inhibition of extracellular FGF2 by neutralizing antibodies. PC-9 cells were treated with 0.1 or 1μM of gefitinib under the presence of various levels of anti-FGF2 antibodies (α-FGF2). After 3day-incubation, cell viability was examined as in b. Significant difference between α-FGF2 and a control IgG was examined by Student’s t-test (two-tailed; *<0.05, **p<0.01, ***p<0.001). (e, f) Effects of extracellular FGF2 on transcription factors. The pGL4-NF-κB-RE and pGL4-SRE plasmids were introduced together with phRL-Tk into PC-9 cells. After 24h-incubation, the cells were treated with various levels of FGF2, and further incubation was carried out. After 6h, dual-luciferase assay and data analyses were carried out as in Fig 1b (n = 3). (g) Influence of PD173074 on NF-κB activation. The pGL4-NF-κB-RE and phRL-Tk plasmids were transfected into PC9 cells as in e, and the cells were treated with PD173074 for 1h prior to gefitinib treatment. After 6h gefitinib treatment, dual-luciferase assay was carried out as in Fig 1b (n = 3).
Fig 3.
Effect of gene silencing against EGFR on NF-κB activation.
(a) RNAi knockdown. PC-9 cells were subjected to gene silencing with siRNAs against total EGFR (siEGFR) or oncogenic mutant EGFR alleles [si746/50_3D10 (si3D10) targeting the 746/750 deletion in exon 19] in PC-9 cells. Non-silencing siRNAs (siControl: siC) were also examined as a negative control. 24h after introduction of siRNAs, EGFR was examined by western blotting. Alpha-tubulin was examined as an internal control. (b) EGFR mRNA levels. Indicated siRNAs were transfected into PC-9 cells as in a. The level of EGFR was examined by RT-qPCR and analyzed by the delta-delta Ct method using the data of Gapdh as a reference. The data were further normalized to the data obtained by siControl (siC) as 1. Data are shown as mean ± SD (n = 3). (c) Activation of NF-κB. The pGL4-NF-κB-RE and phRL-Tk plasmids together with indicated siRNAs (20nM) were introduced into PC-9 cells. After 24h incubation, Dual-luciferase assay was carried out as in Fig 1b (n = 6). (d) Effect of gefitinib on NF-κB activation under EGFR knockdown. Reporter plasmids and siRNAs were introduced into PC9 cells as in c. The cells were treated with 0, 0.1, 1.0 and 10μM of gefitinib (Gef). Dual-luciferase assay was carried out as in Fig 1b (n = 6). Significant difference in cell viability between gefitinib and DMSO treatment was examined by one-way analysis of variance (ANOVA) (Dunnett’s test; *p<0.05). n.s.: not significant.
Fig 4.
Effects of MG132 on gefitinib-induced NF-κB activation and cell viability.
(a) Phosphorylation of IκBα after gefitinib treatment. PC-9 cells were treated with 80μM of gefitinib (gef.), and cell lysates were prepared at indicated times. Phosphorylated IκBα (phospho IκBα) and α-Tubulin as an internal control were examined by western blotting. (b) Inhibition of IκBα degradation by MG132. PC-9 cells were treated with 10μM of MG132 or DMSO (vehicle) for 1h prior to 10μM gefitinib treatment. Total IκBα and α-Tubulin were examined by western blotting. (c) Reporter gene assay. The pGL4-NF-κB-RE and phRL-Tk plasmids were transfected into PC-9 cells. After 24h-incubation, the cells were treated with 1μM of MG132 or DMSO for 1h and then treated with 10μM of gefitinib and/or 1.0μM of PD173074 for 6h. The expression of reporter genes was examined and analyzed as in Fig 1b. Data are shown as mean ± SD (n = 4). (d) Cell viability. PC-9 cells were treated with MG132 or DMSO (as a control) for 1h and then treated with 0, 0.1, 1.0 and 10μM of gefitinib. After 3days of gefitinib treatment, cell viability was examined. Data are shown as mean ± SD (n = 3). Significant differences in cell viability between MG132 and DMSO treatment were analyzed by one-way analysis of variance (ANOVA) (Dunnett’s test; *p<0.05, **p<0.01).
Fig 5.
Schematic drawing of hypothetical early events in EGFR-TKIs-exposed PC-9 cells.
After EGFR-TKIs treatment, many PC-9 cells are killed and the dead cells leak out FGF2 that works as a survival factor; and simultaneously, neighboring (alive) cells manage to survive from the first exposure of EGFR-TKIs by leaked-out FGF2 and also by NF-kB activation. FGF2 can trigger the initiation of the FGFR-signaling pathway that is a surrogate pathway for suppressed EGFR pathway. The FGFR-signaling pathway and NF-kB independently contribute to cells for survival in the early stages of EGFR-TKIs treatment. Cells that survived successfully, thereafter, need to change gene regulation to maintain resistance to EGFR-TKIs; and NF-kB may play an important role as an initial transcription factor there. Finally, the cells acquire complete drug resistance, in which FGF2-FGFR autocrine system may work as a major vital pathway.