Normal human lung epithelial cell line (BEAS-2B), human lung adenocarcinoma cell line (H1650) with EGFR mutation (del L747_E749 and A750P), human lung adenocarcinoma cell line (H1975) with EGFR mutation (L858R), human lung adenocarcinoma cell lines (A549 and H292) with wild EGFR were obtained from American Type Culture Collection (Manassas, VA, USA). All cell culture reagents were purchased from Invitrogen Corporation (Carlsbad, CA, USA). Briefly, the cells were cultured in RPMI-1640 supplemented with 10% fetal bovine serum under standard conditions (37°C in a humidified atmosphere containing 5% CO₂)

To evaluate the effect of AG1478 on the EGFR signaling pathway and miR-145 expression levels, cells were serum starved for 24 h, incubated in the presence or absence of AG1478 (5 μM; Calbiochem) for 2 h, and then for an additional 24 h in the presence or absence of EGF (20 ng/ml; Promega). U0126 (Cell Signaling Technology, Inc., Beverly, MA, USA), a pharmacological MEK1/2 inhibitor, was dissolved in dimethyl sulfoxide (DMSO) at a concentration of 10 mM and stored at −20°C until used (10 μM). LY194002, PI3K kinase inhibitor and AG490, JAK2 inhibitor, were purchased from Calbiochem and dissolved in DMSO at a recommend concentration and stored at −20°C.

Cells were washed once with PBS and then lysed in buffer containing 10 mM Tris, pH 7.6, 150 mM NaCl, 5 mM EDTA, pH 8.0, 10 ml/l Triton X-100, 1 mM DTT and 0.1 mM phenylmethanesulfonyl fluoride (PMSF). After 30 min on ice, lysates were collected and clarified by centrifugation at 15,000 × g for 5 min at 4°C. Protein concentrations were measured by BCA protein detection kit (Pierce, Rockford, IL, USA). Equal amounts of protein (10–30 μg/lane) from whole-cell lysates were separated by gel electrophoresis on 10% gels, transferred to nitrocellulose membranes and were probed with specific primary antibody [phospho-EGFR (Tyr1068) rabbit mAb; phospho-AKT (Ser473) rabbit mAb; phospho-ERK1/2 (Thr202/Tyr204) rabbit mAb; phospho-STAT3 (Tyr705) rabbit mAb; EGFR rabbit mAb; ATK rabbit mAb; ERK1/2 rabbit mAb; STAT3 rabbit mAb; Cell Signaling Technology, Inc.] and then with the appropriate HRP-conjugated secondary antibodies. Proteins were detected using the enhanced chemiluminescence detection kit (Thermal Science, Rockford, IL, USA). For loading control, the membrane was probed with a monoclonal antibody for GAPDH (Kangchen Biotechnology, Shanghai).

Total RNA was isolated from the cultured cells with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions and quantified by a spectrophotometer. To assess the purity of RNA, optical density (OD) was measured at 260 and 280 nm for determination of OD₂₆₀/OD₂₈₀ ratio. The RNAs with >1.8 OD ratios were used in this study. Relative abundance and integrity of 18s and 28s ribosomal bands were assessed with formaldehyde denaturing agarose gel. Those RNAs that exhibited intact 18s and 28s ribosomal bands with 1:1.5 relative abundance ratios were used in the present study.

Expression of mature miRNAs was examined by qRT-PCR analysis using a Real-time PCR Universal Reagent kit according to the manufacturer’s instructions (GenePharma, Co., Ltd., Shanghai). In brief, two-step qRT-PCR procedure was performed. Reverse transcription was performed in 20 μl volume, starting with 10 ng of total RNA. The reaction mixture was initially heated to 25°C for 30 min, 42°C for 45 min, 85°C for 10 min and finally to 4°C for 5 min. In the PCR step, PCR products were amplified from cDNA samples using specific miRNA primers and all assays were performed in triplicate on the MX3000P Real-time PCR Instrument (Stratagene, USA). The primers used were: miR-145, 5′-ATC GTC CAG TTT TCC CAG G-3′ (forward), and 5′-CGC CTC CAC ACA CTC ACC-3′ (reverse); U6 snRNA, 5′-ATT GGA ACG ATA CAG AGA AGA TT-3′ (forward), and 5′-GGA ACG CTT CAC GAA TTT G-3′ (reverse). The assay tubes were initially heated to 95°C for 3 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 60 sec. The expression levels of candidate miRNAs were evaluated by the comparative CT method and were normalized using U6 snRNA as the endogenous control. Relative quantitative expression levels of miRNAs were determined by the 2^−ΔΔCT method.

The human cell lines were seeded in a 6-well plate and cultured to 70% confluence. siRNA oligonucleotides targeting STAT3, AKT, ERK1/2 and negative control were purchased from GenePharma. The siRNA sequences were: STAT3 siRNA sense, 5′-GCA GCA GCU GAA CAA CAU GTT-3′ and antisense, 5′-CAU GUU GUU CAG CUG CUG CTT-3′; AKT siRNA sense, 5′-UGC CCU UCU ACA ACC AGG ATT-3′ and antisense, 5′-UCC UGG UUG UAG AAG GGC ATT-3′; ERK1 siRNA sense, 5′-GAC CGG AUG UUA ACC UUU ATT-3′ and antisense, 5′-UAA AGG UUA ACA UCC GGU CTT-3′; ERK2 siRNA sense, 5′-CAC CAA CCA UCG AGC AAA UGT T-3′ and antisense, 5′-CAU UUG CUC GAU GGU UGG UGT T-3′ and negative control sense, 5′-UUC UCC GAA CGU GUC ACG UTT-3′ and antisense, 5′-ACG UGA CAC GUU CGG AGA ATT-3′. Cells were transfected with siRNA (100 nM) by using Lipofectamine 2000 (Invitrogen) following the manufacturer’s protocol. The selective silencing of STAT3, AKT, ERK1/2 was confirmed by western blot analysis.

All experiments were repeated a minimum of three times. Error bars indicate standard errors of mean. Microsoft Excel or Instat software (GraphPad Prism4; San Diego, CA, USA) was used to analyze the data. A Student’s t-test or one-way analysis of variance (ANOVA) was used for parametric data. Correlation analysis was made by using Pearson’s correlation coefficient. P<0.05 was considered to indicate a statistically significant difference.

The levels of miR-145 in human normal lung bronchial epithelial cells (BEAS-2B) and lung cancer cells (H1975, H1650, A549 and H292) were determined by qRT-PCR. The results showed that the levels of miR-145 in four lung cancer cell lines were significantly lower than in normal lung epithelial cells (P=0.025; Fig. 1A). Although there were no significant changes in EGFR protein between human normal lung bronchial epithelial cells and lung cancer cells, EGFR activity in the form of pi-EGFR expression in lung cancer cells was higher than in normal lung epithelial cells (P=0.032; Fig. 1B). The quantitative comparison of miR-145 and p-EGFR levels showed a significant negative correlation between these two factors (Pearson’s correlation, r=−0.926, P<0.05). These results suggest that the activated EGFR signaling pathway may be functionally associated with miR-145 downregulation.

To further investigate the effect of EGFR status on miR-145 expression, EGFR-mutant H1975 cells, EGFR wild-type A549 cells and human normal lung BEAS-2B cells were first starved for 24 h with serum free culture and then treated with AG1478 (a specific inhibitor of EGFR) in the presence or absence of EGF (12,13). The results showed that there was no change in the total EGFR protein. In the three cells, the levels of p-EGFR protein were significantly increased in the cells treated with EGF, compared with the respective control group. The AG1478 completely blocked the activation of EGF to EGFR (Fig. 2A, C and E). The levels of miR-145 were significantly decreased in the treatment of EGF. The AG1478 may restore the downregulation of EGF to miR-145 in A549, H1975 and BEAS-2B cells (Fig. 2B, D and F).

Due to the mutation of EGFR in H1975 cells, the EGFR signaling pathway is highly activated (14). This was confirmed by detecting the EGFR downstream signaling molecules STAT3, ERK1/2 and AKT. STAT3, ERK1/2 and AKT activity was inhibited when AG1478 blocked EGFR phosphorylation in H1975 cells (Fig. 3A). However, it was not clear whether these signaling molecules were involved in the process of miR-145 downregulation by EGF-EGFR. Therefore, H1975 cells were treated with the three signaling molecule inhibitors AG490, LY294002 and U0126 (15–17). The data obtained indicated that the levels of miR-145 were upregulated after the inhibitors blocked the corresponding signal molecules (Fig. 3B–G).

Although the AG490, a relatively specific chemical inhibitor for STAT3, may not inhibit the phosphorylation of AKT, it reduces the levels of p-ERK1/2 (18). Thus, we hypothesized that the levels of miR-145 in lung cancer cells may be restored by the non-specific inhibition of AG490. In order to determine whether STAT3 signaling molecules are involved in the regulation of miR-145 expression, the siRNA against STAT3 was used to treat H1975 cells. The results showed that the levels of miR-145 presented almost no change while STAT3 expression and its phosphorylation levels were suppressed; there was also no effect on p-ERK1/2 (Fig. 4). These observations suggest that STAT3 signaling molecules were not involved in the downregulation of miR-145 by the activation of EGFR in lung cancer cells.

LY294002 is a potent inhibitor of phosphoinositide 3-kinases (PI3Ks), an upstream molecule of AKT (19). In H1975 cells treated with LY294002, the activation of AKT and ERK1/2 was effectively suppressed, but LY294002 did not inhibit the STAT3 activity (Fig. 5A). At the same time, the miR-145 levels were increased by LY294002 in H1975 cells. Therefore, we hypothesized that the levels of miR-145 in lung cancer cells may be restored by the pi-ERK1/2 inhibition of LY294002. To explore the role of AKT in the regulation of miR-145, the siRNA against AKT was used to treat H1975 cells. The data demonstrated that ERK1/2 phosphorylation levels were blocked after the specific inhibition of the activation of AKT signaling molecules (Fig. 5B). We detected that miR-145 expression was upregulated in the AKT siRNA group (Fig. 5C). This indicated that the miR-145 levels were regulated by AKT via the activation of ERK1/2.

U0126, is a highly selective inhibitor of both MEK1 and MEK2, a type of MAPK/ERK kinase (20,21). In H1975 cells treated with U0126 for 24 h, western blot results confirmed that U0126 only inhibited the activation of ERK1/2, and there were no inhibitory effects on the AKT and STAT3 signaling molecules (Fig. 6A). Similarly, in applications for ERK1/2 siRNA in H1975 cells, no effect on the activation of AKT and STAT3 was observed, while the ERK1/2 activity was specifically silenced (Fig. 6B). The miR-145 levels were upregulated, compared with the control group and negative control group (Fig. 6C). These findings suggested that ERK1/2 mediated the downregulation of miR-145 in the EGF-EGFR signaling pathway in lung cancer cells.