Doxorubicin, paclitaxel, vincristine, cisplatin and 3-(4,5-dimethylthiazol-yl)-2,5-diphenyltetrazolium bromide (MTT) were from Sigma-Aldrich, Inc. (Shanghai, China). Dulbecco's modified Eagle's medium (DMEM) and RPMI-1640 medium were from Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA).

The human H1299, A549 and PC9 NSCLC cell lines, and normal lung cells (as the control), were purchased from the Cancer Research Institute of China Medical University (Shenyang, China). The cells (1×10⁵ cells/ml) were cultured in DMEM supplemented with 10% (m/v) fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin (Sangong Biotech, Shanghai, China) at 37°C in a 5% CO₂ atmosphere with stable humidity.

Cell viability was assessed using the MTT assay, as described in a previous study (20). Briefly, the cells were plated at a density of 3,000 cells/well into 96-well plates. At the end of treatment, the supernatant was removed using a pipette and 20 µl MTT and 270 ml fresh DMEM were added to the supernatant. Following incubation for 4 h at 37°C, 120 µl dimethyl sulfoxide was placed in each well to dissolve the tetrazolium crystals, after which the absorbance at 570 nm was recorded using a Multi-Well Plate Reader (Tecan Group, Ltd., Männedorf, Switzerland). Each experiment was performed four times. The results are presented as the percentage growth inhibition with respect to the untreated cells.

Double-strained small interfering RNA (siRNA) probes targeting the human AXL gene (GenBank accession no. NM-021913) and the Twist gene (GenBank accession no. NM_000474.3) were synthesized using the Silencer siRNA Construction kit (Ambion; Thermo Fisher Scientific, Inc.), as described in a previous study (21). Briefly, single-stranded gene-specific sense and antisense RNA oligomers were synthesized by in vitro transcription using T7 RNA polymerase (Sangong Biotech). In order to promote annealing of the siRNA, sense and antisense single-stranded RNA oligomers were mixed and incubated at 37°C overnight, after which the siRNA was ethanol-precipitated and resuspended in nuclease-free water. The integrity of the siRNA was assessed by gel electrophoresis (Sangong Biotech), and it was quantified of by measuring the absorbance at 260 nm. Subsequently, the cells were transfected with four pooled siRNA duplexes (20 nM; Invitrogen; Thermo Fisher Scientific, Inc.) using the TransIT-TKO^® Transfection Reagent (Mirus Bio, LLC, Madison, WI, USA), in order to target the endogenous AXL and Twist genes. The cells were transfected twice at 3-day intervals, and were then collected at 72 h following the second transfection. Specific silencing of the target genes was confirmed using qPCR and western blot analysis. A mock transfection with the transfection reagent alone served as the control.

Log phase PC9 cells were collected at a final concentration of 2×10⁵ cells/ml, washed with phosphate-buffered saline (PBS) and fixed with 70% ethanol. The cells were centrifuged at 8,000 × g for 5 min at 4°C to remove ethanol, washed with PBS, and stained with propidium iodide (PI; Sangong Biotech) in the dark for 30 min prior to FCM analysis. The FACSCalibur™ platform (BD Biosciences, Franklin Lakes, NJ, USA) was used to detect the cell cycle distribution. The cells were sampled using the CellQuest Pro software, version 3.0 (BD Biosciences), and the proportion of cells in the various stages of the cell cycle were quantified using ModFit LT 3.0 (Verity Software House, Inc., Topsham, ME, USA) (22). Each experiment was performed four times.

PC9 cells (1×10⁶) were collected by centrifugation at 8,000 × g for 5 min at 4°C. The cell pellets were lysed using DNA lysis buffer (10 mM Tris, pH 7.5, 400 mM EDTA and 1% Triton X-100), followed by centrifugation at 6,000 × g for 8 min at 4°C. The supernatant was incubated overnight with proteinase K (0.1 mg/ml; Sangong Biotech), then with RNase (0.2 mg/ml; Sangong Biotech) for 2 h at 37°C. DNA was extracted using phenol:chloroform (1:1), separated by 2% agarose gel electrophoresis (Sangong Biotech), and visualized by ultraviolet illumination after staining with 10% ethidium bromide and washing twice with water. Quantitative assessment of apoptotic cells was conducted using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) method (BD ApoAlert DNA Fragmentation Assay kit; cat. no. 630107; BD Biosciences), which examines DNA-strand breaks during apoptosis.

RNA extraction from the cells was carried out using TRIzol reagent (Life Technologies; Thermo Fisher Scientific, Inc.) according to the manufacturer's recommendations. Genomic DNA was eliminated and RT conducted using the PrimeScript RT Reagent kit with gDNA Eraser, and qPCR was conducted using the SYBR Green PCR kit (both Takara Biotechnology Co., Ltd., Dalian, China) according to the manufacturer's recommendations. The process was conducted using a 7500 Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc. The primers used in the PCR were as follows: AXL forward, 5′-TCGCCAGTGGCATGGAGTATCTG-3′ and reverse, 5′-GGCGATACGTCCCTGGCGGTA-3′; Twist forward, 5′-CATCCACACCGTCCCCTCCC-3′ and reverse, 5′-GACTGGCGAGCTGGACACGTC-3′; and β-actin forward, 5′-TGAAGTACCCCATCGAGCAC-3′ and reverse, 5′-CTTGGGGTTCAGGGGGGCCT-3′ (GenBank accession no. BC004251.1). β-actin was used as an internal reference.

The reaction mix consisted of 4 µl cDNA, 0.5 µl forward and reverse primer mix (20 µM of each), 1 µl 50X ROX Reference Dye II and 25 µl 2X SYBR Green PCR mix in a final volume of 50 µl. All reactions were setup in triplicate and every sample was replicated in parallel three times to ensure statistical relevance. A negative control without cDNA template, and RT control without RT transcription were included. The following thermal conditions were used for all PCR reactions: 30 sec at 95°C, followed by 40 cycles of 30 sec at 95°C and 34 sec at 60°C. Primer specificity was confirmed by RT-PCR amplification prior to qPCR, which generated single amplicons of the expected size for each primer set. Furthermore, the amplicons were sequenced by a commercial sequencing service (BGI, Shenzhen, China) in order to validate their specific amplification, and the specificity of qPCR was confirmed by the presence of dissociation curves with single peaks and the sequencing of its products with unique bands of the expected size. Amplicon dissociation curves were obtained after cycle 40 using default settings suggested by the instrument. Data were analyzed using the SDS software (version 2.0.6; Applied Biosystems; Thermo Fisher Scientific, Inc.). All quantifications were normalized against the quantity of β-actin using the 2^−ΔΔCq method (23).

Cell lysates were prepared in radioimmunoprecipitation assay buffer [50 mM Tris-HCl buffer, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS)], supplemented with 1X Halt protease inhibitor and 1X Halt phosphatase inhibitor cocktails (Pierce Biotechnology, Inc., Rockford, IL, USA). The Bio-Rad protein assay (cat. no. 500-0006; Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used to determine protein concentrations. Proteins (100 ng) were separated by 10–12% SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes (GE Healthcare Life Sciences, Chalfont, UK). The membranes were blocked for 1 h at room temperature using bovine serum albumin (BSA; Sangong Biotech). Subsequently, the membranes were incubated with specific primary antibodies (1:100 dilution) targeting Axl (cat. no. 4566), Twist (cat. no. 46702) and β-actin (cat. no. 4790; all from Cell Signaling Technology, Inc., Danvers, MA, USA) at 4°C overnight. After washing the membrane three times with TBST (5 min per wash), it was incubated with the recommended dilution (1:100) of horseradish peroxidase-conjugated secondary antibodies (cat. no. 7071; Cell Signaling Technology, Inc.) at room temperature for 1 h, and then washed as described above. The protein bands were visualized using the Immobilon Western Chemiluminescent HRP Substrate (EMD Millipore, Billerica, MA, USA). The chemiluminescence of proteins transferred to the PVDF membranes was detected using the ECL Plus Western Blot analysis Reagent (GE Healthcare Life Sciences). Relative protein expression levels were semi-quantified by densitometry using ImageJ software (version 2.1.4.7; National Institutes of Health, Bethesda, MA, USA).

Data were analyzed using SPSS statistical software, version 16.0 (SPSS, Inc., Chicago, IL, USA). The results are expressed as the mean ± standard deviation. The mean was compared using the Student's t-test or by analysis of variance. The statistical significance of the studies was determined using the parametric unpaired Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

To evaluate the expression of AXL in the NSCLC cell lines, RT-qPCR and western blot analysis were conducted. AXL protein expression levels were markedly upregulated in the NSCLC cell lines compared with the control (Fig. 1A). Consistent with this, the mRNA expression levels of AXL were significantly upregulated in the NSCLC cells, as compared with the control (P<0.05; Fig. 1B). PC9 exhibited the highest level of AXL transcripts (~2-fold upregulation), as compared with the control; the AXL transcripts were ~1.7-fold upregulated in the A549 cells and 1.3-fold upregulated in the H1299 cells, as compared with the control (Fig. 1B). These results suggest that AXL is upregulated in NSCLC cells.

To characterize the role of AXL in NSCLC cells, a gene-silencing experiment using double-stranded siRNA to silence AXL in the PC9 cells was conducted. PC9 cells were selected for further experimentation, as the greatest upregulation of AXL expression levels was observed in this cell line. PC9 cells were transfected with the siRNA and, after 3–7 days, RNA was extracted from control and transfected cell lines. A marked reduction was detected in the mRNA and protein expression levels of AXL in the siRNA-transfected cells, as compared with the control cells (Fig. 2A and B), and this was shown to be significant using RT-qPCR (P<0.01; Fig. 2C). In particular, ~1.83-fold downregulation of AXL was detected in the transfected cells, as compared with the control (Fig. 2). These results suggested that knockdown of AXL expression in the PC9 cells was successful.

The viability of the PC9 cells was analyzed using the MTT assay. Downregulation of endogenous AXL in PC9 cells significantly reduced cell survival (~29.3%), as compared with the control (P<0.05; Fig. 3). These results suggest that inhibition of AXL by RNAi may reduce the survival of NSCLC cells.

AXL is involved in NSCLC cell apoptosis. To further investigate the involvement of AXL in the inhibition of cell proliferation, cell-cycle distribution was evaluated using FCM analysis. As compared with the control group, the number of cells in the G₁/G₀ phase was significantly increased (P<0.05), and the number of cells in the S and G₂/M phases was significantly decreased (P<0.05), in the PC9 cells treated with AXL siRNA; the majority of cells were in the G₁ phase (Table I). Furthermore, the proportion of diploid cells slightly increased, with a reduction in the number of aneuploid cells, and the cell debris was markedly increased in the PC9 cells treated with AXL siRNA, as compared with the control group.

The effect of downregulating AXL expression on the induction of apoptosis in PC9 cells was investigated using a DNA fragmentation assay. Agarose gel electrophoresis demonstrated that knockdown of AXL resulted in the formation of DNA fragments in PC9 cells (Fig. 4A). In addition, a quantitative evaluation was conducted using TUNEL to detect DNA-strand breaks. As compared to with the control cells, 30.7% of PC9 G₂ cells treated with AXL siRNA were apoptotic cells (Fig. 4B). These results suggest that interfering with AXL expression may inhibit cell-cycle progression in PC9 G₂ cells, thereby resulting in a marked increase in the percentage of cells in the G₁ phase.

The acquisition of mesenchymal cell characteristics by epithelial cells, in particular the ability to migrate as single cells and invade the extracellular matrix, is the functional hallmark of EMT (24). EMT-inducing genes that have essential roles in EMT, including Twist and Snail, are termed master EMT genes (25). These genes function as transcriptional repressors of the cell-cell adhesion glycoprotein, E-cadherin whose functional loss is one of the hallmarks of EMT (26). Therefore, the status of Twist expression in PC9 cells was investigated in the present study. PC9 cells that were stably expressing Twist exhibited an EMT phenotype characterized by upregulation of the mesenchymal marker N-cadherin (Fig. 5A), and downregulation of the epithelial markers E-cadherin and β-cadherin (Fig. 5B). Notably, Axl was markedly upregulated in the Twist-expressing PC9 cells (Fig. 5C). Conversely, in the PC9 cells treated with AXL siRNA, downregulation of Twist and N-cadherin (Fig. 5A), and upregulation of E-cadherin and β-cadherin (Fig. 5B) were observed. However, knockdown of Twist did not affect the expression levels of AXL (Fig. 5C), despite the fact that E-cadherin and β-cadherin were upregulated (Fig. 5B). These results suggest that downregulation of AXL may abrogate EMT program induction by inhibiting the expression of Twist in NSCLC cells.