Cisplatin sensitivity is enhanced in non-small cell lung cancer cells by regulating epithelial-mesenchymal transition through inhibition of eukaryotic translation initiation factor 5A2

The human NSCLC cell lines, A549 and NCI-H1299, were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and stored following ATCC guidelines. All cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA) and 1% penicillin-streptomycin (Sigma-Aldrich, St. Louis, MO, USA). The cells were maintained at 37°C in a humidified atmosphere of 5% CO₂.

NSCLC cells were transfected with eIF5A-2 siRNA (10 μmol/mL; Santa Cruz Biotechnology, Dallas, TX, USA) or negative control siRNA (Invitrogen) using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. The transfection medium was replaced with culture medium 6 h after transfection. All subsequent experiments were performed 24 h after transfection and repeated in triplicate.

A Cell Counting Kit-8 (CCK8; Dojindo, Kumamoto, Japan) was used to measure relative cell viability after treatment. NSCLC cells (5 × 10³ cells/well) were seeded into 96-well plates and cultured for 24 h. The culture medium was replaced by medium containing the required concentrations of cisplatin or cisplatin combined with GC7, and the cells were incubated for 48 h. CCK-8 solution (10 μL/well) was added, the cells were incubated for a further 4 h, and absorbance was measured at 450 nm using an MRX II microplate reader (Dynex Technologies, Chantilly, VA, USA). Relative cell viability was calculated as a percentage of untreated controls.

The cells were washed twice in ice-cold phosphate buffer solution (PBS) and resuspended in 100 μL cell lysis buffer (Cell Signaling, Danvers, MA, USA) with protease inhibitors (Sigma-Aldrich). The protein concentrations were quantified using a BCA Protein Kit (Thermo Fisher, Rockford, IL, USA). Cell lysates (40 μg/lane) were separated by 10% SDS-PAGE, transferred to polyvinyl diflouride (PVDF) membranes (Millipore, Billerica, MA, USA) and blocked with Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBST) and 5% bovine serum albumin (BSA). The membranes were incubated with anti-E-cadherin, anti-Vimentin (Biovision, Milpitas, CA, USA) or anti-eIF5A-2 (Proteintech, Chicago, IL, USA) antibodies (1:1000) at 4°C overnight, washed three times with TBST and then incubated with the appropriate HRP-conjugated secondary antibodies for 1 h at room temperature. The protein bands were developed by chemiluminescence (GE Healthcare, Piscataway, NJ, USA) and visualized by autoradiography on X-Ray films (Kodak, Rochester, NY, USA). Band densities were estimated using Image-Pro Plus v. 6.0 software (Media Cybernetics, Bethesda, MD, USA) and protein levels were normalized to GAPDH.

Cells were washed with ice-cold PBS, fixed in 4% paraformaldehyde for 30 min followed by incubation with 3% H₂O₂ for 15 min at 37°C and blocked in fetal calf serum for a further 15 min. After incubation with anti-E-cadherin, anti-vimentin or anti-eIF5A-2 antibodies (1:1,000) overnight at 4°C, the cells were washed with ice-cold PBS and incubated for 1 h at room temperature with the appropriate secondary antibodies (1:2000; GE Healthcare): goat anti-mouse FITC-conjugated secondary antibody (E-cadherin) or goat anti-mouse Cy5-conjugated secondary antibody (vimentin). Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) and the cells were observed by fluorescence confocal microscopy (Olympus, Japan).

Cells were seeded into six-well plates at a density of 2 × 10⁵ cells/well and cultured with RPMI-1640 medium containing 10% FBS overnight at 37°C in a humidified atmosphere of 5% CO₂, after which, the medium was changed to RPMI-1640 without FBS and the cells were cultured for a further 24 h until >90% confluence. The cells were harvested by scraping the adherent cells using a plastic 100 μL tip. After transfection with eIF5A-2 siRNA (10 μmol/mL) or treatment with N1-guanyl-1,7-diaminoheptane (GC7; 20 μM) for 6 h, the cells were treated with transforming growth factor-β1 (TGF-β1) at a concentration of 10 ng/mL for 24 h at 37°C in a humidified atmosphere of 5% CO₂. Micrographs were taken using an inverted phase contrast microscope (Olympus; magnification, 40×) at 0 h and 24 h. The ratio of the remaining wound area relative to the initial wound area was calculated and the wound area was quantified using Image-Pro Plus v. 6.0 software.

After transfection with eIF5A-2 siRNA (10 μmol/mL) or treatment with GC7 (20 μM) for 6 h, the cells were treated with TGF-β1 (10 ng/mL) for 48 h. The cells were seeded at a density of 5 × 10⁴ cells/well in the upper chamber of a Transwell 24-insert plate with RPMI-1640 medium. The upper chambers were coated with Matrigel (BD Biosciences, San Jose, CA, USA) and the lower chamber contained RPMI-1640 plus 10% FBS medium. After 24 h, the bottom of the inserts were fixed in methanol for 10 min and stained with hematoxylin and eosin (H&E). The cells that had invaded to the lower surface were measured using an inverted phase contrast microscope (Olympus; magnification, 40×) and photographed.

Western blot analysis was used to determine eIF5A-2 protein expression in A549 and NCI-H1299 cells. The results showed that eIF5A-2 was expressed in the control cells of both cell lines; however expression was higher in NCI-H1299 cells compared to A549 cells (Figure 1A). In order to test the cytotoxicity of GC7 in A549 and NCI-H1299 NSCLC cell lines CCK-8 cell viability assays were performed. The results showed that GC7 had almost no effect on A549 cell viability between 0 and 20 μM, and NCI-H1299 cell viability was well when GC7 concentrations were less than 30 μM, indicating that GC7 had little cytotoxicity against NSCLC cells at low concentrations (Figure 1B,C). Conversely, at GC7 concentrations exceeding 30 μM in A549 cells or exceeding 40 μM in NCI-H1299 cells, cell viability was significantly inhibited (Figure 1B,C). Some studies have reported that low concentrations of GC7 (10 μM) could inhibit the hypusination of eIF5A2 effectively in some tumor cells [15, 19]. In this case, the 20 μM concentration GC7, which has been showed had little cytotoxicity against NSCLC cells but could inhibit the eIF5A2 activation, was chosen for further co-treatments with cisplatin.

CCK-8 assays were carried out to assess the dose-dependence of A549 (epithelial phenotype) and NCI-H1299 (mesenchymal phenotype) cell viability to cisplatin treatment. The results found that increasing doses of cisplatin reduced cell viability in both cell lines (Figure 2A): the IC₅₀ values at 72 h were 3.069 μg/mL (2.735–3.402 μg/mL) and 7.140 μg/mL (6.432–7.848 μg/mL) in A549 and NCI-H1299 cells, respectively (Table 1), showing that A549 cells exhibited higher sensitivity to cisplatin than NCI-H1299 cells. When cisplatin was combined with GC7 treatment (20 μM), cisplatin sensitivity increased in both cell lines compared to cisplatin treatment alone: IC₅₀ values at 72 h decreased to 4.454 μg/mL (3.848–5.060 μg/mL; P <0.0001) in NCI-H1299 (Figure 2B) and 2.360 μg/mL (2.098–2.622 μg/mL; P <0.01) in A549 cells (Figure 2C, In Additional file 1), indicating that GC7 increased cisplatin sensitivity most markedly in NCI-H1299 cells.The difference between phenotypes was examined by western blotting and immunofluorescence to detect expression of E-cadherin (epithelial) and vimentin (mesenchymal) EMT markers in both NSCLC cell lines. The results showed that A549 cells, which were more sensitive to cisplatin, showed higher expression of the epithelial marker E-cadherin, but no expression of the mesenchymal marker vimentin. In contrast, NCI-H1299 showed higher expression of the mesenchymal marker vimentin, but no expression of the epithelial marker E-cadherin (Figure 2D,E).

GC7 can inhibit the activity of eIF5A-2 (In Additional file 2). In order to discover the mechanism by which GC7 enhanced cisplatin sensitivity, we transfected eIF5A-2 siRNA into A549 and NCI-H1299 cells to interfere with eIF5A-2 expression, and found that eIF5A-2 expression was significantly inhibited in both NSCLC cell lines (Figure 3A). We then treated these transfected cells with cisplatin alone, or cisplatin combined with GC7, and carried out CCK-8 cell viability assays. Without GC7, NCI-H1299 cells were the most sensitive to cisplatin after eIF5A-2 siRNA transfection: the IC₅₀ at 72 h was 4.468 μg/mL (4.093–4.842 μg/mL; Table 2). Although A549 cells remained sensitive to cisplatin, the IC₅₀ value was lower: 2.626 μg/mL (2.466–2.785 μg/mL; P = 0.0145 vs. cisplatin alone. Table 2). In contrast, when cisplatin treatment was combined with GC7 after eIF5A-2 siRNA transfection, there was little change in the cisplatin sensitivity of both cell lines: the IC₅₀ values at 72 h were 3.982 μg/mL (3.609–4.356 μg/mL; P = 0.0648) and 2.434 μg/mL (2.307–2.560 μg/mL; P =0.0571) in NCI-H1299 and A549 cells, respectively (Table 2; Figure 3B,C). As GC7 also inhibits eIF5A-1’s activity, we evaluated the role of eIF5A-1 in this process. Western Blot analysis indicated that eIF5A-1 was expressed in the control cells of both cell lines; however the expression of eIF5A-1 was higher in NCI-H1299 cells compared to A549 cells. Moreover, We also evaluated the effect of eIF5A-1 in the siRNA transfected cell. The results showed that when cisplatin treatment was combined with GC7 after eIF5A-1 siRNA transfection, there was little change in the cisplatin sensitivity of both cell lines (In Additional file 3).

Having established that GC7 enhanced the chemotherapeutic effect of cisplatin in NCI-H1299 more than in A549 cells, we wished to determine whether the mechanism was related to EMT. After GC7 treatment for 72 h, A549 cells retained their epithelial characteristics (Figure 4A,B), whereas NCI-H1299 cells displayed a loss of mesenchymal properties and a gain of epithelial properties, appearing a reduction in their migration and invasion capabilities (Figure 4E,F). Furthermore, the NCI-H1299 cells showed increased levels of epithelial marker E-cadherin and lower levels of mesenchymal marker vimentin (Figure 4C,D).Several reports have shown that TGF-β1 could induce epithelial NSCLC cells to undergo EMT. In this study, TGF-β1 exposure (10 ng/mL for 48 h) transformed epithelial A549 cells to mesenchymal phenotype, causing the cells to develop an elongated appearance, irregular pseudopodia, weaker cell-cell junctions (Figure 5A) and stronger migration and invasion capabilities compared to control cells (Figure 5D,E). In addition, the cells showed lower levels of epithelial marker E-cadherin and increased levels of mesenchymal marker vimentin (Figure 5B,C). Conversely, if the A549 cells were pre-treated with GC7 before exposure to TGF-β1, the cells retained their epithelial appearance, levels of EMT markers, and migration and invasion capabilities (Figure 5A–E).In order to verify whether eIF5A-2 was a key factor in GC7 regulation of EMT, we transfected eIF5A-2 siRNA into NCI-H1299 cells without carrying out GC7 treatment. The results showed that the transfected NCI-H1299 cells transformed from mesenchymal phenotype to epithelial phenotype (Figure 6A–D). Conversely, when the transfected cells were treated with GC7, the cells stayed as epithelial phenotype (Figure 6A–D).