The A549 human NSCLC cell line was purchased from The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences. A549-cisplatin resistant (A549-Cis) cells were obtained from Otwo Biotech (Shenzhen) Inc., and were created by screening cells after long-term treatment with 5 µg/ml cisplatin. Therefore, the present study also used a cisplatin concentration of 5 µg/ml. The A549 cells were cultured in RPMI 1640 medium (HyClone; Cytiva) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin. The cells were maintained in a humidified incubator containing 5% CO₂ at 37°C. The A549 cells used in the present experiments were passaged a maximum of 50 times.

GOLPH3 short hairpin (sh)RNAs (shGOLPH3) and GOLPH3 overexpression (GOLPH3-OE) plasmid vectors were designed and obtained from VectorBuilder GmbH. Notably, pLV[shRNA]-EGFP:T2A:Puro-U6 was used as the vector backbone for knockdown and pLV[Exp]-EGFP:T2A:Puro-EF1A was used as the vector backbone for overexpression. Briefly, 1×10⁶/dish 293T cells (The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences) were plated on 10-cm dishes to clone the lentiviruses encoding shGOLPH3, noncoding negative control (NC) shRNA (shNC), GOLPH3-OE or GOLPH3-NC (empty vector). The lentiviral vectors were produced using a 3rd generation system. For each transfection, 10 µg lentiviral plasmid was used along with a 3:2:1 ratio of lentiviral vector to packaging plasmids (psPAX2) and envelope plasmids (pMD2.G). The cells were transfected at 37°C for 48 h using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.). Lentiviral particles were collected 48 h post-transfection, concentrated by ultracentrifugation and stored at −80°C. For stable expression, A549 cells were infected with the lentiviruses encoding shGOLPH3, shNC, GOLPH3-OE and GOLPH3-NC at a multiplicity of infection of 10 at 37°C for 24 h, followed by selection with 2 µg/ml puromycin (cat. no. P8230; Beijing Solarbio Science & Technology Co., Ltd.) at 37°C for 72 h. GOLPH3 expression levels were verified by reverse transcription-quantitative PCR (RT-qPCR) and western blotting. The oligonucleotide sequences for shRNAs were as follows: sh1, 5′-GCTTGTGGAATGAGACGTAAACTCGAGTTTACGTCTCATTCCACAAGC-3′ (target sequence: GCTTGTGGAATGAGACGTAAA); sh2, 5′-GCTTGCTTCAATCATGGTTATCTCGAGATAACCATGATTGAAGCAAGC-3′ (target sequence: GCTTGCTTCAATCATGGTTAT); shNC, 5′-CCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACCTTAGG-3′ (target sequence: CCTAAGGTTAAGTCGCCCTCG; this sequence has no corresponding target was found in both humans and mice). Cells were allowed to recover for 72 h post-transduction before any subsequent experimentation.

Total RNA was isolated from NSCLC cells using a PureLink™ RNA isolation kit (cat. no. 12183018A; Invitrogen; Thermo Fisher Scientific, Inc.). cDNA was generated from the RNA using a PrimeScipt™ RT reagent kit (cat. no. RR037A; Takara Biotechnology Co., Ltd.) according to the manufacturer's protocol. RT was performed at 37°C for 15 min, followed by 85°C for 5 sec to inactivate the reverse transcriptase. qPCR was performed using SYBR Green (Bio-Rad Laboratories, Inc.) on the CFX96 touch real-time PCR system (iQ5; Bio-Rad Laboratories, Inc.). The thermocycling conditions were as follows: Initial denaturation at 95°C for 3 min, followed by 40 cycles at 95°C for 10 sec and 60°C for 30 sec. β-actin was used as a normalization control and relative mRNA expression levels were calculated using the 2^−ΔΔCq method (21). The primer sequences for qPCR were as follows (5′-3′): GOLPH3, forward GATGCTCCAACAGGGGATGT, reverse TGGTGAGGGGATGTGTTGTC; ATP7A forward GCCAGCCTCTGACACAAGAA, reverse GCTCCTCTCAACGTTTCTGGA; ABCG2, forward TGGCTGTCATGGCTTCAGTA, reverse GCCACGTGATTCTTCCACAA; MATE1, forward CTTCAGGCAGGACCCAGAT, reverse CAGATAGTTGGCGAGGGCAT; MATE2K, forward ATCCTAGCCACCAGGCACTA, reverse GTGTCCACCTGCACTAGACC; ALDH1A1, forward TGGACCAGTGCAGCAAATCA, reverse ACGCCATAGCAATTCACCCA; C-myc, forward TACTGCGACGAGGAGGAGAA, reverse CGAAGGGAGAAGGGTGTGAC; β-actin, forward AACTGGGACGACATGGAGAAAA, reverse, GGATAGCACAGCCTGGATAGCA.

Total proteins were extracted from NSCLC cells with RIPA buffer (Roche Diagnostics). The protein concentration was determined using the bicinchoninic acid (BCA) assay (Pierce BCA Protein Assay Kit; Thermo Fisher Scientific, Inc.). Equal amounts of protein (30 µg/lane) were separated by SDS-PAGE on a 10% polyacrylamide gel and were transferred to polyvinylidene fluoride (PVDF) membranes (Bio-Rad Laboratories, Inc.). After blocking in 5% BSA at room temperature for 1 h, the PVDF membranes were probed with primary anti-GOLPH3 (1:1,000; cat. no. A13121; ABclonal Biotech Co., Ltd.) and anti-GAPDH (1:5,000; cat. no. AC026; ABclonal) antibodies overnight at 4°C. Subsequently, the membranes were incubated with horseradish peroxidase-linked secondary antibodies (1:5,000; cat. no. A6154; Sigma-Aldrich; Merck KGaA) at room temperature for 1 h. The ECL Kit (Wanleibio Co., Ltd.) was used for detecting protein bands, and protein levels were semi-quantified with Image Lab software (Bio-Rad Laboratories, Inc.), normalized to GAPDH levels.

Cell viability was evaluated using the Cell Counting Kit-8 (CCK-8) assay (Dojindo Laboratories, Inc.). Briefly, 0.5–1×10⁴ cells/well of NSCLC cells (infected with shNC, sh1, sh2, GOLPH3-OE and GOLPH3-NC) were seeded in 96-well plates. After an overnight incubation, the cells were treated with various concentrations of cisplatin (0, 1, 2.5, 5, 10, 25 and 50 µg/ml; Shanghai Aladdin Biochemical Technology Co., Ltd.) at 37°C for 24 h. Subsequently, CCK-8 solution was added to the phenol red- and serum-free cell medium for another 1 h. Light absorbance at 450 nm was then recorded using an Infinite M200 Pro microplate reader (Tecan Group, Ltd.). Each concentration was tested at least three times.

A total of 5×10⁴ cells/well of NSCLC cells (infected with shNC, sh1, sh2, GOLPH3-OE and GOLPH3-NC) were seeded in 24-well plates. After overnight incubation, the cells were treated with 5 µg/ml cisplatin at 37°C for 24 h. Subsequently, the cells were cultured with EdU reagent (1:1,000 dilution; Beyotime Institute of Biotechnology) for another 2 h, and were fixed with 4% paraformaldehyde at 37°C for 15 min, followed by staining with DAPI (1 µg/ml) at 37°C for 10 min. All cells were detected using fluorescence microscopy (Leica Microsystems, Inc.).

Cell death was evaluated using an Annexin V-FITC/propidium iodide (PI) Staining Kit (cat. no. CA001; Signalway Antibody LLC). Stably transduced NSCLC cells were plated onto 24-well plates at a density of 5×10⁴ cells/well. After incubation for 24 h, the cells were treated with 5 µg/ml cisplatin at 37°C for 24 h. Subsequently, the cells were washed with phosphate-buffered saline (PBS), digested with EDTA-free trypsin, resuspended in 500 µl binding buffer, and stained with 10 µl Annexin V-FITC and 10 µl PI at 37°C for 15 min in the dark. After washing twice with PBS, the cells were immediately analyzed by flow cytometry (FACSCelesta; BD Biosciences) and fluorescence microscopy (Leica Microsystems, Inc.), the results were analyzed using FlowJo-V10-CL software (FlowJo LLC).

Cell cycle progression was examined using a Cell Cycle Analysis Kit (cat. no. C1052; Beyotime Institute of Biotechnology). Stably transduced NSCLC cells were plated onto 24-well plates at a density of 5×10⁴ cells/well and were incubated for 24 h, after which, the cells were treated with 5 µg/ml cisplatin at 37°C for 24 h. After washing twice with PBS to remove excess cisplatin, the cells were fixed with 75% ethanol for 24 h at 4°C, were washed twice with PBS and were treated with RNAse (100 µg/ml) and PI (50 µg/ml) solution at 37°C for 30 min in the dark. After washing twice with PBS, the cells were assessed using flow cytometry (FACSCelesta; BD Biosciences) and the results were analyzed using FlowJo-V10-CL software.

NSCLC cells were plated onto 24-well plates at a density of 5×10⁴ cells/well and were incubated at 37°C for 24 h. The cells were treated with 5 µg/ml cisplatin-Cy5 (Xi'an Qiyue Biotechnology Co., Ltd.) at 37°C for 24 h, and, after washing twice with PBS, intracellular cisplatin-Cy5 concentration was assessed under a fluorescence microscope (Leica Microsystems, Inc.).

Intracellular GSH levels were detected using a GSH Assay Kit (cat. no. S0053; Beyotime Institute of Biotechnology). NSCLC cells were plated onto 24-well plates at a density of 5×10⁴ cells/well and were incubated at 37°C for 24 h. The cells were treated with 5 µg/ml cisplatin at 37°C for 24 h, after which, intracellular GSH levels were measured according to the kit instructions.

Intracellular ROS were detected using a Reactive Oxygen Species Assay Kit (cat. no. S0033S; Beyotime Institute of Biotechnology). NSCLC cells were plated onto 24-well plates at a density of 5×10⁴ cells/well and were incubated for 24 h. The cells were treated with 5 µg/ml cisplatin at 37°C for 24 h, and, after washing twice with PBS, the DCFH-DA probe was added to the culture medium and co-incubated 37°C for another 2 h. After washing twice with PBS to remove excess DCFH-DA, the cells were resuspended in PBS and were detected using fluorescence microscopy (Leica Microsystems, Inc.).

NSCLC cells were seeded into ultra-low cluster 6-well microplates at a density of 500 cells/well and were cultured in serum-free medium (containing 20 ng/ml basic fibroblast growth factor and epidermal growth factor, 5 µg/ml insulin, and 0.4% BSA; Dalian Meilun Biology Technology Co., Ltd.) supplemented with 5 µg/ml cisplatin. Cells were cultured for ~10 days before 3D tumor spheres (tight, spherical, nonadherent masses >50 µm in diameter) were observed and images were captured under a phase-contrast fluorescence microscope (Leica Microsystems, Inc.).

Data are presented as the mean ± SD (n=3). The experimental data were statistically analyzed by one-way analysis of variance using GraphPad Prism 7.0 statistical software (Dotmatics). For post hoc comparisons, the Fisher's least significant difference method was used, or the Tukey honestly significant difference method was used when >3 groups were compared. P<0.05 was considered to indicate a statistically significant difference.

To evaluate the effects of GOLPH3 on cisplatin resistance, knockdown and overexpression of GOLPH3 using lentiviruses encoding shGOLPH3 or GOLPH3-OE was performed. The protein and mRNA expression levels of GOLPH3 were increased in GOLPH3-OE cells, and were decreased in shGOLPH3 cells (Figs. 1A and S1). Subsequently, the inhibitory effects of cisplatin on the viability of these cells were assessed. As shown in Fig. 1B, GOLPH3 knockdown significantly increased the inhibitory effects of 5 µg/ml cisplatin on cell viability by ~2 fold. By contrast, 2.5 µg/ml cisplatin reduced cell viability by ~25% in GOLPH3-knockdown cells, but had minimal impact on control cells. Moreover, the inhibitory effect of cisplatin on A549-Cis cells was significantly lower than that on A549 cells (Fig. S2). By contrast, overexpression of GOLPH3 resulted in minimal impact on the inhibitory effects of cisplatin on cell viability (Fig. 1C), indicating that GOLPH3 overexpression does not affect cisplatin resistance. Since EdU is commonly used to detect cell proliferation, cells with GOLPH3 overexpression or knockdown were treated with EdU in the presence of 5 µg/ml cisplatin. Consistent with the cell viability assay, the numbers of EdU-positive cells were markedly decreased in the GOLPH3 knockdown groups compared with those in the control group (Fig. 1D). By contrast, the numbers of EdU-positive cells were comparatively higher in the GOLPH3 overexpression group than in the control group. These findings indicated that GOLPH3 knockdown may increase cisplatin sensitivity in NSCLC cells.

The present study further investigated whether GOLPH3 knockdown could enhance the apoptosis of cells in the presence of 5 µg/ml cisplatin. A low proportion of apoptotic cells was observed in the control group, whereas the proportion was increased to 10–15% in the GOLPH3 knockdown groups (Fig. 2A and B). Consistent with the results of flow cytometry, immunofluorescence analysis showed that GOLPH3 knockdown markedly increased the proportion of Annexin V⁺ and/or PI⁺ cells (Fig. 2C). Moreover, cisplatin-mediated DNA toxicity was elevated in cells with GOLPH3 knockdown. As shown in Fig. 2D, GOLPH3 knockdown in A549 cells resulted in the prolongation of cisplatin-mediated cell cycle arrest at G₂ phase compared with that in the shNC group. Without cisplatin treatment, the cell cycle distribution was consistent for all cells (Fig. S3). By contrast, GOLPH3 overexpression partially abrogated cisplatin-mediated cell cycle block at G₂ phase compared with that in the GOLPH3-NC group (Fig. 2E), suggesting that GOLPH3 may impair cisplatin-DNA adducts and cytotoxicity. These data indicated that GOLPH3 knockdown not only enhances the pro-apoptotic effect of cisplatin, but may also elevate its DNA toxicity.

Reduced drug accumulation is a main factor involved in cisplatin resistance (14,15). To verify the hypothesis that GOLPH3 decreased intracellular concentrations of cisplatin, the present study assessed the intracellular cisplatin concentration using cisplatin-Cy5 fluorescence. As expected, GOLPH3 knockdown in A549 cells led to enhanced cisplatin-fluorescence signal accumulation (Fig. 3A). By contrast, GOLPH3 overexpression notably decreased intracellular cisplatin concentration (Fig. 3B). Intracellular drug concentrations depend on cellular uptake or efflux, and passive drug diffusion is the main form of drug accumulation. The present study aimed to uncover the biological functions of GOLPH3 on intracellular drug efflux by examining the expression levels of drug efflux transporters. RT-qPCR showed that GOLPH3 knockdown significantly downregulated ATP7A (Fig. 3C), ABCG2 (Fig. 3D), MATE1 (Fig. 3E) and MATE2K (Fig. 3F) mRNA expression levels compared with those in the shNC group, indicating that GOLPH3 knockdown may increase cisplatin sensitivity by downregulating the expression of ABC transporters and decreasing drug efflux.

Antioxidant ability assists in the survival of tumor cells treated with chemotherapy (22); therefore, the present study assessed whether GOLPH3 was associated with redox homeostasis. The levels of GSH were examined in cells with GOLPH3 knockdown or overexpression under 5 µg/ml cisplatin treatment. As shown in Fig. 4A, cells with GOLPH3 knockdown exhibited a reduction in GSH levels compared with in the control cells. By contrast, GOLPH3-overexpressing cells displayed higher levels of GSH compared with in the control cells (Fig. 4B), indicating that GOLPH3 could impair redox homeostasis by maintaining a reduced state. Notably, A549-Cis cells exhibited higher levels of GSH compared with A549 cells, which is consistent with the results of GOLPH3-overexpressing cells (Fig. S4). The present study further assessed the dependence of ROS accumulation on GOLPH3 expression. Enhanced ROS accumulation was detected in cells with GOLPH3 knockdown compared with in the control cells, whereas no notable ROS signal was detected in GOLPH3-overexpressing cells, following treatment with 5 µg/ml cisplatin (Fig. 4C). These data strongly indicated that cisplatin-mediated oxidative stress depends on cellular GOLPH3 expression, and that GOLPH3 overexpression may disrupt the GSH/ROS balance and result in cisplatin resistance.

Tumor stem cell subpopulations are closely associated with chemotherapeutic resistance. Therefore, the present study explored the possibility that intrinsic cisplatin resistance in NSCLC cells is due to the existence of a GOLPH3-mediated stem cell subpopulation. Cancer stem cells (CSCs) are a subpopulation of cancer cells with the ability to self-renew and drive tumorigenesis (23). As expected, the mRNA expression levels of CSC, markers (ALDH1A1, C-myc) were significantly inhibited in NSCLC cells with GOLPH3 knockdown (Fig. 5A). Tumor sphere formation assay further verified this phenomenon. The size and number of sphere colonies were markedly inhibited in cells with GOLPH3 knockdown compared with in the control group (Fig. 5B). These data suggested that GOLPH3 knockdown could suppress the stem cell-like phenotype of NSCLC cells and further elevate cisplatin sensitivity.