Inhibition of non-small cell lung cancer (NSCLC) proliferation through targeting G6PD

Tissue specimens

This study was approved by the Institute Ethics Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (IRB 561/59, COA No. 1034/2016). The study protocol was performed according to the declaration of Helsinki for the participation of human individuals. Lung tissue specimens from biopsies histologically examined for diagnosis prior to treatment were obtained from 64 lung cancer patients who were admitted to the King Chulalongkorn Memorial Hospital (Bangkok, Thailand) during 2009–2014. All tissue samples were examined by a team of pathologists at the King Chulalongkorn Memorial Hospital to determine the type and stage of cancer. Clinical data were collected at the time of the first diagnosis and continued until recurrence, death, or the last follow-up appointment.

Histological and immunohistochemical analyses

Formalin-fixed and paraffin-embedded lung tissue blocks were serially cut into 5 µm thick sections, mounted on slides in serial order and processed following standard procedures for histological evaluation. Hematoxylin and eosin (H&E) staining was performed to determine the cancerous and adjacent non-cancerous areas on the sections. To detect G6PD protein expression, tissue sections adjacent (serial) to H&E-stained sections were heated in an autoclave at 120 °C for 10 min for antigen retrieval, treated with hydrogen peroxide to quench endogenous peroxidase, and blocked with corresponding serum from a secondary antibody raised. Subsequently, the sections were incubated with an anti-G6PD antibody produced in rabbits (HPA000247; Sigma-Aldrich, St. Louis, MO, USA) at a dilution of 1:1000 as the manufacturer’s recommendation at 4 °C overnight. Detection was performed using a biotinylated secondary antibody (Sigma-Aldrich) followed by a streptavidin-biotin complex peroxidase (1:200; Vector, Burlingame, CA, USA) and visualized with 3,3′-diaminobenzidine tetrahydrochloride (Sigma-Aldrich). An isotype IgG antibody was used as a negative control. No counterstain was used. According to data from the Human Protein Atlas database (https://www.proteinatlas.org/ ENSG00000160211-G6PD/tissue), G6PD exhibits high expression in testis tissue, which was utilized as the positive control. The stained sections were evaluated by two pathologists who were blinded to the patients’ clinical information. The G6PD H-score was calculated using the formula 1× (% of weakly-stained as light brown, 1+) +2× (% of moderately stained as medium brown, 2+) +3× (% of strongly stained as dark brown, 3+), as described previously (Leesutipornchai et al., 2020).

Cell culture and treatments

Beas-2B (epithelial cells isolated from normal human bronchial epithelium; ATCC# CRL-3588) and human NSCLC cell lines, including NCI-H1975 (lung epithelial cells derived from adenocarcinoma tissue; ATCC# CRL-5908), and NCI-H292 (human lung mucoepidermoid carcinoma cells; ATCC# CRL-1848), were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Beas-2B were cultured in DMEM, whereas NCI-H1975 and NCI-H292 were cultured and routinely passaged in RPMI-1640 medium containing 10% heat inactivated FBS and 1% penicillin/streptomycin solution (Merck Millipore, MA, USA) at 37 °C in 5% CO₂. Dehydroepiandrosterone (DHEA; Sigma-Aldrich) was prepared as a 1000-fold stock solution in dimethyl sulfoxide (DMSO), thus the final concentration of DMSO did not exceed 0.1%. D-(-)-ribose (R9629; Sigma-Aldrich) was dissolved in culture medium and sterilized by filtration through a 0.45 µm filter before use. After treatment with DHEA for 48 h, the cells were used as indicated in each experiment.

siG6PD and transfection assay

Cells were plated in 12-well plates at 30–50% confluency in complete medium 24 h before transfection. Transfection was performed using Lipofectamine^® 3000 (ThermoFisher Scientific, Waltham, MA, USA), following the manufacturer’s instructions. The siG6PD for G6PD (sense sequence: GGCCGUCACCAAGAACAUU) and non-silencing scramble (sense sequence: GGCACUACCAGACACGAUU) were synthesized and purified by Integrated DNA Technologies, Inc., (IDT, Coralville, IA. USA) and were used at a 100 nM final concentration. After transfection of siG6PD for 24 h, the cells were used in the indicated experiments.

Cell proliferation assay

After treatment, cells were incubated with MTT tetrazolium salt (final concentration 0.5 mg/ml, Sigma-Aldrich) for 2 h at 37 °C. The formazan crystal product was then dissolved with DMSO, and the optical absorbance was measured at 570 nm using a Synergy HT microplate reader (BioTek Instruments Inc., Winooski, VT, USA).

Clonogenic assay

Cells were seeded in a 6-well plate at a density of 2 × 10² cells/well for 24 h before beginning treatment. After treatment, the culture medium was removed and replaced with a fresh complete medium. The cells were then continuously cultured for 7 days. At the end, the colonies were washed with 1X PBS, fixed with glutaraldehyde (0.6% v/v), stained with crystal violet (0.5% w/v) for 30 min, and counted.

G6PD activity assay

After treatments, the cell pellets were collected at the indicated times and washed with ice-cold PBS. The cells were resuspended in cold PBS, sonicated for 10 s (repeated three times), and cooled on ice. Total protein was determined by the BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer’s instructions. G6PD enzyme activity was measured as previously described (Garcia-Nogales et al., 1999) with some modifications. Briefly, 20 µl of the cell lysate (protein at 1mg/ml concentration) was mixed with 980 µl of the reaction buffer, containing 0.38 mM NADP, 6.3 mM MgCl₂, 3.3 mM glucose 6-phosphate, and 5 mM maleimide in 50 mM Tris–HCl (pH 7.5) buffer. The absorbance was kinetically measured at 340 nm for 15 min at 37 °C using a Synergy HT microplate reader (BioTek Instruments Inc.). Enzyme activity was calculated using a standard curve of NADPH and expressed as NADPH units/min/mg of total protein.

Scratch wound assay

Cells were seeded in 24-well plates to a final density of 1 × 10⁵ cells/well and maintained in a CO₂ incubator at 37 °C for 24 h to allow cell adhesion. The confluent monolayer was scratched with a sterile 200-µl pipette tip. Then, culture medium containing dislodged cells was immediately removed and replaced with fresh medium, either alone or containing DHEA or siG6PD. The scratched areas were monitored by collecting digitized images at various time points or until closure of the wound in the control monolayer.

Quantitative RT-PCR

Total RNA was isolated from cell pellets using TRIzol^® Reagent (Thermo Fisher), according to the manufacturer’s instructions. The quality and concentration of RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). cDNAs were synthesized using a RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific), according to the manufacturer’s instructions. Gene expression analysis was performed using PowerUp™ SYBR^® Green Master Mix (Thermo Fisher Scientific) on a StepOnePlus real-time PCR machine (Thermo Fisher Scientific), according to the manufacturer’s protocol. Primer sequences were as follows: G6PD forward primer 5′-GTC AAGGTGTTGAAATGCATC-3′ and reverse primer 5′-CATCCCACCTCTCATTCTCC-3′, Bax forward primer 5′-AACATGGAGCTGCAGAGGAT-3′ and reverse primer 5′- CAGC CCATGATGGTTCTGAT-3′, Bcl-2 forward primer 5′-GGTGGGGTCATGTGTGTG-3′ and reverse primer 5′-CGGTTCAGGTACTCAGTCATC-3′, and ß-actin as a reference gene forward primer 5′-ACTCTTCCAGCCTTCCTTC-3′ and reverse primer 5′-ATCTCCTTC TGCATCCTGTC-3′. The relative abundance of each target gene was calculated relative to ß-actin. The fold change in expression levels was reported as 2^−ΔΔCt.

Statistical analysis

Statistical analyses were performed using SPSS v.22.0 (SPSS Inc., Chicago, IL, USA). Data were obtained from three independent experiments performed in triplicate and presented as the mean ± standard deviation (SD). The chi-square test (χ²) and Student’s t-test were used to examine the differences between categorical and quantitative variables. A two-sided difference with a p-value less than 0.05 was considered statistically significant.

Characteristics of patients

A total of 64 lung cancer patients, consisting of 44 NSCLC patients (68.75%) and 20 SCLC patients (31.25%), were examined in this study. The demographic data of the subjects is summarized in Table 1. The mean age of the patients was 69.5 ± 12.5 years, ranging from 30 to 94 years. The clinical profiles of NSCLC patients were adenocarcinoma (28/44; 63.64%) and squamous cell carcinoma (16/44; 36.36%). Most lung adenocarcinoma patients were female (19/28; 67.86%). Male patients were diagnosed with lung squamous cell carcinoma more often than females (15/16; 93.75%). The stage at the time of diagnosis was determined according to the tumor, node and metastasis (TNM) staging system, as shown in Table 1. There were three lung cancer patients (one from each category: adenocarcinoma, squamous cell carcinoma, and SCLC) for whom we were unable to obtain their TNM stages. In comparison with NSCLCs, most SCLC patients had significantly poor prognosis due to factors such as tumor size (6.3 ± 3.1 cm.; p = 0.009), lymph node metastasis (85.00% of all SCLC cases; p < 0.001), distant metastasis (75.00% of all SCLC cases; p < 0.001), and late stage (III–IV) (94.74% of all SCLC cases; p < 0.001).

G6PD immunostaining

To examine whether G6PD expression could be correlated with the tumorigenesis of lung tissues, we performed immunohistochemistry of G6PD protein in lung tissues obtained from lung cancer patients. Results revealed that the G6PD protein was highly expressed in cancerous areas of lung tissues compared to adjacent normal tissues (Fig. 1). The staining was more intense in NSCLC than in SCLC tissues (p < 0.001) (Fig. 1, Table 1). Consistently, the H-scores of G6PD expression were higher in NSCLC than in SCLC tissues (p = 0.007) (Table 1). Among NSCLC tissues, the H-scores of G6PD in squamous cell carcinoma were higher than those in adenocarcinoma tissues (p = 0.013) (Table 1).

G6PD and NSCLC cell proliferation

In this set of experiments, we aimed to investigate a correlation between G6PD expression and NSCLC growth. The growth of three cell lines with different characteristics was compared. The first cell line was Beas-2B, epithelial cells isolated from normal human bronchial epithelium obtained from autopsies of noncancerous individuals. The second cell line, NCI-H292, was derived from pulmonary mucoepidermoid carcinoma, whereas the third cell line, NCI-H1975, was lung epithelial cells derived from adenocarcinoma tissue. All cells were seeded at the same cell density. After 48 h, the MTT cell proliferation assay was performed to quantify the number of viable cells. Results showed that NCI-H292 cells had a higher proliferation rate than NCI-H1975 and Beas-2B cells (Fig. S1). Then, we compared the mRNA expression levels and activity of G6PD in all cell lines. Results showed that mRNA expression levels and activity of G6PD were significantly higher in NCI-H292 cells than in NCI-H1975 and Beas-2B cells (Fig. S1). These findings suggested that expression levels and activity of G6PD had a positive correlation with the growth of NSCLC cells.

Inhibition of G6PD reduced NSCLC cell proliferation

We challenged the role of G6PD in the growth of NSCLS cells by using DHEA, a known G6PD inhibitor (Preuss et al., 2013), and small interfering RNA for G6PD (siG6PD). Results in the first set of experiments with DHEA demonstrated that, after 48 h of treatment, DHEA exhibited a concentration-dependent reduction of G6PD activity in both NCI-H292 and NCI-H1975 cells (Figs. 2A–2B). Although DHEA at 400 µM showed superior results in the reduction of G6PD activity in both NSCLC cells, as compared to DHEA at 300 µM, we observed its incomplete solubility at this concentration. Therefore, we selected DHEA at the concentration of 300 µM for the evaluation of NSCLC cell proliferation using MTT and colony forming assays. After 48 h of treatment, DHEA significantly reduced cell proliferation of both NSCLC cells (Fig. 2C). If NSCLC cells were allowed to grow in normal culture conditions for 7 days after DHEA treatment, there were no colonies observed at the end of the experiment compared to corresponding controls (Fig. 2D) in NCI-H1975 (Figs. 2E–2F, Fig. S2) and NCI-H292 (Figs. 2G–2H, Fig. S2).

In the next set of experiments, siG6PD was used to suppress G6PD expression in NSCLC cells. The efficacy of siG6PD after 24 h of application was evaluated by measuring G6PD mRNA levels and G6PD activity. Results showed that siG6PD significantly reduced G6PD expression and G6PD activity in NCI-H1975 and NCI-H292 cells by more than 50% compared to scramble controls (Figs. 2I–2J). Given that the G6PD mRNA levels in NCI-H1975 cells were significantly lower than those in NCI-H292 cells (Fig. S1), the knockdown of G6PD mRNA using siG6PD at the same concentration (100 nM) had a more pronounced effect on G6PD mRNA expression levels in NCI-H1975 cells compared to NCI-H292 cells. After 24 h of siG6PD application, the cell proliferation of NSCLC cells was determined. Results showed that siG6PD significantly reduced cell proliferation of both NCI-H1975 cells and NCI-H292 cells by approximately 20% compared to scramble controls (Fig. 2K).

Inhibition of G6PD affected NSCLC cell migration.

In this set of experiments, a scratch-wound assay was performed 24 h post-treatment to assess the effects of G6PD inhibition on the contribution of NSCLC cell migration to wound closure. Results demonstrated that DHEA at 300 µM decreased wound confluence by increasing unmigrated area in cultures of NCI-H1975 and NCI-H292 cells, compared with that of the untreated controls at every time point examined (Figs. 3A–3D, Figs. S3 and S4). Initially, an application of siG6PD slightly decreased wound confluence in cultures of NCI-H292 cells compared with that of scramble controls. However, its effect was significantly observed at 24 h post-application (Figs. 3F and 3H, Fig. S5). In contrast, an application of siG6PD showed less effectiveness in decreasing wound confluence in cultures of NCI-H1975 cells at all time points examined, when compared to scramble controls (Figs. 3E and 3G, Fig. S6).

DHEA and siG6PD induced NSCLC cell apoptosis

To investigate whether the mechanism underlying the inhibitory effects of G6PD on NSCLC cell proliferation involved the induction of apoptosis, we measured the levels of the Bax/Bcl-2 mRNA ratio. Results indicated that the reduction of G6PD activity after 48 h of DHEA treatment and the downregulation of G6PD expression by siG6PD after 24 h post-application significantly induced apoptosis in NCI-H292 cells by increasing the Bax/Bcl-2 mRNA ratio when compared to corresponding controls. However, the same treatments did not induce significant changes in the Bax/Bcl-2 mRNA ratio in NCI-H1975 cells (Figs. 4A–4B).

Co-treatment effects of D-(-)-ribose and DHEA/siG6PD on NSCLC cell proliferation

To highlight the role of PPP in NSCLC cell proliferation, we asked whether the addition of D-(-)-ribose, an end-product of PPP (Mahoney et al., 2018), could reverse the inhibitory effects of DHEA and siG6PD on NSCLC cell proliferation. In this set of experiments, D-(-)-ribose was co-administered with either DHEA or siG6PD. Results demonstrated that addition of D-(-)-ribose did not alter cell proliferation of NCI-H1975 cells that were treated with DHEA or siG6PD, compared to DHEA or siG6PD-treated controls (Figs. 5A–5B). In contrast, D-(-)-ribose increased cell proliferation of NCI-H292 cells treated with DHEA or siG6PD to reach the levels of untreated control cells in a concentration-dependent manner (Figs. 5A–5D).