This article is mentioned in:

Introduction

Cervical cancer is one of the most common malignancies in females worldwide (1). Surgery and concurrent chemoradiotherapy are the two primary treatment options for patients with cervical cancer (2). Traditional chemotherapy drugs cause serious adverse reactions and thus, the development of more effective therapeutics will provide a marked benefit for patients with cervical cancer (3).

Ginsenoside Rh2 (G-Rh2) is a major bioactive component of ginseng (4) that has been indicated to significantly inhibit the proliferation of cancer cells in cell culture and in tumor models of various human cancers (5–8). Previous studies have indicated that apoptosis of cancer cells induced by G-Rh2 is caused by an excessive accumulation of mitochondrial reactive oxygen species (mtROS), accompanied by the occurrence of numerous downstream apoptotic events, such as mitochondrial depolarization, cytochrome c release and activation of caspase enzymes (9–11). However, the upstream mechanisms through which G-Rh2 induces mtROS-mediated apoptosis have remained elusive.

Recently, numerous studies have demonstrated that mitochondrial dysfunction resulting in mtROS overproduction is a potential cause of cancer cell apoptosis (12–14). Mitochondria perform oxidative phosphorylation (OXPHOS) to synthesize ATP through the mitochondrial electron transport chain (ETC), which comprises complexes I–V. The leakage of electrons from the gradient formed by the ETC complexes in the electron transfer process is the main source of mtROS (15–17). As an unavoidable byproduct of metabolism, the mtROS concentration is normally maintained at a low level by the corresponding scavenger system and thereby a steady state is maintained, which is important because mtROS are a significant signal transducer for various molecular pathways (18). However, if an imbalance or decompensation of mtROS production/scavenging occurs, cells become damaged (19,20) with reduced mitochondrial function and they subsequently undergo apoptosis (21–23). In addition, a recent study by our group reported that G-Rh2 induces apoptosis in HeLa cervical cancer cells and that the process was related to ROS accumulation and mitochondrial dysfunction (24). Based on these previous findings, it was hypothesized that G-Rh2 targets ETC complexes, which mediates the generation of mtROS and induces apoptosis of cervical cancer cells. The aim of the present study was to further investigate the novel upstream direct target of mtROS production and apoptosis induction by G-Rh2. The present study provided novel insight into how G-Rh2 induces apoptosis in cervical cancer cells.

Materials and methods

Cell culture

The human cervical cancer cell lines HeLa [human papillomavirus (HPV)-positive] and C33A (HPV-negative) were obtained from the Type Culture Collection of the Chinese Academy of Sciences and cultured in DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin (all from Hyclone; Cytiva). The nontumorigenic (control) End1/e6e7 cell line was purchased from the American Type Culture Collection and cultured in KSFM medium (Gibco; Thermo Fisher Scientific, Inc.) with bovine pituitary extract (0.05 mg/ml), human recombinant epidermal growth factor (0.1 ng/ml), calcium chloride (44.1 mg/l) (all from Shanghai Absin Biotechnology Co., Ltd) and 1% antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin; Hyclone; Cytiva). All cell lines were cultured in a humidified normoxic chamber (Thermo Fisher Scientific, Inc.) with 5% CO2 at 37°C.

Assessment of cell viability

HeLa, C33A and End1/e6e7 cells (2×104 cells/well) were inoculated onto 96-well plates and incubated for 24 h. To determine the 50% growth inhibition concentration (IC50) (25,26), cells were treated with 35, 45, 55 and 65 µM G-Rh2 (Yuanye Biotechnology Co., Ltd) for 24 h. Cell viability was measured using the Cell Counting Kit-8 (CCK-8) assay (Wuhan Boster Biological Technology, Ltd.) according to the manufacturer's protocol. Evaluation was performed using an Infinite M200 PRO plate reader (Tecan Group, Ltd.) at 450 nm.

Flow cytometric analysis of cell death mode

The mode of cell death was assessed using an FITC Annexin V Apoptosis Detection Kit I (BD Biosciences). Cells were inoculated onto 6-well plates and incubated for 24 h. HeLa cells were treated with 35 and 45 µM G-Rh2 for 24 h and C33A cells were treated with 45 and 55 µM G-Rh2 for 24 h. After treatment, the cells were collected by centrifugation, washed twice with ice-cold PBS, suspended in 1X Binding Buffer and then incubated with 5 µl each of propidium iodide (PI) and Annexin V-FITC solution. Probes were incubated at 37°C in the dark for 15 min. Samples were examined with a flow cytometer (Amnis Corporation) and quantified using IDEAS software v6.1 (Amnis Corporation).

Apoptosis morphology detection with Hoechst 33342 staining

The mode of cell death was also demonstrated by Hoechst 33342 staining. After treating cells as above, the culture medium was removed and cells were washed twice with PBS and then submerged in 1 ml Hoechst 33342 staining solution (Beyotime Institute of Biotechnology). The cells were washed twice again after labelling for 30 min at 37°C in the dark and then observed with the EVOS FL Auto Imaging System (Thermo Fisher Scientific, Inc.).

Determination of mitochondrial membrane potential (MMP)

JC-1 (Beijing Solarbio Science and Technology, Co., Ltd.) and rhodamine 123 (Beyotime Institute of Biotechnology) were used to evaluate the MMP. HeLa cells were treated with G-Rh2 at 35 and 45 µM for 24 h and C33A cells were treated with G-Rh2 at 45 and 55 µM for 24 h. After removing the medium and washing the cells twice with cold PBS, 1 ml of cell culture medium supplemented with 1 ml JC-1 dye working solution was added, followed by incubation at 37°C in the dark for 20 min. Cells were then washed twice with JC-1 staining buffer prior to imaging with the EVOS FL Auto Imaging System (Thermo Fisher Scientific, Inc.). Rhodamine 123 was also used to estimate changes in MMP. In brief, cells were stained with rhodamine 123 in the dark at 37°C for 30 min. Cells were washed twice with PBS prior to loading and evaluation by flow cytometry (Amnis Corporation).

Analysis of intracellular ATP levels

An ATP analysis kit (Beyotime Institute of Biotechnology) was used to detect intracellular ATP levels. In brief, cells inoculated onto 6-well cell culture plates were incubated for 24 h and subsequently, HeLa cells were treated with G-Rh2 at 35 and 45 µM for 24 h and C33A cells were treated with G-Rh2 at 45 and 55 µM for 24 h. The culture medium was then removed and replaced with 200 µl ice-cold ATP lysis buffer and the lysate was then centrifuged at 12,000 × g for 10 min at 4°C. The supernatant was then transferred to a new tube. The reaction was initiated at 37°C for 3 min to consume background ATP with 100 µl ATP detection buffer prior to adding 10 µl supernatant for detection. Optical density values were recorded using an Infinite M200 PRO plate reader (Tecan Group, Ltd.) and the ATP content was converted according to the standard curve.

Measurement of oxygen consumption rate (OCR)

The OCR was tested using the Seahorse Bioscience XFp Extracellular Flux Analyzer (Agilent Technologies, Inc.) according to the manufacturer's protocol. The Seahorse XFp Mito Stress Test Kit (Agilent Technologies, Inc.) was applied to measure the OCR. After treatment with G-Rh2 (35 and 45 µM) for 24 h, HeLa cells (1×104/well) were inoculated onto XFp cell culture plates for 24 h. The cells were then washed with analysis medium (supplemented with 10 mM glucose, 2 mM glutamine and 1 mM sodium pyruvate). Subsequently, the plates were placed in a non-CO2 incubator at 37°C for at least 1 h for degassing. After baseline measurement, 1 µM oligomycin (ATP synthase inhibitor), 1 µM carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP; uncoupler) and 0.5 µM rotenone (rot)/antimycin A (AA) mixture (inhibitors of ETC complexes I/III) were used to perform real-time OCR quantifications. Data were evaluated using Wave software version 2.6.3 (Agilent Technologies, Inc.) and OCR was expressed in pmol/min.

Measurement of extracellular acidification rate (ECAR)

The ECAR was detected using the Seahorse XF Glycolysis Stress Test Kit (Agilent Technologies, Inc.) on the Seahorse XFp Extracellular Flux Analyzer (Agilent Technologies, Inc.) according to the manufacturer's protocol. In brief, HeLa cells were treated with 35 or 45 µM G-Rh2 for 24 h and then inoculated onto Seahorse XFp cell culture microtiter plates for 24 h (1×104 cells/well). The cells were then washed with assay medium (supplemented with 2 mM glutamine) prior to degassing treatment at 37°C in a non-CO2 incubator for 1 h. After the ECAR baseline was calibrated, 10 mM glucose, 1 µM oligomycin (OXPHOS inhibitor) and 50 mM 2-deoxy-d-glucose (glycolysis inhibitor) were injected into the detection wells at the specified time points. Data were evaluated using Wave software version 2.6.3 (Agilent Technologies, Inc.) and ECAR was expressed in mpH/min.

Cell fractionation

The Cell Mitochondrial Isolation Kit (Beyotime Institute of Biotechnology) was used to extract mitochondrial protein. HeLa cells were collected and rinsed with cold PBS after treatment with 35 and 45 µM G-Rh2 for 24 h, and then treated with ice-cold mitochondrial separation reagent containing 1 mM PMSF for 10 min. Next, the cells were homogenized and the supernatant was centrifuged at 11,000 × g for 10 min at 4°C. The mitochondrial protein precipitate was lysed with mitochondrial lysate for subsequent western blotting analysis. Finally, total protein was obtained by lysing HeLa cells with RIPA buffer (Beyotime Institute of Biotechnology). The protein concentration of the samples was detected using a BCA Protein Assay Kit.

Western blot analysis

Western blot was applied to analyze protein expression. In brief, denatured protein samples (30 µg) were separated by 12% SDS-PAGE, transferred to PVDF membranes (EMD Millipore), blocked in 5% skimmed milk (Beijing Solarbio Science & Technology, Co., Ltd.) for 2 h at room temperature, washed with PBS containing Tween-20 (PBST) and incubated with primary antibodies [total OXPHOS human WB antibody cocktail (cat no. ab110411); anti-myc tag (cat no. ab206486); anti-Flag (cat no. ab205606); and anti-6×His tag (cat no. ab18184); all 1:1,000 dilution; Abcam] at 4°C overnight on an orbital shaker. PVDF membranes were then washed with PBST and incubated with horseradish peroxidase-conjugated anti-mouse IgG (cat no. sc-2005; 1:1,000 dilution; Santa Cruz Biotechnology, Inc.) at room temperature for 1 h prior to washing with PBST and adding a working solution of Enhanced Chemiluminescence Reagent Kit (Beyotime Institute of Biotechnology). Membranes were imaged using the iBright FL1000 Imaging System (Invitrogen; Thermo Fisher Scientific, Inc.). Image J software 1.53a (National Institutes of Health) was used to quantify the gray value of western blot bands and the protein expression level was normalized as the gray value of the target protein/the loading control protein.

ETC complex activity assay

ETC complex activity detection kits (cat no. BC0515, BC3235, BC3245, BC0945 and BC1445; Beijing Solarbio Science & Technology, Co., Ltd.) were used to measure the enzymatic activity of ETC complexes I–V with an Infinite M200 PRO Plate Reader (Tecan Group, Ltd.) according to the manufacturer's protocol. The oxidation rate of NADH was recorded at 340 nm to assess the activity of ETC complex I. The rate of 2,6-dichloroindolephenol reduction was used to measure the activity of ETC complex II at 605 nm. The activity of ETC complex III was detected by measuring the increase rate of the light absorption of cytochrome c-reduced at 550 nm. The activity of ETC complex IV was monitored by measuring the rate of decrease in the absorbance of reduced cytochrome c at 550 nm. The activity of ETC complex V was quantified by calculating the rate of addition of inorganic phosphate (Pi) at 660 nm.

Molecular docking

Molecular docking analyses were performed for G-Rh2 and the ETC complexes. Data on the structures of ETC complexes were obtained from the RCSB Protein Data Bank (PDB; http://www.rcsb.org/pdb), including ETC complex I (PDB ID: 4G73) (27), ETC complex III (PDB ID: 4PD4) (28) and ETC complex V (1BMF) (29,30). Molecular docking was performed with AutoDock tools (version 4.2.6) and visualized with the Discovery Studio 4.0 Visualizer (BIOVIA).

Measurement of mtROS

The mitoSOX probe (Beyotime Institute of Biotechnology) was used to determine mtROS production. First, cells were seeded onto 6-well cell culture plates. After 24 h of incubation, G-Rh2 specificity was determined by pretreating HeLa cells with mitoQ (5 µM) for 1 h prior to treatment with G-Rh2 (35 and 45 µM) for 24 h. The cells were collected by centrifugation and washed with cold PBS prior to adding 5 µM mitoSOX probe (cat. no. M36008; Invitrogen; Thermo Fisher Scientific, Inc.) and incubating the mixture at 37°C for 20 min in the dark. The cells were then washed twice with PBS, 200 µl cell suspension (1×106 cells/ml) was added to the 96-well black plate and fluorescence units were detected with an Infinite M200 PRO plate reader (Tecan Group, Ltd.) at 510/580 nm excitation/emission wavelengths.

Construction of the overexpression vectors and transfection

According to the PDB ID of ETC complexes I, III and V in molecular docking, the representative subunits of corresponding human genes were selected as NADH-ubiquinone oxidoreductase core subunit S1(NDUFS1), biquinol-cytochrome c reductase core protein 1 (UQCRC1) and ATP synthase F1 subunit beta (ATP5F1B), respectively. The NDUFS1, UQCRC1 and ATP5F1B cDNA fragments were amplified with primers using a cDNA library prepared from HeLa cells as the template. The PCR program was as follows: 98°C for 10 sec, 58°C for 10 sec and then 72°C for 1 min per kb for a total of 35 cycles. The target fragments were digested with double restriction enzymes and then ligated to the corresponding plasmids (Guangzhou RiboBio Co., Ltd.) with T4 Ligase (Takara Bio, Inc.) and then transformed into E. coli turbo (New England Biolabs). Positive recombinant plasmids were identified by restriction digestion and sequencing. All plasmids, primers and enzymes used are provided in Table SI. For transfection, HeLa cells were seeded in 6-well plates and cultured for 24 h. Cells were transfected with 2 µg of the plasmids using Lipofectamine® 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) for 6 h. The culture medium was then replaced with fresh opti-DMEM (Gibco; Thermo Fisher Scientific, Inc.) and the cells were continued to be grown in culture for 48 h. For G-Rh2 treatment, HeLa cells were treated with 45 µM G-Rh2 for 24 h after transfection.

Statistical analysis

Statistical analyses were performed with GraphPad Prism software version 6 (GraphPad Software Inc.). For quantitative analyses, data were collected from three independent experiments and presented as the mean ± standard deviation. One-way ANOVA followed by Dunnett's post-hoc test or Tukey's post-hoc test as indicated within the figure legends were utilized to compare among groups. P<0.05 was considered to indicate a statistically significant difference.

Results

G-Rh2 inhibits the viability of cervical cancer cells but is not cytotoxic to End1/e6e7 cells

To study the impact of G-Rh2 on the viability of cancer and normal cells, two cervical cancer cell lines, including HeLa (integrated HPV18) and C33A (without HPV) and human cervical epithelial cells (End1/e6e7) as a control were exposed to 35, 45, 55 and 65 µM G-Rh2 for 24 h. As presented in Fig. 1A and B, G-Rh2 was cytotoxic to all cervical cancer cell lines assessed, with IC50 values of 45 µM for HeLa and 55 µM for C33A cells at 24 h. However, compared with the effect on the two cervical cancer cell lines, the effect of treatments at all concentrations on the viability of End1/e6e7 cells was negligible, indicating that G-Rh2 was highly selective for cancer cells (Fig. 1C).

Figure 1. G-Rh2 reduces cervical cancer cell viability but has no cytotoxic effect on End1/e6e7 cells. Viability of cervical cancer cell lines (A) HeLa and (B) C33A and of (C) End1/e6e7 cells (control) following exposure to G-Rh2 (35, 45, 55 and 65 µM) for 24 h. Cell Counting Kit-8 assays were performed to measure cell viability, which was compared with that of the control group under 0.12% v/v DMSO. All results were obtained from three independent experiments (n=3). Data are presented as the mean ± standard deviation and were analyzed by one-way ANOVA followed by Dunnet's post-hoc test. *P<0.05; **P<0.01; ***P<0.001 vs. control. G-Rh2, ginsenoside Rh2.

G-Rh2 induces apoptosis in cervical cancer cells

Next, Annexin V-FITC/PI and Hoechst 33342 staining assays were used to analyze the number of apoptotic cells following exposure to different concentrations of G-Rh2. As indicated in Fig. 2A, after 24 h of treatment, the rate of early apoptosis was considerably increased in G-Rh2-treated cervical cancer cells (HeLa and C33A) compared with that in the control group. Furthermore, more intense blue staining and more irregular chromatin were observed in G-Rh2-treated HeLa and C33A cells (Fig. 2B) compared with that in the control group, indicating that G-Rh2 treatment induced apoptosis in cervical cancer cells. Of note, G-Rh2 significantly induced early apoptosis in both cancer cell lines in a dose-dependent manner. However, treatment with high-dose G-Rh2 (55 and 65 µM) slightly increased the rate of cell death in End1/e6e7 cells compared with that in the control group (Fig. S1) and demonstrated that the effect on normal cells was not statistically significant as determined for cancer cells.

Figure 2. Treatment with G-Rh2 promotes apoptosis in cervical cancer cells. (A) Representative flow cytometry images indicating apoptosis of HeLa and C33A cells after G-Rh2 treatment. The early apoptotic rate was quantified by IDEAS software v6.1. (B) Representative image of Hoechst staining indicating the morphological changes to apoptotic HeLa and C33A cells following G-Rh2 treatment (scale bar, 100 µm). White arrows represent nuclei with intense fluorescence and an irregular shapes. Data are presented as the mean ± standard deviation of three independent experiments and were analyzed by one-way ANOVA followed by Dunnet's post-hoc test. *P<0.05; ***P<0.001 vs. control. G-Rh2, ginsenoside Rh2.

HeLa cells are more sensitive to G-Rh2 than C33A cells regarding MMP and ATP generation

According to a previous study by our group suggesting that G-Rh2 possibly induced mitochondrial dysfunction in cancer cells, the mitochondrial defects induced by G-Rh2 may potentially result in apoptosis of cervical cancer cells (24). In the present study, to further examine this, the JC-1 and rhodamine 123 probes were used to detect changes in the MMP, an index of mitochondrial function. After staining with the JC-1 fluorescent probe, HeLa cells had an obvious response, displaying red to green transition with increasing G-Rh2 concentration. However, only a slight change in JC-1 probe fluorescence was identified in C33A cells (Fig. 3A). In addition, flow cytometry was employed to measure the fluorescence intensity of rhodamine 123; G-Rh2 treatment significantly reduced the MMP in HeLa cells. Although 55 µM G-Rh2 was able to significantly reduce the MMP in C33A cells, the degree of reduction was significantly lower than that in HeLa cells (Fig. 3B). Next, the effect of G-Rh2 on cellular ATP levels was evaluated by a chemiluminescence assay. Specifically, G-Rh2 treatment reduced cellular ATP levels by 10.4% (35 µM) and 49.1% (45 µM) in HeLa cells and by 7.5% (35 µM) and 35.2% (45 µM) in C33A cells compared with that in in the control group (Fig. 3C). According to these observations, HeLa cells were more sensitive to G-Rh2 than C33A cells, at least with regard to MMP and ATP production. Therefore, HeLa cells were selected for subsequent mechanistic examination of these effects.

Figure 3. Greater inhibitory effect of G-Rh2 on MMP and ATP production in HeLa cells compared with that in C33A cells. (A) Representative image of MMP as measured by the JC-1 fluorescent probe in HeLa and C33A cells after G-Rh2 treatment (scale bar, 400 µm). The white arrows represent that the fluorescence color changes from red to green and the MMP decreases. (B) Representative flow cytometry histograms indicating the Rhodamine 123 fluorescence intensity of HeLa and C33A cells after G-Rh2 treatment and quantification by IDEAS software v6.1. (C) Changes in intracellular ATP levels in HeLa and C33A cells following exposure to G-Rh2. Data are presented as mean ± standard deviation of three independent experiments and were analyzed by one-way ANOVA followed by Dunnet's post-hoc test. *P<0.05; **P<0.01; ***P<0.001 vs. control. G-Rh2, ginsenoside Rh2; MMP, mitochondrial membrane potential.

Inhibitory effect of G-Rh2 on mitochondrial OXPHOS is greater than that on glycolysis in HeLa cells

OXPHOS is essential for the ability of cells to hold the MMP (7). To study the effect of G-Rh2 on mitochondrial OXPHOS, a Seahorse XFp extracellular flux analyzer was used to monitor mitochondrial aerobic respiration in HeLa cells. As presented in Fig. S2A, G-Rh2 treatment of HeLa cells resulted in a rapid decline of OXPHOS. G-Rh2 treatment also significantly inhibited the cellular response to typical ETC complex inhibitors [e.g., oligomycin (ATP synthase inhibitor), FCCP (mitochondrial uncoupling agent), AA (ETC complex III inhibitor) and rot (ETC complex I inhibitor)]. In addition, G-Rh2 significantly inhibited the basal and maximum mitochondrial OXPHOS capacity. Furthermore, in HeLa cells, ATP-linked proton leak as well as reserve capacity were also more significantly inhibited by G-Rh2 compared with those in the control group (Fig. S2B).

Cancer cells provide energy and produce lactic acid through the glycolysis pathway, via a process termed the ‘Warburg effect’. Therefore, it was investigated whether G-Rh2 was involved in blocking glycolysis. The results indicated that G-Rh2 reduced ECAR, which reflects glycolytic capacity. Glycolysis, glycolytic capacity and glycolytic reserve were all significantly inhibited by G-Rh2 (Fig. S2C and D). Of note, G-Rh2 inhibited both OXPHOS and glycolysis in HeLa cells and had a stronger effect on OXPHOS.

Impairment of OXPHOS by G-Rh2 occurs via direct inhibition of ETC complexes activity, not by downregulation of ETC proteins

Decreasing OXPHOS may contribute to the downregulation of ETC proteins or to the inhibition of ETC complex enzymes (31). Therefore, in the present study, the expression levels of five ETC subunits were detected in G-Rh2-treated cells. As presented in Fig. 4A and B, ETC protein expression was not significantly altered after G-Rh2 treatment, indicating that G-Rh2 damages OXPHOS not by downregulating ETC protein expression but by directly inhibiting the activity of ETC complexes. As expected, treatment with 45 µM G-Rh2 significantly inhibited the activity of ETC complexes I, III and V in HeLa cells but did not affect ETC complexes II and IV (Fig. 4C). These results suggested that the impairment of G-Rh2 on OXPHOS was not caused by downregulation of ETC protein expression but by directly inhibiting the activity of ETC complexes.

Figure 4. G-Rh2 decreases oxidative phosphorylation by directly inhibiting the activity of ETC complexes instead of regulating their protein expression. (A) Representative western blot image indicating ETC protein expression after G-Rh2 treatment. COX4 was used as the sample loading control. (B) Quantitative expression of ETC protein complexes. (C) Activities of ETC complexes I–V after G-Rh2 treatment compared with those in the control group. Quantitative bar graphs were generated separately and then plotted together. Data are presented as the mean ± standard deviation of three independent experiments and were analyzed by one-way ANOVA followed by Dunnet's post-hoc test. **P<0.01; ***P<0.001 vs. control. G-Rh2, ginsenoside Rh2; ETC, electron transport chain; NDUFB8, NADH-ubiquinone oxidoreductase subunit B8; SDHB, succinate dehydrogenase complex iron sulfur subunit B; UQCRC2, ubiquinol-cytochrome C reductase core protein 2; COX II, cytochrome oxidase subunit 2; ATP5A, adenosine triphosphate 5A.

Increased ROS levels in G-Rh2-mediated apoptosis

MtROS is an outgrowth of OXPHOS that is primarily generated by the ETC complexes. As G-Rh2 significantly inhibited the activity of complexes I, III and V, its impact on mtROS production was examined. As presented in Fig. 5A, treatment of HeLa cells with 45 µM G-Rh2 led to overproduction of mtROS, with mtROS levels exhibiting a significant increase of 2.8-fold compared with that in the control group. This burst of mtROS production was partially rescued by the antioxidant mitoQ.

Figure 5. Apoptosis induced by G-Rh2 is related to mtROS. HeLa cells were pretreated with or without mitoQ (5 µM) for 1 h and then exposed to the indicated concentrations of G-Rh2 for 24 h. (A) The mitoSOX fluorescence units in HeLa cells were detected using the Infinite M200 PRO plate reader. Data were analyzed by one-way ANOVA followed by Tukey's post-hoc test and are presented as the mean ± standard deviation of three independent experiments. (B) Cell viability was detected using Cell Counting Kit-8 assays. (C) Annexin V-FITC/propidium iodide staining was applied to measure the percentage of apoptotic cells, which was quantified by flow cytometry. Data are presented as the mean ± standard deviation of three independent experiments and were analyzed by one-way ANOVA followed by Tukey's post-hoc test in A or Dunnet's post-hoc test in B and C. ***P<0.001 vs. without G-Rh2 and Mito Q; ###P<0.001 vs. 45 µM G-Rh2 without mitoQ. G-Rh2, ginsenoside Rh2.

To study whether mtROS was involved in G-Rh2-induced apoptosis, the protective effect of mitoQ on G-Rh2-induced cytotoxicity was detected using CCK-8 assays. As indicated in Fig. 5B, the effect of G-Rh2 on HeLa cells was partially attenuated by mitoQ pretreatment. Furthermore, quantitative measurements of Annexin V/PI-positive cells by flow cytometry (Fig. 5C) verified that mitoQ partially rescued G-Rh2-induced early apoptosis. These results demonstrated that G-Rh2-induced mtROS generation mediated the subsequent apoptosis.

Molecular docking reveals a putative G-Rh2 binding site on ETC complexes I, III and V

Although the above results confirmed that G-Rh2 induced mtROS, triggering apoptosis via suppressing ETC complexes I, III and V, the detailed binding mechanism remained to be further elucidated. Therefore, protein-ligand docking was performed to understand the mode of G-Rh2 binding and to locate the ETC complex residues that may have crucial roles in G-Rh2 binding.

The molecular docking results revealed that the optimal docking conformation binding energy of G-Rh2 and ETC complex III was −10.6 kcal/mol, which was lower than that of ETC complex I (−7.73 kcal/mol) and ETC complex V (−6.48 kcal/mol). The hydroxyl oxygen atom of the six-membered ring of G-Rh2 was indicated to be able to form hydrogen bonds with Thr87 in ETC complex I, Ile122 in ETC complex III and Thr163 and Asp256 in ETC complex V, with hydrogen bond lengths of 3.3, 2.1, 1.9 and 3.1 Å, respectively. The oxygen atom of the six-membered ring of G-Rh2 was also able to form a 2.5 Å hydrogen bond with Arg189 in ETC complex V. In addition, the oxygen of Tyr345 in ETC complex V formed a hydrogen bond with the six-membered ring hydrogen of G-Rh2, the length of which was 1.9 Å (Fig. 6A and B; Table I). The formation of these hydrogen bonds enhanced the ability of G-Rh2 to target mitochondrial ETC complexes by better inhibiting protein activity.

Figure 6. Binding of G-Rh2 to ETC complexes I/III/V. (A) Schematic diagram of G-Rh2 ligand binding to protein components of ETC complex I, III and V. The proteins are displayed in cartoon format. (B) Schematic diagram of the amino acid site of G-Rh2 ligand binding to ETC complex I, III and V proteins. G-Rh2, ginsenoside Rh2; ETC, electron transport chain.

Table I. Binding affinities (free energies of binding) and the types of interactions of the ligand ginsenoside Rh2 with its receptors ETC complex I, III and V.

Table I.

Binding affinities (free energies of binding) and the types of interactions of the ligand ginsenoside Rh2 with its receptors ETC complex I, III and V.

ETC complex	Docking score (kcal/M)	Residues involved in hydrogen bonding	Residues involved in weak interactions
I	−7.73	THR87	PRO84, ASP85, SER86, THR87, VAL112, ASN113, GLU126, GLU213, PHE258, PHE509, PHE510, LYS511, GLY512
III	−10.60	ILE122	ILE125, PHE129, TYR132, MET139, ILE147, PRO271, LEU275, TYR279, PHE278, LEU282, MET295, PHE9, ILE299
V	−6.48	THR163, ASP256, ARG189, TYR345	THR163, VAL164, ARG189, GLU192, ASP256, TYR345, PHE424, ALA421, THR425

[i] ETC, electron transport chain.

Overexpression of UQCRC1 partially rescued G-Rh2-induced mtROS production and apoptosis

In order to verify that the effect of ETC complexes I, III and V in G-Rh2 induced cervical cancer cell apoptosis, overexpression vectors for the representative subunits of ETC complex I (NDUFS1), ETC complex III (UQCRC1) and ETC complex V (ATP5F1B) were constructed. HeLa cells transfected with these overexpression vectors exhibited a significant increase in the expression levels of NDUFS1, UQCRC1 and ATP5F1B protein compared to mock-transfected cells (Fig. 7A). G-Rh2 treatment increased mtROS and apoptosis in HeLa cells compared with that in the control group, which was partially attenuated by overexpression of NDUFS1, UQCRC1 and ATP5F1B and the overexpression of UQCRC1 had a significant effect (Fig. 7B and C).

Figure 7. Overexpression of UQCRC1 partly rescued G-Rh2-induced mtROS and apoptosis. (A) Total protein extracts from HeLa cells expressing flag-tagged NDUFS1, myc-tagged UQCRC1 and 6×-his-tagged ATP5F1B were used for WB analysis with an anti-flag/myc/6×-his-tag antibody. GAPDH was used as the sample loading control. (B) The mitoSOX fluorescence units in HeLa cells were recorded using an Infinite M200 PRO plate reader. (C) Flow cytometry was used to detect the percentage of apoptotic cells, which was quantified by IDEAS software v6.1. Data are presented as the mean ± standard deviation of three independent experiments and were analyzed by one-way ANOVA followed by Dunnet's post-hoc test. ***P<0.001 vs. mock-transfected group. ###P<0.001 vs. 45 µM G-Rh2 without UQCRC1 overexpression. G-Rh2, ginsenoside Rh2; WB, western blot; NDUFS1, NADH dehydrogenase Fe-S protein 1; UQCRC1, ubiquinol-cytochrome c reductase core protein 1; ATP5F1B, ATP synthase F1 subunit beta.

Discussion

In the present study, the cytotoxicity of G-Rh2 in the HeLa (HPV18-positive) and C33A (HPV-negative) cervical cancer cell lines compared with that of the non-tumorigenic spontaneous End1/e6e7 cell line was assessed. The results demonstrated that G-Rh2 exposure had specific dose-dependent cytotoxicity on both cervical cancer cell lines but not on End1/e6e7 cells. It was further reported that G-Rh2-induced apoptosis was closely related to the specific activity inhibition of ETC complexes I, III and V, which led to reduced OXPHOS, decreased oxygen consumption and increased mtROS production; these processes are summarized in Fig. 8.

Figure 8. Schematic diagram of G-Rh2 targeting mitochondria. In the process of reducing mitochondrial metabolism, G-Rh2 primarily targets ETC complexes III and indirectly affects I and V, thereby impairing mitochondrial oxidative phosphorylation, blocking electron transfer related to the TCA cycle and stimulating excess mtROS production, leading to apoptosis. G-Rh2, ginsenoside Rh2; ETC, electron transport chain; mtROS, mitochondrial reactive oxygen species; Cyt c, cytochrome c; Q, ubiquinone; TCA, tricarboxylic acid.

The vast majority of G-Rh2-based anti-cervical cancer studies have been performed in HeLa cells (32,33). The present results are consistent with these reports; in addition, it was indicated that G-Rh2 was cytotoxic to both HeLa and C33A cells, with IC50 values of 45 and 55 µM, respectively. Compared with the effect in C33A cells, HeLa cells had a more markedly decreased viability following treatment with a relatively low concentration of G-Rh2. This may indicate that cervical cancer caused by HPV infection is more sensitive to G-Rh2. By contrast, G-Rh2 did not inhibit the viability of End1/e6e7 cells to a marked degree, although their viability was slightly decreased at higher concentrations tested. Similar to reports on other human cancer cell lines (34,35), the specific effect of G-Rh2 on cervical cancer cells determined in the present study suggests its low intrinsic toxicity and its differential effects on normal cells vs. cancer cells.

One of the major apoptotic pathways in cancer cells involves the mitochondria. The presence of mitochondrial dysfunction was considered as a possible mechanism of G-Rh2-induced cytotoxicity (36). A previous study by our group suggested that G-Rh2 upregulated voltage-dependent anion channel 1 to trigger the mitochondrial translocation of BAX, promoting cytochrome c release and initiating mitochondrial-dependent apoptosis in HeLa cells (24). In the present study, it was revealed that the effect of G-Rh2 on the dissipation of the MMP and the reduction in ATP generation in HeLa cells was much stronger than that in C33A cells, consistent with the cell viability and apoptosis results. It is worth noting that p53 is mutated in C33A cells, while p53 is wild-type in HeLa (HPV-positive cervical cancer cell line) and p53 is involved in maintaining mitochondrial homeostasis. The response of wild-type p53 to mitochondrial stress in HeLa cells may be responsible for the higher sensitivity to decreased MMP and ATP levels (37,38). In summary, mitochondria may be the primary intracellular target of G-Rh2.

Mitochondria primarily synthesize ATP via OXPHOS, which involves the ETC complex along with several carriers that localize to the inner mitochondrial membrane (39). The next goal of the present study was to investigate whether G-Rh2 affected mitochondrial OXPHOS and the underlying mechanism. In these experiments, it was demonstrated that G-Rh2 inhibited mitochondrial respiration (basal OCR, maximal OCR and reserve capacity). Mitochondrial defects may lead to reduced expression of ETC complexes I–V (40). The results suggested that G-Rh2 primarily suppressed ETC by inhibiting the activity of ETC complex enzymes rather than by affecting their expression. It was further demonstrated that G-Rh2 exhibited potent inhibitory effects on ETC complexes I, III and V, but had no impact on ETC complexes II and IV. The suppression of ETC complexes I, III and V by G-Rh2 caused significantly decreased MMP and cellular ATP levels. However, the inhibition of the MMP by G-Rh2 was not consistent with that of ATP, with the inhibitory effect on MMP being stronger. This was probably due to the ATP produced in compensation by the glycolytic pathway. More active glycolysis takes place to compensate for the decreased ATP production due to impaired mitochondrial OXPHOS (41). During homeostasis, ~70% of ATP is synthesized in the mitochondria; however, this percentage declines to 30% under a hypoxic environment (42). In the present study, it was also demonstrated that G-Rh2 does not completely inhibit glycolysis, exerting a stronger inhibitory effect of G-Rh2 on OXPHOS than on glycolysis in HeLa cells.

The overproduction of mtROS is a result of the events that inhibit the activity of ETC complexes (43). In fact, in the process of electron transfer from NADH to coenzyme Q in complex I, mtROS may be generated at the flavin mononucleotide site and the coenzyme Q binding site; the hemiquinone radical (QH-) of complex III produces mtROS by leaking electrons to O2 through a non-enzymatic reaction (17). Although ETC complex V (ATP synthase) is not a direct source of mtROS, it participates in the regulation of energy metabolism, which in turn affects mtROS generation (44). In the present study, a suppressive effect of G-Rh2 on the activity of ETC complexes I, III and V was observed. To this end, the necessity of mtROS was further validated, which was stimulated by the inhibition of ETC complexes I, III and V in the induction of apoptosis. The results indicated that the overproduction of mtROS mediated by G-Rh2 was primarily responsible for the cytotoxicity caused by G-Rh2. After G-Rh2 treatment, cellular mtROS levels increased by ~2.8-fold and this was partially reversed by the mitochondrial target antioxidant Mito Q. Furthermore, Mito Q also inhibits apoptosis induced by G-Rh2. These results are in line with previous reports that suggested G-Rh2 induces apoptosis by promoting mtROS accumulation in human leukemia Jurkat, HepG2 and Hep3B cells (34,36,45).

The molecular docking experiments indicated that due to powerful hydrogen bonding and/or hydrophobic interactions, G-Rh2 exhibited strong affinities towards ETC complex III (−10.6 kcal/mol), ETC complex I (−7.73 kcal/mol) and ETC complex V (−6.48 kcal/mol). Enzymatic activity experiments suggested that G-Rh2 significantly inhibited the activity of ETC complexes I, III and V and the degree of inhibition was basically the same. Conversely, the molecular docking results indicated that G-Rh2 has a higher affinity for ETC complex III. This implied that G-Rh2 may directly bind to ETC complex III and indirectly interfere with ETC complexes I and V. The binding sites of the molecular docking results suggested that post-translational modification of the protein is likely responsible for the ability of G-Rh2 to reduce the activity of the ETC complex enzymes under the condition that the protein expression of the representative subunits of complex I–V did not change. G-Rh2 may catalyze the phosphorylation of corresponding amino acids through hydrogen bonding.

To gain additional insight into the mechanism of action of ETC complex I, III and V in G-Rh2-induced mtROS-mediated apoptosis, overexpression vectors for NDUFS1, UQCRC1 and ATP5F1B, which are representative subunits of ETC complex I, III and V, respectively, were constructed. The results suggested that overexpression of UQCRC1 significantly attenuated G-Rh2-induced mtROS-mediated apoptosis in HeLa cells. Consistent with the molecular docking results, G-Rh2 mainly targets ETC complex III, which promotes the production of mtROS and ultimately induces apoptosis. It has been reported that ETC complex I and ETC complex III are closely connected through the interaction between protein subunits (46), which may be the cause of the decrease in ETC complex I activity. In addition, ETC complexes I–IV not only transfer electrons, but also establish the proton gradient of ETC complex V to synthesize ATP (17). The inhibition of ETC complex III hinders electron transfer and accumulated electrons may be responsible for the decrease of ETC complex V enzyme activity.

The limitation of the present study on G-Rh2 is that the experiments were all performed in cell lines cultured in vitro. It has been reported that G-Rh2 exhibits a variety of anti-tumor activities in vivo with no measurable toxicity even at a dose of 120 mg/kg (47), which means that G-Rh2 possesses a wide tolerance. However, due to the huge difference between in vitro and in vivo experiments, the actual mechanisms of G-Rh2 require to be further studied in vivo.

In conclusion, the present study indicated that G-Rh2 is a major contributor to apoptosis in HeLa cells through the mitochondrial pathway. In this process, G-Rh2 inhibits the activity of ETC complex III and indirectly affects the activity of ETC complexes I and V, leading to the overproduction of mtROS, decreased ATP synthesis and bidirectional inhibition of OXPHOS and glycolysis. To the best of our knowledge, the present study was the first to elucidate that the mechanism through which mtROS overproduction induces apoptosis in HPV-positive cervical cancer cells following G-Rh2 treatment was through inhibiting the activity of ETC complex III. This artificial regulation of mtROS at ETC complex III may be of great significance for cancer treatment.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

The research was supported by key projects at the central government level: The Ability Establishment of Sustainable use for Valuable Chinese Medicine Resources (grant no. 2060302), the National Key Research and Development Program of China (grant no. 2017YFC1702104) and the National Natural Foundation of China (grant no. 81703663).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Authors' contributions

ML and JW conceived and designed the study. YL, SY, XX and JQ performed experiments. YY helped with the collection and assembly of data. YY, WZ and JQ analyzed the data and prepared the figures. ML, YL and WZ drafted and revised the manuscript. All authors read and approved the final manuscript. ML and WZ confirm the authenticity of all the raw data.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References