Curcumin suppresses gastric tumor cell growth via ROS-mediated DNA polymerase γ depletion disrupting cellular bioenergetics

Curcumin, oligomycin, carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP), antimycin A, rotenone, glucose were purchased from sigma (St. Louis, MO). Giemsa and crystal violet were purchased from Solarbio Bioscience & Technology (Shanghai, China). MTT assay kit, CCK-8, DCFH-DA ROS detection kit, JC-1 mitochondrial membrane potential detection kit, horseradish peroxidase (HRP)-conjugated anti-rabbit, anti-mouse immunoglobulin G were obtained from Beyotime (Haimen, China). BCA Protein Assay Kit and Pierce ECL Western Blotting Substrate were obtained from Thermo Scientific (Waltham, MA). The monoclonal antibody against β-actin ((#244586) was from Abmart (Shanghai, China). POLG (ab128899), COXI (ab14705), COXII (ab110258), COXIV (ab140643), p-p38T180/Y182 (ab38238), p38 (ab170099) were purchased from abcam (HKSP, New Territories, HK). p21 (#10355-1-AP), ND1 (#19703-1-AP), ND2 (#19704-1-AP) and CytB (#55090-1-AP) were purchased from Protech Group (Wuhan, China). p-ERK^{Thr202/Tyr204} (#9101), ERK (#9102), p-JNK^{Thr183/Tyr185} (#4668), JNK (#9252), phosphor-p53 antibody sampler kit (#9919), Bax (#2774), Bcl-2 (#2870), p-Akt^ser473 (#4060) and Akt (#9272) were obtained from Cell Signaling Technology. Apoptosis detection assay kit was purchased from BD science.

Human gastric cancer cell lines, SGC-7901 and BGC-823 were purchased from the Cell Bank of Shanghai Institute of Cell Biology (Shanghai, China).and cultured in RPMI 1640 (Life Technologies) supplemented with 10% fetal bovine serum (FBS) (Life Technologies) and antibiotics (100 U/ml penicillin and streptomycin) at 37 °C in a humidified incubator with 5% CO₂.

Three hundred SGC-7901 and BGC-823 cells/well were seeded into 6-well plate and cultured at 37 °C with 5% CO₂. Two weeks later, the cells were washed with pre-warmed PBS for 3 times, fixed with methanol for 20 min, and stained with crystal violet for 15 min. The cells were next washed with ddH₂O to eliminate residual crystal violet, and colony number was calculated by Image J software.

For cell proliferation assay, 2 × 10³ cells/well were seeded into five 96-well plates, treated with 10 μg/ml of curcumin for 0, 24, 48, 72, or 96 h, and the CCK-8 detection kit was used to determine the relative cell number. h. The OD values at 450 nm were read in a plate reader (Thermo Scientific). Cell viability was assessed using the MTT assay. Six × 10³ cells were plated into a 96-well plate, treated with curcumin at 0,2.5,5,10,20, or 40 μg/mL) for 24 h, and assay determined using the MTT kit according to the manufacture’s protocol. In brief, MTT reagent (5 μg/mL) was added to the cells and incubated for 4 h, the crystals produced were dissolved with formazan, and OD values read in plate reader (Thermo Scientific).

For apoptosis analysis, curcumin (10 μg/mL) pre-treated SGC-7901 and BGC-823 cells were washed with ice-cold PBS and collected. Add Annexin V-FITC/PI (BD, San Jose, CA) mixture followed by incubated at room temperature for 20 min protected from light. After that, the samples were subjected to BD Accuri^TM C6 flow cytometer (BD, Franklin Lakes, NJ). Intracellular ROS levels were measured as described using the fluorescence probe 2′, 7′-dichlorodihydrofluorescein diacetate (DCFH-DA) according to the manufacturer’s protocol (Beyotime, Shanghai, China). Cells were collected and washed with pre-warmed PBS, followed by incubated with DCFH-DA which dissolved in FBS-free 1640 medium at 37 °C for 20 min. The cells were washed with FBS-free 1640 medium for 2 times and analyzed by FACS. Similarly, mitochondrial superoxide production was determined from the curcumin-treated cells by MitoSOX staining dye according to manufacturer’s instructions, and analyzed by fluorescence microscopy and FACS; n = 6. For mitochondrial membrane potential analysis, curcumin (10 μg/mL for 4 h) treated SGC and BGC cells were stained with 2.0 μM JC-1 in complete medium, and incubated for 20 min at 37 °C in the dark. Cells were washed with PBS to remove the excess JC-1 dye, and evaluated by fluorescence microscopy.

Total RNA was extracted from SGC-7901 and BGC-823DNA polymerase γ knockdown cells according to the manufacturer’s protocol. cDNA was synthesized using the PrimeScript^TM RT reagent Kit with gDNA Eraser (Takara, Dalian, China). Quantitative real-time PCR assays were performed using 2 μl cDNA/20 μl reaction volumes on a CFX connect^TM real-time system (Bio-Rad) using SYBR Green (Bio-Rad) according to the manufacturer’s protocol. Primer sequences are provided in the. For cDNA amplification, the thermal cycling was performed using the following parameters: 95 °C for 10 min, 45 cycles of denaturation at 95 °C for 10 sec and extension at 60 °C for 30 sec. To quantify the mtDNA copy number, genomic DNA was isolated and mtDNA was probed with primers inside ND1 and nuclear DNA with primer inside β-actin. The threshold cycle number (CT) was recorded for each reaction. Each sample was analyzed in triplicate and repeated 3 times. All results are expressed as means ± SD.

Western blot analysis was performed as described previously. In brief, equivalent amount of protein extracts from whole-cell or tissues were separated by SDS-PAGE followed by electrophoretic transfer onto nitrocellulose membrane in Tris-glycine buffer. Block the membrane with 3% non-fat milk in a shaker at room temperature for 1.5 h followed by incubated with indicated primary antibodies at 4 °C overnight. After that, recycle the primary antibodies and wash the membrane 3 × 10 min, and incubated with corresponding secondary antibodies at room temperature for 1 h. Finally, membrane were washed with 1 × TBST for 3 × 10 min and reacted with ECL reagent according to the manufacturer’s protocol (Thermo Scientific, Rockford, IL) for 1 min followed by exposure to X-ray films.

Real time integrated cellular oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were measured using the Seahorse XF96 Extracellular Flux Analyser (Seahorse Bioscience, North Billerica, MA, USA) as previous described. In brief, SGC-7901 or BGC-823 cells were treated with 10 μg/mL curcumin for 12 h and 10³ cells were plated into the seahorse customized cell plates. After the probes were calibrated, the OCR was detected with sequential injection of the following compounds which regulate mitochondrial respiration: oligomycin (ATP synthase inhibitor; 1 μM), FCCP (uncoupler; 1 μM), rotenone (complex I inhibitor; 1 μM), and antimycin A (complex III inhibitor; 1 μM). Cellular aerobic glycolysis profile was determined by measuring ECAR from SGC-7901 or BGC-823 cells treated with curcumin. Measurements were made after sequential injection of glucose (10 mM), oligomycin (1 μM), or 2-DG (100 mM); n = 6.F. Basal cellular glycolytic rate in SGC-7901 and BGC-823 cells treated with or without 10 μg/mL curcumin. G. Cellular spared glycolytic capacity alteration of curcumin treatment in SGC and BGC cells.

BALB/c-nu/nu mice (6–8 wk old, female, ~20 g bodyweight) were purchased from Wenzhou Medical University laboratory animal center and fed under specific pathogen-free conditions. All experimental protocols and animal care were compiled with the Guide for the Care and Use of Laboratory Animals, Institute of Laboratory Animal Resources, and were approved by the Institutional Animal Care and Use Committee of Wenzhou Medical University. The nude mice were subcutaneously inoculated with 5 × 10⁶ cells of a 0.1 ml BGC cell suspension. And the tumor nodules were allowed to grow to a volume of ~100 mm³ before initiating treatment. The mice were randomized into three groups (six mice for each group). The three groups were intraperitoneally injected with normal saline (control), PBS and 25 mg/kg of curcumin water suspension, respectively. The body weight and tumor volume of each mouse were measured every 2 days over a period of 14 day. The tumor volume was calculated using the following equation: tumor volume = length × width × width/2. On day 14, the mice were sacrificed by overdose of sodium pentobarbital, and its tumor tissues were immediately harvested for further analysis.

All the statistical analysis was performed using SPSS 16.0 (SPSS Standard version 16.0, SPSS Inc., Chicago, IL). Student’s t test was used to compare data of different groups. The data are presented as the mean ± standard deviations (SD) values obtained from at least three independent experiments. P-values <0.05 was considered statistically significant.

To investigate the role of curcumin in regulating gastric tumor cell growth, we determined the effects of curcumin concentrations (0-40 μg/ml) on cell viability, and found a dose-dependent reduction of cell viability (Fig. 1a and b). Moreover, curcumin treatment (10 μg/ml) decreased cell proliferation as monitored for up to 5 days (Fig. 1b), significantly suppressed tumor cell colony formation in both SGC-7901 and BGC-823 gastric tumor cells (Fig. 1c). The impact of curcumin in initiating cell apoptosis was evaluated with the PI/FITC dye to identify apoptotic versus non-apoptotic cells by flow cytometry analysis. Results indicated that curcumin treatment (10 μg/ml) of both gastric cell types induced up to 6-fold increase over control of apoptotic cells (Fig. 1d). Current evidence indicated that curcumin likely promoted cell apoptosis via suppressing the Bcl-2 anti-apoptotic pathway as well as inactivating ERK/MAPK signaling. Thus, we asked whether curcumin-mediated gastric tumor cell apoptosis was also dependent on these pathways. Results indicated that curcumin reduced ERK activation and decreased Bcl-2 level which was accompanied with increased p53 serine phosphorylation (Fig. 1e). Figure 1f shows curcumin dramatically disrupted mitochondrial membrane potential, suggesting possible loss of mitochondrial homeostasis (Fig. 1f). Mitochondria contribute to major reactive oxygen species (ROS) generation, which is attributed to the “electron leak” of the electron transport chain (ETC). Although physiological levels of ROS play an important role in normal cell proliferation and cellular signaling, the excessive ROS released from mitochondria can lead to severe oxidative damage and the consequent cell apoptosis. We observed that curcumin treatment of the gastric tumor cells resulted in >2-fold increase in the cellular ROS (Fig. 1g). Significantly, the curcumin-induced increase in ROS content was localized to mitochondria (Fig. 1h). Collectively, our findings indicated that curcumin dramatically promoted cell apoptosis, which was likely through a mechanism of increased mitochondrial oxidative stress.

Mitochondria has a major role in cellular bioenergetics in most eukaryotic cells, being responsible for producing nearly 95% of cellular ATP through mitochondrial oxidative phosphorylation, thereby controlling cell death or survival. We speculated that curcumin would enhance oxidative damage of mitochondrial integrity, which could further limit cellular bioenergetics. To test this hypothesis, we determined the effects of curcumin on mitochondrial respiration (OXPHOS) and aerobic glycolysis using the seahorse 96XF Extracellular Flux Analyser (Methods). As shown in Fig. 2a, curcumin treatment reduced the intact cell oxygen consumption rate (OCR), indicating reduced OXPHOSin SGC-7901 and BGC-823 gastric cancer cells. Furthermore, we assessed specific mitochondrial functions, in particular, basal respiration, maximal respiration and ATP production. As shown in Fig. 2b, curcumin significantly reduced basal respiration which represents the basal mitochondrial OXPHOS activity. Furthermore, in the presence of the uncoupler agent FCCP, cellular oxygen consumption was dramatically increased, while this change was blocked by ETC inhibitors rotenone and antimycin A, inhibiting electron transfer through complex I and complex III. As well, curcumin markedly reduced maximal respiration (Fig. 2c). Integrated mitochondrial electron transfer chain, which drives the H⁺ pump, powers ATP Synthase to catalyze generation of ATP. In SGC-7901 and BGC-823 cells, we found that 10 μg/mL curcumin significantly reduced ATP production (Fig. 2d). These data indicated that curcumin suppressed mitochondrial respiration, resulting in reduced ATP supplement and thereby, restricted cancer cell growth. Otto Warburg had proposed that cancer is the result from the regression of cells to a more primitive metabolism, which is exhibited by proliferating eukaryotic cells, and high aerobic glycolysis may be the most shared metabolic process in cancer cells. We therefore, asked whether curcumin also regulates aerobic glycolysis to limit tumor growth. We determined the effects of curcumin on glycolysis using the seahorse XF96 Extracellular Flux Analyser (Method). Results indicated that curcumin treatment significantly we decreased aerobic glycolysis as measured by extracellular acidification rate (ECAR) in SGC-7901 and BGC-823 cells (Fig. 2e). To further understand the role of curcumin in regulating aerobic glycolysis, we analyzed indices of the basal glycolytic rate and spared glycolytic capacity. As shown in Fig. 2f, curcumin treatment reduced basal glycolytic rate, and together with the observed suppressed mitochondrial respiration, can dramatically inhibit overall tumor cell growth. Moreover, the spared glycolytic capacity was significantly blocked within curcumin (Fig. 2g). Taken together, our findings suggest that curcumin can suppress tumor cell growth likely by limiting cellular bioenergetics.

Mitochondrial DNA is known to be susceptible to oxidative damage because of a lack of efficient protective mechanisms such as histone protection. We postulated that curcumin-induced elevation of chronic ROS levels could damage mtDNA, leading to disruption of mitochondrial respiration. We investigated this possibility by measuring mtDNA copy number using real-time PCR and found that curcumin treatment resulted in reduced mtDNA level in SGC-7901 and BGC-823 cells (Fig. 3a). Moreover, the curcumin-decreased mtDNA was associated with decreased protein expression of POLG as well as subunits of the respiration complex, such as COXI, COXII, COXIV, CytB (Fig. 3b). POLG is critical for mtDNA replication, transcription and maintenance of mtDNA integrity. To investigate the effects of POLG in modulating mitochondrial respiration in gastric cancer biology, we knockdown POLG using specific target sequences of siRNA targeted which significantly reduced both POLG mRNA and protein levels (Fig. 3c). As expected, POLG knockdown results to reduced expression of subunits of the mitochondrial respiration complex (Fig. 3c). Also, the siRNA-induced knock-down of POLG was associated with reducedmtDNA copy number (Fig. 3d). Based on these findings, we asked whether POLG knockdown also lead to a similar cellular bioenergetics phenotype as that produced by curcumin. Therefore, we determined alterations of OCR (assessment of mitochondrial respiration) and ECAR (assessment of aerobic glycolysis) in gastric cancer cells with siRNA-induced POLG knockdown using the Seahorse 96XF Extracellular Flux Analyser. Results indicated the down-regulated POLG dramatically decrease overall mitochondrial respiration (Fig. 3e) and aerobic glycolysis (Fig. 3f). Additionally, we found that the downregulated POLG was associated with reductions of basal respiration (Fig. 3g), maximal respiration (Fig. 3h), ATP production (Fig. 3i), basal glycolytic rate, (Fig. 3j) and glycolytic capacity (Fig. 3k). We next asked whether excessive ROS generation induced by curcumin was responsible for the POLG depletion. For study, the ROS scavenger, NAC, was used to block curcumin-induced ROS generation in the gastric cancer cells. As predicted, NAC effectively rescued the curcumin-induced POLG depletion as well as the aerobic glycolysis (Fig. 3l-n). Together, our data indicated that curcumin inhibited cancer cell growthvia decreasing POLG, thereby reducing mtDNA content, which further suppressed cellular bioenergetics. These data indicates POLG is significantly involved in the anti-cancer effects of curcumin. Further work is needed to uncover the mechanism by which curcumin downregulates POLG.

To confirm our data in vivo, we used a xenograft nude mice model in which BGC cells were subcutaneously injected (Methods). Curcumin (25 mg/kg) or normal saline was delivered by IP and tumor size evaluated at 14 days. Results indicated that curcumin caused a progressive decrease in tumor volume as measured for up to 15 days (Fig. 4a). The body weight was not affected by curcumin during this treatment period (Fig. 4b), indicating normal growth condition of the nude mice. As expected, the curcumin-induced decrease in tumor volume was also associated with significant decrease in tumor weight (Fig. 4c). In vivo bioluminescence imaging was used for further confirmation by visualizing and quantification of the effects of curcumin on tumor suppression (Fig. 4d and e). Evaluation of the tumor histology (H&E) indicated that the curcumin-treated group presented with necrosis, resulting in lower cellularity compared to the mock or vehicle controls (Fig. 4f). The mtDNA in tumors from the curcumin-treated group was 50% lower than that of controls (Fig. 4g), which was associated with decreased POLG protein expression as well as the indicated proteins we tested (Fig. 4h). Additionally, the curcumin-treated tumors showed similar changes in expression of apoptosis-related proteins, signaling, and proteins of the respiration complex as those in curcumin-treated gastric cell lines. We next isolated tumor cells from the xenograft nude mice and knocked down POLG with siRNA in these tumor cells. Evaluation of the effects on cellular bioenergetics indicated ECAR and OCR were predictably suppressed (Fig. 4i and j, respectively). Collectively, our in vivo data validate the anti-cancer effects and POLG/cellular bioenergetics-involved mechanism of curcumin.

We further determined POLG expression in gastric tumors isolated from human subjects diagnosed with gastric cancer, with surrounding normal gastric tissues as comparison. Results indicated that POLG protein and mRNA were dramatically upregulated in tumor tissues compared to normal tissues (Fig. 5a-c). The tumors were digested for isolation of primary tumor cells for investigation of the effects of curcumin on cellular bioenergetics. We observed that curcumin significantly suppressed cellular OXPHOS (Fig. 5d) and glycolysis (Fig. 5e). Furthermore, curcumin decreased the indices of OCR, i.e., the basal respiration, ATP production and maximal respiration (Fig. 5f, g,-h, respectively). And as expected, the parameters of cellular glycolysis, i.e., basal glycolytic rate and spare glycolytic rate were both reduced with curcumin treatment (Fig. 5i and j, respectively). Furthermore, we depleted POLG using siRNA in the primary gastric cancer cells and determined the mitochondrial respiration and aerobic glycolysis. Consistent with data obtained from tumors in the nude mice and the gastric cancer cell lines, POLG depletion resulted in reduced respiration (Fig. 5k) and glycolysis (Fig. 5l). Together, the data obtained from gastric tumors of human subjects further confirmed that curcumin repressed gastric cancer cell bioenergetics.