Oxidative stress regulates cellular bioenergetics in esophageal squamous cell carcinoma cell

Esophageal squamous cell carcinoma (ESCC) is the most frequent histological subtype of esophageal cancer and is the sixth most common cause of cancer-associated mortality worldwide [1]. Despite advances in early diagnosis, surgery, and chemoradiotherapy, the prognosis of ESCC is still very poor and remains a challenge [2]. Thus, our understanding of mechanisms of ESCC tumorigenesis is urgent, which will facilitate the development of diagnosis marker and novel therapeutic strategies.

Unlike normal cells, tumor cells demand a vast amount of energy (ATP) and metabolites to support their rapid proliferation. Indeed, some tumor cells are addicted to glycolysis rather than oxidative phosphorylation (OXPHOS) even in normoxic conditions and need quick reprogramming of energy metabolism. This phenomenon is often referred to as the Warburg effect [3]. Importantly, glycolysis provides cancer cells with not only energy but also crucial intermediates for biosynthesis of macromolecules, lipids as well as NADPH which is important for redox homeostasis. Metabolic reprogramming is required for cell malignant transformation, tumorigenesis, invasion, metastasis, and resistance to cancer treatments [4]. In solid tumor, hypoxia is a crucial microenvironmental stimuli that leads to significant up-regulation of hypoxia-inducible factor-1α (HIF-1α) in majority of human cancers [5–7]. Cobalt chloride (CoCl₂), a well-known hypoxia-mimicking agent in vitro, which blocks HIF-1α degradation and thus causes HIF-1α accumulation in cells, therefore leading to an intracellular hypoxia-like microenvironment and enhancing tumor malignancy.

Because hydrogen peroxide (H₂O₂) can generate large amounts of oxygen free radicals and cause oxidative stress in tumor cells, it is widely used as an apoptosis inducer [8]. Indeed, many anticancer drugs promote cancer cell death through enhancing the intracellualr H2O2 generation [9,10].Thus, studying the effects of H₂O₂ on cancer cells is helpful to provide novel strategies for the prevention and treatment of cancers.

H₂O₂ is also an important signaling molecule in tumor cells, which can regulate cell signaling pathways and transcription factors at different levels [11,12]. For example, H₂O₂ can induce the activation of EGFR, which leads to the amplification of Ras signaling cascade and activation of mitogen-activated protein kinases (MAPKs) [13]. Moreover, H₂O₂ can indirectly regulate PTK-EGFR-Ras signaling in cancer cells, and finally the activation of MAPK, so that ERK, JNK, p38, three subfamilies signaling pathways are regulated [13,14].

Low concentration of H₂O₂ plays an essential role in regulating cell division and growth as a signaling molecule [15,16]; however, when the amount of H₂O₂ exceeds a certain critical value, the cell cycle will be blocked and even lead to apoptosis [17]. Therefore, the up-regulation of H₂O₂ amount in tumor cells is one of the strategies to kill tumor cells. This process can also be rescued by up-regulation of catalase expression. Overexpression of SOD inhibits tumor cell proliferation, metastasis, and other malignant phenotypes. Similarly, catalase or glutathione reductase can reverse the antitumor effect of SOD [18].

In the present study, we examined the effects of CoCl₂-induced hypoxia and H₂O₂ on the regulation of cellular bioenergy metabolism in ESCC cells and sought to investigate the underlying molecular mechanism.

RPMI-1640 medium and FBS (Life Technologies, Grand Island, NY); oligomycin, FCCP, rotenone, antimycin A, glucose, 2-DG, and CoCl₂ were purchased from Sigma (St. Louis, MO); Annexin V-FITC/propidium iodine (PI) apoptosis detection kit was from BD Pharmingen (San Diego, CA); 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA), horseradish peroxidase (HRP) conjugated anti-rabbit, anti-mouse Ig, trypsin were from Beyotime (Shanghai, China). The following antibodies were used: anti-COXI, anti-COXII, anti-COXIV, and anti-SDHA (Abcam, Cambridge, MA); anti-ND2, anti-NDUFA5, anti-NDUFA6, anti-NDUFA9, anti-HIF-1α, anti-ATP5A, anti-PARP, anti-Caspase-9, anti-Caspase-3, anti-cleaved Caspase-3 (Cell Signaling, Beverly, MA); anti-β-actin (Abmart, Shanghai, China); anti-Cyt. b (Santa Cruz Biotechnology, Santa Cruz, CA). All other chemicals used were of highest analytical grade and obtained from Sigma, unless otherwise stated.

The human ESCC cell line TE-1 was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). TE-1 cells were cultured in RPMI-1640 medium supplemented with 10% FBS and antibiotics (100 units/ml penicillin and 100 μg/ml streptomycin) at 37°C with 5% CO₂.

TE-1 cells were plated in 60-mm dishes at a density of 0.3 × 10⁶ and incubated with series concentrations of H₂O₂ for 24 h at 37°C with 5% CO₂. Inverted microscopy (Nikon, Tokyo, Japan) was used to examine the alterations in cell morphology and cell number.

Cells were harvested by trypsin, washed with PBS, and incubated with DCFH-DA at a final concentration of 10 µM in RPMI-1640 medium for 30 min at 37°C, then cells were washed three times with cold PBS. Intracellular reactive oxygen species (ROS) levels were measured as described previously by using the fluorescence probe DCFH-DA according to the manufacturer’s protocol (Beyotime, Shanghai, China) [19]. The excitation wavelength for DCF fluorescence is 488 nm and the emission wavelength for DCF fluorescence is 525 nm, respectively.

Harvested cells were washed with cold PBS twice, and incubated with 1× binding buffer at a concentration of 1 × 10⁶ cells/ml. Annexin V-FITC/PI (BD, San Jose, CA) was added, followed by incubation in the dark at room temperature for 20 min. Four hundred microliters of 1X binding buffer was added to each tube. The samples were analyzed by flow cytometry within 1 h.

Cells were washed with cold PBS and lysed in RIPA lysis buffer supplemented with protease and phosphatase inhibitors on ice for 20 min, followed by centrifugation at 13000 rpm for 20 min at 4°C, and the supernatants were collected. Protein concentrations of whole cell extracts were determined using the Pierce BCA protein assay kit. Cell extracts equivalent to 20 µg total protein were resolved in 10% SDS/PAGE gels followed by electrophoretic transfer on to nitrocellulose membrane (Bio–Rad, Hercules, CA) in Tris-glycine buffer. Blots were blocked at room temperature for 2 h in 3% non-fat milk in TBS-Tween (TBS-T) on a shaker, and then incubated with the primary antibodies overnight at 4°C. The membrane was washed in TBS-T for at least 3× forw 10 min and then incubated with HRP conjugated anti-rabbit or anti-mouse IgG at room temperature for 1 h with gentle shaking. Signal was detected using ECL according to the manufacturer’s protocol (Thermo Scientific, Rockford, IL) and exposed to X-ray films. β-actin was used as control for equal loading and the optical density was measured using National Institute of Health ImageJ software. Each experiment was repeated at least three times.

Total RNAs were extracted from cells using TRIzol reagent (Life Technologies, Carlsbad, CA) following manufacturer’s instructions. cDNA was synthesized from 2 μg purified total RNA using the PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Dalian, China). Quantitative real-time PCR (qRT-PCR) was performed as described previously [19] using gene-specific primers. The results were normalized to β-actin in the same samples. Each sample was analyzed in triplicate and repeated three times. The primer sequences used were as follows: COXI 5′-TCCGCTACCATAATCATCGCT-3′ (forward), 5′-CCGTGGAGTGTGGCGAGT-3′ (reverse); COXII 5′-CGACTACGGCGGACTAATCT-3′ (forward), 5′-TCGATTGTCAACGTCAAGGA-3′ (reverse); COXIV 5′-GCAGTGGCGGCAGAATG-3′ (forward), 5′-AGTCTTCGCTCTTCACAACA-3′ (reverse); ATPase6 5′-CTGTTCGCTTCATTCATTGC-3′ (forward), 5′-AGTCATTGTTGGGTGGTGAT T-3′ (reverse), ATPase8 5′-AAACTACCACCTACCTCCCTCAC-3′ (forward), 5′-GCAATGAATGAAGC ATTCATTGC-3′ (reverse), ND3 5′-GCGGCTTCGACCCTATATC-3′ (forward), 5′-TGGCAGGTTAGTT GTTTGTAGG-3′ (reverse), ND4 5′-TGAACGCAGGCACATACTTC-3′ (forward), 5′-TCTTGGGCAGT GAGAGTGAG-3′ (reverse), ND5 5′-CATTGTCGCATCCACCTTTA-3′ (forward), 5′-CTGGGTTGTTTGG GTTGTG-3′ (reverse), ND6 5′-GGTGCTGTGGGTGAAAGAGT-3′ (forward), 5′-CTCCCGAATCAA CCCTGAC-3′ (reverse), β-actin 5′-GCCAGTGGACTCCACGAC-3′ (forward), 5′-CAACTACATGGTTTA CATGTTC-3′ (reverse).

The intact cellular oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of TE-1 cells were measured using a Seahorse XF-96 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, MA) as previously described [19]. Results were obtained by performing three independent experiments with eight replicates. After the assay was completed, a BCA Protein Assay Kit was used according to the manufacturer’s instructions to determine the protein concentration to normalize OCR and ECAR.

All experiments were performed at least three times and the data obtained were analyzed using SPSS version 16.0 (SPSS, Chicago, U.S.A.). The values are presented as the mean ± S.D. For the evaluation of two groups, Student’s t test was used. P<0.05 was considered to indicate a statistically significant difference.

To determine the effect of CoCl₂ on TE-1 cells, we first measured the intracellular ROS level in CoCl₂-treated TE-1 cells by flow cytometry. CoCl₂ treatment significantly increased the ROS production of TE-1 cells as shown in Figure 1A. Next, we treated TE-1 cells with the indicated concentration of CoCl₂ (0, 50, 100, and 200 μM) and analyzed the expression of HIF-1α, complex I subunits (ND1, NDUFA5, NDUFS6, and NDUFA9), complex II subunit (SDHA), complex III subunit (Cyt. B), complex IV subunits (COX II and COX IV) and complex V subunit (ATP5A) in TE-1 cells by Western blot. We found that the expression of HIF1-α protein increased in a dose-dependent manner of COCl₂. However, the expression of ND1, cytochrome b and COX II encoded by mtDNA and NDUFA5, NDUFS6, NDUFA9, SDHA, COX IV, and ATP5A encoded by nuclear gene were decreased in a dose-dependent manner of CoCl₂ treatment (Figure 1B). To clarify whether CoCl₂ regulates the protein expression or transcription, we further examined the mRNA levels of these proteins (Figure 1C).

Taken together, our findings indicated that CoCl₂ may inhibit mitochondrial respiration in TE-1 cells.

In order to further study the effect of CoCl₂ on cellular bioenergetics metabolism, we used Seahorse XF96 Extracellular Flux Analyzers to detect the OCR and found that OCR in TE-1 cells decreased significantly after treating with CoCl₂ for 24 h (Figure 2A). The production of ATP, basal respiration, and maximal respiration was markedly reduced and the difference was statistically significant (Figure 2B). In addition, we detected the ability of glycolysis in TE-1 cells when treated with CoCl₂, as result showed that when compared with the negative control, the glycolysis ability of TE-1 cells significantly increased under the treatment of CoCl₂ and the difference was statistically significant (Figure 2C,D).

HIF-1α was one of the important transcription factors in tumor development and progression, contributed to cell survival, and activation of gene expression under hypoxic condition. The target genes mainly related to metabolism of carbohydrates that include glycolytic enzymes, aldolase A, and glucose transporter protein-1 (GLUT-1). We hypothesized that ESCC cell TE-1 may switch cellular energy metabolism from mitochondrial OXPHOS to glycolysis under hypoxic conditions stimulated by CoCl₂. On the one hand, TE-1 cells inhibited the expression of mitochondrial complex subunits by increasing ROS level; on the other hand, TE-1 cell enhanced glycolysis ability by increasing the expression of glucose metabolism related enzymes. To demonstrate our hypothesis, we set three groups: the negative control group, CoCl₂ treated group, and both CoCl₂ and N-acetyl cysteine (NAC, ROS scavenger) treated group. Western blot was used to detect mitochondrial complex subunits protein expression in the three groups. We found that the subunits of mitochondrial complex recovered obviously in the group of TE-1 cells treated with CoCl₂ and NAC simultaneously (Figure 3A). Meanwhile, by using Seahorse Bioenergetics Analyzer to measure OCR, we found that NAC could significantly rescue mitochondrial respiration in TE-1 cells treated with CoCl₂ (Figure 3B). The difference was statistically significant (Figure 3C). Additionally, we found that NAC could rescue CoCl₂ induced up-regulation of aerobic glycolysis. As shown in Figure 3D, NAC treatment decreased aerobic glycolysis enhanced by CoCl₂. Moreover, we assessed the basal glycolytic rate, spare glycolytic, and maximal glycolytic rate, which indicated that NAC treatment could reverse CoCl₂ induced aerobic glycolysis (Figure 3E). Hence these results suggested that ESCC cell TE-1 maintained cell survival by the transformation of energy metabolism under conditions of low oxygen caused by CoCl₂.

As shown in Figure 4A, we treated TE-1 cells with the indicated concentration of H₂O₂. We observed cell morphological changes under phase contrast microscopy. With the increase in the concentration of H₂O₂, the cell viability was decreased compared with the control group as well as the cell size became smaller, indicating that the cells tended to apoptotic state. To explore the effect of H₂O₂ on TE-1 cells, we analyzed the intracellular ROS level by flow cytometry. The result showed that compared with the negative control group, intracellular ROS level increased in a dose-dependent manner in H₂O₂-treated TE-1 cells (Figure 4B). In order to further examine the apoptotic rate of cells, we detected the apoptosis-related proteins by Western blot and the results showed that cleaved-PARP, cleaved-Caspase 3, and cleaved-Caspase 9 were increased in a dose-dependent manner with H₂O₂ treatment from 100 to 400 μM (Figure 4C). These data indicated that H₂O₂ could induce TE-1 cells death that is mediated through caspase activation. We also analyzed the effects of H₂O₂ and CoCl₂ on cell growth, and found that both H₂O₂ and CoCl₂ could suppress cell growth in TE-1 cells. However, H₂O₂ had a stronger suppressive activity than CoCl₂ on TE-1 cells (Supplementary Figure S1). We further detected the cell death of CoCl₂-treated cells, however, the treatment of CoCl₂ had no significant effect on cell death of TE-1 cells (Supplementary Figure S2).

Next, we assessed the protein level in mitochondrial respiratory chain complex subunits in TE-1 cells treated with H₂O₂ by using Western blot analysis. As shown in Figure 5A, the expression of complex І subunits (ND2, NDUFA5, NDUFS6, and NDUFA9), complex III subunit Cyt. B and complex IV subunits (COXI and COXIV) were decreased in TE-1 cells treated with H₂O₂. However, there were no significant difference in the protein expression level of complex II subunit SDHA and complex V subunit ATP5A. To further investigate the effects of H₂O₂ on TE-1 cellular bioenergetics, we analyzed mitochondrial respiration by determining the OCR (Figure 5B), and found that cells under the treatment of H₂O₂ displayed a lower basal respiration and a significantly lower maximum respiratory capacity and accompanied by less ATP production compared with the negative control group (Figure 5C). In addition, we found that the glycolysis rates of TE-1 cells were significantly decreased after H₂O₂ treatment (Figure 5D,E). To further confirm the reduction in respiration and aerobic glycolysis were caused by H₂O₂ treatment, we used ROS scavenger NAC to investigate whether NAC could rescue these effects. As shown in Figure 6A, we found that NAC treatment dramatically reverses H₂O₂ induced mitochondrial respiration reduction, as well as ATP production, basal respiration, and maximal respiration (Figure 6B). Moreover, we detected the alterations in ECAR, and found that NAC could also rescue aerobic glycolysis suppressed by H₂O₂ (Figure 6C). In addition, basal glycolytic rate, spare glycolytic rate, and maximal glycolytic rate were assessed (Figure 6D). Consistently, NAC could reverse these three indexes of aerobic glycolysis. These results indicated that H₂O₂ could cause the reduction in cellular bioenergetics of TE-1 cells via decreasing the expression of mitochondrial respiratory chain complex subunits.

Based on the above data, we speculated that H₂O₂ induced apoptosis in TE-1 cells may be due to cellular energy metabolism disorders that were caused by the increase in ROS. Therefore, we tested the hypothesis by adding ROS scavenger (NAC) simultaneously, and as shown in Figure 7A, TE-1 cells were treated with H₂O₂ in the absence or presence of NAC, and we observed that apoptosis reduced obviously in the presence of NAC. Moreover, we examined the apoptosis by flow cytometry, with the increased concentration of H₂O₂, the apoptosis rate gradually increased, but when TE-1 cells were treated with NAC simultaneously, we found that apoptosis induced by H₂O₂ was significantly reduced (Figure 7B,C). Thus, we believed that H₂O₂ may induce the production of ROS and cause progressive oxidative damaged, mitochondrial dysfunction, cell bioenergetics metabolism disorders, and ultimately cell death.

Accumulating studies have shown that the development of malignant tumors is actually a process of tumor cell metabolism reprogramming [20,21], which is one of the hallmarks of most human tumors [22]. Mitochondria as key actors in cancer metabolic reprogramming, not just because these organelles play a significant role in energy production [23] and biosynthetic intermediates formation, as well as occurrence of mutations in both nuclear and mtDNA encoding metabolic enzymes is associated with different types of cancer [24,25]. Glycolysis is an oxygen-independent metabolic pathway that converts glucose into pyruvate via a series of intermediate metabolites and leads to a net production of ATP [26]. OXPHOS is the metabolic pathway in which cells use enzymes to oxidize nutrients, thereby releasing energy which is accounts for high ATP yield [27]. OXPHOS is a critical part of metabolism, however, it also produces ROS which leads to damaging cells and contributes to diseases and agieng. The Warburg effect suggests that the levels of different cellular energy metabolism are not the same [28]. Cancer cells mainly produce ATP by a high rate of glycolysis followed by lactic acid fermentation in the cytosol, rather than by a relatively low rate of glycolysis followed by oxidation of pyruvate in mitochondria as in most normal cells [29]. Previous studies have shown that mutations in mtDNA, mitochondrial fusion and fission dysfunction and mitochondria-related active enzyme mutation can cause normal cells transformed into cancer cells [30,31]. The enzymes involved in glycolysis are also the target of many anticancer drugs that inhibit their activities or regulate gene expression [32].

It is well known that ROS produced by aerobic cells during metabolic processes and have important roles in cell signaling and homeostasis [33,34]. However, while undergoing a wide variety of environmental stress, ROS levels can increase dramatically thus leading to oxidative damage to cells [35]. Some studies indicate that ROS levels significantly increased in tumor cells and plays a vital role in the occurrence and development of tumors and also participates in many signaling pathways, such as mitochondrial mediated caspase-dependent apoptotic pathway [36] and regulation of Akt/MAPK-related signaling pathways [37], and even affect the proliferation and differentiation of tumor cells [38–40].

Due to mitochondria being the major sources of ATP or ROS, more and more attention has been focussed on the study of the relationship between energy metabolism and tumor [41–43]. Our present study investigated the mechanism of oxidative stress on the biological energy metabolism of ESCC cells. We used the oxidative stress inducer CoCl₂ and H₂O₂ to treat TE-1 cells to study the effects of different stress on cellular bioenergy metabolism.

First of all, we utilized CoCl₂ to simulate hypoxic microenvironments. Under hypoxic conditions, HIF-1α is no longer degraded and translocated to the nucleus regulating hundreds of genes, and many of them play important roles in cancer cell metabolism [44]. Our findings demonstrate that CoCl₂ induced the production of ROS and reduced the expression of mitochondria respiratory chain enzyme complex subunits due to reduced gene transcription and protein synthesis. To further confirm the effects of CoCl₂ on the energy metabolism of TE-1 cells, we detected the OCR and ECAR to evaluate the ability of TE-1 cells OXPHOS and glycolysis, our results indicate that the mitochondrial respiration was significantly decreased, accompanied by the increase in glycolysis rates. This is consistent with the findings in breast cancer [27]. Additionally, we used the ROS scavenger (NAC) to verify our previous hypothesis, and NAC could dramatically rescue from the effects induced by CoCl₂, which suggest that tumor cell may undergo the metabolic reprogramming to survive under adverse conditions.

Second, H₂O₂ as a widely used inducer of cell damage and apoptosis, is capable of producing large amounts of oxygen free radicals that cause cellular oxidative stress [45,46]. Our data proved that H₂O₂ indeed could induce apoptosis and inhibit the expression of mitochondria respiratory chain complex subunits. Moreover, we found that H₂O₂ could dramatically suppress the OCR and ATP production and simultaneously inhibited the cellular glycolytic rate. Hence, we proposed that the apoptosis induced by H₂O₂ may be due to the energy metabolism dysfunction that is caused by the increase in ROS levels. This was further confirmed that NAC pretreatment significantly abolished H₂O₂-induced apoptosis in TE-1 cells. These results demonstrate that oxidative stress caused by H₂O₂ leads to severe impairment in TE-1 cell energy metabolism. Moreover, our findings show that H₂O₂ mediates apoptosis of TE-1 cells through activating caspase 3 and caspase 9. Therefore, H₂O₂ induced the increase in ROS levels, causing cell oxidative damage, mitochondrial dysfunction, and resulting in cell biological energy metabolism disorder, and ultimately induced apoptosis.

In summary, we found that ESCC TE-1 cells exhibit distinct changes in bioenergy metabolism in response to different oxidative stress. Under hypoxic condition, TE-1 cells are able to switch their metabolic phenotype from OXPHOS to glycolysis as an alternative source of biogenetic substrates to maintain cell survival. However, TE-1 cells are unable to reprogram metabolism when they are exposed to H₂O₂, and eventually leads to cell death due to energy deficiency. This may provide as a novel strategy for treatment of ESCC, and which is also helpful to solve the problems of drug resistance of ESCC by combination with current treatment and realize the specific targetted therapy of tumor cells.

We thank Dr Charles Reichman and members of the Lu lab for their suggestions on this manuscript.

The authors declare that there are no competing interests associated with the manuscript.

B.L. and L.L. conceived the idea, supervised the research, and wrote the manuscript. X.Z., L.L., L.N., J.L., C.L., M.G., S.M., and J.L. performed the experiments. Y.L., L.L., and B.L. performed data analyses. All the authors read and approved the final manuscript.

This work was supported by the National Natural Science Foundation of China [grant numbers 31570772, 31070710, 31171345 (to B.L.)]; the Natural Science Foundation of Zhejiang Province [grant number LY17C070005 (to Y.L.)]; and the Wenzhou Municipal Science and Technology Bureau [grant number Y20140670 (to X.Z.)].

Cyt. b

Cytochrome b

DCF

Dichlorofluorescein

DCFH-DA

Dichlorodihydrofluorescein diacetate

2-DG

2-Deoxy-D-glucose

ECAR

extracellular acidification rate

EGFR

epidermal growth factor receptor

ERK

extracellular regulated protein kinases

ESCC

esophageal squamous cell carcinoma

FCCP

carbonylcyanide-p-trifluoromethoxyphenylhydrazone

HIF-1α

hypoxia-inducible factor-1α

HRP

horseradish peroxidase

JNK

c-Jun N-terminal kinase

MAPK

mitogen-activated protein kinase

NAC

N-acetyl cysteine

OCR

oxygen consumption rate

OXPHOS

oxidative phosphorylation

PARP

poly ADP-ribose polymerase

PI

oropidium iodine

PTK

protein tyrosine kinase

qRT-PCR

quantitative real-time PCR

ROS

reactive oxygen species

RPMI--1640

roswell Park Memorial Institute-1640

SOD

superoxide dismutase

TBS-T

TBS-Tween