Advertisement
Tumor and Stem Cell Biology

hTERT Overexpression Alleviates Intracellular ROS Production, Improves Mitochondrial Function, and Inhibits ROS-Mediated Apoptosis in Cancer Cells

Inthrani R. Indran, Manoor P. Hande and Shazib Pervaiz
DOI: 10.1158/0008-5472.CAN-10-1588 Published January 2011

Abstract

The human telomerase reverse transcriptase (hTERT) is the catalytic subunit of the telomerase holoenzyme. Evidence is accumulating to link hTERT to activities other than telomere maintenance and immortalization. Here, we show that hTERT overexpression not only reduces the basal cellular reactive oxygen species (ROS) levels but also inhibits endogenous ROS production in response to stimuli that induce intracellular ROS generation. Conversely, siRNA-mediated gene silencing of hTERT potentiated the increase in cellular ROS levels following exposure to oxidative stress. This antioxidant effect of hTERT is mediated via a significant increase in the ratio of reduced to oxidized glutathione (GSH:GSSG) as well as efficient recovery of the oxidized peroxiredoxin to its nonoxidized form. Our data also provide evidence for mitochondrial localization of hTERT, and a significantly higher activity of cytochrome C oxidase, the rate-limiting enzyme in the mitochondrial electron transport chain, in hTERT overexpressing cells. To ascertain whether the improved mitochondrial function and antioxidant effect of hTERT could provide cancer cells with a survival advantage, the effect of oxidative stress on mitochondrial apoptosis was evaluated. Indeed, hTERT overexpressing cells inhibited cytosolic acidification, translocation of Bax, the drop in mitochondrial transmembrane potential, the release of cytochrome C to the cytosol, and cell death. Taken together, these data demonstrate a hitherto undefined role of hTERT in alleviating cellular ROS levels by way of potentiating the cellular antioxidant defense systems, and in doing so endowing cancer cells with the ability to evade death stimuli. Cancer Res; 71(1); 266–76. ©2010 AACR.

Introduction

Telomeres are specialized structures that cap and preserve chromosomal integrity by protecting chromosomal ends from degradation, end to end fusions, and rearrangement (1). Telomerase, a ribonucleoprotein, is essential for the maintenance of telomeres. Telomerase consists of a catalytic protein subunit, human telomerase reverse transcriptase (hTERT), the RNA component of the telomerase (hTR or hTERC) that is used as the template by the reverse transcriptase to elongate the telomeric ends, and telomerase-associated proteins such as TEP1, POT1, and TPP1 that play a primary role in maintaining the integrity of the telomeres in normal cells (2,3). In recent years, roles deviating from the canonical activity of hTERT, namely cellular immortalization, telomerase activity, and overcoming cellular senescence, have been demonstrated with experimental models (4, 5). Toward that end, hTERT has been implicated in telomerase-mediated DNA repair (5), control of cell growth, mitochondrial function (4), and modulation of cells' response to apoptotic stimuli (6, 7).

There is accumulating evidence that cancer cells' response to apoptotic insults is regulated by the cellular redox status. In this regard, we previously demonstrated divergent effects of the 2 major reactive oxygen species (ROS), superoxide (O2−) and hydrogen peroxide (H2O2), on cell death signaling in cancer cells, whereby a slight tilt in favor of O2− inhibited death execution and provided a survival advantage while an increase in intracellular H2O2 created a permissive milieu for death execution (8–10). Interestingly, intracellular ROS have been shown to regulate hTERT at many levels, such as regulation of its expression and activity as well as modulating its localization and in doing so altering its functions (4, 11, 12). Despite convincing evidence that cellular redox status impacts hTERT biology, the effect of hTERT on cancer cell redox milieu remains less well defined. In this regard, studies have suggested a protective cellular response against oxidative stress in hTERT overexpressing cells (13–15). In addition, reports have also suggested that hTERT expression positively impacts mitochondrial function by improving its calcium buffering capacity and reducing O2− production (16). More recently, it was shown that mitochondrial hTERT exerted a novel protective function by binding to mitochondrial DNA, increasing respiratory chain activity, and protecting against oxidative stress-induced damage (17).

Here we investigated the effect of hTERT expression on intracellular redox status, as well as the cellular response to ROS-induced apoptotic stimuli. In this regard, the effect on cells' antioxidant defenses and the activity of cytochrome C oxidase (COX; complex IV), the rate-limiting step in the mitochondrial electron transport and a reliable endogenous marker of oxidative metabolism (18), were assessed. Results indicate that hTERT expression antagonizes ROS-induced ROS production, thereby preventing toxic increase in intracellular ROS and bestowing survival advantage upon cancer cells.

Materials and Methods

Cell lines, culture conditions, and plasmids

The human cervical cancer cell line HeLa (CCL-2) and the human neuroblastoma cell line SH-SY5Y (CRL-2266) were purchased from American Type Culture Collection (ATCC). Both the cell lines were characterized and authenticated by the source cell bank ATCC. Normal fetal lung fibroblast cell line MRC-5 was obtained from Coriell Cell Repositories. The cell lines used throughout this study were not used for more than 3months and/or 20 passages, following purchase from the authentic repositories. MRC-5-hTERT cells were generously provided by Dr. Hidetoshi Tahara (Hiroshima University, Hiroshima, Japan). HeLa cells and SH-SY5Y cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) and RPMI-1640, respectively, supplemented with 1% l-glutamine, 1% streptomycin-penicillin (Gemini Bio-Products; v/v), and 10% fetal bovine serum. HeLa cells stably transfected with pBabe-Neo (Neo cells) or pBabe-Neo-hTERT (HT1 cells) plasmids were maintained in 500 μg/mL of G418 disulphate salt solution (Roche Diagnostics Corp). Both MRC-5 and MRC-5-hTERT cell lines were grown in DMEM-F12 supplemented with 1% l-glutamine, 1% penicillin/streptomycin, 1mmol/L of sodium pyruvate solution, and 1× nonessential amino acids. All tissue culture–related reagents were purchased from Hyclone unless otherwise stated. For cellular assays, cells were plated in G418-free complete medium over 2days before treatment or analysis. The plasmids pBabe-Neo and pBabe-Neo-hTERT were a generous gift from Dr. Fabrizio d'Adda di Fagagna [Italian Foundation for Cancer Research Institute ofMolecular Oncology/European Institute of Experimental Oncology (Milan, Italy)].

Flow cytometric analysis of intracellular ROS

Cells were loaded with 5-(and-6)-chloromethyl-2-,7-dichlorofluorescin diacetate (DCHF-DA) for cellular ROS measurement or MitoSox Red (Molecular Probes) for intramitochondrial O2− assessment as described previously (19). Briefly, the cells were harvested and washed with 1× PBS, followed by incubation with 5 μmol/L DCHF-DA or 10 μmol/L MitoSox Red for 15 minutes at 37°C in the dark. The cells were then washed in 1× PBS, resuspended in plain DMEM, and analyzed by flow cytometry (Coulter EPICS Elite ESP). At least 10,000 events were analyzed.

Measurement of mitochondrial membrane potential using DiOC6

Cells were washed and loaded with 5,3,3′-dihexyloxacarbocyanine iodide [DIOC6(3); Sigma Aldrich] as described above and previously (20) and analyzed by flow cytometry with excitation and emission wavelengths at 488 and 525nm, respectively.

Measurement of intracellular pH with 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein

Intracellular pH (pHi) was measured by loading cells with membrane-impermeant dye 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester [(BCECF-AM) Sigma Aldrich] as described elsewhere (21). Briefly, the cells were loaded with 40 nmol/L of BCECF-AM for 15 minutes at 37°C in the dark, washed, resuspended in plain DMEM, and analyzed by flow cytometry. The fluorescence ratio of 525:610 nm was used to derive cytosolic pH using a standard pH calibration curve.

Assessment of intracellular GSH:GSSG levels

The glutathione assay was optimized from the protocol by Hissin and Hilf (22). Reduced and oxidized glutathione (GSH and GSSG) and N-ethylamide (NEM) solution were obtained from Sigma Aldrich. Cells (0.2 × 106) were seeded over 2days in a 6-well plate before treatment. Following H2O2 treatment, the cells were washed once with chilled PBS. Three hundred sixty microliters of chilled trichloroacetic acid (6.5%) was added to each well and the plates were left at 4°C for 10 minutes. o-Phthaladehyde [(OPT) Sigma Aldrich] solution was freshly prepared in reagent grade absolute methanol (1 mg OPT/mL of methanol; Merck). After 20 minutes of incubation in the dark, the plates were excited at 350nm and fluorescence emission was detected at 420nm with POLARstar OPTIMA microplate reader (BMG Labtech).

Glutathione peroxidase and reductase assays

The glutathione peroxidase (GPx) assay was optimized from the protocol provided by the manufacturer (Sigma Aldrich), and cumene hydroperoxide was used as the substrate to measure the total GPx activity. NADPH and cumene hydroperoxide were obtained from Sigma Aldrich. The activity was measured by reading the absorbance at 340nm once every minute for 15 minutes in a microplate reader (Tecan Spectrofluoroplus). The glutathione reductase (GR) assay was performed as described in the Cayman glutathione reductase assay kit (Cayman Chemical Company).

Isolation of mitochondrial and cytosolic fractions and assessment of cytochrome c oxidase activity

Mitochondrial and cytosolic fractions were obtained as described elsewhere (23). Cytochrome c oxidase (COX) activity was assessed as described previously (24). For details refer to the Supplementary Methods.

Statistical analysis

All experiments were performed at least 3 times for statistical significance. Numerical data were expressed as mean ± SD. Statistical analysis was performed using the paired Student's t test considering the variances unequal. P < 0.05 was considered significant.

Results

hTERT Expression alleviates basal intracellular ROS levels

To address the effect of hTERT expression on ROS, hTERT was stably expressed in HeLa cells that had a moderate background expression of hTERT (Fig. 1A). The cells stably transfected with pBabe-Neo (Neo) and pBabe-Neo-hTERT (HT1 and HT2) were then analyzed for basal intracellular ROS levels. Flow cytometric analysis revealed that HeLa cells overexpressing hTERT displayed significantly lower basal and mitochondrial ROS levels. The G-mean values of dichlorofluorescein fluorescence obtained from 6 independent experiments were used to calculate the fold difference in ROS levels among Neo, HT1, and HT2 cells (Fig. 1B). In addition, the assessment of mitochondrial O2− levels using MitoSox Red also displayed a 10% to 15% decrease in mitochondrial O2− in HT1 and HT2 cells (Fig. 1C). These findings were strengthened by transient hTERT overexpression in HeLa cells (Supplementary Fig. S1A and S1B) or upon siRNA-mediated gene silencing of hTERT in SH-SY5Y cells that normally express higher levels of hTERT (Supplementary Fig. S1C); hTERT silencing increased basal levels of intracellular ROS. Similarly, intracellular ROS levels were significantly lower in normal human lung fibroblasts (MRC-5) immortalized with hTERT (MRC-5-hTERT), compared with the control MRC-5 cells (Supplementary Fig. S1D).

Figure 1.
Figure 1.

Stable hTERT overexpression reduces basal intracellular ROS. A, hTERT expression in untransfected HeLa cells (ATCC), p-Babe-Neo–transfected cells (Neo), and 2 independent heterogeneous populations of hTERT-transfected cells (HT1, HT2) were assayed by Western blot analysis. β-actin was used as a loading control. Basal levels of intracellular ROS were detected by flow cytometry in HeLa Neo, HT1, and HT2 cells using the probes (B) DCHF-DA and (C) MitoSox Red and the average G-mean values were computed. Neo and HT1 cells (D) were exposed to about 300 μmol/L of H2O2 or 100 μg/mL of C1 for 3 hours following which cellular ROS was assessed using the probe DCHF-DA. The density plot represents data from 4 independentexperiments. *, P < 0.01; **, P < 0.005.

hTERT Overexpression blocks ROS-induced ROS production

Because transient hTERT expression resulted in a decrease in basal levels of intracellular ROS, it was of interest to investigate if hTERT overexpressing cells were better able to cope with stimuli that increased ROS production. To assess that possibility, we evaluated the response of hTERT overexpressing cells to the exogenous addition of H2O2 or upon exposure to a small molecule compound C1 that we previously showed as a strong stimulus for intracellular ROS production (23, 25). Results showed that hTERT overexpressing HT1 cells displayed a significant decrease in intracellular ROS levels following treatment with H2O2 or C1 (Fig. 1D). Of note, the response of HT1 cells to C1 was significantly more pronounced than exogenously added H2O2, thus indicating that the hTERT overexpressing cells were not only resistant to exogenous H2O2 but were also able to resist drug-induced intracellular ROS production. Corroborating these data, an improved resistance to oxidative stress was also observed in HeLa cells transiently transfected with hTERT or MRC-5-hTERT cells following treatment with H2O2 or C1 (Supplementary Fig. S2A, S2B, and S2D). Conversely, hTERT silencing in SH-SY5Y cells resulted in a significant increase in intracellular ROS levels in response to 600 μmol/L H2O2 for 4 hours (Supplementary Fig. S2C).

hTERT Overexpressing cells display enhanced glutathione antioxidant defense capacity

Intrigued by the resistance of hTERT-expressing cells to oxidative stress, we next investigated the mechanism of cellular adaptation/resistance to oxidative stress in this model system. Thus the GSH/GSSG couple was investigated to determine if hTERT overexpression influenced this redox couple, thus altering the cellular capacity to resist oxidative stress. As such, Neo and HT1 cells were exposed to H2O2 from 0 to 8 hours and assayed for their intracellular ratios of GSH to GSSG as described in the Materials and Methods section. Indeed, hTERT-overexpressing cells maintained a higher GSH/GSSG ratio following treatment with H2O2, compared with mock-transfected cells (Fig. 2A). These results were corroborated in MRC-5-hTERT cells (Supplementary Fig. S3Ai). Furthermore, silencing hTERT in SH-SY5Y cells resulted in a rapid decline in the GSH/GSSG ratio upon the exogenous addition of H2O2 (Supplementary Fig. S3Aii). These data provide evidence to implicate the glutathione metabolism in the protective effect of hTERT against oxidative insults.

Figure 2.
Figure 2.

hTERT-overexpressing cells express enhanced glutathione antioxidant defense system and a more rapid recovery of peroxiredoxins from the hyperoxidized state. Neo and HT1 cells were treated with 300 μmol/L of H2O2 from 0 to 8 hours following which (A) intracellular GSH/GSSG ratio was computed or whole cell lysates were prepared and subjected to (B) immunoblot analysis of GCLC or (D) Prx I, II, III, Trx I, and Prx SO3. The antibody recognizesPrx I/II SO3 and Prx III SO3. C, Neo and HT1 cells were used to prepare the sample for the GPx activity assay as described earlier, and the fold difference was calculated. Data shown are mean ± SD of at least 3 independent experiments and (C) was performed in duplicates. *, P < 0.05; **; P < 0.005.

Approximately 10% of cellular GSH is found in the mitochondria. Because mitochondria do not contain catalase, they depend upon GSH peroxidase and nonenzymatic reaction with GSH to protect against ROS and more specifically peroxide-mediated toxicity. Therefore, in addition to the total intracellular glutathione levels we also investigated the effect on mitochondrial GSH. Neo and HT1 cells were treated with 300 μmol/L of H2O2 for 4 and 8 hours following which the mitochondrial fractions were isolated and glutathione levels were assessed. Results show that the rate of reduction of GSH levels following H2O2 treatment was much slower in HT1 cells than in the Neo cells (Supplementary Fig. S3B).

Because the GSH/GSSG ratios were higher in HT1 cells as compared with the Neo cells, the level and activity of enzymes in the glutathione regulation pathway were assayed next. Glutamate cysteine ligase (GCL) catalyzes the rate-limiting reaction in GSH synthesis. The catalytic subunit of this ligase, GCLC, confers the total enzymatic activity to GCL. Thus, its expression was analyzed by Western blotting following treatment with 300 μmol/L of H2O2 from 0 to 8 hours. In HT1 cells, the upregulation of GCLC was detected as early as 2 hours and remained induced for 6 hours after treatment as compared with the Neo cells in which a late and transient upregulation was observed 6 hours following exposure to H2O2 (Fig. 2B). This could contribute to the increased GSH synthesis and the increased GSH/GSSG ratio upon hTERT overexpression.

Next, the activities of GPx and GR, important regulators of glutathione levels, were assessed. The net GPx (Fig. 2C) and GR activity were significantly higher in HT1 cells as compared with Neo cells (Supplementary Fig. S3C).

hTERT Expression regulates redox status of peroxiredoxins

The thiol-dependent peroxiredoxins (Prx) and thioredoxins play a critical role in removing cellular peroxides. Though Prx I and Prx II are catalytically less efficient than GPx or catalase, they are more abundantly found in the cytosol and display a greater affinity for H2O2. This makes them prime candidates as regulators of H2O2 signaling (26). Prx III is localized mainly to the mitochondria and serves as one of the main antioxidant defenses against peroxides at the mitochondria. By studying the rate at which these peroxiredoxins might be returned to their active state in cells the cellular Prx activity could be indirectly assessed (27). We show here that as the levels of Prx I, II, and III remained unchanged following treatment with H2O2 in Neo and HT1 cells, the amount of hyperoxidized peroxiredoxins diminished more rapidly in the HT1 cells as compared with Neo cells (Fig. 2D). The process of recovery from the hyperoxidized state without changes in total expression has been suggested as a process of regeneration of native peroxiredoxins from the hyperoxidized forms (27, 28). Substantiating these findings, hTERT silencing in SH-SY5Y cells demonstrated a slower regeneration of Prx from the hyperoxidized Prx SO3, compared with the cells transfected with scrambled siRNA (Supplementary Fig. S3D).

hTERT Overexpression results in improved mitochondrial function

Given that hTERT-expressing cells exhibited a significantly lower mitochondrial O2−we next asked if hTERT expression improved mitochondrial function. As such, the mitochondrial localization pattern of hTERT, which is largely known to reside in the nucleus was first studied. Indeed, hTERT was localized to the mitochondrial fraction with significantly higher expression in HT1 cells as compared with the Neo cells (Fig. 3A). Furthermore, we provide evidence that hTERT localizes to the inner as well as outer mitochondrial membrane fractions (Fig. 3B). These findings are in agreement with a recent report by Haendeler and colleagues showing a similar localization pattern of hTERT in HEK cells(17).

Figure 3.
Figure 3.

hTERT localizes to the mitochondria and hTERT expression improves mitochondrial COX activity. Neo and HT1 cells were fractionated and the mitochondrial hTERT levels were then assessed in Neo and HT1 cells (A). Mitochondrial fractions of HT1 cells were subjected to 50 μg/mL proteinase K to remove the outer mitochondrial membrane. The fractions were then subjected to Western blot analysis of hTERT, Bcl-2, VDAC, and prohibitin (B). C, 30 × 106 cells were fractionated and the COX activity was assessed in the mitochondrial fractions. The data shown are mean ± SD of at least 3 independent experiments. *, P < 0.05. D, Neo and HT1 cells were treated with 300 μmol/L of H2O2 or 100 μg/mL of C1 for 12 hours and fractionated to obtain the cytosolic and mitochondrial fractions that were subjected to Western blot analysis of Bax and Cyt C with VDAC and Cu/Zn SOD as mitochondrial and cytosolic loading controls WCL, whole cell lysate.

Given the reduction in mitochondrial O2− levels in hTERT-overexpressing cells and the localization of hTERT to the mitochondria coupled with the functional significance of COX in mitochondrial respiration, the role of hTERT in mitochondrial respiration with regard to COX was investigated next. As expected, HT1 cells displayed an increase in COX activity with a more rapid oxidation of ferrocytochrome C substrate (Fig. 3C).

hTERT Expression impairs ROS-induced cytosolic acidification and downstream mitochondrial outer membrane permeabilization

Having established the inhibitory effect of hTERT on intracellular ROS production, we next set out to investigate the mechanism underlying this protective antioxidant effect. We have previously demonstrated that H2O2 added exogenously or triggered endogenously upon drug exposure is a stimulus for cytosolic acidification, which creates a permissive intracellular milieu for apoptotic execution (9, 23). Therefore, the effect of hTERT expression on ROS-induced cytosolic acidification was assessed using the fluorescent pH probe BCECF. Whereas exposure of Neo cells to H2O2 or the ROS-inducing agent, C1, triggered significant cytosolic acidification (ΔpH = −0.80 units), the drop in pHi was much less pronounced in HT1 cells (ΔpH = −0.196 units; Fig. 4A and B).

Figure 4.
Figure 4.

hTERT expression reduces ROS-induced cytosolic acidification, Bax translocation, and release of proapoptotic proteins. Neo and HT1 cells were treated with 300 μmol/L of H2O2 (A) or 100 μg/mL of C1 (B) for 4 hours following which changes in pHi were assessed. C, Neo and HT1 cells were treated with 300 μmol/L of H2O2 or 100 μg/mL of C1 for 6 hours after which the cells the mitochondrial membrane potential was detected by flow cytometry using the probe DIOC6. All data shown are mean ± SD of at least 4 independent experiments. *, P < 0.05.

Cytosolic acidification has been shown to signal the translocation of Bax to the mitochondria followed by the release of proapoptotic factors (23). Following treatment of Neo and HT1 cells with H2O2 or C1, cytosolic- and mitochondrial-enriched fractions of Neo and HT1 cells were prepared and analyzed by Western blot for the release of mitochondrial proteins as well as Bax activation. Results show that overexpression of hTERT significantly blocked mitochondrial translocation of Bax as well as the release of proapoptotic proteins from the mitochondria (Fig. 3D). Because the release of proapoptotic proteins from the mitochondria suggested mitochondrial membrane permeabilization, the mitochondrial transmembrane potential (▵ψm) was assessed next. The drop in ▵ψm following treatment with H2O2 or C1 was greater in Neo cells as opposed to HT1 cells (Fig. 4C). This partial rescue from the drop in ▵ψm with hTERT overexpression provided further support that hTERT expression endows cells with a survival advantage against ROS-mediated insults by inhibiting mitochondrial permeabilization. Corroborating that, hTERT-overexpressing HT1 and HT2 cells were relatively refractory to the death-inducing activity of H2O2 or C1 compared with the control cells (Fig. 5A–D). To further characterize the loss of cell viability, cell-cycle analysis was performed in Neo and HT1 cells following exposure to 150 and 300 μmol/L of H2O2 for 18 hours to determine the sub-G1 pool. Indeed, the percentage of sub-G1 population was lower in HT1 (6.2% and 12.9%) cells as compared with Neo cells (16.27% and 21.84%) at 150 and 300 μmol/L of H2O2 (Fig. 5C). This was further supported by significantly improved ability of HT1 cells to form colonies following H2O2 or C1 treatment in contrast to Neo cells (Fig. 5D).

Figure 5.
Figure 5.

hTERT overexpression inhibits H2O2-mediated apoptosis. HeLa ATCC, Neo, HT1, and HT2 cells were treated with H2O2 (A) or C1 (B) in a dose-dependent manner and cell viability was assessed by crystal violet assay at 18 hours. C, Neo and HT1 cells were treated with 150 μmol/L and 300 μmol/L of H2O2 and stained with propidium iodide for DNA fragmentation analysis by flow cytometry (*, % of sub-G1 fraction) or exposed to (D) 300 μmol/L of H2O2 or 100 μg/mL of C1 following which colony-forming assay was performed as described earlier. All data shown are mean ± SD of at least 3 independent experiments. *, P < 0.05.

Discussion

The recent discovery of the novel antiapoptotic role of hTERT, independent of its telomerase activity has added a new perspective to reevaluate the antitelomerase and specifically anti-hTERT strategies (6, 7, 29, 30). The present study provides strong evidence that increased hTERT expression not only reduces the total basal and mitochondrial ROS levels, but also significantly antagonizes the increase in cellular ROS in response to both exogenous (H2O2) and endogenous (C1) ROS triggers. These data are in agreement with recent reports linking hTERT expression in MRC-5 cells to reduced cellular ROS levels (4, 17, 31).

Mechanisms of redox regulation by hTERT

Given the localization of hTERT to the mitochondria and its ability to antagonize mitochondrial ROS coupled with the functional significance of COX in mitochondrial respiration, here the role of hTERT in mitochondrial respiration with regard to COX was investigated. Interestingly, cells overexpressing hTERT have greater basal COX activity as compared with the mock-transfected Neo cells, which could suggest an improved mitochondrial efficiency; more efficient mitochondria could lead to reduced ROS production and could further account for the reduced O2− levels in hTERT overexpressing cells. Because hTERT has been shown to localize to the inner and outer mitochondria (Fig. 3B), a direct interaction with COX, which is an inner mitochondrial protein, might be feasible too. Alternatively, it is possible that hTERT might modulate COX activity in an indirect manner by binding to other COX interacting proteins such as Bcl-2.

The second mechanism by which hTERT seems to modulate intracellular redox status is by improving cellular antioxidant capacity. In this regard, GSH via its sulfhydryl group has been shown to play a critical role in cellular defense against toxic insults. Indeed, the GSH-GSSG oxidation–reduction cycle is regarded as pertinent in conserving the structural and functional viability of the cells (32), and agents that modulate the cellular GSH concentrations have been shown to alter the transcription of other detoxification enzymes such as glutathione S transferase (GST) and NAD(P)H:quinone oxidoreductase (NQO1) as well as affect cell proliferation and death mechanisms (33, 34). As such, the GSH/GSSG redox state can be viewed as the mirror of cellular redox status and alterations in the cellular GSH/GSSG ratio can have immense significance on cell survival (35, 36). Here, the demonstration of an enhanced GSH/GSSG ratio in a variety of cell lines strongly implicates a role for hTERT in modulating glutathione levels in the cells and conferring survival advantages. The elevated GPx activity and a lack of difference in GR activity in HT1 cells raised the question of how GSH levels might have been regulated to maintain an elevated GSH/GSSG ratio in HT1 cells.

GSH loss and oxidation have been associated with an increased expression of the rate-limiting enzyme of GSH synthesis, GCL, and several other detoxification systems. This prompted further analysis on the GCLC, which catalyzes the rate-limiting step of GSH synthesis (35, 37). The results demonstrated an early and sustained induction of GCLC in response to H2O2 in HT1 cells. This was in clear contrast to the transient and late activation of GCLC observed in mock-transfected cells. Because GCLC is critical for GSH synthesis, an early and sustained induction of GCLC following oxidative stress in hTERT expressing cells would certainly help to antagonize a rapid decline in cellular GSH levels, thus contributing to their higher GSH/GSSG ratios. The increased basal GPx and GR activity could contribute to the lowered basal cellular ROS levels in hTERT-overexpressing cells. Assessment of GPx activity or GR activity showed no further changes even when exposed to oxidative stress triggers (data not shown).

The assessment of Prx I, II, and III reflected no significant changes in their total levels after H2O2 treatment. Interestingly, the study of the hyperoxidized forms of Prx I, II, and III revealed a faster decline in the overall hyperoxidized forms of the Prx in HT1 cells. The hyperoxidation of Prx could lead to the inactivation of Prx and their regeneration would require repair proteins such as sulfiredoxins and/or sestrins. In this regard, several studies have shown the regeneration process by monitoring the total and hyperoxidized forms of Prxs and used these varied forms of Prx to assess cellular Prx activity (27, 38). Because the total amount of Prx remains unchanged over the same time course, our data suggest that the hyperoxidized forms are not targeted for degradation and most probably regenerated to their active forms. Cumulatively, the more rapid decrease in the inactive form, as has been observed in hTERT-overexpressing cells, might also mean a larger pool of active Prx available to target cellular peroxides in the cells.

Perturbation of cellular redox status alters intracellular milieu and cell fate

Studies have shown that an increase in intracellular H2O2 can lead to cytosolic acidification, which can subsequently engage the mitochondria-mediated pathway of death signaling via Bax conformational change and mitochondrial translocation culminating in effective death execution (9, 23). In thecurrent study, H2O2-induced cytosolic acidification was significantly reduced in hTERT-overexpressing cells, which impeded Bax translocation and mitochondrial outer membrane permeabilization. This could have led to the protection of hTERT-overexpressing cells from the apoptotic stimuli as demonstrated by the reduction in cell death and the improvement in colony-forming ability of hTERT-overexpressing cells. This is consistent with a recent report demonstrating the protective effect of hTERT on receptor-mediated death signaling [tumor necrosis factor (TNF) and TNF-related apoptosis-inducing ligand (TRAIL)] in NB4-LR1 cells, but not against cisplatin- or etoposide-induced cytoxicity (39). In contrast, several studies have shown in other cell lines that hTERT expression is able to protect cells from the cytotoxic effects of cisplatin (40–42). As such, it is worth mentioning that ROS generation via p53-mediated p38α mitogen-activatedprotein kinase is a mechanism by which cisplatin induces apoptosis (43). Thus, in a p53 defective cell line such as NB4-LR1, although cisplatin could induce cytotoxic effects via other mechanisms such as DNA damage, its ability to elicit a death response via the induction of ROS could be diminished. As such, our data underscore the significance of hTERT expression in diminishing ROS-induced cytotoxicity, and the protective effect of hTERT could be a function of the intrinsic differences in the cell's ability to engage ROS-mediated death effector pathways triggered by drug treatment.

Taken together, the discovery of hTERT's ability to modulate intracellular ROS by altering antioxidant status and mitochondrial function highlights for the first time a ROS-mediated mechanism by which hTERT expression endows cancer cells with a survival advantage by blunting apoptotic signals. By implication, cancer cells with intrinsically higher levels of hTERT might be refractory to chemotherapeutic regiments that function via ROS-mediated mechanisms for efficient death execution (Fig. 6). Hence, under such circumstances, alternative approaches such as targeting hTERT or the cellular antioxidant defenses might help to enhance chemosensitivity of cancer cells. Although it is early to hazard a conjecture on the mechanism that hTERT employs to regulate these antioxidant defenses, a possible candidate would be NF-κB, a key cellular transcription factor that engages in a unique cross-talk with ROS. NF-κB is modulated by cellular redox status and has been shown to transcribe several antioxidant factors including MnSOD and γ-glutamyl-cysteine synthetase—the rate-limiting enzyme for GSH synthesis (44–46)—and also enhance the transcription of AP1's downstream target genes such as sulfiredoxin (47, 48). Several studies have also reported a physical interaction between hTERT and NF-κB. Akiyama and colleagues have demonstrated that the active NF-κB p65 complex could chaperone hTERT from the nucleus into the cytosol (49, 50). In addition, it has also been shown by Sharma and colleagues (51) that ectopic hTERT expression can upregulate expression of NF-κB by over 6-fold. These authors also showed that this occurred via a mechanism that did not involve direct binding with the promoter regions of NF-κB (51). Our unpublished data also show higher levels of nuclear p65 in HT1 cells as compared with Neo cells. In addition, HT1 cells show a significantly higher level of MnSOD, which is a transcriptional target of NF-κB (data not shown). Given the keen association between the active NF-κB complex and hTERT at the nucleus, it is tempting to speculate that this interaction might be able to influence NF-κB's binding affinity to the DNA and the transcription of the NF-κB–driven antioxidant genes. Data presented here together with recent evidence underscore the broad biological consequences of hTERT expression aside from its essential function in telomere maintenance and clearly illustrate the need for new therapeutic strategies to target the novel functional consequence of increased hTERT expression.

Figure 6.
Figure 6.

Underlying mechanism of hTERT-mediated protection against oxidative stress-induced cell death. At the nucleus, hTERT's interaction with transcription factors (TF) such as NF-κB may offer an explanation for the enhanced antioxidant defenses (e.g., GSH, MnSOD) and lowered cellular ROS levels that are observed with hTERT expression. Exogenous ROS triggers have been shown to trigger nuclear export of hTERT. hTERT can be imported from the cytosol into the mitochondria by interacting with mitochondrial import proteins (IP) such as Tom 20 where it may enhance mitochondrial activity and reduce mitochondrial ROS production. Antagonism of an increase in cellular ROS levels can impede downstream proapoptotic events such as cytosolic acidification, dissipation of mitochondrial membrane potential (▵ψm), and the release of proapoptotic factors (PA) thus protecting from cell death.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work is supported by grants from the National Medical Research Council (NMRC) and Biomedical Research Council (BMRC) of Singapore to S. Pervaiz and a Tier 2 program grant from MOE, Singapore to S. Pervaiz and M.P. Hande.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Footnotes

  • Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

  • Received May 4, 2010.
  • Revision received October 15, 2010.
  • Accepted November 2, 2010.

References

  1. 1.↵
    1. Greider CW
    . Chromosome first aid. Cell 1991;67:645–7.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Greider CW
    . Telomere length regulation. Annu Rev Biochem 1996;65:337–65.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. De Boeck G,
    2. Forsyth RG,
    3. Praet M,
    4. Hogendoorn PC
    . Telomere-associated proteins: cross-talk between telomere maintenance and telomere-lengthening mechanisms. J Pathol 2009;217:327–44.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Ahmed S,
    2. Passos JF,
    3. Birket MJ,
    4. Beckmann T,
    5. Brings S,
    6. Peters H,
    7. et al.
    Telomerase does not counteract telomere shortening but protects mitochondrial function under oxidative stress. J Cell Sci 2008;121:1046–53.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Masutomi K,
    2. Possemato R,
    3. Wong JM,
    4. Currier JL,
    5. Tothova Z,
    6. Manola JB,
    7. et al.
    The telomerase reverse transcriptase regulates chromatin state and DNA damage responses. Proc Natl Acad Sci U S A 2005;102:8222–7.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Del Bufalo D,
    2. Rizzo A,
    3. Trisciuoglio D,
    4. Cardinali G,
    5. Torrisi MR,
    6. Zangemeister-Wittke U,
    7. et al.
    Involvement of hTERT in apoptosis induced by interference with Bcl-2 expression and function. Cell Death Differ 2005;12:1429–38.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Sung YH,
    2. Choi YS,
    3. Cheong C,
    4. Lee HW
    . The pleiotropy of telomerase against cell death. Mol Cells 2005;19:303–9.
    OpenUrlPubMed
  8. 8.↵
    1. Pervaiz S,
    2. Ramalingam JK,
    3. Hirpara JL,
    4. Clement MV
    . Superoxide anion inhibits drug-induced tumor cell death. FEBS Lett 1999;459:343–8.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Hirpara JL,
    2. Clement MV,
    3. Pervaiz S
    . Intracellular acidification triggered by mitochondrial-derived hydrogen peroxide is an effector mechanism for drug-induced apoptosis in tumor cells. J Biol Chem 2001;276:514–21.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Clement MV,
    2. Hirpara JL,
    3. Pervaiz S
    . Decrease in intracellular superoxide sensitizes Bcl-2-overexpressing tumor cells to receptor and drug-induced apoptosis independent of the mitochondria. Cell Death Differ 2003;10:1273–85.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Chou WC,
    2. Chen HY,
    3. Yu SL,
    4. Cheng L,
    5. Yang PC,
    6. Dang CV
    . Arsenic suppresses gene expression in promyelocytic leukemia cells partly through Sp1 oxidation. Blood 2005;106:304–10.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Santos JH,
    2. Meyer JN,
    3. Van Houten B
    . Mitochondrial localization of telomerase as a determinant for hydrogen peroxide-induced mitochondrial DNA damage and apoptosis. Hum Mol Genet 2006;15:1757–68.
    OpenUrlAbstract/FREE Full Text
  13. 13.↵
    1. Armstrong L,
    2. Saretzki G,
    3. Peters H,
    4. Wappler I,
    5. Evans J,
    6. Hole N,
    7. et al.
    Overexpression of telomerase confers growth advantage, stress resistance, and enhanced differentiation of ESCs toward the hematopoietic lineage. Stem Cells 2005;23:516–29.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Mondello C,
    2. Bottone MG,
    3. Noriki S,
    4. Soldani C,
    5. Pellicciari C,
    6. Scovassi AI
    . Oxidative stress response in telomerase-immortalized fibroblasts from a centenarian. Ann N Y Acad Sci 2006;1091:94–101.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Dairkee SH,
    2. Nicolau M,
    3. Sayeed A,
    4. Champion S,
    5. Ji Y,
    6. Moore DH,
    7. et al.
    Oxidative stress pathways highlighted in tumor cell immortalization: association with breast cancer outcome. Oncogene 2007;26:6269–79.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Massard C,
    2. Zermati Y,
    3. Pauleau AL,
    4. Larochette N,
    5. Metivier D,
    6. Sabatier L,
    7. et al.
    hTERT: a novel endogenous inhibitor of the mitochondrial cell death pathway. Oncogene 2006;25:4505–14.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Haendeler J,
    2. Drose S,
    3. Buchner N,
    4. Jakob S,
    5. Altschmied J,
    6. Goy C,
    7. et al.
    Mitochondrial telomerase reverse transcriptase binds to and protects mitochondrial DNA and function from damage. Arterioscler Thromb Vasc Biol 2009;29:929–35.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Wong-Riley MT
    . Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci 1989;12:94–101.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Pervaiz S,
    2. Clement MV
    . Hydrogen peroxide-induced apoptosis: oxidative or reductive stress? Methods Enzymol 2002;352:150–9.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Marzo I,
    2. Brenner C,
    3. Zamzami N,
    4. Susin SA,
    5. Beutner G,
    6. Brdiczka D,
    7. et al.
    The permeability transition pore complex: a target for apoptosis regulation by caspases and bcl-2-related proteins. J Exp Med 1998;187:1261–71.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    1. Koo MK,
    2. Oh CH,
    3. Holme AL,
    4. Pervaiz S
    . Simultaneous analysis of steady-state intracellular pH and cell morphology by automated laser scanning cytometry. Cytometry A 2007;71:87–93.
    OpenUrlPubMed
  22. 22.↵
    1. Hissin PJ,
    2. Hilf R
    . A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal Biochem 1976;74:214–26.
    OpenUrlCrossRefPubMed
  23. 23.↵
    1. Ahmad KA,
    2. Iskandar KB,
    3. Hirpara JL,
    4. Clement MV,
    5. Pervaiz S
    . Hydrogen peroxide-mediated cytosolic acidification is a signal for mitochondrial translocation of Bax during drug-induced apoptosis of tumor cells. Cancer Res 2004;64:7867–78.
    OpenUrlAbstract/FREE Full Text
  24. 24.↵
    1. Chen ZX,
    2. Pervaiz S
    . Bcl-2 induces pro-oxidant state by engaging mitochondrial respiration in tumor cells. Cell Death Differ 2007;14:1617–27.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Wong CH,
    2. Iskandar KB,
    3. Yadav SK,
    4. Hirpara JL,
    5. Loh T,
    6. Pervaiz S
    . Simultaneous induction of non-canonical autophagy and apoptosis in cancer cells by ROS-dependent ERK and JNK activation. PLoS One 2010;5:e9996.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Rhee SG,
    2. Chae HZ,
    3. Kim K
    . Peroxiredoxins: a historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic Biol Med 2005;38:1543–52.
    OpenUrlCrossRefPubMed
  27. 27.↵
    1. Woo HA,
    2. Kang SW,
    3. Kim HK,
    4. Yang KS,
    5. Chae HZ,
    6. Rhee SG
    . Reversible oxidation of the active site cysteine of peroxiredoxins to cysteine sulfinic acid. Immunoblot detection with antibodies specific for the hyperoxidized cysteine-containing sequence. J Biol Chem 2003;278:47361–4.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Chevallet M,
    2. Wagner E,
    3. Luche S,
    4. van Dorsselaer A,
    5. Leize-Wagner E,
    6. Rabilloud T
    . Regeneration of peroxiredoxins during recovery after oxidative stress: only some overoxidized peroxiredoxins can be reduced during recovery after oxidative stress. J Biol Chem 2003;278:37146–53.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Bollmann FM
    . The many faces of telomerase: emerging extratelomeric effects. Bioessays 2008;30:728–32.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Cong Y,
    2. Shay JW
    . Actions of human telomerase beyond telomeres. Cell Res 2008;18:725–32.
    OpenUrlCrossRefPubMed
  31. 31.↵
    1. Santos JH,
    2. Meyer JN,
    3. Skorvaga M,
    4. Annab LA,
    5. Van Houten B
    . Mitochondrial hTERT exacerbates free-radical-mediated mtDNA damage. Aging Cell 2004;3:399–411.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Schafer FQ,
    2. Buettner GR
    . Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic Biol Med 2001;30:1191–212.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Jones DP
    . Redox potential of GSH/GSSG couple: assay and biological significance. Methods Enzymol 2002;348:93–112.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Kirlin WG,
    2. Cai J,
    3. Thompson SA,
    4. Diaz D,
    5. Kavanagh TJ,
    6. Jones DP
    . Glutathione redox potential in response to differentiation and enzyme inducers. Free Radic Biol Med 1999;27:1208–18.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Meister A,
    2. Anderson ME
    . Glutathione. Annu Rev Biochem 1983;52:711–60.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. Storey KB
    . Oxidative stress: animal adaptations in nature. Braz J Med Biol Res 1996;29:1715–33.
    OpenUrlPubMed
  37. 37.↵
    1. Halliwell B,
    2. John M,
    3. Gutteridge C
    . Free radicals in biology and medicine. 3rd ed. New York: Oxford University Press;1999.
  38. 38.↵
    1. Jonsson TJ,
    2. Lowther WT
    . The peroxiredoxin repair proteins. Subcell Biochem 2007;44:115–41.
    OpenUrlCrossRefPubMed
  39. 39.↵
    1. Dudognon C,
    2. Pendino F,
    3. Hillion J,
    4. Saumet A,
    5. Lanotte M,
    6. Segal-Bendirdjian E
    . Death receptor signaling regulatory function for telomerase: hTERT abolishes TRAIL-induced apoptosis, independently of telomere maintenance. Oncogene 2004;23:7469–74.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Biroccio A,
    2. Gabellini C,
    3. Amodei S,
    4. Benassi B,
    5. Del Bufalo D,
    6. Elli R,
    7. et al.
    Telomere dysfunction increases cisplatin and ecteinascidin-743 sensitivityof melanoma cells. Mol Pharmacol 2003;63:632–8.
    OpenUrlAbstract/FREE Full Text
  41. 41.↵
    1. Kondo Y,
    2. Kondo S,
    3. Tanaka Y,
    4. Haqqi T,
    5. Barna BP,
    6. Cowell JK
    . Inhibition of telomerase increases the susceptibility of human malignant glioblastoma cells to cisplatin-induced apoptosis. Oncogene 1998;16:2243–8.
    OpenUrlCrossRefPubMed
  42. 42.↵
    1. Gorbunova V,
    2. Seluanov A,
    3. Pereira-Smith OM
    . Expression of human telomerase (hTERT) does not prevent stress-induced senescence innormal human fibroblasts but protects the cells from stress-induced apoptosis and necrosis. J Biol Chem 2002;277:38540–9.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Bragado P,
    2. Armesilla A,
    3. Silva A,
    4. Porras A
    . Apoptosis by cisplatin requires p53 mediated p38alpha MAPK activation through ROS generation. Apoptosis 2007;12:1733–42.
    OpenUrlCrossRefPubMed
  44. 44.↵
    1. Bubici C,
    2. Papa S,
    3. Dean K,
    4. Franzoso G
    . Mutual cross-talk between reactive oxygen species and nuclear factor-kappa B: molecular basis and biological significance. Oncogene 2006;25:6731–48.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Jang JH,
    2. Surh YJ
    . Bcl-2 attenuation of oxidative cell death is associated with up-regulation of gamma-glutamylcysteine ligase via constitutive NF-kappaB activation. J Biol Chem 2004;279:38779–86.
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    1. Allen RG,
    2. Tresini M
    . Oxidative stress and gene regulation. Free Radic Biol Med 2000;28:463–99.
    OpenUrlCrossRefPubMed
  47. 47.↵
    1. Soriano FX,
    2. Baxter P,
    3. Murray LM,
    4. Sporn MB,
    5. Gillingwater TH,
    6. Hardingham GE
    . Transcriptional regulation of the AP-1 and Nrf2 target gene sulfiredoxin. Mol Cells 2009;27:279–82.
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Wei Q,
    2. Jiang H,
    3. Matthews CP,
    4. Colburn NH
    . Sulfiredoxin is an AP-1 target gene that is required for transformation and shows elevated expression in human skin malignancies. Proc Natl Acad Sci U S A 2008;105:19738–43.
    OpenUrlAbstract/FREE Full Text
  49. 49.↵
    1. Kimura A,
    2. Ohmichi M,
    3. Kawagoe J,
    4. Kyo S,
    5. Mabuchi S,
    6. Takahashi T,
    7. et al.
    Induction of hTERT expression and phosphorylation by estrogen via Akt cascade in human ovarian cancer cell lines. Oncogene 2004;23:4505–15.
    OpenUrlCrossRefPubMed
  50. 50.↵
    1. Akiyama M,
    2. Hideshima T,
    3. Hayashi T,
    4. Tai YT,
    5. Mitsiades CS,
    6. Mitsiades N,
    7. et al.
    Nuclear factor-kappaB p65 mediates tumor necrosis factor alpha-induced nuclear translocation of telomerase reverse transcriptase protein. Cancer Res 2003;63:18–21.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Sharma GG,
    2. Gupta A,
    3. Wang H,
    4. Scherthan H,
    5. Dhar S,
    6. Gandhi V,
    7. et al.
    Sharma GG, Gupta A, Wang H, Scherthan H, Dhar S, Gandhi V, et al. (2003). hTERT associates with human telomeres and enhances genomic stability and DNA repair. Oncogene 2003;22:131–46.
    OpenUrlCrossRefPubMed
Back to top
Cancer Research: 71 (1)
January 2011
Volume 71, Issue 1

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
hTERT Overexpression Alleviates Intracellular ROS Production, Improves Mitochondrial Function, and Inhibits ROS-Mediated Apoptosis in Cancer Cells
Inthrani R. Indran, Manoor P. Hande and Shazib Pervaiz
Cancer Res January 1 2011 (71) (1) 266-276; DOI: 10.1158/0008-5472.CAN-10-1588
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement