Inhibition of Mitochondrial Cytochrome c Release and Suppression of Caspases by Gamma-Tocotrienol Prevent Apoptosis and Delay Aging in Stress-Induced Premature Senescence of Skin Fibroblasts

In this study, we determined the molecular mechanism of γ-tocotrienol (GTT) in preventing cellular aging by focusing on its anti-apoptotic effect in stress-induced premature senescence (SIPS) model of human diploid fibroblasts (HDFs). Results obtained showed that SIPS exhibited senescent-phenotypic characteristic, increased expression of senescence-associated β-galactosidase (SA β-gal) and promoted G0/G1 cell cycle arrest accompanied by shortening of telomere length with decreased telomerase activity. Both SIPS and senescent HDFs shared similar apoptotic changes such as increased Annexin V-FITC positive cells, increased cytochrome c release and increased activation of caspase-9 and caspase-3 (P < 0.05). GTT treatment resulted in a significant reduction of Annexin V-FITC positive cells, inhibited cytochrome c release and decreased activation of caspase-9 and caspase-3 (P < 0.05). Gene expression analysis showed that GTT treatment down regulated BAX mRNA, up-regulated BCL2A1 mRNA and decreased the ratio of Bax/Bcl-2 protein expression (P < 0.05) in SIPS. These findings suggested that GTT inhibits apoptosis by modulating the upstream apoptosis cascade, causing the inhibition of cytochrome c release from the mitochondria with concomitant suppression of caspase-9 and caspase-3 activation. In conclusion, GTT delays cellular senescence of human diploid fibroblasts through the inhibition of intrinsic mitochondria-mediated pathway which involved the regulation of pro- and anti-apoptotic genes and proteins.


Introduction
Replicative senescence is the final in vitro state reached by all primary cells in culture and is characterized by decreased total cell numbers and loss of proliferation capacity, in which they remain alive and metabolically active [1]. It has been widely used to identify aging-associated molecular changes in human cells and correlated with the in vivo aging process [2]. Recently, the term stress-induced premature senescence (SIPS) was introduced to represent an irreversible growth arrest of proliferative cells induced by exogenous oxidative agents.
The imbalance between cellular oxidants and antioxidants leads to oxidative damage to cellular macromolecules thus contributing to the decline in cellular functions and finally progressive organism aging [3]. Vitamin E in particular alpha-tocopherol protects biological membrane from oxidation by acting as a radical chain-breaking molecule. Other biological functions of vitamin E that are unrelated to its antioxidant properties include its roles in cellular signaling, gene expression, immune response, and apoptosis are now considered to be of importance [4].
Recently, an isomer of vitamin E, the tocotrienols have gained increasing interest due to their stronger antioxidant effects and a nonantioxidant activity profile that differs from tocopherols [5]. Tocotrienols were found not only to have antioxidant function, but also modulate signal transduction and gene expression. In cellular aging, tocotrienols have been observed to decrease biomarkers of senescent cells, reduced damaged DNA, decreased cells in G 0 /G 1 phase, and promoted cells to enter S phase [6]. However, the mechanism of action in delaying aging is not fully understood.
Other reports have also shown that tocotrienol administration reduced oxidative protein damage and consequently extended the mean life span of C. elegans [7]. γ-Tocotrienol at low concentration protected against H 2 O 2 -induced cell loss in fibroblast cell lines derived from different age groups. γ-Tocotrienol was also able to protect against H 2 O 2 -induced telomere shortening and telomerase activity loss [8].
Accumulating evidence suggested that dysregulation of apoptosis is associated with the aging process. Oxidative stress and DNA damage, both being important factors in the aging process, induce apoptosis. Previous study had shown that apoptosis was upregulated during aging in various cells [9]. Other report showed that apoptotic changes were increased in senescent endothelial cells [10].
In the present study, we determined the binding of Annexin V-FITC to phosphatidylserine (PS) on plasma membrane. FITC-conjugated Annexin V is widely used to detect apoptotic cells in a diverse range of cell types and in response to many different proapoptotic stimuli [11]. It has been reported that PS translocation to the outer leaflet of plasma membrane is the early and widespread event during apoptosis of cells from numerous lineages and is inhibited by overexpression of apoptosis repressor proteins such as Bcl-2 [11,12].
Caspases are a family of proteases that play major roles in apoptosis process. Caspase-3, caspase-6, and caspase-7, known as executioner caspases, are located down stream of the caspase-8 or caspase-9 activation cascade. Activation of executioner caspases is responsible for the proteolytic cleavage of a broad range of proteins [13].
Bax, a member of the Bcl-2 protein family, is involved in controlling apoptotic events. High level of Bcl-2 inhibits apoptosis by preventing cytochrome c release while high level of Bax induces apoptosis by binding to the mitochondrial membrane and increases membrane permeability allowing the release of cytochrome c [14]. The mitochondrial outer membrane permeability induced either by Bax-or Bakdependent proapoptotic drugs, or by H 2 O 2 results in the intracytosolic release of cytochrome c but the subsequent caspase activation is required to induce the translocation of apoptosis-inducing factor (AIF) into the cytosol. Therefore, mitochondrial response to several proapoptotic stimuli was suggested as a selective process leading to a hierarchical ordering of the effectors involved in cell death induction [15].
To date, there have been very few studies on the antiapoptotic effects of tocotrienol on normal fibroblasts in vitro. In the present study, we first established the SIPS model of primary HDFs by comparing the cellular senescence markers between SIPS and replicative senescent HDFs. We then determined the molecular apoptotic events involved in SIPS and the effects of γ-tocotrienol in modulating the apoptosis pathway and thus delaying cellular aging.

Characterization of HDFs Stress-Induced Premature
Senescence (SIPS). Control young HDFs displayed the normal spindle shape characteristic of fibroblast cells, while prolonged exposure to 20 μM H 2 O 2 resulted in HDFs exhibiting senescence features such as cells were enlarged and flattened with increased size of nucleus. Less spindle-shaped cells were present and accumulation of granular cytoplasmic inclusions was observed in SIPS and control senescent HDFs (Figure 1(a)). SA-β-galactosidase staining was also positive in SIPS and control senescent cells (Figure 1(b)). The percentage of SA-β-gal positive cells was markedly increased in SIPS and control senescent cells as compared to control young HDFs (P < 0.05) (Figure 1(c)).
Analysis on cell cycle profile showed that HDFs population in G 0 /G 1 phase was significantly increased while S phase and G 2 /M phase cells decreased in SIPS as compared to control young HDFs (Figures 2(a) and 2(b)) (P < 0.05). Similar increase and decrease in G 0 /G 1 phase and S phase cells, respectively, were observed in control senescent HDFs (P < 0.05).
Shortening of telomere length was observed in SIPS and control senescent HDFs as compared to young HDFs (P < 0.05) (Figures 3(a) and 3(b)). Similarly, decreased telomerase activity was observed both in SIPS and control senescent HDFs (Figures 4(a) and 4(b)). indicates early apoptotic cells (FITC + /PI − ) which is demonstrated by Annexin V binding and cytoplasmic membrane integrity. The percentage of cells at early apoptotic stage was significantly increased in SIPS and control senescent cells as compared to control young HDFs (P < 0.05) ( Figure 5(b)). GTT treatment in SIPS significantly decreased the percentage of early apoptotic cells as compared to SIPS alone (P < 0.05).

Activation of Caspase-3, Caspase-8, and Caspase-9.
Significant activation of caspase-8 and caspase-9 was observed in SIPS and control senescent cells as compared to control young and GTT-treated control HDFs (P < 0.05) (Figure 6(b)) while caspase-3 was significantly activated in SIPS, but not in control senescent cells. GTT treatment of SIPS inhibited caspase-3 and caspase-9 activation with no notable inhibition observed in caspase-8 activation (P < 0.05). In addition, GTT-treated control HDFs showed suppression of caspase-9 activation as compared to untreated young HDFs (P < 0.05).
Oxidative Medicine and Cellular Longevity . Data are expressed as means ± SD, n = 6. a denotes P < 0.05 compared to control young.

Cytochrome c Release.
Cytochrome c release was increased in SIPS and control senescent cells as compared to control young cells and GTT-treated control HDFs (P < 0.05) (Figure 7). SIPS cells released 47.7% of cytochrome c from the mitochondria into the cytosol as compared to control young HDFs (1%) and control senescent cells (79.5%). Cytochrome c release decreased significantly to 36.5% in SIPS treated with GTT as compared to SIPS alone (P < 0.05).

Analysis of Apoptotic-Associated Genes Expression.
Proapoptotic gene, BAX was markedly increased in SIPS and control senescent cells as compared to control young and GTT-treated control HDFs (P < 0.05) (Figure 8(a)). SIPS treated with GTT; however, showed a significant downregulation of BAX (P < 0.05). No significant difference was observed in the expression of another proapoptotic gene, BID in all treatment groups (Figure 8(b)).
Analysis of antiapoptotic genes showed that GTT treatment to control HDFs caused a significant upregulation of BCL2A1 mRNA (P < 0.05) while no change was observed in other treatment groups (Figure 9(a)). Analysis on the expression of another antiapoptotic gene, BCL2L1 showed no significant change observed in all treatment groups (Figure 9(b)).

Analysis of Bax and Bcl-2 Protein
Expression. The ratio of Bax/Bcl-2 protein expression in SIPS and control senescent cells was markedly increased as compared to untreated control young and GTT-treated control HDFs (P < 0.05) ( Figure 10). Treatment with GTT in SIPS considerably reversed the effect of H 2 O 2 in Bax/Bcl-2 protein expression whereby the ratio was significantly decreased as compared to SIPS alone (P < 0.05).  Figure 2: Flow cytometry analysis of cell cycle progression in young, SIPS, and control senescent HDFs (a). Quantitative analysis of cell cycle progression in control young HDFs, SIPS, and control senescent HDFs. Cell population in the G 0 /G 1 phase was significantly increased while S phase cells decreased in SIPS and control senescent HDFs (b). a denotes P < 0.05 compared to control young HDFs, b P<0.05 compared to SIPS. Comparison was done between HDFs in the same phase of cell cycle. Data is presented as mean ± SD, n = 6.

Discussion
The present study explored the molecular mechanism of GTT in inhibiting apoptosis and delaying cellular aging in H 2 O 2 -induced premature senescence (SIPS) of HDFs.
Our results confirmed that SIPS demonstrated senescence characteristics as evidenced by senescent-like morphological changes, a significant increase in senescence-associated marker SA-β-galactosidase activity, promoted G 0 /G 1 cell cycle arrest, shortening of telomere length, and decreased telomerase activity. Similar changes were observed for cultures at late stages (replicative senescent cells).
Both SIPS and senescent HDFs shared similar apoptotic changes as shown by increase in Annexin V-FITC positive cells, cytochrome c release, and activation of caspase-9 and caspase-3. These findings were supported by previous studies which reported that premature senescence of fibroblasts can be triggered by induction with acute sublethal doses of H 2 O 2 [16,17] or prolonged low dose of H 2 O 2 [1].
In this study, we exposed young HDFs to prolonged low dose of H 2 O 2 which mimics the oxidative stress in vivo to induce premature senescence state instead of acute induction with sublethal doses of H 2 O 2 . Exposure of prolonged low dose of H 2 O 2 to the embryonic human lung HDFs was found not only induced irreversible cell cycle arrest and senescentlike morphology, but also caused accumulation of damaged DNA accompanied by telomere shortening [1].
The initial step to develop SIPS model is to find doses that are subcytotoxic to the cells. We found that doses of H 2 O 2 30-80 μM slowed down cells growth rate which leads to cell death after 1 week culture while concentrations of H 2 O 2 above 100 μM triggered cell death within 24   changes and a significant increase in senescence-associated marker SA-β-galactosidase activity which was similar to replicative senescent cells. Previously, human foreskin fibroblasts which were exposed to H 2 O 2 for 7 days demonstrated positive staining for SA-β-gal [18] indicating cellular senescence. Therefore, based on these findings we decided to use the dose of 20 μM H 2 O 2 for 2 weeks to induce SIPS in HDFs.
One of the mechanisms mediating the development of senescent phenotype is telomere shortening. In this study, we found that both SIPS and senescent cells demonstrated shortening of telomere length with concomitant reduction in telomerase activity. We further determined cell cycle  . Band at 36 bp represents the internal control for the assay (a). Telomerase activity (Total Product Generated, TPG) of control young, SIPS, and control senescent HDFs. Reduction in telomerase activity was observed in SIPS and control senescent HDFs (b). a denotes P < 0.05 compared to control young HDFs. Data is presented as mean ± SD, n = 6. progression which showed that SIPS cells were mainly at G 0 /G 1 phase and decrease cell population in S phase indicating they might enter the irreversible growth arrest state followed by apoptosis. The reduction in S phase cells in SIPS suggested a slowing down of cellular proliferation which was similar to that of senescent cells. Previous in vitro study reported that H 2 O 2 -induced premature senescence and spontaneous senescent fibroblasts activated the apoptosis pathway as evidenced by activation of caspases in both groups of fibroblasts [19]. Human fibroblasts showed increased number of dead cells with increasing passage and increase in DNA fragmentation and mitochondrial dysfunction represented by loss of mitochondrial membrane potential, increased caspase-3, and increased cytochrome c positive cells [20]. Further identification of the molecular apoptotic changes in SIPS and senescent HDFs showed that Annexin V-FITC positive cells were increased both in SIPS and replicative senescent HDFs confirming early apoptotic changes occurred in both groups of cells.
Characterization of the apoptotic events in stress-induced premature senescence of HDFs showed that in SIPS, caspase-3 was activated not only by the activation of caspase-9 but also by caspase-8. These findings are consistent with a study done by Nagase et al. [21] and suggested that apoptosis pathway is activated during stress-induced cellular aging of HDFs. The activation of caspase-3, -8, and-9 indicated that both cell-surface-death receptor pathway and the intrinsic mitochondrial pathway are involved in inducing cell death during stress-induced premature senescence of HDFs [19].
We further investigated the involvement of mitochondrion-initiated pathway for both SIPS and senescent HDFs by determining the percentage of cytochrome c release from the mitochondria to the cytosol. We found that both SIPS and replicative senescent HDFs displayed significantly high percentage of cytochrome c release.
The induction of cytochrome c release can be achieved through various mechanisms, including the involvement of pro-and antiapoptotic proteins such as Bcl-2 family proteins. The 2 classes of proteins are localized to intracellular membranes, particularly mitochondria, and have been shown to interact with each other [22]. Activation of Bax protein resulted in rapid cell death, and this can be inhibited by death-inhibiting proteins; Bcl-2 and Bcl-xL.
Oxidative Medicine and Cellular Longevity  Figure 6: Flow cytometry analysis of active caspase-3, caspase-8, and caspase-9 with FITC/PI double staining in HDFs (a). Quantitative analysis of apoptotic cells population for active caspase-3, caspase-8, and caspase-9. (b). Data are expressed as means ± SD, n = 6. a denotes P < 0.05 compared to control young, b P < 0.05 compared to GTT, c P < 0.05 compared to SIPS, and d P < 0.05 compared to control senescent cells.
Further evaluation of the transcriptional and translational processes involving the expression of Bax and Bcl-2 proteins, the regulators of apoptotic pathway showed that Bax/Bcl-2 protein ratio was increased in SIPS and senescent HDFs indicating activation of apoptotic pathway and inhibition of antiapoptotic pathway with H 2 O 2 induction.
Our data on gene expression analysis showed that BAX expression was significantly increased in SIPS and senescent HDFs while another key proapoptotic gene, BID was not affected by H 2 O 2 induction. As for the antiapoptotic genes BCL2A1 and BCL2L1, our results did not show any significant changes in the expression of both genes in SIPS and cellular senescence of HDFs.
Based on these findings we suggested that both SIPS and replicative senescent HDFs undergo apoptosis. Therefore, further study was carried out to determine the specific stage that GTT intervenes in the apoptosis cascade and thus delaying aging.
Flow cytometry analysis showed that GTT treatment decreased the percentage of Annexin V-FITC positive cells  HDFs. Data are presented as means ± SD; n = 6. a denotes P < 0.05 compared to control young, b P < 0.05 compared to GTT, c P < 0.05 compared to SIPS, and d P < 0.05 compared to control senescent cells.
suggesting the prevention of PS translocation to the outer layer of cell membrane, thus inhibiting the early apoptotic changes of HDFs. Besides, GTT treatment to SIPS was found to inhibit the activation of caspase-9 and subsequently caspase-3. Similar effects however were not observed for caspase-8. These results indicated that GTT possessed antiapoptotic effects in preventing stress-induced premature senescence of HDFs which attenuated the key elements of intrinsic caspases in the mitochondrion-initiated pathway. The involvement of mitochondrion-initiated pathway was further investigated to determine the specific stage that GTT intervenes in the apoptosis cascade. Our results showed a reduction of cytochrome c release with GTT treatment indicating inhibition of stress-induced apoptosis by GTT was probably at the upstream stage in the cascade of events leading to apoptosis. In addition, GTT may also be involved in the regulation of pro-and/or antiapoptotic proteins which induced cytochrome c release from the mitochondria.
Gene expression analysis showed that BCL2A1 was markedly upregulated in control young HDFs treated with GTT. Furthermore, GTT was also found to downregulate BAX in SIPS cells. These findings demonstrated the involvement of GTT in regulating the apoptotic genes expression particularly BAX and BCL2A1 mRNA in normal diploid fibroblast cells.
Treatment of SIPS with GTT was also found to decrease Bax/Bcl-2 protein ratio, suggesting that exogenous antioxidants can modulate Bax and Bcl-2 expression in HDFs exposed to oxidant and thus prevents apoptosis. Several studies have shown that antioxidants may have antiapoptotic or proapoptotic properties, depending on the type of cells and the apoptosis-inducing signal involved, but the trend showed that antioxidants can prevent apoptosis in healthy cells and may promote apoptosis of transformed cells [22,23].
Previously, it had been demonstrated that high levels of Bcl-2 can prevent the release of cytochrome c and thus caspase activation in response to a number of apoptosisinducing stimuli such as UVB, staurosporine and etoposide [24,25]. Therefore, the decrease in Bax/Bcl-2 protein ratio with GGT treatment is suggestive of its action in preventing apoptosis in HDFs subjected to oxidative stress. The antiapoptotic effects of GTT may be attributable not only to its potent antioxidant properties but may also attributed to its nonantioxidant functions.
It is shown that GTT inhibited the apoptosis cascade leading to the prevention of cell death in SIPS. The ability of GTT to inhibit caspases and cytochrome c release upon prolong exposure of H 2 O 2 suggests the underlying mode of action of this compound in preventing oxidative stressinduced apoptosis.

Cell
Culture. This research has been approved by the Ethics Committee of the Universiti Kebangsaan Malaysia (Approval Project Code: FF-313-2009). Primary HDFs were derived from foreskins of three 9-to 12-year-old boys after circumcision. Written informed consents were obtained from parents of all subjects. The samples were aseptically collected and washed several times with 75% alcohol and phosphate buffered saline (PBS) containing 1% antibioticantimycotic solution (PAA, Austria). After removing the epidermis layer, the pure dermis was cut into small pieces and transferred into centrifuge tubes containing 0.03% collagenase type I solution (Worthington Biochemical Corporation, USA). The cells were rinsed with PBS before being cultured in Dulbecco's Modified Essential Medium (DMEM) (Flowlab, Australia) supplemented with 10% bovine fetal serum (FBS) (PAA, Austria) and 1% antibiotic-antimycotic solution at 37 • C in a humidified atmosphere at 5% CO 2 . Cultures were grown to 80-90% confluency in 75-mm culture flasks (Nunc, Denmark) with expansion degree 1 : 4. For subsequent experiments, cells were used at passage 4 (young cells, population doubling; PD < 12) and passage 30 (senescent cell, PD > 55).

Treatment Protocols.
Stress-induced premature senescence model was established by culturing the early passage cells (passage 6) with culture medium containing 20 μM H 2 O 2 for 2 weeks which was replaced every 3 days. Treated cells were preincubated with 1 μM γ-tocotrienol (Malaysian Palm Oil Board, Malaysia) at passage 4 followed by H 2 O 2 induction at passage 6 for 2 weeks. Untreated control cells were cultured without any treatment until similar population doublings were reached.
At the end of the treatment duration, the treated culture medium was replaced by fresh medium containing 10% serum for 3 days to ensure that the observed changes to the cells were not due to lack of mitogenic stimulations [26].

Morphology Analysis and Senescence-Associated β-
Galactosidase (SA-β-gal) Staining. HDFs positive for SA-βgal activity was determined as described by Dimri et al. [27]. SA-β-gal staining was performed with a senescent cell staining kit (Sigma, USA) according to the manufacturer's instructions. A total of 1 × 10 5 cells were seeded  Figure 9: Relative expression value of antiapoptotic genes BCL2A1 (a) and BCL2L1 (b). Data are expressed as means ± SD, n = 6. a denotes P < 0.05 compared to control young and b P < 0.05 compared to GTT.
in six-well plates and incubated with fixation buffer (2% formaldehyde/0.2% glutaraldehyde) for 6-7 min at room temperature. Cells were then rinsed three times with PBS and incubated with 5-bromo-4-chloro-3-indolyl β-D-galactopyranoside at 1 mg/mL in a buffer containing 40 mM citric acid/phosphate (pH 6.0), 5 mM K 3 FeCN 6 , 5 mM K 4 FeCN 6 , 150 mM NaCl, and 2 mM MgCl 2 for 4 h at 37 • C in the absence of CO 2 . Blue staining was visible after incubation, and the percentage of blue cells observed in 100 cells under a light microscope was calculated.

Cell Cycle Analysis by Flow Cytometry.
HDFs were harvested at desired time points after trypsinization and were rinsed 3 times with buffer solution with adjusted concentration 1 × 10 6 cells/mL and prepared using CycleTEST PLUS DNA Reagent Kit (Becton Dickinson, USA) according to the manufacturer's instruction. Cell cycle status was analyzed by flow cytometer using propidium iodide (PI) as a specific fluorescent dye probe. The PI fluorescence intensity of 10,000 cells was measured for each sample using a Becton-Dickinson FACS Calibur Flow Cytometer.  Data are expressed as means ± SD, n = 6. a denotes P < 0.05 compared to control young, b P < 0.05 compared to GTT, c P < 0.05 compared to SIPS and d P < 0.05 compared to control senescent cells. Detection of telomeric restriction fragments (TRFs) was visualized by alkaline phosphatase metabolizing CDP-Star, a highly sensitive chemiluminescence substrate. Smears on the membrane were detected by chemiluminescence signal using Fluorchem FC2 gel documentation system (Alpha Innotech, USA). Densitometry analysis was carried out using Image-Master TotalLab v1.11 software (Amersham, Germany) by comparing signals relative to DNA molecular weight marker. The mean TRF length was calculated for each sample by integrating the signal density above background over the entire TRF distribution as a function of TRF length, using the formula below:

Estimation of
where OD i and L i are, respectively, the signal intensity and TRF length at position i on the gel image, as described in the manufacturer's recommendations.  ). The PCR products were electrophoresed on 10% polyacrylamide gel using 0.5X TBE buffer. The gel was further analysed and quantitated using the ImageMaster TotalLab v1.11 software (Amersham, Germany). Telomerase activity (total product generated; TPG) was determined as the ratio of the intensity of telomerase ladders to the intensity of the 36 bp internal standard based on the formula below:

Estimation of Telomerase
where, X is TRAP product ladder (noninactivated sample); X 0 is TRAP product ladder (enzyme-inactivated sample); r is quantification control TSR 8 (0.5 μL); r 0 is quantification control TSR 8 (1.0 μL); C is assay internal control/S-IC; C R is negative control.

Quantitative Real-Time RT-PCR.
The expression level of BAX, BID, BCL2A1, and BCL2L1 was determined using iScript One-Step RT-PCR Kit with SYBR Green (Biorad, USA) and iQ5 Multicolor Real-Time PCR system (Biorad, USA) with the following reaction profile; cDNA synthesis at 50 • C for 20 min, iScript reverse transcriptase inactivation at 95 • C for 4 min, followed by 38 cycles of denaturation at 95 • C for 10 sec and primer annealing and extension at 61 • C for 30 sec. Melt curve analysis was carried out at 95 • C for 1 min, 55 • C for 1 min, and 60 • C for 10 sec (70 cycles, increase set point temperature after cycle 2 by 0.5 • C). The expression level of each targeted gene was normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene as an internal reference [29]. Agarose gel electrophoresis was performed for confirmation of the PCR products. Relative expression value of target genes was calculated based on the 2 −ΔΔCt method of relative quantification [30] by the following equation: Relative expression value = 2 Ct value of GAPDH−Ct value gene of interest. (3) 4.14. Western Blotting Analysis. Determination of protein expression was carried out by Western blotting using NuPAGE electrophoresis system (Invitrogen, CA, USA). Protein (25 μg) from each sample was loaded on a 4-12% NuPAGE Novex Bis-Tris gel using NuPAGE MES running buffer followed by dry-blotting of protein using iBlot Transfer Device for 7 min. WesternBreeze Chemiluminescent Kit was used to detect the primary antibodies. After transferring onto nitrocellulose membranes, the membranes were blocked for 30 min with blocking solution and probed with antibodies against Bax (1 : 200, overnight), Bcl-2 (1 : 200, overnight), and β-actin (1 : 1000, 1 h) as an internal control. All antibodies were purchased from Santa Cruz Biotechnology Inc., CA, USA. After 3 washing steps of antibody wash, membranes were incubated with secondary antibody solution (anti-mouse-IgG) for 1 h. Blots were developed using chemiluminescence substrate with enhancer for 7 min. The autoradiographies were scanned using gel documentation system Fluorchem FC2 (Alpha Innotech, USA) and band intensities were quantified by Image Master TotalLab software (Amersham Biosciences, UK).

Statistical Analysis.
Each experiment was carried out in triplicates. Data were reported as means ±SD. Comparison between groups was made by Student t-test (two-tailed). P < 0.05 was considered statistically significant

Conclusions
In summary, GTT was found to inhibit the translocation of PS into the outer membrane which is an early event in the apoptosis pathway. The antiapoptotic effect of GTT was through the mitochondria-dependent pathway as shown by GGT preventing the activation of caspase-3 and caspase-9 and reduced the release of cytochrome c from the mitochondria. Genes and proteins expression analysis suggested that GTT stimulated the signaling in the regulation of apoptoticrelated genes expression, prevented H 2 O 2 -induced apoptosis in HDFs, and delayed cellular aging.
In conclusion, GTT delays cellular senescence of human diploid fibroblasts through the inhibition of intrinsic mitochondria-mediated pathway which involved the regulation of pro-and antiapoptotic genes and proteins.