Resveratrol Prevents ROS-Induced Apoptosis in High Glucose-Treated Retinal Capillary Endothelial Cells via the Activation of AMPK/Sirt1/PGC-1α Pathway

Resveratrol (RSV) is used as a protective therapy against diabetic retinopathy. However, the mechanism(s) underlying this protective effect has not been fully elucidated. Bovine retinal capillary endothelial cells (BRECs), an in vitro model, were used to investigate the mechanism of RSV. Our results showed that high glucose induced significant cellular apoptosis in BRECs, which was accompanied by increased intracellular levels of reactive oxygen species (ROS) and cleaved caspase-3. The glucose-induced apoptosis and ROS elevation were both inhibited by RSV. High glucose was found to decrease the levels of phosphorylated AMP-activated protein kinase (p-AMPK), which was accompanied by increased levels of Sirt1 and PGC-1α. These changes were reversed by RSV. We also demonstrated that AMPK regulates the modulations of Sirt1 and PGC-1α using specific inhibitors of AMPK and Sirt1 and small interfering RNAs of PGC-1α. In summary, the current study demonstrates that RSV is effective against high glucose-induced cellular apoptosis and its action is exerted via the inhibition of ROS/AMPK/Sirt1/PGC-1α pathway.


Introduction
Diabetic retinopathy (DR) is one of the major causes of severe vision loss and blindness in the working-age population [1]. Vision loss in DR is caused by macular edema, which is characterized by vascular leakage from increased vascular permeability secondary to the breakdown of the bloodretinal barrier (BRB) [1,2]. The endothelium, which is a single layer of cells that covers the inner surface of all blood vessels, plays a critical role in the regulation of vascular function in the pathological processes of DR [3]. Apoptosis of endothelial cells of the retinal vasculature plays a vital role in the pathogenesis of DR [4,5]. Thus, therapeutic strategies focus on the identification of pharmacological targets that are involved in DR-induced endothelial apoptosis.
In in vitro systems, high glucose (HG), an independent risk factor for diabetes, has been shown to induce apoptosis in retinal capillary endothelial cells [5,6]. A hypothesis has been proposed that high blood glucose induces oxidative stress through the generation of excessive reactive oxygen species (ROS), which play a dominant role in the development of chronic complications caused by diabetes, including retinopathy [7,8]. Several studies suggested that HG can lead to overproduction of ROS in endothelial cells and subsequent apoptosis [9]. Peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α) is an important mediator of the metabolic effects of ROS, where PGC-1α activation results in the increase of mitochondrial energy metabolism and the cellular capacity to detoxify ROS, thereby reprogramming cell metabolism to maintain survival [10][11][12][13].
The AMP-activated protein kinase (AMPK) is a trimeric enzyme that contains a catalytic α-subunit and regulatory βand γ-subunits [14]. AMPK acts to ameliorate the pathogenesis of metabolic disorders, including diabetes, by controlling the expression and activation of various downstream molecules [15]. In addition to regulating energy metabolism, AMPK participates in the regulation of many other cellular processes, including inflammatory responses, autophagy, and apoptosis [16,17]. Our previous study showed that interaction between AMPK and sensor class III histone deacetylase sirtuin 1 (Sirt1), (interact to cause) caused endothelial damage and induced apoptosis in experimental diabetes [18]. Thus, AMPK/Sirt1 pathway plays an important role in microvascular damage associated with diabetes, including DR.
Resveratrol (3,5,4′-trihydroxystilbene (RSV)), a polyphenolic phytoalexin found in red wine and grape skin, has been reported to provide a wide range of health benefits such as reducing oxidative, inflammatory, and apoptotic signals [16,19,20]. RSV has also been found to be beneficial in the treatment of diabetes mellitus and its complications [21,22]. The antidiabetic effects of RSV are believed to occur via the activation of AMPK and Sirt1, and subsequent reduction in the circulatory levels of proinflammatory cytokines and proapoptotic cells, along with enhancement of antioxidant defenses [23,24]. A recent study, however, indicated that RSV improves oxidative stress and protects against diabetic nephropathy in an AMPK/Sirt1-independent manner [25]. To investigate the protective effect of RSV on DR and its dependency on AMPK/Sirt1 pathway, apoptosis was induced in bovine retinal capillary endothelial cells (BRECs) by exposure to HG and demonstrated a critical role of ROS in HG-mediated apoptosis and the regulatory role of AMPK/ Sirt1/PGC-1α pathway in the antiapoptotic activity of RSV.

Cell Viability Assay.
Cell viability was evaluated with the 3-[4.5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay. BRECs (1 × 10 5 cells) were plated in 96-well plates and cultured overnight. The cells were treated with 1, 5, 10, and 20 μM RSV for 48 h. The medium was then replaced with fresh medium containing 0.5 mg/mL MTT for 4 h. After incubation, the medium was carefully removed from the plate, and dimethylsulfoxide (DMSO) was added to solubilize the formazan produced from MTT by the viable cells. The absorbance value was measured in a microplate reader (Bio-Rad Laboratories, Shanghai, China) at 490 nm. Values were expressed as a percentage relative to normal glucose (NG).

Measurement of Intracellular ROS Production.
The level of intracellular ROS was detected by the peroxide-sensitive fluorescent probe dichlorodihydrofluorescein diacetate (DCFH-DA). DCFH-DA is converted by intracellular esterases to DCFH, which is oxidized into fluorescent dichlorofluorescein (DCF) in the presence of a proper oxidant. Samples were incubated with 2 μM DCFH-DA in serumfree DMEM in the dark at 37°C with 5% CO 2 for 30 min. After incubation, the cells were washed twice with PBS and used immediately for flow cytometry with excitation wavelength at 488 nm and emission wavelength at 525 nm. Mean fluorescence was calculated by using the program Cell Quest (Bio-Rad, USA). Each experiment was repeated three times.

Detection of Apoptosis by Flow Cytometry.
After 24 h, cells were washed in Annexin V binding buffer and were incubated in 500 μL Annexin V staining solution (1 : 50 dilution of Annexin V-phycoerythrin in Annexin V binding buffer) for 10 min in the dark at 37°C, 5% CO 2 . The cells were then washed with ice-cold PBS twice and resuspended in the binding buffer, then 10 μL of propidium iodide (PI) was added for 10 min at 4°C. Flow cytometry was used to collect 5000 cells, which were sorted based on whether they were stained with PI (apoptotic or dying cells) or were unstained with either fluorochrome (live cells). All data were collected, stored, and analyzed by Multigraph software (Coulter, Miami, FL, USA).

RNA Extraction and Real-Time RT-PCR.
Total RNA was extracted from BRECs using TRIzol reagent (Invitrogen Life Technologies, Gaithersburg, MD) and stored at −80°C. The DyNAmo Flash SYBR Green qPCR kit (Finnzymes Oy, Espoo, Finland) was used according to the manufacturer's instructions. The reaction conditions were 40 cycles of two-stage PCR consisting of denaturation at 95°C for 30 sec and annealing at 60°C for 30 sec after an initial denaturation step at 95°C for 5 sec. The primer sets are listed in Table 1).
The specificity of the amplification product was determined by performing a melting curve analysis. Standard curves were generated for each gene by using serial dilutions of known quantities of the corresponding cDNA gene template. Relative levels of target gene mRNA expression were calculated using the 2 −ΔΔCT method. Amplification of the target gene cDNA was normalized to β-actin expression.

Protein Preparation and Western Blot.
BRECs were plated at a density of 3 × 10 6 cells/well on 6-well plates and lysed using modified RIPA buffer on ice for 30 min, and the lysates were centrifuged at 12,000 ×g for 15 min at 4°C. Proteins in the supernatant were quantified using the bicinchoninic acid (BCA) protein assay. Equal protein amounts from each sample were run in 10% sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE) in a Bio-Rad miniature slab gel apparatus and transferred onto a nitrocellulose membrane. After blocking with 5% nonfat dried milk solution for 1 h, the membranes were incubated overnight at 4°C with primary antibodies for cleaved caspase 3 (1 : 500), AMPK (1 : 500), p-AMPK Thr172 (1 : 500), Sirt1 (1 : 500), and PGC-1α (1 : 500). Goat anti-rabbit IgG (1 : 1000) was used as the secondary antibody. To detect GAPDH expression, we used a monoclonal antibody (1 : 1000; ProteinTECH Group, Chicago, IL, USA) as an internal control to confirm equivalent total protein loading. All measures are expressed relative to the signal intensities measured in the control lanes.
2.9. Statistical Analysis. Data were presented in mean ± SEM. One-way analysis of variance (ANOVA) was performed followed by Tukey's post hoc test. p value < 0.05 was considered statistically significant. All computations were performed with the SPSS 16.0 (Chicago, IL) software.

Results
3.1. Cell Culture and Identification. BRECs were isolated from tissues obtained from a local slaughterhouse and cultured following protocols described previously [26]. After 3-4 passages, BRECs appeared flat and assumed a cobblestone-shaped morphology (Figure 1(a)). These cells were stained positive for Von Willebrand, a molecular marker for retinal endothelial cells, with a finely granular cytoplasmic staining pattern (Figure 1(b)), and were negative for smooth muscle actin (Figure 1(c)). This immunocytochemical labeling confirms that the cultured cells are retinal capillary endothelial cells.

Effect of RSV on the Viability of Retinal Endothelial Cells.
To investigate the cytotoxic effect of RSV on retinal endothelial cells, we performed an MTT assay after 24 h exposure to RSV at concentrations ranging from 1 to 20 μM. Up to the highest concentration (20 μM), RSV did not affect the viability of the BRECs (Figure 2). These results indicate that 20 μM RSV is safe for BRECs, and thus, we used RSV at 20 μM dose for subsequent studies.

RSV Suppresses HG-Induced BREC Apoptosis through
Reduction of ROS Production. BRECs were cultured with NG, NG + mannitol or HG alone, or in combination with RSV, H 2 O 2 , or NAC (Figures 3 and 4). When cultured with

Effect of RSV on the AMPK/Sirt1/PGC-1α Pathway in
High Glucose-Treated BRECs. Previous studies have shown that the AMPK/Sirt1/PGC-1α pathway plays an important role in the induction of ROS-induced apoptosis in diabetes [18,27,28]. mRNA by the 2 −ΔΔCT method and protein levels by Western blot were therefore quantified. As shown in Figure 5, mRNA levels of Sirt1 and PGC-1α were substantially reduced in BRECs treated with HG as compared to NG (p < 0 05 for each). By Western blot analysis, protein levels p-AMPK, Sirt1, and PGC-1α were substantially reduced in BRECs treated with HG as compared to NG (all p < 0 05). In comparison, nonphosphorylated AMPK levels were unchanged (p > 0 05). When HG-cultured BRECs were treated with RSV, mRNA levels of Sirt1 and PGC-1α approached the NG level and the protein levels of p-AMPK, Sirt1, and PGC-1α protein were also increased to levels near BREC cells cultured in NG.
To explore the relationship between the antiapoptotic effect of RSV and the AMPK/Sirt1/PGC-1α pathway, additional experiments using the AMPK activator AICAR and the Sirt1 activator DCHC were performed. Similar to RSV, addition of AICAR or DCHC to BRECs grown in HG resulted in increased levels of p-AMPK, Sirt1, and PGC-1α protein. In comparison, addition of compound C (an antagonist of AMPK) and EX527 (an inhibitor of Sirt1) reversed the effect of RSV.

RSV Mitigates ROS-Induced BREC Apoptosis by
Modulating AMPK/Sirt1/PGC-1α Pathway. To further explore RSV inhibition of ROS-induced BREC apoptosis through the AMPK/Sirt1/PGC-1α pathway, siRNA was used to reduce PGC-1α levels and specific inhibitors of AMPK and Sirt1. As shown in Figures 6 and 7, culture of cells in HG as compared to NG increased the levels of Annexin V and the number of PI-positive apoptotic cells. HG also increased levels of ROS and cleaved caspase-3 compared with NG. These increases were reversed when 20 μM RSV was administered to HG-cultured cells. Inclusion of the AMPK agonist AICAR or the Sirt1 agonist DCHC mirrored the beneficial effect of 20 μM RSV on HG-induced ROS production, cell apoptosis, and cleaved caspase-3 protein expression. In comparison, the beneficial effects of RSV were abolished by inhibitors of AMPK or Sirt1 or by treatment with PGC-1α siRNA. These results are consistent with a model in which RSV mitigates ROS-induced BREC apoptosis by modulation of the AMPK/Sirt1/PGC-1α pathway.

Discussion
In this study, we cultured BRECs in HG medium to mirror some features of DR related to endothelial cell damage. HG increased ROS levels and induced BREC apoptosis. These changes were reversed by RSV, a derivative of grape skin that can reduce oxidative damage, inflammation, and cellular apoptosis [29]. In BRECs, RSV reduced ROS levels, cleaved caspase-3, and apoptosis. The current study demonstrated that the beneficial effects of RSV could be mirrored by activators of the AMPK/Sirt1/PGC-1α pathway and could be blocked by agents that interfere with AMPK/Sirt1/ PGC-1α signaling.
Microvascular changes, including thickening of capillary basement membranes, apoptosis of microvascular cells, loss     of pericytes, and acellular capillary formation, are major features of DR [2]. Vascular lesions resulting from BRB breakdown are primary causes of vision loss in DR [1,2]. Both human and animal studies indicated that microvascular apoptosis plays a crucial role in the development of early lesions [3,30]. Although retinal microvascular cell loss plays a critical role in the pathogenesis of DR [3,5], the mechanisms by which HG induces endothelial cell apoptosis in DR are not fully understood. In the present study, BREC culture system was used to replicate the impact of HG on endothelial cells and to verify prior findings [4,31] that HG induces ROS production and apoptosis. Brownlee proposed that mitochondrial production of ROS in response to hyperglycemia may be a key initiating step in the pathogenesis of diabetes, including DR [8]. Other studies have established a close relation between intracellular ROS production and retinal endothelial cell apoptosis [9,32,33]. In our BREC system, we confirmed that HG stimulates ROS production leading to cleaved caspase-3 expression and apoptosis. These changes induced by HG were blocked by the antioxidant NAC, but were mirrored in BRECs cultured in NG along with H 2 O 2 . These supported the use of antioxidants as a treatment for DR to target intracellular ROS production. RSV is a potent antioxidant [34,35] that evidence was found to support the effects of RSV in diabetes and other ocular diseases [23,29,36,37]. Tennen et al. proposed that RSV acts by activation of AMPK and Sirt1 [16,38,39], which is consistent with our results. AMPK is a serine/threonine protein kinase, which may sense regulated metabolic homeostasis by energy deficiency through an increased AMP/ATP ratio [15] through controlling the transcriptional regulation of certain genes such as PGC-1α and NF-κB [38,40,41]. Cantó and Auwerx reported that AMPK regulates transcription through direct events (i.e., phosphorylation of transcriptional regulators) and also indirectly by increasing NAD+ and inducing Sirt1 activity [42]. Our results are in line with the model in which RSV acts through the AMPK/Sirt1/PGC-1α pathway to modulate the negative effects of HG on BRECs and support future efforts towards the use of RSV to ameliorate the cellular damage induced by diabetes.

Conclusion
Our data demonstrate that RSV acts to reduce intracellular ROS through the activation of AMPK/Sirt1/PGC-1α pathway and thereby suppresses apoptosis in HG-treated retinal capillary endothelial cells. This study provides insight into the application of RSV as an effective treatment for the early stages of progression in DR.

Conflicts of Interest
The authors declare no competing interests.