Evaluation of Sirtuin Role in Neuroprotection of Retinal Ganglion Cells in Hypoxia

Abstract

Purpose.: Hypoxia-induced apoptosis is responsible for reduced retinal ganglion cell (RGC) viability in a variety of chronic ocular disorders. Sirtuin 1 (SIRT1) plays an important role in preserving cell viability during hypoxia. We investigated the role of SIRT1 in sustaining RGC viability in an in vitro model of hypoxia.

Methods.: Staurosphorine-differentiated RGCs (RGC-5) received varying hypoxic concentrations (100–500 μM) of cobalt chloride (CoCl₂) for 24 hours. Hypoxia-induced cell viability was assessed by WST-1 assay. The role of SIRT1 in promoting viability was determined indirectly via sirtinol (SIRT1 inhibitor). Hypoxia-induced apoptosis was evaluated by measuring stress-activated protein kinase/c-jun N-terminal kinase (SAPK/JNK) and caspase 3 activity. Vascular endothelial growth factor (VEGF) was measured to ascertain the influence of SIRT1.

Results.: CoCl₂ concentrations greater than 100 μM resulted in significantly reduced RGC viability (P = 0.01). CoCl₂ treatment increased SIRT1 levels significantly (P < 0.01): 100 (6.5-fold), 200 (6-fold), 300 (3.5-fold), and 400 μM (4.5-fold). Phosphorylated SAPK/JNK increased 36-fold (200 μM CoCl₂ concentration), then plateaued at the 300- (25-fold) and 400-μM (27.8-fold) CoCl₂ concentrations (P < 0.01). CoCl₂ and sirtinol treatment increased Caspase 3 activity (P < 0.05). VEGF release was significantly higher than control at the 100-μM CoCl₂ concentrations (P < 0.01). Sirtinol reduced RGC viability, SIRT1 levels, and VEGF release (P < 0.01) while having greater effect on SAPK/JNK phosphorylation.

Conclusions.: SIRT1 significantly influences RGC viability. Sirtinol's effect reflects the interaction SIRT1 has with apoptotic signaling proteins. This investigation demonstrated SIRT1 importance in forestalling the effects of hypoxia-induced apoptosis.

Introduction

Chronic neurodegenerative ocular disorders, such as glaucoma and optic neuropathy, are the leading causes of progressive retinal ganglion cell (RGC) loss for individuals older than 40. RGCs, which are located at the innermost layer of the retina, play a crucial role in transmitting light signals. This transmission occurs from the outermost neural retina, where light is captured and converted to an electrical impulse, to the visual-processing centers of the brain. RGC death occurs primarily by an apoptotic mechanism instigated by a number of stimuli such as an elevation of IOP, ischemia, oxidative stress, and deprivation of neurotropic factors. ^1,2  

In vivo, compromised blood flow leads to a state of hypoxia through the reduction of oxygen supply. ³ RGCs are particularly sensitive to transient, mild, and long-term hypoxia. Consequently, as exposure to hypoxia is prolonged, the rate of RGC apoptosis increases. ⁴ However, the mediators of hypoxia-induced cell death are complex and not fully understood. 

Hypoxia alters the cellular redox state and activates a stress responsive class III histone deacetylase known as sirtuin1 (SIRT1). ^5–7 SIRT1's deacetylation of histone proteins modulates nonhistone substrates that play an essential role in neuronal cell survival during oxidative stress and apoptotic states. ⁸ SIRT1 induces a number of transcription factors/proteins such as p53, ⁹ FOXO (Forkhead box), PGC-1alpha (peroxisome proliferator-activated receptor gamma coactivator 1-alpha), NF-κB (nuclear factor-κB), ^10,11 VEGF, and EPO (erythropoietin). ¹² In the eye, SIRT1 is expressed in various retinal cells, such as the retinal pigment epithelium (RPE), choroid, and neural retina. ⁷  

In several disease models, SIRT1 activation led to the cessation of dysfunction and damage to RGCs. In an experimental model of optic neuritis, Shindler et al. ¹³ showed that intravitreal injection of SIRT1 activators established a beneficial effect to RGC survival by attenuating RGC death. Nevertheless, the role of SIRT1 in sustaining RGC survival during hypoxic states in retinal neurodegenerative disorders has not been well defined. 

In this article, we investigated what role SIRT1 has in supporting the viability of RGCs in an in vitro model of hypoxia. 

Materials and Methods

Cell Culture

In this study, the transformed retinal ganglion cell line (RGC-5) was used due to inherent cell viability limitations associated with primary RGC cultures. RGC-5 was obtained as a kind gift from Neeraj Agarwal (University of North Texas Health Science Center, Fort Worth, TX) and maintained under standard culture conditions in Dulbecco's Modified Eagle's Medium (DMEM; Invitrogen Corp., Carlsbad, CA) with 10% fetal bovine serum. Media was supplemented with 100 U/mL of penicillin and 100 μg/mL of streptomycin (Invitrogen Corp.). The cells were cultured in 75 cm² filter-capped flasks in an incubator consisting of 95% air and 5% CO₂ at 37°C. 

RGC Differentiation and Induction of Hypoxia

In order to induce RGC-5 to assume morphologic features of mature RGCs, staurosporine (Sigma-Aldrich, St. Louis, MO) was used. ^14,15 RGC-5 cells were differentiated via kinase-dependent pathway, as described by Frassetto et al. ¹⁴ In brief, 2 × 10³ cells/well of RGCs were plated in 96-well plates and allowed to reach a semiconfluent state within 48 hours. Cells were then differentiated by applying 100 nM of staurosporine to each well for 24 hours, as previously described. 

For the experimental purposes of this study, cobalt chloride (CoCl₂; Sigma-Aldrich) was used to induce hypoxia-related cell injury. CoCl₂ establishes the cellular hypoxic response via stabilization of hypoxia-inducible factors through inhibition of the prolyl hydroxylase domain. ^16,17 Through this mechanism, the hypoxic response can be replicated in the in vitro environment. CoCl₂ concentrations of 100, 200, 300, 400, and 500 μM were added to the media of staurosporine-differentiated RGCs for a period of 24 hours. ¹⁸  

The effect of CoCl₂ on cell viability following hypoxic insult was evaluated using WST-1 assay based on previous literature. ^15,18 WST-1 (4-[3-(4Iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1.3-benzene disulfonate), a colorimetric assay, is based on the cleavage of tetrazolium salts to formazan by mitochondrial dehydrogenases in viable cells (Roche, Mannheim, Germany). The plates were read at 440 nm, with a reference wavelength at 690 nm, using a multidetection microplate reader (BioTek Synergy HT, Winooski, VT). Results were normalized against controls and presented as percentage of cell viability, compared with controls. 

CoCl₂ Dose-Response Analysis on RGC Viability

RGCs were cultured and differentiated as previously mentioned. Cells were then subjected to the varying concentrations of CoCl₂ (100–500 μM). Cell viability was determined via the WST-1 colorimetric assay. Lethal dose 50 (LD₅₀) was also determined in this experiment. CoCl₂ concentrations that were determined to be greater than the LD₅₀ were not used for subsequent experiments. 

SIRT1-Induced Cell Viability

RGC cell viability was assessed by evaluating the effects that sirtinol (Tocris biosciences, Ellisville, MO), a permeable cellular inhibitor of SIRT1, had on tissue sustainability. Sirtinol was added to RGC media, as per manufacturer's instructions, for 24 hours. Different concentrations of sirtinol (100, 150, and 200 μM) were evaluated so as to optimize the maximal inhibitory potential of SIRT1 activity. The 200-μM concentration was determined to be the most effective dose and was implemented for subsequent experiments (data not shown). RGCs were cultured and differentiated as discussed earlier and cells were treated with CoCl₂ (100–400 μM) in combination with sirtinol. 

Immunoblot Analysis: SIRT1 and SAPK/JNK Levels

Differentiated cells were treated with the aforementioned concentrations of CoCl₂ (100–400 μM) and sirtinol overnight. To determine SIRT1 levels and the role of stress-activated protein kinase/c-jun N-terminal kinase (SAPK/JNK) after hypoxic insult, we measured the protein levels of SIRT1 and the phosphorylation state of the SAPK/JNK kinases. The JNK inhibitor (SP600125) was purchased from Biomol International (Plymouth Meeting, PA). After treatment, cells were washed twice with ice-cold Hanks' Buffered Saline Solution and lysed with cell lysis buffer for 15 minutes to extract the proteins as per manufacturer's instructions (Ambion Applied Biosystem, Invitrogen Corp.). Protein concentrations in the resultant supernatant cells were determined using a Bradford reagent (Biorad, Hercules, CA). 

Twenty micrograms of protein samples were run in 6% and 8% SDS-PAGE gel and transferred to nitrocellulose membranes, which were blocked using PBS-Tween 20 (PBST) with 5% BSA at 4°C. The membranes were incubated with anti-SIRT1 (1:3000; Upstate Biotechnology, Billerica, MA), Total SAPK/JNK, anti-phospho-SAPK/JNK (Cell Signaling Technology, Danvers, MA) at 1:1000 in PBS at 4°C overnight. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Upstate Biotechnology) served as a loading control. After washing with PBST, the blots were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit IgG antibodies at a 1:1000 dilution for 2 hours at room temperature. Blots were washed 3 times in PBST and the proteins were detected with an enhanced chemiluminescence method. Bands were scanned and densitometry was performed to quantify the intensity of signal using ImageJ analysis software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/:index.html). The protein levels were expressed either in optical density units (ODU) or as percentages.  

Intracellular Analysis of Caspase 3 Activity

We investigated the role SIRT-1 played in neuroprotection by measuring the activated caspase 3, a protease instrumental in the induction of apoptosis. Differentiated SIRT-1 inhibited RGC-5′s were subjected to CoCl₂ conditions according to methods described earlier. Activation of caspase 3 was measured as per manufacturer's instructions (Thermo Scientific, Logan, UT). In brief, following treatment, cells were fixed using 4% paraformaldehyde and permeabilized for 15 minutes using ×1 surfact-amps X-100. Cell permeabilization was quenched using 30% H₂O₂ followed by the addition of blocking buffer for 30 minutes. Cells were then incubated with 1:500 dilution of anti–caspase 3 and α-tubulin (control), at 4°C overnight. After washing, they were incubated with anti-rabbit HRP conjugate for 30 minutes at room temperature followed by the addition of 3,3',5,5'-Tetramethylbenzidine substrate. Cleaved caspase 3 protein levels were assessed by measuring the absorbance at 450 nm using microplate reader. The results were compared against α-tubulin control after detection with background absorbance and interpreted in percentage. Experiments were duplicated to ensure congruent results. 

VEGF ELISA Assay

The goal for this experiment was to ascertain the role of VEGF in facilitating SIRT1 functionality as a potential neuroprotectant during RGC hypoxia-induced apoptosis. VEGF levels were measured using ELISA after plating 1 × 10⁴ cells on a 6-well culture plate and maintaining standard culture conditions. While under hypoxic mimetic conditions (5% CO₂ in 37°C) and following sirtinol treatment, cell culture media were collected. The concentration of VEGF in the media was measured with an ELISA kit (Invitrogen Corp.), based on the manufacturer's instructions. 

Cell Morphology Analysis

RGC morphological analysis was conducted within 96-well microtiter plates to observe structural changes associated with varying concentrations of CoCl₂ treatment as well as SIRT1 inhibition via sirtinol. Morphological changes to treated RGCs were assessed under bright-field phase-contrast microscopy (Olympus 1X51, Tokyo, Japan). 

Statistics

Mean and SD values were analyzed using Graphpad Instat software (GraphPad Instat3, La Jolla, CA). A two-tailed, unpaired, or paired Student's t-test was used to compare mean statistics. Statistical significance was accepted for P values less than 0.05. 

Results

CoCl₂ Dose-Response Analysis on RGC Viability

CoCl₂ treatment decreased the RGC viability (Fig. 1). This reduction at the 100-μM concentration (85.4% ± 5.7%) was not significantly different than control cells. However, the 200- (81.0% ± 3.4%), 300- (62.0% ± 4.6%), and 400- (50.4% ± 2.1%) μM CoCl₂ concentrations conveyed significantly reduced cell viability in comparison with control cells (P < 0.01). The 500-μM concentration, when contrasted with control, resulted in a reduction of cell viability (37.2% ± 6.7%), which was considered beyond the LD₅₀. Since LD₅₀ was surpassed at the 500-μM concentration, we performed subsequent hypoxia induction experiments within the 100- to 400-μM CoCl₂ concentration range. 

Figure 1. 

 

Hypoxia-induced growth arrest by CoCl₂ at various concentrations (100–500 μM for 24 hours) in differentiated RGCs (staurosporine-treated for 24 hours); the concentration of CoCl₂ used for inducing hypoxia is represented on the x-axis and the percentage of cell viability normalized against control is represented on the y-axis (*P < 0.05).

Effect of SIRT1 Inhibition on Cell Survival

In the presence of CoCl₂, sirtinol significantly decreased the survival of differentiated RGCs to 60.9% ± 4.9% and 55.6% ± 2.9%, at the 100- and 200-μM concentrations, respectively. This decrease in cell survival was exacerbated to 41.7% ± 4.2% (300 μM) and 25.8% ± 3.3% (400 μM) in response to escalating concentrations of CoCl₂ when contrast to the control condition (P < 0.01; Fig. 2). 

Figure 2. 

 

SIRT1-induced cell survival (after sirtinol treatment of 24 hours) on differentiated RGCs after induction of hypoxia using CoCl₂ (100–500 μM for 24 hours). The x-axis represents the concentration of CoCl₂ used for inducing hypoxia, the letter “H” represents hypoxia, and the y-axis represents the percentage of viable cells normalized against control (*P < 0.01).

SIRT1 Levels

In control cells, we observed 0.004 ODU of SIRT1 levels (Fig. 3). During hypoxia, ODU values peaked at the 100-μM (0.026 ODU) CoCl₂ treatment condition and plateaued at the 200-μM (0.024 ODU) CoCl₂ concentration. After that point, SIRT1 ODU values began to decline to 0.014 ODU and 0.017 ODU for the 300- and 400-μM CoCl₂ concentrations, respectively. Despite the changes seen in ODU values among the CoCl₂ concentrations, as a whole, CoCl₂ treatment conditions demonstrated statistically significant more SIRT1 expression than the control (P < 0.01). However, after SIRT1 activity blockade via sirtinol, the levels of SIRT1 decreased to 0.008 ODU, 0.01 ODU, 0.003 ODU, and 0.001 ODU at 100-, 200-, 300-, and 400-μM concentrations of CoCl₂. 

Figure 3. 

 

(A) Immunoblot shows the levels of SIRT1 and the phosphorylated activation of SAPK/JNK, Total SAPK/JNK, and Control GAPDH in differentiated retinal ganglion cells (staurosporine-treated for 24 hours) in absence or presence of sirtinol (24 hours) after hypoxic insult (100–400 μM concentrations of CoCl₂ for 24 hours). “H” represents hypoxia mimetic CoCl₂ treatment; “S” represents sirtinol treatment. (B) Immunoblot shows the levels of SIRT1 after blocking JNK activity using JNK inhibitor SP600125 in the absence or presence of sirtinol after hypoxia mimetic CoCl₂ treatment for 24 hours (top panel: pretreatment of SP600125 (1 μM) for 1 hour; bottom panel: Control GAPDH); “H” represents hypoxia mimetic CoCl₂ treatment; “S” represents sirtinol treatment. (C) Quantitative evaluation of SIRT1 levels (without blocking JNK activity) using densitometry (Image J software); “H” represents hypoxia mimetic CoCl₂ treatment; “S” represents sirtinol treatment. (D) Quantitative evaluation of phosphorylated SAPK/JNK levels using densitometry (Image J software); “H” represents hypoxia mimetic CoCl₂ treatment; “S” represents sirtinol treatment.

Activation of SAPK/JNK

As shown in Figure 3, in the Western blot analysis of SAPK/JNK kinase activation during hypoxic conditions, the phosphorylation state of SAPK/JNK progressively increased to 59.9% ± 8.5% at the 100-μM, 126.4% ± 12.3% at the 200-μM, 90.0% ± 5.6% at the 300-μM, and 97.3% ± 10.4% at the 400-μM CoCl₂ concentration in comparison with the control sample (3.5%). The highly phosphorylated levels at all these treatment conditions were statistically significant, when contrasted to control (P < 0.01). 

Interestingly, SAPK/JNK maintained a high state of phosphorylation despite inhibition of SIRT1 activity. In the hypoxia-induced plus sirtinol conditions, SAPK/JNK phosphorylation measured 112.9% ± 7.3% at the 100-μM CoCl₂ concentration and remained stable when CoCl₂ was increased to the 200-μM concentration. There was further increase in phosphorylation noted at the 300- (131.0% ± 16.1%) and 400-μM concentrations (131.8% ± 12.5%). 

Activated Caspase 3 Evaluation

We observed a gradual increase of activated caspase 3 in hypoxic cells, from 1.4% ± 0.6%, to 3.73% ± 1.2%, to 4.67% ± 2.7% for the 100-, 200-, and 300-μM concentrations of CoCl₂ (Fig. 4). In comparison with control RGCs, the activated caspase 3 levels for the 400-μM concentration elevated to 10.74% ± 5.9% (Fig. 4). However, after inhibiting SIRT1 activity, the rate of activated caspase 3 increased to 5.1% ± 2.0% at the 100-μM concentration of CoCl₂. As the concentrations of CoCl₂ increased from 200 to 400 μM, a significantly increased rate of activated caspase 3 was noticed compared with the control condition: 16.82% ± 5.4% at 200 μM, 21.96% ± 2.9% at 300 μM, and 41.12% ± 9.0% at 400 μM (P < 0.05). We did not observe any significant change in activated caspase3 levels for cells treated with sirtinol alone (Fig. 4). 

Figure 4. 

 

Evaluating the neuroprotective role of SIRT1 by assessing the activated caspase 3, a mediator in the induction of apoptosis. The x-axis represents the varying concentrations of CoCl₂ for 24 hours (μM) in the presence and absence of sirtinol (24 hours); the letter “H” represents hypoxia. The y-axis represents the number of cells showing activated caspase 3 adjusted to an internal control, α-tubulin, and expressed as percentage.

Evaluation of SIRT1 Induced VEGF Release

Analysis of VEGF production showed that at the 100-μM CoCl₂ concentration, VEGF concentration was 37.9 ± 4.5 pg/mL. Control samples produced 6.93 ± 3.2 pg/mL of VEGF. The remaining concentrations of CoCl₂ produced VEGF levels between 10 and 15 pg/mL; they demonstrated statistical significance when compared with control (P < 0.05). However, for sirtinol-treated RGCs, VEGF levels were significantly reduced at lower concentrations of CoCl₂ treatment conditions (P < 0.01, Table 1). 

Table 1. 

 

VEGF Levels in the Presence of Hypoxia Mimetic CoCl₂ (100–400 μM) and in Cotreatment of CoCl₂ with Sirtinol

Experi- mental Condition	Control	H100	H200	H300	H400	H100+ Sirtinol	H200+ Sirtinol	H300+ Sirtinol	H400+ Sirtinol	Sirtinol
VEGF levels, pg/mL	6.93±3.20	37.9±4.5	13.2±5.4	15.08±2.10	10.0±4.7	11.0±2.8	5.3±1.0	13.86±5.60	9.56±2.70	8.56±1.90

Morphological Evaluation

At low hypoxic concentrations of CoCl₂ (100 and 200 μM), RGC cells showed normal cellular morphologic shape and filamentous neuronal processes in comparison with the control cells. The addition of sirtinol did not affect the shape and size of RGCs at the 100- and 200-μM concentrations. However, increased concentrations of cobalt chloride (300 and 400 μM), despite sirtinol's presence or absence, did show irregular elongated shaped cells with irregular cell borders (Fig. 5). 

Figure 5. 

 

(A) Morphological evaluation of retinal ganglion cells after the varying concentrations of CoCl₂ treatment for 24 hours. (B) Hypoxia mimetic CoCl₂ induced RGCs after the inhibition of SIRT1 activity using sirtinol (24 hours); the letter “H” represents hypoxia.

Discussion

In the current study, we investigated the role of SIRT1 on RGC viability while using an in vitro hypoxia mimetic model. The findings from this report revealed that SIRT1 levels and cell viability decline as the amount of hypoxia rises. Moreover, as SIRT1 is inhibited, cell viability further plummets. These results divulge the important role of SIRT1 activity in maintaining RGC viability during anoxic events. 

SIRT1's major role as a neuroprotector has been widely attributed to its ability to inhibit hypoxic response mechanisms deemed instrumental for initiating the cascading events of apoptosis. In an in vitro hypoxia model, Dioum et al. ¹² first reported the association of SIRT1 deacetylation properties to HIF-2 alpha activation. However, since then, the anti-hypoxic stress response nature attributed to SIRT1 has been suggested to reside mainly in its ability to deacetylate HIF-1-alpha subunits. ^19,20 Additionally, more sirtuin proteins with similar abilities to regulate HIF responsive genes have been uncovered. ²¹ HIF proteins are key modulators necessary for activating downstream second messenger and enzymatic proteins, cytokines, and gene regulatory products. HIF activation represents an important regulator step in the pathway of cellular apoptosis stimulated by hypoxia induction. 

During the cascade of cellular changes that follows hypoxia, signal transduction pathways play a vital role in coordinating the apoptotic response. Mitogen-activated protein kinases (MAPK) and JNK represent key cellular regulatory components necessary for the modulation of signaling pathways. ²² The SAPK and c-jun N-terminal kinases (JNK1/2/3) are important regulatory elements that are elevated in a number of neurodegenerative diseases. ^23–26 In a rat model of optic nerve axotomy, increased JNK 1/2/3 activity contributed to decreased RGC survival. ^27,28 In addition, there have been reports which demonstrated a significant correlation between high levels of SAPK/JNK activity and the induction of apoptosis. ^23,29,30 To develop methods to mitigate RGC apoptosis, further characterization of the interaction of SAPK/JNK to other signaling pathways is needed. In our study, induction of hypoxia was associated with a 17-fold elevation of SAPK/JNK from basal levels. This rise in SAPK/JNK eventually peaked at levels 36-fold from baseline before falling to a plateau of about 27.8-fold above basal levels. Similarly, our findings demonstrated that caspase 3 activity increased as SIRT1 protein expression declined during elevated hypoxia. 

This pattern of SAPK/JNK and caspase 3 expression shows an indirect relationship with SIRT1 profiles, as well as RGC viability (Fig. 6). Gao et al. ³¹ investigated the role intracellular kinases had on SIRT1 activity. They were able to show that SIRT1 phosphorylation lead to short-lived SIRT1 activation. SIRT1 phosphorylation was then followed by ubiquination and later degradation, by proteasomes, of the SIRT1 protein. Gao et al. ³¹ were able to correlate persistent JNK1 activation with extensive SIRT1 degradation and eventual hepatic steatosis development in obese murine hepatocytes. Furthermore, they found that as JNK1 activity was suppressed, less SIRT1 deactivation occurred. Confirming with this study, our study also showed inhibition of JNK (SP600125) with higher SIRT1 activation. In the presence of SIRT1 inhibitor, caspase 3 activity exhibited even greater expression profiles compared with experimental conditions with active SIRT1 proteins. In several studies looking at the effects of SIRT1 to mitigate apoptosis, caspase 3 activation was found to be one of the major components deactivated with increased SIRT1 levels. ^32–35 Additionally, in our investigation we were able to establish a 5.5-fold elevation in VEGF levels from baseline with low concentrations of the hypoxia mimetic agent. This elevation subsequently declined with SIRT1 inhibition as well as in the presence of increased hypoxia. VEGF is an important neuroprotectant in the central nervous system. ^36–42 In a study conducted by Nishijima et al., ⁴³ VEGF-A caused a dose-dependent reduction of retinal neuronal apoptosis in a model of ischemia-reperfusion injury. Their findings suggested that VEGF receptor-2 expression had a significant role in providing retinal neuroprotection. Moreover, they were able to show that ischemia precondition led to an elevation of VEGF-A as well as subsequent abatement of retinal cell apoptosis. This was also substantiated with their finding of chronic VEGF-A inhibition, in normal adult mice, causing significant loss of RGCs. The implication of this pattern of increased kinase and caspase expression with decreased SIRT1 and VEGF activity is congruent with the known mechanistic actions of hypoxia-induced factors in swaying the balance of pro-apoptotic agents versus anti-apoptotic agents. 

Figure 6. 

 

Trend analysis for fold change in outcome measures: SIRT1, SAPK/JNK, caspase 3, and cell death. Fold change represented by logarithmic scale.

In our investigation, SIRT1 activity expressed a 6.5-fold increase from basal levels during low states of relative hypoxia. These levels subsequently plateaued as the level of hypoxia increased within the cells. Likewise, this effect was also observed when a SIRT1 inhibitor was included into the milieu of the cells during hypoxic episodes. In this case, SIRT1 activity returned to basal levels as expressed by control RGCs. This clearly shows that as the onset of relative hypoxia occurs in RGC cells, there are underlying cellular mechanisms that perpetuate the activation of SIRT1 from basal conditions. Yet, what is still perplexing is the overall meaning of this initial rise in SIRT1 level. It is not clear whether this is a protective mechanism against programmed cell death or if this is a facet of the cellular cascade that is associated with eventual apoptosis. What is evident from our results is that as relative hypoxia increased for RGCs then the percentage of viability for the RGCs drastically diminished (Fig. 6). This result was further corroborated when SIRT1 inhibitor was added to the media; as the level of relative hypoxia increased for cultured RGCs there was a precipitous decline in the cell viability. This, thus, presents the notion that instead of SIRT1 activity acting in concert with the apoptotic pathway during hypoxia, it serves to promote cellular survival. This may in part be due to its ability to modulate the activity of the HIF-1α and HIF-2α proteins, which contributes to the hypoxic response. ¹⁹  

There are some limitations to the present study. This investigation provides only descriptive information about SIRT1's activity during states of relative hypoxia for in vitro cultured RGCs. SIRT1 expression reaches a plateau as the relative level of hypoxia is increased. This effect may have major implications when considering the point at which pro-apoptotic agents began exerting programmed cell death. SIRT1 regulation is primarily driven by the reduction of SIRT1 mRNA transcription, as well as the alteration of the redox state of nicotinamide adenine dinucleotide (NAD+/NADH) ratio, as elaborated by Lim et al.¹⁹ In terms of transcript regulation, the transcriptional co-repressor C-terminal binding protein (CtBP) has been regarded as the primary factor in modulating SIRT1 transcription. ^19,44 The ratio of NAD+ to NADH has been shown to have two primary roles for regulating SIRT1. First, NAD+ serves as a substrate necessary for deacetylation via sirtuin. Secondly, the NAD+/NADH ratio provides physiologic regulation of SIRT1 activity within the cellular environment. ⁴⁵ Additionally, NAD+/NADH acts to indirectly affect SIRT1 transcription by directly regulating the activity of CtBP. SIRT1 has been shown to inhibit the HIF-1-alpha protein, which is instrumental in the induction and propagation of the hypoxia response. ¹⁹ It is possible that as a consequence of inhibiting the HIF response downstream, apoptotic events may be averted. Although this report shows SIRT1's role in maintaining cell viability during hypoxia, the exact mechanism of how it wields this effect is not understood. 

In terms of potential improvements and possible future research endeavors, investigating the effects of SIRT1 activators would help to further the understanding of whether SIRT1 can use a dose-dependent effect in reducing caspase 3 activity and maintaining cellular viability during increasing hypoxia conditions. Furthermore, a SIRT1 activator study could establish more information as to the interaction of VEGF levels with increasing SIRT1 activation. As conveyed in this study, VEGF level fluctuations seemed to almost mirror the SIRT1 expression profile patterns during increased hypoxic conditions. This may represent either a direct or indirect involvement of SIRT1 in VEGF upregulation. As a future investigation, observing the role of SIRT1 in a knockout or knockdown (i.e., RNA interference [RNAi]) animal model could provide added understanding to the pathophysiologic effects SIRT1 inactivity has on various neuroretinal and ocular tissue types. He et al. (2010) investigated the role SIRT1 had in the kidney by using a knockdown mouse model (RNAi) of SIRT1. ⁴⁶ They found that knockdown of SIRT1 expression reduced cellular resistance to oxidative stress in primary mouse renal medullary interstitial cells. Conversely, they also witnessed that cell survival and response to oxidative stress improved with the addition of SIRT1 activators (i.e., Resveratrol or SRT2183). Additionally, in a unilateral ureteral obstruction model of kidney injury, more renal apoptosis and fibrosis was observed in knockdown mice than wild-type mice; this was attenuated when SIRT1 activator was introduced. Their findings also showed that SIRT1 knockdown led to the attenuation of oxidative stress-induced expression of cyclo-oxygenase 2 (COX2), while SIRT1 activator increased COX2 expression. Importantly, when exogenous PGE2 was delivered to SIRT1 knockdown mice there was a reduction in apoptosis to medullary interstitial cells exposed to oxidative stress. In that study, the authors suggest that not only did SIRT1 have a protective role in the mouse kidney, SIRT1's effect seemed to also be mediated through COX2 induction. He et al.'s study highlighted the fact that tissue (i.e., mouse kidney) viability against oxidative stress insults is greatly compromised as consequence of decreased SIRT1 expression. This in vivo model demonstrates how the utility of investigating SIRT1 activity in concert with other coinciding cellular homeostatic processes can provide a better understanding and appreciation for the general gestalt of SIRT1's function. 

In summary, our results provide a basis of understanding SIRT1's role in maintaining RGC viability during hypoxic episodes. This study is the first of its kind to demonstrate the link between RGC viability and SIRT1 activity within an in vitro model of hypoxia. The implications of this study may have profound effects on understanding programmed cell death in RGCs as well as methods that can be instituted to mitigate this response during hypoxic conditions. 

Footnotes

 Disclosure: S. Balaiya, None; L.R. Ferguson, None; K.V. Chalam, None