SIRT6 Is Required for Normal Retinal Function

The retina is one of the major energy consuming tissues within the body. In this context, synaptic transmission between light-excited rod and cone photoreceptors and downstream ON-bipolar neurons is a highly demanding energy consuming process. Sirtuin 6 (SIRT6), a NAD-dependent deacylase, plays a key role in regulating glucose metabolism. In this study, we demonstrate that SIRT6 is highly expressed in the retina, controlling levels of histone H3K9 and H3K56 acetylation. Notably, despite apparent normal histology, SIRT6 deficiency caused major retinal transmission defects concomitant to changes in expression of glycolytic genes and glutamate receptors, as well as elevated levels of apoptosis in inner retina cells. Our results identify SIRT6 as a critical modulator of retinal function, likely through its effects on chromatin.


Introduction
The mammalian retina is a highly metabolically active tissue and one of the most energy-consuming ones [1]. It requires constant supply of blood glucose to sustain its function and its energy demand is normally met through the uptake of glucose and oxygen [2,3]. Glucose movement across the blood-retinal barrier occurs mainly through the glucose transporter 1 (GLUT1) [4,5] and the need for glucose is evidenced by alterations in electroretinogram (ERG) responses and altered neurotransmitter release observed in hypoglycemic conditions [6,7,8]. It has been shown that acute hypoglycemia decreases rod and cone vision, blurs central vision, and produces temporary central scotomas in humans [6,[9][10][11]. The consequence of sustained hypoglycemia on retinal function is less clear.
Among many neuronal cellular events, action potential-mediated neuronal communication is believed to be a major process of energy consumption where energy cost comes mainly from postsynaptic receptor activation [12]. In the brain, most of the synaptic activity is mediated by glutamate, thus, the excitatory glutamatergic system represents the single largest energy user, consuming 50% of ATP in the brain [13].
Photoreceptors convert light stimuli to electric impulses. Retinal ON bipolar cells receive direct glutamatergic input from photoreceptor cells. These cells exclusively express the class III G 0 -coupled type 6 metabotropic glutamate receptor (mGluR6 or Grm6) as their primary postsynaptic glutamate receptor [14,15]. Activation of mGluR6 initiates an intracellular signaling cascade ultimately leading to closure of cGMP gated cation channels and cell hyperpolarization. Thus, energy requirement and consumption in the retina changes greatly according to neuronal activity [1].
Sirtuins are an evolutionarily conserved family of NAD +dependent deacylases that have been involved in many cellular responses to stress, including chromatin modifications, genomic stability, metabolism, inflammation, cellular senescence and organismal lifespan [16,17,18]. In mammals, 7 sirtuin isoforms have been described that differ in their subcellular localization and substrates. It is currently accepted that sirtuins are crucial regulators of energy metabolism, likely through sensing changes in levels of intracellular NAD+. Among the members of this family of proteins, SIRT6 appears to have particular significance in regulating metabolism, DNA repair and lifespan. SIRT6 knockout mice appear normal at birth, but they rapidly develop a degenerative process that includes loss of subcutaneous fat, lymphopenia, osteopenia, and acute onset of hypoglycemia, leading to death in less than one month of age [19]. Recently, Zhong et al. demonstrated that the lethal hypoglycemia exhibited by SIRT6 deficient mice is caused by an increased in glucose uptake in muscle and brown adipose tissue [20]. At a molecular level, SIRT6 functions as a histone H3K9 and H3K56 deacetylase to control glucose homeostasis by inhibiting multiple glycolytic genes, including GLUT1, and by co-repressing Hif1a, a critical regulator of nutrient stress responses [20].
Given the importance of glucose availability for retinal function, and the critical role of SIRT6 in modulating glycolysis, we undertook this study to characterize SIRT6 in the mouse retina.
We found that SIRT6 is expressed and is active in the mouse retina. SIRT6 deficiency has no effect on retinal structure but KO mice retinas exhibited up-regulation of the glucose transporter GLUT1 and down-regulation of the metabotropic glutamate receptor Grm6, indicating that both neurotransmission and glucose levels in the retina might be regulated by SIRT6. ERG analysis showed that the retina of SIRT6-KO mice was profoundly impaired.

Methods
All animal procedures were in strict accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The ethic committee of the School of Medicine, University of Buenos Aires (Institutional Committee for the Care and Use of Laboratory Animals, (CICUAL) approved this study.
Unless specified, all experiments were performed with retinas from 20-day-old mice.
In situ hybridization (ISH) ISH for Sirt6 transcripts was performed on paraffin-fixed retinas at the service at the Dana Farber/Harvard Cancer Center (DF/ HCC) specialized histopathology research pathology core using RNAscope, according to the manufacturer's instructions.

Morphologic analysis
Anesthetized mice were intracardiacally perfused with 4% paraformaldehyde as previously described [22]. Eyes were removed, immersed for 24 h in the same fixative, embedded in paraffin and sectioned (6 mm). Sections were stained with hematoxylin and eosin and used for morphometric analysis. Digital images were obtained at 1 mm dorsal and ventral from the optic disk. Thickness of the total retina as well as the individual layers was measured for each eye. Results obtained from four separate sections were averaged and the mean of 4 eyes was recorded as the representative value for each group.

RNA Isolation
Retinas were homogenized in phenol-guanidine isothiocyanate (Invitrogen Life Technologies). RNA was extracted with chloroform and precipitated with isopropanol by centrifugation (12,000 g) at 4uC. The RNA pellet was washed twice with 75% ethanol and resuspended in 30 mL of water treated with diethylpyrocarbonate. The concentration and quality of total RNA samples were assessed using a NANODROP 2000 spectrophotometer (Thermo Scientific Inc., Newark, DE). Only RNA samples with adequate purity ratios (A260/A280 J 1.9-2.1) were used for subsequent analyses.

qRT-PCR
For qRT-PCR, total RNA was isolated using RNeasy Mini Kit (QIAGEN) and cDNA was generated using QuantiTect Reverse Transcription Kit (QIAGEN) according to manufacturer instructions. qPCR was carried out using Brilliant SYBR Green qPCR Master Mix Kit (Stratagene). Data were calculated using the DCt method. Primers used were as follow: Sirt6-F GGGAACTT-GAAGGAACCACA,

RNA expression array
Whole retina mRNA from SIRT6 WT and KO mice (n = 3 from each condition) was hybridized onto an Affymetrix Mouse Gene 2.1 ST DNA microarrays. Raw expression values in the form of CEL files were processed and normalized using RMA in the R Bioconductor package (data used in the analysis is available in GEO under accession GSE56563). Clustering of the genes of glutamatergic synapse receptors was performed in R on rownormalized data using the heatmap.2 function with Euclidean distance and the complete agglomeration method. When multiple probe sets were available for a particular gene in a pathway, the probe set with the maximum variance in the data set was chosen.

Chromatin extraction and Western blot analysis
Western blot analysis was carried out as previously described [21]. The antibodies used were as follows: anti-SIRT6 (1:1000, Abcam), anti-AcH3K9 (1:500, Novus Biological), anti-AcH3K56 Homogenates were kept in ice for 20 min and centrifuged at 14.000 rpm for 10 min at 4uC. Supernatant was used for cytoplasmic protein determination and pellets were resuspended in 100 mL of HCl 0,2N. Afer 20 min on ice, samples were centrifuged at 14.000 rpm for 10 min at 4uC. Supernatants were separated and neutralized with 1M TRIS-HCl pH 8. An aliquot was used to determine protein concentration by the Bradford method. Samples (30 mg protein/well) were separated by 4-20% gradient Tris-glycine SDS-PAGE pre casted gel (BioRad). Proteins were transferred to polyvinylidene difluoride (PVDF) membranes, blocked and incubated with the corresponding primary antibodies overnight at 4uC. After washing and incubation for 1 h at room temperature with the indicated secondary antibody they were visualized by enzymatic chemiluminiscence (ECL, Western Blotting Analysis System, Amersham Biosciences, Buenos Aires, Argentina). Developed membranes were scanned and the intensity of bands was determined by using the ImageJ program (National Institutes of Health, Bethesda, MD, USA). Values were expressed as arbitrary units relative to b-actin or total H3.

Immunoflourescence
After deparaffinization, sections were immersed in 0.1% Triton X-100 in 0.1 M PBS for 10 min. Antigen retrieval was performed by heating at 90uC for 30 minutes in citrate buffer (pH 6.3). Sections were pre-incubated with 5% normal horse serum for 1 h, and incubated overnight at 4uC with the corresponding primary antibodies. After several washings, sections were incubated with the corresponding secondary antibody for 1 h at room temperature (Alexa Fluor 555 Goat Anti-Mouse IgG, 1:500, Alexa Fluor 488 Goat Anti-rabbit IgG, 1:500). Regularly, some sections were treated without primary antibodies to confirm specificity. After immunostaining, nuclei were stained with DAPI, mounted with antifade medium and viewed with a fluorescence microscope as described.

TUNEL
The ApopTag Plus Peroxidase In Situ Apoptosis Detection Kit was use to detect apoptotic cells utilizing terminal deoxynucleotidyl transferase (TdT) specific staining in paraffin embedded retinal sections according to manufacturer instructions.

Electroretinography (ERG)
Bilateral flash ERG measurements were performed as previously described [23] using a UTAS-E3000 Visual Electrodiagnostic System running EM for Windows (LKC Technologies, Inc., Gaithersburg, MD). Mice (n = 4 for each group) were dark-adapted overnight and anesthetized with 80 mg/kg ketamine and 15 mg/kg xylazine under dim red illumination for electrode placement and testing. Body temperature was maintained at 3760.5uC with a heating pad controlled by a rectal temperature probe (FHC Inc., Bowdoin, ME). The mouse's head was positioned just inside the opening of the Ganzfeld dome and pupils were dilated with 1.0% atropine sulfate (Bausch & Lomb, Tampa, FL). The recording electrode was a platinum loop 2.0 mm in diameter, positioned in a drop of 1.25% hydroxypropyl methylcellulose (GONAK; Akorn Inc., Buffalo Grove, IL) on the corneal surface of each eye. The reference needle electrode was inserted under the skin at the vertex of the skull. The ground

Statistical analysis
Quantitative data were expressed as means 6 SEM of values obtained from the ''n'' described in each figure. Significance values were calculated using Student's t-test and differences were considered significant at P,0.05.

Results
In order to characterize SIRT6 in the mouse retina, we first determined the subcellular localization of this sirtuin. The expression of SIRT6 was analyzed by in situ hybridization. Characteristic punctuate staining indicative of mRNA expression was observed in the nuclei of cells in all retinal layers of sirt6 +/+ mice (WT) (Fig. 1A). Consistently, in retina extracts from WT mice abundant protein levels were detected by Western blot, where the specificity of the antibody was confirmed by absence of signal in extract from sirt6 2/2 (KO) mice (Fig 1B). We then analyzed the structure of the retina in SIRT6 KO mice. The histological analysis revealed no evidence of structure alteration and no difference in the thickness of the whole retina or the individual layers of the KO mice compared to the WT. Moreover, KO retinas showed no evidence of inflammation or infiltrate cells (Fig. 1C-D and Fig. S1). Retinal vasculature was analyzed by a retina fundus image. No alterations were observed and both the number and caliber of vessels showed no difference between WT and KO eyes. Moreover, the underlying pigment was observed to be homogeneous suggesting no obvious retinal pigment epithelial abnormality in these animals (Fig. S1).
Changes in histone acetylation status can be used as an indication of the level of activity of a histone deacetylase (HDAC). We analyzed the level of acetylation of two of the described substrates of SIRT6: H3K9 and H3K56. Using both immunofluorescence and western blot analysis, we found that both H3K9Ac and H3K56Ac were significantly increased in retinal chromatin preparations from KO mice compared to WT mice, indicating that SIRT6 is active in the mouse retina, and lack of SIRT6 causes a profound change in those chromatin modifications ( Fig. 2 and  Fig. S2).
It has been described that SIRT6 deficiency in mice renders a phenotype characterized by a profound hypoglycemia [19] and that SIRT6-dependent deacetylation of histone H3 at lysines 9 and 56 is required for the regulation of genes associated with glucose metabolism, including GLUT1 [20]. Since this transporter is the main one responsible for the transport of glucose across the blood-retinal barrier (BRB) [4,5], we determined GLUT1 expression in WT and KO retinas. In accordance with previous findings in other tissues [20], GLUT1 was found to be upregulated in KO retinas, using immunofluorescence (Fig.3A), immunoblotting (Fig. 3B) and PCR (Fig. 3C), suggesting that glucose availability may be altered in this tissue. Up-regulation of this transporter was evident in all retinal layers and, interestingly, in inner retinal vessels. Levels of another transporter, GLUT4, were not affected (data not shown), suggesting that the effect of SIRT6 is quite specific.
In order to understand at the molecular level how SIRT6 acts in the retina, we performed a microarray analysis on total retinal RNA from 15 days-old animals. Among the genes that showed a significant variation between WT and KO animals, several genes involved in glutamatergic synapse (Grm4-8 and the kainate/ AMPA-type ionotropic glutamate receptor, GRIK1 and 2) were found to be significantly downregulated in the retinas of KO mice (FIG. 4A). It is worth noting that the most significant difference was found for the metabotropic receptor Grm6, described to be expressed exclusively in ON bipolar cells. This result was confirmed by quantitative PCR and immunostaining (Fig 4B-C). Down-regulation of Grm6 might cause inner retinal neurons to be exposed to a higher amount of endogenous glutamate that might result in severe excitotoxic degeneration [24]. Therefore, we analyzed the presence of apoptotic cells by TUNEL. Notably, a significant increase of dead cells was observed in the inner nuclear layer of SIRT6 KO retinas (Fig 4D).
Synaptic transmission is a highly energy demanding process and multiple molecular events occur during synaptic activation, including receptor trafficking, signaling, and metabolic processes. In order to analyze if the changes in gene expression observed would have a functional consequence we assessed retinal function by electroretinography (ERG). As shown in Fig. 5, both scotopic and photopic aand b-wave amplitudes were severely altered for all the light intensities analyzed, indicating that absence of SIRT6 profoundly impaired retinal function.

Discussion
The retina is one of the major energy consuming tissues within the body, maintaining high glucose metabolism for its energetic needs. Moreover, synaptic transmission is a highly energy demanding process, therefore the retina evolved mechanisms to coordinate synaptic activity with glucose homeostasis.
SIRT6 is a member of the sirtuin family of NAD(+)-dependent deacylases that acts to maintain homeostatic control of glucose metabolism by repressing a number of enzymes involved in this process [20]. Although SIRT6 plays a critical role in modulating bioenergy homeostasis and that metabolic control is essential for vision, very little is known regarding this enzyme's involvement in retinal physiology. In this study, we aimed for the first functional characterization of SIRT6 in the mouse retina.
Recent studies have shown that SIRT6 is expressed in the mouse retina and its levels seem to be higher in this tissue compared to brain, heart, liver or kidney [25]. We confirmed the presence of SIRT6 in the mouse retina and were able to determine that it is expressed in all retinal layers. Moreover, SIRT6 was found to be active in the retina since lack of this enzyme resulted in significantly increased level of acetylation of two of its described substrates, H3K9 and H3K56, in KO retinas compared to WT.
Our previous studies demonstrated that SIRT6 deficiency in mice renders a phenotype characterized by a profound hypoglycemia [19] due to increased glucose uptake. Molecularly, SIRT6 deacetylates H3K9 and H3K56, repressing expression of glycolytic genes, including the glucose transporter GLUT1 [20]. In the absence of SIRT6, those genes are up-regulated, causing uncontrolled glucose uptake and a switch towards glycolytic metabolism. Since GLUT1 is the sole transporter responsible for the transport of glucose across the blood-retinal barrier (BRB) [4,5], we focused on this transporter. Indeed, we found that mRNA and protein levels of GLUT1 were significantly increased in retinas from SIRT6 KO mice, suggesting a glucose imbalance in this tissue. It is expected that GLUT1 upregulation would be accompanied by an increase of glucose availability in the retina making difficult to explain the defects we observed. On the other hand, upregulation of GLUT1 could as well indicate a compen-satory mechanism due to local low levels of glucose since, as we published, blood glucose is being mostly uptake by skeletal muscle and adipose tissue in SIRT6 deficient mice [20]. Whether such chromatin-dependent changes in GLUT1 play a causal role in the functional defects we observed in the SIRT6 KO retinas remains to be determined.
Glutamate is the major excitatory neurotransmitter in the central nervous system, and plays important roles in regulating cell excitability and synaptic transmission. Multiple molecular events are coordinated during synaptic activation, including sodium pump activity, receptor trafficking, cytoskeletal rearrangements, signaling, and metabolic processes, making synaptic activity an energetically costly process [26]. Microarray analysis on total retinal RNA revealed that several members of the class III of the metabotropic glutamate receptor family (Grm4-8), and the kainate/AMPA-type ionotropic glutamate receptor (GRIK1-2), are down-regulated in retinas from SIRT6 KO mice. Synaptic transmission between light-excited rod photoreceptors and downstream ON-bipolar neurons is indispensable for dim vision in the mammalian retina. Indeed, disruption of this process leads to congenital stationary night blindness in human patients [27]. Notably, among the metabotropic receptors analyzed, Grm6 was found to be the most significantly down-regulated in KO retinas.
Once released from the pre-synapsis, glutamate can bind to different glutamate receptors on the post-synapsis or can be removed from the synaptic cleft by high affinity glutamate transporters located on adjacent neurons and surrounding glial cells to prevent cell death. The major pathway of glutamate metabolism consists of glutamate uptake by glutamate transporters followed by enzymatic conversion of glutamate to non-toxic glutamine by glutamine synthetase. When a glutamate transporter is pharmacologically blocked, inner retinal neurons are exposed to a higher amount of endogenous glutamate, resulting in severe excitotoxic degeneration [24]. These observations suggest that glutamate is neurotoxic when the uptake system is impaired rather than when the release is excessive. Alternatively, impaired expression of a post-synaptic receptor may contribute to the accumulation of the neurotransmitter in the inter-synaptic space. In this context, down-regulation of Grm6 could account for the increment of glutamate in the inter-synaptic space exerting a toxic effect that could explain the increase of TUNEL positive cells (bipolar cells or Müller cells or both) found in the inner nuclear Although no structural differences were observed between WT and KO retinas, functional analysis showed a significantly impaired ERG in KO animals where both a and b waves were severely compromised. Interestingly, cones would be less affected than rods since the fold decrease of the a-wave amplitude in the photopic ERG is less than the fold decrease in the scotopic ERG (2.5 vs. 8). Whether lack of SIRT6 is affecting photoreceptor activity specifically requires further studies. Taken together, upregulation of GLUT1, down-regulation of Grm6 and the increased number of apoptotic cells in retinas from sirt6 2/2 mice could account, at least in part, for the visual deficiency observed in these mice.
Overall, our studies demonstrated that SIRT6 deficiency causes major chromatin changes in the retina, which is accompanied by marked changes in expression of metabolic genes and metabotropic receptors, likely explaining the severe functional impairment observed in the SIRT6-KO retinas. Previous studies established sirtuins as critical modulators of metabolism, protecting against metabolic-and age-related diseases, such as diabetes, metabolic syndrome, cancer and neurodegenerative disorders [28]. Our results indicate that SIRT6 plays an important role in maintaining normal retinal function.
Altered methylation patterns and histone modifications have been identified in different ocular diseases like diabetic retinop- athy, glaucoma and age-related macular degeneration [29,30,31]. However, a comprehensively characterized epigenomic signature for any ocular disease remains elusive. Several questions arise: is there any epigenetic mark that would predict the onset or the progression of an ocular disease? Do epigenetic factors regulate other cellular pathways (other that glucose metabolism and glutamatergic synapse) that may alter visual function? Although epigenetic therapies have proven to be effective in cancer applications [32], the benefits of these approaches have not yet been applied for human ophthalmic diseases.

Supporting Information
Figure S1 Histological analysis and eye fundus. a) Representative low magnification (4X) cross-sections of eyes from WT and SIRT6KO mice. General structure, size and optic nerve head can be observed. No distinctive alteration is detected in SIRT6KO retinas b) Representative fundus images were taken from WT and 2-week-old mouse using the endoscopic fundus imaging system. Posterior pole of the fundus can be observed.