Hesperetin rescues retinal oxidative stress, neuroinflammation and apoptosis in diabetic rats

Microvascular Research 87 (2013) 65–74

Contents lists available at SciVerse ScienceDirect

Microvascular Research journal homepage: www.elsevier.com/locate/ymvre

Hesperetin rescues retinal oxidative stress, neuroinflammation and apoptosis in diabetic rats Binit Kumar a, Suresh Kumar Gupta a,⁎, B.P. Srinivasan a, Tapas Chandra Nag b, Sushma Srivastava a, Rohit Saxena c, Kumar Abhiram Jha b a b c

Ocular Pharmacology Laboratory, Department of Pharmacology, Delhi Institute of Pharmaceutical Sciences & Research, University of Delhi, New Delhi-110017, India Department of Anatomy, All India Institute of Medical Sciences, New Delhi-110029, India Dr. R.P. Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi-110029, India

a r t i c l e

i n f o

Article history: Accepted 23 January 2013 Available online 31 January 2013

a b s t r a c t The purpose of the study was to evaluate the effects of hesperetin (Hsp) on diabetes-induced retinal oxidative stress, neuroinflammation and apoptosis in rats. The Hsp treatment (100 mg/kg body weight) was carried for twenty four weeks in STZ-induced diabetic rats and evaluated for antioxidant (Superoxide dismutase; SOD, Catalase; CAT and glutathione; GSH) enzymes, inflammatory cytokines (TNF-α, IL-1β), caspase-3, glial fibrillary acidic protein (GFAP) and aquaporin-4(AQP4) expression. Histological changes were evaluated by light and transmission electron microscopic (LM and TEM) studies. Retinal GSH levels and anti-oxidant enzymes (SOD and CAT) activity were significantly decreased in diabetic group as compared to normal group. However, in Hsp-treated rats, retinal GSH levels were restored close to normal levels and positive modulation of anti-oxidant enzyme activity was observed. Diabetic retinae showed significantly increased expression of Pro-inflammatory cytokines (TNF-α and IL-1β) as compared to normal retinae. While Hsp-treated retinae showed significantly lower levels of cytokines as compared to diabetic retinae. Diabetic retinae showed increased caspase-3, GFAP and AQP4 expression. However, Hsp-treated retinae showed inhibitory effect on caspase-3, GFAP and AQP4 expression. LM images showed edematous Müller cell endfeet, and also degenerated photoreceptor layer; however, protective effect of Hsp was seen on Müller cell processes and photoreceptors. TEM study showed increased basement membrane (BM) thickness in diabetic retina, while relatively thin BM was recorded in Hsp-treated retina. It can be postulated that dietary flavanoids, like Hsp, can be effective for the prevention of diabetes induced neurovascular complications such as diabetic retinopathy. © 2013 Elsevier Inc. All rights reserved.

Introduction Diabetic retinopathy (DR) is defined as disorder of the microvasculature of the retina as a result of uncontrolled hyperglycaemia. Hyperglycaemia results in the production of free radicals, which are actually associated in the pathologies such as cancer, atherosclerosis, diabetic complications, heart diseases etc. When insufficient neutralisation of free radicals occurs due to an imbalance between Pro-oxidants and Anti-oxidants in diabetes mellitus (DM) leading to oxidation of cellular lipids, proteins and nucleic acids (Gürler et al., 2000; Kowluru and Koppolu, 2002a; Kowluru and Kowluru, 2007). As a result oxidative stress exerted on the retina results in the activation of NF-kB (Kowluru et al., 2003; Romeo et al., 2002), an inflammatory cascade, and acts as a trigger for the release of pro-inflammatory cytokines (IL-1β, TNF-α) (Gupta et

⁎ Corresponding author at: Ocular Pharmacology Lab, Department of Pharmacology, DIPSAR, University of Delhi, Pushp Vihar Sec-3, New Delhi - 110017, India. Fax: +91 11 29554503. E-mail addresses: [email protected] (B. Kumar), [email protected] (S.K. Gupta). 0026-2862/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.mvr.2013.01.002

al., 2011, 2012; Zhang et al., 2011). Multiple pro-inflammatory cytokines, chemokines and adhesion molecules activate endothelium to increase expression of adhesion molecules (E-Selectin, ICAM-1) (Masuzawa et al., 2006; Zernecke and Weber, 2010). Further, leukostasis in retinal capillaries leads to capillary occlusion causing non-perfusion and ischemia. Alternatively, Cytokines (IL-1β, TNF-α) are involved in the activation of several apoptosis regulatory genes including caspase (Behl et al., 2008). Moreover, Cytokine induced NF-kB activation in retinal pericytes plays significant role in the hyperglycaemia induced pericyte death as observed in diabetic retinopathy. Pericyte loss is a critical stage in diabetic retina leading to increased vascular permeability and formation of microanurysms (Romeo et al., 2002). That is how retinal inflammation and oxidative pathways are strongly implicated in the pathogenesis of DR. Hesperetin (Hsp) is one of the most common flavonoids found in Citrus fruits (Fig. 1). Hsp has been widely studied for its potential hypolipedemic and anti-cancer properties (Aranganathan and Nalini, 2009; Morin et al., 2008). Hsp has been reported to be a powerful radical scavenger and promoter of cellular antioxidant defence-related enzyme activity (Choi, 2008; Hwang and Yen, 2008; Kim et al., 2004; Pollard

66

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

Fig. 1. Chemical structure of hesperetin (2,3-dihydro-5,7-dihydroxy-2-(3-ydroxy-methoxyphenyl)-4H-1-benzopyran-4-one) and mass spectroscopy chromatogram of product ion scan of hesperetin molecule (MH+, 303.1, Obtained from Cayman chemical Company, USA).

et al., 2006). Further, Hsp has been found to be a potential antiinflammatory (Hirata et al., 2005) and neuroprotective agent (Choi and Ahn, 2008). Earlier, we have found retinal vasculoprotective property of Hsp via its anti-angiogenic actions in experimental diabetic rats (Kumar et al., 2012b). As oxidative stress and inflammation are implicated in the onset and progression of diabetes induced neurovascular complications like DR, therefore, Hsp is expected to provide therapeutic benefits. So far, no scientific studies have been conducted to evaluate the efficacy of Hsp in the diabetes induced retinal neurovascular complication and to correlate them with antioxidant and inflammatory markers. Therefore, in the present study we investigated the effect of oral treatment with Hsp (Purity≥98%, Cayman Chemicals, USA) on hyperglycaemia-induced retinal oxidative stress, inflammation and apoptosis with supported structural and ultra-structural findings.

injection of streptozotocin (STZ 45 mg/kg body weight). Blood glucose was measured prior to the induction of diabetes and 48 hours post STZ/vehicle injection in all groups. The rats showing a blood glucose concentration greater than 300 mg/dl were considered diabetic. Age-matched normal rats served as control. Diabetic rats were divided into 2 groups of 18 rats each: the rats in group 1 received normal diet without Hsp, and group 2 received oral Hsp in a dose of 100 mg/kg body weight by oral gavage soon after establishment of diabetes (48 h after administration of STZ). Blood glucose was estimated regularly with the help of Accu-Chek® Active Glucose Test Strips using an Accu-Chek® metre (Roche Diagnostics India Pvt. Ltd). After 24 weeks of diabetes, the rats were euthanized by an overdose of pentobarbital, the eyes removed, and the retina was isolated and processed as per our previous protocols (Gupta et al., 2011; Kumar et al., 2012a).

Antioxidant parameters Materials & methods Study design Wistar albino rats (either sex) weighing 200 to 250 g were housed in the DIPSAR animal house facility, with prior approval from Institutional Animal Ethics Committee. Treatment of the animals conformed to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and with CPCSEA guidelines for the care and use of Laboratory animals. The doses were selected on the basis of LD50 value and scientific literature available, and on the basis of our present study – 100 mg/kg bodyweight dose is reported in this scientific paper. Diabetes was induced in rats with a single

Estimation of anti-oxidant parameter such as Glutathione (GSH), Superoxide dismutase (SOD) and Catalase (CAT) were performed using commercially available kits from Cayman Chemicals Ltd. All estimations were done as per manufacturer's instructions.

Inflammatory parameters TNF-α and IL-1β levels in retinae were estimated using commercially available enzyme-linked immunosorbent assay (ELISA) kit from Diaclone, France, and Ray Biotech, Inc., respectively as per the manufacturer's instructions.

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

Light microscopy (LM) & immunohistochemistry Cryostat retinal sections (14 μm) were rehydrated in phosphate buffered saline (PBS) for 20 min, blocked with 10% normal goat serum in PBS for one hour, and incubated overnight in a moist chamber with the primary antibody diluted in PBS containing 3% normal goat serum and 0.5% Triton X-100. For rabbit polyclonal anti-Caspase-3 (1:2000 dilution) and GFAP detection (1:1000 dilution, Abcam Plc., UK), the sections underwent heat-induced antigen retrieval with a microwave oven (three 5-minute cycles in 10 mM Tris–EDTA buffer [pH 9] at 650 W). Further, the sections were incubated with the biotinylated secondary antibody and reacted with the avidin–biotinylated peroxidase complex. The reaction product was visualized by incubation in development solution (containing 3, 3-diaminobenzidine tetrahydrochloride (DAB) and hydrogen peroxide). The slides were lightly counterstained with hematoxylin in case of caspase-3. Finally, the sections were rinsed with distilled water, cleared, dehydrated in ethanol, mounted and cover slipped. Cryostat sections (4 μm) were stained with Haemotoxylin & Eosin (H&E) for ganglion cell counting and histopathology. Ganglion cells were counted in central and peripheral retina (both peripheral sides, nasal and temporal), and average ganglion cell number was reported per 100 μm length in total six retinas from each group.

67

Quantification of photoreceptor nuclei Photoreceptor damage was quantified by counting the number of Photoreceptor nuclei per 25 μm Square grid using Digital Micrograph™ Software. Quantification was done in central retina, nasal peripheral retina and temporal peripheral retina in a total of six retinae per group. Statistical analysis All data are expressed as mean±SD. The groups were compared by one-way ANOVA with suitable post hoc multiple comparison test. Data were considered statistically significant at p-valueb 0.05. Results Blood glucose and body weight Blood glucose levels in the diabetic group (541.50± 52.09 mg/dl) were significantly higher than in the normal rats (100.67 ± 5.01 mg/dl) (pb 0.001) at the end of 24 week period. In Hsp-treated rats the blood glucose levels (464.85±37.53 mg/dl) were significantly lower than in the diabetic group (pb 0.05). Body weight in normal group was found to be increased by 54.32% as compared to diabetic group with a weight gain of 20.86%. Rats in Hsp-treated group gained 35.63%.

Aquaporin-4 (AQP4) immunofluorescence Antioxidant parameters Fourteen μm thick cryosections were incubated in 10% normal goat serum for 1 h. The sections were then incubated in polyclonal rabbit anti-rat AQP4 antibody (Abcam Plc., UK, 1:1000) overnight at 4 °C, followed by anti-rabbit secondary antibody conjugated to FITC (Molecular Probes, USA) to identify the localization of AQP4 in the retinal sections. Slides were mounted in anti-fade medium containing DAPI (Vectashield-DAPI; Vector Laboratories, CA, USA) to counterstain the nuclei, and images were obtained with microscope (Leica DM6000 B fluorescence microscope).

Retinal GSH levels were almost four fold lower in diabetic rats as compared to normal rats (p b 0.001). However, in Hsp-treated rats, retinal GSH levels were significantly higher than diabetic group (pb 0.05) (Fig. 2A). The antioxidant enzymes SOD and CAT showed more than three-fold decrease in activity in diabetic retinae as compared to normal retinae (pb 0.001). Both SOD and CAT enzymatic activities were restored close to normal in Hsp-treated diabetic rats (pb 0.05) (Fig. 2B).

Transmission electron microscopy

Inflammatory parameters

Eyes were enucleated from respective groups. Soon after enucleation cornea and lens were removed. The posterior segment was carefully freed from vitreous adhering to the retina gently with a blunt forceps and retina isolated. Further, the retinal tissues were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 6 h at 4 °C. After fixation, the retina was circumcised around 2 mm around the optic nerve head and further trimmed into 1 mm 2 pieces. After fixation in 1% osmium tetraoxide, tissue was dehydrated, and embedded in araldite CY 212. Semithin section (500 nm thick) were stained with 0.5% toluidine blue and examined under a light microscope. The LM images thus obtained were evaluated to study LM changes in the retina of respective groups. After selecting the areas of interest, the sections (70 nm) were cut on an ultra-microtome and placed on the grids. Staining was done with uranyl acetate and lead citrate, and viewed under TECNAI G20 transmission electron microscope (FEI Company, The Netherlands). Digital Micrograph™ software was used for analysing the ultramicrographs.

TNF-α estimations in untreated diabetic retinae showed more than 2-fold higher levels than in the normal retinae (pb 0.001). The TNF-α levels in the retinae from Hsp-treated rats were 1.5-fold lower than untreated diabetic retinae (pb 0.05) (Fig. 3A). Mean IL-1β value in normal rat retinae was found to be more than 2-fold lower than the untreated diabetic retinae (pb 0.001). Mean IL-1β values in Hsp-treated rats were significantly lower than untreated diabetics (pb 0.05) (Fig. 3B).

Measurement of basement membrane thickness We evaluated only the capillaries from the outer plexiform layer. Electron micrographs of six randomly selected capillaries from mid-retina per animal and a total of 6 rat retinae were analysed per experimental group. Only cross-sectioned capillaries were considered. BM thickness was measured as per method described by Siperstein et al. (1966) method. A grid equally divided into 24-clock hours was superimposed on a vessel. The BM thickness was measured at point of intersection by each spoke. We have also included two points having minimal BM thickness.

Immunohistochemistry Generalised pattern of caspase-3 expression was seen in normal rat retina (Fig. 4A). On the other hand, intense expression of caspase-3 was seen in nerve fibre layer (astrocytes) and Müller cells in case of diabetic retina (Fig. 4B). However, Hsp-treated retina showed relatively lesser expression of caspase-3 than diabetic retina (Fig. 4C). Positive GFAP expression was seen in diabetic retina (Figs. 5A and B); however insignificant GFAP expression was seen in treated retina (Fig. 5C). Normal retina served as negative control for GFAP expression. Aquaporin-4 (AQP4) immunofluorescence In diabetic retina, AQP4 was strongly expressed in Müller cell endfeet and in perivascular space (Figs. 6E and F) as compared to normal retina showing moderate immunoreactivity in Müller cell endfeet (Figs. 6B and C). Hsp-treated retina (Figs. 6H and I) showed lesser immunoreactivity compared to diabetic retina and less immunofluorescence in Müller cell endfeet.

68

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

Fig. 2. (A). Effect of Hsp on retinal GSH levels in different study groups after 24 weeks of diabetes in rats. Values are presented as mean ± SD, n = 6. *p b 0.001 compared with normal; #pb 0.05 compared with Hsp-treated diabetic. (B). Effects of Hsp on retinal SOD and CAT activities after 24 weeks of diabetes in rats. Values are presented as mean±SD, n=6. *pb 0.001 compared with normal; #pb 0.05 compared with Hsp-treated diabetic.

Light and electron microscopic studies Diabetic retina showed degenerated outer nuclear layer and photoreceptor layer as a result of separation from retinal pigment epithelium layer (Fig. 7B). Inner detached retina is remained by many Müller cell processes as seen in the form of strands (Figs. 7B and C). Hsp-treated retina showing intact retinal layers without any signs of diabetes induced structural changes (Fig. 7D). The retinal detachment and extensive photoreceptor degeneration was rare in experimental rat and other models, and only detected once out of twelve diabetic retinae studied in the present study(six from each diabetic and Hsp-treated diabetic) after six months of diabetes. Average ganglion cell counts in diabetic retinae were found to be significantly less than normal retinae. However, Hsp-treatment helped in rescuing apoptotic ganglion cell death as compared to diabetic retinae (Fig. 8). LM image of toluidine stained normal retina showed well-organised structural morphology (Fig. 9A). On the other hand, LM image of diabetic retina has shown edematous Müller cell endfeet, increase intercellular spaces in inner nuclear and outer nuclear layers. Further, photoreceptor layer was also shown to be degenerated and damaged (Fig. 9B). However, Hsp-treated retina showed comparatively lesser degenerated changes in ganglion cell layer and inner nuclear layer. Outer nuclear layer and photoreceptor layer appeared to be normal in structure (Fig. 9C).

Fig. 3. (A). Effect of Hsp on retinal TNF-α levels in normal, diabetic and diabetic + Hsp groups after 24 weeks of diabetes. Values are mean ± SD, n = 6. (*p b 0.001) diabetic vs. normal; (#p b 0.05) diabetic vs. Hsp-treated diabetic. (B). Effect of Hsp on retinal IL-1β levels in normal, diabetic and diabetic + Hsp groups after 24 weeks of diabetes. Values are presented as mean ± SD, n = 6. *p b 0.001 diabetic vs. normal; # p b 0.05 diabetic vs. Hsp-treated diabetic.

Further, electron microscopic ultrastructure showed degenerated and pyknotic photoreceptor nuclei; and degenerated rod outer segment in diabetic retina (Figs. 10B, 11B, C and F). On the other hand, Hsptreated retina showed well defined photoreceptor nuclei and rod outer segments (Figs. 10C, 11D and G). Electron microscopic observation recorded a thinner capillary BM (0.085±0.01 μm) in normal retinae as compared to diabetic retinae with a thicker capillary BM (0.180±0.02 μm) (pb 0.001). On the other side, Hsp-treated retinae recorded relatively thin capillary BM (0.105± 0.02 μm) compared to diabetic retina (pb 0.001) (Fig. 12). Discussions Diabetic complications such as DR are a result of multiple metabolic, vascular and inflammatory defects. Uncontrolled hyperglycaemia is the key metabolic abnormality in DM. The early signs of DR in experimental diabetic models include vascular inflammatory response due to oxidative stress results in over-expression of pro-inflammatory cytokines (TNF-α, IL-1β etc.), and consequent over-expression of ICAM-1 leading to leucostasis. As a result occluded retinal capillary leads to ischemia, thereafter causing compromised vascular and neural functions. Uncontrolled hyperglycaemia induced retinal metabolic abnormalities lead to increased oxidative stress, which contribute to development of diabetic complications such as DR (Gürler et al., 2000; Kowluru and Koppolu, 2002a; Kowluru and Kowluru, 2007). The retina has a highly

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

69

Fig. 4. Caspase-3 expression in normal, diabetic and Hsp-treated retina. (A). Normal retina shows normal pattern of Caspase-3 expression. Counterstained with hematoxylin. (B). Diabetic retina showing increased expression of caspase-3 in astrocytes (NFL, arrow heads) and along Müller cell fibres (IPL, arrows), and INL. (C). Hsp-treated retina shows relatively lesser expression of caspase-3 in NFL, IPL and INL. Counterstained with hematoxylin. NFL—nerve fibre layer, INL—inner nuclear layer, IPL—inner plexiform layer, ONL—outer nuclear layer, PRL—photoreceptor layer.

efficient antioxidant defence mechanism comprising of free radical scavengers such as α-tocopherol, glutathione, and ascorbic acid and antioxidant enzymes such as glutathione peroxidase, SOD and CAT (Armstrong et al., 1981; Castorina et al., 1992; Kowluru et al.,

2001). The diabetic rats in our study showed decreased levels of GSH and subnormal activity of SOD & CAT as compared to normal rats (Figs. 2A and B). These changes in antioxidant parameters are in accordance with earlier studies (Gupta et al., 2011; Obrosova et al., 2006).

Fig. 5. GFAP expression in diabetic and treated retina. It is prominent in Müller cell inner processes (arrows) in diabetic retina (A, B), but insignificant in treated retina (C). Normal retina is GFAP negative (not shown).

70

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

Fig. 6. Representative fluorescence micrographs showing AQP4 immunoreactivity in different study groups. (A–C), AQP4 immunoreactivity is moderately present in Müller cell endfeet of normal retina; (D–F) In diabetic retina, AQP4 is strongly expressed in Müller cell endfeet (arrows) and in perivascular space (star); (G–I) Hsp-treated retina showing lesser immunoreactivity compared to that in diabetic retina and less staining in Müller cell endfeet. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; ONL, outer nuclear layer. Scale bar: 25 μm (shown in Figure I, common to all micrographs).

Hsp-treated retinae showed restoration of antioxidant defence system of retina back to normal, thereby preventing retinal inflammatory changes (Figs. 2A and B). Hsp being a flavanone offers powerful anti-oxidant activity which modulates the enzymatic activity effectively (Aranganathan and Nalini, 2009; Aranganathan et al., 2009). Similarly, in our study positive modulation of GSH and antioxidant enzymes was observed.

Pro-inflammatory cytokines (TNF-α & IL-1β) are the key mediator in neurovascular inflammation and upregulated in DR. IL-1β is very well studied in Blood Retinal Barrier Breakdown (Bamforth et al., 1997) and induction of retinal capillary cell apoptosis in diabetic retina (Kowluru and Odenbach, 2004a, 2004b). Further, contribution of TNF-α to the pathogenesis DR is supported by a number of reports (Joussen et al., 2002, 2003) and significantly higher levels were found in the plasma

Fig. 7. Photomicrographs of rat retina in normal and other experimental groups. A. Retinal architecture in normal group, B. diabetic retina showing degenerated outer nuclear layer (ONL) and photoreceptor layer (PRL). Inner retina is still attached by many Müller cell processes as seen in the form of strands (arrows), extensive degenerated photoreceptor layer might be a complication of retinal detachment (this is a rare observation), C. Magnified view of selected region from Fig. B, showing clearly visible Müller cell strands (arrows), D. Hsp-treated retina showing intact retinal layers without any signs of diabetes induced structural changes. NFL, nerve fibre layer; IPL, inner plexiform layer; INL, inner nuclear layer; ONL, outer nuclear layer; PRL, photoreceptor layer; PEL, pigment epithelial layer.

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

71

Fig. 8. Effect of Hsp on retinal ganglion cell apoptosis after 24 weeks of diabetes. Ganglion cell counting was done per 100 μm of retina in normal, diabetic and diabetic + Hsp groups. Values are mean ± SD, n = 6. (*p b 0.05) diabetic vs. normal; (#p b 0.05) diabetic vs. Hsp-treated diabetic.

Fig. 9. Representative LM images of different study groups. (A). Normal retina showing retinal layers without any obvious damage (B). Diabetic retina showing edematous Müller cell endfeet (arrow), increases in intercellular spaces in INL and ONL due to apoptosis and degenerated PRL (arrow) (C). Hsp-treated retina showing lesser edematous changes in GCL and INL. The ONL and PRL appear intact. GCL-ganglion cell layer, IPL- inner plexiform layer, INL-inner nuclear layer, ONL-Outer nuclear layer, PRC-photoreceptor layer. Scale bar — 50 μm.

Fig. 10. Effect of Hsp on photoreceptor degeneration, A. Normal photoreceptor nuclei structure, B. diabetic retina showing degenerated and pyknotic photoreceptor nuclei, C. Hsp-treated retina showing well characterised photoreceptor nuclei, except few pyknotic nuclei (star), D. photoreceptor nuclei were counted per 25 μm square area at mid and periphery retina regions. Values are mean ± SD, n = 6. (p b 0.05) diabetic vs. normal; (p b 0.05) diabetic vs. Hsp-treated diabetic.

72

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

Fig. 11. A. Rod outer segment (ROS) of normal retina, B. degenerated and vacuolated ROS (arrow) seen in diabetic retina, C. magnified view of ROS showing excessive formation of vacuoles (arrow) and degeneration of discs, D. Hsp-treated retina showing normal structured ROS, E. Normal retina showing well-organised ROS, microvilli are long and the RPE layer showing healthy Mitochondria (arrowhead), F. diabetic retina showing degenerated (star) and vacuolated ROS (arrow), microvilli are disorganised in structure and degenerated mitochondria (arrow head) in RPE, G. Hsp-treated retina showing outer of retina with well-defined ROS, long microvilli and normal mitochondria (arrowhead). — ROS, rod outer segments, RPE — retinal pigmented epithelium.

Fig. 12. Retinal capillary endothelial BM thickness in different groups, A. capillary from normal group, showing a thin BM (0.07 μm & 0.07 μm), B. capillary from diabetic group, showing a thick BM (0.16 μm & 0.18 μm), pericyte (P) and pericyte mitochondria (m) are degenerated, muller cell processes are swollen and degenerated, C. retinal capillary from Hsp-treated group, showing a relatively thin BM (0.08 μm & 0.09 μm), pericyte (P) is intact. l, lumen of capillary; n, nucleus of endothelial cell; *, Capillary endothelium; p, Pericyte. Scale Bar — 0.5 μm, C. Average BM thickness in respective groups. Values are mean±SD, n=6. (*pb 0.001) diabetic vs. normal; (#pb 0.001) diabetic vs. Hsp-treated diabetic, D. endothelial cell/pericyte ratio (E/P ratio) in respective groups. Values are mean±SD, n=6. (*pb 0.05) diabetic vs. normal; (#pb 0.05) diabetic vs. Hsp-treated diabetic.

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

of diabetic patients (Foss et al., 1992; Zorena et al., 2007). Similarly, in the present study significantly higher levels of pro-inflammatory cytokines (TNF-α & IL-1β) were found in the retinae of diabetic rats versus normal rat retinae (Figs. 3A and B). However, Hsp-treated diabetic retinae showed significantly lower levels of TNF-α & IL-1β as compared to diabetic retinae (Figs. 3A and B). Similarly, various studies have shown that dietary flavanoids including Hsp inhibits IL-1β (Kowluru and Odenbach, 2004b; Sharma et al., 2007) & TNF-α (Vafeiadou et al., 2009; Xagorari et al., 2001) mediated retinal inflammation and apoptosis. Hyperglycaemia acts as a trigger that causes activation of various apoptotic proteins involved in apoptotic cell death, including members of the caspase family (Allen et al., 2005). Earlier studies showed that caspases play important role in the initiation and execution of apoptosis (Doonan and Cotter, 2004; Katai and Yoshimura, 1999). Moreover, caspase-3 has been studied to play an important role in diabetes and its complications (Kern et al., 2000; Kowluru and Koppolu, 2002b). Similarly, in the present study, we have seen an increased expression of caspase-3 in nerve fibre layer and along Müller cell processes as compared to normal retina (Fig. 4B). However, treatment with Hsp inhibited the expression of caspase-3 as compared to diabetic retina (Fig. 4C). Our findings are consistent with earlier studies that anti-oxidants could play important role in inhibiting caspase-3 mediated retinal apoptosis (Kowluru and Koppolu, 2002b). The Müller (radial glial) cells are the principal glia of all vertebrate retinae.They stabilizes the retinal architecture, helpful in fluid homeostasis and support neuronal survival and information processing. It has been shown that Müller cell destruction causes retinal dysplasia, photoreceptor apoptosis and, finally, retinal degeneration (Bringmann et al., 2004; Dubois-Dauphin et al., 2000; Mizutani et al., 1998). In the present study, diabetic retina showing edematous Müller cell end feet in nerve fibre layer (Figs. 6F & 9B), ganglion cell loss (Fig. 8), increased intercellular spaces in inner-nuclear layer and outer nuclear layer, degenerated outer retina due to apoptotic cell death as a result of over-expression of caspase-3 (Figs. 4 & 7), and also Photoreceptor layer degeneration due to inflammation induced apoptosis and retinal detachment (Figs. 10 and 11). Further, TEM ultramicrograph showing degenerated and swollen Müller cell processes (Fig. 12B). However, LM images of toluidine and H&E stained normal retina showing normal Müller cell structures and all nuclear layers without any obvious damage or degeneration (Figs. 7A and 9A). Similarly, Hsp-treated retina showing lesser edematous Müller cell endfeets in nerve fibre layer and normally organised inner-nuclear layer (Fig. 6I & 9C). The preventive effects observed in Hsp-treated retina are due to its potential anti-oxidant, anti-inflammatory and anti-apoptotic properties. Further, Increased expression of GFAP is the hallmark of glial cell reactivity. Earlier various studies have shown significantly increased expression of GFAP in diabetic retina (Feit-Leichman et al., 2005; Mizutani et al., 1998). Similarly, in the present study we have observed increased GFAP expression along Müller cell processes. However, Hsp-treated retina showed insignificant expression of GFAP (Fig. 5). Previous studies have also shown inhibitory effects of flavanoids on GFAP activation in Müller cells (Yang et al., 2009). AQP4, a water specific membrane-channel protein, is more specifically expressed in Müller cells and astrocytes, and its overexpression has been implicated in neuronal and glial swelling (Bringmann et al., 2005; Da and Verkman, 2004; Liu et al., 2007; Nagelhus et al., 2004). Earlier studies have shown that intense expression of AQP4 at Müller cell endfeets (Da and Verkman, 2004; Nagelhus et al., 1998). Similarly, we have found increased intensity of AQP4 at Müller cell endfeet in diabetic retina as compared to normal rat retina(Figs. 6F and C). However, Hsp treated retina showed decreased expression of AQP4 at Müller cell endfeet (Fig. 6I). Further, various authors have suggested that APQ4 inhibition could be considered as novel target for therapeutic intervention (Bringmann et al., 2005; Da and Verkman, 2004; Nagelhus et al., 1998).

73

Hyperglycaemia results in increased synthesis of extracellular matrix causing thickened BM, loss of pericytes and degeneration of capillary endothelium (Casaroli Marano et al., 1995; Roy et al., 2003; Spirin et al., 1999). Further, Oxidative stress and inflammatory cytokines (TNFα and IL-1β) are strongly implicated in increased basement thickening (Behl et al., 2008; Kumar et al., 2012a). Similarly, in the present study increased levels of inflammatory mediators are directly involved in thickening of BM in diabetic retina compared to normal retina. On the other hand, Hsp treatment has shown preventive effect on thickening of BM (Fig. 12). Similarly, earlier studies have also shown the role of polyphenolic compounds in preventing hyperglycaemia induced thickening of BM (Gupta et al., 2011). Therefore, the present study is first to clearly demonstrate the therapeutic benefits of Hsp in rescuing retinal neuroinflammation, oxidative stress, apoptosis and oedema as a result of chronic uncontrolled hyperglycaemic state. In conclusion, it can be postulated that dietary flavanoids, like Hsp, can be very effective for the prevention of diabetes induced neurovascular complications such as DR. Acknowledgment Financial support from the Department of Science & Technology under DPRP is gratefully acknowledged. Facilities for electron microscopy availed at SAIF (DST), All India Institute of Medical Sciences, New Delhi, are acknowledged. References Allen, D.A., Yaqoob, M.M., Harwood, S.M., 2005. Mechanisms of high glucose-induced apoptosis and its relationship to diabetic complications. J. Nutr. Biochem. 12, 705–713. Aranganathan, S., Nalini, N., 2009. Efficacy of the potential chemopreventive agent, hesperetin (citrus flavanone), on 1,2-dimethylhydrazine induced colon carcinogenesis. Food Chem. Toxicol. 47, 2594–2600. Aranganathan, S., Selvam, J.P., Sangeetha, N., Nalini, N., 2009. Modulatory efficacy of hesperetin (citrus flavanone) on xenobiotic-metabolizing enzymes during 1,2-dimethylhydrazine-induced colon carcinogenesis. Chem. Biol. Interact. 180, 254–261. Armstrong, D., Santangelo, G., Connole, E., 1981. The distribution of peroxide regulating enzyme in canine eye. Curr. Eye Res. 1, 225–242. Bamforth, S.D., Lightman, S.L., Greenwood, J., 1997. Interleukin-1 beta-induced disruption of the retinal vascular barrier of the central nervous system is mediated through leukocyte recruitment and histamine. Am. J. Pathol. 50, 329–340. Behl, Y., Krothapalli, P., Desta, T., DiPiazza, A., Roy, S., Graves, D.T., 2008. Diabetes-enhanced tumor necrosis factor-alpha production promotes apoptosis and the loss of retinal microvascular cells in type 1 and type 2 models of diabetic retinopathy. Am. J. Pathol. 172, 1411–1418. Bringmann, A., Francke, M., Reichenbach, A., 2004. Müller cells in retinopathies. Adv. Mol. Cell Biol. 31, 1117–1132. Bringmann, A., Uckermann, O., Pannicke, T., Iandiev, I., Reichenbach, A., Wiedemann, P., 2005. Neuronal versus glial cell swelling in the ischaemic retina. Acta Ophthalmol. Scand. 83, 528–538. Casaroli Marano, R.P., Preissner, K.T., Vilaro, S., 1995. Fibronectin, laminin, vitronectin and their receptors at newly-formed capillaries in proliferative diabetic retinopathy. Exp. Eye Res. 60, 5–17. Castorina, C., Campisi, A., Di Giacomo, C., 1992. Lipid peroxidation and antioxidant enzymatic systems in rat retina as a function of age. Neurochem. Res. 17, 599–604. Choi, E.J., 2008. Antioxidative effects of hesperetin against 7, 12-dimethylbenz (a) anthracene-induced oxidative stress in mice. Life Sci. 82, 1059–1064. Choi, E.J., Ahn, W.S., 2008. Neuroprotective effects of chronic hesperetin administration in mice. Arch. Pharm. Res. 31, 1457–1462. Da, T., Verkman, A.S., 2004. Aquaporin-4 gene disruption in mice protects against impaired retinal function and cell death after ischemia. Invest. Ophthalmol. Vis. Sci. 45, 4477–4483. Doonan, F., Cotter, T.G., 2004. Apoptosis: a potential therapeutic target for retinal degenerations. Curr. Neurovasc. Res. 1, 41–53. Dubois-Dauphin, M., Poitry-Yamate, C., de Bilbao, F., Julliard, A.K., Jourdan, F., Donati, G., 2000. Early postnatal Müller cell death leads to retinal but not optic nerve degeneration in NSE-Hu-Bcl-2 transgenic mice. Neuroscience 95, 9–21. Feit-Leichman, R.A., Kinouchi, R., Takeda, M., Fan, Z., Mohr, S., Kern, T.S., Chen, D.F., 2005. Vascular damage in a mouse model of diabetic retinopathy: relation to neuronal and glial changes. Invest. Ophthalmol. Vis. Sci. 46, 4281–4287. Foss, M.C., Foss, N.T., Paccola, G.M., Silva, C.L., 1992. Serum levels of tumor necrosis factor in insulin-dependent diabetic patients. Braz. J. Med. Biol. Res. 25, 239–242. Gupta, S.K., Kumar, B., Nag, T.C., Agrawal, S.S., Agrawal, R., Agrawal, P., Saxena, R., Srivastava, S., 2011. Curcumin prevents experimental diabetic retinopathy in rats through its hypoglycemic, antioxidant, and anti-inflammatory mechanisms. J. Ocul. Pharmacol. Ther. 27, 123–130. Gupta, S.K., Kumar, B., Srinivasan, B.P., Nag, T.C., Srivastava, S., Saxena, R., 2012. Retinoprotective effects of Moringa oleifera via antioxidant, anti-inflammatory,

74

B. Kumar et al. / Microvascular Research 87 (2013) 65–74

and anti-angiogenic mechanisms in streptozotocin-induced diabetic rats. J. Ocul. Pharmacol. Ther. http://dx.doi.org/10.1089/jop. 2012.0089. Gürler, B., Vural, H., Yilmaz, N., Oguz, H., Satici, A., Aksoy, N., 2000. The role of oxidative stress in diabetic retinopathy. Eye 5, 730–735. Hirata, A., Murakami, Y., Shoji, M., Kadoma, Y., Fujisawa, S., 2005. Kinetics of radicalscavenging activity of hesperetin and hesperidin and their inhibitory activity on COX-2 expression. Anticancer Res. 25, 3367–3374. Hwang, S.L., Yen, G.C., 2008. Neuroprotective effects of the citrus flavanones against H2O2-induced cytotoxicity in PC12 cells. J. Agric. Food Chem. 56, 859–864. Joussen, A.M., Poulaki, V., Mitsiades, N., Kirchhof, B., Koizumi, K., Döhmen, S., Adamis, A.P., 2002. Nonsteroidal anti-inflammatory drugs prevent early diabetic retinopathy via TNF-alpha suppression. FASEB J. 16, 438–440. Joussen, A.M., Poulaki, V., Mitsiades, N., Cai, W.Y., Suzuma, I., Pak, J., Ju, S.T., Rook, S.L., Esser, P., Mitsiades, C.S., Kirchhof, B., Adamis, A.P., Aiello, L.P., 2003. Suppression of Fas-FasL-induced endothelial cell apoptosis prevents diabetic blood-retinal barrier breakdown in a model of streptozotocin-induced diabetes. FASEB J. 17, 76–78. Katai, N., Yoshimura, N., 1999. Apoptotic retinal neuronal death by ischemia-reperfusion is executed by two distinct caspase family proteases. Invest. Ophthalmol. Vis. Sci. 40, 2697–2705. Kern, T.S., Tang, J., Mizutani, M., Kowluru, R.A., Nagaraj, R.H., Romeo, G., Podesta, F., Lorenzi, M., 2000. Response of capillary cell death to aminoguanidine predicts the development of retinopathy: comparison of diabetes and galactosemia. Invest. Ophthalmol. Vis. Sci. 12, 3972–3978. Kim, J.Y., Jung, K.J., Choi, J.S., Chung, H.Y., 2004. Hesperetin: a potent antioxidant against peroxynitrite. Free Radic. Res. 38, 761–769. Kowluru, R.A., Koppolu, P., 2002a. Termination of experimental galactosemia in rats, and progression of retinal metabolic abnormalities. Invest. Ophthalmol. Vis. Sci. 43, 3287–3291. Kowluru, R.A., Koppolu, P., 2002b. Diabetes-induced activation of caspase-3 in retina: effect of antioxidant therapy. Free Radic. Res. 36, 993–999. Kowluru, V., Kowluru, R.A., 2007. Increased oxidative stress in diabetes regulates activation of a small molecular weight G-protein, H-Ras, in the retina. Mol. Vis. 13, 602–610. Kowluru, R.A., Odenbach, S., 2004a. Role of interleukin-1beta in the development of retinopathy in rats: effect of antioxidants. Invest. Ophthalmol. Vis. Sci. 45, 4161–4166. Kowluru, R.A., Odenbach, S., 2004b. Role of interleukin-1beta in the pathogenesis of diabetic retinopathy. Br. J. Ophthalmol. 88, 1343–1347. Kowluru, R.A., Tang, J., Kern, T.S., 2001. Abnormalities of retinal metabolism in diabetes and experimental galactosemia. VII. Effect of long-term administration of antioxidants on the development of retinopathy. Diabetes 50, 1938–1942. Kowluru, R.A., Koppolu, P., Chakrabarti, S., Chen, S., 2003. Diabetes-induced activation of nuclear transcriptional factor in the retina, and its inhibition by antioxidants. Free Radic. Res. 37, 1169–1180. Kumar, B., Gupta, S.K., Nag, T.C., Srivastava, S., Saxena, R., 2012a. Green tea prevents hyperglycemia-induced retinal oxidative stress and inflammation in streptozotocininduced diabetic rats. Ophthalmic Res. 47, 103–108. Kumar, B., Gupta, S.K., Srinivasan, B.P., Nag, T.C., Srivastava, S., Saxena, R., 2012b. Hesperetin ameliorates hyperglycemia induced retinal vasculopathy via antiangiogenic effects in experimental diabetic rats. Vascul. Pharmacol. 57, 201–207. Liu, X.Q., Kobayashi, H., Jin, Z.B., Wada, A., Nao-i, N., 2007. Differential expression of Kir4. 1 and aquaporin 4 in the retina from endotoxin-induced uveitis rat. Mol. Vis. 13, 309. Masuzawa, K., Goto, K., Jesmin, S., Maeda, S., Miyauchi, T., Kaji, Y., Oshika, T., Hori, S., 2006. An endothelin type A receptor antagonist reverses upregulated VEGF and

ICAM-1 levels in streptozotocin-induced diabetic rat retina. Curr. Eye Res. 31, 79–89. Mizutani, M., Gerhardinger, C., Lorenzi, M., 1998. Müller cell changes in human diabetic retinopathy. Diabetes 47, 445–449. Morin, B., Nichols, L.A., Zalasky, K.M., Davis, J.W., Manthey, J.A., Holland, L.J., 2008. The citrus flavonoids hesperetin and nobiletin differentially regulate low density lipoprotein receptor gene transcription in HepG2 liver cells. J. Nutr. 138, 1274–1281. Nagelhus, E.A., Veruki, M.L., Torp, R., Haug, F.M., Laake, J.H., Nielsen, S., Ottersen, O.P., 1998. Aquaporin-4 water channel protein in the rat retina and optic nerve: polarized expression in Müller cells and fibrous astrocytes. J. Neurosci. 18, 2506–2519. Nagelhus, E.A., Mathiisen, T.M., Ottersen, O.P., 2004. Aquaporin-4 in the central nervous system: cellular and subcellular distribution and coexpression with KIR4. 1. Neuroscience 129, 905–913. Obrosova, I.G., Drel, V.R., Kumagai, A.K., Szábo, C., Pacher, P., Stevens, M.J., 2006. Early diabetes-induced biochemical changes in the retina: comparison of rat and mouse models. Diabetologia 49, 2525–2533. Pollard, S.E., Whiteman, M., Spencer, J.P., 2006. Modulation of peroxynitrite induced fibroblast injury by hesperetin: a role for intracellular scavenging and modulation of ERK signalling. Biochem. Biophys. Res. Commun. 347, 916–923. Romeo, G., Liu, W.H., Asnaghi, V., Kern, T.S., Lorenzi, M., 2002. Activation of nuclear factor-kB induced by diabetes and high glucose regulates a proapoptotic program in retinal pericytes. Diabetes 51, 2241–2248. Roy, S., Sato, T., Paryani, G., Kao, R., 2003. Downregulation of fibronectin overexpression reduces basement membrane thickening and vascular lesions in retinas of galactosefed rats. Diabetes 52, 1229–1234. Sharma, V., Mishra, M., Ghosh, S., Tewari, R., Basu, A., Seth, P., Sen, E., 2007. Modulation of interleukin-1β mediated inflammatory response in human astrocytes by flavonoids: implications in neuroprotection. Brain Res. Bull. 73, 55–63. Siperstein, M.D., Norton, W., Unger, R.H., Madison, L.L., 1966. Muscle capillary basement membrane width in normal, diabetic and prediabetic patients. Trans. Assoc. Am. Physicians 79, 330–347. Spirin, K.S., Saghizadeh, M., Lewin, S.L., Zardi, L., Kenney, M.C., Ljubimov, A.V., 1999. Basement membrane and growth factor gene expression in normal and diabetic human retinas. Curr. Eye Res. 18, 490–499. Vafeiadou, K., Vauzour, D., Lee, H.Y., Rodriguez-Mateos, A., Williams, R.J., Spencer, J.P., 2009. The citrus flavanone naringenin inhibits inflammatory signalling in glial cells and protects against neuroinflammatory injury. Arch. Biochem. Biophys. 484, 100–109. Xagorari, A., Papapetropoulos, A., Mauromatis, A., Economou, M., Fotsis, T., Roussos, C., 2001. Luteolin inhibits an endotoxin-stimulated phosphorylation cascade and proinflammatory cytokine production in macrophages. J. Pharmacol. Exp. Ther. 296, 181–187. Yang, L.P., Sun, H.L., Wu, L.M., Guo, X.J., Dou, H.L., Tso, M.O., Zhao, L., Li, S.M., 2009. Baicalein reduces inflammatory process in a rodent model of diabetic retinopathy. Invest. Ophthalmol. Vis. Sci. 50, 2319–2327. Zernecke, A., Weber, C., 2010. Chemokines in the vascular inflammatory response of atherosclerosis. Cardiovasc. Res. 86, 192–201. Zhang, W., Liu, H., Al-Shabrawey, M., Caldwell, R.W., Caldwell, R.B., 2011. Inflammation and diabetic retinal microvascular complications. J. Cardiovasc. Dis. Res. 2, 96–103. Zorena, K., Myśliwska, J., Myśliwiec, M., Balcerska, A., Hak, Ł., Lipowski, P., Raczyńska, K., 2007. Serum TNF-alpha level predicts nonproliferative diabetic retinopathy in children. Mediators Inflamm. 2007, 92196.