Conflict/competing interest: No stated conflict of interest.
Funding sources: This work was supported by grant from Inje University, 2010.
Dr Hyun Woong Kim, Kekum-Dong, Busanjin-Ku, Paik Hospital, Busan 614-735, Korea. Email: email@example.com
Background: This study evaluated the effect of a lipoic acid on reactive oxygen species formation and the simultaneous changes of several angiogenic factors in an experimental diabetic rat retina.
Methods: Diabetes was induced chemically by intraperitoneal injection of streptozotocin in 30 Sprague–Dawley rats. After inducing diabetes, lipoic acid (10 mg/kg) was administered to 10 rats orally. The rats were divided into normal, diabetes mellitus, and lipoic acid-treated groups (each group n = 10). The eyeballs were harvested 8 weeks after inducing diabetes. The expression of vascular endothelial growth factor, erythropoietin, angiopoietin 1 and 2 and NADPH oxidase was examined in the rat retina using reverse transcription-polymerase chain reaction and Western blot. Superoxide formation was examined using dihydroethidium stain.
Results: Dihydroethidium analyses showed increased superoxide formation in the retina of the diabetic group. The superoxide formation was suppressed with lipoic acid treatment. Western blot analysis showed that NADPH oxidase was decreased in the diabetic group and returned to normal level in the lipoic acid-treated group. Treatment with lipoic acid blocked hyperglycaemia induced increases of vascular endothelial growth factor, angiopoietin 2 and erythropoietin shown by reverse transcription-polymerase chain reaction and Western blot analysis.
Conclusions: Lipoic acid treatment suppressed expression of vascular endothelial growth factor, angiopoietin 2 and erythropoietin via blockade of superoxide formation. Antioxidant treatment is suspected to have an antiangiogenic effect.
Vascular endothelial growth factor (VEGF), a key angiogenic growth factor, plays an important role in the angiogenic process and pathogenesis of diabetic retinopaty.1–3 Hypoxia and hyperglycaemia stimulate VEGF expression and increase reactive oxygen species (ROS). Oxidative stress represents an imbalance of excess formation and impaired removal of ROS. In diabetes mellitus, the activities of antioxidant defense enzymes such as superoxide dismutase (SOD), glutathione reductase, glutathione peroxidase and catalase responsible for scavenging free radicals and maintaining redox homeostasis are diminished in the retina.4,5
High concentrations of ROS cause apoptosis and cell death.6 Oxidative stress not only creates a vicious cycle of damage to macromolecules by amplifying the production of more ROS, but it also activates other metabolic pathways that induce the development of diabetic retinopathy. These include the multiple pathway,7 the advanced glycation end product pathway,8 protein kinase C (PKC) pathway,9,10 the hexosamine biosynthesis pathway,11 alteration in the expressions of VEGF12 and insulin-like growth factor-1,13, and elevation in mitochondrial production of superoxide and mitochondrial dysfunctions.14
The major source of ROS in endothelial cells is an NADPH oxidase.15 NADPH oxidase is activated by numerous stimuli including growth factors such as VEGF, angiopoietin 1, erythropoietin, cytokines, shear stress, hypoxia and G-protein coupled receptor agonists including angiotensin 2 in endothelial cells.6,16–18
Although oxidative stress may contribute to the development of diabetic retinopathy and elevated expression of VEGF, it is still unknown whether antioxidants as therapeutics can prevent or delay the progression of diabetic retinopathy. Many different types of antioxidant, such as vitamin C, vitamin E, β-carotene, lipoic acid and others, have been studied in cultured vascular cells, in animal models of diabetes and in diabetic patients.19–22 It is known that ROS promotes expression of VEGF in experimental study,6 but it has not been clearly established whether other angiogenic factors correlate with ROS overproduction.
Lipoic acid is an antioxidant capable of thiol-disulfide exchange. It is able to scavenge ROS and reduce metabolites, such as glutathione, to maintain a healthy cellular redox state.23 In this study, we investigated the effect of lipoic acid that prevents early diabetes-related biochemical changes and retinal vascular histopathology on the activity of NADPH oxidase, production of ROS and sequential suppression of various angiogenic growth factors such as VEGF, erythropoietin and angiopoietin.
Inbred male 30 Sprague–Dawley rats weighing 200–230 g at 6 weeks of age were purchased from Orient Bio Laboratory Animals (Daejeon, Korea). Animals had free access to deionized water and standard rat chow for 7 days after arrival. All protocols involving animals were approved by the Institutional Animal Care and Use Committee at the Inje University and were consistent with the Guidelines for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH).
Induction and assessment of diabetes
Diabetes was induced by a single intraperitoneal injection of streptozotocin (STZ, Wako Pure Chemical Industries, Osaka, Japan: 60 mg/kg body weight) in a 0.1 mol/L citrate buffer (pH 4.5) in rats (at 7 weeks of age). The age-matched control rats received an equivalent amount of citrate buffer. Diabetic rats having a blood glucose level of >300 mg/dL at 2 days following STZ injection were used for the present study. We monitored blood glucose level weekly throughout the experimental period. Blood samples were obtained after cutting the tip of the tail, and blood glucose levels were determined with AccuCheck ActiveTM (Roche Diagnostics, GmbH., Germany).
Two days after STZ or vehicle injection, rats were divided into three groups: non-diabetic rats without treatment (normal group, n = 10), diabetic rats without lipoic acid treatment (diabetic group, n = 10) and diabetic rats with lipoic acid (Thioctacid, Bukwang Pharma, Seoul, Korea) treatment (lipoic acid-treated group, n = 10). The untreated rats (diabetic group) were given 10 mg/kg sodium nitrite dissolved in deionized drinking water. Lipoic acid-treated diabetic rats were given lipoic acid (10 mg/kg each day per oral) dissolved in deionized drinking water. At the end of the 8-week experimental period, all the rats were sacrificed with CO2 gas. Twenty eyeballs were collected from each experimental group, and were randomly divided into three groups for each different study method: histologic analysis (four eyeballs, n = 4), polymerase chain reaction (PCR) analysis (four eyeballs, n = 4), Western blot analysis (four eyeballs, n = 4).
Primary antibodies for Western blot were as follows: rat anti-angiopoietin 1, anti-angiopoietin 2, anti-erythropoietin, anti-erythropoietin receptor, and anti-VEGF antibodies (Santa Cruz Biotechnology Inc., CA, USA). Secondary antibodies for Western blot was biotin-SP-conjugated donkey anti-rabbit IgG (Jackson Immuno Research Laboratories, PA, USA), HRP conjugated-Rabbit polyclonal anti-goat IgG.
Eyeballs collected from rats (both control and experimental groups) were fixed in 10% neutral buffered formalin solution and 25% sucrose solution, and samples were embedded in frozen medium. Cryosections (8 µm) were stained with hematoxylin and eosin (H&E) staining for light microscopic observation. Digital images were obtained using a high-resolution digital camera system (C3040-AD6, Olympus, Tokyo, Japan) linked to a microscope (CH30, Olympus, Tokyo, Japan) and desktop computer. For all eyes, images were obtained at a similar distance from the optic nerve.
Dihydroethidium (DHE) assay for detection of superoxide anions
The unfixed frozen retinal segments were sectioned with a cryostat and placed on glass slides. DHE (1 µmol/L; Molecular Probe, Eugene, USA) was applied to each tissue section, and then the sections were cover-slipped. The slides were incubated in a light-protected humidified chamber at 37°C for 30 min before measurement of red fluorescence labelling by a Confocal Laser Scanning Microscope (510 META, Carl ZEISS, Jena, Germany).
Immunofluorescence staining for NADPH oxidase
Fresh-frozen retinal sections were first incubated with Goat anti-rat NADPH oxidase antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) for 30 min in a light protected humidified chamber at 37°C. Samples were then incubated with FITC conjugated anti-Goat antibody (Sigma-Aldrich, ST. Louis, MO, USA) for 30 min in the same condition as earlier before measurement of green fluorescence labelling by a Confocal Laser Scanning Microscope (510 META, Carl ZEISS, Jena, Germany).
Total RNA was isolated using the RNeasy Mini Kit (QIAGEN, Hilden, Germany). RNA was transcribed into cDNA using oligo (dT) primers (Bioneer, Daejeon, Korea) and reverse transcriptase. PCR amplification was performed using specific primer sets (Bioneer, Daejeon, Korea) for Angiopoietin 1, forward primer 5′-GGA GCA TGT GAT GGA AAA TTA-3′, backward 5′-TGT GTT TTC CCT CCA TTT CTA-3′ (360 bp-product); angiopoietin 2, forward primer 5′-AAA GAG TAC AAA GAG GGC TTC-3′, backward primer 5′-TCC AGT AGT ACC ACT TGA TAC-3′ (415 bp-product); erythropoietin forward primer 5′-CCG TCC CAG ATA CCA AAG TC-3′, backward primer 5′-TGC AGA AAG TAT CCG CTG TG-3′(280 bp-product); erythropoietin receptor, forward primer 5′-CTA TGG CTG TTG CAA CGC GA-3′, backward primer 5′-CCG AGG GCA CAG GAG CTT AG-3′ (395 bp-product); Tie 2, forward primer 5′-ATG GAC TCT TTA GCC GGC TTA-3′, backward primer 5′-CCT TAT AGC CTG TCC TCG AA-3′ (337 bp-product); VEGF, forward primer 5′-CAG CTA TTG CCG TCC AAT TGA-3′, backward primer 5′-CCA GGG CTT CAT CAT TGC A-3′ (131-bp product); VEGF receptor 1, forward primer 5′-GCT CTG TGG TTC TGC GTG GA-3′, backward primer 5′-CAT GGG ATC ACC ACA GTT TT-3′ (422-bp product); VEGF receptor 2, forward primer 5′-GCA GCA CCT TGA CCT TGA AC-3′, backward primer 5′-AGG ATT GTA TTG GTC TGC CG-3′ (424-bp product). For control, a specific primer set for β-actin, forward primer 5′-AGG GAA ATC GTG CGT GAC AT-3′, backward primer 5′-AAC CGC TCA TTG CCG ATAT GT-3′ (149 bp-product) was used. PCR (25 cycles: 20 s at 94°C, 10 s at 60°C, 30 s at 72°C) was performed using AccuPower PCR premix (Bioneer, Daejeon, Korea). PCR products were analyzed by agarose gel electrophoresis and visualized with ethidium bromide under UV light using Multiple Gel-DOC system (Fujifilm, Tokyo, Japan). Densitometry was performed using Multi Gauge version 2.3 (Fujifilm, Tokyo, Japan).
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) and Western blot analysis
Western blot analysis was performed to determine the protein expression of VEGF, VEGF receptor, erythropoietin, erythropoietin receptor, angiopoietin 1, angiopoietin 2 and NADPH oxidase. Extracted retina tissues were lysed in RIPA buffer (EIPiS, Daejeon, Korea), and the proteins (10 µg/sample) were immediately heated for 5 min at 100°C. Total cell lysates (5 × 106 cells/sample) were subjected to SDS-PAGE on gels containing 15% (wt/vol) acrylamide under reducing conditions. Separated proteins were transferred to nitrocellulose membranes using semidry technique at 80 mA for 2 h. Membranes were blocked by treatment with 5% skim milk in TBS supplemented with 0.1% Tween-20 (TBST) for 1 h and subsequently incubated with primary monoclonal or polyclonal antibodies (Abs) in a final concentration of 1 µg/mL or at a final dilution of 1:1000, respectively, overnight in TBS. After three washes in TBST, membranes were incubated with peroxidase-conjugated secondary Abs (final dilution, 1:3000) in TBS for 1 h and subsequently washed as described earlier. Detection was performed by chemiluminescence using ECL-kit (Enhanced Chemiluminescence; Amersham Life Science, Braunschweig, Germany) and subsequent Multiple Gel-DOC system (Fujifilm, Tokyo, Japan). The following primary antibodies were used: anti-VEGF, anti-VEGF receptor, anti-erythropoietin, anti-erythropoietin receptor, anti-angiopoietin 1, anti-angiopoietin 2 antibodies as described earlier.
Changes of blood sugar level and weight in normal and diabetes-induced rat
Blood sugar level is elevated 2 days after injection of STZ into the abdominal cavity. After 1 week, blood sugar level was checked for diabetes induction, and the normal group had an average level of 138.1 ± 12.6 mg/dL, the diabetes group had an average level of 557.8 ± 51.4 mg/dL, and the lipoic acid-treated group had an average level of 484.2 ± 61.4 mg/dL, which was over 300 mg/dL, so diabetes induction was confirmed. In the diabetic rat, blood glucose levels were not significantly altered by lipoic acid (P > 0.05, diabetic vs. diabetic + lipoic acid). On weekly monitored glucose level, none of the animals showed a restoration of blood glucose level.
Retinal H&E staining
Two months after the induction of diabetes, microscopic examination of H&E stained retinas showed that all of the retinal layers were normal, with no morphological differences among the groups (Fig. 1). The retinal thickness and each layer thickness were qualitatively similar among the groups.
Superoxide production: DHE staining results of retinal tissue
DHE is converted to ethidium bromide in the presence of superoxide. Ethidium bromide intercalates into DNA and has been used as an indirect measure of superoxide generation. Retinas from diabetic rats showed more fluorescence than samples from normal control. This increase was down-regulated in lipoic acid-treated rats (Fig. 2).
Superoxide production: Visualization of NADPH oxidase
It is known that NADPH oxidase contributes to total superoxide production in retinas of diabetic animals. Thus, we visualized NADPH oxidase with immunofluorescence staining. Significantly more fluorescence was present in the retina from lipoic acid-treated group than in the retina from normal group, but decreased immunofluorescence was seen in the retina from the diabetic group (Fig. 3).
RT-PCR results for VEGF and VEGF receptors
Four eyeballs from four rats in each three groups were chosen, respectively, and the retinas were detached and used for the following RT-PCR experiments. RT-PCR showed that VEGF and VEGF receptor mRNA levels were higher in retinas from the diabetic group than from the normal group, and VEGF mRNA levels were lower in the retinas from the lipoic acid-treated group than from the diabetic group (P < 0.05 Kruskal–Wallis test, Fig. 4a).
RT-PCR results for erythropoietin and erythropoietin receptor
RT-PCR showed that erythropoietin and erythropoietin receptor mRNA levels were higher in retinas from the diabetic group than from the normal group, and erythropoietin and erythropoietin receptor mRNA levels were lower in the retinas from the lipoic acid-treated group than from the diabetic group (P < 0.05 Kruskal–Wallis test, Fig. 5a).
RT-PCR results for angiopoietin 1 and 2
RT-PCR showed that angiopoietin 2 mRNA levels were higher in retinas from the diabetic group than from the normal group, and angiopoietin 2 mRNA levels were lower in the retinas from the lipoic acid-treated group than from the diabetic group (P < 0.05). However, angiopoietin 1 is decreased in the diabetic group and increased in the lipoic acid-treated group compared with the diabetic group (P > 0.05 Kruskal–Wallis test, Fig. 6a).
Western blot for NADPH oxidase
Four out of 20 eyes from the 10 diabetic rats were chosen, and the retinas were detached and processed for Western blot analysis. Results show that levels of NADPH oxidase are decreased in the diabetic group and increased in the lipoic acid-treated group compared with the normal group (P < 0.05 Kruskal–Wallis test, Fig. 7).
Western blot results for VEGF
Four out of 20 eyes from the 10 diabetic rats were chosen, and the retinas were detached and processed for Western blot analysis. Results show that levels of VEGF are increased in the diabetic group and decreased in the lipoic acid-treated group compared with the diabetic group (P < 0.05 Kruskal–Wallis test, Fig. 4b).
Western blot results for erythropoietin and erythropoietin receptor
Protein levels of erythropoietin and erythropoietin receptor were checked by Western blot. Results show that levels of erythropoietin and erythropoietin receptor are increased in the diabetic group and decreased in the lipoic acid-treated group compared with the diabetic group (P < 0.05 Kruskal–Wallis test, Fig. 5b).
Western blot results for angiopoietin 2
Protein levels of only angiopoietin 2 were checked by Western blot. Results show that levels of angiopoietin 2 are increased in the diabetic group and decreased the in lipoic acid-treated group compared with the diabetic group (P < 0.05, Kruskal–Wallis test, Fig. 6b). Angiopoietin 1 was not detected in four different times.
It has been suggested that the correlation between hyperglycaemia and oxidative stress is the key element in the pathogenesis of diabetic retinopathy.24 Animal studies have demonstrated that oxidative stress contributes not only to the development of diabetic retinopathy but also to the resistance of retinopathy to reverse after good glycemic control is reinstituted.25,26 Because oxidative stress represents an imbalance between excess formation and/or impaired removal of ROS, the antioxidant defense system of the cell is an essential part of the overall oxidative stress experienced by a cell. In diabetes, the activities of antioxidant defense enzymes in the retina, such as SOD, glutathione reductase, glutathione peroxidase, and catalase responsible for scavenging free radicals and maintaining redox homeostasis are diminished.4,5 Apart from the antioxidant defense enzymes, nonenzymic antioxidants such as vitamin C, vitamin E and β-carotene that exist biologically for the regulation of redox homeostasis are also decreased during hyperglycaemia-induced oxidative stress.27
Because there is a strong understanding that oxidative stress may be the instigator of all other dysmetabolisms implicated in the pathogenesis of diabetic retinopathy, the use of appropriate antioxidants may have therapeutic potential on the metabolic and functional abnormalities found in diabetic retinopathy. Antioxidants may act at different levels of the redox system in cells and may inhibit the formation of ROS, scavenge free radicals or increase the capabilities of antioxidant defense enzymes.
Lipoic acid is an antioxidant capable of thiol-disulfide exchange. It is able to scavenge ROS and reduce metabolites, such as glutathione, to maintain a healthy cellular redox state. It distributes to the mitochondria and serves as a critical cofactor for the mitochondrial enzyme complexes23, and is regenerated via glycolytic flux. Lipoic acid attenuates the apoptosis of rat retinal capillary cells and decreases the levels of 8-OHdG and nitrotyrosine.28 Lipoic acid supplementation completely prevents diabetes-induced increase in nitrotyrosine and activation of nuclear transcription factor (NF-kB), which decreases the levels of VEGF and oxidatively modified proteins in the rat retina.28,29 Some researchers have shown that long-term administration of lipoic acid prevents the development of diabetic retinopathy in rats and decreases the number of apoptotic capillary cells and acellular capillaries in the retina of diabetic rats.28,30
The major findings in this study are that: (i) the enzyme that affects ROS production, NADPH oxidase, is not down-regulated by lipoic acid administration; but (ii) lipoic acid inhibits superoxide production that indicated an ROS level; and (iii) lipoic acid has an effect on VEGF, erythropoietin, angiopoietin-2 down-regulation, and angiopoietin-1 up-regulation by inhibition of ROS formation.
We certainly proved that lipoic acid has the beneficial effect of down-regulating superoxide, but our current Western blot results for NAPDH oxidase show that lipoic acid did not have an inhibitory effect of NADPH oxidase. Recent evidence shows that an NADPH oxidase is a major source of ROS in endothelial cells.18 Furthermore, one study proved that inhibition of NADPH oxidase activity blocks VEGF overexpression and neovascularization in ischemic retinopathy, and prevents lipid peroxidation.15 This study, however, was performed under hyperglycaemic conditions rather than hypoxic conditions. NADPH oxidase was not elevated in hyperglycaemia-induced diabetic retina, and administration of lipoic acid did not inhibit NADPH oxidase activity. We think this means that NAPDH oxidase is not a major source of superoxide in hyperglycaemia, NADPH oxidase expression might be protective or NADPH oxidase expression might be increased by negative feedback.
VEGF, an angiogenesis inducer, plays a pivotal role in diabetic retinopathy and is implicated as the mediator and initiator of non-proliferative and proliferative diabetic retinopathies.3 Retinal expression of VEGF is induced by ROS,31 and VEGF can also interact with other metabolic pathways important to the development of retinopathy, such as the PKC and polyol pathways.7
Furthermore, VEGF is a potent stimulus for angiopoietin 2 mRNA expression and release.32 This strong interaction reflects the close relationship between these factors, suggesting a tight and preferential control mechanism for angiogenesis and vascular permeability. However, VEGF appears to have only a limited effect on angiopoietin 1 expression.33 Selective up-regulation of VEGF and angiopoietin 2 may promote vascular permeability, destabilization and sprouting. These effects are normally antagonized by angiopoietin 1.34
Hence, our finding of elevated expression of angiopoietin 1, but not VEGF and angiopoietin 2 levels in the lipoic acid treated group, suggests that the effect of lipoic acid treatment improved an adverse angiogenic environment in rats with diabetes.35
The glycoprotein erythropoietin stimulates the formation of red cells by enhancing both their proliferation and differentiation, and by preventing apoptotic death of erythropoietin-responsive erythroid precursor cells.36 Hypoxia is a major signal that regulates the production of erythropoietin in tissues. Furthermore, erythropoietin shows angiogenic activity in vascular endothelial cells by stimulating proliferation, migration and angiogenesis in vitro, probably by means of the erythropoietin receptor expressed in those cells.37 Because erythropoietin is an ischemia-induced paracrine factor that promotes angiogenesis, we sought to identify whether antioxidant therapy could suppress the activity of erythropoietin.
Some studies show that erythropoietin expression is up-regulated in experimental in vivo models of retinal angiogenesis and erythropoietin is independent of VEGF expression.38 Our studies showed that decrease of oxidative stress down-regulated erythropoietin production.
Our results show that lipoic acid treatment reduces ROS overproduction, which in turn leads to inhibition of the VEGF, erythropoietin, and angiopoietin 2 expression. Many studies have focused on the therapeutic effect of VEGF suppression but have found that this incompletely inhibits neovascularization. Therefore, it is likely that other neovascular growth factors play an important role in neovascularization. Recently, many growth factors were found to have a role in the angiogenesis process. It is known that ROS is associated with production of VEGF, however, other angiogenetic factors, such as erythropoietin and angiopoietin 2, have not previously been shown to have this effect. These results indicate that lipoic acid reduces ROS overproduction, which suppresses the production of VEGF, erythropoietin, and angiopoietin 2. Lipoic acid as antioxidant represented multiple angiogenic factors not only VEGF, and it is necessary to consider antioxidant as new treatment modality in various pathologic angiogenic condition. Thus it is assumed that lipoic acid may have a preventive and therapeutic effect on diabetic retinopathy as adjunctive treatment.