Molecular Vision 2020; 26:311-325
Received 09 January 2020 | Accepted 22 April 2020 | Published 24 April 2020
S. Sreenivasa Reddy, Y. K. Prabhakar, Ch. Uday Kumar, P. Yadagiri Reddy, G. Bhanuprakash Reddy
Biochemistry Division, National Institute of Nutrition, Hyderabad, India
Purpose: Diabetic retinopathy (DR) is the most common complication of diabetes involving microvasculature and neuronal alterations in the retina. Previously, we reported that vitamin B12 deficiency could be an independent risk factor for DR in humans. However, the effect of vitamin B12 supplementation in experimental DR is unknown. Thus, in this study, we investigated the impact of dietary supplementation of vitamin B12 on retinal changes in diabetic rats.
Methods: Diabetes was induced in 2-month-old Sprague-Dawley rats and maintained for 4 months. One group of diabetic rats were fed normal levels of vitamin B12, and one group double the quantity of vitamin B12 (50 µg/kg diet). Vitamin B12 and homocysteine levels in the plasma were analyzed with radioimmunoassay (RIA) and high-performance liquid chromatography (HPLC), respectively. At the end of 4 months of experimentation, the eyeballs were collected. Retinal changes were analyzed with hematoxylin and eosin (H&E) staining, immunoblotting, and immunofluorescence methods.
Results: Dietary supplementation of vitamin B12 had no effect on food intake, bodyweight, fasting blood glucose, and plasma homocysteine levels in the diabetic rats. However, vitamin B12 supplementation prevented loss of rhodopsin, and overexpression of VEGF, and completely prevented overexpression of HIF1α, GFAP, and endoplasmic reticulum (ER) stress markers (GRP78, ATF6α, XBP1, CHOP, and caspase 12) in the diabetic rat retina. Further, vitamin B12 ameliorated apoptosis in the retina as shown with terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) and prevented retinal thinning.
Conclusions: Vitamin B12 supplementation of diabetic rats appeared to be beneficial by circumventing retinal hypoxia, VEGF overexpression, and ER stress-mediated cell death in the retina. The present study adds another potential therapeutic strategy of vitamin B12 in diabetes.
Diabetes has become one of the most challenging health problems worldwide. Prolonged exposure to chronic hyperglycemia in diabetes can lead to various short- and long-term complications. Approximately 463 million adults globally live with diabetes, and by 2045, this will rise to 700 million International Diabetes Federation. With the increase in the incidence of diabetes, its complications have also increased accordingly, impairing quality of life and causing socioeconomic burdens. Poor glycemic control and long disease duration are significant risk factors of the microvascular and macrovascular complications that cause most of the morbidity and mortality associated with diabetes. Diabetic retinopathy (DR) is the most common microvascular complication of diabetes, and is the leading cause of blindness and visual impairment affecting one-third of people with diabetes worldwide . DR leads to vision loss through two main mechanisms: leakage of fluid in the macula and growth of new blood vessels and mechanical damage to the retina. Persistent hyperglycemia disrupts the microvasculature and neuronal functioning in the retina leading to vision loss. DR is a common complication in both types of diabetes, and the symptoms include blurred vision, floaters, difficulty seeing colors, and even total loss of vision. The early stages of DR (non-proliferative DR, NPDR) are characterized by the presence of microaneurysms, dot and blot hemorrhages, cotton wool spots, and venous abnormalities, depriving blood supply to areas of the retina. In the advanced stage of DR known as proliferative DR (PDR), neovascularization is seen that is fragile and leaky causing further damage to the retina. The prevalence of PDR is 50% in type 1 and 15% in type 2 diabetics with 25 years of disease . According to the World Health Organization (WHO), almost 32 million Indians have DR, and this number is estimated to increase to almost 80 million by 2030 , the highest number in any country.
Multiple factors are likely to be involved in the onset and progression of DR. Age, duration of diabetes, lower body mass index, higher fasting plasma glucose, and higher HbA1c levels have been identified as the risk factors most strongly associated with the development of DR [4-7]. Studies have reported ethnic differences in the prevalence and severity of DR even after controlling for systemic risk factors. A cross-sectional study conducted in the United Kingdom (UK) showed that DR was much more prevalent in people of Afro-Caribbean descent and South Asians compared to Caucasians, and they were at higher risk for sight-threatening complications . Studies suggested that patients with diabetes are at higher risk for deficiency of micronutrients [9-12]. Recently, we reported a high prevalence of multiple subclinical micronutrient deficiencies, dietary inadequacies (along with hyperhomocysteinemia) in apparently healthy adults (30–70 years old), particularly B vitamins, including vitamin B12 [13-15]. However, to date, only a few studies have evaluated the possible role of nutritional factors in the development of DR [11,12]. Most importantly, our previous studies suggested that vitamin B12 deficiency could be an independent risk factor for DR .
Vitamin B12 or cobalamin is a water-soluble vitamin that plays a fundamental role in DNA synthesis, optimal hemopoiesis, and neuronal and vascular functions. Several studies [11,16-19] have shown vitamin B12 deficiency in diabetes. A meta-analysis showed that treatment with vitamin B12 improved nerve conduction velocity in patients with diabetic peripheral neuropathy . Mizukami et al. reported that vitamin B12 supplementation improved nerve conduction velocity in diabetic rats by preventing impaired neural signaling of protein kinase C and oxidative stress-induced damage . Another study concluded that exogenous vitamin B12 delayed the onset of diabetic peripheral neuropathy via upregulation of sciatic nerve IGF-1 gene expression . Rathod et al. proved that fortification of foods with vitamin B12 helps improve brain development . However, the effect of vitamin B12 supplementation in experimental DR is unknown. Therefore, in the present study, we investigated whether vitamin B12 supplementation could influence DR in diabetic rats.
Two-month-old male Sprague-Dawley rats with an average bodyweight of 200 g were obtained from the National Center for Laboratory Animal Sciences (Hyderabad, India). All animals were fed with the AIN-93 rodent diet obtained from Research Diets Inc. (New Brunswick, NJ). The rats were housed in individual cages with temperature (22±2.0 °C), humidity (55±5.0%), and light (12-h:12-h light-dark) controlled conditions. Diabetes was induced in rats with a single intraperitoneal injection of streptozotocin (STZ; 38 mg/kg) in citrate buffer (pH 4.5), whereas the control group of rats received vehicle alone (n = 8). The STZ-induced diabetic rat is the most widely used animal model to study DR . Rats with fasting blood glucose ≥150 mg/dl were included in the study. Half of the animals (n = 8) were fed a vitamin B12-supplemented diet, and the remaining half (n = 8) were fed a normal diet similar to that of the control group of rats. The vitamin B12 -supplemented diet consists of 50 µg/kg diet, double the vitamin B12 content compared to that of the normal diet (25 µg/kg diet). All rats had ad libitum access to water and food. Bodyweight and fasting blood glucose levels were measured weekly until the end of the 4-month experimentation period. Animal care protocols were in accordance with and approved by the Institutional Animal Ethics Committee (IAEC). The study adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. At the end of the experimentation period, overnight fasted rats were euthanized by CO2 inhalation to collect the eyeballs. Four eyeballs per group, each from a different animal, were fixed in 4% paraformaldehyde for immunohistochemical analysis, and the remaining were dissected to collect the retinas, snap-frozen in liquid nitrogen, and stored at −80 °C for protein analysis.
Plasma total vitamin B12 levels were analyzed using a commercially available solid-phase radioimmunoassay kit (MP Biomedicals, Diagnostic Division, New York, NY) according to the manufacturer’s instructions. A gamma counter equipped with dual channels for determining the radioactivity of 57Co simultaneously was used (Perkin Elmer, Waltham, MA; 3 wizard 1480). Plasma total homocysteine was determined by engaging a unique reversed phase column for separating the analytes, supplied in the commercially available reverse phase high-performance liquid chromatography (HPLC) kit with fluorescence detection (Recipe Chemicals and Instruments. GmbH, Germany) .
Retinal cellular protein was extracted in Tris-lysis buffer (pH 7.5) with protease inhibitors (Sigma, St. Louis, MO), and the protein concentration was measured with the Lowry method. Equal amounts of protein from each group were separated with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to nitrocellulose membrane. The primary antibodies for HIF1α (ab187524, Abcam antibodies, Cambridge, UK), GRP78/Bip (PA5–19503, ThermoScientific, Waltham, MA), ATF6 α (MA5–16172, ThermoScientific), XBP1 (ab37152, Abcam), CHOP (2895s, Cell Signaling Technologies, Danvers, MA), caspase 12 (ab62484, Abcam), BAX (2772s, CST), VEGF (MA1–16629, tubulin (ThermoScientific), and the respective secondary antibodies were added sequentially, and the protein was detected with an enhanced chemiluminescence reagent (Bio-Rad Laboratories, Inc., Berkeley, CA). Tubulin served as a loading control. Band intensity was measured with ImageJ software (National Institutes of Health [NIH], Bethesda, MD).
Eyeballs fixed in paraformaldehyde were oriented and embedded in paraffin. The eyeballs were sectioned sagitally, and sections passing through the optic nerve and center of the cornea (4 µm thickness) were stained with either hematoxylin and eosin (H&E) or immunofluorescence for Rho, GFAP, XBP1, and CHOP as reported previously . Briefly, the ocular sections were dewaxed and rehydrated for heat-induced antigen retrieval and incubated overnight with primary antibodies for rhodopsin, GFAP, XBP1, and CHOP. The secondary antibody incubation with Alexa Fluor (488 and 568) was at room temperature for 2 h. 4′, 6-Diamidino-2-phenylin-dole (DAPI) staining was performed to label the cell nuclei. Labeled sections were visualized in a Leica fluorescence microscope (Leica Microsystems, Germany) at 400X. Images were analyzed and measured using ImageJ software (NIH). The extent of fluorescence intensity was measured as the mean gray value for rhodopsin and GFAP (as the fluorescence is prominent across the section area), and as the number of positive cells per sectional area for XBP1 and terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay.
H&E-stained ocular sections were used to measure retinal thickness. LAS v4.8 software (Leica Microsystems, Wetzlar, Germany) was employed to measure retinal thickness on 400X magnified images, at either side of the optic nerve (three on each side), approximately at equal distance from the optic nerve, and the mean was calculated. Similarly, four different animals per group were measured, and the means of all groups were subjected to statistical analysis.
TUNEL assay was performed on the ocular sections to investigate apoptosis in the retina. The In Situ Cell Death Detection Kit (Roche Diagnostics; Basel, Switzerland) was employed following the manufacturer’s protocol . After the antigens were retrieved, the sections were incubated with the TUNEL reaction mixture for 1 h at 37 °C. The cell nuclei were labeled with DAPI staining. The labeled sections were visualized in a Leica fluorescence microscope at 400X. TUNEL positive cells were analyzed and quantitated using ImageJ software (NIH).
Values are represented as the mean ± standard error of mean (SEM). Statistical analysis was performed using one-way analysis of variance (ANOVA) with a post hoc Tukey test. A p value of less than 0.05 was considered statistically significant.
In line with previous studies , the food intake and fasting blood glucose levels of the diabetes group were higher, and the bodyweights were lower, than those of the normal control rats (Figure 1). Further, vitamin B12 supplementation of diabetic rats showed no influence on any of the three parameters.
Although there was no statistically significant change in the plasma vitamin B12 levels, the plasma homocysteine levels were found to be lower in the STZ-induced diabetic rats (Figure 1). However, vitamin B12 supplementation of diabetic rats for 4 months led to increased plasma vitamin B12 levels over and above those of the control rats. Although the homocysteine levels were slightly higher in the vitamin B12 -supplemented diabetic rats than in the diabetic control rats, the level was not statistically significant.
Systematic morphological examination of the H&E-stained ocular sections as shown in Figure 2 revealed differences in retinal thickness among the groups. Although the inner plexiform layer was affected predominantly, the remaining layers, including the outer nuclear layer, were also affected significantly in the diabetic rats. The overall thickness of the retina in the diabetic rats was decreased significantly when compared with that of the control rats. Vitamin B12 supplementation in the diabetic rats partially prevented retinal thinning.
The rod cell visual photoreceptor rhodopsin (Rho) mediates the transformation of light into vision. We analyzed Rho levels in diabetic rats with and without vitamin B12 supplementation after 4 months and in age-matched non-diabetic control rats with immunofluorescence staining. Figure 3 shows reduced Rho staining (red) in the retinas of the diabetic rats, but vitamin B12 treatment considerably prevented loss of Rho in the diabetic rats. GFAP is an intermediate filament protein present in retinal glial cells. Retinal glial cells respond to retinal injury and have been shown to be activated in diabetes. Immunofluorescence staining for GFAP on the rat retinal sections showed minimal staining (red) in the control rats, especially in the ganglionic cell layer and the nerve fiber layer (Figure 4). In the diabetic rats, the GFAP staining spanned all the retinal layers indicating activation of Müller cells or gliosis. Nevertheless, vitamin B12 treatment in the diabetic rats for 4 months completely prevented diabetic gliosis.
HIF1α is a sub-unit of HIF1, a master regulator of the cellular response to hypoxia. VEGF is a prominent angiogenic factor that induces vascular permeability. Immunoblotting for HIF1α and VEGF displayed higher protein levels in diabetic rats indicative of hypoxia and elevated vascular permeability in the diabetic rat retina (Figure 5). Vitamin B12 intervention in the diabetic rats prevented overexpression of HIF1α and VEGF.
As endoplasmic reticulum (ER) stress has been shown to cause DR pathology , we examined ER stress markers such as GRP78, ATF6, XBP1, CHOP, and caspase 12 with immunoblotting (Figure 6), and XBP1 and CHOP with immunofluorescence (Figure 7 and Figure 8). The diabetic rats showed higher levels of GRP78, ATF6, and XBP1 proteins indicative of ER stress. Immunofluorescence for XBP1 and CHOP showed more staining (green and red, respectively) in the outer nuclear and inner nuclear layers of the retina in the diabetic rats. Further, CHOP and caspase 12 overexpression is indicative of maladaptive ER stress, as they trigger apoptosis. Interestingly, vitamin B12 supplementation of the diabetic rats prevented overexpression of ER stress markers in the retina (Figure 6, Figure 7, and Figure 8).
After observing maladaptive ER stress, we next examined the retinas for apoptosis. Immunoblotting for BAX protein showed higher levels in diabetic rats (Figure 6). Further, TUNEL assay was performed to confirm apoptosis, and the results showed increased TUNEL positive cells (green) as shown in Figure 9. Vitamin B12 supplementation of diabetic rats considerably prevented apoptosis in the retina.
Diabetic retinopathy is the most common microvascular complication of diabetes, and long-term complications of diabetes, including DR, represent the main cause of morbidity and mortality in diabetic patients. Nearly all individuals with type 1 diabetes and more than 60% of people with type 2 diabetes show some degree of DR after 20 years of disease. In particular, India has a high rate of diabetes and thus, DR. Established risk factors for DR include glycemic control, duration of diabetes, hypertension, and dyslipidemia [1,27]. Although strict glycemic control is expected to prevent DR, perfect glycemic control is not always possible. Multiple factors are likely to be involved in predisposing patients with diabetes to complications, as evidenced by the fact that many but not all diabetic patients develop one or more microvascular complications. In addition to genetic factors, nutritional factors (in particular, micronutrients) could play a role in the development of DR. Previously, we reported that vitamin B12 deficiency could be an independent risk factor for DR . Therefore, we investigated the effect of vitamin B12 supplementation on DR using the STZ-induced diabetic rat model.
In line with previous reports, the plasma vitamin B12 levels in the diabetic rats showed a decreased trend with no statistical significance when compared with those of the control rats [28,29]. The diabetic rats consumed double the amount of food, and thus, double the vitamin B12 intake, when compared to the control rats. However, vitamin B12 supplementation of the diabetic rats was reflected in higher plasma levels compared to those in the control rats. In agreement with previous studies in humans and rodents, there was a decrease in the circulating homocysteine levels in the diabetic rats [21,27,30-35]. This is due to deficit levels of insulin that plays a key role in homocysteine metabolism. A diabetic state, characterized by reduced insulin levels, hyperglycemia, and elevated circulating counter-regulatory hormones, has been shown to be a condition that results in the perturbation of homocysteine and methyl group metabolism. For type 1 and type 2 diabetes without renal complications, hypohomocysteinemia has been found in humans and rats. This altered methyl metabolism drastically affects various cellular processes, including DNA methylation, an epigenetic mechanism to control gene expression that has been implicated in many pathologies. Further, previous reports showed that there exists no correlation between plasma homocysteine and retinopathy .
Four months of diabetes resulted in drastic retinal thinning. Interestingly, vitamin B12 supplementation prevented retinal thinning. Rhodopsin was decreased in the diabetic rats demonstrating disturbed retinoid metabolism and photoreceptor death. This may cause visual function defects, such as delayed dark adaptation in early DR. Chronic deficiency in visual pigment formation may contribute to photoreceptor degeneration and irreversible retinal pathologies in advanced DR. The present study results suggested that restoration of the normal retinoid level with vitamin B12 supplementation may represent a potential therapeutic strategy for early DR.
Retinal Müller cells are principal macroglial cells that spread a considerable width of the retina and play a crucial role in supporting the neurons and their functions. Activation of these cells, as revealed by increased GFAP expression, is an indication of human early DR pathology. In normal conditions, GFAP expression is observed only in astrocytes of the ganglion cell layer and the nerve fiber layer of the retina. GFAP levels are regulated by hormones such as insulin and glucocorticoids, cytokines, and growth factors which are greatly altered in diabetes. However, in DR due to cellular stress and injury, Müller cells start expressing GFAP, which is considered as a marker for reactive gliosis [36,37]. Glial activation may disturb the metabolism and functioning of neurons causing neurodegeneration . In the present study, we observed that vitamin B12 supplementation in the diabetic rats prevented overexpression of GFAP in the retina. Vitamin B12 is also a good scavenger of reactive oxygen species, can pass through the blood–brain barrier, and is suggested to be a good neuroprotectant [39,40]. Previous studies showed antioxidant, anti-inflammatory, antiapoptotic, and antinecrotic effects of vitamin B12 on neurons [41,42]. Further, studies suggested that vitamin B12 treatment had a preventive effect on peripheral nerve lesions in experimental diabetic neuropathy  and a beneficiary effect in repairing the damaged nerves of rats .
The increased expression of HIF1α in the retina of diabetic rats is indicative of retinal hypoxia, which is another contributory element in DR. As the retina is a metabolically active tissue, it is sensitive to hypoxic conditions. HIF1α further promotes transcription of its target gene VEGF (Gene ID: 83785; OMIM 192240), leading to neovascularization and vascular hyperpermeability in the retina [45-48]. Müller cells are the primary source of retinal VEGF, in healthy as well as DR conditions [49,50]. Müller cell-derived VEGF protein overexpression plays a central role in DR pathology by causing blood–retina barrier breakdown, vascular leakage, acellular capillaries, retinal neovascularization and vaso-obliteration, and reduction of preretinal neovascular endothelial cells. Thus, VEGF inhibitors are the most successful therapeutic strategies at present. Although we did not show the effects of VEGF in the present study rat model, we observed higher VEGF protein expression in diabetes that is partially ameliorated by vitamin B12 supplementation. The status of DR in the present rat model could be in between severe non-proliferative retinopathy and early proliferative retinopathy.
We and others reported ER stress-mediated cell death in diabetic rat retinas and implicated it as one of the major proteostasis factors for DR [26,51]. Studies have shown involvement of ER stress in neuronal and vascular abnormalities, such as pericyte loss and neovascularization [52,53]. ATF4 that is upregulated during ER stress increases VEGF expression in the retina. ER stress also increases inflammation in the retina [51,54]. As shown in the present study, persistent ER stress induces apoptosis mediated by CHOP and caspase 12. CHOP is a key mediator of ER stress-induced cell death, and previous studies showed silencing of CHOP expression prevents ER stress-mediated cell death . Caspase 12 is an ER resident caspase, activated explicitly by ER stress to facilitate apoptosis. Interestingly, vitamin B12 supplementation in the present study abolished adaptive (GRP78, ATF6, and XBP1) and maladaptive (CHOP and caspase 12) ER stress responses in the diabetic rat retina.
Supplementation with some antioxidants and micronutrients has shown encouraging results in experimental models of DR and human studies [56,57]. Interestingly, a study showed that age-related eye disease study (AREDS)-based micronutrients proven to be beneficial in ameliorating the lesions associated with DR in experimental rats . Low levels of vitamin B12 have been recognized in Indians for a long time, and recent studies confirmed vitamin B12 deficiency and its implications in diabetes and cardiovascular diseases in India [13,15,59]. In conclusion, vitamin B12 supplementation of diabetic rats was shown to be beneficial by preventing retinal hypoxia, VEGF overexpression, and ER stress-mediated cell death in the retina. The present study adds another facet of vitamin B12 in diabetes. Considering the general prevalence of micronutrient deficiency, and its contribution to many metabolic and age-related disorders (such as diabetes) and cardiovascular diseases in India [59,60], the ameliorative effects of vitamin B12 on DR merit attention.