Zingiber officinale attenuates retinal microvascular changes in diabetic rats via anti-inflammatory and antiangiogenic mechanisms

Study design

Diabetes was induced in rats with a single intraperitoneal injection of STZ 45 mg/kg bodyweight, in 0.1 M citrate buffer, pH 4.5. Animals with a blood glucose concentration greater than 300 mg/dl after 48 h of STZ injection were considered diabetic and included in the study. Age-matched healthy rats that were injected with an equal volume of citrate buffer served as the normal control (NC). The diabetic rats were randomly divided into diabetic control (DC) and ginger extract treatment groups (6-G; n = 15; 30 eyes per group). The 6-G-treated group received the ginger extract (75 mg/kg/day) evenly dispersed in 0.5% Tween-80 by oral gavage in a volume of 0.5 ml/kg bodyweight. The DC group similarly received vehicle. The dose of ginger extract selected for this study was based on previous studies that investigated the effects of ginger extract on the blood glucose level in rats [23,26,27].

During the 24-week experimental period, bodyweight and blood glucose levels were monitored weekly using Accu-Chek® Active Glucose Test Strips and Accu-Chek® meters (Roche Diagnostics, Mumbai, India). Fundus photography was performed every 4 weeks, and the last photographs were used to calculate vessel diameter.

At the end of the experimental period, blood was collected for HbA1c. Subsequently, the rats were euthanized in a CO₂ chamber. Eyes were enucleated, and retinas were isolated to estimate TNF-α (n = 6), VEGF (n = 6), and phosphorylated NF-κB (n = 6) using ELISA, western blot analysis for NF-kB p65 (n = 4), histopathological examination using hematoxylin and eosin (H&E) staining (n = 4), and immunohistochemistry for NF-κB (n = 4). H&E staining and immunohistochemistry were performed using the same eye. Additionally, we performed transmission electron microscopy (TEM; n = 4) for accurate measurement of the retinal vascular membrane thickness.

Fundus photography and vessel diameter

The fundus photographs were taken after pupillary dilation with 1% tropicamide (Sunways, Mumbai, India) in conscious rats using a fundus camera (Genesis-Df, Kowa, Tokyo, Japan). The arteriolar and venular diameters in the fundus photographs were estimated as described previously [28]. Briefly, the vessel diameters of the three most prominent vessels were estimated at three sites in its widest portion. Before the diameter estimation, the retinal photographs from all groups were randomized, and three independent observers performed the estimations. An average of three estimates was taken as the final vessel diameter.

Inflammatory and angiogenic parameters

For the estimation of TNF-a α and VEGF, isolated retinas were homogenized in ice cold phosphate buffer saline (PBS; pH 7.4). Subsequently, TNF-α (n = 6) and VEGF levels in retinas (n = 6) were estimated using commercially available ELISA kits according to the manufacturer’s instructions.

Histopathological studies

The retinal tissues (n = 4) were fixed in 4% phosphate buffered formalin. Paraffin-embedded tissue was cut into 4-µm sections; slides were prepared and stained with H&E. Histological features were studied by an experienced pathologist unaware of the identity of the samples.

Immunohistochemistry for NF-ƙB

For the immunohistochemistry, 4-μm formalin-fixed, paraffin-embedded sections of the retinal tissue (n = 4 retinas) were placed on polylysine coated slides. The sections were deparaffinized and rehydrated, and endogenous peroxidase activity was blocked with H₂O₂ in methanol. The sections were pretreated in citrate buffer (pH 6.0) for antigen retrieval in a microwave oven [29]. Incubation was performed overnight under humid conditions using NF-ƙB rabbit polyclonal antibody (dilution 1:1,000, Abcam, Cambridge, UK) at 4 °C. The slides were then washed three times in Tris buffer for 5 min each. The mouse/rabbit ImmunoDetector horseradish peroxidase/3, 3-diaminobenzidinetetrahydrochloride (HRP/DAB) detection System (Bio SB) was used in which the sections were incubated with biotinylated anti-mouse and anti-rabbit immunoglobulin solution and then with streptavidin conjugated to HRP solution and finally with DAB plus chromogen. The slides were then counterstained with hematoxylin. Immunopositivity expression was assessed under a light microscope.

Western blot for NF-ƙB and ELISA for phosphorylated NF-ƙB

The retinal tissues were homogenized in ice-cold protein extraction buffer RIPA (3M Tris-HCl, 5M NaCl, 0.01% Nonidet P40, 0.1% glycerol, 100 mM sodium vanadate, and 0.05% protease inhibitor cocktail; Sigma-Aldrich Corporation, Bangalore, India) and incubated at 4 °C for 1 h followed by centrifugation at 10,000 ×g for 30 min at 4 °C. The supernatant was collected immediately after centrifugation and stored at −80 °C. Protein concentrations of the samples were estimated with the Bradford method. Twenty-five µg of protein was loaded in each lane, along with protein markers (molecular weight range: 10–250 kDa, Bio-Rad, Hercules, CA) in 10–12% sodium dodecyl sulfate (SDS) polyacrylamide gels and electrophoresed at 80 V. The separated proteins were then transferred on the nitrocellulose membranes at 65 V for 50 min. Later, the membranes were treated with 3% non-fat milk for 1 h to block the non-specific binding and incubated with primary antibodies against NF-kB p65 (dilution: 1:1,000, rabbit polyclonal, Abcam) and β-actin (1:5,000, mouse monoclonal, Sigma-Aldrich) for 12 h at 4 °C. The blots were then incubated in the anti-mouse and anti-rabbit secondary antibodies (dilution: 1:200) followed by incubation in preformed avidin-biotin-peroxidase complex for 2 h. The protein bands were visualized by incubation in the development solution containing DAB and hydrogen peroxide. Protein extracts from the rat retina were used as a positive control, and β-actin was used as the loading control for proteins. For densitometric analysis, the blots were scanned in a gel documentation system, using Quantity 1 software (Bio-Rad). All blots were scanned simultaneously and normalized with β-actin. The activation of NF-κB was also quantified by measuring the phosphorylated NF-κB p65 levels using the phosphorylated NF-κB p65 ELISA kit (Cell Signaling Technology, Danvers, MA), according to the manufacturer’s instructions and as described previously [30].

Transmission electron microscopy

For TEM, the retinal tissues were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). After 6 h of fixation at 4 °C, the retinas were cut around 2 mm from the region of the optic nerve head and further trimmed into 1–2 mm pieces. The 1–2 mm pieces were again fixed in 1% osmium tetraoxide, dehydrated, and embedded in araldite CY 212. The semithin sections (500 nm thick) were examined under a light microscope to select the areas of interest after staining with 0.5% toluidine blue. Thin sections (70 nm) were cut using an ultramicrotome, contrasted with uranyl acetate and lead citrate and viewed under a TECNAI G20 transmission electron microscope (FEI, Eindhoven, the Netherlands).

Measurement of vascular basement membrane thickness

The capillaries from the nerve fiber layer (NFL) and the ganglion cell layer (GCL) of the retinas were selected for the evaluation of the basement membrane (BM) thickness. Five sections were selected at 500 µm from the mid-retina. A total of four rat retinas were analyzed per group. Only cross-sectioned capillaries were considered for the BM thickness, which was measured using the Siperstein method [31]. The cross-sectioned capillaries were superimposed on a vessel, and the BM thickness was measured at the points of intersection by each beam divided symmetrically into 24 clock hours [32].

Statistical analysis

All data are expressed as mean ± standard deviation (SD). Statistical comparisons were made using one-way ANOVA (ANOVA) with the post-hoc analysis using the Dunnett multiple comparison test (Graph Pad Prism, Version 5). A p value of less than 0.05 was considered statistically significant.

Bodyweight and blood glucose

The mean bodyweight of the rats in three groups was comparable at the start of the study (p>0.05). However, at the end of the 24-week experimental period, the mean bodyweight of the animals in the DC group (210.66±11.30 g) was significantly (p<0.0001) lower than that of the NC group (328.16±20.68 g). The 6-G-treated group showed significantly greater bodyweight (298.83±29.89 g) when compared with the DC group (p<0.0001; Figure 1).

At 24 weeks, the mean blood glucose level in the DC group (533.66±113.12 mg/dl) was significantly higher (p<0.0001) than that in the NC group (99.16±8.36 mg/dl). Oral administration of the ginger extract reduced the blood glucose levels to a mean value of 361.33±93.44 mg/dl, which was significantly lower compared to that in the DC group (p<0.001) but remained significantly higher than in the NC group (p<0.001; Figure 2). In line with the blood glucose levels, HbA1c in the 6-G-treated group (6.66±1.11) was also significantly lower than in the DC group (9.91±1.56; p<0.0001) but remained significantly higher than in the NC group (4.95±0.68; p<0.01).

Fundus photography and vessel diameter

The retinal blood vessels in the DC group showed a significantly greater diameter compared to that in the NC group (p<0.0001), whereas in the 6-G-treated group the vessel diameter was significantly reduced compared to that in the DC group (arterioles p<0.001, venules p<0.0001). When compared to the NC group, the vessel diameter in the 6-G-treated group remained significantly higher (arterioles p<0.0001, venules p<0.0001; Figure 3).

Proinflammatory and angiogenic markers

The levels of the proinflammatory marker, TNF-α, in the retinal homogenate from the DC group (32.88±1.42 pg/ml) were significantly higher compared to that in the NC rats (13.75±1.67 pg/ml). The same in the 6-G-treated group (19.47±1.23) showed significant reduction compared to the DC group; however, remained higher compared to that in the NC group (Figure 4A). The VEGF levels in the DC group (14.10±1.8 pg/ml) were significantly higher compared to that in the NC group (4.71±0.8 pg/ml), whereas the 6-G rats showed mean VEGF levels of 6.6±0.85 pg/ml, which was significantly lower than in the DC rats but remained higher than in the NC rats (p<0.0001 versus DC; p<0.01 versus NC; Figure 4B). The significantly improved TNF-α and VEGF levels in the 6-G-treated group compared to the DC group but not the NC group were in accordance with the effects observed on the vessel diameter in this group.

Histopathological changes

The H&E-stained sections showed well-organized retinal layers in the NC group. In the DC group, we observed edematous retinal tissue, particularly in the inner retina. Observation of the retinal blood vessel (Ret BV) was in line with the observations made earlier of vessel diameters in the fundus photography as we observed a thinner vascular basement membrane in the 6-G-treated group as well as in the NC group compared to the DC group indicating considerable thickening of the vascular basement membrane in the DC group and a reduction in the thickness after treatment with 6-G (Figure 5A).

Immunohistochemistry

The retinas in the DC group showed intense expression indicating higher expression of NF-kB in the nerve fiber layer (NFL), inner plexiform layer (IPL), and inner nuclear layer (INL). Less intense expression of NF-kB was visualized in the NFL, IPL, and INL in the 6-G retinas (Figure 5B).

Western blot for NF-kB and ELISA for phosphorylated NF-kB

Western blot analysis was performed to assess the protein level of NF-κB p65 in the retinas of the three groups. We observed little expression of NF-κB P65 in the NC retinas whereas in the DC retinas the amount of NF-κB P65 had notably increased. However, in the retinas of the diabetic rats treated with 6-G, the expression of NF-κB p65 was markedly reduced. Quantification of P-NF-κB using ELISA showed significantly elevated levels in the DC rats compared to the NC rats, while the same was significantly lower in the 6-G-treated rats compared to the DC rats (Figure 6). These observations were in agreement with those made in the immunohistochemistry.

Basement membrane thickness

Electron microscopy revealed that the mean vascular basement membrane thickness in the DC group (173.64±10.85 nm) was significantly higher compared to the NC group (77.46±9.83 nm; p<0.0001). However, in the 6-G-treated group, the basement membrane thickness (120.14±10.50 nm) was significantly lower compared to the DC group (p<0.0001) but remained higher compared to the NC group (p<0.001; Figure 7). The electron microscopic measurements of the basement membrane thickness supported our observations made on the histopathological examination, which also showed that the basement membrane thickness was greater in the DC group compared to the NC group and the 6-G-treated group, and the same was comparable in the 6-G-treated group and the NC group.