Background
Diabetes-induced retinal dysfunction is a common and serious microvascular complication [
1]. Hyperglycemia induces retinal vascular injury and blood–retinal barrier (BRB) breakage, which can cause macular edema in patients with diabetes [
2]. The blood vessel endothelium is composed of a single layer of endothelial cells, and the cleft between neighboring endothelial cells is tightly sealed by tight junctions of trans membrane protein complexes such as occludins, claudins, and zona occludens [
3]. These complexes contribute to the paracellular barrier, such as the blood–brain barrier and BRB [
4]. This endothelial cell barrier can be used as a therapeutic target for vascular permeability-associated diabetic retinopathy [
5].
The medical care strategy for patients with diabetic retinopathy is focused on early diagnosis and tight control of blood glucose levels to slow the onset of the disease. Although several classes of glucose-lowering drugs have been developed, the prevalence of diabetic retinopathy is still high [
6]. Recently, there has been considerable focus on botanical products that can potentially prevent or treat diabetes and its complications.
Aster koraiensis (Korean starwort) is a valuable perennial Korean native plant. This herb is used in food and in folk medicine to treat diseases such as pneumonia, chronic bronchitis, diabetes, and pertussis [
7,
8]. In our previous studies, we reported that the extract of
A. koraiensis inhibits the formation of AGEs and its cross-links with proteins in vitro [
9,
10]; and prevented AGE deposition and podocyte apoptosis in the renal tissues of streptozotocin (STZ)-induced diabetic rats [
11]. Recently, the extract of
A. koraiensis was also shown to inhibit retinal pericyte apoptosis in STZ-induced diabetic rats [
9]. Although the various effects of
A. koraiensis extract on diabetes-induced renal and retinal injury in a type 1 diabetes animal model have been reported, the effect on diabetes-induced retinal vascular injury in a type 2 diabetes animal model is unknown. To elucidate this, we investigated the inhibitory effect of
A. koraiensis extract on the retinal vascular dysfunction in Spontaneously Diabetic Torii (SDT) rats, which is a type 2 diabetes animal model. We also determined the possible mechanism of AKE on the loss of tight junction protein associated with retinal vascular hyperpermeability in this animal model.
Methods
Preparation of a. Koraiensis extract
Aerial parts including the flowers, leaves, and stems of
A. koraiensis were purchased from Gongju, Chungchengnamdo, South Korea in August 2007. The
A. koraiensis extract (AKE) was prepared according to a previously reported method [
11]. Briefly, 2.5 kg of
A. koraiensis was extracted with EtOH (3 × 20 L) by maceration at room temperature for 3 days. The extracted solution was concentrated to AKE powder (303 g) in vacuum at 40 °C. Voucher specimens of
A. koraiensis were deposited at the Herbarium of the Korea Institute of Oriental Medicine, Korea (Herbarium No. KIOM-83A). AKE was standardized using high-performance liquid chromatography (HPLC) with, chlorogenic acid and 3, 5-di-
O-caffeoylquinic acid (Sigma, MO, USA) as reference compounds. The HPLC fingerprint and contents of chlorogenic acid and 3, 5-di-O-caffeoylquinic acid of AKE are described previously [
9].
Animals and experimental design
Male SDT rats and age-matched Sprague-Dawley rats were purchased from CLEA Japan (Tokyo, Japan). At 25 weeks, the rats were randomly divided into four groups of 10 rats, as follows: (1) normal rats (NOR), (2) SDT rats, (3) and (4) SDT rats treated with AKE (SDT + AKE, 50 and 100 mg/kg body weight, respectively). AKE was orally administered for 6 weeks. At necropsy, all rats were sacrificed by CO2 asphyxiation. The animals were housed at the animal facility in Qu-Best Co. Ltd. (Seoul, Korea). All experimental procedures were approved by the Institutional Animal Care and Use Committee at Qu-Best Co. Ltd., Seoul, Korea (IACUC Approval No. QE10072).
Blood glucose
Blood glucose levels were consistently monitored. Blood samples were collected from the tail vein after a 16 h fast. Blood glucose levels were measured using an automated analyzer (Wako, Tokyo, Japan).
At the end of the study, rats were anesthetized with isoflurane. Then, 50 mg/mL fluorescein-dextran (Sigma) in phosphate-buffered saline (PBS) was injected into the left ventricle. The tracer dye was allowed to perfuse for 15 min and the eyeballs were then placed in 4% paraformaldehyde for 1.5 h. Blood fluorescein concentration was measured with a fluorometer (Synergy™ HT, Bio-Tek, VT, USA). Dissected retinas were placed on a microscope slide and the whole-mount retinas were observed by fluorescence microscopy (Olympus, Tokyo, Japan). Fluorescence intensity was determined by Image J software (National Institutes of Health, Bethesda, MD, USA) and normalized to the blood fluorescein intensity for each rat.
Preparation of trypsin-digested vessels
The dissected retinas were fixed in 10% formalin for 24 h. This was followed by incubation in sodium phosphate buffer containing 3% trypsin and 100 mM NaF to inhibit the DNase activity for 1 h. The retinal microvessels were separated from the retinal cells by gentle rinsing in PBS and mounted on slides.
Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining
The retinal microvessels were stained with a TUNEL fluorescein kit (Promega, Madison, WI, USA) according to the manufacturer’s protocols. TUNEL-positive cells were counted per mm2 of capillary area in five fields.
Immunofluorescence staining
The trypsin digests were hydrated and incubated with either antibodies raised against AGEs (Transgenic Inc., Kobe, Japan) or occludin (Invitrogen, CA, USA) for 1 h at room temperature. Signal detection was achieved using a rhodamine-conjugated goat anti-mouse antibody (Santa Cruz Biotechnology, Paso Robles, CA, USA). As a negative control, tissue sections were incubated with serum from non-immunized animals, instead of the primary antibody. The signal intensity was analyzed in five randomly selected mm2 of capillary area from each rat using Image J software.
Western blot analysis
The retinas were homogenized in (radio-immunoprecipitation assay) RIPA lysis buffer with 1% protease inhibitor cocktail and total protein concentrations were quantified. Protein lysates were separated using SDS–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA, USA). For immunoblotting, the following antibodies were used: rabbit anti-occludin antibody (Invitrogen Life Technologies) and mouse β-actin antibody (Sigma). The immunoreactive bands were visualized and analyzed using an enhanced chemiluminescence detection system (Luminograph II system, Atto Corp., Tokyo, Japan).
Statistical analysis
Group data were analyzed via one-way analysis of variance followed by Tukey’s multiple comparison test or via unpaired Student’s t-test, using GraphPad Prism 6.0 software (GraphPad, San Diego, CA, USA). Differences with a p value <0.05 were considered statistically significant.
Discussion
Currently, several dietary herbal supplements with beneficial effects on health are sold worldwide. In this study, we showed that AKE, as a dietary herbal supplement, ameliorates diabetes-induced retinal vascular dysfunction in a type 2 diabetes animal model. AKE-treated SDT rats showed significant improvements in retinal vascular function markers, such as vascular permeability. Additionally, AKE reduced apoptosis and AGE accumulation in retinal vascular cells.
In our previous study, we reported that chlorogenic acid is a bioactive compound of AKE and that the content of chlorogenic acid in AKE is 1.24%. An oral dose of 100 mg/kg AKE contained 1.24 mg/kg chlorogenic acid is recommended [
9]. Qi et al. reported that the peak plasma concentration of chlorogenic acid after oral administration of 50 mg/kg chlorogenic acid to rats was 550 μg/ml [
12]. Although we used a relatively low dose of chlorogenic acid, a concentration of 52.55 μg/ml chlorogenic acid is sufficient to inhibit AGE formation in vitro [
9]. Indeed, an oral administration of AKE (100 mg/kg body weight) in STZ-induced diabetic rats reduced the AGEs deposited in the retinal microvascular cells. Thus, we chose the oral doses of 50 and 100 mg/kg AKE to evaluate their inhibitory effects on retinal AGEs accumulations in the SDT rats.
The SDT rat is a newly developed non-obese type 2 diabetes animal model that spontaneously develops hyperglycemia resulting from reduced insulin secretion [
13]. An SDT rat is the most appropriate animal model to investigate diabetic retinopathy. Retinal vascular leakage, vascular cell loss, and proliferative neovascularization are characteristics of SDT rats, thereby resembling the clinical features of human diabetic retinopathy [
14,
15].
The BRB is critical for the maintenance of normal microvascular function, and its breakdown is an early clinical feature of diabetic retinopathy [
16]. The retinal vascular endothelium has high trans-endothelial resistance due to intercellular tight junctions [
17]. Occludin was the first identified component of tight junction strands and has an important role in the regulation of endothelium electrical resistance, permeability, and barrier function [
18]. In this study, fluorescein-dextran microscopy demonstrated that AKE significantly decreased fluorescein leakage, suggesting that AKE might prevent BRB breakdown. Moreover, the loss of occludin was markedly inhibited after the administration of AKE. These results suggest that AKE can prevent retinal vascular leakage under diabetic conditions.
In the retinal tissues, extensive formation and accumulation of AGEs is considered responsible for the pathogenesis of diabetic retinopathy [
19,
20]. Repeated intravenous (i.v.) injection of exogenous AGEs into animals with normal blood glucose levels resulted in BRB breakdown in the retinas [
21]. It has been reported that AGEs elicit pro-apoptotic cytokine release or reactive oxygen species (ROS) generation through the AGE/RAGE (receptor for AGE) interaction, resulting in apoptotic cell death of retinal vascular cells [
22‐
24]. Moreover, anti-AGE agent, aminoguanidine, prevented the diabetes-induced retinal vascular injury [
25]. In our previous study, we reported that AKE has a potent anti-AGE activity. In STZ-induced diabetic rats, a type 1 diabetes animal model, oral administration of AKE (100 mg/kg body weight) for 4 months reduced the AGEs deposited in the retinal microvascular cells [
9]. Similarly, the present study showed that AKE (50 and 100 mg/kg body weight) also dose-dependently prevented AGE accumulation and vascular cell apoptosis in the retinal vessels of SDT rats. Based on these findings, we suggest that AKE potentially ameliorates diabetic retinopathy through anti-AGE activity and that the minimum effective dose of AKE was 50 mg/kg body weight.
Chlorogenic acid and 3, 5-di-
O-caffeoylquinic acid are two major compounds present in AKE [
9]. Kim et al. showed that chlorogenic acid could inhibit the formation of AGEs and the AGE/RAGE interaction in vitro [
9,
10]. Chlorogenic acid has a chelating activity towards methylglyoxal [
26], which can disturb the formation of AGEs. 3, 5-Di-
O-caffeoylquinic acid prevented the formation of AGEs with approximately 20-fold higher activity than chlorogenic acid [
9]. These findings suggest that the preventive effects of AKE on diabetic retinopathy may be ascribed to the synergistic effects of these two active components.