Background
Diabetic nephropathy is the leading cause of end-stage renal disease and is a growing global health problem. Despite developments in pharmacological strategies to modulate diabetes, diabetic nephropathy remains a major microvascular complication in many patients with diabetes. However, there is still insufficient understanding about the full mechanisms involved in progressive diabetic renal disease. One of the causal factors for the progression of diabetic kidney injury is lipotoxicity. Increasing evidence suggests that renal ectopic lipid accumulation and dysregulated lipid metabolism in the kidney are involved, in part, in the pathophysiology of diabetic nephropathy [
1,
2].
Anthocyanins, flavonoids within the polyphenol class, are most abundant in various coloured fruits, vegetables, red wine, and grains, and possess a basic skeleton of 2-phenylbenzopyrylium or flavylium glycoside [
3,
4]. Anthocyanins are the most oxidized flavonoids with a fully unsaturated C ring and a hydroxyl at position 3, and are potential therapeutic agents as antioxidants [
5]. Their mechanisms of action have been investigated in numerous experimental models in vivo and in vitro. Interest has focused on their potential antioxidant activity, which is proposed as a key mechanism for the prevention of many chronic diseases including metabolic disorders and cancer [
6‐
9]. Anthocyanins have potential antidiabetic activity to protect against pancreatic cell damage and improve insulin sensitivity, and have a potent therapeutic effect on diabetic complications such as diabetic retinopathy [
10,
11]. However, it has barely been established whether these natural compounds have beneficial effects on the pathogenesis of diabetic nephropathy.
Adenosine monophosphate-activated protein kinase (AMPK) is a crucial metabolic energy sensor involved in a wide range of biological activities that normalise lipid, glucose, and energy imbalances [
12]. AMPK is activated by low energy status (AMP/ATP ratio) and is dysregulated in patients with metabolic syndromes such as diabetes and obesity. It has been reported that AMPK is abundantly expressed in the kidney and its activity is reduced in the diabetic kidney [
13]. Experimental evidence suggests that AMPK activation attenuates lipotoxicity, reactive oxygen species generation, inflammation, and endothelial dysfunction in type 2 diabetes [
13‐
15].
The collective results favour the view that anthocyanins have beneficial effects on lipid metabolism dependent on activation of the AMPK pathway in metabolic disorders including type 2 diabetes, obesity, and non-alcoholic fatty liver disease [
4,
16]. Anthocyanins ameliorate hyperglycaemia and insulin resistance via activation of AMPK, with inactivation of acetyl-CoA carboxylase (ACC) in hepatocytes, white adipose tissue, and skeletal muscle in diabetic animal models [
17].
We hypothesised that anthocyanin-rich Seoritae extract (SE;
Glycine max L.) may ameliorate renal oxidative stress and lipotoxicity, which are the principal causes of renal damage in diabetic nephropathy, via activation of AMPK and the consequent effects on its target molecules. Glomerular endothelial dysfunction is commonly the initiating insult that predisposes tissue to injury in diabetes [
18]. We evaluated the effects of anthocyanins on high glucose-induced oxidative stress and apoptosis related to AMPK activation and its downstream molecules in cultured human glomerular endothelial cells (HGECs).
Methods
Preparation of SE and analysis of anthocyanins in SE
The SE used was produced by the following method. Seoritae samples (1,500 g) were extracted with 12,000 mL of 30% ethanol for 3 h at 90–100°C. The solution was filtered twice through 50 and 1-µm filters and concentrated in a vacuum evaporator (60°C) to 70 brix. The residual solvent was removed in a drying machine for 18 h at 60°C in a vacuum. The resulting powder was stored in a plastic bag until use. The anthocyanin content in SE was analysed by HPLC using a Dionex Ultimate 3000 series dual low-pressure ternary gradient pump (Dionex Softron GmbH, Germering, Germany) and an Ultimate 3000 series photodiode array detector for anthocyanin analysis. The crude anthocyanin extract was analysed by its HPLC chromatogram. Three principal anthocyanin peaks were detected in the chromatogram by diode array detection at 530 nm. The major anthocyanins were identified as delphinidin-3-
O-glucoside (25.2%), cyanidin-3-
O-glucoside (68.3%), and petunidin-3-
O-glucoside (6.5%) by comparison with the HPLC retention times of standard compounds, as previously described [
19].
Experimental methods
Male 6-week-old C57BLKS/J db/m and db/db mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA). The mice were fed a regular chow diet, provided with water ad libitum, and allowed to acclimatise for 1 week before experiments. The mice were divided into four groups. Control db/m mice (n = 6) and control db/db mice (n = 6) received drinking water only, while anthocyanin db/m mice (n = 8) and anthocyanin db/db mice (n = 8) received 10 mg/kg body weight anthocyanin-rich SE daily for 12 weeks. For measurement of 24-h urinary albumin, the mice were placed in individual mouse metabolic cages (Nalgene, Rochester, NY, USA) every 4 weeks. At week 20, the mice were anaesthetised by intraperitoneal injection of a mixture of Rompun (10 mg/kg; Bayer Korea, Ansan, Gyeonggi-Do, Korea) and Zoletil (30 mg/kg; Virbac, Carros, France). Blood was collected from the left ventricle and centrifuged, and the resulting plasma was stored at −70°C for analyses. The kidneys were rapidly dissected and stored in 10% buffered formalin for immunohistochemical analyses. HbA1c was measured from red cell lysates by HPLC (Bio-Rad, Richmond, CA, USA). Triglyceride and total cholesterol concentrations were determined using an automatic analyser (Model 917; Hitachi, Tokyo, Japan) and commercial kits (Wako, Osaka, Japan). NEFA levels were measured with a JCA-BM1250 automatic analyser (JEOL, Tokyo, Japan).
Ethics statement
All animal experiments were in accordance with the Laboratory Animals Welfare Act, Guide for the Care and Use of Laboratory Animals, and Guidelines and Policies for Rodent Experiments provided by the Institutional Animal Care and Use Committee at the School of Medicine, The Catholic University of Korea (Approval No. YEO20131601FA). All procedures complied with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication No. 85–23, revised 1996).
Assessment of albuminuria, renal oxidative stress, and intra-renal lipid content
Twenty-four-hour urine collection for measurement of albuminuria was performed using metabolic cages (Nalgene) at week 20, and urinary albumin concentrations were obtained by immunoassay (Bayer, Elkhart, IN, USA). To evaluate oxidative stress, we measured 24-h urinary 8-epi-prostaglandin F
2α (OxisResearch, Foster City, CA, USA). Intra-renal lipids were extracted using the method of Bligh and Dyer with slight modifications [
20].
Light microscopy study
Kidney samples were fixed in 10% buffered formalin and embedded in paraffin. Histology was assessed after periodic acid-Schiff staining. The mesangial matrix and glomerular tuft were quantified for each glomerular cross-section, as previously reported [
21]. More than 30 glomeruli, cut through the vascular pole, were counted per kidney, and the average of the measured areas was used for analysis.
Immunohistochemistry for TGF-β1, type IV collagen, and TUNEL assay
For immunohistochemistry, 4-μm-thick sections were deparaffinised, hydrated in ethanol, treated with an antigen-unmasking solution containing 10 mmol/L sodium citrate buffer (pH 6.0), and washed with PBS. The sections were incubated with 3% H2O2 in methanol to block endogenous peroxidase activity, and then with 10% normal goat serum in PBS to block non-specific binding. The sections were incubated overnight with anti-TGF-β1 (1:100; R&D Systems, Minneapolis, MN, USA) and anti-COL IV (1:200; Biodesign International, Saco, ME, USA) antibodies in a humidified chamber at 4°C. The bound antibodies were localised with a peroxidase-conjugated secondary antibody using a Vector Impress Kit (Vector Laboratories, Burlingame, CA, USA) and 3,3-diaminobenzidine substrate solution. Finally, the sections were dehydrated in ethanol, cleared in xylene, mounted without counterstaining, and examined in a blinded manner using light microscopy (Olympus BX-50; Olympus Optical, Tokyo, Japan).
For quantification of proportional areas of staining, ~20 views (×400 magnification) were randomly located in the renal cortex and corticomedullary junction of each section and images were taken. The images were analysed to determine the density X-positive area/glomerular total area using a computer image analysis program (Scion Image Beta 4.0.2; Frederick, MD, USA).
Detection of apoptotic cells in formalin-fixed, paraffin-embedded tissue was performed by in situ TUNEL using an ApopTag In Situ Apoptosis Detection Kit (Chemicon-Millipore, Billerica, MA, USA). The TUNEL reaction was assessed in a whole glomerular biopsy under ×400 magnification.
Western blot analysis
Total protein from renal cortical tissues was extracted with Pro-Prep Protein Extraction Solution (Intron Biotechnology, Gyeonggi-Do, Korea), following the manufacturer’s instructions. Protein concentrations were determined using the Bradford reagent (Bio-Rad Laboratories, Hercules, CA, USA). Western blot analysis was performed to further confirm the responses using antibodies recognising specific epitopes. Proteins were separated by SDS-PAGE, transferred to nitrocellulose membranes, and detected with the following antibodies: anti-phosphorylated (phospho)-Thr172 AMPK (Cell Signaling Technology, Danvers, MA, USA), anti-total AMPK (Cell Signaling Technology), anti-peroxisome proliferator-activated receptor (PPAR) α (Abcam, Cambridge, UK), anti-PPARγ, anti-phospho-ACC (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-total ACC (Santa Cruz Biotechnology), anti-sterol regulatory element-binding protein 1 (SREBP-1; Santa Cruz Biotechnology), anti-B cell leukaemia/lymphoma 2 (BCL-2; Santa Cruz Biotechnology), anti-BCL-2-associated X protein (BAX; Santa Cruz Biotechnology), and anti-β-actin (Sigma-Aldrich, St Louis, MO, USA). After washing, the membranes were incubated with anti-mouse IgG or anti-rabbit IgG horseradish peroxidase (HRP)-linked secondary antibodies (Cell Signaling Technology) or a rabbit anti-goat IgG HRP-peroxidase secondary antibody (Sigma-Aldrich). The membranes were developed using an ECL Plus Detection Kit (Amersham International, Buckinghamshire, UK) to produce chemiluminescence signals, which were captured on X-ray films. Band densities were quantified with Quantity One software (Bio-Rad Laboratories).
Cell culture and small interfering RNA (siRNA) transfection
HGECs were purchased from Anigio-Proteomie (Boston, MA, USA) and subcultured in endo-growth media (Angio-Proteomie). After treatment with medium containing different concentrations of
d-glucose (5 mmol/L
d-glucose (low glucose), 40 mmol/L
d-glucose (high glucose), or 5 mmol/L
d-glucose plus 35 mmol/L mannitol (osmotic control)) and 1, 10, or 50 µg/mL anthocyanin for 6 h, western blotting was performed for phospho-Thr
172 AMPK, total AMPK, PPARα, PGC-1α, ERR-1α, PPARγ, phospho-ACC, total ACC, superoxide dismutase (SOD)-1, SOD-2, BCL-2, BAX, and β-actin with specific antibodies. To examine the effects of anthocyanins on other renal cells in high-glucose medium, we also used cultured NMS2 mesangial cells (see Additional file
1 for further details).
siRNAs targeted toward Ampkα1, Ampkα2, and Sirt1, and a non-specific scrambled siRNA (siRNA control) were complexed with a transfection reagent (Lipofectamine 2000; Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions. The sequences of the siRNAs were: α1-AMPK, 5′-GCAUAUGCUGCAGGUAGAU-3′; α2-AMPK, 5′-CGUCAUUGAUGAUGAGGCU-3′; SIRT1, 5′-CACCUGAGUUGGAUGAUAU-3′; and scrambled siRNA, 5′-CCUACGCCACCAAUUUCGU-3′ (Bioneer, Daejeon, Korea). At 24 h after transfection, HGECs were exposed to 40 mmol/L glucose and 50 µg/mL anthocyanins for 24 h.
Statistical analysis
Data are expressed as mean ± SD. Differences between groups were examined for statistical significance by ANOVA with the Bonferroni correction using SPSS version 19.0 (IBM Corp., Armonk, NY, USA). Values of p < 0.05 were considered statistically significant.
Discussion
The present results suggest that anthocyanin-rich SE ameliorated renal function and phenotype in the examined mouse model of type 2 diabetes, and was related to renal lipid accumulation, apoptotic renal cell injury, and oxidative stress through restoration of decreased AMPK activity and its target molecules in diabetic nephropathy. Downstream regulators of AMPK including ACC, SREBP-1, and PPAR were all translated into restoration of diabetes-induced lipotoxicity by anthocyanins in the mouse model and in vitro experiments. Using siRNAs for AMPKΑ1 and AMPKΑ2 in HGECs, we observed beneficial effects of anthocyanins on lipotoxicity in an AMPK-dependent manner.
The ultrasensitive energy sensor AMPK activates the catabolic pathway, inhibits the anabolic pathway, and restores energy homeostasis by phosphorylating multiple substrates in many metabolic syndrome-associated diseases [
13,
15,
22,
23]. Among the target molecules of AMPK, ACC is related to fatty acid oxidation and inhibition of fatty acid synthesis in many organs, such as the liver and kidneys. Phosphorylation of ACC1 at Ser
79 and ACC2 at Ser
218 by AMPK leads to inhibition of ACC activity and decreased malonyl-CoA, which is pivotal in controlling the rate of fatty acid β-oxidation. It has been reported that the reduced AMPK activity in diabetic nephropathy is linked to increased triglyceride accumulation because of reduced inhibitory phosphorylation of ACC [
15,
24].
SREBP, another target molecule controlled by AMPK, is a transcription factor regulating cellular lipogenesis and lipid homeostasis. Mammalian SREBP has three isoforms (SREBP-1a, SREBP-1c, and SREBP-2) and distinct but overlapping lipogenic transcriptional processes [
25]. The nutrient sensor AMPK negatively regulates SREBP to limit lipogenesis by directly phosphorylating SREBP. In our previous studies, we found that suppressed AMPK and the subsequent increase in SREBP activity in diabetic nephropathy lead to lipotoxicity in the pathogenesis of diabetic nephropathy [
26,
27]. In the present study, anthocyanin-mediated AMPK activation increased phosphorylation of ACC and decreased SREBP, which were correlated with decreased lipid contents in the diabetic kidney.
PPARs have noteworthy biological functions as sensors for fatty acid derivatives and control important metabolic pathways involved in lipid and energy metabolism [
28,
29]. Renal PPARα and PPARγ play important roles in modulating energy utilisation in the kidney through regulation of renal fatty acid oxidation. Stimulation of fatty acid oxidation through PPAR activation provides a potential mechanism by which the lipid content of tissues can be reduced to prevent lipid accumulation. In light of this, PPAR upregulation by anthocyanins seems to be another therapeutic target for improvement of lipotoxicity in diabetic nephropathy [
30‐
32].
Our in vitro study demonstrated that another critical mediator of the effects of PPARα on lipotoxicity is PGC-1α, a coactivator protein that interacts with several transcription factors, including ERR-1α, an important mediator in mitochondrial biogenesis. Notably, ERR-1α activated by PGC-1α induces genes that play roles in lipid transport and fatty acid oxidation [
33]. Meanwhile, SIRT1 is another molecule that responds to a variable nutrient status and is considered an AMPK partner [
34]. The results for the siRNA for
SIRT1 in the present in vitro study suggest that the effect of anthocyanins operates in a SIRT1-independent pathway.
The mechanism of lipotoxic injury in diabetic nephropathy could be explained by downregulation of fatty acid oxidation with altered expression related to energy sensor AMPK activity and its downstream lipogenic transcription factors, including PPAR and SREBP-1 [
2]. Although this may not entirely explain the complex pathogenesis of diabetic nephropathy, it has not been explored to its fullest potential. Recently, some investigators reported that altered renal lipid metabolism and accumulation in diabetic nephropathy were observed in humans [
1]. They clarified that heavy lipid deposition and increased intracellular lipid droplets were associated with dysregulation of lipid metabolism genes, and suggested that renal lipid metabolism may serve as a target for specific therapies of human diabetic nephropathy [
1,
26].
Anthocyanins are natural phytochemicals and the most oxidized flavonoids with a fully unsaturated C ring and a hydroxyl at position 3. They can be found as glycosylated forms of polyhydroxy and polymethoxy derivatives of 2-phenylbenzopyrylium, acylated or non-acylated with aliphatic acids, thus providing a rich source of potential therapeutic agents as antioxidants. Most experimental data have verified that anthocyanins serve as antioxidants and have anti-inflammatory properties in metabolic diseases. Takikawa et al. showed that anthocyanin-rich bilberry extract ameliorates hyperglycaemia and insulin sensitivity via AMPK activation in white adipose tissue, skeletal muscle, and the liver in type 2 diabetic mice [
17,
35]. These effects were accompanied by inactivation of ACC and upregulation of decreased PPARα expression. Valenti et al. [
4] proposed a possible mechanism for the beneficial effects of anthocyanins in non-alcoholic fatty liver disease and non-alcoholic steatohepatitis. They explained that the effect of anthocyanins on lipid metabolism is dependent on activation of the AMPK pathway and its target molecules, SREBP-1c and PPARα, in hepatocytes. In the present study, we found that diabetic nephropathy exhibited profound metabolic abnormalities, including increased intra-renal lipid accumulation in kidney tissue as well as circulating NEFA and triglyceride concentrations, while all levels were restored with anthocyanin treatment via AMPK activation and its target molecules such as ACC, SREBP-1, and PPAR. Taken together, these results demonstrated that anthocyanins improved decreased AMPK activity and the accompanying lipotoxicity in the examined mouse model of diabetes. Further studies are needed to verify the mechanisms.
It is uncertain whether the target of anthocyanins is solely AMPK or whether it overlaps with upstream molecules, such as adiponectin. Little is known about the roles of upstream molecules of AMPK such as adiponectin, LKB1, and CaMKKβ in the regulation of AMPK in the kidney [
36]. Recently, Liu et al. [
3] reported that anthocyanins enhance adiponectin secretion from adipose tissue and ameliorate diabetes-related endothelial dysfunction. Accordingly, we cannot exclude the possibility that anthocyanins restore decreased AMPK activity, in part, through increased activity of adiponectin [
3,
16]. Furthermore, we could not determine the specific contributions of the bioactive compounds in SE to the observed effects or the amounts of these compounds incorporated into the kidney. As cyanidin-3-O-glucoside was the most abundant anthocyanin in SE, we cautiously supposed that the main therapeutic effect of SE was mediated by cyanidin-3-O-glucoside based on previous studies [
9,
31], but future investigations are required to verify this notion.
We used glomerular endothelial cells under high-glucose conditions in an in vitro study. The results showed that the earliest indicator of kidney damage in diabetes is albuminuria, which is widely considered to reflect underlying endothelial dysfunction [
18]. The overall effect of AMPK in the endothelium has been proposed as the potential improvement of endothelial dysfunction, although the specific roles of endothelial AMPK in the kidney have not been shown [
36]. Considering the marked differences in the metabolic profiles of the individual cell types within the kidney, it could be important to identify the differences in AMPK expression and the effects of anthocyanins on other renal cell populations, such as podocytes. The results for cultured mesangial cells in high-glucose medium with anthocyanin treatments showed similar patterns in the present study (Additional file
2).
In conclusion, reduced AMPK activity was associated with lipotoxicity, which could be related to apoptosis and oxidative stress in diabetic nephropathy. Anthocyanins, as new and interesting AMPK activators, attenuated lipotoxicity through inhibition of ACC and SREBP-1 and restoration of PPARα and PPARγ, which mediated the removal of lipid accumulation in the kidney. This was accompanied by improvement of glomerular matrix accumulation and albuminuria. The data suggest that high glucose-induced HGEC damage translated into reduced AMPK activity and exacerbated oxidative stress and apoptotic cell damage in the diabetic setting. All changes were reversed with anthocyanin treatment.