European Journal of Nutrition

, Volume 52, Issue 6, pp 1607–1619

Aspalathin improves hyperglycemia and glucose intolerance in obese diabetic ob/ob mice

Authors

  • Myoung Jin Son
    • Department of Applied Life Science, Graduate School of Agricultural ScienceTokyo University of Agriculture and Technology
  • Miki Minakawa
    • Department of Applied Life Science, Graduate School of Agricultural ScienceTokyo University of Agriculture and Technology
  • Yutaka Miura
    • Division of Applied Biological Chemistry, Institute of AgricultureTokyo University of Agriculture and Technology
    • Division of Applied Biological Chemistry, Institute of AgricultureTokyo University of Agriculture and Technology
    • Graduate School of MedicineThe University of Tokyo
Original Contribution

DOI: 10.1007/s00394-012-0466-6

Cite this article as:
Son, M.J., Minakawa, M., Miura, Y. et al. Eur J Nutr (2013) 52: 1607. doi:10.1007/s00394-012-0466-6

Abstract

Purpose

Although several researches have demonstrated that rooibos extract has hypoglycemic effect, the role of aspalathin, a main polyphenol in the extract, remains unclear. Our aims were to find specific mechanisms for anti-diabetic action of aspalathin employing a rat skeletal muscle-derived cell line (L6 myocytes) and a rat-derived pancreatic β-cell line (RIN-5F cells) and to investigate its effect in type 2 diabetic model ob/ob mice.

Methods

We investigated in vitro the effect of aspalathin on the glucose metabolism through the studies on molecular mechanisms of glucose uptake using cultured L6 myotubes. We also measured the antioxidative ability of aspalathin against reactive oxygen species (ROS) generated by artificial advanced glycation end product (AGE) in RIN-5F cells. In vivo, ob/ob mice were fed 0.1 % aspalathin-containing diet for 5 weeks, and the effect of aspalathin on fasting blood glucose level, glucose intolerance, and hepatic gene expression was studied.

Results

Aspalathin dose dependently increased glucose uptake by L6 myotubes and promoted AMP-activated protein kinase (AMPK) phosphorylation. Aspalathin enhanced GLUT4 translocation to plasma membrane in L6 myoblasts and myotubes. In RIN-5F cells, aspalathin suppressed AGE-induced rises in ROS. In vivo, aspalathin significantly suppressed the increase in fasting blood glucose levels and improved glucose intolerance. Furthermore, aspalathin decreased expression of hepatic genes related to gluconeogenesis and lipogenesis.

Conclusions

Hypoglycemic effect of aspalathin is related to increased GLUT4 translocation to plasma membrane via AMPK activation. In addition, aspalathin reduces the gene expression of hepatic enzymes related to glucose production and lipogenesis. These results strongly suggest that aspalathin has anti-diabetic potential.

Keywords

AspalathinAMPKGLUT4 translocationHyperglycemiaAdvanced glycation end products (AGEs)

Abbreviations

PEPCK

Phosphoenolpyruvate carboxykinase

G6Pase

Glucose-6-phosphatase

GS

Glycogen synthase

LGP

Liver glycogen phosphorylase

ACC

Acetyl-CoA carboxylase

FAS

Fatty acid synthase

SCD

Stearoyl-CoA desaturase 1

AMPK

Adenosine monophosphate-activated protein kinase

GLUT4

Glucose transporter 4

Introduction

The total number of people with diabetes is increasing worldwide by population growth, aging, urbanization, and increasing physical inactivity and prevalence of obesity [1]. Under diabetic conditions, hyperglycemia and subsequent augmentation of reactive oxygen species (ROS) increase insulin resistance and deteriorate β-cell functions, which lead to the aggravation of type 2 diabetes [2]. The skeletal muscles, which account for the majority of insulin-mediated glucose uptake in post-prandial state, play an important role in maintaining glucose homeostasis [3]. In skeletal muscle, insulin increases glucose uptake through a signaling that leads to activation of phosphatidylinositol-3 kinase (PI3K) and Akt, resulting in increased translocation of glucose transporter 4 (GLUT4) to the plasma membrane [3]. Another GLUT4 translocation promoter is AMP-activated protein kinase (AMPK) [4]. Two distinct signals exist as AMPK activators; transient increase in intracellular Ca2+ concentration contributes to activation of AMPK, and AMPK-dependent pathway mediated by LKB1 leads to contraction-induced glucose uptake [5], which is related to enhanced GLUT4 genesis and its translocation to the cell membrane [6, 7]. Among the several sources of ROS production, advanced glycation end products (AGEs), the products of non-enzymatic glycation and oxidation of proteins and lipids, result in diabetes complications [8, 9]. Several studies have clearly demonstrated that AGEs and binding to their receptors link to various pathogenic causes in diabetic nephropathy, retinopathy, atherosclerotic disease, cardiomyopathy, and peripheral arterial disease [10].

Recently, numerous studies have focused on the prevention and treatment of various diseases including metabolic syndrome and its associated health risks using natural polyphenol compounds [11]. As one representative compound of the dihydrochalcone subgroup of polyphenols, aspalathin is only discovered in Rooibos (Aspalathus linearis), which grows exclusively in South Africa. Rooibos contains strong antioxidative substances; in particular, aspalathin has higher antioxidative abilities in comparison with other contents [12]. Despite many studies on rooibos and properties of its flavonoids, for instance, protective effects against DNA damage [13, 14], inflammation [15], cancer promotion [16], respiratory ailments [17], and hyperlipidemia [18], few studies have been attempted to elucidate their anti-diabetic effect. We have reported that aspalathin itself possesses a potential to suppress rises in the fasting blood glucose levels and to improve glucose intolerance in type 2 diabetic model db/db mice, which lack leptin receptor [19].

The present study investigated the effect of aspalathin on glucose metabolism through the study on molecular mechanisms for glucose uptake using cultured L6 myocytes. We also examined whether or not aspalathin would protect pancreatic β-cells from ROS stress in culture. To determine the effect of aspalathin on type 2 diabetes in vivo, we examined its effect on the fasting blood glucose levels and impaired glucose tolerance using obese diabetic ob/ob mice, another type 2 diabetic animal model that lacks leptin.

Materials and methods

Materials

Aspalathin was supplied by Tama Biochemical Co., Ltd., Tokyo, Japan. L6 myoblasts derived from a rat and RIN-5F cells derived from rat pancreatic β-cells were obtained from American Type Culture Collection, Manassas, VA; ATCC® numbers: CRL-1458 and CRL-2058, respectively. Dulbecco’s modified Eagle medium (DMEM) and RPMI 1640 medium were purchased from Nissui Pharmaceutical Co., Tokyo, Japan. Fetal bovine serum (FBS) was from JRH Biosciences, Lenexa, KS, USA. Streptomycin and penicillin G were from Nacalai Tesque, Inc., Kyoto, Japan. Bovine serum albumin (BSA, fatty acid free) and Triton X-100 were purchased from Sigma Chemical Co., St. Louis, MO, USA. Carboxymethyl cellulose sodium salt (CMC) was from Wako Pure Chemical Industries, Ltd., Osaka, Japan. Glucose, total cholesterol (T-Ch), and triglyceride (TG) assay kits (Glucose CII-Test Wako, Cholesterol E-Test Wako, and Triglyceride E-Test Wako, respectively) were from Wako Pure Chemical Industries, Ltd., Osaka, Japan. Serum lipid peroxide was estimated as thiobarbituric acid-reactive substances (TBARS) with a commercial kit (TBARS Assay Kit), ZeptoMetrix Corporation, Buffalo, NY, USA. Adiponectin assay kit was from Otsuka Pharmaceutical Co., Ltd, Tokyo, Japan. Tumor necrosis factor (TNF)-α assay kit was from Shibayagi Co., Ltd., Gunma, Japan. Compound C, a selective AMPK inhibitor, was from Wako Pure Chemical Industries, and 5-aminoimidazole-4-carboxamide 1-β-d-ribofuranoside (AICAR) was from Toronto Research Chemicals, Toronto, ON, Canada. The anti-phospho-AMPKα (Thr172) and anti-AMPK antibodies were obtained from Cell Signaling Technology, Inc., Beverly, MA, USA. Horseradish peroxidase-conjugated anti-rabbit IgG antibodies were from Amersham Biosciences, Buckinghamshire, UK. All other chemicals were of the best grade commercially available, unless otherwise noted. Plastic multiwell plates and tubes were obtained from Nunc A/S, Roskilde, Denmark, or Iwaki brand, Asahi Glass Co., Ltd., Tokyo, Japan.

Determination of glucose uptake by cultured L6 myotubes

Stock cultures of L6 myoblasts were maintained in DMEM supplemented with 10 % (v/v) FBS, streptomycin (100 μg/ml), and penicillin G (100 U/ml) (10 % FBS/DMEM) under an atmosphere of 5 % CO2/95 % humidified air at 37 °C as described previously [20]. Effect of aspalathin on glucose uptake was examined by the procedure described previously [21, 22] with slight modifications. Briefly, L6 myoblasts (5 × 104 cells/well) were subcultured into Nunc 24-place multiwell plates and grown for 11 days to form myotubes in 0.4 ml of 10 % FBS/DMEM. The medium was replaced every 3 days. The 11-day-old myotubes were kept for 2 h in Krebs-Henseleit buffer (pH 7.4) containing 0.1 % BSA, 10 mM Hepes, and 2 mM sodium pyruvate (KHH buffer). The myotubes were then cultured in KHH buffer containing 11 mM glucose in the absence or presence of aspalathin (0–100 μM) for another 4 h. Aspalathin was dissolved in KHH buffer. Glucose concentrations in KHH buffer were determined with a glucose assay kit, and the amounts of glucose consumed were calculated from the differences in glucose concentrations between before and after culture. This assay system with L6 myotubes is comparable to that with soleus muscles and [14C] 2-deoxyglucose as mentioned previously [23].

Western blot analysis

Western blot analysis was carried out as described previously [21]. In brief, L6 myotubes were solubilized in a lysis buffer (10 mM Tris–HCl pH 7.4, 150 mM NaCl, 1 % NP-40, 0.5 % sodium deoxycholate, 0.1 % sodium dodecyl sulfate (SDS), 0.5 mM dithiothreitol, 0.2 mg/ml Pefabloc SC, 1 mM Na3VO4) for 30 min at 4 °C. The lysates were centrifuged at 12,000×g for 20 min at 4 °C. The supernatant was electrophoresed on SDS-PAGE (10 %) and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were incubated in a blocking solution, 3 % BSA in Tris-buffered saline (TBS) containing 0.05 % Tween 20, for 2 h. After the incubation, the membranes were washed three times with TBS and incubated with primary antibodies overnight at 4 °C. The membranes were then washed three times with TBS containing 0.05 % Tween 20 (TBST) and incubated with horseradish peroxidase-conjugated anti-rabbit IgG antibodies for 2 h at room temperature. Immunoreactive bands were detected using SuperSignal West pico chemiluminescent substrate. The intensity of each band was analyzed with a lumino image analyzer (Model LAS-4000 Mini; Fujifilm, Tokyo, Japan) coupled with image analysis software (Multi Gauge Ver. 3.0; Fujifilm).

Transfection and immunocytochemistry of pFN21A (HaloTag®7)-glut4 in L6 myoblasts

Effect of aspalathin on GLUT4 translocation was examined by the procedure described previously using HaloTag-mock and HaloTag-glut4 expression vectors that were constructed in our laboratory [23]. To transfect the expression vector (pFN21A-rat glut4) and control vector (pFN21A-mock), L6 myoblasts (5 × 104) were cultured in a 24-well culture plate (Nunc) for glucose uptake assay or an 8-well chamber slide (Nunc) for immunocytochemistry. To support cell attachment and growth, an 8-well chamber slide was coated with collagen (Cellmatrix Type I–C, Nitta Gelatin Co., Ltd., Osaka, Japan). After 24 h, at an approximately 60 % confluency, they were transfected with pFN21A-glut4 or pFN21A-mock using FuGENE 6 (Roche Diagnostics, Indianapolis, IN, USA) according to the manufacturer’s instruction. The amounts of transfected DNA were 0.3 μg/well for a 24-well culture plate and 0.2 μg/well for an 8-well chamber slide. Cells were used for the glucose uptake assay at 48 h after transfection and for immunocytochemistry at 36 h after transfection. At 36 h after transfection, cells were washed twice with phosphate-buffered saline (PBS) and fixed with 3.7 % formaldehyde in PBS for 10 min at room temperature. After washing twice with PBS containing 0.05 % Tween 20 (PBS-T), cells were blocked with 3 % non-fat dried skim milk in PBS for 1 h and incubated with anti-HaloTag® rabbit polyclonal antibody (Promega KK, Tokyo, Japan) and anti-caveolin-3 goat polyclonal IgG (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) overnight at 4 °C. After washing three times with PBS-T, cells were incubated with Alexa Fluor 555-conjugated anti-rabbit IgG (Invitrogen, Carlsbad, CA, USA) and FITC-conjugated anti-goat IgG (Santa Cruz Biotechnology Inc.) antibodies for 1 h at room temperature. Finally, after washing three times with PBS-T, cells were mounted using Vectashield Mounting Medium with DAPI (Vector Laboratories, Burlingame, CA, USA) and examined with an Axiovert 200 M microscope (Carl Zeiss, Oberkochen, Germany).

Synthesis of advanced glycation end product (AGE)

Artificial AGE was generated by co-incubation of BSA with D-glyceraldehyde according to the method of Kume et al. [24]. BSA was incubated at 37 °C for 2 weeks in the presence of D-glyceraldehyde (designated as AGE). BSA alone was similarly incubated at 37 °C for 2 weeks under conditions without any carbohydrates [21] and employed as the control for AGE. This BSA was designated as CNT.

Measurement of intracellular reactive oxygen species (ROS) in RIN-5F cells

Intracellular ROS levels were measured as described previously [21]. Briefly, RIN-5F cells derived from rat pancreatic β-cells were maintained in RPMI 1640 supplemented with 10 % (v/v) FBS, streptomycin (100 μg/ml), and penicillin G (100 U/ml) (10 % FBS/RPMI 1640) under an atmosphere of 5 % CO2/95 % humidified air at 37 °C. The cells (5 × 105 cells/well) were cultured into Nunc 12-place multiwell plates. After being cultured for 72 h in 1 ml of 10 % FBS/RPMI 1640, the medium in each well was removed. Thereafter, RIN-5F cells received 1 ml of fresh medium (1 % FBS/RPMI 1640) without or with aspalathin and AGE for another 4 h. After the 4-h incubation, RIN-5F cells were incubated with DCFH-DA (final concentration of 25 μM) in 1 % FBS/RPMI 1640 at 37 °C for 20 min. At the end of the incubation, DCFH fluorescence of the cells from each well was measured at an emission wavelength of 530 nm and an excitation wavelength of 488 nm with a flow cytometer (EPICS ELITE EPS, Beckman-Coulter, Hialeah, FL, USA). The intensity of fluorescence reflects enhanced oxidative stress.

Effect of aspalathin on blood glucose levels in ob/ob mice

All animal experiments were conducted in accordance with guidelines established by the Animal Care and Use Committee of Tokyo University of Agriculture and Technology and were approved by this committee (ethics approval number: 23–22). To determine the effect of aspalathin on fasting blood glucose levels, ob/ob mice were used as a model of type 2 diabetes. Twenty-one male C57BL/6 J-ob/ob (ob/ob) and six male C57BL/6 J (normal) mice (5 weeks old) were obtained from Charles River Japan, Kanagawa, Japan. Animals were individually housed in stainless steel cages with wire bottoms in an air-conditioned room with a temperature of 22 ± 2 °C, a relative humidity of 60 ± 5 %, and an 8:00 a.m.–8:00 p.m. light cycle. All mice were maintained on a stock CE-2 pellet diet (CLEA Japan, Tokyo, Japan) for 3 days and thereafter a basal 20 % casein diet (20C) for 4 days. The composition of the 20C diet (AIN-93 formula) was described elsewhere [19, 21]. After preliminary feeding for 1 week, mice were deprived of their diet at 9:00 a.m. but allowed free access to water until blood collection from tail vein 3 h later. Blood (10 μl) was burst in water (40 μl), 20 % (w/v) trichloroacetic acid aqueous solution (50 μl) was added, and test tubes containing the mixture were kept in ice-cold water. The mixture was then centrifuged at 13,000×g and 4 °C for 5 min. The resultant supernatant (80 μl) was subjected to glucose determination with a commercial kit and a microplate reader (Appliskan, Thermo Fisher Scientific Inc., MA, USA) at 508 nm. Then, ob/ob mice (6 weeks old) were divided into two groups of similar fasting blood glucose levels and body weights (0 week). Diabetic mice of each of the two groups were given either the 20C as the control group or the 20C supplemented with 0.1 % aspalathin as the test group for 5 weeks. By adding 0.1 % aspalathin to the 20C, we intended to attain to an approximate dose of 100 mg/kg/day/mouse. Aspalathin was supplemented to the 20C at the expense of β-cornstarch. We prepared aspalathin-containing 20C diet every week and kept the diet in black vinyl bags at −30 °C in a freezer to prevent or minimize oxidation of aspalathin. One bag for 1 day was used every day and distributed into each glass pot for mice. After weighing the residual diets, they were replaced with fresh ones. The color of aspalathin-containing diet was unchanged. Likewise, C57BL/6 J (normal) mice were given the 20C as the normal group for 5 weeks. Water and each diet were available at all times except for the experiments to determine fasting blood glucose levels, which were carried out every week as mentioned above. After blood collection from tail vein for the final determination of fasting blood glucose levels at 5th week of feeding, whole blood was collected from all mice (11 weeks old) by cardiac puncture under pentobarbital anesthesia. Blood was left to clot at room temperature to obtain serum. Concentrations of serum lipids, glucose, insulin, adiponectin, and TNF-α were measured with commercial kits. Liver T-Ch and TG contents were also measured as described previously [25].

Intraperitoneal glucose tolerance test

Intraperitoneal glucose tolerance test (IPGTT) was performed after the determination of fasting blood glucose levels at 3rd week of feeding. Briefly, two groups of ob/ob mice (9.5 weeks old) were deprived of their diet at 8:00 p.m. but allowed free access to water. After fasted for 15 h, aspalathin (test group) was orally given at a dose of 10 mg/ml/100 g body weight. Similarly, 0.1 % CMC alone (1 ml/100 g body weight) was orally given to the control group as a vehicle. Two hours later, blood was collected from the tail vein of ob/ob mice (0 min). Immediately after blood collection, diabetic mice received an intraperitoneal injection of glucose (0.2 g/ml/100 g body weight). Blood samples were successively collected at appropriate time intervals (30, 60, 90, 120, and 180 min), and blood glucose levels were determined as mentioned above.

Quantitative RT-PCR analysis

Total RNA was isolated from the liver using modified AGPC method [26]. cDNA was synthesized using SuperScript™ III Reverse Transcriptase kit (Invitrogen) according to the manufacturer’s instructions. Real-time PCR was performed with FastStart SYBR green Master Mix (Roche Diagnostics, Mannheim, Germany) in Thermal Cycler Dice® Real Time System II (TAKARA Bio, Inc., Tokyo, Japan). The PCR reaction was performed in duplicate. The relative expression levels of the target genes to the expression level of the endogenous reference gene β-actin were calculated using the delta cycle threshold (Ct) method. The primer sequences are listed in supplementary Table 1.

Statistical analyses

All data are presented as means ± standard errors of means (SEM). Multigroup comparisons were carried out by one-way analysis of variance followed by Tukey–Kramer multiple comparisons test, and differences between two group means were compared by Student’s t-test. Values of P < 0.05 were considered statistically significant.

Results

Effect of aspalathin on glucose uptake in cultured L6 myotubes

Aspalathin dose dependently and significantly promoted glucose uptake in L6 myotubes at concentrations up to 100 μM under the condition of insulin absence (Fig. 1aA). Compound C, an ATP-competitive inhibitor of AMPK [27], significantly canceled the aspalathin-induced increase in glucose uptake at a concentration of 10 μM, where compound C alone exerted no influence on glucose uptake (Fig. 1aB).
https://static-content.springer.com/image/art%3A10.1007%2Fs00394-012-0466-6/MediaObjects/394_2012_466_Fig1_HTML.gif
Fig. 1

Effect of aspalathin treatment on glucose uptake (a), phosphorylation of AMPK and Akt (b), and GLUT4 localization (c) in cultured L6 myocytes. a For glucose uptake assay, L6 myotubes were preincubated in 24-place multiwell plates in Krebs-Henseleit-Hepes buffer (KHH buffer) without glucose for 2 h. They were then incubated in KHH buffer containing 11 mM glucose without or with aspalathin (0, 25, 50, 100 μM) (aA) and Compound C (10 μM) (aB) for 4 h, and the glucose uptake was measured using a Glucose CII-Test Kit. Each value represents the mean ± SEM for 6 wells. Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test. b L6 myotubes were preincubated in KHH buffer without glucose for 2 h. They were then incubated in KHH buffer for 4 h. Concentrations of aspalathin and AICAR were 50 μΜ and 1 mM, respectively. Total cell lysates were analyzed by immunoblotting with anti-phospho-AMPK, anti-AMPK (bA), anti-phospho-Akt, and anti-Akt antibodies (bB). c Glucose uptake for 4 h was measured without or with aspalathin (0, 50 μM) in L6 myoblasts 48 h after transfection of pFN21A-mock vector or pFN21A-glut4 vector (cA). Each value and bar represents the mean ± SEM for 4 wells. Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test. To visualize translocation of GLUT4 and caveolin-3 (cB), 36 h after transfection, L6 myoblasts transiently expressing proteins of HaloTag (mock ASP− and mock ASP+) or Halo-GLUT4 (glut4 ASP− and glut4 ASP+) were processed for immunocytochemistry using anti-HaloTag antibody and anti-caveolin-3 antibody. Cellular localization of HaloTag and HaloTag-GLUT4 proteins is shown in red fluorescence (left) and that of caveolin-3 is shown in green fluorescence (center). Merged image is also shown (right). L6 myoblasts (mock ASP+ and GLUT4 ASP+) were exposed to aspalathin (50 μM) for 30 min

Effect of aspalathin on phosphorylation of AMPK and Akt in cultured L6 myotubes

L6 myotubes were treated with 50 μM aspalathin for 0–240 min and 1 mM AICAR, an AMPK activator, for 120 min. Aspalathin stimulated the phosphorylation of AMPK (the ratio of phosphorylated AMPK to total AMPK, p-AMPK/AMPK) from 30 min after treatment, which peaked after 120 min (Fig. 1bA). However, aspalathin did not affect the phosphorylation of Akt (p-Akt/Akt) in the absence of insulin (Fig. 1bB).

Bioimaging analysis of aspalathin effect on GLUT4 translocation to plasma membrane

Glucose uptake for 4 h was examined in L6 myoblasts 48 h after transfection of pF21A-mock or pF21A-glut4 vector (Fig. 1cA). Glucose uptake was significantly higher in pF21A-glut4 vector-transfected cells than in pF21A-mock vector-transfected cells (mock ASP− vs. glut4 ASP−), suggesting that Halo-GLUT4 derived from transfected pF21A-glut4 could operate in glucose uptake. Aspalathin (50 μM) significantly promoted glucose uptake in L6 myoblasts transfected with both the pF21A-mock and pF21A-glut4 vectors (mock ASP− vs. mock ASP+, glut4 ASP− vs. glut4 ASP+), indicating that aspalathin could promote glucose uptake even under the condition of GLUT4 overexpression. Caveolin-3 is involved in spatial and temporal regulation of GLUT4 translocation to plasma membrane and hence glucose uptake in skeletal muscle cells [28]. Figure 1cB (left) shows cellular localization of HaloTag protein alone (mock ASP− and mock ASP+) and Halo-GLUT4 protein (glut4 ASP− and glut4 ASP+), Fig. 1cB (center) shows cellular localization of caveolin-3, and Fig. 1cB (right) shows their merging. In the cells transfected with pFN21A-mock vector, HaloTag protein and caveolin-3 were expressed in the whole area but did not co-localize (mock ASP− and mock ASP+). Likewise, HaloTag protein and caveolin-3 were similarly but more strongly expressed in the cells transfected with pFN21A-glut4 vector than in those transfected with pFN21A-mock vector, and co-localization of two proteins was recognized. Aspalathin treatment for 30 min strengthened their co-localization in the plasma membrane compartment as shown by yellowish color (glut4 ASP− vs. glut4 ASP+).

Changes of intracellular ROS levels in cultured RIN-5F cells by aspalathin treatment

Aspalathin dose dependently and significantly reduced the basal ROS levels of RIN-5F cells (Fig. 2a) as compared with control cells (0 μM aspalathin). As seen in Fig. 2b, AGE was found to significantly enhance the production of ROS in RIN-5F cells. However, aspalathin significantly suppressed the AGE-induced rise in the ROS level.
https://static-content.springer.com/image/art%3A10.1007%2Fs00394-012-0466-6/MediaObjects/394_2012_466_Fig2_HTML.gif
Fig. 2

Effect of aspalathin on ROS levels in RIN-5F cells. Intracellular ROS levels in RIN-5F cells were measured with a DCFH-DA fluorescent probe and flow cytometer. a Effect of aspalathin on basal ROS levels in RIN-5F cells after 3 h treatment of aspalathin. b Effect of aspalathin on artificial AGE-induced oxidative stress for 4-h treatment with AGE and BSA (CNT). Each value represents the mean ± SEM of 3 wells. Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test

Effect of aspalathin on fasting blood glucose levels and glucose intolerance in ob/ob mice

Based on the above-mentioned in vitro experiments, we next examined effect of aspalathin under diabetic condition using leptin-deficient type 2 diabetic model C57BL/6 J-ob/ob mice. Initial body weights, body weight gains, and food intakes in obese diabetic control mice were significantly higher than those in normal mice, despite the same age and ingesting the same diet (Table 1). Aspalathin commenced to suppress significantly the rises in the fasting blood glucose levels in diabetic ob/ob mice after treatment with the polyphenol for 2 weeks and thereafter up to 5 weeks (Fig. 3aA). There were no differences in body weight gain and food intake between aspalathin-treated and control ob/ob mice during the feeding period (Table 1), indicating that the suppressive effect of aspalathin on blood glucose levels is not due to reduced food intakes but is due to its pharmacological action. The serum glucose level at the end of experimental period was significantly higher in diabetic mice than in normal mice. Aspalathin significantly suppressed this rise without affecting the serum insulin levels (Table 1). In intraperitoneal glucose tolerance test (IPGTT), aspalathin significantly suppressed the rises in the blood glucose levels 60, 90, and 120 min after glucose administration. Area under the concentration–time curve (AUC) of aspalathin group significantly decreased as compared with that in the control group (Fig. 3aB).
Table 1

Effect of aspalathin feeding on the food intake and body weight, serum and liver cholesterol, and serum insulin and glucose levels in ob/ob mice

Measurement

NOR

CNT

ASP

Initial body weight (g)

20.0 ± 0.4a

31.5 ± 0.5b

31.6 ± 0.5b

Final body weight (g)

26.0 ± 0.5a

41.4 ± 1.1b

41.7 ± 1.0b

Weight gain (g/5 weeks)

6.0 ± 0.2a

9.9 ± 1.0b

10.0 ± 0.7b

Food intake (g/5 weeks)

125.4 ± 2.7a

158.0 ± 3.8b

153.8 ±2.0b

Serum cholesterol level (mg/dl)

135.3 ± 8.1a

307.9 ± 16.6b

291.6 ± 16.4b

Liver cholesterol level (mg/g liver)

2.10 ± 0.28a

6.31 ± 1.26b

5.60 ± 1.22b

Serum insulin level (ng/ml)

4.83 ± 0.01

5.82 ± 0.21

6.93 ± 0.32

Serum glucose level (mg/dl)

83.1 ± 10.6a

366.7 ± 18.2b

295.6 ± 20.8c

Overall food intake, average body weight, and body weight gain were measured once a day for 5 weeks. The levels of serum total cholesterol, insulin and glucose, and liver total cholesterol were measured at 5 weeks of aspalathin feeding. The average of aspalathin ingested for 5 weeks was 106.4 ± 1.3 mg/kg body weight/day/mouse. Each value represents the mean ± SEM (NOR: n = 6, CNT: n = 9, ASP: n = 11). Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test

https://static-content.springer.com/image/art%3A10.1007%2Fs00394-012-0466-6/MediaObjects/394_2012_466_Fig3_HTML.gif
Fig. 3

Effect of aspalathin feeding on fasting blood glucose level and glucose intolerance (a), serum TBARS and serum and liver triglyceride levels (b), and serum adiponectin and TNF-α levels (c) in ob/ob mice. a During 5 weeks of aspalathin feeding, fasting blood glucose level was measured once a week after mice were fasted for 3 h before blood collection. Each value represents the mean ± SEM (NOR: n = 6, CNT: n = 9, ASP: n = 11). Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test (aA). After fasted for 15 h, aspalathin (test group) was given orally at a dose of 10 mg/ml CMC/100 g body weight (n = 11). CMC solution (1 ml/100 g body weight) alone was orally given to the diabetic control group (n = 9). Two hours later, blood was collected from the tail vein of ob/ob mice (0 min). Immediately after blood collection, diabetic mice received an intraperitoneal injection of glucose (0.2 g/ml/100 g body weight). Blood samples were successively collected at appropriate time intervals (30, 60, 90, 120, and 180 min), and blood glucose levels were determined. Asterisk means statistical significance (P < 0.05) by a two-tailed Student’s t-test as compared with CNT group (aB). b The levels of serum thiobarbituric acid-reactive substances (TBARS) and serum and liver triglyceride (TG) were measured at 5 weeks of aspalathin feeding (bA, bB, bC). Mice were deprived of their diets 4 h before the measurements. TBARS are substances formed as a byproduct of lipid peroxidation. Each value represents the mean ± SEM (NOR: n = 6, CNT: n = 9, ASP: n = 11). Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test. c The levels of serum adiponectin (cA) and TNF-α (cB) were measured after 5 weeks of aspalathin feeding. Each value represents the mean ± SEM (NOR: n = 6, CNT: n = 9, ASP: n = 11). Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test

Effect of aspalathin on serum and liver lipid levels in ob/ob mice

To investigate whether or not aspalathin would affect lipid metabolism, we investigated the levels of lipids in the serum and liver of normal and ob/ob mice (Fig. 3b). The levels of liver triglyceride (TG) as well as serum thiobarbituric acid-reactive substances (TBARS) and TG in control ob/ob mice were significantly higher than those in normal mice, but aspalathin significantly suppressed these rises (Fig. 3bA, bB, and bC). As shown in Table 1, the serum and liver total cholesterol (T-Ch) levels in control ob/ob mice were also significantly higher than those in normal mice. Although not statistically significant, aspalathin tended to suppress these rises (Table 1).

Effect of aspalathin on serum adiponectin and TNF-α levels in ob/ob mice

Because adiponectin and TNF-α are known for its important role in insulin sensitivity, we measured the levels of serum adiponectin and TNF-α in ob/ob mice (Fig. 3c). The serum adiponectin level in the diabetic control ob/ob mice significantly decreased as compared with that in the normal mice, whereas aspalathin treatment significantly prevented this decrease in the serum adiponectin level (Fig. 3cA). In contrast, the serum TNF-α level in the diabetic control ob/ob mice significantly increased as compared with that in the normal mice, and aspalathin treatment tended to suppress this increase in the serum level of TNF-α (Fig. 3cB).

Effect of aspalathin on the gene expression of enzymes related to gluconeogenesis, glycogenesis, glycogenolysis, and lipogenesis in the liver of ob/ob mice

We next examined the gene expression levels of the key enzymes related to glucose metabolism, phosphoenolpyruvate carboxykinase (PEPCK), glucose-6-phosphatase (G6Pase), glycogen synthase (GS), liver glycogen phosphorylase (LGP), and those related to lipogenesis, acetyl-coenzyme A carboxylase 1 (ACC), fatty acid synthase (FAS), and stearoyl-coenzyme A desaturase 1 (SCD), in the liver of ob/ob mice (Fig. 4). The expression of both PEPCK and G6Pase was increased in the control mice as compared with the normal ones, whereas it was decreased by aspalathin feeding as compared with the control mice (Fig. 4a). Moreover, the expression of glycogen synthase (GS) was lower in the control mice than in the normal mice, but the decrease was returned to the normal level by aspalathin treatment (Fig. 4b). In contrast, glycogen phosphorylase (LGP) was increased in the control mice as compared with the normal mice, but the increase was significantly suppressed upon aspalathin feeding in ob/ob mice (Fig. 4b). The expression of lipogenic genes such as ACC, FAS, and SCD was significantly higher in the control mice than in the normal ones. However, aspalathin significantly (SCD) suppressed or tended to suppress (ACC, FAS) the increase in the expression of such lipogenic genes as compared with the control (Fig. 4c).
https://static-content.springer.com/image/art%3A10.1007%2Fs00394-012-0466-6/MediaObjects/394_2012_466_Fig4_HTML.gif
Fig. 4

Effect of aspalathin feeding on the expression of glucose and lipid metabolism-related genes in the liver. Real-time PCR was used to measure relative mRNA levels of PEPCK and G6Pase in gluconeogenesis (a), GS and LGP in glycogenesis and glycogenolysis (b), and ACC, FAS, and SCD in fatty acid synthesis (c). (NOR: n = 6, CNT: n = 9, ASP: n = 11). Each value represents the mean ± SEM. Values not sharing a common letter are significantly different at P < 0.05 by Tukey–Kramer multiple comparisons test. Abbreviations: PEPCK phosphoenolpyruvate carboxykinase; G6Pase glucose 6-phosphatase, GS glycogen synthase, LGP liver glycogen phosphorylase, ACC acetyl-coenzyme A carboxylase 1, FAS fatty acid synthase, SCD stearoyl-coenzyme A desaturase 1

Discussion

In our previous study, aspalathin was found to suppress the increases in fasting blood glucose levels and improve the glucose intolerance in type 2 diabetic model db/db mice [19]. The present study was attempted to elucidate the mechanisms for anti-diabetic effect of aspalathin at the molecular and cellular levels employing L6 myocytes and RIN-5F pancreatic β-cells. In addition, the modes of action of aspalathin were studied at the whole-body level using other type 2 obese diabetic model ob/ob mice.

AMP-activated protein kinase is suggested to play a role in improving insulin insensitivity by direct stimulation of glucose uptake in muscle independently of insulin signaling pathway [29, 30]. In the present study, we found that aspalathin did not affect Akt activation in vitro, which is included in insulin signaling pathway. However, aspalathin activated AMPK signaling and promoted endogenous GLUT4 translocation to plasma membrane in L6 myocytes, which was visually demonstrated by immunocytochemical bioimaging method; this stimulatory tendency on AMPK was also observed in the muscle of ob/ob mice treated with aspalathin (data not shown). These results indicate that regulation of GLUT4 translocation and hence glucose uptake by aspalathin is due to activation of AMPK but not due to Akt activation.

Hyperglycemia encourages the accumulation of advanced glycation end products (AGEs). It is well recognized that AGEs produce ROS through their receptor and play a significant role in the development of diabetic complications [810]. The serum level of AGEs is reportedly associated with insulin resistance even in non-obese, non-diabetic subjects [31]. Thus, reducing AGEs-induced ROS as well as the circulating AGEs appears to protect against the development of diabetes and its complications. We examined whether or not aspalathin can protect β-cells from AGEs-induced ROS level using artificial AGE. Aspalathin was found to reduce the artificial AGE-induced increase in the intracellular ROS level in RIN-5F cells. Aspalathin also suppressed the rise in the serum TBARS level in ob/ob mice. From these findings, aspalathin is demonstrated to show its antioxidative function both in vitro and in vivo. It has been suggested that AGEs can impair insulin sensitivity in adipocytes and skeletal muscles [3234]. The limitation of our study is that we only have demonstrated that aspalathin has AGE-induced ROS scavenging activity. Further intensive studies such as measurement of the circulating level of AGEs are required to verify that aspalathin increases insulin sensitivity and improves diabetic complications caused by AGEs.

Adiponectin is the most widely studied adipokine in the research of metabolic disease, and experimental and clinical studies have shown that adiponectin improves insulin sensitivity [35]. Adiponectin helps to increase glucose uptake and GLUT4 translocation in rat skeletal muscle cells [36]. AMPK has been reported to be activated by adiponectin in skeletal muscle, where it stimulates glucose uptake and fatty acid oxidation, and inhibits the expression of gluconeogenic genes in the liver [4]. In this study, serum adiponectin levels were significantly increased by dietary feeding of aspalathin. This result suggests that aspalathin may, at least partly, play a role in the modulation of insulin sensitivity dependent on adiponectin.

Hyperglycemia is particularly related to hepatic glucose synthesis. Excessive hepatic gluconeogenesis is responsible for the high blood glucose level, leads to insulin resistance and elevated glucagon levels. Thus, suppression of gluconeogenesis in the liver is suggested as an intervention for type 2 diabetes. In the present experiment, we tested the expression of the genes for the hepatic gluconeogenic enzymes PEPCK and G6Pase in aspalathin-fed ob/ob mice. Aspalathin suppressed the gene expression of PEPCK and G6Pase, increased glycogen synthesis by inducing the gene expression of GS and suppressing LGP. Insulin serves as the most important role that inhibits gluconeogenesis by suppressing both of PEPCK and G6Pase through insulin signaling pathway [37]. In contrast, insulin-independent pathway is related to inhibition of gluconeogenesis. Treatment of H4 rat hepatoma cells with AMPK inhibitor, AICAR, was found to mimic the effect of insulin on G6Pase and PEPCK gene expression, indicating that the activation of AMPK is able to suppress G6Pase and PEPCK gene expression independently of insulin [38]. Another AMPK activator, metformin, stimulated CREB binding protein (CBP) phosphorylation in a hepatocyte cell line, even in the presence of the PI3K inhibitor LY294002, and inhibited gluconeogenesis [39]. Furthermore, there has been a report that adiponectin suppresses PEPCK and G6Pase gene expression by the activation of AMPK [40]. In our study, the results of increased AMPK activation in muscle and increased level of serum adiponectin provide the possibility that aspalathin can suppress hepatic glucose production independently of insulin action. Govorko et al. [41] reported that one of representative dihydrochalcones, 2′,4′-dihydroxy-4-methoxydihydrochalcone from Artemisia dracunculus L. (A. dracunculus), decreased PEPCK overexpression in H4IIE hepatoma cell line through insulin-independent AMPK pathway, that is, the effect of the dihydrochalcone is not regulated by PI3K, but it is dependent on activation of the AMPK pathway. As mentioned above, our results on the effect of aspalathin on GLUT4 translocation through AMPK pathway without involvement of PI3K in L6 myocytes are in good agreement with these previous findings on PEPCK expression in H4IIE hepatoma cells.

To determine the effect of aspalathin on lipid metabolism, the concentration of T-Ch and TG was measured in the serum and liver of ob/ob mice from each experimental group. TG concentrations in both the serum and liver of aspalathin-fed mice were significantly reduced as compared with those in diabetic control mice. Reduction in TG levels might be associated to genes and transcriptional regulators of hepatic lipogenesis. In our experiment, the expression of ACC, FAS, and SCD mRNAs in the livers of aspalathin-fed mice decreased as compared with that of diabetic control mice. These results suggest that the reduction of glucose and TG levels by aspalathin is related to decreases in glucogenesis, glycogenolysis, and lipogenesis as well as an increase in glycogenesis.

The improvement of glucose metabolism by aspalathin is likely to be associated with its antioxidative function. TBARS is widely used as an indicator of oxidative stress in animals. In our result of serum TBARS measurement in vivo, its level was suppressed by aspalathin administration. Excessive ROS level can disturb physiological functions of cellular macromolecules, such as DNA, proteins, or lipids. Therefore, ROS may be thought to play an important role in the study of diabetic complications. Progressive β-cell failure gives rise to impaired insulin secretion, insulin resistance, and causes type 2 diabetes. Several researches on protecting β-cells from oxidative stress by antioxidative drugs have been performed in β-cell lines and isolated pancreatic islets from rodents [42, 43]. In our experiments in vitro, aspalathin was demonstrated to decrease intracellular ROS level in a β-cell line as mentioned above. These results support the previous reports by other researchers that aspalathin possesses strong antioxidative activity [12, 44].

Considering the absorption and metabolism of aspalathin, human study results showed that aspalathin is metabolized in the body through the absorption in intestine, and its metabolites were found in urine [45]. According to Reagan-Shaw et al. [46], effective doses in humans are suggested to be one-eighth to one-twelfth of those in mice. Although the necessary amounts of aspalathin to apply its hypoglycemic effect in human body have not been established yet, our research provides the importance and basis of its glucose-lowering effect for further researches in humans.

In summary, aspalathin promoted glucose uptake through GLUT4 translocation to plasma membrane via AMPK activation in L6 myocytes. Aspalathin scavenged artificial AGE-induced ROS in RIN-5F cells. Aspalathin suppressed the increases in fasting blood glucose levels, improved the impaired glucose tolerance, and reduced hypertriglyceridemia and serum TBARS level in ob/ob mice. Aspalathin prevented the decrease in the serum adiponectin level. Aspalathin suppressed the gene expression of enzymes related to gluconeogenesis, glycogenolysis, and lipogenesis and reversed the decrease in gene expression of the enzyme related to glycogenesis in the liver of ob/ob mice, contributing at least partly to the reduction in blood glucose and serum TG levels. These results strongly suggest that aspalathin has potential anti-diabetic value.

Acknowledgments

This research was supported in part by the Japan Society for the Promotion of Science and in part by Rooibos Marketing Ltd., Japan.

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

394_2012_466_MOESM1_ESM.ppt (151 kb)
Supplementary material 1 (PPT 151 kb)

Copyright information

© Springer-Verlag Berlin Heidelberg 2012