Original Paper
Cellular Physiology and Biochemistr Biochemistryy
Cell Physiol Biochem 2007;20:925-934
Accepted: July 04, 2007
The Flavonoid Silibinin Decreases Glucose-6Phosphate Hydrolysis in Perifused Rat Hepatocytes by an Inhibitory Effect on Glucose6-Phosphatase Bruno Guigas1*, Roula Naboulsi2*, Gloria R. Villanueva2, Nellie Taleux1,3, José M. Lopez-Novoa2, Xavier M. Leverve3 and MohamadYehia El-Mir2 1 Université catholique de Louvain, Institute of Cellular Pathology, Hormone and Metabolic Research Unit, Brussels, Belgium, 2Departamento de Fisiologia y Farmacologia, Facultad de Farmacia, Universidad de Salamanca, Campus Miguel de Unamuno, Salamanca, Spain, 3INSERM-E0221 “Bioénergétique Fondamentale et Appliquée”, Université J. Fourier, Grenoble, France, *The first two authors contributed equally to this study
Key Words Silibinin • Flavonoid • Hepatocytes • Gluconeogenesis • Glucose-6-phosphatase • Type 2 diabetes
Abstract Background/Aims: The flavonoid silibinin has been reported to be beneficial in several hepatic disorders. Recent evidence also suggests that silibinin could be beneficial in the treatment of type 2 diabetes, owing to its anti-hyperglycemic properties. However, the mechanism(s) underlying these metabolic effects remains unknown. Methods: The effects of silibinin on liver gluconeogenesis were studied by titrating hepatocytes from starved rats with sub-saturating concentrations of various exogenous substrates in a perifusion system. Hepatocytes from fed rats were also used to investigate glycogenolysis from endogenous glycogen. The effect of silibinin on glucose-6phosphatase kinetics was determined in intact and permeabilized rat liver microsomes. Results: Silibinin induced a dose-dependent inhibition of gluconeogen-
Fax +41 61 306 12 34 E-Mail
[email protected] www.karger.com
© 2007 S. Karger AG, Basel 1015-8987/07/0206-0925$23.50/0 Accessible online at: www.karger.com/cpb
esis associated with a potent decrease in glucose-6phosphate hydrolysis. This effect was demonstrated whatever the gluconeogenic substrates used, i.e. dihydroxyacetone, lactate/pyruvate, glycerol and fructose. In addition, silibinin decreased the glucagoninduced stimulation of both gluconeogenesis and glycogenolysis, this being associated with a reduction of glucose-6-phosphate hydrolysis. Silibinin inhibits glucose-6-phosphatase in rat liver microsomes in a concentration-dependent manner that could explain the decrease in glucose-6-phosphate hydrolysis seen in intact cells. Conclusion: The inhibitory effect of silibinin on both hepatic glucose-6-phosphatase and gluconeogenesis suggests that its use may be interesting in treatment of type 2 diabetes. Copyright © 2007 S. Karger AG, Basel
Introduction Flavonoids are the most abundant polyphenolic compounds present in the human diet, with a daily consumption estimated at 1 g [1]. Thousands flavonoids have been identified in fruits, vegetables and plant-derived Dr. M. Yehia El-Mir Universidad de Salamanca Facultad de Farmacia, Departamento de Fisiologia y Farmacologia Campus Miguel de Unamuno, E.D. S-11, 37007-Salamanca (Spain) Tel. +34 -923-294472, Fax +34 -923-294669, E-Mail
[email protected]
925
beverages, such as tea and red wine, and have been also associated with a wide variety of biochemical and pharmacological properties beneficial to health. Silymarin, one of the oldest traditional herbal medicines, is a lipophilic extract from the milk thistle Silybum marianum which is predominantly composed of four flavonolignan isomers: silibinin, isosilibinin, silidianin and silichristin. Silibinin constitutes 50-70% of the silymarin mixture and has been identified as the major biologically active component [2]. While for centuries extracts of milk thistle have been empirically used to combat multiple organ disorders, a growing body of evidence shows that silibinin could be clinically efficient in several hepatic diseases, such as cirrhosis or acute and chronic viral hepatitis [2-4]. The most remarkable use of silibinin (Legalon ®) remains in the treatment of acute Amanita phalloides mushroom poisoning where its hepatoprotective effects after accidental ingestion have been demonstrated repeatedly in humans [4]. In addition, silibinin has been reported to protect the liver against toxicity induced by a wide range of chemical and environmental agents. A large part of these effects are attributed to its antioxidant properties (see [4] for review). Silibinin also received attention because of other effects not directly related to its antioxidant properties, i.e. the flavonoid exerting potential anti-carcinogenic [5, 6], anti-inflammatory [7] or neuroprotective [8] activities. Finally, silibinin has been recently proposed to be beneficial in type 2 diabetes since treatment with an oral formulation induced a significant decrease in both fasting and mean daily glycemia [9-11] and in HbA1c, triglyceride and total cholesterol levels [11] in patients. Furthermore, aqueous extract of Silybum marianum exhibits potent hypoglycaemic and antihyperglycaemic activities in both streptozotocin- [12] and alloxan-induced diabetic rats [13]. The underlying mechanism could be related to an inhibition of hepatic glucose production but remains unknown to date. Glucose-6-phosphatase (G6Pase) is associated with the endoplasmic reticulum and catalyses the reaction of glucose-6-phosphate (Glc-6-P) hydrolysis [14, 15]. The particular location of this enzyme at the crossroads of multiple pathways generating Glc-6-P confers a central role on G6Pase, notably in the regulation of hepatic glucose production [16]. While regulation of G6Pase gene expression is well characterised [17], a post-transcriptional regulation of the enzyme is still a matter of debate. Indeed, before the first demonstration of a short-term effect on G6Pase kinetics [18], it was commonly thought that enzyme activity was exclusively modulated by intracellular Glc-6-P levels. Other acute changes in G6Pase activity
were then reported, mediated by either insulin, metabolites such as proline or α-ketoglutarate, or fatty acids (see [19] for review). In addition, we have recently demonstrated that acute stimulation of gluconeogenesis by glucagon could be also mediated, at least in part, by a temperature-sensitive activation of Glc-6-P hydrolysis [20]. In the present study we investigated the metabolic effects of silibinin in rat hepatocytes perifused with different carbohydrates, and more particularly on gluconeogenic pathway by focusing on Glc-6-P hydrolysis at the level of G6Pase.
926
Guigas/Naboulsi/Villanueva/Taleux/Lopez-Novoa/Levere/El-Mir
Cell Physiol Biochem 2007;20:925-934
Materials and Methods Hepatocytes isolation Liver cells were prepared between 8.00 and 10.00 am from fed or 24 h-starved male Wistar rats (200-300 g). Animals were anaesthetized with intraperitoneal injection of sodium thiopental (125 mg.kg-1) and hepatocytes were isolated in the presence of collagenase according to the method of Berry and Friend [21] modified by Groen et al. [22]. All procedures were performed in accordance with the principles and guidelines established by the European Convention for the Protection of Laboratory Animals. Perifusion of hepatocytes Hepatocytes from 24 h-starved rats (200-250 mg dry cells) were perifused at 37°C with Krebs-bicarbonate-calcium (120 mM NaCl, 4.8 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 24 mM NaHCO3, 1.3 mM CaCl2, pH 7.4) saturated with O2/CO2 (19:1) at a flow rate of 5 ml.min-1, according to the method of Van der Meer and Tager [23] as modified by Groen et al. [22]. After 35 min, when a steady state has been reached, the cells were titrated with increasing substrate concentrations (0.15, 0.3, 0.6, 1.2, 2.4 and 4.8 mM) and in the presence of 25 or 100 µM silibinin, 10-7 M glucagon or vehicle, alone or in combination. At the end of each steady state of 15 min, extracellular medium (“perifusate”) and cell samples were collected at the exit or inside the perifusion chamber, respectively. For the time-course experiments, hepatocytes from fed rats (120 mg dry cells) were perifused as described above without addition of exogenous substrate and in the presence of 100 µM silibinin or vehicle. 107 M glucagon was added in both chambers after 25 min and perifusate was collected at the indicated time. Cell samples were also removed from the chamber for determination of intracellular Glc-6-P level at the indicated time. Determination of metabolic fluxes and intracellular intermediates and adenine nucleotides concentrations The proteins in perifusate were denatured by heating in a warm water bath (80°C for 10 min) and glucose, lactate and pyruvate concentrations were then measured by spectrophotometric assays [20]. The net rates of glucose (JGlucose) or lac-
tate-plus-pyruvate production (JL+P) were calculated from the total cell content of the perifusion chamber, the flow rate through the system and the concentration of each metabolite in the perifusate, and were expressed as µmoles.min-1.g dry cells-1. The total metabolic flux from exogenous substrate (JSubstrate) was expressed as three-carbon equivalents and calculated from the sum of 2xJGlucose and JL+P. Once removed from the chamber, the cell sample was centrifuged through a layer of silicone oil (Rhodorsil 640 V 100) into HClO4 (10% mass/vol) to separate intracellular content from extracellular medium. The acid-soluble fraction was then neutralized with KOH/Mops (2/ 0.3 M) and frozen at -20°C for subsequent determination of intracellular metabolites. Dihydroxyacetone phosphate (DHAP), Glc-6-P and fructose-6-phosphate (Fru-6-P) concentrations were measured fluorimetrically, as described [20]. Simultaneously, a second sample was also removed from the chamber and the mitochondrial and cytosolic contents were separated using the digitonin fractionation technique [24]. The cell suspension was placed in an eppendorf tube on top of a layer of an isotonic medium containing 2 mM of digitonin. After 15 sec the tube was centrifuged for 30 sec and the mitochondria were precipited through a layer of silicon oil (Wacker AR 200) into HClO4 (10% mass/vol.) containing 25 mM EDTA. Immediately, the supernatant was quenched with HClO4 (5% mass/vol.). Finally, the neutralized extracts of each compartment were frozen at -20°C for subsequent determination of adenine nucleotides by high performance liquid chromatography [20].
and supplemented with 20 mM DHA. After 45 min of incubation in the presence of 100 µM silibinin or vehicle, the cell suspension was saturated again with O2/CO2 for one min and immediately transferred into a stirred oxygraph vessel equipped with a Clark oxygen electrode. The oxygen consumption rate was measured at 37°C and expressed in µmol O2.min-1.g dry cells-1. Materials Enzymes were purchased from Roche, substrates from Merck, collagenase A, Glc-6-P, α-glycerophosphate, βglycerophosphate, histone 2A and silibinin from Sigma. Silibinin was dissolved in a stock water-based solution containing 50% DMSO. Rhodorsil silicone oil was purchased from RhonePoulenc. Statistics The results are expressed as means ± SEM for the indicated number of separate experiments. The statistical significance of differences was calculated using Student’s test for unpaired samples.
Results
Determination of oxygen consumption rate Hepatocytes (7.5 mg dry cells.ml-1) were incubated in a shaking water bath at 37°C in closed vials containing 2 ml KrebsRinger-bicarbonate-calcium buffer saturated with O2/CO2 (19:1)
Silibinin decreases both gluconeogenesis and glycolysis in hepatocytes perifused with dihydroxyacetone In a first set of experiments we studied the effects of two different concentrations of silibinin (25 or 100 µM) on both glycolysis and gluconeogenesis rates in hepatocytes perifused with dihydroxyacetone (DHA) as exogenous substrate. Indeed, the rates of glucose (Jglucose) and lactate-plus-pyruvate (JL+P) production strictly reflect the respective gluconeogenic and glycolytic fluxes since glycogen content is negligible in hepatocytes from 24 hstarved rats. As shown in Figure 1A, silibinin induced a dosedependent inhibition of gluconeogenesis from DHA, JGlucose above 1.2 mM DHA exhibiting a mean reduction by 33% and 49% compared with control-vehicle in the presence of 25 and 100 µM silibinin, respectively. Since gluconeogenesis and glycolysis are branched pathways in DHA metabolism, the decrease in JGlucose could be due to a concomitant increase in JL+P. However, JL+P was also decreased by silibinin (Figure 1B), thus leading to a drop in the overall DHA metabolism rate expressed as threecarbon equivalents (JDHA, Figure 1C). This indicates that inhibition of gluconeogenesis by the flavonoid was not only related to a simple redistribution of carbon from one pathway to another. Furthermore, an increase in L/P ratio (Figure 1D), which reflects a more reduced cytosolic
Silibinin Inhibits Liver Glucose-6-Phosphatase
Cell Physiol Biochem 2007;20:925-934
Measurement of glucose-6-phosphatase activity G6Pase activity was measured on microsomal fractions prepared as described [25]. Briefly, livers of overnight fasted rats were homogenized in 3 vol of 50 mM Hepes, 0.2 M sucrose, 0.5 mM EDTA, pH 7.2 in a Potter device. Microsomes were obtained by two successive centrifugations (13000 g, 20 min and 100000 g, 60 min) and kept frozen at -80°C. The microsomes were gently resuspended with a Dounce in 25 mM Hepes, 0.5 mM EDTA, pH 7.2 to reach a protein concentration of 10 mg/ ml, corresponding to about 0.5 g of liver/ml. G6Pase or mannose6-phosphatase (M6Pase) activity was measured at 30°C in a final vol of 0.2 ml containing the indicated concentrations of substrate and of silibinin (or vehicle, i.e. 0.4 % DMSO), 50 mM Hepes pH 7.2 and 0.25 mg of microsomal protein. The reaction was stopped after 20 min of incubation by the addition of 1 ml 10% TCA and the released Pi was measured colorimetrically [26]. Microsomes were permeabilized by incubation in 0.4% taurodeoxycholate for 30 min at 0°C. Microsome intactness was evaluated by measuring M6Pase activity, which was about 10-times (9.8 ± 1.4, n=5; p
