Báo cáo khoa học: The role of glucose 6-phosphate in mediating the effects of glucokinase overexpression on hepatic glucose metabolism pdf

The role of glucose 6-phosphate in mediating the effects of glucokinase overexpression on hepatic glucose metabolism Linda Ha ¨ rndahl 1 , Dieter Schmoll 2 , Andreas W. Herling 2 and Loranne Agius 1 1 School of Clinical Medical Sciences-Diabetes, The University of Newcastle upon Tyne, Medical School, Newcastle upon Tyne, UK 2 Aventis Pharma Deutschland GmbH, TD Metabolism, Frankfurt, Germany Glucose 6-phosphate (Glc6P) is the first intermediate in glucose metabolism and is generated by the hexo- kinase-catalysed reaction. Hepatocytes express all four hexokinase isoenzymes (EC 2.7.1.1), but predominantly hexokinase IV, commonly known as glucokinase [1]. Glucokinase differs from the other isoenzymes by its lack of inhibition by physiological concentrations of Glc6P [2] and by its low affinity for glucose and sig- moidal kinetics [1]. Accordingly, in the hepatocyte, the cellular content of Glc6P responds to changes in both glucose concentration and glucokinase activity [3]. The latter is regulated by insulin and glucagon at the tran- scriptional level [4], and by glucose, precursors of fruc- tose 1-phosphate and hormones at post-transcriptional Keywords glucokinase activators; glucokinase; glucose 6-phosphate; glycogen; liver Correspondence L. Agius, School of Clinical Medical Sciences-Diabetes, The Medical School, University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4HH, UK Fax: +44 191 2220723 Tel: +44 191 2227033 E-mail: Loranne.Agius@ncl.ac.uk (Received 25 August 2005, revised 15 October 2005, accepted 18 November 2005) doi:10.1111/j.1742-4658.2005.05067.x Pharmacological activation or overexpression of glucokinase in hepatocytes stimulates glucose phosphorylation, glycolysis and glycogen synthesis. We used an inhibitor of glucose 6-phosphate (Glc6P) hydrolysis, namely the chlorogenic derivative, 1-[2-(4-chloro-phenyl)-cyclopropylmethoxy]-3, 4-di- hydroxy-5-(3- imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyloxy)-cyclohexane- carboxylic acid (also known as S4048), to determine the contribution of Glc6P concentration, as distinct from glucokinase protein or activity, to the control of glycolysis and glycogen synthesis by glucokinase overexpres- sion. The validity of S4048 for testing the role of Glc6P was supported by its lack of effect on glucokinase binding and its nuclear ⁄ cytoplasmic distri- bution. The stimulation of glycolysis by glucokinase overexpression corre- lated strongly with glucose phosphorylation, whereas glycogen synthesis correlated strongly with Glc6P concentration. Metabolic control analysis was used to determine the sensitivity of glycogenic flux to glucokinase or Glc6P at varying glucose concentrations (5–20 mm). The concentration control coefficient of glucokinase on Glc6P (1.4–1.7) was relatively inde- pendent of glucose concentration, whereas the flux control coefficients of Glc6P (2.4–1.0) and glucokinase (3.7–1.8) on glycogen synthesis decreased with glucose concentration. The high sensitivity of glycogenic flux to Glc6P at low glucose concentration is consistent with covalent modification by Glc6P of both phosphorylase and glycogen synthase. The high control strength of glucokinase on glycogenic flux is explained by its concentration control coefficient on Glc6P and the high control strength of Glc6P on gly- cogen synthesis. It is suggested that the regulatory strength of pharmacolo- gical glucokinase activators on glycogen metabolism can be predicted from their effect on the Glc6P content. Abbreviations Fru2,6P 2 , fructose 2,6-bisphosphate; Glc6P, glucose 6-phosphate; GKRP, glucokinase regulatory protein; MEM, minimal essential medium; PP-1, protein phosphatase-1; S4048, 1-[2-(4-chloro-phenyl)-cyclopropylmethoxy]-3, 4-dihydroxy-5-(3-imidazo[4,5-b]pyridin-1-yl-3-phenyl- acryloyloxy)-cyclohexanecarboxylic acid. 336 FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS levels through the interaction of glucokinase with bind- ing proteins [5,6]. The glucokinase regulatory protein (GKRP) is a specific inhibitor of glucokinase [5] and also a nuclear receptor that sequesters glucokinase in the nucleus at low concentrations of glucose [7]. Its affinity for glucokinase is decreased by fructose 1-phosphate, which causes translocation of glucokinase from the nucleus to the cytoplasm [5,7]. Conditions that cause dissociation of glucokinase from GKRP are associated with a parallel increase in the cell content of Glc6P, confirming the regulatory role of GKRP on glucokinase activity [8]. Recent studies have identified small molecule activators of glucokinase that bind to an allosteric site and markedly increase the affinity for glucose [9–11]. These compounds, unlike fructose 1-phosphate, do not dissociate glucokinase from GKRP in solution, but nonetheless cause transloca- tion of glucokinase from the nucleus. They thereby impact hepatocyte glucose metabolism by direct activation of glucokinase and by altering its subcellular location [10]. In the liver cell, Glc6P is an intermediate of various metabolic pathways and a substrate for glucose 6-phosphatase [12]. It is also a key regulator of glyco- gen synthase [13] and phosphorylase [14], and is impli- cated in the transcriptional control of glycolytic and lipogenic genes [15]. The effect of Glc6P on glycogen synthase, which is regulated by multisite phosphoryla- tion, involves a hierarchy of mechanisms [13]. Glc6P is an allosteric activator of glycogen synthase and thereby increases the enzyme affinity for its substrate, uridine 5¢-diphosphoglucose (UDP-glucose), but it also favours a conformation of the enzyme that is a better substrate for dephosphorylation by protein phospha- tase-1 (PP-1). The affinity of glycogen synthase for Glc6P increases with dephosphorylation, leading to further activation [16]. Glucose has an analogous (but converse) role on phosphorylase. It causes allosteric inhibition of phosphorylase-a (the phosphorylated act- ive form) by favouring the tense (T) conformation, which is also a better substrate for dephosphorylation by PP-1. As the dephosphorylated form (phosphory- lase-b) is catalytically inactive in liver [17], glucose cau- ses both allosteric inhibition and covalent inactivation [17]. Glc6P, like glucose, favours the T conformation [18] and regulates phosphorylase-a by allosteric inhibi- tion and promoting its dephosphorylation [14]. The actions of glucose and Glc6P on dephosphorylation are synergistic [19], indicating an enhancing role for Glc6P. Phosphorylase-a is a potent allosteric inhibitor of glycogen synthase phosphatase activity by binding to the glycogen targeting protein of PP-1, designated GL [17]. Thus, Glc6P-mediated depletion of phos- phorylase-a by dephosphorylation impacts glycogen synthase by a cascade mechanism (Fig. 1). Adenoviral vectors for the overexpression of gluco- kinase and hexokinase I in hepatocytes have been very useful tools for demonstrating the impact of over- expression of these isoenzymes on Glc6P and glycogen synthesis [3,20,21] and for applying metabolic control analysis to titrated enzyme overexpression to determine the control exerted by glucokinase on hepatic glucose metabolism [22,23]. Although both glucokinase and hexokinase I cause an increase in the cell content of Glc6P, the former, but not the latter, causes activation of glycogen synthase [3]. This anomaly can be explained by sequestration of Glc6P derived from glu- cokinase and hexokinase I in distinct pools [3,24], or by the involvement of mechanisms, additional to Glc6P, in mediating the effects of glucokinase over- expression. As glucokinase binds to a dual-specificity phosphatase [25], and also associates with PP1 in a multiprotein complex [26], it might promote dephos- phorylation of glycogen synthase through scaffolding of protein complexes. The question of whether glucokinase impacts meta- bolism exclusively through its catalytic activity or also through macromolecular interactions is now of particular relevance in the context of predicting the Glucose UDPG Glycogen GSA Phos-a GSB Phos-b G6Pase 1 1 2 3 S4048 GK G L PP PP G6P UDP GK expression Gluconeogenesis CP-91149 Fig. 1. Glucose 6-phosphate (Glc6P) regulates glycogen synthase by a cascade mechanism. Phosphorylase (phos) exists as inactive (Phos-b) and active (Phos-a) forms, which are dephosphorylated and phosphorylated, respectively. Phos-a is a potent inhibitor of activation of glycogen synthase (GS) by binding to the glycogen tar- geting protein, G L , in association with protein phosphatase-1 (PP-1), which converts less active GSB to more active GSA by dephospho- rylation. Glc6P (G6P) activates glycogen synthase by a cascade mechanism, whereby (a) it favours the conversion of phos-a to phos-b by a substrate-directed mechanism that is synergistic with glucose, (b) depletion of phos-a relieves the inhibition of G L , and (c) Glc6P stimulates the conversion of GSB to GSA by a substrate- directed mechanism. L. Ha ¨ rndahl et al. Control strength of Glc6P on glycogenesis FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS 337 metabolic effects of small-molecule glucokinase activa- tors, which may be a future treatment for type-2 diabe- tes [9–11]. A wide range of these molecules have been reported in patents [27,28]. However, there have been few published studies of their metabolic effects in hepatocytes [10], leading to extrapolation of their expected effects from studies of glucokinase overex- pression [29,30]. The aim of the present study was to provide a better understanding of the contribution of Glc6P to the metabolic effects of glucokinase overexpression in hepatocytes and to test for evidence for additional mechanisms. We used an inhibitor of glucose 6-phos- phate hydrolysis [31,32] to modulate the cell content of Glc6P independently of glucokinase protein expression and used metabolic control analysis to define the con- trol exerted by glucokinase and Glc6P. Results Effects of S4048 on hepatocyte Glc6P content and glucose metabolism To test the validity of 1-[2-(4-chloro-phenyl)-cyclopro- pylmethoxy]-3, 4-dihydroxy-5-(3-imidazo[4,5-b]pyridin- 1-yl-3-phenyl-acryloyloxy)-cyclohexanecarboxylic acid (S4048), a selective inhibitor of Glc6P hydrolysis [31], for evaluating the role of Glc6P, we determined the effect of different inhibitor concentrations on the Glc6P content and the rate of glycogen synthesis in hepatocytes incubated with 15 mm glucose (Fig. 2). The maximum effect on glycogen synthesis occurred at 0.2–2 lm S4048 and the half-maximum effect at <30 nm, consistent with the low 50% inhibitory con- centration (IC 50 )( 10 nm) of S4048 on glucose 6-phosphate hydrolysis in microsomes [31]. We next tested the effects of 2 lm S4048 at different glucose concentrations (Fig. 3). The cell content of Glc6P was increased fourfold (0.2–1.2 nmol mg )1 ) between 5 and 35 mm glucose. S4048 caused a progres- sive increase in Glc6P content with increasing glucose concentration, which was significant at 10–35 mm glu- cose after 1 h (Fig. 3A) and at 15–35 mm glucose after 3 h (results not shown). S4048 increased the fructose 2,6-bisphosphate (Fru2,6P 2 ) content by twofold at 5–15 mm glucose and by 50–70% at higher glucose concentrations (Fig. 3B). Glycogen synthesis was increased by S4048 by between two- and threefold at glucose concentrations of > 10 mm (Fig. 3C), whereas glycolysis was modestly increased (24–48%) at 10–20 mm glucose, but not at higher glucose concen- trations (Fig. 3D). S4048 had no effect on either glucokinase binding or on the metabolism of [2- 3 H]glucose, which approximates the rate of glucose phosphorylation (Fig. 3E,F). The distribution of glu- cokinase between the nucleus and the cytoplasm, expressed as the nuclear ⁄ cytoplasmic ratio, was also not affected by S4048 (2 lm) at either 5 mm glucose (2.0 ± 0.1 vs. 2.0 ± 0.1, mean ± SE; n ¼ 18) or 20 mm glucose (1.5 ± 0.1 vs. 1.5 ± 0.1). These results suggest that the increase of Glc6P in the presence of S4048 was caused by the inhibition of Glc6P hydro- lysis in the absence of detectable changes in glucose phosphorylation. S4048 causes inactivation of phosphorylase As the elevation of Glc6P causes inactivation of phos- phorylase in hepatocytes [14], the increase in Glc6P caused by S4048 would be expected to inactivate phos- phorylase. We therefore tested the effects of S4048 on the activity of phosphorylase-a. S4048 (2 lm) did not inactivate phosphorylase at either 5 or 35 mm glucose (Fig. 4). However, it caused a leftward shift in the glucose concentration, which resulted in the half- maximal inactivation (depletion) of phosphorylase-a, from 18.6 ± 1.0 mm to 8.2 ± 0.6 mm (Fig. 4). Similar results were obtained after 1 h of incubation (22.9 ± 2.2 mm to 12.8 ± 0.6 mm). To determine whether the leftward shift in the glu- cose-induced inactivation of phosphorylase is caused Fig. 2. Effects of 1-[2-(4-chloro-phenyl)-cyclopropylmethoxy]-3, 4-di- hydroxy-5-(3-imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyloxy)-cyclo- hexanecarboxylic acid (S4048) concentration on hepatocyte glucose 6-phosphate (Glc6P) and glycogen synthesis. Hepatocyte monolay- ers were incubated for 3 h in minimal essential medium (MEM) containing 15 m M glucose, and the concentrations of S4048 are shown for the determination of glycogen synthesis and Glc6P. Data represent the mean ± SE from four experiments. Control strength of Glc6P on glycogenesis L. Ha ¨ rndahl et al. 338 FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS by the elevation of Glc6P (Fig. 3A), rather than syner- gism between glucose and S4048 itself, we tested the effect of S4048 at 5 mm glucose in the presence of a gluconeogenic precursor. S4048 increased the Glc6P content in the presence of 2 mm dihydroxyacetone (control, 0.4 ± 0.1; 2 lm S4048, 2.2 ± 0.6 nmolÆmg )1 of protein), but not in its absence (from 0.3 ± 0.1 nmolÆmg )1 of protein to 0.2 ± 0.1 nmolÆmg )1 of protein) and, likewise, S4048 (0.05–2 lm) caused inac- tivation of phosphorylase in the presence, but not in the absence, of dihydroxyacetone (Fig. 5A), confirming that the inactivation by S4048 is caused by the elevated Glc6P. A plot of phosphorylase-a activity against the respective Glc6P content in the incubations with dihydroxyacetone and varying concentrations of S4048 showed saturation of phosphorylase inactivation at  1.4 nmolÆmg )1 of Glc6P and a half-maximal effect at 0.48 ± 0.15 nmol )1 Æmg )1 (Fig. 5B). The saturation of phosphorylase inactivation by Glc6P could explain the lack of further inactivation by S4048 at 35 mm glucose (Fig. 4). Comparison of the metabolic effects of glucokinase overexpression and S4048 As the effects of S4048 on glucose metabolism are not associated with changes in glucokinase activity or com- partmentation, and the inactivation of phosphorylase can be explained by the elevation in Glc6P, this sup- ports the validity of S4048 as a tool to modulate the cell content of Glc6P. We next compared the effects of S4048 (0.02 and 0.5 lm) and glucokinase overexpres- sion (1.7-, 2.5- and 4.2-fold), using recombinant adeno- virus [20], on glycogen synthesis and glycolysis at 5, 10, 15 or 20 mm glucose, and correlated the changes in flux with glucose phosphorylation and cell content of 0 200 400 600 0 100 200 300 400 500 20 40 60 80 010203040 0 200 400 600 800 010203040 0 100 200 300 400 * *** * *** *** *** *** ** ** ** *** ** 0 2 4 6 8 10 Control S4048 *** ** ** ** A B D F E C Glucose (mM) Glucose (mM) Free GK (%) Glycogen Synthesis Glucose 6-P Phosphorylation Glycolysis Fructose 2,6-P ** Fig. 3. Effects of 1-[2-(4-chloro-phenyl)- cyclopropylmethoxy]-3, 4-dihydroxy-5- (3-imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyl- oxy)-cyclohexanecarboxylic acid (S4048) on glucose metabolism at varying concentra- tions of glucose. Hepatocytes were incubated for 1 or 3 h with the glucose concentrations indicated, in the absence or presence of 2 l M S4048. (A) Glucose 6-phosphate (Glc6P) (nmolÆmg )1 ), 1 h. (B) Fructose 2,6-bisphosphate (pmolÆmg )1 ), 3 h. (C) Glycogen synthesis (nmolÆ3h )1 Æmg )1 ). (D) Glycolysis (nmolÆ3h )1 Æmg )1 ). (E) Free glucokinase activity (% total), 3 h. (F) Glucose phosphorylation (nmolÆ3h )1 Æmg )1 ). Data represent the mean ± SE from four to six experiments. *P < 0.05; **P < 0.01; ***P < 0.001, effect of S4048. Fig. 4. 1-[2-(4-Chloro-phenyl)-cyclopropylmethoxy]-3, 4-dihydroxy- 5-(3-imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyloxy)-cyclohexanecarb- oxylic acid (S4048) potentiates the inactivation of phosphorylase-a by glucose. Hepatocytes were incubated for 3 h with the concen- trations of glucose shown, in the absence or presence of 2 l M S4048. Phosphorylase-a is expressed as munitsÆmg )1 of protein. Data are the mean ± SE from five experiments. *P < 0.05; **P < 0.01, effect of S4048. L. Ha ¨ rndahl et al. Control strength of Glc6P on glycogenesis FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS 339 Glc6P (Fig. 6). Glucokinase overexpression (4.2-fold) caused a threefold increase in glucose phosphorylation at 5 mm glucose and a twofold increase at 10–15 mm glucose (determined from the detritiation of [2- 3 H]glu- cose), and the rates of glycolysis showed very similar trends (Fig. 6A,C). In contrast, the Glc6P content and rates of glycogen synthesis during incubation with 10–15 mm glucose were increased by 7–11-fold by glu- cokinase overexpression (Fig. 6B,D). When the combined data for incubations with S4048 and overexpression of glucokinase were analysed (Fig. 6E–H), the rates of glycolysis correlated more strongly with glucose phosphorylation than with Glc6P content (r, 0.97 vs. 0.79, Fig. 6E,F) whilst the rates of glycogen synthesis correlated more strongly with the Glc6P content than with glucose phosphoryla- tion (r, 0.92 vs. 0.82) (Fig. 6G,H). Metabolic control analysis defines the control by Glc6P Metabolic control analysis [33,34] was used to deter- mine (a) the sensitivity of the Glc6P content to gluco- kinase activity, expressed as the concentration control coefficient of glucokinase (C G6P GK ), (b) the sensitivity of glycogen synthesis to Glc6P, expressed as the flux control coefficient of Glc6P (C J G6P ), and (c) the sensi- tivity of glycogen synthesis to glucokinase activity, expressed as the flux control coefficient of glucokinase (C J GK ) (Table 1). The concentration control coefficient of glucokinase on Glc6P, determined from the slope of double log plots of Glc6P against the respective glucokinase activ- ity, was relatively independent of glucose concentration (1.4–1.7). However, the flux control coefficient of Glc6P on glycogen synthesis was more than twofold higher at 5 mm compared with 20 mm glucose, as was the flux control coefficient of glucokinase on glycogen synthesis (Table 1). Dual control of glycogen synthase by Glc6P and phosphorylase The higher flux control coefficient of Glc6P on glyco- gen synthesis at 5 mm compared with 20 mm glucose (Table 1) may be in part caused by a dual effect of Glc6P on phosphorylase and glycogen synthase at low Glc6P and by an effect on glycogen synthase at eleva- ted Glc6P. To test the latter possibility, we determined the separate and combined effects of S4048 and gluco- kinase overexpression on phosphorylase inactivation and glycogen synthase activation and compared this with the effects of a phosphorylase inhibitor (CP-91149), which causes depletion of phosphorylase-a and sequen- tial activation of glycogen synthase [35,36]. The latter inhibitor caused a greater depletion of phosphorylase-a than the combined effects of glucokinase overexpres- sion and S4048 (Fig. 7A), which had additive effects on Fig. 5. Inactivation of phosphorylase by glucose 6-phosphate (Glc6P). Hepatocytes were incubated for 1 h in minimal essential medium (MEM), containing 5 m M glucose, in the absence (open symbol) or presence (closed symbol) of 2 m M dihydroxyacetone and at the concentrations of 1-[2-(4-chloro-phenyl)-cyclopropyl- methoxy]-3, 4-dihydroxy-5-(3-imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyl- oxy)-cyclohexanecarboxylic acid (S4048) shown, and phosphorylase-a (A) and Glc6P content were determined. (B) Phosphorylase-a activ- ity against the corresponding Glc6P content in the incubations with dihydroxyacetone. Data shown represent the mean ± SE from four experiments. *P < 0.05. Control strength of Glc6P on glycogenesis L. Ha ¨ rndahl et al. 340 FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS the Glc6P content (Fig. 7B). Activation of glycogen synthase by S4048 and ⁄ or glucokinase overexpression increased with increasing Glc6P (Fig. 7B), consistent with the activation of glycogen synthase at elevated Glc6P. Half-maximal activation occurred at a higher Glc6P concentration (5 nmolÆ mg )1 ) than for inactiva- tion of phosphorylase-a (0.5 nmolÆmg )1 ). It is notewor- thy that activation of glycogen synthase by CP-91149 was independent of changes in Glc6P, consistent with the role of phosphorylase-a as a negative modulator of synthase phosphatase [17]. Discussion Glc6P is an allosteric regulator of glycogen synthase and phosphorylase and it also promotes the de- phosphorylation of both enzymes, causing activation of synthase and inactivation of phosphorylase (con- version of phosphorylase-a to phosphorylase-b, Fig. 1). As phosphorylase-a is a potent allosteric inhibitor of glycogen synthase phosphatase, depletion of phosphorylase-a by Glc6P leads to the further activation of glycogen synthase through a cascade Fig. 6. Glycogen synthesis correlates with glucose 6-phosphate (Glc6P), and glycolysis correlates with glucose phosphorylation. Hepatocytes were either untreated (control, S0.02, S0.5) or treated with varying titres of adenoviral vector encoding rat liver glucokin- ase (Ad-LGK) for overexpression of glucokin- ase by 1.7-fold (GK1.7), 2.5-fold (GK2.5) and 4.2-fold (GK4.2). They were incubated with 5, 10, 15 or 20 m M glucose, in the absence (Con) or presence of 1-[2-(4-chloro-phenyl)- cyclopropylmethoxy]-3, 4-dihydroxy-5-(3- imidazo[4,5-b]pyridin-1-yl-3-phenyl-acryloyl- oxy)-cyclohexanecarboxylic acid (S4048), at either 0.02 l M (S0.02) or 0.5 lM (S0.5) for determination of glucose phosphorylation (A), Glc6P content (B), glycolysis (C) and glycogen synthesis (D). (E) Glycolysis vs. glucose phosphorylation. (F) Glycolysis vs. Glc6P. (G) Glycogen synthesis vs. glucose phosphorylation. (H) Glycogen synthesis vs. Glc6P. (E)–(H) Closed symbols, control and S0.02 or S0.5; open symbols, glucokinase overexpression. Data represent the mean ± SE from three experiments. L. Ha ¨ rndahl et al. Control strength of Glc6P on glycogenesis FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS 341 mechanism (Fig. 1). Glc6P also determines the cell content of Fru2,6P 2 , an allosteric activator of phos- phofructokinase-1 [37], and is implicated in the con- trol of gene expression of various enzymes involved in either glycolysis or lipogenesis [15,38]. Modulation of hepatic glucokinase activity in vitro or in vivo is associated with large perturbations of the cell content of Glc6P [3] as well as glycogenic flux and lipid accumulation [21,30,39]. The glycogenic effects are attributed to a regulatory role of Glc6P, based on correlation studies [13]. However, evidence that the overexpression of hexokinase I causes a smaller acti- vation of glycogen synthase than expected from the increment in Glc6P [3] has raised questions on the extent by which the steady-state cell content of Glc6P, as distinct from either glucose phosphoryla- tion or glucokinase protein through macromolecular interactions, accounts for the metabolic effects of glu- cokinase overexpresssion. In this study we established the validity of S4048, a potent inhibitor of Glc6P transport into the endoplas- mic reticulum and thereby of Glc6P hydrolysis [31], as an experimental tool to modulate the Glc6P content in the absence of changes in rates of glucose phosphory- lation. We show, using S4048 to modulate Glc6P, that during glucokinase overexpression, glycogenic flux cor- relates closely with Glc6P content rather than glucose phosphorylation, whereas the converse is true for gly- colysis. We used metabolic control analysis to deter- mine the quantitative relationship between glucokinase and the cell content of Glc6P, and also between Glc6P and metabolic flux. S4048 causes large perturbations in Glc6P concen- tration in liver cells [31,32,38], as well as secondary changes in gene expression, including up-regulation of lipogenic enzymes and down-regulation of glucokinase mRNA levels [38]. In the experimental conditions used in this study, involving incubation for up to 3 h, S4048 Table 1. Control coefficients of glucokinase and glucose 6-phos- phate (Glc6P ) on glycogen synthesis. Coefficients were determined for each glucose concentration, from the data in Fig. 6, in cells overexpressing glucokinase by up to 4.2-fold. The concentration control coefficients (C G6P GK ) of glucokinase on the Glc6P content were determined from the slope of double log plots of Glc6P against glucokinase activity. These were linear (r ¼ 1.0 in all cases) up to 4.2-fold glucokinase overexpression. The Glc6P concentration range of these plots is shown in column 2. The flux control coefficients of Glc6P (C J G6P ) and glucokinase (C J GK ) on glycogen synthesis (J) were determined from the slope of double log plots of glycogen synthesis against Glc6P content or glucokinase activity, respectively. These plots were curvilinear (r-values between 0.94 and 0.97). The initial slope of these plots is shown in parentheses. Glucose (m M) Glc6P (nmolÆmg )1 )C G6P GK C J G6P C J GK 5 0.2–1.1 1.4 1.7 (2.4) 2.6 (3.7) 10 0.4–2.2 1.4 1.3 (1.9) 2.0 (2.8) 15 0.5–3.7 1.7 0.8 (1.1) 1.7 (2.4) 20 0.6–4.3 1.7 0.7 (1.0) 1.1 (1.8) Fig. 7. Combined effects of glucokinase overexpression and 1-[2-(4- chloro-phenyl)-cyclopropylmethoxy]-3, 4-dihydroxy-5-(3-imidazo[4,5- b]pyridin-1-yl-3-phenyl-acryloyloxy)-cyclohexanecarboxylic acid (S4048) on phosphorylase and glycogen synthase. Hepatocytes were either untreated (Control, S, CP) or treated with adenoviral vector encoding rat liver glucokinase (Ad-LGK) (glucokinase or GKS), and incubated for 1 h without or with 2 l M S4048 (S) or 10 l M CP-91149 (CP) at either 10 mM (open bars ⁄ symbols) or 25 m M glucose (solid bars ⁄ symbols), as indicated. Phosphorylase-a (A), glycogen synthase and Glc6P were determined as described in the Experimental procedures. (B) Glycogen synthase (activity ratio) relative to the corresponding Glc6P. Data represent the mean ± SE of three experiments. *P < 0.05, **P < 0.01, relative to the control. Control strength of Glc6P on glycogenesis L. Ha ¨ rndahl et al. 342 FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS did not affect the total glucokinase activity, suggesting that changes in protein expression are minimal within this time scale. S4048 also did not affect the distribu- tion of glucokinase between the nucleus and cyto- plasm, or the rate of glucose phosphorylation. This establishes the validity of S4048 as a tool to specifically modulate Glc6P, independently of changes in glucose phosphorylation. The stimulation of glycogen synthesis by S4048 was associated with both inactivation of phosphorylase and activation of glycogen synthase, as expected from the dual role of Glc6P in regulating both of these enzymes (Fig. 1). The selectivity of the effects of S4048 (at < 2 lm) was further supported by its lack of effect on phosphorylase-a at low glucose (5 mm) when the concentration of Glc6P was unchanged. Comparison of the glycogenic effects of S4048 and glucokinase overexpression established a major role for Glc6P in accounting for the stimulation of glycogen synthesis by glucokinase overexpression. As the glyco- genic effect of glucokinase overexpression did not exceed that of S4048 for a corresponding cell content of Glc6P, our results do not support a significant role for glucokinase protein through macromolecular inter- actions in accounting for the greater glycogenic potency of glucokinase compared with hexokinase I overexpression. Thus, other explanations must be sought for the lower glycogenic potency of overexpres- sion of hexokinase I [3]. One suggestion that remains untested is that the elevated Glc6P, derived from hexo- kinase I, is sequestered in a compartment that is not accessible to the regulatory mechanisms involved in sti- mulation of glycogen synthesis [13]. Another possibility is that hexokinase I may have an inhibitory effect on glycogen synthesis. Glycolysis, unlike glycogen synthesis, correlated more strongly with glucose phosphorylation than with Glc6P concentration. Glc6P has a dual role in the con- trol of glycolysis through changes in the cell content of Fru2,6P 2 , an activator of phosphofructokinase-1 [37], and up-regulation of pyruvate kinase gene expression [38]. Although S4048 increased the Fru2,6P 2 content, as expected [37], the stimulation of glycolysis was small and only observed at low glucose concentrations. This is consistent with recent findings that the overexpres- sion of phosphofructokinase-2 ⁄ fructose bisphospha- tase-2 does not increase glycolytic flux, despite the increase in Fru2,6P 2 concentration, indicating that the endogenous Fru2,6P 2 concentration is saturating [6]. Changes in protein expression are probably minimal under the short-term incubation conditions of the present study (see above). It can be inferred therefore that under short-term conditions, the rate of glucose phosphorylation, rather than the elevated concentra- tion of Glc6P or Fru2,6P 2 , determines the increased glycolysis during activation of glucokinase. However, following chronic elevation of Glc6P, an increase in glycolysis and lipogenesis would be expected as a result of increased expression of pyruvate kinase and lipogen- ic enzymes, as demonstrated following the in vivo administration of S4048 [32,38]. Metabolic control analysis provides a framework for a quantitative description of how changes in enzyme activities and concentrations of metabolic intermedi- ates influence the flux through a pathway [33,34]. The flux control coefficient of an enzyme or metabolite des- cribes the sensitivity of flux to changes in enzyme activity or metabolite concentration, respectively, and, likewise, concentration control coefficients describe the sensitivity of metabolite concentrations to perturba- tions in enzyme activity. Previous work has shown that the rate of glycogen synthesis in hepatocytes is a sig- moidal function of glucokinase activity and that glu- cokinase has a very high flux control coefficient on glycogen synthesis at low glucose concentrations [22,23]. Three main findings emerged from the metabolic control analysis: first, glucokinase has a very high con- centration control coefficient (> 1) on Glc6P, which is relatively independent of glucose concentration (1.4–1.7); second, the rate of glycogen synthesis is a sigmoidal function of Glc6P concentration when deter- mined over a range of glucose concentrations and activities of glucokinase; and, third, the flux control coefficient of Glc6P on glycogen synthesis shows very similar trends to the flux control coefficient of gluco- kinase on glycogen synthesis and is more than twofold higher at 5 mm glucose than at 20 mm glucose. The large increase in Glc6P content (5.5–7-fold) dur- ing glucokinase overexpression (fourfold) is consistent with the prediction that perturbations in enzyme activ- ity cause large changes in metabolite concentrations [40] and occurs despite the presence of multiple mecha- nisms that buffer the Glc6P concentration, including glucose 6-phosphate hydrolysis. The sigmoidal relationship between the rate of gly- cogen synthesis and Glc6P indicates a high sensitivity at low Glc6P concentration. This can be explained by the combined effects of Glc6P on phosphorylase and glycogen synthase (Fig. 1). Glc6P causes allosteric activation and enhances dephosphorylation of glyco- gen synthase through a substrate-directed mechanism. The affinity of glycogen synthase for Glc6P as an allosteric activator is a sigmoidal function of the degree of dephosphorylation [16]. However, the sensi- tivity of glycogen synthase to Glc6P as an allosteric L. Ha ¨ rndahl et al. Control strength of Glc6P on glycogenesis FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS 343 activator is an inverse function of the affinity for Glc6P [41]. Thus, at low Glc6P, when the phosphory- lation state of glycogen synthase is high, sensitivity to Glc6P as an allosteric activator would also be high. Nevertheless, Glc6P-mediated activation of glycogen synthase, whether by allosteric mechanisms or by de- phosphorylation, cannot explain the high sensitivity of glycogenic flux, because activation of glycogen syn- thase by dephosphorylation with inhibitors of glyco- gen synthase kinase-3 has a negligible impact on glycogen synthesis in the absence of independent inac- tivation of phosphorylase ([42], Fig. 6). Thus, inacti- vation of phosphorylase by Glc6P is essential for the high sensitivity of glycogenic flux to Glc6P. This study provides further support for this mechanism from the leftward shift of the glucose-induced inacti- vation of phosphorylase in the presence of S4048 (Fig. 4) and from the higher affinity of phosphorylase inactivation compared with synthase activation to Glc6P (Fig. 4 vs. Fig. 7B). The mechanisms by which glucokinase activation or overexpression control hepatic glycogenesis, glyco- lysis and lipogenesis are of major interest in view of the potential use of glucokinase activators for the treatment of Type 2 diabetes [9–11]. Hepatic gluco- kinase overexpression in vivo causes increased glyco- gen accumulation and dysregulation of lipid metabolism [30,43]. As yet, few studies have reported the effects of glucokinase activators on hepatic meta- bolism [9,10]. In isolated hepatocytes, the glucokinase activators cause a leftward shift in the affinity of the hepatocyte for glucose and a similar stimulation of glycogenesis as that induced by physiological stimuli (precursors of fructose 1-phosphate) but a larger sti- mulation of glucose phosphorylation and glycolysis [10]. In contrast, glucokinase overexpression causes a larger increase in both glycogen synthesis and Glc6P content than either physiological or pharmacological activation of endogenous glucokinase [3]. This can be explained by the progressive increase in Glc6P caused by increasing glucokinase expression and by the high flux control coefficient of Glc6P on glyco- gen synthesis. Based on the results of the present study, it is suggested that the regulatory strength of glucokinase activators on glycogen metabolism can be predicted from the magnitude of their effect on the cell content of Glc6P. If dysregulation of lipid metabolism by glucokinase overexpression in vivo [30,43] is caused by Glc6P, as suggested by the lipid accumulation caused by S4048 [38], then it could be argued that the impact of glucokinase activators on lipid metabolism can also be predicted from the Glc6P content. Experimental procedures Materials S4048 [31] was synthesized at Aventis (Pharma GmbH, Frankfurt, Germany), and CP-91149 [35] was a kind gift from Pfizer Global Research and Development (Pfizer, New London Laboratories, Groton, CT, USA). Sources of other reagents were as described previously [6]. Hepatocyte isolation and monolayer culture Male Wistar rats (180–300 g body weight) were either from B & K (Hull, UK) or from Charles River (Margate, UK). Hepatocytes were isolated by collagenase perfusion of the liver [22] and suspended in minimal essential medium (MEM) with 5% (v ⁄ v) neonatal calf serum and plated in multiwell plates or on gelatin-coated coverslips [10]. After cell attachment (2 h), the medium was replaced by serum-free MEM containing varying titres of adeno- viral vector encoding rat liver glucokinase, for the over- expression of rat liver glucokinase [20,21]. After a further 2 h of incubation, the medium was replaced with serum- free MEM containing 5 mm glucose and 10 nm dexameth- asone, and the hepatocyte monolayers were cultured for 16–18 h. All experiments were carried out in accordance with EC Council Directive (86/609/EEC). Metabolic incubations Hepatocytes were incubated for 1 or 3 h in MEM contain- ing the glucose concentrations and additions indicated. S4048 and CP-91149 were dissolved in dimethylsulfoxide (final concentration 0.1%, v ⁄ v). The medium was supple- mented with [U- 14 C]glucose, [3- 3 H]glucose or [2- 3 H]glucose (1–2 lCiÆmL )1 ) for the determination of glycogen synthesis, glycolysis or glucose phosphorylation, respectively. Glyco- gen synthesis was determined from the incorporation of 14 C label into glycogen [22], and glycolysis and glucose phos- phorylation from the formation of 3 H 2 O [8]. Parallel incuba- tions were performed to determine enzyme activities and metabolites. Rates of metabolic flux were similar in 1 h and 3 h incubations, and the cell content of Glc6P and the activ- ity of phosphorylase were also similar after 1 h and 3 h. Metabolite and enzyme assays Glc6P and Fru2,6P 2 were determined as described previ- ously [8]. Glucokinase (free and bound activity) was deter- mined by the digitonin permeabilization assay and free activity is expressed as the percentage total [22]. Phosphory- lase-a, assayed spectrometrically [42], is expressed as munitsÆ mg )1 protein, and glycogen synthase, assayed in the absence or presence of Glc6P [42], is expressed as the activ- ity ratio. Control strength of Glc6P on glycogenesis L. Ha ¨ rndahl et al. 344 FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS Glucokinase immunostaining Hepatocyte monolayers were washed in NaCl ⁄ P i , fixed in 4% (w ⁄ v) paraformaldehyde in NaCl ⁄ P i and stained for glu- cokinase with a rabbit IgG against human glucokinase:318– 405 (H88, sc-7908; Santa Cruz) and fluorescein isothiocya- nate (FITC)-labelled anti-rabbit IgG [6]. Imaging for FITC fluorescence was performed using a Nikon Eclipse E400 epi- fluorescence microscope (Nikon Corporation, Tokyo, Japan) and the nuclear ⁄ cytoplasmic ratio was analysed from the Gray images using Lucia G ⁄ F Analysis Software [6]. Data analysis Results are presented as mean ± SE for the number of hepatocyte preparations indicated. Statistical analysis was carried out by using the Student’s paired t-test. Flux con- trol coefficients of glucokinase or Glc6P on glycogen syn- thesis (J) were determined from the slopes of the double log plots of rates of glycogen synthesis against glucokinase activity or Glc6P, respectively [34]. The concentration con- trol coefficient of glucokinase on Glc6P was determined from the slope of the double log plot of Glc6P content against glucokinase activity [34]. Acknowledgements We thank the ‘Emma Ekstrands, Hildur Teggers and Jan Teggers Foundation’ and the ‘Wenner-Gren Foun- dation’ for postdoctoral fellowships to LH; Diabetes UK and the Medical Research Council (Joint Research Equipment Initiative) for equipment grants to LA; Dr Chris Newgard for the adenoviral vector and Dr Judith Treadway for helpful advice. References 1 Cardenas ML (1995) Glucokinase: Its regulation and role in liver metabolism. RG Landes, Austin. 2 Storer AC & Cornish-Bowden A (1977) Kinetic evi- dence for a ‘mnemonical’ mechanism for rat liver glu- cokinase. Biochem J 165, 61–69. 3 Seoane J, Gomez-Foix AM, O’Doherty RM, Gomez- Ara C, Newgard CB & Guinovart JJ (1996) Glucose 6- phosphate produced by glucokinase, but not hexokinase I, promotes the activation of hepatic glycogen synthase. J Biol Chem 271, 23756–23760. 4 Iynedjian PB (1993) Mammalian glucokinase and its gene. Biochem J 293, 1–13. 5 Van Schaftingen E, Detheux M & Veiga da Cunha M (1994) Short-term control of glucokinase activity: role of a regulatory protein. FASEB J 8, 414–419. 6 Payne VA, Arden C, Wu C, Lange AJ & Agius L (2005) Dual role of phosphofructokinase-2 ⁄ fructose bisphosphatase-2 in regulating the compartmentation and expression of glucokinase in hepatocytes. Diabetes 54, 1949–1957. 7 Brown KS, Kalinowski SS, Megill JR, Durham SK & Mookhtiar KA (1997) Glucokinase regulatory protein may interact with glucokinase in the hepatocyte nucleus. Diabetes 46, 179–178. 8 Agius L (1997) Involvement of glucokinase transloca- tion in the mechanism by which resorcinol inhibits gly- colysis in hepatocytes. Biochem J 325, 667–673. 9 Grimsby J, Sarabu R, Corbett WL, Haynes NE, Bizzar- ro FT, Coffey JW, Guertin KR, Hilliard DW, Kester RF, Mahaney PE et al. (2003) Allosteric activators of glucokinase: potential role in diabetes therapy. Science 301, 370–373. 10 Brocklehurst KJ, Payne VA, Davies RA, Carroll D, Vertigan HL, Wightman HJ, Aiston S, Waddell ID, Leighton B, Coghlan MP et al. (2004) Stimulation of hepatocyte glucose metabolism by novel small molecule glucokinase activators. Diabetes 53, 535–541. 11 Kamata K, Mutsoya M, Nishimura T, Eiki J & Nagata Y (2004) Structual basis for allosteric regulation of the monomeric allosteric enzyme human glucokinase. Struc- ture 12, 429–438. 12 van de Werve G, Lange A, Newgard C, Mechin MC, Li Y & Berteloot A (2000) New lessons in the regulation of glucose metabolism taught by the glucose 6-phospha- tase system. Eur J Biochem 267, 1533–1549. 13 Ferrer JC, Favre C, Gomis RR, Fernandez-Novell JM, Garcia-Rocha M, de la Iglesia N, Cid E & Guinovart JJ (2003) Control of glycogen deposition. FEBS Lett 546, 127–132. 14 Aiston S, Andersen B & Agius L (2003) Glucose 6-phosphate regulates hepatic glycogenolysis through inactivation of phosphorylase. Diabetes 52, 1340–1346. 15 Dentin R, Pegorier JP, Benhamed F, Foufelle F, Ferre P, Fauveau V, Magnuson MA, Girard J & Postic C (2004) Hepatic glucokinase is required for the synergis- tic action of ChREBP and SREBP-1c on glycolytic and lipogenic gene expression. J Biol Chem 279, 20314– 20326. 16 Roach PJ, Takeda Y & Larner J (1976) Rabbit skeletal muscle glycogen synthase. I. Relationship between phos- phorylation state and kinetic properties. J Biol Chem 251, 1913–1919. 17 Bollen M, Keppens S & Stalmans W (1998) Specific fea- tures of glycogen metabolism in liver. Biochem J 336, 19–31. 18 Newgard CB, Hwang PK & Fletterick RJ (1989) The family of glycogen phosphorylases: structure and func- tion. Crit Rev Biochem Mol Biol 24, 69–99. 19 Aiston S, Green AR, Mukhtar M & Agius L (2004) Glucose 6-phosphate causes translocation of phosphorylase in hepatocytes and it inactivates the L. Ha ¨ rndahl et al. Control strength of Glc6P on glycogenesis FEBS Journal 273 (2006) 336–346 ª 2005 The Authors Journal compilation ª 2005 FEBS 345 [...]... Seoane J, Gomez-Foix AM, Guinovart JJ & Newgard CB (1996) Differential metabolic effects of adenovirus-mediated glucokinase and hexokinase I overexpression in rat primary hepatocytes J Biol Chem 271, 20524–20530 Agius L, Peak M, Newgard CB, Gomez-Foix AM & Guinovart JJ (1996) Evidence for a role of glucoseinduced translocation of glucokinase in the control of hepatic glycogen synthesis J Biol Chem 271,... J, Guinovart JJ & Agius L (2000) The role of the regulatory protein of glucokinase in the glucose sensory mechanism of the hepatocyte J Biol Chem 275, 10597–10603 Gomis RR, Cid E, Garcia-Rocha M, Ferrer JC & Guinovart JJ (2002) Liver glycogen synthase but not the muscle isoform differentiates between glucose 6-phosphate produced by glucokinase or hexokinase J Biol Chem 277, 23246–23252 Munoz-Alonso... Acad Sci USA 101, 1485– 1490 Aiston S, Coghlan MP & Agius L (2003) Inactivation of phosphorylase is a major component of the mechanism by which insulin stimulates hepatic glycogen sythesis in hepatocytes Eur J Biochem 270, 2773–2781 Ferre T, Riu E, Franckhauser S, Agudo J & Bosch F (2003) Long-term overexpression of glucokinase in the liver of transgenic mice leads to insulin resistance Diabetologia 46,... 36 37 38 39 40 41 42 43 PA, McPherson RK et al (1998) Plasma glucose levels are reduced in rats and mice treated with an inhibitor of glucose- 6-phosphate translocase Diabetes 47, 1630– 1636 Herling AW, Burger H, Schubert G, Hemmerle H, Schaefer H & Kramer W (1999) Alterations of carbohydrate and lipid intermediary metabolism during inhibition of glucose- 6-phosphatase in rats Eur J Pharmacol 386, 75–82... HJ, Herling AW, Kuipers F, Meijer AJ & Reijngoud DJ (2001) Acute inhibition of hepatic glucose- 6-phosphatase does not affect gluconeogenesis but directs gluconeogenic flux toward glycogen in fasted rats A pharmacological study with the chlorogenic acid derivative S4048 J Biol Chem 276, 25727– 25735 Brown AM, Wiggins D & Gibbons GF (1999) Glucose phosphorylation is essential for the turnover of neutral... lipid and the second stage assembly of triacylglycerolrich ApoB-containing lipoproteins in primary hepatocyte cultures Arterioscler Thromb Vasc Biol 19, 321– 329 Cornish-Bowden A & Cardenas ML (2001) Silent genes given voice Nature 409, 571–571 Schafer JR, Fell DA, Rothman D & Shulman RG (2004) Protein phosphorylation can regulate metabolite concentrations rather than control flux: the example of glycogen... 2103–2106 Couzin J (1993) Medicine Drug deals diabetes a onetwo punch Science 301, 290 O’Doherty RM, Lehman DL, Telemaque-Potts S & Newgard CB (1999) Metabolic impact of glucokinase overexpression in liver: lowering of blood glucose in fed rats is accompanied by hyperlipidemia Diabetes 48, 2022–2027 Parker JC, VanVolkenburg MA, Levy CB, Martin WH, Burk SH, Kwon Y, Giragossian C, Gant TG, Carpino 346 32...Control strength of Glc6P on glycogenesis 20 21 22 23 24 25 26 27 28 29 30 31 L Harndahl et al ¨ enzyme synergistically with glucose Biochem J 377, 195–204 Becker TC, Noel RJ, Johnson JH, Lynch RM, Hirose H, Tokuyama Y, Bell GI & Newgard CB (1996) Differential effects of overexpressed glucokinase and hexokinase I in isolated islets Evidence for functional segregation of the high and low... S, Hampson L, Gomez-Foix AM, Guinovart JJ & Agius L (2001) Hepatic glycogen synthesis is highly sensitive to phosphorylase activity: evidence from metabolic control analysis J Biol Chem 276, 23856–23866 Hue L & Rider MH (1987) Role of fructose 2,6-bisphosphate in the control of glycolysis in mammalian tissues Biochem J 245, 313–324 van Dijk TH, van der Sluijs FH, Wiegman CH, Baller JF, Gustafson LA,... Guillemain G, Kassis N, Girard J, Burnol AF & Leturque A (2000) A novel cytosolic dual specificity phosphatase, interacting with glucokinase, increases glucose phosphorylation rate J Biol Chem 275, 32406–32412 Danial NN, Gramm CF, Scorrano L, Zhang CY, Krauss S, Ranger AM, Datta SR, Greenberg ME, Licklider LJ, Lowell BB et al (2003) BAD and glucokinase reside in a mitochondrial complex that integrates . The role of glucose 6-phosphate in mediating the effects of glucokinase overexpression on hepatic glucose metabolism Linda Ha ¨ rndahl 1 ,. through the interaction of glucokinase with bind- ing proteins [5,6]. The glucokinase regulatory protein (GKRP) is a specific inhibitor of glucokinase [5]