INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

MUSCLE GLYCOGEN SYNTHESIS is a major determinant in glucose homeostasis. Glycogen synthesis, in resting glycogen-containing muscle, appears to be limited by glucose transport/phosphorylation, which determines glucose 6-phosphate (6-P) availability. Glucose 6-P levels, besides constituting a precursor pool for glycogenesis, have been positively correlated with the activation state of glycogen synthase (GS) (24), the enzyme generally considered to limit the glycogenetic rate. However, Shulman and colleagues (21, 22), on the basis of metabolic control analysis, proposed that the activation of GS in muscle does not control the glycogenic flux, but rather adapts its activity to changes in the flux of glucose 6-P generation. In accordance with this model, GS activation and glycogen have a passive role in the control of muscle glucose metabolism, and glycogen is synthesized to buffer the increases in glucose 6-P, which is mostly dependent on insulin stimulation of glucose uptake. This hypothesis is well supported by NMR experiments in resting muscle in vivo (21); nevertheless, there is no evidence that this model applies to glycogen synthesis during recovery after glycogen exhaustion (16). Glycogen resynthesis in muscle is biphasic (11): a rapid insulin-independent resynthesis (19) occurs during the 1st h, which drops to a slow rate (17) in the following period, reaching basal levels or above-basal levels over the next few days (14). GS has been shown to be highly activated after glycogen-exhausting exercise and to decline from early to late recovery (6). The regulation of glycogenesis and GS activation during glycogen recovery could follow the model of control by glucose transport phosphorylation, provided that glucose 6-P rises during recovery (6). However, quantification of glucose 6-P by NMR in human muscle during glycogen-depleting exercise and recovery (16) showed that glucose 6-P increased immediately after initiation of exercise and during the period of glycogen breakdown; but, during early insulin-independent recovery of glycogen, glucose 6-P levels progressively declined to resting concentrations. Therefore, these data do not seem to support an increase in glucose 6-P concentrations as the leading cause of glycogen synthesis during recovery, because both parameters were changing inversely over time. Alternatively, on the basis of the inverse correlation between glycogen content and glycogen synthesis rate (3, 10, 15, 17), it has been proposed that muscle glycogen repletion during the initial phase is dependent on the extent of glycogen depletion. Glycogen resynthesis may also depend on the balance between the activities of degrading glycogen phosphorylase (GP) and synthesizing GS enzymes. The proportion of active GP may be a determinant for glycogen recovery because activated glycogenolysis may prevent glycogen deposition. Thus the transient inactivation of the enzyme at the onset of glycogen replenishment was proposed to be a determinant in the progression of the synthesis (7). Nevertheless, the fact that GP inactivation reverses within 20 min, whereas glycogen deposition continues, contradicts this assumption and suggests allosteric inhibition of the enzyme.

In this work, the role of acute glycogen depletion versus increments in glucose 6-P in the control of glycogen recovery was evaluated in cells overexpressing GP with enhanced glycogenolytic capacity (4). GP-engineered cells showed lower glycogen content and identical glucose 6-P levels after glucose deprivation but increased GS activation compared with controls. On reincubation with glucose, GP-engineered cells showed enhanced glycogen resynthesis without differences in glucose 6-P. In summary, our results support the hypothesis that transient activation of GS in response to glucose 6-P increments has no impact on the glycogenic flux, whereas the activation of GS induced by acute deglycogenation strongly correlates with the glycogenic rate during early recovery of glycogen stores.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Human muscle primary cultures. Human muscle primary cultures were initiated from satellite cells of muscle biopsies from nine patients considered free of muscle disease after diagnostic studies were reported (biopsies were obtained with informed consent and approval of the Human Use Committee of the Hospital Clínic, Barcelona). Aneural muscle cultures were established in a monolayer according to the technique described by Askanas and Gallez-Hawkins. (2). Cultures were grown in DMEM-M-199 medium, 3:1, supplemented with 10% fetal bovine serum (FBS), 10 µg/ml insulin (Sigma), 2 mM glutamine (Sigma), 25 ng/ml fibroblast growth factor (FGF), and 10 ng/ml epidermal growth factor (EGF) (Becton and Dickinson). Immediately after myoblast fusion, cells were rinsed in Hanks' balanced salt solution, and the medium was replaced by a medium devoid of FGF, EGF, and glutamine. Cultures were fed twice a week and examined daily by phase-contrast inverted microscopy. Muscle cultures were maintained in this medium for up to 4 wk. All cultures were kept at 37°C in a humidified 5% CO₂ atmosphere.

Transduction with recombinant adenovirus. Control recombinant adenovirus and adenovirus inserting a 2.56-kb fragment of rabbit muscle GP cDNA including the entire coding region (AdCMV-MGP) have been described elsewhere (13). The recombinant viruses were amplified in 293 cells, and viral stocks of 8 × 10⁸ plaque-forming units /ml were prepared in 10% FBS-DMEM. Gene delivery to muscle cultures was achieved by exposing 10-day-old fibers induced to fuse by removal of growth factors to the virus for 2 h at a multiplicity of infection of 10. All studies were performed 7 days after infection.

Enzyme activity assays. To measure enzyme activities, 100 µl of homogenization buffer consisting of (in mM) 10 Tris · HCl (pH 7.0), 150 KF, 15 EDTA, 600 sucrose, 15 2-mercaptoethanol, 1 benzamidine, 1 phenylmethylsulfonyl fluoride, and 10 µg/ml leupeptin was used to scrape frozen plates containing the cell monolayers and for subsequent sonication. Homogenates were collected in Eppendorf tubes and centrifuged at 10,500 g at 4°C for 15 min, and the resulting supernatants were used for the determination of enzyme activities. Protein concentration was measured as described by Bradford (6a) using the Bio-Rad protein assay reagent. GP activity was determined by the incorporation of [U-¹⁴C]glucose 1-phosphate into glycogen in the absence or presence of the allosteric activator AMP (5 mM) (12). GS activity was measured in the absence or presence of 10 mM glucose 6-P as described (23).

Metabolite determinations. To measure glycogen content, cell monolayers were scraped into 30% KOH and the homogenates were boiled for 15 min and centrifuged at 5,000 g for 15 min. An aliquot of the supernatants was used for the measurement of protein concentration. Supernatants were spotted onto Whatman 31ET paper, and glycogen was precipitated by immersing the papers in ice-cold 66% ethanol. Dried papers containing precipitated glycogen were incubated in 0.4 M acetate buffer (pH 4.8) with 25 U/ml of -amyloglucosidase (Sigma) for 90 min at 37°C. Glucose released from glycogen was measured enzymatically in a Cobas-Bio autoanalyzer with a GlucoQuant (Boehringer Mannheim) kit. Glucose 6-P concentration was measured enzymatically in neutralized HClO₄ extracts.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Glycogenolysis and glycogen resynthesis. Glycogen content and mobilization in response to glucose deprivation were evaluated in cells exposed to control virus or AdCMV-MGP virus, showing GP total activity levels ~10 times higher (715 ± 33 mU/mg protein) than controls (62 ± 2 mU/mg protein). Glycogen concentration in AdCMV-MGP cells continuously incubated with 25 mM glucose on gene transfer was not different from that of controls (Fig. 1). Deprivation of glucose for 6 h caused a 40% reduction in control cells in the concentration of glycogen, which reached a value of 159 ± 14 µg glucose/mg protein. In AdCMV-MGP cells, a faster and greater decline in the glycogen content was observed, reaching levels (59 ± 3 µg glucose/mg protein) 80% lower at 6 h than basal levels.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Glycogen mobilization on glucose and insulin deprivation. Cells were treated with control (, ) or AdCMV-MGP (, ) virus and incubated with 25 mM glucose and 1 µM insulin. Seven days after transduction, glycogen content was determined (t0). Cells incubated with 25 mM glucose were then deprived of insulin for 24 h (, ), then they were deprived of glucose for indicated times. Data are means ± SE of 3 independent experiments performed in triplicate.

Resynthesis of glycogen on reincubation with 25 mM glucose was monitored (Fig. 2). Glycogen resynthesis was nonlinear. In control cells, there was no rapid recovery, because during the 1st h postloading no resynthesis of glycogen occurred. From 1 to 4 h of recovery, glycogen was slowly replenished at a constant rate of 0.23 µg · mg protein¹ · min¹, reaching 60% of predepletion levels after 4 h. During this period, addition of insulin stimulated the glycogen repletion rate up to a value of 0.40 µg · mg protein¹ · min¹ and led to glycogen levels 80% of predepletion at 4 h. After 4 h, glycogen resynthesis dropped to 0.05 µg · mg protein¹ · min¹, which was not modified by insulin treatment. At 24 h, glycogen levels were slightly lower than those before glucose depletion, whereas over the next day, glycogen reached initial levels in noninsulin-treated cells or was 24% higher than initial levels in insulin-treated cells (data not shown). In AdCMV-MGP cells, the initial (0-1 h) glycogen resynthesis rate was the highest (1.41 µg · mg protein¹ · min¹) and essentially independent of insulin (Fig. 2). This first rapid phase progressed until glycogen concentration reached levels close to those in glucose-deprived control cells (~154 µg glucose/mg protein). The rate over the next 1-4 h of recovery declined to a constant value of 0.53 µg · mg protein¹ · min¹. During this phase, insulin slightly stimulated the rate of resynthesis up to 0.63 µg · mg protein¹ · min¹. Over the next period, 4-48 h, the resynthesis rate dropped further to a mean value of 0.1 µg · mg protein¹ · min¹ with no insulin effect. As a result of increased resynthesis, glycogen levels at 4 h postrecovery were already slightly higher in AdCMV-MGP cells than in controls. Moreover, cells expressing higher GP exhibited supercompensated glycogen levels at 48 h that were 2.2- and 2.7-fold above predepletion in the absence or presence of insulin, respectively (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Recovery of glycogen after depletion. Glycogen-depleted control (, ) or AdCMV-MGP (, ) cells after 6 h deprivation of glucose and insulin were reincubated with 25 mM glucose in presence (, ) or absence (, ) of 100 nM insulin. Glycogen content was determined at different reincubation times with glucose. Data are means ± SE of 5 independent experiments performed in triplicate. A magnification of data corresponding to first 2 h is included.

GP and GS activity during glycogen recovery. AdCMV-MGP cells showed higher levels of active GP (independent of AMP activation) (450 ± 20 mU/mg protein) than controls (43 ± 16 mU/mg protein) after 6 h of glucose deprivation, although GP activity ratio was very similar in both cell systems (Fig. 3). Reincubation with 25 mM glucose resulted in the transient inactivation of GP, with maximal inactivation occurring at 15 min and recovery of basal levels after ~120 min of glucose incubation. The degree of glucose-induced GP inactivation was also very similar (~20-30%) in control (Fig. 3A) and AdCMV-MGP cells (Fig. 3B). Insulin treatment did not cause an additional inactivation of GP compared with that induced by 25 mM glucose in either cell type.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Time course of glycogen phosphorylase inactivation during recovery. Cells treated with control (, ) (A) or AdCMV-MGP (, ) virus (B) were depleted of insulin for 24 h and glucose for 6 h and then incubated with 25 mM glucose in absence (, ) or presence (, ) of 100 nM insulin. Results are means ± SE of 4 independent experiments performed in duplicate.

Total GS activity in cells incubated persistently with 25 mM glucose was higher in AdCMV-MGP cells (19 ± 3 mU/mg of protein) than in controls (11 ± 1 mU/mg of protein), as previously shown (4). Before glucose deprivation, GS activity ratio, defined as fractional activity independent of glucose 6-P, was similar in controls (0.16 ± 0.007) and AdCMV-MGP cells (0.20 ± 0.01). In contrast, after glucose deprivation the activity ratio was twofold higher in engineered cells versus control. The time course of GS activity ratio after reincubating glycogen-depleted cells with 25 mM glucose was determined (Fig. 4). In control cells, glucose caused a time-dependent increase in GS activity ratio that peaked within 15-30 min after glucose addition (increment of 85%) and returned to near basal values after 2 h. In AdCMV-MGP cells, a much smaller increase in GS activity ratio (12%) was found, which peaked within 15-45 min. Thereafter, GS activation progressively declined, reaching values similar to those in controls 16 h after the glucose addition. However, because total GS activity was higher in AdCMV-MGP cells, the levels of active GS were increased in these cells (3.33 ± 0.25 mU/mg protein) compared with controls (2.26 ± 0.17 mU/mg protein) along the time studied (16 h). Incubation with 100 nM insulin together with 25 mM glucose induced in control cells an additional activation of 10% above that induced by glucose alone; the additional activation by glucose and insulin did not reach statistical significance in cells overexpressing GP. The correlation between the activity ratio of GS and glycogen levels was studied during recovery from 0 to 16 h, leaving out the transient short-term (15-45 min) GS activation induced by glucose incubation. An inverse correlation was observed when plotting the values for AdCMV-MGP cells (Fig. 5B). In these cells, changes in GS activation were found in conjunction with marked changes in the glycogen content from ~60 to 500 µg glucose/mg protein. In contrast, no correlation between GS activity ratio and glycogen content was found in control cells (Fig. 5A). This observation appears to be related to the slight glycogen depletion in response to glucose deprivation to 150 µg glucose/mg in control cells and a minor increase during recovery up to 250 µg glucose/mg protein.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Time course of glycogen synthase activation during recovery. Cells treated with control (, ) or AdCMV-MGP (, ) virus were depleted of insulin for 24 h and glucose for 6 h and then incubated with 25 mM glucose in absence (, ) or presence (, ) of 100 nM insulin. Results are means ± SE of 4 independent experiments performed in duplicate.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Relationship between GS activity ratio and glycogen concentration. Cells treated with control (, ) or AdCMV-MGP (, ) virus were depleted of insulin for 24 h and glucose for 6 h and then incubated with 25 mM glucose in absence (, ) or presence (, ) of 100 nM insulin for different times (0, 1, 2, 4, and 16 h). Data points represent individual observations in control cells (A; n = 29, from 4 independent experiments) and AdCMV-MGP-treated cells (B; n = 29 from 4 independent experiments).

Glucose 6-P concentration after glucose reload. The changes in glucose 6-P concentration in glycogen-depleted cells after reincubation with 25 mM glucose were monitored (Fig. 6). In control cells, glucose 6-P levels increased very rapidly, reaching maximal levels about fivefold already at 8 min after glucose reincubation, which persisted along the time studied up to 2 h. No differences in glucose 6-P accumulation in AdCMV-MGP cells compared with controls were observed. The effect of insulin on glucose 6-P accumulation was determined 8 min after glucose addition. Treatment with insulin caused a very small increase in the accumulation of glucose 6-P 8 min after glucose addition in control (3.9 ± 0.3 vs. 3.3 ± 0.2 nmol/mg protein) and AdCMV-MGP cells (4.8 ± 0.3 vs. 3.2 ± 0.2 nmol/mg protein). The higher insulin effect on AdCMV-MGP cells compared with controls may be related to the higher content of GLUT-4 determined in these cells (5).

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Time course of glucose 6-P levels during glycogen repletion. Cells treated with control () or AdCMV-MGP () virus were depleted of insulin (24 h) and glucose (6 h) and then incubated with 25 mM glucose for varying times. Results are means ± SE of 4 independent experiments performed in duplicate.

Dose dependence of glucose 6-P, GS, and glycogen content. The dose dependence of glucose uptake and glycogen synthesis on glucose concentration during recovery was analyzed (Fig. 7). Glucose 6-P was accumulated in control cells with a positive relation to the increase in extracellular glucose concentration (Fig. 7A). A biphasic dose response was determined, with maximal accumulation occurring as glucose increased from 0 to 5 mM, and very little additional accumulation occurred at higher concentrations. This may be explained by the autoinhibition of hexokinase II by glucose 6-P (26) and the autoinhibition of glucose transport (20). AdCMV-MGP cells exhibit no differences in glucose 6-P levels in response to increasing glucose concentrations compared with controls. The GS activity ratio peak rose in control cells with a similar glucose concentration pattern to glucose 6-P (Fig. 7B). In AdCMV-MGP cells, a dose-dependent effect was also determined, although, as observed in the time course study, the glucose-dependent effect was markedly reduced because of high basal GS activation. Glycogen content at 16 h of glucose reincubation increased with extracellular glucose concentration, but the dose-response correlated with glucose 6-P accumulation both in control and AdCMV-MGP cells (Fig. 7C). Glycogen accumulation in AdCMV-MGP cells was higher than in controls at glucose concentrations >5 mM, whereas at 2.5 mM glucose it was lower, probably because of the insufficient allosteric inhibition of the active phosphorylase present at higher levels in AdCMV-MGP cells.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Glucose concentration dependence of glucose 6-P, glycogen synthase, and glycogen content. Cells treated with control () or AdCMV-MGP () virus were depleted of insulin for 24 h and glucose for 6 h, and then they were incubated with no glucose or varying glucose concentrations. Glucose 6-P levels (A), glycogen synthase activity ratio (B), and glycogen content (C) were determined at 2 h, 30 min, and 16 h, respectively. Results are means ± SE of at least 3 independent experiments performed in triplicate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this study, the contribution of glycogen exhaustion versus glucose 6-P in the control of muscle glycogen resynthesis and GS activation was analyzed. A cultured human muscle model overexpressing GP (AdCMV-MGP cells) was used to achieve extensive glycogen depletion. Despite exhibiting 10 times higher active GP levels, these cells showed unaltered net glycogen content compared with controls when incubated with 25 mM glucose, but responded to glucose deprivation (6 h) with a more acute glycogen depletion, reaching levels (59 ± 3 µg glucose/mg protein) 80% lower than predepletion compared with levels (159 ± 14 µg/mg protein) only 40% lower in controls. Glycogen resynthesis, initiated by reincubating the cells with 25 mM glucose, was nonlinear, as has been characterized in vivo in rat (11) and human muscle (16, 17). In control cells, there was no immediate resynthesis of glycogen (during the 1st h) and the highest rate of resynthesis (0.2 µg · mg protein¹ · min¹) was observed during 1-4 h postloading, which markedly dropped up to 48 h. Insulin increased (1.7-fold) the resynthesis rate during the first 4 h, whereas it had no effect on the resynthesis rate during the next period (4-48 h). In contrast, in acutely deglycogenated GP-overexpresssing cells, glycogen recovery started immediately at the highest rate (1.41 µg · mg protein¹ · min¹) corresponding to the 0-1 h period, whereas the second phase (1-4 h) proceeded at a constant rate of 0.43 µg · mg protein¹ · min¹. This pattern strongly resembles the recovery after glycogen-depleting exercise in rats (11) and humans (17), where a rapid early (0-1 h) recovery has been described. Moreover, the rapid recovery phase in AdCMV-MGP cells similar to that defined in in vivo human muscle in normal (17) or diabetic individuals (16) did not require insulin stimulation, although addition of insulin increased the recovery rate by 35%. The rapid phase progressed as glycogen increased to a concentration, estimated at ~160 µg glucose/mg protein, which was approximately the same as before the glucose load in control cells. Therefore, the rapid early resynthesis appeared to depend on the depletion of glycogen below this threshold and progression to the recovery of this level. Consistently, control cells, which did not decrease glycogen below this proposed threshold, did not show rapid glycogen recovery. Similar conclusions were obtained by Price and colleagues (17) when studying glycogen recovery after differing intensity exercise in humans by NMR. They found different resynthesis rates depending on the degree of glycogen depletion, suggesting that a glycogen concentration threshold is required for rapid initial glycogen resynthesis.

Glycogen metabolizing enzymes were, in both cell types, inversely regulated by glucose, as anticipated, with a maximal effect within 30 min that resumed 60 min after glucose exposure. GP was inactivated to a similar degree in control and AdCMV-MGP cells, although active GP levels in GP-engineered cells were ~10-fold higher due to the higher GP expression. Despite this, AdCMV-MGP cells showed net glycogen synthesis on incubation with glucose in the range from 2.5 to 25 mM, and the glycogen content was higher than in controls at glucose concentrations >5 mM. Therefore, it is shown that active GP may coexist with glycogen deposition as occurs in resting muscle where active GP does not induce glycogen mobilization. A possible explanation is that the dephosphorylated GP is allosterically inhibited by intracellular metabolites such as glucose or glucose 6-P (8). Concomitant addition of glucose plus insulin did not exert any additional effect on the glucose-induced GP inactivation, indicating that the stimulation of glycogen synthesis by insulin is not due to an enhanced inactivation of the glycogenolytic enzyme. A lack of effect of insulin on GP activation in human muscle biopsies was also reported by Yki-Jarvinen and colleagues (28) when studying the effect of varying combinations of glycemia and insulinemia in vivo. GS activity was markedly different in GP-overexpressing cells from controls. A moderate elevation in the total activity of the enzyme was associated with the higher expression of GP, which is not due to a transcriptional effect as we previously described (4). Importantly, AdCMV-MGP-treated cells showed two times higher GS activity ratio than controls after glucose deprivation. This change correlated with lower glycogen content, whereas no differences in glucose 6-P levels between both cell types were found at the end of the glucose deprivation period or after reincubation with 25 mM glucose. GS was transiently activated on glucose reincubation in both controls and AdCMV-MGP-treated cells, and the peak of activation dose dependently correlated with glucose 6-P concentration, as consistently demonstrated (24). Insulin caused a slight additional activation of GS, which corresponded to an additional increase in glucose 6-P. However, the glucose-dependent activation was blunted in AdCMV-MGP cells, probably because GS was almost maximally activated in AdCMV-MGP-treated cells and very little additional effect might be achieved by raising glucose 6-P. In these cells, GS activation correlated inversely with glycogen content along 16 h of recovery. Therefore, AdCMV-MGP cells showed the same inverse correlation between glycogen and GS activation in muscle as has been reported after glycogen-depleting exercise (3, 9, 17, 27) and during recovery (7).

Therefore, the control of glycogen resynthesis rate in cultured muscle was not correlated to the transient covalent changes in GS induced by the increase in glucose 6-P after glucose reload. Although a positive dose-dependent relationship was found between glucose 6-P levels, GS activation, and glycogen content in control cells, there was no time relationship between GS activation and glycogenesis. Although GS activation resumed during the 1st h, glycogen synthesis was maximal from 1 to 4 h in controls. These observations may be concordant with the hypothesis of Shulman and colleagues (21, 22) that GS phosphorylation is a mechanism of adapting to glucose 6-P availability rather than controlling glycogen synthesis flux. In contrast, after acute deglycogenation, as in glucose-deprived AdCMV-MGP cells, GS activation, which inversely correlated with glycogen content, had a positive impact on the control of glycogen resynthesis flux. Thus the marked drop in GS activation 1 h after glucose incubation coincided with a marked decline in the glycogen resynthesis rate. No control effect could be attributed to glucose 6-P as a precursor, because glucose 6-P concentration was identical in AdCMV-MGP cells and controls. Similarly, no support for the role of glucose 6-P as a leading cause of glycogen synthesis during recovery was found by NMR examination of the time courses of intramuscular glucose 6-P concentration and glycogen synthesis (16), indicating that glucose 6-P levels declined over the period of early rapid glycogen recovery in vivo.

In summary, we show that in cultured muscle, lowering glycogen below a certain threshold leads to the activation of GS and that the activated GS controls the glycogenic flux. Therefore, during recovery from exhaustion, glycogen is not playing a mere buffer role for maintaining glucose 6-P homeostasis but rather has an active role in regulating GS activity and glycogenesis. Finally, we show that whatever the mechanism by which acute deglycogenation activates GS, it promotes the control of the glycogenic flux. Such effect is enhanced by the overexpression of GP, and, as a result, a rapid recovery of glycogen, which is not dependent on glucose 6-P increments, takes place.

Perspectives