| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Center for Designing Foods to Improve Nutrition, Department of Food Science and Human Nutrition, Iowa State University, Ames, Iowa; 2Department of Surgery, Vanderbilt University Medical Center, Nashville, Tennesse; and 3AminoScience Laboratories, Ajinomoto Co. Inc., Kawasaki, Japan
Submitted 15 March 2005 ; accepted in final form 14 July 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
isotopic tracer; glucose kinetics; net organ balance; gluconeogenic amino acid
Whether glutamine availability is limiting during and after the stress of exercise is not uniformly clear. This may partially be due to differences in study design, including the length and intensity of exercise, or the fact that some investigators report values for blood and some for muscle and a glutamine concentration gradient exists between blood and muscle. Circulating glutamine has been demonstrated to increase with exercise (4). However, others have shown that muscle glutamine progressively decreases with exercise in swimming rats (10), circulating glutamine decreases in running dogs (13), and the net loss of glutamine from muscle is greater with exercise in both dogs (35) and humans (12). After exercise, however, the majority of data suggest that glutamine availability is reduced, particularly after strenuous exercise (13). This has led to the hypothesis that reduced circulating glutamine may be an indicator of exercise stress and overtraining (21, 29).
Even less is known about the role glutamine availability plays in exercise-related metabolism. Recent studies have suggested important interactions between glutamine and carbohydrate homeostasis (3, 5, 20, 23, 30). Glutamine carbon has potential to enter the Krebs cycle through 
-ketoglutarate, thereby providing carbon for gluconeogenesis (1). Glutamine has also been shown to interact with glucose utilization, stimulating whole body glucose utilization and hindlimb glucose uptake during hyperinsulinemic euglycemia in postabsorptive dogs (5).
Exercise elevates whole body glucose utilization to meet fuel demands in both exercising and nonexercising tissues (14, 31, 33). On the other hand, the liver responds to exercise by increasing glucose production to meet these increased demands for glucose (31, 33). After exercise, glucose production and utilization initially remain elevated, and insulin sensitivity is enhanced to facilitate glycogen replenishment in the liver and skeletal muscle tissues (22, 26, 28). Although glutamine appears to interact with glucose production and utilization, it has not been elucidated whether glutamine influences glucose production and utilization during and after exercise. Furthermore, while there is considerable commercial interest in using glutamine as an ergogenic aid, few studies have investigated the interactions of glutamine with glucose metabolism during and after exercise (7, 32).
Therefore, the following study was conducted to test the hypotheses that 1) increased glutamine availability would further stimulate glucose production and utilization during exercise and 2) increased glutamine availability after exercise would enhance glucose production and insulin-mediated glucose utilization. Therefore, to investigate the interactions between glutamine availability and glucose metabolism during and after exercise, isotope-dilution and organ-balance techniques were used in exercising multicatheterized dogs. The exercise period represents the influence of glutamine during a situation when both glucose production and utilization are stimulated. In the postexercise period, a hyperinsulinemic-euglycemic clamp technique was used to represent a condition similar to postexercise carbohydrate supplementation, when accelerated glycogen repletion is beneficial but glucose production is blunted by insulin.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
31% protein, 52% carbohydrate, 11% fat, and 6% fiber. Body weight was unchanged from the day of surgery to the day of the experiment. Surgical procedures were conducted as previously described (5). Briefly, the dogs were placed under anesthesia (20 mg/kg pentothal sodium and 1.0% isoflurane inhalation) and intubated, mechanical ventilation was provided, and a midline laparotomy was performed for placement of indwelling catheters and flow probes. Silastic catheters (0.04 in. ID; Dow Corning, Midline, MI) were inserted into the left femoral artery and into the left common hepatic, portal, and right common iliac veins for the sampling of blood. A catheter was also placed in the vena cava for infusions of D-[3-3H]glucose (New England Nuclear, Boston, MA) and glutamine (Ajinomoto, Kawasaki, Japan). Catheters were also placed in the splenic and jejunal veins for intraportal delivery of insulin. The catheters were filled with heparinized saline (200 U/ml; Abbott Laboratories, North Chicago, IL), and their free ends were knotted and placed in a subcutaneous pouch until the day of the metabolic study. Doppler blood flow cuffs (Transonic Systems, Ithaca, NY) were placed around the right external iliac artery, the portal vein, and the hepatic artery to measure hindlimb, portal vein, and hepatic artery blood flows, respectively. These leads also were placed in a subcutaneous pocket.
Each dog was allowed to recover from surgery for 18 days before the metabolic study. One week after surgery, dogs were trained to exercise on a motorized treadmill. Animals were not exercised during the 48 h preceding the metabolic study. Before study, the dogs had to exhibit good health, including 1) consumption of daily food rations, 2) normal stools, 3) blood leukocyte count <18,000/ mm, and 4) hematocrit >38%. The dogs were fasted overnight (18 h) before the metabolic study.
On the morning of the metabolic study, all catheters and blood flow cuff leads were removed from the subcutaneous pouches using local anesthesia (2% lidocaine; Astra Pharmaceutical, Worcester, MA). The catheters were aspirated to remove any contents and were flushed with saline. The dogs were then placed in a Pavlov harness, and infusion of 1 U/ml heparinized saline (0.1 ml/min) was started via the arterial catheter to maintain patency and to replace the volume of blood removed during experimentation. In addition, 18-gauge angiocatheters (Becton-Dickinson, Sandy, UT) were inserted percutaneously into a cephalic vein for infusion of glucose as needed.
Experimental protocol. The experimental protocol used in the metabolic study is depicted in Fig. 1. Each metabolic study consisted of a 120-min equilibration period (–150 to –30 min), a 30-min basal sampling period (–30 to 0 min), a 90-min moderate-intensity exercise period (0 to 90 min), and a 150-min postexercise period (90 to 240 min). A bolus infusion of D-[3-3H]glucose (30 µCi) was given to initiate the study. This was followed by a constant infusion of D-[3-3H]glucose (20 µCi/h) throughout the 390-min study. Arterial, femoral venous, portal venous, and hepatic venous blood samples were drawn every 30 min starting at the end of the equilibration period (time = –30, 0, 30, 60, 90, 120, 150, 180, 210, and 240 min). At simultaneous time points, blood flow measurements from Doppler flow probes were recorded from continuous online data. Data for plasma AA and net organ AA balance are taken from the last sample of each sampling period (0, 90, and 240 min). The other parameters are average of all samplings in each period. Plasma flow was determined from blood flow using hematocrit. Caloric expenditures were measured by indirect calorimetry (2900 Metabolic Measurement Cart, SensorMedics, Loma Linda, CA) during the basal and postexercise periods.
|
Each dog was subjected to two test protocols on separate days: 1) a constant infusion of glutamine (Gln) at 12 µmol·kg–1·min–1 in normal saline (0.12 ml/min) or 2) a constant infusion of the same volume of normal saline (Con; 0.12 ml/min) during the exercise and postexercise periods. There were at least 2 wk between the two experimental treatment days, and the order in which these treatments were given was random.
Analytic procedures. Analytical methods were conducted as previously described (5). Briefly, the blood samples were collected in heparinized syringes and placed in tubes containing Na2EDTA for analyses of plasma. The blood was spun in a refrigerated (4°C) desktop centrifuge at 3,000 g for 10 min. Plasma was collected and immediately placed on ice until analysis. Blood glucose concentration was determined by a glucose analyzer (Beckman Instruments, Palo Alto, CA). Plasma glucose specific activity was determined by liquid scintillation counting after cold precipitation of plasma protein with 0.066 N barium hydroxide and 0.066 N zinc sulfate. Aliquots of arterial plasma were transferred to separate tubes and stored at –70°C for later determination of insulin and glucagon concentrations. Immunoactive insulin and glucagon were determined using a disequilibirum double-antibody RIA system (Linco Research, St. Charles, MO) with a coefficient of variation of 5%. Plasma AA concentrations were determined by using a EZ:faast gas chromatography-mass spectrometry analysis kit (Phenomenex, Torrance, CA). The kit consists of a solid phase extraction step, followed by a derivatization procedure. Derivatized samples were analyzed by gas chromatography-mass spectrometry (model 6890/5973, Agilient Technologies).
Calculations. Tracer-determined glucose rates of appearance (Ra) and disappearance (Rd) as estimates of whole body glucose production and utilization were calculated according to the isotope-dilution method outlined by Altszuler et al. (2). Whole body endogenous glucose Ra was calculated by subtracting the exogenous glucose infusion rate from the total Ra. Note that endogenous glucose production represents both hepatic and extrahepatic (e.g., renal) glucose production, and it thus could be a slight overestimate of hepatic glucose production. Because of the constant tracer infusion and the increases in glucose from increased glucose production and glucose infusion, the specific activity of plasma glucose fell during the exercise and postexercise periods (8,589 ± 1,062, 5,791 ± 644, and 3,612 ± 189 dpm/mg for basal, exercise, and postexercise periods for Con and 6,131 ± 1,000, 3,737 ± 857, 2,514 ± 492 dpm/mg for Gln, respectively).
Net hepatic and hindlimb glucose balances were calculated by the same calculations as previously described using arterial-venous differences and blood flow measurements (5). Hepatic flow of blood into the liver is carried by both the portal vein and the hepatic artery, and the entire flow of blood out of the liver is carried by the hepatic vein. Blood flow rates in both the portal vein and hepatic artery were measured by Doppler flow probes. Thus the equation for net hepatic balance is as follows:
 |
Statistical analysis. All data analyses were done using SAS statistical software packages (SAS Institute, Cary, NC). The variables from the time points within each period were averaged across the period for each experimental unit similar to an area under the curve. The data are expressed as means ± SE for each period and treatment. Two-way ANOVA (treatment x period) with repeated measures was employed to identify overall significant model effects. Significant differences among means were determined by pairwise contrast comparisons between Con and Gln (treatment) or between basal and either exercise or postexercise periods (period within treatment). Differences were considered significant at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Exercise period. During exercise, plasma glucose significantly fell by 7 mg/dl in Con, but this fall was blunted in Gln (Table 1). Insulin decreased (P < 0.05), glucagon increased (P < 0.05), and hindlimb blood flow was accelerated (P < 0.05) similarly in both treatment groups during exercise. Hepatic artery and portal vein blood flow was not different between treatment groups during exercise.
|
4-fold) in both treatment groups (Fig. 5).
|
|
|
|
|
|
|
The rate of exogenous glucose infusion needed to maintain euglycemia during hyperinsulinemia was not significantly different between Con and Gln (8.32 ± 0.77 vs. 7.10 ± 0.47 mg·kg–1·min–1). Endogenous glucose production with Con significantly decreased from basal to near zero due to the hyperinsulinemic-euglycemia (Fig. 2). Conversely, glucose production did not fall below basal levels and was significantly higher in Gln. Glucose utilization was significantly increased during the postexercise period in both groups (Fig. 3), but glucose utilization was 16% greater in Gln than Con (P < 0.05). The liver switched from net production to net utilization of glucose similarly in both groups, suggesting an active replenishment of hepatic glycogen stores (Fig. 4). Net uptake of glucose by the hindlimb was significantly increased from basal for Gln, but there was no significant change from basal with Con (Fig. 5).
Plasma glutamine concentration was twofold greater in Gln than Con during the postexercise period (Table 2). Other AA were not significantly different between treatments.
Net hepatic uptake of glutamine was over threefold greater for Gln than Con during the postexercise period (Table 3). However, the net uptake of other individual AA was not significantly different between treatments. Net hepatic uptake of total gluconeogenic AA was significantly higher in Gln than in Con (Table 3), primarily due to differences in Gln uptake. Net hindlimb balance of AA was not significantly different between groups (Table 4).
Resting energy expenditure was not significantly different for Gln vs. Con (Table 5). Furthermore, nonprotein respiratory quotients were similar between groups.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
During exercise, hormonal and neural conditions are suitable for stimulating glucose production. Circulating concentrations of plasma glucagon and epinephrine are increased, and insulin is slightly decreased (33). In our study, exercise increased glucose production by 48% in Con. Increased glutamine availability further stimulated glucose production 24% during exercise without differences in the hormonal milieu vs. Con. As far as we know, this is the first evidence that glutamine interacts with exercise to increase endogenous glucose production. Stoichiometrically, the amount of glutamine infused during exercise had the theoretical potential to form glucose at the rate of 1.1 mg·kg–1·min–1. The additional glucose produced was 1.3 mg·kg–1·min–1, suggesting that not all of the glucose could be accounted for via gluconeogenesis from glutamine.
It is important to maintain plasma glucose during exercise, because circulating glucose provides energy for muscle contraction and is the primary source of energy used by neural tissue during exercise. In the present study, the plasma glucose pool gradually decreased as exercise continued in the Con group. However, plasma glucose was maintained at basal levels with Gln, suggesting that an important fate of the increased glucose production was to sustain the total body glucose pool during exercise.
Gluconeogenesis normally contributes 20% of total glucose production during 90 min of exercise in overnight-fasted dogs (33). It would be anticipated that the fractional contribution of gluconeogenesis to total hepatic glucose production would be greater, if gluconeogenic precursors were supplemented during exercise. We found that increased glutamine availability during exercise enhanced hepatic glutamine and alanine uptake, net hepatic glucose production, and total whole body glucose production. Roef et al. (27) demonstrated that infusion of another glucogenic precursor (lactate) during low-intensity exercise increased both total glucose production and contribution of gluconeogenesis to glucose production compared with bicarbonate infusion. In addition, Miller et al. (17) demonstrated that incorporation of 13C from tracer lactate into blood glucose was increased with a lactate clamp (4 mM) during moderate exercise. Together these previous studies and the present study support the concept that gluconeogenic precursors can be significant sources of glucose during exercise. The efficacy of providing glucose precursors vs. glucose itself, however, cannot be established from the present study and requires further evaluation.
It has been speculated that the rate of Krebs cycle flux and concentration of Krebs cycle intermediates may be limiting to exercise performance (6). However, when oral supplementation of glutamine was used to increase the Krebs cycle intermediate pool in muscle, neither endurance capacity nor the degree of phosphocreatine depletion or lactate accumulation was altered (6). In a similar study, an increase in Krebs cycle intermediate pool did not have a significant increase in oxidative energy production (8), suggesting that glutamine may promote anaplerosis, but its potential translation to direct improvements in performance may be limiting.
Net hepatic glutamine and alanine uptake was greater in Gln than in Con during exercise. Surprisingly, those differences between groups existed despite the fact that plasma levels of alanine were not different between treatments. These data suggest that interaction of glutamine and exercise may promote hepatic extraction of the gluconeogenic AA. Perriello et al. (24) also demonstrated that glutamine supplementation increased glucose production from glutamine and also from alanine even in the absence of a change in plasma alanine concentrations. Glutamine also has been suggested to activate gluconeogenic enzymes, such as phosphoenolpyruvate carboxykinase in rat kidney or fructose-1,6-bisphosphatase in cultured HeLa cells and fibroblasts (30). The enzymes of glucogenic pathway in the liver may be stimulated by glutamine. While the present study does not provide such data, one may speculate that glutamine regulates gluconeogeneis via stimulation of hepatic gluconeogenic AA transport systems.
The extension of these results to relevant situations in humans needs to be done with care. Glucose metabolism in canines and humans responds to exercise and euglycemic hyperinsulinemia similarly (34), and certainly qualitative changes and conclusions would be expected to be applicable to exercising humans. However, extending these results in healthy dogs undergoing intravenous infusions of glutamine to humans under the conditions of postsurgical stress, trauma, or diabetes mellitus would require further research to determine whether glucose homeostasis could be altered with glutamine under these specific conditions.
Furthermore, there were several options available for the delivery of glutamine. Previous human studies have used both an intravenous (32) and an oral (7) route to deliver glutamine. Oral intake of glutamine is certainly the closest to what would occur under practical conditions of exercise training. However, glutamine delivered orally is subject to differences in absorption and gastrointestinal tract degradation. Oral glutamine also could potentially create gastrointestinal distress. Therefore, glutamine was delivered via a peripheral vein in the present study to provide the best route for a controlled metabolic study, and the data need to be considered in this context.
In the present study, whole body glucose utilization was higher with Gln than Con during exercise, but where this increased glucose utilization occurred is unclear. The main fate of glucose taken up by tissues during exercise is oxidation, and skeletal muscle is quantitatively the most important site of glucose utilization during exercise. However, net hindlimb glucose uptake was not significantly increased with Gln during exercise in the present study, suggesting that the rate of hindlimb glucose utilization is regulated more by exercise itself than by substrate availability. This casts doubt on the potential for increased glutamine availability during exercise to directly improve performance.
The liver is also a metabolically active tissue during exercise, and hepatic glucose utilization has been shown to be increased during exercise (33). The liver needs fuel to facilitate its metabolic activity during exercise, and it is possible that increased hepatic utilization of glucose with Gln during exercise was responsible for the enhanced whole body glucose utilization observed. However, this cannot be assessed because only net hepatic balance was measured. Another important fuel during exercise, fat, has been shown to be spared from hepatic oxidation due to glutamine supplementation (9, 11). While Randle et al. (25) proposed that glucose oxidation is inversely proportional to the quantity of fatty acids available for oxidation, further work is required to understand if glutamine promotes changes in hepatic oxidative substrate metabolism during exercise.
In the present study, a hyperinsulinemic-euglycemic clamp technique was employed during the postexercise period to simulate, in a controlled manner, a postexercise situation where glycogen repletion is accelerated via glucose intake. As would be expected in this situation, glucose production was inhibited to near zero in the control. However, glutamine infusion blunted the inhibitory effect of insulin on glucose production, which is consistent with our previous report using postabsorptive nonexercising dogs (5). Bowtell et al. (7) investigated the effect of glutamine on glucose production after exercise in humans, providing either glutamine alone, glutamine plus glucose polymer, or glucose alone after exercise. In this case, glucose production was greatest in glutamine plus glucose. In the present study, glutamine infusion could theoretically account for only 1.1 of 2.9 mg·kg–1·min–1 endogenous glucose production during the postexercise period, suggesting that other mechanisms made significant contributions to this increase. The increased endogenous glucose production allowed less glucose to be infused (1.2 mg·kg–1·min–1; nonsignificant) and increased whole body glucose utilization (1.4 mg·kg–1·min–1). Interestingly, these total changes (2.6 mg·kg–1·min–1) were very close to the increase in endogenous glucose production (2.9 mg·kg–1·min–1). It may be speculated that if glucose were infused at equal rates in both groups, glucose utilization would have been even greater in the Gln group.
In conclusion, glutamine infusion during exercise increased whole body glucose production, resulting in increased whole body glucose utilization and maintenance of circulating glucose levels. Whether this has potential to translate to improvements in exercise performance needs to be assessed in future studies, but hindlimb glucose utilization was not altered by increased glutamine availability during exercise. Glutamine promoted postexercise glucose production despite hyperinsulinemic euglycemia, and it also augmented insulin's action on whole body glucose utilization. Glutamine enhanced hindlimb glucose uptake after exercise, but the potential for this increased uptake to translate to faster muscle recovery and improved subsequent exercise performance needs to be assessed by future studies.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
DISCLOSURES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A Bassini-Cameron, A Monteiro, A Gomes, J P S Werneck-de-Castro, and L Cameron Glutamine protects against increases in blood ammonia in football players in an exercise intensity-dependent way Br. J. Sports Med., April 1, 2008; 42(4): 260 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Iwashita, C. Mikus, S. Baier, and P. J. Flakoll Glutamine Supplementation Increases Postprandial Energy Expenditure and Fat Oxidation in Humans JPEN J Parenter Enteral Nutr, March 1, 2006; 30(2): 76 - 80. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) were taken every 30-min during basal, exercise, and postexercise periods. Indirect calorimetry (
) was used to measure oxygen consumption and carbon dioxide production rates during basal and postexercise periods. Glutamine (Gln) or saline (Con) were infused intravenously during exercise and postexercise.
P < 0.05 vs. basal.
