INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

GLUTAMATE IS A UNIQUE AMINO acid that is central to many metabolic processes in skeletal muscle. Not only does glutamate participate in the transamination of the branched-chain amino acids, but it is also pivotal to the formation of ammonia, aspartate, alanine, and glutamine (14, 17, 24, 32). Although glutamate is abundant in the skeletal muscle free amino acid pool, it is the primary amino acid taken up by resting and active muscle (10, 14, 17). At the onset of exercise, intramuscular glutamate decreases by ~40-80% and remains low during exercise, despite its constant uptake from circulation (7, 10, 16). During exercise, there is an elevated release of ammonia, glutamine, and alanine from the working muscle, suggesting that glutamate plays an important role in the transfer of amino groups and in the TCA cycle in skeletal muscle (7, 10, 12, 16).

Whereas glutamate is central to numerous reactions, it has rarely been manipulated. Plasma glutamate levels can be elevated by administering large doses of monosodium glutamate (MSG). However, most studies that have used this manipulation have focused on gut and liver metabolism and did not study muscle metabolism. After ingestion of ~100-150 mg/kg body wt of MSG, plasma glutamate levels can be elevated up to 700-800% (4, 9, 15, 25-29). Graham et al. (15) demonstrated that ingesting similar doses of MSG can elevate intramuscular glutamate concentrations at rest by ~40%. In resting subjects, Thomassen et al. (31) infused MSG and observed an elevated influx of glutamate into the leg, as well as a small but significant increase in femoral venous alanine. Both of these studies indicate that an increased availability of circulating glutamate can result in increased intramuscular glutamate at rest. A rise in insulin shortly after MSG administration was also a common observation to both studies (15, 31), implying that glutamate may stimulate insulin secretion at rest, as suggested in previous work (19, 23). Aoki et al. (1) have suggested that this rise in insulin may enhance the uptake of glutamate into muscle in humans. However, it is not known whether glutamate would also moderate the decline in insulin that is normally seen during exercise.

During exercise, when the extraction of glutamate from circulation is increased, it is not known whether elevated circulating glutamate levels from MSG ingestion can be maintained. Thomassen and coworkers (30) administered MSG orally and intravenously to coronary patients, who then exercised at a low intensity for variable periods of time. They showed that glutamate levels remained elevated for the duration of the short-exercise bout and plasma alanine was elevated with oral ingestion; however, these were the only amino acids measured in this study, and ammonia measurements were not included. Because glutamate is central to numerous reactions in amino acid metabolism, elevated glutamate levels could potentially affect ammonia, glutamine, and alanine production; provide more aspartate for the purine nucleotide cycle; and affect anaplerosis of the TCA cycle. As such, increasing circulating glutamate levels via MSG ingestion could ultimately serve as a tool to investigate the role of glutamate in muscle metabolism during exercise.

To further understand glutamate's role in rest and exercise metabolism, normal glutamate metabolism was manipulated by enhancing glutamate availability via MSG ingestion. Because glutamate availability may be functionally important to the TCA cycle flux, the decline in intramuscular glutamate at the onset of exercise needs to be understood. By providing glutamate in excess, glutamate availability in circulation is increased, which, subsequently, may be utilized by the skeletal muscle. In this study, MSG was administered to explore glutamate's effects on whole body metabolism and, indirectly, on muscle metabolism near the onset of exercise. The study was designed to 1) examine whether MSG can successfully elevate plasma glutamate levels and maintain these levels during exercise, 2) examine the shifts in plasma amino acids resulting from glutamate transamination after MSG ingestion during exercise, and 3) further examine the changes that MSG ingestion exerts on insulin at rest and during exercise. As such, we hypothesize that the ingestion of glutamate will result in an increase of circulating glutamate levels at rest and during exercise. This increase in circulating glutamate availability should also lead to an increase in plasma alanine, glutamine, and ammonia because these compounds intimately interact with intramuscular glutamate. Potential shifts in plasma amino acids will indirectly suggest specific alterations in muscle metabolism as a result of enhanced glutamate availability and direct future investigation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Seven healthy men with a mean age, body mass, and maximal oxygen consumption (O_2 max) of 26.7 ± 2.0 yr, 75.8 ± 3.5 kg, and 56.8 ± 2.3 ml · kg¹ · min¹, respectively, volunteered to participate in the study. All subjects were recreationally active to moderately trained individuals. Subjects were informed verbally and in writing of the purpose and risks involved with the experimental protocol, which was approved by the Human Ethics Committee of the University of Guelph.

Experimental protocol. Subjects underwent four different trials: 1) exercise + MSG, 2) exercise + placebo (Plb), 3) rest + MSG, and 4) rest + Plb. The trials were conducted in a double-blind crossover manner and were each separated by at least 1 wk. For each trial, subjects arrived after an overnight fast. Subjects were asked to refrain from caffeine, aspartame, alcohol, and exercise for 2 days before each trial, and they used a checklist food record to maintain a consistent diet. They recorded their diets for 2 days before their first trial and referred to these records to maintain the same diet 2 days before each subsequent trial.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Experimental design. A: rest trials. B: exercise trials. MSG, monosodium glutamate; O2, oxygen consumption.

Analyses. For the analysis of amino acids, ammonia, sodium, chloride, and potassium, blood was collected in tubes containing heparin. These were centrifuged, and the plasma was stored at 80°C. Amino acids were measured by using high-performance liquid chromatography, as previously described (18). The electrolytes were measured by using the NovaStat Profile 9+ (Nova Biomedical, Waltham, MA). For lactate and glucose analysis, 200 µl of whole blood were deproteinized by using 1 ml of 0.6 M perchloric acid and stored at 80°C. A portion of the blood sample was centrifuged, and the serum was stored at 80°C for the analysis of insulin (Coat-A-Count, DPC, Los Angeles, CA), C peptide (Human C-Peptide RIA Kit, Linco Research), free fatty acids (WAKO Chemical USA), and glycerol. Ammonia, glucose, lactate, and glycerol were analyzed by using fluorometric methods (5).

Statistics. Peak values for each amino acid were determined by averaging the peak values of a given amino acid from all subjects. Total amino acids were calculated as the sum of all amino acids measured. The sum of threonine, valine, methionine, isoleucine, leucine, phenylalanine, and lysine was referred to as total essential amino acids. Tryptophan was not included because it could not be measured precisely. Branched-chain amino acids were calculated as the sum of valine, leucine, and isoleucine. All data were analyzed by using a two-way repeated-measures ANOVA (trial × condition). Statistical significance was accepted at P  0.05, and analysis was conducted by using Tukey's post hoc test. All data are expressed as means ±SE. During exercise, the 45-, 55-, and 65-min time points (refer to Fig. 1) do not have time points that correspond to the rest trials. Thus statistical differences between exercise and rest within the same treatment at these particular time points were analyzed in a slightly different manner. The average between 40 and 50 min of the rest trial was compared with 45 min of the exercise trial. Similarly, the average between 50 and 60 min of the rest trial was compared with 55 min of the exercise trial, whereas the time-weighted average between 60 and 75 min was compared with 65 min of the exercise trial. To confirm our statistical results, these exercise time points (45, 55, and 65 min) were also compared with the original, resting trial time points that occur before and after the given exercise time point (i.e., 45 min of the exercise trial was compared with 40 and 50 min of the rest trial).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory and electrolyte data. There were no changes in pulmonary measurements at rest; however, MSG ingestion resulted in a mean O₂ increase of ~2.5 ml · kg¹ · min¹ [49.6 ± 2.7 ml · kg¹ · min¹ (MSG) vs. 47.1 ± 1.7 ml · kg¹ · min¹ (Plb); P < 0.05] during exercise. It is likely that this increase in O₂ is a result of elevated O₂ in the exercising muscle, because this increase was not seen during the rest trials. The mean relative O₂ workload at 10 and 15 min of exercise was ~84.2 ± 2.3% O_2 max during Plb, whereas the relative workload after MSG ingestion was ~87.9 ± 2.4% O_2 max. There were no differences observed in respiratory exchange ratio. Changes in sodium, chloride, and potassium did not result with MSG ingestion (data not presented).

Plasma amino acids. MSG ingestion resulted in at least an 18-fold increase in plasma glutamate (Fig. 2). During the rest trial, the mean peak concentration was 470 ± 52 µM and occurred from 40 to 60 min postingestion. However, during the exercise trial, the increase in plasma glutamate after MSG ingestion was similar but larger than that during the resting trial (P < 0.05), with a mean peak concentration of 630 ± 63 µM (P < 0.001). This larger increase that was observed in the exercise trial coincided with the onset of exercise.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Plasma glutamate response to MSG ingestion. Values are means ± SE. , Rest + placebo (R/Plb); , rest + MSG (R/MSG); , exercise + placebo (E/Plb); , exercise + MSG (E/MSG). Symbols depict a treatment effect: * E/MSG vs. E/Plb,  R/MSG vs. R/Plb, and  E/MSG vs. R/MSG, P  0.05. Significant difference from a 0 min and b 40 min, P  0.05. Dashed line, postexercise time.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Plasma aspartate (A), taurine (B), and alanine (C) responses to MSG ingestion. Values are means ± SE. Group symbols, superscripted symbols and letters, and dashed lines are as defined in Fig. 2 legend.

Ammonia. Because ammonia levels are low and variable at rest, the ammonia data were normalized to the Plb resting baseline values (Fig. 4). Plasma ammonia did not change throughout the rest trial, nor did it change in the preexercise phase. However, in contrast to the increases in glutamate, glutamine, aspartate, alanine, and taurine during exercise, the rise in plasma ammonia was attenuated during exercise after MSG ingestion (P < 0.05). These ammonia data, together with the results of glutamine and alanine, demonstrate a shift in nitrogen distribution during exercise (Fig. 5). Our results imply that an increase in glutamate availability leads to a perturbed balance in the nitrogen carriers.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Plasma ammonia response to MSG ingestion during the exercise trials. Values are means ± SE. Group symbols and superscripted symbols and letters are as defined in Fig. 2 legend.

View larger version (9K): [in this window] [in a new window]   Fig. 5.   Histogram depicting percent change in area under the curve between placebo and MSG for alanine (solid bar), glutamine (open bar), and ammonia (shaded bar) during exercise trial (55 min).

Lactate, glucose, insulin, and C peptide. These data are summarized in Tables 3 and 4. During the rest trials, there were no changes in lactate and glucose after ingestion of MSG. Insulin was modestly but significantly elevated from 30 to 60 min after MSG ingestion, coinciding with the peak increase of plasma glutamate. C peptide followed a similar trend.

Glycerol and free fatty acids. Glycerol and free fatty acid data are also summarized in Tables 3 and 4. During the rest trials, there were no changes in glycerol. Glycerol levels were elevated during the MSG exercise trial, but significance was only reached at 10 min of exercise. Free fatty acid concentrations were not different in any of the four trials.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Enhanced glutamate availability was achieved via MSG ingestion to investigate glutamate's effects on resting and exercise metabolism. Previous studies have used MSG ingestion to examine cardiac and pulmonary physiology during exercise; this is the only study, to our knowledge, that has examined the metabolic implications of increased glutamate availability during exercise. Providing glutamate in excess resulted in a further elevation of plasma glutamate as well as aspartate during exercise compared with rest. This increase in circulating glutamate availability, as well as the resulting elevated insulin levels, potentially led to increased intramuscular glutamate levels, which ultimately may have induced a rise in plasma alanine during exercise. MSG ingestion also resulted in specific exercise responses of increased taurine levels and O₂. Although the increase in O₂ was small, it may imply elevated TCA cycle flux during exercise after glutamate ingestion. Contrary to our expectations, the shift in nitrogen distribution that occurred after glutamate provision entailed a larger increase in plasma alanine, a small increase in glutamine, and a decrease in ammonia over the entire exercise period (Fig. 5).

MSG ingestion resulted in an ~18-fold increase in plasma glutamate during rest, which is similar to results of previous studies that administered MSG to humans (2, 15, 26-28). Surprisingly, there was a greater increase in plasma glutamate during exercise than at rest (mean peak increase was 601 ± 65 µM during exercise vs. 470 ± 52 µM during rest trials). The pronounced and rapid increase occurred with the onset of exercise and could be due to increased absorption at the gut or decreased clearance at the liver. However, with 40 min between the time that MSG is ingested and the time that exercise commences, absorption of glutamate at the gut should be completed. In addition, as exercise commences, blood flow is shifted away from the splanchnic region and shifted toward the exercising muscle; thus it is unlikely that there is an increased absorption of glutamate at the gut and more likely that there is a decreased clearance of glutamate, possibly by the liver.

We assume that glutamate is taken up by skeletal muscle, because the increased availability of plasma glutamate has been shown to increase glutamate arteriovenous differences during exercise (31) and increase intramuscular glutamate at rest (15). After a glutamate administration, Thomassen et al. (31) demonstrated an increase in glutamate arteriovenous difference across the leg. Graham and colleagues (15) observed intramuscular glutamate levels rise ~40% at rest within 40-60 min of MSG ingestion. Because the subjects in the present study rested for 40 min before exercise, it is very likely that intramuscular glutamate levels were elevated before the onset of exercise. It is also known that there is a greater uptake of glutamate by skeletal muscle during exercise (10, 17). Combined with a large glutamate concentration gradient from the circulation to muscle and the increase in insulin, there may be a further rise in intramuscular glutamate in the active muscle.

Although glutamate plays an integral role in numerous metabolic reactions, relatively few plasma amino acid concentrations (i.e., aspartate, taurine, alanine, and glutamine) were affected by glutamate ingestion. The most interesting effects were seen in plasma alanine, which did not change during rest trials and preexercise but notably increased during exercise. It is well established that plasma alanine increases during exercise (10, 17, 20). In the present study, we observed a rise in alanine during exercise; however, MSG enhanced this response. The rise in plasma alanine during exercise is likely attributed to the elevated production of alanine from the active muscle. If exercise time were extended, one may expect alanine to continue to rise if glutamate uptake continually increased intramuscular glutamate in the exercising muscle. Alanine, along with 2-oxoglutarate (a TCA cycle intermediate), is formed by the transamination of glutamate and pyruvate via alanine aminotransferase (AAT). Because AAT transamination is a near-equilibrium reaction, elevated levels of intramuscular glutamate from MSG administration, as well as elevated levels of pyruvate from an exercise-induced increase in glycolysis, could produce more 2-oxoglutarate and alanine in the muscle, whereby alanine would be released into circulation.

It has been proposed that 2-oxoglutarate is critical to oxidative metabolism (10, 11, 32). At the onset of exercise, there is an expansion of the TCA cycle intermediate pool (anaplerosis), but 2-oxoglutarate is the only TCA cycle intermediate to decrease after 1 min of exercise (10-13). One could speculate that an increase in 2-oxoglutarate could result after MSG ingestion, which could potentially modify TCA cycle flux. This may be supported by the ~2.5 ml · kg¹ · min¹ increase in O₂ that was observed during exercise after MSG administration. Although 2.5 ml · kg¹ · min¹ appears to be a small increase in O₂, it is important to reiterate that this is a whole body measurement and can be translated to ~170 ml/min of O₂. Because O₂ did not change during rest with MSG ingestion and with the large increase in blood flow to the exercising muscle at this high-intensity exercise, it would be assumed that the majority of this increase in O₂ occurred in the exercising legs. Therefore, this elevated O₂ may be greater per kilogram of active muscle and have more direct implications on the TCA cycle flux of the exercising muscle. It is also important to note that the fitness level of our subjects was high, which would also increase the O₂ of the exercising muscles, and thus may affect TCA cycle anaplerosis. However, one would also predict lower lactate production, which, in contrast, tends to be elevated with MSG. These data warrant future study using MSG ingestion as a tool to investigate glutamate's role on TCA cycle anaplerosis.

Glutamine and ammonia also demonstrated interesting changes. Plasma ammonia and glutamine did not change during the rest trials after MSG ingestion. However, ammonia was attenuated with glutamate ingestion during exercise, similar to the results of Systrom et al. (29), whereas glutamine increased gradually. Previous work has shown that alanine, glutamine, and ammonia are released in equal amounts during exercise (20). Thus we hypothesized that the greater availability of glutamate from MSG administration would elevate these particular nitrogen carriers (alanine, glutamine, and ammonia) equally because they are involved in various metabolic reactions in which glutamate has an integral role. However, it appears that increased glutamate levels have led to a shift in the normal distribution of nitrogen, as MSG induced a pronounced increase in alanine and a modest increase in glutamine while plasma ammonia was attenuated during exercise. This shift in nitrogen balance is represented in Fig. 5, where percent change in area under the curve between Plb and MSG trials has been calculated for alanine, glutamine, and ammonia during the exercise trials. This shift in nitrogen distribution emphasizes the importance of glutamate in various transamination reactions. Glutamate and glutamine are known to have intimate interactions during exercise; whereas our results suggest that glutamate may have a more important role in alanine metabolism during exercise. These data depict a much greater shift toward the production of alanine vs. glutamine, which further supports glutamate's role in TCA cycle flux.

Aspartate substantially increased during the rest and exercise trials and was remarkably similar in time course and pattern with the increase in glutamate in both MSG trials. These increases, at rest, are similar to the increases in aspartate reported previously for resting subjects (15, 26-28). The elevated aspartate levels that result after MSG ingestion may be attributed to the interrelated metabolism of glutamate and aspartate. The increases seen in aspartate at rest and during preexercise potentially originate from transamination reactions in the gut as part of a "flooding effect." Glutamate is known as an important nutrient of splanchnic metabolism (3, 21) and can be sequestered, almost entirely, by the splanchnic bed when provided in small amounts (3, 21). Because there is such a large inflow of glutamate into the intestinal tract at rest, the intestinal cells most likely sequester and use what they need of the incoming glutamate, whereas the remainder may be released into circulation as glutamate or transaminated (via aspartate aminotransferase) into aspartate before being released. This may explain the similar time course of increase demonstrated by aspartate and glutamate. During exercise, the mechanisms responsible for the increases in aspartate are most likely different from those at rest. As the blood flow to the gut declines during exercise, the continued increase in plasma aspartate is likely due to a decreased clearance from the liver, as previously suggested with glutamate.

It has been debated whether the purine nucleotide cycle is important during exercise. With a greater availability of plasma aspartate and glutamate, one would expect an increase in intramuscular aspartate via increased uptake of aspartate from circulation or from transamination of elevated intramuscular glutamate, respectively. Graham et al. (15) have shown that MSG ingestion tends to increase intramuscular aspartate at rest. Therefore, there would be more substrate provided for the purine nucleotide cycle. This, in turn, would predict that ammonia would be elevated with increased purine nucleotide cycling. However, our data show that MSG results in an attenuated increase in ammonia during exercise, suggesting either that the purine nucleotide cycle is not active during exercise or that aspartate is directed toward other metabolic reactions (i.e., the production of oxaloacetate).

Surprisingly, taurine was another amino acid that was elevated after glutamate ingestion during the rest trials and to a much greater extent during exercise. The further increase during exercise may be due to a decrease in clearance, as suggested for glutamate and aspartate. Taurine is purported to have insulin-like actions on glucose utilization, although its actions are suggested to be independent of insulin (8, 22). It is possible that taurine, together with the rise in insulin, seen in the present study, may have prevented a rise in glucose during exercise after MSG ingestion. The physiological implications of glutamate ingestion on taurine metabolism are poorly understood. Although glutamate is central to numerous reactions, few changes in plasma amino acids resulted after glutamate ingestion. However, not only was taurine one of the few amino acids that was elevated with MSG ingestion, but it increased in a similar manner as plasma glutamate. Because glutamate and taurine appear to affect insulin independently (8, 22), the potential interrelationship between these two amino acids may further support a possible role toward insulin. Future investigations are certainly warranted to examine further glutamate's importance in taurine metabolism.

As seen in previous work, insulin was elevated after MSG ingestion (6, 15, 31). C peptide, which has not been measured in previous MSG studies, increased in a similar manner to insulin but was not significant. Several studies have demonstrated the existence of glutamate receptors in pancreatic cells (23, 33), and it has been suggested that glutamate may stimulate insulin secretion through these cells (19, 23). However, our data suggest that, whereas glutamate ingestion may have, in part, increased insulin secretion, glutamate ingestion may have also resulted in inhibiting the clearance of insulin.

In summary, MSG ingestion has several specific effects on amino acids such as glutamate, aspartate, glutamine, taurine, and alanine. Whereas glutamate and aspartate are elevated at rest after glutamate ingestion, these amino acids are elevated to a greater extent during exercise. MSG induces an enhanced increase in plasma alanine only during exercise, which may be a result of elevated intramuscular alanine production via the AAT transamination of elevated intramuscular glutamate from MSG and pyruvate from exercise-induced glycolysis. Glutamine also appears to be elevated during exercise after glutamate ingestion. Conversely, ammonia displays an attenuated increase with greater glutamate availability and exercise. These data suggest that increased glutamate availability during exercise alters its distribution, resulting in elevated alanine, modest increases in glutamine, and decreased ammonia.