INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

GLUCOSE PRECURSORS, SUCH AS lactate and glycerol, when ingested during prolonged exercise, are readily available for oxidation (5, 18). Alanine is also a glucose precursor that could be ingested in large amounts without gastrointestinal discomfort. For example, Carlin et al. (6) and Koeslag and co-workers (11, 12) have shown that up to 100 g of alanine can be ingested with minimal side effects, before or after exercise. In these situations, alanine reduced ketosis, presumably because of anaplerosis in the liver. Recent studies have also shown that muscles can take up significant amounts of alanine when a large load of an amino acid mixture is infused (3) or ingested (22) during recovery from exercise.

In the present experiment, the metabolic fate of a large load of alanine (1 g/kg) ingested immediately before and during prolonged exercise was described. Both ¹³C and ¹⁵N labeling were used to separately follow decarboxylation of the carbon skeleton and excretion of the amino-N of exogenous alanine, respectively, from the production of ¹³CO₂ at the mouth and from [¹⁵N]urea production in urine and sweat. White and Brooks (28), using [¹⁴C]alanine in rats, have shown that endogenous alanine is readily decarboxylated/oxidized during prolonged exercise. However, their data, as well as data from Wolfe et al. (30), suggest that the metabolic fate of the two moieties of the alanine molecule could be different, with decarboxylation of the carbon skeleton but reincorporation of amino-N in amino acids and proteins. We thus hypothesized that exogenous alanine will be catabolized during the exercise period but that decarboxylation of the carbon skeleton computed from ¹³CO₂ production at the mouth will be comparatively larger than deamination computed from [¹⁵N]urea excretion.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. The experiment was conducted on six active and healthy male subjects who gave their informed, written consent to participate in the study, which was approved by the Institutional Board on the Use of Human Subjects in Research. None of the subjects were smokers, heavy drinkers, under medication, or using recreational drugs. Their age, height, weight, and maximal O₂ uptake (O_2 max) on cycle ergometer were 21.8 ± 0.5 yr, 175.0 ± 1.5 cm, 73.7 ± 2.0 kg, and 54.0 ± 1.8 ml · kg¹ · min¹, respectively (mean ± SE).

Experiments. O_2 max and experimental workload on cycle ergometer (Ergomeca, La Bayette, France) were determined for each subject during a preliminary test session using open-circuit spirometry (1100 medical gas analyzer, Marquette Electronics, Milwaukee, WI). One week later, the subjects exercised for 180 min (from ~9:00 to ~12:00 noon) at 132 ± 8 W, corresponding to 45% of the maximal workload (294 ± 17 W), and 53 ± 2% O_2 max. The last evening meal before the experiment (7:00 PM; ~1,250 kcal: ~70% carbohydrates, ~15% lipids, ~15% proteins) and the morning breakfast (7:30 AM; ~500 kcal: ~50% carbohydrates, ~35% lipids, ~15% proteins) were standardized and were both poor in ¹³C. In addition, to keep a low background ¹³C enrichment of expired CO₂, ingestion of foods from plants with the C₄ photosynthetic cycle, which are naturally enriched in ¹³C (14), was avoided 1 wk before the experiment. Subjects also refrained from exercising and from drinking coffee and alcohol for 2 days before the experiment.

¹³C and ¹⁵N labeling. During the experiment, the subjects ingested 1 g/kg body mass of alanine dissolved in water (10 ml/kg): 3 ml/kg with 0.3 g of alanine/kg 20 min before the beginning of exercise and 1 ml/kg with 0.1 g of alanine/kg every 20 min thereafter up to minute 120 (seven doses). The total amount of alanine ingested was 73.7 ± 2 g in 737 ± 20 ml of water. The subjects were also encouraged to ingest water throughout the experiment (total amount ingested: 1,537 ± 238 ml). The alanine [ICN Pharmaceuticals, Biomedical Research Products, Costa Mesa, CA; ¹³C/¹²C = 22.0 difference () ¹³C-Pee Dee Belemnitella-1 (PDB₁) and ¹⁵N/¹⁴N = 5.9  ¹⁵N-air N] was artificially enriched with [U-¹³C]alanine and [¹⁵N]alanine (¹³C/¹²C and ¹⁵N/¹⁴N both > 99%; Isotec, Miamisburg, OH) to achieve final isotopic compositions close to 14  ¹³C-PDB₁ and 110  ¹⁵N-air N (actual values measured by mass spectrometry: 14.2  ¹³C-PDB₁ and 115  ¹⁵N-air N).

Measurements and analysis. Oxygen uptake (O₂) and carbon dioxide production (CO₂) were measured at rest before ingestion of the first dose of alanine and every 20 min thereafter with open-circuit spirometry (1100 medical gas analyzer, Marquette Electronics), and the urine produced over the experiment was collected. In addition, the amount of sweat produced over the exercise period was estimated from changes in body mass. For this purpose, body mass was measured after emptying of the bladder, immediately before ingestion of the first dose of alanine, and immediately after exercise, after thorough drying. The observed change in body mass is the balance between, on one hand, loss of mass through sweat and urine, water loss through the lungs (computed from pulmonary ventilation and the ambient relative humidity), and CO₂, and, on the other hand, gain of mass through fluid and alanine intake and O₂ (15). Samples of expired gases (10 ml) were also collected in vacutainers (Becton-Dickinson, Franklin Lakes, NJ) at rest before ingestion of the first dose of alanine and every 20 min thereafter for the measurement of ¹³C/¹²C in expired CO₂. Finally, a 5-ml sample of sweat was collected at the end of exercise from a plastic pouch fixed on the back of the subjects. The urea in urine and sweat was purified for measurement of ¹⁵N/¹⁴N. For this purpose, a 1-ml sample of urine or sweat was diluted into 12 ml of distilled water and 25 ml of glacial acetic acid, and the urea was precipitated by adding 2.5 ml of methanol containing 0.125 g of xanthydrol (C₃H₁₀O₂) (24).

Computations. The amounts of the various substrates (exogenous alanine, endogenous proteins, glucose, and lipids) oxidized over the exercise period were computed as follows. First the amount of labeled urea excreted in urine and sweat was computed from the total amount of urea excreted (i.e., volume of fluid × urea concentration) and from the fraction of urea derived from labeled alanine in urine and sweat over the exercise period

In this equation, R is the ¹⁵N/¹⁴N ratio in urea extracted from urine or sweat during exercise (obs), in urea extracted from a sample of urine collected before ingestion of the first dose of alanine (ref), and in ingested alanine (exo). The amount of exogenous alanine deaminated was then computed, taking into account that 1 mmol of alanine provides 0.5 mmol of urea. Second, on the basis of the amount of unlabeled urea excreted computed by the difference between the total amount of urea excreted and the amount of labeled urea excreted, the amount of endogenous proteins deaminated was computed (2.9 g of proteins/g of urea excreted) (16). Third, on the basis of ¹³C/¹²C in expired CO₂ in response to exercise with ingestion of alanine (Robs_C) and in ingested alanine (Rexo_C), the amount of exogenous alanine decarboxylated was computed

In this equation, Rref_C is the background ¹³C enrichment of expired CO₂ and 0.629 is the volume of CO₂, in liters, produced when 1 g of alanine is decarboxylated. The value of Rref_C has been shown to change in response to exercise and to substrate ingestion and oxidation (19). In the present experiment, changes in Rref_C have been estimated by having the subjects exercise for 180 min at the same workload (after the standardized evening meal and breakfast) but with ingestion of water only (see Fig. 2). This method neglects the small changes that could be due to alanine ingestion and oxidation. It should also be recognized that the oxidation of exogenous alanine could be slightly underestimated because of the presence of a large bicarbonate pool that delays equilibration between ¹³CO₂ production in tissues and at the mouth at the beginning of exercise (17). In addition, a fraction of ¹³C entering the tricarboxylic acid cycle under the form of acetate could be irreversibly lost (21, 23). When labeled acetate is infused in trace amounts, this fraction, which is high at rest (50-75%), is much lower during exercise (10-35%) (21, 23), but no data are presently available when ¹³C-labeled substrates are ingested in large amounts during prolonged exercise. Finally, on the basis of CO₂ and O₂ corrected for endogenous protein oxidation (1.010 liters of O₂ and 0.843 liters of CO₂/g of proteins oxidized) (16) and for the decarboxylation and oxidation of exogenous alanine (0.755 and 0.629 liters of O₂ and CO₂/g of alanine oxidized), the amounts of glucose and fat oxidized were computed (18)

The contribution of the oxidation of each substrate to the total energy yield was computed from their respective energy potential at 37°C (3.87, 9.75, 4.70, and 3.72 kcal/g for glucose, lipids, proteins, and alanine) (16, 27). In these computations, it was assumed that the carbon skeleton of the endogenous proteins and of the exogenous alanine deaminated was entirely decarboxylated and oxidized.

Statistics. Data are presented as means ± SE. Comparisons were made by using analysis of variance for repeated measures (Statistica package) and Newman-Keuls post hoc tests to identify the location of significant differences (P  0.05) when the analysis of variance yielded a significant F-ratio.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory exchanges (O₂ and CO₂) were stable over the 3-h exercise period (Fig. 1), but the respiratory exchange ratio corrected for endogenous protein and exogenous alanine oxidation slowly decreased from a peak value observed at minute 20 until the end of exercise period.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Respiratory exchanges (O2 and CO2) and respiratory exchange ratio (RER) corrected for endogenous protein oxidation and exogenous alanine decarboxylation and oxidation during the 3-h exercise period.

As shown in Table 1, urea concentration in urine was ~15 times that in sweat, but the amount of urea excreted in urine was only ~4.5 times that excreted in sweat because of the much larger volume of sweat than urine produced during the exercise period. The isotopic composition of urea before ingestion of labeled alanine was 4.81 ± 0.10  ¹⁵N-air N. The isotopic composition of urea excreted in sweat and urine over the exercise period was about four times higher and was not significantly different in the two fluids (Table 1). The percentage of urea deriving from labeled alanine was ~12% in both urine and sweat, and the total amount of urea excreted that derived from labeled alanine was 13.0 ± 3.2 mmol corresponding to the deamination of 26.0 ± 3.2 mmol of exogenous alanine, or 2.3 ± 0.3 g. A much larger amount of urea derived from endogenous proteins: ~88% or 97.3 ± 9.0 mmol, corresponding to the deamination of 16.9 ± 1.6 g of endogenous proteins.

The isotopic composition of expired CO₂ at rest was not significantly different in the control and experimental situations (23.2 ± 0.5 and 23.4 ± 0.2  ¹³C-PDB₁; Fig. 2). In response to exercise with ingestion of water only, a small transient increase in the isotopic composition of expired CO₂ was observed over the first hour of exercise. When labeled alanine was ingested, the isotopic composition of expired CO₂ was slightly but significantly increased immediately before the beginning of exercise, 20 min after ingestion of the first dose. A much higher increase was observed in response to exercise, with a plateau at about 19  ¹³C-PDB₁ over the last 100 min of exercise. The rate of decarboxylation of the carbon skeleton of exogenous alanine progressively increased over the first 2 h of exercise and leveled ~350 mg/min thereafter (Fig. 2). Over the exercise period, 50.6 ± 3.5 g of alanine were decarboxylated (568 ± 39 mmol), or 68.7 ± 4.5% of the ingested load.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   13C/12C in expired CO2 during the 3-h exercise period with ingestion of water and labeled alanine (A) and exogenous alanine decarboxylation and oxidation (B). Horizontal bars indicate data points significantly different from the basal value observed 20 min before the beginning of exercise, P < 0.05. PDB1, Pee Dee Belemnitella-1.

Table 2 summarizes the oxidation of the various substrates and their respective contributions to the energy yield over the 3-h period of exercise. Total protein oxidation contributed ~15% of the energy yield: 4.8 ± 0.4% from endogenous protein oxidation and 10.0 ± 0.6% from exogenous alanine oxidation. Glucose and lipids were the main substrates oxidized, contributing 47.6 ± 4.3 and 37.4 ± 4.7% to the energy yield, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Several types of amino acids, proteins, and protein hydrolysates have been used as oral supplements in relation to exercise in an attempt to reduce central fatigue (4), to favor anaplerosis (6, 11, 12), to increase the immune response (2), to reduce the effect of aspirin on the permeability of the gut (13), and to increase protein synthesis (3, 22) and glycogen resynthesis (25). The purpose of the present experiment was to describe the metabolic fate of a large dose of alanine ingested during prolonged exercise at a comparatively low workload. Alanine is the main amino acid released by the muscle during exercise (20). Williams et al. (29) have shown that alanine release from the muscle increases ~2.5 times during prolonged exercise. In this situation, alanine is taken up by the liver and converted into glucose (10, 26). Data from Carlin et al. (6) and Koeslag and co-workers (11, 12) also show that alanine ingested in large amounts at rest (100 g) (6, 12) or exercise (50 g) (11) reduces ketogenesis, presumably by providing oxaloacetate in the liver, not only for gluconeogenesis but also for citrate synthesis.

In the present experiment, the large load of alanine ingested immediately before and during exercise was well tolerated, as already reported by Carlin et al. (6), and the production of ¹³CO₂ at the mouth showed that the carbon skeleton of exogenous alanine was readily available for decarboxylation and oxidation. Indeed, ¹³C/¹²C in CO₂ produced at the mouth was already significantly elevated 20 min after ingestion of the first dose of labeled alanine, while the subjects were still at rest. A further marked increase was observed in response to exercise and ingestion of the subsequent doses of [¹³C]alanine. The cumulative volume of ¹³CO₂ produced indicates that over the 180-min period of exercise, 50.6 ± 3.5 g of exogenous alanine were decarboxylated, providing 10.0 ± 0.6% of the energy yield. At the end of the exercise period, the oxidation rate of exogenous alanine peaked at 0.35 ± 0.07 g/min, providing 11.8 ± 1.0% of the energy yield. These figures are only slightly lower than those reported when similar amounts of glucose are ingested [e.g., 0.43 g/min with 75 g of glucose ingested (18)].

It is generally accepted that endogenous alanine released from the muscle during exercise is taken up by the liver and converted into glucose (glucose-alanine cycle) (10). However, Biolo et al. (3) have shown that when alanine was infused at rest, along with other amino acids, alanine balance across the leg reverted from net release to net uptake, particularly during the recovery from exercise. A similar observation was made by Tipton et al. (22) after oral alanine administration. Data from Abumrad et al. (1) indicate that muscle could remove 28% of an intravenous alanine load. These data suggest that in the present experiment with a large amount of exogenous alanine ingested, alanine transport in the muscle instead of alanine release from the muscle could have occurred and that exogenous alanine could, thus, have been directly oxidized in the muscle without prior conversion into glucose by the liver. Results from the present experiment do not allow, however, conclusions about the possible routes followed by the carbon skeleton of alanine before being decarboxylated and oxidized.

As discussed by Williams et al. (29), data from White and Brooks (28) in rats and from Wolfe et al. (30) in humans indicate that the metabolic fate of the amino-N of alanine does not follow that of the carbon skeleton and remains unclear. Results from the present experiment are in line with these observations. Indeed, the amount of [¹⁵N]urea excreted in urine and sweat over the period of exercise was very low (13.0 ± 3.2 g), corresponding to the deamination of only 2.3 ± 0.3 g of exogenous alanine vs. 50.6 ± 3.5 g that underwent decarboxylation. Figure 3 summarizes the possible flux of substrates and the metabolic pathways that could be followed by the carbon skeleton and the amino-N of exogenous alanine over the exercise period, as inferred from ¹³CO₂ and [¹⁵N]urea production. Whereas most of the carbon skeleton of exogenous alanine was excreted under the form of CO₂, there was no parallel excretion of ¹⁵N into urea. Accordingly, most of the amino-N provided by exogenous alanine was retained in the body and could have been redistributed into nonessential amino acids and/or incorporated into proteins. Under this hypothesis, ¹⁵N could be found in various nonessential amino acids in the plasma. In addition, under the hypothesis discussed above that the carbon skeleton of alanine was at least in part directly oxidized in muscle, at least a portion of the amino-N released in the process should have been carried from the muscle to the liver under the form of [¹⁵N]-glutamine. As shown in Fig. 3, the conversion of glutamate into glutamine, which requires an additional NH₃ mainly provided from protein breakdown, could actually be the limiting factor in the excretion of the amino-N of exogenous alanine. Obviously, these putative pathways followed by the amino-N moiety of ingested alanine remained speculative, because, in the present experiment, no blood samples were taken, and the distribution of ¹⁵N in amino acids and proteins was not directly measured.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Possible flux of substrates and metabolic pathways followed by the carbon skeleton and the amino-N of exogenous alanine over the exercise period, as inferred from 13CO2 and [15N]urea production.

A large amount of nitrogen was provided under the form of alanine (11.6 ± 0.3 g), but nitrogen loss in urine and sweat was small (110.3 ± 10.4 mmol of urea or 3.1 ± 0.3 g of nitrogen). The positive nitrogen balance was thus large (8.5 ± 0.3 g over the 180-min period). Results from several studies indicate that muscle protein synthesis is increased in response to protein ingestion at rest (1, 3) and after exercise (20, 22). Results from the present experiment suggest that muscle protein synthesis could also be increased during exercise at a comparatively low workload, when a large amount of a single amino acid is ingested. As discussed by Carraro and colleagues (7, 8), the synthesis of acute phase proteins in response to exercise could also explain in part the positive nitrogen balance.

In conclusion, results from the present experiment with ingestion of a large alanine load during prolonged moderate exercise in man, are in line with previous data concerning the metabolic fate of the two moieties of the molecule (28, 30). Excretion of ¹³CO₂ at the mouth indicated that the carbon skeleton of alanine was readily oxidized, either after conversion into glucose in the liver or directly, providing ~10% of the energy yield. In contrast, only a small percentage of the ingested nitrogen was excreted in the form of urea in urine and sweat, suggesting that in this situation the amino-N of alanine could be preferentially incorporated in amino acids and proteins.