Contents - Previous - Next

Exercise, aging and protein metabolism


* Human Physiology Laboratory, USDA Human Nutrition Research Center on Aging, Tufts University, 711 Washington Street, Boston, MA 02111, U.S.A.

1. Body composition changes with age and their consequences
2. Fuels used to meet various components of energy requirements
3. Age and dietary protein needs
4. Exercise-induced muscle damage and acute phase response
5. Exercise and protein metabolism
6. Summary

1. Body composition changes with age and their consequences

Aging is associated with a decrease in skeletal muscle size and strength. LOSS of muscle mass with age in humans has been demonstrated both indirectly and directly. The excretion of urinary creatinine, reflecting muscle creatine content and total muscle mass, decreases by nearly 50% between the ages of 20 and 90 (TZANKOFF and NORRIS, 1978). Muscle and non-muscle mass can be calculated from total potassium and nitrogen; about 60% of the body's potassium is found in skeletal muscle, and the ratio of nitrogen is higher in muscle than in non-muscle lean tissue (COHN et al., 1978). Using total body K and N. Cohn and co-workers determined that skeletal muscle protein is reduced and non-muscle protein is maintained with advancing age.

Computerized tomography of individual muscles shows that, after age 30, there is a decrease in cross-sectional areas of the thigh along with decreased muscle density associated with increased intramuscular fat. These changes are more pronounced in women (IMAMURA et al., 1983). Muscle atrophy may result from a gradual and selective loss of muscle fibers. The number of muscle fibers in the midsection of the vastus lateralis of autopsy specimens is lower by about 110000 in elderly men (age 70-73) than in young men (age 19-37), a 23% difference (LEXELL et al., 1983). The decline is more marked in Type 11 muscle fibers, which fall from an average 60% of total fibers in sedentary young men to below 30% after the age of 80 (LARSSON, 1983), and is significantly associated with age-related decreases in strength (r=0.54, p<0.001).

Declining muscle mass is associated with a number of age-related changes. For example, declining skeletal muscle mass is closely associated with the age-related decrease in basal metabolic rate (TZANKOFF, 1978). As activity decreases, energy requirements are decreased in the elderly. Unfortunately, energy intake does not decrease to the same extent as energy requirements, with a resultant increase in body fat content with advancing age (EVANS and MEREDITH, 1989). Reduced muscle mass and activity are likely to be an important cause of age-related loss in bone mineral, resulting in osteoporosis (SANDIER, 1989).

Muscle atrophy and weakness are more prevalent among elderly individuals who develop hip fractures than those of similar age who do not (ANIANSSON et al., 1984). We have shown that resistance training is an effective way to increase both the size and strength of muscles in the elderly (FRONTERA et al., 1988; FIATARONE et al., 1990). Not only can resistance training increase muscle size, but a recent report indicated that long-term resistance training may prevent age-associated changes in histochemical fibre-type distribution, myosin heavy chain isoforms and tropomyosin isoforms (KLITGAARD et al., 1990a; b).

Regularly performed submaximal exercise by elderly men and women can result in important improvements in aerobic and muscle oxidative capacity (MEREDITH et al., 1989a), glucose tolerance and insulin sensitivity (HUGHES et al., in review) and has been demonstrated to prevent the onset of type II diabetes. Because of these positive outcomes seen in the elderly, increased and vigorous activity is recommended for men and women at all ages (EVANS and MEREDITH, 1989; EVANS, ROSENBERG and THOMSON, 1991; FIATARONE and EVANS, 1990; FRONTERA and EVANS, 1986). This review will focus on the interaction between exercise, aging, and protein and energy metabolism.

2. Fuels used to meet various components of energy requirements

The substantial energy requirements of endurance exercise are primarily met by the oxidation of skeletal muscle glycogen and triglycerides as well as hepatic glycogen and adipocyte triglyceride stores as blood-born glucose and free fatty acids. These fuels can satisfy 90 to 95% of the total energy requirement. The remainder is met by the oxidation of protein which, unlike fats and carbohydrate, is a non-renewable source of energy.

This review will focus on the relationship between endurance and strengthening exercise and protein metabolism, and the evidence for increased dietary protein requirements. For the most part, two experimental approaches have been used. Early studies used the appearance of nitrogen-containing waste products, especially urea, in blood and urine as an index of protein oxidation. Changes in the oxidation of individual amino acids using isotope labels as tracers have been employed more recently.

The theory of the great German chemist JUSTUS VON LIEBIG (1870) that protein was the primary fuel of working muscle was tested by a number of 19th century scientists. FICK and WISCLICENUS (1866) hiked up the Faulhorn in 1865 and collected their urine for nitrogen analysis before and during the hike. Unfortunately, they confounded their results by placing themselves on a protein-free diet the day be fore their climb, thereby certainly reducing their urinary nitrogen excretion. They also stopped their urine collection upon completion of the climb and, therefore, eliminated the detection of any post-exercise rise in urinary nitrogen loss. They concluded that protein oxidation provided only a small portion of the energy required to make their climb. In a review, CATHCART (1925) also reached the conclusion that a slight increase in nitrogen excretion during and following work confirms the notion that protein oxidation does not provide the major source of energy for the working muscles.

The availability of carbohydrate as a fuel during exercise influences the oxidation of protein. LEMON and MULLIN (1980) estimated that the oxidation of protein (from serum, urine, and sweat urea losses) during prolonged submaximal exercise, when their subjects were glycogen-depleted, was double (up to 12% of the total energy demand) when compared to similar exercise in the glycogen-loaded state.

Paradoxically, exercise in the heat (30C) is associated with a lower rate of protein utilization when compared to that seen during and for two days following similar exercise in 5 and 20C. Exercise in a hot environment has been demonstrated to increase the rate of glycogen utilization (FINK, COSTILL and VAN HANDEL, 1975), a condition which should be associated with increased rates of protein utilization.

Gender differences in substrate utilization during prolonged submaximal exercise have also been seen. TARNOPOLSKY and coworkers (1990) have shown that a group of female athletes, matched to males by age and VO2 max, used significantly less glycogen and protein (estimated from 24-hour urinary nitrogen excretion) than did men while running on a treadmill at 65% VO2 max.

A number of studies have used the primed, constant infusion of 13C-leucine to measure the rate of oxidation of this essential amino acid during exercise. These studies indicate that submaximal exercise does not alter leucine flux, but substantially increases the rate of whole-body leucine oxidation (KNAPIK et al., 1991; MILLWARD et al., 1982; RENNIE et al., 1980). RENNIE and co-workers (1981) demonstrated that the increase in leucine oxidation rates was directly related to the intensity of exercise. KNAPIK and co-workers (1991) found that a complete 3.5-day fast did not cause an increase in leucine flux during exercise but caused a 44% increase in the rate of leucine oxidation.

We demonstrated that high-intensity exhaustive exercise caused a significant 48% accumulation of muscle -ketoisocaproic acid (KIC) (FIELDING et al., 1986). The elevation in muscle KIC concentration was not reflected in simultaneous changes in plasma KIC levels, suggesting a limited diffusion rate from muscle to blood. This study indicates that brief high-intensity exercise is associated with accelerated transamination of leucine.

DEVLIN et al. (1990) examined recovery from 3 hours of cycling at 75 % of VO2 max and found that whole-body protein breakdown was not increased above resting levels, leucine oxidation was decreased and non-oxidative leucine disposal (synthesis) was increased when compared to pre-exercise resting values. Using a much lower exercise intensity (4 hours at 40% VO2 max), CARRARO and co-workers (1990) also examined post-exercise recovery. They found a significant increase in the muscle fractional synthetic rate during the recovery period.

The concentration of urea in plasma and urine increases during submaximal exercise and remains high for some time later, also in proportion to the intensity and duration of the exercise (LEMON, DOLNY and YARESHESKI, 1984; HARALAMBIE and BERG, 1976). The increased oxidation of indispensable amino acids during submaximal exercise must, therefore, increase the need for dietary protein.

While the experiments examining protein metabolism using 13C-leucine (or other labeled amino acids) indicate increased oxidation during exercise, these studies do not directly show an increased requirement for dietary protein. GONTZEA, SUTZESCU and DUMITRACHE (1974) conducted a carefully controlled study in which 30 healthy young men consumed a diet containing 1.0 g protein/kg body weight. Nitrogen balance determinations were performed for three periods, a sedentary adaptation period, a 4-day exercise period, and a 4-day sedentary post-exercise period. The daily exercise consisted of six 20-minute intervals on a cycle ergometer at an intensity of 8-10 kcal/min, separated by 30-minute breaks. Energy intake was adjusted during the exercise period to provide an extra 50 kcal/kg body weight/d. Sweat nitrogen losses were included in the calculation of nitrogen balance. The mean became negative during the exercise period and did not become positive, even when the dietary protein intake was increased to 1.5 g/kg/d. A follow-up study examined the effect of a longer training period on nitrogen balance using a similar exercise load and a diet containing 1.0 protein/kg/d. Nitrogen balance became negative with the onset of the exercise period, but approached equilibrium by 2 weeks of training. The subjects in these studies were initially sedentary, and therefore the studies of GONTZEA, SUTZESCU and DUMITRACHE (1974; 1975) do not address the question of whether athletes, who have adapted to a high-level training intensity and duration, have a high protein requirement when energy demands are met.

TARNOPOLSKY, MacDOUGALL and ATKINSON (1988) attempted to determine the protein requirements in body builders and endurance-trained men. They examined nitrogen balance in these athletes on two different dietary protein intakes, both of which were above the amount required for achievement of balance. By extrapolating a line connecting the balance figures on two levels of protein intake, they estimated that bodybuilders required 1.12 times and endurance athletes required 1.67 times more daily protein than did sedentary controls. However, these protein requirements were extrapolated from nitrogen balance obtained with protein intakes of 1.7 and 2.65 g/kg/d and almost certainly overestimated protein requirements when compared with results that were closer to zero and included negative values.

Using nitrogen balance to estimate dietary protein requirement at three different dietary intakes (0.6, 0.9 and 1.2 g/kg/d of high-quality protein over three separate 10-day periods), we (MEREDITH et al., 1989b) found that habitual endurance exercise was associated with dietary protein needs greater than the current Recommended Dietary Allowance of 0.8 g/kg/d and average 0.94 0.05 g/kg/d. No age-related differences in protein requirements were seen in the six young (20-30 years) and six middle-aged (48-59 years) men examined in this study. Whole-body protein turnover, using 15N glycine as a tracer, and 3-methylhistidine excretion were not different from values reported for sedentary men, and were not different between the two age groups. Protein requirements expressed as a percent of energy needs (averaging 3910 240 kcal/d) showed that these subjects needed only 6.9 0.5% of total dietary calories as protein. These results suggest that well-trained individuals, consuming an average American diet (12-15% protein) and adequate amounts of energy, are not likely to have an inadequate dietary protein intake.

We examined the dietary intake of a group of eumenorrheic and amenorrheic athletes with similar exercise habits (NELSON et al., 1986). There were no differences between the two groups in VO2 max, number of miles run per week, or in body composition.

When compared to the eumenorrheic athletes, the amenorrheic women reported consuming less energy (1730 152 vs 2250 141 kcal) and protein (0.7 0.1 vs 1.0 0.1 g/kg/d). This implicates inadequate dietary energy and/or protein as a potential cause of athletic amenorrhea. When compared to the eumenorrheic women, the amenorrheic athletes have lower estradiol, estrone, LH (and LH pulse amplitudes), FSH, and T3 levels (FISHER et al., 1986), a hormone profile often seen in women on severely hypocaloric diets or women suffering from protein-energy malnutrition.

3. Age and dietary protein needs

Estimates of dietary protein needs of the elderly using nitrogen balance range from 0.59 to 0.8 g/kg/d (GERSOVITZ et al., 1982; UAUY, SCRIMSHAW and YOUNG, 1978; ZANNI, CALLOWAY and ZEZULKA 1979). However, the low value was' reported by ZANNI, CALLOWAY and ZEZULKA (1979), who preceded their 10-day dietary protein feeding with a 17-day protein-free diet, which was likely to improve nitrogen retention during the 10-day balance period.

UAUY, SCRIMSHAW and YOUNG (1978) examined both elderly men and women receiving one of three dietary protein intakes (0.52, 0.65 and 0.8 g/kg/d for 10 days. They reported an average protein requirement of 0.8 and 0.83 g/kg/d for the men and women, respectively. They recommended a dietary protein intake of and 0.8 g/kg/d; however, this recommendation does not include an allowance for differences in protein quality.

GERSOVITZ et al. (1982) examined nitrogen balance for a 30-day period in older men and women who consumed the RDA of 0.8 g/kg/d of egg protein. They found that, on average, their subjects were in negative N balance (-7.4 g N/d) during the first 10 days. The men and women in this study significantly improved their N balance during the second and third 10-day period. The authors concluded that the current RDA was inadequate to meet the needs of 97.5% of the elderly and therefore not a safe level for the majority of people above the age of 70.

Using stable isotope probes to examine differences in whole-body amino acid and protein turnover, a number of authors (YOUNG, 1990; GERSOVITZ et al., 1980; CAMPBELL et al., 1992) have seen no differences in metabolism when the results are expressed as a function of body weight. However, when the turnover data are expressed as a function of fat-free mass (FFM), the elderly show a greater rate of amino acid flux. This increase in amino acid flux/kg FFM may result from the greater dietary protein intake/kg FFM and necessary adaptation to maintain N balance.

The impact of age-related differences in body composition on protein metabolism and nutrition in the elderly are not fully known. The selective loss of muscle mass represents a major shift in body protein distribution and partial loss of a primary protein reservoir that may be utilized to supply amino acids for synthesis of new body proteins (YOUNG, 1990).

4. Exercise-induced muscle damage and acute phase response

Muscle contraction and shortening produces a concentric action. However, when skeletal muscle lengthens as it produces force, the result is an eccentric muscle action. An example of this is lifting a weight (concentric action) and lowering it (eccentric action). At the same power output, the oxygen cost of eccentric exercise is lower than that of concentric exercise (ASMUSSEN, 1956), but eccentric exercise has been demonstrated to be a potent cause of muscle damage (NEWHAM et al., 1983; O'REILLY et al., 1987), and increased circulating creatine kinase (CK) activity (EVANS et al., 1986).

Running a marathon can cause extensive skeletal muscle damage (HIKIDA et al., 1983; WARHOL et al., 1985). WARHOL and coworkers (1985) showed a characteristic pattern of muscle damage, with tearing of sarcomeres at the Z-band level followed by movement of fluid into the muscle cells in biopsies taken in the days following the race. Mitochondrial and myofibrillar damage showed progressive repair by 3 to 4 weeks after the marathon. Late biopsies (8 to 12 weeks after the race) showed central nuclei and satellite cells characteristic of a regenerative response. The damage seen by these investigators is very similar to the ultrastructural changes in skeletal muscle resulting from eccentric exercise.

The extent of the ultrastructural evidence of damage is greater well after the initial damaging exercise. FRIDEN et al., (1983) found more damaged muscle fibers 3 days than one hour after high-tension eccentric exercise. NEWHAM and co-workers (1983) also showed that eccentric exercise caused immediate damage, but that biopsies taken 24 to 48 hours after the exercise had more marked damage. These data are indicative of an ongoing process of skeletal muscle repair, consisting of increased degradation of damaged proteins and increased rate of protein synthesis.

Following only one bout of high-intensity eccentric exercise (EVANS, 1986), previously sedentary men showed a prolonged increase in the rate of muscle protein breakdown, evidenced by an increase in urinary 3-methylhistidine/creatinine which peaked 10 days later. In addition, an increase in circulating interleukin-1 (IL-I) levels in these subjects was seen 3 hours after the exercise. Endurance-trained men, performing the same exercise, did not display increased circulating IL-1 levels. However, their pre-exercise plasma IL-1 levels were significantly higher than those seen in the untrained subjects.

Damage to tissue, as well as infection, stimulates a wide range of defense reactions, known as the acute phase response (KAMPSCHMIDT, 1981). The acute phase response is critical for its antiviral and antibacterial actions as well as for promoting the clearance of damaged tissue and subsequent repair. Within hours of injury or exercise (CANNON et al., 1990), the number of circulating neutrophils can increase many-fold. Neutrophils migrate to the site of injury where they phagocytize tissue debris and release factors known to increase protein breakdown, such as lysozyme and oxygen radicals (BABIOR, KIPNES and CURNUTTE, 1973). Greater neutrophil increases have been observed after eccentric exercise than after concentric exercise (SMITH et al., 1989).

While neutrophils have a relatively short half-life (one or two days within tissue (BAINTON, 1988)), the life span of monocytes may be one to two months after migration to damaged tissue (JOHNSTON, 1988). Substantial monocyte accumulation in skeletal muscle was found after completion of a marathon. Following eccentric exercise, monocyte accumulation in muscle was not seen until 4 to 7 days later (JONES et al., 1986; ROUND, JONES and CAMBRIDGE, 1987). In addition to the capability to phagocytize damaged tissue, monocytes secrete cytokines such as IL-1 and tumor necrosis factor (TNF). These and other cytokines mediate a wide range of metabolic events, having an effect on virtually every organ system in the body.

Elevated cytokine levels during infection or injury have different and selective effects. IL-1 mediates an elevated core temperature during infection (CANNON and KLUGER, 1983). In laboratory animals, IL-1 and TNF increase muscle proteolysis and liberation of amino acids (NAWABI et al., 1990), possibly providing substrate for increased hepatic protein synthesis. While circulating IL-1 has been shown to increase acutely as a result of eccentric exercise (EVANS et al., 1983), by 24 hours after the exercise it returned to resting levels. Biopsies of the vastus lateralis taken before, immediately after, and 5 days after downhill running, showed an immediate and prolonged increase in IL-1b (CANNON et al., 1989). This study implicates muscle IL-1b in the post-exercise change in protein metabolism.

5. Exercise and protein metabolism

Eccentric exercise induced increases in muscle hydrolase activity (VIKHO, SALMINEN and RANTAMAKI, 1979); intracellular [Ca2+ ] (DUAN et al., 1990), IL-1b, and urinary 3-methylhistidine levels indicate that muscle protein turnover is also increased. FIELDING et al. (1991) used a primed, constant infusion of 1-13C-leucine before, immediately after, and 10 days after a single bout of high-intensity eccentric exercise in previously sedentary old and young men to estimate whole-body protein metabolism. Unlike the studies of DEVLIN et al. (1990) and CARRARO and co-workers (1990) who used concentric exercise, FIELDING et al. (1991) found that leucine oxidation and flux were significantly elevated (compared to pre-exercise samples) at both post-exercise time points, indicating a prolonged increase in protein turnover.

These results tend to support those showing that previously sedentary young men consuming 1 g protein/kg/d showed an increased urinary nitrogen excretion and a prolonged period of negative nitrogen balance when beginning a vigorous exercise program (GONTZEA, SUTZESCU and DUMITRACHE, 1974). In this study, the responses of elderly and young men were compared. While there was no difference in leucine flux before or after the exercise between the young and older men, the older subjects showed a significantly greater rise in 3-methylhistidine/creatinine in the days following the eccentric exercise.

In addition, muscle biopsies of the vastus lateralis, taken before, immediately after and 10 days after the exercise, showed that the older men had a significantly greater number of damaged muscle fibers than the young subjects (MANFREDI et al., 1991). These studies indicate that the need for dietary protein may be higher at the initiation of training. They also demonstrate a greater increase in myofibrillar protein breakdown as a percentage of total body breakdown in response to training in older subjects.

FRONTERA et al. (1988) demonstrated that older (age 60-72 years) sedentary men have the capacity to significantly increase both the size and strength of their muscles. Using a progressive resistance training (PRT) program (80% of the one repetition maximum, 3 days per week for 12 weeks), we demonstrated that muscle hypertrophy was associated with a significant post-training elevation in urinary 3-methylhistidine/creatinine. This PRT program had a substantial eccentric component, which almost certainly resulted in significant damage in the knee extensor and flexor muscles.

Half of the men who participated in this study were given a daily protein-calorie supplement (S) providing an extra 560 16 kcal/d (16.6% as protein, 43.3% as carbohydrate, and 40.1% as fat) in addition to their normal ad lib diet. The rest of the subjects received no supplement (NS) and consumed an ad lib diet. By the twelfth week of the study, dietary energy (2960 230 in S vs 1620 80 kcal in NS) and protein (118 10 in S vs 72 11 g/d in NS) intake were significantly different between the S and NS groups.

Composition on the midthigh was estimated by computerized tomography and showed that the S group had greater gains in muscle than did the NS men. In addition, urinary creatinine excretion was greater at the end of the training in the S group when compared to that of the men in the NS group (MEREDITH, FRONTERA and EVANS, 1992), indicating a greater muscle mass in the S group. The change in energy and protein intake (beginning vs 12 weeks) was correlated with the change in midthigh muscle area (r=0.69, p=0.019; r=0.63, p=0.039, respectively). There was no difference in strength gains between the two groups. These data suggest that a change in total food intake, or perhaps, selected nutrients, in subjects beginning a strength-training program can affect muscle hypertrophy.

It is clear that exercise-induced muscle damage leads to a long-term increase in protein breakdown and synthesis (FRONTERA et al., 1988; FIELDING et al., 1991; CANNON et al., 1991; EVANS, 1986). Few studies have compared the longitudinal effect of high-intensity eccentric and concentric exercise training. Most progressive resistance training devices and lifting free weight have substantial concentric and eccentric components. KOMI and BUSKIRK (1972) measured arm circumference before and after training either eccentrically or concentrically. They found that arm circumference increased only in the arms of men who trained eccentrically.

CIRIELLO et al. (1983) examined the effects of 4 months of high-intensity strength training on a Cybex isokinetic dynamometer which has little or no eccentric component. Although the strength of the subjects increased significantly, there was no evidence of hypertrophy of Type I or Type II muscle fibers. PEARSON and COSTILL. (1988) examined the effects of a progressive resistance weight-training protocol (with eccentric and concentric components) on one leg and isokinetic training (with no eccentric component) on the contralateral leg. Only the leg trained with a significant eccentric component increased in size as a result of the training.

Recently, a direct comparison of the effects of strength training with eccentric and concentric or concentric exercise alone was made (DUDLEY et al., 1991; COLLIANDER and TESCH, 1990). These studies showed that increases in strength were greater following a program of maximum concentric and eccentric muscle actions than resistance training using concentric muscle actions only. The evidence suggests that eccentric exercise-induced skeletal muscle damage and its subsequent repair are important for increasing muscle fiber size in respone to strength training. Not only can resistance training increase muscle size, but a recent report indicated that long-term resistance training may prevent age-associated changes in histochemical fibre-type distribution, myosin heavy chain isoforms and tropomyosin isoforms (KLITGAARD et al., 1990a; b).

6. Summary

Prolonged submaximal exercise increases the oxidation of indispensable amino acids which almost certainly increases the dietary requirement for protein. For this reason, 1.2 g/kg/d (twice the standard deviation seen by MEREDITH et al., 1989b) should serve as a recommended dietary protein intake for endurance athletes or those performing physically demanding occupations consuming a eucaloric diet. Those who attempt to lose weight by increasing their activity levels and decreasing their energy intake will have an even greater dietary protein requirement.

In previously sedentary individuals, the initiation of a weight-lifting program may also increase the need for dietary protein by increasing the rate of skeletal muscle protein. This seems to be particularly true in the elderly. However, there is no evidence that strength and power athletes already adapted to their sport, and who perform little or no endurance exercise, have a high dietary protein requirement.

Contents - Previous - Next