Weight, food intake, and body composition: effects of exercise and of protein deficiency

Transcription

1 AMERICAN JOURNAL OF PHYSIOLOGY Vol. 216, No. 2, February Printed in U.S.A. Weight, food intake, and body composition: effects of exercise and of protein deficiency EUGENE L. CREWS III, K. WILLIAM FUGE, LAWRENCE B. OSCAI, JOHN 0. HOLLOSZY, AND ROBERT E. SHANK Department of Preventive Medicine, Washington University School of Medicine, St. Louis, Missouri CREWS, EUGENE L. III, K. WILLIAM FUGE, LAWRENCE B. OSCAI, JOHN 0. HOLLOSZY, AND ROBERT E. SHANK. kt eight, food intake, and body cornposition: effects of exercise and protein dejiciency. Am. J. Physiol. 216(2): The effects of exercise and of protein deficiency, both separately and together, on food intake, weight gain, and body composition were studied. Exercising animals gained less weight than free-eating sedentary controls as a result of both an increase in caloric expenditure and a decrease in food intake. Exercise resulted in a significant decrease in the percentage of the carcass composed of fat, with a proportional relative increase in lean body mass. Sedentary calorie-restricted animals had a higher fat content than exercising animals of the same weight, suggesting a specific fat-mobilizing effect of exercise. Growing rats eating a lowprotein diet (8y0 casein) were markedly stunted but otherwise appeared healthy. They had a greater capacity for prolonged running than the rats on the normal protein diet. Protein deficiency resulted in a decrease in the percentage of the carcass composed of fat and an increase in the proportion made up of water. Because of the small size of the protein-deficient animals, total amounts of all four body constituents were markedly decreased. caloric restriction; dietary adequacy; longevity R ELATIVELY LITTLE inforlnation is available regarding the effects of regularly performed endurance exercise, such as prolonged running, on weight gain, food intake, and body composition in growing animals. Stevenson et al (16) f ound that prolonged exercise decreases the voluntary food intake of rats. Jones et al (6) found that growing rats subjected to a program of swimming gained weight more slowly and had a lower body-fat content than sedentary controls. Whether or not appetite suppression contributed to this effect of swimming on bodyfat content is not known, as food intake was not measured. A reduction of the number of calories available for growth can markedly affect body size and composition ( 1, 7). It appeared of considerable interest, therefore, to 359 determine whether or not exercise has a specific effect on body composition separate from an associated relative caloric deficit. One purpose of the present study was to elucidate this point and to obtain further information regarding the effects of prolonged exercise on food intake and weight gain. Another variable which can have a profound effect on growth and on body composition is the availability of protein ( 12, 18). In previous investigations on the effects of dietary protein restriction on body composition, very severe degrees of protein deficiency, incompatible with growth or prolonged survival were studied ( 12, 18). In the present study, the effects of protein deficiency severe enough to cause marked stunting, yet compatible with prolonged survival and good exercise tolerance, on food intake, weight gain, and body composition were investigated. Information was also obtained regarding the interacting effects of protein deficiency and exercise, and of protein deficiency and caloric restriction. The purpose of the latter studies was to obtain leads regarding the physiological mechanisms by which the effects of protein deficiency and of exercise on body composition are mediated. METHODS Animal care, diets, and exercise program. Male rats of a Wistar strain (specific-pathogen-free CFN rats), 7 weeks of age, were obtained from Carworth Farms and kept in individual cages. One group was fed a diet low in protein content composed, in terms of percent of total weight, of 8 % casein, 74 % sucrose, 4 % corn oil, 10 % Crisco, 3 % Hegsted salt mixture, and 1 % vitamin-fortification mixture. This diet is subsequently referred to as the CCproteindeficient diet. A second group was fed a diet normal in protein content containing 22 % casein, 60 % sucrose, 4 % corn oil, 10 % Crisco, 3 % Hegsted salt mixture, and 1 % vitamin-fortification mixture. This diet is subsequently referred to as the normal protein diet. The food was placed in circular aluminum containers with tightly fitting covers which had a 1.8-inch diam round

2 360 CREWS ET AL. opening. The covers and containers were securely fastened to the walls of the cages to prevent the animals from tipping them over. A perforated steel disk was placed in the container on top of the food to prevent the animals from scattering it. A careful record was kept of the weight of the food eaten per day. Each dietary group was subdivided into an exercising and two sedentary groups. The exercising rats were trained to run on a motor-driven treadmill as described previously (5, 11). The treadmill was set at an 8 incline. The animals were exercised 5 days/week. They were subjected to the U-week training program used routinely in this laboratory ( 11). Th e initial work load consists of 10 min of running at 22 m/min, twice daily, 4 hr apart. The amount of exercise is progressively increased over the 1 st 3-week period to 40 min of running at 28 m/min, twice daily, 4 hr apart; over the 2nd 3-week period to 60 min at 31 m/min, twice daily, 4 hr apart; over the 3rd 3-week period to 120 min at 3 1 m/min once daily; over the final 3 weeks to 120 min at 31 m/min, with 12 sprints at 42 m/min, lasting 1 min each, interspersed at lo-mm intervals through the exercise period. They were maintained at the final work load until they were sacrificed; this period varied from 12 to 18 days. The free-eating sedentary group was provided with food ad libitum. The paired-weight sedentary group had their food intake adjusted so as to maintain their body weights approximately the same as those of the exercised animals. The sedentary animals were not subjected to treadmill running. The initial weights of the six subgroups were approximately the same at the beginning of the study. Carcass analysis. The rats were fasted for 16 hr and then killed with chloroform. Hair was removed with a mixture consisting of barium sulfide (200 g), detergent (Tide, 50 g), and glycerol ( 500 ml). Analyses were performed on the carcasses, from which feces had been removed, by the method of Mickelsen and Anderson (9) with the following modifications. After homogenizing the autoclaved rat carcass in a Waring Blendor, a stable suspension was obtained by sonicating the homogenate in a Blackstone ultrasonic cleaning tank model CT.5 at maximum intensity for 10 min, instead of passing it through a colloid mill. Instead of petroleum ether, alcohol-ether 3 z 1 (v/v) was used for the fat extraction as described by Entenman (2). For the purpose of this study, the term lean body mass is defined as the total body weight minus total body fat. A4ateriaZ.s. Vitamin-free test casein, Hegsted salt mixture, and GBI vitamin-fortification mixture were obtained from General Biochemicals. program outlined under METHODS. They became severely fatigued after min of continuous running at 30 m/min, and appeared unable to increase their exercise capacity beyond this point. They, therefore, had to be maintained at this work level after the 1st 6 weeks of the training period. In contrast, the protein-deficient rats went through the training program without difficulty, and showed no obvious signs of fatigue after the exercise sessions, which involved 2 hr of running per day after the 1st 6 weeks of the study. Because of the differences in the duration of the exercise sessions no attempt is made, in the following sections, to directly compare the exercising rats in the two dietary groups. However, the greater body weights of the animals on the normal protein diet do compensate, to some extent, for their shorter exercise sessions, in terms of the total amount of work performed. Effects of the exercise on weight gain and caloric intake. The exercising rats in both groups gained significantly less weight than their sedentary free-eating controls (Table 1). Part of this effect could be ascribed to the decreased appetite of the exercising rats, which was evidenced by a significantly lower caloric intake (Table 2). This appetite-depressing effect of the exercise was much more evident in the animals on the normal protein diet (Table 2). The caloric intake of the sedentary paired-weight controls was significantly lower than that of the exercising rats in both groups; this difference was much greater in the protein-deficient animals that, as mentioned above, exercised for longer periods of time. Effects of protein dejciency on weight gain and caloric intake in sedentary rats. Rats permitted unrestricted access to the TABLE I. Weight gains ouer a 3-month period Group Weight Gain, g Normal protein 319zt21 184H9 135 <O.OOl 187ztlO Protein deficient 1 116ztll 1 75zk7 ( 41 1 <O.Ol ( 63zt6 TABLE Values are expressed as & the SE of the mean for six rats. Group 2. Caloric intakes over a 3-month period Food Intake, Calories Sedentw, pairedweight rats P Diff t P RESULTS Response to diets and to exercise program. The proteindeficient rats had somewhat coarse coats and were markedly stunted, but otherwise appeared healthy and alert. The rats on the normal protein diet appeared sleek and healthy, but were unable to complete the training Normal tein Protein ficient pro- de- 7,638 6,090 1,548 <O.Ol 5, zk224 zk261 *43 6,037 5, <0.05 4,370 1,170 zt230 MO3 It53 Values are expressed as means -+I the SE of the mean for six rats. * Sedentary free-eating minus exercising. t Exercising minus sedentary paired-weight.

3 EXERCISE, PROTEIN DEFICIENCY, AND BODY COMPOSITION 361 protein-deficient diet gained weight at a much slower rate than the free-eating animals on the normal protein diet. The total weight gain of the free-eating, proteindeficient animals, over a 3-month period, was only about one-third as great as that of the comparable control group (P < 0.001) (Table 1). This difference can be partially explained on the basis of the significantly lower caloric intake of the protein-deficient animals (P < 0.001) (Table 2). Effects of exercise on body composition. In both dietary groups, exercise resulted in a decrease (P < 0.01) in the percentage of the carcass composed of fat (Table 3). Most of the difference between the body weights of the exercising and the sedentary free-eating, protein-deficient rats can be accounted for on the basis of the lower fat content of the exercisers (Table 4). The exercising animals on the normal protein diet similarly had a markedly reduced total body-fat content (Table 4); in addition, they had a slightly reduced lean body mass (300 =t 13 vs g for the sedentary free eaters; P < 0.05). Efects of protein dejciency on body composition. A considerable change in the relative proportions of body fat and water resulted from protein deficiency, as can be seen from a comparison of the sedentary, free-eating animals in the two dietary groups (Table 3). The percentage of the carcass composed of fat was decreased (P < O.Ol), while the proportion of the body made up of water increased (P < 0.01). Tabulation of the data in terms of the total amounts of protein, fat, ash, and water in the carcasses provides a different perspective on the effects of protein deficiency. As shown in Table 4, the bodies of the sedentary, freeeating animals on the protein-deficient diet contained approximately two-thirds as much protein, water, and minerals (P < O.OOl), and only 40 % as much fat (P < O.OOl), as those of the comparable animals on the normal protein diet. Effects of caloric restriction on body composition. In the TABLE 3. Effects of exercise, of caloric restriction, and of protein dejciency on body composition Diet Group Subgroup Body Protein, wt, g % Protein de- Runners ficien t *lo Iko.9 Sed paired wt St6 Ito. Sed free eating zt16 ho.6 Normal pro- Runners tein rt18 2to.4 Sed paired wt zt Sed free eating Fat, % *$ *I.3 zko *I.9 zto zk2.6 zto h1.2 zto zt1.1 *O.l ho.6 zko.1 ztl8 zko.3 - Values are expressed as means b the SE of the mean rats. WFer zto zt ~ *I zko zto.3 for six TABLE 4. Effects of exercise, of caloric restriction, and of protein dejciency on total-body content of protein, fat, minerals, and water Diet Group Subgroup Body w g Protein, g Fat, g Protein ficien Normal tein det pro- Runners Sed paired wt Sed free eating Runners Sed paired wt Sed free eating 238 *lo 229 zk6 273 & zt *4 465 zt zt zk zt zt * st zt zt zt1o.c 40.3 zt zt &7.2 Ash, g 9.1 zto ztzo ZtO zko ho zko.5 Water g zt zt zt It zt zt9.7 values are expressed as means & the SE of the mean for six rats. animals on the normal protein diet, caloric restriction produced a considerable change in the relative proportions of the various body constituents (Table 3). The percentage of the body composed of fat was significantly lower (P < O.Ol), and the relative proportions of ash and water were significantly greater (P < 0.01 ), in the paired-weight rats than in the free-eating sedentary controls (Table 3). Because of their smaller size, the carcasses of the paired-weight sedentary rats on the normal protein diet contained significantly lower total amounts of protein (P < O.Ol), minerals (P < O.Ol), and water (P < O.Ol), as well as of fat (P < O.OOl), than those of the comparable, free-eating sedentary animals (Table 4). In the protein-deficient group, there was no significant difference between the calorie-restricted and the freeeating sedentary animals with respect to their relative proportions of body protein, fat, minerals, and water (Table 3). Th e 1 ower weights of the paired-weight animals relative to the free-eating, sedentary ones, were due to both a lower total fat content and a smaller lean body mass (Table 4). DISCUSSION Jones et al have reported that growing rats subjected to a program of endurance exercise gain weight more slowly than sedentary controls (6). The results of the present study confirm this finding and demonstrate that both an increase in caloric expenditure and a decrease in caloric intake are responsible for the lower body weights of the exercising animals. The reduction in food intake was much more marked in the exercising animals on the normal protein diet than in those on the proteindeficient diet. The rats on the normal protein diet were exhausted at the end of their exercise sessions in contrast to the smaller, protein-deficient animals that did not appear fatigued, despite a longer period of running. This raises the possibility that the degree of appetite suppres-

4 362 CREWS ET AL. sion may be related to the amount of stress evoked by the work load, rather than to the amount of exercise per se. It appears possible that the appetite suppression is mediated by the increased levels of catecholamines associated with the exercise stress ( 15, 17). In the study by Jones et al the difference between the final body weights of the exercising and the sedentary free-eating animals, of approximately 60 g, could be explained entirely by the lower fat content of the carcasses of the exercising animals (6). In the present study, fat was also the body constituent most markedly affected by exercise. However, in contrast to the finding of Jones et al (6), total lean body mass was also significantly lower in the exercising than in the sedentary free-eating rats on the normal protein diet. It appears likely that a larger relative caloric deficit in our rats is responsible for this apparent discrepancy. Evidence for this explanation is provided by the greater difference in body weight of 126 g, between the exercising and sedentary free-eating animals on the normal protein diet in the present study (vs. 60 g in the study by Jones et al). The exercising animals were leaner than the sedentary calorie-restricted animals of the same weight. This finding points to a specific lipid-mobilizing effect of exercise in addition to the increase in caloric expenditure and the decrease in caloric intake. It has been demonstrated that vigorous exercise results acutely in a large increase in fatty acid mobilization (3, 13). This lipolytic action appears to be mediated, in large part, by the increase in catecholamine levels associated with exercise ( 10, 15). In light of the foregoing, the difference in body-fat content between the sedentary, calorie-restricted and the exercising animals of the same weight could reflect the cumulative effect of daily episodes of lipolysis induced by exercise. The marked stunting associated with protein deficiency was found to be due to a decrease in all four of the body constituents measured. The protein-deficient diet caused a marked suppression of appetite, with a voluntary decrease in food intake. It appears likely that the low body-fat content of the protein-deficient animals was caused in part by this decrease in caloric intake. As also occurs in response to exercise, total body-fat content was the variable most markedly lowered by protein deficiency, being almost as low in the sedentary, free-eating, protein-deficient animals as in the exercisers on the normal protein diet. However, protein deficiency had a much more markedly depressing effect on total body protein, mineral, and water content than did exercise. Decreased availability of protein per se, rather than of calories, appears to be the most important factor limiting growth of the lean body component of the proteindeficient rats. This is evidenced by the finding that the paired-weight sedentary animals on the normal protein diet had a significantly greater total-body protein content, despite having fewer calories available for growth than the sedentary free-eating, protein-deficient animals. Furthermore, the lean body mass of the exercising, protein-deficient rats was the same as that of their sedentary free-eating controls despite a lower caloric intake and greater caloric expenditure. Meyer and Hargus (8) have reported that rats on a low-protein diet subjected to 4 hr of swimming daily, ate significantly more food and gained more weight than comparable sedentary controls on the same diet. This is in contrast to our finding that exerci sing, protei n-defici ent rats ate less food and gained less weight than sedentary protein-deficient controls. Conceivably, this discrepancy could be due to the difference in the type and amount of exercise used; further work is needed to clarify this point. In the protein-deficient group, the caloric deficit of the exercising animals, relative to the free-eating controls, was reflected only in a reduction of body-fat content. In contrast, the caloric deficit of the food-restricted sedentary, protein-deficient animals was reflected in a reduction of both lean body mass and total fat content. As a result, the exercising animals had a greater lean body mass than sedentary rats of the same weight. Similarly, the exercising animals on the normal protein diet had a greater lean body mass than the sedentary, calorierestricted rats of the same weight. These findings must reflect a difference in the manner in which the limited number of calories available for growth are utilized by the exercising and the sedentary calorie-restricted animals. It seems possible that the fat-mobilizing effect of exercise, mentioned earlier, could result in a greater channeling of available calories into the synthesis of lean tissue, with less storage of calories in the adipose cells. The greater lean body mass of the exercising animals, as compared to the pair-weighed controls, could also reflect a specific stimulus to amino acid conservation and protein synthesis. An additional factor which may have contributed to the lower lean body mass of the calorie-restricted sedentary animals is a lower protein intake secondary to their smaller total food consumption. In dietary studies on animals, the criterion most commonly used to judge the adequacy of a diet is the rate of weight gain. By this standard, the protein-deficient diet was grossly inferior to the normal protein diet in the present study. However, when other criteria are used, such as the effect of a diet on exercise capacity or life span, a different picture emerges. Ross (14) has shown that, under ad libitum conditions of feeding, animals eating a diet low in protein (8 CT0 casein) have a considerably longer life-span than rats eating a diet containing 21.6 % casein (832 vs. 600 days). In the present study, using comparable diets, it was found that the stunted, protein-deficient anirnals have a far greater capacity for prolonged running than the heavier animals on the normal protein diet. It should not be assumed from the foregoing that these beneficial effects of the low-protein diet are mediated by protein deficiency per se. The studies of Ross (14) and the results of work in this laboratory suggest that the decrease in caloric intake and slow rate of weight gain associated with a moderate degree of protein deficiency are responsible for the longer life-span and greater

5 EXERCISE, PROTEIN DEFICIENCY, AND BODY COMPOSITION 363 capacity for prolonged exercise. Thus, when the caloric intake of rats eating diets normal or high in protein content was markedly restricted, their life-span was approximately 100 days longer than that of the rats eating the protein deficient diet ad libitum (14). Similarly, when the caloric intake of rats eating a diet with a protein content of was restricted so as to keep their weights in the same range as those of the protein-deficient animals, they were able to run at least as long as the proteindeficient animals in an all-out treadmill run to exhaustion (4). The authors express their appreciation to Mrs. May Chen for technical assistance, and to Mrs. Lenore Alberty for help in preparation of the manuscript. This investigation was supported by Public Health Service Research Grant HDO16 13 from the National Institute of Child Health and Human Development and by Training Grant AM from the National Institute of Arthritis and Metabolic Diseases. E. L. Crews III was a Student Summer Research Fellow supported by a Norman Jollife Fellowship. L. B. Oscai was a Postdoctoral Trainee in Nutrition supported by Training Grant AM Reprint requests and inquiries should be directed to J. 0. Holloszy. A / Received for oublication 13 Mav REFERENCES ELKINTON, J. R., AND E. M. WIDDOWSON. Effect of chronic undernutrition on body composition in the rat. MetaboZism 8: , ENTENMAN, C. General procedures for separating lipid components of tissue. In: kfethods in Enzymology, edited by S. P. Colowick and N. 0. Kaplan. New York: Academic, 1957, vol. 3, p FRIEDBERG, S. J., W. R. HARLAN, JR., D. L. TROUT, AND E. H. ESTES, JR. The effect of exercise on the concentration and turnover of plasma nonesterified fatty acids. J. Clin. Invest. 39 : , FUGE, K. W., E. L. CREWS III, P. K. PATTENGALE, J. 0. HOLLOSZY, AND R. E. SHANK. Effects of protein deficiency on certain adaptive responses to exercise. Am. J. Physiol : , HOLLOSZY, J. 0. Biochemical adaptations in muscle. J. BioZ. Chem. 242: , JONES, E. M., H. J. MONTOYE, P. B. JOHNSON, S. M. J. M. MARYSIN, W. D. VAN Huss, AND D. CEDERQUIST. Effects of exercise and food restriction on serum cholesterol and liver lipids. A?n. J. Physiol. 207 : , MCCANCE, R. A., AND E. M. WIDDOWSON. Nutrition and growth. Proc. Roy. Sot., London, Ser B. 156: , MEYER, J. H., AND W. A. HARGUS. Factors influencing food intake of rats fed low-protein rations. Am. J. Physiol. 197 : , MICKELSEN, O., AND A. A. ANDERSON. A method for preparing intact animals for carcass analyses. J. Lab. CZin. Med. 52: 28% 290, MUIR, G. G., D. A. CHAMBERLAIN, D. T. PEDOE. Effects of p- sympathetic blockade on non-esterified-fatty-acid and carbohydrate metabolism at rest and during exercise. Lancet 2: , PATTENGALE, P. K., AND J. 0. HOLLOSZY. Augmentation of skeletal muscle myoglobin by a program of treadmill running. Am. J. Physiol. 213 : , POND, W. G., R. H. BARNES, R. B. BRADFIELD, E. KWONG, AND L. KROOK. Effect of dietary energy intake on protein deficiency symptoms and body composition of baby pigs fed equalized but suboptimal amounts of protein. J. Nutr. 85: 57-66, RODAHL, K., H. I. MILLER, AND B. ISSEKUTZ, JR. Plasma free fatty acids in exercise. J. ApjZ. Physiol. 19 : , Ross, M. H. Length of life and nutrition in the rat. J. Nutr. 75: , RUSSEK, M., AND S. PINA. Conditioning of adrenalin anorexia. Nature 193 : 1296-l 297, STEVENSON, J. A. F., B. M. Box, V. FELEKI, AND J. R. BEATON. Bouts of exercise and food intake in the rat. J. AppZ. Physiol, 21: , VENDSALU, A. Plasma concentrations of adrenaline and noradrenaline during muscular work. Acta. Physiol. Stand. SupfiZ. 173 : 57-69, WIDDOWSON, E. M., AND R. A. MCCANCE. Effect of a lowprotein diet on the chemical composition of the bodies and tissues of young rats. Brit. J. Nutr. 11 : , 1957.