Mechanisms of Ageing and Development, 61 (1991) 123--133 Elsevier Scientific Publishers Ireland Ltd.
123
AGEING, EXERCISE AND FOOD RESTRICTION: EFFECTS ON SKELETAL MUSCLE GLUCOSE UPTAKE
J.L. IVY*, J.C. YOUNG**, B.W. CRAIG***, W.M. KOHRT AND J.O. HOLLOSZY Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110 (U.S.A.) (Received March 14th, 1991) (Revision received April 20th, 1991)
SUMMARY
We investigated the effects of ageing, exercise and food restriction on glucose uptake by muscles perfused with physiological concentrations of insulin and glucose in male Long-Evans rats. The rate of glucose uptake by hindlimb muscles perfused with medium containing 50 #U//ml insulin and 8 mM glucose was the same in 9-10 month-, 18 month-, and 24-month-old rats. Rats exercised by means of swimming 3 h/day, 5 days/week, had significantly higher rates of muscle glucose uptake than did the sedentary freely eating rats. Paired-weight sedentary rats, that were food restricted so as to keep their weights in the same range as those of the swimmers, had a hindlimb glucose uptake rate similar to that of the swimmers, and greater than that of the sedentary freely eating rats. The 24-month-old sedentary freely eating rats showed a trend toward a higher plasma glucose response and a lower plasma insulin response to a glucose tolerance test. The 24-month-old swimmers showed no deterioration in glucose tolerance compared to the 9-10 month-old rats. Our findings argue against the concept that ageing results in skeletal muscle insulin resistance in rats.
Key words." Ageing; Exercise; Food restriction; Glucose uptake; Muscle Correspondence to: John O. Holloszy, Department of Internal Medicine, Campus Box 8113, Washington University School of Medicine, 4566 Scott Avenue, St. Louis, MO 63110, U.S.A. *Present address: Department of Kinesiology and Health Education. University of Texas, Belmont Hall 222, Austin, TX 78712. **Present address: Department of Health Sciences, Boston University, 635 Commonwealth Avenue, Boston, MA 02215. Present address." Human Performance Laboratory, Ball State University, Muncie, IN 47306. 0047-6374/91/$03.50 Printed and Published in Ireland
© 1991 Elsevier Scientific Publishers Ireland Ltd.
124
INTRODUCTION Studies on human subjects using the euglycemic clamp procedure have provided evidence that ageing is associated with development of insulin resistance [1-4]. Skeletal muscle is the major site of insulin-mediated glucose disposal [5] and the slower rates of glucose disposal in older people during the euglycemic clamp is thought to be primarily due to development of skeletal muscle insulin resistance [1-4]. The results of studies on rats, the species most commonly used in studies of ageing, have also been interpreted by some investigators to indicate that ageing results in insulin resistance [6-8]. The role of ageing is not clearly established, however, because the studies that showed a decrease in insulin action involved comparisons of juvenile rats with mature [6] or old [7,8] rats. In this context, we have investigated the effect of ageing by comparing the rates of glucose uptake by skeletal muscles perfused with physiological concentrations of insulin and glucose in young adult (9-10 months), middle-aged (18 months), and old (24 months) freely eating sedentary male L o n g - E v a n s rats. Obesity is frequently associated with development of insulin resistance [9], while exercise can result in increased insulin sensitivity [10,11]. It, therefore, seems possible that weight gain and physical inactivity could contribute to insulin resistance in older individuals. We, therefore, also examined the interactions between the effects of ageing and regularly performed exercise or food restriction on glucose uptake by muscles perfused under the same conditions, and in the same age groups, as for the sedentary freely eating rats. MATERIALS AND METHODS Animal care and exercise program
Three groups of male specific-pathogen-flee L o n g - E v a n s rats were obtained from Charles River (Wilmington, MA) at age 3 months. Subgroups of these rats were necropsied shortly after arrival. Histopathologic examination, culture of the respiratory tract, tympanic bullae and gastrointestinal contents, and testing of serum for titers of antibodies against pathogenic viruses and mycoplasma confirmed that they were pathogen free. Additional rats from our aging rat colony were subjected to the same procedures at ages of - 12 months and - 2 4 months, and were also found to be pathogen free. Groups of rats were studied at the ages of 9-10 months, 18 months and 24 months. The 18- and 24- month-old rats were from the initial group of animals. The two additional groups of rats were obtained 12 and 18 months after the initial group so that young adult (9-10 month) rats could be studied at the same times as the 18- and 24-month-old animals. The rats were housed in individual stainless steel cages measuring 7 x 14 x 8 in, in a temperature and light controlled animal room with its own ventilation system, with 15 air exchanges per hour (100% intake and 100% exhaust) in a facility in which
125 no other rats were housed. To avoid introducing infections into the rat colony, the animal caretakers did not work with other rodents or in areas where they could be exposed to other rodents. The rats were fed a diet of Purina Rat Chow and water. All of the rats were fed ad libitum until age 5 months, when they were randomly assigned to either a sedentary freely eating group, a freely eating exercise group, or a sedentary paired-weight group that was food restricted to keep their body weights in the same range as those of the exercisers. The exercise program consisted of swimming 5 days/week, in groups of six, in water maintained between 33 and 35°C in stainless steel barrels filled to a depth of 60 cm. The duration of the exercise sessions was gradually increased over 4 weeks until the rats were swimming 3h/day.
Hindlimb perfusion After an overnight fast, and for the swimmers 45-52h after their last exercise session, the rats were anesthetized with pentobarbital sodium (5 mg/100 g body wt) injected intraperitoneally. The rationale for the overnight fast was to standardize as much as possible the milieu to which the skeletal muscles of the rats in the different groups were exposed in the period immediately preceeding the hindlimb perfusion study, especially with respect to insulin and glucose. Surgical preparation of the hindquarter was performed as described by Ruderman et ai. [12]. The perfusion procedure and apparatus have been described previously [13]. The perfusion medium consisted of Krebs-Henseleit bicarbonate (KHB) buffer containing 4 g/100 ml of dialyzed fatty acid poor bovine serum albumin (Cohn fraction V; Miles Laboratory), blood bank time-expired human erythrocytes, rejuvenated by the procedure of Valeri [ 14], at a concentration of 12 g hemoglobin/100 ml, 50/zU/ml of purified pork insulin (Squibb), and, when present, 8 mM glucose. The left iliac artery was ligated, limiting perfusate flow to the right hindlimb. The hindlimb preparation was initially washed out with 50 ml of KHB. A flow-through perfusion at 8 ml/min with medium containing 50 ttU/ml insulin, but no glucose was started. After 15 min of perfusion without glucose, perfusion with medium containing 8 mM glucose and 50/zU/ml of insulin was begun. The hindlimb was then perfused with this medium for 20 min, and glucose uptake was measured during the last 15 min during which the venous effluent was collected on ice, and arterial and venous samples were taken at 5-rain intervals. Rates of glucose uptake were calculated from the perfusate flow rate and the arteriovenous difference in glucose concentration. Glucose concentration was measured enzymatically [15]. After the perfusions, the muscles that were perfused were identified by their staining after injection of bromophenoi blue (Mallinckrodt Chemicals) into the arterial circulation; the stained muscles were dissected out and weighed. Measurement of glucose tolerance After an overnight fast, the rats were anesthetized with 5 mg/100 g body wt. of sodium pentobarbital intraperitoneally. Respiration was maintained with a small
126 animal respirator (Phipps and Bird, Richmond, VA). A catheter was placed in the left carotid artery, tied in place, and used for taking blood samples. A baseline blood sample (0.8 ml) was taken. The iliac vein was then exposed, and glucose, 150 rag/100 g body wt., was injected intravenously (60% glucose in water). Additional 0.8 ml blood samples were taken (15, 30, 60 and 90 min) after the glucose load. Alter each blood sample was taken, 0.8 ml of 0.9% saline was injected via the carotid catheter to maintain blood volume. The blood samples were placed in tubes containing 0.1 ml EDTA (24 mg/ml) and kept on ice. Plasma was separated by centrifugation at 4°C and stored at -20°C until analysis. Glucose concentrations were determined by a glucose oxidase method (YSI glucose analyzer; Yellow Springs, OH). Plasma insulin was determined by radioimmunoassay [16]. The rat insulin standard was kindly given to us by Eli Lilly. After completion of the glucose tolerance tests, the anesthetized rats were killed by exsanguination, and various organs and skeletal muscles were dissected out and stored at - 70°C for use in other studies.
Calculations and statistics The total areas under the plasma glucose and insulin curves obtained during the intravenous glucose tolerance test (IVGTT) were determined by computer analysis. The statistical significance of differences between groups was assessed by a two-way analysis of variance (ANOVA). When statistical significance was indicated by ANOVA, differences were isolated using the Neuman-Keul post hoc test. RESULTS
Body weights and hindlimb muscle weights The body weights of the swimmers and paired-weight sedentary rats were 20-30% lower than those of the freely eating sedentary rats (Table I). There were no significant differences in body weight between the 9-10 month-, 18 month- and 24-monthold rats in any of the treatment groups. Average food intake between 8 and 24 months was similar for the sedentary freely eating animals (26 ~ 3 g/day) and the swimmers (28 ± 3 g/day); the food intake of the sedentary paired-weight rats, which was restricted to keep their body weights in the same range as those of the swimmers, averaged 22 -4- 2 g/day. The average weight of the perfused hindlimb muscles was significantly greater in the sedentary freely eating animals than in the swimmers or sedentary paired-weight rats (Table I). The weight of the hindlimb muscles decreased with age, being approximately 10% lower in the 24-month-old than in the 9-10-month-old rats.
Glucose uptake by perfused hindlimbs The effect of age on muscle permeability to glucose in the presence of an insulin concentration in the physiological range found in the fed state was evaluated in rat hindlimbs perfused with 8 mM glucose and 50 ~U/ml insulin. As shown in Table II,
127
TABLE 1 BODY WEIGHTS A N D PERFUSED H I N D L I M B MUSCLE WEIGHTS
Age (months)
Group Sedentary Freely eating
Swimmers Paired weight
Body weight(g) 9-10 18 24
656 4. 698 a (8) 689 4. 19 a (5) 701 ± 52 a (5)
522 ± 43 (7) 518 4- 54 (5) 493 4- 46 (5)
508 4- 46 (10) 552 4. 44 (4) 494 4. 43 (5)
23.8 4. 2.7 (7) 22.0 4. 3.4 (5) 21.6 4. 2.4 (5)
25.6 4. 1.9 (10) 23.5 4. i.8 (4) 23.2 4. 1.0 c (5)
Weight of perfused hindlimb muscles (g) 9-10 18 24
Values are means ap < 0.01 versus bp < 0.02 versus cp < 0.05 versus
30.1 4- 2.0 b (8) 28.0 4- 1.5 b (5) 26.9 4. 2.5 b'c (5) 4- S.D. for the number sedentary paired-weight sedentary paired-weight 9-month-old rats in the
of rats given in parentheses. and swimmer groups of the same age. and swimmer groups of the same age. same group.
when glucose uptake is expressed per hindlimb, there is a slight downward trend for glucose uptake with ageing, that did not attain statistical significance in any of the three treatment groups. Even this trend disappears when glucose uptake is expressed per gram of muscle to eliminate the effect of the - 10% decrease in the mass of the muscles with ageing (Table III). The rate of glucose uptake was lower in the hindlimb muscles of the sedentary freely eating rats than in those of the swimmers and the sedentary paired-weight rats.
TABLE I! GLUCOSE UPTAKE (p,mol " hindlimb -l • min -I) BY PERFUSED HINDLIMB
Age (months)
Group Sedentary
9-10 18 24
Swimmers
Freely eating
Paired-weight
6.94 4. 1.38 (8) 6.78 4. 1.04 (5) 6.54 4. 2.10(5)
8.57 4. 2.01 (7) 8.20 4. 1.85 (5) 7.81 ± 3.04(5)
Values are means ± S.D. for the number of animals given in parentheses. ap < 0.05 versus sedentary freely eating.
9.92 4. 3.88 a (10) 8.32 4. 2.60 (4) 8.66 4. 3.11 (5)
128 TABLE I11 GLUCOSE UPTAKE (t~mol"g-l. 15min-I) PER GRAM OF PERFUSED MUSCLE
Age (months)
Group Sedentary
9-10 18 24
Swimmers
Freely eating
Paired-weight
3.40 ± 0.89 (8) 3.65 ± 0.71 (5) 3.67 ± 1.22 (5)
5.40 + 1.23 (7)a 5.71 ± 1.62 (5)a 5.30 ± 1.68 (5)
5.74 ± 2.02b (10) 5.28 ± 1.55 (4) 5.62 + 2.09 (5)
Values are means ± S.D. for the number of animals given in parentheses. ap < 0.05 versus sedentary freely eating. bp < 0.01 versus sedentary freely eating.
When rates of glucose uptake expressed per hindlimb are compared in the age subgroups of the three treatment groups, only the 9-10-month-old swimmers are significantly different from the sedentary freely eating rats of the same age (Table II). When glucose uptake is expressed per gram of muscle, only the 9-10-month-old and 18-month-old paired-weight rats, and the 9-10-month-old swimmers, are significantly different from the sedentary freely eating rats (Table III). However, when the data are analyzed by treatment group without regard to age, i.e. when the three age subgroups are pooled, the rates of glucose uptake are significantly lower in the freely eating sedentary rats than in either the paired-weight sedentary rats (per hindlimb P < 0.05; and per gram muscle P < 0.01), or swimmers (P < 0.01, both per hindlimb and per gram muscle). There were no significant differences in hindlimb muscle glucose uptake between the swimmers and sedentary paired-weight rats.
Intravenous glucose tolerance tests ( I V G T T ) Plasma glucose response. Insulin resistance could be due to humoral or other systemic factors that are no longer present in washed out, perfused skeletal muscles. We therefore, performed IVGTTs on additional 9-10-month-old and 24-month-old swimmers and freely eating sedentary controls to evaluate the possibility that insulin resistance might be present in vivo in old rats. The weights of the rats used for the IVGTTs are shown in Table IV. As shown in Fig. 1, the plasma glucose responses to the I V G T T were similar in the young swimmers and the young sedentary rats; the areas under the glucose tolerance curves were also nearly identical (Table V). There was essentially no difference in the plasma glucose response during the IVGTT between the young and
129
T A B L E IV BODY W E I G H T S O F RATS USED F O R M E A S U R E M E N T O F G L U C O S E T O L E R A N C E
Age (months)
Sedentary freely eating
Swimmers
9-10 24
630 .4- 60 a (11) 618 .4- 62 b (8)
576 .4- 36 (10) 496 .4- 51 (8)
Values are means 4. S.D. for the number of rats given in parentheses. ap < 0.05 versus swimmers of same age. bp < 0.01 versus swimmers of same age.
FREELY FATING GROUP
SWIMMING GROUP 20
2O
18
18
10
10
14 I&J
.J 0
12
12
lo
10
a
e !
6 4
- ~ o MONTH --i- 24 MONTH
2 0
I
I
I
I
I
I
4
2 I
0
0 16 30 46 6O 76 00
I
r
TIME (,.in)
J
I
r
I
TiE (,.in) S'6 $3 $'6 2.7
3.6 $'6 S.O 2.7
I
I
0 16 30 46 O0 76 90
2.4 |.1
2.4
:L1
1.8
1.6 1.2
1.6
0"6
OJ O.I 0.1
1.2
OA 0.3
0.0
r
~
~ r
i
i
J
0 16 30 46 60 76 gO
TiME (mk)
0.0
l l l l
I I
O m ~ U ~ M
T~
(.:in)
Fig. 1. Plasma glucose and insulin responses to an intravenous glucose tolerance test. After an overnight fast, rats were anesthetized and given glucose 150 mg/100 g body wt intravenously. Points are means .4S.E.M. for 8-11 rats.
130 TABLE V AREAS UNDER THE IVGTT GLUCOSE AND INSULIN RESPONSE CURVES Age (months)
Sedentary freely eating
Swimmers
Glucose areas(mM 90 min "t)
9-10 24
751 ± 221 (11) 905 ± 373 (8)
769 ± 222 (10) 778 ~ 156 (8)
Insulin areas(ng" m1-1" 90 rain "l)
9-10 24
201 + 68 (11) 152 + 75b (8)
117 ± 65a (10) 74 ± 32~ (8)
Values are means + S.D. for the numbers of rats given in parentheses. ap < 0.05 versus sedentary rats in the same age group. bp < 0.10 versus 9-10 month-old sedentary group.
old swimmers (Fig. 1, Table V). In the freely eating sedentary animals the average plasma glucose concentrations were -20°/,, higher in the 24-month-old rats than in the 9 - 1 0 - m o n t h - o l d rats during the I V G T T . However, because o f the variability in response, neither the differences between the glucose concentrations at the individual time points nor the areas under the glucose curves, attained statistical significance. Insulin responses. There was a tendency for the plasma insulin levels to be lower in the 24-month-old than in the 9 - 1 0 - m o n t h - o l d rats (Fig. 1; Table V). A N O V A showed significant effects o f age on plasma insulin concentration at the 30 min and 60 min time points, and for the areas under the insulin curves (P < 0.05). However, the differences between the y o u n g and old rats within each treatment g r o u p did not attain statistical significance, except for the values at the 30-min time point in the sedentary freely eating group (P < 0.05). The swimmers had lower insulin responses than the sedentary freely eating rats, as reflected in significantly smaller areas under the insulin curves, both at 9 - 1 0 - m o n t h s - o l d and 24-months-old (Table V). Plasma insulin concentrations at all o f the time points except 90-min were significantly lower in the swimmers than in the sedentary freely eating rats both at 9 - 1 0 - m o n t h s and at 24-months (Fig. 1). DISCUSSION Glucose tolerance deteriorates in a high proportion of men and women as they grow older [17]. This decline in glucose tolerance appears to be due in large part to development o f insulin resistance o f insulin sensitive tissues [1-4]. Some investigators have interpreted the results o f studies on rats to indicate that ageing is also associated with development o f insulin resistance in this species [6-8]. The present results suggest that glucose tolerance does tend to deteriorate with ageing in free-
131 ly eating sedentary male rats (Fig. 1, upper panel). However, this trend toward higher glucose levels in the old rats, which was not seen in rats that exercised regularly, appears to have been due to a reduced insulin response to the glucose challenge [Fig. 1, lower panel], rather than to insulin resistance. Our finding of a smaller increase in plasma insulin concentration in response to the IVGTT in the 24-month-, compared to the 9-10-month-old rats, is in keeping with the finding of Elahi et al. [18] that isolated pancreas preparations from 23-month-old rats secreted less insulin than those of 12-month-old rats when perfused with 8 mM glucose. The results of the glucose tolerance tests fits well with our finding that the rate of glucose uptake in the presence of a physiological insulin concentration was not decreased in the perfused hindlimb muscles of the 18-month- or 24-month-old rats as compared to the 9-month-old rats. Skeletal muscle is the major site of insulinmediated glucose transport [5]. Our findings are, therefore, at odds with the concept that ageing results in development of insulin resistance in the rat. However, examination of the studies cited as showing development of insulin resistance with ageing in rats shows that what is actually a decrease in insulin action during growth and development has been misinterpreted as being due to ageing. In some of these studies 6-8-week-old rats were compared to 12-month-old rats [6,19]. In other studies, in which a wider range of ages was compared, it was found that insulin action decreased during growth and development, with no further deterioration attributable to ageing [7,20]. Goodman et al. [20] found, for example, that resistance to the action of insulin of perfused hindlimb muscles increased markedly in rats between 3 weeks and 24 weeks of age, with no further significant change between 24 and 96 weeks. Although the authors clearly stated that the decrease in insulin action occurred during early development [20], others have cited this study as evidence for development of insulin resistance due to ageing. Nishimura et al. [7], using the euglycemic clamp technique, also found that most of the decline in insulin action occurs between 2 months and 4 months of age in rats, and that there is no significant decline in the rate of insulinmediated glucose disposal between 10 months and 20 months; nevertheless, these authors refer to this decrease in insulin action as being due to ageing [7, 8]. Our finding that glucose uptake by hindlimb muscles perfused with a physiological concentration of insulin is the same in 9-10-month-, 18-month- and 24-month-old rats is in good agreement with the data of Goodman et al. [20] and Nishimura et ai. [7]. Another finding that has been cited as evidence that ageing results in insulin resistance in rats is that the adipocytes of older rats are markedly resistant to the action of insulin compared to those of juvenile animals [19]. However, this is also not an ageing phenomenon, but a preventable consequence of obesity and fat cell hypertrophy [21 ]. Exercise-training results in an increase in insulin sensitivity in humans [11] and rats [10]. An interesting aspect of this relatively short-lived adaptation is a blunted plasma insulin response to a glucose stimulus as the result of decreased insulin secre-
132 tion by the pancreas [22-24]. The increase in insulin sensitivity and the decrease in insulin secretion are closely matched so that normoglycemia is maintained despite the increased insulin sensitivity [25]. In keeping with these findings, the swimmers had a normal glucose tolerance despite a significantly blunted plasma insulin response to a glucose load. Furthermore, the swimmers' hindlimb muscles had a significantly greater rate o f glucose uptake when exposed to a submaximally effective insulin stimulus than did the muscles o f the sedentary freely eating rats. It is interesting that food restriction sufficient to keep the b o d y weights o f sedentary rats in the same range as those o f the swimmers was as effective as the swimming program in improving hindlimb muscle glucose uptake. In conclusion, glucose uptake by hindlimb muscles perfused with 8 m M glucose and 50/~U/ml insulin was the same in 9 - 1 0 - m o n t h , 18-month, and 24-month-old rats. This finding argues against the concept that ageing causes skeletal muscle insulin resistance in rats. Rats that were exercised regularly by swimming, and pairedweight sedentary rats that were food restricted to keep their b o d y weights in the same range as the swimmers', had higher rates o f glucose uptake by hindlimb muscles than did the sedentary freely eating rats. The swimmers also had normal glucose tolerance despite a blunted insulin response to a glucose tolerance test, providing further evidence for an increased susceptibility to the action o f insulin. ACKNOWLEDGEMENTS This research was supported by National Institutes o f Health Research G r a n t AG00425 from the National Institute on Ageing. J.L. Ivy, J.C. Y o u n g and B.W. Craig were the recipients o f N.I.H. National Research Service Awards AM06159, AG05216 and AM06123, respectively. REFERENCES 1 R.A. DeFronzo, Glucose intolerance and aging. Evidence for tissue insensitivity to insulin. Diabetes. 28(1979) 1095-1101. 2 R.I. Fink, O.G. Kolterman, J. Griffin and J.M. Olefsky, Mechanisms of insulin resistance in aging. J~ Clin. Invest. 71 (1983) 1523-1535. 3 J.W. Rowe, K.L. Minaker, J.A. Pallotta and J.S. Flier, Characterization of the insulin resistance of aging. J. Clin. Invest. 71 (1983) 1581-1587. 4 M. Chen, R.N. Bergman, G. Pacini and D. Porte, Jr., Pathogenesis of age-related glucose intolerance in man: insulin resistance and decreased A-cell function, Z Clin. EndocrinoL Metab. 60 (1985)13. 5 R.A. DeFronzo, E. Jacot, E. Jequier, E. Maeder, J. Wahren and J.P. Felber, The effect of insulin on the disposal of intravenous glucose. Diabetes 30 (1981) 1000-1007. 6 E. Reaven, D. Wright, C.E. Mondon, R. Solomon, H. Ho and G.M. Reaven, Effect of age and diet on insulin secretion and insulin action in the rat. Diabetes 32 (1983) 175-180. 7 H. Nishimura, H. Kuzuya, M. Okamoto, Y. Yoshimasa, K. Yamada, T. Ida, T. Kakehi and H. Imura, Change of insulin action with aging in conscious rats determined by euglycemicclamp. Am. J. Physiol. 254 (1988) E92-E98. 8 S. Kono, H. Kuzuya, M. Okamoto, H. Nishimura, A. Kosaki, T. Kakehi, M. Okamoto, G. Inoue,
133
9
10 11
12 13 14
15
16 17
18
19 20
21 22
23 24 25
I. Maeda and H. lmura, Changes in insulin receptor kinase with aging in rat skeletal muscle and liver. Am. J. Physiol. 259 (1990) E27-E35. B. Jeanrenaud, Hyperinsulinemia in obesity syndromes: its metabolic consequences and possible etiology. Metabolism 27 (1978) 1881-1892. D.E. James, E.W. Kraegen and D.J. Chisholm, Effects of exercise training on in vivo insulin action in individual tissues of the rat. J. Clin. Invest. 76 (1985) 657-666. D.S. King, G.P. Dalsky, W.E. Clutter, D.A. Young, M.A. Staten, P.E. Cryer and J.O. Holloszy, Effects of exercise and lack of exercise on insulin sensitivity and responsiveness. J. Appl. Physiol. 64 (1988) 1942-1946. N.B. Ruderman, C.R.S. Houghton and R. Hems, Evaluation of the isolated perfused rat hindquarter for the study of muscle metabolism. Biochem. J. 124 (1971) 639-651. M.J. Rennie and J.O. Holloszy, Inhibition of glucose uptake and glycogenolysis by availability of oleate in well-oxygenated perfused skeletal muscle. Bioehem. J. 168 (1977) 161-170. C.R. Valeri, Salvaging and rejuvenation of outdated red cells and cells with elevated 2,3-DPG levels and decreased affinity for oxygen. In D.M. Surgenor (ed.), The Red Blood Cell, 2nd edn., Vol. I, Academic Press, New York, 1974, pp. 528-532. H.U. Bergmeyer, E. Bernt, F. Schmidt, and H. Stork, D-glucose. Determination with hexokinase and glucose-6-phosphate dehydrogenase. I n H.U. Bergmeyer (ed.), Methods q f En=ymatic Analysis, 2nd edn., Academic Press, New York, 1974, pp. 1196-1201. V. Herbert, K.S. Lau, C.W. Gottlieb and S.J. Bleicher, Coated charcoal immunoassay of insulin. J. Clin. Endocrinol. 25 (1965) 1375-1384. M.B. Davidson, The effect of aging on carbohydrate metabolism: a review of the English literature and a practical approach to the diagnosis of diabetes mellitus in the elderly. Metabolism 28 (1979) 686-705. D. Elahi, D.C. Muller, D.K. Anderson, J.D. Tobin and R. Andres, The effect ofage and glucose concentration on insulin secretion by the isolated perfused rat pancreas. Endocrinology 116 (1985) 11-16. S. Narimiya, C.B. Dolkas, C.E. Mondon, C. Sims, D.W. Wright and G.M. Reaven, Insulin resistance in older rats. Am. J. Physiol. 246 (1984) E397-E404. M.N. Goodman, S.M. Dluz, M.A. McElaney, E. Belur and N.B. Ruderman, Glucose uptake and insulin sensitivity in rat muscle: changes during 3-96 weeks of age. Am. J. Physiol. 244 (1983) E93-EI00. B.W. Craig, S.M. Garthwaite and J.O. Holloszy, Adipocyte insulin resistance: effects of aging, obesity, exercise, and food restriction. J. Appl. Physiol. 62 (1987) 95-100. D.S. King, G.P. Dalsky, M.A. Staten, W.E. Clutter, D.R. Van Houten and J.O. Holloszy, Insulin action and secretion in endurance-trained and untrained humans. J. Appl. Physiol. 63 (1987) 2247-2252. D.S. King, M.A. Staten, W.M. Kohrt, G.P. Dalsky, D. Elahi and J.O. Holloszy, Insulin secretory capacity in endurance-trained and untrained young men. Am. J. PhysioL 259 (1990) EI55-EI81. H, Galbo, C.J. Hedeskov, K. Capito and J. Vinten, The effect of physical training on insulin secretion of rat pancreatic islets. Acta Physiol. Seand. 11 (1981) 75-79. J. LeBlanc, A. Nadeau, D. Richard and A. Tremblay, Studies on the sparing effect of exercise on insulin requirements in human subjects. Metabolism 3(I (1981) 1119-1124.
123
AGEING, EXERCISE AND FOOD RESTRICTION: EFFECTS ON SKELETAL MUSCLE GLUCOSE UPTAKE
J.L. IVY*, J.C. YOUNG**, B.W. CRAIG***, W.M. KOHRT AND J.O. HOLLOSZY Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110 (U.S.A.) (Received March 14th, 1991) (Revision received April 20th, 1991)
SUMMARY
We investigated the effects of ageing, exercise and food restriction on glucose uptake by muscles perfused with physiological concentrations of insulin and glucose in male Long-Evans rats. The rate of glucose uptake by hindlimb muscles perfused with medium containing 50 #U//ml insulin and 8 mM glucose was the same in 9-10 month-, 18 month-, and 24-month-old rats. Rats exercised by means of swimming 3 h/day, 5 days/week, had significantly higher rates of muscle glucose uptake than did the sedentary freely eating rats. Paired-weight sedentary rats, that were food restricted so as to keep their weights in the same range as those of the swimmers, had a hindlimb glucose uptake rate similar to that of the swimmers, and greater than that of the sedentary freely eating rats. The 24-month-old sedentary freely eating rats showed a trend toward a higher plasma glucose response and a lower plasma insulin response to a glucose tolerance test. The 24-month-old swimmers showed no deterioration in glucose tolerance compared to the 9-10 month-old rats. Our findings argue against the concept that ageing results in skeletal muscle insulin resistance in rats.
Key words." Ageing; Exercise; Food restriction; Glucose uptake; Muscle Correspondence to: John O. Holloszy, Department of Internal Medicine, Campus Box 8113, Washington University School of Medicine, 4566 Scott Avenue, St. Louis, MO 63110, U.S.A. *Present address: Department of Kinesiology and Health Education. University of Texas, Belmont Hall 222, Austin, TX 78712. **Present address: Department of Health Sciences, Boston University, 635 Commonwealth Avenue, Boston, MA 02215. Present address." Human Performance Laboratory, Ball State University, Muncie, IN 47306. 0047-6374/91/$03.50 Printed and Published in Ireland
© 1991 Elsevier Scientific Publishers Ireland Ltd.
124
INTRODUCTION Studies on human subjects using the euglycemic clamp procedure have provided evidence that ageing is associated with development of insulin resistance [1-4]. Skeletal muscle is the major site of insulin-mediated glucose disposal [5] and the slower rates of glucose disposal in older people during the euglycemic clamp is thought to be primarily due to development of skeletal muscle insulin resistance [1-4]. The results of studies on rats, the species most commonly used in studies of ageing, have also been interpreted by some investigators to indicate that ageing results in insulin resistance [6-8]. The role of ageing is not clearly established, however, because the studies that showed a decrease in insulin action involved comparisons of juvenile rats with mature [6] or old [7,8] rats. In this context, we have investigated the effect of ageing by comparing the rates of glucose uptake by skeletal muscles perfused with physiological concentrations of insulin and glucose in young adult (9-10 months), middle-aged (18 months), and old (24 months) freely eating sedentary male L o n g - E v a n s rats. Obesity is frequently associated with development of insulin resistance [9], while exercise can result in increased insulin sensitivity [10,11]. It, therefore, seems possible that weight gain and physical inactivity could contribute to insulin resistance in older individuals. We, therefore, also examined the interactions between the effects of ageing and regularly performed exercise or food restriction on glucose uptake by muscles perfused under the same conditions, and in the same age groups, as for the sedentary freely eating rats. MATERIALS AND METHODS Animal care and exercise program
Three groups of male specific-pathogen-flee L o n g - E v a n s rats were obtained from Charles River (Wilmington, MA) at age 3 months. Subgroups of these rats were necropsied shortly after arrival. Histopathologic examination, culture of the respiratory tract, tympanic bullae and gastrointestinal contents, and testing of serum for titers of antibodies against pathogenic viruses and mycoplasma confirmed that they were pathogen free. Additional rats from our aging rat colony were subjected to the same procedures at ages of - 12 months and - 2 4 months, and were also found to be pathogen free. Groups of rats were studied at the ages of 9-10 months, 18 months and 24 months. The 18- and 24- month-old rats were from the initial group of animals. The two additional groups of rats were obtained 12 and 18 months after the initial group so that young adult (9-10 month) rats could be studied at the same times as the 18- and 24-month-old animals. The rats were housed in individual stainless steel cages measuring 7 x 14 x 8 in, in a temperature and light controlled animal room with its own ventilation system, with 15 air exchanges per hour (100% intake and 100% exhaust) in a facility in which
125 no other rats were housed. To avoid introducing infections into the rat colony, the animal caretakers did not work with other rodents or in areas where they could be exposed to other rodents. The rats were fed a diet of Purina Rat Chow and water. All of the rats were fed ad libitum until age 5 months, when they were randomly assigned to either a sedentary freely eating group, a freely eating exercise group, or a sedentary paired-weight group that was food restricted to keep their body weights in the same range as those of the exercisers. The exercise program consisted of swimming 5 days/week, in groups of six, in water maintained between 33 and 35°C in stainless steel barrels filled to a depth of 60 cm. The duration of the exercise sessions was gradually increased over 4 weeks until the rats were swimming 3h/day.
Hindlimb perfusion After an overnight fast, and for the swimmers 45-52h after their last exercise session, the rats were anesthetized with pentobarbital sodium (5 mg/100 g body wt) injected intraperitoneally. The rationale for the overnight fast was to standardize as much as possible the milieu to which the skeletal muscles of the rats in the different groups were exposed in the period immediately preceeding the hindlimb perfusion study, especially with respect to insulin and glucose. Surgical preparation of the hindquarter was performed as described by Ruderman et ai. [12]. The perfusion procedure and apparatus have been described previously [13]. The perfusion medium consisted of Krebs-Henseleit bicarbonate (KHB) buffer containing 4 g/100 ml of dialyzed fatty acid poor bovine serum albumin (Cohn fraction V; Miles Laboratory), blood bank time-expired human erythrocytes, rejuvenated by the procedure of Valeri [ 14], at a concentration of 12 g hemoglobin/100 ml, 50/zU/ml of purified pork insulin (Squibb), and, when present, 8 mM glucose. The left iliac artery was ligated, limiting perfusate flow to the right hindlimb. The hindlimb preparation was initially washed out with 50 ml of KHB. A flow-through perfusion at 8 ml/min with medium containing 50 ttU/ml insulin, but no glucose was started. After 15 min of perfusion without glucose, perfusion with medium containing 8 mM glucose and 50/zU/ml of insulin was begun. The hindlimb was then perfused with this medium for 20 min, and glucose uptake was measured during the last 15 min during which the venous effluent was collected on ice, and arterial and venous samples were taken at 5-rain intervals. Rates of glucose uptake were calculated from the perfusate flow rate and the arteriovenous difference in glucose concentration. Glucose concentration was measured enzymatically [15]. After the perfusions, the muscles that were perfused were identified by their staining after injection of bromophenoi blue (Mallinckrodt Chemicals) into the arterial circulation; the stained muscles were dissected out and weighed. Measurement of glucose tolerance After an overnight fast, the rats were anesthetized with 5 mg/100 g body wt. of sodium pentobarbital intraperitoneally. Respiration was maintained with a small
126 animal respirator (Phipps and Bird, Richmond, VA). A catheter was placed in the left carotid artery, tied in place, and used for taking blood samples. A baseline blood sample (0.8 ml) was taken. The iliac vein was then exposed, and glucose, 150 rag/100 g body wt., was injected intravenously (60% glucose in water). Additional 0.8 ml blood samples were taken (15, 30, 60 and 90 min) after the glucose load. Alter each blood sample was taken, 0.8 ml of 0.9% saline was injected via the carotid catheter to maintain blood volume. The blood samples were placed in tubes containing 0.1 ml EDTA (24 mg/ml) and kept on ice. Plasma was separated by centrifugation at 4°C and stored at -20°C until analysis. Glucose concentrations were determined by a glucose oxidase method (YSI glucose analyzer; Yellow Springs, OH). Plasma insulin was determined by radioimmunoassay [16]. The rat insulin standard was kindly given to us by Eli Lilly. After completion of the glucose tolerance tests, the anesthetized rats were killed by exsanguination, and various organs and skeletal muscles were dissected out and stored at - 70°C for use in other studies.
Calculations and statistics The total areas under the plasma glucose and insulin curves obtained during the intravenous glucose tolerance test (IVGTT) were determined by computer analysis. The statistical significance of differences between groups was assessed by a two-way analysis of variance (ANOVA). When statistical significance was indicated by ANOVA, differences were isolated using the Neuman-Keul post hoc test. RESULTS
Body weights and hindlimb muscle weights The body weights of the swimmers and paired-weight sedentary rats were 20-30% lower than those of the freely eating sedentary rats (Table I). There were no significant differences in body weight between the 9-10 month-, 18 month- and 24-monthold rats in any of the treatment groups. Average food intake between 8 and 24 months was similar for the sedentary freely eating animals (26 ~ 3 g/day) and the swimmers (28 ± 3 g/day); the food intake of the sedentary paired-weight rats, which was restricted to keep their body weights in the same range as those of the swimmers, averaged 22 -4- 2 g/day. The average weight of the perfused hindlimb muscles was significantly greater in the sedentary freely eating animals than in the swimmers or sedentary paired-weight rats (Table I). The weight of the hindlimb muscles decreased with age, being approximately 10% lower in the 24-month-old than in the 9-10-month-old rats.
Glucose uptake by perfused hindlimbs The effect of age on muscle permeability to glucose in the presence of an insulin concentration in the physiological range found in the fed state was evaluated in rat hindlimbs perfused with 8 mM glucose and 50 ~U/ml insulin. As shown in Table II,
127
TABLE 1 BODY WEIGHTS A N D PERFUSED H I N D L I M B MUSCLE WEIGHTS
Age (months)
Group Sedentary Freely eating
Swimmers Paired weight
Body weight(g) 9-10 18 24
656 4. 698 a (8) 689 4. 19 a (5) 701 ± 52 a (5)
522 ± 43 (7) 518 4- 54 (5) 493 4- 46 (5)
508 4- 46 (10) 552 4. 44 (4) 494 4. 43 (5)
23.8 4. 2.7 (7) 22.0 4. 3.4 (5) 21.6 4. 2.4 (5)
25.6 4. 1.9 (10) 23.5 4. i.8 (4) 23.2 4. 1.0 c (5)
Weight of perfused hindlimb muscles (g) 9-10 18 24
Values are means ap < 0.01 versus bp < 0.02 versus cp < 0.05 versus
30.1 4- 2.0 b (8) 28.0 4- 1.5 b (5) 26.9 4. 2.5 b'c (5) 4- S.D. for the number sedentary paired-weight sedentary paired-weight 9-month-old rats in the
of rats given in parentheses. and swimmer groups of the same age. and swimmer groups of the same age. same group.
when glucose uptake is expressed per hindlimb, there is a slight downward trend for glucose uptake with ageing, that did not attain statistical significance in any of the three treatment groups. Even this trend disappears when glucose uptake is expressed per gram of muscle to eliminate the effect of the - 10% decrease in the mass of the muscles with ageing (Table III). The rate of glucose uptake was lower in the hindlimb muscles of the sedentary freely eating rats than in those of the swimmers and the sedentary paired-weight rats.
TABLE I! GLUCOSE UPTAKE (p,mol " hindlimb -l • min -I) BY PERFUSED HINDLIMB
Age (months)
Group Sedentary
9-10 18 24
Swimmers
Freely eating
Paired-weight
6.94 4. 1.38 (8) 6.78 4. 1.04 (5) 6.54 4. 2.10(5)
8.57 4. 2.01 (7) 8.20 4. 1.85 (5) 7.81 ± 3.04(5)
Values are means ± S.D. for the number of animals given in parentheses. ap < 0.05 versus sedentary freely eating.
9.92 4. 3.88 a (10) 8.32 4. 2.60 (4) 8.66 4. 3.11 (5)
128 TABLE I11 GLUCOSE UPTAKE (t~mol"g-l. 15min-I) PER GRAM OF PERFUSED MUSCLE
Age (months)
Group Sedentary
9-10 18 24
Swimmers
Freely eating
Paired-weight
3.40 ± 0.89 (8) 3.65 ± 0.71 (5) 3.67 ± 1.22 (5)
5.40 + 1.23 (7)a 5.71 ± 1.62 (5)a 5.30 ± 1.68 (5)
5.74 ± 2.02b (10) 5.28 ± 1.55 (4) 5.62 + 2.09 (5)
Values are means ± S.D. for the number of animals given in parentheses. ap < 0.05 versus sedentary freely eating. bp < 0.01 versus sedentary freely eating.
When rates of glucose uptake expressed per hindlimb are compared in the age subgroups of the three treatment groups, only the 9-10-month-old swimmers are significantly different from the sedentary freely eating rats of the same age (Table II). When glucose uptake is expressed per gram of muscle, only the 9-10-month-old and 18-month-old paired-weight rats, and the 9-10-month-old swimmers, are significantly different from the sedentary freely eating rats (Table III). However, when the data are analyzed by treatment group without regard to age, i.e. when the three age subgroups are pooled, the rates of glucose uptake are significantly lower in the freely eating sedentary rats than in either the paired-weight sedentary rats (per hindlimb P < 0.05; and per gram muscle P < 0.01), or swimmers (P < 0.01, both per hindlimb and per gram muscle). There were no significant differences in hindlimb muscle glucose uptake between the swimmers and sedentary paired-weight rats.
Intravenous glucose tolerance tests ( I V G T T ) Plasma glucose response. Insulin resistance could be due to humoral or other systemic factors that are no longer present in washed out, perfused skeletal muscles. We therefore, performed IVGTTs on additional 9-10-month-old and 24-month-old swimmers and freely eating sedentary controls to evaluate the possibility that insulin resistance might be present in vivo in old rats. The weights of the rats used for the IVGTTs are shown in Table IV. As shown in Fig. 1, the plasma glucose responses to the I V G T T were similar in the young swimmers and the young sedentary rats; the areas under the glucose tolerance curves were also nearly identical (Table V). There was essentially no difference in the plasma glucose response during the IVGTT between the young and
129
T A B L E IV BODY W E I G H T S O F RATS USED F O R M E A S U R E M E N T O F G L U C O S E T O L E R A N C E
Age (months)
Sedentary freely eating
Swimmers
9-10 24
630 .4- 60 a (11) 618 .4- 62 b (8)
576 .4- 36 (10) 496 .4- 51 (8)
Values are means 4. S.D. for the number of rats given in parentheses. ap < 0.05 versus swimmers of same age. bp < 0.01 versus swimmers of same age.
FREELY FATING GROUP
SWIMMING GROUP 20
2O
18
18
10
10
14 I&J
.J 0
12
12
lo
10
a
e !
6 4
- ~ o MONTH --i- 24 MONTH
2 0
I
I
I
I
I
I
4
2 I
0
0 16 30 46 6O 76 00
I
r
TIME (,.in)
J
I
r
I
TiE (,.in) S'6 $3 $'6 2.7
3.6 $'6 S.O 2.7
I
I
0 16 30 46 O0 76 90
2.4 |.1
2.4
:L1
1.8
1.6 1.2
1.6
0"6
OJ O.I 0.1
1.2
OA 0.3
0.0
r
~
~ r
i
i
J
0 16 30 46 60 76 gO
TiME (mk)
0.0
l l l l
I I
O m ~ U ~ M
T~
(.:in)
Fig. 1. Plasma glucose and insulin responses to an intravenous glucose tolerance test. After an overnight fast, rats were anesthetized and given glucose 150 mg/100 g body wt intravenously. Points are means .4S.E.M. for 8-11 rats.
130 TABLE V AREAS UNDER THE IVGTT GLUCOSE AND INSULIN RESPONSE CURVES Age (months)
Sedentary freely eating
Swimmers
Glucose areas(mM 90 min "t)
9-10 24
751 ± 221 (11) 905 ± 373 (8)
769 ± 222 (10) 778 ~ 156 (8)
Insulin areas(ng" m1-1" 90 rain "l)
9-10 24
201 + 68 (11) 152 + 75b (8)
117 ± 65a (10) 74 ± 32~ (8)
Values are means + S.D. for the numbers of rats given in parentheses. ap < 0.05 versus sedentary rats in the same age group. bp < 0.10 versus 9-10 month-old sedentary group.
old swimmers (Fig. 1, Table V). In the freely eating sedentary animals the average plasma glucose concentrations were -20°/,, higher in the 24-month-old rats than in the 9 - 1 0 - m o n t h - o l d rats during the I V G T T . However, because o f the variability in response, neither the differences between the glucose concentrations at the individual time points nor the areas under the glucose curves, attained statistical significance. Insulin responses. There was a tendency for the plasma insulin levels to be lower in the 24-month-old than in the 9 - 1 0 - m o n t h - o l d rats (Fig. 1; Table V). A N O V A showed significant effects o f age on plasma insulin concentration at the 30 min and 60 min time points, and for the areas under the insulin curves (P < 0.05). However, the differences between the y o u n g and old rats within each treatment g r o u p did not attain statistical significance, except for the values at the 30-min time point in the sedentary freely eating group (P < 0.05). The swimmers had lower insulin responses than the sedentary freely eating rats, as reflected in significantly smaller areas under the insulin curves, both at 9 - 1 0 - m o n t h s - o l d and 24-months-old (Table V). Plasma insulin concentrations at all o f the time points except 90-min were significantly lower in the swimmers than in the sedentary freely eating rats both at 9 - 1 0 - m o n t h s and at 24-months (Fig. 1). DISCUSSION Glucose tolerance deteriorates in a high proportion of men and women as they grow older [17]. This decline in glucose tolerance appears to be due in large part to development o f insulin resistance o f insulin sensitive tissues [1-4]. Some investigators have interpreted the results o f studies on rats to indicate that ageing is also associated with development o f insulin resistance in this species [6-8]. The present results suggest that glucose tolerance does tend to deteriorate with ageing in free-
131 ly eating sedentary male rats (Fig. 1, upper panel). However, this trend toward higher glucose levels in the old rats, which was not seen in rats that exercised regularly, appears to have been due to a reduced insulin response to the glucose challenge [Fig. 1, lower panel], rather than to insulin resistance. Our finding of a smaller increase in plasma insulin concentration in response to the IVGTT in the 24-month-, compared to the 9-10-month-old rats, is in keeping with the finding of Elahi et al. [18] that isolated pancreas preparations from 23-month-old rats secreted less insulin than those of 12-month-old rats when perfused with 8 mM glucose. The results of the glucose tolerance tests fits well with our finding that the rate of glucose uptake in the presence of a physiological insulin concentration was not decreased in the perfused hindlimb muscles of the 18-month- or 24-month-old rats as compared to the 9-month-old rats. Skeletal muscle is the major site of insulinmediated glucose transport [5]. Our findings are, therefore, at odds with the concept that ageing results in development of insulin resistance in the rat. However, examination of the studies cited as showing development of insulin resistance with ageing in rats shows that what is actually a decrease in insulin action during growth and development has been misinterpreted as being due to ageing. In some of these studies 6-8-week-old rats were compared to 12-month-old rats [6,19]. In other studies, in which a wider range of ages was compared, it was found that insulin action decreased during growth and development, with no further deterioration attributable to ageing [7,20]. Goodman et al. [20] found, for example, that resistance to the action of insulin of perfused hindlimb muscles increased markedly in rats between 3 weeks and 24 weeks of age, with no further significant change between 24 and 96 weeks. Although the authors clearly stated that the decrease in insulin action occurred during early development [20], others have cited this study as evidence for development of insulin resistance due to ageing. Nishimura et al. [7], using the euglycemic clamp technique, also found that most of the decline in insulin action occurs between 2 months and 4 months of age in rats, and that there is no significant decline in the rate of insulinmediated glucose disposal between 10 months and 20 months; nevertheless, these authors refer to this decrease in insulin action as being due to ageing [7, 8]. Our finding that glucose uptake by hindlimb muscles perfused with a physiological concentration of insulin is the same in 9-10-month-, 18-month- and 24-month-old rats is in good agreement with the data of Goodman et al. [20] and Nishimura et ai. [7]. Another finding that has been cited as evidence that ageing results in insulin resistance in rats is that the adipocytes of older rats are markedly resistant to the action of insulin compared to those of juvenile animals [19]. However, this is also not an ageing phenomenon, but a preventable consequence of obesity and fat cell hypertrophy [21 ]. Exercise-training results in an increase in insulin sensitivity in humans [11] and rats [10]. An interesting aspect of this relatively short-lived adaptation is a blunted plasma insulin response to a glucose stimulus as the result of decreased insulin secre-
132 tion by the pancreas [22-24]. The increase in insulin sensitivity and the decrease in insulin secretion are closely matched so that normoglycemia is maintained despite the increased insulin sensitivity [25]. In keeping with these findings, the swimmers had a normal glucose tolerance despite a significantly blunted plasma insulin response to a glucose load. Furthermore, the swimmers' hindlimb muscles had a significantly greater rate o f glucose uptake when exposed to a submaximally effective insulin stimulus than did the muscles o f the sedentary freely eating rats. It is interesting that food restriction sufficient to keep the b o d y weights o f sedentary rats in the same range as those o f the swimmers was as effective as the swimming program in improving hindlimb muscle glucose uptake. In conclusion, glucose uptake by hindlimb muscles perfused with 8 m M glucose and 50/~U/ml insulin was the same in 9 - 1 0 - m o n t h , 18-month, and 24-month-old rats. This finding argues against the concept that ageing causes skeletal muscle insulin resistance in rats. Rats that were exercised regularly by swimming, and pairedweight sedentary rats that were food restricted to keep their b o d y weights in the same range as the swimmers', had higher rates o f glucose uptake by hindlimb muscles than did the sedentary freely eating rats. The swimmers also had normal glucose tolerance despite a blunted insulin response to a glucose tolerance test, providing further evidence for an increased susceptibility to the action o f insulin. ACKNOWLEDGEMENTS This research was supported by National Institutes o f Health Research G r a n t AG00425 from the National Institute on Ageing. J.L. Ivy, J.C. Y o u n g and B.W. Craig were the recipients o f N.I.H. National Research Service Awards AM06159, AG05216 and AM06123, respectively. REFERENCES 1 R.A. DeFronzo, Glucose intolerance and aging. Evidence for tissue insensitivity to insulin. Diabetes. 28(1979) 1095-1101. 2 R.I. Fink, O.G. Kolterman, J. Griffin and J.M. Olefsky, Mechanisms of insulin resistance in aging. J~ Clin. Invest. 71 (1983) 1523-1535. 3 J.W. Rowe, K.L. Minaker, J.A. Pallotta and J.S. Flier, Characterization of the insulin resistance of aging. J. Clin. Invest. 71 (1983) 1581-1587. 4 M. Chen, R.N. Bergman, G. Pacini and D. Porte, Jr., Pathogenesis of age-related glucose intolerance in man: insulin resistance and decreased A-cell function, Z Clin. EndocrinoL Metab. 60 (1985)13. 5 R.A. DeFronzo, E. Jacot, E. Jequier, E. Maeder, J. Wahren and J.P. Felber, The effect of insulin on the disposal of intravenous glucose. Diabetes 30 (1981) 1000-1007. 6 E. Reaven, D. Wright, C.E. Mondon, R. Solomon, H. Ho and G.M. Reaven, Effect of age and diet on insulin secretion and insulin action in the rat. Diabetes 32 (1983) 175-180. 7 H. Nishimura, H. Kuzuya, M. Okamoto, Y. Yoshimasa, K. Yamada, T. Ida, T. Kakehi and H. Imura, Change of insulin action with aging in conscious rats determined by euglycemicclamp. Am. J. Physiol. 254 (1988) E92-E98. 8 S. Kono, H. Kuzuya, M. Okamoto, H. Nishimura, A. Kosaki, T. Kakehi, M. Okamoto, G. Inoue,
133
9
10 11
12 13 14
15
16 17
18
19 20
21 22
23 24 25
I. Maeda and H. lmura, Changes in insulin receptor kinase with aging in rat skeletal muscle and liver. Am. J. Physiol. 259 (1990) E27-E35. B. Jeanrenaud, Hyperinsulinemia in obesity syndromes: its metabolic consequences and possible etiology. Metabolism 27 (1978) 1881-1892. D.E. James, E.W. Kraegen and D.J. Chisholm, Effects of exercise training on in vivo insulin action in individual tissues of the rat. J. Clin. Invest. 76 (1985) 657-666. D.S. King, G.P. Dalsky, W.E. Clutter, D.A. Young, M.A. Staten, P.E. Cryer and J.O. Holloszy, Effects of exercise and lack of exercise on insulin sensitivity and responsiveness. J. Appl. Physiol. 64 (1988) 1942-1946. N.B. Ruderman, C.R.S. Houghton and R. Hems, Evaluation of the isolated perfused rat hindquarter for the study of muscle metabolism. Biochem. J. 124 (1971) 639-651. M.J. Rennie and J.O. Holloszy, Inhibition of glucose uptake and glycogenolysis by availability of oleate in well-oxygenated perfused skeletal muscle. Bioehem. J. 168 (1977) 161-170. C.R. Valeri, Salvaging and rejuvenation of outdated red cells and cells with elevated 2,3-DPG levels and decreased affinity for oxygen. In D.M. Surgenor (ed.), The Red Blood Cell, 2nd edn., Vol. I, Academic Press, New York, 1974, pp. 528-532. H.U. Bergmeyer, E. Bernt, F. Schmidt, and H. Stork, D-glucose. Determination with hexokinase and glucose-6-phosphate dehydrogenase. I n H.U. Bergmeyer (ed.), Methods q f En=ymatic Analysis, 2nd edn., Academic Press, New York, 1974, pp. 1196-1201. V. Herbert, K.S. Lau, C.W. Gottlieb and S.J. Bleicher, Coated charcoal immunoassay of insulin. J. Clin. Endocrinol. 25 (1965) 1375-1384. M.B. Davidson, The effect of aging on carbohydrate metabolism: a review of the English literature and a practical approach to the diagnosis of diabetes mellitus in the elderly. Metabolism 28 (1979) 686-705. D. Elahi, D.C. Muller, D.K. Anderson, J.D. Tobin and R. Andres, The effect ofage and glucose concentration on insulin secretion by the isolated perfused rat pancreas. Endocrinology 116 (1985) 11-16. S. Narimiya, C.B. Dolkas, C.E. Mondon, C. Sims, D.W. Wright and G.M. Reaven, Insulin resistance in older rats. Am. J. Physiol. 246 (1984) E397-E404. M.N. Goodman, S.M. Dluz, M.A. McElaney, E. Belur and N.B. Ruderman, Glucose uptake and insulin sensitivity in rat muscle: changes during 3-96 weeks of age. Am. J. Physiol. 244 (1983) E93-EI00. B.W. Craig, S.M. Garthwaite and J.O. Holloszy, Adipocyte insulin resistance: effects of aging, obesity, exercise, and food restriction. J. Appl. Physiol. 62 (1987) 95-100. D.S. King, G.P. Dalsky, M.A. Staten, W.E. Clutter, D.R. Van Houten and J.O. Holloszy, Insulin action and secretion in endurance-trained and untrained humans. J. Appl. Physiol. 63 (1987) 2247-2252. D.S. King, M.A. Staten, W.M. Kohrt, G.P. Dalsky, D. Elahi and J.O. Holloszy, Insulin secretory capacity in endurance-trained and untrained young men. Am. J. PhysioL 259 (1990) EI55-EI81. H, Galbo, C.J. Hedeskov, K. Capito and J. Vinten, The effect of physical training on insulin secretion of rat pancreatic islets. Acta Physiol. Seand. 11 (1981) 75-79. J. LeBlanc, A. Nadeau, D. Richard and A. Tremblay, Studies on the sparing effect of exercise on insulin requirements in human subjects. Metabolism 3(I (1981) 1119-1124.
