Amylin (also known as islet amyloid polypeptide) is a 37-amino acid glucoregulatory peptide that is costored with insulin in β-cell secretory granules and co-secreted with insulin from pancreatic β-cells (22,24). The expression of amylin mRNA and protein release is stimulated by glucose (22); moreover, amylin works in concert with insulin to regulate blood glucose concentration by reducing gastric emptying and inhibiting glucagon secretion, thereby reducing hepatic glycogenolysis (12). Amylin analogs are used to treat patients with type 1 and 2 diabetes (7,19,24) and are effective in reducing postprandial glucose excursions. Recent evidence also suggests that amylin regulates energy homeostasis by increasing energy expenditure and reducing appetite (18,27), having a synergistic effect with leptin (23,28). Moreover, leptin concentrations have been shown to play a role in glycemic control and are affected by insulin levels (1). Collectively, these data indicate that amylin plays a significant role in regulating glucose homeostasis and energy balance, both of which can be altered during exercise.
Requirements for sustaining long-term (>1 h) moderate aerobic exercise include the ability to use substantial amounts of carbohydrate. This requires effective blood glucose regulation over time to replace intramuscular glycogen stores used during muscle contraction. Only two studies, both from our laboratory, have investigated the effects of exercise on amylin and other glucoregulatory hormones. In the first study, we documented the effects of treadmill exercise intensity on amylin and other glucoregulatory hormones and reported that running at 90% and 100% of V˙O2max elicited similar increases in both amylin and insulin as blood glucose rose (12). Lower exercise intensities (60% and 75% V˙O2max) did not alter these hormones. In a subsequent study, resistance exercise, which included only concentric or only eccentric muscle actions, elevated insulin without significantly increasing amylin concentrations (17). There are no data on the effects of long-term steady state exercise on amylin. Given that 1) amylin plays an important role in maintaining blood glucose levels postprandially (12) and 2) amylin analog administration reduces glucose excursions in patients with type 1 and 2 diabetes (19,24), and 3) because patients with type 2 diabetes are treated with physical exercise, it is important to determine whether prolonged exercise affects amylin in healthy subjects to properly treat patients with glucoregulatory deficiencies. Thus, the purpose of our study was to determine the effects of 90 min of steady-state exercise on amylin, insulin, C-peptide, glucagon, cortisol, leptin, and blood glucose responses. It has been demonstrated that extended fasting reduces insulin that, in turn, affects leptin levels (26). Because we were interested in the effect of exercise and not fasting on these hormones, we examined glucoregulatory responses under realistic conditions in which subjects were in a postprandial state before completing a long-term exercise bout. We hypothesized that compared with a resting control trial, an exercise trial would result in suppression of amylin, insulin, and C-peptide concentrations with no change in leptin and increases in glucagon and cortisol concentrations.
Eight young (mean ± SD, 22.6 ± 1.5 yr) healthy male subjects were recruited from the university community and gave informed consent to participate in the study. On the basis of a medical history and a 3-d food record, subjects met the following criteria: 1) between the ages of 18 and 35 yr, 2) not taking any prescription medications, 3) no history of cardiovascular or metabolic disease, and 4) no adherence to a diet that would affect metabolic responses to exercise. All subjects were physically active and completed regular weekly (minimum of three to four times per week, 30-60 min per session) aerobic exercise regimens. The study was approved by the Southeastern Louisiana University Institutional Review Board. Subjects completed a preliminary trial followed by an experimental (exercise) trial and a control (resting) trial in counterbalanced manner with 1 month between trials. Descriptive data are shown in Table 1.
Subjects reported to the laboratory and body composition (via seven-site skinfold assessment (30)) and maximal oxygen consumption (V˙O2max) were determined. Subjects completed a graded treadmill exercise test that began at 2.5 miles·h−1 and 4% grade and progressed by 1 mile·h−1 every 2 min until exhaustion (which we have named the Kraemer protocol). Ventilatory volumes, FEO2, and FECO2 were measured, and V˙O2 was calculated using a metabolic cart (ParvoMedics 2400, Sandy, UT). V˙O2max was reached when either of the following was met: the primary criterion of a plateau in V˙O2 with an increase in workload or two of three secondary criteria: 1) reaching predicted maximal HR, 2) RER >1.1, or 3) RPE (15-point Borg scale) of 19 or 20.
Exercise and control trials.
For the exercise and control trials, subjects were instructed to: 1) maintain their normal diet, 2) refrain from exercising for 48 h before sessions, 3) refrain from drinking alcohol for 48 h before sessions, 4) fast from at least 12 midnight the night before, and 5) report to the laboratory at 8:00 a.m. after consuming a liquid meal (Ensure Plus™) at 7:00 a.m. Ingestion of a pretrial liquid meal allowed determination of glucoregulatory responses to exercise under realistic conditions without effect of prolonged fasting contributing to the metabolic and endocrine responses. The macronutrient content percentages from the 350 kcal in the Ensure Plus™ liquid meal consumed by the subjects before the exercise and control trials was 15% protein, 28% fat, and 57% carbohydrate. We calculated that the preexercise time period would be associated with a normal daily insulin/blood glucose level and that the 350-kcal liquid meal would allow the subjects to complete the prolonged exercise in which they would expend an estimated 700-900 kcal. An intravenous catheter was inserted into a dorsal vein of the hand at 8:30 a.m., and a saline lock was used to maintain patency. Two resting blood samples (28 mL per sample) were collected at 9:00 a.m. (−30 min) and 9:30 a.m. (0 min), at which time subjects began exercising on the treadmill. V˙O2 was continually monitored during the 90 min of exercise, and treadmill speed and grade were adjusted to maintain a V˙O2 of 60% of V˙O2max. Blood samples were also collected during (18, 36, 54, 72, and 90 min) and after exercise (20, 40, and 60 min). For the control trial, the blood sampling protocol was identical with the exercise trial; however, the subjects rested in a seated position with no exercise or measurement of V˙O2. Blood samples for glucagon analysis were collected in chilled EDTA tubes with protease inhibitor and samples for insulin, C-peptide, amylin, and leptin were collected in the same manner without protease inhibitor. Blood samples for glucose analysis were collected in a sodium oxalate/potassium fluoride tube. Owing to the assay error, plasma samples for glucagon were expended for one subject, and only seven of eight complete sample sets were analyzed for glucagon.
Plasma concentrations of amylin, insulin, C-peptide, and leptin were determined by ELISA (Millipore Corp., St. Charles, MO). Cortisol and glucagon concentrations were determined by Immulite (Siemens, Los Angeles, CA) and radioimmunoassay (Millipore), respectively. The interassay coefficients of variation for insulin, C-peptide, amylin, leptin, glucagon, and glucose were 13.5%, 4.69%, 14.4%, 8.9%, 12.7%, and 1.9%, respectively. The intra-assay coefficients of variation for insulin, C-peptide, amylin, leptin, glucagon, and glucose were 3.1%, 2.2%, 7.2%, 4.3%, 4.0%, and 0.6%, respectively. Glucose concentrations were determined by an enzymatic spectrophotometric analyses (Pointe Scientific, Canton, MI). Hematocrit was determined using the microhematocrit method, and hemoglobin concentrations from the whole blood were determined using an enzymatic colorimetric method (Pointe Scientific). Hemoglobin and hematocrit were used to determine plasma volume shifts across time (6).
Statistical analyses were performed at the alpha level of 0.05 using SPSS PASW Statistics 18 (IBM Corp., Somers, NY). Two different statistical approaches were used. First, a 2 × 10 ANOVA was used to examine hormone changes over time and between trials. Post hoc dependent t-tests were applied where appropriate. Second, to determine the total response of the hormones and metabolites to exercise, integrated area-under-the-curves (AUC) were computed using a trapezoidal method after subtracting averaged baseline hormone concentrations for each subject. Dependent t-tests were used to determine differences between experimental and control trials. Although power was sufficient to detect main effect of time (0.77 for amylin), it was lower for the trial and interaction effects. However, this lower power for trial and interaction was offset by the control of individual variability through multiple time measures from the same subject. Thus, the within-subjects design over multiple time points in a counterbalanced manner greatly reduced the variance of the endocrine and glucose measures.
Plasma volume shifts did not exceed −7.5% between the second resting blood sample and all subsequent exercise samples during exercise. There was no time, trial, or time × trial effect for glucose (Fig. 1), and glucose AUC values were not different between exercise and control trials. Glucose concentrations remained stable across the exercise and control trials, and although there was a trend for glucose to be higher in the exercise trial, the differences were not significant between exercise and control trials (Fig. 2). For amylin, insulin, and C-peptide, there was a significant time effect (F 9,126 = 7.64, P = 0.01; F 9,126 = 5.29, P < 0.001; F 9,126 = 3.52, P < 0.01, respectively) but no significant time × trial interaction. Amylin, insulin, and C-peptide concentrations declined in a similar manner during exercise and control trials and remained suppressed during recovery (Fig. 3). There was no significant time effect for glucagon (F 9,108 = 2.09, P = 0.12), but there was a significant time × trial interaction (F 9,108 = 4.08, P = 0.01) for glucagon. There was also a significant time effect for cortisol (F 9,126 = 3.39, P = 0.046) but not a significant time × trial interaction (F 9,126 = 1.78, P = 0.19). However, there was a significant trial effect for cortisol (F 1,14 = 10.85, P = 0.005), indicating greater cortisol concentration across time in the exercise trial. Glucagon AUC was significantly greater (P = 0.001) in the exercise trial (1057.64 ± 322.63 pg·min−1·mL−1) than the control trial (−862.79 ± 460.98 pg·min−1·mL−1), indicating a higher total glucagon response to the exercise trial, whereas there were no differences in AUC for amylin, insulin, and C-peptide during the two trials.
This is the first study to determine the effects of prolonged steady-state exercise on amylin and other glucoregulatory hormone responses. In this study, amylin concentrations declined during the 90-min period in both the exercise and control trials and glucagon rose during the exercise trial, which likely contributed to sustained blood glucose concentration during the exercise session. Insulin and C-peptide declined in a similar manner during the exercise and control trials. Data suggest that, although amylin is suppressed during long-term steady-state exercise, the suppression is not significantly greater than a smaller decline in a nonexercise condition. A previous study has shown that amylin responds to intense running in a similar manner to that of insulin (12). The present study suggests that the pattern of amylin and insulin secretion is similar in a 2.5-h postprandial state during long-term steady-state exercise and resting conditions. It also indicates that glucagon and cortisol are important in the regulation of blood glucose levels during prolonged exercise in a postprandial state.
After being cleaved from proinsulin, C-peptide is stored in β-cell secretory granules and subsequently released in equimolar quantities with those of insulin (29). Unlike insulin, there is negligible hepatic extraction of C-peptide, and thus, this peptide is thought to be a better indicator of insulin secretion (20). Thus, we chose to measure both C-peptide and insulin in the present study. Insulin and C-peptide responded in a similar manner, declining in both the exercise and the control trials, and confirming a reduction in insulin release in both trials. There was a similar amylin response to that of insulin and C-peptide during prolonged, steady-state exercise, indicating reduced secretion as well. These endocrine responses, coupled with sustained blood glucose concentration, suggest that steady blood glucose concentration during moderate-intensity, long-term exercise, plays a role in preventing increases in β-cell peptide secretion of amylin and insulin. This prolonged exercise response of blood glucose is different from previously reported increases in response to higher exercise intensities (90% and 100% V˙O2max) that resulted in both circulating insulin and amylin spikes during and after exercise (12). Moreover, the lower insulin concentration in the present study would reduce insulin-induced inhibition of lipolysis in subcutaneous adipocytes (11), which is consistent with the steady decline in RER that was observed across 90 min of exercise in the present study (data not shown).
Glucagon concentration was elevated during the exercise versus the control trial. Glucagon is important in stimulating hepatic glycogenolysis and thus providing the replacement of the blood glucose being used progressively to a greater extent over the exercise duration (5). This suggests the importance of glucagon in sustaining long-term exercise. Although we did not measure catecholamine concentrations in this study, from previous reports it has been shown that steady-state prolonged exercise stimulates increases in epinephrine and norepinephrine (10,21). In the resting control trial, it can be assumed that catecholamine concentration did not increase across time (2). Glucagon secretion is stimulated by catecholamines (9) and suppressed by amylin. Thus, under conditions of exercise but not resting conditions, stimulation of glucagon secretion by catecholamines could be opposed to a greater degree without decline in circulating amylin concentration. Therefore, it is possible that gradual reduction in amylin concentration may be especially vital during exercise, providing a physiological environment that allows greater increases in glucagon secretion (which occurred during the latter portion of the exercise trial) and thus allows hepatic glycogenolysis to sustain blood glucose concentration to counter enhanced glucose uptake by contracting skeletal muscle during exercise. The elevation of blood lactate during the prolonged exercise (data not shown) also provided an additional source of carbohydrate to sustain blood glucose concentrations via intercellular (e.g., Cori cycle) and intracellular lactate shuttles (3).
During the control trial, there was a pattern of cortisol decline that is typical of well-known diurnal reduction in cortisol during the morning hours. However, during the exercise trial, cortisol decline was attenuated during the first 54 min and then began to rise, peaking after 20 min of recovery. Elevated cortisol concentration could facilitate maintenance of blood glucose concentration because one of its effects is to block tissue uptake of glucose (12).
Evidence suggests that leptin is important in glycemic regulation (1) and that leptin infusion suppresses insulin (25). Data from the present study revealed a decline in amylin during long-term exercise and control trial in a postprandial state condition, but leptin concentration was sustained across time. We previously reported a decline in leptin in response to acute exercise in postmenopausal women that was associated with normal diurnal reductions (16). These responses are different than increases that may occur in some subjects in response to high-intensity exercise (13-15). The lack of decline in leptin in the present study may be due to the 2.5-h postprandial state of the subject before 90 min of sustained exercise. Sustained leptin concentrations may have had a positive effect on blood glucoregulation during exercise, but further investigation is required to determine this.
There are several possible reasons that would explain why amylin as well as insulin and C-peptide concentrations did not change differently across the exercise and control trials. First, training has been shown to attenuate reduction in insulin concentration and increase in glucagon in response to exercise, which has been attributed to training adaptations of catecholamines (8). It is possible that the trained state of the individuals attenuated the insulin, C-peptide, and amylin responses to the prolonged exercise and thus contributed to the lack of a difference in these hormone responses during the control and exercise trials. Moreover, it has been shown that exercise in a 2-h postprandial condition elicited greater reductions in insulin after subjects consumed a high-glycemic index meal than a low-glycemic index meal (4). Insulin responses to exercise after the high- and low-glycemic index meals in that study were greater than insulin responses to no exercise (control) after a low-energy/low-glycemic jelly ingestion. In the present study, the same caloric content meal was consumed in the exercise and control trials, which could explain the discrepancies between insulin responses in the present study and previous study. Moreover, the liquid meal consumed in the exercise and control trials in the present was associated with similar blood glucose concentration over time, and blood glucose plays a major role in insulin and amylin secretion.
In summary, this is the first study to determine the effect of long-term (>1 h) exercise on amylin and other glucoregulatory hormones. Our results indicate that, in healthy young men in a postprandial state, insulin, C-peptide, and amylin concentrations decline in the same manner at rest as during prolonged, moderate-intensity exercise; however, glucose concentrations are maintained, while glucagon and cortisol increase after the first hour of exercise. These glucoregulatory endocrine adjustments may enable blood glucose concentration to be sustained during long-term exercise. Future studies should investigate the effects of acute exercise on these glucoregulatory hormones in healthy subjects and diabetic patients in varying stages of alimentation.
This work was supported by faculty development grants (grants 1385 (2007-2008) and 60053 (2008-2009)) from the Center for Faculty Excellence, Southeastern Louisiana University, and by a Conduits Program Grant (grant 52301 (2008-2009)) from the Office of Research and Graduate Studies, Southeastern Louisiana University.
The authors thank the subjects for their participation in the study. The authors also thank Tahir Khan for his assistance in the laboratory.
There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Aas AM, Hanssen KF, Berg JP, Thorsby PM, Birkeland KI. Insulin-stimulated increase in serum leptin
levels precedes and correlates with weight gain during insulin therapy in type 2 diabetes. J Clin Endocrinol Metab
2. Acevedo EO, Kraemer RR, Kamimori GH, Durand RJ, Johnson LG, Castracane VD. Stress hormones, effort sense, and perceptions of stress during incremental exercise: an exploratory investigation. J Strength Cond Res
3. Brooks GA. Cell-cell and intracellular lactate shuttles. J Physiol
4. Burke LM, Claassen A, Hawley JA, Noakes TD. Carbohydrate intake during prolonged cycling minimizes effect of glycemic index of preexercise meal. J Appl Physiol
5. Coggan AR. Plasma glucose
metabolism during exercise in humans. Sports Med
6. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol
7. Edelman SV, Caballero L. Amylin replacement therapy in patients with type 1 diabetes. Diabetes Educ
. 2006;32(3 suppl):119-27S.
8. Gyntelberg F, Rennie MJ, Hickson RC, Holloszy JO. Effect of training on the response of plasma glucagon to exercise. J Appl Physiol
9. Harvey WD, Faloona GR, Unger RH. The effect of adrenergic blockade on exercise-induced hyperglucagonemia. Endocrinology
10. Howley ET, Cox RH, Welch HG, Adams RP. Effect of hyperoxia on metabolic and catecholamine responses to prolonged exercise. J Appl Physiol
11. Koppo K, Larrouy D, Marques MA, et al. Lipid mobilization in subcutaneous adipose tissue during exercise in lean and obese humans. The roles of insulin and natriuretic peptides. Am J Physiol Endocrinol Metab
12. Kraemer RR, Acevedo EO, Synovitz LB, et al. Glucoregulatory endocrine responses to intermittent exercise of different intensities: plasma changes in a pancreatic beta-cell peptide, amylin. Metabolism
13. Kraemer RR, Acevedo EO, Synovitz LB, Hebert EP, Gimpel T, Castracane VD. Leptin
and steroid hormone responses to exercise in adolescent female runners over a 7-week season. Eur J Appl Physiol
14. Kraemer RR, Chu H, Castracane VD. Leptin
and exercise. Exp Biol Med (Maywood)
15. Kraemer RR, Durand RJ, Acevedo EO, et al. Effects of high-intensity exercise on leptin
and testosterone concentrations in well-trained males. Endocrine
16. Kraemer RR, Kraemer GR, Acevedo EO, et al. Effects of aerobic exercise on serum leptin
levels in obese women. Eur J Appl Physiol Occup Physiol
17. Kraemer RR, Durand RJ, Hollander DB, Tryniecki JL, Hebert EP, Castracane VD. Ghrelin and other glucoregulatory hormone responses to eccentric and concentric muscle contractions. Endocrine
18. Lutz TA. The role of amylin in the control of energy homeostasis. Am J Physiol Regul Integr Comp Physiol
19. Nyholm B, Orskov L, Hove KY, et al. The amylin analog pramlintide improves glycemic control and reduces postprandial glucagon concentrations in patients with type 1 diabetes mellitus. Metabolism
20. Polonsky K, Jaspan J, Pugh W, et al. Metabolism of C-peptide in the dog. In vivo
demonstration of the absence of hepatic extraction. J Clin Invest
21. Powers SK, Howley ET, Cox R. A differential catecholamine response during prolonged exercise and passive heating. Med Sci Sports Exerc
22. Qi D, Cai K, Wang O, et al. Fatty acids induce amylin expression and secretion by pancreatic beta-cells. Am J Physiol Endocrinol Metab
23. Ravussin E, Smith SR, Mitchell JA, et al. Enhanced weight loss with pramlintide/metreleptin: an integrated neurohormonal approach to obesity pharmacotherapy. Obesity (Silver Spring)
24. Schmitz O, Brock B, Rungby J. Amylin agonists: a novel approach in the treatment of diabetes. Diabetes
. 2004;53(3 suppl):S233-8.
25. Seufert J, Kieffer TJ, Habener JF. Leptin
inhibits insulin gene transcription and reverses hyperinsulinemia in leptin
mice. Proc Natl Acad Sci
26. Sonnenberg GE, Krakower GR, Hoffmann RG, Maas DL, Hennes MM, Kissebah AH. Plasma leptin
concentrations during extended fasting and graded glucose
infusions: relationships with changes in glucose
, insulin, and FFA. J Clin Endocrinol Metab
27. Trevaskis JL, Lei C, Koda JE, Weyer C, Parkes DG, Roth JD. Interaction of leptin
and amylin in the long-term maintenance of weight loss in diet-induced obese rats. Obesity (Silver Spring)
28. Turek VF, Trevaskis JL, Levin BE, et al. Mechanisms of amylin/leptin
synergy in rodent models. Endocrinology
29. Wahren J, Ekberg K, Johansson J, et al. Role of C-peptide in human physiology. Am J Physiol Endocrinol Metab
30. Whaley MH, Brubaker PH, Otto RM, editors. ACSM's Guidelines for Exercise Testing and Prescription
. 7th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2006. p. 62-3.