Approach to the Problem and Experimental Design
A preliminary experiment was conducted to determine whether 30 min of moderately intense continuous treadmill running altered adiponectin concentrations in healthy males. The subjects served as their own controls. In the second experiment, we examined whether adiponectin concentrations were affected by a strenuous intermittent treadmill protocol (using four different exercise intensities) designed to be a provocative stimulus for glucose and glucoregulatory hormone responses. One month later, each of the same subjects completed a control trial that excluded exercise and followed an identical blood draw protocol as the exercise trial. The exercise and control trial responses were then compared. The latter experiment was an extension of a previous study of intermittent, intense exercise on glucoregulatory hormones (18). The main outcome measures of the study were serum concentrations of adiponectin and insulin, and plasma concentrations of glucose and lactate.
The subjects were six healthy young males recruited from the university and surrounding communities. The mean (±SEM) age, weight, height, percent fat, and V̇O2max were 23.0 ± 1.34 yr, 78.56 ± 3.33 kg, 174.83 ± 4.17 cm, 13.75 ± 1.13%, and 50.0 ± 1.9 mL·kg−1·min−1, respectively. Criteria for participation in the study included 1) no past history of any metabolic or cardiovascular diseases, 2) not taking any prescription medications, 3) no history of smoking, 4) not taking any over-the-counter nutritional supplements, and 5) able to complete 30 min of moderately intense treadmill exercise. The investigation was approved by the Southeastern Louisiana University Institutional Review Board and conducted in accordance with the policies of the American College of Sports Medicine.
Subjects completed written informed consent and a medical history questionnaire. Height and weight of the subjects were measured and body composition was assessed using skinfold measurements (13). Subjects were familiarized with treadmill running and exposed to the methods used in the experimental trial. Subjects completed a graded treadmill exercise test at a constant 3% grade to determine V̇O2max. V̇O2 was determined every 30 s using a metabolic system (ParvoMedics, Sandy, UT).
Subjects reported to the exercise physiology lab at 0730 h after an overnight fast. After a 15-min rest period in a seated position, a 15-mL blood sample was collected by venipuncture into a 10-mL whole-blood tube, 2-mL tube with EDTA, and a 3-mL tube with sodium fluoride and potassium oxalate. Then subjects began a 30-min treadmill exercise bout at a speed predicted to elicit a workload of 80% of the V̇O2max determined in the preliminary trial. V̇O2 was assessed during the exercise session to verify exercise workload. Subjects exercised at 79 ± 1.8% V̇O2max. Another 15-mL blood sample was collected in a seated position immediately after exercise and after 30 min of seated recovery.
Seven trained male runners were recruited from the university community and provided written consent for participation in the study. A description of the subjects and exercise protocol that was conducted is described in our earlier report (18). In brief, the subjects had a mean (±SEM) age, weight, height, percent fat, and V̇O2max of 28.71 ± 2.91 yr, 73.39 ± 4.14 kg, 179.80 ± 2.53 cm, 11.08 ± 1.01%, and 61.01 ± 2.37 mL·kg−1·min−1, respectively. Criteria for participation were the same as for experiment 1, with the exception that subjects were well trained (V̇O2max ≥52.0 mL·kg−1·min−1 and personal record for a 10-km run ≤37 min) in order to ensure the completion of a more intense running protocol.
Subjects completed a preliminary trial to determine standard fitness measures that included body composition (skinfold measures) and cardiorespiratory fitness (V̇O2max). Subjects completed a graded exercise test to exhaustion at a constant grade. The treadmill speeds (at 4% treadmill grade) that corresponded with 60, 75, 90, and 100% V̇O2max were calculated from a regression equation generated from the relationship between V̇O2 and treadmill speed in the preliminary trial.
Exercise and control trials.
An intravenous catheter (Travenol, 22 g, 32 mm) was inserted into an antecubital vein, and a normal saline lock was attached. At 0830 h, 40 min before exercise (−40) and at 0900 h, 10 min before exercise (−10), resting blood samples were collected from the catheter while the subjects were in a seated position. Subjects then completed an intermittent treadmill exercise protocol at four speeds predicted to elicit the specific workloads of 60% V̇O2max for 10 min, 75% V̇O2max for 10 min, 90% V̇O2max for 5 min, and 100% V̇O2max for 2 min. After each workload was completed at the prescribed intensity and duration, treadmill speed was reduced to a walking speed (for 3.5–4 min) while a blood sample was collected. Gas samples were collected continuously and confirmed that the actual V̇O2 corresponded with the predicted V̇O2 for each workload. The total exercise time, including the 3.5–4 min of activity between the four exercise intensities, was 38–39 min. One month later, subjects reported to the lab at 0745 h after an overnight fast for a control trial. Blood samples were drawn (in the seated position) at the identical times of the exercise trial, but exercise was excluded in this trial.
In addition to blood samples collected from the intravenous catheter after each workload (60, 75, 90, and 100% V̇O2max), samples were also collected every 15 min during a 1-h recovery in a seated position (R15, R30, R45, and R60).
For both experiments, serum and plasma aliquots were stored at −80°C until assayed. Blood samples were analyzed for glucose and lactate using an enzymatic method (Sigma Chemical, St. Louis, MO) and for adiponectin using a radioimmunoassay (Linco Research, Inc., St. Charles, MO) and insulin using a sensitive chemiluminescent assay (Immulite, Diagnostic Products Corp., Los Angeles, CA). Hemoconcentration was calculated (5) from hematocrit (microhematocrit method) and hemoglobin levels determined from a colorimetric method (Sigma Chemical). The interassay coefficient of variation for adiponectin and insulin was 14.79 and 10.6%, respectively, and the intra-assay coefficient of variation <5.0%. The minimum sensitivity for the adiponectin and insulin assays was 1 ng·mL−1 and 2 μIU·mL−1, respectively.
For experiment 1, hormone and substrate concentrations were analyzed using a repeated measures ANOVA. For experiment 2, hormone and substrate concentrations were analyzed using a trial (exercise and control) × time (1–10) repeated measures ANOVA. Data were also converted to percent change from baseline and analyzed to account for variability in adiponectin concentrations. An alpha level of P < 0.05 was considered statistically significant. Post hoc power analyses were conducted after each ANOVA to provide values for eta squared, a statistic indicating power to detect observed differences as significant. Eta-squared values indicated power to detect significant differences, given the research design and number of participants, was adequate when a factor accounted for approximately 30% or more of the variance in the data.
There was a shift in plasma volume of −8.55 ± 1.85% from pre- to postexercise. Lactate rose from 0.79 ± 0.23 mM to 3.34 ± 0.79 mM immediately postexercise and declined to 1.21 ± 0.37 mM 30 min postexercise, indicating a moderately rigorous protocol. Glucose, insulin, and adiponectin responses are shown in Table 1. There was no significant change in glucose and insulin across time, although there was a trend for glucose to increase postexercise. There was a significant increase in adiponectin across time. Adiponectin rose significantly from pre- to postexercise; however, the increase was relatively small, constituting a 9.8% change. Post hoc analyses indicated 59% of the variance in adiponectin was attributed to the time effect, and observed power for detecting these differences as significant was 0.83. When individual adiponectin values from experiment 1 were corrected for plasma volume shifts, a repeated measures ANOVA revealed no significant differences across time.
As described previously (18), the greatest reduction in plasma volume between any two time points was −9.9% during exercise (after 60% V̇O2max) and −2.2% during recovery (60 min recovery). Lactate values rose from preexercise of 0.96 ± 0.18 mM to peak after 100% V̇O2max at 10.19 ± 1.2 mM, verifying that the protocol in experiment 2 produced considerably more metabolic stress on skeletal muscle than experiment 1. Glucose and insulin levels changed significantly across time with resting values of 86.13 ± 5.8 mg·dL−1 and 4.76 ± 0.88 μIU·mL−1 to peak values after 100% V̇O2max at 140.42 ± 12.56 mg·dL−1 and 14.37 ± 3.46 μIU·mL−1, respectively (18). Trial × time ANOVA revealed significant increases in adiponectin concentrations across time for both the exercise and control trials with no significant difference between the two trials (Fig. 1). Eta-squared values indicated 29% of the variance in adiponectin was due to the time factor, 18% to the trial factor, and only 2% to the trial by time interaction. Post hoc power calculations were 0.84 for the time factor, 0.33 for the phase factor, and 0.09 for the interaction. When individual adiponectin values were corrected for plasma volume shifts from experiment 2, a trial × time repeated measures ANOVA revealed the same findings as without correction, i.e., a significant time effect but no significant time × trial interaction.
To account for the variability among subjects, the exercise and recovery values were converted to percent change values, and exercise and control trials compared. Percent change during exercise and recovery was calculated for each subject as the difference between their resting adiponectin concentration (average of the individual’s values at −40 and −10) and adiponectin values at each time point during exercise and recovery, divided by the individual’s resting concentration. The greatest percent increase in adiponectin during exercise occurred from baseline to 10 min after 60% V̇O2max. There was a significant time effect as in experiment 1 but no significant difference between the exercise and control trials (Fig. 2).
Exercise is known to increase insulin sensitivity (2,8) and higher levels of adiponectin are associated with increased insulin sensitivity (9,20). The aim of the present study was to determine whether acute exercise affects adiponectin. The first experiment demonstrated that a moderate running protocol (79.0% V̇O2max for 30 min) was associated with a small (10%), but significant, immediate postexercise increase in adiponectin levels, but after correcting for hemoconcentration, the increase was no longer significant. There was little change in glucose and insulin concentrations in healthy young men. For the second experiment, previous data from our laboratory revealed that a more rigorous running protocol, consisting of progressive, intermittent high-intensity exercise, elicited large increases in glucose and insulin immediately postexercise (18) in trained male runners, but in the present study, there were moderate increases (−23%) in adiponectin after 10 min at 60% V̇O2max, followed by reductions at higher intensities (−19.0% reduction after 90 and 100% V̇O2max), and a rebound (∼11.0%) during recovery. The changes in adiponectin, however, paralleled and were not significantly different from the adiponectin changes in the same runners in a resting control trial indicating no increased stimulation and release from the exercise. Moreover, after correcting individual adiponectin concentrations for hemoconcentration and reanalyzing the data statistically, results remained the same. The data suggest that 30 min of heavy continuous running or more strenuous intermittent running does not stimulate an increase in production and release of adiponectin, and small increases in adiponectin concentrations resulting from the exercise may be attributed to normal plasma volume shifts. This is the first study to report acute effects of exercise on circulating adiponectin concentrations.
Glucose and insulin concentrations were not significantly elevated after exercise in the first experiment (30 min at 79% V̇O2max), but they were significantly elevated (170% increase in insulin immediately after 100% V̇O2max) in the more rigorous second experiment (18), which is in agreement with prior findings. Yale et al. (29) have demonstrated that sustained cycling exercise at 80% of V̇O2max for approximately 7 min will produce little or no change in plasma glucose during exercise in lean subjects, quickly followed by a sharp elevation in recovery. It has been reported that during exercise at 100% of V̇O2max (the same as the fourth work load of experiment 2), there is a 700-800% increase in production and approximately 400% increase in utilization of glucose, resulting in increase circulating glucose levels (3,17,26). Hyperglycemia is associated with hyperinsulinemia soon after a brief bout of exercise greater than 85% of V̇O2max (16,22,24,26), which is consistent with insulin increases postexercise in the second experiment. Further, there is evidence that insulin is an inhibitor of adiponectin gene expression (7), yet with an increase in circulating insulin concentrations in the present study, no discernible changes in adiponectin were observed. Perhaps a longer exercise duration would affect adiponectin or more time (longer than 1 h postexercise) may be required for hormonal alterations during exercise to affect the expression of the adiponectin gene.
One very recent study examined the effects of exercise training on adiponectin by comparing a pretraining with a posttraining blood sample (11). Hulver et al. (11) trained three females and five males using different forms of aerobic exercise for 6 months and found a significant relationship between posttraining, resting insulin, and adiponectin concentrations. However, they did not find significant changes in adiponectin concentrations from pre- to posttraining and suggested that adiponectin does not contribute to exercise improvements in insulin sensitivity.
Although adiponectin concentrations appeared to decline during exercise in the second experiment and rebound during recovery, there was a parallel pattern for the resting control trial. However, we do not think this pattern was due to circadian rhythm. One published report has concluded no daily diurnal rhythm in adiponectin for both diabetic and nondiabetic subjects (9). However, few sampling time points were included in their study: seven samples within 18 h. Recent evidence from adiponectin concentrations frequently sampled from normal-weight females over 24 h also suggests that adiponectin does not have a diurnal rhythm (personal communication, Peter Havel, University of California, Davis). In experiment 2 of the present study, the greatest alteration in PV change, −9.9%, occurred between the second rest period blood sample and the first exercise blood sample. This also was the point of greatest change in adiponectin (see Figs. 1–2). It appears that effects from hemoconcentration due to PV shifts during the exercise trial account for much of the increase in adiponectin in response to the exercise. Data from experiment 1 also supports this contention, because the significant increase in adiponectin after exercise was not significant when values were corrected for plasma volume shifts. Possible effects of increased exposure of tissue receptors to greater adiponectin levels from hemoconcentration will require further investigation.
There was considerable difference in the metabolic stress on skeletal muscle produced by the exercise bouts in the two experiments, yet no differences in adiponectin responses were observed. Peak lactate values immediately after exercise in the second experiment were three times greater than those after exercise in the first experiment, suggesting a clear difference in rate of glycogen utilization without a clear alteration in adiponectin. This is an interesting observation because it has been speculated that the acute effects of exercise on increased insulin sensitivity may be related to depletion of muscle glycogen, which might indirectly stimulate GLUT4 translocation (2,12).
It is known that acute exercise can increase insulin sensitivity, although the mechanisms and threshold are not well defined (27), and in the present study, it is possible that insulin sensitivity did not change. However, it has been demonstrated that in rodents, 30 min of heavy endurance exercise increases insulin sensitivity measured at 0.25 and 2 h after exercise (19). Moreover, a single 36-min bout of heavy endurance cycling in humans has been shown to improve the rate of glucose disappearance for as long as 11 h after exercise (8). The exercise protocols used in both experiments in the present study were similar or of greater duration and intensity than used in the protocols in those previous investigations. Additionally, some prior data suggest that a minimum exercise intensity of 70% V̇O2max is required to ensure exercise effects on glucose metabolism (24,17), and both protocols in the present study exceeded this exercise intensity.
In conclusion, we have demonstrated that 30 min of heavy continuous running or more strenuous intermittent running does not stimulate an increase in production and release of adiponectin, and small increases in adiponectin concentrations resulting from the exercise may be attributed to normal plasma volume shifts. Future studies are required to determine whether a longer duration of exercise may affect adiponectin, and additional studies are needed to determine whether adiponectin expression shows a delayed increase after exercise.
We wish to thank the subjects who participated in the study. We are also grateful to Ginger Kraemer, R.N., M.A., and Linda Synovitz, R.N., Ph.D., for their help with phlebotomy, as well as Matt McClain, Cooper Chadick, and Hongnan Chu for their assistance with data collection. We are especially grateful to Terry Gimpel for her work in the radioimmunoassay laboratory.
The study was funded by a Faculty Development Grant and the B.C. Purcell Endowed Professorship in Kinesiology from Southeastern Louisiana University.
1. Arita, Y., S. Kihara, N. Ouchi, et al Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem. Biophys. Res. Commun. 257: 79–83, 1999.
2. Borghouts, L. B., and H. A. Keizer. Exercise and insulin sensitivity: a review. Int. J. Sports Med. 21: 1–12, 2000.
3. Calles, J., J. J. Cunningham, L. Nelson, et al. Glucose turnover during recovery from intensive exercise. Diabetes 32: 734–738, 1983.
4. Commuzzie, A. G., T. Funahashi, G. Sonnenberg, et al. The genetic basis of plasma variation in adiponectin, a global endophenotype for obesity and the metabolic syndrome. J. Clin. Endocrinol. Metab. 86: 4321–4325, 2001.
5. Dill, D. B., and D. L. Costill. Calculation percentage changes in volumes of blood, plasma, and red cell dehydration. J. Appl. Physiol. 37: 247–248, 1974.
6. Fasshauer, M., J. Klein, S. Neumann, M. Eszlinger, and R. Paschke. Adiponectin gene expression is inhibited by beta-adrenergic stimulation via protein kinase A in 3T3-L1 adipocytes. FEBS Lett. 507: 142–146, 2001.
7. Fasshauer, M., J. Klein, S. Neumann, M. Eszlinger, and R. Paschke. Hormonal regulation of adiponectin gene expression in 3T3-L1 adipocytes Biochem. Biophys. Res. Commun. 290: 1084–1089, 2002.
8. Higaki, Y., J. Kagawa, A. Fujitana, et al. Effects of a single bout of exercise on glucose effectiveness. J. Appl. Physiol. 80: 754–759, 1996.
9. Hotta, K., T. Funahaski, Y. Arita, et al. Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 d, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler. Thromb. Vasc. Biol. 20: 1595–1599, 2000.
10. Hu, E., P. Liang, and B. M. Spiefelman. ADIPOQ is a novel adipose-specific gene dysregulated in obesity. J. Biol. Chem. 271: 10697–10703, 1996.
11. Hulver, M. W., D. Zheng, C. J. Tanner, et al. Adiponectin is not altered with exercise training despite enhanced insulin action. Am. J. Physiol. Endocrinol. Metab. 283: E861–E865, 2002.
12. Ivy, J. L., and C. H. Kuo. Regulation of GLUT4 protein and glycogen synthase during muscle glycogen synthesis after exercise. Acta Physiol. Scand. 162: 295–304, 1998.
13. Jackson, A. S., and J. H. Wilmore. Generalized equations for predicting body density of man. Br. Med. J. 40: 499–504, 1978.
14. Kang, J., R. J. Robertson, J. M. Hagberg, et al. Effect of exercise intensity on glucose and insulin metabolism in obese individuals and obese NIDDM patients. Diabetes Care 19: 341–349, 1996.
15. Kappes, A., and G. Loffler. Influences of ionomycin, dibutyryl-cycloAMP and tumour necrosis factor-alpha on intracellular amount and secretion of apM1 in differentiating primary human preadipocytes. Horm. Metab. Res. 32: 548–554, 2000.
16. Kjaer, M., P. A. Farrell, N. J. Christensen, and H. Galbo. Increased epinephrine response and inaccurate glucoregulation in exercising athletes. J. Appl. Physiol. 61: 1693–1700, 1986.
17. Kjaer, M., B. Kiens, M. Hargreaves, and E. A. Richter. Influence of active muscle mass on glucose homeostasis during exercise in humans. J. Appl. Physiol. 71: 552–557, 1991.
18. Kraemer, R. R., E. O. Acevedo, L. B. Synovitz, et al. Glucoregulatory endocrine responses to intermittent exercise of different intensities: plasma changes in a pancreatic beta-cell peptide, amylin. Metabolism 51: 657–663, 2002.
19. Langfort, J., L. Budohoski, and E. A. Newsholme. Effect of various types of acute exercise and exercise training on the insulin sensitivity of rat soleus muscle measured in vitro. Pflugers Arch. 41: 101–105, 1988.
20. Maeda, N., M. Takahashi, T. Funahashi, et al. PPAR-gamma ligands increase expression and plasma concentrations of adiponectin, an adipose-derived protein. Diabetes 50: 2094–2099, 2001.
21. Maeda, K., K. Okubo, I. Shimomura, T. Funahashi, Y. Matsuzawa, and K. Matsubara. cDna cloning and expression of a novel adipose specific collagen-like factor, apm1 (adipose most abundant gene transcript 1). Biochem. Biophys. Res. Commun. 221: 286–289, 1996.
22. Marliss, E. B., S. H. Kreisman, A. Manzon, J. B. Halter, M. Vranic, and S. J. Nessim. Gender differences in glucoregulatory responses to intense exercise. J. Appl. Physiol. 88: 457–466, 2000.
23. Mikines, K. J., B. Sonne, P. A. Farrell, B. Tronier, and H. Galbo. Effect of physical exercise on sensitivity and responsiveness to insulin in humans. Am. J. Physiol. Endocrinol. Metab. 254: E248–E259, 1988.
24. Purdon, C., M. Brousson, S. L. Nyveen, et al. The roles of insulin and catecholamines in the glucoregulatory response during intense exercise and early recovery in insulin-dependent diabetic and control subjects. J. Clin. Endocrinol. Metab. 76: 566–573, 1993.
25. Scherer, P. E., S. Williams, M. Fogliano, G. Baldini, and H. F. Lodish. A novel serum protein similar to C1q, produced exclusively in adipocytes. J. Biol. Chem. 270: 26746–26749, 1995.
26. Sigal, R. J., C. Purdon, D. Bilinski, M. Vranic, J. B. Halter, and E. B. Marliss. Glucoregulation during and after intense exercise: effects of beta-blockade. J. Clin. Endocrinol. Metab. 78: 359–366, 1994.
27. Thompson, P. D., S. F. Crouse, B. Goodpaster, D. Kelley, N. Moyna, and L. Pescatello. The acute versus the chronic response to exercise. Med. Sci. Sports Exerc. 33: S438–S445, 2001.
28. Weyer, C., T. Funahashi, S. Tanaka, et al. Hypoadiponectinemia in obesity and type 2 diabetes: close association with insulin resistance and hyperinsulinemia. J. Clin. Endocrinol. Metab. 86: 1930–1935, 2001.
29. Yale, J. F., L. A. Leiter, and E. B. Marliss. Metabolic responses to intense exercise in lean and obese subjects. J. Clin. Endocrinol. Metab. 68: 438–445, 1989.
30. Yamauchi, T., J. Kamon, H. Waki, et al. The fat-derived hormone adiponectin reverses insulin resistance associated with both lipoatrophy and obesity. Natl. Med. 7: 941–946, 2001.
31. Yang, W. S., W. J. Lee, T. Funashashi, et al. Weight reduction increase plasma levels of an adipose-derived anti-inflammatory protein, adiponectin. J. Clin. Endocrinol. Metab. 86: 3815–3819, 2001.