0021-972X/00/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2000 by The Endocrine Society
Vol. 85, No. 3 Printed in U.S.A.
Endocrine Response to High-Intensity Exercise: DoseDependent Effects of Dexamethasone* PATRICIA A. DEUSTER, JOHN S. PETRIDES†, ANITA SINGH†, GEORGE P. CHROUSOS, AND MERRILY POTH Departments of Military and Emergency Medicine (P.A.D.) and Pediatrics (M.P.), Uniformed Services University of the Health Sciences; and Developmental Endocrinology Branch, National Institutes of Child Health and Human Development (G.P.C.), Bethesda, Maryland 20814 ABSTRACT We recently reported that in 30 –50% of healthy men and women the release of ACTH and cortisol stimulated by exercise is not suppressed by prior administration of a 4-mg dose of dexamethasone (DEX). We now explore other potential differences between these subjects and those whose exercise response was suppressed by examining the effect of a smaller, 1-mg, dose of DEX on exercise-stimulated ACTH and cortisol. Men (n ⫽ 15) and women (n ⫽ 9) were studied during three high intensity exercise tests: one after taking placebo, one after taking 1 mg DEX, and one after taking 4 mg DEX. Before participation, subjects underwent a test for classification as either a high (HR; n ⫽ 10) or low (LR; n ⫽ 14) reactor and a maximal exercise test to assess maximal aerobic capacity. Distinct dose-related
T
HE REGULATORY and adaptive processes of the hypothalamic-pituitary-adrenal (HPA) axis are becoming increasingly important as more individuals are being diagnosed with stress-related disorders (1–3). Mechanisms of HPA axis regulation in response to stress, including the magnitude of the response, coordination of the response, and differential responses to various stimuli, are all of interest. We previously reported differential metabolic and neuroendocrine reactivities to the stress of exercise among the normal population; a subgroup of men and women exhibited high reactivity, as indicated by exaggerated exercise-induced increases in plasma lactate, glucose, arginine vasopressin (AVP), ACTH, and cortisol, whereas the other group displayed low to moderate reactivity (4, 5). Interestingly, high intensity exercise promoted escape of ACTH and cortisol from dexamethasone (DEX) suppression in the high (HR), but not the low/moderate (LR), reactors. Importantly, the strength of the stimulus can dictate the magnitude of ACTH and cortisol escape. Exercise at 100% of maximal capacity promoted escape in more individuals than exercise at 90% Received August 30, 1999. Revision received November 10, 1999. Accepted November 23, 1999. Address all correspondence and requests for reprints to: Patricia A. Deuster, Ph.D., M.P.H., Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, Maryland 20814-4799. E-mail:
[email protected]. * The opinions and assertions expressed herein are those of the authors and should not be construed as reflecting those of the Uniformed Services University of the Health Sciences or the Department of Defense. This project was supported by Uniformed Services University of the Health Sciences Project RO9142. † Formerly with Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences.
reductions in plasma concentrations of ACTH, cortisol, and dehydroepiandrosterone (DHEA) were noted for HR under the treatment conditions, whereas both doses of DEX blocked ACTH, cortisol, and DHEA release in LR. Furthermore, basal plasma cortisol, DHEA, and DHEA sulfate were significantly higher in HR compared to LR. Thus, there are inherent basal and stress-reactive differences in HR and LR, and these differences may be useful in constructing a model for the mechanisms and physiological regulation of hypothalamic-pituitaryadrenal axis activation. The question of whether these differences in reactivity of the ACTH-cortisol axis between the HR and LR groups have implications for individual short term function or long term health remains to be answered. (J Clin Endocrinol Metab 85: 1066 – 1073, 2000)
(6). In addition, we recently showed that those classified as HR also exhibit heightened adrenal activity after a mental stress paradigm, which included both public speaking and arithmetic, compared to LR (7). In the aforementioned studies, a 4-mg dose of DEX was used, a higher dose than that used clinically in the DEX suppression test (1 mg). We sought to determine whether the magnitude of cortisol and ACTH escape in response to exercise would be modified by the lower 1-mg dose of DEX and whether the differences between the HR and LR groups would persist with the smaller dose. We used our standardized exercise paradigm (4 – 8) to compare the metabolic and HPA axis responses of persons characterized by low (LR) and high (HR) neuroendocrine reactivity under conditions of placebo and two doses of DEX (1 and 4 mg). Moreover, it was of interest to determine whether there were differences in baseline levels of the adrenal hormones dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) between the HR and LR groups and whether these steroids would be suppressed by DEX under both basal and exercise conditions. Finally, given the known interactions between the HPA axis and the immune system (2, 3), we sought to compare the release of interleukin 6 (IL-6) in LR and HR, both basally and after exercise, under conditions of placebo and DEX treatment. Materials and Methods Subjects and protocol The study was approved by the institutional review board of the Uniformed Services University of the Health Sciences, and informed, written consent was obtained from all participants. Healthy, medicationfree men (n ⫽ 15) and women (n ⫽ 9) subjects were recruited to complete
1066
ENDOCRINE RESPONSES TO EXERCISE AND DEXAMETHASONE all phases of this study. A medical history, physical examination, and resting 12-lead electrocardiogram were obtained from each volunteer before entry into the study. Subjects reported to the laboratory on five occasions for 1) a classification exercise test, 2) a progressive maximal treadmill test to volitional exhaustion to determine maximal oxygen uptake (VO2 max), and 3) three standardized exercise tests, each on separate occasions, under conditions of placebo and 1 and 4 mg DEX. The Borg Perceived Exertion Scale was used at the end of each exercise session to document perceived effort and stress and relative intensity (9).
1067
cool-down of jogging/walking (3.3 mph) followed the run. The speeds and grades of the treadmill for a given subject were identical under each experimental condition. Blood samples were collected in heparinized tubes (15 IU heparin/mL blood) containing fluoride (1 mg fluoride/mL blood) for lactate and glucose determinations and in chilled ethylenediamine tetraacetate tubes (1.6 mg ethylenediamine tetraacetate/mL blood) for hemoglobin, hematocrit, ACTH, cortisol, AVP, DHEAS, DHEA, and IL-6 measurements. Plasma was separated and stored at ⫺70 C for later analyses.
Biochemical assays Visit 1: classification test for HR and LR To identify HR and LR, each volunteer underwent a high intensity exercise test to elicit an intensity approximating 90% of each subject’s VO2 max 8 h after receiving 4 mg DEX (Pathway Pharmacy, Bethesda, MD). The speed during the high intensity test was based on VO2 max results estimated from an 8-min submaximal cycle ergometry test as described by Lockwood et al. (10). The cycle ergometry test took place before the high intensity test, with a 20-min rest period between tests. Blood samples were obtained for measurement of ACTH 5 min before exercise, at the end of high intensity exercise (time ⫹20), and at the end of cool-down (time ⫹25). Subjects who showed a net increase in plasma ACTH (peak minus baseline) of more than 1.1 pmol/L over baseline levels were termed HR (n ⫽ 10), whereas those who failed to show a net rise in plasma ACTH were termed LR (n ⫽ 14).
Visit 2: determination of maximal oxygen uptake Each subject underwent a progressive maximal treadmill test to volitional exhaustion for quantifying VO2 max. The maximal exercise test, as previously described by Kyle et al. (11), was conducted on a motorized treadmill and began with a 10-min warm-up walk at 3.5 mph on a 10% grade. Treadmill speed was then increased to 5, 6, or 7 mph (depending on the usual running pace); the initial grade was set at 0% and every 2 min thereafter the treadmill grade was increased by 2.5%. Oxygen uptake and CO2 production during all exercise tests were determined with a Metabolic Measurement Cart 2900c (SensorMedics, Yorba Linda, CA). Electrocardiograms and heart rates were monitored continuously throughout all exercise protocols. The results of the maximal treadmill test were used to determine the treadmill speed (at 10% grade) to elicit an exercise intensity of 90% of each individual’s VO2 max for each of the treatment exercise tests. Verification that each subject actually achieved V02 max consisted of meeting three of the following five criteria: 1) achieving predicted maximal heart rate, 2) Borg’s perceived exertion scale rating of 17 or higher, 3) a respiratory exchange ratio of 1.10 or more, 4) an increase in oxygen uptake of 150 mL or more for an increase in workload, and/or 5) lactate concentration of 10.0 mmol/L or more.
Visits 3–5: standardized exercise tests The third, fourth, and fifth visits consisted of treadmill exercise at an intensity equivalent to 90% of each subject’s V02 max. All subjects took DEX (4 and 1 mg, orally) or placebo (150 mg lactose, orally) 8 h before the standardized exercise tests in a randomized, double blinded manner. The placebo and DEX pills were prepared by Pathway Pharmacy; only the pharmacist knew the composition of the pills. Each subject participated in all treatments, and no adverse reactions were reported. Tests were separated by at least 1 week to allow for drug metabolism and washout. Subjects abstained from caffeine and alcohol consumption and running or other strenuous activity for 15 h before testing. After arriving at the laboratory, subjects were uniformly hydrated by having them drink water (5 mL/kg BW); next an iv catheter for blood sampling was inserted into one forearm vein 50 min before exercise. Blood was drawn for baseline measurements at ⫺10 and 0 min relative to the start of exercise, 20 min after the start of exercise, at the end of exercise, and every 10 min after exercise for 50 min. Heart rate was also recorded before each blood draw. The exercise test consisted of 25 min of jogging/running. The initial 5 min served as a warm-up, during which each subject jogged at an intensity equivalent to 50% of his/her VO2 max. After the warm-up, the treadmill grade was increased to 10%, and the subject exercised at 70% VO2 max for 10 min and at 90% for the subsequent 5 min; a 5-min
Lactate and glucose concentrations were determined in duplicate (analyzer model 27, YSI, Inc., Yellow Springs, OH). Hemoglobin and hematocrit were determined in triplicate by the cyanomethemoglobin and microcapillary methods, respectively. Plasma cortisol was measured by RIA (Diagnostic Products, Webster, TX), and plasma ACTH concentrations were determined by a two-site immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). DHEAS and DHEA (Diagnostics Systems Laboratories, Inc., Webster, TX) were determined by RIA. Plasma AVP was extracted and assayed by RIA as previously described by Rittmaster et al. (12). The recovery using this procedure was greater than 90%. Detection limits for the cortisol, ACTH, DHEAS, and DHEA assays were 8.3 nmol/L, 0.22 pmol/L, 10.3 nmol/L, and 3.1 pmol/L, respectively; that for the AVP assay was 0.3 nmol/L. Intraassay coefficients of variation (CVs) for cortisol and ACTH, respectively, were less than 6% and 8%, whereas interassay CVs were less than 10% and 15%. Intraassay CVs for DHEAS and DHEA were 8.5% and 1.6%, respectively, whereas interassay CVs were 7.6% and 8.4%. Intra- and interassay CVs for AVP were 7.5% and 12%, respectively. Plasma IL-6 was measured by an enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN). Intra- and interassay CVs were 5.6% and 8.6%, respectively. The dynamic range of this high sensitivity assay was 0.156 –10 pg/mL. For all assays, all samples from a single subject were analyzed in one assay to eliminate interassay variations.
Statistical analyses The statistical software program, SAS (SAS Institute, Inc., Cary, NC), was used for all data analyses. Data were analyzed as a factorial design with repeated measures (group/treatment/time); a multivariate analysis of variance, general linear model was used. When significant effects were detected by multivariate analysis of variance, Duncan’s multiple range test was used to identify differences across time, group, and treatments. Significance was set at the 0.05 level. Areas under the curve (AUCs) were calculated by the trapezoidal method after subtracting the baseline. Data are presented as the mean ⫾ sem.
Results Subject characteristics: maximal exercise and classification test
Table 1 presents the general characteristics of the subjects and peak responses to the maximal exercise test by group. No TABLE 1. General characteristics of subjects and responses to maximal exercise testing
Age (yr) Wt (Kg) Ht (cm) Body fat (%) Max VO2 (mL/kg 䡠 min) Max heart rate (beats/min) Borg scale rating (RPE) Peak lactate (mmol/L) Peak ACTH (pmol/L) Peak cortisol (nmol/L) a
LR (n ⫽ 14)
HR (n ⫽ 10)
30.4 ⫾ 1.9 68.4 ⫾ 3.6 171.3 ⫾ 2.5 19.1 ⫾ 2.2 49.5 ⫾ 4.1 188 ⫾ 3.1 18.9 ⫾ 0.3 9.9 ⫾ 1.4 31 ⫾ 4.3 776 ⫾ 66.6
28.8 ⫾ 1.5 68.4 ⫾ 2.8 173.5 ⫾ 1.8 19.0 ⫾ 1.7 50.2 ⫾ 2.5 194 ⫾ 1.5 18.4 ⫾ 0.5 13.1 ⫾ 1.3a 43 ⫾ 7.4a 1127 ⫾ 83.2a
Significantly different from LR.
1068
significant differences in age, weight, height, body fat, VO2 max, or ratings of perceived exertion were noted between LR and HR. However, peak plasma lactate, ACTH, and cortisol were significantly lower in LR compared to HR subjects. Table 2 presents the responses of LR and HR to the classification test. Despite similar basal levels of ACTH (⬍1 pmol/ L), peak concentrations after exercise were clearly lower in LR compared to HR. Basal metabolic, neuroendocrine, and cytokine profiles
Table 3 presents basal plasma levels of various metabolic, neuroendocrine, and cytokine parameters. With respect to placebo conditions, HR had significantly higher concentrations of plasma cortisol, DHEAS, and DHEA and lower concentrations of plasma IL-6. With respect to the DEX treatments, plasma glucose was significantly (P ⬍ 0.01) enhanced, and plasma ACTH, cortisol, and DHEA levels were significantly suppressed by both 1 and 4 mg DEX in LR and HR compared to the effect of placebo pretreatment. In contrast, no significant change in either plasma lactate or DHEAS levels was noted after the administration of either DEX dosage. Moreover, DEX had minimal effects on plasma IL-6. When HR and LR were compared under the 1 and 4 mg DEX conditions, plasma ACTH and cortisol for the two groups were not significantly different after either dose. In contrast, DHEA levels remained significantly higher in HR compared to LR for both doses. This occurred despite a clear suppression of plasma concentrations for HR and LR. Likewise, plasma DHEAS levels were significantly higher in HR compared to LR across both doses of DEX, with only a hint of suppression in both groups. Mean basal plasma AVP concentrations were between 0.50 ⫾ 0.01 and 0.52 ⫾ 0.02 nmol/L for all treatment conditions. Importantly, plasma DEX concentrations were not significantly different for HR (1 mg, 343 ⫾ 98; 4 mg, 926 ⫾ 153 ng/dL) and LR (1 mg, 276 ⫾ 36; 4 mg, 1113 ⫾ 106 ng/dl) after either dose of DEX. Exercise-induced metabolic, neuroendocrine, and cytokine profiles
For all subjects combined, the mean speed of the treadmill at a 10% grade during the 90% high intensity exercise averaged 8.4 ⫾ 0.2 mph. Within each group, heart rate, relative VO2, and respiratory exchange ratio values averaged over the last 4 min of high intensity exercise were unaffected by DEX pretreatment, and HR and LR values were not different (data not shown). Figure 1 presents the patterns of change in plasma lactate and glucose (upper and lower panels, respectively) across the three treatment conditions. As expected, significant exerciseTABLE 2. Responses of low responders (LRs) and high responders (HRs) to classification exercise test after 4 mg dexamethasone
Max heart rate (beats/min) Borg scale rating (RPE) Peak lactate (mmol/L) Basal ACTH (pmol/L) Peak ACTH (pmol/L)
JCE & M • 2000 Vol 85 • No 3
DEUSTER ET AL.
LR (n ⫽ 14)
HR (n ⫽ 10)
186 ⫾ 3 17 ⫾ 1 7.9 ⫾ 0.8 ⬍1 1 ⫾ 0.5
195 ⫾ 4 17 ⫾ 1 12.5 ⫾ 1.3 ⬍1 4 ⫾ 1.2
TABLE 3. Basal concentrations of plasma lactate, glucose, ACTH, cortisol, DHEAS, DHEA, and IL-6 of low and high responders (LR/HR) after treatment with placebo and 1 and 4 mg dexamethasone (DEX)
Plasma LR HR Plasma LR HR Plasma LR HR Plasma LR HR Plasma LR HR Plasma LR HR Plasma LR HR a
lactate (mmol/L) glucose (mmol/L) ACTH (pmol/L) cortisol (nmol/L) DHEAS (mol/L) DHEA (nmol/L) IL-6 (pg/mL)
Placebo
1 mg DEX
4 mg DEX
1.1 ⫾ 0.2 1.2 ⫾ 0.1
0.9 ⫾ 0.1 1.3 ⫾ 0.1
1.2 ⫾ 0.1 1.4 ⫾ 0.2
4.8 ⫾ 0.1 4.8 ⫾ 0.1
5.2 ⫾ 0.1 5.2 ⫾ 0.2
6.4 ⫾ 0.2 6.5 ⫾ 0.2
3.8 ⫾ 0.7 4.3 ⫾ 0.6
0.7 ⫾ 0.1 0.8 ⫾ 0.2
0.5 ⫾ 0.1 0.5 ⫾ 0.1
469 ⫾ 47.1 603 ⫾ 38.6a
30 ⫾ 2.3 57 ⫾ 25.2
33 ⫾ 2.3 34 ⫾ 3.7
4.2 ⫾ 0.5 7.2 ⫾ 1.2a
3.1 ⫾ 0.5 6.2 ⫾ 0.8a
3.2 ⫾ 0.4 6.0 ⫾ 0.8a
26.7 ⫾ 2.3 37.2 ⫾ 4.1a
11.9 ⫾ 0.90 19.6 ⫾ 1.9a
11.1 ⫾ 1.0 15.4 ⫾ 1.6a
3.8 ⫾ 0.9 1.9 ⫾ 0.7a
3.8 ⫾ 0.8 2.1 ⫾ 0.7a
2.9 ⫾ 0.7 1.9 ⫾ 0.8a
Significantly different from LR under same condition.
induced increases in plasma lactate and glucose were noted for both LR and HR. Net integrated and peak plasma glucose responses were significantly higher in HR compared to LR under all three conditions (data not shown; P ⬍ 0.05). Administration of 4 mg DEX resulted in significantly higher plasma glucose values compared to placebo for both HR and LR, whereas the 1-mg dose was without significant effects. No effect of DEX was noted for lactate, but net integrated and peak plasma lactate responses to high intensity exercise were significantly greater in HR compared to LR across all treatment conditions (data not shown; P ⬍ 0.01). Both LR and HR showed significant increases in plasma ACTH (upper panel) and cortisol (lower panel) in response to high intensity exercise after placebo pretreatment (Fig. 2). These responses were significantly greater in HR compared to LR (P ⬍ 0.01). Pretreatment with DEX significantly attenuated exercise-induced increases in plasma ACTH and cortisol for both LR and HR compared to placebo (P ⬍ 0.05), but a clear response was seen for both the 1- and 4-mg doses in HR. In contrast, no increase in plasma ACTH or cortisol was noted for LR after either dose of DEX. Net integrated plasma ACTH and cortisol responses to exercise after DEX (Fig. 2, far right panels) were significantly higher in HR compared to LR under all conditions. Interestingly, the disparity of the ACTH and cortisol responses between HR and LR under the two DEX conditions was most evident during the 1-mg DEX test. The AUC for cortisol during the 1-mg test run was comparable to that for the placebo run for HR (placebo, 13.0 ⫾ 2.1; 1 mg, 12.9 ⫾ 3.1 nmol/80 min䡠L), but not for LR. As basal plasma cortisol levels in the groups were not significantly different after 1 mg DEX, potential differences in the absorption and metabolism of drug treatment were discounted. Patterns of change in DHEA (upper panel) and DHEAS (lower panel) are presented in Fig. 3. As stated above, basal plasma DHEA and DHEAS were significantly higher in HR
ENDOCRINE RESPONSES TO EXERCISE AND DEXAMETHASONE
1069
FIG. 1. Mean (⫾SEM) time courses for plasma lactate and glucose responses to exercise (upper and lower panels, respectively) after administration of placebo (left panels) and two doses of DEX [right panels; 1 mg (circles) and 4 mg (triangles)] in HR (open symbols) and LR (solid symbols). For all conditions, peak values for lactate and glucose were significantly greater for HR than for LR (P ⬍ 0.01).
compared to LR, and this pattern was maintained throughout exercise. Although neither exercise nor DEX administration had a significant impact on plasma DHEAS, plasma DHEA was significantly increased by exercise and significantly suppressed by DEX (Fig. 3). Moreover, DHEA values at all time points during exercise were significantly higher for HR compared to LR. In addition, a clear dose response was noted for DHEA in HR; plasma DHEA was significantly higher with 1 mg compared to 4 mg DEX as previously noted with ACTH and cortisol. Finally, no exercise-induced increase in DHEA was noted for LR, as had been observed with ACTH and cortisol. Lastly, plasma concentrations of AVP under conditions of placebo and DEX for HR and LR are presented in Fig. 4. As demonstrated in previous studies, exercise-induced plasma AVP levels were significantly higher in HR compared to LR. A trend toward enhanced AVP release during DEX treatments was noted, but due to the variability, significance was not achieved. Discussion
The present investigation revealed that although both 1 and 4 mg DEX suppressed the HPA axis response to exercise in the LR subjects, an incomplete and dose-dependent suppression was seen in the HR group. That is, in the HR group, the 4-mg dose suppressed more than the 1-mg dose, but neither dose totally prevented exercise-stimulated release of
ACTH and cortisol. Specifically, after the administration of 1 and 4 mg DEX, escape of pituitary ACTH and adrenal cortisol was observed after exercise at 90% of VO2 max in HR, but the magnitude of escape was significantly greater after 1 mg DEX. In contrast, LR maintained suppression of the HPA axis after both 1 and 4 mg DEX. Importantly, other new differences between HR and LR were also noted. In particular, HR had significantly higher basal and exercise-induced DHEA responses and higher basal DHEAS values compared to LR. We have previously reported that among the general population of healthy men and women, two groups can be identified based on their neuroendocrine responses to exercise after pretreatment with the glucocorticoid agonist, DEX (4, 5). The HR group mounts marked pituitary-adrenal responses to exercise in the absence of DEX and escapes pituitary adrenal suppression of HPA activation by 4 mg DEX when the same stimulus of exercise is applied (4, 5). In contrast, LR exhibit modest HPA responses to exercise in the absence of DEX, and complete suppression when exercise is used to activate the HPA axis after administration of 4 mg DEX (4, 5). More recently, we showed that HR also exhibit heightened adrenal activity after mental stress (7). In the present study we demonstrated that when a 1-mg dose of DEX is used, HR mount a marked response to exercise, whereas LR do not. Interestingly, in examining the AUCs for cortisol under 1 mg DEX and placebo, little suppression is
1070
DEUSTER ET AL.
JCE & M • 2000 Vol 85 • No 3
FIG. 2. Mean (⫾SEM) time courses for plasma ACTH and cortisol responses to exercise (upper and lower panels, respectively) after administration of placebo (left panels) and two doses of DEX [center panels; 1 mg (circles) and 4 mg (triangles)] in HR (open symbols) and LR (solid symbols). For all conditions, peak values for ACTH and cortisol were significantly greater for HR than for LR (P ⬍ 0.01). AUCs for placebo (open), 1 mg DEX (hatched), and 4 mg DEX (solid) conditions are shown in the right panels.
seen in the HR. This finding of differential sensitivity of the HPA axis to negative feedback in the two groups along with greater baseline and stress reactivities may have important clinical implications for individual health and vulnerability to stress-related disease. Whereas the site of regulation cannot be determined by these studies, it seems likely that target tissue sensitivities to glucocorticoids may be involved. Thus, glucocorticoid receptor sensitivity, density, synthesis, and function may be important. Consistent with previous findings, HR exhibited markedly greater lactate and AVP responses under conditions of DEX and placebo compared to LR (4, 5). Moreover, DEX did not appear to serve any decisive modulatory role for lactate. Although there are clear trends in the present and an earlier study (5) of enhanced AVP release under conditions of DEX in HR, the wide variability in the AVP response in the current study precludes achieving statistical significance. However, the implications of these consistent findings remain an area of considerable interest given the importance of lactate in metabolic regulation and of AVP in osmotic, volemic, pressure (12–14) and possibly, ACTH regulation (12–15). Because of our consistent findings of increased pituitary and adrenal reactivity in HR compared to LR, we also mea-
sured the adrenal hormone, DHEA, and its primary precursor and metabolite, DHEAS. DHEA and DHEAS are the most abundant steroid hormones in blood (16, 17) and can be converted to testosterone and aromatized to estrogens in the tissues of both men and women (18). Although the biological actions of DHEA and DHEAS have not been conclusively identified, affective, behavioral, cognitive, cardiovascular, and immune actions have been proposed (16 –23). Interestingly, significant differences between HR and LR were noted for these two hormones; basal DHEA and DHEAS concentrations were higher, and the magnitudes of the exerciseinduced increases in DHEA were significantly greater in HR compared to LR under all treatment conditions. Moreover, pretreatment with DEX blocked the DHEA response to exercise in LR, but not HR. These findings are of interest given current data from both animal and human studies with regard to DHEA. Data from animal studies indicate that DHEA may have important physiological roles in brain neuroendocrine systems (20, 24 –28). In particular, DHEA may function as a neurosteroid by binding to the ␥-aminobutyric acidA (GABAA) receptor complex in brain (20, 24 –27) and/or as an antiglucocorticoid in glucocorticoid-sensitive systems
ENDOCRINE RESPONSES TO EXERCISE AND DEXAMETHASONE
1071
FIG. 3. Mean (⫾SEM) time courses for plasma DHEA and DHEAS responses to exercise (upper and lower panels, respectively) under conditions of placebo (left panels) and two doses of DEX [right panels; 1 mg (circles) and 4 mg (triangles)] in HR (open symbols) and LR (solid symbols). For all conditions, basal and peak values for DHEA were significantly greater for HR than for LR (P ⬍ 0.01).
FIG. 4. Mean (⫾SEM) time course for plasma AVP responses to exercise (upper and lower panels, respectively) under conditions of placebo (left panels) and two doses of DEX [right panels; 1 mg (circles) and 4 mg (triangles)] in HR (open symbols) and LR (solid symbols). For all conditions, peak values for AVP were significantly greater for HR than for LR (P ⬍ 0.01).
(29 –32). If DHEA serves to restrain GABAergic activity and potentially overrides glucocorticoid negative feedback, it may normally regulate HPA reactivity via a GABAAergic mechanism. It is possible that DHEA/ DHEAS may facilitate activation of the HPA axis by antagonizing GABAA-induced restraint of CRH and AVP release. The findings in the current study lend support to the possibility that DHEA may serve in any one of those two capacities. Clearly, this will require further investigation. Also of interest to these findings is the knowledge that
HPA axis and metabolic patterns serve important functions with respect to behavior, cognition, and physical performance (1–3, 33, 34). The hormone ACTH is known to influence mood and cognitive performance (33, 34), and extensive evidence from animal and human studies indicates that stress and glucocorticoids influence cognitive function (35– 38). For these reasons, the neuroendocrine differences and reactivity to exercise and mental stress between HR and LR are intriguing. Does HPA and metabolic (re)activity contribute to differences in behavioral and cognitive functioning? Do the inherent neuroendocrine patterns of an individual
1072
DEUSTER ET AL.
dictate physical and cognitive performance characteristics and limitations? Do these specific neuroendocrine patterns reflect vulnerability to future health and/or disease processes? A number of diseases are associated with high cortisol values, and the side-effects of corticosteroid therapy are well documented (39). Recently, it has been suggested that the high cortisol values associated with various disease states may be an initiator of rather than a response to a particular disease (39). For this reason, antiglucocorticoid therapies, such as DHEA, are being used to mediate cortisol-induced changes (29, 31, 39). Interestingly, several studies indicate that under a number of conditions, such as selected diseases, aging, and food ingestion, DHEA/DHEAS levels are low relative to cortisol levels (40 – 44). Thus, DHEA/DHEAS may vary independently of circulating cortisol. One important question is whether the higher basal levels of DHEA observed in the HR reflect a protective response to minimize the effects of high cortisol responses. Clearly, the results of this study raise many intriguing and important questions that cannot be addressed by the current study. Perhaps, basal values are of no consequence, and if HR and LR were followed over a 24-h period or if diurnal differences were compared (45), the differences would be minimized. It is also possible that studies of 24-h secretion would show that differences between groups persist or even are accentuated. Future studies can be undertaken to make these determinations. In summary, the results of this study provide one model for studying mechanisms and physiological regulation of the HPA axis and its activation. This model is based on the findings that two distinct groups can be identified based on their neuroendocrine responses to exercise after pretreatment with the glucocorticoid agonist, DEX. The neuroendocrine response patterns of these two groups differ significantly, both at baseline and after exercise and with and without DEX pretreatment. The potential clinical importance of these differences in inherent hypothalamic-pituitary-adrenal reactivity, sensitivity to negative feedback, and levels of circulating DHEA/DHEAS remains uncertain.
9. 10. 11. 12.
13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
25. 26. 27. 28. 29.
References
30.
1. Chrousos GP, Gold PW. 1992 The concepts of stress and stress system disorders: overview of physical and behavioral homeostasis. JAMA. 267:1244 –1252. 2. Chrousos GP. 1995 The hypothalamic-pituitary adrenal axis and immune mediated inflammation. N Engl J Med. 332:1351–1362. 3. Chrousos GP. 1998 Editorial: ultradian, circadian, and stress-related hypothalamic-pituitary-adrenal axis activity–a dynamic digital-to-analog modulation. Endocrinology. 139:437– 440. 4. Petrides JS, Mueller GP, Singh A, Chrousos GP, Kalogeras K, Gold PW, Deuster PA. 1997 Differential responses to exercise-induced activation of the hypothalamic-pituitary-adrenal axis and arginine vasopressin stimulation. J Appl Physiol. 82:1979 –1988. 5. Petrides JS, Mueller GP, Chrousos GP, Kalogeras K, Deuster PA. 1994 Exercise-induced activation of the hypothalamic-pituitary-adrenal (HPA) axis: differential sensitivity to glucocorticoid suppression. J Clin Endocrinol Metab. 79:377–383. 6. Deuster PA, Petrides JS, Singh A, Lucci EB, Chrousos GP, Gold PW. 1998 High intensity exercise promotes escape of acth and cortisol from suppression by dexamethasone: sexually dimorphic responses. J Clin Endocrinol Metab. 83:3332–3338. 7. Singh A, Petrides JS, Gold PW, Chrousos GP, Deuster PA. 1999 Differential hypothalamic-pituitary-adrenal axis reactivity to psychological and physical stress. J Clin Endocrinol Metab. 84:1944 –1948. 8. Luger A, Deuster PA, Kyle SB, et al. 1987 Acute hypothalamic-pituitary-
31. 32. 33. 34. 35. 36. 37. 38. 39.
JCE & M • 2000 Vol 85 • No 3
adrenal responses to the stress of treadmill exercise: physiologic adaptations to physical training. N Engl J Med. 316:1309 –1315. Borg GAV. 1982 Psychological bases of perceived exertion. Med Sci Sports Exerc. 14:377–381. Lockwood PA, Yoder JE, Deuster PA. 1997 Comparison and cross-validation of cycle ergometry estimates of VO2 max. Med Sci Sports Exerc. 29:1513–1520. Kyle SB, Smoak BL, Douglass LW, Deuster PA. 1989 Variability of responses across training levels to maximal treadmill exercise. J Appl Physiol. 67:160 –165. Rittmaster RS, Cutler GB, Gold PW, Brandon DD, Tomai T, Loriaux DL, Chrousos GP. 1987 The relationship of saline-induced changes in vasopressin secretion to basal and corticotropin-releasing hormone stimulated adrenocorticotropin and cortisol secretion in man. J Clin Endocrinol Metab. 64:371–376. Verbalis JG. 1993 Osmotic inhibition of neurohypophysial secretion. Ann NY Acad Sci. 689:146 –160. Elias LLK, Antunes-Rodrigues J, Elias PCL, Moreira AC. 1997 Effect of plasma osmolality on pituitary-adrenal responses to corticotropin-releasing hormone and atrial natriuretic peptide changes in central diabetes insipidus. J Clin Endocrinol Metab. 82:1243–1247. Debold CR, Sheldon WR, DeCherney GS, et al. 1984 Arginine vasopressin potentiates adrenocorticotropin release induced by ovine corticotropin-releasing factor. J Clin Invest. 73:533–538. Baulieu E-E. 1996 Dehydroepiandrosterone (DHEA): a fountain of youth? J Clin Endocrinol Metab. 81:3147–3151. Longcope C. 1996 Dehydroepiandrosterone metabolism. J Endocrinol. 150:S125–S127. Watson RR, Huls A, Araghinikuam M, Chung S. 1996 Dehydroepiandrosterone and diseases of aging. Drugs Aging. 9:274 –291. Barrett-Connor E, Khaw K-T, Yen SSC. 1986 A prospective study of dehydroepiandrosterone, mortality, and cardiovascular disease. N Engl J Med. 315:1519 –1524. Baulieu E-E, Robel P. 1996 Dehydroepiandrosterone and dehydroepiandrosterone sulfate as neuroactive neurosteroids. J Endocrinol. 150:S221–S239. Loria M, Padgett DA, Huynh PN. 1996 Regulation of the immune response by dehydroepiandrosterone and its metabolites. J Endocrinol. 150:S209 –S220. Wolkowitz OM, Reus VI, Roberts E, et al. 1997 Dehyrdoepiandrosterone (DHEA) treatment of depression. Biol Psychiatry. 41:311–318. Khaw K-T. 1996 Dehydroepiandrosterone, dehydroepiandrosterone sulphate and cardiovascular disease. J Endocrinol. 150:S149 –S153. Kimonides VG, Khatibi NH, Svendsen CN, Sofroniew MV, Herbert J. 1998 Dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEAS) protect hippocampal neurons against excitatory amino acid-induced neurotoxicity. Proc Natl Acad Sci USA. 95:1851–1857. Baulieu EE. 1997 Neurosteroids: of the nervous system, by the nervous system, for the nervous system. Recent Prog Horm Res. 52:1–32. Sousa A, Ticku MK. 1997 Interactions of the neurosteroid DHEAS with the GABA(A) receptor complex. J Pharmacol Exp Ther. 282:827– 833. Steffensen SC. 1995 Dehydroepiandrosterone sulfate suppresses hippocampal recurrent inhibition and synchronizes neuronal activity to theta rhythm. Hippocampus. 5:320 –328. Reddy DS, Kulkarni SK. 1997 Differential anxiolytic effects of neurosteroids in the mirrored chamber behavior test in mice. Brain Res. 752:61–71. Arneo B, Daynes R. 1995 Dehydroepiandrosterone functions as more than an antiglucocorticoid in preserving immunocompetence after thermal injury. Endocrinology. 136:393– 401. Browne ES, Wright BE, Porter JR, Svec F. 1992 Dehydroepiandrosterone: antiglucocorticoid action in mice. Am J Med Sci. 303:366 –371. Fleshner M, Pugh CR, Tremblay D, Rudy JW. 1997 DHEA-S selectively impairs contextual-fear conditioning: support for the antiglucocorticoid hypothesis. Behav Neurosci. 111:512–517. Blauer KL, Poth M, Rogers M, Bernton EW. 1991 Dehydroepiandrosterone antagonizes the suppressive effect of dexamethasone on lymphocyte proliferation. Endocrinology. 129:3174 –3179. Deijen JB, de Boer H, Blok GJ, van der Veen EA. 1996 Cognitive impairments and mood disturbances in growth hormone deficient men. Psychoneuroendocrinology. 21:313–322. Wolkowitz OM, Tinklenberg JR, Weingartner H. 1985 A psychopharmacological perspective of cognitive functions. II. Specific pharmacologic agents. Neuropsychobiology. 14:133–156. de Quervain DJ-F, Roozendaal B, McGaugh JL. 1998 Stress and glucocorticoids impair retrieval of long-term spatial memory. Nature. 394;787–790. Lupien S, Lecours AR, Lussier I, Schwartz G, Nair NP, Meaney MJ. 1994 Basal cortisol levels and cognitive deficits in human aging. J Neurosci. 14:2893–2903. Lupien SJ, Gaudreau S, Tchiteya BM, et al. 1997 Stress-induced declarative memory impairment in healthy elderly subjects: relationship to cortisol reactivity. J Clin Endocrinol Metab. 82:2070 –2075. Lupien SJ, McEwen BS. 1997 The acute effects of corticosteroids on cognition: integration of animal and human model studies. Brain Res Rev. 24:1–27. Sapse AT. 1997 Cortisol, high cortisol diseases, and anti-cortisol therapy. Psychoneuroendocrinology. 22:S3–S10.
ENDOCRINE RESPONSES TO EXERCISE AND DEXAMETHASONE 40. Parker LN, Eugene J, Farber D, Lifrak ET, Lai M, Juler G. 1985 Dissociation of adrenal androgen and cortisol levels in acute stress. Horm Metab Res. 17:209 –212. 41. Cunningham SK, McKenna TJ. 1994 Dissociation of adrenal androgen and cortisol secretion in Cushing’s syndrome. Clin Endocrinol (Oxf). 41:795– 800. 42. Korbonits M, Trainer PJ, Nelson ML, et al. 1996 Differential stimulation of cortisol and dehydroepiandrosterone levels by food in obese and normal subjects: relation to body fat distribution. Clin Endocrinol (Oxf). 45:699 –706. 43. Pavlov EP, Harman SM, Chrousos GP, Loriaux DL, Blackman MR. 1986
1073
Responses of plasma adrenocorticotropin, cortisol, and dehydroepiandrosterone to ovine corticotropin-releasing hormone in healthy aging men. J Clin Endocrinol Metab. 62:767–772. 44. Spratt DI, Longcope C, Cox PM, Bigos ST, Wilbur-Welling C. 1993 Differential changes in serum concentrations of androgens and estrogens (in relation with cortisol) in postmenopausal women with acute illness. J Clin Endocrinl Metab.76:1542–1547. 45. Heuser I, Deuschle M, Luppa P, Schweiger U, Standhardt H, Weber B. 1998 Increased diurnal plasma concentrations of dehydroepiandrosterone in depressed patients. J Clin Endocrinol Metab. 83:3130 –31133.
