| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Internal Medicine I (P.W., H.L.F., C.D.) and Neuroendocrinology (J.B.), University Hospital Schleswig Holstein, D-23538 Luebeck, Germany
Address all correspondence and requests for reprints to: Peter Wellhoener, M.D., University Hospital Schleswig Holstein, Campus Luebeck, Department of Internal Medicine, D-23538 Luebeck, Germany. E-mail: peter.wellhoener{at}innere1.uni-luebeck.de.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Recent studies suggest that limbic MRs are involved in the feedback regulation of the human HPA axis (7, 8), and experiments in rats have shown that corticosterone exerts an inhibitory control on HPA activity via hippocampal MR (9). In humans, MR-mediated effects on HPA axis regulation have been examined after MR blockade with canrenoate. Canrenoate potassium is the active metabolite of spironolactone and a specific MR antagonist (10) that penetrates the blood-brain barrier (11). In humans canrenoate has a half-life of 3.7 ± 1.2 h and is metabolized by the liver to inactive compounds and excreted via the kidneys (12). MR blockade increases adrenocortical activity in response to an iv injection of CRH and vasopressin, indicating that MRs mediate effects on pituitary function (7). Additionally, MR blockade in humans increases cortisol levels during nocturnal sleep (13). Cortisol secretion is actively inhibited during this time, and MRs may play a key role in this inhibition (14). It is unclear, however, whether MRs also affect HPA activation in response to physiological stimuli. Furthermore, it is unknown whether such MR-related alteration of HPA activity is associated with changes in the somatotropic activity in humans as well. A previous study (13) showed slightly increased human GH (hGH) concentrations during nocturnal sleep after MR blockade by canrenoate.
The present study aims to clarify the role of MRs in the regulation of cortisol and hGH secretion during activation of the HPA axis by a physiological stressor. Physical exercise has repeatedly been shown to stimulate the somatotropic and corticotropic axis (15, 16, 17, 18, 19, 20, 21). It is likely that this complex stimulus involves neuroendocrine centers including hypothalamic and hippocampal regions that mediate regulatory MR effects in animals (9, 22). Consequently, such a stressor may sensitively reveal the role of MRs in the regulation of the stimulated HPA activity, which is not detected by stimulation of the pituitary or the adrenocortex after iv injection of exogenous CRH, vasopressin, and ACTH, respectively (23). Exercise also has the advantage that it enables an extended duration of HPA stimulation. It can be expected that feedback inhibition of cortisol during prolonged stimulation becomes particularly clear because inhibitory influences, launched at the beginning of the stressor, develop during the continuous stimulation and may be more evident toward the end of the stimulatory period.
Therefore, we examined the influence of MRs on the physiologically stimulated stress axis during a submaximal nonexhaustive treadmill exercise of 1-h duration in healthy men. For the evaluation of MR influences under this stimulus on ACTH, cortisol and hGH secretion receptors were blocked by the MR antagonist canrenoate.
| Subjects and Methods |
|---|
|
|
|---|
To evaluate the effect of a subchronic treatment (24), canrenoate (200 mg) or placebo was iv injected in two separate doses 24 and 8 h before the treadmill exercise after a randomized, double-blind, cross-over protocol. The two experimental sessions were divided by a washout period of 46 wk. On the experiment day, all participants were allowed to have a standard meal at 1200 h and drink plain water until the start of the experiment. The experiments started at 1515 h with the insertion of an iv catheter (Vasofix, Braun, Melsungen, Germany) into the antecubital vein of the right arm. At 1530 h participants rested comfortably on a bench in a supine position and blood samples for the determination of baseline values of ACTH, hGH, and cortisol were sampled every 20 min. After 60 min the participants started exercising on a treadmill (Jaeger, Hoechberg, Germany). To enable a steady physical stress over 1 h, the workload was set to 160 W. This corresponds to a speed of 4.3 km/h and an inclination of 15% according to the criteria of the American Heart Association (25). Heart rate was monitored electrocardiographically (Hellige, Freiburg, Germany). During the exercise period, blood was sampled every 10 min. After a 60-min interval of exercise, participants were asked to rest comfortably in a horizontal position for the next 90 min. During the rest period, blood samples were taken every 20 min.
Blood was immediately centrifuged for 10 min at 4 C and 4000 rpm, and plasma and serum was stored at 80 C until final analysis.
Hormone assays
Cortisol was determined by RIA (DPC Biermann GmbH, Bad Nauheim, Germany). The sensitivity was 5.5 nmol/liter, the intraassay coefficient of variation was 3% between 28.0 and 1380.0 nmol/liter. There was no detectable cross-reactivity against canrenoate (13). ACTH was determined by immunoluminometric assay (Lumitest; Brahms, Berlin, Germany). The sensitivity was 0.44 pmol/liter, the intraassay coefficient of variation was less than 8%. The interassay coefficient of variation for the ACTH and cortisol assay was less than 10%. Human GH was determined by RIA (hGH-RIA, DPC Biermann GmbH) with a sensitivity of 0.9 ng/ml (1.8 mIU/liter) [inter- and intraassay variation less than 5.9% for concentrations between 1.7 and 10.9 ng/ml (3.4 and 21.8 mIU/liter)]. All samples were analyzed in duplicate in the same assay.
Statistical evaluation
Average values of the different variables were calculated for the three experiment periods: period 1, rest before exercise; period 2, treadmill exercise; period 3, rest after exercise. Statistical analysis of the effects of canrenoate vs. placebo and the different phases of physical activity and rest relied on ANOVA with a repeated measure factor of time for the three experiment periods and the condition of treatment (canrenoate vs. placebo). When ANOVA showed significant treatment effects, subsequent post hoc pairwise comparisons were performed. Degrees of freedom were corrected according to the method of Greenhouse-Geisser. A value of P < 0.05 was considered significant. Results are presented as mean ± SEM.
| Results |
|---|
|
|
|---|
Cortisol
Exercise induced an increase in cortisol in all subjects and all treadmill sessions irrespective of the treatment (P < 0.05 for main effect of time). After canrenoate cortisol levels reached a maximum 20 min after the end of exercise (416.99 ± 60.62 nmol/liter at t = 140 min), whereas cortisol after placebo peaked somewhat earlier at the end of the exercise period with no further increase thereafter (maximum cortisol 334.42 ± 54.15 nmol/liter at t = 140 min). Whereas baseline-to-peak increases in cortisol on exercise were comparable for both treatment conditions (
= 156.40 ± 61.25 nmol/liter, 136.39 ± 67.40 nmol/liter for canrenoate and placebo, P = n.s. for time x treatment), canrenoate was accompanied with a persistently increased cortisol concentration throughout the whole experiment (P < 0.05 for the condition of treatment). This effect was observed for all experiment periods (rest period before and after the exercise and for the exercise period itself). There was no significant interaction between the repeated-measures factor of time and the condition of treatment (Fig. 1
, lower panel).
|
ACTH secretion was significantly stimulated by the treadmill procedure (P < 0.05 for main effect, time). During the exercise, levels increased steadily and a maximum was approached at the end of the treadmill period (t = 120 min). Thereafter, levels immediately declined to rest values. Canrenoate, however, did not affect ACTH concentrations, either during the rest period or during the stimulatory period (Fig. 1
, middle panel). The peak-to-baseline difference between maximum ACTH concentrations and concentrations directly before exercise for the canrenoate and placebo-treated group was
= 3.22 ± 1.15 pmol/liter and 3.78 ± 0.96 pmol/liter (P = n.s.), respectively.
hGH
Treadmill exercise was a strong stimulus for hGH secretion, and serum levels were significantly increased (P < 0.001 for main effect of time). Maximum hGH levels were observed at the end of the treadmill period and declined immediately during the subsequent rest period. Canrenoate had no additional effect on hGH concentrations, either during rest or on the stress-induced secretory peak (Fig. 1
, upper panel).
| Discussion |
|---|
|
|
|---|
To stimulate the release of hGH, ACTH, and cortisol, volunteers performed a treadmill exercise for 1 h. The exercise was submaximal according to criteria of the literature (25, 26). Accordingly, heart rate remained stable during the whole 60-min period. The chosen protocol induced a clear hormonal response both affecting the somatotropic and the corticotropic axis. Hormone concentrations were steadily increasing toward the end of the exercise when they approached a maximum for ACTH and hGH, whereas the cortisol concentration continued to rise for a further 20 min.
The use of a physical stressor for the evaluation of MR-mediated effects on the HPA axis adds to the knowledge derived from stimulating pituitary-adrenal activity by exogenous activation through hypothalamic secretagogues (7, 14). Those tests primarily assess the secretory capacity of the stimulated pituitary gland and the administered supraphysiological stimulatory dose typically overrides inhibitory influences on HPA activity. Like other physiological maneuvers, e.g. eating (27) or psychological stressors (18, 28), the hormonal secretion during exercise is most likely initiated via higher centers in hypothalamic and limbic regions (15). Thus, these maneuvers are suitable to evaluate drug effects on this supraordinate regulation of the HPA activity. Furthermore, exercise allows for stimulation of hormonal secretion during an extended period of time. Principally, feedback restraints on cortisol secretion are more likely to be detected when adaptation to a successive increase in cortisol levels occurs continuously during a prolonged phase of stimulation. Thus, a 1-h physical exercise enables a particularly sensitive assessment of feedback influences on HPA secretory activity. The fact that the elevation of cortisol levels were enhanced after canrenoate, regardless of whether the subjects rested or exercised, points to a MR blockade in the hypothalamic or hippocampal region that generally lowers tonic feedback inhibition in this system. Animal experiments have also shown that limbic areas are of particular importance in the mediation of tonic inhibitory signals via MRs on corticosterone secretion (9).
Increased cortisol levels after canrenoate have been observed during the early hours of nighttime sleep (13), and it was hypothesized that MRs mediate the central nervous inhibition of the HPA axis during this phase. A subsequent study demonstrated that the inhibition of the HPA axis during nocturnal sleep can also be observed after stimulation of the HPA axis with exogenous CRH and vasopressin and again is reversed after MR blockade with canrenoate (14). Most recently, Arvat et al. (7) reported that cortisol secretion in response to exogenous CRH and vasopressin is enhanced after administration of canrenoate. Additionally, Grottoli et al. (8) could demonstrate that the stimulatory effect of canrenoate on cortisol and ACTH secretion is abolished by alprazolam, which suggests that the HPA axis is also under control of the secretion of
-aminobutyric acid. In combination, these studies show that MRs play an important role in the regulation of cortisol secretion and that their blockade stimulates HPA secretory activity.
ACTH concentration was not significantly affected by the canrenoate treatment in our study (Fig. 1
, middle panel). This is in accordance with the findings by Arvat et al. (7), who stimulated with CRH at the pituitary level. However, it could be argued that the presence of significantly increased cortisol levels throughout the experiment period would reinforce feedback inhibition, thereby leading to a stronger suppression of ACTH secretion (29). The lack of change in ACTH thus supports the view that MR activation adjusts the set point of cortisol feedback to a lower level (22). Vice versa, blockade of MRs elevates this set point. It does not, however, change the secretory capacity of the pituitary or the adrenocortex, which explains our observations that the baseline-to-peak levels of cortisol and ACTH in response to exercise were identical in both treatment conditions, despite the elevated cortisol level before the stressor.
Alternative explanations for a cortisol increase in the presence of an unchanged ACTH might derive from data regarding an additional non-ACTH-mediated regulation of the adrenal cortex. In situations of chronic or subacute stress, for example, ACTH levels do not regularly respond to the chronically elevated concentrations of glucocorticoids (30). Among several factors that cause a non-ACTH mediated rise in cortisol, the autonomous nervous system seems to have an important influence on adrenal steroid release because it has been shown that, for example, splanchnic nerve stimulation enhances glucocorticoid production, probably by increasing the sensitivity of the adrenal cortex to ACTH (31, 32). However, the role of MRs in this complex interaction of the HPA axis and the autonomous nervous system is unclear, and presently there is no evidence that canrenoate can influence this interaction.
Furthermore, we cannot rule out that alterations in HPA activity, secondary to electrolyte disturbances, could have influenced our results because chronic treatment with canrenoate can lead to hyperkalemia in patients with renal or cardiac failure, especially when combined with angiotensin-converting enzyme inhibitors (33). However, our experiments were performed in healthy individuals who did not take any further medications. We did not control for serum potassium before and 8 h after canrenoate iv (immediately before the treadmill exercise), but a change in serum potassium has not been reported in previous studies using similar doses of canrenoate in healthy subjects (7, 8). However, in patients with congestive heart failure, a single bolus of 200 mg canrenoate caused only a slight elevation (0.02 mmol/liter) in serum potassium (34), and it seems unlikely that such a small increase would lead to alterations of the stimulated HPA activity 8 h later.
Also, we cannot completely rule out GR antagonistic activity of canrenoate due to the high dose that was chosen for the experiment. Theoretically, a partial additional blockade of GRs could have dampened the GR-mediated feedback inhibition and thus might have contributed to our findings. We are, however, not aware of any study describing a cross-reactivity of canrenoate at any dose on GRs. Finally, a disturbed metabolism of cortisol by canrenoate seems less likely because it has been ruled out before (35) and should, anyway, have reduced ACTH secretion in our study.
Our findings underline the importance of MRs in the tonic regulation of the HPA axis and could explain previous observations regarding cortisol secretion in response to physical exercise. Those studies indicated that HPA activation in this context is controlled only partially by a glucocorticoid feedback because the dexamethasone that in vivo binds preferentially to GRs does not reliably suppress cortisol secretion in each individual (16, 36). One could hypothesize that in those subjects who do not respond to GR activation, MRs predominate HPA regulation.
Mineralocorticoid antagonists are frequently used in the treatment of heart failure in humans (37). According to our results and those of other groups, this treatment can activate the HPA axis, which, in turn, could result in the development of weight gain (38), depression, and anxiety (39, 40) or impairment of memory and learning (41). Together this could be of major consequence for the treated patients but has, so far, not been reported for the low doses used in these conditions. The dosage, for example, used in the Randomized Aldactone Evaluation Study was 2550 mg and thus, substantially lower than the one we used in our study (37). Furthermore, it is not clear whether the effects of canrenoate observed here will sustain for longer periods. Nevertheless, results from stimulation tests to assess adrenal-pituitary function should be interpreted cautiously in patients treated with MR antagonists. Likewise, dexamethasone suppression tests might be altered in the presence of MR antagonist treatment.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received January 19, 2004.
Accepted July 8, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Peters, M. Conrad, C. Hubold, U. Schweiger, B. Fischer, and H. L. Fehm The principle of homeostasis in the hypothalamus-pituitary-adrenal system: new insight from positive feedback Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2007; 293(1): R83 - R98. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Rozeboom, H. Akil, and A. F. Seasholtz Mineralocorticoid receptor overexpression in forebrain decreases anxiety-like behavior and alters the stress response in mice PNAS, March 13, 2007; 104(11): 4688 - 4693. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. B. Goldman, G. Wood, M. B. Goldman, M. Gavin, S. Paul, S. Zaheer, G. Fayyaz, and R. S. Pilla Diminished Glucocorticoid Negative Feedback in Polydipsic Hyponatremic Schizophrenic Patients J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 698 - 704. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. H. DeRijk, S. Wust, O. C. Meijer, M.-C. Zennaro, I. S. Federenko, D. H. Hellhammer, G. Giacchetti, E. Vreugdenhil, F. G. Zitman, and E. R. de Kloet A Common Polymorphism in the Mineralocorticoid Receptor Modulates Stress Responsiveness J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 5083 - 5089. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Giordano, M. Bo, M. Pellegrino, M. Vezzari, M. Baldi, A. Picu, M. Balbo, L. Bonelli, G. Migliaretti, E. Ghigo, et al. Hypothalamus-Pituitary-Adrenal Hyperactivity in Human Aging Is Partially Refractory to Stimulation by Mineralocorticoid Receptor Blockade J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5656 - 5662. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |