| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Internal Medicine I (B.F.-S., W.K., W.B., P.W., H.L.F., A.P.) and Clinical Neuroendocrinology (J.B.), University of Luebeck, D-23538 Luebeck; and Department of Diabetes and Metabolism (W.K.), Klinikum Karlsburg, D-17495 Karlsburg, Germany
Address all correspondence and requests for reprints to: Bernd Fruehwald-Schultes, M.D., Medical University Luebeck Department of Internal Medicine I, Ratzeburger Allee 160, D-23538 Luebeck, Germany. E-mail: fruehwal{at}kfg.mu-luebeck.de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
4.1 mmol/L;
P < 0.05). Our results demonstrate that insulin
acutely stimulates the HPA secretory activity in humans. The pattern
suggests an effect of insulin at both peripheral and central levels of
the HPA axis. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Thirty young, healthy men participated in the experiments. Exclusion criteria were chronic or acute illness, current medication of any kind, smoking, alcohol or drug abuse, adiposity, and diabetes or hypertension in first-degree relatives. Each volunteer gave written informed consent, and the study was approved by the local ethics committee.
Experimental design
Each subject underwent a hyperinsulinemic hypoglycemic clamp experiment and a hyperinsulinemic euglycemic clamp experiment, separated by an interval of at least 4 weeks. The order of conditions was balanced across subjects, and experiments were performed in a single-blind fashion. The 30 subjects were randomly assigned to 2 different groups of 15 subjects each. In the first group, insulin was infused at a rate of 1.5 mU min-1kg-1 during both clamp sessions (low-insulin group), whereas in the second group, insulin was infused at a rate of 15.0 mU min-1 kg-1 during both clamp sessions (high-insulin group). The low-insulin group had a mean age (± SEM) of 26.0 ± 1.0 yr (range, 22–32 yr) and a mean BMI of 22.5 ± 0.6 kg m-2 (18.6–25.5 kg m-2); the high-insulin group had a mean age of 25.4 ± 0.6 yr (23–29 yr), and a mean BMI of 23.2 ± 0.5 kg m-2 (19.9–26.0 kg m-2).
All subjects were requested to abstain from alcohol, not to perform any kind of exhausting physical activity, and to go to bed no later than 2100 h on the day preceding the study. On the day of the study, subjects came to the medical research unit at 0800 h, after an overnight fast of at least 10 h. The experiments took place in a sound-attenuated room with the subjects sitting with their trunk in an almost upright position (about 60 degrees) and their legs in a horizontal position on the bed. A cannula was inserted into a vein on the back of the hand, which was placed in a heated box (50–55 C) to obtain arterialized venous blood. A second cannula was inserted into an antecubital vein of the contralateral arm. Both cannulas were connected to long thin tubes, which enabled blood sampling and adjustment of the rate of dextrose infusion from an adjacent room without the knowledge of the subject. After a 1-h baseline period, insulin (H-insulin, Hoechst Marion Roussel, Inc., Frankfurt, Germany) was infused at a continuous rate of either 1.5 mU min-1kg-1 or 15.0 mU min-1kg-1, respectively, depending upon the group. A 20% dextrose solution was simultaneously infused at a variable rate to control plasma glucose levels. Arterialized blood was drawn at 5-min intervals to measure plasma glucose concentration (Glucose Analyser, Beckman Coulter, Inc., Munich, Germany). In euglycemic clamp sessions, plasma glucose was held stable between 5.0 and 5.5 mmol/L. In hypoglycemic clamp sessions, plasma glucose levels were reduced, in a stepwise manner, to achieve four respective plateaus of 4.1, 3.6, 3.1, and 2.6 mmol/L. Each plateau was maintained for a 45-min period, and the next lower plateau was induced gradually within the next 45 min. Blood samples for determination of plasma or serum levels of insulin, cortisol, and ACTH were collected every 30 min. During high-dose insulin infusion, potassium concentrations were monitored at 30-min intervals, and substitution was given whenever the level fell below 4.0 mmol L-1.
Analytical methods
All blood samples were immediately centrifuged, and the supernatants were stored at -24 C until assay. Serum insulin was measured by RIA (Pharmacia Insulin RIA 100, Pharmacia Diagnostics, Uppsala, Sweden) with an interassay coefficient of variation (CV) < 5.8% and an intraassay CV < 5.4%. Serum cortisol was measured by enzyme-linked immunosorbent assay (Enzymun-Test Cortisol, Roche Molecular Biochemicals Immundiagnostica, Mannhein, Germany) with an interassay CV < 3.0% and an intraassay CV < 4.2%. Plasma ACTH was measured by immunoluminometric assay (LUMI test ACTH, Brahms Diagnostica, Berlin, Germany) with an interassay CV < 12% and an intraassay CV < 8%.
Statistical methods
All values are presented as means ± SEM. Statistical analysis was based on analyses of covariance (ANCOVAs) for repeated measurements, including the factors group (low vs. high insulin) and time (duration of the clamp), with baseline values (0 min) serving as the covariate. The interaction effect of these two factors were termed: group by time. Serum insulin concentrations were compared using the unpaired Student’s t test. A P-value < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 1
presents the levels of all
measurements at baseline and the end of the clamp. Plasma glucose
concentrations did not differ between the high- and the low-insulin
groups during either the euglycemic (Fig. 1A) or the hypoglycemic (Fig. 2A) clamp sessions. Mean serum insulin
concentrations were approximately 40-fold greater in the high- than in
the low-insulin group during the euglycemic clamp sessions (543 ±
34 vs. 24,029 ± 1,595 pmol/L; P <
0.0001; Fig. 1B) and during the hypoglycemic clamp sessions (622
± 32 vs. 23,624 ± 1,587 pmol/L; P <
0.0001; Fig. 2B).
|
|
|
During the euglycemic clamp sessions, plasma ACTH concentrations
in the low-insulin group did not essentially change (Fig. 1C
), whereas
plasma ACTH concentrations in the high-insulin group increased from
4.3 ± 0.3 to 7.6 ± 1.4 pmol/L (group by time,
P < 0.01). Serum cortisol concentrations in the
low-insulin group gradually decreased, whereas serum cortisol
concentrations in the high-insulin group slightly increased (Fig. 1D).
ANCOVA revealed a clear difference between the groups in regard to
serum cortisol, with higher concentrations in the high-insulin group
(effect of group, P < 0.02); this difference increased
with duration of the clamp (group by time, P <
0.005).
During the hypoglycemic clamp sessions, plasma levels of ACTH increased
in a similar pattern in both groups (Fig. 2C; effect of time,
P < 0.0001). Likewise, in both groups, serum levels of
cortisol increased to comparable peak values (Table 1; effect of time,
P < 0.0001). However, separate analyses of the
different hypoglycemic plateaus revealed higher levels of cortisol in
the high- than low-insulin group for the first hypoglycemic plateau,
i.e. 4.1 mmol/L (240 ± 30 vs. 350 ±
30; P < 0.05).
Dependency of cortisol on ACTH
To determine to what extent the changes in cortisol could be reduced to effects of insulin on plasma ACTH levels, ANCOVAs were performed on cortisol levels (at each point in time) with corresponding ACTH concentrations serving as covariates, i.e. adjusting for ACTH levels. Values of ACTH concentrations were log-transformed before this analysis, taking into account the logarithmic relationship between cortisol release after exogenous ACTH administration found in previous studies (18). The present analyses revealed that cortisol levels during the euglycemic and the hypoglycemic clamp sessions were strongly related to ACTH (with ACTH as covariate at each time point, P < 0.001). At most points in time during the euglycemic and the hypoglycemic clamp sessions, differences in serum cortisol between the two groups were completely explained by differences in plasma ACTH (with residual effects of the group being not significant).
However, the differences in serum cortisol concentrations between the
high- and the low-insulin groups during the second hour of the
euglycemic (Fig. 1) and the hypoglycemic (Fig. 2) clamp sessions
occurred without differences in ACTH levels between the groups. At 90
min, e.g. differences in serum cortisol between the two
groups remained significant, even after adjusting for plasma ACTH
(effect of group: euglycemia, P < 0.02; hypoglycemia,
P < 0.05, respectively).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
4.1 mmol/L) of
hypoglycemia. The finding that supraphysiological hyperinsulinemia acutely stimulates HPA secretory activity during euglycemia agrees with previous observations that cortisol concentrations increase from baseline during hyperinsulinemic euglycemic clamp experiments in dogs (19) and in humans (20, 21). However, because those studies had no control condition, they did not allow for a straightforward conclusion as to a causative role of insulin for the increased release of cortisol. Other undefined factors, such as experimental stress, also could have caused the increase in cortisol levels of those studies. Here, different supraphysiological degrees of hyperinsulinemia were induced to investigate the effect of insulin per se on plasma concentrations of ACTH and cortisol. This approach seems to be superior to a comparison of the effects of insulin infusion and fasting without insulin infusion, i.e. placebo condition. During fasting, plasma glucose decreases in conjunction with various neuroendocrine changes (22), which may per se act on HPA activity (23), and thereby prevent a clear-cut interpretation of changes after insulin administration. Such metabolic and neuroendocrine changes are absent or attenuated during supraphysiological hyperinsulinemic euglycemic clamps. Moreover, it must be emphasized, in this context, that in the present experiments both low- and high-insulin groups were subjected to exactly the same experimental conditions, concerning potential stress factors, which excludes a confounding influence by these factors.
Insulin seems to be a minor stimulus for HPA activation, as compared with neuroglucopenia, because an additional stimulatory effect of insulin on ACTH/cortisol release was not detectable during more pronounced hypoglycemia, i.e. neuroglucopenia. The strong stimulatory effect of hypoglycemia may have covered the moderate stimulatory effect of insulin on HPA secretory activity. This view seems to be supported by several foregoing studies investigating the effects of hypoglycemic clamps at different levels of hyperinsulinemia in humans, which provided completely inconsistent results. The cortisol response to hypoglycemia in those studies was found to be attenuated (24), not different (25, 26), and amplified (27, 28) by high-dose insulin infusion, as compared with a low-dose insulin infusion. However, a selective increase in the level of insulin in the blood perfusing the brain has been shown to amplify the cortisol response to mild hypoglycemia (3.2 mmol/L) in dogs (29). This experimental data suggests that insulin also increases HPA secretory response to moderate hypoglycemic-stress. In addition, in a recent study (30), we demonstrated a prolonged stimulatory effect of insulin on HPA secretory activity, preventing the development of hypoglycemia-associated counterregulatory failure.
The magnitude of the stimulatory effect of insulin on HPA activity cannot be determined directly from the present results because we compared effects of two different supraphysiological degrees of hyperinsulinemia. However, in the light of findings indicating a sigmoid dose-response relationship for the effects of insulin on various other physiological parameters (31), one may assume that the effects of insulin on HPA activity will follow a similar dose-response curve. Because serum insulin levels were supraphysiological under both sets of experimental conditions, the differences between the effects of the two different degrees of hyperinsulinemia were expected to be small. To detect small differences, the study included a relatively large number of rather homogeneous subjects. Although the magnitude of insulin effect on HPA activity cannot be defined exactly here, the present results provide evidence that insulin is capable of stimulating HPA activity.
Because the present study did not include a placebo condition (without
insulin infusion), for the above mentioned reasons, it remains unclear
whether the low rate of insulin infusion also had a stimulatory effect
on HPA secretory activity during the euglycemic clamp. However, the
rapid fall in ACTH and cortisol levels, observed during the baseline
period, abruptly stopped 30–60 min after starting the low rate of
insulin infusion (Fig. 1). This finding may hint at an effect of the
low rate of insulin infusion on the diurnal variations of HPA secretory
activity, attenuating or delaying the commonly observed decrease in
ACTH and cortisol levels throughout daytime. Support for this view
derives from a comparison of the present data with those obtained
during a previous study (32) investigating the diurnal plasma cortisol
pattern in healthy male subjects. In this study, cortisol levels
decreased by about 60 nmol/L between 1000 h and 1200 h,
whereas in the present study, cortisol levels decreased only by 20
nmol/L during this time interval, i.e. the first 2 h of
low-rate insulin infusion.
Interestingly, ACTH levels increased during the high-rate insulin
infusion of the euglycemic clamp, whereas cortisol levels essentially
did not change (Table 1). Correspondingly, ACTH levels did not change
during the low rate of insulin infusion of the euglycemic clamp,
whereas cortisol levels decreased. This finding may be explained by the
diurnal variations in adrenal sensitivity to the effects of ACTH, which
have been demonstrated to be highest during the time of greatest
spontaneous HPA secretory activity in rats (33). Furthermore, there is
also a diurnal change in ACTH sensitivity to corticosterone feedback
(34). Together, these mechanisms may likely explain the changes in the
ACTH-to-cortisol relationship observed during the euglycemic
clamps.
Because it is well established that insulin crosses the blood-brain barrier (35), the effect of insulin on ACTH release could be exerted at 3 levels, i.e. the hippocampus, the hypothalamus, and the pituitary. The hippocampus is one of the brain regions with the highest density of insulin receptors (36) and a mediator of negative feedback effects of corticosteroids on HPA activity (37). By inhibiting hippocampal activity (36), insulin may impair this feedback regulation, thereby enhancing ACTH and cortisol release. The hypothalamus is another possible site for insulin to exert its effect on ACTH release. Evidence for this view derives from experiments by Grunstein et al. (38), demonstrating, in rats, that hyperinsulinemia suppresses glucose use in the medial basal hypothalamus, accompanied by an increase in serum corticosterone. Thus, even with normal plasma glucose concentration, hyperinsulinemia may induce neuroglucopenia in the medial basal hypothalamus and counterregulatory hormone release. At the pituitary, insulin may directly influence ACTH release, because insulin receptors have been found widely dispersed throughout the pituitary gland (39).
The present data also suggest that insulin activates cortisol release at peripheral sites independently of its effect on ACTH release. This insulin effect could be localized at the adrenals or at extraadrenal sites. The adrenals are a likely site for such an insulin effect, given that Penhoat and colleagues (40) have found and characterized insulin receptors in bovine adrenal fasciculata cells. Furthermore, they demonstrated that insulin enhances the steroidogenic and cAMP response to ACTH, in a series of in vitro experiments. Based on this background, insulin may also, in vivo, enhance adrenocorticotropic action of ACTH. Therefore, the stimulation of cortisol secretion by insulin does not seem to be entirely independent of ACTH release, but rather reflects an increased sensitivity of the adrenals to the action of ACTH.
At extra adrenal sites, insulin may also increase plasma cortisol concentration, considering that Bujalska et al. (41) recently demonstrated that adipose stromal cells from omental fat have a large capacity to convert inactive cortisone to active cortisol through the expression of 11ß-HSD1 (11ß-hydroxysteroid dehydrogenase isoform 1). In vitro, insulin increased the expression and activity of this enzyme (41). This mechanism of converting cortisone to cortisol could be likewise relevant for the in vivo increase in cortisol concentrations after insulin administration.
In summary, this study provides evidence for an acute stimulatory effect of insulin on HPA secretory activity in humans. This finding may explain, at least in part, the frequently found hyperactivity of the HPA axis in hyperinsulinemic subjects, such as patients with diabetes or abdominal adiposity.
|
Acknowledgments |
|---|
Received November 12, 1998.
Revised April 6, 1999.
Accepted May 17, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. Roelfsema, P. Kok, M. Frolich, A. M. Pereira, and H. Pijl Disordered and Increased Adrenocorticotropin Secretion with Diminished Adrenocorticotropin Potency in Obese in Premenopausal Women J. Clin. Endocrinol. Metab., August 1, 2009; 94(8): 2991 - 2997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Neu, M. L. Laudenslager, and J. Robinson Coregulation in Salivary Cortisol During Maternal Holding of Premature Infants Biol Res Nurs, January 1, 2009; 10(3): 226 - 240. [Abstract] [PDF]
|
|
|
|
 |
|
 K. Jauch-Chara, S. M. Schmid, M. Hallschmid, J. Born, and B. Schultes Altered Neuroendocrine Sleep Architecture in Patients With Type 1 Diabetes Diabetes Care, June 1, 2008; 31(6): 1183 - 1188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Fisher, J. C. Bruning, S. Lannon, and C. R. Kahn Insulin Signaling in the Central Nervous System Is Critical for the Normal Sympathoadrenal Response to Hypoglycemia Diabetes, May 1, 2005; 54(5): 1447 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Chan, K. Inouye, E. Akirav, E. Park, M. C. Riddell, M. Vranic, and S. G. Matthews Insulin Alone Increases Hypothalamo-Pituitary-Adrenal Activity, and Diabetes Lowers Peak Stress Responses Endocrinology, March 1, 2005; 146(3): 1382 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Inouye, O. Chan, J. T. Y. Yue, S. G. Matthews, and M. Vranic Effects of diabetes and recurrent hypoglycemia on the regulation of the sympathoadrenal system and hypothalamo-pituitary-adrenal axis Am J Physiol Endocrinol Metab, February 1, 2005; 288(2): E422 - E429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Sandoval, L. Ping, A. R. Neill, S. Morrey, and S. N. Davis Cortisol Acts Through Central Mechanisms to Blunt Counterregulatory Responses to Hypoglycemia in Conscious Rats Diabetes, September 1, 2003; 52(9): 2198 - 2204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Vicennati, L. Ceroni, L. Gagliardi, A. Gambineri, and R. Pasquali Response of the Hypothalamic-Pituitary-Adrenocortical Axis to High-Protein/Fat and High-Carbohydrate Meals in Women with Different Obesity Phenotypes J. Clin. Endocrinol. Metab., August 1, 2002; 87(8): 3984 - 3988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Chan, S. Chan, K. Inouye, K. Shum, S. G. Matthews, and M. Vranic Diabetes Impairs Hypothalamo-Pituitary-Adrenal (HPA) Responses to Hypoglycemia, and Insulin Treatment Normalizes HPA but not Epinephrine Responses Diabetes, June 1, 2002; 51(6): 1681 - 1689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Inouye, K. Shum, O. Chan, J. Mathoo, S. G. Matthews, and M. Vranic Effects of recurrent hyperinsulinemia with and without hypoglycemia on counterregulation in diabetic rats Am J Physiol Endocrinol Metab, June 1, 2002; 282(6): E1369 - E1379. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Oltmanns, B. Fruehwald-Schultes, W. Kern, J. Born, H. L. Fehm, and A. Peters Hypoglycemia, But Not Insulin, Acutely Decreases LH and T Secretion in Men J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4913 - 4919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Reynolds, B. R. Walker, H. E. Syddall, C. B. Whorwood, P. J. Wood, and D. I. W. Phillips Elevated Plasma Cortisol in Glucose-Intolerant Men: Differences in Responses to Glucose and Habituation to Venepuncture J. Clin. Endocrinol. Metab., March 1, 2001; 86(3): 1149 - 1153. [Abstract] [Full Text]
|
|
|
|
|
|
V. Vicennati and R. Pasquali Abnormalities of the Hypothalamic-Pituitary-Adrenal Axis in Nondepressed Women with Abdominal Obesity and Relations with Insulin Resistance: Evidence for a Central and a Peripheral Alteration J. Clin. Endocrinol. Metab., November 1, 2000; 85(11): 4093 - 4098. [Abstract] [Full Text]
|
|
|
|
|
|
L. Ibáñez, C. Valls, N. Potau, M. V. Marcos, and F. de Zegher Sensitization to Insulin in Adolescent Girls to Normalize Hirsutism, Hyperandrogenism, Oligomenorrhea, Dyslipidemia, and Hyperinsulinism after Precocious Pubarche J. Clin. Endocrinol. Metab., October 1, 2000; 85(10): 3526 - 3530. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
;
dotted line) and the high-insulin group (•;
solid line). Each group consisted of 15 healthy men.
SEMs of serum insulin concentrations were smaller than the
size of symbols.