Journal of Applied Physiology
Vol. 82, No. 6, pp. 1979-1988, June 1997
METABOLISM

Marked differences in functioning of the hypothalamicpituitary-adrenal axis between groups of men

John S. Petrides¹, Philip W. Gold⁴, Gregory P. Mueller², Anita Singh¹, Costas Stratakis³, George P. Chrousos³, and Patricia A. Deuster¹

¹ Department of Military and Emergency Medicine and ² Department of Physiology, Uniformed Services University of the Health Sciences, and ³ Developmental Endocrinology Branch, National Institute of Child Health and Development, and ⁴ Clinical Neuroendocrinology Branch, National Institute of Mental Health, Bethesda, Maryland 20814

ABSTRACT

INTRODUCTION

METHODS AND DESIGN

RESULTS

DISCUSSION

FOOTNOTES

REFERENCES

ABSTRACT

Petrides, John S., Philip W. Gold, Gregory P. Mueller, Anita Singh, Costas Stratakis, George P. Chrousos, and Patricia A. Deuster. Marked differences in functioning of the hypothalamic-pituitary-adrenal axis between groups of men. J. Appl. Physiol. 82(6): 1979-1988, 1997.To compare profiles of hypothalamic-pituitary-adrenal (HPA) responsiveness, healthy, moderately trained men (n = 15) were classified as high (n = 7) or low responders (n = 8) on the basis of plasma adrenocorticotropic hormone (ACTH) responses to strenuous treadmill exercise 4 h after 4 mg of dexamethasone (Dex). These groups were then evaluated to compare 1) HPA and growth hormone responses to exercise at 90% maximal oxygen uptake 4 h after placebo, Dex (4 mg), and hydrocortisone (100 mg); 2) pituitary-adrenal responses to infusion of arginine vasopressin (AVP); 3) plasma cortisol after a Dex suppression test (1 mg); and 4) behavioral characteristics. In comparison to low responders, high responders exhibited significantly 1) higher plasma ACTH responses to exercise after placebo and Dex; 2) higher plasma AVP secretion with exercise after placebo and marked Dex- and hydrocortisone-induced enhancement of exercise-induced AVP secretion; 3) lower Dex-induced increases in basal and stimulated growth hormone secretion; 4) higher plasma ACTH responses to infusion of AVP; and 5) a trend (P = 0.09) for higher trait anxiety ratings. Similar suppression of plasma cortisol was noted after 1 mg Dex. We conclude that subgroups of healthy male volunteers exhibit unique profiles of HPA responsiveness. We also believe that glucocorticoid pretreatment combined with strenuous exercise allows functional HPA responsiveness to be distinguished between subgroups of healthy controls and may be useful in the determination of susceptibility to disorders characterized by hyper- and hypo-HPA activation.

stress; adrenocorticotropic hormone; cortisol; dexamethasone and hydrocortisone

INTRODUCTION

THE HYPOTHALAMIC-PITUITARY-ADRENAL (HPA) axis is the principal neuroendocrine system responsible for physiological adaptation to "stress." However, like any adaptive system, sustained alterations in the regulation of HPA function can predispose individuals to many illnesses, ranging from inflammatory disease to major depression. Recent data suggest that one characteristic of major depression and other psychological illnesses is hyperarousal of hypothalamic neurons, which secrete corticotropin-releasing hormone (CRH); such a state appears to result from an escape of counterregulatory influences (13, 14). Conversely, susceptibility to inflammatory disease has been linked to hyporesponsiveness of hypothalamic CRH neurons to inflammatory mediators, with a resultant decrease in glucocorticoid-mediated restraint of the immune response (30). Because the HPA axis is highly sensitive to both internal and external stimuli, it has proven difficult to ascertain the relevance of large interindividual variations in HPA responsiveness in overall health and disease. Accordingly, functional assessment of the HPA axis would be improved by the use of a natural stressor that causes dose-dependent activation of the HPA axis and is independent of the subject's prior experience.

Graded treadmill exercise is such a stressor, in that pituitary-adrenal responses to relative exercise intensities of 50, 70, and 90% maximal oxygen uptake (O_{2 max}) are similar among individuals despite marked differences in physical training (19). A preliminary study from our laboratory showed that moderately trained, healthy men can be segregated into two groups on the basis of their pituitary-adrenal responses to high-intensity exercise after high-dose dexamethasone (Dex) pretreatment (25). Subjects designated as high responders showed persistence of pituitary-adrenal responses to exercise despite pretreatment with 4 mg of Dex while those designated as low responders showed abolishment of exercise-induced pituitary-adrenal responses. Interestingly, high responders also showed significantly higher plasma arginine vasopressin (AVP) responses during exercise compared with low responders.

In the present investigation, we substantially extended our preliminary study by exploring whether 1) healthy controls can be distinguished on the basis of pituitary-adrenal responses to graded treadmill exercise either before or after pretreatment with the type II glucocorticoid-receptor agonist Dex and/or the type I glucocorticoid-receptor agonist hydrocortisone and 2) high and low responders would show a more generalized tendency in their pituitary-adrenal responsiveness to other stimuli, as well as alterations in either plasma AVP or growth hormone (GH). We also determined whether alterations in HPA responsiveness to a variety of stimuli were associated with differences in indexes of behavioral arousal. Ultimately, identification of subgroups of healthy controls characterized by either relatively high or low responsiveness of HPA function may aid in the identification and further characterization of individuals predisposed to the development of disorders involving HPA dysregulation such as depression or inflammatory disease.

METHODS AND DESIGN

Subjects

Protocols

O_{2 MAX} TEST.

1

1

1

1

1

1

1

1

Order of Experimental Procedures

Visit 2: Psychological tests and determination of each subject's

Visit 2: Psychological tests and determination of each subject's O_{2 max}.

Biochemical Assays

Lactate and glucose concentrations were determined in duplicate (YSI analyzer model 27, Yellow Springs Instruments, Yellow Springs, OH). Hb and Hct were determined immediately in triplicate by the cyanomethemoglobin and microcapillary methods, respectively. Relative changes in plasma volume were calculated by using Hb and Hct values from each test, according to the method described by Dill and Costill (11). Plasma ACTH concentrations were assayed by a two-site immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA), and plasma cortisol concentrations were measured by radioimmunoassay (RIA; Diagnostic Products, Los Angeles, CA); plasma GH was measured by Hazelton Laboratories. Plasma AVP concentrations were assayed by RIA, as previously described (26). Intra-assay variation was 7% for ACTH, 5% for cortisol, 7% for AVP, and 5% for GH. All samples from a single subject were analyzed in one assay to eliminate interassay variations.

Statistical Analyses

RESULTS

Screening Test and Subject Characteristics

Table  1.   General characteristics of subjects Low Responders High Responders Age, yr 28 ± 1  30 ± 1  Weight, kg 80.0 ± 3.0  82.0 ± 3.0  Height, cm 178 ± 2  182 ± 2  Maximal O2, l/min 4.05 ± 0.17  3.96 ± 0.15  Plasma cortisol, nmol/l 43.5 ± 4.6  47.3 ± 10.7 Values are means ± SE; n = 15 subjects. Plasma cortisol is for standard dexamethasone suppression test. O2, oxygen uptake.

Basal Responses to Drug Treatments

Table  2.   Basal responses to placebo, dexamethasone, or hydrocortisone pretreatment Low Responders High Responders Placebo Dex Hco Placebo Dex Hco ACTH, pmol/l 3.2 ± 0.6  0.7 ± 0.1* 0.7 ± 0.1* 3.7 ± 0.5  0.8 ± 0.1* 0.8 ± 0.1* Cortisol, µmol/l 0.22 ± 0.04  0.09 ± 0.01* 1.39 ± 0.17* 0.24 ± 0.03  0.08 ± 0.01* 1.49 ± 0.08* AVP, µmol/l 0.32 ± 0.07  0.26 ± 0.01  0.26 ± .01  0.29 ± 0.04  0.28 ± 0.02  0.55 ± 0.15  GH, µg/l 0.79 ± 0.49  2.11 ± 0.71* 1.6 ± 0.51  0.42 ± 0.19  0.76 ± 0.27  1.43 ± 0.88  Lactate, mmol/l 0.9 ± 0.1  1.1 ± 0.1  1.0 ± 0.1  1.1 ± 0.1  1.0 ± 0.1  1.0 ± 0.1  Glucose, mmol/l 91.7 ± 2.0  105.4 ± 2.3* 102.0 ± 2.7* 96.1 ± 2.5  109.6 ± 2.7* 105.9 ± 1.8* Values are means ± SE; n = 15 subjects. Dex, dexamethasone; Hco, hydrocortisone; AVP, arginine vasopressin; GH, growth hormone. * Significantly different from placebo pretreatment, P < 0.05.

Exercise-Induced Endocrine and Metabolic Responses

Table  3.   Physiological parameters during high-intensity intermittent running with dexamethasone, hydrocortisone, or placebo pretreatment Low Responders High Responders Placebo Dex Hco Placebo Dex Hco Heart rate, beats/min 179 ± 4  180 ± 4  181 ± 5  183 ± 3  183 ± 3  184 ± 4  Absolute O2, l/min 3.7 ± 0.2  3.6 ± 0.2  3.5 ± 0.2  3.7 ± 0.1  3.6 ± 0.1  3.6 ± 0.1  Relative O2, %  91 ± 2  90 ± 3  88 ± 3  93 ± 2  90 ± 2  90 ± 2  RER 0.93 ± 0.05  0.89 ± 0.09  0.86 ± 0.09  0.93 ± 0.07  0.92 ± 0.08  0.89 ± 0.09 Values are means ± SE; n = 15 subjects. Relative O2 is percentage of maximal oxygen uptake achieved. RER, respiratory exchange ratio.

Both high and low responders showed significant increases in net integrated AUC and peak plasma ACTH and cortisol in response to high-intensity exercise and placebo pretreatment. These responses were significantly greater in high responders compared with low responders (P < 0.05). Pretreatment of Dex or hydrocortisone significantly suppressed the exercise-induced plasma ACTH response in both groups compared with placebo (P < 0.05; Fig. 2). After treatment with Dex, plasma ACTH responses to high-intensity exercise were significantly increased above baseline in high responders but not in low responders (Fig. 2, left). Both net integrated AUC and peak plasma ACTH responses to exercise after Dex pretreatment were significantly higher in high compared with low responders. Interestingly, high responders showed a trend toward greater suppression of plasma ACTH after hydrocortisone pretreatment compared with Dex (P = 0.07). In contrast, low responders showed a similar suppression of the exercise-induced plasma ACTH response after both Dex and hydrocortisone (Fig. 2, right). Dex and hydrocortisone pretreatments completely suppressed the exercise-induced increase in plasma cortisol in both high and low responders compared with the placebo pretreatment (Fig. 3, right). Because plasma levels of cortisol were similar for both groups after hydrocortisone pretreatment and were unaffected by exercise, potential differences in the absorption and metabolism of drug treatment were discounted.

Both high and low responders exhibited significant exercise-induced increases in plasma AVP. Net integrated AUC and peak plasma AVP responses were significantly greater in high than in low responders (P < 0.05; Fig. 4). Overall, there was a significant positive correlation between net integrated AUC plasma ACTH and AVP responses to high-intensity exercise (r = 0.55, P < 0.03). A significant positive correlation also was found when plasma ACTH and AVP responses were expressed as  (r = 0.67, P < 0.01). Net integrated AUC plasma AVP responses to high-intensity exercise were significantly higher in high responders compared with low responders with Dex and hydrocortisone pretreatments (P < 0.05; Fig. 4, right). In high responders, both Dex and hydrocortisone pretreatment actually increased the exercise-induced plasma AVP response compared with the placebo pretreatment (P < 0.05). This accentuation was not seen in the low responders (Fig. 4, right).

The net integrated AUC GH responses to exercise under all three treatment conditions are presented in Fig. 5. Exercise induced a significant increase in plasma GH concentrations in both high and low responders (P < 0.05). No differences in GH responses were noted between low and high responders after placebo alone (Fig. 5). Dex pretreatment elicited a significantly greater exercise-induced GH response than did placebo in the low responders (P < 0.05). This Dex-induced enhancement of GH concentration during exercise was not observed in high responders. Hydrocortisone pretreatment failed to augment the GH response to high-intensity exercise compared with the placebo pretreatment in either high or low responders (Fig. 5).

Both high and low responders showed significant increases in plasma lactate and glucose responses to exercise after placebo. Moreover, net integrated AUC and peak plasma lactate responses were significantly higher in high compared with low responders with placebo, whereas glucose responses were similar in high and low responders (P < 0.05; Table 4). Net integrated AUC and peak plasma lactate responses to high-intensity exercise were significantly higher in high compared with low responders pretreated with Dex and/or hydrocortisone (P < 0.05). Neither Dex nor hydrocortisone appreciably altered peak lactate responses to exercise in either group. Peak exercise-induced glucose responses were significantly elevated by Dex and hydrocortisone pretreatments compared with the placebo pretreatment and were significantly different between low and high responders (Table 4). In contrast, no significant differences in the net integrated AUC glucose response were observed between high and low responders.

Table  4.   Metabolic responses to exercise and placebo, dexamethasone, or hydrocortisone pretreatment Low Responders High Responders Placebo Dex Hco Placebo Dex Hco Lactate, mmol/l Peak 11.2 ± 0.6  11.8 ± 0.6  11.5 ± 0.7  13.9 ± 0.5 14.7 ± 0.6 13.7 ± 0.7 AUC 260.8 ± 24.8  289.8 ± 22.9  260.7 ± 24.4  384.7 ± 17.1 410.1 ± 28.0 382.4 ± 30.3 Glucose, mmol/l Peak 127.1 ± 3.4  146.4 ± 4.3*, 145.2 ± 6.9*, 138.3 ± 6.0  166.6 ± 6.1*, 160.8 ± 8.8*, AUC 1,459.2 ± 154.1  1,872.1 ± 338.1  2,137.1 ± 504.2  1,665.3 ± 362.9  2,758.9 ± 441.1* 2,963.7 ± 542.7* Values are means ± SE; n = 15 subjects. AUC, area under the curve. * Significantly different from placebo pretreatment, P < 0.05.  Significant difference between low and high responders, P < 0.05.

Responses to AVP Infusion

1

1

1

1

Lactate concentrations rose slightly during the AVP infusion period and then declined after the infusion was terminated (data not shown). Furthermore, the AUC for plasma lactate revealed no significant difference in the lactate responses to AVP stimulation between low (4.35 ± 1.12 mmol/l) and high responders (6.75 ± 2.4 mmol/l). Circulating plasma glucose, in contrast, did increase significantly (P < 0.05) in response to the AVP infusion in both high and low responders. Similar to the ACTH and cortisol responses, all subjects exhibited peak levels of plasma glucose by the end of the drug infusion (time = 60 min). Analysis of the AUC for plasma glucose indicated that the responses were equivalent for both low (1,446 ± 352 mmol/l) and high responders (1,435 ± 202).

Psychological Profiles

Of all the psychological categories evaluated with SCL-90-R, the high responders scored higher in phobic anxiety (high responders, 0.56 ± 0.17; low responders, 0.25 ± 0.13) and lower in interpersonal sensitivity compared with low responders, although these differences were not significant (high responders, 0.11 ± 0.09; low responders, 0.620 ± 0.41). No significant differences in mood were found by using the Bipolar POMS questionnaire under any of the three conditions. Additionally, scores on the Beck Depression Inventory indicated that no subjects had signs of depression.

DISCUSSION

Healthy moderately trained men can be divided into two groups, high and low responders, on the basis of neuroendocrine and metabolic responses to high-intensity treadmill exercise with and without glucocorticoid pretreatment. Compared with low responders, high responders showed the following profile: 1) higher plasma ACTH responses to exercise at an intensity of 90% O_{2 max} after placebo and Dex pretreatments; 2) higher plasma cortisol and AVP responses during exercise after placebo administration; 3) enhanced Dex- and hydrocortisone-induced augmentation of plasma AVP responses during exercise; and 4) significantly higher plasma ACTH responses to an AVP infusion. In contrast, high responders did not exhibit a significant Dex- or hydrocortisone-induced augmentation of either basal or stimulated GH secretion. Through utilization of an identical exercise and Dex pretreatment protocol, the present study confirmed our previous findings in a larger number of subjects and extended them to include pituitary-adrenal responses to continuous AVP infusion, indexes of basal and exercise-induced GH secretion, and assessment of trait anxiety.

It is well known that glucocorticoid negative feedback is mediated through two distinct corticosteroid ligands: type I receptors and type II receptors (5, 9). These receptors exhibit different affinities for cortisol and different tissue distributions. To investigate the possibility that differential sensitivity to glucocorticoid suppression during exercise results from intrinsic differences in the functioning of glucocorticoid type I and type II receptors, hydrocortisone and Dex were used to preferentially elicit type I and type II receptor mechanisms, respectively.

In contrast to significant differences between high and low responders in the post-Dex plasma ACTH responses to exercise, both groups showed similar post-hydrocortisone plasma ACTH responses to exercise at 90% O_{2 max}. Similarly, whereas six of seven high responders escaped Dex suppression during high-intensity exercise, only three of seven high responders escaped suppression by hydrocortisone. Moreover, these three high responders showed considerably lower peak ACTH responses after hydrocortisone pretreatment than after Dex. Although these results could suggest that high responders are more sensitive to hydrocortisone than to Dex, the results are inconclusive regarding potential differential receptor mechanisms in the control of negative feedback during stress within this small subset of individuals. In addition, there is no clear consensus regarding the relative potency of Dex and hydrocortisone in human subjects. Hence, we cannot rule out the possibility that the dose of hydrocortisone used here was more potent than that of Dex.

Despite the apparent specific resistance of the HPA axis in high responders to type II glucocorticoid agonists, high responders showed exaggerated plasma AVP and deficient GH responses after both Dex and hydrocortisone pretreatment. In addition, no differences in basal ACTH and cortisol plasma levels were noted between high and low responders after Dex pretreatment. Furthermore, all high and low responders showed normal suppression of plasma cortisol levels after a standard 1-mg DST. Hence, the present differentiation of high and low responders on the basis of post-Dex exercise-induced cortisol responses would not have emerged on the basis of standard endocrine screening.

Although increases in plasma AVP are not thought to be an inherent component of generalized stress, plasma AVP has been shown to rise consistently in response to hypotension, hemorrhage, and intense exercise (7, 27, 32). In the present study, high responders had markedly greater exercise-induced AVP responses compared with low responders. Moreover, high responders showed an augmented plasma ACTH response to AVP infusion in the morning compared with low responders. It is difficult to ascertain why high responders exhibited enhanced AVP function compared with low responders. Although no differences in changes in plasma volume, nausea, and dizziness were noted between the groups during test procedures, high responders may have had a higher plasma osmolarity, which could have induced an increase in AVP secretion.

We believe that high responders have an enhanced hypothalamic drive for AVP and possibly CRH release. Interestingly, studies in experimental animals have demonstrated that HPA function can be sustained and even enhanced despite profound negative-feedback inhibition by high concentrations of plasma glucocorticoids. In the context of repeated stress, some investigators have speculated that this mechanism involves downregulation of central glucocorticoid receptors, whereas others have demonstrated a paradoxical glucocorticoid-mediated accentuation in HPA responsiveness (1, 8, 21, 33). Whether the increased hypothalamic drive in high responders is due to a possible type II glucocorticoid-receptor resistance, behavioral arousibility, chronic stress, or physiological perceptions of stress remains speculative. Furthermore, while high and low responders exhibited similar conditioning levels based on O_{2 max} values, differences in anaerobic threshold may provide an additional physiological explanation for differential HPA responsiveness and need to be carefully addressed in future studies.

It is well known that although AVP alone is a weak stimulus to ACTH release, it also profoundly potentiates CRH-induced ACTH release (12). Thus, it is theoretically possible that the augmented exercise-induced plasma AVP concentration observed in high responders contributes to their significantly higher pituitary-adrenal activation after both placebo and glucocorticoid pretreatments. It should be noted, however, that postexercise plasma AVP concentrations are much lower than those observed in hypophyseal portal blood and would be insufficient to influence pituitary ACTH secretion directly. This statement is in agreement with the generally accepted concept that AVP secreted from magnocellular neurons, which serve as the primary source of plasma AVP, does not play a physiological role in ACTH release. Rather, the AVP released into hypophyseal portal blood by paraventricular parvocellular neurons is thought to modulate pituitary-adrenal function (2). Despite current lines of thought, recent anatomic data from rat studies suggest that under some circumstances magnocellular AVP could potentially contribute to ACTH release through neuronal pathways and blood supply (3, 15, 24). Thus magnocellular AVP-secreting neurons could potentially influence plasma ACTH secretion. In addition, plasma AVP levels could be an indirect marker for AVP within hypophyseal portal blood emanating from magnocellular neurons.

Kalogeras and colleagues (16) have recently reported a potential mechanism in humans for enhancing pituitary-adrenal function during stressful conditions, which involves a CRH-mediated release of AVP measured in petrosal sinus blood draining the anterior pituitary (16). In the present study, compared with placebo pretreatment, glucocorticoids significantly enhanced the exercise-induced AVP response in high responders. Whether a paradoxical glucocorticoid-mediated enhancement of the magnocellular contribution to hypophyseal portal AVP concentrations proves to be an additional mechanism by which humans maintain robust exercise-induced HPA activation remains to be determined. Moreover, the presence of this mechanism and its apparent accentuation in a group of healthy controls would have implications for the regulation of the HPA axis in both health and illnesses characterized by HPA dysfunction. The clinical availability of a vasopressin antagonist that does not penetrate the blood-brain barrier could potentially clarify these neuroendocrine mechanisms.

Previous studies in healthy controls have shown that acute glucocorticoid administration increases both basal and stimulated GH release (6, 31). The present project documented this phenomenon with exercise-induced GH secretion in low responders, but high responders failed to show either a basal or stress-induced Dex-mediated GH increase. Whether the failure of high responders to respond with increased basal and exercise-induced GH secretion after Dex reflects a type II glucocorticoid resistance, a subtle suppression of GH release by increased glucocorticoid secretion in the context of chronic stress, or other undefined factors remains to be determined.

Interestingly, although not statistically significant (P = 0.09), the higher scores obtained by high responders on the Spielberger Trait Anxiety Scale suggests a concordance between hyperresponsiveness of the HPA axis during the stress of exercise and a bias toward basal behavioral arousal. If borne out in future studies, these data suggest that although the central pathways mediating pituitary-adrenal responses to physical and emotional stressors differ in several respects, there is sufficient overlap to demonstrate concordance between a physical stressor and an index of behavioral arousal.

The separation of healthy controls into high and low responders is not simply a reflection of their differential responsiveness to an acute novel stressor. Six of seven high responders identified from the screen test continued to show exaggerated plasma ACTH responses to exercise at 90% O_{2 max}. In addition, higher plasma ACTH responses to AVP administration were seen in the high responders despite the fact that the exercise and infusion procedures were carried out in each subject at least 6 mo apart. Moreover, multiple-regression analysis that controlled for differences in relative exercise intensities (percentage of O_{2 max}) revealed that exercise-induced peak and AUC AVP responses were significantly different for high responders compared with low responders (P < 0.05). These results support the hypothesis that high responders have inherently exaggerated neuroendocrine responses to exercise, rather than their simply having reached a higher relative exercise intensity, which could trigger a greater neuroendocrine response.

In summary, we have demonstrated that a subgroup of healthy, moderately trained men (high responders) can be identified on the basis of plasma ACTH responses to vigorous exercise after high-dose Dex administration. High responders show a confluence of other neuroendocrine characteristics, including a Dex-induced facilitation of exercise-induced plasma AVP secretion, pituitary hyperresponsiveness to AVP, and an unresponsiveness to glucocorticoid-induced GH secretion. Whereas exaggerated HPA responsiveness in the high responders seems to reflect a confluence of possible relative type II glucocorticoid resistance, enhanced CRH secretion during the stress of exercise and a glucocorticoid-mediated facilitation of stress-induced plasma AVP secretion, the precise mechanisms for their characteristic patterns of neuroendocrine secretion are currently unknown. Further exploration of the Dex-exercise paradigm used in the present study is indicated to further determine whether 1) a bimodal population exists that shows disparate HPA responsiveness to stressful stimuli, 2) these responses are concordant with indexes of behavioral arousal, and 3) these responses either predict those who may be predisposed to illnesses associated with HPA dysfunction or are associated with frank abnormalities in individuals with full-blown depressive or inflammatory disorders.

FOOTNOTES

Address for reprint requests: P. A. Deuster, Dept. of Military and Emergency Medicine, USUHS, 4301 Jones Bridge Rd., Bethesda, MD 20814-4799 (E-mail: pdeuster{at}usuhs.mil).

Received 10 June 1996; accepted in final form 19 February 1997.

REFERENCES

1.	Aguilera, G. Regulation of pituitary ACTH secretion during chronic stress. Front. Neuroendocrinol. 15: 321-350, 1994 . [Medline]
2.	Antoni, F. A. Vasopressinergic control of pituitary adrenocorticotropin secretion comes of age. Front. Neuroendocrinol. 14: 76-122, 1993 . [Medline]
3.	Baertschi, A. J. Portal vascular route from hypophysial stalk/neural lobe to adenohypophysis. Am. J. Physiol. 239 (Regulatory Integrative Comp. Physiol. 8): R463-R469, 1980 .
4.	Beck, A. T., C. H. Ward, and M. Mendelson. Inventory for measuring depression. Arch. Psychol. 4: 53-61, 1961.
5.	Bradbury, M. J., S. F. Akana, and M. F. Dallman. Roles of type I and type II corticosteroid receptors in regulation of basal activity in the hypothalamic-pituitary-adrenal axis during the diurnal trough and the peak: evidence for a nonadditive effect of combined receptor occupation. Endocrinology 134: 1286-1296, 1994 . [Abstract/Free Full Text]
6.	Casanueva, F. F., B. Burguera, C. Muruais, and C. Dieguez. Acute administration of corticoids: a new and peculiar stimulus of growth hormone secretion in man. J. Clin. Endocrinol. Metab. 70: 234-237, 1990 . [Abstract/Free Full Text]
7.	Convertino, V. A., L. C. Keil, E. M. Bernauer, and J. E. Greenleaf. Plasma volume, osmolality, vasopressin, and renin activity during graded exercise in man. J. Appl. Physiol. 50: 123-128, 1981. [Abstract/Free Full Text]
8.	Dallman, M. F. Stress update: adaptation of the hypothalamo-pituitary-adrenal axis to chronic stress. Trends Endocrinol. Metab. 4: 62-69, 1993.
9.	De Kloet, E. R. Brain corticosteroid receptor balance and homeostatic control. Front. Neuroendocrinol. 12: 95-164, 1989.
10.	Derogatis, L. R., and P. A. Cleary. Conformation of the dimensional structure of the SCL-90: a study in construct validation. J. Clin. Psychiatry 33: 981-989, 1977.
11.	Dill, D. B., and D. L. Costill. Calculation of percentage change in volumes of blood, plasma, and red cells in dehydration. J. Appl. Physiol. 37: 247-248, 1979.
12.	Gillies, G. E., E. A. Linton, and P. J. Lowry. Corticotropin-releasing activity of the new CRH is potentiated several times by vasopressin. Nature 299: 355-357, 1982 . [Medline]
13.	Gold, P. W., F. K. Goodwin, and G. P. Chrousos. Clinical and biochemical manifestations of depression: relation to the neurobiology of stress (pt. 1 of 2). N. Engl. J. Med. 319: 348-353, 1988 . [Medline]
14.	Gold, P. W., F. K. Goodwin, and G. P. Chrousos. Clinical and biochemical manifestations of depression: relation to the neurobiology of stress (pt. 2 of 2). N. Engl. J. Med. 319: 413-420, 1988 . [Abstract]
15.	Holmes, M. C., F. A. Antoni, G. Aguilera, and K. J. Catt. Magnocellular axons in passage through the median eminence release vasopressin. Nature 319: 326-329, 1986 . [Medline]
16.	Kalogeras, K. T., L. K. Nieman, T. C. Friedman, J. L. Doppman, G. B. Cutler, G. P. Chrousos, R. L. Wilder, P. W. Gold, and J. A. Yanovski. Inferior petrosal sinus sampling in healthy subjects reveals a unilateral corticotropin-releasing hormone-induced arginine vasopressin release associated with ipsilateral adrenocorticotropin secretion. J. Clin. Invest. 97: 2045-2050, 1996 . [Medline]
17.	Kyle, S. B., B. L. Smoak, L. W. Douglass, and P. A. Deuster. Variability of responses across training levels to maximal treadmill exercise. J. Appl. Physiol. 67: 160-165, 1989 [Abstract/Free Full Text] .
18.	Lorr, M., and D. M. McNair. Profile of Mood States Manual: Bi-Polar Form (POMS-BI). San Diego, CA: Educational and Industrial Testing Service, 1988.
19.	Luger, A., P. A. Deuster, P. W. Gold, D. L. Loriaux, and G. P. Chrousos. Hormonal responses to the stress of exercise. Adv. Exp. Med. Biol. 245: 273-280, 1988 . [Medline]
20.	Luger, A., P. A. Deuster, S. B. Kyle, W. T. Gallucci, L. C. Montgomery, P. W. Gold, D. L. Loriaux, and G. P. Chrousos. Acute hypothalamic-pituitary-adrenal responses to the stress of treadmill exercise. Physiologic adaptations to physical training. N. Engl. J. Med. 316: 1309-1315, 1987 . [Abstract]
21.	Makino, S., M. A. Smith, and P. W. Gold. Increased expression of corticotropin-releasing hormone and vasopressin messenger ribonucleic acid (mRNA) in the hypothalamic paraventricular nucleus during repeated stress: association with reduction in glucocorticoid receptor mRNA levels. Endocrinology 136: 3299-3309, 1995 . [Abstract]
22.	Meikle, A. W., and F. H. Tyler. Potency and duration of glucocorticoids: effects of hydrocortisone, prednisone and dexamethasone on human pituitary-adrenal function. Am. J. Med. 63: 200-207, 1977 . [Medline]
23.	Michelson, D., L. Stone, E. Galliven, M. A. Magiakou, G. P. Chrousos, E. M. Sternberg, and P. W. Gold. Multiple sclerosis is associated with alterations in hypothalamic-pituitary-adrenal axis function. J. Clin. Endocrinol. Metab. 79: 848-853, 1994 . [Abstract]
24.	Page, R. B. Directional pituitary blood flow: a microcinephotographic study. Endocrinology 112: 157-165, 1983 . [Abstract/Free Full Text]
25.	Petrides, J. S., G. P. Mueller, K. T. Kalogeros, G. P. Chrousos, P. W. Gold, and P. A. Deuster. Exercise-induced activation of the hypothalamic-pituitary-adrenal axis: marked differences in the sensitivity to glucocorticoid suppression. J. Clin. Endocrinol. Metab. 79: 377-383, 1994 . [Abstract]
26.	Rittmaster, R. S., G. J. Cutler, P. W. Gold, D. D. Brandon, T. Tomai, D. L. Loriaux, and G. P. Chrousos. 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, 1987 . [Abstract/Free Full Text]
27.	Robertson, G. L. The regulation of vasopressin function in health and disease. Recent Prog. Horm. Res. 33: 333-385, 1977.
28.	Sherman, B., B. Pfohl, and G. Winokur. Circadian analysis of plasma cortisol levels before and after dexamethasone administration in depressed patients. Arch. Gen. Psychiatry 41: 271-275, 1984 . [Abstract/Free Full Text]
29.	Spielberger, C. D., R. L. Gorsuch, R. Lushene, P. R. Vagg, and G. A. Jacobs. Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press, 1983.
30.	Sternberg, E. M., R. L. Wilder, P. W. Gold, and G. P. Chrousos. A defect in the central component of the immune systemhypothalamic-pituitary-adrenal axis feedback loop is associated with susceptibility to experimental arthritis and other inflammatory diseases. Ann. N. Y. Acad. Sci. 594: 289-292, 1990 . [Medline]
31.	Thakore, G. D., and T. H. Dinan. Subnormal growth hormone responses to acutely administered dexamethasone in depression. Clin. Endocrinol. 40: 623-627, 1994. [Medline]
32.	Wade, C. E. Response, regulation and actions of vasopressin during exercise: a review. Med. Sci. Sports Exercise 16: 506-511, 1984 . [Medline]
33.	Young, E. A., and H. Akil. Paradoxical effect of corticosteroids on pituitary ACTH/-endorphin release in stressed animals. Psychoneuroendocrinology 13: 317-323, 1988 . [Medline]