ABSTRACT. The enzyme 11-Ã-hydroxysteroid dehydrogenase (11-Ã-HSD) reg- ulates glucocorticoid activity by converting cortisol into cortisone and vice versa.
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.
Increased Diurnal Plasma Concentrations of Cortisone in Depressed Patients BETTINA WEBER, SABINA LEWICKA, MICHAEL DEUSCHLE, MICHAEL COLLA, PAUL VECSEI, AND ISABELLA HEUSER Central Institute of Mental Health, J5 (B.W., M.D., M.C., I.H.), 68195 Mannheim, Germany; and Department of Pharmacology, University of Heidelberg, INF 366 (S.L., P.V.), 69120 Heidelberg, Germany ABSTRACT The enzyme 11--hydroxysteroid dehydrogenase (11--HSD) regulates glucocorticoid activity by converting cortisol into cortisone and vice versa. Frequent signs of major depression are elevated concentrations of circulating cortisol and ACTH. However, no information is available about the activity of 11--HSD in this disorder. Therefore, we compared diurnal plasma concentrations of cortisol and cortisone and their ratios, reflecting 11--HSD activity, in 25 severely depressed patients (Hamilton Depression Scale, 29 ⫾ 6; 14 men, 11 women, age 22–77 yr; mean, 47 ⫾ 16) and 30 control persons (20 men, 10 women age 23– 85 yr; mean, 51 ⫾ 19). Cortisol and cortisone were measured at 0900 h, 1100 h, 1300 h, 2000 h, 2200 h, 0100 h, 0300 h, and 0700 h with specific RIAs after extraction. Both cortisol and
M
AJOR DEPRESSION is associated with several endocrine signs, especially hypercortisolemia and increased ACTH release (1). Elevated hypothalamus-pituitaryadrenal system activity is accompanied by a flattened diurnal variation of cortisol and a reduced period of quiescence in depressed patients, suggesting a dysregulation of central glucocorticoid and mineralocorticoid receptor function (2). Hypercortisolemia exerts a wide range of effects on central and peripheral tissues. Typical clinical features of excess glucocorticoids are found in Cushing’s syndrome, similar symptoms could be demonstrated in hypercortisolemic depressed patients: impairment of glucose tolerance (3; Weber, B., U. Schweiger, M. Deuschle, and I. Heuser, submitted for publication) , acceleration of loss of bone mineral density (4), and a decrease in testosterone secretion in male patients (5). The biological activities of glucocorticoids are mediated via specific intracellular receptors (mineralocorticoid receptor and glucocorticoid receptor). The access to these receptors is regulated on different levels: 1) by the circulating concentrations of steroid hormones and corticosteroid-binding globulins; 2) by the density of the two receptor types; and 3) by the enzyme 11--hydroxysteroid dehydrogenase (11-HSD). Two isoforms of this enzyme have been identified, 11--HSD type 1 and type 2, mediating a prereceptor metabolism of ligands by converting cortisol into its inactive form cortisone and vice versa. Received September 2, 1999. Revision received November 30, 1999. Accepted December 6, 1999. Address correspondence and requests for reprints to: Bettina Weber, Klinik fu¨r Psychiatrie und Psychotherapie, Central Institute of Mental Health, J5, Mannheim, Germany 68159.
cortisone concentrations were significantly increased in patients compared with controls (cortisol, 251.7 ⫾ 113.1 vs. 160 ⫾ 96.6 nmol/L; cortisone, 32.8 ⫾ 10.9 vs. 21.9 ⫾ 10.9 nmol/L). The calculated ratios of cortisol to cortisone were similar in controls and patients. Similar to cortisol, the circadian variation of cortisone was flattened in patients with the ratio of maximal cortisone to minimal cortisone being 1.9-fold higher in controls than in patients. There was no genderspecific difference in cortisone values neither in patients nor in controls. We conclude that in major depression increased cortisol is not due, at least partly, to an altered 11--HSD activity or to a decrease in cortisone. (J Clin Endocrinol Metab 85: 1133–1136, 2000)
The adult plasma contains cortisol and cortisone in a ratio of 10:1, and cortisone exerts only a 10% glucocorticoid-like effect (6). In humans cortisone derives from the oxidation of the 11--hydroxyl group of cortisol into 11-ketone by 11-HSD type 2. This type 2 enzyme is active mainly in aldosterone-selective organs such as kidney, colon, and salivary gland to ensure aldosterone specificity for the nonspecific mineralocorticoid receptor, but also in other tissues (e.g. adrenal cortex, placenta, and testes) (7, 8). In contrast, 11--HSD type 1 is expressed ubiquitously and functions bidirectionally depending on tissue-specific expression with predomination as an 11--reductase (9). In the liver, fetal lung and intra-abdominal fat, as well as in the central nervous system, especially in hippocampus, cortex, cerebellum, and pituitary, 11--reductase amplifies the effect of cortisol by regenerating active cortisol from cortisone, whereas in testes and placenta it acts as an 11--dehydrogenase inactivating cortisol into cortisone and, thus, protecting these organs from deleterious effects of the excess of glucocorticoids (8). The ratio of cortisol to cortisone in circulating blood reflects the activity of the enzyme 11--HSD type 1 and type 2. Little is known about mechanisms regulating the expression of 11--HSD. Stimulation of an expression of 11--HSD type 1 in liver and hippocampus was shown in animal studies after chronic treatment with corticosterone and the stress of arthritis (10), whereas chronic restraint stress failed to change the activity of the enzyme (11). ACTH was shown to suppress the activity of 11--HSD type 2 by an indirect saturation mechanism in patients with ectopic ACTH syndrome and healthy probands (12, 13). A
1133
1134
JCE & M • 2000 Vol 85 • No 3
WEBER ET AL.
reduced activity of 11--HSD type 2 could also be demonstrated in depressed women who were hypercortisolemic (14). Thus, the aim of this study was to assess whether hypercortisolemia in depressed patients parallels hypocortisonemia as a result of a dysfunction in 11--HSD activity.
MAX) and minimal cortisone (cortisone MIN) and the ratio of cortisone MAX to cortisone MIN. Postprandial values at 0900 h and 1300 h were excluded from calculation because of the well-known stimulating effect of food intake on cortisol secretion. Additional unpaired t tests were used to compare the cortisol to cortisone ratio at different time points between patients and controls. All results are reported as means ⫾ sd.
Subjects and Methods
Results
Subjects The study was approved by the local ethics committee, and all subjects gave informed written consent. In-patients with major depression were included in this study. Inclusion criteria were: 1) major depression according to DSM-IV (15); 2) at least 18 points on the 21-item Hamilton Depression Scale (16); 3) no history of substance abuse or dependency; 4) no neurological or relevant medical disease; and 5) no psychotropic drugs for at least 7 days prior to the study, except zolpidem given in cases of sleep difficulties. The healthy control group was acquired by newspaper advertisement. A standardized psychiatric interview gave no evidence for an individual or family history of psychiatric disorders. In all subjects a thorough physical examination and routine laboratory tests, including structural magnetic resonance imaging of the brain, electrocardiogram, and electroencephalogram, revealed no sign of physical illness. Twenty-six patients with major depression (11 females, 15 males) and 33 sex-matched healthy controls (11 females, 22 males) of similar age [47 ⫾ 16 yr (sd) vs. 51 ⫾ 19 yr (sd)] and body mass index (23 kg/m2 ⫾ 4 vs. 23 kg/m2 ⫾ 3) participated in the study. Mean Hamilton Depression Scale in patients was 29 ⫾ 6 (sd). All probands were free of glucocorticoid intake within the past year.
Methods Blood sampling. All subjects underwent a blood sampling at 0900 h, 1100 h,1300 h, 2000 h, 2200 h, 0100 h, 0300 h, and 0700 h. Blood was drawn through an indwelling forearm catheter for measurement of cortisol and cortisone concentrations. Between blood samplings, the tubing system was kept patent by saline infusion at a rate of 50 mL/h. Each sample was immediately centrifuged and stored at ⫺20 C for cortisol and cortisone determinations. Subjects remained sedentary in bed; food was given at 0835 h, 1235 h, 1835 h ad libitum; and lights were off at 2300 h. Patients and controls spent the time mostly reading or watching television. Daytime napping was not allowed. Hormone assays. Cortisol was determined with specific RIA after breaking protein binding with absolute ethanol applying the method of Vecsei and Penke (17). Intra-assay variation was less than 5%, interassay variation was less than 7%, and limit of detection was 5 pg. Cortisone was measured with specific RIA after extraction with chloroform. Briefly, to 20 L of serum, 3000 dpm tritiated cortisone (as recovery tracer), 1.5 mL phosphate buffer, and 5 mL chloroform were added. After 1 h of mixing, organic phase was separated, evaporated to dryness, and taken up in 500 L phosphate buffer. For the recovery determination, a 100-L aliquot was measured in a scintillation counter. RIA was carried out in triplicate, 3 ⫻ 100-L aliquots were incubated overnight at 4 C with excess of tritiated cortisone and specific antibody raised to cortisone-3-oxime-bovine serum albumin complex (cross-reactivity with cortisol ⬍0.2%, progesterone and 17-hydroxyprogesterone ⬍0.03%). Intra-assay variation was less than 7%, interassay variation was less than 8%, and limit of detection was 5 pg. Unbound steroids were then precipitated with dextrane-coated charcoal, and antibody-bound radioactivity was measured in a scintillation counter. Each result was corrected with individually determined recovery. Statistical analysis. Repeated measurement ANOVA with diagnosis (depressed patients vs. healthy controls) and gender (female vs. male) as factors and cortisol, as well as cortisone, and the ratio of cortisol to cortisone as dependent variables were used. Furthermore, unpaired t tests were used to compare the cortisol to cortisone ratio at different time points between patients and controls. To describe the circadian pattern of cortisone secretion, data were condensed into the following parameters: maximal cortisone (cortisone
Both cortisol and cortisone concentrations were elevated at all time points in depressed patients compared with healthy controls (Figs. 1 and 2). Repeated measurements of ANOVA revealed a significant impact of diagnosis on the dependent variables cortisone and cortisol (cortisone, F1,29 ⫽ 39.8, P ⬍ 0.002; cortisol, F1,29 ⫽ 12.9, P ⬍ 0.002). The calculated ratios of cortisol to cortisone were similar in patients and controls (P ⬎ 0.05) (Fig. 3). In patients, the circadian variation of cortisone secretion was flattened: cortisone MAX (46.2 nmol/L) was only 2.4fold higher than the cortisone MIN (22.7 nmol/L), in contrast to controls who had a 4.3-fold higher cortisone MAX (33.9 nmol/L) in comparison to cortisone MIN (9 nmol/L; P ⬍ 0.05). There was no effect of gender on hormonal secretion (cortisone, F1,28 ⫽ 0.76, P ⫽ 0.4; cortisol, F1,28 ⫽ 0.27, P ⫽ 0.6). Discussion
The main finding of this study is that circulating concentrations of cortisone parallels increased cortisol in depressed patients compared with healthy controls. This is the first study demonstrating elevated spontaneous cortisone concentrations in major depression, which is in agreement to an earlier study showing elevated cortisone concentrations after stimulation with corticotropin releasing factor and suppression with dexamethasone (18). Second, cortisone profiles in both groups of individuals studied show a circadian rhythm similar to that of cortisol, with highest values in the morning (0700 h) and a nadir in the evening (2200 h). This is in agreement with earlier studies demonstrating a diurnal rhythm of circulating cortisone concentrations in healthy probands (19, 20). Hence, the time points chosen in our study for the quantification of cortisone
FIG. 1. Cortisol concentrations in depressed patients and controls.
INCREASED CORTISONE IN DEPRESSED PATIENTS
FIG. 2. Cortisone concentrations in depressed patients and controls.
FIG. 3. Ratio of cortisol to cortisone concentrations in depressed patients and controls.
were sensitive enough to demonstrate the variation of cortisone over the course of the day. There is only one study that did not find a variation in serum cortisone concentrations (6), but this was probably due to the very small number of healthy probands included (n ⫽ 7). Similar to cortisol, as we have shown in a previous study (2), diurnal variation of cortisone is flattened in depressed patients with a 1.9-fold lower ratio of MAX to MIN than those of controls. This underscores the tight relationship between cortisol and cortisone and points to an unchanged activity of the enzyme 11--HSD, as evidenced by a normal ratio of cortisol to cortisone in depressed patients. In humans, cortisone concentrations in plasma derive from oxidation of cortisol by the 11--HSD type 2, which is mainly located in the kidney and other mineralocorticoid target tissues (7). Cortisol as substrate stimulates the activity of type 2 enzyme, so that an elevation of cortisol secretion is followed by a raise in cortisone production. Time of substrate stimulation seems to be crucial for 11-HSD activity. A decrease in 11--HSD type 2 activity was seen after
1135
infusion of short-acting ACTH for 1 h, resulting in blunted cortisone plasma concentrations. This could not be demonstrated after infusion of long-acting ACTH for 4 h (18). However, in both conditions an increase in cortisol to cortisone ratio was observed (18). Extending these findings, other studies demonstrated that the dampening of the 11--HSD type 2 enzyme activity after ACTH was mainly due to a saturation of this enzyme (12, 13). No information is available on mechanisms that compensate the enzyme saturation of short-acting vs. long-acting ACTH. One might speculate that chronic elevation of cortisol as the substrate increases the initially suppressed activity of the enzyme. Our results are in agreement with such an assumption showing no changes in 11-HSD type 2 activity in severely depressed patients who have chronic elevation of ACTH and cortisol. It has to be mentioned, however, that the serum measurements reflect overall activity of both isoforms of the enzyme. The activity of 11--HSD type 2 can be assessed more accurately from the urinary concentrations of cortisol and cortisone, which we have not measured for the well-known problems with proper collection of urine in depressed patients. Whereas 11--HSD type 2 converts cortisol into cortisone, type 1 mainly regenerates active cortisol from cortisone. Animal data suggest that during stress, but not after exogenous application of equivalent amounts of corticosteroids, the type 1 activity is unchanged (10, 11). This is similar to our results showing no difference in the calculated ratio of cortisol to cortisone in hypercortisolemic depressed patients, reflecting an unchanged activity of the enzyme 11--HSD type 1. In summary, the calculated ratio of serum cortisol to cortisone reflects the overall activity of the enzyme 11--HSD. Our results suggest that the activity of 11--HSD type 1 relative to 11--HSD type 2 has not changed in depressed patients compared with controls. Therefore, it seems unlikely that hypercortisolemia in depression results, at least partly, from an increased 11--reductase (11--HSD type 1) or a decreased 11--dehydrogenase activity (11--HSD type 2). However, one has to take into account that there could be increased activity of both enzymes in depression. Most importantly, however, this is the first study to show that not only cortisol but also cortisone is increased in major depression. This elevation in cortisone concentrations in depressed patients might further increase the pool for exposure of glucocorticoid receptors to cortisol and consequently reinforce deleterious effects of hypercortisolemia, such as loss of bone mineral density (4) and impairment of glucose tolerance (3; Weber, B., U. Schweiger, M. Deuschle, and I. Heuser, submitted for publication) in these patients. References 1. Heuser I. 1998 The hypothalamic-pituitary-adrenal system in depression. Pharmacopsychiatry. 31:10 –13. 2. Deuschle M, Schweiger U, Weber B, et al. 1997 Diurnal activity and pulsatility of the hypothalamus-pituitary-adrenal system in male depressed patients and healthy controls. J Clin Endocrinol Metab. 82:234 –238. 3. Winokur A, Maislin G, Phillips JL, Amsterdam JD. 1988 Insulin resistance after oral glucose tolerance testing in patients with major depression. Am J Psychiatry. 145:325–330. 4. Schweiger U, Deuschle M, Ko¨rner A, et al. 1994 Low lumbar bone mineral density in patients with major depression. Am J Psychiatry. 151:1691–1693. 5. Schweiger U, Deuschle M, Weber B, et al. 1999 Testosterone, gonadotropin and cortisol secretion in male patients with major depression. Psychosomatic Med. 61:292–296.
1136
WEBER ET AL.
6. Walker BR, Campbell JC, Fraser R, Stewart PM, Edwards CRW. 1992 Mineralococorticoid excess and inhibition of 11-hydroxysteroid dehydrogenase in patients with ectopic ACTH syndrome. Clin Endocrinol. 37:483– 492. 7. Mazzocchi G, Rossi GP, Neri G, Malendowicz LK, Albertin G, Nussdorfer GG. 1998 11-Hydroxysteroid dehydrogenase expression and activity in the human adrenal cortex. FASEB J. 12:1533–1539. 8. Monder C, Miroff Y, Marandici A, Hardy MP. 1994 11- hydroxysteroid dehydrogenase alleviates glucocorticoid-mediated inhibition of steroidogenesis in rat leydig cells. Endocrinology. 134:1193–1204. 9. Low SC, Chapman KE, Edwards CRW, Wells T, Robinson IC, Seckl JR. 1994 Liver-type 11-hydroxysteroid dehydrogenase cDNA encodes reductase not dehydrogenase activity in intact mammalian COS-7 cells. J Mol Endocrinol. 13:167–174. 10. Low S, Moisan MP, Noble JM, Edwards CRW, Seckl JR. 1994 Glucocorticoids and chronic stress up-regulate 11-hydroxysteroid dehydrogenase activity and gene expression in the hippocampus. J Neuroendocr. 6:285–290. 11. Jellinck PH, Dhabhar FS, Sakai RR, McEwen BS. 1997 Long-term corticosteroid treatment but not chronic stress affects 11-hydroxysteroid dehydrogenase type I activity in rat brain and peripheral tissues. J Steroid Biochem Mol Biol. 60:319 –323. 12. Ulick S, Wang JZ, Blumenfeld JD, Pickering TG. 1992 Cortisol inactivation overload: a mechanism of mineralocorticoid hypertension in the ectopic adrenocorticotropin syndrome. J Clin Endocrinol Metab. 74:963–967.
JCE & M • 2000 Vol 85 • No 3
13. Diederich S, Quinkler M, Miller K, Heilmann P, Scho¨nesho¨fer M, Oelkers W. 1996 Human kidney 11- hydroxysteroid dehydrogenase: regulation by adrenocorticotropin? Eur J Endocrinol. 134:301–307. 14. Raven PW, Taylor NF. 1998 11 -HSD and 17-HSD as biological markers of depression: sex differences and correlation with symptom severity. Endocr Res. 24:659 – 662. 15. American Psychiatric Association. 1987 Diagnostic and statistical manual of mental disorders, 3rd ed. Washington, DC: American Psychiatric Association. 16. Hamilton M. 1960 A rating scale for depression. J Neurol Neurosurg Psychiatry. 23:56 – 62. 17. Vecsei P, Penke B. 1972 Radioimmunological determination of plasma cortisol. Experientia. 28:1104 –1105. 18. Holsboer F, Muller OA, Doerr HG, et al. 1984 ACTH and multisteroid responses to corticotropin-releasing factor in depressive illness: relationship to multisteroid responses after ACTH stimulation and dexamethasone suppression. Psychoneuroendocrinology. 9:147–160. 19. Morineau G, Boudi A, Barka A, et al. 1997 Radioimmunoassay of cortisone in serum, urine, and saliva to assess the status of the cortisol-cortisone shuttle. Clin Chem. 43:1397–1407. 20. Nomura S, Fujitaka M, Sakura N, Ueda K. 1997 Circadian rhythms in plasma cortisone and cortisol and the cortisone/cortisol ratio. Clin Chem Acta. 266: 83–91.
