- Email: [email protected]
Neuroactive steroid concentrations following metyrapone administration in depressed patients and healthy volunteers
Neuroactive Steroid Concentrations following Metyrapone Administration in Depressed Patients and Healthy Volunteers Rainer Rupprecht, Andreas Stro¨hle, Bettina Hermann, Flavia di Michele, Gianfranco Spalletta, Augusto Pasini, Florian Holsboer, and Elena Romeo Background: There is evidence that treatment with the 11b-hydroxylase inhibitor metyrapone may represent an alternative treatment strategy in major depression. As a consequence of inhibition of cortisol synthesis the overdrive of corticotropin leads to an accumulation of precursor steroids. However, the effects of metyrapone on the concentrations of endogenous neuroactive steroids that modulate ion channels, e.g., the GABAA receptor, have not yet been studied systematically. Methods: Therefore, we quantified the concentrations of an array of neuroactive steroids following administration of 1.5g metyrapone before and after pretreatment with 1 mg dexamethasone in 19 patients suffering from severe depression in comparison to 13 healthy controls by means of a highly sensitive gas chromatography/mass spectrometry analysis. Results: The administration of metyrapone induced a pronouced increase in all neuroactive steroids studied both in patients and controls that was prevented by dexamethasone pretreatment. Conclusions: Thus, the psychotropic properties of endogenous neuroactive steroids may contribute to the antidepressant properties of metyrapone in the treatment of major depression. Biol Psychiatry 1998;44:912–914 © 1998 Society of Biological Psychiatry Key Words: Neuroactive steroids, metyrapone, depression, antidepressants, treatment
by which metyrapone exerts its antidepressant effects remain unclear at present. It has been shown that acute administration of metyrapone can elevate glucocorticoid receptor density in lymphocytes in control subjects but not in patients with major depression (Rupprecht et al 1991). Moreover, a preliminary report on urinary steroid profiles of 6 patients following metyrapone treatment suggested that an increase in 11-deoxycortisol metabolites in addition to the suppression of cortisol synthesis may contribute to its psychotropic properties (Raven et al 1996). Certain precursors and metabolites of progesterone and deoxycorticosterone, respectively, modulate neuronal excitability through both genomic mechanisms and nongenomic interactions with neurotransmitter receptors (Paul and Purdy 1992; Rupprecht 1997). In animal studies and initial clinical investigations these neuroactive steroids displayed a variety of psychotropic effects, including sleep modulating, memory enhancing, and anxiolytic properties (Paul and Purdy 1992; Romeo et al 1994; Rupprecht 1997); however, the effects of metyrapone on the concentrations of neuroactive steroids have not up to now been studied. Therefore, we employed a highly sensitive gas chromatography–mass spectrometry method (Romeo et al 1994) to investigate the effects of a single dose of metyrapone, with and without dexamethasone pretreatment, on plasma concentrations of an array of neuroactive steroids in patients suffering from major depression and in healthy controls.
Introduction
S
uppression of cortisol synthesis using the 11b-hydroxylase inhibitor metyrapone is discussed as an alternative treatment strategy in major depressive disorder (Murphy 1991; O’Dwyer et al 1995); however, the mechanisms From the Max Planck Institute of Psychiatry, Clinical Institute, Munich, Germany (RR, AS, BH, FH); and Universita Tor Vergata, Department of Experimental Medicine, Rome, Italy (FdM, GS, AP, ER). Address reprint requests to Dr. R. Rupprecht, Max Planck Institute of Psychiatry, Clinical Institute, Kraepelinstr. 10, 80804 Munich, Germany. Received March 11, 1997; revised May 26, 1997; accepted September 30, 1997.
© 1998 Society of Biological Psychiatry
Methods and Materials Five male and 14 female subjects with unipolar major depressive disorder or bipolar depression according to DSM-III-R aged between 18 and 72 years (mean 6 SD, 50.9 6 15.2 years) were studied. Two men and 11 women aged between 33 and 57 years (mean 6 SD, 47.5 6 7.3 years) served as controls. Severity of depression was assessed by the 21-item Hamilton Depression Rating Scale (HDRS). The mean (6SD) HDRS score of the patients was 26.8 6 4.8, with values ranging from 18 to 36. Patients were studied after a drug-free period of at least 72 hours. 0006-3223/98/$19.00 PII S0006-3223(97)00521-0
Neuroactive Steroids in Depression
913
BIOL PSYCHIATRY 1998;44:912–914
Table 1. Effects of Metyrapone and Metyrapone plus Dexamethasone (DEX) Treatment on the Concentrations (Mean 6 SD) of Neuroactive Steroids and Corticotropin in Patients during Depression and in Healthy Controls Baseline
Corticotropin Pregnenolone Progesterone 11-deoxycortisol 5a-DHDOC 3a,5a-THDOC 5a-DHP 3a,5a-THP 3a,5b-THP 3b,5a-THP DHEA
Metyrapone
Metyrapone plus DEX
Patients
Controls
Patients
Controls
Patients
Controls
3.7 6 1.3 14.4 6 8.5 9.1 6 5.8 4.4 6 2.2 1.2 6 0.6 1.1 6 0.9 2.9 6 1.1 2.2 6 1.0 1.6 6 1.2 1.0 6 0.7 16.8 6 6.2
3.5 6 1.1 23.9 6 9.6 11.9 6 6.5 4.5 6 2.0 0.3 6 0.2 0.2 6 0.1 3.7 6 2.3 2.9 6 1.9 3.4 6 4.2 0.8 6 0.6 15.9 6 5.7
20.4 6 16.8 39.7 6 38.0 17.8 6 10.9 240.2 6 140.3 2.4 6 1.4 2.0 6 1.2 5.3 6 2.1 4.7 6 2.9 3.4 6 2.1 1.8 6 1.6 41.0 6 18.1
25.3 6 18.4 80.2 6 74.8 30.6 6 20.2 284.3 6 150.1 1.4 6 1.5 0.7 6 0.7 6.1 6 4.1 5.1 6 3.8 7.3 6 9.2 1.8 6 1.2 55.4 6 37.3
2.9 6 2.0 18.7 6 22.6 9.3 6 6.2 37.3 6 34.2 1.2 6 0.7 0.9 6 0.8 3.1 6 1.3 2.5 6 1.3 2.1 6 1.6 1.1 6 0.8 18.7 6 10.0
1.9 6 1.8 29.0 6 17.0 16.6 6 10.1 23.1 6 31.3 0.8 6 0.9 0.4 6 0.3 4.1 6 2.6 3.2 6 2.1 2.6 6 2.4 1.0 6 0.7 22.1 6 17.1
Further details of clinical and demographic data of the patients and the medication given before the drug-free interval have been given elsewhere (Rupprecht et al 1991). For quantification of neuroactive steroids, plasma samples were drawn at 4:00 PM on 3 consecutive days after obtaining informed consent. On day 1 baseline values were determined. On day 2 at 9:00 AM the subjects were pretreated with 1.5 g metyrapone administered orally with milk to avoid severe gastric symptoms. One milligram dexamethasone was given orally at 11:00 PM. On day 3 administration of 1.5 g metyrapone was performed as on day 2. Neuroactive steroids were quantified using a combined gas chromatography–mass spectrometry after extraction with ethyl acetate and separation by thin layer chromatography. A detailed description of the method is given elsewhere (Romeo et al 1994). Briefly, after extraction with ethyl acetate and separation by thin layer chromatography, the eluate containing 3a,5a-THP (5apregnan-3a-ol-20-one), 3a,5b-THP (5b-pregnan-3a-ol-20-one), 3b5a-THP (5a-pregnan-3b-ol-20-one), 3a,5a-THDOC (5apregnane-3a,21-diol-20-one), 5a-DHP (5a-pregnan-3,20-dione), 5a-DHDOC (5a-pregnan-21-ol-3,20-dione), progesterone, pregnenolone (5-pregnen-3b-ol-20-one), and dehydroepiandrosterone (DHEA) (5-androsten-3b-ol-17-one) was lyophilized and derivatized. The derivatives were analyzed by gas chromatography–mass spectrometry using an HP 5971 mass selective detector coupled to an HP 5890A gas chromatograph. The steroids were assayed in the electron impact mode and the ions at m/z 510, 496, 298, 270, 301, and 343 were selectively monitored. Cortisol, corticotropin, and 11-deoxycortisol were measured by immunoradiometric assays as described previously (Rupprecht et al 1991). As there was a cross-reactivity of our cortisol antibody with 11-deoxycortisol of about 15%, determinations of cortisol following metyrapone administration would not be accurate and were therefore excluded from further analysis. For statistical comparisons of the mean steroid and corticotropin concentrations between the patient and the control group throughout the different treatments, multivariate analysis of variance (MANOVA) with a repeated measurement design was performed. Treatment was the within-subjects factor, and group was the between-subjects factor. When MANOVA indicated significant main or interaction effects, univariate F tests followed
to identify the variables that contributed significantly to these effects. When significant treatment or treatment by group effects occurred, tests with contrasts followed to locate the pairs of treatments with significant differences. The data were transformed by log transformation [x* 5 ln(x)] before analysis to approximate their homogeneity and normal distribution. As a nominal level of significance, alpha 5 .05 was accepted. For keeping the type I error ,0.05, all univariate F tests and tests with contrast were performed at a reduced level of significance (adjusted alpha according to the Bonferroni procedure). Correlations between steroid concentrations and the HDRS score were obtained by Spearman’s rank correlation.
Results MANOVA revealed only a significant treatment effect [F(24,7) 5 76.0, significance of F 5 0.001; Table 1], to which corticotropin and all steroids investigated contributed (univariate F tests, p , .05). There was a significant increase in the concentrations of corticotropin and the steroids studied following metyrapone administration (tests with contrast, p , .05; Table 1), which was prevented by pretreatment with dexamethasone, in both the patient and the control group with the exception of 11-deoxycortisol. This direct precursor of cortisol remained significantly elevated above baseline levels in spite of pretreatment with dexamethasone (tests with contrast, p , .05). Neither group, i.e., patient or control, nor “group by treatment” interaction exerted significant effects on steroid concentrations. Cortisol baseline values were slightly though not significantly elevated in depressed patients (mean 6 SD, 286.9 6 129.7 nmol/L) when compared with healthy controls (mean 6 SD, 212.4 6 77.3 nmol/L). There was a significant negative correlation between DHEA concentrations and the HDRS score (r 5 2.54, p 5 .04). The other neuroactive steroids studied did not correlate with severity of depression.
914
R. Rupprecht et al
BIOL PSYCHIATRY 1998;44:912–914
Discussion The major finding of the present study is that a single dose of metyrapone is sufficient to increase the concentrations of all neuroactive steroids studied, both in healthy controls and in depressed patients. Raven et al (1996) have recently reported an increase in urinary excretion of 11-deoxycorticosteroids including one putative gamma-aminobutyric acid (GABA)ergic neuroactive steroid following treatment of depression with metyrapone, which is in line with our results. Enhanced levels of progesterone and 3a-reduced neuroactive steroids have been shown to exert benzodiazepinelike sleep-modulating and anxiolytic effects (Lancel et al 1996; Rupprecht 1997). On the other hand, pregnenolone and DHEA may have memory-enhancing properties (Rupprecht 1997). A preliminary report suggested an improvement of depressive symptoms and cognitive performance after oral administration of DHEA (Wolkowitz et al 1997). Thus, the increase in neuroactive steroids may contribute to the amelioration of certain clinical symptoms such as anxiety, sleep disturbances, and cognitive deficits during treatment of depression with metyrapone. There is considerable evidence from electrophysiological studies that the 3a-reduced neuroactive steroids 3a,5aTHP, 3a,5b-THP, and 3a,5a-THDOC are potent positive allosteric modulators of the GABAA receptor, whereas 3b,5a-THP, pregnenolone sulfate, and DHEA sulfate act as functional antagonists (Paul and Purdy 1992; Rupprecht 1997). An increase in 3a,5a-THP and 3a,5b-THP with a concomitant decrease in 3b,5a-THP has been reported during treatment of depression with fluoxetine (Romeo et al 1998) and tri- or tetracyclic antidepressants in view of unchanged progesterone, pregnenolone and DHEA concentrations (Romeo et al 1998). Thus, in contrast to metyrapone, antidepressants do not enhance the concentrations of GABA-antagonistic steroids that might counteract the psychotropic effects of GABA-agonistic steroids to a certain extent. We could not detect significant differences in the concentrations of neuroactive steroids between patients and controls. Recently, decreased levels of pregnenolone in the cerebrospinal fluid (George et al 1994) and of plasma 3a,5a-THP and 3a,5b-THP (Romeo et al 1998) have been reported in patients during depression; however, the patients in these studies were completely medication-free for at least 2 weeks (George et al 1994) or had never been treated before (Romeo et al 1998), whereas the drug-free interval of the patients presented here ranged from 4 days to several weeks (Rupprecht et al 1991). Thus, the antidepressant therapy prior to the study may have influenced the baseline concentrations of neuroactive steroids.
Because a suppression of hypercortisolemia and an increase in glucocorticoid receptors have also been observed after treatment with various antidepressants (Holsboer and Barden 1996), these mechanisms may also contribute to the antidepressant effects of metyrapone. Nevertheless, the putative antidepressant properties of neuroactive steroids deserve further consideration.
This work was supported by the Gerhard-Hess Program of the Deutsche Forschungsgemeinschaft to R.R. The authors thank A. Yassouridis for valuable statistical advice.
References George MS, Guidotti A, Rubinow D, Pan B, Mikalauskas K, Post RM (1994): CSF neuroactive steroids in affective disorders: Pregnenolone, progesterone and DBI. Biol Psychiatry 35: 775–780. Holsboer F, Barden N (1996): Antidepressants and hypothalamic-pituitary-adrenocortical regulation. Endocr Rev 17:187– 205. Lancel M, Faulhaber J, Holsboer F, Rupprecht R (1996): Progesterone induces changes in sleep EEG comparable to those of agonistic GABAA receptor modulators. Am J Physiol 271:E763–E772. Murphy BEP (1991): Treatment of major depression with steroid suppressive drugs. J Steroid Biochem Mol Biol 39:239 –244. O’Dwyer A-M, Lightman SL, Marks MN, Checkley SA (1995): Treatment of major depression with metyrapone and hydrocortisone. J Affect Disord 33:123–128. Paul SM, Purdy RH (1992): Neuroactive steroids. FASEB J 6:2311–2322. Raven PW, O’Dwyer A-M, Taylor NF, Checkley SA (1996): The relationship between the effects of metyrapone treatment on depressed mood and urinary steroid profiles. Psychoneuroendocrinology 21:277–286. Romeo E, Cheney DL, Zivkovic I, Costa E, Guidotti A (1994): Mitochondrial diazepam-binding inhibitor receptor complex agonists antagonize dizocilpine amnesia: Putative role for allopregnanolone. J Pharmacol Exp Ther 270:89 –96. Romeo E, Strohle A, Spaletta G, di Michele F, Hermann B, Holsboer F, Pasini A, Rupprecht R (1998): Effects of antidepressant treatment on neuroactive steroids in major depression. Am J of Psychiatry 155:910 –913. Rupprecht R (1997): The neuropsychopharmacological potential of neuroactive steroids. J Psychiatr Res 3:297–314. Rupprecht R, Kornhuber J, Wodarz N, Lugauer J, Go¨bel C, Haack D, et al (1991): Disturbed glucocorticoid receptor autoregulation and corticotropin response to dexamethasone in depressives pretreated with metyrapone. Biol Psychiatry 29:1099 –1109. Wolkowitz OM, Reus VI, Roberts E, Manfredi F, Chan T, Raum WJ, et al (1997): Dehydroepiandrosterone (DHEA) treatment of depression. Biol Psychiatry 41:311–318.
by which metyrapone exerts its antidepressant effects remain unclear at present. It has been shown that acute administration of metyrapone can elevate glucocorticoid receptor density in lymphocytes in control subjects but not in patients with major depression (Rupprecht et al 1991). Moreover, a preliminary report on urinary steroid profiles of 6 patients following metyrapone treatment suggested that an increase in 11-deoxycortisol metabolites in addition to the suppression of cortisol synthesis may contribute to its psychotropic properties (Raven et al 1996). Certain precursors and metabolites of progesterone and deoxycorticosterone, respectively, modulate neuronal excitability through both genomic mechanisms and nongenomic interactions with neurotransmitter receptors (Paul and Purdy 1992; Rupprecht 1997). In animal studies and initial clinical investigations these neuroactive steroids displayed a variety of psychotropic effects, including sleep modulating, memory enhancing, and anxiolytic properties (Paul and Purdy 1992; Romeo et al 1994; Rupprecht 1997); however, the effects of metyrapone on the concentrations of neuroactive steroids have not up to now been studied. Therefore, we employed a highly sensitive gas chromatography–mass spectrometry method (Romeo et al 1994) to investigate the effects of a single dose of metyrapone, with and without dexamethasone pretreatment, on plasma concentrations of an array of neuroactive steroids in patients suffering from major depression and in healthy controls.
Introduction
S
uppression of cortisol synthesis using the 11b-hydroxylase inhibitor metyrapone is discussed as an alternative treatment strategy in major depressive disorder (Murphy 1991; O’Dwyer et al 1995); however, the mechanisms From the Max Planck Institute of Psychiatry, Clinical Institute, Munich, Germany (RR, AS, BH, FH); and Universita Tor Vergata, Department of Experimental Medicine, Rome, Italy (FdM, GS, AP, ER). Address reprint requests to Dr. R. Rupprecht, Max Planck Institute of Psychiatry, Clinical Institute, Kraepelinstr. 10, 80804 Munich, Germany. Received March 11, 1997; revised May 26, 1997; accepted September 30, 1997.
© 1998 Society of Biological Psychiatry
Methods and Materials Five male and 14 female subjects with unipolar major depressive disorder or bipolar depression according to DSM-III-R aged between 18 and 72 years (mean 6 SD, 50.9 6 15.2 years) were studied. Two men and 11 women aged between 33 and 57 years (mean 6 SD, 47.5 6 7.3 years) served as controls. Severity of depression was assessed by the 21-item Hamilton Depression Rating Scale (HDRS). The mean (6SD) HDRS score of the patients was 26.8 6 4.8, with values ranging from 18 to 36. Patients were studied after a drug-free period of at least 72 hours. 0006-3223/98/$19.00 PII S0006-3223(97)00521-0
Neuroactive Steroids in Depression
913
BIOL PSYCHIATRY 1998;44:912–914
Table 1. Effects of Metyrapone and Metyrapone plus Dexamethasone (DEX) Treatment on the Concentrations (Mean 6 SD) of Neuroactive Steroids and Corticotropin in Patients during Depression and in Healthy Controls Baseline
Corticotropin Pregnenolone Progesterone 11-deoxycortisol 5a-DHDOC 3a,5a-THDOC 5a-DHP 3a,5a-THP 3a,5b-THP 3b,5a-THP DHEA
Metyrapone
Metyrapone plus DEX
Patients
Controls
Patients
Controls
Patients
Controls
3.7 6 1.3 14.4 6 8.5 9.1 6 5.8 4.4 6 2.2 1.2 6 0.6 1.1 6 0.9 2.9 6 1.1 2.2 6 1.0 1.6 6 1.2 1.0 6 0.7 16.8 6 6.2
3.5 6 1.1 23.9 6 9.6 11.9 6 6.5 4.5 6 2.0 0.3 6 0.2 0.2 6 0.1 3.7 6 2.3 2.9 6 1.9 3.4 6 4.2 0.8 6 0.6 15.9 6 5.7
20.4 6 16.8 39.7 6 38.0 17.8 6 10.9 240.2 6 140.3 2.4 6 1.4 2.0 6 1.2 5.3 6 2.1 4.7 6 2.9 3.4 6 2.1 1.8 6 1.6 41.0 6 18.1
25.3 6 18.4 80.2 6 74.8 30.6 6 20.2 284.3 6 150.1 1.4 6 1.5 0.7 6 0.7 6.1 6 4.1 5.1 6 3.8 7.3 6 9.2 1.8 6 1.2 55.4 6 37.3
2.9 6 2.0 18.7 6 22.6 9.3 6 6.2 37.3 6 34.2 1.2 6 0.7 0.9 6 0.8 3.1 6 1.3 2.5 6 1.3 2.1 6 1.6 1.1 6 0.8 18.7 6 10.0
1.9 6 1.8 29.0 6 17.0 16.6 6 10.1 23.1 6 31.3 0.8 6 0.9 0.4 6 0.3 4.1 6 2.6 3.2 6 2.1 2.6 6 2.4 1.0 6 0.7 22.1 6 17.1
Further details of clinical and demographic data of the patients and the medication given before the drug-free interval have been given elsewhere (Rupprecht et al 1991). For quantification of neuroactive steroids, plasma samples were drawn at 4:00 PM on 3 consecutive days after obtaining informed consent. On day 1 baseline values were determined. On day 2 at 9:00 AM the subjects were pretreated with 1.5 g metyrapone administered orally with milk to avoid severe gastric symptoms. One milligram dexamethasone was given orally at 11:00 PM. On day 3 administration of 1.5 g metyrapone was performed as on day 2. Neuroactive steroids were quantified using a combined gas chromatography–mass spectrometry after extraction with ethyl acetate and separation by thin layer chromatography. A detailed description of the method is given elsewhere (Romeo et al 1994). Briefly, after extraction with ethyl acetate and separation by thin layer chromatography, the eluate containing 3a,5a-THP (5apregnan-3a-ol-20-one), 3a,5b-THP (5b-pregnan-3a-ol-20-one), 3b5a-THP (5a-pregnan-3b-ol-20-one), 3a,5a-THDOC (5apregnane-3a,21-diol-20-one), 5a-DHP (5a-pregnan-3,20-dione), 5a-DHDOC (5a-pregnan-21-ol-3,20-dione), progesterone, pregnenolone (5-pregnen-3b-ol-20-one), and dehydroepiandrosterone (DHEA) (5-androsten-3b-ol-17-one) was lyophilized and derivatized. The derivatives were analyzed by gas chromatography–mass spectrometry using an HP 5971 mass selective detector coupled to an HP 5890A gas chromatograph. The steroids were assayed in the electron impact mode and the ions at m/z 510, 496, 298, 270, 301, and 343 were selectively monitored. Cortisol, corticotropin, and 11-deoxycortisol were measured by immunoradiometric assays as described previously (Rupprecht et al 1991). As there was a cross-reactivity of our cortisol antibody with 11-deoxycortisol of about 15%, determinations of cortisol following metyrapone administration would not be accurate and were therefore excluded from further analysis. For statistical comparisons of the mean steroid and corticotropin concentrations between the patient and the control group throughout the different treatments, multivariate analysis of variance (MANOVA) with a repeated measurement design was performed. Treatment was the within-subjects factor, and group was the between-subjects factor. When MANOVA indicated significant main or interaction effects, univariate F tests followed
to identify the variables that contributed significantly to these effects. When significant treatment or treatment by group effects occurred, tests with contrasts followed to locate the pairs of treatments with significant differences. The data were transformed by log transformation [x* 5 ln(x)] before analysis to approximate their homogeneity and normal distribution. As a nominal level of significance, alpha 5 .05 was accepted. For keeping the type I error ,0.05, all univariate F tests and tests with contrast were performed at a reduced level of significance (adjusted alpha according to the Bonferroni procedure). Correlations between steroid concentrations and the HDRS score were obtained by Spearman’s rank correlation.
Results MANOVA revealed only a significant treatment effect [F(24,7) 5 76.0, significance of F 5 0.001; Table 1], to which corticotropin and all steroids investigated contributed (univariate F tests, p , .05). There was a significant increase in the concentrations of corticotropin and the steroids studied following metyrapone administration (tests with contrast, p , .05; Table 1), which was prevented by pretreatment with dexamethasone, in both the patient and the control group with the exception of 11-deoxycortisol. This direct precursor of cortisol remained significantly elevated above baseline levels in spite of pretreatment with dexamethasone (tests with contrast, p , .05). Neither group, i.e., patient or control, nor “group by treatment” interaction exerted significant effects on steroid concentrations. Cortisol baseline values were slightly though not significantly elevated in depressed patients (mean 6 SD, 286.9 6 129.7 nmol/L) when compared with healthy controls (mean 6 SD, 212.4 6 77.3 nmol/L). There was a significant negative correlation between DHEA concentrations and the HDRS score (r 5 2.54, p 5 .04). The other neuroactive steroids studied did not correlate with severity of depression.
914
R. Rupprecht et al
BIOL PSYCHIATRY 1998;44:912–914
Discussion The major finding of the present study is that a single dose of metyrapone is sufficient to increase the concentrations of all neuroactive steroids studied, both in healthy controls and in depressed patients. Raven et al (1996) have recently reported an increase in urinary excretion of 11-deoxycorticosteroids including one putative gamma-aminobutyric acid (GABA)ergic neuroactive steroid following treatment of depression with metyrapone, which is in line with our results. Enhanced levels of progesterone and 3a-reduced neuroactive steroids have been shown to exert benzodiazepinelike sleep-modulating and anxiolytic effects (Lancel et al 1996; Rupprecht 1997). On the other hand, pregnenolone and DHEA may have memory-enhancing properties (Rupprecht 1997). A preliminary report suggested an improvement of depressive symptoms and cognitive performance after oral administration of DHEA (Wolkowitz et al 1997). Thus, the increase in neuroactive steroids may contribute to the amelioration of certain clinical symptoms such as anxiety, sleep disturbances, and cognitive deficits during treatment of depression with metyrapone. There is considerable evidence from electrophysiological studies that the 3a-reduced neuroactive steroids 3a,5aTHP, 3a,5b-THP, and 3a,5a-THDOC are potent positive allosteric modulators of the GABAA receptor, whereas 3b,5a-THP, pregnenolone sulfate, and DHEA sulfate act as functional antagonists (Paul and Purdy 1992; Rupprecht 1997). An increase in 3a,5a-THP and 3a,5b-THP with a concomitant decrease in 3b,5a-THP has been reported during treatment of depression with fluoxetine (Romeo et al 1998) and tri- or tetracyclic antidepressants in view of unchanged progesterone, pregnenolone and DHEA concentrations (Romeo et al 1998). Thus, in contrast to metyrapone, antidepressants do not enhance the concentrations of GABA-antagonistic steroids that might counteract the psychotropic effects of GABA-agonistic steroids to a certain extent. We could not detect significant differences in the concentrations of neuroactive steroids between patients and controls. Recently, decreased levels of pregnenolone in the cerebrospinal fluid (George et al 1994) and of plasma 3a,5a-THP and 3a,5b-THP (Romeo et al 1998) have been reported in patients during depression; however, the patients in these studies were completely medication-free for at least 2 weeks (George et al 1994) or had never been treated before (Romeo et al 1998), whereas the drug-free interval of the patients presented here ranged from 4 days to several weeks (Rupprecht et al 1991). Thus, the antidepressant therapy prior to the study may have influenced the baseline concentrations of neuroactive steroids.
Because a suppression of hypercortisolemia and an increase in glucocorticoid receptors have also been observed after treatment with various antidepressants (Holsboer and Barden 1996), these mechanisms may also contribute to the antidepressant effects of metyrapone. Nevertheless, the putative antidepressant properties of neuroactive steroids deserve further consideration.
This work was supported by the Gerhard-Hess Program of the Deutsche Forschungsgemeinschaft to R.R. The authors thank A. Yassouridis for valuable statistical advice.
References George MS, Guidotti A, Rubinow D, Pan B, Mikalauskas K, Post RM (1994): CSF neuroactive steroids in affective disorders: Pregnenolone, progesterone and DBI. Biol Psychiatry 35: 775–780. Holsboer F, Barden N (1996): Antidepressants and hypothalamic-pituitary-adrenocortical regulation. Endocr Rev 17:187– 205. Lancel M, Faulhaber J, Holsboer F, Rupprecht R (1996): Progesterone induces changes in sleep EEG comparable to those of agonistic GABAA receptor modulators. Am J Physiol 271:E763–E772. Murphy BEP (1991): Treatment of major depression with steroid suppressive drugs. J Steroid Biochem Mol Biol 39:239 –244. O’Dwyer A-M, Lightman SL, Marks MN, Checkley SA (1995): Treatment of major depression with metyrapone and hydrocortisone. J Affect Disord 33:123–128. Paul SM, Purdy RH (1992): Neuroactive steroids. FASEB J 6:2311–2322. Raven PW, O’Dwyer A-M, Taylor NF, Checkley SA (1996): The relationship between the effects of metyrapone treatment on depressed mood and urinary steroid profiles. Psychoneuroendocrinology 21:277–286. Romeo E, Cheney DL, Zivkovic I, Costa E, Guidotti A (1994): Mitochondrial diazepam-binding inhibitor receptor complex agonists antagonize dizocilpine amnesia: Putative role for allopregnanolone. J Pharmacol Exp Ther 270:89 –96. Romeo E, Strohle A, Spaletta G, di Michele F, Hermann B, Holsboer F, Pasini A, Rupprecht R (1998): Effects of antidepressant treatment on neuroactive steroids in major depression. Am J of Psychiatry 155:910 –913. Rupprecht R (1997): The neuropsychopharmacological potential of neuroactive steroids. J Psychiatr Res 3:297–314. Rupprecht R, Kornhuber J, Wodarz N, Lugauer J, Go¨bel C, Haack D, et al (1991): Disturbed glucocorticoid receptor autoregulation and corticotropin response to dexamethasone in depressives pretreated with metyrapone. Biol Psychiatry 29:1099 –1109. Wolkowitz OM, Reus VI, Roberts E, Manfredi F, Chan T, Raum WJ, et al (1997): Dehydroepiandrosterone (DHEA) treatment of depression. Biol Psychiatry 41:311–318.