Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1 Sex Differences in the ACTH Response to 24H Metyrapone in Depression. Elizabeth A Young MD1, 2 and Saulo C Ribeiro MD1. From The Department of Psychiatry1 and Molecular and Behavioral Neuroscience Institute2 (EAY), the University of Michigan Text word count: 4118 Abstract word count: 199 Corresponding Author: Elizabeth A Young Molecular and Behavioral Neurosciences Institute (MBNI) 205 Zina Pitcher Place Ann Arbor MI 48109-0720 [email protected] Fax: 734-647-4130 2 Abstract Increased hypothalamic-pituitary adrenal (HPA) axis activation has been observed in major depression. Increased corticotropin-releasing hormone (CRH) is hypothesized to drive corticotropin (ACTH) secretion leading to increased ACTH and cortisol secretion throughout 24H. Contradicting data exist as to whether the increased drive is present throughout the day or is restricted to the late afternoon and evening. To determine if increased HPA axis activation occurs during a specific circadian phase or is found throughout the 24H, we studied 26 healthy drug-free depressed patients and 26 healthy age and sex matched control subjects under metyrapone inhibition of cortisol synthesis for 24H beginning at 4PM. Blood was drawn every 10 minutes for 24H and assayed for ACTH and cortisol. Gender differences in response to metyrapone were seen in both patients and controls. Depressed women demonstrated increased ACTH concentrations between 4PM and 10PM compared to control women. Maximal ACTH response over time was identical between depressed and control women. Depressed men demonstrated significantly decreased ACTH secretion between 4-10PM as well as decreased maximal ACTH response compared to control men or depressed women. These data support a circadian phase specific increase in CRH drive in depressed women, but overall decreased CRH drive in depressed men. 3 INTRODUCTION Activation of the hypothalamic pituitary adrenal (HPA) axis is one of the most robust findings in major depression. Studies examining 24H cortisol secretion have found that 25-33% of patients show elevated mean cortisol (Rubin et al, 1987; Halbreich et al, 1985; Pfohl et al, 1985; Linkowski et al, 1985; Young et al, 2001). In addition, multiple studies have demonstrated failure to suppress cortisol to dexamethasone challenge (Arana et al, 1985; Ribeiro et al, 1993; Carroll et al, 1981); this finding has been linked to elevated mean cortisol basally (Poland et al, 1987). Failure to suppress pituitary -endorphin secretion with dexamethasone challenge was also linked to increased basal cortisol secretion (Young et al, 1993). This HPA axis hyperactivity is generally believed to reflect increased central CRH and AVP secretion from the PVN. In support of this hypothesis, increased CRH in the CSF of depressed patients has been demonstrated (Nemeroff et al, 1984); furthermore, post-mortem studies have found increased CRH mRNA levels and increased number of CRH and AVP co-expressing neurons in the PVN of depressed patients (Raadsheer et al, 1994, 1995). Studies examining 24H secretion have generally found cortisol secretion is higher in depressed patients than normal subjects around the clock (Rubin et al 1987; Halbreich et al, 1985; Young et al, 2001.). Other studies have suggested that the late afternoon is a particularly sensitive time to detect cortisol increases (Halbreich et al, 1982). Early studies by Sachar et al (1973) found late evening to show the maximum difference between depressed and control patients in cortisol secretion. Likewise, late afternoon has been found to be the most sensitive time to detect non-suppression of cortisol or -endorphin to dexamethasone 4 (Carroll et al, 1981; Young et al, 1993). We have found that administration of metyrapone, a 11- hydroxylase inhibitor of cortisol synthesis, in the period between 4PM and 10PM, leads to increased -endorphin and ACTH secretion in depressed patients compared to matched controls (Young et al, 1994, 2004) while administration of metyrapone in the morning (8AM to 2PM) leads to increased ACTH secretion in both patients and normal subjects equally (Young et al, 1997). Consequently, our previous metyrapone studies suggest a specific circadian time of increased activation of basal secretion – 4PM and 10PM (Young et al, 1994, 1997, 2004). However, our own 24H studies undertaken in depressed women under basal condition suggested increased ACTH and cortisol were found around the clock (Young et al, 2001). Our previous studies of morning and evening metyrapone blockade were conducted in separate groups of patients and controls, and we did not assess overnight ACTH secretion with metyrapone blockade of cortisol synthesis. Thus, the current studies were undertaken to test the following alternative hypotheses: 1) that while blocking cortisol synthesis and feedback with metyrapone, we would observe increased ACTH secretion (reflecting increased CRH and AVP drive) across the 24H or 2) that under the influence of metyrapone, we would observed increased secretion of ACTH at a specific circadian phase only – from 4PM until 10PM in the evening. METHODS Subjects: We studied 52 subjects in total, 26 depressed patients and 26 age and sex matched control subjects. All subjects were recruited by advertising and were untreated for the current episode of depression at the time of recruitment. After initial telephone screening, patients were administered the Structured Clinical Interview for DSM-IV by 5 our trained research nurse and all met criteria for current major depression as determined by the SCID and confirmation by a psychiatrist (EAY or SCR). All subjects were physically healthy as determined by screening blood work and physical exam and were on no other medications. Four depressed women and four control women were post menopausal. The remaining women were studied at random phases of the menstrual cycle, since our previous studies comparing ACTH response to metyrapone in control women at follicular and luteal phase showed no menstrual cycle effect (Altemus et al., in press). By cycle day estimates, 4 patients were studied in the follicular phase and 9 in the luteal phase; 4 controls were studied in the follicular phase, 5 in the luteal phase and for 4 controls cycle phase is unknown. All subjects were untreated for their current episode of depression and on no medications of any type including herbal supplements. No minimum severity level was required other than meeting MDD by SCID. All subjects signed an informed consent and were free to withdraw at any time. Procedure: The study was approved by the University of Michigan Health system IRB. All subjects were admitted to the General Clinical Research Center at 2:30PM and given a snack at this time. An intravenous catheter for blood drawing was inserted at 3:00 and the first dose of metyrapone was administered at 4:00PM immediately following the first blood sample. The starting time of 4PM was chosen to replicate our previous study of increased ACTH response to metyrapone in depressed women (Young et al, 1004), and to determine if this increased drive continued throughout the 24 hours. Blood was drawn every 10 minutes for the next 24 hours (145 samples total). Metyrapone, 750 mg, was administered every 4H from 4PM through noon the following day. Standardized meals were given at 7:30PM, 7:30 AM, 11:30 AM and a snack at 3:30PM. The midnight and 6 4AM doses of metyrapone were given with milk. All samples were drawn via syringe and placed in polypropylene tubes with EDTA, then placed on ice until centrifuged and separated, a maximum of 2 hours after collection. Following centrifugation, plasma was separated and frozen on dry ice. Samples were stored at -80 C until assay. Hormone Assay: All samples were assayed for ACTH and cortisol with the Diagnostic Products Corporation (Los Angeles, CA) Immulite chemiluminescent based system. All samples for each subject were run in one assay, and the matched pairs were assayed in adjacent runs with the same assay kits. Intrassay variability was 4.8% for ACTH and 5% for cortisol. Interassay variability for ACTH was 7% and 8% for low (30pg/ml) and high (430pg/m) ACTH samples respectively; interassay variability was 7% and 9% for low (4µg/dl) and high (34µg/dl) cortisol samples across assays. Minimum detection limit for ACTH was 10 pg/ml and for cortisol was 1µg/dl. Quiescent period onset and offset, secretory onset and circadian maximum; The 24H profile of ACTH secretion did not follow the classical circadian rhythm since ACTH levels were still greatly elevated at the end of the study (4PM, day2), as reported previously (Veldhuis et al, 2001). To estimate periods of nocturnal quiescence and the onset of the circadian secretory surge we adapted the procedures of Linkowski et al (1985). We referenced all values to the mean plasma ACTH concentration from 1600h to 0400h in each individual and periods of at least 60 minutes duration in which all values were below this reference mean were designated quiescent periods. Onset of the circadian secretory ACTH surge was identified as the time from which all samples for at least a one hour block were above the reference mean. Ultimately, this algorithm defined the point in which all subsequent samples were above the reference mean for the duration of the 7 study. Circadian maximum was defined using a moving average of 3 samples as the time of the highest ACTH level. (Rubin et al, 1987) Statistical analyses: Data were analyzed using a repeated measure analysis of variance (RM-ANOVA) with time as one factor and group as a second factor. Because a significant sex difference in response to metyrapone was observed in both control subjects and depressed patients, data for men and women were also analyzed separately. Some samples, particularly at the trough of the circadian rhythm, were below the detection limit for ACTH assay (10pg/ml). These samples were set to the assay limit (10 pg/ml) for analysis; since average peak levels under metyrapone blockade were greater than 400 pg/ml, the relative contribution of these extremely low levels was small. Likewise, cortisol levels reading below 1µg/dl were set to 1. Cortisol was assayed to assure that escape from metyrapone blockade did not occur. For the purpose of determining adequate blockade of cortisol levels, actual values below 1µg/dl were not needed. Finally, in order to include all subjects in the RM-ANOVA , missing values were interpolated by taking the mean of the two adjacent samples for any subject with missing values. Missing values were les than 1% of all samples. RESULTS Mean patient and control demographic data are presented in Table 1. As would be expected by disease prevalence twice as many of the patients and controls were women. The patients were moderately depressed as measured by the 17 item Hamilton Depression Rating Scale (HDRS). Thirteen (9F, 4M) of the 26 patients met DSM-IV criteria for melancholia and 14 patients (11F, 3M) were recurrent, while 12 patients (6F, 6M) were 8 first episode. In two subjects, dysthymia preceded the onset of major depression. Four patients had a history of anxiety disorders on the SCID (PTSD in two; social anxiety disorder in three; two with panic disorder). In addition, five patients had a past history of alcohol abuse which resolved 2, 8, 10, 17 and 18 years earlier. The mean episode length was 15 ±18 months with a range of 1 month to 5 years. 15 of the patients were depressed longer than 1 year. Figure 1 shows the cortisol data for the patients and control subjects. The y-axis is set to the usual circadian maximum, 20µg/dl. As can be observed, metyrapone successfully blocked cortisol secretion to approximately 2 µg/dl throughout the 24 hours, although small increases are observed immediately before each scheduled dose of metyrapone. There was no difference between patients and controls in mean cortisol levels (F=0.45, df=1/42, p=.5 for group and F=0.79, df=144/6048, p=.96 for interaction of group and time). There were no significant sex differences in mean cortisol levels (F=0.19, df=1/42, p=.66; interaction F=0.62, df=144/6048, p=.998); nor was there a significant sex difference within the patients (F=.016, df=1/42, p=.9 for sex, interaction F=.858, df=144/6048, p=.88) or controls (F=.557, df=1.42, p=.46 for sex, interaction F=.805, df=144/6048, p=.95). The ACTH response to this low cortisol feedback is shown in Figure 2. Following the steep reduction of cortisol levels after 4 PM, no major secretory rise of ACTH was observed. Beginning at around 4:00 to 4:30 AM, the circadian surge of ACTH commenced. Under the prevailing low feedback condition, this expected circadian ACTH 9 secretory phase was exaggerated (peak values > 400 pg/ml) and prolonged to around noon (see Table 2), in agreement with Veldhuis et al (2001). Initial analyses of the ACTH time series demonstrated no significant difference between patients and controls (F=.27, df=1/42, p=.6), but a significant effect of gender in both patients and normal subjects (F = 5.08, df = 1/42, p = .029 for gender for all subjects; F = 5.37, df = 1/20, p = .03 for gender in patients (F>M); F = 0.8, df = 1/20, p = .36 for gender in controls, but a significant interaction between gender and time in controls F = 1.977, df = 144/2880, p = .0001(F>M)). Subsequent analyses of patients vs. controls were conducted separately by sex. In women, there was no significant difference in the ACTH response to 24H metyrapone (Fig 2a) (F = .0002, df = 1/30, p = .98 for group; F = 108, df = 144, p = .0001 for time and F = .76, df=144/4320, p = .98 for interaction). In contrast, depressed men showed a significantly smaller maximal response to 24H metyrapone blockade then their matched controls (Fig 2b) (F = 3.13, df = 1/10, p = .09 for group; F = 40.93, df = 144, p = .0001 for time and F = 3.268, df = 144/1440, p = .0001 for interaction of group with time). Since our previous studies had shown an increased response to metyrapone in the 410PM time frame, we examined the data from the early time points (4-10 PM) to determine if we had replicated the increased drive in the early evening in these patients. If we had been unable to replicate this increased evening drive in these subjects, then the demonstration of a normal maximal ACTH response to metyrapone could be due to differences in the depressed patient populations studied. The data are shown in Fig 3. In depressed women, (Fig 3a) we again observed a significant increase in ACTH secretion 10 in the evening compared to these matched controls (F = 5.82, df = 1/32, p = .02 for group; F = 10.79, df=12, p = .001 for time; no significant interaction). In contrast, depressed men showed a significant decrease from their matched controls (F = 6.3, df = 1/16, p = .02 for group; F = 11.02, df=12, p = .0001 for time; F=2.42, df=12/192, p=.006 for interaction), (Fig 3b). We explored whether the subset of patients who met criteria for melancholia showed greater HPA axis activation compared to controls but their evening and 24H data looked the same as all patients (data not shown). Subjects were asked to record their bedtime and awakening for four days prior to the study. Comparison of group and sex differences in bedtime showed a significant effect of sex (F=10.56, df=1/1, p=.002), but no significant difference between patients and controls (F=1.04, df=1, p=.31) and no significant interaction (F=.65, df=1/40, p=.4). (Control men=1:58±58min; control women=0:44±72min; depressed men=1:32±67min; depressed women=0:39±83min.). The women were on average older (35±14.2) that the men (27.7±9.4) although this difference was not significant (p=.18 unpaired t). There was a significant negative correlation between age and average bedtime in patients (r=0.588, F=10.6, df=1, p=.004) and controls (r=.56, F=9.67, p=.005). Consequently, the significant sex difference in bedtime appeared to be a function of age differences between men and women with more subjects over 40 in the female groups. Finally, since depressed women showed increased ACTH response in the early evening, but a normal ACTH response during the circadian maximum, we asked at what time the ACTH secretion returned to normal in women. At 1AM the depressed women and control women show equal ACTH response to metyrapone blockade. 11 The ACTH data following metyrapone blockade do not show the classic circadian pattern which looks like a cosine function, since the low levels of secretion continue from onset for 12 hours or longer and then rise dramatically beginning around 4AM and are still near maximal at the end of the study (Fig 2). Rather than using harmonic analysis, we defined evening and overnight quiescent period onset and offset and secretory drive onset as described in the methods. All controls met criteria for a quiescent period within an hour of the initiation of sampling at 4PM. In all but 2 controls, this quiescent period was interrupted prior to midnight. The average time of this initial quiescent period offset (excluding 2 controls with offset after midnight) is shown in Table 2. Among patients, 6 did not demonstrate an initial quiescent period. For the remaining 20 patients the offset of this period is shown in Table 2. In no patients did the initial quiescent period extend past midnight. All subjects except one patient and one control showed at least one other quiescent period with onset around midnight. The data for onset and offset of this quiescent period are shown is shown in Table 2. Although there is a suggestion that patients showed shorter initial and overnight quiescent periods because of wide individual variations there are no significant differences. The ACTH secretory onset time is also similar in patients and controls. This onset time was correlated with average bedtime in controls (r=.475, F=6.1, p=.02) and in patients (r=.72, F=21.8, p=.0001). The time of ACTH secretory onset was also correlated with age in controls (r=.643, F=16.95, p=.0004) and patients (r=.73, f=27.2, p=.0001). DISCUSSION These studies were conducted to determine if HPA axis hyperactivity was present throughout the 24H in major depression or was specific to the evening. We utilized 12 metyrapone blockade for 24H beginning at 4PM in the evening – the time when we had previously observed increased evening pituitary -endorphin (Young et al, 1994) and ACTH (Young et al, 2004) response to metyrapone blockade. The answer to this question was completely gender dependent. In women, we observed increased evening ACTH secretion under metyrapone blockade, in agreement with our two previous studies, one involving men and women and the other involving premenopausal women only (Young et al, 1994, 2004). However, depressed women showed a normal maximal response to metyrapone at the circadian acrophase (Fig 2a). This agrees with our previous work looking at a 6 hour metyrapone challenge in the morning (8AM), which showed normal ACTH secretion in depressed patients (Young et al, 1997). In contrast, depressed men showed decreased ACTH drive in the evening (Fig 3b), and decreased maximal ACTH response to metyrapone (Fig 2b). We have previously observed a significant sex differences in the HPA axis in depressed men and women, including the observation that depressed men did not show increased -endorphin/-LPH activation in response to evening metyrapone (Young, 1995). Thus, these data replicate our previous observations of a sex difference in response to evening metyrapone. We have also observed that depressed men did not show evidence of basal cortisol hypersecretion between 8AM- 10AM while depressed women did (Young 1995). However, we did not expect that the depressed men would show a smaller ACTH response to metyrapone blockade. This same pattern was observed in 6 of 9 men, so it was not due to one individual only. Although our ACTH assay did not detect levels below 10pg/ml, this was unlikely to affect the overall estimates of maximal response since the maximal ACTH secretion was in the 300-400 pg/ml range for the depressed patients. Interestingly, a 13 study by Rubin et al (1999) examining ACTH and cortisol responses to low dose physostigmine challenge found an significantly greater response to the challenge in depressed women compared to matched controls and significantly smaller response to the physostigmine in depressed men compared the their matched control men. Their data point to increased cholinergic sensitivity in depressed women and decreased sensitivity in depressed men. Furthermore, the ACTH and cortisol responses to physostigmine appeared to act through arginine vasopressin release (Rubin et al, 1999). Studies of the ACTH response to vasopressin have predominantly demonstrated a normal response in both depressed men and depressed women (Rubin et al, 1999; Carroll et al, 1993) Despite the smaller ACTH response to metyrapone in depressed men, our data should not be interpreted to suggest that depressed men show no HPA axis dysregulation, since studies of melancholic depressed men have similarly shown elevated cortisol around the clock. (Deuschle et al, 1997). In addition, we have found that younger depressed men (<50) were more likely to show non-suppression of -endorphin to dexamethasone than younger (< 50) depressed women (Young, 1995). Since only 4 of the depressed men met criteria for melancholia, they may not have shown the classic HPA axis dysregulation picture. In neither case do our data suggest that CRH drive is increased across the 24 H in patients with major depression. Rather, the data support increased ACTH and CRH drive during the evening only, and then only in women. 14 Studies in rodents suggest that the circadian rhythm of CRH persists in the absence of glucocorticoid negative feedback (Kwak et al, 1993), although data with metyrapone support that cortisol feedback in the morning shifts the acrophase (Veldhuis et al, 2001) . Our data suggest continued HPA axis activation in the evening in depressed women. However, during the time of maximal circadian drive, the peak ACTH secretion was normal. We do not know the mechanisms through which this increased evening secretion is occurring. Although metyrapone successfully blocks cortisol secretion, it cannot suppress cortisol below 2 µg/dl, a range in which mineralocorticoid receptors (MR) are active. Similar to our finding, elevated cortisol immediately before sleep onset has been found in adolescents with major depression (Rao et al, 1996; Forbes et al, 2006). We do not know if MR dysfunction could be involved in the increased evening drive in women, or if a different neural inhibitory pathway is not properly functioning in the depressed women. One possibility is that there is just phase delay in the circadian nadir of ACTH in depressed women, since the two groups eventually were equal by 1AM. However, we found an earlier bedtime in women compared to men and no difference in bedtime between patients and controls, so it is unlikely that phase delay can explain the findings. Furthermore, using our definitions of quiescent period onset and offset, we found no difference between patients and controls, although we did note that 6 patients but no controls showed no quiescent period at the initiation of the study. Finally, a similar pattern of elevated evening cortisol secretion has been found with advanced aging (Van Cauter et al, 1996) and also following experimenter imposed partial sleep deprivation (Leproult et al, 1997). It is possible that sleep loss played a role in the increased evening activation seen in our depressed patients. However, based upon the HDRS administered 15 on the day of the study, only 5 (4F, 1M) of the 26 patients had experienced middle or terminal insomnia in the week of the study, which suggests that disturbed sleep does not account for the increased drive seen in women. With regards to depressed men, again we are unsure of the mechanism involved in the smaller ACTH response to metyrapone. Although speculative, the findings of Rubin et al, (1999) showing a smaller AVP response to physostigmine in depressed men may mean that tonic AVP release under metyrapone is less in depressed men compared to depressed women. Such a change would result in a smaller ACTH response under the open loop system produced by metyrapone inhibition of cortisol synthesis and release, and would account for our findings. Several limitations are noted in this study. We attempted to recruit subjects that were as representative of the general population of depression as possible, including twice as many women as men and including premenopausal and postmenopausal women. However, this results in limited ability to address issues of effects of menopause or menstrual cycle on the HPA axis drive under metyrapone. With respect to menstrual cycle phase we previously observed no menstrual cycle difference in 24H basal ACTH and cortisol secretion in either depressed or control women (Young et al, 2001). Furthermore, using a within subject design we observed no follicular luteal phase differences in ACTH following metyrapone (Altemus et al., in press) administration. The small number of men is a clear limitation and suggests separate studies focusing on men should be undertaken. The mild nature of the depression and the few subjects with 16 melancholia among the males may account for the absence of HPA axis activation in depressed men. We do want to note that other studies have found significantly decreased basal HPA axis activity (Posener et al, 2000; Geracioti et al, 1992) in non-psychotic major depression, although neither study examined possible sex differences in their data. While atypical depression has been proposed to show a pattern of reduced central drive, we did not observe this in our previous 24 study of the HPA axis in depressed women (Young, 2001) nor did any of the men meet criteria for atypical depression. Another limitation was the need to awaken subjects at midnight and 4AM to administer metyrapone. Obviously, this caused sleep disturbances that may have influenced the HPA axis. The studies of Leproult et al (1997) showed that sleep disruption leads to increased afternoon/evening cortisol. In addition, in order to collect samples for pulsatility analysis, blood was drawn every 10 minutes, which may also cause sleep disruption. While we do not know the degree to which sleep disruption contributed to the HPA axis picture, we do not believe it would have contributed differentially between groups. In our previous study of ACTH and cortisol pulsatility over 24H, which included sampling every 10 minutes even while subjects were asleep, we were still able to observe increased ACTH and cortisol in major depression. Furthermore, the mean cortisol in control women was similar to mean cortisol in other studies using less frequent sampling. Finally, in this study, the evening time period was evaluated before the possible sleep disturbances induced by nighttime metyrapone administration and overnight blood sampling. 17 In conclusion, the use metyrapone allowed us to examine the maximal secretion of ACTH in the absence of glucocorticoid feedback. Our data suggest that metyrapone is useful to determine the true timing and activity of central CRH systems of the HPA axis. This strategy revealed unexpected findings with regards to depressed men. Future analyses will address the pulsatile and ultradian rhythm of ACTH secretion in these subjects. But, our data highlight the importance of gender in evaluating the HPA axis in depression. 18 Acknowledgements: The authors acknowledge the support of MH 50030 (EAY), K05 MH 0193 to EAY and MO1 RR00042 (General Clinical Research Center of the University of Michigan) and the assistance of the nursing staff of the G-CRC in blood drawing, and Kathleen Singer RN in recruiting and interviewing patients. We also acknowledge the contributions of our consultant, Bernard J Carroll in the design of the study and helpful comments upon the manuscript. 19 References: Arana GW, Baldessarini RJ, Ornsteen M. (1985) The dexamethasone suppression test for diagnosis and prognosis in psychiatry. Commentary and review. Arch Gen Psychiatry. 42:1193-204 Carroll BJ, Feinberg M, Greden JF, Tarika J, Albala AA, Haskett RF, James NM, Kronfol Z, Lohr N, Steiner M, de Vigne JP, Young E.A (1981) specific laboratory test for the diagnosis of melancholia. Standardization, validation, and clinical utility. Arch Gen Psychiatry. 38:15-2 Deuschle M, Schweiger U, Weber B, Gotthardt U, Korner A, Schmider J, Standhardt H, Lammers CH, Heuser I. (1997) Diurnal activity and pulsatility of the hypothalamuspituitary-adrenal system in male depressed patients and healthy controls. J Clin Endocrinol Metab 82:234-8 Forbes EE, Williamson DE, Ryan ND, Birmaher B, Axelson DA, Dahl RE. (2006) Perisleep-onset cortisol levels in children and adolescents with affective disorders. Biol Psychiatry. 59:24-30. Geracioti TD Jr, Orth DN, Ekhator NN, Blumenkopf B, Loosen PT. (1992) Serial cerebrospinal fluid corticotropin-releasing hormone concentrations in healthy and depressed humans. J Clin Endocrinol Metab. 74:1325-30 20 Halbreich U, Zumoff B, Kream J, Fukushima DK. (1982) The mean 1300-1600 h plasma cortisol concentration as a diagnostic test for hypercortisolism. J Clin Endocrinol Metab. 54:1262-4 Halbreich,U., Asnis, G.M., Schindledecker, R., Zurnoff, B., Nathan, R.S.(1985) Cortisol secretion in endogenous depression I. Basal plasma levels. Arch. Gen. Psychiatry 42:909-914 Carroll BT, Meller WH, Kathol RG, Gehris TL, Carter JL, Samuelson SD, Pitts AF. (1993) Pituitary-adrenal axis response to arginine vasopressin in patients with major depression.Psychiatry Res.;46(2):119-2; Kwak SP, Morano MI, Young EA, Watson SJ, Akil H. (1993) Diurnal CRH mRNA rhythm in the hypothalamus: decreased expression in the evening is not dependent on endogenous glucocorticoids. Neuroendocrinology. 5:96-105 Leproult R, Copinschi G, Buxton OM, Van Cauter E. (1997) Nocturnal sleep deprivation results in an elevation of cortisol levels the next evening. Sleep 20:865-70. Linkowski P, Mendelwicz J, LeClercq R,.Brasseur M, Hubain P, Goldstein J, Copinschi G, van Cauter, E. (1985) The 24 hour profile of ACTH and cortisol in major depressive illness. J Clin. Endocrinol.Metab. . 61:429-34. 21 Nemeroff CB, Widerlov E, Bissette G, Walleus H, Karlsson I, Eklund K, Kilts CD, Loosen PT, Vale W. (1984) Elevated concentrations of CSF corticotropin-releasing factor-like immunoreactivity in depressed patients. Science. 226:1342-4. Pfohl, B., Sherman, B., Schlecte, J., Stone, R. (1985.) Pituitary/adrenal axis rhythm disturbances in psychiatric patients. Arch. Gen. Psychiatry 42:897-903. Poland RE, Rubin RT, Lesser IM, Lane LA, Hart PJ. (1987) Neuroendocrine aspects of primary endogenous depression. II. Serum dexamethasone concentrations and hypothalamic-pituitary-adrenal cortical activity as determinants of the dexamethasone suppression test response. Arch Gen Psychiatry. 44:790-5 Posener JA, DeBattista C, Williams GH, Chmura Kraemer H, Kalehzan BM, Schatzberg AF. (2000) 24-Hour monitoring of cortisol and corticotropin secretion in psychotic and nonpsychotic major depression. Arch Gen Psychiatry. 57:755-60 Raadsheer, F.C., Hoogendijk, W.J., Stam, F.C., Tilders, F.J. & Swaab, D.F. (1994) Increased numbers of corticotropin-releasing hormone expressing neurons in the hypothalamic paraventricular nucleus of depressed patients. Neuroendocrinology 60: 436-444 Raadsheer, F.C., van Heerikhuize, J.J., Lucassen, P.J., Hoogendijk, W.J., Tilders, F.J. & Swaab, D.F. (1995) Corticotropin-releasing hormone mRNA levels in the paraventricular 22 nucleus of patients with Alzheimer's disease and depression. Am J Psychiatry 152: 13721376. Rao U. Dahl RE. Ryan ND. Birmaher B. Williamson DE. Giles DE. Rao R. Kaufman J. Nelson B. (1996) The relationship between longitudinal clinical course and sleep and cortisol changes in adolescent depression. Biological Psychiatry. 40:474-84. Ribeiro SC, Tandon R, Grunhaus L, Greden JF. (1993) The DST as a predictor of outcome in depression: a meta-analysis. Am J Psychiatry. 150:1618-29 Rubin RT, Poland RE, Lesser IM, Winston RA, Blodgett N.(1987) Neuroendocrine aspects of primary endogenous depression I. Cortisol secretory dynamics in patients and matched controls. Arch. Gen. Psychiatry 44:328-336 Rubin RT, O'Toole SM, Rhodes ME, Sekula LK, Czambel RK (1999) Hypothalamopituitary-adrenal cortical responses to low-dose physostigmine and arginine vasopressin administration: sex differences between major depressives and matched control subjects. Psychiatry Res 89:1-20 Sachar EJ, Hellman L, Roffwarg H, Halpern F, Fukushima D, Gallagher T. (1973) Disrupted 24-hour patterns of cortisol secretion in psychotic depression. Arch General Psychiatry;28:19-24 23 Van Cauter E, Leproult R, Kupfer, DJ. (1996) Effects of gender and age on the levels and circadian rhythmicity of plasma cortisol. J Clin Endocrinol Metab 81:2468-2473. Veldhuis JD. Iranmanesh A. Naftolowitz D. Tatham N. Cassidy F. Carroll BJ. (2001) Corticotropin secretory dynamics in humans under low glucocorticoid feedback. Journal Clin Endocrinol Metab. 86:5554-63 Young EA, Lopez JF, Murphy-Weinberg V, Watson SJ and Akil H. (1987) Normal Pituitary Response to Metyrapone in the Morning in Depressed Patients: Implications for Circadian Regulation of CRH Secretion, Biological Psychiatry 41: 1149-1155 Young EA, Altemus M, Lopez JF, Kocsis JH, Schatzberg AF, DeBattista C, Zubieta JK. (2004) HPA axis activation in major depression and response to fluoxetine: a pilot study. Psychoneuroendocrinology 29:1198-20 Young EA, Carlson NE, Brown MB. (2001) 24 Hour ACTH and Cortisol Pulsatility in Depressed Women, Neuropsychpoharmacology, 25:267-276. Young EA, Haskett RF, Grunhaus L, Pande A, Weinberg VM, Watson SJ, Akil H. (1994) Increased evening activation of the hypothalamic-pituitary-adrenal axis in depressed patients.Arch Gen Psychiatry. 51:701-7. 24 Young EA, Kotun J, Haskett RF, Grunhaus L, Greden JF, Watson SJ, Akil H. (1993) Dissociation between pituitary and adrenal suppression to dexamethasone in depression. Arch Gen Psychiatry. 50:395-403 Young EA, Lopez JF, Murphy-Weinberg V, Watson SJ, Akil H (2003) Mineralocorticoid receptor function in major depression. Arch Gen Psychiatry. 60:24-8. Young, EA. (1995) The glucocorticoid cascade hypothesis revisited: role of gonadal steroids, Depression 3:20-27 25 Figure Legends Figure 1. Mean cortisol levels in patients and control subjects over 24H following metyrapone blockade of cortisol synthesis. The upper panel shows the data for women and the lower panel for men. The first dose of metyrapone is given following drawing the first blood sample at 4PM. Repeat doses are given every 4H and are indicated by arrows. Following the initial dose of metyrapone, cortisol levels decline to approximately 2 µg/dl and remain low and constant throughout the 24H. Small increases are seen near the end of the 4 hours when the effects of metyrapone blockade is waning. Figure 2. 24H ACTH profile following metyrapone blockade of cortisol synthesis in depressed women (n=17, upper panel) and depressed men (n=9, lower panel) and their matched controls (17, 9 respectively). Depressed women demonstrate a normal maximal ACTH response to metyrapone blockade (upper panel) while depressed men demonstrate a significantly smaller response than matched control subjects (lower panel). Figure 3. Initial ACTH response to metyrapone blockade of cortisol synthesis between 4PM and 10 PM in depressed women (n=17, upper panel) and depressed men (n=9, lower panel) and their matched controls (17, 9 respectively). Depressed women demonstrate significantly greater ACTH secretion following metyrapone, while depressed men show similar responses to metyrapone as their matched control subjects, 26 27 Table 1 Baseline Description of Patients and Controls (Mean ±SD) Patients Controls Sex 17F; 9M 17F; 9M Age 32.0 ± 13.5 31.8± 12.7 HDRS 17.3 ± 3.3 (range 9-24) 0.8 ± 1.1 Carroll Depression Rating 20.2 ± 5.6 (range 11-26) 0.8 ± 1.2 Melancholic 13 Atypical 4 Recurrent 14 28 Table 2 Circadian Secretory Parameters (Mean ±SD) Controls Patients 19:45±94 min 18:56±105min 23:15±85 min 23:50±100 min Last Quiescent Offset 3:29±159 min 3:08±109 min Secretory Drive Onset 4:25±106 min 4:08±80 min Maximum Secretion 11:51±128min 10:55±129min Initial Quiescent Period Offset Last Quiescent Period Onset 29 Cortisol under Metyrapone in Depressed Female Patients and Control Women 20 Patients Depressed Women Control Women Controls Cortisol, µg/dl 15 10 5 0 0 4PM 12 24 8PM 36 48 MN 60 72 84 4AM Fig 1a 96 8AM 10 8 12 0 Noon 13 2 14 4 4PM 30 Cortisol under Metyrapone in Depressed Male Patients and Control Men 20 Depre ssed men Patients Control m en Controls Cortisol, µg/dl 15 10 5 0 0 4PM 12 24 8PM 36 48 MN 60 72 4AM 84 96 8AM Figure 1b 10 8 12 0 Noon 13 2 14 4 4PM 31 Response of Depressed Women and Their Matched Controls to 24H Metyrapone Blockade 500 Patients Depressed Controls Control 400 ACTH, pg/ml 300 200 100 0 0 4PM 12 24 8PM 36 48 MN 60 72 4AM 84 96 8AM Figure 2a 10 8 12 0 Noon 13 2 14 4 4PM 32 Response of Depressed Men and Their Matched Controls to 24H Metyrapone Blockade 500 Controls Controls Depressed Patients 400 ACTH, pg/ml 300 200 100 0 0 4PM 12 24 8PM 36 48 MN 60 72 4AM 84 96 8AM Figure 2b 10 8 12 0 Noon 13 2 14 4 4PM 33 Evening Metyrapone Challenge in Women Depressed Patients and Matched Controls 35 Controls Patients Dinner 30 ACTH, pg/ml 25 20 15 10 5 0 2 4 6 8 Clock Time, PM Figure 3a 10 12 34 Evening Metyrapone Challenge in Male Depressed Patients and matched Controls 60 Controls Patients 50 Dinner ACTH, pg/ml 40 30 20 10 0 2 4 6 8 Clock Time, PM Figure 3b 10 12 35 Response to Metyrapone in Depressed Women and Matched Control Subjects Over 12 Hours 45 Controls 40 Patients 35 ACTH, pg/ml 30 25 20 15 10 5 0 2 4 4PM 6 6PM 8 8PM Figure 4 10 10PM 12 MN 14 2AM 16 4AM