The role of mineralocorticoid receptor function in treatment-resistant depression

Abstract

Background:

Treatment-resistant depression patients show both reduced glucocorticoid receptor function and a hyperactive hypothalamic-pituitary-adrenal axis. However, few studies have examined the role of the mineralocorticoid receptor. This study aimed to evaluate the functional activity of the mineralocorticoid receptor system in regulating the hypothalamic-pituitary-adrenal axis in well-defined treatment-resistant depression patients.

Material and method:

We recruited 24 subjects divided into: (a) treatment-resistant depression; (b) healthy controls. We evaluated: (a) the effect of combined glucocorticoid receptor/mineralocorticoid receptor stimulation with prednisolone; (b) the effect of prednisolone with the mineralocorticoid receptor antagonist spironolactone; and (c) the effect of spironolactone alone. The response of the hypothalamic-pituitary-adrenal axis was measured using salivary cortisol and plasma levels of drugs were also measured.

Results:

Treatment-resistant depression patients had higher cortisol compared with controls after all challenges. In controls, spironolactone increased cortisol compared to placebo. The co-administration of spironolactone with prednisolone in controls decreases the suppressive effects of prednisolone. In contrast, in treatment-resistant depression, spironolactone did not increase cortisol compared to placebo and spironolactone with prednisolone had no effect on the suppressive effects of prednisolone. Patients with treatment-resistant depression had a reduction in the conversation of spironolactone to the active metabolite canrenone.

Conclusion:

Our data confirmed that treatment-resistant depression is associated with hypercortisolism and these patients no longer show an hypothalamic-pituitary-adrenal response to the administration of a mineralocorticoid receptor antagonist, suggesting that there is a mineralocorticoid receptor malfunctioning, such as a down regulation, however, pharmacokinetics and pharmacodynamics in these subjects could also have had an effect on the lack of mineralocorticoid receptor response.

Keywords

Cortisol hypothalamic-pituitary-adrenal axis mineralocorticoid receptor prednisolone treatment resistant depression mineralocorticoid receptor antagonist

Introduction

Endogenous corticosteroids such as cortisol and cortiocosterone have multiple effects, including feedback regulation of the hypothalamic-pituitary-adrenal (HPA) axis, and act through two distinct intracellular corticosteroid receptors: the mineralocorticoid receptor (MR) and the glucocorticoid receptor (GR) (De Kloet et al., 1998; McEwen, 2000).

The MR has a high affinity for endogenous glucocorticoids: the in vitro dissociation constant/ionization constant (Kd/Ki) is 0.13 nM for cortisol binding to human MR (Rupprecht et al., 1993). In contrast, the GR has a low affinity for endogenous glucocorticoids: the in vitro Kd/Ki is 15 nM for cortisol binding to human GR (Rupprecht et al., 1993). Thus, under basal cortisol levels, especially at the circadian nadir, negative feedback is mediated predominately through MR in the hippocampus, whereas under stress and high cortisol concentrations, the less sensitive GR in the hippocampus, hypothalamus and pituitary come into play. The balance in these MR- and GR- mediated effects on the stress system is of crucial importance to the set point of the HPA axis activity (De Kloet et al., 1998). De Kloet et al. (1998) and Spencer et al. (1998) have clarified that GR activation is necessary for the HPA axis feedback regulation when levels of glucocorticoids are high (e.g. response to stress and the circadian peak), but that MR also plays an important role by modulating GR-dependent regulation (Spencer et al., 1998). A decrease in MR in the anterior but not posterior hippocampus of major depressive disorder patients emphasizes the important functional role of the anterior hippocampus in neuroendocrine regulation in humans (Medina et al., 2013). Klok et al. (2011) demonstrated the effect of common functional MR gene variants on the cortisol awakening response, which is often disturbed in stress-related disorders like depression.

A multitude of studies have shown that some depressed patients do not suppress cortisol secretion following administration of the exogenous glucocorticoid, dexamethasone (DEX) (Ribeiro et al., 1993; Juruena et al., 2004). However, DEX has pharmacodynamic and pharmacokinetic features that are very distinct from those of the main endogenous glucocorticoid cortisol. Recently, a novel suppressive test of the HPA axis using prednisolone has been developed (Pariante et al., 2002). Prednisolone is a synthetic glucocorticoid that is similar to cortisol in its pharmacodynamics and pharmacokinetics in that it binds both to the GR and MR. This prednisolone suppression test (PST) diverges from the traditional DEX suppression test (DST) in that, while DEX only investigates the function of GR, prednisolone probes both GR and MR. Since endogenous HPA axis feedback involves both GR and MR, and given the evidence reviewed above that MR can modulate and compensate for altered GR function (De Kloet et al., 1998), prednisolone should provide a more physiological test of the HPA axis in depression. In a preliminary study on 18 subjects with treatment-resistant depression (TRD), we found a normal suppressive response to prednisolone (5 mg) even though the same subjects demonstrated resistant suppression to DEX (Juruena et al., 2006), suggesting that there may be changes in MR function that are compensating for reduced GR function in these patients. The PST was administered to 45 inpatients with depression assessed as resistant to two or more antidepressants and to 46 controls. The TRD group had higher cortisol levels than controls and non-response to inpatient treatment was predicted by a more dysfunctional HPA axis (higher cortisol levels post-prednisolone and lower percentage suppression). These data suggested that in TRD patients HPA axis activity is reset at a higher level, although feedback remains intact. Moreover, prospectively determined TRD is associated with an impaired feedback response to combined GR and MR activation by prednisolone (Juruena et al., 2009).

Relatively few studies have specifically probed MR function in depression. The early studies showing raised cortisol levels and insensitivity to DEX, together with the potential for glucocorticoids to down-regulate both MR and GR (Halbreich et al., 1985; Pfohl et al., 1985) led to conclusions that major depression is accompanied by decreased MR activity. However, MR function was in fact not directly tested by the DST, given the pharmacological effects of DEX described earlier, and direct tests of the hypothesis were not undertaken until sometime later. The first indication came from a post mortem study on depressed suicide victims (Lopez et al., 1998), which demonstrated decreased MR messenger RNA in the hippocampus compared with healthy controls. Individuals with psychotic major depression compared to healthy control subjects have diminished feedback inhibition of the HPA axis in response to the MR agonist fludrocortisone (Lembke et al., 2013). GR and MR differed in their response to the concentration of corticosterone in neural cells and non-neural cells. In the nuclear region, Nishi and Kawata (2007) detected GR-MR heterodimers, which were affected by changes in corticosterone concentrations in response to various hormonal milieus, such as circadian rhythm and stress.

In order to test directly the hypothesis that MR function is reduced in depression, Young et al. (2003) administered spironolactone, a MR antagonist, to patients with major depression and matched control subjects and assessed adrenocorticotropic hormone (ACTH) and cortisol secretion in response to this acute challenge. This study in fact found that the function of MR is normal or possibly hyperactive in depression (Young et al., 2003). Young et al.’s (2003) results demonstrate that, despite high baseline cortisol levels, patients with major depression display a heightened response to MR antagonism and thus increased functional activity of the MR system. Taken together with the data showing decreased sensitivity to GR agonists like DEX, these data suggest the possibility of an imbalance in the MR/GR ratio in depression.

The neurobiology of TRD remains poorly understood. Indeed, few studies have specifically focused on biological factors associated with a poor response to standard treatments in depression. One potential factor that could be associated with treatment resistance is the chronically hyperactive HPA axis, with high cortisol levels contributing to the development or maintenance of depressive symptoms (Juruena et al., 2006, 2009).

Our hypotheses were based on the view that there is preserved or up-regulated function of the MR receptor in TRD, which is able to counteract the down-regulation of GR function. On this basis, we predicted that in comparison to healthy controls TRD patients would show the following pattern of responses: (1) normal cortisol suppression by prednisolone; (2) normal or enhanced cortisol activation by spironolactone, because of normal or hyperactive MR; (3) in the presence of MR antagonism, prednisolone would act as a pure GR agonist and therefore patients would show cortisol non-suppression because prednisolone would only be able to activate the (resistant) GR.

Method

Study design

Subjects entered a single-blind, placebo controlled study in which the effects of prednisolone (5 mg), spironolactone (400 mg) and the combination of spironolactone (400 mg) and prednisolone (5 mg) on salivary cortisol secretion were measured. Saliva samples were taken from TRD patients and controls hourly between 09:00 and 17:00 hours in a 4-day protocol as follows. On day 1 (placebo), saliva samples were collected after subjects had taken one capsule containing placebo at 22:00 hours the night before. On day 2 saliva samples were collected after subjects had taken one capsule containing prednisolone at 22:00 hours the night before. On day 3, undertaken at least 48 h after the administration of prednisolone, saliva was collected after subjects has received spironolactone at 22:00 hours the night before. On day 4, undertaken at least 1 week after day 3, saliva was collected after the administration of a capsule containing both spironolactone and prednisolone at 22:00 hours the night before.

Subjects

Two groups of subjects were recruited. The patient group consisted of 12 patients with major depression who were inpatients on the National Affective Disorders Unit (ADU) of the Bethlem Royal Hospital (South London and Maudsley NHS Trust). Patients were included in this study if they were diagnosed as having unipolar recurrent major depressive disorder or a major depressive episode according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 1994). All were treatment-resistant based on a retrospective history of failure to respond to two or more prior treatment trials of adequate dose and duration using the Antidepressant Treatment History Form (ATHF; Sackeim et al., 2001). A thorough medical examination was performed in order to assess comorbidity, physical disorders, general medical conditions, lifestyle, psychosocial problems and stress. For practical reasons, it was not possible to test patients in a drug-free state, but patients taking steroids or anti-glucocorticoid drugs such as ketoconazole were excluded. Exclusion criteria for patients were: history of hypersensitivity to corticosteroids or steroid use; heavy smokers (i.e. more than 25 cigarettes/d); viral illnesses during the preceding 2 weeks; pregnant or lactating women; alcohol dependence; significant physical illnesses (for example, severe allergies, autoimmune diseases, hypertension, malignancy, haematological, endocrine, pulmonary, renal, hepatic, gastrointestinal or neurological disease). Patients with bipolar affective disorder, psychotic symptoms unrelated to their depressive disorder or an organic aetiology were excluded.

Clinical assessment

For diagnostic assessment, we utilized the Structured Clinical Interview for DSM-IV (American Psychiatric Association, 1994) for Axis I (SCID-I; First et al., 1997) and Axis II (SCID-II) disorders (First et al., 1995). In addition to the ATHF, we also applied the Thase and Rush staging criteria (Thase and Rush, 1997), which recognize five stages of treatment resistance according to the number of treatment trials adequately delivered. For clinical severity of depression, we utilized a structured version of the 21-item Hamilton Rating Scale for Depression (HAMD-21; Williams, 1988) and the Beck Depression Inventory-II (Beck et al., 1996). In order to assess early environmental stress, we utilized the Childhood Experience of Care and Abuse Interview (Bifulco et al., 1994) and the Childhood Experience of Care and Abuse Questionnaire (CECA-Q; Bifulco et al., 2005) .

As expected, there was a wide range of comorbidity. According to the SCID-I, two patients had generalized anxiety disorder, one had atypical depression, one had social phobia, three had post-traumatic stress disorder, one had panic disorder and agoraphobia and one had somatization disorder. According to the SCID-II, nine of 12 patients had current features of one or more comorbid Axis II conditions. Of these, six patients presented with avoidant traits, five obsessive compulsive, six depressive, seven dependent, two schizoid, one borderline, one narcissistic, one passive aggressive and one schizotypal. We did not attempt formally to diagnose a lifelong personality disorder given the difficulty of so doing in the presence of current depression. In addition, eight of the 12 patients had a history of early life stress according to the CECA-Q (Bifulco et al., 2005); of these, five had experienced parental neglect or emotional abuse, four had also experienced physical abuse and five had also experienced sexual abuse.

All 12 patients were taking medication during the assessment. Ten patients were taking mood stabilizers (lithium, sodium valproate, lamotrigine), nine were taking benzodiazepines (diazepam, clonazepam, nitrazepam), nine were taking antipsychotics (chlorpromazine, risperidone, amilsupride, quetiapine, flupenthixol), six were taking tricyclic and related antidepressants (imipramine, amitriptyline, trazodone), six were taking other drugs (zopliclone, paracetamol, promethazine), five were taking other antidepressants (bupropion, mirtazapine, venlafaxine), four were taking selective serotonin reuptake inhibitors (SSRIs; fluoxetine, escitalopram, paroxetine) and two were taking monoamine oxidase inhibitors (phenelzine).

The control group comprised 12 healthy controls recruited from our database of controls from hospital staff, students and the local community. Patients and controls were matched according to age (to within a limit of 5 years), gender and body mass index (BMI; within a range of ± 5 kg/m²). Controls were physically healthy on the basis of complete medical history and examination, were not taking any psychotropic medications, were not taking any hormonal medications (including oral contraceptives) and had no history of hypersensitivity to corticosteroids. Urine tests for illicit drug use and pregnancy were conducted before starting the study. Healthy controls were excluded if they had a personal history or first degree relative history of a DSM-IV Axis I disorder. The Beck Depression Inventory-II (Beck et al., 1996) and the structured HAMD-21 (Williams, 1988) provided information on the severity of control subjects’ depressive symptoms. All healthy controls had to have Beck Depression Inventory scores of <9 (and in fact none was >6: see Table 1).

Table 1.

Demographic and clinical measures in patients with treatment-resistant depression (TRD, n=12) and in healthy controls (n=12).

	TRD patients (mean (SEM) or percentage)	Healthy controls (mean ((SEM) or percentage)	F (d.f.) values	p values
Gender (% female)	10 f (83%)	9 f (75%)	–	–
Age (years)	50.7 (2.8)	51.4 (3.1)	0.03 (22)	0.87
BMI (kg/m²)	31.2 (1.1)	28.6 (1.1)	3.0 (22)	0.10
Years of education	13.2 (0.9)	12.8 (0.6)	0.09 (22)	0.79
Age at onset of 1st MDE (years)	32.8 (2.8)	–	–	–
Age at onset of the current episode (years)	45.2 (3.4)	–	–	–
Number of MDEs	8.7 (1.8)	–	–	–
Duration of illness (years)	17.0 (4.3)	–	–	–
TRD stage (%)	Stage 5: 09 (75%) Stage 4: 02 (17%) Stage 3: 01 (08%)	–	–	–
Number of previous hospital admissions	4.6 (1.2)	–	–	–
Number of medications for depression tried prior to admission	13.0 (1.9)	–	–	–
Duration of current MDE (months)	52.5 (9.5)	–	–	–
Duration of current admission (weeks)	20.3 (2.2)	–	–	–
ECT in the past (%)	11 (91%)	–	–	–
HAMD-21	25.7 (1.9)	0.75 (0.3)	164.8 (22)	<0.001
BDI-II	35.2 (2.8)	1.1 (0.3)	147.9 (22)	<0.001

d.f.: degrees of freedom; f: female; BMI: Body Mass Index; HC: healthy controls; SCID-I: Structured Clinical Interview for DSM–IV Axis I Disorders; MDE: major depressive episode; ECT: Electroconvulsive therapy; HAMD-21: 21-item Hamilton Rating Scale for Depression; BDI-II: Beck Depression Inventory-II.

The study protocols were all approved by the Research Ethics Committee of the Institute of Psychiatry and South London and Maudsley NHS Trust. Written informed consent was obtained from all subjects.

Endocrine protocol

Both patients and controls were admitted to the research rooms of the ADU, where they spent the period 08:45–17:15 hours engaged in sedentary activities. Subjects remained fasting from midnight the previous night until after the first sample collection at 09:00 hours. Snacks, meals and drinks were then standardized throughout the day. Saliva samples were collected using Salivettes (Sarstedt, Leicester, UK) containing untreated cotton swabs. Individuals were asked to place this swab in their mouths and move it around with their tongue until it was saturated with saliva; the cotton swab was then replaced in the vial without being touched. Saliva was separated from cotton roll by quick centrifugation (3500 rpm, 10 min) and samples stored in a freezer at −40°C until assayed.

Peripheral blood was collected by venepuncture each day at 09:00 hours in order to measure plasma cortisol, plasma spironolactone, plasma canrenone and plasma prednisolone as appropriate. The saliva samples were always collected before blood sampling or meals to avoid confounding effects of blood collection or eating.

Saliva cortisol assay

Salivary cortisol was measured using a time-resolved immunofluorescent assay, as previously described (Juruena et al., 2006; Pariante et al., 2002). The intra-assay precision was 8.8% at 0.3 nmol/L, 8.9% at 1.0 nmol/L and 6.6% at 4.6 nmol/L. The inter-assay precision was 7.7% at 2.1 nmol/L and 5.9% at 9.2 nmol/L. The minimal detection concentration was 0.1 nmol/L and there was no ‘drifting’ evident in assays up to 200 wells. The cross-reactivity of the antiserum was prednisolone (28%), 11-deoxycortisol (10%), cortisone (1%), corticosterone (1%) and there was no cross-reactivity with DEX.

Plasma prednisolone, spironolactone and canrenone assays

Plasma levels of prednisolone, spironolactone and canrenone (one of the main active metabolites of spironolactone) were measured by high performance liquid chromatography (HPLC) with UV detection performed with a Hewlett Packard model 1100 HPLC fitted with a HP 1100 variable wavelength detector linked to a Chemstation collection system. Calibration graphs were linear for prednisolone in the range 5–500 ng/ml, spironolactone 10–500 ng/ml and canrenone and 5–500 ng/ml. The intra-assay precision for prednisolone was 11.2% at 5 ng/ml, 5.2% at 18 ng/ml and 2.0% at 225 ng/ml. The inter-assay precision was 10.7% at 5 ng/ml, 9.6% at 18 ng/ml and 3.1% at 225 ng/ml.

Statistical analysis and data presentation

In the main analysis the following factors were entered into a general linear model (GLM): challenge (placebo vs. prednisolone vs. spironolactone vs. prednisolone + spironolactone; group (TRD patients vs. healthy controls); and time (9:00 to 17:00 hours). We also used as summary measures the total salivary cortisol output, calculated as the area under the curve (AUC) using the trapezoidal method, and further calculated the percentage suppression of salivary cortisol for each individual. The percentage suppression represented the AUC as a percentage of the AUC_PLACEBO based on the formula:

Percentage suppression = - [({AUC}_{PLACEBO} - AUC) / {AUC}_{PLACEBO}) * 100] % .

Further analyses were conducted using one-way analysis of variance for the salivary cortisol AUCs (from 9:00 to 17:00 hours), for cortisol values at individual time-points and for prednisolone, spironolactone and canrenone plasma levels. All values are presented as means (± SEM). All p-values reported are two-tailed. A value of p<0.05 was considered statistically significant.

Results

Clinical assessment

Patients and controls were matched for age, gender and BMI. There were three males and nine females among controls and two males and 10 females among patients (χ²=0.242; degrees of freedom (d.f.)=1; p=0.62), see Table 1.

Endocrine assessment

The salivary cortisol profiles after each challenge are shown in Figure 1 for controls and in Figure 2 for TRD patients.

The GLM analysis showed a main effect of group (TRD patients vs. healthy controls; F=83.3; d.f.=1,198; p<0.001), indicating that TRD patients had higher salivary cortisol levels compared to controls on all 4 days. The GLM analysis also showed a main effect of time (09:00 to 17:00 hours; F=13.5; d.f.=8,198; p<0.001), indicating higher cortisol concentration in the morning than in the afternoon, and a group×time interaction (F=4.5; d.f.=8,198; p<0.001), indicating that the fall in cortisol levels over time was larger in the patient group owing to the higher 09:00 hour values. Pairwise data analyses confirmed the differences between patients and controls after placebo (p<0.001), after prednisolone (p<0.001), after spironolactone (p<0.001) and after spironolactone + prednisolone (p<0.001). These differences between the daily cortisol profiles of the patients and those of the controls were confirmed by the analysis of the AUCs (see Table 2). Patients had significantly larger AUCs compared with controls after placebo (approximately two-fold; p=0.012), prednisolone (approximately two-fold; p=0.001) and spironolactone (approximately 1.7 fold; p=0.001) and trends towards larger AUCs after prednisolone + spironolactone (approximately 1.7 fold; p=0.069).

Table 2.

Total salivary cortisol output from 09:00 to 17:00 hours, measured as the area under the curve (AUC) and presented as the absolute mean (SEM) and the percentage suppression relative to placebo.

	Controls (n=12)	TRD patients (n=12)	t (d.f.) values	p values
AUCs after placebo (nmol*h/L)	23.3 (1.7)	54.8 (11.4)	−2.7.(22)	0.012
AUCs after prednisolone (nmol*h/L)	11.6 (1.2)	29.2 (4.7)	−3.6 (22)	0.001
AUCs after spironolactone (nmol*h/L)	36.7 (2.4)	61.9 (6.0)	−3.9 (22)	0.001
AUCs after spironolactone + prednisolone (nmol*h/L)	18.5 (1.7)	31.8 (6.7)	−1.9 (22)	0.069
Percentage suppression by prednisolone	−49.9% (4.0)	−39.9% (8.6)	−1.1 (22)	0.303
Percentage stimulation by spironolactone	62.9% (11.2)	39.4% (13.9)	1.3 (22)	0.203
Percentage suppression by spironolactone + prednisolone	−18.8% (6.3)	−32.1% (11.1)	1.0 (22)	0.310

d.f.: degrees of freedom.

Data shown for patients with treatment-resistant depression (TRD) and healthy controls after administration of placebo, prednisolone (5mg), spironolactone (400mg) or the combination of prenisolone and spironolactone.

Subsequent GLM analyses were conducted separately in TRD patients and controls. In the healthy controls the GLM analysis showed a main within-group effect of challenge (placebo vs. prednisolone vs. spironolactone vs. prednisolone + spironolactone; F=119.2; d.f.=3,297; p<0.001) and a challenge×time interaction (F=4.2; d.f.=31,363; p<0.001). Subsequent pairwise analyses in healthy controls showed that all challenges induced differential effects (all p<0.001). Analysis of the AUCs between 09:00 and 17:00 hours confirmed that there was a significant differential effect on cortisol levels between placebo vs. prednisolone (suppressed approximately two-fold; p=0.001), placebo vs. spironolactone (stimulated approximately 1.6 fold; p=0.001) and placebo vs. prednisolone + spironolactone (suppressed approximately 1.3 fold; p=0.001). These analyses were confirmed by undertaking t-tests on the results of the AUCs between 09:00 and 17:00 hours, which showed that all four challenges induced different levels of suppression or activation.

In TRD patients GLM analysis showed a main within-group effect of challenge (placebo vs. prednisolone vs. spironolactone vs. prednisolone + spironolactone; F=46.7; d.f.=3,297; p<0.001) and a challenge×time interaction (F=4.15; d.f.=31,363; p<0.001). Subsequent pairwise analysis indicated that there was no difference between placebo and spironolactone in their effects on salivary cortisol (p=0.4) and, similarly, there was no difference between prednisolone and prednisolone + spironolactone (p=0.3). The other pairwise comparisons yielded significant differences (p<0.001).

The GLM pairwise analyses in TRD were confirmed by the results of the AUCs between 09:00 and 17:00 hours in that there was no differential effect on cortisol levels between placebo and spironolactone or between prednisolone and the prednisolone + spironolactone combination.

Next, responses to each challenge were analysed as the percentage suppression or stimulation in relation to the placebo response. The values are shown in Table 2; essentially, the degree of suppression did not differ between TRD patients and controls. This is consistent with our previous data in a larger sample of TRD patients (Juruena et al., 2006, 2009). We also then compared the effect of each challenge measured as percentage suppression within healthy controls and within TRD patients (Table 3). The main finding was that, while in controls the suppressive response to prednisolone was significantly attenuated by the co-administration of spironolactone, in TRD patients the co-administration of spironolactone did not significantly alter the suppressive response.

Table 3.

Comparison between challenges in terms of mean (SEM) percentage suppression relative to placebo in 12 healthy controls and 12 patients with treatment resistant-depression (TRD).

% Suppression by A vs. B	Percentage suppression by A	Percentage suppression by B	t (d.f.) values	p values
Healthy controls
Prednisolone vs. spironolactone + prednisolone	−49.9% (4.0)	−18.8% (6.3)	−4.6 (11)	0.001
Spironolactone vs. spironolactone + prednisolone	62.9% (11.2)	−18.8% (6.3)	8.1 (11)	<0.001
TRD patients
Prednisolone vs. spironolactone + prednisolone	−39.9% (8.6)	−32.1% (11.1)	−0.7 (11)	0.479
Spironolactone vs. spironolactone + prednisolone	39.4% (13.9)	−32.1% (11.1)	5.6 (11)	<0.001

d.f.: degrees of freedom. A is the first listed challenge and B is the second listed challenge for each comparison.

Plasma drug levels

Comparison between prednisolone plasma levels after 5 mg prednisolone showed similar plasma levels in patients and controls. Values for 5 mg prednisolone were 66.5 (10.9) ng/ml in patients vs. 56.1 (5.1) ng/ml in controls (t= −0.86; d.f.=22; p=0.40). Again, this is consistent with our previous study in TRD patients (Juruena et al., 2006, 2009).

Levels of spironolactone and its active metabolite canrenone are shown in Table 4. In some patients, spironolactone levels were undetectable the morning after administration. Overall, TRD patients had numerically much larger levels of spironolactone (8–11-fold higher), although this failed to reach statistical significance in the small numbers, and lower levels of canrenone, which was detectable in all patients and did reach significance. This suggests that some TRD patients metabolized the conversion of spironolactone to canrenone more slowly than did healthy controls.

Table 4.

Mean (SEM) and range of plasma levels of spironolactone and canrenone at 09:00 hours in treatment-resistant depression (TRD) patients and controls after receiving spironolactone (400 mg) or spironolactone (400 mg) + prednisolone (5 mg) at 22:00 hours the previous night.

	Controls		TRD patients		t (d.f.) values	p values
	Mean	Range (minimum–maximum)	Mean	Range (minimum–maximum)	t (d.f.) values	p values
Spironolactone plasma levels (ng/ml) after spironolactone	16.2 (5.5)^a	Not detectable−68.7	123.4 (68.7)^b	Not detectable−689.0	−1.5 (22)	0.148
Canrenone^e plasma levels (ng/ml) after spironolactone	198.0 (14.3)	120.1–275.9	129.5 (16.6)	11.8– 321.4	3.1 (22)	0.005
Spironolactone plasma levels (ng/ml) after spironolactone + prednisolone	11.2 (6.2)^c	Not detectable−73.9	124.4 (67.9)^d	Not detectable−661.8	−1.6 (22)	0.124
Canrenone^e plasma levels (ng/ml) after spironolactone + prednisolone	350.3 (36.6)	204.7–571.8	202.0 (26.7)	106–419.4	3.4 (22)	0.003

d.f.: degrees of freedom.

Detectable levels in 9/12 controls.

Detectable levels in 5/12 TRD patients.

Detectable levels in 5/12 controls.

Detectable levels in 8/12 TRD patients.

Canrenone was detected in all 24 subjects.

We also compared the canrenone plasma levels between spironolactone and spironolactone + prednisolone. In healthy controls and TRD patients, canrenone levels were significantly lower after spironolactone than after spironolactone + prednisolone (t= −4.256, d.f. =11, p=0.001; t= −3.089, d.f.=11, p=0.01, respectively). This suggests that prednisolone enhances the conversion of spironolactone to canrenone in healthy controls and TRD patients.

Discussion

This study was designed to clarify the status of MR in TRD and the role of MR in the differential response to prednisolone and DEX (Juruena et al., 2006). The specific aim of the study was to evaluate functional activity of the MR system and the relative role of MR and GR in regulating the HPA axis in TRD. Although HPA axis dysfunction is recognized to be an important factor in the aetiology of severe depressive illnesses, and specifically a factor implicated in treatment resistance, very little data exist about the status of the MR receptor in this illness. On the other hand, there are several studies that assess GR with DEX; for instance, Barden et al. (1997) demonstrated that impaired GR function implied in major disturbances in HPA axis regulation seem to be caused by the primary defect in conjunction with secondary modifications in, among others, pituitary corticotrophin releasing hormone (CRH) receptor system(s), sympathetic output and adrenal development. The GR antagonist mifepristone (RU486) exerts therapeutic effects in the treatment of cognitive dysfunction in bipolar disorder (Young et al., 2004), but the data on core depressive symptoms in psychotic depression are less promising (Flores et al., 2006).

For the first time, we have analysed the extent to which MR receptors contribute to HPA axis suppression in TRD patients.

We previously showed that, in control subjects, there was a correlation between suppression by prednisolone and suppression by DEX, indicating that control subjects are equally sensitive to both glucocorticoids. In contrast, no such correlation was present in depressed patients, confirming the dissociation between sensitivity to prednisolone and resistance to DEX in depression (Juruena et al., 2006). Our preferred interpretation of these findings is that depressed patients (or, specifically, that sample of depressed patients: TRD patients) show a selective impairment of GR sensitivity, probed by DEX, whereas MR sensitivity, additionally probed by prednisolone, is retained (Juruena et al., 2006). This interpretation is consistent with the only study that has so far examined MR function in depression, using the MR antagonist spironolactone (Young et al., 2003). In this study, the authors administered spironolactone in the morning and found that depressed patients showed a larger activation of cortisol secretion compared with control subjects. This suggests that MR activity in depressed patients is preserved, or possibly higher than, in control subjects to compensate for the reduced GR function (Young et al., 2003). Our previous study used an MR agonist (prednisolone) to test directly the ability of the MR to suppress the HPA axis, rather than inferring this ability by blocking MR with an antagonist, and it is striking that these two different approaches reached the same conclusion. It is intriguing to speculate that the (hyper)functional MR could represent a protective mechanism that prevents further biological and clinical deterioration in depressed patients.

In humans, while MRs are thought to be involved in the tonic inhibitory activity within the HPA axis, GRs appear to ‘switch off’ cortisol production at times of stress (Reul and De Kloet, 1985). According to Pace and Spencer (2005), MR may be necessary for glucocorticoid regulation of HPA axis activity during mild stressors, but not during stressors that result in a stronger HPA axis response (Pace and Spencer, 2005). Some studies had demonstrated that, because of the low levels of circulating cortisol in the nadir, MR is more important in the regulation of HPA axis drive in the evening. Thus, Dodt et al. (1993) described enhanced secretory activity of the HPA system in humans after the administration of the selective MR antagonist, canrenoate. After short-term administration in humans they found cortisol, but not ACTH concentrations, to be increased during the night (Dodt et al., 1993). However, studies have established that MR is active throughout the circadian rhythm and that blocking MR feedback results in increased cortisol secretion, both in the morning and evening (Young et al., 1998). Similarly, pre-treatment with the MR antagonist spironolactone induces a heightened response to the combined DEX/CRH test, suggesting that MR antagonism can decrease the degree of negative feedback (Heuser et al., 2000). Stimulation of MR with fludrocortisone as adjunct to escitalopram accelerated the response in a depressive group of treatment responders (Otte et al., 2010).

Direct assessment of MR receptor function using the MR antagonist spironolactone

As described in the introduction, most studies in normal subjects have demonstrated increased basal plasma cortisol levels after receiving a MR antagonist (spironolactone) (Deuschle et al., 1998; Heuser et al., 2000; Kellner et al., 2002;Young et al., 1998). Also, after CRH stimulation (Arvat et al., 2001; Kellner et al., 2002) as well as in the combined DEX suppression/CRH stimulation test (Heuser et al., 2000) and during exercise (Wellhoener et al., 2004), significant elevations of plasma cortisol after MR antagonists treatment have been reported. In genetically predisposed subjects, an imbalance in the MR/GR control mechanism can introduce a bias towards stress-related brain disease after adverse experiences. (De Kloet et al., 2005). The increased availability of cortisol after spironolactone treatment implies enhanced GR activation, which, in combination with MR blockade, presumably resulted in a decreased MR/GR activation ratio. This condition influences both selective attention and performance in various memory tasks (Cornelisse et al., 2011). Moreover, an association of the inter-individual variability in neuroendocrine and/or autonomic activity between the I180V single nucleotide polymorphism (SNP) and symptoms of geriatric depression has been suggested, in that, the MR I180V polymorphism carriers had higher scores of depressive symptoms than non-carriers (Kuningas et al., 2007). The MR I180V SNP is associated with increased cortisol and has a specific effect on depressive symptoms, regardless of cognitive function (Kuningas et al., 2007). This may explain why human carriers of a MR gene variant display enhanced neuroendocrine and autonomic responsiveness to a psychological stressor (Joëls et al., 2008).

We found elevated cortisol levels in saliva after spironolactone administration in healthy controls, thus confirming the previous findings with different methodology. However, in contrast to the picture in healthy controls, we find no difference between placebo and spironolactone on salivary cortisol in TRD patients. We were therefore unable to demonstrate any effect of MR on endogenous glucocorticoids in TRD subjects, suggesting that in respect of HPA activity there was a down-regulation of MR.

It is important to look for possible explanations for these intriguing findings. The first possibility is that TRD is truly associated with a loss of MR effect in the HPA feedback loop. The second is that the intensive pharmacotherapy in these patients could itself result in MR down-regulation. Third, a differential metabolism of spironolactone might result in lower effective levels of circulating MR antagonist in the TRD subjects.

Spironolactone has a short plasma half-life, probably due to rapid hepatic metabolism, and it is an inducer of hepatic microsomal drug metabolizing enzymes in man, notably CYP3A ( Kocarek et al., 1995; Schuetz et al., 1998). Many psychotropic drugs can inhibit CYP3A, such as SSRIs, norfluoxetine, tricyclics, venlafaxine, mirtazapine, reboxetine and others (Zhou et al., 2007). Other psychotropic drugs are also metabolized by CYP3A, such as aripiprazole, benzodiazepines, buspirone, SSRIs, clozapine, mirtazapine, paracetamol, quetiapine, reboxetine, risperidone, tricyclics, topiramate, valproate, ziprasidone, zolpidem and zopliclone. Thus, from a pharmacokinetic perspective, the most likely interpretation of why there was no cortisol response to MR antagonism by spironolactone was that TRD patients receiving intensive pharmacotherapy had increased inhibition of CYP3A and as a result increased plasma levels of drugs that are metabolized by CYP3A and reduced levels of their metabolites. Therefore, given the evidence that spironolactone has active metabolites including canrenone that are more potent MR antagonists than spironolactone (Overdiek and Merkus, 1987), the net antagonistic effects of spironolactone on MR in TRD patients were reduced by this lower rate of metabolism into canrenone and other active metabolites and could have contributed to the apparent lack of MR activity.

Pharmacodynamic explanations are also possible. Grottoli and colleagues found that the stimulatory effect of canrenoate, another MR antagonist, on the activity of the HPA axis is attenuated by benzodiazepines in humans (Grottoli et al., 2002a) and suggested that the inhibitory effect of GABAergic activation by benzodiazepines overrides the stimulatory effect of mineralocorticoid blockade by canrenoate on the HPA axis (Grottoli et al., 2002b). Since three-quarters of our TRD sample were receiving benzodiazepines in some degree, the stimulatory effects on salivary cortisol by spironolactone and canrenone could have been attenuated by this mechanism. It would seem likely therefore that a combination of pharmacokinetic and pharmacodynamic effects played an important role in the lack of stimulatory effect of spironolactone in this group of TRD patients.

One previous study (Young et al., 2003) has looked at the response of the HPA axis to spironolactone in depression and found a heightened cortisol response indicative of increased MR receptor sensitivity. It is important to note that there are significant differences in the sample studied by Young et al. (2003), in that their patients were not treatment resistant, were younger, were less severely depressed, were less chronically ill, had lower rates of comorbidity and were mostly drug free.

Another important factor to note is that we assessed a group of severe treatment-resistant hypercortisolaemic depressed patients, with a high incidence of childhood maltreatment, with a mean duration of 52.5 (9.5) months in a current episode (75% more than 24 months). The degree of hypercortisolaemia was marked and there is a possibility that there was a ceiling effect such that further activation of the HPA axis in response to MR blockade was difficult to produce. Thus, these differences in the psychopathological characteristics and the degree of endocrine dysfunction are additional possible explanations for the different results after spironolactone between our study and that of Young et al. (2003).

There is evidence that the expression and role of MRs in the hippocampus is controlled by the stimulus of 5-hydroxytryptamine (5-HT) receptors (Semont, et al., 1999). Stressful stimuli increase 5-HT release and turnover in the hippocampus and we can suggest that some of the alterations in mineralocorticoid and GR expression may be facilitated, in part at least, by the increase in 5-HT (Storey and Balfour, 2002). These findings have led some researchers to propose that post-synaptic 5-HT_1A and 5-HT_2A receptors have functionally contrasting outcomes that an imbalance of these receptors may be causal to the pathophysiology of stress, melancholic depression, anxiety disorders and impusivity and that re-establishing this balance is essential for therapeutic action (Young, et al., 2003).

One application of the PST to date has been to demonstrate an increase in negative feedback induced by acute antidepressant treatment (Pariante et al., 2004). Healthy individuals were treated with the SSRI citalopram (20 mg/day) for 4 days, with the PST administered before and after citalopram treatment. The authors found that citalopram treatment increased the degree of suppression induced by prednisolone from approximately 22% before citalopram to 45% after citalopram. Thus, the antidepressant was able to amplify glucocorticoid-mediated negative feedback on the HPA axis after just 4 days of treatment. These data support the idea that one of the mechanisms by which antidepressants exercise their effects is by normalizing HPA axis hyperactivity due to an increased function of the corticosteroid receptors (Pariante et al., 2004).

Effect of blocking the MR activity of prednisolone on the endocrine effects of prednisolone

The evidence summarized suggests that prednisolone is similar to cortisol in its ability to probe both the GR and the MR and theoretically provides a more naturalistic probe than DEX. Indeed, prednisolone is particularly useful in examining the suppression of salivary cortisol (Pariante et al., 2002), which represents the bioavailable fraction (5–10%) of plasma cortisol and, therefore, more accurately reflects the hormone that reaches and binds the corticosteroid receptors (Kirschbaum et al., 1994).

As described earlier, one hypothesis being tested was that, in the absence of MR effects, prednisolone would produce similar suppressive effects on cortisol to the pure GR agonist DEX. In healthy controls, blocking MR activity alongside the administration of prednisolone resulted in less suppression than when prednisolone was administered alone. By contrast, in TRD patients the findings indicate that there was no effect of blocking MR activity on the effects of prednisolone. Thus, contrary to our hypothesis, patients did not become resistant to prednisolone suppression in the presence of a MR antagonist. However, the factors discussed in the previous paragraphs that may have limited the effects of spironolactone also apply to the interpretation of the results here. It is likely that the blocking effects of spironolactone on the MR in TRD patients were reduced by lower conversion into canrenone or/and masked by benzodiazepines. For these reasons, the present results do not alIow a definitive answer to the question as to whether the additional MR effects of prednisolone account for the different responses observed in TRD patients to DEX and prednisolone.

Attracted by the hypothesis that amplified efficacy of MR signalling is a first important step in antidepressant action, a clinical trial was conducted in which the MR antagonist spironolactone or a placebo was dispensed under controlled conditions to depressed patients who were being treated with amitriptyline. The patients who were given spironolactone as an add-on responded less positively to amitriptyline than those who were given placebo. Spironolactone triggers the HPA system by blocking hippocampal MRs (Holsboer et al., 1999).

The results of our previous studies (Juruena et al, 2006, 2009) raised the possibility that up-regulated MR receptors in TRD could be compensating for down-regulated GR receptors, hence explaining why the suppressive response to prednisolone was normal whereas the response to DEX was attenuated. By co-administering the MR antagonist spironolactone with the dual GR and MR receptor agonist prednisolone, we aimed to disentangle the MR effects of prednisolone from its GR effects. We have not been able to demonstrate directly that MR receptors are upregulated in TRD, although as discussed above there may be other pharmacokinetic and pharmacodynamic reasons for this, and it may be that it is the relative balance between MR and GR that is most important. We do note that a previous study using intravenous prednisolone found that predisolone-mediated fast feedback was only reduced by GR antagonist pretreatment and not by MR antagonism, suggesting a predominantly GR-mediated pathway. The intravenous PST provides another potential route to investigate HPA abnormalities underlying metabolic and psychiatric disease states (Russell et al., 2010). Thus, we propose that the PST may offer specific biological and clinical information, related to its action at both the GR and the MR (Juruena et al., 2010a, 2010b).

MR in subtypes of depression

As described earlier, a study that has investigated the effects of spironolactone in depressed patients has found that the response to spironolactone is increased in these patients compared to controls (Young et al., 2003). It is intriguing to speculate that the subgroup of depressed patients with a chronic course could have different MR sensitivity to non-chronic depressed patients (i.e. those assessed by Young et al., 2003) independently of the degree of treatment resistance present. The PST was previously administered to 18 TRD patients (Juruena et al., 2006). These previously studied patients had been in a depressive episode for a shorter period (24 months compared to 52 months in the present study) and were less likely to have been chronically depressed (33% had been depressed 2 years or more compared to 75% in the present study).

Catalán et al. (1998) studied endocrine differences in patients affected with major depression, dysthymic disorder and healthy subjects. They found that patients with major depressive disorder exhibited the highest plasma cortisol levels; dysthymic disorder values were very similar to those in mild or moderate depression and the lowest values were seen in the healthy subjects (Catalan et al., 1998). Oshima et al. (2000) studied the combined DEX/CRH test on patients with major depressive and dysthymic disorders and healthy controls showing that cortisol responses in dysthymia were identical to those in normal controls (Oshima et al., 2000). Watson et al. (2002) noted that the response to DEX suppression in patients with chronic depression had earlier been shown to be normal. They found that chronically depressed patients showed a normal cortisol response both to the DST and to the DEX/CRH test (Watson et al., 2002). O’Keane and colleagues described results suggesting that chronic depression, in contrast to acute melancholic depression, might be characterized by increased ACTH in response to CRH challenge (O’Keane et al., 2005).

Limitations

Some limitations must be considered in this study and measures to help further elucidate the role of MR function in depression and in TRD. First, it will be important to examine the results after MR antagonism in a larger sample, particularly to investigate the effect of certain illness characteristics such as chronicity on the MR. It might also be useful to allow comparison of male and female subjects to ascertain whether sex steroids and menopausal status can influence HPA axis dysfunction. Second, other hormones like ACTH and aldosterone, which is the most selective hormone to bind to MR, could be measured concomitantly to improve the overall assessment of MR sensitivity and function (Grossmann et al., 2004). Third, no GR antagonist was administered in this study. In future studies, we need to dissect the mechanism underlying HPA axis dysregulation by testing the effects of co-treatment with GR and MR antagonists in the same depressed patients, testing the theory that there is imbalance in the MR/GR ratio. Interestingly, whilst a reduction in both MR and GR function has been described in patients with major depression and both MR and GR are influenced by antidepressant treatment (Holsboer, 2000; Reul et al., 2000), little evidence exists about changes in the balance between GR and MR.

Conclusions

This study adds to the growing body of work suggesting that in patients with severe TRD there is marked hypercortisolism (Bauer et al., 2003; Juruena et al., 2006, 2009). Although relatively little studied, data are also emerging to suggest that structural and functional brain changes are also associated with treatment resistance, helping to establish that there is a significant biological element to the depression in a proportion of these patients. The precise role of MR and GR in the HPA axis dysfunction of depression, and in the aetiology of treatment resistance, remains unclear, although evidence is emerging that the degree of HPA axis dysfunction is a determinant of the degree of treatment resistance in some severely depressed patients (Juruena et al., 2009; Markopoulou et al., 2009). This has a potential role in improving the treatment options available for patients who are resistant to our current therapeutic options and is, we believe, a promising avenue for future translational research to that end.

Footnotes

Conflict of interest

The authors declare no conflict of interest.

Funding

This research has been supported by a 2003 and a 2005 NARSAD Young Investigator Award and a 2004 MRC Clinician Scientist Fellowship to C.M. Pariante; by a 2003 CAPES Fellowship Award and a 2006 NARSAD Young Investigator Award to M.F. Juruena; and by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust & [Institute of Psychiatry] Kings’ College London; CNPq and FAPESP grants. This report presents independent research partly funded by the National Institute for Health Research (NIHR). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

References

American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Press.

Arvat

Maccagno

Giordano

. (2001) Mineralocorticoid receptor blockade by canrenoate increases both spontaneous and stimulated adrenal function in humans. J Clin Endocrinol Metab 86: 3176–3181.

Barden

Stec

Montkowski

. (1997) Endocrine profile and neuroendocrine challenge tests in transgenic mice expressing antisense RNA against the glucocorticoid receptor. Neuroendocrinology 66: 212–220.

Bauer

Papadopoulos

Poon

. (2003) Altered glucocorticoid immunoregulation in treatment resistant depression. Psychoneuroendocrinology 28: 49–65.

Beck

Steer

Ball

. (1996) Comparison of Beck Depression Inventories -IA and -II in psychiatric outpatients. J Pers Assess 67: 588–597.

Bifulco

Bernazzani

Moran

. (2005) The childhood experience of care and abuse questionnaire (CECA.Q): validation in a community series. Br J Clin Psychol 44: 563–581.

Bifulco

Brown

Harris

(1994) Childhood Experience of Care and Abuse (CECA): a retrospective interview measure. J Child Psychol Psychiatry 35: 1419–1435.

Catalan

Gallart

Castellanos

. (1998) Plasma corticotropin-releasing factor in depressive disorders. Biol Psychiatry 44: 15–20.

Cornelisse

Joels

Smeets

(2011) A randomized trial on mineralocorticoid receptor blockade in men: effects on stress responses, selective attention, and memory. Neuropsychopharmacology 36: 2720–2728.

10.

De Kloet

Joels

Holsboer

(2005) Stress and the brain: from adaptation to disease. Nat Rev Neurosci 6: 463–475.

11.

De Kloet

Vreugdenhil

Oitzl

. (1998) Brain corticosteroid receptor balance in health and disease. Endocr Rev 19: 269–301.

12.

Deuschle

Weber

Colla

. (1998) Mineralocorticoid receptor also modulates basal activity of hypothalamus-pituitary-adrenocortical system in humans. Neuroendocrinology 68: 355–360.

13.

Dodt

Kern

Fehm

. (1993) Antimineralocorticoid canrenoate enhances secretory activity of the hypothalamus-pituitary-adrenocortical (HPA) axis in humans. Neuroendocrinology 58: 570–574.

14.

First

Spitzer

Gibbon

. (1995) The Structured Clinical Interview for DSM-III-R Personality-Disorders (SCID-II) 2. Multisite test-retest reliability study. J Personal Disord 9: 92–104.

15.

First

Spitzer

Gibbon

. (1997) Structured Clinical Interview for DSM-IV Axis I Disorders- Clinician Version (SCID-CV). Washington, DC: American Psychiatric Press.

16.

Flores

Kenna

Keller

. (2006) Clinical and biological effects of mifepristone treatment for psychotic depression. Neuropsychopharmacology 31: 628–636.

17.

Grossmann

Freudinger

Mildenberger

. (2004) Evidence for epidermal growth factor receptor as negative-feedback control in aldosterone-induced Na+ reabsorption. Am J Physiol Renal Physiol 286: F1226–F1231.

18.

Grottoli

Giordano

Maccagno

. (2002a) The stimulatory effect of canrenoate, a mineralocorticoid antagonist, on the activity of the hypothalamus-pituitary-adrenal axis is abolished by alprazolam, a benzodiazepine, in humans. J Clin Endocrinol Metab 87: 4616–4620.

19.

Grottoli

Maccagno

Ramunni

. (2002b) Alprazolam, a benzodiazepine, does not modify the ACTH and cortisol response to hCRH and AVP, but blunts the cortisol response to ACTH in humans. J Endocrinol Invest 25: 420–425.

20.

Halbreich

Asnis

Shindledecker

. (1985) Cortisol secretion in endogenous depression. II. Time-related functions. Arch Gen Psychiatry 42: 909–914.

21.

Heuser

Deuschle

Weber

. (2000) Increased activity of the hypothalamus-pituitary-adrenal system after treatment with the mineralocorticoid receptor antagonist spironolactone. Psychoneuroendocrinology 25: 513–518.

22.

Holsboer

(1999) The rationale for corticotropin-releasing hormone receptor (CRH-R) antagonists to treat depression and anxiety. J Psychiatr Res 33: 181–214.

23.

Holsboer

(2000) The corticosteroid receptor hypothesis of depression. Neuropsychopharmacology 23:477–501.

24.

Joels

Karst

DeRijk

. (2008) The coming out of the brain mineralocorticoid receptor. Trends Neurosci 31: 1–7.

25.

Juruena

Cleare

Pariante

(2004) The hypothalamic pituitary adrenal axis, glucocorticoid receptor function and relevance to depression. Rev Bras Psiquiatr 26: 189–201.

26.

Juruena

Cleare

Papadopoulos

. (2006) Different responses to dexamethasone and prednisolone in the same depressed patients. Psychopharmacology 189: 225–235.

27.

Juruena

Pariante

Papadpoulos

. (2009) Prednisolone suppression test in depression: prospective study of the role of HPA axis dysfunction in treatment resistance. Br J Psychiatry 194: 342–349.

28.

Juruena

M.F

Pariante

Papadopoulos

. ( 2010a) The development and application of the prednisolone suppression test in psychiatry: a novel tool for assessing glucocorticoid and mineralocorticoid receptor function. Mind Brain 1: 115–122.

29.

Juruena

Cleare

Papadopoulos

. (2010b) The prednisolone suppression test in depression: dose–response and changes with antidepressant treatment. Psychoneuroendocrinology 35: 1486–1491.

30.

Kellner

Baker

Yassouridis

. (2002) Mineralocorticoid receptor function in patients with posttraumatic stress disorder. Am J Psychiatry 159: 1938–1940.

31.

Kirschbaum

Hellhammer

(1994) Salivary cortisol in psychoneuroendocrine research: recent developments and applications. Psychoneuroendocrinology 19: 313–333.

32.

Kocarek

Schuetz

Strom

. (1995) Comparative analysis of cytochrome P4503A induction in primary cultures of rat, rabbit, and human hepatocytes. Drug Metab Dispos 23: 415–421.

33.

Kuningas

De Rijk

Westendorp

. (2007) Mental performance in old age dependent on cortisol and genetic variance in the mineralocorticoid and glucocorticoid receptors. Neuropsychopharmacology 32: 1295–1301.

34.

Lembke

Gomez

Tenakoon

. (2013) The mineralocorticoid receptor agonist, fludrocortisone, differentially inhibits pituitary-adrenal activity in humans with psychotic major depression. Psychoneuroendocrinology 38: 115–121.

35.

Lopez

Chalmers

Little

. (1998) A.E. Bennett Research Award. Regulation of serotonin1A, glucocorticoid, and mineralocorticoid receptor in rat and human hippocampus: implications for the neurobiology of depression. Biol Psychiatry 43: 547–573.

36.

McEwen

(2000) Allostasis and allostatic load: Implications for neuropsychopharmacology. Neuropsychopharmacology 22: 108–124.

37.

Markopoulou

Papadopoulos

Juruena

. (2009) The ratio of cortisol/DHEA in treatment resistant depression. Psychoneuroendocrinology 34: 19–26.

38.

Nishi

Kawata

(2007) Dynamics of glucocorticoid receptor and mineralocorticoid receptor: implications from live cell imaging studies. Neuroendocrinology 85: 186–192.

39.

O’Keane

Dinan

Scott

. (2005) Changes in hypothalamic-pituitary-adrenal axis measures after vagus nerve stimulation therapy in chronic depression. Biol Psychiatry 58: 963–968.

40.

Oshima

Yamashita

Owashi

. (2000) The differential ACTH responses to combined dexamethasone/CRH administration in major depressive and dysthymic disorders. J Psychiatr Res 34: 325–328.

41.

Otte

Hinkelmann

Moritz

. (2010) Modulation of the mineralocorticoid receptor as add-on treatment in depression: a randomized, double-blind, placebo-controlled proof-of-concept study. J Psychiatr Res 44: 339–346.

42.

Overdiek

Merkus

(1987) The metabolism and biopharmaceutics of spironolactone in man. Rev Drug Metab Drug Interact 5: 273–302.

43.

Pace

Spencer

(2005) Disruption of mineralocorticoid receptor function increases corticosterone responding to a mild, but not moderate, psychological stressor. Am J Physiol Endocrinol Metab 288: E1082–E1088.

44.

Pariante

Papadopoulos

Poon

. (2002) A novel prednisolone suppression test for the hypothalamic-pituitary- adrenal axis. Biol Psychiatry 51: 922–930.

45.

Pfohl

Sherman

Schlechte

. (1985) Differences in plasma ACTH and cortisol between depressed patients and normal controls. Biol Psychiatry 20: 1055–1072.

46.

Reul

Gesing

Droste

. (2000) The brain mineralocorticoid receptor: greedy for ligand, mysterious in function. Eur J Pharmacol 405: 235–249.

47.

Reul

JMHM

De Kloet

(1985) Two receptor systems for corticosterone in rat brain: Microdistribution and differential occupation. Endocrinology 117: 2505–2511.

48.

Ribeiro

Tandon

Grunhaus

. (1993) The DST as a predictor of outcome in depression: a meta-analysis. Am J Psychiatry 150: 1618–1629.

49.

Rupprecht

Reul

Van Steensel

. (1993) Pharmacological and functional characterization of human mineralocorticoid and glucocorticoid receptor ligands. Eur J Pharmacol 247: 145–154.

50.

Russell

Henley

Leendertz

. (2010) Glucocorticoid receptor-mediated inhibition of hypothalamic–pituitary–adrenal ultradian activity in healthy males. J Neurosci 30: 6106 – 6115.

51.

Sackeim

Haskett

Mulsant

. (2001) Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA 285: 1299–1307.

52.

Schuetz

Brimer

Schuetz

(1998) Environmental xenobiotics and the antihormones cyproterone acetate and spironolactone use the nuclear hormone pregnenolone X receptor to activate the CYP3A23 hormone response element. Mol Pharmacol 54: 1113–1117.

53.

Spencer

Kim

Kalman

. (1998) Evidence for mineralocorticoid receptor facilitation of glucocorticoid receptor-dependent regulation of hypothalamic-pituitary-adrenal axis activity. Endocrinology 139: 2718–2726.

54.

Thase

Rush

(1997) When at first you don’t succeed: sequential strategies for antidepressant nonresponders. J Clin Psychiatry 58 (Suppl. 13) :23–29.

55.

Watson

Gallagher

Del-Estal

. (2002) Hypothalamic-pituitary-adrenal axis function in patients with chronic depression. Psychol Med 32: 1021–1028.

56.

Wellhoener

Born

Fehm

. (2004) Elevated resting and exercise-induced cortisol levels after mineralocorticoid receptor blockade with canrenoate in healthy humans. J Clin Endocrinol Metab 89: 5048–5052.

57.

Williams

JBW

(1988) Structured Interview Guide for the Hamilton-Depression And Anxiety Scales (SIGH-AD). New York: New York State Psychiatric Institute.

58.

Young

Gallagher

Watson

. (2004) Improvements in neurocognitive function and mood following adjunctive treatment with mifepristone (RU-486) in bipolar disorder. Neuropsychopharmacology 29: 1538–1545.

59.

Young

Lopez

Murphy-Weinberg

. (1998) The role of mineralocorticoid receptors in hypothalamic-pituitary- adrenal axis regulation in humans. J Clin Endocrinol Metab 83: 3339–3345.

60.

Young

Lopez

Murphy-Weinberg

. (2003) Mineralocorticoid receptor function in major depression. Arch Gen Psychiatry 60: 24–28.

61.

Zhou

Xue

. (2007) Clinically important drug interactions potentially involving mechanism-based inhibition of cytochrome P450 3A4 and the role of therapeutic drug monitoring. Ther Drug Monit 29: 687–710.