Effect of antidepressant drugs on the hypothalamic-pituitary-adrenal axis activity and glucocorticoid receptor function

Hyperactivity of hypothalamic-pituitary-adrenal (HPA) axis is the main biochemical change, besides disturbed monoaminergic neurotransmission, observed in the patients suffering from a major depression. High incidence of depression in Cushing's syndrome as well as antidepressant effects of adrenocortical enzyme inhibitors in major depression support hypothesis that hyperactivity of HPA axis may be involved in pathogenesis of depression.     

Post-receptor-mediated increases in adenylate cyclase activity after chronic antidepressant treatment: relationship to receptor desensitization

Administration to rats of chronic electroconvulsive shock (ECS) or chronic desipramine (DMI, 10 mg/kg daily i.p. for 3 weeks) did not affect either basal or forskolin-stimulated adenylate cyclase activity in membranes prepared from the caudate nucleus. In cerebellar membranes prepared from rats which had received chronic ECS, forskolin-stimulated activity was significantly increased compared to activity in sham- or single ECS treated rats. Forskolin-stimulated adenylate cyclase was increased in both hippocampal and cerebellar membranes from rats which received chronic DMI, compared to saline-treated animals. In cerebellar membranes, increases comparable to those with forskolin were also obtained with guanylyl-5'-imidodiphosphate (GppNHp) after both treatments, while with Mn2+ ions, either alone or in the additional presence of forskolin, the changes observed were similar to those previously reported in cortical membranes. A possible mechanism for these effects was investigated by studying antidepressant-induced and in vitro desensitization of the cyclic AMP response in slice preparations from the various brain areas. In slices from caudate nucleus, chronic DMI did not alter stimulation of cyclic AMP formation by either noradrenaline or forskolin, while in cerebellar slices the noradrenaline response was significantly reduced, and in hippocampal slices both responses were reduced (heterologous desensitization). In vitro incubation of cortical slices with noradrenaline also resulted in a reduction in the response to both agents. However, in membranes prepared from the desensitized cortical slices, there was no change in the degree of activation of adenylate cyclase by either NaF or forskolin. Thus, the increase in these activities, observed in certain brain areas after chronic antidepressant treatment may not necessarily be related to beta-adrenoceptor desensitization.     

Effects of chronic antidepressant treatment on serotonin receptor activity in mice

The effect of acute and chronic treatments with conventional and atypical antidepressant drugs on serotonin receptor activity was assessed by the responsiveness of mice to the serotonin receptor agonist 5-methoxy-N,N-dimethyltryptamine. Acute treatment with 10 mg/kg of amitriptyline, imipramine, trazodone, mianserin or viloxazine reduced the head twitch response measured 1 h following a challenged dose of the serotonin agonist. Acute iprindole and desmethylimipramine, however, had no effect on the serotonergic response. Chronic treatment with the clinically effective antidepressants amitriptyline, imipramine, desmethylimipramine, iprindole, and trazodone produced an enhanced responsiveness to 5-MeODMT. The enhanced responsiveness was first observed 24 h after cessation of treatment with most drugs. The effect lasted for at least 48 h. Chronic treatment with the neuroleptic haloperidol did not result in altered responsivity to the serotonin agonist. Brain accumulation of imipramine and amitriptyline and their deaminated metabolites were measured. Brain drug and metabolite levels peaked 1 h following both acute and chronic treatments. Brain accumulations of amitriptyline and its metabolite were much greater than those of imipramine and its metabolite. This pharmacokinetic data is consistent with an early (1 h) antagonism of the 5-MeODMT response and the emergence of hightened responsiveness to 5-MeODMT after chronic treatment, when brain drug levels are reduced. These findings are also consistent with the greater inhibitory effect found after treatment with amitriptyline than with imipramine. It is concluded that enhanced serotonin neurotransmission which develops during chronic treatment with antidepressant drugs may be related to the clinical action of these drugs.     

Tryptophan hydroxylase gene associated with paroxetine antidepressant activity.

The possible association of the A218C tryptophan hydroxylase (TPH) gene variant with the antidepressant activity of paroxetine was investigated in a sample of 121 inpatients affected by a major depressive episode and treated with paroxetine 20-40 mg with either placebo or pindolol in a double blind design for 4 weeks. The severity of depressive symptoms was weekly assessed with the Hamilton Rating Scale for Depression. TPH allelic variants were determined in each subject using a PCR-based technique. TPH*A/A and TPH*A/C variants were associated with a poorer response to paroxetine treatment when compared to TPH*C/C (P=0.005); this difference was not present in the pindolol augmented group. Other variables, such as sex, diagnosis, presence of psychotic features, severity of depressive symptomatology at baseline and paroxetine plasma level, were not associated with the outcome. TPH gene variants are therefore a possible modulator of paroxetine antidepressant activity.     

Increased GABAB receptor function in mouse frontal cortex after repeated administration of antidepressant drugs or electroconvulsive shocks.

1 Addition of baclofen to a medium containing slices of mouse frontal cortex inhibited the potassium-evoked release of 5-hydroxytryptamine (5-HT) in a concentration-dependent manner. The degree of inhibition was increased in frontal cortex tissue taken from animals treated for 14 days with amitriptyline (10 mg kg-1, twice daily) at all concentrations of baclofen tested (10(-6) M-10(-4) M). 2 Administration of either desipramine, mianserin or zimeldine (10 mg kg-1 daily) for 14 days also approximately doubled the degree of inhibition evoked by addition of baclofen (10(-5) M) to the medium. 3 One day of treatment with the antidepressant drugs did not alter the inhibitory effect of baclofen on K+-evoked 5-HT release. 4 Addition of the antidepressant drugs to the medium had no effect on the K+-evoked release of 5-HT. 5 Repeated administration of electroconvulsive shock (5 seizures spread out over 10 days), like amitriptyline, produced a significant enhancement of the baclofen-induced inhibition of 5-HT release over the range of baclofen concentrations studied. A single electroconvulsive shock had no effect. 6 These data suggest that repeated administration of the antidepressant drugs or electroconvulsive shock increases the function of the gamma-aminobutyric acid (GABA)B receptor in the frontal cortex modulating 5-HT release and are consistent with the finding of increased GABAB receptor number in this region following various antidepressant treatments.     

Effect of antidepressant drugs on the human corticotropin-releasing-hormone gene promoter activity in neuro-2A cells

In order to test if antidepressant drugs can directly influence corticotropin-releasing hormone (CRH) gene expression, their effect on CRH gene promoter activity was evaluated in neuro-2A cells stably transfected with a human CRH - chloramphenicol acetyl-transferase plasmid. Forskolin (an activator of adenylate cyclase), but not phorbol 12-myristate 13-acetate (an activator of protein kinase C), ca. 3-fold increased reporter gene activity, which confirms the critical role of the cAMP-responsive element in regulation of the CRH gene. Imipramine and fluoxetine present in the medium for 5 days, in a concentration-dependent manner (3-30 microM) inhibited the basal activity of CRH gene promoter, while tianeptine was inactive. The obtained results indicate that inhibition of the human CRH gene promoter activity by imipramine and fluoxetine, but not tianeptine, may play a role in a mechanism by which the former drugs attenuate HPA axis activity.     

Antidiabetic activity of passive nonsteroidal glucocorticoid receptor modulators

Much has been learned about the consequences of glucocorticoid receptor antagonism by studying steroidal active antagonists such as RU-38486 (1). In the liver glucocorticoid receptor antagonism suppresses hepatic glucose production decreasing plasma glucose levels; however, extrahepatic antagonism produces several undesirable side effects including activation of the hypothalamic pituitary adrenal axis. A series of nonsteroidal passive N-(3-dibenzylamino-2-alkyl-phenyl)-methanesulfonamide glucocorticoid receptor modulators was discovered. Liver selective and systemically available members of this series were found and characterized in diabetes and side effect rodent models. A highly liver selective member of this series, acid 14, shows efficacy in the ob/ob model of diabetes. It lowers plasma glucose, cholesterol, and free fatty acid concentrations and reduces the rate of body weight gain. The structurally related systemically available passive modulator 12 lowers glucose, HbA(1c), triglyceride, free fatty acid, and cholesterol levels. Interestingly, it did not acutely activate the hypothalamic pituitary adrenal axis in unstressed CD-1 mice or have the abortive effects observed with 1. These results indicate that passive GR antagonists may have utility as antidiabetic agents.     

Increased tyrosine hydroxylase activity in central adrenaline neurons after reserpine treatment

By using preferential microdissection of the adrenaline (A) and noradrenaline (NA) neurons within the A2-C2 region of the rat medulla oblongata, it was possible to study the biochemical response of these two neuronal populations to reserpine administration. Three days after reserpine injections (10 mg/kg s.c., per day for 3 days), tyrosine hydroxylase (TH) activity was increased in the adrenergic C2 region whilst no change was observed in the noradrenergic A2 region. The response of the A neurons to reserpine was of lesser magnitude than the increase in TH activity observed under the same conditions in the NA neurons of the locus coeruleus and of the A5 region, and was likely to have originated in the A cell bodies. In contrast with previous studies, this work suggests that the A-containing neurons are responsive to reserpine administration, despite the lack of change in phenylethanolamine-N-methyltransferase activity.     

Increased central 5-hydroxytryptamine receptor mechanisms in rats after chronic neuroleptic treatment

1 The behavioural responses of drugs known to act through central 5-hydroxytryptamine (5-HT) mechanisms have been investigated in rats receiving a neuroleptic (trifluoperazine) in their drinking water for 4 to 6 months.

2 5-Hydroxytryptophan (5-HTP) induced 5-HT-dependent behaviours including head bobbing and lateral head weaving, reciprocal forepaw treading, tremor, backward walking, body writhing and `wet-dog' shakes. In doses of 50 to 150 mg/kg, 5-HTP induced more intense behavioural effects in neuroleptic-treated rats than in the control animals.

3 Similarly the putative 5-HT agonist, quipazine (1 to 20 mg/kg) and the 5-HT releasing drug, fenfluramine (5 to 20 mg/kg), both induced significantly greater motor responses in the chronically neuroleptic-treated rats.

4 A 5-HT uptake inhibitor (femoxetine, 2.5 to 10 mg/kg) had little behavioural effect in either control or trifluoperazine-treated rats.

5 Total specific high-affinity binding of radiolabelled 5-HT was significantly increased in crude membrane fractions prepared from the cortex, striatum and substantia nigra of neuroleptic-treated rats compared to control animals.

6 High-affinity uptake of radiolabelled 5-HT into striatal slices was similar in experimental and control animals.

7 Behavioural and biochemical data would indicate that postsynaptic 5-HT mechanisms are enhanced in rats treated chronically with trifluoperazine. Chronic neuroleptic therapy may thereby induce cerebral 5-HT receptor supersensitivity in addition to the well-documented cerebral dopamine receptor supersensitivity.

     

Increased 5-HT2A receptor expression and function following central glucocorticoid receptor knockdown in vivo

Central glucocorticoid receptor function may be reduced in depression. In vivo modelling of glucocorticoid receptor underfunctionality would assist in understanding its role in depressive illness. The role of glucocorticoid receptors in modulating 5-HT(2A) receptor expression and function in the central nervous system (CNS) is presently unclear, but 5-HT(2A) receptor function also appears altered in depression. With the aid of RNAse H accessibility mapping, we have developed a 21-mer antisense oligodeoxynucleotide (5'-TAAAAACAGGCTTCTGATCCT-3', termed GRAS-5) that showed 56% reduction in glucocorticoid receptor mRNA and 80% down-regulation in glucocorticoid receptor protein in rat C6 glioma cells. Sustained delivery to rat cerebral ventricles in slow release biodegradable polymer microspheres produced a marked decrease in glucocorticoid receptor mRNA and protein in hypothalamus (by 39% and 80%, respectively) and frontal cortex (by 26% and 67%, respectively) 5 days after a single injection, with parallel significant up-regulation of 5-HT(2A) receptor mRNA expression (13%) and binding (21%) in frontal cortex. 5-HT(2A) receptor function, determined by DOI-head-shakes, showed a 55% increase. These findings suggest that central 5-HT(2A) receptors are, directly or indirectly, under tonic inhibitory control by glucocorticoid receptor.     

Increased tilt-cage activity after serotonin depletion by 5,7-dihydroxytryptamine.

Administration of 5,7-dihydroxytryptamine (5,7-DHT, 8 μg) into the medial forebrain bundle produced a 78% decrease in forebrain serotonin (5-HT) and a 2-fold increase in 24 hr tilt-cage activity. Intraventricular administration of 5.7-DHT (200 μg) resulted in a 4-fold increase in 24 hr activity and 5-HT depletions of 83%, 58% and 71% in forebrain, brainstem and spinal cord, respectively. The diurnal index of activity was identical for 5,7-DHT and control groups, suggesting preservation of the normal light-dark activity rhythm which contrasts with the disruption of this rhythm reported in p-chlorophenylalanine (p-CPA)-induced hyperactivity. Furthermore, unlike rats with median raphe lesions, 5,7-DHT-treated-rats were hypoactive when tested in the open-field. Administration of p-CPA produced a dose-dependent increase in tilt-cage activity in normal rats, but had relatively little effect on activity in 5,7-DHT-treated-rats. It is concluded that different methods of 5-HT depletion can result in different effects on various behavioral measures, but that this does not appear to be the case for tilt-cage hyperactivity. That is, this effect occurs following electrolytic raphe lesions, p-CPA or 5,7-DHT. It is, therefore, suggested that this effect is not the result of non-specific damage but rather is due to subnormal 5-HT synaptic activity in the central nervous system.     

Novel glucocorticoid receptor agonists in the treatment of asthma

Introduction: Inhaled corticosteroids are the only drugs that effectively suppress the airway inflammation, but they can induce considerable systemic and adverse effects when they are administered chronically at high doses. Consequently, the pharmaceutical industry is still searching for newer entities with an improved therapeutic index.

Areas covered: Herein, the authors review the research in the glucocorticoid field to identify ligands of the glucocorticoid receptor (GR). These ligands preferentially induce transrepression with little or no transactivating activity, in order to have a potent anti-inflammatory action and a low side-effects profile.

Expert opinion: Several agents have been synthesized, but few have been tested in experimental models of asthma. Furthermore, only three (BI-54903, GW870086X and AZD5423) have entered clinical development, although the development of at least one of them (BI-54903) was discontinued. The reason for the limited success so far obtained is that the model of transactivation versus transrepression is a too simplistic representation of GR activity. It is difficult to uncouple the therapeutic and harmful effects mediated by GR, but some useful information that might change the current perspective is appearing in the literature. The generation of gene expression ‘fingerprints’ produced by different GR agonists in target and off-target human tissues could be useful in identifying drug candidates with an improved therapeutic ratio.