Francis Chaouloff

Physiopharmacological interactions between stress hormones and central serotonergic systems

The present review tries to delineate some mechanisms through which the sympathetic nervous system (SNS) and the hypothalamo-pituitary-adrenal (HPA) interact with central serotonergic systems. The recent progress in 5-hydroxytryptamine (5-HT) receptor pharmacology has helped to define the means by which central serotonergic activity may alter the respective activities of the SNS (sympathetic nerves and adrenomedulla) and of the HPA axis. These pharmacological findings have also helped to characterize the differential effects of central 5-HT upon different branches of the SNS and the numerous sites at which 5-HT exerts stimulatory influences upon the HPA axis. Although relevant to stress-related neuroendocrinology, the extent to which these interactions are involved in the antidepressant/anxiolytic properties of some serotonergic agents still remains to be clarified. Beside these findings, there is also abundant evidence for a tight control of central serotonergic systems by stress hormones. Activation of the SNS increases, by numerous means, central availability of tryptophan, whereas glucocorticoids exert differential actions upon the intra- and the extraneuronal regulation of 5-HT function. Actually, a significant number of these mechanisms is involved in the maintenance of homeostasis during stressful events, thereby conferring to these mechanisms a key role in adaptation processes.

Effects of the 5-HT1c/5-HT2 receptor agonists DOI and α-methyl-5-HT on plasma glucose and insulin levels in the rat

Abstract Administration of the 5-HT 1C /5-HT 2 receptor agonist 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI, 0.125–2.0 mg/kg i.v.) triggered dose-dependent increases in plasma insulin levels remained unchanged. Pretreatment with the 5-HT 1C /5-HT 2 receptor antagonists LY 53857, risanserin, or the mixed 5-HT 2 / α 1 -adrenoceptor antagonist ketanserin either diminished or prevented the hyperglycemic effect of DOI (0.5 mg/kg). Administration of the mixed 5-HT 1C receptor agonists/5-HT 2 receptor antagonists 1-(3-chlorophenyl)-piperazine (mCPP) or 1-(3-(trifluoromethyl) phenyl)piperazine (TFMPP) did not affect plasma glucose levels. However, pretreatment with mCPP or TFMPP decreased DOI-induced hyperglycemia in a dose-dependent manner. The α 2 -adrenoceptor antagonist idazoxan and the ganglionic blocker hexamethonium both decreased DOI-induced hyperglycemia, whilst the α 1 -adrenoceptor antagonist prazosin amplified the rise in plasma glucose elicited by DOI. The peripherally acting 5-HT 1C /5-HT 2 receptor agonist α-methyl-5-HT (0.5–1.0 mg/kg i.v.) triggered a rise in plasma glucose levels that was associated with an increase in plasma insulin levels. Pretreatment with LY 53857 diminished α-methyl-5-HT-induced hyperglycemia. These data indicate that 5-HT 2 receptors, but not 5-HT 1C receptors, and catecholaminergic systems, mediate DOI-induced hyperglycemia. Moreover, it is suggested that the inhibition of insulin release by DOI is centrally mediated, and that activation of peripheral 5-HT 2 receptors may affect glycemia.

Ganglionic transmission is a prerequisite for the adrenaline-releasing and hyperglycemic effects of 8-OH-DPAT

The present study concerned the influence of ganglionic nicotinic blockade (by hexamethonium, 5 mg/kg and 10 mg/kg) of the 8-OH-DPAT-induced elevations in plasma adrenaline, corticosterone and glucose levels in conscious rats. Prior treatment with hexamethonium dose dependently decreased the basal glucose levels and increased the basal corticosterone levels but did not significantly affect the basal adrenaline levels. Hexamethonium decreased in a dose-dependent manner the elevations in plasma adrenaline, corticosterone and glucose levels that were elicited by the 5-HT1A receptor agonist, 8-OH-DPAT (0.25 mg/kg). These results indicate that acetylcholine release is a prerequisite for both the adrenaline-releasing and hyperglycemic effects of 8-OH-DPAT. The data also point to the possibility that adrenaline release participates in the elevation in plasma corticosterone levels that is elicited by 5-HT1A receptor activation.

Mechanisms involved in the hyperglycemic effect of the 5-HT1C/5-HT2 receptor agonist, DOI

Previous experiments have indicated that 5-HT2 receptors and catecholaminergic systems mediate the rise in plasma glucose levels elicited by acute administration of the 5-HT1c/5-HT2 receptor agonist, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI). On this basis, we investigated the location of these serotonin receptors and the nature of this catecholaminergic involvement. Administration of DOI (0.4 mg/kg i.v.) to conscious rats (bearing jugular catheters) elicited a rapid rise in plasma glucose which was associated with a decreased insulin response to a glucose bolus (300 mg/kg i.v.). Pretreatment with the peripherally acting 5-HT1c/5-HT2 receptor antagonist, BW 501C67 (0.5 mg/kg i.v. 10 min beforehand) prevented the rise in plasma glucose triggered by the peripherally acting 5-HT1c/5-HT2 receptor agonist, alpha-methyl-5-HT (0.75 mg/kg i.v.), but amplified the rise elicited by DOI. Pretreatment with chlorisondamine (1 mg/kg i.v. 10 min beforehand) or adrenalectomy 20 h beforehand prevented the DOI-induced hyperglycemia. On the other hand, pretreatment with dexamethasone (0.35 mg/kg s.c. 2 h and 20 min beforehand) did not affect the DOI-induced hyperglycemia. It is concluded that the hyperglycemic effect of DOI administration is mediated by centrally located 5-HT2 receptors and, in turn, adrenal epinephrine release.

Effects of chlorisondamine and restraint on cortical [3H]ketanserin binding, 5-HT2A receptor-mediated head shakes, and behaviours in models of anxiety

A recent study has indicated that ganglionic transmission mediates acute restraint-elicited increases in brain tryptophan (5-HT precursor) levels, 5-HT synthesis and (possibly) release. Because restraint-induced release of 5-HT has been shown to be associated with a paradoxical increase in cortical 5-HT2A receptor binding, we have examined the influence of 5-HT synthesis/release upon cortical 5-HT2A receptor binding and 5-HT2A receptor-mediated head shakes in 3-hr restrained rats pretreated with the ganglionic blocker chlorisondamine. In keeping with past reports regarding the effects of restraint and ganglionic blockade upon anxiety, we have also measured the behavioural effects of restraint and/or chlorisondamine in two animal models of anxiety, the elevated plus-maze and the social interaction test. Chlorisondamine pretreatment (2.5 mg/kg, 20 min beforehand) prevented restraint-elicited defaecation and body weight decreases. Although stress amplified the head shake response to the injection of the 5-HT2A/5-HT2C receptor agonist 1-(4-iodo-2,5-dimethoxyphenyl)-2-aminopropane (DOI, 1 or 2 mg/kg 2 hr after the end of restraint), cortical [3H]ketanserin binding remained unaltered. Chlorisondamine treatment was inactive, except for the amplification of the head shake response to DOI (2 mg/kg) in restrained rats. When exposed to the social interaction test, neither restraint nor chlorisondamine affected social interaction, locomotion, or rearings. In the elevated plus-maze, the percent number of open arms entered and the total number of arms entered were decreased by acute restraint, whilst chlorisondamine pretreatment was inactive.

Influence of physical exercise on 5-HT1A receptor- and anxiety-related behaviours

Treadmill exercise (1 h at 20 m/min) has been shown to increase brain serotonin (5-HT) synthesis/metabolism in 4-day trained rats. This work has analysed whether these changes are associated with (5-HT-dependent or -independent) alterations in pre- and postsynaptic 5-HT1A receptor-mediated behaviours; in addition, the controversial hypothesis that exercised rats are less anxious than sedentary rats was tested in the elevated plus-maze and the social interaction models. Acute administration of the 5-HT1A receptor agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT; 0.125-0.5 mg/kg) triggered hyperphagia, forepaw treading, and flat body posture, the amplitude of which were not affected by training or acute exercise. In the elevated plus-maze, training, but not acute exercise, reduced the number of total arm entries. In the social interaction test, trained and acutely exercised rats displayed increased locomotion and decreased defecation, compared to resting rats, but social interaction and rearings were not affected. It is suggested that the aforementioned effects of acute exercise upon 5-HT synthesis/metabolism in short-term trained rats are not associated with changes in 5-HT1A receptor- and anxiety-related behaviours.

Influence of the novel antidepressant tianeptine on neurochemical, neuroendocrinological, and behavioral effects of stress in rats

Tianeptine is a novel tricyclic agent that activates the neuronal uptake of serotonin (5-hydroxytryptamine, 5-HT). Taking into account the antidepressant effect of tianeptine in animals and humans, we have measured the influence of a pretreatment with tianeptine (10 mg/kg IP 1 hr beforehand) on some consequences of a single 2-hr restraint stress session in male rats. Thus, we have analyzed (1) 5-HT metabolism in various brain regions and plasma glucose (an index of sympathoadrenal activity) and corticosterone levels at the end of stress, and (2) open field scores 18-19 hr after immobilization in saline- or tianeptine-pretreated rats. Tianeptine was found to leave unaltered stress-induced increases in cortical, hippocampal, hypothalamic, midbrain, and striatal serotonergic metabolism. Similarly, stress-elicited elevations in plasma glucose and corticosterone levels were not affected by tianeptine pretreatment. On the other hand, tianeptine pretreatment reversed stress-induced deficit in exploratory activity. To test whether the latter positive effect of tianeptine was associated with changes in plasma glucose and corticosterone levels during the early phase of stress, we have measured plasma glucose and corticosterone levels (at 0, 5, 15, 30, and 60 min) in resting and stressed (catheterized) rats. The results indicated that tianeptine pretreatment did not alter plasma glucose and corticosterone early responses to stress. Although this study confirmed the antidepressant effect of tianeptine, the neurochemical and neuroendocrinological mechanisms underlying this positive effect remain to be determined.

Behavioural and biochemical evidence that glucocorticoids are not involved in DOI-elicited 5-HT2 receptor down-regulation

Numerous studies have brought evidence for reciprocal relationships between glucocorticoids and 5-HT2 receptors; however, whether glucocorticoids affect 5-HT2 receptor regulation is still unknown. Herein, we have analyzed whether 5-HT2 receptor down-regulation following repeated administration of the 5-HT2/5-HT1C receptor agonist 1-(4-iodo-2,5-dimethoxyphenyl)-2-aminopropane (DOI) is affected by glucocorticoid removal. Compared with sham surgery, adrenalectomy (11-15 days beforehand) did not affect either frontal cortex [3H]ketanserin binding nor the number of head shakes elicited by a single administration of DOI (2.5 mg/kg s.c.). Pretreatment with DOI (2.5 mg/kg s.c. x 4 in 48 h) decreased to similar extents the head shake response to DOI injection in sham (-88%) and adrenalectomised (-95%) rats. Confirmingly, this paradigm was found to diminish the Bmax for [3H]ketanserin binding in sham and adrenalectomised rats by 64% and 46%, respectively. From these data, it is concluded that glucocorticoid removal does not alter 5-HT2 receptor binding and function nor does it affect 5-HT2 receptor down-regulation.

Psychoneuroendocrine outcomes of short-term crowding stress

One-week exposure to crowding stress has been reported to affect body weight and pituitary-adrenal axis activity. This study has first evaluated whether the aforementioned changes are associated with altered exploratory activity. One-week crowding (16 rats/cage) diminished body weight gains, compared to the control situation (four rats/cage): actually, this difference was mainly due to a marked loss in food intake and body weight following the first day of crowding. On the other hand, 1-week crowded rats were not different from their controls when placed for the first time in an open field, as shown by the analysis of their exploratory activities, their number of grooming episodes, and their defecation. Psychoneuroendocrine effects of 1-day crowding were then investigated: actually, open field behaviors were not modified but basal plasma glucose levels were decreased. Moreover, neither plasma ACTH and corticosterone levels nor 2-min cold swim-induced increases in corticosterone levels were affected. Besides, cold swim-induced increases in plasma ACTH levels were amplified in crowded rats, and these increases were associated with hyperglycemia in crowded, but not in control rats. It is suggested that 1-day exposure to crowding has differential consequences upon feeding and exploratory behaviors while potentiating the responsiveness of stress hormones.

Effects of cold stress on some 5-HT1A, 5-HT1C and 5-HT2 receptor-mediated responses

The purpose of the present study was to analyze the influence of stress (24-h cold exposure) on presynaptic 5-HT1A receptors, and on postsynaptic 5-HT1A, 5-HT1C and 5-HT2 receptors. Cold exposure for 24 h affected neither pargyline-induced decreases in 5-hydroxyindoleacetic acid (5-HIAA) levels in midbrain and rest of brain, nor plasma glucose and corticosterone levels. Treatment with the 5-HT1A receptor agonist, 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT; 0.5-1 mg/kg), 3-5 h after the end of cold exposure triggered less intense flat body posture and forepaw treading in cold-exposed rats than in controls. On the other hand, 15- and 30-min plasma glucose responses to 8-OH-DPAT (0.25-0.5 mg/kg, 3-5 h after cold) or to the alpha 2-adrenoceptor agonist, clonidine (0.025 mg/kg), were not affected by cold, while the 15-min, but not the 30 min, plasma corticosterone response to 8-OH-DPAT was slightly amplified in cold-exposed rats. Cold exposure affected neither the inhibitory effect of 8-OH-DPAT (0.25-0.5 mg/kg, 3-5 h after cold) on midbrain 5-HIAA levels, nor the hypothermic effect of 8-OH-DPAT (0.5-1 mg/kg, 3-5 h after cold). Lastly, the hypoactivity elicited by the 5-HT1C receptor agonist, m-chlorophenyl-piperazine (1.5-3 mg/kg, 3-5 h after cold), or head shakes elicited by the 5-HT2 receptor agonist, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (1-2 mg/kg, 3-5 h after cold), were of similar intensities in control and in cold-exposed rats.