Social isolation-induced changes in the hypothalamic–pituitary–adrenal axis in the rat

Abstract

Social isolation of rats both reduces the cerebrocortical and plasma concentrations of 3α-hydroxy-5α-pregnan-20-one (3α,5α-TH PROG) and 3α,5α-tetrahydrodeoxycorticosterone and potentiates the positive effects of acute stress and ethanol on the concentrations of these neuroactive steroids. We now show that social isolation decreased the plasma level of adrenocorticotropin (ACTH), moreover, intracerebroventricular administration of corticotropin releasing factor (CRF) induced a marked increase in the plasma corticosterone level in both isolated and group-housed rats, but this effect was significantly greater in the isolated rats (+121%) than in the group-housed rats (+86%). In addition, in isolated rats, a low dose of dexamethasone had no effect on the plasma corticosterone concentration, whereas, a high dose significantly reduced it; both doses of dexamethasone reduced plasma corticosterone in group-housed rats. Furthermore, the corticosterone level after injection of dexamethasone at the high dose was significantly greater in the isolated animals than in the group-housed rats. These results suggest that social isolation increased sensitivity of the pituitary to CRF and impaired negative feedback regulation of the hypothalamic–pituitary–adrenal (HPA) axis.

• Ethanol
• HPA axis
• corticosterone
• rat
• social isolation
• stress

Introduction

Long-term social isolation after weaning markedly affects the behavior of rats. Isolated animals are aggressive, neophobic and highly reactive to human handling. They appear nervous and show both an anxiety-like profile in the elevated plus-maze test and increased locomotor activity in response to novel situations (Hatch et al. 1963; Parker and Morinan 1986; Wongwitdecha and Marsden 1996). Social isolation is thus thought to be stressful for these normally gregarious animals and their abnormal reactivity to environmental stimuli when reared under this condition is thought to be a product of prolonged stress. Although, the underlying mechanisms remain poorly understood, similar social conditions are thought to contribute to the etiology of psychiatric disorders such as schizophrenia, depression and anxiety in humans (Heim and Nemeroff 2001).

Several acute stressful stimuli, as well as ethanol, increase the brain and plasma concentrations of neuroactive steroids, which are endogenous steroids that affect the excitability of central neurons in a manner independent of nuclear hormone receptors (for review see Biggio and Purdy 2001). Some of these compounds, such as 3α-hydroxy-5α-pregnan-20-one (allopregnanolone or 3α,5α-TH PROG) and 3α,5α-tetrahydrodeoxycorticosterone (3α,5α-TH DOC), are among the most potent positive allosteric modulators of type A receptors for the inhibitory neurotransmitter γ-aminobutyric acid (GABA) (Majewska et al. 1986; Belleli and Lambert 2005). Thus, their acute administration in pharmacological doses elicits anxiolytic, anticonvulsant and sedative-hypnotic effects in rodents; moreover, physiological or pharmacologically induced changes in the levels of 3α,5α-TH PROG are implicated in the regulation of GABA_A receptor plasticity in addition to modulation of receptor function (for reviews see Biggio and Purdy 2001; Smith 2004).

We found that social isolation of rats for 30 days immediately after weaning, in the absence of any additional stressor, resulted in a decrease in the cerebrocortical and plasma concentrations of 3α,5α-TH PROG and 3α,5α-TH DOC compared with the corresponding values for group-housed animals, an effect prevented by handling of the animals twice daily (Serra et al. 2000). The molecular mechanism responsible for the persistent decrease in the abundance of neuroactive steroids induced by social isolation in rats remains unclear. The observations that adrenalectomy both markedly reduces the brain content of neuroactive steroids (Purdy et al. 1991; Khisti et al. 2002; O'Dell et al. 2004) and prevents the increase in the plasma and brain concentrations of these compounds induced by acute stress (Barbaccia et al. 1997) suggest that adrenal steroidogenesis plays an important role in maintaining the abundance of neuroactive steroids in both brain and plasma. An altered regulation of the hypothalamic–pituitary–adrenal (HPA) axis might thus contribute to the reduction in the amounts of neuroactive steroids apparent in isolated animals. We have previously shown that the increases in the brain and plasma concentrations of 3α,5α-TH PROG and 3α,5α-TH DOC induced by foot shock (Barbaccia et al. 1996, 1997) used in this instance as a novel acute stressor were markedly greater on a percentage basis in socially isolated rats (395% and 292%, respectively) than in group-housed animals (78% and 107%, respectively; Serra et al. 2000). These results suggest that social isolation induced a change in regulation of the HPA axis rather than a decrease in secretory capability per se. This conclusion is consistent with the notion of development during exposure to chronic stress of a “facilitatory trace”, characterized by hyperresponsiveness of the HPA axis to new stimuli (Akana et al. 1992). To examine the mechanism responsible for the reduction in the basal concentrations of neuroactive steroids and the increased sensitivity of the production of these steroids to stress induced by social isolation, we have now investigated the effect of social isolation on neuroendocrine state.

Materials and methods

Animals

Male Sprague-Dawley CD rats at 30 days of age, immediately after weaning, were housed for 30 days either in groups of 6–8 per cage or individually in smaller cages. They were maintained under an artificial 12-h-light, 12-h-dark cycle (lights on at 07:00 h) at a constant temperature of 23° ± 2°C and 65% humidity. All experiments were performed between 08:30 h and noon. Food and water were freely available at all times. Animal care and handling throughout the experimental procedures were in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC).

Treatments

Corticotrophin releasing factor (CRF)

Seven days before the end of the 30-day period, a polyethylene cannula (SP-10 PE) was implanted into the right lateral ventricle of rats anesthetized with equithesin (propylene glycol 20%, ethanol 10%, pentobarbital 0.2 M, 0.3 ml/kg, i.p.). At the end of the 30-day period, CRF (500 ng in 5 μl of physiological saline) or saline was injected into the lateral cerebral ventricle of the experimental animals with the use of a 10-μl microsyringe and an injection cannula inserted into the guide cannula. The animals were killed 30 min later. Correct placement of the cannula was verified histologically.

Dexamethasone

Rats were injected intraperitoneally (i.p.) with dexamethasone (3 or 500 μg/kg body weight) or physiological saline (0.9%) vehicle and killed 150 min later.

Extraction and assay of corticosterone

Rats were killed by decapitation with a guillotine. Blood was collected from the trunk of killed rats into heparinized tubes and centrifuged at 900g for 20 min at room temperature. The resulting plasma was frozen at − 80°C until assayed for steroids. Corticosterone was extracted from the plasma with ethyl acetate (recovery of 70–80% as monitored by addition of a trace amount of [³H]corticosterone) and then quantified by radioimmunoassay, as described previously (Serra et al. 2000), with specific antibodies (ICN, Costa Mesa, CA).

Adrenocorticotrophin hormone (ACTH) radioimmunoassay

Blood was collected from the trunk of killed rats into prechilled (4°C) tubes containing EDTA and then centrifuged within 60 min at 900g for 10 min in a refrigerated centrifuge (4°C). The resulting plasma was frozen ( − 80°C) until assayed for radioimmunoassay with a kit obtained from ICN, Costa Mesa.

Results

We postulated that an altered regulation of the HPA axis might contribute to the reduction in the amounts of neuroactive steroids found in isolated animals. Consistent with this hypothesis, we found that the basal concentration of ACTH in plasma was significantly decreased in isolated rats (1023 ± 148 pg/ml) compared with group-housed rats (1495 ± 210 pg/ml) (data are means ± SD of values from 36 animals, P < 0.01, Student's t-test).

We previously found that foot-shock stress or ethanol injection increased the cerebrocortical and plasma concentrations of neuroactive steroids by a greater percentage in isolated rats than in group-housed animals (Serra et al. 2000, 2003). This indicates a hyperresponsiveness of the HPA axis in isolated rats, and indirectly suggests that in spite of the decreased basal plasma level of ACTH, full secretory capacity of the pituitary corticotrophs is maintained in these animals. Therefore, we examined the effect of central administration of CRF on the plasma concentration of corticosterone (Figure 1). CRF was injected into the lateral ventricle to gain access to the primary capillary plexus of the hypothalamo–pituitary portal system, although there may have also been central actions of CRF leading to HPA axis activation. CRF induced a marked increase in the plasma corticosterone concentration in both isolated and group-housed rats, but this effect was significantly greater (P < 0.01) in the isolated animals (+121%) than in the group-housed rats (+86%).

Figure 1 Effect of exogenous CRF on the plasma concentration of corticosterone in socially isolated rats. Rats were housed in groups or in isolation for 30 days. Data represent the percentage increase in the plasma concentration of corticosterone in the rats given intracerebroventricular (i.c.v.) CRF, relative to the corresponding values for control (saline-injected) rats and are means ± SEM of values from 14 animals. Basal values: group-housed rats, 99 ± 11 ng/ml; isolated rats, 127 ± 16 ng/ml. *P < 0.01 vs group-housed rats (Student's t-test).

Next, we investigated the effect of intraperitoneal injection of dexamethasone on the basal concentration of corticosterone in the plasma of socially isolated rats. As shown in Figure 2, we found that the plasma corticosterone concentration in group-housed rats was significantly reduced by injection of the low or high dose of dexamethasone ( − 38 and − 81% of basal values, as given in Figure 2 legend). In isolated rats, however, the low dose of dexamhasone had no effect on the plasma corticosterone concentration, whereas the high dose significantly reduced it ( − 51%). The plasma corticosterone concentration after injection of dexamethasone at the high dose was nevertheless significantly greater in the isolated animals than in the group-housed rats (Figure 2).

Figure 2 Effect of i.p. dexamethasone on the plasma concentration of corticosterone in socially isolated rats. Rats were housed in groups or in isolation for 30 days. Data represent the plasma concentration of corticosterone expressed as a percentage of the corresponding values for control (saline-injected) rats and are means ± SEM of values from 20 animals. Basal values: group-housed rats, 113 ± 10 ng/ml; isolated rats, 131 ± 14 ng/ml. *P < 0.01 vs corresponding control rats; ^†P < 0.01 vs corresponding group-housed rats (two-way analysis of variance followed by Newman-Keuls test).

Discussion

We have shown that social isolation decreased the plasma concentration of ACTH, evidently increased sensitivity of the pituitary to CRF and impaired glucocorticoid negative feedback regulation of corticosterone secretion.

A decrease in the plasma concentration of ACTH, despite the continuous presence of the stressor, has been described for animals exposed to various chronic stressful stimuli and several mechanisms for this effect, in addition to a reduction in pituitary responsiveness to modulators of ACTH secretion (CRF, AVP), have been proposed (Keller-Wood and Dallman 1984; Rivier and Vale 1987; Hauger et al. 1988). Rivier and Vale (1987) suggested that both a decrease in the readily releasable pool of ACTH and the negative feedback exerted by corticosterone may account for the diminished responsiveness of the HPA axis of rats exposed to chronic intermittent electroshock.

In contrast, we have previously shown that social isolation increases the responsiveness of the HPA axis to new stimuli. Thus, the increases in the brain and plasma concentrations of 3α,5α-TH PROG and 3α,5α-TH DOC induced by foot shock (Barbaccia et al. 1996, 1997), used in this instance as a novel acute stressor, or by systemic injection of ethanol (Van Doren et al. 2000), were markedly greater on a percentage basis in socially isolated rats than in group-housed animals (Serra et al. 2000, 2003). The enhanced effects of acute stress and ethanol on the brain and plasma concentrations of neuroactive steroids in isolated rats may be related to an abnormal reactivity of the HPA axis that develops as an adaptive response to chronic stress. Abnormalities in the behavioral response of isolated rats to distinct challenges have been associated with functional changes in the endocrine response, although differences in social isolation procedures or test environments among studies have led to apparently discrepant results. For example, the basal level of corticosterone in plasma was found to be either unchanged (Morinan and Leonard 1980; Viveros et al. 1988; Haller and Halàsz 1999), increased (Rivier and Vale 1987; Greco et al. 1990; Genaro et al. 2004; Sandstrom and Hart 2005) or decreased (Miachon et al. 1993; Sanchez et al. 1998; Chida et al. 2005) in socially isolated animals.

Given that CRF is the main stimulator of ACTH release (Axelrod and Reisine 1984), and the stimulatory effect of ethanol on the corticotrophs requires the release of endogenous CRF (Lee et al. 2004), either enhanced CRF release in response to stress and to ethanol, or an increased pituitary sensitivity to CRF might be responsible for the exaggerated response of isolated rats to a novel stress.

The results presented suggest that the enhanced corticosteroid secretion apparent in response to a novel acute stress in socially isolated rats may be attributable, at least in part, to an increased sensitivity of the pituitary corticotrophs to CRF, although, an augmented release of CRF and AVP from the hypothalamic paraventricular nucleus or an increased POMC primary transcript level (Lee et al. 2004) cannot be ruled out.

Studies on HPA axis sensitivity during chronic stress have generated apparently contradictory findings as a result of the large variation in the intensity and duration of exposure to stressors and in the doses of administered CRF. Pituitary–adrenocortical responses to CRF have been found to be unaffected by chronic stress associated with immobilization (Hashimoto et al. 1988; Culman et al. 1991) or crowding (Bugajski et al. 1994) in rats, whereas chronic shock-avoidance stress resulted in an attenuated ACTH response to CRF (Odio and Brodish 1990). In contrast, the ACTH response to intravenous administration of CRF was significantly increased in rats stressed by cold adaptation (Uehara et al. 1989) or by social defeat (Buwalda et al. 1999). The latter study also showed that levels both of the glucocorticoid receptor in the hippocampus and hypothalamus and of the mineralocorticoid receptor in the hippocampus were significantly decreased in the stressed animals, resulting in reduced feedback inhibition of the HPA axis (Buwalda et al. 1999). Expression of glucocorticoid and mineralocorticoid receptors in the brain has been found not to be markedly affected by social isolation in rats (Holson et al. 1991; Olsson et al. 1994; Weiss et al. 2004; Filipovic et al. 2005). Nevertheless, it is possible that the increased responsiveness of socially isolated rats to acute stimuli is attributable in part to decreased negative feedback by corticosterone. The negative feedback exerted by corticosterone on its own release after exposure of animals to stress is mediated by glucocorticoid receptors in the pituitary, hypothalamus and hippocampus (Keller-Wood and Dallman 1984). A gradual decrease in the number of glucocorticoid receptors in specific brain areas in response to social isolation might result in a reduced effectiveness of feedback inhibition of the HPA axis, thereby leading to an increased ACTH response.

The data obtained with the dexamethasone suppression test suggest that the chronic mild stress associated with social isolation impairs negative feedback. Several studies have demonstrated a pituitary rather than a brain site of action in the suppression of HPA axis activity if moderate amounts of dexamethasone are administered (De Kloet et al. 1975; Miller et al. 1992; Cole et al. 2000). Low doses of the synthetic steroid in the drinking water were previously found to induce selective activation of glucocorticoid receptors in the pituitary, with mineralocorticoid and glucocorticoid receptors in the brain being unaffected; in contrast, high doses of dexamethasone activate glucocorticoid receptors in the brain (Miller et al. 1992). Although, we used a different route (i.p.) of administration of dexamethasone, we selected a low and a high dose of the steroid, in an attempt to examine separately the effects of glucocorticoid receptor activation in the pituitary and in the brain (hypothalamus, hippocampus, cerebral cortex). The low dose of dexamethasone, which effectively reduced the plasma corticosterone level in group-housed rats, presumably by acting primarily at the glucocorticoid receptors in the anterior pituitary, failed to affect the plasma level of corticosterone in isolated animals. Moreover, the partial suppression of corticosterone secretion induced by the high dose of dexamethasone in the isolated rats is suggestive of a partial down-regulation of glucocorticoid receptors in brain areas responsible for feedback inhibition. This hypothesis is supported by the results of studies showing that most procedures for the induction of chronic stress result in down-regulation of glucocorticoid receptors in both the hippocampus and hypothalamus as well as in a consequent hyperresponsiveness of the HPA axis (Makino et al. 1995; Kitraki et al. 1999).

Acknowledgements

Our studies were supported by grants from RAS (Prevenzione ed Educazione Sanitaria 2004) and by GIO.I.A. Foundation (Pisa, Italy).