Effect of different doses of corticosterone on depression-like behavior and HPA axis responses to a novel stressor

Abstract

Stress is recognized to precipitate depressive illness, yet the specific relationship between stress, glucocorticoids and depression is not well understood. We have recently shown that repeated corticosterone (CORT) injections reliably increase depression-like behavior on the forced-swim test in rats, suggesting that glucocorticoids can precipitate depressive symptomatology. The purpose of this experiment was to determine the extent to which the effects of CORT on depression-like behavior depend on (1) the dose-injected and (2) the duration of treatment. Rats received either acute or repeated injections of vehicle, 10, 20 or 40 mg/kg of CORT, and then were subjected to the forced-swim test. Serum CORT levels were assessed after the 21-day injection period, and 30 and 60 min after the onset of forced-swim testing. Repeated, but not acute, CORT injections decreased body weight and increased immobility behavior in the forced-swim test in a dose-dependent manner. In addition, all doses of repeated CORT injections suppressed CORT release after the novel stress of forced-swim testing. Our results demonstrate that glucocorticoids increase depression-like behavior in rats in a dose-dependent manner and disrupt normal HPA axis function. These results support the hypothesis that high levels of cortisol contribute to the etiology of depressive symptomatology in humans.

Introduction

The human stress experience contributes to the pathogenesis of depression, and may also play a role in the severity and recurrence of this debilitating illness. The connection between stress and depression was initially drawn from observations of elevated cortisol levels and disrupted cortisol rhythmicity in depressed patients [23], [28], [29], [55], [57], and from the diagnostic utility of the dexamethasone (DEX) suppression test, in which administration of the synthetic glucocorticoid DEX generally fails to suppress hypothalamic–pituitary–adrenal (HPA) axis activity in depressed patients [13], [14], [15]. Cortisol is further linked to depressive symptomatology by observations that patients experiencing elevated glucocorticoid levels as a result of Cushing's disease or synthetic glucocorticoid therapy develop psychiatric and cognitive symptoms consistent with those observed in major depression [1], [9], [10], [11], [35]. In addition, several existing classes of effective antidepressant drugs act on the neuroendocrine substrates that regulate cortisol secretion [45], [48], [49], [50], and novel antidepressant therapies that inhibit cortisol secretion have shown promise in clinical trials [3], [43], [44], [69].

In working towards an understanding of the complex pathophysiology of depression, much study has been devoted to describing the molecular and neuronal effects of exposure to repeated stress in laboratory animals. Several groups have shown that chronic exposure to stress or exogenous corticosterone (CORT; the rodent stress hormone that is equivalent to cortisol in humans) leads to neuronal remodeling in key brain regions associated with depressive illness, namely the hippocampus [40], [66], [68], amygdala [63], [64] and medial prefrontal cortex [17], [54], [58], [67]. Importantly, treatment with certain antidepressant drugs can reverse these effects [26], [39], [65]. These findings are consistent with observations of morphological changes within the hippocampus [6], [38], [59], amygdala [24], [60] and orbitofrontal cortex [7], [36] of human depressed patients, but they do not address the question of whether exposure to stress or elevated glucocorticoid levels produces a depressive phenotype in the laboratory rodent, either through altered behavioral or endocrine status.

Originally described by Porsolt et al. [52], [53], the forced-swim test is the most commonly employed behavioral assay for screening antidepressant compounds (for a review, see [20]). Multiple classes of antidepressant drugs and several non-pharmacological antidepressant therapies, including transcranial magnetic stimulation, rapid eye movement sleep deprivation and electroconvulsive shock, reduce the amount of time a rat spends immobile, delay the onset of immobility, and increase the amount of time a rat spends actively swimming or climbing upon exposure to this inescapable swim task (see [5]). Further validation for use of the forced-swim test to assess depression-like behavioral states has been provided by studies showing that ‘depressogenic’ manipulations, such as exposure to chronic stress and repeated administration of glucocorticoids, increase the amount of time a rat spends immobile, encourage the onset of immobility, and decrease the amount of time a rat spends engaged in active, escape-directed behaviors [4], [12], [42], [51], [56].

We recently reported that repeated CORT injections at a dose of 40 mg/kg consistently increase immobility behavior on the forced-swim test in both male and female rats [30], [34]. These findings suggest that a repeated CORT injection paradigm could provide a useful model within which to study glucocorticoid-induced changes in emotional behavior in the context of depressive illness. Accordingly, the primary objective of the present experiment was to determine the extent to which these behavioral effects depend on the amount of exogenous CORT that is administered to the animals. This question was addressed by comparing the effects of 21 daily low, intermediate or high dose (i.e., 10, 20 or 40 mg/kg) CORT injections on forced-swim behavior. For control purposes, we also assessed the effects of a single low, intermediate or high dose CORT injection on forced-swim behavior.

A second objective of this experiment was to determine the effect of different doses of repeated CORT injections on serum CORT levels. The impact of this injection paradigm on basal serum CORT levels and HPA axis reactivity to a novel stressor has not yet been studied. Thus, we assessed basal serum CORT levels on day 21 of the CORT injection paradigm, as well as novel-stressor-induced CORT levels 30 and 60 min after the onset of the forced-swim test.