Regular physical exercise is beneficial for both physical and mental health. By contrast, stress is associated with deleterious effects on health and there is growing evidence that regular physical exercise counteracts some of the effects of stress. However, most previous studies have suggested that prior exercise does not alter the acute hypothalamic pituitary adrenal (HPA) axis responses to stress. The present series of studies provides evidence that in rats, 6 weeks (but not 1 or 3 weeks) of voluntary wheel running reduces the HPA axis responses to lower-intensity stressors such as an i.p. saline injection, exposure to a novel environment or exposure to moderate intensity noise, but not to more intense stressors such as predator odour exposure or restraint. Daily exercise does not appear to be necessary for the reduction in HPA axis responses, with intermittent access (24 h out of each 72-h period) to a running wheel for 6 weeks, resulting in similar decrements in adrenocorticotrophic hormone and corticosterone release in response to 85 dBA noise exposure. Data from in situ hybridisation for c-fos mRNA are consistent with the hypothesis that voluntary exercise results in a decrease in HPA axis responsiveness to a low-intensity stressor at a central level, with no changes in primary sensory processing. Together, these data suggest that 6 weeks of daily or intermittent exercise constrains the HPA axis response to mild, but not more intense stressors, and that this regulation may be mediated at a central level beyond the primary sensory input.

We have shown previously that unconditioned stressors inhibit neurons of the lateral/capsular division of the central nucleus of the amygdala (CEAl/c) and oval division of the bed nucleus of the stria terminalis (BSTov), which form part of the central extended amygdala. The current study investigated whether conditioned fear inhibits c-fos mRNA expression in these regions. Male rats were trained either to associate a visual stimulus (light) with footshock or were exposed to the light alone. After training, animals were replaced in the apparatus, and 2 hours later injected remotely, via a catheter, with amphetamine (2 mg/kg i.p.), to induce c-fos mRNA and allow inhibition of expression to be measured. The rats were then presented with 15 visual stimuli over a 30 minute period. As expected, fear conditioned animals that were not injected with amphetamine, had extremely low levels of c-fos mRNA in the central extended amygdala. In contrast, animals that were trained with the light alone (no fear conditioning) and were injected with amphetamine had high levels of c-fos mRNA in the CEAl/c and BSTov. Animals that underwent fear-conditioning, and were re-exposed to the conditioned stimulus after amphetamine injection had significantly reduced levels of c-fos mRNA in both the BSTov and CEAl/c, compared to the non-conditioned animals. These data suggest that conditioned fear can inhibit neurons of the central extended amygdala. Because these neurons are GABAergic, and project to the medial CEA (an amygdaloid output region), this may be a novel mechanism whereby conditioned fear potentiates amygdaloid output.

The hippocampal formation receives extensive noradrenergic projections and expresses high levels of mineralocorticoid (MR) and glucocorticoid (GR) receptors. Considerable evidence suggests that the noradrenergic system influences hippocampal corticosteroid receptors. However, there is relatively little data describing the influence of glucocorticoids on noradrenergic receptors in the hippocampal formation. α 1d adrenergic receptor (ADR) mRNA is expressed at high levels in the hippocampal formation, within cells that express MR or GR. In order to determine whether expression of α 1d ADR mRNA is influenced by circulating glucocorticoids, male rats underwent bilateral adrenalectomy (ADX) or sham surgery, and were killed after 1, 3, 7 or 14 days. Levels of α 1d ADR mRNA were profoundly decreased in hippocampal subfields CA1, CA2 and CA3 and the medial and lateral blades of the dentate gyrus, as early as 1 day after ADX, as determined by in situ hybridization. The effect was specific for the hippocampal formation, with levels of α 1d mRNA unaltered by ADX in the lateral amygdala, reticular thalamic nucleus, retrosplenial cortex or primary somatosensory cortex. Additional rats underwent ADX or sham surgery and received a corticosterone pellet (10 or 50 mg) or placebo for 7 days. Corticosterone replacement prevented the ADX-induced decrease in hippocampal α 1d ADR mRNA, with the magnitude of effect depending on corticosterone dose and hippocampal subregion. These data indicate that α 1d ADR mRNA expression in the hippocampal formation is highly sensitive to circulating levels of corticosterone, and provides further evidence for a close interaction between glucocorticoids and the noradrenergic system in the hippocampus.