haut de page

mise à jour du 31 juillet 2003
Brain Research
1985; 347; 404-408
Stress-induced activation of the hippocampal cholinergic system and the pituitary adrenocortical axis
Gilad G., Mahon B., et al
Center of neurosciences and behavioral research, Weizmann institute of science, Israel


It is well known that the pituitary adrenocortical axis is activated in response to stressful stimuli in animals and man. In addition, it is well recognized that the limbic system is involved in both neuroendocrine regulation and emotional responsivity to stress. Accordingly, a limbic-hypothalamic-pituitary-adrenocortical system bas been conceptualized as an integrating unit which controls the physiological and behavioral response to stress. Little is known about the neuroanatomical basis for this response in the forebrain. However, evidence from psychopharmacological and lesion studies have suggested that the septo-hippocampal cholinergic system is an important part of a neuronal network in the brain which controls the response to stress, a response which is adaptive in nature. Indeed, we have recently demonstrated that in rats the septo-hippocampal cholinergic system is actively involved in the response to stress. This is expressed by a reduction in high affinity choline uptake and increased muscarinic binding capacity in the hippocampus. These changes are more pronounced and occur carlier in rat strains which are more reactive to stress.

Furthermore, it has been demonstrated that the hippocampus is a main target region for corticosterone and that the electrical activity of pyramidal neuror in the hippocampus is affected by corticosterone. These data indicate that the hippocampus can integrate both endocrine and neural signals of stress, possibly through participation of the cholinergic system. In the present study therefore we have examined the effects of stress, ACTH and corticosterone treatments on the dynamics of hippocampal cholinergic synaptic mechanisms.

Sprague-Dawley female rats, 10-12 weeks old, were housed 5 to a cage, supplied freely with food and water and maintained at 24°C with a 12 h light-dark cycle. Stress consisted of immobilization in a prone position for 10 min, as reported before. Rats left undisturbed in their home cage served as controls.

For adrenalectomy, rats were anesthetized with halothane and the two adrenal glands were removed through small incisions made on both sides of the back. Operated rats were maintained on 0.9% NaCl (saline) drinking water. Sham operated rats served as controls.

Drugs were injected intraperitoneally (i.p.) in 200 µl of solvent (vehicle): corticosterone (35 mg/kg) in peanut oil and ACTH (1.5 u/kg) in saline. Vehicle injected rats served as controls.

After 10 min of stress, or 10 min after drug treatments rats were decapitated, their brains rapidly excised and the hippocampus dissected. The dissected tissues were immediately homogenized in 10 vols. (w/v) of ice-cold 0.32 M sucrose in a glass homogenizer fitted with a Teflon pestle. Homogenates were centrifuged at 1000 g for 10 min and the resulting supernatant centrifuged at 17,000 g for 20 min. The rude synaptosomal pellet (P2 fraction) was resuspended in the original volume of 0.32 M sucrose and used for biochemical assays.

High affinity sodium-dependent uptake of [3H]choline (0.4 µCi, spec. act. 80 Ci/mmol), measured for 4 min, was assayed according to the method of Atweh et al.

Spontaneous release of newly synthesized [3H]aceiylcholine (ACh) was measured in synaptosomal fractions following incubation for 4 min in the presence of [3H]choline (0.4,uCi, spec. act. 80 Ci/mmol). Immediately after the incubation 3 ml of ice-cold Krebs-Ringer phosphate buffer, pH 7.4, were added and synaptosomal fractions were centriuged at 20,000 g for 15 min. The pellets were washed ,wice with 1 ml and then resuspended in 1 ml of the same buffer containing 5µm eserine sulfate. Release of [3H]ACh was measured at 37 'C for up to 8 min. The tubes were then rapidly centrifuged at 20,000 g for 15 min at 4 'C and the supernatant removed for [3H]ACh measurements according to the method of McCaman and Stetzler.

To assess muscarinic cholinergic binding, the antagonist [3H]quinuclidinyl benzilate (ONB) (spec. act. 33 Ci/mmol) served as a ligand. Binding was assayed according to Yamamura and Snyder in P2 fractions, using 10 µM atropine sulfate to determine non-specific binding. The values of QNB binding measurements were plotted according to Scatchard.

Protein content was measured by the method of Lowry et al. Effects of stress, ACTH or corticosterone treatments. Injections of high doses of ACTH (1.5 u/kg) or corticosterone (35 mg/kg) led 10 min later to increases in both [3H]choline uptake and in newly synthesized [3H]ACh release, as did a 10 min session of immobilization stress. Analysis of the variance indicated that there was no difference between treatments with regard to [3H]choline uptake elevations (FO.05 (3,28)=2.4).

Effects of adrenalectomy. The effects of adrenalectomy were examined 2 days after the operation. Both [3H]choline uptake and [3H]ACh release were increased as compared to unhandled or sham operated controls. Adrenalectomy led to the largest observed increase in 13HIACh release. Interestingly, [-3H]ACh release but not [-3H]choline uptake, was increased in sham operated animals (2 days postoperative) as compared to unhandled controls. This could have been the result of the long term stress caused by the operation.

Effects of stress, ACTH or corticosterone treatments on adrenalectomized rats. The effects of stress, or single injections of ACTH or corticosterone on [3H]choline uptake and [3H] ACh release were examined after 10 min in animals which were adrenalectomized 2 days earlier.

Uptake of [3H] choline was elevated after all treatments to the same extent (F0, 05 (6,40) = 2. 1). The increases in [3H]ACh release observed after adrenalectomy alone were not altered by acute stress or ACTH treatments. However, corticosterone treatment of adrenalectomized rats did result 10 min later in a significant reduction; yet these reduced levels were still significantly higher than unhandled controls. In contrast, corticosterone did not alter choline uptake after adrenalectomy.

The present data demonstrate that the hippocampal cholinergic synaptic mechanisms are activated within minutes after stress and that similar changes can be produced shortly after treatments with high doses of ACTH or corticosterone. An increase in [3H] choline uptake was observed before in hippocampal synaptosomes 3 h after a single intravenous injection of a high dose of corticosteroids. These findings indicate that the septo-hippocampal cholinergic system can be activated secondary to an activation of the pituitary-adrenocortical axis, an activation which occurs rapidly when stressful conditions are encountered.

The hippocampal cholinergic system appears to be sensitive to either an increase (i.e. corticosterone injection) or decrease (i.e. adrenalectomy) in circulating corticosterone levels. Both conditions activate presynaptic cholinergic terminals as indicated by increased high affinity choline uptake and newly synthesized ACh release. While choline uptake was increased to comparable levels both after adrenalectomy and corticosterone, ACh release was more enhanced after adrenalectomv. Injection of corticosterone, while having no effect on adrenalectomy-induced alteration in choline uptake, can at the same time attenuate the enhanced levels of ACh release induced by adrenalectomy.
In contrast, ACTH, which can produce similar effects as does corticosterone in intact animals, has no effect on the cholinergic induced changes after adrenalectomy. Comparable sensitivity to alterations in the pituitary-adrenocortical hormonal status was demonstrated by Azmitia et al. for the septal driving of hippocampal thetarhythm. These authors have attributed the change. to be the result of interaction between the serotonine input to the hippocampus and hippocampal neurons which are able to concentrate corticosterone8,21. Our findings demonstrate that stress-induced activation of the pituitary-adrenocortical axis, which results in enhanced circulating corticosterone or treatment with high doses of corticosterone, can in turn activate the presynaptic cholinergie terminals in the hippocampus. This indicates that septo-hippocampal cholinergic neurons themselves are sensitive to major changes in circulating levels of corticosterone. Overproduction of corticosterone can occur after stress or in disease (e.g. Cushing's syndrom) while deficiency may occur in adrenocortical insufficiency (e.g. Addison's syndrom).

The finding that corticosterone treatment reduces the adrenalectomy-induced increase in ACh release, with no significant effect on the enhanced choline uptake levels, indicates that these two mechanisms which control presynaptic cholinergic activity are differentially sensitive to corticosterone. So far corticosterone effects in the brain were correlated mostly with intracellular binding within neurons. However, the rapid effects demonstrated in the present study indicate that corticosterone may act directly on these presynaptic cholinergic mechanisms in the hippocampus. This possibility is currently being examined by measuring the effects of corticosterone on synaptosomal preparations in vitro.

Preliminary data indicate that values of QNB binding to hippocampal synaptosomal preparations were not altered after acute stress. ACTH, or corticosterone treatments and remained at levels of unhandled control values which were: Bmax = 2.35 ± 0.12 (pmol/mg protein); Ad = 0.61 ± 0.03 (nM). At 2 days after adrenalectomy, however, Kd values were increased to 124.6% of control, with no change in Bmax Mis change together with the previous demonstration that chronic intermittent stress leads to increased Bmax values indicate that muscarinic receptors can be modified when circulating corticosteroid levels are altered over prolonged periods. In support of this view is the recent report on corticosteroid iodulation of muscarinic receptors in rat myocardial membranes.

In summary, the present study demonstrates that after short term stress the septo-hippocampal cholinergic system is activated secondary to activation of the pituitary-adrenocortical axis. The demonstration that major changes in circulating levels of corticosteroids can modulate the activity of the hippocampal cholinergic synapse implies that the septo-hippocampal system may be a common output for various neuronal and endocrine systems which convey signals of stress.