The effect of incubating the hypothalamus of adult male rats with various neurotransmitters upon the release of corticotrophin-releasing hormone (CRH) was studied. The CRH activity in the incubation medium was assayed in 48 h median eminence-lesioned rats and the corticosteroidogenesis of excised adrenals in vitro was used as the end-point. 5-Hydroxytryptamine (100 pg/ml-10ng/ml) caused a dose-dependent release of CRH which was antagonized by methysergide (30-100 ng/ml). The response to 5-hydroxytryptamine was also inhibited by hexamethonium and atropine which indicated that it was acting through a cholinergic interneurone. Melatonin (10 ng) did not alter the basal release of CRH but inhibited the action of both 5-hydroxytryptamine (10 ng) and acetylcholine (3 pg). Thus it appears that both 5-hydroxytryptamine and melatonin play a role in the control of CRH release. Noradrenaline blocked the release of CRH induced by both acetylcholine and 5-hydroxytryptamine and presumably this inhibition was caused by direct action on the CRH neurone. gamma-Aminobutyric acid (GABA) also inhibited the release of CRH and may also be involved in the regulation of CRH secretion. The inhibitory neurotransmitters, noradrenaline, GABA and melatonin, act via independent receptor mechanisms. A model based on the above data is presented.

SUMMARY1. The hypophysiotrophic area of the rat hypothalamus was studied in vitro. The preparation remained viable for at least 3 hr and showed oxygen consumption varying between 68-9-120 ,umole/g. hr. The tissue potassium ion content (per unit wet weight) fell to about 50 % ofthe in vivo concentration during this time compared with 15 % in the presence of ouabain (104 M). Histological examination of tissue incubated for 3 hr showed variable perineuronal oedema but the nuclei were of normal appearance and none showed the pyknotic changes that would be associated with cell degeneration.2. Corticotrophin releasing hormone (CRH) in the medium pooled from five to twenty hypothalami was assayed in five to twelve rats which were median-eminence lesioned 48 hr earlier. In vitro corticosterone production of quartered adrenals was used as the end point of the assay. Regression lines of the dose-response curves for ACTH, crude CRH and different volumes of medium from electrically stimulated hypothalami were parallel.CRH output was maximal at 75 Hz and 100,uA when the square-wave pulses lasted for 1 msec. No CRH activity was found on stimulation of cerebral cortex or thalamic tissue pieces of equivalent size.3. Hypothalami taken from rats, adrenalectomized 7-14 days previously, released several-fold more CRH into the medium during electrical stimulation than the initial content of the tissue, showing that the tissue was capable of synthesizing CRH in vitro. The hypothalami taken from intact rats released considerably less CRH into the medium than tissue taken from 12 to 14 day adrenalectomized rats. The hyper-secretion of CRH observed in hypothalami taken from adrenalectomized rats was abolished by pre-treatment with 5 mg/100 g s.C. of corticosterone 24 hr before removal of the tissue. It is therefore proposed that the delayed negative feed-back action of corticosterone at the hypothalamic level is by the suppression of CRH synthesis and that the effect of secretion is secondary to the effect on synthesis.4. The presence of Ca2+ in the medium was essential for the release of CRH.5. CRH secretion increases linearly with doses of acetylcholine from 5.5 X 10-1k-5-5 X 10--14 M. Cerebral cortex incubated with acetylcholine showed no CRH activity. The effect of acetylcholine was reduced by atropine (3.5 x 10-13 M). Median eminence-pituitary stalk fragments (which contain mainly terminal axons of neurones) incubated with acetylcholine showed no CRH stimulation in the doses that activate the release of CRH using the hypophysiotrophic hypothalamus. Acetylcholine may act as a neurotransmitter at the dendritic level in the CRH neurone.

Corticosteroid feedback mechanisms were investigated at the hypothalamic level using the rat hypothalamus in vitro and the pituitary level using basal hypthalamic-lesioned rats. Both fast and delayed corticosteroid feedback effects were demonstrated at the level of the hypothalamus and pituitary gland with doses of corticosteroids within or near the physiological range. These two phases of feedback were separated temporally by a 'silent period' during which no feedback was apparent. Studies on the mechanism of action of corticosteroids at the hypothalamic level showed that the fast feedback mechanism acts by inhibition of release whilst the delayed feedback mechanism acts by inhibition of both synthesis and release. The fast feedback action of corticosterone does not appear to act by excitation of neuroinhibitory pathways since neither picrotoxin nor phentolamine prevented the feedback action of corticosteroids in vitro. Corticosterone inhibition of corticotrophin releasing factor release was overcome by depolarization of the membrane with K+ suggesting that the mechanism of action of the fast feedback of corticosteroids is by membrane stabilization.

The effect of various putative neuro-transmitters upon the release of corticotrophin releasing hormone (CRH) was investigated using the hypothalamus of the rat in vitro. Acetylcholine (1–5 pg) caused a dose-dependent release of CRH which was antagonized by hexamethonium (1–10 ng) and partially antagonized by atropine (300 pg). Neither noradrenaline, dopamine nor histamine had any effect on the basal secretion of CRH. Noradrenaline (10 ng), however, was able to inhibit the release of CRH in response to acetylcholine (3 pg) and this action of noradrenaline was reduced by phentolamine (100 ng), an α adrenergic blocking agent.

Structure-activity studies on the corticosteroid fast and delayed feedback receptor mechanisms controlling the secretion of corticotrophin releasing factor (CRF) were carried out with the rat hypothalamus in vitro. The secretion of CRF was induced by acetylcholine (3 pg/ml). The fast feedback receptor appears highly specific, and the structure essential for efficacy involves an 11beta-hydroxyl group and an unblocked 21-hydroxyl group. Several steroids showed antagonism and so the binding site is not very specific. 18-hydroxy, 11-deoxycorticosterone, progesterone, 17alpha-hydroxyprogesterone and 11-deoxycorticosterone were antagonists of fast feedback. The delayed feedback receptor required either an 11beta-or a 21-hydroxyl group for efficacy. The binding site required a 17-hydroxyl group when the 11beta- or 21-hydroxyl groups were absent. Binding also involved the 3-oxo,4,5-ene structure since steroids in which these are absent were inactive.