In pregnancy metabolic actions of stress-stimulated glucocorticoids may compromise energy supply to the fetuses. I.C.V. injection of orexin-A or neuropeptide Y (NPY), signalling energy lack, stimulates the HPA axis in virgin but not late pregnant rats [1,2]. Here we studied responses to I.C.V. ghrelin, which acts like orexin via NPY neurones, and to insulin-induced hypoglycaemia (IIH). We tested roles of endogenous opioid [3] in suppressing responses to NPY and orexin. Rats (n=5-7/group) were implanted (halothane anaesthesia) 5 days before experiment (pregnancy day 21) with a jugular vein cannula, and an I.C.V. cannula if required for 2 µl aCSF ± peptide injections. Rats were housed singly, in standard conditions (lights on 07.00h, off 19.00h), and cannulae connected ca 07.30h, 2h before experiment. Blood samples (replaced by 0.9% saline) were taken for ACTH and corticosterone assay pre- and post-treatment; 240 or 90 min later brains were collected after decapitation or formaldehyde fixation-perfusion (pentobarbitone anaesthesia, 42 mg/kg I.P.), for quantitative in situ hybridisation or Fos immunocytochemistry on coronal cryostat or frozen sections, respectively. Data were analysed by ANOVA. In virgins I.C.V. ghrelin (2 nmol) increased plasma ACTH and corticosterone concentrations from (mean±s.e.m.) 42.6±6.8 to 59.7±5.3 pg/ml and from 77.7±0.86 to 181.3±29.1 ng/ml, respectively, at 10 min (n=6, p<0.05), but had no significant effects in late pregnant rats; similarly, ghrelin increased parvocellular paraventricular nucleus (pPVN) Fos expression (3.9x, p<0.05) only in virgins. As expected, NPY (1 nmol I.C.V.) alone only activated the HPA axis in virgins (increased plasma ACTH, pPVN CRH and vasopressin (VP) mRNA [see 2] and Fos (3.5x) expression). Naloxone (5 mg/kg I.V.) had no effect in virgins, but after naloxone in pregnant rats, NPY increased plasma ACTH from 37.1±7.8 to 128.7±29.3 pg/ml at 15 min, pPVN CRH and VP mRNA expression 1.6x and 15.2x, and Fos counts by 2.3x (all p<0.05). I.C.V. orexin-A (1.4 nmol) increased pPVN Fos counts (2.1x) and plasma ACTH (1.9x at 15 min; both p<0.05) only in virgins [1]. Naloxone had no effects in virgins, but in pregnant rats it restored orexin responses: increase in plasma ACTH from 77.0±3.7 to 159.2±34.1 pg/ml at 15 min (p<0.05), and increases in pPVN CRH and VP mRNA expression (2.2x and 5.5x vs. orexin alone, respectively, p<0.05). Thus endogenous opioid suppresses actions of NPY, and hence of orexin, and maybe ghrelin (since it acts via NPY), on the HPA axis in late pregnancy. In contrast, IIH induced by I.V. insulin injection (10 U/kg) in virgin and pregnant rats similarly reduced blood glucose (to ca 2 mmol/l), increased plasma ACTH (by, mean±s.e.m., 46.1±17.8 and 35.6±7.8 pg/ml) and pPVN VP mRNA expression (by 3.3x and 2.3x; all p<0.05). Evidently activation of the HPA axis by IIH is intact in late pregnancy, and may not be mediated by NPY, orexin or ghrelin.