The physiological and pathophysiological role of cortisol in pulsatile LH release was investigated in 14 patients (5 men, 6 premenopausal women, and 3 postmenopausal women) with Addison's disease. The explicit effect of cortisol in relation to the effect of corticotropin-releasing factor (CRF), ACTH, and opioids was ensured by hypo-, normo-, and hypercortisolism. Hypocortisolism was obtained by 24-h discontinuation of hydrocortisone (HC) followed by 23-h saline infusion. Eucortisolism was secured by infusion of HC (0.5 mg/kg) over 23 h. Stress-appropriate hypercortisolism was obtained by infusion of HC (2.0 mg/kg) over 23 h, preceded by treatment for 5 days with dexamethasone (1.5 mg/day). To imitate the normal diurnal rhythm for serum cortisol, HC was infused in graduated doses. Blood sampling was performed every 10 min during the last 10 h of the study period, followed by a LH-releasing hormone test (5 microg, i.v.) and a TRH test (10 microg, i.v.). In pre- and postmenopausal women, the mean LH level and the LH pulsatility pattern were similar on the 3 occasions. In contrast, the mean LH level in men was significantly reduced during hypocortisolism compared to that during eucortisolism (3.26 +/- 0.68 vs. 4.49 +/- 0.83 U/L; P < 0.05) and was associated with a clear decrease in LH pulse amplitude (1.09 +/- 0.33 vs. 1.96 +/- 0.53 U/L; P < 0.05). During high doses of glucocorticoids, the mean LH level in men was significantly lower than that during eucortisolism (3.81 +/- 0.88 vs. 4.49 +/- 0.83 U/L; P < 0.05). In both men and women, the mean PRL levels increased significantly (P < 0.05) during hypocortisolism, whereas high glucocorticoid doses suppressed the mean PRL level (P < 0.05). The LH and PRL responses to LH-releasing hormone and TRH were, however, similar during low, medium, and high cortisol levels in both men and women. In conclusion, our data suggest that the attenuation of pulsatile LH secretion in men during hypo- and hypercortisolism is due to variations in the hypothalamic opioid activity secondary to alterations in serum cortisol levels. A higher level of opioid receptor activity in men than in low estrogen women may explain the gender differences.