Although p-adrenergic control of another steroidogenic tissue, the ovary, has been established (Godkin et al., 1977; Jordan et al, 1978; Hunzicker-Dunn, 1980; Hunzicker-Dunn et al., 1979; Kliachko & Zor, 1981), there have been no reports that catecholamines can directly stimulate steroidogenesis in Leydig cells. During the course of our investigations on factors controlling Leydig activity in culture we found that catecholamines do stimulate testosterone production. Because of the possible physiological implications and also potential use in studying hormonal control of adenylate cyclase in Leydig cells (Dix & Cooke, 1981) more detailed investigations have been carried out. The time and dose-responses of the p-adrenergic hormone agonist, isoprenaline and adrenaline and the pantagonist, propranolol, on testosterone production have been determined and the role of cyclic AMP has been investigated. The results obtained indicate that physiological amounts of catecholamines stimulate testosterone production via cyclic AMP in mouse Leydig cells and that this response ‘develops’ in the Leydig cells after 1 day in culture and is then maintained for approx. 2 days. Adult mouse Leydig cells were isolated by mechanical dissociation of the testes followed by purification on a &90% Percoll gradient as described previously (Schumacher et al., 1978; Hunter et al., 1982). The Leydig cells (100000 cells/well) were cultured in Dulbecco’s modified Eagle’s medium (1 ml) with Kanamycin (0.1 ng/ml) (Gibco) and 10% foetal-calf serum (Gibco). The medium was changed daily and the cells were stimulated on day 2 of culture with catecholamines or lutropin as indicated for 2h in the presence of isobutylmethylxanthine (0.2 mM), unless otherwise stated. In a series of experiments carried out it was found that isoprenaline (10-0.1 PM) added to freshly prepared Leydig cells 2h after being plated out in the culture wells had little or no effect on testosterone production. However, addition of this catecholamine analogue on days 1, 2 and 3 of culture consistently stimulated testosterone production. The results from a representative experiment are shown in Fig. 1. In this experiment two preparations of Leydig cells were made each from 10 mice killed either by cervical dislocation or by gassing with CO,. There were no detectable differences in the results obtained from these two preparations. In contrast with testosterone production, isoprenaline did stimulate cyclic AMP production on day 0 of culture, which increased on each subsequent day of culture. The capacity of the Leydig cells to respond to isoprenaline in the stimulation of testosterone production varied with different preparations from 20 to 90% of that obtained with lutropin. Further experiments were carried out to determine if inclusion of foetal-calf serum in the culture medium influenced the response to catecholamines. The results are given in Table 1. In agreement with the results in Fig. 1, little or no stimulation by isoprenaline of testosterone production was obtained on day 0 of culture in the presence of either foetal-calf serum or 1.5% (w/v) albumin. However, stimulation of cyclic AMP levels on day 0 was approx. three times higher in the cells cultured with foetal-calf serum compared with albumin. On subsequent days, stimulation by isoprenaline of both testosterone and cyclic AMP was obtained with albumin in the culture medium but much higher levels were attained in the presence of foetal-calf serum. Higher lutropin-stimulated testosterone but not cyclic-AMP levels were obtained in the presence of foetal-calf serum compared with albumin on all days of culture. Dose-response experiments for isoprenaline showed that a small increase in testosterone production was obtained with 10nwisoprenaline and reached a maximum with approx. 1 PM. A parallel dose-response curve for isoprenaline-stimulated cyclic AMP production was obtained, the only difference being that a stimulation was not observed with IOnM. Isoprenaline stimulation of testosterone production was inhibited by the