Interleukin 6 is a pleiotropic cytokine produced in the central nervous system (CNS) that has been involved in both direct neurotrophic activities and in the regulation of the production of acute phase proteins both at peripheral and central levels. In rat cortical type I astrocytes, interleukin 6 release is under the control of cAMP-protein kinase A and calcium-phospholipids-protein kinase C systems. Somatostatin is a neuropeptide, acting as a neurotransmitter, highly concentrated within the CNS, where it has been involved in the modulation of learning and memory processes. The aim of this study was to characterize the effects of somatostatin on the release of interleukin 6 from rat cortical type I astrocytes and the intracellular mechanisms involved in this activity. Our results show that somatostatin, in a concentration-dependent manner, inhibited basal and forskolin-stimulated interleukin 6 release from rat cortical type I astrocytes in culture. The EC50 of the inhibitory action was calculated to be approximately 10 nM. Furthermore, this effect of somatostatin was completely abolished by pretreating cortical astrocytes with pertussis toxin that, uncoupling, by ADP-rybosylating, the inhibitory GTP-binding protein from the receptors, prevents the activation of the intracellular effectors such as the adenylyl cyclase enzyme. To identify the intracellular mechanism mediating the effects of somatostatin on the interleukin 6 release, we evaluated the peptide modulation of basal and stimulated intracellular accumulation of cAMP. In our experimental conditions somatostatin significantly inhibited both basal and forskolin-stimulated cAMP accumulation. Conversely, somatostatin did not affect the increase of interleukin 6 release induced by dibutyryl-cAMP, a nonhydrolizable cAMP analog that, bypassing the effects of somatostatin on adenylyl cyclase activity, directly activated protein kinase A. These observations support the hypothesis that somatostatin inhibitory activity on interleukin 6 release is mediated by its effects on cAMP production. Somatostatin analog SMS 201-995 did not affect interleukin 6 production either in basal or stimulated conditions. Since, SMS 201-995 was reported to bind with high affinity only to somatostatin receptors type 2, 3 and 5, the lack of effect of this compound on interleukin 6 release suggests that the inhibitory action of somatostatin could be mediated by the activation of either type 1 or type 4 somatostatin receptors. In conclusion, our data demonstrate that the release of interleukin 6 from rat cortical type I astrocytes is inhibited by somatostatin through the activation of a somatostatin receptor coupled to the inhibition of adenylyl cyclase via a G-protein sensitive to pertussis toxin.