Progesterone rapidly inhibits glucose oxidation of isolated rat adipocytes. Because this inhibition is triggered by endogenous adenosine, the present study was designed to examine the effect of the steroid on cyclic adenosine monophosphate (cAMP) accumulation, its relation to lipolysis, and the possible participation of adenosine. The results strongly indicate that physiological concentrations of progesterone increase the release of adenosine by isolated adipocytes, with maximal release at the end of a 20-minute incubation. Progesterone decreased both cAMP levels and lipolysis in quiescent adipocytes or in adipocytes stimulated by isoproterenol. The increase of endogenous adenosine may explain the decline of cAMP and glycerol levels observed with progesterone. The effects of the steroid on lipolysis disappeared when adenosine was hydrolyzed by adenosine deaminase (ADA). On the other hand, in the absence of endogenous adenosine, the effect of progesterone on the cAMP level was decreased only in isoproterenol-stimulated cells. The inhibitory effects of progesterone on cAMP and glycerol production seem not to be related directly to the adenosine A1 receptor, for selective A1 receptor antagonists (8-cyclopentyl-1,3-dipropylxanthine [DPCPX] and CP 68,247) did not counteract these effects. However, mechanisms mediated by guanyl nucleotide binding proteins cannot be excluded. The decrease of cAMP and of lipolysis may be related to a stimulation of phosphodiesterases (PDEs). When PDEs I [Ca(2+)-calmodulin-regulated PDE family) were blocked by a selective inhibitor (CP 41,757), the progesterone inhibitory effect persisted, suggesting that PDEs I are not regulated by the steroid. On the other hand, the progesterone effect on cAMP accumulation but not on lipolysis disappeared in the presence of a selective inhibitor of the PDE IV family (cAMP-dependent-specific family). Ro 20.1724. When the specific inhibitor of PDE I or PDE IV was combined with ADA, the progesterone effect on cAMP disappeared. Taken together, these results suggest that the progesterone inhibitory action on cAMP levels was not mediated through A1 receptors or through activation of PDE I, but may be related to PDE IV activities. The progesterone effect on lipolysis seemed not to be directly related to changes in cAMP levels; an effect on PDE III activities in relation with the increase of adenosine release cannot be excluded.