Despite the existence of a large body of information on the subject, the mechanisms of opiate tolerance and dependence are not yet fully understood. Although the traditional mechanisms of receptor down-regulation and desensitization seem to play a role, they cannot entirely explain the phenomena of tolerance and dependence. Therefore, other mechanisms, such as the presence of antiopiate systems and the coupling of opiate receptors to alternative G-proteins, should be considered. A further complication of studies of opiate tolerance and dependence is the multiplicity of endogenous opiate receptors and peptides. This review will focus on the endogenous opioid system--peptides, receptors, and coupling of receptors to intracellular signaling via G-proteins--in the context of their roles in tolerance and dependence. Opioid peptides include the recently discovered endomorphins and those encoded by three known genes--pro-opiomelanocortin, pro-enkephalin, and pro-dynorphin. They bind to three types of receptors--mu, delta, and kappa. Each of the receptor types is further divided into multiple subtypes. These receptors are widely known to be coupled to G-proteins of the Gi and Go subtypes, but an increasing body of results suggests coupling to other G-proteins, such as Gs. The coupling of opiate receptors to Gs, in particular, has implications for tolerance and dependence. Alterations at the receptor and transduction level have been the focus of many studies of opiate tolerance and dependence. In these studies, both receptor down-regulation and desensitization have been demonstrated in vivo and in vitro. Receptor down-regulation has been more easily observed in vitro, especially in response to morphine, a phenomenon which suggests that some factor which is missing in vitro prevents receptors from down-regulating in vivo and may play a critical role in tolerance and dependence. We suggest that antiopiate peptides may operate in vivo in this capacity, and we outline the evidence for the antiopiate properties of three peptides: neuropeptide FF, orphanin FQ/nociceptin, and Tyr-W-MIF-1. In addition, we provide new results suggesting that Tyr-W-MIF-1 may act as an antiopiate at the cellular level by inhibiting basal G-protein activation, in contrast to the activation of G-proteins by opiate agonists.