Based on the differential abilities of the opioid antagonists naltrexone and D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Pen-Thr-NH2 (CTP) to antagonize the antinociceptive action of beta-endorphin and morphine in the rat periaqueducatal gray (PAG), three pharmacologically distinct mechanisms were determined to mediate the antinociceptive effect of beta-endorphin. Two of these mechanisms are unique to beta-endorphin, possess a high affinity for CTP and can be discriminated based on their differential sensitivity to naltrexone. The third mechanism displays characteristics common to that activated by morphine. The results of radioligand binding studies were consistent with these observations. [125I]-beta-Endorphin labeled a population of sites in the PAG which (compared to those labeled by [3H]morphine) displayed a significantly higher affinity for CTP. In addition, a naltrexone-insensitive binding component was identified in the [125I]-beta-endorphin, but not [3H]morphine assays. Furthermore, comparable competitor affinities were determined across assays, suggesting an interaction of the radioligands with common PAG sites. A naltrexone-insensitive component to beta-endorphin antinociception also was identified in studies which evaluated the ability of the antagonist to shift the beta-endorphin dose-response curve. Interestingly, the ability of low doses of CTP and naltrexone to inhibit increasing doses of beta-endorphin was described by a U-shaped dose effect curve. The response to low and high, but not intermediate, doses of beta-endorphin were antagonized by picomole doses of both antagonists. As there was no evidence for allosteric interactions between [125I]-beta-endorphin binding sites in the PAG, it appears that beta-endorphin also may activate pain facilitory mechanisms which counterbalance its overall antinociceptive effect.