Agonist and antagonist effects of mixed action opioids in the pigeon drug discrimination procedure: influence of training dose, intrinsic efficacy and interanimal differences.

The stimulus effects of selective, high efficacy mu opioids and mixed action opioids with varying degrees of intrinsic efficacy at the mu receptor were examined in pigeons trained to discriminate between saline and either 0.056 (low) or 0.18 (high) mg/kg of fentanyl. The stimulus profiles produced by the various opioids could be separated into three groups: 1) opioids that substituted completely for both training doses of fentanyl, with steep slopes and little interanimal differences in the lowest dose (lowest discriminable dose) that produced complete substitution (fentanyl, morphine and l-alpha-acetylmethadol); 2) opioids that substituted completely for the low training dose and produced high levels of substitution for the high training dose, with relatively shallow slopes and interanimal differences in the lowest discriminable dose (butorphanol, buprenorphine, ethylketocyclazocine, ketocyclazocine, proxorphan, (-)-pentazocine and (-)-metazocine); and 3) opioids that substituted completely for the low training dose, with relatively shallow slopes and large interanimal differences in the lowest discriminable dose. Each of these opioids also antagonized the high-dose fentanyl stimulus with large interanimal differences in the lowest antagonist dose (nalbuphine, nalorphine, (-)-cyclorphan, (-)-cyclazocine, (-)-n-ally-normetazocine and levallorphan). These patterns of substitution and antagonism most likely reflect differences in the intrinsic efficacy of these drugs at the mu receptor, with low intrinsic efficacy associated with shallow dose-effect functions, large interanimal differences in the drug's lowest discriminable dose and low levels of substitution for the high-dose fentanyl stimulus. During antagonism tests with naloxone, two patterns were observed: 1) opioids against which naloxone had apparent pA2 values of approximately 7.0, with little interanimal differences and with the slopes of the Schild plots approximating -1.0 (fentanyl and morphine) and 2) opioids against which naloxone had apparent pA2 values an order of magnitude higher, with large interanimal differences and with the slopes of Schild plots being relatively shallow (butorphanol, nalbuphine, nalorphine and levallorphan). The present findings emphasize the importance of training dose, intrinsic efficacy and interanimal differences when analyzing drug discrimination data.