The involvement of delta opioid receptors in supraspinal analgesia was investigated. With this aim, opioids that produced analgesia in the tail immersion test were administered i.c.v. to mice a few minutes before the irreversible antagonist, beta-funaltrexamine (beta-FNA). Protection of the respective analgesic effects from beta-FNA blockade was obtained when evaluated 24 h later. Moreover, mu ligands protected the analgesia evoked by ED50s of morphine, [D-Ala2,N-Me-Phe4,Met-(o)5-ol]enkephalin (FK 33-824), [D-Ala2,N-Me-Phe4,Gly-ol5]enkephalin (DAGO) and human beta-endorphin at doses (ED50s) lower than those required for delta ligands (approximately ED90s) to reach a similar protection. delta Preferential ligands effectively protected the analgesia induced by ED50s of [D-Ala2,D-Leu5]enkephalin (DADLE), [D-Thr2,Leu5]enkephalin-Thr6 (DTLET) and [D-Pen2,D-Pen5]enkephalin (DPDPE) from the beta-FNA-deteriorating effect. FK 33-824 and DAGO also provided good protection of the analgesia elicited by these delta ligands whereas morphine protected much less. Binding studies after i.c.v. injection of beta-FNA showed that its alkylating effect on opioid receptors was restricted to periventricular areas. In PAG, where the mu/delta receptor ratio is about 10, [3H]DADLE specific binding was still present after ED50s of DPDPE, DAGO, morphine and DADLE as protecting agents. [3H]Dihydromorphine [( 3H]DHM) binding was well protected by ED90s of morphine and DAGO, and to a lesser extent by DPDPE and DADLE. These results suggest that delta ligands, after binding to delta receptors, also need to act upon mu receptors to produce high levels of supraspinal analgesia in the tail immersion test.