Activation of mu-opioid receptors (MORs) transfected into human embryonic kidney 293 cells, caused a multiphasic increase in cytosolic free Ca(2+) levels (Ca(2+)i). The first Ca(2+)i maximum (peak 1) between 5 and 7 s depended on the presence of extracellular Ca(2+) (Ca(2+)e). The second phase peaking at approximately 15 s (peak 2) was independent of Ca(2+)e and thus represents Ca(2+) release from intracellular stores. A decrease in temperature from 37 to 25 degrees C also caused reduction of peak 1 but not peak 2, suggesting that the two responses arise from mechanistically distinct pathways. A delayed Ca(2+)e-dependent third response phase is thought to represent capacitative Ca(2+)e influx evoked after release of Ca(2+) from internal stores. Agonists and antagonists of two major classes of opioid ligands, oxymorphinans (morphine and naloxone) and oripavines (etorphine and diprenorphine), had differential effects on Ca(2+) currents. Although morphine activated both phases with equal potency, etorphine was 20-fold less potent at stimulating peak 1 over peak 2. Similarly, the antagonists, naloxone and diprenorphine, blocked the Ca(2+) response to each agonist with greatly varying potencies. Specifically, concomitant injection of diprenorphine failed to affect peak 1 (thought to represent rapid Ca(2+)e influx) stimulated by morphine while fully blocking peak 2 (intracellular Ca(2+) release). However, diprenorphine potently inhibited peak 1 as well when added to the cells before morphine, indicating limited or slow access of diprenorphine to these morphine binding sites. The existence of multiple, functionally distinct binding site conformations could account for these findings. In conclusion, different opioid ligands can differentially affect Ca(2+) response patterns resulting from MOR activation.