This study investigates the mechanism of CGRP-induced relaxation in intramural coronary arteries by determining the effect of CGRP on cytosolic Ca(2+) concentration (Ca(2+)) using FURA-2 technique. 2. CGRP concentration-dependently (10 pM - 100 nM) decreased the Ca(2+) and tension of coronary arteries precontracted with either U46619 or BAY K 8644, and also of resting coronary arteries in PSS. In 36 mM K(+)-depolarized arteries, CGRP reduced only the tension without affecting the Ca(2+). 3. In 300 nM U46619- precontracted arteries, pretreatment with 10 microM thapsigargin significantly (P<0.05) attenuated the CGRP-induced reduction in the tension (but not Ca(2+)). 4. In 300 nM U46619-precontracted arteries, pretreatment with either 100 nM charybdotoxin or 100 nM iberiotoxin or 10 nM felodipine significantly (P<0.05) attenuated the CGRP-induced reduction in both Ca(2+) and tension. In contrast, 1 microM glibenclamide did not affect the CGRP-induced responses in these coronary arteries. 5. In resting coronary arteries, only pretreatment with the combination of 1 microM glibenclamide and 100 nM charybdotoxin attenuated the CGRP-induced decrease in the Ca(2+) and tension, suggesting a different mechanism of action for CGRP in resting coronary arteries. 6. We conclude that CGRP relaxes precontracted rat coronary arteries via three mechanisms: (1) a decrease in Ca(2+) by inhibiting the Ca(2+) influx through membrane hyperpolarization mediated partly by activation of the large conductance Ca(2+)-activated potassium channels, (2) a decrease in Ca(2+) presumably by sequestrating cytosolic Ca(2+) into thapsigargin-sensitive Ca(2+) storage sites and (3) a decrease in the Ca(2+)-sensitivity of the contractile apparatus. In resting coronary arteries, however, there seems to be an interplay between different types of K(+) channels.