Analysis of relaxation and repolarization mechanisms of nicorandil in rat mesenteric artery.

1. The mechanisms by which nicorandil causes relaxation of rat isolated small mesenteric arteries mounted on a Mulvany myograph was investigated by use of a combination of putatively mechanism-specific antagonists. 2. In arteries precontracted by the thromboxane-mimetic, U46619, the EC50 for cromakalim and levcromakalim-induced relaxation curves were raised by 36 and 17 fold by glibenclamide (3 microM) while the EC50 for nicorandil-induced relaxation was unaffected by either glibenclamide or methylene blue (10 microM). A combination of these antagonists raised the EC50 for nicorandil by 8 fold. 3. In U46619-contracted arteries, nifedipine (100 nM) did not affect the cromakalim relaxation curve but it raised the EC50 for nicorandil by 5 fold. The combination of methylene blue, glibenclamide and nifedipine further inhibited the maximum relaxation to nicorandil. 4. In separate experiments, membrane potential (Em) and force responses were measured simultaneously. In arteries depolarized and contracted by U46619 both nicorandil and cromakalim repolarized (delta Em, 35 mV) and relaxed (100%) the vessels. Glibenclamide (3 microM) did not alter the relaxation-concentration curve to nicorandil but reduced the maximum repolarization to delta 10.8 mV. In contrast to Em and relaxation-response curves to cromakalim were shifted to the right by glibenclamide by 30-100 fold. 5. In unstimulated arteries, nicorandil (but not cromakalim) -induced hyperpolarization was significantly antagonized by methylene blue (10 microM) (6.6 fold rightward shift) or nifedipine (100 nM) (2.6 fold). In depolarized arteries (U46619), nifedipine but not methylene blue inhibited the nicorandil-induced hyperpolarization. 6. In arteries precontracted to 50% tissue maximum by either KCl or U46619, nifedipine (100 nM) relaxed the artery but failed to repolarize the Em. Presumably voltage-operated calcium channels (VOCC) were blocked preventing contraction but the artery remained depolarized, presumably through non VOCC mechanisms. 7. These data suggest that nicorandil may relax small arteries through 3 parallel pathways, (i) NO-donor mediated stimulation of guanylate cyclase and increase in cyclic GMP, (ii) K+ATP channel opening, and (iii) nifedipine-sensitive VOCC inhibition. Em data suggest that nicorandil-induced repolarization is caused principally through opening K+ATP channels. Blockade of this hyperpolarization by glibenclamide is not sufficient to alter the relaxation, indicating dissociation of nicorandil-induced changes in membrane potential and relaxation. 8. These results highlight the 'chameleon' actions of nicorandil where there is no apparent association of Em repolarization with relaxation, in contrast to the parallel responses for cromakalim.