Thimerosal blocks stimulated but not basal release of endothelium-derived relaxing factor (EDRF) in dog isolated coronary artery.

1. The effect of an acetly-coA lysolecithin acyltransferase inhibitor, thimerosal, on the release of endothelium-derived relaxing factor (EDRF) was examined in the greyhound isolated coronary artery. 2. Thimerosal (1-10 microM) relaxed fully, ring segments of coronary artery which were contracted with the thromboxane A2-mimetic, U46619 (30 nM). The response was endothelium-dependent, slow in both onset and time to reach maximum. The maximum relaxation to the highest concentration of thimerosal (10 microM) was maintained for 10-20 min before the tissue slowly regained active force (1-2 h) to the same or higher level as that prior to the addition of thimerosal. At this time the endothelium-dependent relaxation responses to acetylcholine (ACh), substance P (SP), bradykinin (BK) and the calcium ionophores, ionomycin and A23187 were abolished. The endothelium-dependent contractions to the nitric oxide synthase inhibitors, NG-nitro-L-arginine (L-NNA; 10-100 microM) and NG-monomethyl-L-arginine (L-NMMA: 10-100 microM), however, were unaffected. 3. Thimerosal (10 microM) did not affect the relaxation curve to sodium nitroprusside (SNP) nor the contraction curve to the thromboxane A2-mimetic, U46619. 4. Both the relaxation response to thimerosal and the selective block of the relaxation responses to stimulated EDRF release were unaffected by either indomethacin (10 microM) or superoxide dismutase (150 u ml-1). 5. L-NNA (100 microM) significantly blocked the relaxation curves to thimerosal and A23187 but not that to SNP.6. Abolition of stimulated EDRF-mediated responses with thimerosal was unlikely to result from maximal and maintained stimulation of EDRF release even when active U46619-induced force had returned to pre-thimerosal levels, since the relaxation curves to glyceryl trinitrate (GTN) and SNP were markedly attenuated in the presence of SNP and GTN respectively when active force was restored with endothelin-1 (ET-1).7. Melittin (1 microM), ionomycin (1 microM) and A23187 (1 microM) each had selective effects on stimulated but not basal EDRF responses, similar to those of thimerosal.8. We propose that stimulated but not 'basal' release of EDRF is dependent on the release of arachidonic acid or one of its non-cyclo-oxygenase metabolites, possibly by Ca2'-dependent activation of phospholipase A2.