Although it is generally accepted that the generation of an EDHF (endothelium derived hyperpolarizing factor) response follows calcium dependent endothelial cell hyperpolarization, recent data from rabbit mesenteric arteries has challenged this concept (Vequaud & Thorin, 2001). We have modified a technique for cellular loading of large membrane impermeant molecules for use in pressurized arteries, and selectively loaded heparin into endothelial cells in rat mesenteric artery to block IP₃-receptors. We demonstrate directly that increases in endothelial cell calcium are necessary for EDHF mediated vasodilatation in this vessel.

Male Wistar rats (200-250 g) were killed humanly and sections of a third order branch of the superior mesenteric artery (i.d 250-300 µm) then mounted in a pressure myograph at 50 mmHg and perfused with MOPS at 37 °C. In some experiments, endothelial cells were loaded with fluo-4 AM. The endothelial cells were selectively loaded with low molecular weight heparin (15 kDa) using an adaptation of the osmotic loading protocol developed for cells in culture by Okada & Rechsteiner (1982). Arteries were then contracted with phenylephrine (PE, 3 µM) and vasodilatation stimulated with acetylcholine (ACh, 10 nM-3 µM) or cyclopiazonic acid (CPA, 15 µM) in the presence of the NO synthase inhibitor, 100 µM L-NAME.

Increasing concentrations of ACh evoked a concentration-dependent dilatation (EC₅₀ 177 nM; n = 8), which was unaltered by the osmotic loading protocol (EC₅₀ without heparin present 170 nM; n = 8). The incorporation of heparin caused a significant rightward shift in the concentration response curve to ACh (EC₅₀ 822 nM; n = 5; P < 0.0001, Student’s t test) and reduced the maximum response. Dilatation to 300 nM ACh was reduced from 76 ± 4 % (n = 8) to 22 ± 5 % (n = 5; P < 0.001) or 14 ± 5 % (n = 7; P < 0.001) by 10 mg ml^-1 or 20 mg ml^-1 external heparin, respectively, while contractions to 3 µM PE were unaltered. The ability of endothelial cells to store and release Ca²⁺ was not modified by loading heparin, as dilatation to CPA (15 µM) was unaltered (n = 3). However, heparin (20 mg ml^-1) did reduce the rise in intracellular Ca²⁺ normally evoked by 300 nM ACh. Peak increases in cytoplasmic Ca²⁺ after 10 seconds exposure to ACh (53 ± 10 %) were greatly attenuated by the incorporation of heparin (10 ± 2 %) (n = 4; P < 0.05). These data indicate that selectively loading endothelial cells with heparin significantly reduces the calcium increase normally evoked by ACh. This reduction was associated with marked surpression in EDHF mediated dilatation.

Together, these data provide direct evidence that EDHF evoked smooth muscle relaxation requires Ca²⁺ release, presumably from IP₃ sensitive intracellular stores in endothelial cells in rat mesenteric small arteries.

This work was supported by the Wellcome Trust