1 We have identi®ed the P 2 receptors mediating vasomotor responses in the rabbit pulmonary artery. 2 Neither ATP nor UTP contracted intact or endothelium-denuded rings. However, both relaxed intact rings of rabbit pulmonary artery that had been preconstricted with phenylephrine (pD 2 5.2 and 5.6, respectively). 3 The vasodilator e ect of UTP was endothelium-dependent and abolished by the nitric oxide synthaseThe vasodilator e ect of ATP was only partially inhibited by removal of endothelium or addition of L-NOARG, suggesting an additional direct e ect on vascular smooth muscle. 5 The endothelium-dependent vasodilator responses to UTP and ATP were competitively antagonized by suramin. 6 Preconstricted, endothelium-denuded rings were also relaxed by 2-methylthio ATP (pD 2 6.6), a P 2Y receptor agonist. 7 Ca 2+ -mobilizing P 2U receptors were identi®ed on smooth muscle cells on the basis of single cell responses to ATP (pD 2 7.8) and UTP (pD 2 7.9; 6.7 in the presence of 100 mM suramin). 8 There was no evidence of a Ca 2+ -mobilizing P 2Y receptor in these cultured cells. 9 The data suggest the presence of (i) a suramin-sensitive P 2U receptor on endothelial cells that induces vasorelaxation through NO release, (ii) a suramin-sensitive P 2U receptor on cultured smooth muscle cells that mobilizes Ca 2+ but is not coupled to vasomotor responses and (iii) a putative P 2Y receptor on vascular smooth muscle cells that induces relaxation via a Ca 2+ -independent signal transduction pathway.

The endothelin B receptor (ETB) is an endothelial cell receptor found in caveolae. Studies with GFP-tagged ETB have suggested that the protein is constitutively endocytosed and targeted to lysosomes where it is rapidly degraded. We report that iodinated endothelin-1 ligand (ET-1) is taken up by cells transfected with ETB and remains undegraded for at least 17 h. Analysis of the intracellular traffic of endocytosed ET-1 on isotonic Ficoll gradients shows that it is rapidly internalised to lysosomes by a chloroquine sensitive and cholesterol dependent pathway. Low-temperature nonreducing SDS gels show that the ET-1 initially binds to full-length GFP-tagged ETB, which is rapidly clipped at the amino-terminus and is then stable for at least 6 h. Analysis of GFP tagged ETB on reducing SDS gels shows that it is proteolytically cleaved with a half time of approximately 3 h. However, nonreducing gels show that the receptor is virtually intact, suffering only a similar cleavage to the liganded receptor. We conclude that the ETB receptor shows remarkable stability in lysosomes, held together by disulfide bonds, and maintaining ligand binding for long periods of time.

1. The purpose of the present study was to examine the effects of papaverine-HCl, administered into the lumen of the human internal mammary artery (IMA) during harvesting of this vessel, on vascular reactivity in vitro and to specifically test the hypothesis that intraluminal administration of papaverine-HCl impairs endothelium-dependent vasodilation. 2. The present study measured in vitro dilator and constrictor responses of terminal segments of human IMA. Internal mammary artery segments were obtained either prior to routine administration of intraluminal papaverine (pre-P) or after papaverine administration (post-P) in patients undergoing coronary artery bypass grafting. In addition, the viability of cultured human saphenous vein endothelial cells exposed to papaverine-HCl was examined. 3. Cumulative concentrations of U46619, 5-hydroxytryptamine and phenylephrine (PE) produced active contractions in post-P IMA rings. Contractile responses to low concentrations of endothelin-1 were significantly enhanced in post-PIMA compared with pre-P IMA segments. 4. Maximal endothelium-dependent vasodilator responses of pre-P IMA segments to cumulative concentrations of acetylcholine (ACh) and the calcium ionophore A23187 were 49 +/- 7 and 66 +/- 4%, respectively, of the initial active tension induced by PE (1 mumol/L). 5. Maximal endothelium-dependent vasodilator responses were markedly attenuated in post-P IMA (6 +/- 6 and 11 +/- 10% for ACh and A23187, respectively; P< 0.0001 for both vasodilators compared with pre-P). Post-P IMA relaxed completely to the endothelium-independent vasodilator sodium nitroprusside. 6. Exposure of cultured human saphenous vein endothelial cells to papaverine-HCl (1.2 and 12.0 mg/mL) for 1 h resulted in the reduced viability of these cells. 7. The loss of endothelium-dependent relaxation could dangerously predispose the IMA graft to vasospasm in the postoperative period.

Interactions between vascular endothelial cells and blood platelets have been investigated using a model microcirculation consisting of microcarrier beads colonized with human umbilical vein endothelial cells (HUVECs) and perfused with washed platelet suspensions. To simulate the effects of endothelial desquamation and exposure of subendothelium, fibrillar collagen in suspension was coinjected with the platelets. In this model, neither the passage of platelets alone nor collagen alone stimulated prostacyclin (PGI2) production by the HUVECs. Platelets activated by coinjection with collagen released thromboxane A2 (TXA2), and this was associated with the simultaneous production of PGI2 by the HUVECs. By means of double-isotope experiments with [3H]arachidonic acid (AA) incorporated into platelets and [14C]-AA into HUVECs, it was shown that all the PGI2 generated was derived from platelet AA and/or endoperoxides. This interpretation was strengthened by the finding that PGI2 production was not prevented by treatment of HUVECs with indomethacin followed by perfusion with collagen-stimulated platelets. AA metabolites in double-isotope label experiments were further characterized by reverse-phase chromatography, and it was shown that both cyclooxygenase and lipoxygenase products of the HUVECs were derived from platelet membrane lipid. Thrombin regularly produced transient PGI2 release, but showed rapid tachyphylaxis. Platelet-derived compounds including ADP, ATP, and platelet-activating factor (PAF) did not produce PGI2 release by HUVECs in this system. Thus, the transfer of AA and metabolites from collagen- stimulated platelets is likely to be the mechanism for PGI2 production in the context of minor degrees of endothelial desquamation.