Endothelium-derived vasoconstriction by purines and pyrimidines.

The article by Tölle et al1 claims to show for the first time that adenosine 5 -tetraphosphate (AP4) is released on mechanical stimulation from human microvascular endothelial cells in the perfused rat kidney and, further, that AP4 is the most potent mediator of vascular smooth muscle constriction via P2X1 receptors or, indeed, via noradrenaline. It is well established that ATP and UTP released from endothelial cells in response to sheer stress produced by changes in blood flow act largely on P2Y receptor subtypes (but also some P2X receptor subtypes) on endothelial cells to release nitric oxide (NO), leading to vasodilatation (Figure).2–5 Endothelium-derived contracting factors have been identified, notably endothelin-1, prostaglandin H2, thromboxane A2, and superoxide anions.6,7 Tölle et al1 present compelling evidence for the release of AP4 from endothelial cells in response to mechanical stimulation, which then acts as a vasoconstrictor of the smooth muscle of microvessels in the kidney via P2X1 receptors. The presence of P2X1 receptors on vascular smooth muscle is well established, and they have been shown to respond to ATP released as a cotransmitter with noradrenaline from perivascular sympathetic vasoconstrictor nerves.8 However, P2X1 receptors have also been described on endothelial cells of human internal mammary and radial arteries and saphenous vein.9 Occupation of endothelial P2X1 receptors in rat mesenteric arteries resulted in a small vasoconstriction, followed by a profound and sustained endothelium-dependent vasodilatation, although not via NO.10 In P2X1 knockout mice, the vasoconstrictor response to ATP released by nerve stimulation is abolished.11 However, whether the constrictive responses to AP4 are also abolished in P2X1 knockout mice has not been examined, nor are there any studies of changes in P2X1 receptor-mediated endothelium-dependent vasodilatation. Presumably there are no P2X1 receptors on endothelial cells of the kidney microvessels; otherwise, there would be competing vasodilator effects of AP4. A study of the role of P2X1 receptors in renal microvascular autoregulatory behavior in response to increases in renal perfusion pressure suggested that ATP released from macula densa cells12 was mediated by P2X1 receptors and the reduction of the autoregulatory responses in P2X1 knockout mice supported this hypothesis.13 AP4 activates P2Y, as well as P2X receptors.14 The decreasing effect of AP4 on blood pressure is mediated by P2Y receptors on endothelial cells, but under certain conditions, such as hemorrhage, AP4 produces vasoconstriction via smooth muscle P2X receptors, where it was noted to be more potent than ATP.15 In addition to AP4, uridine adenosine tetraphosphate (Up4A) was also identified as a highly potent purinergic endothelium-derived vasoconstrictor by this group.16 However, in their present article, the researchers show that AP4 is more potent than Up4A, being an active vasoconstrictor in nanomolar concentrations. It is puzzling why the actions of ATP released from endothelial cells in response to shear stress are directed largely to endothelial cell P2 receptors, leading to vasodilatation (rather than smooth muscle P2X1 receptors), whereas AP4 released from endothelial cells is