ATP: a double‐edged signalling molecule regulating the flow of oxygen

Abstract

The flow of blood and oxygen to exercising muscles of humans and animals is closely coupled to the oxygen demand of the contracting myocytes. This holds true over a wide range of exercise modalities, intensities and environmental or experimental conditions that alter blood oxygenation such as hypoxia (Gonzalez-Alonso et al. 2002; Saltin, 2007). This circulatory response matching muscle oxygen delivery to oxygen demand under most exercise conditions is thought to mainly involve the competing influences of locally released vasodilator signals and sympathetic neural vasoconstrictor activity, the contributions of which are altered with increases in the amount of muscle mass engaged during exercise. Clearly, sympathetic activation can diminish muscle perfusion when unopposed by vasodilator signals. In contracting muscle, however, sympathetic vasoconstriction is attenuated or even abolished presumably via increased concentration of metabolic byproducts (metabolites), allowing elevations in muscle blood flow despite increases in sympathetic vasoconstrictor activity (Remensnyder et al. 1962). Metabolites modulate sympathetic vasoconstriction by interfering with signal transduction pathways subservient to the activation of postjunctional α1- and α2-adrenoreceptors located on the vascular smooth muscle surrounding the skeletal muscle microvessels (McGillivray-Anderson & Faber, 1990). Over the years, a variety of muscle, interstitial and circulating metabolites (e.g. H+, Pi, K+, prostaglandins, adenosine, NO) have been proposed to be involved in the modulation of local sympathetic vasoconstrictor tone and the matching of muscle perfusion to oxygen demand. To date, ATP is the only metabolite shown to be capable of having the dual function of causing substantial vasodilatation and blunting sympathetic vasoconstriction in skeletal muscle. While the vasodilatory capacity of ATP has long been recognized (Folkow, 1949; Ellsworth, 2004), its sympatholytic effect was only recently discovered. In 2004, Rosenmeier and colleagues provided the first evidence that circulating ATP could be a signal involved in the attenuation of sympathetic neural vasoconstriction in the skeletal muscle microcirculation. Using intra-arterial administration of tyramine to evoke endogenous noradrenaline release from the sympathetic nerve endings, they found evidence that circulating ATP can override sympathetic vasoconstriction in the resting human leg circulation in a similar manner to that observed during exercise (Rosenmeier et al. 2004). This ‘sympatholysis’ occurs despite equal increase in noradrenaline and in the absence of metabolites from contracting muscle, suggesting that the modulatory effect of ATP happens at the level of postjunctional α-receptors. Whether circulating ATP directly modulates postjunctional α-adrenoceptor responsiveness and whether this is selective for α1- or α2-adrenoreceptors could not be addressed in that study. In this issue of the Journal of Physiology, Kirby et al. (2008) report several key observations in the human forearm using selective α1- and α2-adrenoreceptor agonists instead of indirect stimulation via tyramine, which shed more light into the role of ATP as a sympatholytic agent in skeletal muscle. The results clearly show that the forearm vasoconstrictor responses to direct α1- and α2-adrenoreceptor stimulation with phenylephrine or dexme-detomidine infusion were abolished during ATP infusion, but were only blunted during adenosine infusion and moderate rhythmic handgrip exercise. The differential response of adenosine and ATP as modulators of postjuntional α-adrenergic vasoconstriction was also cleverly demonstrated in a follow-up protocol showing that the α1-mediated vasoconstrictor responses to graded increases in ATP infusion were significantly blunted during moderate and high ATP infusion doses, whereas the vasoconstrictor responses with increasing adenosine dose were progressively greater. The important contributions of the study of Kirby et al. (2008) are the demonstrations that exogenous ATP can abolish both postjunctional α1- and α2-adrenergic vasoconstriction and that this response is graded with the rate of ATP infusion and thus the magnitude of ATP-induced hyperaemia. It must be emphasized, however, that the present results do not address whether the sympatholytic effects of ATP are mediated via ATP itself or its degradation products ADP and AMP. Ruling out a significant contribution of ATP dephosphorylated metabolites, experiments in the human leg model (Rosenmeier, et al. 2008) show that neither ADP nor AMP or adenosine infusion abolishes tyramine-mediated increases in vasoconstrictor drive. This contrasts sharply with the total blunting of sympathetic vasoconstriction produced during ATP and UTP infusion. We postulate that the sympatholytic effects of ATP in the skeletal muscle vasculature are largely mediated via ATP itself. Two issues that warrant further inquiry are the precise signalling mechanisms implicated in ATP-induced sympatholysis and the implications of the present findings for the vascular control in contracting muscle across the full range of exercise hyperaemia. Regarding the latter, it is important to bear in mind that the present results during moderate forearm hyperaemia might not be directly applicable to conditions of intense and maximal exercise hyperaemia where sympathetic vasoconstrictor activity is very high and muscle vasoconstriction is apparent (Mortensen et al. 2008). Notwithstanding, the paper of Kirby et al. (2008) clearly highlights that ATP's dual vasodilator and sympatholytic properties make it a very attractive signalling molecule for controlling the flow of blood and oxygen to skeletal muscle.