A_2b adenosine receptors can change their spots

Abstract

Recently, a central role for the A_2b adenosine receptor in a variety of cardiovascular functions including inflammation, erectile function, coronary artery dilation, asthma and cardioprotection has been demonstrated. Despite this evidence, the low-affinity A_2b adenosine receptor is still poorly understood. This receptor appears to be very promiscuous in its coupling. In most tissues, it couples to G_s much like its cousin, the A_2a adenosine receptor, but in mast cells and now, most recently, in cardiac fibroblasts, the A_2b receptor also couples to G_q. Because of its low affinity, this receptor was originally thought unlikely to play any important physiological role. But the sensitivity of A_2b adenosine receptors can be greatly increased by interaction with protein kinase C (PKC) making this receptor, under various conditions, both an activator and a target of PKC. We have recently documented a third coupling involving G_i. This plasticity and versatility of A_2b adenosine receptors position them as potential triggers of signalling in multiple signalling cascades in many physiological responses, making this a most interesting receptor indeed.

This article is a commentary on Feng et al., pp. 1598–1607 of this issue. To view this paper visit http://dx.doi.org/10.1111/j.1476-5381.2009.00558.x

Adenosine is a purine nucleoside that is widely distributed in all tissues and body fluids. Adenosine agonists and antagonists are being targeted for treatment of a variety of clinical conditions. These actions are attributed to binding of the purine to four distinct receptors: A₁, A_2a, A_2b and A₃. Although all four receptors have been characterized pharmacologically and have been cloned, our familiarity and appreciation of the A_2b adenosine receptor is still very limited because selective agonists and antagonists have only recently become available. And what we think we know about A_2b adenosine receptors may not be accurate.

All four receptors have been found in cardiac tissue, which is a mixed tissue composed of cardiomyocytes, fibrous tissue and fibroblasts, vascular endothelial and smooth muscle cells, mast cells, etc. A₁ and A₃ adenosine receptors have been shown to be present in cardiomyocytes (Grdeńet al., 2005). A_2a and A_2b adenosine receptors are both present in the coronary vasculature (Mubagwa and Flameng, 2001). Message for A_2a receptor has been identified in cardiomyocytes (Xu et al., 1996). The presence of A_2b receptors in cardiomyocytes has been controversial (Liang and Haltiwanger, 1995; Morrison et al., 2002; Grdeńet al., 2005; Yang et al., 2006), although we have detected A_2b receptor message in both rabbit and rat ventricular myocytes (unpublished observation). Given this new evidence that A_2b adenosine receptors were expressed in cardiomyocytes, we were very surprised when A_2a receptors, but not A_2b receptors, were found on the sarcolemma in single cardiomyocytes in which cAMP was monitored (Xin et al., 2009).

Binding of agonists to A₁ and A₃ adenosine receptors results in activation and cleavage of G_i leading to a decrease in cAMP and activation of PKC (Mubagwa and Flameng, 2001), whereas A_2a and A_2b receptors stimulate adenylyl cyclase and increase cAMP production by liberating components of G_s (Mubagwa and Flameng, 2001). The A_2a and A_2b adenosine receptors were first identified by their differential ability to stimulate cAMP production in brain slices at low (0.1–1 µM) and high (>10 µM) adenosine concentrations (Schulte and Fredholm, 2003). Thus the A_2b receptor became known as the low-affinity receptor. Using human embryonic kidney (HEK) 293 cells overexpressing human A_2b receptors Linden et al. (1999) found that A_2b receptors coupled not only to G_s but also to G_q/11 leading to activation of phospholipase C and mobilization of calcium in the transfected kidney cells. Although G_s coupling mediated vasodilation, the physiological significance of G_q/11 coupling was demonstrated in the degranulation of mast cells. Thus the A_2b adenosine receptor could trigger at least two distinct signalling cascades.

This receptor fickleness was confirmed by Feng et al. (2009) in the current edition of the British Journal of Pharmacology. They examined production of the proinflammatory cytokine interleukin-6 (IL-6) by mouse cardiac fibroblasts following exposure to 5′-(N-ethylcarboxamido) adenosine (NECA), a potent, albeit not selective, A_2b receptor agonist. Interestingly these fibroblasts expressed all four adenosine receptor subtypes. In response to NECA these cells produced significant amounts of IL-6, which was not mimicked by selective A₁, A_2a or A₃ receptor agonists. And only the selective A_2b receptor antagonist MRS 1754 could interfere with the response to NECA. Furthermore silencing of A_2b receptors with siRNA also suppressed NECA-induced IL-6 secretion. Clearly IL-6 production by cardiac fibroblasts was greatly influenced by stimulation of A_2b receptors. When Feng et al. (2009) studied the downstream signalling they found involvement of G_q and not G_s. Pretreatment with either the cAMP-competitive analog Rp-cAMPS or 1 of two protein kinase A (PKA) inhibitors, H-89 or KT 5720, before exposure to NECA failed to block NECA-induced IL-6 production. Nor could a cAMP analogue trigger IL-6 release. Therefore, the G_s-cAMP-PKA pathway was not involved in the regulation of IL-6 production by A_2b receptors. Although the authors did not directly test for G_q coupling, it was presumed to be responsible. These data confirm the plasticity of A_2b adenosine receptors, and demonstrate their complicated biology. This evidence of G_q coupling of A_2b receptors is probably as important as the observation that the heart can be its own source of inflammatory cytokines.

Interest in adenosine in the heart was stimulated when we found that adenosine released by a transient period of ischaemia (preconditioning) was a trigger for a signalling cascade that resulted in protection of myocardium from a subsequent more prolonged coronary occlusion (Liu et al., 1991). Although A₁ adenosine receptor agonists were powerful preconditioning agents, they had little clinical value. For treatment of patients with acute myocardial infarction, an intervention was needed that could be applied at the time of reperfusion. We first reported that A_2b receptors were critical to signalling at the onset of reperfusion in the preconditioned heart (Philipp et al., 2006). Thus both ischaemic preconditioning (Solenkova et al., 2006) and post-conditioning (repeated episodes of very brief coronary reocclusion in the first minutes of reperfusion) (Philipp et al., 2006) were aborted if A_2b receptors were blocked at reperfusion. But these observations posed a significant theoretical quandary. Under basal conditions, the extracellular concentration of adenosine ranges from 30 to 300 nM. Although the concentration increases substantially during ischaemia to 1–4 µM, it rarely exceeds 10 µM (Schulte and Fredholm, 2003). Yet the K_i of A_2b receptors for adenosine may be as high as 24 µM (Gao and Jacobson, 2007). Therefore, how could endogenous adenosine released from ischaemic heart muscle bind significantly to A_2b receptors? And why only in preconditioned hearts? In the study by Feng et al. in which the end point was IL-6 production, they also found that A_2b receptors activated PKC. However, in investigations of cardioprotection in which the end point was anatomical infarct size, Philipp et al. (2006) showed that in the cardioprotective signal transduction pathway, PKC is actually upstream of A_2b adenosine receptors, the reverse of the sequence reported by Feng et al. in cardiac fibroblasts. It is well known that PKC activity can sensitize A_2b receptor signalling, although the mechanism has never been elucidated nor has a physiological function been attributed to it (Feoktistov and Biaggioni, 1997). We (Kuno et al., 2007) tested whether this sensitization phenomenon might be involved in preconditioning. We found that a subthreshold dose of NECA, which, by itself could not increase phosphorylation of the survival kinases Akt and ERK in rabbit myocardium could activate them if the PKC activator phorbol 12-myristate 13-acetate (PMA) had been previously administered. This same effect was reproduced by a preconditioning cycle of ischaemia/reperfusion that would also have activated endogenous PKC. Therefore, we proposed that, in a preconditioned heart, PKC increased the sensitivity of A_2b receptor signalling to the point that endogenous adenosine released during ischaemia could now bind to A_2b receptors at reperfusion and initiate protective intracellular signalling. This observation again emphasized the versatility of A_2b receptors and justified the impression that this adenosine receptor could play an important physiological role in the myocardium.

Most recently Kuno et al. (2009) further examined A_2b receptor coupling to survival kinases in HEK 293 cells stably transfected with human A_2b receptors. As in the heart NECA caused a dose-dependent phosphorylation of Akt and ERK. Furthermore NECA increased cAMP consistent with the previously reported G_s-coupling of A_2b receptors. Yet when G_s was inhibited with cholera toxin or the G_s antagonist NF449, the ability of NECA to phosphorylate Akt and ERK was actually enhanced. On the other hand, Pertussis toxin abolished NECA-triggered kinase phosphorylation indicating either a G_i or G_o coupling. These studies have now been repeated with the A_2b adenosine receptor selective agonist BAY 60-6583, and the observations are identical (submitted for publication). So here we have yet a third coupling pathway for cardiac A_2b receptors. The biology of the A_2b receptor is much more complicated than that of the other adenosine receptors and may be the key to a host of clinically relevant therapies. Unlike the leopard, A_2b adenosine receptors can change their spots.

Of course, additional questions must still be addressed. Where are the A_2b receptors located if they are not on the sarcolemma? Are they associated with some intracellular organelle? Are they functional? Does the cardioprotective signalling cascade occur in the cardiomyocyte, or does A_2b receptor signalling occur on outer cell membranes of another tissue, perhaps vascular endothelium, with production of a diffusible messenger that then targets cardiomyocytes?

Glossary

Abbreviations:

IL-6

interleukin-6

NECA

5′-(N-ethylcarboxamido) adenosine