Adenosine and inosine increase cutaneous vasopermeability by activating A3 receptors on mast cells (original) (raw)

Adenosine and inosine can have dramatic effects on vascular permeability, with edema formation occurring rapidly upon cutaneous exposure to these nucleosides. We show here that these physiological changes are mediated through the binding of adenosine and inosine to the A3 adenosine receptor. Lack of plasma protein extravasation after the intradermal administration of adenosine in mast cell–deficient mice fails to support a model in which adenosine induces changes in vascular permeability by acting directly on adenosine receptors expressed by endothelial cells. Rather, these findings are consistent with the hypothesis that adenosine and inosine mediate changes in vascular permeability indirectly by activation of tissue mast cells, which in turn release secondary mediators that act on endothelial cells. The experiments described here also show that adenosine and inosine alone can provide, through activation of the A3 receptor, sufficient signal for degranulation of intradermal mast cells in vivo. These findings contrast with in vitro studies using BMMCs, in which exposure to adenosine or inosine alone does not result in mast-cell degranulation.

Previous work has shown that adenosine can influence vascular permeability in multiple tissues, and our results showing edema formation in the mouse after the intradermal injection of adenosine are consistent with these studies (9–11, 17). It has previously been difficult to definitively assign a specific adenosine receptor to a given physiological response owing to the incomplete selectivity of adenosine receptor agonists and antagonists. Our studies show that plasma protein extravasation in response to adenosine is mediated entirely through activation of the A3 receptor, as adenosine cannot provoke this physiological response in mice lacking this receptor. These findings are consistent with reports showing that pharmacologic reagents that preferentially bind to the A3 receptor are effective at inducing edema formation (30, 38).

To determine whether adenosine-mediated increases in cutaneous vasopermeability are due to direct effects on the vasculature or occur indirectly through mast cell–mediator release, we used mast cell–deficient mice and their congenic littermate controls. Control mice showed the expected increase in vascular permeability in response to adenosine, whereas mast cell–deficient mice showed no response. These results suggest that mast cells are required for the induction of plasma protein extravasation by adenosine and that adenosine does not act directly on the vasculature to produce these physiological changes. Earlier in vitro studies using microvascular endothelial cells have shown adenosine to have direct effects on permeability (17). It is possible that these discrepancies reflect a difference in the site and cell type exposed to adenosine. In the experiments described here, adenosine was delivered to the interstitial space. It is possible that high levels of adenosine delivered to the luminal surface of the endothelial cell may directly alter postcapillary permeability. In vivo studies have also suggested that adenosine may have direct effects on the vasculature; however, the results of most of these studies are confounded by the presence of perivascular mast cells (9, 10). Recently, studies of vascular permeability in the skin of conscious rats after exposure to adenosine analogues have suggested an indirect action through mediator release from mast cells, as plasma protein extravasation after the intradermal administration of an adenosine analogue was nearly eliminated when animals were pretreated with either a histamine and serotonin antagonist or compound 48/80 that depletes mast-cell mediators (30). Our results are in agreement with these observations and provide evidence that adenosine’s effects on cutaneous vascular permeability are mast-cell mediated. Formal proof that these effects are indeed mast-cell mediated awaits successful and unsuccessful restoration of the response to adenosine in mast cell–deficient mice reconstituted with A3AR+/+ and A3AR–/– BMMCs, respectively.

Given our findings that the edema response to adenosine is mast cell–dependent, and that this response is absent in the A3AR-deficient mice, 2 possible explanations need to be considered. One possibility is that the failure to see a response in the A3AR-deficient mice is due to the nonresponsiveness of mast cell to adenosine because of the absence of A3AR receptor expression. An alternative hypothesis is that the A3AR plays an important role in the development, migration, or overall survival of mast cells and that loss of this receptor renders mice mast-cell deficient. Two lines of experimental data support the former hypothesis. First, mast cells were found in approximately normal numbers in all tissues examined and could not be distinguished from wild-type controls based on morphological criteria. Perhaps more convincing data supporting the normal development of mast cells in A3AR–/– mice are the demonstration that loss of this receptor has only a minimal impact on systemic and cutaneous IgE-mediated anaphylaxis in these animals. A small decrease in response to cutaneous anaphylaxis was observed, but this difference did not reach statistical significance. It is possible that this trend may be related to the in vitro demonstration that adenosine, acting through the A3 receptor, potentiates IgE-mediated mast-cell degranulation. The role of adenosine in this response may be difficult to measure when mast cells are maximally stimulated in passive anaphylaxis, and further studies with suboptimal doses of antigen or antibody may reveal a greater role for adenosine in these mast cell–mediated inflammatory responses. Moreover, further insight into the contribution of adenosine to disease processes such as asthma, in which mast-cell activation is believed to play a role, may be gained by examining A3AR-deficient mice in established models of this disease.

Previous investigations using murine BMMCs have shown adenosine’s potentiation of degranulation in these cells to be partially pertussis-toxin sensitive (39). These same investigators also showed that adenosine-mediated rises in inositol trisphosphate (IP3) and intracellular calcium were pertussis-toxin insensitive (39). Because A3-mediated signal transduction has been shown to be sensitive to pertussis toxin (40, 41), it has been speculated that another adenosine receptor, such as the A2B receptor, plays a role in adenosine’s actions on BMMC degranulation. Our results showing the potentiation of antigen-stimulated BMMC degranulation by adenosine through the A3 receptor do not support these observations. A3 receptors have been shown to interact with multiple G proteins including Gi and Gq, and it has been suggested that A3AR stimulation of phospholipase C may have a pertussin toxin–insensitive component in some systems (42).

While adenosine can potentiate antigen-induced mast-cell degranulation in vitro, it cannot initiate degranulation independent of an additional stimulus. In contrast, adenosine alone appears to be sufficient to activate mast cells in vivo. Studies carried out in several species support this observation (23, 24, 29–31). Several different hypotheses can explain the profound effect of adenosine on mast-cell function in vivo. First, BMMCs in tissue culture may be immature and lack the necessary signaling mechanisms required to initiate degranulation in response to adenosine. Second, the ability of adenosine to initiate degranulation may vary between different mast-cell types. Tissue mast cells can be classified into either connective tissue or mucosal mast cells based on certain morphological and histochemical characteristics, and it has been suggested that BMMCs more closely resemble the latter (21). The in vivo studies described here examine the cutaneous response to adenosine and, therefore, reflect activation of A3AR on connective tissue mast cells. Finally, in vivo, low levels of antigen may be bound to IgE receptors occupied by circulating IgE, providing the additional signaling necessary for activation of mast cells by adenosine. The availability of mouse lines lacking a functional FcεR1 receptor has enabled us to test this hypothesis directly. Intradermal administration of adenosine to these mice results in a similar degree of plasma protein extravasation as observed in wild-type controls, establishing that adenosine-induced mast-cell degranulation in vivo occurs independently of the presence of signal transduction by the high affinity IgE receptor.

Although it is well accepted that adenosine is a paracrine and autocrine mediator in a broad spectrum of physiological responses, less information is available concerning the functions of inosine, a primary metabolite of adenosine. Two different routes for metabolism of adenosine have been described. First, adenosine can be used as a substrate for nucleotide synthesis producing ADP and ATP, which themselves have potent receptor-mediated biologic reactions (43). Alternatively, adenosine can be converted by adenosine deaminase to inosine. The activity of these 2 pathways is believed to be regulated, at least in part, by the substrate availability (44). When levels of adenosine are low, most adenosine is converted to AMP by adenosine kinase. However, when adenosine levels increase as a result of trauma, shock, exercise, hypoxia, or endotoxin, adenosine deamination predominates, leading to significant increases in inosine production and resultant interstitial levels of inosine that can rise to greater than 1 mM (44, 45). The biologic significance of these high tissue levels of inosine has not been established, nor is it known whether this metabolite mediates it actions solely by binding to adenosine receptors or whether it mediates its effects through yet undescribed inosine receptors. In vitro studies have shown that inosine can potentiate antigen-induced degranulation of both rat serosal mast cells and rat RBL-2H3 mastlike cells (5, 46). We show here that inosine can also potentiate the degranulation of BMMCs. Furthermore, this response is not seen in A3AR-deficient mast cells, demonstrating that inosine’s actions on BMMCs are mediated through the A3 receptor. These findings are consistent with recent pharmacologic studies that showed inosine to preferentially bind to recombinant rat A3 receptors (5). These studies also suggest that adenosine is more effective than inosine in mediating this response and are consistent with earlier studies with rat mast cells that showed inosine to be some 10 times less potent that adenosine at enhancing antigen-induced degranulation (46).

Intradermal injection of inosine was at least as effective as adenosine at eliciting an edema response. This finding is consistent with pharmacologic studies that suggested that inosine can activate the A3 receptor (5). Thus, the studies reported here support a physiological role for inosine in the acute inflammatory response and show that these effects of inosine are mediated solely through an adenosine receptor, specifically the A3 receptor. This does not rule out the possibility that other biologic responses of inosine may be mediated by other adenosine receptors or yet uncharacterized receptors.

In summary, we have shown that both adenosine and its principal metabolite inosine promote plasma protein extravasation through activation of A3 adenosine receptors. Lack of any changes in vascular permeability in mast cell–deficient mice after exposure to adenosine suggests that adenosine acts indirectly through A3 receptors on mast cells to produce these physiological changes. These actions of adenosine in vivo occur independently of the presence of the expression of the high-affinity IgE receptor, suggesting a more profound role for adenosine as a modifier of the inflammatory response.