Regulation of Atherosclerosis and Associated Risk Factors by Adenosine and Adenosine Receptors
- First Online:
- Cite this article as:
- Koupenova, M., Johnston-Cox, H. & Ravid, K. Curr Atheroscler Rep (2012) 14: 460. doi:10.1007/s11883-012-0263-y
- 293 Views
Adenosine is an endogenous metabolite that has an anti-inflammatory effect across the vasculature. Extracellular adenosine activates 4 G-protein coupled receptors (A1, A3, A2A, and A2B) whose expression varies in different cells and tissues, including the vasculature and blood cells. Higher levels of adenosine are generated during stress, inflammation, and upon tissue damage. Some of the adenosine receptors (AR), such as the A2BAR, are further up-regulated following such stresses. This review discusses the role of adenosine and adenosine receptors in the development of atherosclerosis and some of the risk factors associated with this pathology. These include adenosine receptor-regulated changes in atherosclerosis, blood pressure, thrombosis, and myocardial infarction. Potential therapeutic applications are reviewed, as well as reasons for phenotypic differences occasionally observed between receptor knockout and pharmacological inhibition via drug administration.
KeywordsAdenosineAdenosine receptorsFoam cellsAtherosclerosisMyocardial infarctionIschemic preconditioning
The beginning of adenosine research dates back to 1929 when scientists from the University of Cambridge injected intravenously tissue extract from heart, which led to an immediate decrease in heart rate, impaired heart conduction, and arrested experimentally induced arrhythmia of the auricle . The scientists were Drury and Szent-Gyorgi, and the metabolic substance causing these effects on the vasculature and blood flow was adenosine. Adenosine can either be transported out of the cell or generated as a result of catabolic break down of adenine nucleotides during stress [2•, 3], inflammation , and tissue damage [3, 5–9]. Released adenosine affects the behavior of the surrounding cells by activating 4 G-protein coupled receptors. Two of the adenosine receptors (AR), A2A and A2B, couple with Gαs, activate adenylate cyclase, and increase intracellular 3’-5’-cyclic adenosine monophosphate (cAMP) levels. Whereas, A1 and A3, couple with Gαi, inhibit adenylate cyclase, and decrease cAMP levels. All of the ARs are activated by nanomolar concentration of adenosine, with the exception of the A2BAR, which in many cases requires micromolar concentrations , or possible oligomerization with the A2AAR . Expression of these receptors varies in different cells throughout the vasculature and in the blood.
Atherosclerosis is a disease of the large and medium arteries and its ultimate outcome is viewed as the number 1 killer in United States. Atherosclerosis is a result of a complex network of processes typically involving a high lipid profile, and hemodynamic factors in the blood circulation that interact with the innate immune system in the arterial wall [12–14]. Diets rich in cholesterol and saturated fats, lack of exercise, smoking, and high blood pressure all contribute to the development of atherosclerosis and to its impact on the vasculature.
This review describes the role of adenosine and its receptors in atherosclerosis, and in the development of other cardiovascular events that contribute to atherosclerosis or are amplified by it.
Foam Cells and Adenosine Receptors
Diets rich in cholesterol and saturated fatty acid intake combined with genetic factors lead to an increase in circulating cholesterol and triglycerides, and a consequent activation of the vascular endothelial layer . Endothelial cell activation is characterized by increased expression of cell surface adhesion molecules, such as vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule (ICAM1), P-selectin, and E-selectin. Up-regulation of adhesion molecules results in recruitment and attachment of mononuclear leukocytes, some of which are monocytes and neutrophils . The attachment and locally produced chemoattractant molecules, such as monocyte chemotactic protein (MCP-1) , stimulate monocyte migration into the sub-endothelial space, where they differentiate to macrophages . The differentiation process, which is governed by macrophage colony stimulating factor 1 (MSF-1), subsequently results in up-regulation of macrophage scavenger receptors and to the internalization of lipoprotein particles. The result is accumulation of large droplets of lipids and cholesterol esters, leading to the foamy appearance of macrophage cells and the formation of foam cells . Large accumulation of foam cells in the intima leads to fatty streaks along the artery. Fatty streaks that continue to be exposed to high levels of lipids and inflammation give rise to atherosclerosis.
As foam cell enlargement and accumulation progresses, hypoxia and inflammation increases and adenosine levels rise . Adenosine receptors are expressed on the surface of the endothelial cells, on vascular smooth muscle cells  as well as on the surface of different type of leukocytes including macrophages . Released adenosine, however, can render its anti-inflammatory effect on the vasculature by activating the A2 adenosine receptors (A2A, A2B). Endothelial cells lacking the A2BAR have elevated levels of ICAM-1, P-selectin, and E-selectin . Signals coming from either bone marrow- A2BAR or A2AAR lower the levels of vascular adhesion molecules and reduce inflammation [5, 20]. There is no clearly defined role for adenylate cyclase-inhibitory adenosine receptors in the regulation of adhesion molecule expression in vascular endothelial cells, although renal cortices, post-renal ischemic injury, show an increase in the message for ICAM-1, and TNF-alpha in the A1AR knockout (KO) mice . Thus, it seems that adenosine receptors affect the initiation of plaque formation by reducing inflammation and adhesion molecule expression. These observations suggest a protective role of adenosine signaling in the endothelial layer primarily through the A2-type adenosine receptors, and possible delay in early atherosclerotic events.
The monocyte/macrophage lineage represents a crucial player in the development of atherosclerosis. All 4 of the adenosine receptors are expressed on the macrophage surface , and macrophages are major contributors to overall systemic inflammation. Activation of the A2AAR in this lineage leads to elevation of the anti-inflammatory cytokine IL-10 and to reduced levels of pro-inflammatory cytokines such as TNF-alpha . In accordance, macrophages depleted of the A2BAR secrete much higher levels of TNF-alpha and IL-6 . In vitro studies using murine  and human macrophage  cell lines (J774.1 and U937 respectively) demonstrate that A3AR activation inhibits LPS-stimulated TNF-alpha production. This result is confirmed in vivo , suggesting that the adenylyl cyclase-inhibitory receptors can also have an anti-inflammatory function.
As a result of its anti-inflammatory properties, adenosine provides protection against foam cell formation and development of atherosclerosis. In activated human THP-1 monocytic cell line and in murine peritoneal macrophages, activation of the A2AAR by specific ligand inhibits foam cell formation, as antagonism or lack of this receptor has no effect [26, 27]. Mechanistically, activation of the A2AAR reduces foam cell formation by increasing reverse cholesterol efflux through elevated levels of the lipoprotein transporter ABCA1 [28•]. Reiss et al observed the effects on foam cell formation in peritoneal macrophages expressing or lacking the A3AR , as the KO cells could not reduce foam cell formation upon stimulation or inhibition with A3AR specific ligands . Interestingly, in the human macrophage cell line U937, pharmacological inhibition of the A3AR (with MRE 3008 F20) and antagonism of the A2BAR (with MRE 2029 F20) inhibited foam cell formation , suggesting further attention to the interpretation of cell lines studies in this field of research.
Advanced Vascular Lesions and Adenosine Receptors
Advanced vascular lesions begin with accumulation of vast amount of foam cells that start to necrotize and initiate T-cell infiltration to the intima. T-cells, together with neutrophils, contribute to the signaling mechanism that drives the vascular smooth muscle cells (VSMC) to move into the intima. There, VSMC de-differentiate, proliferate, and start secreting extracellular matrix proteins. The result of this process is a formation of a fibrous cap around the necrotic foam cell mass (reviewed in ). No one has yet determined whether adenosine levels in or around the vascular lesion are elevated. On the other hand, inflammation rises as atherosclerosis progresses, and inflammatory cytokines increase expression of adenosine receptors [4, 19, 29]. Indeed, studies done in mouse models lacking adenosine receptors show that the role of adenosine in atherosclerosis, whether protective or deleterious, depends on the ablated receptor.
The first glimpse of the role of adenosine in atherosclerosis started with the elimination of the A3AR on an Apolipoprotein E (ApoE) null background. ApoE KO mice are a useful model to study atherosclerosis, since mice do not naturally develop all stages of atherosclerotic lesions along the arterial tree [30–33]. In the case of the A3AR, although aortic vascular smooth muscle cells exhibit decreased proliferative potential upon receptor-elimination and A3AR contributes to the process of inflammation, there is no significant effect on the chronic development of atherosclerosis .
As mentioned earlier, A2AAR KO macrophages have increased foam cell formation potential and A2AAR is anti-inflammatory. It is surprising, however, that A2AAR, ApoE double KO mice on high fat diet have smaller atherosclerotic lesions, compared with the control mice used in the study. The major contributors to this anti-atherosclerotic profile were the macrophages; in vivo, these cells were highly apoptotic and vascular lesions were smaller [35•]. Overall, elimination of the A2AAR results in increased levels of plasma cholesterol, concentrated predominantly in the LDL particles [35•]. These findings are quite intriguing in the context of the observed atherosclerotic profile, considering that increased inflammatory state and elevated plasma cholesterol typically lead to atherosclerosis. Further, in a model of restenosis post arterial angioplasty, elimination of the A2AAR results in increased neointima, raising the possibility that A2AAR is actually protective against vascular lesion formation [36•]. In this case, there was an increase in the homing ability of leukocytes to the sub-endothelial space in the A2AAR KO, which led to a consequent increase in neointima thickness [36•]. It is quite possible that under chronic inflammation (atherosclerosis) A2AAR is harmful, but under more acute conditions (post-angioplasty) it becomes protective. This defines the A2AAR signaling as more complex than expected, and this receptor might be more challenging as a therapeutic target in the context of vascular disease.
The role of the A2BAR in vascular diseases has also been extensively studied. As described earlier, endothelial A2BARs regulate the expression of surface adhesion molecules and improve leukocyte rolling . In a similar fashion to the A2AAR, in a model of restenosis post angioplasty, the A2BAR protects against increased neointima formation. This protective effect is due to A2BAR-mediated signals originating in bone marrow cells . Vascular smooth muscle cells lacking expression of this receptor have an increased proliferation rate , and activated macrophages show elevation in levels of matrix metalloproteinase-9 (MMP-9) . Vascular-derived matrix metalloproteinases, such as MMP-9 play an important role in neointima formation after vascular injury and in de-stabilizing advanced atherosclerotic plaques .
In summary, with respect to atherosclerosis, A2BAR is protective against atherosclerosis and its administration of its ligand by injection reduces vascular lesion formation. In contrast, A2AAR elimination is protective and A3AR has no effect (Fig. 1). The role of A1AR in this regard is not known. As noted in the concluding remarks, considering that some of these receptors affect inflammation, future studies could explore the influence of the microbiome in different settings and strains on receptor knockout outcomes.
Adenosine and Thrombosis
The rupture of an advanced atherosclerotic lesion can lead to an impairment of vessel blood flow, giving rise to myocardial infarction . As a result, there is exposure of matrix proteins, such as collagen, which can result in platelet accumulation and cross-linking, creating a thrombus that can give rise to acute coronary events. Platelet activation can contribute to atherosclerosis and other vascular impact, while excessive thrombosis can increase the risk of a pathological outcome in this vascular disease. In the pathogenesis of a myocardial infarction, platelets are a major player and adenosine and adenosine receptors are known to affect platelet function. Ex vivo studies using platelets isolated from A2AAR KO mice have shown that these cells exhibit mildly higher aggregation  and platelet-platelet interaction, which is important for the formation of tertiary structure of the clot. The mechanism involved is cAMP-dependent. Consequently, ligand-specific activation of the A2AAR on human platelets results in an increase in cAMP levels and consequent decrease in platelet aggregation . More recent studies have shown that the other cAMP-elevating receptor, the A2BAR, has a modest effect on aggregation as well. In this case, platelets isolated from the A2BAR KO mice have decreased levels of cAMP and elevated expression of the ADP receptor P2Y1, which is a potent activator of aggregation . Platelets do not express the A1AR or A3AR . Thus, in case of thrombosis, adenosine protects against excessive thrombus formation through the A2-type adenosine receptors.
Adenosine and Myocardial Infarction
Myocardial infarction occurs when the blood flow in the heart is impaired due to obstruction caused by rupture of atherosclerotic plaque, and or thrombi development around that plaque. This obstruction often leads to ischemia and consequently to impaired oxygen supply, which, if untreated, can lead to cell death and damage to the heart. Restoration of blood flow and oxygen supply or reperfusion post myocardial infarction causes significant injury to the tissue and elevated inflammatory response . The inflammatory signals in the course of reperfusion originate from cardiomyocytes and endothelial cells, and these cells express A1AR and A2-type ARs, respectively. Ischemia leads to decreased oxygen supply that causes hypoxia and affects endothelial function by augmented cellular permeability and induction of pro-coagulant properties . There is also a decrease in cAMP levels as a result of reduction of endothelial adenylate cyclase activity . Adenosine, in turn, regulates levels of cAMP in vascular endothelial cells through its receptors.
Studies examining the role of selective A2AAR agonist (ATL-146e) on myocardial infarction, using a canine model, show that infarct size is significantly reduced upon activation of this receptor. In the infracted myocardium of ATL-146e-treated dogs there is reduced accumulation of neutrophils . Neutrophils are the major contributor to cardiac damage in this process by causing damage or apoptosis of the tissue and impaired endothelial function . This suggests that adenosine has a beneficial effect on reduction of myocardial infarction, and this effect is mediated by the A2AAR.
Adenosine signaling has also a beneficial effect on ischemic preconditioning, a process that renders myocardium resistant to the detrimental effects of ischemia . In this setting, activation of the A1AR results in reduction of infarct size, as inhibition of A1AR with specific antagonist (8-cyclopentyl-1,3-dipropylxanthine) blocks the protective effect of preconditioning. The mechanism by which A1AR exhibits its protective effect during preconditioning is by opening the myocardial KATP channels . Selective activation of A3AR by Cl-IB-MECA (2-chloro-N(6)-(3-iodobenzyl)adenosine-5'-N-methylcarboxamide) also protects against myocardial ischemia/reperfusion injury . Recently, Dr. Eltzschig’s group has shown that at least in mice, A2BAR is the major receptor that acts as protective in ischemic setting . During ischemic preconditioning, the A2BAR is selectively up-regulated, and from all of the AR KOs, A2BAR KO mice are the only ones in which this process is not protective . Injection of A2BAR specific agonist (BAY 60-6583) significantly reduces infarct size after ischemia . These studies suggest that adenosine, signaling mostly through the cAMP-elevating A2AAR or A2BAR plays an important protective role during ischemic preconditioning and can decrease myocardial infarction.
Adenosine and Blood Pressure Homeostasis
Atherosclerotic plaque becomes destabilized as a function of weakened fibrous cap. This destabilization is a complex balance among lipid content in the plaque, distribution of vascular smooth muscle cells, shear stress coming from the blood flow, and weakened endothelium [51, 52]. High blood pressure as a result of narrowing of the arteries due to the plaque formation has been associated with high risk of vulnerable plaque rupture and consequent predictor of myocardial events.
Adenosine, in turn, is beneficial against blood pressure elevation, and it can decrease heart rate (reviewed in ). Studies in rats have shown that adenosine exhibits its effect on heart rate and homeostasis of blood pressure (baroreflex) by acting on the brain Nucleus Tractus Solitarii (NTS) . In this part of the brain, activation of extracellular signal-regulated kinases 1, 2, and the endothelial nitric oxide synthase (eNOS) are the downstream targets of adenosine, possibly through the A2AAR . In addition, adenosine signaling through the A2AAR , and the A2BAR , provides a potent vasodilatory effect on mean arterial pressure. In the vessel, endothelial A2AAR leads to an increase in nitric oxide (NO) production as a result of activation of the eNOS .
Adenosine receptors knockout (KO) mice and their contribution to atherosclerosis and related vascular outcomes
Lipid/foam cell effects
Effect on platelet function
A1AR, ApoE dKO
↓foam cell formation 
No effect (tail cuff) 
No effect 
No effect 
A3AR, ApoE dKO
No effect 
↑BP; ↑HR 
No effect 
↑Foam cell formation 
A2AAR, ApoE dKO
↓Aortic plaque formation 
No effect on BP 
No effect [2•]
↑Aggregation  (modest effect)
A2BAR, ApoE dKO
No effect on BP or HR [2•]
↑Aortic plaque formation [2•]
The other major adenosine receptor involved in blood pressure homeostasis is the A1AR [56, 62]. Activation of cardiac A1AR reduces cAMP levels, hyperpolarizes the KATP channels and increases potassium levels. The result of this signaling cascade leads to transient block of the Atrioventricular node (AV) and a consequent reduction in blood pressure . A1AR null mice have elevated blood pressure at baseline on low sodium diets . Elimination of the A1AR also results in an increase in heart rate . These studies have been done using the A1AR KO model (created by Fredholm’s group ) and using implant transmitter in the carotid artery on either mixed background (C57BL,129/OlaHsd) , or in the same mice outbred to C57BL/6J . Blood pressure measured by tail cuff method in an A1AR KO created by Schnermann group (mice on C57BL/6J) showed no significant difference between A1AR KOs compared with wild type controls . It is possible that tail cuff method is not sensitive enough to account for transient changes, or the background of the mouse strain, and disguises the effect that adenosine might have on blood pressure.
The role of the A3AR or the A2BAR on blood pressure homeostasis, if any, is not comparable to the effect of the A1AR and the A2AAR. Elimination of either of these former receptors has no effect on blood pressure measured by tail cuff method in young, twelve-week old mice [5, 34]. Studies with older mice have not been carried out. Although there is no significant effect on blood pressure, A2BAR KO mice (on C57BL/6J background) have significantly reduced heart rate compared with wild type after 7 weeks on a high fat diet [2•]. This influence of adenosine on heart rate regulated by the A2BAR could be a consequence of the action of this receptor in the brain, which is yet to be explored. Further, it is possible that signaling of adenosine through the A2BAR has a modest effect on blood pressure, which is not detected by the less sensitive tail cuff method.
In summary, adenosine acting simultaneously in the central nervous system and in the periphery can maintain favorable blood pressure levels and thus provide a beneficial effect against plaque rupture.
Conclusion: Adenosine Receptors as Therapeutic Targets for Cardiovascular Disease
There is no doubt that adenosine has a crucial role in the development of atherosclerosis, myocardial infarction, and blood pressure homeostasis. All 4 receptors can potentially be beneficial targets for different aspects during the pathogenesis of cardiovascular disease. Agonists of A1AR and A3AR could be targeted for regulation of ischemia and preconditioning. However, the first drug designed to specifically inhibit A1AR, Rolofylline (KW-3902; Merck and Co, Whitehouse Station, NJ) for the treatment of compensated heart failure did not result in a favorable outcome. There were increased neurological outcomes in the treated group vs placebo, as well as increased incidences of stroke . Adenosine receptor polymorphism was partially responsible for the undesired response to this drug, as infarct size was associated with 3 variants in the 3’-untranslated region of A1 and 1 in the coding region of A3AR .
Based on mouse studies, one might envision development of antagonists for the A2AAR to halt the development of atherosclerosis, but not vascular restenosis. On the other hand, A2AAR being a potent vasodilator could be targeted for lowering blood pressure using an agonist (rather than antagonist). Indeed, an A2AAR specific agonist has been developed to cause vasodilatation. The agonist, also known as Regadenson (Astellas Pharma Inc, Deerfield, IL) causes an increase in blood flow to different tissues, enabling myocardial perfusion imaging . A2AAR agonists can also be used to inhibit platelet activation and thrombosis.
If mouse studies translate to patients, targeting of the A2BAR could be beneficial, because of its uniform protective effect on atherosclerosis, restenosis, and thrombosis. In vivo injection into mice of a specific A2BAR agonist reduces atherosclerosis without change of western diet intake. That is, the pathology can be significantly reversed without modifying life style. Furthermore, cholesterol and triglyceride levels decrease upon receptor activation, as synthesis of liver lipids drops down. Also, A2BAR activation is a potent reducer of ischemia during ischemic preconditioning. Thus, future development of A2BAR agonists could constitute initial steps towards examining the therapeutic potential of this receptor in humans with respect to atherosclerosis and cardiovascular disease.
Some final points to consider in studies involving any of these receptors, and in contemplating therapeutic applications: (1) Adenosine receptors regulate inflammation, and as such, future studies might need to consider the impact of the microbiome on phenotypic outcomes monitored in knockout and wild type mice housed in different laboratories, and laboratory conditions (further highlighting the importance of comparing littermates and accounting for strain-background); (2) the mode of administration of a ligand might also influence the result, with digested drugs affecting the gut, while injected ones potentially yielding a somewhat different result; (3) finally, a developmental knockout might lead to a different phenotype than pharmacological inhibition by an antagonist, considering adaptation in the case of a total body knockout, and potentially variable impacts of a drug on different organs (eg, depending on accessibility) and, hence, on the whole body phenotype.
The authors apologize to all whose work they could not cite due to limited space required by the journal. This work was supported by National Heart, Lung, and Blood Institute grant HL93149, and by the Boston Nutrition Obesity Research Center (DK046200) Pilot Grant to Katya Ravid; Milka Koupenova was supported by postdoctoral Training Grant from NIH (HL007224); Hillary A. Johnston-Cox was supported by pre-doctoral Training Grant from NIH (HL007969).
No potential conflicts of interest relevant to this article were reported.