Adenosine and the Cardiovascular System

Abstract

Adenosine is an endogenous nucleoside with a short half-life that regulates many physiological functions involving the heart and cardiovascular system. Among the cardioprotective properties of adenosine are its ability to improve cholesterol homeostasis, impact platelet aggregation and inhibit the inflammatory response. Through modulation of forward and reverse cholesterol transport pathways, adenosine can improve cholesterol balance and thereby protect macrophages from lipid overload and foam cell transformation. The function of adenosine is controlled through four G-protein coupled receptors: A₁, A_2A, A_2B and A₃. Of these four, it is the A_2A receptor that is in large part responsible for anti-inflammatory effects of adenosine as well as defense against excess cholesterol accumulation. A_2A receptor agonists are the focus of efforts by the pharmaceutical industry to develop new cardiovascular therapies, and pharmacological actions of the atheroprotective and anti-inflammatory drug methotrexate are mediated via release of adenosine and activation of the A_2A receptor. Also relevant are antiplatelet agents that decrease platelet activation and adhesion and reduce thrombotic occlusion of atherosclerotic arteries by antagonizing adenosine diphosphate-mediated effects on the P2Y12 receptor. The purpose of this review is to discuss the effects of adenosine on cell types found in the arterial wall that are involved in atherosclerosis, to describe use of adenosine and its receptor ligands to limit excess cholesterol accumulation and to explore clinically applied anti-platelet effects. Its impact on electrophysiology and use as a clinical treatment for myocardial preservation during infarct will also be covered. Results of cell culture studies, animal experiments and human clinical trials are presented. Finally, we highlight future directions of research in the application of adenosine as an approach to improving outcomes in persons with cardiovascular disease.

1. INTRODUCTION

The endogenous, ubiquitous purine-nucleoside adenosine exerts multiple biochemical effects that serve important roles in cardiac and vascular biology (1–3). Adenosine is known to regulate myocardial and coronary circulatory functions and exerts potent vasodilatory effects in most vascular beds of mammalian species (4, 5). Adenosine acts by at least four major types of G protein-coupled cell surface receptors, A₁, A_2A, A_2B and A₃ (6, 7) which are encoded by distinct genes and are differentiated based on their affinities for adenosine agonists and antagonists (8). All four receptors are N-linked glycoproteins. Adenosine receptors are ubiquitous and are activated by different ranges of endogenous adenosine concentrations (8). A₁ and A₃ receptors are negatively coupled to adenylyl cyclase via interaction with pertussis toxin-sensitive G proteins of the Gi and Go family, A₂ subtypes are cyclic AMP-elevating, Gs protein-coupled receptors positively coupled to adenylyl cyclase (9, 10).

The widespread actions of adenosine include effects on multiple organs and systems including the heart (11), nervous system (12–14), lungs (15), gastrointestinal system (16), kidneys (17–19) and reproductive organs (20, 21), as well as on blood cells (22), adipocytes (23, 24), and the immune system (25, 26). This review examines the role of adenosine in cardiovascular processes, both pathological and physiological. There is a focus on how they change lipid transport and platelet aggregation because these are two major factors in development and progression of atherosclerosis that are the targets for many current therapies (27–30).

2. SYNTHESIS AND METABOLISM

Adenosine is released in tissues at times of cellular stress such as hypoxia, ischemia and inflammation. With ischemic insult, when metabolic demands exceed oxygen supply, endogenous levels of adenosine increase rapidly (31). Cell hypoxia is a potent stimulus for adenosine release. Adenosine is formed via dephosphorylation of ATP both inside and outside the cell (32) (Figure 1). It can be formed intracellularly from ATP, ADP or AMP by activity of cytoplasmic 5’-nucleotidases or extracellularly from ATP or ADP by the sequential action of ecto-nucleoside triphosphate diphosphohydrolase (ecto-NTPDase-1 [CD39])—or possibly other NTDPases— that form AMP and ecto-5′-nucleotidase (CD73) which converts AMP to adenosine. Adenosine can also be generated from S-adenosylhomocysteine (SAH) via SAH hydrolase (33). A biochemical mechanism responsible for significant adenosine production from cyclic adenosine 3',5'-monophosphate (cAMP) is referred to as the cyclic AMP-adenosine pathway (34). This pathway involves the conversion of cyclic AMP to AMP by the enzymes phosphodiesterase (PDE, or exonuclease) followed by dephosphorylation of 5'-AMP by intra- and extracellular 5'-nucleotidases. Adenosine is able to travel across cell membranes to maintain equilibrium between intracellular and extracellular adenosine concentrations. Extracellular adenosine is rapidly taken into cells via both sodium-dependent and sodium-independent transporters for subsequent metabolism. Very rapid uptake of adenosine takes place via endothelial cells, erythrocytes, and adjacent tissues, where adenosine can move across the plasma membrane space and be utilized within the cell. Once adenosine is taken up by endothelium, it is phosphorylated by adenosine kinases to form AMP or degraded by adenosine deaminase to inosine (35). The physiological concentration of adenosine in human plasma is 0.1–1 μM (36). The half-life of adenosine in human plasma is very short, ranging from 0.6 to 1.5 seconds (37). The short half-life is attributed to the rapid elimination of adenosine from the extracellular space which depends to a large extent on nucleoside transporters. Adenosine moves by means of these nucleoside transporters because it is hydrophilic and not permeable to cell membranes. The receptor-mediated effects of adenosine are terminated by uptake into cells by these transporters, categorized as equilibrative nucleoside transporters (ENT) and concentrative nucleoside transporters (CNT) (38, 39). ENT act via facilitative diffusion, are bidirectional, sodium-independent, have a large selectivity and exist in multiple cell types, while CNT are nucleoside-sodium symporters that are seen in only certain types of cells and have a more limited specificity. The human ENT1 is an important transporter in the vasculature. ENT1 and ENT2 are found predominantly in the plasma membrane. Certain CNT can interact with adenosine, and human CNT1 has a high affinity for adenosine, which acts as a competitive inhibitor for other permeants. This solidifies the phenomena that adenosine is involved in different interactions with various selective transporters, and inhibitors of nucleoside transporter would block uptake and metabolism of extracellular adenosine, augmenting the receptor-mediated physiological effects of adenosine. (40, 41).

Figure 1. Intracellular and Extracellular Biogenesis of Adenosine.

Adenosine is formed via dephosphorylation of ATP both inside and outside the cell. It can be formed intracellularly from ATP via adenylyl cyclase-mediated conversion of ATP to cAMP, which is then converted to AMP by phosphodiesterases. Subsequent activity of cytoplasmic 5’-nucleotidases convert AMP into adenosine, which can in turn be converted back to AMP through adenosine deaminase and adenosine kinase activity, with an inosine intermediate. A second intracellular pathway exists in which adenosine can be generated from S-adenosylhomocysteine (SAH) via SAH hydrolase. Extracellularly, ATP is converted to adenosine through the sequential action of ecto-nucleoside triphosphate diphosphohydrolase (ecto-NTPDase-1 [CD39]) that forms AMP, and ecto-5′nucleotidase (CD73) which converts AMP to adenosine

3. ADENOSINE, ELECTROPHYSIOLOGIC EFFECTS

Adenosine has potent cardiac electrophysiologic effects on the heart (11) (Figure 2). A₁, A_2A, A_2B, and A₃ adenosine receptors have been identified in the heart (42). The A₁ receptor on cardiomyocytes mediates inotropic inhibitory actions of adenosine on contractility (43). Adenosine acts at receptors in the sinus node and atrioventricular (AV) node. Stimulation of specific cell-surface A₁ receptors shortens the duration, depresses the amplitude, and reduces the rate of rise of the action potential of AV nodal cells, slowing impulse conduction through the AV node (44, 45). Adenosine decreases spontaneous depolarization in the sinus node and conduction velocity in the AV node. The effect on both the sinus and AV nodes is dose-dependent and of very short duration. The rapid negative dromotropic action on conduction speed in the AV node is the basis for the clinical administration of adenosine by intravenous bolus injection as an antiarrhythmic agent for the acute management of paroxysmal supraventricular tachycardia mediated by a reentrant mechanism involving the AV node (46, 47). Adenosine interrupts conduction through the AV node and abruptly terminates the reentry wave as it approaches nodal tissue (48).

Figure 2. Electrophysiological Effects of Adenosine on the Heart.

Adenosine acts at receptors in the atrium, sinus node, and atrioventricular (AV) node. A) The A1 adenosine receptor on cardiomyocytes mediates inotropic inhibitory actions of adenosine on contractility that are opposed by A2 adenosine receptor activation. Stimulation of specific cell-surface A1 receptors shortens the duration, depresses the amplitude, and reduces the rate of rise of the action potential of AV nodal cells, slowing impulse conduction through the AV node. Adenosine decreases spontaneous depolarization in the sinus node and conduction velocity in the AV node.

The drug can also be used diagnostically to distinguish a supraventricular tachycardia with aberrant conduction - which may terminate with adenosine, from a ventricular tachycardia - which will not (49–51). In a study by Flyer et al, adenosine was used as a therapy to treat supraventricular tachycardia to alter the AV node action potential. In the 80 heart transplant patients treated, none experienced asystole after adenosine administration, with AV block induced in 96% of patients at a dose of 200µg/kg, indicating that the treatment was both safe and effective (52).

Adenosine is contra-indicated in Wolff-Parkinson-White syndrome with atrial fibrillation because degeneration to ventricular fibrillation can result (53). During intravenous infusion into conscious humans, adenosine has a number of known side effects such as hypotension and bronchospasm, facial flushing and headache. These occur very briefly due to the short half-life of the medication and resolve upon termination of infusion (54, 55).

4. ADENOSINE ACTIONS AT THE CELLULAR LEVEL - ARTERIAL ENDOTHELIUM AND SMOOTH MUSCLE

Adenosine is able to regulate vascular tone in the arterial tree by relaxing arterial smooth muscle (56, 57). This relaxation decreases vascular resistance thereby facilitating blood flow and oxygen delivery (58). The effects of adenosine on coronary blood flow are thought to be mediated primarily by activation of A_2A receptors (59). Adenosine activates A_2A receptors, triggering the opening of K_v and K_ATP channels on smooth muscle cells. This leads to membrane hyperpolarization, and relaxation (60, 61). Other adenosine receptors and mechanisms may also contribute, particularly in pathologic conditions such as diabetes and cardiovascular disease (62). It has also been suggested that adenosine acts on endothelium to cause nitric oxide release which, in turn, dilates coronary arteries (63).

Although an exercise stress test is the preferred stimulus to detect ischemic heart disease, pharmacologic vasodilators are the next best option for those who cannot exercise adequately (64). Adenosine, dipyridamole, or regadenoson (an A_2A receptor agonist) can be used in myocardial perfusion imaging studies for pharmacologic stress testing (65). Dipyridamole increases levels of intrinsic adenosine because it inhibits adenosine deaminase, the enzyme that breaks down adenosine. During myocardial perfusion imaging, adenosine increases coronary blood flow, partly through cAMP production, and with increased blood flow comes enhanced radionuclide uptake in myocardium (66). In myocardium with an impaired coronary flow reserve, which can occur as a result of coronary artery narrowing, adenosine-mediated increases in blood flow and radionuclide uptake are blunted compared to the normal physiologic response (67).

Vascular tone is modulated by activation of adenosine receptors primarily on endothelial cells and, to a lesser extent, on vascular smooth muscle (11). Endothelial cells that line the luminal surface of blood vessels function as a complex metabolically active organ system involved in regulation of blood flow, exchange of nutrients, passage of waste products and control of thrombosis/thrombolysis (68). The endothelial monolayer resting upon a basement membrane comprises the intima. The endothelium is actively involved in maintaining vascular homeostasis and adenosine A_2A receptors expressed on these cells participate in the process by inducing vasodilation and vascular relaxation. In both human and porcine arterial endothelial cells, adenosine A_2A receptors increase while adenosine A₁ receptors decrease production of the vasodilator nitric oxide (NO). NO is produced by endothelial cells lining the vasculature in a reaction catalyzed by NO synthase via a five electron oxidation of the guanidine nitrogen terminal of the amino acid L-arginine. Stimulation of A_2A receptors triggers the activation of endothelial nitric oxide synthase (eNOS) leading to increased synthesis of NO in human endothelium (69). In rat aortic endothelium, A_2A-mediated release of NO requires extracellular Ca²⁺ and Ca^2+-activated K⁺ channels (70). NO released by endothelium can diffuse to adjacent smooth muscle cells where it binds to and activates the enzyme soluble guanylate cyclase by removing the histidine residue on its axial position. Soluble guanylate cyclase catalyzes the conversion of GTP to cyclic GMP which relaxes smooth muscle (71).

Further rodent studies confirm the importance of the A_2A receptor in vasorelaxation. Ponnoth et al compared aortic vasorelaxation in wild type mice versus A_2A knockout mice and found reduced relaxation in response to adenosine analogs in the knockout mice. (72). Mouse studies show that A_2A receptor ligation induced relaxation of the aorta via K_ATP channels and this response is blunted to a similar extent by removal of the endothelial layer or in A_2A knockout mice (73).

In order to determine the effect of A_2A receptors on vasorelaxation in mice, aortic rings prepared from wild type and A_2A receptor knockout mice fed a high (4%) or normal salt diet were exposed to adenosine agonists and antagonists in their water bath (74, 75). This approach showed that the A_2A receptor protein was upregulated by about 30% in the high salt diet-derived wild type aortas compared to the normal salt diet aortas. Under both dietary conditions, a nonspecific adenosine agonist enhanced relaxation and the effect was blocked by an A_2A receptorspecific antagonist. An A_2A receptor agonist caused relaxation of aortic rings from either high or normal salt diet mice as long as they were wild type. High salt diet-fed wild type mouse aortae exhibited exaggerated vascular relaxation to the selective adenosine A_2A receptor agonist, and this response is lost in A_2A receptor null mice, indicating that the A_2A receptor is important for adaptive relaxation under high salt conditions.

A_2B receptors have a low affinity for adenosine and are thus activated only under pathologic conditions in which high concentrations at the micromolar level are achieved such as ischemia (76). A_2B receptors are upregulated by hypoxia inducible factor-1α (HIF-1α), a transcription factor that is stabilized by inflammatory/hypoxic conditions (77–79). In humans, examination of adenosine receptor transcript levels in cardiac tissue from persons with ischemic heart disease showed a selective induction of A_2B receptors in comparison with healthy controls and this may be a myocardial adaptation to ischemia (80). Although there is some controversy over the cardioprotective value of A_2B receptors, mice lacking these receptors display increased vulnerability to myocardial ischemia (81) and the anti-inflammatory effect of A_2B receptors present specifically on polymorphonuclear leukocytes is crucial to limiting injury (82). A_2B receptors may also mediate coronary artery dilation in humans, but the contribution of the A2B receptor may be minimal under normal conditions and more prominent in disease states such as metabolic syndrome (83–85).

Endothelial cells produce adenosine when injured (86). Adenosine has direct effects on endothelial barrier function, which is critically important in maintaining the arterial lumen and preventing inciting events in development of the fatty streak. Inflammation-induced increases in vascular permeability are blunted by adenosine, thus maintaining cell-to-cell adhesion and vascular integrity (87, 88). Which specific adenosine receptors are involved in these effects on the endothelial barrier is unclear and either both A_2A and A_2B receptor activation or A_2A activation alone is required (89–91).

Adenosine stimulates proliferation of endothelial cells and the A_2B receptor is involved in this effect (92, 93). A_2B receptor ligation promotes production of angiogenic factors, including VEGF (94, 95).

In vitro studies of cultured human coronary artery smooth muscle cells have shown that these cells express predominantly A₁ and A_2B adenosine receptors and, combined with A_2B receptor silencing experiments, further demonstrate that anti-proliferative effects of adenosine on coronary artery smooth muscle proliferation are A_2B-mediated, likely via the adenylyl cyclase/ cyclic adenosine monophosphate (cAMP)/ protein kinase A (PKA) axis (96). The limiting of proliferation of smooth muscle through A_2B activation may protect against luminal narrowing and stenosis post-injury (97).

5. ADENOSINE AS A TREATMENT FOR MYOCARDIAL PRESERVATION

Adenosine is released under conditions of ischemia or increased myocardial energy demand and in animal models, preconditioning with adenosine significantly decreases ischemiareperfusion-induced myocardial damage, but in humans, preconditioning is impractical (98–100). The germane issue is whether adenosine administration to humans during or immediately after myocardial infarction can limit damage and extend life expectancy. Unfortunately, studies in humans have not been as promising as might be anticipated based on results in mammalian models, although there are some potentially encouraging findings.

An early human trial of adenosine, Asymptomatic Myocardial Ischemia in STroke and Atherosclerotic Disease (Amistad) 1, was designed to explore whether, during the time of thrombolysis, adenosine (70 μg/kg/min over 3 hours) could significantly reduce extent of a myocardial infarction if given within 6 hours of infarction onset (101). This prospective, openlabel trial randomized 236 patients to adenosine or placebo with a primary end point of infarct size as determined by Tc-99 imaging. In AMISTAD 1, reduction in infarct size was observed only in cases of anterior infarction, while underpowering did not allow determination of whether overall clinical outcomes were improved. A second, larger trial, AMISTAD 2, randomized 2118 patients with electrocardiographic evidence of anterior ST-segment elevation myocardial infarction to 3 hours of either 50 or 70ug/kg/min adenosine along with fibrinolysis or percutaneous coronary intervention (102). AMISTAD 2 confirmed that adenosine lowers infarct damage in patients suffering an acute anterior infarct undergoing reperfusion therapy. In a posthoc analysis, adenosine was found to have positive impact on clinical outcomes when reperfusion was initiated early (within 3.17 hours of symptom onset) in the course of anterior myocardial infarction. Patients reperfused after the median 3.17 hours received no clinical outcome benefit from adenosine, giving insight into the optimal timing and methods of this adjunct therapy. Adenosine added more value when administered along with thrombolytic therapy than when it was given with percutaneous coronary intervention (103).

Marzilli et al (104) used adenosine as an adjunct to a percutaneous coronary intervention in acute myocardial infarction. A total of 54 patients within 3 hours of onset of acute myocardial infarction referred for percutaneous coronary intervention were randomized to receive intracoronary adenosine or placebo (saline). Those who received adenosine had a 6.4% reduction in residual diameter stenosis versus the saline group, and each patient receiving adenosine achieved TIMI 3 blood flow in the occluded artery. In contrast, 19 patients had TIMI 3 and 8 patients had TIMI 2 flow in the saline group, revealing a decrease in blood-flow strength. The study also found improved left ventricular wall motion after one week in 64% of the initially dyssynergic segments in the adenosine group and in only 36% of segments in the saline group (P=0.001). Regarding clinical outcomes, 5 patients died in the saline group, compared to none in the adenosine group (P<0.02). There was an improvement in clinical outcomes in the adenosine group versus saline, although the study included a small sample size.

Quintana et al (105) led the Attenuation by Adenosine of Cardiac Complications (ATTACC) study to evaluate the possible beneficial effect on left ventricular function of adenosine adjuvant therapy in patients between age 18 and 80 with chest pain of less than 12 hours duration and acute myocardial infarction treated with thrombolysis. Patients received adenosine (10µg/kg/min) or saline before or at the initiation of thrombolysis with either tissue plasminogen activator or streptokinase. The 10 µg/kg/min adenosine dose was better tolerated in comparison to 40 µg/kg/min adenosine in an earlier pilot study. Enrollment consisted of 302 patients given adenosine and 306 given placebo. The primary end-point for ATACC was indices of left ventricular systolic function by echocardiogram. Secondary end-points reported during a 12-month follow-up period included all-cause and cardiovascular death, non-fatal myocardial infarction, and the composite endpoint which combined cardiac mortality and non-fatal acute myocardial infarction. Short-term side effects including hypotension, bradycardia, atrioventricular block, recurrent chest pain, and congestive heart failure, were similar between adenosine and placebo groups. There were no significant differences in left ventricular function between the two groups, even when analyzed via infarct location (wall motion score indices and ejection fraction data showed similar trends when comparing adenosine and placebo groups). Non-fatal acute myocardial infarction and composite endpoint were alike at the 12-month follow-up. 12-month follow-up data of cardiovascular death showed lower levels of mortality in the adenosine group versus placebo (risk reduced by greater than 3%). Furthermore, when the patients were sorted by specific infarct damage location, total and cardiovascular mortality improved in patients with anterior acute myocardial infarction after a 12- month follow-up and during the complete follow-up (absolute risk reduction exceeding 6%). Although these trends were non-significant in this sample size and study, the potential for therapeutic benefit may warrant a larger study.

In a double blind study of 201 patients with ST segment elevation myocardial infarction receiving percutaneous coronary intervention, subjects were randomized to receive either intracoronary adenosine (4.5 mg) or saline control prior to reperfusion (106). The study failed to show an effect of adenosine on infarct size, but suggested possible myocardial preservation in those with short ischemia duration of less than 200 minutes.

Two randomized, double blind, placebo controlled, parallel group, dose finding phase 2b clinical trials are newly initiated to study the effect of a partial adenosine A₁ receptor agonist, neladenoson bialanate, in patients with heart failure and a preserved ejection fraction (≥45%) or heart failure with a reduced ejection fraction (≤35%) (107, 108). These two studies are designed to find the optimal dose of neladenoson bialanate for a potential phase 3 trial. Possible benefits of an A₁ partial agonist in heart failure include better mitochondrial function, halting of ventricular remodeling with reduced fibrosis and protection against ischemic injury.

6. THE ROLE OF ADENOSINE AND THIENOPYRIDINE DERIVATIVES AS ANTI-PLATELET AGENTS

Intracoronary atherothrombosis, a major cause of acute coronary events, is largely a consequence of platelet activation, aggregation and adhesion (109, 110). Platelet inhibition is an essential part of optimal medical therapy in the treatment of acute coronary events (111–113).

Platelets express A_2A and A_2B receptors, but do not express the A₁ or A₃ adenosine receptor (114, 115). When adenosine binds to A_2A receptors expressed on platelets, it stimulates adenylate cyclase, leading to an increase in intracellular cAMP within the platelet, which robustly inhibits platelet activation (116). Not surprisingly, thus effect is seen with highly selective A_2A receptor agonists, including CGS-21680 and ATL-146e (117–119). CGS-21680 is the classical gold standard A_2A agonist (120). To date, no A_2A agonist has been used directly in humans for anti-platelet effects.

Although platelet activation by adenosine is thought to be primarily mediated through the A_2A receptor, mice deficient in the A_2b receptor show increased platelet aggregation accompanied by upregulation of the ADP receptor P2Y1 (121). The platelet activation process in part is the result of the interaction between ADP and platelet P2Y purinergic receptors. The P2Y1 receptor is required for induction of platelet aggregation by ADP and can initiate the process of platelet aggregation but is not sufficient for a full aggregation response to ADP (122).

The P2Y12 receptor in particular is integrally involved in the ADP stimulated glycoprotein IIb/IIIa receptor activation that prompts increased platelet degranulation, thromboxane production, completion of the aggregation process and prolonged platelet aggregation (123, 124). Platelet inhibitors including, but not limited to, thromboxane inhibitors (aspirin), ADP antagonists (or P2Y12 inhibitors), the thienopyridines (clopidogrel and prasugrel), and nonthienopyrdines, (ticagrelor), are medical therapy options for the inhibition of thrombotic formation and processes (123). Thienopyridines inhibit platelet activation and aggregation due to their abilty to antagonize the P2Y12 receptors (123, 125). Dual anti-platelet therapy of aspirin and clopidogrel in multiple large clinical trials has proven efficacious in the prevention of complications in patients post-acute coronary syndrome events and percutaneous coronary intervention stent revascularization. (123, 126). Despite the proven efficacy of this combination, clopidogrel has pharmacodynamic limitations that result in reduced effectiveness (127, 128). Clopidogrel, a prodrug, requires a 2-step metabolic activation process catalyzed by a cytochrome P450 (CYP) to generate the active metabolite that inhibits the P2Y12 receptor (129–131). 85% of the prodrug is hydrolyzed before intestinal absorption (132). These Clopidogrel properties cause a delay of several hours from ingestion of drug to attainment of therapeutic drug levels (123). Further, clopidogrel has demonstrated a variability in response between patients due to variation in the CYP gene that codes for CYP-450 enzymes (133). These enzymes are involved in hepatic conversion of clopidogrel prodrug to its active metabolite. Another consideration is an association between polymorphisms of the CYP2C19 allele with reduced clopidogrel activity. (123, 134).

In contrast, newer P2Y12 agents like the third generation P2Y12 inhibitor ticagrelor act faster and have a stonger, more consistent anti-platelet effect (135, 136). Ticagrelor is a direct-acting and reversibly binding P2Y12 ADP receptor blocker that also inhibits adenosine uptake via the ENT (137). This can prolong the short half-life of adenosine and increase extracellular concentration. Incubation of human erythrocytes, and human, canine, and rodent cell lines in ticagrelor leads to inhibition of adenosine uptake, but in vivo studies at physiologically relevant concentrations of ticagrelor show no appreciable effect on adenosine levels in healthy human subjects (135, 138–141).

Ticagrelor has greater efficacy than clopidogrel, leading to improved prognosis for patients with acute coronary syndrome (142, 143). In a study by Li et al (135), ticagrelor and clopidogrel were compared in their effect on adenosine, adenosine deaminase (ADA), and cAMP levels in the blood, as well as anti-platelet effect, on patients with non-ST-segment elevation acute coronary syndrome receiving dual antiplatelet therapy, who were reperfused via percutaneous coronary intervention. There were no significant differences in clinical baseline qualities, in adenosine, ADA or cAMP levels between patient groups. After percutaneous coronary intervention, patients either received a loading dose of 180 mg of ticagrelor followed by 90 mg every two days, or 300 mg loading dose of clopidogrel and then 75 mg daily. As measured by immunoassay upon admission and 48 hours after coronary angiography, there was a significant increase in cAMP and adenosine levels in the ticagrelor group, but no difference in adenosine deaminase levels between the patients receiving ticagrelor or clopidogrel. Thrombelastograph Hemostasis Analyzer measures showed minimal correlation between inhibition of platelet aggregation and adenosine/cAMP concentration in the plasma. However, the average inhibition of platelet aggregation was 93.5% greater in patients given ticagrelor compared to those given clopidogrel. Major adverse cardiac events were noted after a 30-day follow-up and were higher in the clopidogrel group (5.8%) compared to ticagrelor (2.9%). However, rate of hospitalization showed no distinction between groups, nor did significant bleeding, dyspnea, or AV block at the 30-day follow-up.

Prasugrel, a new-generation thienopyridine anti-platelet agent unaffected by genetic variations in CYP2C19, has been also been shown to be superior to clopidogrel in preventing similar clincal outcomes in patients with ACS, who underwent percutaneous coronary intervention in the TRITON (Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel) trial (144, 145). Compared with dipyridamole, an ENT1 inhibitor, which was mentioned earlier, ticagrelor has a lower affinity for the ENT transporter. Other P2Y12 antagonists, cangrelor, elinogrel, and metabolites of clopidogrel and prasugrel, did not show any notable activity or interaction with the transporters. In patients with P2Y12 deficiency, ticagrelor has still been shown to be superior to dipyramidole in anti-platelet effects in whole blood, possibly because ticareglor is not only an ENT1 antagonist, but can oppose P2Y12-mediated inhibition of adenylyl cyclase (138).

The PLATO study (Platelet Inhibition and Patient Outcomes) supplied further evidence of reduced cardiovascular mortality with ticagrelor as compared to clopidogrel (146, 147). This study did show that ticagrelor had a greater incidence of dyspnea and ventricular pauses. (138, 148). These combined considerations led investigators to hypothesize that Ticagrelor may have pleiotropic properties with clinically pertinent, non-platelet related effects (149, 150). Increased adenosine half-life and plasma concentrations indicate potential for clinically relevant consequences because of the biological effects associated with adenosine as described throughout this review. Most recently, dyspnea induced as a side effect of Ticagrelor was found to be unrelated to adenosine levels (151).

A recent in vitro study combined adenosine agonists, testing both selective and nonselective compounds, with a P2Y12 inhibitor and found that both non-selective and A_2A selective agonists, but not an A_2B selective agonist were effective in amplifying the anti-aggregatory effect of the P2Y12 antagonist (152). This opens up the possibility of dual therapy targeting platelet aggregation involving adenosine.

7. ADENOSINE IN INFLAMMATION AND CHOLESTEROL HOMEOSTASIS

Methotrexate, a drug that raises adenosine levels, is used as first line therapy to treat rheumatoid arthritis due to its anti-rheumatic and anti-inflammatory properties (153, 154). In addition, it can improve survival in rheumatoid arthritis patients because of its beneficial antiatherogenic effects, reducing atherosclerosis-mediated cardiovascular complications, which are common in the rheumatoid arthritis population. Micha et al (155) compiled data from 10 observational studies totaling 66,334 subjects, and showed that methotrexate at a median dose of 13–15 mg/week can reduce cardiovascular disease risk by 21% and lower myocardial infarction risk by 18%, in patients with systemic inflammation. A meta-analysis by Roubille et al (156) examined cardiovascular outcomes in persons with rheumatoid arthritis, psoriasis or psoriatic arthritis on anti-rheumatic drugs. The primary outcome assessed was the link between treatment and all cardiovascular events. In the rheumatoid arthritis patients, those treated with methotrexate saw a 28% reduction in the risk of cardiovascular events while TNF inhibitors led to a 30% reduction. Studies have shown that methotrexate can confer up to a 70% reduction in mortality in patients with rheumatoid arthritis, with a substantial decrease in cardiovascular death (157).

There is much evidence to support the importance of adenosine release as a mechanism through which methotrexate confers cardiovascular benefits. The inflammatory state in rheumatoid arthritis and other autoimmune disorders may promote the onset of vascular injury and atherosclerosis and some of the consequences may be mitigated by methotrexate via adenosine (158). Methotrexate can increase extracellular adenosine at sites of inflammation, decrease attachment of leukocytes and neutrophils to endothelium and fibroblasts, as well as decrease the expression of inflammatory cytokines and complement pathways that exacerbate lesions and fatty plaque build-up (159–161). It has been shown that inflammatory processes involving leukocytes and neutrophils play a key role in formation of atheroma and, in the case of neutrophils, in clot formation (162–164). Anti-inflammatory effects of methotrexate are seen in a rabbit model of atherosclerosis with carotid artery stenting where methotrexate treated rabbits displayed reduced neointimal thickness and decreased serum cytokines and adhesion molecules (165, 166).

Although adenosine has many anti-inflammatory effects, activation of adenosine receptors on cells of the immune system can induce both pro- and anti-inflammatory responses (31, 32). The concentration of adenosine, affinity of the receptor, proportion of each receptor subtype and cell type may all contribute to determining the relative pro-inflammatory/anti-inflammatory response to adenosine (167, 168). For example, the A_2B receptor, a low affinity adenosine receptor, is anti-inflammatory when inflammation is acute, but may contribute to the inflammatory state when high adenosine levels persist in chronic diseases such as pulmonary fibrosis where A_2B receptors in lung fibroblasts may be pro-fibrotic (169–171). The biological effects of adenosine occur as a result of the interaction with 4 G-protein–coupled receptors and the Gi inhibitory G protein or the Gs the stimulatory G protein, resulting in decreased or increased cAMP (138). The A_2B receptor has been proposed to interact not only with Gs, but with Gq proteins to activate phospholipase C causing increased PKC activation and elevation of intracellular calcium, leading to an inflammatory response (172).

Since the A₁ receptor, associated with pro-inflammatory activity, displays high affinity for adenosine, at low adenosine concentrations the pro-inflammatory effect of the A₁ receptor may predominate, whereas at higher plasma concentrations the anti-inflammatory effect of other receptors predominates. In this manner increasing levels of adenosine would serve as a negative feedback on the immune system. (173, 174). Underscoring this phenomenon, while adenosine has been shown to inhibit neutrophil adhesion and migration through activation of the A_2A and A_2B receptors, activation of the A₁ and A₃ receptors on neutrophils enhances chemotaxis and promotes inflammation. (175, 176). In monocytes/macrophages, adenosine effects are largely anti-inflammatory. Adenosine decreases secretion of the pro-inflammatory cytokines IL-2, TNF-α and IFN-γ, thus promoting differentiation of monocytes to the anti-inflammatory M2 phenotype (177–179). This is consistent with findings that adenosine is atheroprotective via modulation of macrophage behavior in the arterial wall as discussed here.

Anti-inflammatory effects of methotrexate resulting from A_2A receptor activation may lead directly to its vasculoprotective properties and also to its ability to improve lipid handling (180). A number of atheroprotective proteins involved in reverse transport of cholesterol from the periphery to the liver for excretion, are upregulated by methotrexate via adenosine (Figure 3). One of these key mediators in cholesterol homeostasis is the membrane transporter adenosine 5′ triphosphate (ATP) binding cassette transporter (ABCA)A1, involved in active transport of cholesterol and phosopholipids to extracellular apolipoprotein (apo)A-I (181, 182). In humans, homozygous or compound heterozygous mutations in the ABCA1 gene manifest as Tangier disease, described as profoundly reduced levels of high-density lipoprotein, hepatosplenomegaly, cholesterol deposition in various tissues, and a high occurrence of atherosclerotic cardiovascular disease (183, 184).

Figure 3. Methotrexate-Induced Adenosine Upregulation Attenuates Atherosclerotic Risk.

Treatment with methotrexate (MTX) increases adenosine production which acts via the A2A receptor to upregulate various cholesterol efflux transporter proteins found on the macrophage cell membrane, namely: A) ATP-binding cassette subfamily A member 1 (ABCA1), which effluxes intracellular cholesterol in the form of apolipoprotein A-1, B) ATP binding cassette subfamily G member 1 (ABCG1), which effluxes intracellular cholesterol in the form of high-density lipoprotein (HDL), and C) 27-hydroxylase (27OH), which effluxes intracellular cholesterol in the form of 27-hydroxycholesterol. The formation and subsequent efflux of each of these cholesterol byproducts decrease the risk of lipid overload and foam cell formation of the macrophage, thereby decreasing the risk of atherosclerosis and eventual onset of cardiovascular disease.

Another important mediator in cholesterol homeostasis, ABCG1, promotes cholesterol efflux to nascent HDL (185, 186). Adenosine agonist treatment of macrophages increases ABCG1-dependent efflux and lowers foam cell formation (187). This transporter works together with ABCG1 in reverse cholesterol transport (188). ABCA1 may covalently modify lipid-deficient apoA-I to create HDL, thus allowing ABCG1 to transport cholesterol to this new HDL (189).

The cytochrome P450 cholesterol 27-hydroxylase is a third crucial reverse cholesterol transport protein involved in extra-hepatic cholesterol metabolism. This enzyme modifies cholesterol into oxygenated forms such as 27-hydroxycholesterol, a more polar product that easily exits the cell (190). ABCA1, ABCG1 and cholesterol 27-hydroxylase help move cholesterol out of cells to be degraded by the liver into bile acids for excretion and all are upregulated by adenosine acting via the A_2A receptor (187, 191). Further, oxysterols such as 27hydroxycholesterol serve yet another role by functioning as signaling molecules that enhance expression of ABCA1 and ABCG1 (192). These oxysterols activate the ABCA1 promoter and induce ABCA1 mRNA expression via the nuclear transcription factor Liver X receptor (LXR) (193, 194). Costet et al. (195) showed that ABCA1 mRNA levels increase in macrophages treated with oxysterols. Fu et al (196) found a dose-dependent rise in ABCA1 and ABCG1 via addition of exogenous LXR ligands and cholesterol loading, confirming that oxysterols upregulate reverse cholesterol transport proteins via LXR (197). In the THP-1 human macrophage cell line, addition of the pro-inflammatory cytokine interferon (IFN)-γ, induces an activated, atherogenic state with suppression of 27-hydroxylase and ABCA1 and stimulation of foam cell formation (198). Occupancy of A_2A receptors by adenosine agonists counteracts the suppression of cholesterol 27-hydroxylase and ABCA1 levels, increasing both 27-hydroxylase and ABCA1 mRNA expression by about 80% in THP-1 differentiated macrophages. Foam cell formation also decreased by between 25–40%. This occurs via a pathway involving PKA and cAMP (199, 200).

The multiple advantageous effects of adenosine A_2A receptor agonism on cholesterol transport and processing through ABC genes and 27-hydroxylase, coupled with antiinflammatory actions of A_2A receptor ligands, open the possibility of using specific A_2A agonists as treatment for atherosclerotic cardiovascular disease (201, 202). The structure of the receptor is well-characterized and a number of agonists have been developed (203) the selective A_2A agonist, regadenoson, is the preferred vasodilator stress agent used in the United States (204). Newer drugs that specifically target the A_2A receptor and can be administered orally may be candidates for future cardioprotection trials (205, 206).

8. CONCLUSIONS

Adenosine is an endogenous purine nucleoside with an extremely short half-life that functions as an extracellular signaling molecule. Under physiologic conditions, adenosine is present at very low concentrations. However, conditions of hypoxia, inflammation and cellular stress lead to generation of this molecule and a consequent rise in interstitial levels, conferring protection against tissue damage. Among its cardioprotective effects are induction of coronary artery vasodilation, reduction of myocardial oxygen demand and anti-platelet activity.

Adenosine and related compounds are commonly used vasodilators for determination of coronary fractional flow reserve in myocardial perfusion imaging for the detection of obstructive coronary artery disease. It has potent negative chronotropic (slows sinus rate) and dromotropic (slows AV node conduction) effects on the heart. It decreases peripheral resistance and arterial pressure. It is eliminated by cellular uptake and metabolism (deamination and phosphorylation).

In rodents and other animal models, adenosine enhances post-ischemic functional recovery of the myocardium with decreased infarct size. While highly effective in these models of reperfusion, it has failed to achieve dramatic clinical benefits in human studies. Further, the exact pathways and receptors involved in cardioprotection remain unresolved even though selective agonists and antagonists have been developed for each adenosine receptor subtype. Although the adenosine receptors and particularly the A_2A receptor are still considered promising drug targets for the heart, results thus far have been disappointing. UK-432,097, an A_2A agonist developed by Pfizer, was tested for use in asthma and chronic obstructive pulmonary disease, but failed in clinical trials due to lack of efficacy and is not available for use in humans (202, 207, 208). While new compounds and delivery techniques are in development, the future is uncertain (209–212).

KEY POINTS.

Adenosine is an endogenous purine nucleoside with cardioprotective properties.

It is the primary drug used in the treatment of stable narrow-complex supraventricular tachycardia, but has been disappointing in human trials as a myocardium-sparing treatment during acute myocardial infarction.

Adenosine actions are mediated by specific cell surface receptors (A₁, A_2A, A_2B, and A₃ adenosine receptors) with many anti-inflammatory and atheroprotective actions resulting from A_2A receptor activation.

Blockade of the P2Y12 component of adenosine diphosphate receptors on the platelet surface is responsible for the anti-aggregatory effect of some commonly used anti-platelet therapies