Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Adenosine is naturally occurring nucleoside that modulates a variety of physiological functions including the regulation of coronary blood flow (Kusachi et al., 1983; Mustafa, 1980). Adenosine elicits its physiological actions by binding to specific cell surface receptors that were originally classified as A₁ and A₂ subtypes based on their agonist potency profiles and their ability to modulate adenylate cyclase (Williams, 1990; Baer and Vriend, 1985; Londos et al., 1980; Hussain and Mustafa, 1992; Sabouni et al., 1989, 1990, 1991). Adenosine receptor has been pharmacologically identified in coronary artery from many species such as dog, bovine and human to be of A₂ receptor subtype (Kusachi et al., 1983; Mustafa and Askar, 1985; Ramagopal et al., 1988). Adenosine A₂ receptor is believed to cause vasorelaxation in coronary artery (Ramagopal et al., 1988). The A₂ receptors have been further classified into A_2A and A_2B subclasses. The A_2A receptors are believed to be high affinity receptors, whereas, A_2B are low affinity (Londos et al., 1982; Jarvis et al., 1989).

The A₂ receptor has not been as well characterized as A₁ receptor especially in human cardiovascular tissues. Recently, several laboratories have reported the presence of adenosine A_2A receptors in cardiac myocytes from several species (Romano et al., 1989; Xu et al., 1992; Liang and Haltiwanger, 1995; Neuman et al., 1989; Stein et al., 1994). Adenosine A_2A receptors causes an increase in inotropy of the ventricular myocytes (Dobson and Fenton, 1997).

The A₂ (A_2A subclass) receptor cDNA (RDC8) from canine thyroid has been isolated and expressed (Libert et al., 1989; Maenhaut et al., 1990). The deduced amino acid sequence of RDC8 indicates that it encodes a protein with a predicted molecular weight of 45 kDa. The cloned A_2A receptor has been shown to contain seven transmembrane helices and exhibits appropriate A_2A receptor pharmacology such as ligand binding characteristics and stimulation of cyclic AMP.

Our study was undertaken to ascertain the presence of adenosine A_2A receptors in human cardiovascular system. We used a direct immunological approach by developing antipeptide antibody to adenosine A_2A receptor. Using this antibody we directly demonstrated the presence of the receptor subtype in human atria and ventricle in addition to coronary vasculature and platelets. Similar results were observed with porcine cardiovascular system. The use of antibodies against A_2A receptor provides us with yet another tool to study the adenosine receptor.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials

Amino acids and activators were purchased from Amicon (Beverly, MA). [¹²⁵I]Anti-IgG was ordered from Amersham Corp. (Arlington Heights, IL). [¹²⁵I]PAPA-APEC (catalog no. NEX 322) was a kind gift from Du Pont (Boston, MA). Human platelet membranes were a kind gift from Dr. K. C. Aggarwal (Department of Molecular and Biochemical Pharmacology, Brown University, Providence, RI). SANPAH was purchased from Pierce Chemical Company (Rockville, IL). Membranes from cells transfected with recombinant rat A₁ (catalog no. A-232), human A_2A (catalog no. A-237) and human A₃ (catalog no. A-233) adenosine receptors were purchased from Research Biochemicals International (Natick, MA).

Methods

Synthesis of adenosine A_2A receptor peptide. Peptide corresponding to the deduced amino acid sequence (361-390) of RDC-8, the presumed adenosine A_2A receptor cDNA from canine, was synthesized with an additional cysteine at the carboxy end to facilitate the conjugation of the peptide to KLH. The peptide (YTLGLVSGGIAPESHGDMGLPDVELLSHEL) was synthesized at the Biotech Core Facility at East Carolina University, Greenville, NC by a fluorenylmethoxycarbonyl method. One rational for selecting this sequence was that none of the other adenosine receptors extend that far. This sequence is present only in adenosine A_2A receptor. A "BLAST" search was performed on the sequence of the peptide to check for sequence homology with other proteins. As expected, the peptide sequence had very high homology with adenosine A_2A receptors from different mammalian species: human, 80%; guinea pig, 72% and rat, 54%. No homology with adenosine A₁, A_2B or A₃ was obvious. Homologies with other mammalian proteins were not significant. Automated amino acid sequence analysis was performed by Dr. Donald R. Hoffman, Department of Pathology and Laboratory Medicine, East Carolina University, Greenville, NC. The majority of the peptide constituted of the entire sequence, whereas, secondary and trace contained entire sequence with one (Y) and two (YT) amino acids missing, respectively.

Antibody production. The peptide was coupled to KLH using gluteraldehyde. Briefly, 500 µl of 1% aqueous gluteraldehyde solution were added dropwise over a period of 1 hr into 5 ml of peptide solution (containing 1 mg/ml each of the peptide and KLH) with constant stirring. The mixture was dialyzed overnight at 4°C. The coupling efficiency was estimated to be about 80% by this method (estimated using tracer radiolabeled peptide). The peptide coupled to KLH (corresponding to 300 µg peptide calculated) was injected into two rabbits subcutaneously in the back region. The rabbits were boosted twice with 300 µg peptide-KLH at approximately 2-wk intervals. The rabbits were bled through ear vein and serum separated.

Preparation of crude membranes. Accidental death donor human heart (female, 16 yr old) was obtained from LifeNet Transplant Services (Virginia Beach, VA). The warm ischemia time was 1 hr. Porcine hearts were obtained from local slaughter house. Branches of left anterior descending and circumflex were dissected out and cleaned of the adherent tissue and fat; this constituted coronary artery for membrane preparation. Similarly, pieces from ventricle and atria were dissected out and cleaned of any fat and major vascular tissue. All steps were conducted at 4°C. The tissue was minced with scissors followed by homogenization using a polytron (four times at a setting of 6) with six volumes of buffer A (5 mM Tris-HCl, pH 7.5, 0.4 mM MgCl₂, 0.25 M sucrose, containing protease inhibitors 5 mM phenylmethylsulfonyl fluoride, 1 mM benzamidine HCl and 1 mM trypsin inhibitor). The homogenate was centrifuged at 200 × g for 10 min and the supernatant filtered through a double gauze. The supernatant was centrifuged at 105,000 × g for 1 hr. The pellet was washed twice with buffer A followed by resuspension in buffer B (50 mM Tris-HCl, pH 7.5, 4 mM MgCl₂, containing the protease inhibitors described above) at a concentration of 3 to 4 mg/ml protein and aliquots were stored at 80°C for later use. The protein concentrations were determined using Bio-Rad (Hercules, CA) protein assay that is based on Bradford method using bovine serum albumin as standard (Bradford, 1976). Membranes from porcine coronary artery, ventricle and atria were prepared using the same method.

Cultured porcine coronary artery smooth muscle cells. Smooth muscle cells from porcine coronary artery were isolated and cultured as described by Fehr et al. (1990) with minor modifications. Briefly, left circumflex coronary arteries were dissected from porcine hearts and cleaned of the adherent fat and connective tissue. The arteries were cut open longitudinally and the endothelium was removed by rubbing the intimal surface gently with scalpel blade. Tissue was rinsed twice with sterile HBSS containing (in mM): KCl 5.0, KH₂PO₄ 0.3, NaCl 138, NaHCO₃ 4.0, Na₂HPO₄.7H₂O 0.3, D glucose 5.6, and HEPES 10.0. Tissue was then soaked for 10 min at ambient temperature in sterile Hanks' balanced salt solution containing antibiotic/antimyoctic solution. Arteries were minced with scissors and incubated for 3 to 4 hr at 37°C with vigorous shaking in sterile Hanks' balanced salt solution containing 1 mg/ml collagenase type I (300 U/mg), 0.5 mg/ml Soya bean trypsin inhibitor and 3% bovine serum albumin. It was then filtered through six layers of sterile gauze and centrifuged at 100 × g for 10 min. The pellet was washed twice with Dulbecco's modified Eagle's medium (GIBCO) and finally resuspended in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and cultured at 37°C in a 95% O₂ + 5% CO₂ incubator. Media was changed twice a week and when confluent (about 1 wk) were split at 1:3 ratio using trypsin (0.25%).

Isolation of cardiac ventricular myocytes. Cardiac myocytes from porcine ventricle were isolated according to the method of Glick et al. (1974), and membranes were prepared. Briefly, left ventricular tissue was cleaned of any visible major blood vessels and minced into about 3-mm cubes and soaked in sterile PBS for 30 min to remove any blood. Tissue was placed in digestion buffer (1 mg/ml collagenase type I and 2 mg/ml hyaluronidase in sterile PBS) and incubated for 20 min at 37°C with vigorous shaking (100 strokes/min). Sterile PBS was added (three volumes) and filtered through four layers of gauze. Tissue was place again in digestion buffer and mixed vigorously again. This procedure was repeated five times. Filtrate from last three steps were collected and pooled. Filtrate was centrifuged at 60 × g for 3 min and pellet resuspended in 1 ml of PBS. Cell suspension was overlaid on 10 ml of ice cold 3% Ficoll solution in PBS and centrifuged at 60 × g for 3 min. Pellet was washed three times. Membranes were prepared from myocytes immediately after isolation as described above for smooth muscle cells.

Western blot analysis. Crude membrane fractions (50 µg protein) from tissue or cells were subjected to 10% SDS-PAGE under either reducing (in the presence of -mercaptoethanol) or nonreducing conditions (in the absence of -mercaptoethanol) as described by Laemmli (1970). After electrophoresis, the proteins were electroblotted to an Immobilon membrane (Amicon). The free protein binding sites were blocked by incubating the membrane with PBS containing 5% skimmed milk (PBS-milk) for 2 hr at room temperature or overnight at 4°C. The membranes were then incubated 1 to 2 hr at room temperature with PBS-milk containing antipeptide antiserum (1:1000 final dilution) with gentle mixing. The membranes were washed 4 × 10 min with PBS containing 0.05% Tween 20. The membranes were incubated with [¹²⁵I]goat antirabbit IgG (500,000 cpm/ml) for 2 hr at room temperature with gentle mixing. The membranes were washed 4 × 10 min. with PBS containing 0.05% Tween 20, air dried and autoradiographed.

Photoaffinity labeling of A_2A adenosine receptor. Crude membrane fractions (250-300 µg protein) from various tissues suspended in 2 ml of labeling buffer [50 mM HEPES (pH 7.5), 5 mM MgCl₂, 0.01% (w/v) CHAPS (3-[(3-chloramidopropyl)dimethylammonio]-1-propanesulfonate) and 0.4 U/ml adenosine deaminase]. Incubation was carried out for 1 hr at 37°C with ~0.8 nM [¹²⁵I]PAPA-APEC in the presence or absence of unlabeled CGS-21680 (10⁵ M). After incubation the membranes were washed twice with labeling buffer and centrifuged at 43,000 × g for 5 min. The labeled membranes were then suspended in 2 ml of labeling buffer and poured in six-well culture plate and exposed to UV light (model UVCG-25 mineral light) for 15 min at a distance of 1 cm. The photolabeled fractions were washed twice with labeling buffer with centrifugation at 43,000 × g for 5 min and finally resuspended in 500 µl of labeling buffer. Photoaffinity labeled membranes (50 µg) were subjected to 10% SDS-PAGE under reducing conditions as described by Laemmli (1970). The radiolabeled bands were visualized by autoradiography.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Titer of A_2A adenosine receptor antibody. Antisera from both the rabbits showed similar antibody characteristics. The antisera from two rabbits were not pooled together and all the data presented are from rabbit 1. The titer of the adenosine A_2A receptor antibody was determined by solid phase radioimmunoassay using [¹²⁵I]anti IgG as the secondary antibody. The A_2A receptor antibody showed a positive signal even at a dilution of 1:50,000 (fig. 1).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Titer of the anti-A2A adenosine receptor peptide antibody determined by solid phase radioimmunoassay. ELISA plate wells were coated with A2A-receptor peptide (100 ng/well) and incubated with serial dilution of the antipeptide serum or normal rabbit serum. The wells were washed and incubated with [125I]anti IgG (10,000 cpm/well). The bound radioactivity was counted after washing the wells.

Western blot analysis of A_2A receptor. Whole antisera was used for Western blotting. Because bovine brain striatal membranes have been shown by our studies as well as by other investigators (Barrington et al., 1990; Marala and Mustafa, 1993) to contain high density of A_2A receptors, we chose this tissue as a positive control. Initially, the blots were incubated with the primary antisera (A_2A receptor antisera) overnight at 4°C which showed a few nonspecific bands in addition to the A_2A receptor. To reduce the nonspecific bands, the time of incubation was reduced to 1 to 2 hr at room temperature. Under these incubation conditions, the Western blot using bovine brain striatal membranes showed one major immunoreactive band corresponding to a molecular weight of 45 ± 1 kDa (fig. 2). Membranes from porcine coronary artery, ventricle and atria showed a single major immunoreactive band migrating as 45 ± 1 kDa size (fig. 2). Human coronary artery, ventricle, atria and platelet membranes also showed a major immunoreactive band with a molecular weight of about 45 ± 1 kDa (fig. 2). In addition, sometimes these tissues also showed a minor band that migrated as a high molecular weight protein. Human atria and ventricle sometimes showed a 90-kDa protein in addition to the major 45 ± 1 kDa immunoreactive band (fig. 2). These high molecular weight bands could be either nonspecific bands or the A_2A receptor aggregates. The latter seems likely since freshly prepared membranes did not depict the high molecular weight bands (data not shown).

View larger version (44K): [in this window] [in a new window]   Fig. 2.   Western blot analysis of the membrane fraction from human and porcine cardiovascular tissues using anti A2A adenosine receptor peptide antibody. A 50-µg portion of membrane fractions from (A) bovine brain striatum (lanes 1 and 2), porcine coronary artery (lanes 3 and 4), human coronary artery (lanes 5 and 6), human ventricle (lanes 7 and 8) and human atria (lanes 9 and 10), (B) bovine brain striatum repeated (lanes 11 and 15), human platelet (lanes 12 and 16), porcine atria (lanes 13 and 17) and porcine ventricle (lanes 14 and 18) were subjected to 10% SDS-PAGE under reducing (lanes 1, 3, 5, 7, 9, 11, 12, 13 and 14) or nonreducing conditions (lanes 2, 4, 6, 8, 10, 15, 16, 17 and 18). These proteins were electroblotted and analyzed by Western blotting using anti A2A-receptor peptide antibody (1:1000 final dilution) as the primary antibody and [125I]goat antirabbit IgG as the secondary antibody. The immunoreactive bands were visualized by autoradiography. Bovine brain striatum membranes were used as controls. The additional band in lane 6 at about 40 kDa does not represent a protein immunoreactive band but instead is a scratch on the autoradiographic film. Representative autoradiograms are shown and the experiment was repeated at least three times with identical results.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Western blot analysis of the membrane fraction from cultured porcine coronary artery smooth muscle cells and ventricular myocytes. A 50-µg portion of membrane fractions from cultured porcine coronary artery smooth muscle cells (lane 1) and porcine ventricular myocytes (lane 2) were subjected to 10% SDS-PAGE under reducing conditions followed by Western blotting using anti A2A-receptor peptide antibody (1:1000 final dilution) as the primary antibody and [125I]goat antirabbit IgG as the secondary antibody. The immunoreactive bands were visualized by autoradiography.

View larger version (50K): [in this window] [in a new window]   Fig. 4.   Displacement of immunoreactive A2A receptor band in the presence of excess peptide. A, For displacement of immunoreactive band with excess peptide, increasing concentrations of bovine striatal membranes (4-120 µg protein) were analyzed by Western blot as described in "Materials and Methods." B, Immunoreactive band (at 7 µg/lane protein) was competed with increasing concentrations of peptide (lane 1 = no peptide, lane 2 = 50 µg/ml, lane 3 = 100 µg/ml and lane 4 = 200 µg/ml) as described in "Materials and Methods."

View larger version (68K): [in this window] [in a new window]   Fig. 5.   Western blot analysis of the membrane fraction from cells transfected with A2A adenosine receptor peptide antibody. A 5-µg portion of membrane fraction from cells transfected with recombinant A2A (lanes 1 and 2); A1 (lane 3) and A3 (lane 4) receptors were subjected to 10% SDS-PAGE under reducing conditions and analyzed by Western blotting using anti A2A-receptor peptide antibody (1:1000 final dilution) in the absence (lanes 1, 3 and 4) or presence of 100 µg/ml peptide (lane 2) as described in "Materials and Methods." The immunoreactive band was visualized by autoradiography.

Photoaffinity cross-linking. Cross-linking was conducted in the human tissues (atria and ventricle) using [¹²⁵I]PAPA-APEC and SANPAH as the photoaffinity cross-linker. All the tissues showed a 45-kDa protein that was specifically labeled with the radioactive ligand (fig. 6). The presence of 10⁵ M CGS-21680 drastically reduced the labeling of the 45-kDa receptor (fig. 6) suggesting that this band is the A_2A adenosine receptor.

View larger version (97K): [in this window] [in a new window]   Fig. 6.   Autoradiography of [125I]PAPA-APEC labeled A2A adenosine receptor in human cardiac tissues. Membrane fractions from human ventricle (lanes 1 and 2) and human atria (lanes 3 and 4) were photoaffinity labeled with [125I]PAPA-APEC with SANPAH as the photoaffinity cross-linker, in the absence (lanes 1 and 3) or presence (lanes 2 and 4) of 105 M unlabeled CGS 21680. The photoaffinity labeled fractions were subjected to 10% SDS-PAGE and labeled bands were visualized by autoradiography. Representative autoradiogram is shown and the experiment was repeated at least two times.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

In this study, we raised an antipeptide antibody against the canine A_2A adenosine receptor clone RDC8 (Libert et al., 1989). We used this antibody to immunologically characterize the A_2A receptor in human and porcine cardiovascular tissues. The use of the specific antibody provides us with a new tool to gain an insight in the study of the adenosine receptor in various tissues especially vascular and other tissues that are particularly difficult to work with.

Traditionally, adenosine receptor has been identified in various tissues by radiolabeled ligands binding studies and/or by pharmacological techniques. The recent development of relatively specific receptor ligands like CGS-21680, PAPA-APEC, APNEA, etc. (Barrington et al., 1990; Jarvis et al., 1989; Stiles et al., 1985), have provided us with tools to identify the receptor subtype in various tissues including vascular tissues. Traditionally, tritiated radioligands failed to bind the membrane fractions prepared from arteries. This could be because of 1) low specific activity of the tritiated ligands; 2) relatively low abundance of the receptor subtype in the tissue or, 3) receptor uncoupling during membrane preparation from coronary artery. However, radioiodinated ligands (¹²⁵I-APE and ¹²⁵I-azidoAPE) have been used to characterize A_2A receptors in porcine coronary artery (Hussain et al., 1996). Radioiodinated ligands are preferred for such experiments due to the relatively high specific activity of ¹²⁵I. More recently, [³H]SCH58261, a new, selective A_2A receptor antagonist, was used to characterize A_2A receptor in porcine coronary artery membranes (Belardinelli et al., 1996). Unfortunately, [³H]SCH58261 is not commercially available, therefore, the development of a specific antibody against the A_2A adenosine receptor provided us with a unique tool to characterize the receptor in membranes prepared from arteries of various sources.

Recent photoaffinity cross-linking studies using [¹²⁵I]PAPA-APEC indicate that the A_2A adenosine receptor in bovine striatal membranes is a glycoprotein with an apparent molecular weight of 45 kDa (Barrington et al., 1990). The A_2A adenosine in various species including PC12 cells, frog erythrocytes and rabbit striatum has also been identified as a 45-kDa protein (Nanoff et al., 1990). In this study, we demonstrated using immunological techniques that A_2A receptor exists as a 45-kDa protein in bovine striatum, porcine and human cardiovascular tissues. The size of the adenosine A_2A receptor (45 kDa) seems to be conserved in all the three species tested. Western blotting of membranes from cardiac tissues (atria and ventricle) from human and porcine demonstrated the presence of 45-kDa receptor. However, this raises an important issue. Is the immunoreactive A_2A receptor present in the ventricular tissue or on the coronary vasculature supplying the ventricle? To answer this question, we isolated ventricular myocytes by collagenase method (Glick et al., 1974) and prepared membranes. These isolated ventricular myocyte membranes also showed the presence of 45-kDa immunoreactive band indicating the presence of A_2A receptor in ventricle. Additionally, cultured smooth muscle cells from porcine coronary artery (Fehr et al., 1990) also showed the presence of the A_2A receptor indicating the presence of the receptor on the smooth muscle of the artery and not a contamination from adhering adipose and connective tissues. Control blots were incubated with preimmune rabbit serum (under identical conditions) instead of antiserum to study the specificity of antibody. Blots incubated with preimmune serum did not show any immunoreactive band at 45-kDa position indicating the specificity of the antiserum (data not shown). The 45-kDa immunoreactive band could be competed by excess peptide indicating the specificity of the antibody. The antibody also identified the 45-kDa immunoreactive band in membranes from transfected cells. The authenticity of the A_2A receptor was also confirmed by photoaffinity cross-linking studies using [¹²⁵I]PAPA-APEC that specifically labeled a 45-kDa protein in human tissues.

Earlier reports on the distribution of A_2A receptors in heart have been controversial. Adenosine A_2A receptors were reported to be absent in ventricular myocytes from guinea pig, rat and rabbit (Wilken et al., 1991; Shryock et al., 1993). Other reports show the presence of both the A₁ and A_2A adenosine receptor subtypes in cultured ventricular myocytes from rat (Romano et al., 1989), fetal chick (Xu et al., 1992; Liang and Haltiwanger, 1995), guinea pig (Neuman et al., 1989; Stein et al., 1994). Adenosine A_2A receptors in rat ventricular myocytes have been shown to cause increase in inotropy via cAMP-dependent and -independent mechanisms (Dobson and Fenton, 1997). Dixon et al. (1996) recently presented direct evidence for the presence of A_2A receptor transcript in rat heart by reverse transcriptase-polymerase chain reaction. It is now well accepted that adenosine A_2A receptors are present in ventricular myocytes.

However, the presence of adenosine A₁ receptors in both atria and ventricles have been shown in various species using pharmacological and radioligand binding studies (Linden et al., 1985; Lohse et al., 1985; Martens et al., 1987, 1988). In atria, adenosine and its analogs produce a negative ionotropic effect even in the absence of -adrenergic receptor agonist; however, the negative ionotropic effect in ventricle is apparent only upon the pretreatment by -adrenergic receptor agonist (Belardinelli et al., 1989; Liang, 1989; Martens et al., 1987). These negative ionotropic effects in atria and ventricle is believed to be through the activation of A₁ adenosine receptor.

At present time it is believed that both A₁ and A_2A receptor subtypes coexist in the ventricle, whereas, the atrium contains exclusively A₁ receptor subtype. Our data provide evidence for the presence of A_2A receptor in both the ventricle as well as in atria of human and porcine. We identified the receptor subtype in these tissues by Western blot analysis to exist as a 45-kDa protein. It appears that the 45-kDa receptor is present in very high abundance in both the ventricle and the atria from both human and porcine when compared with all other tissues investigated here. The reason for such an uneven distribution of A_2A receptor in various tissues is not yet clearly understood. It can be speculated that because adenosine is produced in the cardiac myocytes and has been shown to be very active in regulating various aspects of cardiovascular functions, that it would have a very high density of the receptors localized in these tissues. However, the application of Western blotting to study the biological distribution of a certain molecule has its limitations. Western blotting will only provide information on the presence and molecular size of the immunoreactive protein(s) but provides no information on its functionality. The study indicates, by Western blotting, the presence of A_2A receptors in both atria and ventricle of human and porcine heart but provides no information on their functionality. The presence of A_2A receptors in atria is in contradiction with the earlier report (Xu et al., 1992). It is possible that the A_2A receptors are present in atria but are nonfunctional which would explain why they went undetected by functional studies.

Pharmacological studies have identified adenosine receptor in coronary artery from many species such as dog, bovine and human to be of A_2A receptor subtype (Kusachi et al., 1983; Mustafa and Askar, 1985; Ramagopal et al., 1988). Our study is in agreement with the previous reports describing the presence of A_2A receptor in human and porcine coronary artery by indirect pharmacological methods (Ramagopal et al., 1988; Sabouni et al., 1989, 1990). We present the direct evidence for the presence of adenosine A_2A receptor with an apparent molecular weight of 45 kDa in human and porcine coronary artery.

From our data we conclude that adenosine A_2A receptors are present in various cardiovascular tissues from human including atria, ventricle, coronary artery and last but not least, platelets. The receptor is present in high density in atria and ventricle, the reason for which is not yet clear. Similar results were obtained from porcine tissues (porcine platelets were not investigated in this study). This is the first report providing direct evidence of A_2A receptor localization in human cardiovascular tissues. Due to the limited availability of the human tissue, every bit of information available is crucial.