Adenosine clearly regulates coronary blood flow (CBF); however, contributions of specific adenosine receptor (AR) subtypes (A1, A2A, A2B, A3) to CBF in swine have not been determined. ARs generally decrease (A1, A3) or increase (A2A, A2B) cyclic adenosine monophosphate, a major mediator of vasodilation. We hypothesized that A1 antagonism potentiates coronary vasodilation and coronary stent deployment in dyslipidemic Ossabaw swine elicits impaired vasodilation to adenosine that is associated with increased A1/A2A expression. The left main coronary artery was accessed with a guiding catheter allowing intracoronary infusions. After placement of a flow wire into the left circumflex coronary artery the responses to bolus infusions of adenosine were obtained. Steady-state infusion of AR-specific agents was achieved by using a small catheter fed over the flow wire in control pigs. CBF was increased by the A2-nonselective agonist 2-phenylaminoadenosine (CV1808) in a dose-dependent manner. Baseline CBF was increased by the highly A1-selective antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), but not changed by other AR-specific agents. The nonselective A2 antagonist 3,7-dimethyl-1-propargylxanthine and A2A-selective antagonist 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl)phenol (ZM241385) abolished adenosine-induced CBF, whereas A2B and A3 antagonism had no effect. Dyslipidemia and stenting decreased adenosine-induced CBF ∼70%, whereas A1, A2A, and A2B mRNA were up-regulated in dyslipidemic versus control >5-fold and there was no change in the ratio of A1/A2A protein in microvessels distal to the stent. In control Ossabaw swine A1 antagonism by DPCPX positively regulated basal CBF. Impaired adenosine-induced CBF after stenting in dyslipidemia is most likely caused by the altered balance between A1 and A2A signaling, not receptor expression.
Adenosine contributes to the maintenance of coronary blood flow (CBF), especially during myocardial ischemia (Fredholm et al., 2001; Tune et al., 2004; Duncker and Bache, 2008) and is used clinically to elicit maximal vasodilation to diagnose flow-limiting conduit artery stenosis and microvascular dysfunction (Kern et al., 2006). Four different subtypes of adenosine receptors (ARs) have been cloned and pharmacologically characterized: A1, A2A, A2B, and A3 (Fredholm et al., 2001; Shen et al., 2005). All are G protein-coupled and either inhibit (A1, A3, Gi/q) or stimulate (A2A, A2B, Gs) adenylyl cyclase (Fredholm et al., 2001). Adenosine predominantly dilates arterioles <100 μm in diameter (Camici and Crea, 2007; Duncker and Bache, 2008). The vasodilatory mechanism involves A2 activation and consequent elevation of cAMP-dependent protein kinase (PKA), resulting in vasodilation mainly via opening of ATP-sensitive K+ (KATP) channels and voltage-dependent K+ (Kv) channels (Standen and Quayle, 1998; Hein et al., 1999; Bender et al., 2009). Because A1 decreases, A2 increases cAMP (Fredholm et al., 2001) and the roles of A1 and A2 subtypes are opposing (Sato et al., 2005), A1 antagonism could positively modulate coronary vasodilation. Tawfik et al. (2006) provided compelling evidence for A1 regulation of CBF in an A1 knockout mouse, but, surprisingly, studies have not been done on A1 in large animals that more closely mimic human coronary physiology. AR may be responsible for the dysregulated adenosine-induced CBF in dyslipidemia.
Deployment of a stent in a flow-limiting coronary atherosclerotic lesion of a conduit coronary artery results in substantial improvement in functional capacity and quality of life. However, almost one-third of patients undergoing coronary stenting eventually experience recurrent angina (Abbate et al., 2007). Beyond in-stent thrombosis/restenosis that is the most common cause of poststent angina (Abbate et al., 2007), downstream microvascular dysfunction has been implicated in causing exertional ischemia (van Liebergen et al., 1998; Kern et al., 1999; Monnink et al., 2003; Werner et al., 2004; Camici and Crea, 2007). This phenomenon was termed poststent “iatrogenic coronary microvascular dysfunction” (Camici and Crea, 2007). The mechanisms underlying microvascular dysfunction associated with stenting are unclear and deserve careful study. Patients with dyslipidemia exhibit impaired coronary flow reserve (Camici and Crea, 2007), which has also been shown convincingly in dyslipidemic swine models by in vivo flow measures (Zhu et al., 2009). Because AR functions are highly specific based on AR subtype, vascular bed, and species (Shen et al., 2005; Edwards et al., 2008), the vasodilatory effect of adenosine involves A2 activation (Standen and Quayle, 1998; Hein et al., 1999; Bender et al., 2009), and A1 and A2 often mediate opposite effects of adenosine (Shen et al., 2005), the balance of A1 and A2 may contribute to impaired adenosine-induced CBF in chronic dyslipidemia after stenting.
We tested two central hypotheses in the present study: 1) A1 antagonism potentiates coronary vasodilation in healthy control Ossabaw miniature swine, and 2) coronary stent deployment in dyslipidemic Ossabaw swine elicits impaired vasodilation to adenosine, which is associated with increased A1/A2A expression in the microvessels downstream from stented conduit coronary arteries.
Materials and Methods
Animal Care and Coronary Stenting.
All protocols involving animals were approved by an Institutional Animal Care and Use Committee and complied fully with standards (National Research Council, 1996; AVMA Panel on Euthanasia, 2001). Ten lean male Ossabaw miniature swine (age = 14 ± 4 months) were fed a diet consisting of standard chow with 22% kcal from protein, 70% kcal from carbohydrates, and 8% kcal from fat. The data in Figs. 1, 2, and 3 were obtained from those pigs. Another 10 male Ossabaw swine (age = 14 months; n = 5/group) were randomized to lean control or dyslipidemic groups in the stenting study. The data in Figs. 4 and 5 were obtained from the pigs in the stenting study. Please see the details in the flow diagram of the study in Supplemental Fig. 1. The dyslipidemic pigs were fed chow supplemented with (percentage by weight): cholesterol 2.0, coconut oil 17, corn oil 2.5, and sodium cholate 0.7. This mixture yielded a diet of 13% kcal from protein, 40% kcal from carbohydrates, and 47% kcal from fat (Mokelke et al., 2005). Control and dyslipidemic groups were calorie-matched (3200 kcal/day) for 43 weeks until sacrifice, the intended experimental design to avoid development of obesity and metabolic syndrome (Sturek et al., 2007; Edwards et al., 2010). After 40 weeks on the diets, control and dyslipidemic pigs in the stenting study received a bare metal stent (2.5–4.0 mm in diameter by 8 mm in length; Express2, Boston Scientific, Minneapolis, MN) in the circumflex coronary artery at a stent/artery ratio of 1.0 with optimal inflation pressure and recovered for 3 weeks before sacrifice as described previously (Sturek et al., 2007; Edwards et al., 2008, 2010; Lloyd et al., 2008; Long et al., 2010; Neeb et al., 2010). Swine received 325 mg of aspirin as antiplatelet therapy starting the day before the stenting and continuing for 3 weeks after stenting until completion of the study. There was no evidence of microvascular embolization in all animals. Cephalexin (1000 mg) was given twice a day for 6 days after the stent procedure. All pigs were housed and fed in individual pens under a 12-h light/12-h dark cycle. Water was provided ad libitum.
Coronary Blood Flow.
The pigs were preanesthetized by using telazol/xylazine and maintained by using 2 to 4% isoflurane with oxygen via tracheal intubation and ventilation at a tidal volume of typically ∼700 ml (10–15 ml/kg) and respiratory rate of 15/min. Heart rate, blood pressure, oxygen saturation, and electrocardiographic data were continuously monitored. These precautions confirmed normal oxygen saturation and no hypercarbia or acidosis among the pigs based on blood gas and pH analysis. The right femoral artery was accessed by surgical cut-down and an 8 F vascular introducer sheath was inserted followed by administration of heparin (200 U/kg). A Cordis (Bridgewater, NJ) 8 F Amplatz L (sizes 0.75–2.0) guiding catheter was advanced into the aortic arch, and a guiding catheter was engaged with the ostium of the left main artery. A 0.014-inch diameter Doppler flow wire (JoMed Inc., Rancho Cordova, CA) or combination pressure and flow wire (ComboMap; Volcano Therapeutics, Inc., Rancho Cordova, CA) was advanced down the circumflex artery under angiographic guidance and a nonbranching section was selected for baseline and drug-induced flow measurements (Mokelke et al., 2005; Sturek et al., 2007; Long et al., 2010; Neeb et al., 2010). Anatomical landmarks were noted in case the wire moved and required repositioning. In all lean control/stent (see Fig. 4, n = 5), all dyslipidemia/stent (see Fig. 4, n = 5), lean/nonstent during A2A-, A2B-, and A3-selective antagonist administration (Fig. 2, n = 5) an intravascular ultrasound (IVUS; 3.2 F 30 MHz; Boston Scientific, Inc.) catheter was then advanced over the flow wire and positioned in the circumflex to align the IVUS imaging transducer with the flow wire tip to obtain accurate diameter measurement of the artery at the location where the flow velocity measurements were acquired (Mokelke et al., 2005). In the experiments in which IVUS was not used to determine conduit artery area, angiography was used to confirm that conduit artery diameter did not change in response to adenosine receptor modulation; nonetheless, we conservatively expressed the coronary blood flow changes as velocity, not volume flow. For constant infusion of drugs the IVUS catheter was placed with only the tip in the most proximal circumflex and drugs were infused through the IVUS catheter. Flow velocity signals were allowed to stabilize for several minutes for determination of CBF velocity responses. For the pigs with stent deployment, coronary blood flow analysis of in vivo microvascular function was done 3 weeks after stenting and in a nonbranching section of the circumflex artery proximal to the stent.
Hyperemia was induced with 3-ml bolus doses of the endothelium-independent vasodilator adenosine (0.167, 0.33, and 1 μg/kg) given via the guiding catheter into the coronary artery as described previously (Mokelke et al., 2005) and is identical to clinical catheterization assessment in humans (Kern et al., 2006). Adenosine injection was followed by 10 ml of saline flush. Saline artifact was tested in each animal and, if present (usually less than 10% peak flow induced by adenosine), was subtracted from subsequent flow measurements. Peak average peak velocity (APV), heart rate, and blood pressure were recorded for each adenosine administration. Each APV value was calculated on-line by using the automated data acquisition system as an average of instantaneous CBF velocity over two consecutive cardiac cycles. Subsequent doses of adenosine were administered when APV, heart rate, and blood pressure had returned to baseline and stabilized, at which time the baseline parameters were again documented followed by the administration of the next adenosine dose. The effect of adenosine typically lasted 15 to 30 s.
The adenosine A2-nonselective agonist 2-phenylaminoadenosine (CV1808) (10−6, 10−5, and 10−4 M) was administered via bolus injection as was done with adenosine. The IVUS catheter was placed in the proximal circumflex coronary artery in control swine for steady-state infusion of selected drugs at a rate of 1 ml/min to observe their effects over a relatively longer period of time: 2-chloro-N(6)-cyclopentyladenosine (CCPA), 10−5 M, A1-selective agonist, Tocris Bioscience, Ellisville, MO; 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), 10−4 M, A1-selective antagonist, Tocris Bioscience; 3,7-dimethyl-1-propargylxanthine (DMPX), 10−4 M, A2-nonselective antagonist, Sigma-Aldrich, St. Louis, MO; 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl)phenol (ZM241385), 10−7 M, A2A-selective antagonist, Tocris Bioscience; N-(4-acetylphenyl)-2-[4-(2,3,6,7-tetrahydro-2,6-dioxo-1,3-dipropyl-1H-purin-8-yl)phenoxy]acetamide (MRS1706), 10−7 M, A2B-selective antagonist, Tocris Bioscience; N-[9-chloro-2-(2-furanyl)[1,2,4]-triazolo[1,5-c]quinazolin-5-yl] benzene acetamide (MRS1220), 10−7 M, A3-selective antagonist, Tocris Bioscience. This was followed by concurrent bolus injection of adenosine. Vehicle control (1% dimethyl sulfoxide) for DMPX was not used in parallel, which should be used for future experiments. The vehicle controls for other reagents were either water or phosphate-buffered saline. Continuous saline infusion was done between each constant infusion treatment to make sure the flow returned to baseline. The above-mentioned drug infusions were done in random order in each pig.
The analog Doppler signals were continuously digitized both as instantaneous CBF velocity and APV values. All flow data were stored on videotapes and personal computers for additional off-line analysis. Data are also expressed as percentage of APV increase, which equals [peak APV − base APV]/base APV × 100. Volumetric CBF was calculated as: CBF (in ml/min) = (artery cross-sectional area in cm2) × (velocity in cm/s) × 0.5 × 60 s/min (Mokelke et al., 2005). Vessel cross-sectional area was calculated off-line from recorded IVUS images by using the commercially available Sonos Intravascular Imaging software package (Hewlett Packard, Palo Alto, CA). Angiography and IVUS showed no change in conduit artery diameter during any drug exposures, thus coronary flow velocity represented microvascular effects.
Plasma Lipid Assays.
Venous blood samples were obtained after overnight fasting and were analyzed for triglyceride and total cholesterol [fractionated into high-density lipoprotein (HDL) and low-density lipoprotein (LDL) components] (Edwards et al., 2010).
Real-Time Reverse Transcription-Polymerase Chain Reaction.
RNA was extracted from coronary microvessels downstream of left circumflex coronary artery by using TRIzol (Invitrogen, Carlsbad, CA), then treated with DNase (DNA free; Ambion, Austin, TX) to remove contaminating genomic DNA, and analyzed by NanoDrop spectrophotometer (ND-1000; Thermo Fisher Scientific, Wilmington, DE) to assess purity (A260/280) and concentration. RNA (1 μg) was converted to cDNA by using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA) in a standard thermocycler (DNA Engine PTC-200; MJ Research, Watertown, MA). A1, A2A, A2B, and A3 mRNA expression was assessed by using quantitative real-time reverse transcription-PCR in an ABI 7500 instrument (Applied Biosystems, Foster City, CA). Primers for PCR amplification were as follows: A1, forward, 5′-GGCCATGCTGGCAATTG-3′, reverse, 5′-CCTGAGCGGGATCTTGACA-3′; A2A, forward, 5′-CCCCTTCATCTATGCCTACCG-3′, reverse, 5′-CATTCCCTCACACTCCCTCCAC-3′; A2B, forward, 5′- TGTGCTGGCTGCCTCTTCAC-3′, reverse, 5′-ACACGATGGGGTTGACGACC-3′; A3, forward, 5′-GAACCTCACCTTCCTTTCCTGC-3′, reverse, 5′-GAACTCCCGTCCATAAAATGC-3′ (Integrated DNA Technologies, Coralville, IA). Taqman probes were used to detect A1 gene amplification and were as follows: 5′-/6-FAM/TCGACCGCTACCTCC/MGBNFQ/-3′ (Applied Biosystems). SYBR Green Universal PCR Master Mix (Applied Biosystems) was used to detect A2A, A2B, and A3 gene expression followed by a dissociation curve to rule out nonspecific amplifications and primer dimers. All results were normalized to the signal of 18s RNA, which was amplified in separate reactions by using TaqMan Universal PCR Master Mix and TaqMan 18s Detection Reagents (Applied Biosystems).
Immunohistochemistry for AR.
Left ventricular myocardium was preserved in 10% formalin solution, embedded in paraffin, sectioned, and treated with antibodies similar to our previous methods (Turk et al., 2003). Primary polyclonal antibodies to A1 and A2A were applied and incubated overnight (1:100 dilution; Novus Biologicals Inc., Littleton, CO). A biotinylated secondary antibody (Vectastain Elite; Vector Laboratories, Burlingame, CA) was used for the detection process. Quantification of AR staining in each microvessel was accomplished with commercially available software (Image Pro Plus, 4.1; MediaCybernetics Inc., Bethesda, MD).
Data are presented as means ± S.E.M. Statistical analysis was performed with commercially available software [(SPSS version 12 (SPSS Inc., Chicago, IL) and Prism version 4.0 (GraphPad Software Inc., San Diego, CA)]. Means of more than two groups were compared by one-way analysis of variance with least significant difference post hoc analyses (by SPSS) or two-way analysis of variance followed by Bonferroni post-test (by Prism). Means of two groups were compared by Student's t test (unpaired, two-tailed). In all cases, p < 0.05 was considered statistically significant.
Metabolic Characteristics of Control and Dyslipidemic Pigs with Coronary Stent Deployment.
The atherogenic diet resulted in significantly higher plasma total cholesterol, LDL, HDL, and LDL/HDL versus control pigs fed standard chow diet (Table 1). There were no differences in body weight, plasma triglycerides, fasting plasma insulin, fasting plasma glucose (Table 1), or resting blood pressure (Table 2). The plasma lipid profile defined the atherogenic diet-fed group as dyslipidemic, but not having metabolic syndrome.
Contribution of A2 to Coronary Blood Flow Regulation.
The A2-nonselective antagonist DMPX did not change base APV response (Fig. 1A), but abolished peak APV response (Fig. 1B) to adenosine at 0.33 and 1.0 μg/kg. The A2-nonselective agonist CV1808 elevated the percentage of APV increase ([peak APV − base APV] / base APV × 100) in a dose-dependent manner in the absence of exogenous adenosine (Fig. 1C). The CV1808-induced percentage of APV increase at 10−4 M was comparable with adenosine-induced percentage of APV increase at 1.0 μg/kg, which was calculated from the base and peak APV in Fig. 1, A and B separately.
Contribution of A2A, A2B, and A3 to Coronary Blood Flow Regulation.
Figure 2 displays data as absolute blood flow in pigs with conduit diameter accurately measured by using IVUS before flow measures. Bolus injection of adenosine in the presence of vehicle control increased peak APV versus base in a dose-dependent manner (Table 3). Continuous intracoronary infusion of the A2A-selective antagonist ZM241385 abolished the adenosine-induced peak CBF (Fig. 2A), while having no effect on baseline CBF (data not shown). The A2B-selective antagonist MRS1706 (Fig. 2A) and A3-selective antagonist MRS1220 (Fig. 2B) had modest effects on the adenosine-induced peak CBF compared with ZM241385 and no effect on base CBF (data not shown).
Contribution of A1 to Coronary Blood Flow.
Constant infusion of the highly A1-selective antagonist DPCPX increased base APV without exogenous adenosine and increased peak APV response to adenosine at 1.0 μg/kg (Fig. 3). Intracoronary infusion of the highly A1-selective agonist CCPA did not alter base APV in the absence of exogenous adenosine or peak APV response to adenosine (Fig. 3). There was no change of systemic parameters (i.e., blood pressure, heart rate) by the administration of AR agonists/antagonists in the nonstented lean control pigs (data not shown; Mokelke et al., 2005; Long et al., 2010).
Comparison of Coronary Blood Flow in Control and Dyslipidemic Pigs Undergoing Stenting.
There were no significant differences in hemodynamic characteristics across groups (Table 2), e.g., heart rate, mean arterial pressure, rate pressure product, and baseline CBF. The CBF (Fig. 4) response to adenosine infusion at 0.167 and 1.0 μg/kg was blunted in dyslipidemic versus control pigs 3 weeks after stenting.
Expression of AR in Coronary Microvessels of Control and Dyslipidemic Pigs 3 Weeks after the Stent Deployment.
A1, A2A, and A2B mRNA expression was up-regulated >5-fold in coronary microvessels downstream from the stented left circumflex coronary artery in dyslipidemic pigs compared with control pigs (Fig. 5, A–C). A3 mRNA level was not different between groups (Fig. 5D). Immunohistochemistry showed that although there was a trend for an increase in A1/A2A protein expression in microvessels of hyperlipidemic pigs (1.12 ± 0.04) versus lean control pigs (1.0 ± 0.05), the difference was not significant (P > 0.05).
We present several key findings on adenosine in regulation of CBF. First, adenosine-induced vasodilation was mediated almost exclusively by A2A in Ossabaw swine. Second, A1 antagonism augments vasodilatory effects of some vasodilators other than adenosine in porcine coronary microvessels under basal conditions in vivo. Third, bare metal stent deployment in coronary conduit arteries in dyslipidemic Ossabaw swine elicited microvascular dysfunction, which was associated with increased A1 and, paradoxically, increased A2A and A2B mRNA expression. Collectively, these data emphasize the delicate balance between A1 and A2A signaling in control of CBF in the setting of dyslipidemia and stenting.
Adenosine was shown as a vasodilator in porcine coronary microcirculation via activation of A2A (Hein and Kuo, 1999). The A2B subtype was suggested to mediate vasodilation of adenosine in human small coronary arteries (Kemp and Cocks, 1999), rat coronary circulation (Hinschen et al., 2003), and murine heart (Talukder et al., 2003). A3 either inhibited or negatively modulated the vasodilatory effect of adenosine in mouse coronary circulation (Talukder et al., 2002). Studies in pigs (Wang et al., 2005) and humans (Sato et al., 2005) demonstrated A1 mRNA and protein expression in coronary arterioles. In our current study, the nonselective A2 antagonist DMPX did not change base APV (Fig. 1A); therefore A2 were not tonically activated under normal, basal conditions. The A2A-selective antagonist ZM241385 decreased peak CBF induced by exogenous adenosine (Fig. 2A), whereas the A2B-selective antagonist MRS1706 and A3-selective antagonist MRS1220 showed relatively little effect on CBF (Fig. 2), thus providing evidence that mainly A2A mediates vasodilatory effects of exogenous adenosine in coronary microcirculation of Ossabaw swine.
The highly A1-selective antagonist DPCPX increased base APV without exogenous adenosine (Fig. 3), whereas A2 were not tonically activated, indicating A1 antagonism by DPCPX might potentiate vasodilatory effects of some physiologically active vasodilators other than adenosine. In contrast, the highly A1-selective agonist CCPA did not change APV with or without exogenous adenosine. There are several explanations for the phenomena. First, CCPA effects could be masked by endogenous adenosine. Because the binding affinity of adenosine for A1 is higher than for A2, and the desensitization of A1 is much slower than A2 (t1/2 = 10 h versus 20 min) (Stiles, 1992), it could be that tonically A1 was activated by adenosine, whereas A2 was not. Therefore, endogenous adenosine activation of A1 could shadow exogenous CCPA activation of A1. Second, we used 10−5 M CCPA for steady-state intracoronary infusion (1 ml/min), which was diluted by continuous coronary blood flow (∼60 ml/min). Therefore the actual intracoronary concentration of CCPA might not be effective enough to regulate CBF. Research done in isolated hearts from A1 knockout and wild-type mice showed that lower concentrations (10−9 to 10−7 M) of CCPA did not significantly change CBF (Tawfik et al., 2006). Intracoronary infusion of CCPA at different concentrations may help to address that in the future. Third, because A2 were not tonically activated, adenosine might not play an important role in the maintenance of CBF under basal conditions, which was echoed by numerous studies done in normal hearts of dogs, swine, or humans under basal conditions or exercise hyperemia (Tune et al., 2002, 2004; Duncker and Bache, 2008). However, adenosine played an important role in coronary vasodilation when the myocardium was ischemic or hypoxic, which is accompanied by abundant adenosine release (Tune et al., 2002, 2004; Sato et al., 2005; Duncker and Bache, 2008).
Compared with other animal models Ossabaw swine are predisposed to metabolic syndrome (Sturek et al., 2007; Neeb et al., 2010) and have substantial diffuse and focal, flow-limiting coronary atherosclerosis (Edwards et al., 2010); therefore, stenting in the Ossabaw swine is quite clinically relevant. Dyslipidemia in the present study was manifested as described previously (Sturek et al., 2007; Bender et al., 2009; Edwards et al., 2010). The reason metabolic syndrome or obesity was avoided in the animals was to reduce the factors that may complicate the results, thus allowing the focus on dyslipidemia. It is noteworthy that A1, A2A, and A2B mRNAs all were up-regulated in coronary microvessels distal to the stent in dyslipidemic swine versus control (Fig. 5). Why was the adenosine-induced CBF response attenuated with A2A and A2B mRNA up-regulation in dyslipidemic swine? There are several explanations. First, mRNA does not always correspond directly to protein, as shown by the decrease in porcine coronary arteriolar A2B protein expression in metabolic syndrome (Bender et al., 2009). Second, A2A and A2B proteins may not function normally. The similar coronary microvascular A1/A2A protein expression determined by immunohistochemistry in control versus dyslipidemic pigs suggests an altered balance between A1 and A2A signaling. The muted response to adenosine challenge is much more likely to be the result of signaling such as altered receptor coupling (e.g., via G proteins), cAMP formation, or other downstream messengers. Future studies should address downstream signal transduction pathways. Third, possible elevation of adenosine in dyslipidemia, stenting, and ischemia could result in high-affinity binding and activation of the A1 receptor that overwhelms the vasodilatory effect mediated by A2A activation. This is feasible because the affinity of adenosine for A1 is higher than A2 and the desensitization of A1 is much slower than A2 (t1/2 = 10 h versus 20 min) (Stiles, 1992). The net result is greater coupling of adenosine to A1 than to A2 to regulate CBF. Fourth, analogous to other disease states, vasodilator signaling pathways can be up-regulated to partially compensate for increased vasoconstrictor or decreases in other vasodilator signaling pathways. For example, Ca2+-dependent K+ channel function is up-regulated in hypertension, but does not fully compensate for vasoconstrictor influences (Liu et al., 1995).
Adenosine has been found to increase postischemic recovery of function in humans in clinical interventions, such as angioplasty (Mubagwa and Flameng, 2001). These effects may be largely caused by cardiac muscle A1. Furthermore, the modulation/attenuation of heart rate (especially β-adrenergic stimulated) is also a classic finding (Belardinelli et al., 1989). There was no change of systemic parameters (i.e., blood pressure, heart rate) by the administration of AR agonists/antagonists (Mokelke et al., 2005; Long et al., 2010); therefore, under the conditions used for our flow measures (anesthetized, near resting heart rate and blood pressure) there was no direct effect of adenosine on cardiac muscle to change cardiac metabolism and, in turn, coronary blood flow. These data do not entirely rule out a direct inhibitory effect of adenosine on cardiac myocytes at higher heart rates elicited by β-adrenergic stimulation and the possibility of different adenosine reduction of heart rate in lean control versus dyslipidemic groups. Future studies should involve adrenergic provocation challenge with adenosine administration in the dyslipidemic pigs.
To our knowledge, this is one of the few studies examining the effects of stenting on coronary microcirculation in a porcine model of dyslipidemia (Long et al., 2010). Because isoflurane was used in each group of animals during the procedures, its activation of coronary KATP channels will elicit more baseline coronary vasodilation perhaps compared with other anesthetics (Kersten et al., 1997); however, the differences between groups should be caused more by the dyslipidemia and stent treatment than isoflurane administration. Stenting and dyslipidemia are two possible causes for the dysregulated adenosine-induced CBF in dyslipidemic Ossabaw swine 3 weeks after stent deployment (Fig. 4). Despite the findings that adenosine receptor-mediated coronary flow was impaired in dyslipidemic patients and patients with metabolic syndrome (Camici and Crea, 2007), as well as in dyslipidemic swine models (Zhu et al., 2009), previous study in our laboratory showed that diabetic dyslipidemic Yucatan pigs (Mokelke et al., 2005) and dyslipidemic Ossabaw pigs (Bender et al., 2009; Neeb et al., 2010) did not exhibit impaired microvascular dilation to adenosine. Coronary stenting has been shown to mechanically damage vascular cells in the target conduit artery segment and endothelium in peristent segments (Kim et al., 2009; Long et al., 2010) and induce downstream microvascular dysfunction (van Liebergen et al., 1998; Kern et al., 1999; Monnink et al., 2003; Werner et al., 2004; Camici and Crea, 2007). However, none of those stent studies were done in a swine model of dyslipidemia prone to metabolic syndrome and the microcirculatory dysfunction was either attributable to endothelium (Monnink et al., 2003) or unclarified (van Liebergen et al., 1998; Kern et al., 1999; Werner et al., 2004). Further studies are needed to clarify 1) whether dyslipidemia or stenting alone or the combination is the cause for the impaired adenosine-induced flow in the setting of dyslipidemia and stenting, 2) which adenosine receptor downstream signaling events are responsible for the adenosine-induced coronary flow dysregulation in dyslipidemia and stenting, and 3) whether the classic adenosine attenuation of β-adrenergic receptor-induced increases in heart rate affects coronary blood flow differently in control versus dyslipidemic pigs.
We conclude that A1 antagonism by DPCPX promotes vasodilation in “healthy” anesthetized Ossabaw swine. An imbalance of A1/A2A signaling in coronary microcirculation distal to a stent was associated with dysregulated adenosine-induced CBF in dyslipidemia. The use of novel adenosine analogs for vasodilation in disease deserves further pharmacological and mechanistic study.
We thank James P. Byrd and James W. Wenzel for excellent technical assistance.
This work was supported by the National Institutes of Health National Center for Research Resources [Grant RR013223] (to M.S.), the National Institutes of Health National Heart, Lung, and Blood Institute [Grant HL062552] (to M.S.); Boston Scientific, Inc.; Research Animal Angiography Laboratory of Indiana Center for Vascular Biology and Medicine; Purdue-Indiana University Comparative Medicine Program; Fortune-Fry Ultrasound Research Fund; an Indiana University School of Medicine Translational Research Fellowship (to J.M.E. and Z.P.N.); National Institutes of Health Translational Research Fellowship [Grant UL1-RR025761] (to Z.P.N.); and an American Heart Association Predoctoral Fellowship (to X.L.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- coronary blood flow
- adenosine A1 receptor
- adenosine A2A receptor
- adenosine A2B receptor
- adenosine A3 receptor
- adenosine receptor
- intravascular ultrasound
- average peak velocity
- high-density lipoprotein
- low-density lipoprotein
- polymerase chain reaction
- N-[9-chloro-2-(2-furanyl)[1,2,4]-triazolo[1,5-c]quinazolin-5-yl]benzene acetamide
- Received May 27, 2010.
- Accepted September 16, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics