Altered purinergic signaling in uridine adenosine tetraphosphate-induced coronary relaxation in swine with metabolic derangement

We previously demonstrated that uridine adenosine tetraphosphate (Up4A) induces potent and partially endothelium-dependent relaxation in the healthy porcine coronary microvasculature. We subsequently showed that Up4A-induced porcine coronary relaxation was impaired via downregulation of P1 receptors after myocardial infarction. In view of the deleterious effect of metabolic derangement on vascular function, we hypothesized that the coronary vasodilator response to Up4A is impaired in metabolic derangement, and that the involvement of purinergic receptor subtypes and endothelium-derived vasoactive factors (EDVFs) is altered. Coronary small arteries, dissected from the apex of healthy swine and swine 6 months after induction of diabetes with streptozotocin and fed a high-fat diet, were mounted on wire myographs. Up4A (10−9–10−5 M)-induced coronary relaxation was maintained in swine with metabolic derangement compared to normal swine, despite impaired endothelium-dependent relaxation to bradykinin and despite blunted P2X7 receptor and NO-mediated vasodilator influences of Up4A. Moreover, a thromboxane-mediated vasoconstrictor influence was unmasked. In contrast, an increased Up4A-mediated vasodilator influence via P2Y1 receptors was observed, while, in response to Up4A, cytochrome P450 2C9 switched from producing vasoconstrictor to vasodilator metabolites in swine with metabolic derangement. Coronary vascular expression of A2A and P2X7 receptors as well as eNOS, as assessed with real-time PCR, was reduced in swine with metabolic derangement. In conclusion, although the overall coronary vasodilator response to Up4A was maintained in swine with metabolic derangement, the involvement of purinergic receptor subtypes and EDVF was markedly altered, revealing compensatory mechanisms among signaling pathways in Up4A-mediated coronary vasomotor influence in the early phase of metabolic derangement. Future studies are warranted to investigate the effects of severe metabolic derangement on coronary responses to Up4A.


Introduction
Diabetes mellitus is the most prevalent endocrine disorder worldwide, and diabetes mellitus and associated metabolic derangement constitute an important risk factor for development of cardiovascular disease including atherosclerosis and diabetic heart disease. The latter is the consequence not only of proximal obstructive coronary artery disease but also of coronary microvascular disease [1,2]. Endothelial dysfunction is an important determinant of altered vascular reactivity and plays a major role in the etiology of diabetes-induced macrovascular and microvascular complications [3,4]. This endothelial dysfunction encompasses an imbalance between the secretion of endothelium-derived relaxing factors (such as NO and prostacyclin) and endothelium-derived constricting factors (such as endothelin and thromboxane) [4,5].
Uridine adenosine tetraphosphate (Up 4 A) was initially identified as an endothelium-derived vasoconstrictor, exerting its constrictor influence in various vascular beds [6][7][8][9][10]. The vasoconstriction was shown to involve the generation of thromboxane [11] and reactive oxygen species [12]. A subsequent study reported that Up 4 A-mediated vascular contraction was enhanced in renal, basilar, and femoral arteries of DOCAsalt-induced hypertensive rats compared to normal rats [9]. Moreover, Up 4 A-induced contraction was increased in renal arteries from rats with type 2 diabetes through activation of the cyclooxygenase-thromboxane pathway [13], suggesting that disease states, including diabetes-associated metabolic derangement [14], may aggravat e Up 4 A-mediated vasoconstriction.
Similar to other extracellular nucleotides, Up 4 A exerts its vasomotor influence by binding to purinergic receptors [15,16]. The purinergic receptor family consists of P1 (adenosine receptors) and P2 receptor subtypes that can be further divided into P2X and P2Y receptors [17]. There is evidence from rodent models and humans that vascular purinergic signaling is altered in metabolic disorders [18]. The vasoconstrictor response to ATP, a putative P2 receptor agonist [19], was increased, while in preconstricted mesenteric arteries from rats with diabetes, the vasodilator response to ATP was decreased [20,21]. Similarly, the vasodilation to ATP, UTP, and adenosine was impaired in femoral arteries of patients with type 2 diabetes [22].
Contrary to the vasoconstrictor effect of Up 4 A in most vascular beds [6][7][8][9][10], Up 4 A acts as a potent vasodilator in the coronary microcirculation [15]. We recently demonstrated that Up 4 A-induced coronary relaxation was blunted in swine after myocardial infarction possibly via downregulation of P1 receptors [23]. In the present study, we tested the hypothesis that the coronary vasodilator response to Up 4 A is also impaired in swine with metabolic derangement (diabetes mellitus and dyslipidemia) and investigated the involvement of altered signaling through purinergic receptor subtypes and endothelium-derived vasoactive factors (EDVFs). Our findings surprisingly demonstrate that, despite the presence of endothelial dysfunction, the coronary vasodilator response to Up 4 A was maintained in metabolic derangement. However, the contribution of purinergic receptor subtypes and endotheliumderived factors was markedly altered.

Animals
Diabetes was induced in 10 female swine (2-3 months old, 25.1 ± 1.2 kg) with intravenous injections of streptozotocin (50 mg/kg/day, Bioconnect, The Netherlands, AG-CN2-0046) for three days. This dose of streptozotocin resulted in a stable hyperglycemia, as evidenced by glucose levels of 19.5 ± 1.4 (week 1), 21.6 ± 2.9 (week 3), 18.8 ± 2.0 (week 18), and 18.7 ± 2.1 mmol/l (week 20). One to two weeks after diabetes induction, a high-fat diet (25% saturated fats and 1% cholesterol) was gradually introduced [3]. Swine were housed in metabolic cages with ad libitum access to food for 1 h per meal, twice daily for the entire 6-month study duration. The diabetic status was regularly monitored by measurement of glucose and ketone levels in urine and blood. At sacrifice, swine with metabolic derangement weighed 104 ± 8 kg. Five healthy female crossbred Yorkshire × Landrace swine, purchased from the same breeder two weeks prior to sacrifice and matched for age and weight to swine with metabolic derangement (8-9 months old, 119 ± 5 kg), served as a control group.

Blood and tissue sampling
At sacrifice, animals were sedated with an intramuscular injection of Zoletil (Tiletamine/Zolazepam; 5 mg/kg) and Xylazine (2.25 mg/kg), anesthetized with pentobarbital (20 mg/kg/h i.v.) and artificially ventilated. Catheters were inserted for arterial blood sampling [24]. The collected blood samples were stored for later determination of lipids, glucose and insulin levels. Following thoracotomy, hearts were arrested and immediately excised and placed in cold, oxygenated Krebs bicarbonate buffer solution.

Myograph studies
Left ventricular apices from female swine with metabolic derangement (n = 10) and normal swine (n = 5) were collected at the time of sacrifice. Additional apices were collected from 8 hearts of healthy swine (3 female + 5 unknown gender) obtained from a local slaughterhouse. When the vascular response to Up 4 A (Biolog Life Science, Germany, U008) in the 8 female swine did not differ from the 5 swine of unknown gender from the slaughterhouse, data were pooled into a single control group (normal, n = 13), otherwise only data from coronary small arteries of female swine were used.
Coronary small arteries (diameter~150 μm) were dissected out from the apex of 7 swine with metabolic derangement and 13 normal swine and stored overnight in cold, oxygenated Krebs bicarbonate solution of the following composition (mM): NaCl 118, KCl 4.7, CaCl 2 2.5, MgSO 4 1.2, KH 2 PO 4 1.2, NaHCO 3 25, and glucose 8.3; pH 7.4 [15,25]. The next day, coronary arteries were cut into segments of~2 mm length and mounted in microvascular myographs (Danish Myo Technology) in separate 6 ml organ baths containing Krebs bicarbonate solution aerated with 95% O 2 /5% CO 2 and maintained at 37°C. Changes in contractile force were recorded with a Harvard isometric transducer. Following a 30-min stabilization period, the internal diameter was set at a tension equivalent to 0.9 times the estimated diameter at 100 mmHg effective transmural pressure [15,25]. At the end of the stabilization period, the vessels were exposed to 30 mM KCl twice to check the contractility. Endothelial integrity was verified by observing relaxation to 10 nM substance P after preconstriction with 100 nM of the stable thromboxane A 2 analogue 9,11-dideoxy-11α,9α-epoxymethanoprostaglandin F2α (U46619). Then, vessels were subjected to 100 mM KCl to determine the maximal vascular contraction. Thereafter, vessels were allowed to equilibrate in fresh Krebs solution for 30 min before initiating different experimental protocols [15,25]. In experiments where the effect of an antagonist on the response to Up 4 A was assessed, antagonists were added to the organ baths 30 min before preconstriction with U46619 and were present throughout the experiments. Only one protocol was executed per vessel segment, and, within one protocol, each vessel was obtained from a different animal.

Experimental protocols
The coronary small arteries from both normal swine and swine with metabolic derangement were subjected to increasing Up 4 A concentrations (10 −9 -10 −5 M). Since no vasoconstrictor influence was observed in vessels from both normal swine and swine with metabolic derangement (data not shown), vessels were preconstricted with U46619 to study the vasodilator effect of Up 4 A and the signaling pathways involved.
MRS2578 and indomethacin were firstly dissolved in DMSO and further diluted in distilled water. Sulfaphenazole was dissolved in ethanol and further diluted in distilled water. Other drugs were dissolved in distilled water. PPADS and MRS2159 were protected from light.

Quantitative real-time PCR analysis
Following dissection, endothelium-intact coronary small arteries were snap-frozen in liquid nitrogen to be used for detection of A 2A , P2X 1 , P2X 4 , P2X 7 , P2Y 1 , P2Y 2 , P2Y 4 , and P2Y 6 receptors mRNA. In addition, the expression of endothelial NOS (eNOS) was measured [33]. Total RNA was extracted from 5 to 7 frozen samples per group using a Qiagen RNA kit. cDNA was synthesized from 100 ng of total RNA with iScript Reverse Transcriptase (Bio-Rad). Quantitative real-time PCR (MyIQ, Bio-Rad) was performed with SYBR Green (Bio-Rad) [15]. Target gene mRNA levels were expressed relative to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an endogenous control [34]. Primer sequences are shown in Table 1.

Data analysis and statistics
Data are presented as mean ± SEM. Statistical analysis of plasma lipids, glucose and insulin was performed using an unpaired two-tailed t test. Vascular relaxation responses to Up 4 A were expressed as percentage of contraction to U46619 [15]. The effects of drug treatment on Up 4 A responses were analyzed using two-way ANOVA for repeated measures. Statistical significance was accepted when P < 0.05 (two-tailed).

Characteristics of diabetes and metabolic derangement
Metabolic derangement was present six months after induction of diabetes with streptozotocin and start of a high-fat diet. Plasma levels of glucose, total cholesterol, high density lipoprotein (HDL), low density lipoprotein (LDL), and triglycerides in swine with metabolic derangement were significantly elevated as compared to normal swine, while insulin levels were similar (Table 2). Fatty streaks were observed in the proximal coronary arteries, but plaque burden never amounted more than 10% of the vascular lumen; fatty streaks were not observed in the coronary microvessels.

Vasoactive influence of Up 4 A on coronary vasculature in swine with metabolic derangement
Up 4 A failed to elicit any vasoconstrictor response in coronary small arteries from either normal swine or swine with metabolic derangement (data not shown). Following preconstriction with U46619, Up 4 A produced concentrationdependent relaxation up to 100%, which was virtually identical between the two groups ( Fig. 1). In contrast, bradykinin (10 −10 -10 −6 M)-induced endothelium-dependent relaxation in coronary small arteries was impaired in swine with metabolic derangement (−logEC 50 8.61 ± 0.12 in normal (n = 6) vs 7.89 ± 0.19 in metabolic derangement (n = 5); P < 0.05).

Effect of metabolic derangement on function and expression of purinergic receptor subtypes
Non-selective P1 receptor blockade with 8PT and, to a lesser extent, non-selective P2 receptor blockade with PPADS attenuated the coronary vasodilator response to Up 4 A (Fig. 2). These responses were however not different between the two groups. In contrast, while the effect of combined administration of 8PT and PPADS was similar to the effect of 8PT alone in normal swine (Fig. 2a), the effect of 8PT and PPADS was significantly greater than the effect of 8PT alone in swine with metabolic derangement with unmasking of a mild vasoconstrictor response to Up 4 A (Fig. 2b).
A 2A receptor blockade resulted in attenuation of the response to Up 4 A that was comparable with the attenuation by 8PT and was not different between coronary small arteries from normal swine and swine with metabolic derangement (Fig. 3a, f), despite slightly lower A 2A mRNA expression levels in coronary small arteries of metabolic derangement (Fig. 4). P2X 1 receptor blockade with MRS2159 attenuated Up 4 A-induced relaxation to a similar extent in coronary small arteries from normal swine and swine with metabolic derangement (Fig. 3b, g), which was consistent with the unaltered P2X 1 mRNA expression between the two groups (Fig. 4). P2X 7 receptor blockade with A438079 attenuated the vasodilator response to Up 4 A in coronary small arteries from normal swine ( Fig. 3c) but not from swine with metabolic derangement (Fig. 3h), which was consistent with the reduced coronary P2X 7 mRNA expression in swine with metabolic derangement (Fig. 4). In contrast, P2Y 1 blockade with MRS2179 attenuated Up 4 A-induced relaxation only in coronary small arteries from swine with metabolic derangement (Fig. 3i) but not from normal swine (Fig. 3d) despite the unaltered mRNA expression of P2Y 1 receptors (Fig. 4). The attenuation by P2Y 6 blockade with MRS2578 of Up 4 A-induced vasorelaxation and the P2Y 6 mRNA expression were comparable between coronary small arteries of normal swine and swine with metabolic derangement (Figs. 3e, j and 4).

Role of endothelium in Up 4 A-induced coronary relaxation
Endothelial denudation, as confirmed by the absence of a vasodilator response to substance P (data not shown), attenuated the vasodilator response to Up 4 A in coronary small arteries from normal swine (Fig. 5a) but not from swine with metabolic derangement (Fig. 5d). Similarly, eNOS inhibition with LNAME attenuated the response to Up 4 A in vessels from normal swine (Fig. 5a) but not from swine with metabolic derangement (Fig. 5d), which was consistent with the reduced eNOS expression in swine with metabolic derangement, from 0.52 ± 0.14 in normal to 0.19 ± 0.03 in metabolic derangement (P < 0.05). Additional COX-inhibition with indomethacin had no effect in coronary small arteries from either normal swine (Fig. 5b) or swine with metabolic derangement (Fig. 5e). In the presence of combined eNOS and COX blockade, inhibition of CYP 2C9 with sulfaphenazole enhanced the relaxation to Up 4 A in coronary small arteries from normal swine, (Fig. 5c), whereas in vessels from swine with metabolic derangement, sulfaphenazole attenuated the vasorelaxation to Up 4 A (Fig. 5f). These findings suggest that CYP 2C9 switches from the production of vasoconstrictor metabolites in normal swine to vasodilator metabolites in swine with metabolic derangement. The thromboxane synthase inhibitor ozagrel had no effect on Up 4 A-induced relaxation in coronary small arteries from either normal swine (Fig. 6a) or swine with metabolic derangement (Fig. 6c). However, while in normal swine, in the presence of LNAME, ozagrel failed to significantly affect Up 4 A-induced relaxation of coronary small arteries (Fig. 6b, P = 0.17), ozagrel significantly enhanced the vasodilator response to Up 4 A in coronary small arteries from swine with metabolic derangement (Fig. 6d).

Discussion
The present study is the first to investigate the coronary vascular effects of the dinucleotide Up 4 A in a large animal model of metabolic derangement. The main findings were that (1)  Up 4 A-induced relaxation was not altered in coronary small arteries isolated from swine with metabolic derangement as compared to normal swine, despite the presence of endothelial dysfunction; (2) combined P1 and P2 receptor blockade had an additive effect in swine with metabolic derangement but not in normal swine; (3) there was a shift in involvement of purinergic receptors from P2X 7 receptors in normal swine to P2Y 1 receptors in swine with metabolic derangement; (4) endothelial denudation as well as eNOS inhibition attenuated the response to Up 4 A in normal swine but not in swine with metabolic derangement, and eNOS inhibition unmasked thromboxane production in swine with metabolic derangement, but not normal swine, in response to Up 4 A; and (5) CYP 2C9 metabolites exerted a vasoconstrictor influence in response to Up 4 A in normal swine but exerted a vasodilator influence on the Up 4 A responses in swine with metabolic derangement. The implications of these findings will be discussed below.
The dinucleotide Up 4 A was initially identified as a novel endothelium-derived vasoconstrictor [6]. Subsequent studies demonstrated that Up 4 A induces vascular contraction in most vascular beds, including mouse renal arterioles [35], mouse aortas [11,36,37], rat aortas [12], rat mesenteric arteries [9], rat pulmonary arteries [7], and rat renal arteries [6], particularly in the presence of diabetes [13]. In addition to its vasoconstrictor effects in most vascular beds under basal tone, Up 4 A can act as a vasodilator when vascular tone is elevated in the rat aortas and rat perfused kidney [12,38]. In contrast, we previously found that Up 4 A is purely a vasodilator in the healthy porcine coronary microcirculation, without any vasoconstrictor activity [15]. We also recently found that this vasodilator response to Up 4 A is reduced after myocardial infarction in coronary microvessels supplying the remote myocardium [23]. Surprisingly, we did not observe any blunting of the coronary vasodilator response to Up 4 A in swine with metabolic derangement as compared to normal swine in the present study (Fig. 1), despite evidence of endothelial dysfunction as indicated by the blunted endothelium-dependent relaxation to bradykinin both at 2 months in our previous study [3] as well as 6 months of metabolic derangement in the present Values are mean ± SEM. *P < 0.05 vs. control; †P < 0.05 effect of sulfa after LNAME + indo study. However, we did observe marked changes in the contribution of various purinergic receptor and endothelial mediators to the relaxation produced by Up 4 A.
In coronary small arteries from healthy swine, the vasodilator effect of Up 4 A is principally mediated through P1 receptors, although the P2X 1 and P2Y 1 receptors also contribute in vessels of approximately 150 μm in diameter [23] but not in vessels of 250-300 μm [15]. In coronary small arteries from swine with metabolic derangement, expression of the A 2A receptor was slightly reduced, but a major part of the vasodilator effect of Up 4 A was still exerted through this purinergic P1 receptor subtype. The effect of combined P1 and P2 receptor blockade was markedly enhanced in the coronary small arteries from swine with metabolic derangement, even unmasking a modest vasoconstrictor effect in response to Up 4 A, suggesting that the P2 receptor subtypes involved in response to Up 4 A may differ between the two groups. Indeed, we found that both the expression and contribution of P2X 7 receptors to Up 4 A-induced relaxation were decreased. In contrast, the contribution of P2Y 1 receptors was increased in coronary small arteries from swine with metabolic derangement, even though expression of the P2Y 1 receptors was not altered. The apparent discrepancy between receptor expression and function is likely due to the fact that receptor mRNA does not equate to receptor function. In addition, expression was studied in whole vessels so that we cannot distinguish expression in endothelial cells from vascular smooth muscle cells. P2 receptors in the vasculature are present on endothelial cells as well as on vascular smooth muscle cells. Activation of P2X 1 receptors and, in some vascular beds, P2X 2 , P2X 4 and P2Y 1 , P2Y 2 , and P2Y 6 receptors on smooth muscle cells, generally results in vasoconstriction, although vasodilation is sometimes also observed [39]. Activation of P2Y 1 , P2Y 2 and possibly also P2Y 4 , P2Y 11 , P2X 1 , P2X 2 , P2X 3 , P2X 4 , and P2X 7 receptors on endothelial cells leads to the production of nitric oxide (NO) and subsequent vasodilation [39,40]. It is therefore possible that a shift in expression of the P2Y 1 receptors from vascular smooth muscle to endothelium may have resulted in an increased contribution of these receptors to the vasodilator effect of Up 4 A. Interestingly, with age, a shift in P2Y 1 receptors from endothelial cells to vascular smooth muscle cells was observed in the rat cerebral vasculature [41]. Consistent with these findings, we observed an attenuation of Up 4 A-induced vasorelaxation by P2Y 1 -receptor blockade in our previous study in healthy swine of approximately 4 months of age [23] but not in the present study in healthy swine of approximately 9 months of age. These findings also suggest that metabolic derangement may have prevented this age-dependent shift from endothelial to smooth muscle cell expression of P2Y 1 . Future studies are needed to investigate the metabolic disorder-induced changes in purinergic receptor subtype distribution between coronary endothelial and vascular smooth muscle cells in more detail. Metabolic derangement is accompanied by microvascular dysfunction and a shift in the balance between endotheliumderived vasodilators and vasoconstrictors [4,5]. In the present study, endothelial denudation blunted Up 4 A-induced relaxation in coronary small arteries from normal swine but not swine with metabolic derangement. This effect of endothelial denudation was similar to the effect of eNOS inhibition with LNAME, indicating a reduced involvement of NO in the vasodilator response to Up 4 A in metabolic derangement. Reduced eNOS activity, which is a hallmark of coronary endothelial dysfunction in humans with metabolic derangement [42], was also observed in our previous study in coronary small arteries [3] and is consistent with the reduced eNOS expression in swine with metabolic derangement observed in the present study.
Endothelial dysfunction often results in a shift in production of endothelium-derived vasodilators, e.g., from NO to prostanoids and EDHF [43], and an increase in endotheliumderived vasoconstrictors such as thromboxane [44,45]. In the present study, we found that inhibition of COX had no effect in coronary small arteries from either normal swine or swine with metabolic derangement while the thromboxane synthase inhibitor ozagrel significantly enhanced the vasodilator response to Up 4 A in the presence of eNOS inhibition in swine with metabolic derangement but not in normal swine, consistent with thromboxane production in response to Up 4 A. However, although NO-production was reduced in the coronary microvasculature of swine with metabolic derangement, there was no effect of thromboxane synthase inhibition in coronary small arteries with active eNOS, suggesting that even in metabolic derangement, NO suppresses the production of thromboxane. An increased contribution of thromboxane to vascular contraction in response to Up 4 A was also recently shown in renal arteries from type 2 diabetic rats, although in that study the increased contribution of thromboxane appeared to be due to an increased sensitivity rather than an increased production of thromboxane [13]. However, in the present study, constriction to the stable thromboxane mimetic U46619 was not different between normal swine and swine with metabolic derangement (data not shown), indicating that sensitivity to thromboxane was not altered. A potential mechanism that has been proposed to underlie the increased thromboxane production in metabolic derangement is that lowgrade inflammation results in eNOS uncoupling, which subsequently leads to the production of reactive nitrogen species such as peroxinitrite, that is capable of inactivating PGI 2 -synthase and cause a shift in production from prostacyclin to thromboxane [44,45].
EDHF has been demonstrated to compensate for the loss of NO in metabolic disorder [46], and CYP 2C9 has been reported to generate EETs that can act as an EDHF in the coronary vasculature [47]. In addition to these vasodilator EETs, CYP 2C9 can also produce ROS [48], which act as vasoconstrictors in the porcine coronary vasculature [49]. Although we have previously shown in pre-adolescent swine that inhibition of CYP 2C9 alone [15], or in the presence of inhibition of eNOS/ COX (unpublished data), does not affect Up 4 A-induced relaxation, we found in the present study in adult swine that inhibition of CYP 2C9, in the presence of inhibition of eNOS and COX, potentiated the vasodilator response to Up 4 A in coronary small arteries from normal swine. These data are consistent with the observations that CYP 2C9 activity increases with age [50] but predominantly produces ROS in healthy porcine coronary small arteries [48]. In contrast, in coronary small arteries from swine with metabolic derangement, inhibition of CYP 2C9 in the presence of eNOS-and COXinhibition reduced the vasorelaxation in response to Up 4 A, suggesting that CYP 2C9 switches from the production of vasoconstrictor to vasodilator metabolites in metabolic derangement. Such production of vasodilator metabolites by CYP 2C9 is consistent with other studies showing that CYP 2C9 metabolites can act as an EDHF [47] and that EDHF(s) is able to compensate for the chronic loss of NO in coronary vasculature in metabolic derangement [46].
In conclusion, while the overall vasodilator response to Up 4 A was maintained in coronary small arteries of swine with metabolic derangement, the purinergic receptor subtypes as well as the EDVFs mediating this response were markedly altered. These findings suggest the presence of compensatory mechanisms among signaling pathways in Up 4 A-mediated coronary vasomotor influences in the early phase of metabolic derangement. Future studies are warranted to investigate the effects of severe metabolic derangement on coronary responses to Up 4 A.