Learn more: PMC Disclaimer | PMC Copyright Notice
Mechanisms underlying uridine adenosine tetraphosphate-induced vascular contraction in mouse aorta: role of thromboxane and purinergic receptors
Abstract
Uridine adenosine tetraphosphate (Up4A), a novel endothelium-derived vasoactive agent, is proposed to play a role in cardiovascular disorders and induces aortic contraction through activation of cyclooxygenases (COX). We and others demonstrated that activation of A1 or A3 adenosine receptors (AR) results in vascular contraction via thromboxane (TX) A2 production. However, the mechanisms of Up4A-induced vascular contraction in mouse aorta are not understood. We hypothesize that Up4A-induced aortic contraction is through COX-derived TXA2 production, which requires activation of A1 and/or A3AR. Concentration responses to Up4A were conducted in isolated aorta. The TXB2 production, a metabolite of TXA2, was also measured. Up4A (10−9–10−5 M) produced a concentration-dependent contraction >70%, which was markedly attenuated by COX and COX1 but not by COX2 inhibition. Notably, Up4A-induced aortic contraction was blunted by both TX synthase inhibitor ozagrel and TXA2 receptor (TP) antagonist SQ29548. Surprisingly, A3AR deletion had no effect on Up4A-induced contraction. Moreover, A1AR deletion or antagonism as well as A1/A3AR deletion potentiated Up4A-induced aortic contraction, suggesting a vasodilator influence of A1AR. In contrast, non-selective purinergic P2 receptor antagonist PPADS significantly blunted Up4A-induced aortic contraction to a similar extent as selective P2X1R antagonist MRS2159, the latter of which was further reduced by addition of ozagrel. Endothelial denudation almost fully attenuated Up4A-induced contraction. Furthermore, Up4A (3 µM) increased TXB2 formation, which was inhibited by either MRS2159 or ozagrel. In conclusion, Up4A-induced aortic contraction depends on activation of TX synthase and TP, which partially requires the activation of P2X1R but not A1 or A3 AR through an endothelium-dependent mechanism.
Graphical Abstract
1. Introduction
The endothelium releases a variety of vasodilators, such as nitric oxide (NO) and prostacyclin, and vasoconstrictors, such as endothelin and thromboxane (TX) A2 (1, 2). A novel endothelium-derived vasoactive factor uridine adenosine tetraphosphate (Up4A) has been initially identified as a vasoconstrictor in isolated kidney of rats (3). Up4A is the first dinucleotide found in living organisms that contains both purine and pyrimidine moieties and therefore can exert its vasoconstrictor effects through both the purinergic P1 (also known as ARs) and P2 receptors (P2R) (3, 4). Indeed, several in vitro studies have indicated that Up4A induces vascular contraction through purinergic receptors. Thus, vasoconstriction was observed in rat renal artery through P2X1R (3), in rat aorta through P1 and P2XR (5), and in rat pulmonary arteries through P2YR (6). Furthermore, vasoconstriction was also observed in mouse renal arterioles (7), mouse aorta (8) and rat mesenteric arteries (9), although Up4A has been subsequently found to be able to exert a vasodilator effect in rat aorta (5), isolated perfused rat kidney (10) and porcine coronary arteries (11, 12) as well as to induce hypotension in conscious rats (8). The observations that plasma concentrations of Up4A detected in juvenile hypertensive subjects are elevated (13), as well as intra-aortic injection of Up4A increases mean arterial blood pressure in intact animal (3), suggest a role for Up4A in the pathogenesis of hypertension. At post-receptor levels, Up4A-induced vascular contraction was markedly attenuated by cyclooxygenase (COX) inhibition in mouse aorta (8), implying a modulatory role of vasoconstrictor prostanoids in Up4A-mediated vascular tone.
TXA2, one of the vasoconstrictor prostanoids, is produced from arachidonic acid by COX pathway (1, 14). Subsequently, COX converts arachidonic acid into intermediate endoperoxide, the latter of which is catalyzed by TX synthase eventually leading to the production of TXA2 (1). By activating TXA2 receptors (TP), TXA2 has been reported to exert a vasoconstrictor effect in various vascular beds, thereby potentially contributing to the development of cardiovascular diseases such as hypertension (1). Previous studies have shown that TXA2 is produced and TP is activated by activation of P2XR and P2YR, which subsequently leads to vasoconstriction in canine basilar arteries (15), human umbilical and chorionic vessels (16) as well as rat aorta (17). In addition to P2R, involvement of A1AR has been observed in TXA2-mediated vasoconstriction in feline pulmonary circulation (18). Recently, we demonstrated that activation of A3AR results in TXA2 production, which leads to vascular contraction in mouse aorta (19). All these observations indicate that TXA2 can be generated via activation of purinergic receptors resulting in vascular contraction. More importantly, the observations that Up4A activates TP resulting in renal vasoconstriction and Up4A increases TXA2 production in renal arteries of rats (20), suggest a link between Up4Amediated vascular contraction and the involvement of TXA2. Although Up4A-induced vasoconstriction in mouse aorta was attenuated by COX inhibition (8), the underlying mechanisms remain obscure.
Consequently, with particular focus on mouse aorta, we hypothesized that Up4A-induced vascular contraction in mouse aorta is through COX-derived TXA2 production, which requires activation of A1AR and/or A3AR. Specifically, we aimed to firstly investigate whether TXA2 is produced and/or TP is activated in response to Up4A that contributes to vascular contraction in mouse aorta, and secondly, to determine the A1AR and/or A3AR involvement in Up4A-induced aortic contraction. Our findings from this study indicate that neither A1AR nor A3AR contributes to Up4A-induced aortic contraction. Since activation of P2R, particularly P2X1R, has been shown to be a vasoconstrictor purinergic receptor contributing to Up4A-mediated vasoconstriction (3, 21), the third aim of our study was to further explore the involvement of P2R, particularly P2X1R, in Up4A-induced vascular contraction in mouse aorta.
2. Materials and Methods
2.1 Drugs and solutions
Acetylcholine, 8-Cyclopentyl-1,3-dipropylxanthine (DPCPX), indomethacin, MRS2159, N-[2-(Cyclohexyloxy)-4-nitrophenyl]-methanesulfonamide (NS398), (E)-3-[4-(Imidazol-1-ylmethyl)phenyl]propenoic acid hydrochloride hydrate (ozagrel), phenylephrine, pyridoxalphosphate-6-azophenyl-2',4'-disulfonic acid (PPADS) and 5-(4-Chlorophenyl)-1-(4-methoxyphenyl)-3-trifluoromethyl pyrazole (SC560) were all purchased from Sigma-Aldrich (St. Louis, MO, USA). Up4A was obtained from Biolog Life Science (Bremen, Germany). [1S-[1α,2α(Z),3α,4α]]-7-[3-[[2-[(phenylamino)carbonyl]hydrazino]methyl]-7-xabicyclo[2.2.1] hept-2-yl]-5-heptenoic acid (SQ29548) and thromboxane B2 enzyme immunoassay (EIK) kit were purchased from Cayman Chemical (Ann Arbor, MI, USA). Indomethacin, DPCPX, NS398, SC560 and SQ29548 were firstly dissolved in DMSO. All subsequent dilutions (at least 1000 fold) and other drugs were obtained with distilled water. PPADS and MRS2159 were protected from light.
2.2 Animals
All experimental protocols were approved by the Institutional Animal Care and Use Committee at School of Medicine, West Virginia University. Wild type (WT) mice (C57BL/6 background) were purchased from The Jackson Laboratory (Bar Harbor, ME). A1AR knockout (KO), A3AR KO and A1/A3AR double KO (DKO) mice were obtained from Dr. Stephen Tilley (University of North Carolina, Chapel Hill, NC). A1AR and A3AR KO mice, both backcrossed 12 generations to the WT, were bred to generate A1/A3AR double heterozygotes. Double heterozygotes were intercrossed, and 1/16th of the offspring were A1/A3AR DKO. A1/A3AR DKO breeding pairs were then established. Mice were caged in a 12:12-h light-dark cycles with free access to standard chow and water. Mice with age from 14 to 16 wk of either sex were used in this study.
2.3 Tissue preparation and isometric force measurement
Mice were euthanized by anesthesia with pentobarbital sodium (65 mg/kg i.p.) followed by thoracotomy and removal of aorta. The aorta was cleaned by removing fat and connective tissues that was then cut transversely into 3 to 4 mm rings as described previously (19, 22). In a subset of rings, the endothelium was removed mechanically with a piece of thin wire by gently rolling it back and forward. Subsequently, the aortic rings were mounted vertically between two wire hooks and then suspended in 10 ml organ baths containing Krebs-Henseleit buffer. The Krebs-Henseleit buffer (pH 7.4) containing (in mM) 118 NaCl, 4.8 KCl, 1.2 MgSO4, 1.2 KH2PO2, 25 NaHCO3, 11 glucose and 2.5 CaCl2 was maintained at 37°C aerated with 95% O2/5% CO2. For measurement of isometric force response (with fixed range precision force transducer, TSD, 125 C; Biopac system), aortic rings were equilibrated for 60 min with a resting force of 1 g (22, 23) and change of the Krebs solution at a 15 min interval. At the end of equilibration period, tissues were exposed to 50 mM KCl twice to check the contractility of individual aortic rings. Endothelial integrity was verified by observing dilation to 1 µM acetylcholine after preconstriction with 1 µM phenylephrine (22). The non-denuded rings that respond appropriately (relaxation >50%) and denuded rings that did not respond to acetylcholine were used in this study. Thereafter, rings were allowed to equilibrate in fresh organ bath fluid for 30 min before initiating different experimental protocols. Since previous study (8) as well as our current observation showed a very mild and transient contraction by Up4A in aortic rings at basal condition (data not shown), all the experiments were performed upon a steady contraction by 1 µM phenylephrine. The aortic rings that did not elicit reproducible and stable contraction with phenylephrine were excluded from the study. In experiments where the effect of an antagonist was measured, it was added 30 min before preconstriction with phenylephrine and was present throughout the experiments. Only one protocol was executed per ring and within one protocol all rings were obtained from different animals.
2.4 Organ bath experimental protocols
Preconstricted aortic rings were subjected to Up4A concentration responses (10−9–10−5 M). In accordance with previous studies (8), Up4A, at a dose of 10−5 M, produced vasoconstriction followed by vasodilation in mouse aorta, indicating a biphasic effect of Up4A. Since the aim of the present study was to elucidate the mechanisms of Up4A-induced contraction in mouse aorta, the observation that Up4A-induced vasodilation at a dose of 10−5 M was excluded.
Up4A-induced aortic contraction was observed to be attenuated by COX inhibition (8), to investigate which COX subtype was involved in this process, rings were subjected to Up4A (10−9–10−5 M) in the absence and presence of non-selective COX inhibitor indomethacin (10 µM) (19), selective COX1 inhibitor SC560 (10 nM) (19) and selective COX2 inhibitor NS398 (1 µM) (19). To test whether the involvement of vasoconstrictor prostanoids was through TXA2 production or TP activation, Up4A concentration responses were performed in aortic rings in the absence and presence of TX synthase inhibitor ozagrel (10 µM) (24) and TP antagonist SQ29548 (1 µM) (19), respectively.
Since both A1AR and A3AR have been shown to be involved in TXA2-mediated vascular contraction (18, 19), aortic rings from WT, A1AR KO, A3AR KO and A1/A3AR DKO mice were exposed to Up4A concentration responses (10−9–10−5 M). In addition, aortic rings of WT were subjected to Up4A in the absence and presence of selective A1AR antagonist DPCPX (10 nM) (25). To investigate the involvement of P2R in Up4A-induced vascular contraction, aortic rings of WT were exposed to Up4A in the absence and presence of non-selective P2R antagonist PPADS (10 µM) (12) and selective P2X1R antagonist MRS2159 (30 µM) (26), respectively.
To assess whether TXA2 production requires activation of P2X1R, aortic rings were treated with Up4A in the absence and presence of MRS2159, ozagrel and the combination of MRS2159 and ozagrel.
To investigate the possible endothelial involvement of Up4A-induced aortic contraction, Up4A concentration responses were conducted in endothelium –non-denuded and –denuded rings. To identify the compartment in which P2X1R locates to generate TXA2 in response to Up4A, denuded aortic rings were exposed to Up4A in the absence and presence of MRS2179 and ozagrel, respectively. The various antagonists used in the present study do not interfere with 1 µM phenylephrine response within each figure panels and there is no significant difference in 1 µM phenylephrine response between WT and KO mice (data not shown).
2.5 Measurement of TXA2 release
TXB2, a metabolite of TXA2, was measured according to protocols from previous studies (20, 27), with slight modification. Briefly, cleaned mouse aortic rings were incubated at 37°C for 60 min in a siliconized tube containing 1 ml of Krebs-Henseleit buffer, as described for organ bath studies. After incubation, rings were treated with MRS2159 (30 µM) or ozagrel (10 µM), and incubated for another 30 min in 500 µl Krebs buffer. Thereafter, Up4A (3 µM) or vehicle (water) was added for 10 min. Then the aortic rings were removed and the tubes were freeze-clamped in liquid nitrogen and stored at −80°C for subsequent analysis. TXB2 was measured using an EIA kit (Cayman Chemicals, Ann Arbor, MI, USA). The undiluted 50 µl samples were used. The various assays were performed as described in the manufacturer’s booklet. The amount of TXB2 is expressed in picograms per milligram wet weight of the aortic rings.
2.6 Data analysis and statistics
Vascular contraction responses to Up4A were expressed as percentage of contraction to phenylephrine. The effects of drug treatment on the Up4A concentration responses were assessed using two-way ANOVA for repeated measures. The effects of drug treatment on Up4A-induced TXA2 production were analyzed using one-way ANOVA. Statistical significance was accepted when P < 0.05 (two-tailed). Data are presented as means ± SEM.
3. Results
3.1 Involvement of COX1 in Up4A-induced aortic contraction
In preconstricted aortic rings, Up4A produced a concentration-dependent contraction over 70% (Fig. 1). In accordance with previous studies (8), Up4A-induced aortic contraction was markedly attenuated by non-selective COX inhibitor indomethacin (Fig. 1A). Furthermore, Up4A-induced aortic contraction was significantly attenuated by selective COX1 inhibitor SC560 (Fig. 1B), but not by selective COX2 inhibitor NS398 (Fig. 1C). These observations indicate that Up4A-induced contraction in mouse aorta is attributed to vasoconstrictor prostanoids produced by COX1 activation.
Effects of non-selective cyclooxygenase (COX), selective COX1 and COX2 inhibition on Up4A-induced vasoconstriction. Shown are the concentration-responses to Up4A in mouse aorta from wild-type (WT) mice in the absence and presence of 10 µM of the non-selective COX inhibitor indomethacin (Indo) (n=13 for both Con and Indo; panel A), 10 nM of the selective COX1 inhibitor SC560 (n=11 for Con; n=12 for SC560; panel B), and 1 µM of the selective COX2 inhibitor NS398 (n=11 for Con; n=10 for NS398; panel C). The aortic rings in different protocols within panels are paired; the Up4A concentration responses for each control are similar to those within figures or in other figures. Values are means ± SEM. * P < 0.05, effect of drug vs. control (Con). * represents the concentration-response relation by two-way ANOVA.
3.2 Involvement of TXA2 in Up4A-induced aortic contraction
To investigate whether the involvement of vasoconstrictor prostanoids in Up4A-induced aortic contraction is TXA2, we performed Up4A concentration responses in isolated aortic rings in the absence and presence of either TX synthase inhibition or TP antagonism. Up4A-induced aortic contraction was dramatically attenuated by TX synthase inhibitor ozagrel (Fig. 2A) and TP antagonist SQ29548 (Fig. 2B), respectively. These findings indicate that the involvement of vasoconstrictor prostanoids in Up4A-induced aortic contraction appears to be through TX synthase and TP activation.
Effects of thromboxane (TXA2) synthase inhibition and TXA2 receptor (TP) antagonism on Up4A-induced vasoconstriction. Shown are concentration-responses to Up4A in mouse aorta from WT in the absence and presence of 10 µM of the TX synthase inhibitor ozagrel (n=11 for both Con and ozagrel; panel A), and 1 µM of the TP antagonist SQ29548 (n=8 for Con; n=6 for SQ29548; panel B). The aortic rings in different protocols within panels are paired; the Up4A concentration responses for each control are similar to those within figures or in other figures. Values are means ± SEM. * P < 0.05, effect of drug vs. Con. * represents the concentration-response relation by two-way ANOVA.
3.3 Involvement of purinergic P1 receptors in Up4A-induced aortic contraction
Since both A1AR and A3AR have been shown to be involved in TXA2-mediated vascular contraction (18, 19), we next studied the involvements of A1AR and A3AR in Up4A-induced aortic contraction. Up4A induced a concentration-dependent aortic contraction in WT mice, which was comparable to that in A3AR KO mice (Fig. 3A). Surprisingly, Up4A-induced aortic contraction was potentiated in A1AR KO as compared to WT mice (Fig. 3B). A similar observation was also found in A1/A3AR DKO mice in which Up4A-induced aortic contraction was enhanced as compared to WT mice (Fig. 3C). These findings suggest that A3ARs does not appear to be involved in Up4A-induced contraction in mouse aorta. However, A1AR may exert a vasodilator rather than vasoconstrictor influence in Up4A-induced aortic contraction.
Effects of A1 adenosine receptor (AR) deletion or antagonism, A3AR deletion and combined A1AR and A3AR deletion on Up4A-induced vasoconstriction. Shown are concentration-responses to Up4A in mouse aorta from A3AR knockout (KO) mice (n=15 in WT; n=6 in A3AR KO; panel A), A1AR KO mice (n=15 in WT; n=6 in A1AR KO; panel B), and A1/A3AR double KO (DKO) mice (n=15 in WT; n=11 in A1/A3AR DKO; panel C), as well as in mouse aorta from WT in the absence (n=10) and presence of 10 nM of selective A1AR antagonist DPCPX (n=4; panel D). The aortic rings in different protocols within panels are paired; the Up4A concentration responses for each control are similar to those within figures or in other figures. Values are means ± SEM. * P < 0.05 vs. Con or WT. * represents the concentration-response relation by two-way ANOVA.
The observation that Up4A-induced aortic contraction was enhanced in A1AR KO or A1/A3AR DKO mice may be due to a compensatory mechanism that often occurs using genetically modified mice (28). To further confirm the role of A1AR in Up4A-induced aortic contraction, we performed concentration responses to Up4A in isolated aortic rings from WT mice in the absence and presence of pharmacological inhibition by the selective A1AR antagonist. Similar to the observations from A1AR KO or A1/A3AR DKO, selective A1AR antagonist DPCPX significantly potentiated Up4A-induced contraction in mouse aorta (Fig. 3D). This finding supports the observation that A1AR may exert a vasodilator influence, which suppresses Up4A-induced vascular contraction in mouse aorta.
3.4 Involvement of purinergic P2 receptors in Up4A-induced aortic contraction
Based on the findings that neither A1AR nor A3AR exerts a vasoconstrictor influence contributing to Up4A-induced aortic contraction, we next investigated the involvement of P2R in this process. Non-selective P2R antagonist PPADS (Fig. 4A) and selective P2X1R antagonist MRS2159 (Fig. 4B) significantly attenuated Up4A-induced vascular contraction in mouse aorta, respectively. To further study whether Up4A-induced aortic contraction via TXA2 production requires activation of P2X1R, isolated aortic rings were exposed to MRS2159, ozagrel or the combination of both MRS2159 and ozagrel. Up4A-induced aortic contraction was significantly reduced by MRS2159, which was slightly but significantly further attenuated by addition of ozagrel (Fig. 5). However, the inhibition of Up4A-induced aortic contraction by a combination of MRS2159 and ozagrel was similar in magnitude to that obtained by ozagrel alone (Fig. 5). All together, these observations indicate that Up4A-induced vascular contraction in mouse aorta is through TXA2 production, which partially requires the activation of P2X1R.
Effects of non-selective purinergic P2 receptor (P2R) antagonism and selective P2X1R antagonism on Up4A-induced vasoconstriction. Shown are concentration-responses to Up4A in mouse aorta from WT in the absence and presence of 10 µM of non-selective P2R antagonist PPADS (n=8 for Con; n=7 for PPADS; panel A) and 30 µM of the selective P2X1R antagonist MRS2159 (n=6 for Con; n=7 for MRS2159; panel B). The aortic rings in different protocols within panels are paired; the Up4A concentration responses for each control are similar to those within figures or in other figures. Values are means ± SEM. * P < 0.05, effect of drug vs. Con. * represents the concentration-response relation by two-way ANOVA.
Effects of P2X1R antagonism, TX synthase inhibition and combined P2X1R and TXA2 inhibition. Shown are concentration-responses to Up4A in mouse aorta from WT in the absence (n=6) and presence of 30 µM of the P2X1R antagonist MRS2159 (n=7), 10 µM of the TXA2 antagonist ozagrel (n=6) and combine MRS2159 and ozagrel (n=6). The control curve and MRS2159 curve are the same as those in Fig. 4B. The aortic rings in different protocols in the figure are paired; the Up4A concentration responses for control are similar to those in other figures. Values are means ± SEM. * P < 0.05 vs. Con; † P < 0.05 vs. MRS2159. * represents the concentration-response relation by two-way ANOVA.
3.5 Endothelium-dependency of aortic contraction produced by Up4A
To elucidate the role of endothelium in Up4A-induced aortic contraction, endothelium-denuded aortic rings were subjected to Up4A concentration responses in the absence and presence of P2X1R antagonist MRS2159 and TX synthase inhibitor ozagrel. Up4A-induced aortic contraction was almost completely blunted by endothelium denudation either with or without MRS2159 or ozagrel (Fig. 6), indicating that Up4A-induced contraction in mouse aorta is mainly endothelium-dependent.
Endothelium-dependency of vasoconstriction induced by Up4A. Shown are concentration-responses to Up4A in endothelium-non-denuded (n=7), endothelium-denuded rings and endothelium-denuded aortic rings from WT in the absence (n=5) and presence of 30 µM of the selective P2X1R antagonist MRS2159 (n=5) or 10 µM of the TX synthase inhibitor ozagrel (n=7). The aortic rings in different protocols in the figure are paired; the Up4A concentration responses for control are similar to those in other figures. Values are means ± SEM. * P < 0.05 vs. Con. * represents the concentration-response relation by two-way ANOVA.
3.6 TXA2 production generated by Up4A in isolated aortic rings
To further test our finding that TXA2 is involved in Up4A-induced aortic contraction, TXB2, a metabolite of TXA2, was measured in isolated intact aortic rings upon Up4A stimulation. The level of TXB2 was greatly increased in Up4A (3 µM)-treated tissues as compared to vehicle (Fig. 7). Furthermore, Up4A-increased TXB2 level was dramatically inhibited by either P2X1R antagonist MRS2159 or TX synthase inhibitor ozagrel (Fig. 7). These findings suggest that Up4A is capable of generating TXA2, which partially requires activation of P2X1R in mouse aorta.
Effects of Up4A on TXA2 production. Shown are effects of Up4A (3 µM) on production of TXB2, a metabolite of TXA2, in intact isolated aortic rings from WT in the absence and presence of 30 µM of the selective P2X1R antagonist MRS2159 or 10 µM of the TX synthase inhibitor ozagrel. N=6 in each group. Values are means ± SEM. * P < 0.05 vs. Vehicle; † P < 0.05 vs. Up4A by one-way ANOVA.
4. Discussion
The main findings of the present study are: 1) Up4A produced a concentration-dependent vascular contraction (>70%), which was attenuated by non-selective COX and selective COX1 but not COX2 inhibition; 2) Up4A-induced aortic contraction was markedly blocked by TX synthase inhibition and TP antagonism, respectively; 3) A3AR deletion had no effect on Up4A-induced aortic contraction, whereas A1AR deletion or antagonism as well as A1/A3AR deletion potentiated Up4A-induced contraction; 4) In contrast, non-selective P2R antagonist PPADS attenuated Up4A-induced contraction, while selective P2X1R antagonist MRS2159 similarly attenuated Up4A-induced aortic contraction, which was further reduced by TX synthase inhibitor ozagrel; 5) Furthermore, endothelial denudation almost fully blunted Up4A-induced aortic contraction either with or without MRS2159 or ozagrel; 6) and finally, TXA2 production was significantly increased by Up4A in intact isolated aortic rings, which was inhibited by either MRS2159 or ozagrel. The implications of these findings are discussed below.
Up4A was first identified as a potent endothelium-derived vasoconstrictor in rat perfused kidney (3). Several subsequent studies have confirmed that Up4A produces vasoconstriction in rat renal artery (3), aorta (5), gastric smooth muscle (29) and pulmonary arteries (6). Furthermore, vasoconstriction was observed in mouse renal arterioles (7) and aorta (8). Although, there is evidence that Up4A can produce vasodilation in isolated aortic rings of rats (5), porcine coronary small arteries (11, 12), human and mouse colon (30) as well as induce hypotension in conscious rats (8). All together, these studies suggest that the vascular responses to Up4A likely depend on the type of vascular beds and species studied (4).
In mouse aorta, Hansen and colleagues (8) demonstrated that in preconstricted aortic rings, cumulative addition (10−7–10−5 M) of Up4A produced vasoconstriction followed by relaxation at a dose of 10−5 M, indicating a biphasic effect of Up4A. Consistent with the vasodilator effect induced by Up4A at a dose of 10−5 M, we observed a similar relaxation of 41 ± 0.06% in the present study as compared to the report by Hansen et. al. (46 ± 6% relaxation) (8). However, cumulative addition of Up4A (10−9–10−5 M) produced a concentration-dependent contraction over 70%, which was inconsistent with the previous study in which cumulative addition of Up4A (10−7–10−5 M) caused aortic contraction of 18 ± 2%, 76 ± 16% and 18 ± 6%, respectively (8). This disparity may not be attributed to the different strain of animals (the same C57BL/6 mice with similar age range), but might due to the different number of animals (n) used in the studies in which variable vascular responses to Up4A between each individual animals may occur, as a large number of control animals were pooled in the present study (n=49) as opposed to the study by Hansen et. al. with an ‘n’ of only six (8). For the sake of comparing vasoconstrictor effect of Up4A (>70%), another vasoconstrictor agonist U46619 (TXA2 analogue 9,11-Dideoxy-11α, 9α -epoxymethanoprostaglandin F2α) concentration responses (10−9 M–10−6 M) were performed in the same preparation that produced a dose-dependent contraction up to ~240%, which almost reaches the maximal contraction. Together with the vasoconstrictor effects of ATP (~50%) (31) and activation of ARs (~30%) in isolated aorta, this may indicate that Up4A is a relatively potent vasoconstrictor, which submaximally contracts mouse aorta. Notably, infusion of Up4A (10−9–10−5 M) into isolated mouse hearts results in a dose-dependent reduction in coronary flow by ~40% (unpublished) as well as intra-aortic injection of Up4A (100 nmol) increases mean arterial blood pressure by ~25% in anesthetized rats (3), suggesting a potential vasoconstrictor influence of Up4A in vivo.
In accordance with previous finding by Hansen et. al. (8), Up4A-induced aortic contraction was markedly attenuated by non-selective COX inhibition with indomethacin, suggesting an involvement of vasoconstrictor prostanoids. Subsequent observations that Up4A-induced aortic contraction was blunted by selective-COX1 but not selective COX2 inhibition further supports the concept that activation of COX1 but not COX2 primarily contributes to generation of vasoconstrictor prostanoids in mouse aorta (32, 33). On the other hand, the lack of involvement of COX2 in Up4A-mediated aortic contraction could be due to the fact that COX2, an inducible enzyme, predominantly plays a role under inflammatory /diseased conditions (34). Indeed, one of the recent studies in diabetes showed that COX2 but not COX1 contributes to Up4A-increased renal contraction (20). Future studies on various disease models are needed to further explore our understanding of these mechanisms. As reported earlier that TXA2 production is abundant in mouse aorta at basal conditions (32) and Up4A is capable of producing TXA2 in renal arteries of rats (20), Up4A may also potentially increase TXA2 in mouse aorta causing vasoconstriction. Indeed, Up4A-induced aortic contraction was significantly reduced by both TP antagonism and TX synthase inhibition, suggesting that Up4A-induced contraction in mouse aorta is not only attributed to TP activation but also to TXA2 production. These findings are further supported by the observation that Up4A-increased TXA2 level in intact isolated mouse aorta was largely decreased by TXA2 sysnthase inhibition. Together with the involvement of COX1, these findings support our hypothesis that TXA2, likely derived from COX1 activation, contributes to Up4A-mediated aortic contraction in mouse aorta.
As a novel extracellular dinucleotide, the vascular effects of Up4A observed in previous studies have been challenged to be through its degradation to ATP or adenosine (3, 5, 11). Although both ATP and activation of ARs have been shown to produce aortic contraction through generation of TXA2 (19, 31), ATP (31) and activation of ARs (19) affected vascular tone much less (~50% and ~30%, respectively) than Up4A (>70%), indicating that these degradation products can only account for a small part, if any, of the Up4A effects, as proposed that Up4A might also be a stimulator of endothelial ATP release (3).
Like other nucleotides, Up4A exerts its vascular influence through activation of purinergic receptors (3, 4). Thus, Up4A induces vasoconstriction in rat perfused kidney through P2X1R and P2Y1R (10), in rat aorta through P1R and P2XR (5), and in rat pulmonary arteries through P2YR (6), while Up4A produces vasodilation in rat perfused kidney through P2Y1R and P2Y2R (10), in porcine coronary arteries through A2AAR, P2X1R and P2Y1R (12) and in human and mouse colon through P2Y1R (30). Surprisingly, Up4A-mediated vascular contraction in mouse aorta in the present study was not affected by A3AR deletion, although one of our previous observations showed that activation of A3AR contributes to mouse aortic contraction (19). Moreover, Up4A-induced aortic contraction was significantly potentiated by either A1AR or A1/A3AR deletion, which was further supported by an enhancing effect with pharmacological inhibition of A1AR (DPCPX). The enhancement with DPCPX in Up4A-induced aortic contraction in WT was greater than that in A1AR KO or A1/A3AR DKO mice. This difference in enhancement between pharmacological agent in WT and genetically modified mice is likely due to the upregulation of other purinergic receptors in A1AR KO or A1/A3AR DKO that compensates for the loss of A1AR (28), as we previously showed that there is an upregulation of A2BAR in A1AR KO mice (28). The vasodilator A2BAR is likely activated by Up4A to counteract the unveiled vasoconstrictor effect due to the loss of A1AR, thereby leading to a less enhancement of Up4A effect in A1AR KO or A1/A3AR DKO as compared to that with DPCPX in WT. Taken together, these findings suggest that A3AR does not appear to be involved in Up4A-induced aortic contraction, and activation of A1AR may exert a vasodilator influence in this process. In contrast to the generally accepted role of A1AR exerting a vasoconstrictor influence in the cardiovascular system (28, 35, 36), several previous studies have shown that selective A1AR agonist induces vasodilation in porcine coronary arteries (37), adenosine-mediated activation of A1AR leads to vasodilation in rat diaphragmatic arterioles (38) and activation of A1AR by adenosine contributes to hypoxia-induced vasodilation in perfused rabbit hearts (39). Moreover, at cellular levels, activation of A1AR has been shown to mediate release of NO from the endothelium of rat aorta (40). However, we could not exclude the involvement of vasodilator influence of A1AR in Up4A-induced aortic contraction is through other indirect mechanisms, as the interaction between A1AR and P2Y1R exists (41). Activation of P2Y1R has been shown to exert a vasodilator influence in vasculature. Thus, Up4A activates P2Y1R to produce vasodilation in rat perfused kidney (10), porcine coronary arteries (12), as well as human and mouse colons (30). Future studies addressing the role of A1AR in this process will bring greater insights into the purinergic receptor-regulated vascular tone.
In contrast to the observations that neither A1AR nor A3AR exerts a vasoconstrictor influence contributing to Up4A-induced aortic contraction, activation of P2X1R, at least in part, accounts for Up4A-induced vascular contraction in mouse aorta, as evidenced by a significant attenuation of Up4A-induced aortic contraction by both non-selective P2R antagonist PPADS as well as selective P2X1R antagonist MRS2159. The involvement of P2X1R in this process is in agreement with previous studies that activation of P2X1R contributes to both Up4A-mediated contraction in renal arteries (3) and NADPH-mediated contraction in aorta (42). Furthermore, there is some evidence linking TXA2 production or TP activation to activation of P2R. Thus, P2XR-agonist-induced vasoconstriction was attenuated by TXA2 antagonist in canine basilar arteries through endothelium-dependent mechanism (15), activation of both P2Y1R and P2Y2R produced vasoconstriction in human umbilical and chorionic vessels, which was blunted by TP antagonist (16), and ATP-induced contraction in rat aorta was attenuated by TP antagonist (31). Consistent with these observations, in the present study, Up4A increased TXA2 production in isolated aorta was inhibited by P2X1R antagonist MRS2159. Of note, Up4A-induced aortic contraction was significantly attenuated by MRS2159, which was further blunted by addition of TX synthase inhibitor ozagrel. Our findings suggest that Up4A-induced aortic contraction mediated by TXA2 is, in part, through activation of P2X1R. The reduction of Up4A-induced aortic contraction by P2X1R antagonist MRS2159 was similar in magnitude to that obtained by non-selective P2R antagonist PPADS, which could not exclude the possibility of involvement of additional P2R subtypes (both P2XR and P2YR) other than P2X1R in Up4A-mediated aortic contraction. Since PPADS blocks most, but not all, of P2R subtypes, and PPADS does not inhibit P2X1R, as evidenced by precious studies that PPADS blocked P2X2R, P2X3R, P2X5R (43), P2X7R (44), P2Y1R (45), P2Y2R, P2Y4R (46) and P2Y6R (47). Notably, P2X1R-mediated renal vasoconstriction in response to Up4A was not blocked by PPADS but blocked by another non-selective P2R antagonist suramin (10). The limitation for the relation between TXA2 generation and activation of P2X1R is that whether the generation of TXA2 in response to Up4A was truly of endothelial origin or from the transcellular metabolism of endoperoxide is unknown, as various prostaglandins contract the smooth muscle cells by activating TP (31). On the other hand, methodological intricacies are compounded by lack of selective P2X1R agonist to explore in detail the coupling of TXA2 production to activation of P2X1R (16). Future studies regarding this issue will confirm the coupling between TXA2 production and P2X1R activation.
P2R in the vasculature are present on endothelial cells as well as vascular smooth muscle cells (VSMC). Activation of P2R on endothelial cells is generally accepted to produce vasodilation, whereas activation of P2R on VSMC leads to vasoconstriction (48). Indeed, previous studies have shown that activation of endothelial P2X1R in response to Up4A results in vasodilation in porcine coronary arteries (12), whereas activation of P2X1R by Up4A on VSMC produces vasoconstriction in rat perfused kidney (3, 4). In contrast to this concept, the observation that endothelium-denudation almost fully attenuated Up4A-induced aortic contraction both in the absence and presence of P2X1R or TXA2 inhibition indicates that Up4A-induced contraction is mainly endothelium-dependent, and Up4A generates TXA2 production requiring activation of P2X1R on endothelial cells. Similar phenomenon has been observed in rat aorta in which activation of endothelial P2XR by ATP leads to endothelial TXA2 production (31), and activation of P2X1R located on endothelial cells has been shown to contribute to NADPH-mediated vascular contraction in mouse aorta (42).
5. Conclusions
The present study demonstrates that Up4A-induced vascular contraction in mouse aorta is attributed to TXA2 production, which partially requires activation of P2X1R through an endothelium-dependent mechanism. As plasma levels of Up4A in juvenile hypertensive subjects are significantly elevated (13) and intra-aortic injection of Up4A increases mean arterial blood pressure in intact animal (3), suggesting a role for Up4A in pathogenesis of hypertension, further studies investigating the altered vascular responses to Up4A in hypertensive conditions and the contribution of TXA2 to Up4A-mediated changes in blood pressure will provide novel therapeutic targets in this process.
Acknowledgments
This study was supported by HL 027339, HL 094447, U54GM104942 (WV-CTSI) (to SJM), HL071802 (to SLT) and MEPSCKL201301 (to ZZ).
Abbreviations
| AR | adenosine receptor |
| COX | cycloxygenase |
| DPCPX | 8-Cyclopentyl-1,3-dipropylxanthine |
| DMSO | dimethyl sulphoxide |
| EDHF | endothelium-derived hyperpolarizing factor |
| KO | knockout |
| NS398 | N-[2-(Cyclohexyloxy)-4-nitrophenyl]-methanesulfonamide |
| ozagrel | (E)-3-[4-(Imidazol-1-ylmethyl)phenyl]propenoic acid hydrochloride hydrate |
| PPADS | pyridoxalphosphate-6-azophenyl-2',4'-disulfonic acid |
| SC560 | 5-(4-Chlorophenyl)-1-(4-methoxyphenyl)-3-trifluoromethyl pyrazole |
| SQ29548 | [1S-[1α,2α(Z),3α4α]]-7-[3-[[2-[(phenylamino)carbonyl]hydrazino]methyl]-7-oxabicyclo[2.2.1]hept-2-yl]-5-heptenoic acid |
| TXA2 | thromboxane |
| TP | thromboxane receptor |
| Up4A | uridine adenosine tetraphosphate |
| U46619 | 9,11-Dideoxy-11α, 9α -epoxymethanoprostaglandin F2α |
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.