FEBS Letters 588 (2014) 2936–2943 journal homepage: www.FEBSLetters.org Characterization of the contractile P2Y14 receptor in mouse coronary and cerebral arteries ⇑ Kristian Agmund Haanes , Lars Edvinsson Department of Clinical Experimental Research, Glostrup Research Institute, Copenhagen University Hospital, Nordre Ringvej 59, 2600 Glostrup, Denmark article info abstract Article history: Extracellular UDP-glucose can activate the purinergic P2Y14 receptor. The aim of the present study Received 13 February 2014 was to examine the physiological importance of P2Y14 receptors in the vasculature. The data pre- Revised 13 May 2014 sented herein show that UDP-glucose causes contraction in mouse coronary and basilar arteries. Accepted 21 May 2014 The EC values and immunohistochemistry illustrated the strongest P2Y14 receptor expression in Available online 6 June 2014 50 the basilar artery. In the presence of pertussis toxin, UDP-glucose inhibited contraction in coronary Edited by Ned Mantei arteries and in the basilar artery it surprisingly caused relaxation. After organ culture of the coro- nary artery, the EC50 value decreased and an increased staining for the P2Y14 receptor was observed, showing receptor plasticity. Keywords: Purinergic receptors Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. P2Y14 Coronary artery Basilar artery Receptor upregulation P2Y2 KO 1. Introduction vasoconstriction in porcine pancreatic arteries [12]. This study was designed to examine the physiological importance of the Nucleotides are extracellular signal molecules that can activate P2Y14 receptor in the vasculature, specifically in the brain (basilar purinergic receptors in different tissues. There exist two main fam- artery) and heart (left anterior descending artery). Since UDP-glu- ilies of nucleotide-activated purinergic receptors: the P2X receptor cose has been shown to also activate P2Y2 receptors [13], experi- family of ligand-gated ion channels and the P2Y receptor family of ments were performed on wild type (WT) and on P2Y2 receptor G-protein coupled receptors [1]. Freemann and colleagues showed knock out (KO) mice [14]. that the orphan G-protein receptor GPR105 was activated by UDP- The A2A adenosine receptor acts through the stimulatory G-pro- glucose [2]. UDP-glucose can be released by cells, and thus extra- tein (Gs) and increases cellular cAMP levels leading to an endothe- cellular UDP-glucose can be present [3]. All signalling components lium independent relaxation [15]. We aimed to investigate whether for this receptor were present, and it was classified as a purinergic the P2Y14 receptor could be involved in smooth muscle contraction receptor and given the name P2Y14 [4]. Later, Fricks and colleagues via Gi activation and if it could be counteracted by the Gs relaxing showed that the intracellular pathway for the P2Y14 receptor is effect of adenosine. The experiments presented here show that coupled to the inhibitory G-protein (Gi) that inhibits the adenylyl the P2Y14 receptor is indeed a contractile receptor that is expressed cyclase, thereby lowering cellular cAMP levels [5]. in the cerebral and coronary arteries, and that the P2Y2 receptor The P2Y14 receptor is expressed widely in the body, e.g., in interference is minor. Organ culture of coronary artery segments brain, heart, skeletal muscles and spleen [6]. Most of the work so revealed enhanced expression of the P2Y14 receptor, demonstrat- far has been done on neutrophils where UDP-glucose is a chemo ing receptor plasticity. attractant [7–9]. Nucleotides have a well-established physiological role in vascular smooth muscle cells, where the purinergic recep- 2. Materials and methods tors modulate the contraction and relaxation [10,11]. Very recently, Alsaqati and colleagues reported that UDP-glucose elicited 2.1. Myography Mixed female and male BalbC mice (WT or P2Y2 KO [backcrossed for 7 generations]; 20 g, n = 52) were sedated with 70% CO2 in 30% ⇑ Corresponding author. Fax: +45 38633983. O and killed by decapitation. Heart and brain were immediately E-mail address: [email protected] (K.A. Haanes). 2 http://dx.doi.org/10.1016/j.febslet.2014.05.044 0014-5793/Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. K.A. Haanes, L. Edvinsson / FEBS Letters 588 (2014) 2936–2943 2937 excised and placed into ice cold oxygenated calcium free physiolog- HPO4) overnight, followed by 25% sucrose for 24 h. The tissue was ical saline solution (Ca2+-free PSS) composed of (in mM): 119 NaCl, covered with tissue TEKÒ (Gibco) and sectioned at a thickness of 4.7 KCl, 2.5 CaCl2, 25 NaHCO3, 1.17 MgSO4, 1.18 KH2PO4, 5.5 glucose, 10 lm on a cryostat (Leica Microsystems GmBH, Germany). All sec- and 0.03 EDTA, pH 7.4. The LAD (left anterior descending) and basilar tions were blocked with 5% donkey serum (Jackson ImmunoRe- artery (segments length 2 mm) were dissected in ice cold, oxy- search Europe, Suffolk, UK) for 30 min, and incubated with genated Ca2+-free PSS and mounted on a Mulvany–Halpern wire primary antibodies directed against the P2Y14 receptor (rabbit myograph (Danish Myo Technology, Denmark). The myographs anti-GPR105/P2Y14, Ab136264, 1:200, Abcam, UK) in PBS contain- were connected to a PowerLab Unit and responses were sampled ing 1% BSA and 0.25% Triton X-100, overnight. Secondary antibody in LabChart™ (ADInstruments, UK). The mounted artery segments (Alexa Fluor 488, donkey anti-rabbit 1:400, Jackson ImmunoRe- were heated to 37 °C, and buffer changed for PSS with calcium. After search Europe, Suffolk, UK) was added after washing with PBS-Tri- 15 min equilibration, the vessels were stretched to their optimal ton X-100. Slides were mounted with Vectashield with DAPI. lumen diameter L1 = 0.9 Â L100, where L100 is an estimate of the Immunoreactivity was visualized and photographed with a Nikon diameter of the vessel under a passive transmural pressure 6 kPa microscope (EZ-c1, Germany), with the following filters: DAPI for the basilar artery and 9 kPa for the coronary arteries, in order (EX340-38, DM400, BA/EM435-485) and Alexa488 (EX465-495, to obtain optimal conditions for active tension development. Subse- DM505, BA/EM515-555). quently, the vessels were allowed to stabilize for 20–30 min. The vessels were kept at this standard tension during the entire period 2.5. Western blot in the myograph. To ensure a stable pH and bicarbonate concentra- tion at 25 mM (pH 7.40), all myograph baths were continuously bub- Dissected arteries were homogenized immediately, in boiling bled with 12% CO2 in 88% O2, to compensate for the large surface to SDS loading buffer (Expedeon, Kem-en-tec, Denmark), sonicated, volume ratio. frozen and re-sonicated before being heated to 70 °C for 10 min. The vascular smooth muscle cell contractile function was con- 10 ll of the sample was loaded into each well on a 4–20% gradient firmed by challenging the segments two times with 125 mM K+ gel. The proteins were transferred to PVDF membranes (GE Health- (KPSS, similar to PSS except that NaCl was exchanged for KCl on care, Denmark) using wet electroblotting. The membranes were an equimolar basis). All contractile responses are expressed as per- blocked overnight using 2% ECL™ Advance Blocking Agent (GE centage of the average maximal contraction induced by the K+ Healthcare, Denmark), incubated overnight at 4 °C in primary anti- response, with either normal baseline or PGF2a (Prostaglandin body solutions (rabbit anti-GPR105/P2Y14, Ab136264, 1:500 or goat F2a) contraction as baseline. The vessels where pre-contracted anti-SM22 1:2,000, both from Abcam, UK), washed briefly in TBS-T À6 À6 with 2 Â 10 M PGF2a for coronary arteries and 3 Â 10 M PGF2a (Tris buffered saline + 0.01% Tween-20) and then incubated with for cerebral arteries, producing about 50% of the contraction seen secondary antibody (ECL™ donkey anti rabbit 1:20,000, or DAKO À5 with 10 M PGF2a. This allows for fine tuning of the contractile donkey anti goat 1:2000) for 1.5 h at room temperature. Membranes response, e.g. by cAMP. All concentration response curves were were washed (5 Â 5 min in TBS-T) and developed using ECL™ Select made in a cumulative manner. Western Blotting Detection Kit (GE Healthcare, Denmark). Image capture was done using a LAS-1000 digital camera unit and quanti- 2.2. Endothelial influence fication using Multi gauge 3.2 (Fujifilm, Denmark). The addition of 10À5 M carbachol after precontraction by 2.6. Chemicals, data analyses and statistics À5 10 M PGF2a, was used to assess the endothelial function. The endothelium dependent carbachol effect value is expressed in %, All drugs and chemicals for myograph studies were acquired where 100% is a relaxation back to the baseline. The vessels used from Sigma–Aldrich, Germany. in this study are very small; therefore inserting the myograph wire Data were analysed using GraphPad Prism software (GraphPad will sometimes cause a partial or full removal of the endothelium. Software Inc., USA) and expressed as mean ± SEM (n equals the Since we have a varying degree of endothelium (determined by the number of animals). EC50 (or as negative logarithm: pEC50) repre- carbachol induced relaxation) we can correlate the amount of func- sents the agonist concentration required to produce 50% of the tional endothelium cells with an agonist induced contraction [16]. maximum response, calculated using non-linear regression. Where The influence of the endothelium on the Emax values of UDP-glu- no clear maximum contraction was observed, the max value was + cose was analysed by constructing a correlation curve of the Emax set to be 60% K ; this could lead to underestimates of the EC50 val- and the endothelium-dependent carbachol effect. ues. Paired and unpaired Student’s t-test was used to compare two groups.