Loss of Mouse P2Y4 Nucleotide Receptor Protects against Myocardial Infarction through Endothelin-1 Downregulation

This information is current as Michael Horckmans, Hrag Esfahani, Christophe Beauloye, of September 25, 2021. Sophie Clouet, Larissa di Pietrantonio, Bernard Robaye, Jean-Luc Balligand, Jean-Marie Boeynaems, Chantal Dessy and Didier Communi J Immunol published online 16 January 2015

http://www.jimmunol.org/content/early/2015/01/16/jimmun Downloaded from ol.1401364 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 25, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published January 16, 2015, doi:10.4049/jimmunol.1401364 The Journal of Immunology

Loss of Mouse P2Y4 Nucleotide Receptor Protects against Myocardial Infarction through Endothelin-1 Downregulation

Michael Horckmans,*,1 Hrag Esfahani,†,1 Christophe Beauloye,‡ Sophie Clouet,* Larissa di Pietrantonio,* Bernard Robaye,x Jean-Luc Balligand,† Jean-Marie Boeynaems,*,{ Chantal Dessy,†,2 and Didier Communi*,2

Nucleotides are released in the heart under pathological conditions, but little is known about their contribution to cardiac inflam-

mation. The present study defines the P2Y4 nucleotide receptor, expressed on cardiac microvascular endothelial cells and involved in postnatal heart development, as an important regulator of the inflammatory response to cardiac ischemia. P2Y4-null mice displayed smaller infarcts in the left descending artery ligation model, as well as reduced neutrophil infiltration and fibrosis. Gene

profiling identified inter alia endothelin-1 (ET-1) as one of the target genes of P2Y4 in ischemic heart. The reduced level of ET-1 was correlated with reduction of microvascular hyperpermeability, neutrophil infiltration, and endothelial adhesion molecule expression, Downloaded from

and it could be explained by the decreased number of endothelial cells in P2Y4-null mice. Expression analysis of metalloproteinases and their tissue inhibitors in ischemic heart revealed reduced expression of matrix metalloproteinase (MMP)-9, reported to be potentially regulated by ET-1, and MMP-8, considered as neutrophil collagenase, as well as reduction of tissue inhibitor of MMP-1

and tissue inhibitor of MMP-4 in P2Y4-null mice. Reduction of cardiac permeability and neutrophil infiltration was also observed in P2Y4-null mice in LPS-induced inflammation model. Protection against infarction resulting from loss of P2Y4 brings new therapeutic perspectives for cardiac ischemia and remodeling. The Journal of Immunology, 2015, 194: 000–000. http://www.jimmunol.org/

ardiac injury is a great challenge owing to the limited the recruitment of leukocytes into the infarcted area. Neutrophils regenerative ability of the heart. Myocardial infarction and macrophages clear the wound from dead cells and matrix C (MI) is the most common source of cardiac injuries debris (2). However, MI is a multifactorial process that leads also leading to ischemic death of a large number of cardiomyocytes (1). in the first hours to a postischemic edema that develops as a result After this phase, dying cells trigger an inflammatory reaction with of increased microvascular permeability. Myocardial edema con- tributes to vessel collapse and impaired heart function, including reperfusion arrhythmias and stunning, and could affect ventricular by guest on September 25, 2021 *Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´culaire, Uni- remodeling by altering myocardial stiffness (3). Among the pro- † versite´ Libre de Bruxelles, 1070 Brussels, Belgium; Unite´ de Pharmacologie et de teins regulating the inflammatory response to cardiac ischemia, The´rapeutique, Universite´ Catholique de Louvain, 1200 Brussels, Belgium; ‡Poˆle de Recherche Cardiovasculaire, Institut de Recherche Expe´rimentale et Clin- endothelin-1 (ET-1) plays an important role. Indeed ET-1 is in- ique, Universite´ Catholique de Louvain, 1200 Brussels, Belgium; xInstitut de volved in neutrophil trafficking and increases cardiac microvas- Recherche Interdisciplinaire en Biologie Humaine et Mole´culaire, Universite´ Libre { cular permeability (4, 5). After the edema step, the healing process de Bruxelles, 6041 Gosselies, Belgium; and De´partement de Me´decine de Labora- toire, Hoˆpital Erasme, Universite´ Libre de Bruxelles, 1070 Brussels, Belgium proceeds to the formation of granulation tissue, which is charac- 1M.H. and H.E. contributed equally to this work. terized by the presence of fibroblasts, macrophages, myofibro- 2C.D. and D.C. contributed equally to this work. blasts, new blood vessels, and extracellular matrix proteins (6, 7), ultimately resulting in scar formation, which is characterized by Received for publication May 28, 2014. Accepted for publication December 8, 2014. acellular and cross-linked collagen-rich regions (8–11). This work was supported by an Action de Recherche Concerte´e of the Communaute´ Franc¸aise de Belgique, an Interuniversity Attraction Pole grant from the Politique Cardiac endothelial cells and cardiomyocytes are an important Scientifique Fe´de´rale (Grant IAP-P6/30), Prime Minister’s Office, Federal Service for source of extracellular nucleotides within the heart, especially Science, Technology and Culture, by grants of the Fonds de la Recherche Scientifique under ischemia (12). Extracellular nucleotides are involved in the Me´dicale, the Fonds d’Encouragement a` la Recherche, the Fonds Emile Defay, the Fonds de la Recherche Scientifique Me´dicale of Belgium, the Walloon Region (Pro- protection of rat cardiomyocytes from hypoxic stress through gramme d’Excellence CIBLES), and by LifeSciHealth Program of the European activation of P2Y2 nucleotide receptors (13). UTP administration Community Grant LSHB-2003-503337. D.C. and C.D. are Senior Research Associ- ates of the Fonds National de la Recherche Scientifique. M.H. and S.C. are supported to rats reduced infarct size and enhanced myocardial function by the Fonds National de la Recherche Scientifique/Fonds pour la Recherche dans (14). The implication of P2Y2 receptor in the cardioprotection l’lndustrie et l’Agriculture, Belgium. The funders had no role in study design, data mediated in vivo by UTP has been recently demonstrated (15). collection and analysis, decision to publish, or preparation of the manuscript. We investigated in the present study the potential role of the Address correspondence and reprint requests to Dr. Didier Communi, Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´culaire, Universite´ Libre P2Y4 receptor in cardiac inflammation and myocardial infarction. de Bruxelles, Building C (5th Floor), Campus Erasme, 808 Route de Lennik, 1070 The human P2Y4 receptor is a UTP receptor coupled to the Brussels, Belgium. E-mail address: [email protected] phosphoinositide/Ca2+ pathway (16). Recombinant rat and mouse Abbreviations used in this article: EF, ejection fraction; ET-1, endothelin-1; HPRT, orthologs of human P2Y4 receptor are activated by UTP and hypoxanthine phosphoribosyltransferase; LAD, left anterior descending artery; LVED, left ventricular end-diastolic volume; LVES, left ventricular end-systolic ATP (17–19), as described for the ubiquitously expressed mouse volume; MI, myocardial infarction; MMP, matrix metalloproteinase; MPO, myelo- P2Y2 receptor (20). Interestingly, P2Y2 and P2Y4 receptors were peroxidase; TIMP, tissue inhibitor of matrix metalloproteinase; VEGF, vascular en- abundantly detected in the rat developing heart (21). P2Y -null dothelial growth factor. 4 mice have been generated in our laboratory (22) and we demon- Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 strated P2Y4 expression in cardiac endothelial cells but not in

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1401364 2 LOSS OF MOUSE P2Y4 PROTECTS AGAINST MYOCARDIAL INFARCTION cardiomyocytes (23). We showed recently that this receptor is in- Allylamine-induced heart fibrosis volved in postnatal heart development (23), as well as in exercise 0/+ 0/2 Male P2Y4 and P2Y4 mice were given 0.1% allylamine-HCl as tolerance and exercise-induced cardiac hypertrophy (24). drinking fluid ad libitum. These mice demonstrated a sustained decreased In this study, we demonstrated that loss of mouse P2Y4 receptor fluid consumption. For histopathological study, mice were given allyl- is associated with an unexpected protection against infarction and amine for 3, 7, 14, or 30 d and were killed by cervical dislocation. The a reduction of cardiac inflammation, permeability, and fibrosis. hearts were rapidly removed and frozen. Quantification of fibrosis in ischemic or allylamine-treated Materials and Methods mice +/+ 2/2 Ischemia experiments: LAD ligation of P2Y4 and P2Y4 Heart sections of ischemic or allylamine-treated mice were cut, fixed in mice Bouin’s solution, and stained with the Masson trichrome. The fibrosis was quantified as the relative surface of the blue staining on five cryosections P2Y4-null mice were generated in the host laboratory (22). The P2Y4 gene (7 mm) of the heart using ImageJ software. is located on the X chromosome: male and female P2Y4-null mice are thus 0/2 2/2 respectively designated as P2Y4 or P2Y4 mice. Female mice were LPS-induced inflammation model preferably used for ischemia experiments because they display lower 2 +/+ 2/2 P2Y 0/+ and P2Y 0/ mice were injected i.p. with 100 mg/kg LPS sus- ligation-induced mortality. P2Y4 and P2Y4 female mice were 4 4 anesthetized with avertin (240 mg/kg), intubated, and ventilated. A left pended in sterile PBS. After 4 h, the peritoneal cavity was washed with side thoracotomy was performed, and the pericardium was incised. MI was 10 ml sterile PBS. Cell number was determined using a counting chamber. then induced by the permanent ligation of the left anterior descending Thereafter, cytospin preparations were prepared using a Shandon CytoSpin artery (LAD) proximal to its bifurcation from the main stem. For some III cytocentrifuge (Thermo Fisher Scientific, Leicestershire, U.K.) and 2 2 +/+ / Diff-Quick staining (Dade Behring, Deerfield, IL). For inflammation Downloaded from experiments, 10 min before ligation, P2Y4 and P2Y4 female mice were injected with vehicle or ET-1 (1 nmol/kg). Twenty-four hours after analysis, mice were coinjected with LPS (100 mg/kg) and ET-1 (1 nmol/kg) LAD ligation, hearts were washed and dissected, and slices were incubated or a selective ETA receptor antagonist, BQ-123 (1 mg/kg). immediately into 2 g/100 ml TTC triphenyltetrazolium chloride solution at 37˚C for 15 min. Slices were then placed into 4% formol for tissue fixation Permeability experiments and stored at 4˚C overnight. Plasma leakage was measured in vivo using a modification of the technique employed by Hele et al. (25). Evan’s blue dye was used as a marker of albumin Echocardiographic analysis of P2Y +/+ and P2Y 2/2 ischemic 4 4 extravasation. In practice, vehicle, ET-1 (1 nmol/kg), or BQ-123 (1 mg/kg) http://www.jimmunol.org/ mice were given 5 min prior to Evan’s blue (20 mg/kg, iv.) with or without LPS (30 mg/kg, iv.). After 15 min, animals were perfused with 50 ml sterile saline Transthoracic echocardiography was performed on mildly anesthetized (0.9%). The vascular leakage, corresponding to the quantity of the dye in spontaneously breathing mice (sedated by inhalation of 1% isoflurane, 1 l/ tissues, was quantified as the relative surface of fluorescence on five cryo- min oxygen), using a Vevo 2100 high-resolution imaging system equipped sections (7 mm) of the heart using ImageJ software. with a 40 MHz transducer (VisualSonics, Toronto, ON, Canada). The mice were placed on a heated electrocardiogram platform. Left parasternal long Statistical analysis axis view and left midpapillary, apical, and basal short axis views were acquired. End-diastolic and end-systolic volumes and ejection fraction were Data are expressed as means 6 SEM for in vitro and in vivo studies. evaluated on the left parasternal long axis view. Endpoint comparisons with two groups were performed using an unpaired two-tailed Student t test (Prism software version 5; GraphPad Software, La by guest on September 25, 2021 Immunohistochemistry experiments Jolla, CA). For parallel repeated-measures studies, two-way ANOVA was used with Bonferroni post hoc evaluations to determine the significance for +/+ 2/2 P2Y4 and P2Y4 ischemic mice were killed by cervical dislocation individual time points (Prism software). For all studies, a p value ,0.05 and dissected. Hearts were harvested, weighed, frozen directly in Tissue- was considered significant. Tek OCT compound (VWR Scientific). Heart sections were cut at 7 mm thickness and fixed with methanol. The sections were then washed in PBS and stained with H&E, DAPI, and Abs against myeloperoxidase (MPO). Results Neutrophil density staining was quantified using ImageJ software in P2Y4-null mice are protected against infarction in the LAD transversal cross-sections by examining 10 fields per section at original ligation model magnification 320, in a blinded fashion. Images of immunostaining were acquired at room temperature in FluorSave reagent (Calbiochem) using an We performed a mouse model of myocardial ischemia, based on +/+ 2/2 Axio Observer Z1 microscope, a high-resolution charge-coupled device the ligation of the LAD, in P2Y4 and P2Y4 female mice. camera, and AxioVision 4.6.3 software (Carl Zeiss). Analysis of whole hearts from base to apex revealed smaller infarcts in P2Y 2/2 hearts 24 h after ligation (Fig. 1A). We ob- Quantitative RT-PCR experiments 4 served a 25% reduction of infarct size, defined as weight of the Total RNAs were extracted using TRIzol reagent (Life Technologies, 2/2 infarcted area divided by left ventricle weight, in P2Y4 mice Groningen, the Netherlands) and an RNeasy kit column (Qiagen, Hilden, +/+ 0/+ 0/2 compared with P2Y4 mice (Fig. 1B). Effectively, 24 h after Germany) from total heart isolated from P2Y4 and P2Y4 mice. RNA 6 6 +/+ was reverse transcribed using random hexamers and SuperScript II reverse ligation, infarct size was 20.2 1.6 and 14.7 1.4% in P2Y4 2/2 transcriptase. For MPO, ICAM-1, E-selectin, VCAM-1, matrix metal- and P2Y4 hearts, respectively (means 6 SEM; p , 0.05) +/+ loproteinase (MMP)-8, MMP-9, tissue inhibitor of MMP (TIMP)-1, TIMP- (Fig. 1B). These ratios were obtained for 21 P2Y4 mice and 17 4, and vascular endothelial growth factor (VEGF) quantification, 24 h after P2Y 2/2 mice. MI, total RNA was isolated from the ischemic area or remote region of 4 +/+ 2/2 We then performed echocardiography experiments to compare P2Y4 and P2Y4 mice injected or not with ET-1 (1 nmol/kg) by using TRIzol (Invitrogen). RT-PCR amplification mixtures contained 2 ng tem- cardiac parameters such as left ventricular end-diastolic volume plate cDNA. Reactions were run on a 7500 fast real-time PCR system (LVED), left ventricular end-systolic volume (LVES), and ejection (Applied Biosystems). Means 6 SEM of normalized ratios were obtained +/+ 2/2 fraction (EF) in P2Y4 and P2Y4 ischemic mice (Fig. 1C–E). using qBase software. PCR data were normalized for each gene to RPL13 LVED and LVES values were significantly increased 7 and 14 d and hypoxanthine phosphoribosyltransferase (HPRT) housekeeping genes, +/+ 2/2 and they are displayed as the ratio of expression compared with the level after ligation in P2Y4 ischemic mice, but not in P2Y4 obtained for the remote area of wild-type mice. ischemic mice (Fig. 1C, 1D). EF was slightly but not significantly 2/2 increased at 7 and 14 d in P2Y4 ischemic mice (Fig. 1E). ELISA ET-1 is a P2Y4 target gene in mouse heart Sera of P2Y4 wild-type and knockout mice were collected and ET-1 (Enzo Life Sciences) or VEGF (R&D Systems) level was measured by ELISA, To investigate the potential mechanism of P2Y4 involvement in according to the manufacturer’s procedure. cardiac ischemia, its target genes were identified by gene profiling The Journal of Immunology 3

+/+ 2/2 FIGURE 1. P2Y4-null mice display a protection against myocardial infarction. (A) Infarcted hearts of P2Y4 and P2Y4 mice. Images of whole Downloaded from +/+ 2/2 P2Y4 and P2Y4 hearts were taken from base to apex 24 h after ligation. (B) Reduced infarct size in P2Y4-null mice. Quantification of the infarct size, +/+ corresponding to the ratio between infarcted area weight and left ventricular weight (infarct/LV) 24 h after LAD ligation. Ratios were obtained in 21 P2Y4 2/2 0/+ 0/2 and 17 P2Y4 mice (means 6 SEM). *p , 0.05. (C–E) Echocardiography analysis of P2Y4 and P2Y4 ischemic mice. Echocardiography experiments 0/+ 0/2 were performed to evaluate cardiac parameters such as LVED (C), LVES (D), and EF (E) in P2Y4 and P2Y4 ischemic mice. The data were obtained 0/+ 0/2 before (0 d) and 1, 7, and 14 d after LAD ligation for five P2Y4 and five P2Y4 ischemic mice. **p , 0.01. http://www.jimmunol.org/ 2/2 in ischemic heart. Microarray experiments revealed tens of dif- was observed in the infarcted area of P2Y4 ischemic hearts +/+ ferences in gene expression in the ischemic heart of P2Y4-null compared with P2Y4 ones (Fig. 3C–3F). A significant effect of compared with wild-type mice (data not shown). Among the in- ET-1 injection was observed only on ICAM-1 expression in +/+ flammatory genes that were regulated, ET-1 was downregulated P2Y4 ischemic heart (Fig. 3D). with a ratio of 0.5. We have focused on this particular target gene because, as described below, we identified in parallel an inflam- Impaired fibrosis in P2Y4-deficient mice +/+ 2/2 matory defect in P2Y4-null mice. It is known that ET-1 is involved We have stained sections of P2Y4 and P2Y4 infarcted hearts in cardiac inflammation, heart failure, and ventricular remodeling with the Masson trichrome method (Fig. 4A). As shown in Fig. 4B

+/+ by guest on September 25, 2021 (26). Therefore, we evaluated ET-1 serum level in P2Y4 and and 4C, Masson trichrome blue staining revealed a highly sig- 2/2 P2Y4 mice during infarction (Fig. 2A). Serum concentration of nificant decrease of the fibrotic area in the infarcted heart of 2/2 +/+ 2/2 +/+ ET-1 was lower in P2Y4 mice than in P2Y4 mice both in P2Y4 mice compared with P2Y4 mice 14 and 30 d after basal conditions and 6 h after coronary ligation (Fig. 2A). LAD ligation. We investigated then the potential effect of reduced ET-1 ex- Members of the MMP/TIMP system such as MMP-2, MMP-8, 2/2 pression in P2Y4 mice in the LAD ligation model. ET-1 or PBS MMP-9, and TIMP-1 are reported to be involved in cardiac was administered via a single i.v. administration 10 min before the remodeling (27). Quantitative PCR experiments were performed ischemic period onset, and infarct size was assessed after 24 h. to quantify expression MMPs (MMP-2, MMP-9, and MMP-8) and +/+ +/+ Injection of ET-1 increased infarct size in both P2Y4 mice and TIMPs (TIMP-1, TIMP-2, TIMP-3, and TIMP-4) in P2Y4 and 2/2 2/2 P2Y4 mice (Fig. 2B). Interestingly, it appeared that infarct size P2Y4 ischemic heart. Lower expression of MMP-8, MMP-9, 2/2 was comparable in P2Y4 ischemic mice injected with ET-1 and TIMP-1, and TIMP-4 was observed in the infarcted area of +/+ 6 6 2/2 +/+ P2Y4 ischemic untreated mice (23.5 1.3 versus 20.2 1.6%, P2Y4 ischemic hearts compared with P2Y4 ones (Fig. 3D– respectively) (Fig. 2B). G), whereas no difference was observed for MMP-2, TIMP-2, and TIMP-3 (data not shown). Defective neutrophil infiltration and endothelial adhesion molecule expression in P2Y4-deficient mice during myocardial infarction

To investigate whether loss of P2Y4 affected inflammation during ischemia, we quantified neutrophil infiltration in the infarction area 24 h after ligation (Fig. 3A, 3B). We observed that loss of + P2Y4 significantly reduced MPO cells infiltrating the infarcted 2 +/+ heart: from 885 6 117 neutrophils/mm in P2Y4 heart to 456 6 2 2/2 90 neutrophils/mm in P2Y4 heart (Fig. 3B). We also observed 2 2 2/2 A a reduction of TNF-a serum level in P2Y / mice 24 h after FIGURE 2. Regulation of ET-1 in P2Y4 ischemic mice. ( ) Quan- 4 +/+ 2/2 ligation (data not shown). tification of ET-1 serum level in P2Y4 and P2Y4 ischemic mice. Sera of P2Y +/+ and P2Y 2/2 mice were collected before (ctrl) or 6 or 24 h after ET-1 is known to regulate neutrophil adhesion and expression 4 4 LAD ligation for ELISA measurements of ET-1. Results represent the of endothelial adhesion molecules such as ICAM-1 and E-selectin +/+ means 6 SEM of eight mice. ***p , 0.001. (B) ET-1 injection in P2Y4 (5). Quantitative PCR experiments were performed to quantify 2/2 and P2Y4 ischemic mice. Quantification of the infarct zone was per- expression of MPO and adhesion molecules (ICAM-1, VCAM-1, formed 24 h after LAD ligation. Ratios were measured with or without +/+ 2/2 +/+ 2/2 and E-selectin) in P2Y4 and P2Y4 ischemic heart (Fig. 3C– injection of ET-1 (1 nmol/kg) in 7 P2Y4 and 7 P2Y4 mice (means 6 F). Lower expression of MPO, ICAM-1, E-selectin, and VCAM-1 SEM). *p , 0.05. 4 LOSS OF MOUSE P2Y4 PROTECTS AGAINST MYOCARDIAL INFARCTION

Similarity between the effects of P2Y4 knockout and endothelin antagonist on LPS-induced cardiac microvascular permeability and neutrophil infiltration SYBR Green experiments were then performed using three in- dependent preparations of RNAs isolated from the heart of male 0/+ 0/2 P2Y4 and P2Y4 mice injected or not with LPS. ET-1 is up- regulated in response to LPS, with a lower level of ET-1 in LPS- 0/2 0/+ treated P2Y4 mice compared with LPS-treated P2Y4 mice (Fig. 5A). ET-1 is also quantified by ELISA in the corresponding mouse sera (Fig. 5B). In basal conditions, we observed that ET-1 0/2 0/+ concentration was lower in P2Y4 mice than in P2Y4 mice. Fig. 5B shows that LPS injection increased more significantly the 0/+ 0/– serum concentration of ET-1 in P2Y4 than in P2Y4 mice: 31.8 6 1.3 and 23.8 6 2.2 pg/ml (means 6 SEM), respectively. Intraperitoneal injection of LPS induced a massive recruitment of immune cells such as neutrophils (15- 6 6-fold) in the peri- 0/+ toneal cavity of P2Y4 mice (Fig. 5C). A similar injection of LPS produced a significant but smaller (6- 6 2-fold) neutrophil infil- 0/– Downloaded from tration in P2Y4 mice. Interestingly, when mice were coinjected with LPS (100 mg/kg) and ET-1 (1 nmol/kg), neutrophil recruit- 0/+ 0/– ment was comparable in P2Y4 and P2Y4 mice (Fig. 5C). Injection of mice with the selective ETA receptor antagonist BQ- 123 (1 mg/kg) markedly attenuated neutrophil recruitment in 0/+ 0/– LPS-injected P2Y4 and P2Y4 mice (Fig. 5C).

0/+ http://www.jimmunol.org/ We then investigated vascular permeability in the heart of P2Y4 0/2 and P2Y4 mice using the protocol described by Hele et al. (25). Evan’s blue dye was used after exposure to LPS (30 mg/kg, iv.) alone or combined with ET-1 (1 nmol/kg) or BQ-123 (1 mg/kg). After injection of LPS, vascular leakage was markedly increased FIGURE 3. Defective neutrophil infiltration and endothelial adhesion 0/+ 0/2 in P2Y4 compared with P2Y4 mice (Fig. 5D). As described molecule expression in P2Y4-deficient mice during myocardial infarction. + +/+ above for neutrophil recruitment (Fig. 5C), when mice were (A and B) Quantification of MPO neutrophils in the hearts of P2Y4 and 2/2 coinjected with LPS and ET-1 (1 nmol/kg), cardiac permeability P2Y4 mice 24 h after MI. Heart sections were stained with an Ab 0/+ 0/2 was no longer significantly different in P2Y4 and P2Y4 against MPO. Counting was performed using ImageJ software blindly at by guest on September 25, 2021 original magnification 320 in 10 fields in the infarct area. Results rep- mice (Fig. 5D). Preinjection of mice with the selective ETA resent the means 6 SEM of 10 P2Y +/+ and 10 P2Y 2/2 hearts. Scale bar, receptor antagonist BQ-123 (1 mg/kg) markedly attenuated 4 4 0/+ 0/2 50 mm. *p , 0.05. (C–F) Quantification of MPO, ICAM-1, E-selectin, and vascular permeability in LPS-injected P2Y4 and P2Y4 +/+ 2/2 VCAM-1 in the hearts of P2Y4 and P2Y4 mice 24 h after MI. mice (Fig. 5D, 5E). Quantitative RT-PCR experiments were performed using specific primers of mouse MPO, ICAM-1, E-selectin, and VCAM-1. cDNA was obtained from RNA extracted from the ischemic area or remote region of wild-type Discussion and P2Y4 knockout mice injected or not with ET-1 (1 nmol/kg). PCR data Cardiac endothelial cells and cardiomyocytes are an important were normalized for each gene to RPL13 and HPRT housekeeping genes. source of extracellular nucleotides within the heart, especially *p , 0.05, ***p , 0.001. under ischemia or pressure overload. Release of UTP has been observed in human heart during myocardial infarction (28). The CD31 staining was performed in both remote and infarcted areas involvement of P2Y2 receptors in UTP-mediated cardioprotection +/+ 2/2 of P2Y4 and P2Y4 infarcted hearts, and it revealed a re- in vivo has been previously demonstrated (13–15). P2Y4, de- 2/2 duction of the number of cardiac microvessels in P2Y4 com- scribed as a UTP receptor (16), displays a very restricted tissue +/+ pared with P2Y4 ischemic heart 30 d after LAD ligation (Fig. 4H, distribution including heart and lung, and its physiological role 4I). We also observed a reduction of VEGF expression in heart and has been poorly investigated. We have previously demonstrated 2/2 +/+ serum of P2Y4 ischemic mice compared with P2Y4 ischemic that adult P2Y4-null mice display a microcardia phenotype related mice (Fig. 4J, 4K). to a cardiac angiogenic defect (23) and a reduced exercise toler- To investigate further the role of P2Y4 in cardiac fibrosis, we ance (24). performed the allylamine-induced fibrosis model (Fig. 4F, 4G). The present study identifies a protection against myocardial Allylamine was added in the drinking fluid of male mice during 3, infarction in P2Y4-null mice. The discrepancy with the previously 7, 14 or 30 d. Myocardial fibrosis was then quantified by Masson reported cardioprotective effect of UTP (13–15) can be explained trichrome staining. In allylamine-treated male mice, the magni- by cell-specific actions of UTP. The cardioprotective effect of 0/2 tude of fibrosis was lower in P2Y4 hearts than in wild-type UTP is mediated by P2Y2 receptors expressed on cardiomyocytes, hearts at all tested periods (Fig. 4F). The strongest difference was whereas P2Y4 is expressed on cardiac endothelial cells but is 0/2 observed after 7 d: 3.0 6 0.8 of fibrosis area for P2Y4 hearts versus absent on cardiomyocytes. 0/+ 8.5 6 2.5% for P2Y4 hearts (means 6 SEM; p , 0.001) (Fig. 4F). To investigate further the mechanisms involved in the car- + MPO neutrophils were blindly counted in the hearts of allylamine- dioprotection resulting from P2Y4 deletion, gene profiling treated mice at days 3, 7, 14, and 30 (Fig. 4G). A significant reduction experiments were performed and identified differences in gene +/+ 2/2 of neutrophils was observed after 30 d in P2Y4-null hearts compared regulation between P2Y4 and P2Y4 hearts in ischemia and 0/+ with P2Y4 hearts (Fig. 4G). inflammation models. Among P2Y4 target genes, we focused our The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

+/+ 2/2 FIGURE 4. Impaired fibrosis in P2Y4-null ischemic mice. (A) Typical hearts of P2Y4 and P2Y4 mice 30 d after MI. Frozen sections (7 mm) of +/+ 2/2 whole hearts of P2Y4 and P2Y4 mice were stained with Masson trichrome. Images were taken at original magnification 32.5 (whole hearts) and +/+ 2/2 original magnification 320. (B and C) Quantification of fibrosis in P2Y4 and P2Y4 hearts 14 and 30 d after MI. Fibrosis expressed as collagen staining in blue was quantified by color image analyzer ImageJ in whole hearts (B) and in remote (noninfarcted) and infarcted areas (C) 14 and 30 d after ligation. +/+ Results are expressed as the ratio between collagen area (stained in blue) and the total area. Results represent the means 6 SEM of five P2Y4 mice and 2/2 +/+ 2/2 five P2Y4 mice. *p , 0.05, **p , 0.01. (D–G) Quantification of MMP-8, MMP-9, TIMP-1, and TIMP-4 in the hearts of P2Y4 and P2Y4 mice 24 h after MI. Quantitative RT-PCR experiments were performed using specific primers of mouse MMP-8, MMP-9, TIMP-1, and TIMP-4. cDNA was obtained from RNA extracted from the ischemic area or remote region of wild-type and P2Y4 knockout mice. PCR data were normalized for each gene to RPL13 and HPRT +/+ 2/2 housekeeping genes. *p , 0.05, **p , 0.01. (H and I) Vessel quantification in P2Y4 and P2Y4 hearts 30 d after MI. CD31 staining was quantified by color +/+ 2/2 image analyzer ImageJ in remote and infarcted areas 30 d after ligation. Results represent the means 6 SEM of 8 P2Y4 mice and 8 P2Y4 mice. Scale bars, +/+ 2/2 +/+ 50 mm. **p , 0.01, ***p , 0.001. (J and K) Quantification of VEGF in heart and serum of P2Y4 and P2Y4 ischemic mice. Heart cDNA and sera of P2Y4 2/2 and P2Y4 mice were collected 24 h after LAD ligation for quantitative PCR analysis (J) or ELISA measurements (K)ofVEGF. (Figure legend continues) 6 LOSS OF MOUSE P2Y4 PROTECTS AGAINST MYOCARDIAL INFARCTION

Lower ET-1 level in P2Y4-null mice is thus compatible with a protection against infarction. Reduced levels of ET-1, which are mainly produced by endothelial cells (32), could be explained by 0/2 the defective cardiac angiogenesis observed in P2Y4 heart observed in the present study. Identification of ET-1 as a P2Y4 target gene led us to study in vivo neutrophil trafficking in P2Y4-null mice. Indeed, ET-1 is known to increase vascular permeability and neutrophil adhesion on cardiac endothelial cells through increased expression of ad- hesion molecules such as ICAM-1 and E-selectin (5). Exposure to LPS is associated with changes in vascular reactivity and elevated plasma levels of ET-1 (33). A defective neutrophil accumulation linked to a reduced ET-1 increase was observed in LPS-injected P2Y4-null mice and was restored by ET-1 injection. Administra- tion of the ETA antagonist BQ-123 had an inhibitory effect on neutrophil level in peritoneal cavity, confirming that ET-1 is markedly involved in neutrophil trafficking (5). Nucleotides and especially ATP were previously described as major mediators of

inflammation. Chen et al. (34) described that ATP release and Downloaded from autocrine feedback through P2Y2 and A3 receptors provide signal amplification, controlling gradient sensing and migration of neu- trophils. In the present study, we observed a defect in neutrophil infiltration in the P2Y4-null mice compared with P2Y4 wild-type mice in all tested models, that is, LAD ligation, LPS-induced

inflammation, and allylamine-induced fibrosis models. These http://www.jimmunol.org/ FIGURE 5. Similarity between the effects of P2Y4 knockout and data define P2Y as a major regulator of neutrophil recruitment in endothelin antagonist on LPS-induced cardiac microvascular perme- 4 inflammatory and ischemic conditions. Reduced neutrophil infil- ability and neutrophil infiltration. (A) Quantitative RT-PCR analysis of 0/+ 0/2 trate was correlated with reduced endothelial adhesion molecule ET-1 expression in the heart of P2Y4 and P2Y4 mice. Ratios were obtained in basal conditions and 6 h after LPS injection for three inde- expression in P2Y4-null compared with P2Y4 wild-type ischemic pendent preparations of mRNA (means 6 SEM) using SYBR Green heart. Defective expression of ET-1 and endothelial adhesion technology and normalized using two housekeeping genes (RPL13 and molecules, which play a pivotal role in neutrophil infiltration, HPRT). *p , 0.05, **p , 0.01. (B) Quantification of ET-1 serum level in appears to be the consequence of reduced angiogenesis observed 0/+ 0/2 0/+ 0/2 P2Y4 and P2Y4 mice. Sera of P2Y4 and P2Y4 mice, treated or in P2Y4-null mice. not with LPS, were collected for ELISA measurements of ET-1. Results Cardiac fibrosis induced by ischemia was strongly reduced in by guest on September 25, 2021 represent the means 6 SEM of eight mice. *p , 0.05, ***p , 0.001. (C) 0/+ P2Y4-null mice compared with P2Y4 wild-type mice. The mag- Neutrophil infiltration in the peritoneal cavity of LPS-injected P2Y4 0/2 nitude of that reduction was larger than the decrease in infarct size and P2Y4 mice. ET-1 (1 nmol/kg) or BQ-123 (1 mg/kg) were injected 0/+ 0/2 and is likely to be the consequence of the decreased infiltration with LPS (100 mg/kg, i.p.) in P2Y4 and P2Y4 mice. Four hours later, neutrophils harvested after peritoneal lavage were counted for of neutrophils and reduced expression of members of the MMP/ 0/+ 0/2 TIMP system such as MMP-9 and neutrophil collagenase, MMP-8. P2Y4 and P2Y4 mice (means 6 SEM, n =6).*p , 0.05, ***p , 0.001. (D and E) Effect of ET-1 and BQ-123 on LPS-induced protein MMPs and TIMPs play major roles in cardiac ischemia through 0/+ 0/2 extravasation in P2Y4 and P2Y4 hearts. LPS (30 mg/kg) was their action in extracellular matrix remodelling (27). MMP-8 injected with or without ET-1 (1 nmol/kg) or BQ-123 (1 mg/kg) 5 min contributes to the generation of chemotactic tripeptide of the ex- 0/+ 0/2 before injection of Evan’s blue dye (20 mg/kg). P2Y4 and P2Y4 tracellular matrix Pro-Gly-Pro that binds CXCR2 and mediates mice were killed 10 min after injection. The values are means 6 SEM of neutrophil migration (27). MMP-9, also named gelatinase B, is an m , , five experiments. Scale bars, 50 m. **p 0.01, ***p 0.001. important player in cardiovascular diseases such as atherosclero- sis, stroke, and heart failure (35). Interestingly, regulation of attention on ET-1, a gene regulated in P2Y4-null ischemic and MMPs such as MMP-9 by ET-1 has previously been reported (36). inflamed hearts. We confirmed by quantitative PCR and ELISA The initial increase in TIMP-1 and TIMP-4 mRNAs observed in 2/2 that ET-1 level was decreased in P2Y4 hearts compared with wild-type mice at 24 h after MI was previously reported and de- +/+ P2Y4 hearts in both LAD ligation and LPS-induced inflam- creased in the ischemic myocardium after 1 wk (27). Lower levels mation models. Interestingly, this protein displays effects on in- of TIMPs in P2Y4-null mice could contribute to the observed flammation and vasoconstriction (29), particularly during myocardial reduced fibrosis. However, even if reduced MMP-9 and MMP-8 infarction. ET-1 plays an important role in myocyte hypertrophy levels in P2Y4-null mice are compatible with reduction of ET-1 after myocardial infarction (30, 31). Studies have shown that and neutrophil infiltration, the link with lower fibrosis quantified plasma ET-1 levels are elevated in the coronary circulation during after several weeks is complex and depends on MMP/TIMP time ischemia and that the coronary artery is very sensitive to ET-1 courses during ischemia experiments. (31). More recently, an association between ET-1 and chronic The fact that there was no significant beneficial effect on EF in heart failure was described (26). P2Y4-null ischemic mice may suggest a contractile adaptation to

Results represent the means 6 SEM of eight mice. *p , 0.05, **p , 0.01. (L and M) Quantification of cardiac fibrosis and neutrophil infiltration in allylamine-induced fibrosis model. Fibrosis was quantified by ImageJ. Results are expressed as the ratio between collagen area and the total area (D). MPO+ 0/+ 0/2 neutrophils were blindly counted at original magnification 320 in 10 fields on transversal sections in the hearts of P2Y4 and P2Y4 mice (E). Results 0/+ 0/2 represent the means 6 SEM of 6 P2Y4 mice and 6 P2Y4 mice. *p , 0.05, **p , 0.01, ***p , 0.001. The Journal of Immunology 7 preserve cardiac output. The echocardiography experiments per- 10. Carmeliet, P., and D. Collen. 2000. Transgenic mouse models in angiogenesis and cardiovascular disease. J. Pathol. 190: 387–405. formed 7 and 14 d after ligation revealed reduction of LVED 11. Matsui, Y., J. Morimoto, and T. Uede. 2010. Role of matricellular proteins in and LVES in P2Y4-null compared with wild-type ischemic mice. cardiac tissue remodeling after myocardial infarction. World J Biol Chem 1: 69– These data suggest an attenuated remodeling in P2Y4-null mice 80. 12. Dutta, A. K., R. Z. Sabirov, H. Uramoto, and Y. Okada. 2004. Role of ATP- that is compatible with the reduction of inflammation, perme- conductive anion channel in ATP release from neonatal rat cardiomyocytes in ability, and fibrosis observed in these mice. ischaemic or hypoxic conditions. J. Physiol. 559: 799–812. ET-1 can also enhance vascular permeability in selected rat 13. Yitzhaki, S., V. Shneyvays, K. A. Jacobson, and A. Shainberg. 2005. Involve- ment of uracil nucleotides in protection of cardiomyocytes from hypoxic stress. vascular beds, including the pulmonary, renal, and mesenteric Biochem. Pharmacol. 69: 1215–1223. circulation (37). It appeared that LPS-induced cardiac microvas- 14. Yitzhaki, S., A. Shainberg, Y. Cheporko, B. A. Vidne, A. Sagie, K. A. Jacobson, cular hyperpermeability, measured after Evan’s blue injection, and E. Hochhauser. 2006. Uridine-59-triphosphate (UTP) reduces infarct size and improves rat heart function after myocardial infarct. Biochem. Pharmacol. 72: was reduced in P2Y4-null mice compared with wild-type 949–955. mice. Moreover, injection of ET-1 could restore vascular hyper- 15. Cohen, R., A. Shainberg, E. Hochhauser, Y. Cheporko, A. Tobar, E. Birk, L. Pinhas, J. Leipziger, J. Don, and E. Porat. 2011. UTP reduces infarct size and permeability in the coronary circulation of P2Y4-null mice to improves mice heart function after myocardial infarct via P2Y2 receptor. Bio- a level comparable to that of wild-type mice. As for neutrophil chem. Pharmacol. 82: 1126–1133. infiltration, BQ-123 administration had an inhibitory effect, con- 16. Communi, D., S. Pirotton, M. Parmentier, and J. M. Boeynaems. 1995. Cloning firming ETA receptor involvement in the observed permeability and functional expression of a human uridine nucleotide receptor. J. Biol. Chem. 270: 30849–30852. increase. The reduced cardiac permeability observed in P2Y4-null 17. Bogdanov, Y. D., S. S. Wildman, M. P. Clements, B. F. King, and G. Burnstock. mice appears thus to be the consequence of lower ET-1 expres- 1998. Molecular cloning and characterization of rat P2Y4 nucleotide receptor. sion. Early myocardial edema is a characteristic feature of the Br. J. Pharmacol. 124: 428–430. Downloaded from 18. Webb, T. E., D. J. Henderson, J. A. Roberts, and E. A. Barnard. 1998. Molecular inflammatory reaction associated with acute myocardial ischemia cloning and characterization of the rat P2Y4 receptor. J. Neurochem. 71: 1348–1357. (38). Weis et al. (39) have demonstrated that, prior to promoting 19. Lazarowski, E. R., L. G. Rochelle, W. K. O’Neal, C. M. Ribeiro, B. R. Grubb, revascularization, VEGF, whose expression is reduced in P2Y V. Zhang, T. K. Harden, and R. C. Boucher. 2001. Cloning and functional 4 characterization of two murine uridine nucleotide receptors reveal a potential ischemic mice, has a profound and immediate effect on ischemic target for correcting ion transport deficiency in cystic fibrosis gallbladder. tissues by inducing a permeability response leading to edema that J. Pharmacol. Exp. Ther. 297: 43–49. 20. Lustig, K. D., A. K. Shiau, A. J. Brake, and D. Julius. 1993. Expression cloning can be detrimental to the ischemic tissue. They demonstrated that http://www.jimmunol.org/ of an ATP receptor from mouse neuroblastoma cells. Proc. Natl. Acad. Sci. USA mice deficient in pp60Src show a normal angiogenic response to 90: 5113–5117. VEGF, but no VEGF-mediated permeability response, and they 21. Cheung, K. K., M. Ryten, and G. Burnstock. 2003. Abundant and dynamic expression of G protein-coupled P2Y receptors in mammalian development. Dev. consequently develop minimal infarcts. In a similar way, the P2Y4 Dyn. 228: 254–266. receptor appears to be able to regulate both cardiac angiogenesis 22. Robaye, B., E. Ghanem, F. Wilkin, D. Fokan, W. Van Driessche, S. Schurmans, (23) and permeability. J. M. Boeynaems, and R. Beauwens. 2003. Loss of nucleotide regulation of epithelial chloride transport in the jejunum of P2Y4-null mice. Mol. Pharmacol. The present study defines P2Y4 as an important regulator of cardiac 63: 777–783. inflammation and fibrosis in response to ischemia. Our data identify 23. Horckmans, M., B. Robaye, E. Le´on-Gόmez, N. Lantz, P. Unger, F. Dol-Gleizes, ET-1, a protein involved in cardiac ischemia and associated in chronic S. Clouet, D. Cammarata, P. Schaeffer, P. Savi, C. Gachet, J.L. Balligand,

C. Dessy, J.M. Boeynaems, and D. Communi. 2012. P2Y4 nucleotide receptor: by guest on September 25, 2021 heart failure, as P2Y4 target genes in the heart. The protection against a novel actor in post-natal cardiac development and function. Angiogenesis 15: ischemia observed in mice deficient in this nucleotide receptor sug- 349–360. 24. Horckmans, M., E. Le´on-Go´mez, B. Robaye, J. L. Balligand, J. M. Boeynaems, gests a therapeutic potential for P2Y4 antagonists to limit the early C. Dessy, and D. Communi. 2012. Gene deletion of P2Y4 receptor lowers ex- damage of myocardial infarction. ercise capacity and reduces myocardial hypertrophy with swimming exercise. Am. J. Physiol. Heart Circ. Physiol. 303: H835–H843. Acknowledgments 25. Hele, D. J., M. A. Birrell, S. E. Webber, M. L. Foster, and M. G. Belvisi. 2000. We thank Sophie De Man and Anne Lefort for help and technical assistance Effect of endothelin antagonists, including the novel ETA receptor antagonist LBL 031, on endothelin-1 and lipopolysaccharide-induced microvascular leak- in echocardiographic and microarray experiments, respectively. age in rat airways. Br. J. Pharmacol. 131: 1129–1134. 26. Yin, W. H., Y. H. Chen, J. Wei, H. L. Jen, W. P. Huang, M. S. Young, D. C. Chen, and P. L. Liu. 2011. Associations between endothelin-1 and adiponectin in Disclosures chronic heart failure. Cardiology 118: 207–216. The authors have no financial conflicts of interest. 27. Spinale, F. G. 2007. Myocardial matrix remodeling and the matrix metal- loproteinases: influence on cardiac form and function. Physiol. Rev. 87: 1285–1342. 28. Wihlborg, A. K., J. Balogh, L. Wang, C. Borna, Y. Dou, B. V. Joshi, References E. Lazarowski, K. A. Jacobson, A. Arner, and D. Erlinge. 2006. Positive ino- tropic effects by uridine triphosphate (UTP) and uridine diphosphate (UDP) via 1. Saraste, A., K. Pulkki, M. Kallajoki, K. Henriksen, M. Parvinen, and L. M. Voipio- P2Y2 and P2Y6 receptors on cardiomyocytes and release of UTP in man during Pulkki. 1997. Apoptosis in human acute myocardial infarction. Circulation 95: 320–323. myocardial infarction. Circ. Res. 98: 970–976. 2. Frangogiannis, N. G. 2006. Targeting the inflammatory response in healing 29. Giannessi, D., S. Del Ry, and R. L. Vitale. 2001. The role of endothelins and myocardial infarcts. Curr. Med. Chem. 13: 1877–1893. their receptors in heart failure. Pharmacol. Res. 43: 111–126. 3. Garcia-Dorado, D., and J. Oliveras. 1993. Myocardial oedema: a preventable 30. Yousef, Z. R., S. R. Redwood, and M. S. Marber. 2000. Postinfarction left cause of reperfusion injury? Cardiovasc. Res. 27: 1555–1563. ventricular remodeling: a pathophysiological and therapeutic review. Car- 4. Lo´pez Farre´,A.,A.Riesco,G.Espinosa,E.Digiuni,M.R.Cernadas,V.Alvarez, diovasc. Drugs Ther. 14: 243–252. M. Monto´n, F. Rivas, M. J. Gallego, J. Egido, et al. 1993. Effect of endothelin-1 on 31. Masaki, T., S. Kimura, M. Yanagisawa, and K. Goto. 1991. Molecular and neutrophil adhesion to endothelial cells and perfused heart. Circulation 88: 1166–1171. cellular mechanism of endothelin regulation. Implications for vascular function. 5. Zouki, C., C. Baron, A. Fournier, and J. G. Filep. 1999. Endothelin-1 enhances Circulation 84: 1457–1468. neutrophil adhesion to human coronary artery endothelial cells: role of ETA 32. Kohno, M., K. Yokokawa, T. Horio, K. Yasunari, K. Murakawa, M. Ikeda, and receptors and platelet-activating factor. Br. J. Pharmacol. 127: 969–979. T. Takeda. 1992. Release mechanism of endothelin-1 and big endothelin-1 after 6. Detillieux, K. A., F. Sheikh, E. Kardami, and P. A. Cattini. 2003. Biological activities stimulation with thrombin in cultured porcine endothelial cells. J. Vasc. Res. 29: of fibroblast growth factor-2 in the adult myocardium. Cardiovasc. Res. 57: 8–19. 56–63. 7. Virag, J. I., and C. E. Murry. 2003. Myofibroblast and endothelial cell prolif- 33. Fabricio, A. S. C., G. A. Rae, P. D’Orle´ans-Juste, and G. E. P. Souza. 2005. eration during murine myocardial infarct repair. Am. J. Pathol. 163: 2433–2440. Endothelin-1 as a central mediator of LPS-induced fever in rats. Brain Res. 1066: 8. Dobaczewski, M., M. Bujak, P. Zymek, G. Ren, M. L. Entman, and 92–100. N. G. Frangogiannis. 2006. Extracellular matrix remodeling in canine and 34. Chen, Y., R. Corriden, Y. Inoue, L. Yip, N. Hashiguchi, A. Zinkernagel, V. Nizet, mouse myocardial infarcts. Cell Tissue Res. 324: 475–488. P. A. Insel, and W. G. Junger. 2006. ATP release guides neutrophil chemotaxis 9. Tyagi, S. C., S. Kumar, S. Cassatt, and J. L. Parker. 1996. Temporal expression of via P2Y2 and A3 receptors. Science 314: 1792–1795. extracellular matrix metalloproteinases and tissue plasminogen activator in the 35. Chow, A. K., J. Cena, and R. Schulz. 2007. Acute actions and novel targets of development of collateral vessels in the canine model of coronary occlusion. matrix metalloproteinases in the heart and vasculature. Br. J. Pharmacol. 152: Can. J. Physiol. Pharmacol. 74: 983–995. 189–205. 8 LOSS OF MOUSE P2Y4 PROTECTS AGAINST MYOCARDIAL INFARCTION

36. Wang, H. H., H. L. Hsieh, and C. M. Yang. 2011. Nitric oxide production by 38. Entman, M. L., L. Michael, R. D. Rossen, W. J. Dreyer, D. C. Anderson, endothelin-1 enhances astrocytic migration via the tyrosine nitration of matrix A. A. Taylor, and C. W. Smith. 1991. Inflammation in the course of early metalloproteinase-9. J. Cell. Physiol. 226: 2244–2256. myocardial ischemia. FASEB J. 5: 2529–2537. 37. Filep, J. G., M. G. Sirois, A. Rousseau, A. Fournier, and P. Sirois. 1991. 39. Weis, S., S. Shintani, A. Weber, R. Kirchmair, M. Wood, A. Cravens, Effects of endothelin-1 on vascular permeability in the conscious rat: H. McSharry, A. Iwakura, Y. S. Yoon, N. Himes, et al. 2004. Src blockade interactions with platelet-activating factor. Br.J.Pharmacol.104: 797– stabilizes a Flk/cadherin complex, reducing edema and tissue injury following 804. myocardial infarction. J. Clin. Invest. 113: 885–894. Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021