Methods and Pharmaceutical Composition for the Preservation of Vascular Endothelial Cell Barrier Integrity

Abstract

The invention relates to an ANGPTL4 polypeptide for use in the preservation of vascular endothelial cell barrier integrity and reduction in no-reflow phenomenon with myocardial infarction.

Claims

1-11. (canceled)

12. A method of treating acute myocardial infarction in a subject in need thereof by reducing infarct size, hemorrhage and no-reflow, comprising the steps of i) restoring blood supply in cardiac ischemic tissue of said subject, and ii) preserving the vascular endothelial cell barrier integrity of said cardiac ischemic tissue by administering to said subject a therapeutically effective amount of human recombinant ANGPTL4 polypeptide or a function conservative variant thereof, said therapeutically effective amount being sufficient to reduce infarct size, hemorrhage and no-reflow in said subject.

13. The method of claim 12, wherein steps i) and ii) are performed sequentially.

14. The method of claim 12, wherein steps i) and ii) are performed concomitantly.

15. The method of claim 12, wherein human recombinant ANGPTL4 polypeptide is full-length ANGPTL4.

16. The method of claim 12 wherein said subject is human.

17. The method of claim 12 wherein said subject is a non-human animal.

18. A method of testing a patient thought to have or be predisposed to microvascular dysfunction, which comprises the step of analyzing a biological sample from said patient for: (i) detecting the presence of a mutation in the gene encoding for ANGPTL4 and/or its associated promoter, and/or (ii) analyzing the expression of the gene encoding for ANGPTL4.

19. The method according to claim 18 wherein said biological sample is blood or serum.

Description

FIGURES

[0089] FIG. 1: Infarct sizes (expressed as a percentage of the area at risk) in control rabbits and in 100 ng/ml human recombinant ANGPTL4 injected-rabbits. Open circles represent individual infarct size and closed circles represent meant SEM. *p<0.05 (A). Extent of the no-reflow zone expressed as a percentage of infarct size (B) or expressed as a percentage of the area at risk (C) in control rabbits and in 100 ng/ml human recombinant ANGPTL4 injected-rabbits. Open circles represent individual no-reflow zone and closed circles represent mean±SEM. *p<0.05. Microscopic images of rabbit infarcted myocardium (D, scale bar=250 μm). HE staining revealed diffused ischemic areas characterized by hemorrhages shown by arrow (A) which were more prominent in control rabbits compared to 100 ng/ml human recombinant ANGPTL4 injected-rabbits.

EXAMPLE 1: CARDIOPROTECTION THROUGH PRESERVATION OF VASCULAR ENDOTHELIAL CELL BARRIER INTEGRITY BY ANGIOPOIETIN-LIKE 4

[0090] Material & Methods:

[0091] The experiments were performed in accordance with the official regulations edited by the French Ministry of Agriculture. This study conforms to the standards of INSERM (the French National Institute of Health) regarding the care and use of laboratory animals, was performed in accordance with European Union Council Directives (86/609/EEC).

[0092] Animals and Genotyping:

[0093] Genotype was determined by PCR of tail genomic DNA using the following conditions: denaturation at 94° C. for 30 seconds, annealing at 56° C. for 45 seconds, and extension at 72° C. for 1 minute and 15 seconds, 30 cycles. Wild-type LacZ LacZ (angptl4.sup.+/+), angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ knock-out C57BL/6 mice, 8 to 12 weeks of age, were subjected to myocardial infarction protocols or used as control in basal conditions.

[0094] Myocardial Ischemia-Reperfusion Experiments:

[0095] Male angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ mice were anesthetized by an intraperitoneal injection of sodium pentobarbital. Myocardial infarction with transient occlusion of the left coronary artery was performed for 45 mn and then tissues were reperfused for 1 h to 3 weeks. For angptl4 expression study WT mice underwent 1 h, 3 h, 24 h, 48 h, 72 h, 1 week, 2 weeks, 3 weeks of reperfusion (2 animals per group). To assess infarct size and for immunohistochemistry (IHC) or ultrastructural studies, male angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ mice were reperfused during either 4 h or 48 h after ischemia (4 or 5 animals per group). For permeability analyses, only 4 h of reperfusion were performed with angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ mice (3 to 5 animals per group). The area at risk was identified by Evans blue staining, and the infarct area was identified by 2,3,5-triphenyltetrazolium chloride (TTC) staining. The area at risk was identified as the nonblue region and expressed as a percentage of the left ventricle weight. The infarcted area was identified as the TTC-negative zone and expressed as a percentage of the area at risk. Ultrastructural analyses were performed on a Hitachi H-9500 electron microscope.

[0096] Rabbits Experiments:

[0097] New Zealand rabbits (2.5-3.0 kg) were anesthetized using zolazepam, tiletamine and pentobarbital (all 20-30 mg/kg i.v.). The animals were intubated, mechanically ventilated and a left thoracotomy was performed. A suture was passed beneath a major branch of the left coronary artery. The ends of the ligature were passed through a short propylene tubing to form a snare. Rabbits then randomly received either vehicle or human recombinant angptl4 (10 μg/kg i.v.; 100 ng/ml). Five minutes after, coronary artery occlusion (CAO) was induced during 30-min by pulling the snare through the tubing. Reperfusion was subsequently induced by releasing the snare. The chest was then closed in layers. Four hours after the onset of reperfusion, the chest was reopened and thioflavine S (4%; 1.5 ml/kg) was rapidly infused through the left atrium. Rabbits were then sacrificed using pentobarbital followed by potassium chloride. After excision, the hearts were perfused retrogradely with Alcian blue (0.5%) and cut into slices. Slices were photographed using UV light to identify the region of no-reflow. The areas of no-reflow, infarct and risk zone were determined as in mice.

[0098] In Situ Hybridization (ISH) and Immunohistochemistry (IHC) Analysis:

[0099] Full length human or mouse angptl4 ORF was amplified by PCR, subcloned in pGem.T (Promega) and used to generate a .sup.35S-RNA antisense mouse angptl4 probe. A sense probe was used as a negative control.and used as a template. ISH using sense probe, antisense human or mouse angptl4 probes and IHC immunolabellings anti-CD45, -Mac3, and -CD31 experiments were performed as previously described (Brechot, N. et al. Modulation of macrophage activation state protects tissue from necrosis during critical limb ischemia in thrombospondin-1-deficient mice. PLoS ONE 3, e3950 (2008)). ISH detecting human angptl4 combined to IHC using anti-CD34 antibody (R&D system) were performed on human infarcted hearts sections.

[0100] Modified Miles Assay:

[0101] Male angptl4.sup.−/− and angptl4.sup.+/− mice were anesthetized with pentobarbital. For basal conditions, mice were injected into the tail vein with 1% Evans blue (200 μl) and sacrificed 4 h later. For the ischemia-reperfusion conditions, mice were subjected to coronary occlusion for 45 mn and intravenously injected with 1% Evans blue (200 μl) prior the 4 h of reperfusion. At sacrifice, mice were perfused through the aorta with citrate buffer pH4 using a 18-gauge cannula. Blood, dye and buffer exited through an opening in the right atrium. Evans Blue was eluted for 18 h at 70° C. in 1 ml formamide. After centrifugation, absorbance at 620 nm was measured using a spectrophotometer. Extravasated Evans blue (ng) was determined from a standard curve and normalized to tissue weight (g). Data are presented as mean±s.e.m. of 3 or 5 mice per genotype.

[0102] Immunofluoresence Study and Confocal Analyzis on Cryosections:

[0103] Mice subjected to ischemia and 4 h of reperfusion were anesthetized with ketamine and xylazine injected intraperitonally. FITC-beads (20 μl) were injected into the femoral vein as previously described (Baffert, F., Le, T., Thurston, G. & McDonald, D. M. Angiopoietin-1 decreases plasma leakage by reducing number and size of endothelial gaps in venules. Am J Physiol Heart Circ Physiol 290, H107-18 (2006)). The chest was opened rapidly, and the vasculature was perfused for 2 minutes at a pressure of 120 mmHg with 1% paraformaldehyde as described for the modified Miles assay. The heart was then placed into 1% paraformaldehyde for 1 h at room temperature, rinsed several times with PBS, infiltrated overnight with 30% sucrose, and frozen for cryostat sectioning. Sections 80 m in thickness were incubated in 5% normal goat serum in PBS containing 0.3% Triton X-100, 0.2% bovine serum albumin, for 1 hour at RT to block nonspecific antibody binding and were incubated O/N with the primary antibodies. Endothelial cells, pericytes and adherens junctions were identified with rat anti-CD31 (BD Pharmingen), rabbit anti-NG2 (Chemicon) and rat anti-VE-Cadherin (personal gift from E. Dejana, IFOM) antibodies, respectively. After several rinses with PBS, specimens were incubated for 3 h at room temperature with fluorescent (Alexa 555 and Alexa 647) secondary antibodies (anti-rat or anti-rabbit; Molecular Probes) diluted 1:500. Finally specimens were mounted in Vectashield (Vector Laboratories). Confocal images were acquired by using a Leica TCS SP5 microscope. Confocal sections were imaged on a Leica SP5 microscope (Leica Microsystems GmbH) using a 63× (NA=1.4) oil objective. An increment of 0.117 m between each section was used. 3D reconstruction of the different structures were obtained using the LABELVOXEL and the SURFACEGEN modules in Amira 5.2.1 software (Visage Imaging GmbH).

[0104] Isolation of Cardiomyocytes and Viability Assay:

[0105] Under anesthesia, the heart was removed from the chest and was cannulated. The heart was perfused for 4 min with tyrode buffer ([mM] NaCl 113 KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; with pH adjusted to 7.46 with NaOH 1N) at constant pressure and 37° C. Perfusion was switched to an enzyme solution ([mM] NaCl 113; KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; CaCl.sub.2 0.0125; with pH ajusted to 7.46 with NaOH 1N) containing 0.1 mg/ml liberase blendzyme IV, (Roche diagnostics) and 0.14 mg/ml trypsin (Sigma). When hearts became swollen and turn slightly pale, the atria and aorta were removed; the left ventricle were cut into small pieces and gently triturated. Cell suspension was transferred into a stopping buffer ([mM] NaCl 113; KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; CaCl.sub.2 0.0125; calf serum 5%; with pH adjusted to 7.46 with NaOH 1N). Extracellular calcium was added incrementally up to 1.0 mM. All cells studied were rod-shaped, had clear cross-striations and lacked any visible vesicles on their surfaces under observations with an optical microscope. Cardiac myocytes were incubated in an anaerobic chamber containing a humidified atmosphere of 5% CO.sub.2 and 95% N.sub.2 for 3 h. Experimental medium was changed to serum-free, glucose-free. Normoxic experimental medium was equilibrated in water-jacketed incubators in a humidified atmosphere of 5% CO.sub.2 and air. Cardiac myocyte survival was measured by staining cells in tissue culture dishes with trypan blue solution (Sigma, Saint-Quentin Fallavier, France) diluted to a final concentration of 0.4%. Myocytes were visualized using brightfield microscopy at 100× magnification. The number of viable (unstained) and non-viable (blue stained) cardiac myocytes in three random microscopic fields was recorded, and at least 100 cells were counted in each field. Percent survival was defined as the number of unstained myocytes counted in each dish divided by the number of total (unstained and stained) myocytes in each condition.

[0106] Ultrasound Analyzis of Cardiac Parameters:

[0107] Mice were subjected to ultrasound measurements using an echocardiograph (Vivid 7, GE Medical Systems Ultrasound) equipped with a 12-MHz linear transducer.

[0108] Statistical Analyses:

[0109] Analyses were performed using StatView Software (version 5.0; Abacus Concepts, Inc., Berkeley, Calif.).

[0110] Results:

[0111] ANGPTL4 Expression after Myocardial Infarction in Humans:

[0112] As ANGPTL4 is induced by hypoxia and expressed in different pathological conditions in humans, such as critical hindlimb ischemia and in numerous types of tumours, we here sought to investigate whether ANGPTL4 is expressed in the human heart after myocardial ischemia. Tissue sections from patients who died from an acute myocardial infarction, obtained from the Pathology Department of Georges Pompidou European Hospital, Paris, France, allowed to visualize the infarcted area. Using ISH, angptl4 mRNA expression was analyzed in following sections. A patchy staining pattern surrounding the infarcted areas was observed. Analysis at higher magnification showed angptl4 mRNA expression was induced in cardiomyocytes, in mononuclear inflammatory cells, likely macrophages and in endothelial cells positive for CD34.

[0113] Infarct Size and Tissue Damages are Increased in Angptl4.sup.LacZ/LacZ Mice:

[0114] angptl4 mRNA expression was then analyzed in a mouse model of myocardial ischemia reperfusion in C57/B16 mice. It could be detected in situ in the infarcted area by ISH as soon as 3 h following ischemia and up to 2 weeks after reperfusion. To determine the functional role that ANGPTL4 might play during myocardial infarction, angptl4 knockout mice in which the angptl4 locus was replaced by a lacZ reporter gene were generated and intercrossed in C57/B16 mice for more than 8 generations. Some lethality was observed during development but surviving angptl4.sup.LacZ/LacZ neonates, which were obtained at ˜25% of the expected frequency, were viable and fertile and did not exhibit functional cardiac defects.

[0115] Myocardial ischemia-reperfusion was performed in angptl4.sup.Lac/+ and in angptl4.sup.LacZ/LacZ mice. Infarct size expressed as a percentage of the risk zone (IS, % area at risk) was increased in angptl4.sup.LacZ/LacZ mice (47±3%) compared to control mice (36±3%).

[0116] We then quantified hemorrhage, edema, inflammation on HE slides from infarcted heart sections in both genotypes. In accordance with increased infarct size, histological analyses of HE-stained sections revealed that tissue necrosis was largely increased in angptl4.sup.LacZ/LacZ compared to control mice (histological score of 2.5 in angptl4.sup.LacZ/LacZ versus 1.2 in angptl4.sup.LacZ/+ mice). A more severe tissue injury in angptl4.sup.LacZ/LacZ compared to angptl4.sup.LacZ/+ mice was also quantified in term of edema (2.3 versus 0.8), hemorrhages (2.3 versus 1.1) and inflammation (2.8 versus 1.1).

[0117] Post-Ischemic Inflammation but not Capillary Density is Modulated in angptl4.sup.LacZ/LacZ Mice:

[0118] As post-ischemic inflammation response may also influence tissue damage, we then analyzed macrophage density using Mac3 immunostainings both in infarcted and non-infarcted areas delineated on adjacent HE-stained sections, in both genotypes. Macrophage density was significantly higher in infarcted areas in angptl4.sup.LacZ/LacZ versus angptl4.sup.LacZ/+ mice (mean macrophage density reached 1822±117 versus 438±71/mm.sup.2, p=0.017) whereas no statistical difference was observed between both groups in control non-infarcted areas (157±120 versus 111±48/mm.sup.2, p=0.33).

[0119] Adult cardiac vascular network was also analyzed using CD31 and no difference was quantified between both genotypes neither in basal conditions nor in the infarcted areas (4510±124 versus 4380±117 mean capillary density/mm.sup.2, respectively, p=0.22). In the most altered infarcted areas (central infarcted areas), a decreased microcapillary density was similarly observed in both genotypes (mean capillary density reached 1115±56 in angptl4.sup.LacZ/LacZ versus 1030±48/mm.sup.2 in angptl4.sup.LacZ/+, p=0.16).

[0120] Cardiomyocyte Viability from Angptl4.sup.LacZ/LacZ Mice is not Affected by Hypoxlia:

[0121] In vitro and in vivo experiments were then performed in order to decipher which mechanisms might be responsible for increased infarct size and tissue damages in angptl4.sup.LacZ/LacZ mice. As angptl4 mRNA was induced by hypoxia mediated by hypoxia-inducible factor 1 in cardiomyocytes, we first hypothezised a direct autocrine effect of ANGPTL4. Cardiomyocytes from both genotypes were therefore isolated and subjected to an hypoxia in vitro survival assay. Cardiomyocytes viability was determined either in normoxia (4 h at 20% O.sub.2) or in hypoxia (3 h at 1% O.sub.2 followed by 1 h at 20% O.sub.2). Contrasting with the in vivo results on myocardial infarct size, no difference in survival was observed between both groups either in normoxia or in hypoxia suggesting that the protective effects of ANGPTL4 upon infarction was not related to a direct protection on cardiomyocytes but rather related to an effect on the coronary vasculature during ischemia-reperfusion.

[0122] Ischemia-Reperfusion Induced Severe Vascular Alterations in Angptl4 Ko Mice:

[0123] Transmission electron microscopy studies were then performed in order to compare the post-ischemic tissue injury in angptl4.sup.LacZ/LacZ mice and angptl4.sup.LacZ/+ mice at the ultrastructural level. Reperfused cardiomyocytes exhibited no differences between both groups. Ultrastructural analysis showed large edematous areas, poor in inflammatory cells after 48 h reperfusion in infarcted areas in angptl4.sup.LacZ/+ mice, whereas inflammatory cells had invaded the edematous region in angptl4.sup.LacZ/LacZ mice. Polynuclear neutrophils, macrophages, lymphocytes and fibrinogen deposits were observed. Whereas angptl4.sup.LacZ/+ mice showed normal pericytes coverage around endothelial cells, a large edematous space between endothelial cells and pericytes was observed in vessels from angptl4.sup.LacZ/LacZ mice. Altogether, these data show increased vascular alterations that correlate with increased inflammatory infiltrate and infarct size in angptl4.sup.LacZ/LacZ mice.

[0124] Assessment of No-Reflow Areas and Therapeutical Potential of ANGPTL4:

[0125] As vascular alterations are associated with no-reflow, we further analyzed no-reflow in angptl4.sup.LacZ/LacZ and angptl4.sup.LacZ/+ mice. The anatomic zone of no-reflow was delineated by thioflavine S staining which binds the intact endothelium causing perfused tissue to fluoresce when exposed to ultraviolet light. The zone of no-reflow expressed as a percentage of the necrotic zone, was more important in the angptl4.sup.LacZ/LacZ group (19±1%) compared to the angptl4.sup.LacZ/+ group (11±2%) (p<0.05).

[0126] We next hypothesized that attenuation of vascular alterations by recombinant ANGPTL4 may lead to the enhancement of endothelial barrier function, which ultimately could protect mice from consequences ischemia-reperfusion in this model. Nevertheless, myocardial ischemia-reperfusion does not induce massive no-reflow in mice in these conditions, which renders it a difficult model to assess any potential therapeutic effect. As a proof of concept, we therefore sought to analyze the therapeutic potential of ANGPTL4 in a second non-rodent specie, i.e., in an open-chest rabbit model of myocardial ischemia reperfusion in which the no-reflow phenomenon has been well established, as previously described (Hale, S. L., Mehra, A., Leeka, J. & Kloner, R. A. Postconditioning fails to improve no reflow or alter infarct size in an open-chest rabbit model of myocardial ischemia-reperfusion. Am J Physiol Heart Circ Physiol 294, H421-5 (2008)). Whereas the levels of circulating ANGPTL4 in ischemic pathologies has never been studied neither in man nor in any other species, a recent publication identify physiological determinants of plasma ANGPTL4 levels in humans, such as energy restriction and plasma FFAs and circulating levels of 20 to 100 ng/ml (Kersten, S. et al. Caloric restriction and exercise increase plasma ANGPTL4 levels in humans via elevated free fatty acids. Arterioscler Thromb Vase Biol 29, 969-74 (2009)). We therefore performed I.V. injection of 100 ng/ml human recombinant ANGPTL4 in rabbits, 5 min. prior to ischemia-reperfusion. Infarct size expressed as a percentage of the risk zone (IS, % area at risk) was 57±5% in the control group and 34±7% in the ANGPTL4-treated group (FIG. 1A); thus ANGPTL4 conferred cardioprotection. Then, the zone of no-reflow was studied. When expressed as a percentage of the area at risk, it was 41±2% in the control group and 19±6% in the ANGPTL4-treated group (FIG. 1B). More importantly, when expressed as a percentage of the infarct size, it was 73±4% in the control group and 55±7% in the ANGPTL4-treated group (FIG. 1C). Histological analyzes show that myocardial infarction consisted in a core of necrosis and huge hemorrhage within interstitial spaces in control group. In the ANGPTL4-treated group, the extent of hemorrhages, expressed as a percentage of total heart section area was decreased as compared to control group (5.7±2% vs. 21.9±6.4%, p<0.05). Thus, these results therefore show that ANGPTL4 induces preservation of vascular network integrity as compared to controls and reduced the extent of no-reflow.

[0127] Early Post-Ischemic Control of Vascular Permeability is Responsible for Cardiac Tissue Protection:

[0128] As vasculoprotection was observed in ANGPTL4-treated rabbits, we tested the hypothesis that early post-ischemic vascular hyperpermeability might constitute a mechanism for increased vascular alterations, inflammation and myocardial damage observed in angptl4.sup.LacZ/LacZ mice. We therefore analyzed vascular permeability after 45 min ischemia and only 4 h of reperfusion in the heart of both angptl4.sup.LacZ/LacZ and angptl4.sup.LacZ/+ mice. Vascular density was similar in both genotypes in basal conditions, and histological studies of HE slides and immunolabellings of CD31 and Mac3 after 45 min ischemia and 4 h reperfusion showed that neither tissues necrosis nor vascular density or inflammation were different in both groups. Extravasation of Evans blue dye from the vascular cardiac beds and accumulating in the interstitium was then measured in angptl4.sup.LacZ/LacZ and in angptl4.sup.LacZ/+ mice. Angptl4.sup.LacZ/LacZ mice displayed an increased cardiac vascular leakage compared to angptl4.sup.LacZ/+ mice 4 h after ischemia (117.5±15.2 versus 84.8±2.7, p=0.032), whereas no significant difference was observed in basal conditions.

[0129] Then, in order to better characterize events responsible for increased permeability after 4 h of reperfusion, we then performed injection of fluorescent microspheres that allowed localization of zones of vascular leakage thereby allowing us to conduct a detailed analysis of these areas by confocal microscopy. Microvessels and pericytes densities were similar between both groups as determined by CD31 and NG2 immunoreactivities. Since stability of VE-cadherin at adherens junctions is critical for maintenance of endothelial permeability and integrity, we therefore sought to investigate endothelial adherens junctions. An heterogenous pattern of VE-cadherin staining was observed in angptl4.sup.LacZ/+ mice; both intense and linear signals were adjacent to thinner signals. In contrast, in angptl4.sup.LacZ/LacZ mice, ischemia-reperfusion injury induced more severe damages in endothelial junctions which were mainly disrupted as shown by a more discontinuous VE-Cadherin staining. In addition, these disrupted junctions massively allowed extravasation of FITC-beads. Comparatively, absence of FITC-beads in angptl4.sup.LacZ/+ mice indicated that the remaining stabilized junctions were sufficient to maintain barrier function in these mice.

[0130] In parallel, similar analyses were performed in basal conditions. No difference was observed in term of vascular density and pericytes coverage between both genotypes. We showed a strong VE-cadherin linear signal that labels a dense vascular network in angptl4.sup.LacZ/+ mice whereas this signal appeared also continuous but thinner in angptl4.sup.LacZ/LacZ mice. In accordance with data obtained in the modified Miles assay which did not reveal any difference of vascular leakage in basal conditions, FITC-labelled beads extravasation was absent in both groups. These observations therefore suggest that whereas VE-cadherin distribution appears less continuous and thinner in angptl4.sup.LacZ/LacZ mice compared to control mice in basal conditions, this was not sufficient to increase basal vascular permeability in the heart. In contrast, ischemic conditions further induced junctional disassembly that eventually led to a higher vascular permeability in angptl4.sup.LacZ/LacZ mice compared to control mice.

[0131] These results suggest that cardioprotection might be achieved through vasculoprotection and preservation of vascular endothelial cell barrier integrity. Altogether, we here describe ANGPTL4 as a new relevant physiological target during acute myocardial ischemia.

[0132] Discussion:

[0133] Upon myocardial ischemia, hypoxia-inducible factor (HIF) proteins, the principal transcription factors involved in the regulation of transcriptional responses to hypoxia are rapidly activated, and induce VEGF-A expression that participates in regulating the angiogenic response but also causes vascular permeability and edema resulting in extensive injury to ischemic tissues. The first wave of VEGF-A release induces increase in endothelial cell permeability and tissue edema, contributing to ischemic-reperfusion injury, whereas the second peak might relate to the angiogenic reparative response. HIF also plays an essential role in triggering cellular protection and metabolic alterations from the consequences of oxygen deprivation. Here, we show that angptl4 mRNA which has previously been shown to be induced by hypoxia in endothelial cells and in cardiomyocytes in vitro as well as in critical hind limb ischemia and stroke is also expressed in cardiac tissue from patients who died from an AMI. We further show that angptl4 mRNA is also expressed in mouse in a model of AMI and we provide evidence that ANGPTL4 mediates post-ischemic tissue damage protection through preservation of vascular endothelial cell barrier integrity that limits no-reflow and the extent of AMI.

[0134] In pathological ischemic conditions, increased permeability which is predominantly controlled by endothelial junction stability is responsible for altered vascular integrity. Yang et al. showed that myocardial tissue levels of VE-cadherin were significantly decreased in the reflow and no-reflow myocardium compared to those in the non-ischemic myocardium, suggesting that microvascular structural integrity was damaged by ischemia/reperfusion (Yang, Y. J. et al. Post-infarction treatment with simvastatin reduces myocardial no-reflow by opening of the KATP channel. Eur J Heart Fail 9, 30-6 (2007)). In the present in vivo study, we show for the first time that angptl4 ko mice display reduced adherens junctions stability and increased vascular permeability thereby leading to increased myocardial infarct size. VE-cadherin which constitutes the major component of the adherens junctions in endothelial cells, is required in vivo in the post-natal vasculature in order to maintain endothelial junction integrity and barrier function. VE-cadherin associates with VEGFR2 and regulates permeability. Indeed, VEGF-A stimulation induces dissociation of the VEGFR2/VE-cadherin complex, that can be prevented by blockade of Src, an essential molecule required for promoting the disruption of endothelial cell-cell contacts and paracellular permeability. We here showed that VE-cadherin distribution is disorganized in angptl4 knockout mice, leading to destabilized adherens junctions, decreased vascular integrity and endothelial cell barrier function following cardiac ischemia-reperfusion. Other members of the angiopoietins family also display a role in the regulation of vascular permeability. In the mature vasculature, Angiopoietin-1 inhibits permeability induced by VEGF-A or mustard oil whereas Angiopoietin-2 causes inflammation in vivo by promoting vascular leakage. Interestingly, Angiopoietin-1 has been shown to phosphorylate Tie-2 and phosphatidylinositol 3-kinase, inducing activation of the GTPase Rac1, necessary to maintain cell-cell adhesion and to prevent hyperpermeability. Angiopoietin1 also promotes the activation of mDia through RhoA, resulting in the association of mDia with Src, thereby interfering with the ability of VEGF-A to initiate the activation of a Src-dependent intracellular signaling. Whether regulation of Src and/or Rac1 signaling pathways by ANGPTL4 may affect intracellular VE-cadherin distribution by stabilizing it at cell junctions through post-translational modification is beyond the scope of the present study but would deserve further investigation.

[0135] In addition, reperfused myocardial infarction is associated with cellular infiltration and acute inflammatory response. Cardiac repair after myocardial infarction is a dynamic and complex biological process that can be divided into the inflammatory phase, the proliferative phase, and the maturation phase. Whereas the inflammatory cascade is a prerequisite for healing of the infarcted myocardium, effective cardiac repair depends on mechanisms that also suppress the inflammatory response during vascular remodelling and that limit expansion of fibrosis to the noninfarcted myocardium. A critical point in the therapy against post-ischemia injury remains to contain deleterious persistent and expanding inflammatory response. Although numerous studies have focused on the expression and role of inflammatory mediators in the infracted myocardium, the cellular and molecular events responsible for downregulation and containment of the inflammatory cascade remain unknown. We here show that modulation of vascular permeability by ANGPTL4 might constitute a point of control that participates in decreasing the post-infarction inflammatory response and thus limit expansion of infarcted area.

[0136] We further show that early protection of vascular integrity by ANGPTL4 induces preservation of the microcirculatory network and lesser extent of hemorrhages that both participate in limiting the extent of no-reflow. This phenomenom is the result of yet uncompletely characterized anatomical changes of coronary microcirculation in which ANGPTL4 might play a crucial role through its vasculoprotective effect.

[0137] ANGPTL4 is also an inhibitor of LPL which is highly expressed in the heart and participates in lipoproteins hydrolysis as a source of fatty acids for uptake. Yu et al. have shown that mice that overexpress ANGPTL4 in the heart exhibit fasting hypertriglyridemia and develop left-ventricular dysfunction. This could mean that mice in which angptl4 is deleted should have higher LPL activity which could participate in preventing cardiomyocyte death during infarction. Interestingly, if this event occurs in vivo in the ischemic heart, this effect is overcome by deleterious vascular effects observed in angptl4 knockout mice that we describe here, rendering those mice more sensitive to myocardial ischemia/reperfusion injury. In addition, products generated from lipoprotein lipase-mediated hydrolysis of triglyceride-rich lipoproteins are reported to increase endothelial layer permeability through zonula occludens-1 (ZO-1) radial rearrangement and concurrent redistribution of F-actin from the cell body to the cell margins in human aortic endothelial cells. In addition, recent findings place VE-cadherin upstream of claudin-5, a key component of tight endothelial junction, in the maintenance of endothelial cell-cell junctions. Therefore, our results on the disorganization of endothelial adherens junctions in angptl4 knockout mice suggest that ANGPTL4 could promote endothelial barrier function at multiple levels.

[0138] Finally, whereas cardiomyocytes were primarily recognized as a clinically therapeutic target of myocardial ischemia, few studies have focused on the importance of the cardiac vessels. More recently, clinical efforts are underway to block VEGF-A-mediated leak in patients after acute myocardial infarction or stroke. Such strategy may have a significant impact on reducing tissue injury and thereby minimizing the long-term consequences. Our findings suggest that ANGPTL4, by counteracting ischemia-induced disruption of endothelial cell junctions and subsequent increase in permeability observed in reperfused AMI might be promising for improved myocardial infarction therapy.

EXAMPLE 2: PROTECTION AGAINST MYOCARDIAL INFARCTION AND NO-REFLOW THROUGH PRESERVATION OF VASCULAR INTEGRITY BY ANGIOPOIETIN-LIKE 4 MATERIAL AND METHODS

[0139] The experiments were performed in accordance with the official regulations edicted by the French Ministry of Agriculture. This study conforms to the standards of INSERM (the French National Institute of Health) in accordance with European Union Council Directives (86/609/EEC).

[0140] Myocardial Ischemia-Reperfusion Experiments:

[0141] Ischemia-reperfusion protocol was performed on angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ mice or rabbits using a standard technique described in Supplemental Materials. Rabbits randomly received either vehicle or human recombinant 55 kDa full-length ANGPTL4 (rhANGPTL4 10 μg/kg i.v.).

[0142] Modified Miles Assay:

[0143] Male angptl4.sup.LacZ/LacZ and angptl4.sup.LacZ/+ mice were anesthetized using pentobarbital. For basal conditions, mice were injected into the tail vein with 1% Evans blue (200 μl) and sacrificed 4 h later. For the ischemia-reperfusion conditions, mice were subjected to coronary occlusion for 45 min and intravenously injected with 1% Evans blue (200 μl) prior the 4 h of reperfusion. At sacrifice, mice were perfused through the aorta with citrate buffer pH4. Blood, dye and buffer exited through an opening in the right atrium. Evans Blue was eluted for 18 h at 70° C. in 1 ml formamide. After centrifugation, absorbance at 620 nm was measured using a spectrophotometer. Extravasated Evans blue (ng) was determined from a standard curve and normalized to tissue weight (g).

[0144] Immunofluorescence study and confocal analyzis on cryosections: Immunofluorescence staining was performed as previously described (Brechot N, Gomez E, Bignon M, Khallou-Laschet J, Dussiot M, Cazes A, Alanio-Brechot C, Durand M, Philippe J, Silvestre J S, Van Rooijen N, Corvol P, Nicoletti A, Chazaud B, Germain S. Modulation of macrophage activation state protects tissue from necrosis during critical limb ischemia in thrombospondin-1-deficient mice. PLoS ONE. 2008; 3:e3950.) and confocal analysis on cryosections is detailed in Supplemental Material.

[0145] Immunoprecipitation and Immunoblotting Analyses:

[0146] For in vivo samples, mice subjected to 45 min ischemia-4 h reperfusion were anesthetized, injected into tail vein with 1 mM Na.sub.3VO.sub.4 and 2 mM H.sub.2O.sub.2 and dissected to remove left ventricle. For HUAECs experiments, 40000 cells/cm.sup.2 were seeded in complete culture medium (Promocell) for 72 h. Cells were starved overnight and treated for 5 min with 100 ng/ml human recombinant VEGF.sub.165 (Sigma) or with a mix containing 100 ng/ml (10 nM) VEGF and 5 μg/ml (360 nM) human recombinant ANGPTL4 (Chomel C, Cazes A, Faye C, Bignon M, Gomez E, Ardidie-Robouant C, Barret A, Ricard-Blum S, Muller L, Germain S, Monnot C. Interaction of the coiled-coil domain with glycosaminoglycans protects angiopoietin-like 4 from proteolysis and regulates its antiangiogenic activity. Faseb J. 2009; 23:940-949.) before to be washed twice with Ca/MgPBS. Proteins were extracted, immunoprecipitated for VEGFR2 and analyzed by Western blotting as described in Supplementary Material.

[0147] Isolation of Cardiomyocytes and Viability Assay:

[0148] Cardiac myocytes were isolated as described in Supplemental Material and incubated in an anaerobic chamber containing a humidified atmosphere of 5% CO.sub.2 and 95% N.sub.2 for 3 h. Experimental medium was changed to serum-free, glucose-free. Cardiac myocyte survival was measured by staining cells with trypan blue (Sigma).

[0149] Statistical Analyses:

[0150] Mann-Whitney or Student's tests were used to assess the statistical differences between groups or conditions (GraphPad Prism 4, GraphPad Software). Error bars indicate SEM and *, P<0.05; **, P<0.001; ***, P<0.0001.

[0151] Supplemental Materials

[0152] Expanded Methods and Results

[0153] Animals and Genotyping:

[0154] Genotype was determined by PCR of tail genomic DNA as previously described (Gomez, 2010, submitted). Eight to 12 weeks of age angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ a knock-out male mice, intercrossed in C57/Bl6 mice for more than 8 generations, were subjected to myocardial infarction protocols or used as control in basal conditions.

[0155] Mice Myocardial Ischemia-Reperfusion Experiments:

[0156] Mice were anesthetized by an intraperitoneal injection of sodium pentobarbital. Myocardial infarction with occlusion of the left coronary artery was performed for 45 mn and tissues were reperfused for 1 h to 3 weeks. For angptl4 expression study WT mice underwent 1 h, 3 h, 24 h, 48 h, 72 h, 1 week, 2 weeks, 3 weeks of reperfusion. To assess infarct size and for immunohistochemistry (IHC) or ultrastructural studies, male angptl4.sup.LacZ/+ and angptl4.sup.LacZ/LacZ mice were reperfused during either 4 h or 48 h after ischemia. The area at risk was identified by Evans blue staining at 48 h after ischemia, and the infarct area was identified by 2,3,5-triphenyltetrazolium chloride (TTC) staining. The area at risk was identified as the non blue region and expressed as a percentage of the left ventricle weight. The infarcted area was identified as the TTC-negative zone and expressed as a percentage of the area at risk. To measure no-reflow, the chest was reopened and thioflavine S (4%; 1.5 ml/kg) was infused through the left atrium four hours after the onset of reperfusion. The hearts were then perfused retrogradely with Alcian blue (0.5%) and cut into slices. Slices were photographed using UV light to identify the region of no-reflow. The areas of infarct and risk zone were determined as defined above. Ultrastructural analyses were performed on a Hitachi H-9500 electron microscope.

[0157] Immunofluorescence study and confocal analyzis on cryosections: Mice subjected to ischemia and 4 h of reperfusion were anesthetized with ketamine and xylazine injected intraperitonally. FITC-beads (20 μl) were injected into the femoral vein as previously described.sup.44. The chest was opened rapidly, and the vasculature was perfused for 2 min at a pressure of 120 mmHg with 1% paraformaldehyde. The heart was then placed into 1% paraformaldehyde for 1 h at room temperature, rinsed with PBS and frozen for cryostat sectioning. Endothelial cells, pericytes and adherens junctions were identified with rat anti-CD31 (BD Pharmingen), rabbit anti-NG2 (Chemicon) and rat anti-VE-Cadhcrin (personal gift from E. Dejana, IFOM) antibodies, respectively. Confocal sections were imaged on a Leica SP5 microscope (Leica Microsystems GmbH) using a 63× (NA=1.4) oil objective. An increment of 0.117 μm between each section was used. 3D reconstruction of the different structures was obtained using the LABELVOXEL and the SURFACEGEN modules in Amira 5.2.1 software (Visage Imaging GmbH).

[0158] Immunoprecipitation and Immunoblotting Analyses:

[0159] Proteins were extracted on ice in 20 mM Tris-HCl, pH 7.6, 150 mM NaCl, 0.1% DOC, 0.5% NP-40, 10% glycerol, 1 mM β-glycerophosphate, 1 mM NaF, 2.5 mM Na pyrophosphate, 1 mM Na.sub.3VO.sub.4 and a cocktail of protease inhibitors (Calbiochem). Lysates were split for immunoprecipitation and for total extracts immunoblottings. For immunoprecipitation, extracts were precleared for 60 min with protein A-agarose beads, incubated overnight with anti-VEGFR-2 (Cell signaling), and the immunocomplexes were collected on protein A-agarose beads for 3 h. Proteins were eluted by boiling for 10 min in reducing laemmli sample buffer. Samples were analyzed by SDS-PAGE followed by Western blotting on nitrocellulose membrane. Anti-VEGFR2 (Cell signaling), anti-VE-Cadherin (Santa Cruz), anti-Src kinase family (Cell signaling), anti-phospho Src family Tyr-416 (Cell signaling) antibodies were used. Signal was revealed by Attophos chemiluminescence (Promega) and band intensity was quantified by Quantity One 1-D Analysis Software (Biorad).

[0160] Isolation of Cardiomyocytes and Viability Assay:

[0161] Under anesthesia, the heart was removed from the chest and was cannulated. The heart was perfused for 4 min with tyrode buffer ([mM] NaCl 113; KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; with pH adjusted to 7.46 with NaOH 1N) at constant pressure and 37° C. Perfusion was switched to an enzyme solution ([mM] NaCl 113; KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; CaCl.sub.2 0.0125; with pH adjusted to 7.46 with NaOH 1N) containing 0.1 mg/ml liberase blendzyme IV, (Roche diagnostics) and 0.14 mg/ml trypsin (Sigma). When hearts became swollen and turn slightly pale, the atria and aorta were removed; the left ventricle were cut into small pieces and gently triturated. Cell suspension was transferred into a stopping buffer ([mM] NaCl 113; KCl 4.7; KH.sub.2PO.sub.4 0.6; Na.sub.2HPO.sub.4 0.6; HEPES 10; MgSO.sub.4 1.2; NaHCO.sub.3 12; KHCO.sub.3 10; taurine 30; phenol red 0.032; glucose 5.5; CaCl.sub.2 0.0125; calf serum 5%; with pH adjusted to 7.46 with NaOH 1N). Extracellular calcium was added incrementally up to 1.0 mM. All cells studied were rod-shaped, had clear cross-striations and lacked any visible vesicles on their surfaces.

[0162] Rabbit Experiments:

[0163] New Zealand rabbits (2.5-3.0 kg) were anesthetized using zolazepam, tiletamine and pentobarbital (all 20-30 mg/kg i.v.). The animals were intubated, mechanically ventilated and a left thoracotomy was performed. A suture was passed beneath a major branch of the left coronary artery through a short propylene tubing to form a snare. Rabbits then randomly received either vehicle or human recombinant 55 kDa full-length ANGPTL4.sup.12 (10 μg/kg i.v.). Five minutes after, coronary artery occlusion (CAO) was induced during 30-min by pulling the snare through the tubing. Reperfusion was subsequently induced by releasing the snare. The chest was then closed in layers. Four hours after the onset of reperfusion, the chest was reopened and thioflavine S (4%; 1.5 ml/kg) was infused through the left atrium. Rabbits were then sacrificed using pentobarbital followed by potassium chloride. After excision, the hearts were perfused retrogradely with Alcian blue (0.5%) and cut into slices. Slices were photographed using UV light to identify the region of no-reflow. The areas of infarct and risk zone were determined as in mice.

[0164] Ultrasound Analysis of Cardiac Parameters:

[0165] Mice were subjected to ultrasound measurements using an echocardiograph (Vivid 7, GE Medical Systems Ultrasound) equipped with a 12-MHz linear transducer.

[0166] Real-Time Quantitative PCR Analysis (RT-qPCR):

[0167] Mice subjected to ischemia and 4 h or 18 h reperfusion were anesthetized, injected into tail vein and intracardiacly with 1 mM Na.sub.3VO.sub.4 and 2 mM H2O2 and dissected to remove left ventricle. Total RNA was isolated by extraction with TRIzol (Invitrogen). Reverse transcription, quantitative PCR (in triplicate) and analysis were performed as previously described (Xu Y, Yuan L, Mak J, Pardanaud L, Caunt M, Kasman I, Larrivee B, Dcl Toro R, Suchting S, Medvinsky A, Silva J, Yang J, Thomas J L, Koch A W, Alitalo K, Eichmann A, Bagri A. Neuropilin-2 mediates VEGF-C-induced lymphatic sprouting together with VEGFR3. J Cell Biol. 2010; 188:115-130.). mRNA expression level was normalized to the housekeeping gene encoding glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Fold changes were calculated using the comparative Ct method.

[0168] In Situ Hybridization (ISH) and Immunohistochemistry (IHC) Analyses:

[0169] Paraffin blocks of human myocardial infarcts were obtained from the Pathology Department of Georges Pompidou European Hospital, Paris, France. The presence of infarcted areas was assessed on standard HE staining and adjacent slides were used for ISH and IHC analyses. ISH using human or mouse angptl4 probes and IHC immunolabellings anti-CD45, -Mac3, and -CD31 were performed as previously described (Brechot N, Gomez E, Bignon M, Khallou-Laschet J, Dussiot M, Cazes A, Alanio-Brechot C, Durand M, Philippe J, Silvestre J S, Van Rooijen N, Corvol P, Nicoletti A, Chazaud B, Germain S. Modulation of macrophage activation state protects tissue from necrosis during critical limb ischemia in thrombospondin-1-deficient mice. PLoS ONE. 2008; 3:e3950.).

[0170] Results

[0171] Early Post-Ischemic Vascular Integrity is Altered in Angptl4.sup.LacZ/LacZ Mice

[0172] angptl4.sup.LacZ/LacZ mice in which the angptl4 locus was replaced by a lacZ reporter gene were generated. We first analyzed vascular permeability after 45 min ischemia and 4 h reperfusion in the heart of both angptl4.sup.LacZ/LacZ and angptl4.sup.LacZ/+ mice. Histological analyzes showed that tissue damage was equivalent in both groups at this early time point. Angptl4.sup.LacZ/LacZ mice displayed an increased vascular leakage compared to angptl4.sup.LacZ/+ mice 4 h after ischemia (117.5±15.2 versus 84.8±2.7 μg/ml, p<0.05), whereas no significant difference was observed in basal conditions. Importantly, angptl4.sup.LacZ/LacZ mice did not exhibit any functional cardiac defects or abnormal vascular morphology both in basal or in ischemic conditions.

[0173] Then, fluorescent microspheres were injected thus allowing the localization of areas of vascular leakage. In basal conditions, FITC-beads did not extravasate in both groups. After 45 min ischemia-4 h reperfusion, no FITC-bead was observed in angptl4.sup.LacZ/+ mice whereas extravasation of fluoresecent microspheres was detected in angptl4.sup.LacZ/LacZ mice indicating that endothelial integrity was altered. Since stability of adherens junctions is critical for maintenance of endothelial permeability and integrity, we sought to investigate VE-Cadherin distribution at endothelial adherens junctions in basal conditions and in ischemia-reperfusion model. In non-ischemic myocardium from both genotypes, a VE-Cadherin linear signal labelling a dense vascular network was observed. In contrast, after ischemia-reperfusion, a heterogeneous pattern of VE-Cadherin staining was observed in angptl4.sup.LacZ/+ mice; both intense and linear signals were adjacent to thinner signals. In angptl4.sup.LacZ/LacZ mice, ischemia-reperfusion injury induced more severe damages in endothelial junctions, which were mainly disrupted as shown by a more systematic discontinuous VE-Cadherin staining. 3D-images rebuilt from confocal pictures of angptl4.sup.LacZ/LacZ heart sections stained with anti-CD31 and anti-NG2 antibodies further confirmed extravasated FITC-beads from blood vessels.

[0174] These observations suggest that coronary vascular integrity is fragile and junction disassembly is more frequent in angptl4.sup.LacZ/LacZ mice during ischemic conditions, thereby leading to increased vascular permeability.

[0175] Post-Ischemic Decrease in VEGFR2 and VE-Cadherin Expression Combined to Increase Src Kinase Phosphorylation Downstream the VEGFR2 in Angpt4.sup.LacZ/LacZ Mice

[0176] In the vasculature, VEGFR2 and VE-Cadherin form complexes that are transiently dissociated upon VEGF binding to VEGF-R2. During myocardial infarction, ischemia promotes VEGF expression that leads to vascular permeability and edema. We therefore investigated whether enhanced VEGFR2/VE-Cadherin complex disassembly might constitute the mechanism responsible for an increased junctional disruption in angptl4.sup.LacZ/LacZ mice after ischemia-reperfusion injury.

[0177] Using RT-qPCR, vegfr2 and ve-cadherin mRNA expression were quantified in the left ventricle in angptl4.sup.LacZ/LacZ and angptl4.sup.LacZ/+ mice, in control conditions or after 4 h or 18 h reperfusion. ve-cadherin and vegfr2 mRNA expression was similar in both groups in basal conditions. After 4 h reperfusion, massive decrease in vegfr2 and ve-cadherin mRNA expression was observed in angptl4.sup.LacZ/LacZ mice compared to control mice (58±3 versus 34±3% for ve-cadherin and 70±2 versus 7±5% for vegfr2, p<0.001). This down-regulation was maintained after 18 h reperfusion for the vegfr2 mRNA (decrease of 66±2% for angptl4.sup.LacZ/LacZ mice, p<0.001). Protein levels were also affected as shown by western blot analyzes performed using total extracts from left ventricles in control and ischemic conditions.

[0178] Src kinase is required in VEGF-mediated permeability through its role in dissociating the VEGFR2/VE-Cadherin complex. To further determine the mechanism leading to early post-ischemic junctions alteration in angptl4.sup.LacZ/LacZ mice, Src kinase signaling downstream of VEGFR2 was analyzed in control conditions and after ischemia-reperfusion. Left ventricles lysates were immunoprecipitated for VEGFR2 followed by immunoblotting for VEGFR2, VE-Cadherin, Src and phospho-Src. In basal conditions, VEGFR2/VE-Cadherin formed complexes in both genotypes. A transient destabilization of VEGFR2/VE-Cadherin complexes was observed at 4 h reperfusion and was restored after 18 h reperfusion in angptl4.sup.LacZ/+ mice whereas VE-Cadherin remains dissociated from VEGFR2 in angptl4.sup.LacZ/LacZ mice. In addition, immunoblottings showed an increased Src kinase recruitment and phosphorylation after 4 h and 18 h of reperfusion in angptl4.sup.LacZ/LacZ compared to angptl4.sup.LacZ/+ mice.

[0179] These results show that a decrease in vegfr2 and ve-cadherin expression combined to an increase in Src kinase phosphorylation downstream the VEGFR2 lead to VEGFR2/VE-Cadherin complex dissociation responsible for massive disorganisation of VE-Cadherin in endothelial adherens junctions in angptl4.sup.LacZ/LacZ mice following ischemia-reperfusion.

[0180] Infarct Size, No-Reflow and Post-Ischemic Inflammation are Increased in Angptl4.sup.LacZ/LacZ Mice

[0181] We next hypothesized that alteration of vascular integrity in angptl4.sup.LacZ/LacZ mice might translate to abnormal myocardial reperfusion and major heart tissue damage at 48 h reperfusion. Indeed, infarct size was increased in angptl4.sup.LacZ/LacZ mice compared to angptl4.sup.LacZ/+ mice (47±3 versus 36±3%, p<0.01). In addition, the no-reflow was more important in the angpt4.sup.LacZ/LacZ group compared to the angptl4.sup.LacZ/+ mice, when expressed as a percentage of the necrotic zone (19±1 versus 11±2%, p<0.05).

[0182] Necrosis, hemorrhages and edema were also quantified (score 1 to 3) on HE-stained sections from infarcted hearts of both genotypes. In accordance with increased infarct size, tissue necrosis was largely increased in angptl4.sup.LacZ/LacZ compared to control mice (2.5±0.6 versus 1.2±0.2). Assessment of hemorrhages and edema revealed a more severe tissue injury in angptl4.sup.LacZ/LacZ (2.3±0.6 versus 1.1±0.2 and 2.3±0.2 versus 0.8±0.2, respectively). The post-ischemic inflammatory response was also analyzed in both genotypes. Macrophage density was significantly higher in infarcted areas in angptl4.sup.LacZ/LacZ a versus angptl4.sup.LacZ/+ mice, whereas no statistical difference was observed between both groups in control non-infarcted areas.

[0183] We then analyzed vascular density in the core infarct area and in the periphery using CD31 staining. A similar diminished microcapillary density was quantified in both genotypes in the central infarcted areas as compared to the periphery. No difference was quantified between both genotypes in both areas.

[0184] Transmission electron microscopy study was further performed in order to assess tissue injury at the ultrastructural level. Analysis of reperfused infarcted areas did not show cardiomyocyte structural differences between both groups (see <<C.sub.1 to 4>> in FIGURE IVA and IVB) but large edematous areas, poor in inflammatory cells were observed in angptl4.sup.LacZ/+ mice, whereas inflammatory cells had already invaded the edematous region in angptl4.sup.LacZ/LacZ mice. Polynuclear neutrophils, macrophages, lymphocytes and fibrinogen deposits were only observed in angptl4.sup.LacZ/LacZ mice. Altogether, these data indicate increased vascular alterations that correlate with increased inflammatory infiltrate in angpt4.sup.LacZ/LacZ mice.

[0185] As hypoxic activation of angptl4 mRNA has been reported in cardiomyocytes, likely mediated by hypoxia-inducible factor 1, we also hypothesized ANGPTL4 might also affect cardiomyocytes survival. Indeed, LacZ staining performed on whole-mount, HE-stained and CD31-immunostained sections from angptl4.sup.LacZ/LacZ mice also revealed that both cardiomyocytes and ECs express angptl4 after ischemia-reperfusion injury. In situ hybridization (ISH) further showed angptl4 mRNA expression was induced as soon as 3 h following ischemia and up to 2 weeks after reperfusion, whereas angptl4 mRNA was not expressed in the non-ischemic area. ISH in cardiac samples from patients who died from AMI also revealed angptl4 mRNA expression in cardiomyocytes and in ECs. Cardiomyocytes from both genotypes were therefore isolated and subjected to an in vitro survival assay. No difference in cardiomyocytes survival was observed in vitro between both groups either in normoxia or in hypoxia suggesting that ANGPTL4 does not have a direct effect on cardiomyocytes.

[0186] Recombinant ANGPTL4 Stabilizes VEGFR2/VE-Cadherin Complex in Response to VEGF

[0187] We next investigated whether rhANGPTL4 could confer protection of VEGFR2/VE-Cadherin complexes disassembly in ECs. Confluent HUAECs were stimulated for 5 min with VEGF alone or with rhANGPTL4. Cell lysates were immunoprecipitated for VEGFR2 followed by immunoblotting for VEGFR2, VE-Cadherin, Src and phospho-Src. The pre-existing VEGFR2/VE-Cadherin observed in control conditions is rapidly disrupted within 5 mn VEGF stimulation. This complex was protected from dissociation in cells treated with both VEGF and rhANGPTL4 (. Src kinase and phospho-Src immunoblottings revealed that VEGF-mediated VEGFR2/VE-Cadhcrin destabilization was correlated with an increased Src phosphorylation downstream VEGFR2 which was partially blocked in cells treated with both VEGF and rhANGPTL4.

[0188] These in vitro experiments provide evidence that rhANGPTL4 protects from VEGF-induced VEGFR2/VE-Cadherin complex destabilisation through inhibition of Src kinase signaling.

[0189] Assessment of Therapeutic Cardioprotective Effect of rhANGPTL4

[0190] We next hypothesized that attenuation of vascular alterations by rhANGPTL4 may lead to the enhancement of endothelial barrier function, which ultimately could protect from ischemia-reperfusion. As myocardial ischemia-reperfusion does not induce massive no-reflow in mice in these conditions, we therefore sought to analyze the therapeutic potential of ANGPTL4 in a non-rodent specie, i.e., in an open-chest rabbit model of myocardial ischemia-reperfusion in which the no-reflow phenomenon has been well established.

[0191] We therefore performed intravenous injection of 10 mg/kg rhANGPTL4, 5 min. prior to ischemia-reperfusion. Infarct size expressed as a percentage of the risk zone (IS, % area at risk) was 57±5% in the control group and 34±7% in the rhANGPTL4-treated group (p<0.01). Then, the zone of no-reflow was studied. When expressed as a percentage of the area at risk, it was 41±2% in the control group and 19±6% in the rhANGPTL4-treated group (p<0.05). More importantly, when expressed as a percentage of the infarct size, it was 73±4% in the control group and 55±7% in the rhANGPTL4-treated group (p<0.05). Histological analysis showed that myocardial infarction consisted in a core of necrosis and huge hemorrhage within interstitial spaces in control group. In the rhANGPTL4-treated group, the extent of hemorrhages was decreased (5.7±2% versus 21.9±6.4%, expressed as a percentage of total heart section area, p<0.05).

[0192] Thus, these results show that rhANGPTL4 induces preservation of vascular integrity that reduces infarct size, hemorrhage and no-reflow and thus confers cardioprotection.

Discussion

[0193] Upon AMI, hypoxia-inducible factor (HIF) proteins, the major transcription factors involved in the regulation of responses to hypoxia are rapidly activated, and induce VEGF-A expression that participates in regulating the angiogenic response but also causes vascular permeability and edema resulting in extensive injury. We showed here that angptl4 mRNA which has previously been shown to be induced by hypoxia in ECs and in cardiomyocytes in vitro as well as in critical hind limb ischemia and stroke is also expressed in cardiac tissue from patients who died from AMI. We further provided evidence that ANGPTL4 mediates post-ischemic damage protection through preservation of vascular endothelial cell barrier integrity that limits no-reflow and the extent of AMI.

[0194] In pathological ischemic conditions, increased permeability, which is controlled by endothelial junction stability, is responsible for altered vascular integrity. VE-cadherin which constitutes the major component of the adherens junctions between ECs, is required in vivo in the post-natal vasculature in order to maintain endothelium integrity and barrier function. It was showed that myocardial VE-Cadhcrin is significantly decreased in the ischemic myocardium, suggesting that microvascular integrity is damaged by ischemia/reperfusion. VE-cadherin associates with VEGFR2 and regulates permeability. Indeed, systemic VEGF-A injection, thereby activating VEGFR2, induces dissociation of the VEGFR2/VE-Cadherin complex. Here we showed durable dissociation of VEGFR2/VE-Cadherin complexes and altered VE-cadherin distribution in angptl4 knockout mice, causing disrupted adherens junctions and decreased ECs barrier function following AMI. Gene expression analysis revealed i) a more prominent diminished ve-cadherin mRNA levels, ii) a prolonged decrease of vegfr2 mRNA levels in angptl4 knockout mice subjected to AMI. Decreased levels of ve-cadherin and vegfr2 gene expression, and therefore of VEGFR2/VE-Cadherin complexes in response to ischemia might participate to junctions disruptions and altered endothelial integrity following AMI in angptl4.sup.LacZ/LacZ mice.

[0195] Src is an essential molecule required for promoting the disruption of ECs contacts and paracellular permeability. We here provide evidence for an enhanced i) Src kinase recruitment at the VEGFR2/VE-Cadherin complex and ii) Src kinase phosphorylation, leading to a more severe destabilization of the VEGFR2/VE-Cadherin complex in angptl4.sup.LacZ/LacZ mice subjected to AMI. Other members of the angiopoietin family also display a role in the regulation of vascular permeability. Angiopoietin-1 phosphorylates Tie-2 and phosphatidylinositol 3-kinase, inducing activation of the GTPase Rac1, necessary to maintain cell-cell adhesion and also activates mDia, resulting in the sequestration Src. Whether regulation of Src/mDia or Rac1 signaling pathways by ANGPTL4 may affect intracellular VE-cadherin distribution by stabilizing it at cell junctions through post-translational modification would deserve further investigation.

[0196] In addition, reperfused myocardial infarction is associated with cellular infiltration and acute inflammatory response. A critical point in post-ischemic therapy remains to contain deleterious persistent and expanding inflammatory response. We here show that altered vascular integrity in angptl4.sup.LacZ/LacZ mice might suppress a point of control that participates to limiting both the post-infarction inflammatory response and the expansion of infarcted area. We also showed that recombinant ANGPTL4 induces both preservation of the microcirculatory network and lesser extent of hemorrhages that both participate in limiting the extent of no-reflow. This phenomenon is the result of yet incompletely characterized anatomical changes of coronary microcirculation (Kloner R A, Ganote C E, Jennings R B. The “no-reflow” phenomenon after temporary coronary occlusion in the dog. J Clin Invest. 1974; 54:1496-1508.) in which ANGPTL4 might play a crucial role through its vasculoprotective effect. In addition, recent findings place VE-cadherin upstream of claudin-5, a key component of tight endothelial junction, in the maintenance of endothelial cell-cell junctions. Therefore, our results on the disorganization of endothelial adherens junctions in angptl4.sup.LacZ/LacZ mice suggest that ANGPTL4 could promote endothelial barrier function at multiple levels.

[0197] Finally, whereas cardiomyocytes were primarily recognized as a clinically therapeutic target of myocardial ischemia, few studies have focused on the importance of the cardiac vessels. Our findings show that ANGPTL4 counteracts increase in permeability observed in reperfused AMI. Clinical efforts are also underway to block VEGF-A-mediated leak in patients after acute myocardial infarction or stroke. The search for combined strategies will certainly have a significant impact on reducing tissue injury, improving coronary microcirculation and thereby improving myocardial infarction therapy.

REFERENCES

[0198] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.