1*1
Transcription
1*1
Universite de Sherbrooke Effets du peroxyde d'hydrogene sur la fonction placentaire : implication dans la physiopathologie de la preeclampsie par Samuel Leblanc Departement d'Obstetrique et Gynecologie Memoire presente a la Faculte de medecine en vue de l'obtention du grade de maitre es sciences (M.Sc.) en physiologie Avril 2009 Evaluateurs : Dr Aziz Aris, departement d'obstetrique gynecologie Dr Jean-Patrice Baillargeon, departement de medecine (endocrinologie) Dr Guilain Boissonneault, departement de biochimie 1*1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street Ottawa ON K1A 0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-65630-3 Our file Notre reference ISBN: 978-0-494-65630-3 NOTICE: AVIS: The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduce, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par ('Internet, prefer, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. 1+1 Canada RESUME Des travaux anterieurs effectues dans notre laboratoire ont demontre une correlation entre les nivaux circulants de peroxyde d'hydrogene (H2O2) et d'hormone gonadotrophine chorionique humaine (hCG) chez des femmes souffrant de preeclampsie (Kharfi et al,. 2005). Le peroxyde d'hydrogene etant surtout connu pour ses effets cytotoxiques, nous avons determine quels etaient ses effets sur des cellules placentaires in vitro, en rapport aux conditions retrouvees in vivo chez des femmes preeclamptiques. Pour ce faire, nous avons developpe un modele d'etude de cellules trophoblastiques in vitro, avec lequel nous avons pu evaluer les effets du peroxyde d'hydrogene sur la fonction placentaire via la production d'hCG, sur la production de TNF-a comme marqueur inflammatoire et sur l'oxyde nitrique (NO) comme facteur vasodilatateur. Nos resultats nous ont permis de constater divers effets du peroxyde d'hydrogene sur les cytotrophoblastes in vitro, effets qui peuvent etre correles avec les niveaux constates chez la femme preeclamptique. Premierement, nous avons pu observer une augmentation de la production d'hCG sous l'effet de faibles concentrations de peroxyde d'hydrogene (<50 uM), alors que des concentrations plus elevees diminuent Pactivite placentaire en diminuant la secretion d'hCG, demontrant son action cytotoxique a plus forte dose. Ceci a egalement ete constate par une augmentation de l'apoptose. Le peroxyde d'hydrogene a aussi demontre ses effets sur la production par les cytotrophoblastes de facteurs pro-inflammatoires, tel le TNF-a, ainsi que de son recepteur, le TNFR2, qui sont tous deux augmentes apres traitement. Le peroxyde d'hydrogene semble aussi avoir une influence sur la production de mediateurs vasodilatateurs tel l'oxyde nitrique, dont la diminution est observee in vitro avec des concentrations croissantes d'FfcC^. Le stress oxydatif dans la preeclampsie, represente dans notre modele in vitro par le peroxyde d'hydrogene, semble done etre central dans la physiopathologie de cette maladie, puisqu'on peut lui attribuer en partie la variation d'importants facteurs. Toutefois, les moyens de la prise en charge de cette condition pathologique doivent etre soigneusement determines, puisque le peroxyde d'hydrogene parait avoir tout de meme un role physiologique important a jouer, d'ou l'interet de parler d'equilibre oxydatif. Mots cles: preeclampsie, placenta, stress oxydatif, peroxyde d'hydrogene et TNF-a. TABLE DES MATIERES Table des Matieres Ill Liste des Illustrations VI Liste des Abbreviations VII Resume VIII 1 Introduction 1 1.1 Grossesse 1 1.1.1 Fecondation 1 1.1.2 Implantation 2 1.2 Placenta 3 1.2.1 Developpement Placentaire 5 1.2.1.1 Trophoblastes Villositaires 5 1.2.1.2 Trophoblastes Extra-Villositaires 6 1.2.2 Fonctions du Placenta 7 1.2.2.1 L'hCG 9 1.2.3 Structures du Placenta a Terme 10 1.2.3.1 Plaque Chorionique 11 1.2.3.2 Espace Intervilleux 13 1.2.3.3 Plaque Basale 13 1.2.3.4 Cordon Ombilical 14 1.3 Hypertension Gestationnelle et Preeclampsie 1.3.1 Preeclampsie 1.3.1.1 Definition 16 16 16 III 1.3.1.2 Historique 17 1.3.1.3 Etiologie 18 1.3.1.4 Physiopathologie 19 1.3.1.5 Facteurs de Risque 22 1.3.1.6 Prevention Prediction et Traitement de la Preeclampsie....24 1.4 Stress Oxydatif. 26 1.4.1 Facteurs Prooxydants 1.4.1.1 Derives Reactifs de l'Oxygene 1.4.1.1.1 Peroxyde d'Hydrogene 1.4.1.2 Derives Reactifs Azotes 1.4.2 Facteurs Antioxydants 26 27 27 30 31 1.4.2.1 Antioxydants Enzymatiques 31 1.4.2.2 Antioxydants Non-Enzymatiques 32 1.4.2.2.1 VitamineE 32 1.4.2.2.2 VitamineC 33 1.4.3 Stress Oxydatif et Preeclampsie 1.5 TNF-a et ses Recepteurs l . S l Structure du TNF-a 34 35 35 1.5.2 Recepteurs du TNF-a 36 1.5.3 TNF-a et Preeclampsie 37 1.6 0bjectifs 38 2 Resultats 40 IV 2.1 Article 1 : Effect of H2O2 on placental regulation of TNF-a system: considerations for preeclampsia 40 2.1.1 Auteurs 40 2.1.2 Role joue par le candidat dans cette etude 40 2.1.3 Resume 41 2.1.4 Article 42 2.2 Article 2: Dual action of H2O2 on placental hCG secretion: Implications for oxidative stress in preeclampsia 66 2.2.1 Auteurs 66 2.2.2 Role joue par le candidat dans cette etude 66 2.2.3 Resume 67 2.2.4 Article 68 2.3 Article 3: Potential biomarkers of preeclampsia : inverse correlation between hydrogen peroxide and nitric oxide early in maternal circulation and at term in placenta of women with preeclampsia 85 2.3.1 Auteurs 85 2.3.2 Role joue par le candidat dans cette etude 85 2.3.3 Resume 86 2.3.4 Article 87 3 Discussion 116 4 Conclusions et Perspectives 124 5 Remerciements 126 6 References 127 V LISTE DES ILLUSTRATIONS Figures Figure 1 : L'oeuf implante a la fin de la deuxieme semaine 4 Figure 2 : Placenta a terme 12 Figure 3 : Les principales structures du placenta a terme 15 Figure 4 : Exposition des cellules eucaryotiques au peroxyde d'hydrogene 29 Tableaux Tableau I : Facteurs de risque de la preeclampsie 23 VI SIGLES, ABREVIATIONS ET SYMBOLES ADN: Acide desoxyribonucleique BCRP: Proteine de resistance du cancer du sein FSH: Hormone folliculostimulante GLUTs : Transporters du glucose GnRH: Gonado-liberine GPx: Gluthation peroxidase hCG: Hormone gonadotrophine chorionique humaine HLA-C : Complexe majeur d'histocompatibilite C KIRs: Recepteurs inhibiteurs cytotoxiques LH: Hormone luteinisante MDR1 : Proteine multi-drogue resistante NF- Facteur nucleaire KB KB : NO: Oxyde nitrique NOS: Oxyde nitrique synthase eNOS: Oxyde nitrique synthase endotheliale Nox: NADPH oxydase NADPH: Nicotinamide adenine dinucleotide phosphate PE: Preeclampsie RCIU: Retard de croissance intra-uterin ROS: Derives reactifs de l'oxygene SOD: Superoxyde dismutase Syk: Proteine tyrosine kinase TACE: EnzymeTNF-alpha convertase TNF-a : Tumor necrosis factor alpha TNFR1 : Receptor du tumor necrosis factor 1 TNFR2: Receptor du tumor necrosis factor 2 TRADD: TNFR1 -associated death domain protein TSH: Hormone de stimulation thyroidienne uNK: Natural killer uterine 1 INTRODUCTION 1.1 Grossesse La grossesse est le processus physiologique au cours duquel la progeniture vivante d'une femme se developpe dans son corps depuis la conception jusqu'a ce qu'elle puisse survivre hors du corps de la mere. La grossesse et la reproduction sont done, par leur importance, l'objet d'investigation de tres nombreux chercheurs et medecins partout au monde, car ces processus et les maladies qui y sont rattachees sont encore loin d'etre expliques et parfaitement compris. La grossesse implique de nombreuses etapes, allant de la fecondation a 1'accouchement, en passant par l'implantation et la placentation, qui seront decrites dans ce memoire. 1.1.1 Fecondation La fecondation proprement dite est 1'ensemble des phenomenes qui resultent de la rencontre entre le gamete male (spermatozo'ide) et le gamete femelle (ovocyte). Cette rencontre resulte en une fusion des deux gametes qui forment un noyau diploi'de, le zygote, au niveau de la trompe uterine. Le zygote migre ensuite vers la cavite uterine tout en debutant sa segmentation, une suite de divisions cellulaires. D'abord, les deux premieres cellules, ou blastomeres, apparaissent environ 24 heures apres la fusion, donnant ensuite chacune des cellules filles conduisant, avec un certain asynchronisme, a la formation d'un oeuf de 8, 16, 32 cellules, de taille de plus en plus petite, puis 64 cellules, ou le stade de morula est atteint environ 96 heures apres la fecondation. Entre le quatrieme et le cinquieme jour, les blastomeres 1 entrent dans la cavite uterine ou ils seront libres et s'appellent maintenant le blastocyste. II est limite par une couche cellulaire peripherique continue, le trophoblaste, qui deviendra le placenta. A un pole de la sphere, appele pole embryonnaire, les cellules constituant le bouton embryonnaire forment une masse cellulaire restant en contact avec le trophoblaste. A l'autre pole, les deux ensembles cellulaires sont separes par une cavite remplie de liquide appelee le blastocele (RABINEAU et al, 2003). 1.1.2 Implantation La fixation du blastocyste a Pendometre se produit vers le septieme jour apres la fecondation, et le phenomene d'implantation debute. Le blastocyste entre done en contact par son pole embryonnaire avec Pepithelium de Pendometre et le trophoblaste, la couche superficielle, prolifere activement au point de fixation. Les divisions nucleaires (cytodierese). II en successives resulte un interviennent syncytium sans derive division du cytoplasmique trophoblaste, le syncytiotrophoblaste qui commence a secreter Phormone chorionique gonadotrope (hCG). Le reste du trophoblaste, qui separe le bouton embryonnaire du syncytiotrophoblaste, reste constitue de cellules bien individualists et prend le nom de cytotrophoblaste. Vers la fin de la deuxieme semaine, le blastocyste sera completement enfonce dans Pendometre, ou la differentiation trophoblastique s'etendra tout autour de Poeuf qui sera completement cerne par une couche de syncytiotrophoblastes entourant les cytotrophoblastes. Des differentes structures qu'on y retrouve, les cytotrophoblastes resulteront en membranes choriales et 2 placenta, l'endometre en differentes deciduales, le disque embryonnaire, provenant du pole embryonnaire, sera a l'origine de Pembryon. On peut aussi y distinguer la cavite amniotique (Figure 1) (RABINEAU et al, 2003). 1.2 Placenta Lorsque la vie a commence a se developper a l'exterieur des oceans, de nouvelles strategies de reproduction ont emerge pour compenser la perte des nutriments et de l'oxygene normalement fournis par cet environnement. Malgre le fait que plusieurs exemples de viviparite existent chez les invertebres, les poissons, les amphibiens et les reptiles, la placentation est relativement nouvelle dans revolution. L'etablissement du lien etroit trophique entre la mere et l'embryon est une caracteristique des mammiferes. La placentation est done une nouvelle strategic dans la reproduction, permettant le developpement d'un petit nombre de foetus a l'interieur meme de l'organisme maternel protecteur. Comme les mammiferes ont probablement evolue d'animaux semblables aux petits rongeurs ayant de courts temps de gestation a des animaux plus gros avec des periodes gestationnelles prolongees, le placenta a du s'adapter aux besoins croissants du foetus. Le placenta est done devenu plus gros, augmentant considerablement la surface de contact entre la circulation maternelle et foetale et a acquis toute une panoplie de fonctions metaboliques, hormonales et immunologiques specialisees. C'est done un organe indispensable pour le developpement de l'embryon, maintenant un systeme de communication etroit entre la mere et le foetus, permettant le bon deroulement de la grossesse (BISCHOF et IRMINGER-FINGER, 2005). 3 Endometre — • r •'<•' /• I / Af ; J , Figure 1 : L'oeuf implante a la fin de la deuxieme semaine. On y retrouve les syncytiotrophoblastes entourant les cytotrophoblastes, 1'endometre et le disque embryonnaire. On peut aussi y distinguer la cavite amniotique. 4 1.2.1 Developpement Placentaire Une fois 1'implantation achevee, une differentiation et une proliferation des cellules trophoblastiques en deux structures distinctes surviennent, chacune ayant des fonctions differentes : les trophoblastes villositaires et extravillositaires. 1.2.1.1 Trophoblastes Villositaires Au cours de la deuxieme semaine post-fecondation, le syncytiotrophoblaste devient lacunaire et ces lacunes se remplissent de sang maternel par erosion des capillaires de l'endometre. A partir du treizieme jour, le syncytiotrophoblaste emet dans toutes les directions de l'espace des travees radiaires qui penetrent dans l'endometre et entrainent avec elles des cordons de cellules cytotrophoblastiques qui constituent l'axe des villosites primaires. Au cours de la troisieme semaine, les villosites se developpent et leur axe est envahi par le mesenchyme extraembryonnaire de la lame choriale et deviennent des villosites secondaires. Entre le 18e et le 21 e jour, des llots vasculo-sanguins se constituent dans l'axe mesenchymateux des villosites, comme dans tout le reste du mesenchyme extraembryonnaire; les villosites deviennent alors des villosites tertiaires (Figure 2). Les villosites tertiaires libres, dont Pextremite flotte dans la chambre intervilleuse, s'arboriseront pour former une importante surface de contact avec le sang maternel lorsque la circulation maternelle y sera etablie (RABINEAU et al, 2003). Les cytotrophoblastes des autres villosites tertiaires continueront de proliferer a leur 5 extremite pour emprunter la voie de differentiation des trophoblastes extravillositaires. 1.2.1.2 Trophobastes extravillositaires Les cytotrophoblastes de ces dernieres villosites, appelees villosites crampon, vont migrer et envahir l'endometre maternel. lis vont poursuivre leur migration dans la decidue et le premier tiers du myometre, realisant 1'invasion interstitielle. lis migrent notamment de facon specifique vers les parois des arteres uterines qu'ils vont eroder dans leur tiers superieur. lis transforment ainsi la tunique elastique arterielle en une paroi fibreuse atone, n'offrant que peu de resistance au flux sanguin maternel, et remplacent les cellules endotheliales maternelles dans le conduit arteriel (ALSAT et al, 1999). Des trophoblastes dits interstitiels vont s'agreger autour des arteres spiralees ou ils preparent les vaisseaux pour faciliter leur invasion par les trophoblastes endovasculaires (PIJNENBORG et al, 1981; PIJNENBORG et al, 1983). Ces derniers entrent dans la lumiere des arteres spiralees, formant un bouchon cellulaire qui empeche le sang maternel d'entrer dans l'espace intervilleux du placenta. Les trophoblastes detruisent ensuite Pendothelium vasculaire, en ne laissant qu'une mince couche de cellules endotheliales et aucune couche de muscles lisses. La consequence de ce phenomene est d'inhiber la reponse des vaisseaux maternels aux agents vasoactifs. Ce sont maintenant les vaisseaux utero-placentaires. 6 Le developpement des vaisseaux utero-placentaires se fait en deux vagues. La premiere se produit avant la 12e semaine post-fecondation et consiste en 1'invasion et la modification des arteres spiralees de la decidue et de Pendometre. La deuxieme vague a cours entre la 12e et la 16e semaine post-fecondation, resultant en une invasion et une modification des arteres spiralees du myometre en des arteres dilatees, et de faible resistance (RAMSEY et DONNER, 1980). Ce n'est qu'a partir de la 12e semaine, ou, parfois, un peu plus tard, que le sang maternel entre dans l'espace intervilleux (BURTON et al, 1999). 1.2.2 Fonctions placentaires Les principales fonctions du placenta peuvent etre classees sous les categories suivantes : transport et metabolisme, protection et fonctions endocriniennes. II permet Papport de tout ce que le foetus a besoin : oxygene, eau, glucides, acides amines, lipides, vitamines, mineraux et autre nutriments, tout en retirant le dioxyde de carbone (CO2) et autres dechets de la circulation foetale. II metabolise un nombre important de substances, pouvant etre liberees autant dans la circulation maternelle que fcetale (GUDE et al, 2004). Par exemple, le foetus, dont la principale source d'energie est le glucose, a une faible capacite de gluconeogenic Ce sucre doit done provenir de la circulation maternelle. Son transport a travers le placenta y est facilite par plusieurs transporteurs de glucose (GLUTs). Le recaptage du glucose maternel se fait par les syncytiotrophoblastes, et peut etre transports via la membrane basale, qui possede aussi des transporteurs de glucose, vers les capillaires foetaux. Le glucose peut aussi y etre converti en glucose-6-phosphate ou 7 en glycogene placentaire pour un usage ulterieur (BAUMANN et al, 2002). Par des processus semblables, des acides amines peuvent y etre convertis en proteines, et les lipides en acides gras et autres precurseurs. Les molecules plus petites, l'eau, les ions, les vitamines, peuvent y passer par transport passif, facilite, ou encore actif (STULC, 1997). Le placenta peut aussi proteger le foetus contre certaines substances ou bacteries pouvant circuler dans le sang maternel. Plusieurs peuvent traverser le placenta par diffusion simple, ou encore par les divers transporters qui ne sont pas completement specifiques a leur molecule endogene habituellement transportee. Toutefois, le placenta possede plusieurs systemes de protection pouvant aider a reduire le transfer! placentaire de substances potentiellement toxiques, incluant la proteine de resistance multidrogue 1 (MDR1) et la proteine de resistance au cancer du sein (BCRP) (MARIN et al, 2003). Le placenta contient aussi des enzymes cytochrome P450 qui peuvent metaboliser certaines drogues et autres xenobiotiques (PASANEN, 1999). Des proteines maternelles, dont des immunoglobulines de type G, y sont transportees a travers le placenta qui fournissent une immunite passive au foetus (MOFFETT et LOKE, 2004). Le placenta etant denue de nerfs, toute communication, tant avec la mere que le foetus, passe par la diffusion de substances a travers le sang. Ces facteurs endocrines, paracrines ou autocrines incluent les cestrogenes, la progesterone, la gonadotrophine chorionique humaine (hCG), le lactogene placentaire, l'hormone de 8 croissance placentaire, plusieurs facteurs de croissance, des cytokines, chemokines, agents vasoactifs, GnRH, ainsi que de nombreuses autres substances (GUDE et al, 2004). L'hCG, une hormone placentaire majeure dans nos recherches, est maintenant decrite plus en detail. 1.2.2.1 L'hCG Au debut du developpement embryonnaire, la presence de progesterone est requise pour le maintien de l'endometre, et est produite par le corps jaune, developpe a partir des cellules de la theque interne du follicule de De Graaf apres Povulation. L'hormone responsable de Pintegrite du corps jaune est 1'hCG, deja transcrite dans l'embryon a 8 cellules, et secretee par les trophoblastes, est detectable des le 8e jour apres Povulation dans le serum maternel. Sa concentration y augmentera exponentiellement jusqu'a la 10e semaine et decline progressivement par la suite, le placenta etant capable de produire suffisamment de progesterone. L'hCG est composee de deux sous-unites. La sous-unite a, commune a toutes les hormones glycoproteines (FSH, LH, TSH), est un polypeptide de 92 acides amines avec deux oligosaccharides lies en N-terminal. La sous-unite beta, specifique a 1'hCG, est un polypeptide de 145 acides amines, comprenant deux oligosaccharides en N-terminal et quatre en C-terminal. En plus de son role majeur de maintenance du corps jaune, 1'hCG est impliquee dans des mecanismes de regulation autocrines et paracrines de la differentiation trophoblastique (EVAIN-BRION et MALASSINE, 2003). 9 Comme les anomalies trophoblastiques jouent un role central dans la physiopathologie de maladies compliquant la grossesse dont la preeclampsie (hypertension gestationnelle), et qu'elles precedent ses signes cliniques, il est envisageable que le profil hormonal de 1'hCG y soit modifie dans la circulation maternelle, demontrant la defectuosite de la fonction placentaire (HSU et al, 1994). L'augmentation de la production d'hCG par les placentas preeclamptiques est done observee (BARROS et al, 2002). II a ete suggere que 1'hCG joue un role important dans la periode entourant Pimplantation en augmentant le flot sanguin uterin via la vasodilatation, l'angiogenese (DANTZER et al, 2000) et l'invasion trophoblastique (ZYGMUNT et al, 1998), resultant en de possibles effets therapeutiques benefiques (TOTH etal, 2001). 1.2.3 Structures du placenta a terme A terme, le placenta (Figure 2) est, dans plus de 90% des cas, un disque plat d'un diametre d'environ 22 centimetres, ayant une epaisseur au centre de 2.5 centimetres en moyenne, et ayant un poids d'environ 470 grammes. Bien entendu, ces dimensions sont tres variables, selon, entre autres, le type de mesure. II est compose a 85% d'eau, 12% de proteines, 0.11% de tissu adipeux et 1% de particules autres. II peut contenir jusqu'a 250 millilitres de sang. Le placenta est divise en quatre parties. D'abord, la surface foetale ou chorionique, qui comprend la plaque chorionique et la membrane entourant le foetus et contenant le liquide amniotique. La plaque basale, en contact direct avec le tissu uterin, est la surface maternelle du placenta. Entre les deux se confine Pespace intervilleux, ou le sang maternel circule 10 en contact avec les villosites choriales. Le cordon complete le placenta et relie les parties principales du placenta au foetus pour completer les echanges maternofoetaux. A la surface maternelle, 10 a 40 lobes sont presents et separes a l'interieur par des septa qui sont a l'origine de la resistance mecanique du developpement placentaire (BERNIRSCHKE et KAUFMANN, 2000). 1.2.3.1 Plaque chorionique La plaque chorionique, qui fait face au foetus, est une structure de plusieurs couches. Elle comprend l'amnios situe du cote fcetal, et le chorion. L'amnios comprend 5 couches : Perithelium amniotique, en contact direct avec le liquide du meme nom, la couche compacte, la couche fibroblastique, la couche spongieuse et la couche intermediate, qui ne sont pas visibles a l'ceil nu. Le chorion est une couche reticulee, et est la membrane basale et contient des cellules trophoblastiques, qui contribuent a Pattachement de la membrane a la decidue maternelle. La plaque chorionique joue un role de protection contre les infections provenant du tractus genital, et participe aussi au developpement foetal et a la progression normale de la grossesse. En effet, en plus de ses activites autocrines, cette membrane secrete des substances tant dans le liquide amniotique, affectant son homeostasie, que vers Puterus, ce qui peut influencer la physiologie cellulaire maternelle (GUDE et al, 2004). 11 Figure 2 : Placenta a terme. A : Cote foetal, comprenant la plaque chorionique, la membrane et le cordon. B : Cote maternel, comprenant la plaque basale. 12 1.2.3.2 Espace intervilleux L'espace intervilleux, situe entre la plaque basale et le chorion, contient le sang maternel qui baignent les villosites choriales permettant les echanges foetomaternels (Figure 3). L'espace intervilleux sert done de lieu d'echange, et contient enormement de villosites qui, sous leur forme arborescente, permettent un contact maximum entre le sang maternel et les syncytiotrophoblastes recouvrant les villosites (BERNIRSCHKE et KAUFMANN, 2000). 1.2.3.3 Plaque basale La plaque basale est la portion placentaire qui est du cote maternel. Sa couche interieure, ou deciduale basale, est formee de residus de quelques syncytiotrophoblastes, pratiquement inexistants a terme dans cette couche, et est recouverte par la couche fibrino'ide de Rohr. Vient ensuite la couche principale de la plaque basale. Elle est composee de cytotrophoblastes extravillositaires, de cellules deciduales et de tissu conjonctif. La couche fibrinoi'de uteroplacental recouvre le tout a la maniere d'un filet, et est constitute d'un melange de cellules maternelles et foetales. Cette portion est le site de nombreux processus immunologiques. La zone de separation vient ensuite delimiter le placenta et le tissu uterin. Elle est formee de cellules deciduales, de cellules stromales endometrials et de trophoblastes (BERNIRSCHKE et KAUFMANN, 2000). 13 1.2.3.4 Cordon ombilical La derniere portion placentaire, reliant le placenta au foetus, est le cordon ombilical. II est de longueur variable, allant de 24 a 146 cm avec une moyenne de 59 cm, et possede un diametre de 0.8 a 2 cm. II y a normalement 2 arteres et une veine dans le cordon ombilical humain, et c'est cette derniere qui amene Poxygene et les nutriments du placenta vers le foetus. Ces vaisseaux sont entoures de tissu conjonctif appele gelee de Wharton, substance composee de polysaccharides et contenant des fibroblastes. Le cordon est ensuite recouvert d'un epithelium amniotique, d'une epaisseur decroissante du placenta au foetus (BERNIRSCHKE et KAUFMANN, 2000). 14 Le placenta et les cotyledons. Figure 3 : Les principales structures du placenta a terme. Du tissu uterin vers le cordon on retrouve la plaque basale, ou deciduale basale, l'espace intervilleux contenant les villosites, et la plaque chorionique, qui contient les vaisseaux principaux qui se ramifient vers le cordon. 15 1.3 Hypertension gestationnelle et preeclampsie Dans certaines pathologies de grossesse, le placenta peut avoir un role important. Parmi celles-ci, l'hypertension est la plus commune des complications de grossesse. Elle est definie par une pression systolique d'au moins 140 mm de Hg et ou d'une pression diastolique d'au moins 90 mm de Hg, diagnostiquee apres 20 semaines de grossesse chez des femmes prealablement normotensives. Cette categorie inclut l'hypertension de grossesse legere et severe, et une autre forme avec une proteinuric qu'on appelle « preeclampsie » et ses variantes plus severes : Peclampsie, et le HELLP (Hemolytic anemia, Elevated Liver Enzymes and Low Platelet count), un syndrome associant hemolyse, enzymes hepatiques elevees et une diminution des plaquettes sanguines (Report of the national high blood pressure education program, working group on high blood pressure in pregnancy, 2000). Entre 6 a 17% des femmes nullipares et 2 a 4% des femmes multipares souffriront d'hypertension gestationnelle. Parmi celles-ci, 70% peuvent developper des complications reliees a la preeclampsie (SIBAI, 2003). 1.3.1 Preeclampsie 1.3.1.1 Definition La preeclampsie est definie, de facon primaire, comme etant une hypertension de grossesse (pression sanguine >140/90 mmHg) avec une proteinuric (300 mg ou plus par 24h de collection des urines) apres 20 semaines d'amenorrhee. Cette maladie affecte entre 2 et 7% des femmes nullipares prealablement en sante et est la cause 16 principale de morbidite et de mortalite perinatale (SIBAI, 2003, ROBERTS et GAMMILL, 2005). La preeclampsie peut aussi etre associee a des femmes souffrant d'hypertension chronique. Cette pathologie se divise en deux categories, soit la preeclampsie legere et la preeclampsie severe. Cette derniere peut inclure des criteres de severite tels qu'un oedeme pulmonaire, un gain de poids important, des convulsions, une oligurie, une hyperuricemie, une thrombocytopenic, des enzymes hepatiques eleves, une hypoalbuminemie, des douleurs epigastriques, de violentes migraines, des troubles visuels et un statut mental altere. Certaines de ces complications peuvent survenir apres le diagnostique de la preeclampsie, et plus elles arriveront tot en grossesse, par exemple, avant la 33 e semaine de grossesse, plus le pronostique est mauvais (SIBAI, 2003). La preeclampsie est aussi associee a de nombreuses autres pathologies de grossesse touchant le foetus, comme l'accouchement premature (30-44%), le retard de croissance intra-uterin (16-29%) (TUFFNELL et al, 2005, YUCESOY et al, 2005, MAGEE et al, 2003, ALMULHIM et al, 2003). 1.3.1.2 Historique II y a 2000 ans, Celse, l'Hippocrate latin, a decrit des convulsions chez les femmes enceintes qui disparaissaient avec l'accouchement. Puisque cette condition semblait survenir sans avertissement, elle fut nommee eclampsie, derive du mot grec eclair. Pour les 1900 annees suivantes, l'eclampsie etait consideree comme un desordre unique a la grossesse. Ce n'est qu'a la moitie du 19e siecle que des similitudes entre la femme souffrant d'eclampsie avec les patients atteints de la 17 maladie de Bright (glomerulonephrite aigue) ont mene a la decouverte que ces femmes avaient frequemment des proteines dans leur urine et que cette condition precedait l'eclampsie. Cinquante ans plus tard, le developpement de techniques non-invasives de mesure de la pression sanguine permit de faire le meme type d'association avec 1'augmentation de la pression sanguine (CHESLEY, 1978). Rapidement, il fut reconnu que ce syndrome, meme sans les convulsions, presentait des risques autant pour la mere que pour le bebe. La preeclampsie fut alors reconnue comme un syndrome maternel ou les convulsions n'etaient qu'une manifestation de la maladie. Malheureusement, la reconnaissance que ce syndrome n'etait pas qu'un probleme de convulsions a ete suivi par le concept que la preeclampsie etait une maladie hypertensive. Les tentatives pour comprendre ce desordre pour les 70 annees suivantes furent grandement concentrees sur l'etude de la pression sanguine, ignorant toute une serie d'autres changements physiologiques. Plus recemment, l'attention fut redirigee vers l'ensemble du syndrome, resultant en une meilleure comprehension de la maladie (ROBERTS et REDMAN, 1993). 1.3.1.3 Etiologie Bien que de nombreuses hypotheses aient ete enoncees sur l'etiologie de la preeclampsie, les causes exactes de cette maladie sont toujours inconnues, lui valant le surnom de la maladie des theories, dont la plupart n'ont pas traverse l'epreuve du temps. Les theories potentielles toujours en cours incluent 1'invasion trophoblastique deficiente des vaisseaux uterins, une intolerance immunologique entre le placenta et les tissus maternels, une mesadaptation aux changements 18 cardiovasculaires ou inflammatoires de la grossesse, une deficience par rapport a la diete, ou encore des anomalies genetiques (SIBAI, 2003). 1.3.1.4 Physiopathologie Un modele en deux etapes a ete propose pour conceptualiser la physiopathologie de la preeclampsie. Le stade 1 consiste en une mauvaise placentation et une irrigation placentaire reduite, pouvant, chez certaines femmes, conduire au stade 2 : le syndrome maternel multi-systemique de la preeclampsie. Bien que beaucoup d'incertitudes demeurent dans ces deux stades, les questions principales sont: pourquoi la reduction de 1'irrigation placentaire resulte en preeclampsie chez certaines femmes seulement, et qu'est-ce qui lie ces deux stades. Stade 1 de la preeclampsie: Reduction de Pirrigation placentaire II y a environ 70 ans, Ernest Page a formule le concept que la perfusion placentaire etait reduite dans la preeclampsie (PAGE, 1939). Pour expliquer ce phenomene, le remodelage des vaisseaux uterins au site d'implantation a ete cible. Comme explique plus tot, il appert que le remodelage de ces vaisseaux est le resultat de l'invasion trophoblastique, et en particulier, l'invasion endovasculaire. Le controle cellulaire de l'invasion cytotrophoblastique depend des interactions entre la deciduale maternelle et les trophoblastes. La tension en oxygene locale et les interactions immunologiques sont particulierement determinantes dans ce processus, dont l'apoptose serait aussi impliquee (HUNG et al, 2002). 19 La tension en oxygene dans l'espace intervilleux est faible jusqu'a environ 12 semaines post-conception, ou les vaisseaux maternels commencent a perfuser cet espace. La tension en oxygene y augmente dramatiquement, tout comme les derives reactifs de Poxygene. II est done propose que si la capacite maternelle en antioxydants est insuffisante, cela nuit a l'invasion, resultant en une faible perfusion placentaire et, possiblement, en preeclampsie (JAUNIAUX et al, 2000). La regulation de la differentiation et de l'invasion des trophoblastes implique aussi des mecanismes immunologiques. Les cellules immunitaires predominates dans la deciduale sont les cellules uNK (natural killer uterines). Les recepteurs KIRs (killer immunoglobulin receptors) presents sur ces cellules interagissent avec des marqueurs trophoblastiques specifiques, influencant l'invasion. L'antigene leucocytaire trophoblastique humain de classe C (HLA-C) est central dans ses interactions avec la deciduale (KING et al, 2000). Des combinaisons genotypiques specifiques de KIRS et HLA-C resultent en un risque accru de preeclampsie (HIBY et al, 2004). L'apoptose placentaire pourrait etre la derniere voie associant ces deux mecanismes. Les placentas de patientes preeclamptiques ont une apoptose generalisee plus elevee que les controles (ALLAIRE et al, 2000). L'apoptose mene aussi au relachement de fragments syncytiotrophoblastiques dans la circulation maternelle, accelerant la preeclampsie (KNIGHT et al, 1998). 20 Stade 2 : Syndrome maternel de la preeclampsie Tous les changements physiologiques presents chez les femmes souffrant d'eclampsie ont un point commun : une reduction de la perfusion dans pratiquement tous les organes. Par exemple, les changements pathologiques dans les reins consistent en une endofheliose glomerulaire caracterisee par une tumefaction des cellules glomerulaires endotheliales suffisante pour bloquer la lumiere capillaire, accompagnee de legers changements dans les podocytes renaux (MCCARTNEY, 1969). Ce changement est important, puisqu'il est absent dans les autres types d'hypertension. La diminution de la perfusion est aussi reduite en raison de spasmes vasculaires, de 1'activation de la coagulation intravasculaire, et de la perte de fluide de l'espace intravasculaire (ROBERTS et LAIN, 2002). Les femmes preeclamptiques sont aussi singulierement sensibles aux agents vasoactifs (CHESLEY, 1966). Tous ces changements ont mene au concept de la dysfonction endotheliale comme point central physiopathologique de la preeclampsie (ROBERTS et LAIN, 2002), qui peut expliquer tous les changements observes chez la mere. La question centrale dans le modele impliquant deux stades de la preeclampsie est le lien qui unit ces deux stades. Cette question demeure toutefois le saint graal de la recherche sur la preeclampsie, puisque ce phenomene semble impliquer de nombreux facteurs, soit genetiques, immunologiques et biochimiques. De ces facteurs, le stress oxydatif, via les ROS, semble jouer un role preponderant dans la preeclampsie, et qui pourrait soit contribuer ou stimuler le lien entre les deux stades (ROBERTS et GAMMILL, 2005). En effet, les cytokines provoquent le 21 relachement de radicaux libres (CONNER et GRISHAM, 1996), alors que les monocytes et les neutrophiles actives en relachent aussi quand ils entrent en contact avec de Pendothelium active. Le stress oxydatif entraine aussi l'apoptose placentaire, provoquant le relachement possible de particules microvillositaires contenant des lipides oxydes, qui peuvent agir sur tout le systeme (ROBERTS et GAMMILL, 2005). 1.3.1.5 Facteurs de risque De nombreux facteurs de risque presents dans la periode de preconception ont etes associes avec une augmentation des risques de preeclampsie (Tableau I). Les femmes ayant des problemes medicaux avant la grossesse auront plus de risques d'avoir une preeclampsie que toute autre grossesse. Le niveau de risque depend bien entendu de la condition medicale specifique de chaque femme, ainsi que de la severite de sa condition. Par exemple, une femme souffrant d'hypertension chronique, en melant tous les cas, a un risque de souffrir de preeclampsie de 25% durant sa grossesse. Ce risque ne sera que de 15% pour les femmes souffrant d'hypertension legere avant ou au debut de la grossesse, alors qu'il frise les 50% dans les cas d'hypertension severe (BARTON et SIBAI, 2008). II en va de meme pour la plupart des facteurs de risque, dont l'obesite, ou les risques augmentent avec l'indice de masse corporelle, probleme lui-meme associe a la resistance a Pinsuline qui est aussi un facteur de risque de la preeclampsie (CATALANO, 2007). Etrangement, l'usage de la cigarette diminuerait les risques de souffrir de 22 preeclampsie durant la grossesse de 33% (CONDE-AGULEDO et al, 1999). Les mecanismes associes sont toutefois toujours a expliquer. Tableau I : Facteurs de risque de la preeclampsie Nulliparity Grossesse multiple Obesite Historique familial de preeclampsie/eclampsie Preeclampsie dans une grossesse precedente Diabete pregestationnel de type 2 Thrombophilie Infections urinaires durant la grossesse Hypertension chronique ou gestationnelle, problemes renaux Age maternel de plus de 40 ans Exposition limitee au sperme paternel (nouveau partenaire, insemination, fecondation in vitro) Pere ayant engendre une grossesse preeclamptique chez une autre femme Femme ayant elle-meme souffert de retard de croissance intra-uterin Complications dans une grossesse precedente (retard de croissance intra-uterin, hematome retroplacentaire, mort fcetale) (Adapte de SIBAI, 2003, BARTON et SIBAI, 2008) 23 1.3.1.6 Prevention, prediction et traitement de la preeclampsie Comme mentionne precedemment, il y a plusieurs facteurs qui peuvent entrainer une augmentation des risques de developper une preeclampsie. Ces risques, identifies avant la grossesse, peuvent amener a une intervention qui constitue une prevention primaire de la preeclampsie. Par exemple, des etudes epidemiologiques ont demontre qu'un controle du poids pouvait affecter Tissue de la gestation (VILLAMOR et CNATTINGIUS, 2006, BAETEN et al, 2001). Pour une femme obese, l'obtention d'un poids ideal avant la conception pourrait done etre une bonne tentative de reduire les risques de preeclampsie. Les femmes avec diabete seraient aussi encouragees a avoir leurs enfants le plus tot possible, avant que leur situation ne se deteriore, et a controler leur diabete et leur hypertension, si presente, de facon appropriee, et ce, plusieurs mois avant la grossesse, et durant celle-ci (BARTON et SIBAI, 2008). Pour la prevention secondaire de la preeclampsie, il y a eu de nombreux essais cliniques decrivant l'usage de nombreuses methodes pour prevenir ou reduire les risques de preeclampsie (SIBAI et al, 2005). L'etiologie de cette maladie etant inconnue, ces interventions ont ete effectuees pour corriger des anomalies physiopathologiques theoriques de la preeclampsie durant la grossesse. L'usage de drogues anti-hypertensives pour les femmes hypertendues, de supplementation en huile de poisson, de calcium, de vitamines C et E et d'heparine se sont averees inefficaces pour la prevention de la preeclampsie. En plus d'avoir des effets secondaires nefastes potentiels, dans le cas des huiles de poisson, et des hautes doses de vitamines C et E (BARTON et SIBAI, 2008). Par exemple, un essai 24 clinique realise en utilisant des fortes doses de vitamines C et E pour retablir Pequilibre oxydatif chez des femmes a risque, n'a pas change 1'incidence de la preeclampsie, mais il a resulte en une augmentation significative des bebes de petit poids a la naissance (POSTON et ah, 2006). Toutefois, l'usage de faibles doses d'aspirine pourrait, chez certaines femmes a risque, par exemple celles ayant deja eu une preeclampsie resultant en un accouchement premature, aider a reduire P incidence de la preeclampsie (ASKIE et al, 2007). Pour ce qui est de la prediction de l'apparition de la preeclampsie, de nombreux tests cliniques, biophysiques et biochimiques ont ete proposes pour la predire ou la detecter de facon precoce. Toutefois, la plupart de ces tests souffrent d'un manque de sensibilite ou d'un faible pourcentage de prediction, et la majorite d'entre eux ne sont pas faits pour un usage de routine en clinique (SIBAI, 2003). Le diagnostic repose done toujours sur la mesure de la pression arterielle et la mesure des proteines dans une collecte urinaire de 24 heures, ainsi que sur le suivi des femmes les plus a risque. Cette methode a aussi ses limites, par exemple pour les femmes qui ne font qu'une legere hypertension. Une fois la preeclampsie diagnostiquee, les traitements utilises sont nombreux, mais peu efficaces, et souvent controverses. Ces traitements peuvent soulager certains symptomes, mais aucun ne peut guerir la maladie, si ce n'est que Paccouchement. Toutefois, les symptomes peuvent quand meme continuer a s'aggraver apres Paccouchement. L'usage d'agents antihypertenseurs, un repos complet ou partiel font partie des traitements recommandes, et peuvent aider a reduire les risques d'hypertension severe. Dans les cas de preeclampsies severes, le 25 sulfate de magnesium est utilise pour eviter les convulsions et le risque d'aggravation en eclampsie, mais la delivrance reste le seul traitement efficace (SIBAI, 2003). 1.4 Stress oxydatif Le stress oxydatif est decrit comme etant un debalancement entre la production de derives reactifs de Poxygene (ROS) et la capacite des antioxydants de les neutraliser. Ce desequilibre peut provenir soit d'une augmentation de la production des ROS et/ou d'une diminution des antioxydants. Le plus souvent, ces ROS sont des radicaux libres et provoquent des dommages cellulaires en reagissant avec des proteines, lipides ou avec les acides nucleiques (MYATT et CUI, 2004). 1.4.1 Facteurs prooxydants Les facteurs prooxydants sont surtout composes de radicaux libres, qui sont des molecules instables et tres reactives. lis deviennent stables en faisant l'acquisition d'un electron provenant d'acides nucleiques, de lipides, de proteines ou toute autre molecule, provoquant une reaction en chaine resultant en un possible dommage cellulaire et possiblement en une pathologie (PIERCE et ah, 2004, WARD et CROFT, 2006). II y a deux categories majeures de radicaux libres : les derives reactifs de l'oxygene (ROS) et les derives reactifs azotes. 26 1.4.1.1 Derives reactifs de l'oxygene Les trois principaux types de ROS sont l'anion superoxyde (O2"), le peroxyde d'hydrogene (H2O2) et le radical hydroxyle (OH). L'anion superoxyde est forme lorsque des electrons s'echappent de la chaine de transport des electrons (HALLIWELL et al, 1992). La dismutation de cet anion resulte en la formation de peroxyde d'hydrogene. Le radical hydroxyle est aussi tres reactif et peut modifier les purines et pyrimidines et causer des dommages a l'ADN (MELLO FILHO et al, 1984). Toutefois, bien qu'ils soient surtout connus pour leurs effets nefastes, les ROS auraient aussi un role important dans le remodelage tissulaire, la signalisation, le cycle cellulaire, l'apoptose, les fonctions des gametes et la reproduction (RILEY et BEHRMAN, 1991, NOSE, 2000) 1.4.1.1.1 Peroxyde d'hydrogene Depuis quelques annees, il ne fait maintenant aucun doute que le peroxyde d'hydrogene (H2O2), bien connu pour ses effets cytotoxiques, joue un role important dans de nombreux processus biologiques dans la plupart, si ce n'est pas dans tous les organismes vivants. Chez les procaryotes et levures, des mecanismes de detection et de reponse a l ' E L ^ dans l'environnement ont ete developpes, et de nombreux organismes vivants produisent intentionnellement de l ' L L ^ dans divers buts, comme lors de la defense immunitaire ou dans la chaine respiratoire mitochondriale. De plus, il y a maintenant de fortes evidences suggerant que PH2O2 27 pourrait aussi avoir certaines fonctions dans des voies de signalisation cellulaire, en particulier chez les vertebres (STONE et YANG, 2006). La source majeure d'F^C^ implique la reaction spontanee ou catalytique de l'anion superoxyde (O2-), produit par la reduction partielle de l'oxygene durant la respiration aerobique et suivant l'exposition des cellules a de nombreux agents physiques, chimiques et biologiques (Figure 4). Par exemple, il est etabli que les cellules phagocytaires activent des complexes NADPH oxydase (Nox) pour generer l'anion superoxyde, par consequent PH2O2, comme agent cytotoxique durant la destruction des microbes. Toutefois, il est maintenant demontre que ces Nox ne sont pas seulement presentes chez les cellules phagocytaires, mais aussi dans de nombreux types de cellules et tissus. En fait, dans les cellules non-immunitaires, les Nox sont associes a la generation de derives reactifs de l'oxygene (ROS) comme molecules de signalisation (VEAL et al, 2007). 28 phagocytes receplor/ligand interaction immunoqlobulin G 0 m ^ Si h *•*!, ^ ^ tnlegrins,, cytoplasm ss-if^i o, ^ O2- • • * H 2 O z ampn growth factors cytokines integrifis o. O2" iiOf Cj'teCtlfWB*! oxtdsM Vi-" mitochondria Figure 4 : Exposition des cellules eucaryotiques au peroxyde d'hydrogene. Le peroxyde d'hydrogene peut etre produit l'oxydation de l'eau par les immunoglobulines G, par des interactions recepteur-ligand, et par les cellules phagocytaires. L'anion superoxyde est lui aussi rapidement converti en peroxyde d'hydrogene par Taction des superoxyde dismutases. Des facteurs de croissance, cytokines et integrines stimulant aussi l'activation de la NADPH oxydase ou 5'lipoxygenase. Le peroxyde d'hydrogene peut aussi diffuser a travers la membrane. 29 1.4.1.2 Derives reactifs azotes La production d'oxyde nitrique (NO) implique la NO synthase (NOS) qui, en la presence suffisante du cofacteur tetrahydrobiopterin, recoit des electrons du NADPH pour convertir la L-arginine et Poxygene moleculaire en NO et L-citruline. II y a trois isoformes de NOS : la NOS neuronale (nNOS), la NOS inductible (iNOS) et la NOS endothelial (eNOS), qui sont respectivement exprimees dans les neurones, dans les globules blancs, et dans les cellules endothelials et le placenta (SCHIESSL et al, 2005). D'un autre cote, les arginases jouent un role important dans le metabolisme du NO en catabolisant la L-arginine en uree et en L-ornithine. II y a deux formes d'arginases. L'arginase I, une enzyme cytosolique, et l'arginase II, qui est une enzyme mitochondriale (CEDERBAUM et al, 2004, MORRIS, 2002). Ces deux arginases sont exprimees par les cytotrophoblastes dans le placenta en debut de grossesse et a terme (ISHIKAWA et al, 2007). Le NO a plusieurs roles physiologiques importants, comme la maintenance de l'equilibre vasculaire systemique, un mediateur de la reponse inflammatoire et la promotion des proprietes antithrombiques de 1'endothelium. L'implication du NO dans la preeclampsie est largement etudiee, en depit de la controverse entourant sa cinetique et les doutes entourant sa regulation. Quelques etudes ont done constate une augmentation du NO dans la circulation maternelle (BHATNAGAR et al, 2007, VURAL, 2002), et dans le placenta (SHAAMASH et al, 2000), alors que d'autres ont constate une diminution dans le serum maternel (SANDRIM et al, 2008) et le placenta (SCHONFELDER et al, 2004), ou encore des niveaux normaux (ORANGE et al, 2003). L'anion superoxyde peut aussi reagir avec le NO, 30 resultant en un puissant agent oxydant, l'anion peroxynitrite (ONOO) qui peut interagir avec de nombreuses molecules (BECKMAN et al, 1990). 1.4.2 Antioxydants Dans des conditions normales, les antioxydants convertissent les ROS en eau pour prevenir la surproduction des ROS. II y a deux types d'antioxydants dans le corps humain : les antioxydants enzymatiques et les antioxydants non-enzymatiques. 1.4.2.1 Antioxydants enzymatiques Les antioxydants enzymatiques sont composes principalement des families des superoxydes dismutases (SOD), des catalases et des gluthation peroxydases (GPx). La dismutation de l'anion superoxyde en peroxyde d'hydrogene est souvent appelee la premiere defense contre les ROS, pour prevenir la formation subsequente d'autres radicaux libres. Les SOD sont classifies en trois classes, selon l'ion metallique qu'elles contiennent, soit les Cu/Zn SOD, les Mn SOD et les Fe SOD. Leur activite est surtout intracellulaire, et leur presence varie selon les tissus (MILLER, 2004). Les catalases sont principalement responsables de l'acceleration de la decomposition de PH2O2 en H2O. Les catalases sont distributes un peu partout dans l'organisme, avec une presence variable selon les tissus. Les GPx, une famille comportant 5 formes, chacune ayant des fonctions semblables, mais des modes et des sites d'action differents (BRIGELIUS-FLOHE et al, 1994). Elles sont aussi responsables de la catalyse de l ' L L ^ et des peroxydes organiques en H2O, et ce, a 31 des concentrations plus basses de peroxyde d'hydrogene que la catalase, cette derniere etant plus active a des concentrations plus elevees (YU, 1994). 1.4.2.2 Antioxydants non-enzymatiques Les antioxydants non-enzymatiques proviennent principalement de l'alimentation et comprennent, entre autres, la vitamine C, la vitamine E, le selenium, le zinc, la glutathione, le p-carotene et la taurine (AGARWAL et ah,2003). Leur mode d'action est varie, et ils sont presents tant au niveau intracellulaire qu'extracellulaire, dependamment de leurs proprietes, comme leur solubilite aqueuse ou lipophile. Les vitamines C et E sont de tres puissants antioxydants, et sont interessants, car ils peuvent etre facilement utilises comme supplements alimentaires pour tenter de corriger un desequilibre oxydatif. 1.4.2.2.1 Vitamine E La vitamine E est Pantioxydant le plus distribue dans la nature, tant chez les plantes que dans le monde animal. La vitamine E est un terme generique qui refere a au moins huit isomeres structuraux de tocopherol, dont Pa-tocopherol, le plus connu, possede l'activite antioxydante la plus importante (BURTON et al., 1982). Les isomeres de tocopherol sont des molecules lipophiles, ce qui fait que la vitamine E est le plus important antioxydant dans les environnements lipophiles, comme dans les lipoproteines plasmatiques. Au niveau intracellulaire, la vitamine E est associee aux membranes riches en lipides comme les mitochondries et le 32 reticulum endoplasmique. L'action antioxydante de cette vitamine est done importante pour contrer la peroxydation des membranes lipidiques en reagissant avec les lipides peroxydes et radicaux alkoxyl (YU, 1994). 1.4.2.2.2 Vitamine C La vitamine C, ou acide ascorbique, est hydrophile et est plus efficace dans un environnement aqueux que la vitamine E. C'est un agent reducteur et antioxydant, qui reagit directement avec l'anion superoxyde ou le radical hydroxyle et divers lipides peroxydes. II peut aussi restaurer les proprietes antioxydantes de la vitamine E oxydee (WITTING et HORWITT, 1964), fournissant a cette vitamine une fonction importante dans le recyclage de la vitamine E. La vitamine C est largement distribute dans les tissus des mammiferes, mais moins presente dans certains organes comme le foie et le cerveau. Toutefois, la vitamine C peut aussi avoir des effets prooxydants. A de plus fortes concentrations, soit environ 1 mM, et en presence de metaux de transition comme le FE3+ ou le Cu2+, elle peut generer des cofacteurs de ROS et promouvoir la peroxydation lipidique. La complexite de son role est mise en evidence en presence de vitamine E, ou la peroxydation lipidique des microsomes chez des animaux deficients en vitamine E est augmentee, mais diminuee par la vitamine C dans des conditions normales. Son role semble done etre ambivalent, selon les conditions environnantes (YU, 1994). 33 1.4.3 Stress oxydatif et preeclampsie II y a de nombreuses evidences d'un debalancement oxydatif chez les femmes souffrant de preeclampsie, incluant une augmentation plasmatique de la concentration des produits d'oxydation des radicaux libres (WICKENS et ah, 1981), de lipides peroxydes (HUBEL et ah, 1996) et une diminution de la concentration d'antioxydants comme les vitamines C et E (MIKHAIL et ah, 1994). Notre laboratoire a aussi demontre une augmentation d'H202 dans le serum de femmes preeclamptiques par rapport aux femmes ayant une grossesse normale (KHARFI et ah, 2005). Les consequences de ce stress oxydatif sont nombreuses. Par exemple, la severite d'un debalancement en faveur des facteurs prooxydants a ete correlee avec une augmentation de la pression sanguine (MADAZLI et ah, 2002). Le TNF-a (tumor necrosis factor- a), est anormalement eleve dans les placentas preeclamptiques (CONRAD et ah, 1998), probablement en reponse au stress oxydatif. Le TNF-a luimeme peut aussi contribuer au stress oxydatif en induisant la production des ROS (WANG et WALSH, 1996). Les etudes de recherche sur le stress oxydatif dans la preeclampsie, bien qu'elles identifient clairement la presence d'un debalancement entre prooxydants et antioxydants, ne parviennent pas a elucider les causes. 34 1.5 TNF-a et ses recepteurs Le TNF-a, une cytokine pleiotrope pro-inflammatoire, est surtout exprimee par une variete de cellules inflammatoires activees telles que les monocytes, les macrophages, les neutrophiles, les cellules NK, les lymphocytes T, mais aussi par des cellules non-inflammatoires, comme les astrocytes, les adipocytes, les keratinocytes et les fibroblastes (TRACEY et CERAMI, 1993). Elle est connue pour affecter la croissance, la differentiation, la survie et les fonctions physiologiques de nombreuses cellules (BEUTLER et CERAMI, 1988; BEUTLER et CERAMI. 1989). 1.5.1 Structure du TNF-a Le TNF-a est exprime sous la forme d'une proteine homotrimerique transmembranaire, qui est ensuite relachee sous sa forme homotrimerique soluble par l'enzyme TNF-a convertase (TACE), une metalloproteinase liee a la membrane (WAJANT et al, 2003). Les protomeres du TNF-a sont de 17 kDa chacun sous la forme soluble, et sont composes de deux feuillets beta anti-paralleles (ECKS et SPRANG, 1980). Chacune des formes solubles et membranaires (26 kDa) existe, et se lie principalement au recepteur 1 du TNF (TNFR1) et au recepteur 2 du TNF (TNFR2) (IDRISS et NAISMITH, 2000). 35 1.5.2 Recepteurs du TNF-a Le TNF-a exerce ses effets via les recepteurs TNFR1 (p55) et TNFR2 (p75), ayant des poids moleculaires de 55 et 75 kDa, respectivement (SHU et al, 1996). Ces deux recepteurs ne partagent que peu d'homologie, ce qui indique qu'ils peuvent induire differentes fonctions (BEYAERT et FIERS, 1994). Contrairement au TNFR1 qui est communement exprime dans la plupart des cellules, le TNFR2 est plus regule et trouve principalement dans les cellules immunitaires et hematopoietiques (WAJANT et al, 2003). Les signaux medies par les recepteurs du TNF-a sont divers, et peuvent avoir des effets similaires via des mecanismes distincts. L'activation d'un ou des deux recepteurs depend de la voie de signalisation mise en jeu (THOMMESEN et LAEGREID, 2005). Des differences fonctionnelles entre le TNFR1 et le TNFR2 concernant la cytotoxicite, les dommages a l'ADN, et la differentiation des macrophages ont ete observes (HIGUCHI et AGGARWA1, 1994). Dans certains cas, les deux recepteurs peuvent avoir des effets opposes. Tandis que le TNFR2 promeut la survie cellulaire, le TNFR1 favorise plutot la mort cellulaire (FONTAINE et al, 2002). Des lignees cellulaires exprimant soit le TNFR1 ou le TNFR2 ont ete utilisees pour confirmer que la cytotoxicite est principalement mediee par le TNFR1 (HIGUCHI et AGGARWAL, 1994). L'absence d'un domaine cytoplasmique de mort cellulaire chez le TNFR2, le domaine de liaison a la proteine TRADD (TNFR1-associated death domain protein), est attribue a sa signalisation de survie cellulaire. 36 Bien que ces recepteurs soient principalement lies a la membrane, ils peuvent aussi etre retrouves dans une forme soluble active, mimant une action paracrine (LOTZ et al, 1996). Les domaines extracellulaires des recepteurs peuvent aussi etre clives par la TACE, et deviennent des fragments solubles qui ont la capacite de neutraliser le TNF-a circulant (KEITH et al, 2000, WAJANT et al, 2003). Les formes solubles des deux recepteurs ont la capacite de Her, mais aussi de relacher le TNF-a (REDDY et al, 1994), ce qui peut entrainer deux effets. Ils peuvent agir comme inhibiteurs en liant le TNF-a, le rendant non disponible, ou agir comme reservoir, relachant lentement le TNF-a et prolongeant ses effets biologiques. La premiere condition survient a de hautes concentrations de recepteurs solubles, alors qu'a de plus faibles concentrations, les recepteurs solubles stabilisent la structure du TNF-a et augmentent son activite (ADERKA et al, 1992). Le sTNFRl, contrairement au sTNFR2, voit son expression etre indifferente aux niveaux de TNF-a circulant (VANDENABEELE et al, 1995). 1.5.3 TNF-a et preeclampsie Lors de la grossesse, le TNF-a joue un role tres important (HUNT et al, 1996). Dans le deroulement normal de la grossesse, il semble etre benefique au developpement foetal, puisqu'il est exprime dans de nombreux organes. Dans le placenta humain, l'augmentation de cette cytokine entraine l'apoptose des cytotrophoblastes isoles a partir de placenta a terme, suggerant un role important dans le renouvellement des trophoblastes (YUI et al, 1994). II a ete demontre que le TNF-a joue un role dans la regulation de la production de la progesterone, de 37 Pestradiol, et de 1'hCG sur des explants placentaires, des cellules JEG-3 (lignee cancereuse trophoblastique) ou des cytotrophoblastes isoles (LI et al, 1992), suggerant qu'il est aussi implique dans la differentiation et la fonction des trophoblastes. L'alteration des niveaux de TNF-a a ete constatee dans le sang de femmes souffrant de preeclampsie (BENYO et al, 2001), et il y a actuellement une incertitude sur la provenance de ce TNF-a, soit de la mere, ou du placenta. La theorie qui prevaut presentement veut que l'augmentation de l'expression du TNF-a dans le placenta puisse mener a l'activation des leucocytes circulants, liant ainsi le stress oxydatif du placenta a un stress oxydatif generalise et a une dysfonction endothelial chez la mere (WALSH, 1998). 1.6 Objectifs Une des conditions physiopathologiques majeures de la preeclampsie est un stress oxydatif eleve. Des travaux anterieurs effectues dans notre laboratoire ont demontre une correlation entre les nivaux circulants du peroxyde d'hydrogene (H2O2) et de P hormone gonadotrophine chorionique humaine (hCG) chez les femmes souffrant de preeclampsie (KHARFI et al,. 2005). II est done concevable que ce stress oxydatif a des effets directs sur le placenta. Afin de determiner les effets du peroxyde d'hydrogene sur les cellules placentaires in vitro, en rapport aux conditions retrouvees in vivo chez des femmes preeclamptiques, nous avons developpe un modele d'etude de cellules trophoblastiques primaires humaines in vitro, avec lequel nous avons evalue les effets du peroxyde d'hydrogene suivants : 38 1 Effet du peroxyde d'hydrogene sur la production placentaire de 1'hCG, explorant ainsi la reponse endocrinienne du placenta au stress oxydatif. 2 Effets du peroxyde d'hydrogene sur la production placentaire du TNF-a, explorant ainsi la reponse inflammatoire du placenta au stress oxydatif. 3 Effet du peroxyde d'hydrogene sur la production placentaire du NO, explorant ainsi la reponse vasoactive du placenta au stress oxydatif. Ces investigations in vitro nous permettent de faire des liens entre le stress oxydatif et certaines conditions rencontrees in vivo lors de la preeclampsie. II est a noter que Pensemble de ces travaux nous ont permis de preparer 3 articles dont deux sont publies aujourd'hui, il s'agit de l'effet sur 1'hCG et sur le NO. L'effet de l'H202 sur le TNF-a constitue mon article principal dans ce memoire et qui est soumis au journal « Free Radical Research ». 39 2 RESULTATS 2.1 Article 1 : Effect of H2O2 on placental regulation of TNF-a system: considerations for preeclampsia 2.1.1 Auteurs : Samuel Leblanc, Jean-Marie Moutquin, Regen Drouin, Yves Giguere, Jean-Claude Forest, Aziz Aris 2.1.2 Role joue par le candidat dans cette etude : Design experimental, realisation des experiences, compilation des donnees, analyse statistique des donnees, interpretation des resultats et ecriture de l'article scientifique. Soumis le 7 avril 2009 au journal « Free Radical Research ». 40 2.1.3 Resume Nos recherches precedentes ont etabli le role du peroxyde d'hydrogene (H2O2) comme un biomarqueur liable du stress oxydatif de la preeclampsie (PE). La PE est caracterisee par une elevation de mediateurs oxydatifs et inflammatoires dans la circulation maternelle et le placenta. Toutefois, les mecanismes de regulation reliant le placenta et le stress inflammatoire ne sont pas encore bien compris. Le but de cette etude etait done d'etudier les effets du H2O2 sur l'expression placentaire du TNF-a et de ses recepteurs. L'expression et la synthese de novo du TNF-a et de ses recepteurs (TNFR1 et TNFR2) ont ete evaluees suivant l'exposition de cytotrophoblastes humains provenant de grossesses normales a H2O2 in vitro. Nos resultats demontrent que l'FfcCh induit simultanement l'activation du TNF-a et du TNFR2. Etonnamment, le TNF-a soluble a ete davantage detecte dans le milieu de culture apres neutralisation du TNFR2. Ceci est du a la liaison du sTNF-a aux formes solubles et membranaires du TNFR2. Nos resultats impliquent done que les cytotrophoblastes repondent au stress oxydatif en augmentant non seulement le TNF-a, mais aussi le TNFR2, qui peut diminuer les exces de TNF-a et reduire ses effets autocrines et paracrines. Ce processus peut representer une nouvelle cible de prevention des maladies associees au stress inflammatoire comme la PE. De plus, nos resultats semblent aussi fournir une explication sur l'inconstance de detection complete du TNF-a soluble dans les milieux biologiques. 41 Submitted April 7tn 2009 2.1.4 Article 1 Free Radical Research Effect of H2O2 on placental regulation of TNF-ct system: considerations for oxidative stress in preeclampsia SAMUEL LEBLANC 12 , JEAN-MARIE MOUTQUIN 1 , REGEN DROUIN 3 , YVES GIGUERE4, JEAN-CLAUDE FOREST4, & AZIZ ARIS 1 2 1 Departement d'Obstetrique-Gynecologie, Service Laboratory of Research in Reproductive and Gestational Health. de Genetique, Centre Hospitalier Sherbrooke. Unite de Perinatalogie et Centre de Recherche de I 'Hopital Saint-Francois d'Assise, CHUQ, Universite Laval. Quebec, Canada Universitaire de Abstract Our previous studies have established the role of hydrogen peroxide (H2O2) as a reliable biomarker of oxidative stress in preeclampsia (PE). PE is characterized by an elevation of oxidative and inflammatory mediators in maternal circulation and placenta. However, regulatory mechanisms involved in these processes are not fully understood. The aim of this study was to investigate the effects of H2O2 on placental expression of TNF-a and its receptors. The expression and the de novo synthesis of TNF-a and its receptors (TNFR1 and TNFR2) were assessed following the exposure of human cytotrophoblasts to H2O2. H202 increases simultaneously the production of TNF-a and TNFR2. In conclusion, cytotrophoblasts respond to oxidative stress by enhancing synthesis of not only TNF-a, but also TNFR2 which may clear the excessive levels of TNF-a and reduce its autocrine and paracrine effects. Keywords: Oxidative stress, hydrogen peroxide (H2O2), TNF-a and its receptors, placenta, cytotrophoblasts 43 Introduction The trophoblast is the major component of human placenta and it is involved in all stages of pregnancy, from the blastocyst implantation stage until delivery. Human trophoblast is formed by two components which are functionally distinct: 1) the villous trophoblast, bathing in maternal blood of intervillous spaces and involved in materno-fetal exchanges and in placental endocrine functions; 2) the extra-villous trophoblast involved in uterine spiral arteries remodelling and in the placental anchorage into the uterine wall. Human villous trophoblasts are composed of two layers of cells: the inner mononuclear cytotrophoblast Cytotrophoblasts cells are and the highly outer multinucleated proliferative which syncytiotrophoblast exert invasive [1]. actions. Syncytiotrophoblasts are highly differentiated, which display many functions necessary for the maintenance of pregnancy. Defective trophoblast development and/or function lead to shallow placentation, and is a common characteristic involved in the pathophysiology of preeclampsia (PE) [2-4]. Oxidative stress and inflammation have been shown associated with shallow placentation and subsequent PE. Aetiology of these associations is still unclear, but it is now postulated to result from a combination of immunologic, environmental and genetic factors that leads to the failure of normal trophoblastic invasion and remodelling of the uterine spiral arteries [2-4]. This disorder causes underperfusion, ischemia and hypoxia in placenta [5, 6], which is then thought to release in maternal circulation a variety of mediators including proinflammatory cytokines such as tumor necrosis-alpha (TNF-a), interleukins (IL-1 and IL-6), interferon gamma (INF-y) and reactive oxygen species (ROS) such as nitric oxide (NO), superoxide anion (02~) and hydrogen peroxide (H2O2) 44 [7, 8]. These mediators are thought to cause endothelial dysfunction and permanent systemic vasoconstriction leading to preeclampsia [9-11]. Placental trophoblasts, similarly to all mammalian cells, possess an ever-expanding list of cellular signalling molecules [12], including immune-mediated extracellular ligands and receptors, such as TNF-a and its receptors [13]. In the human placenta, TNF-cc is produced in villous of the placenta throughout pregnancy and has been implicated in the regulation of placental apoptosis [14]. TNF-a exerts its biological effects via two cell surface receptors, namely TNFR1 (p55, 55kD) and TNFR2 (p75, 75kD). These receptors are encoded by two independent genes, and each receptor is capable of mediating distinct intracellular signals [15, 16]. TNF-a is strongly associated with T helper 1 and its expression has been shown to increase in maternal serum of preeclamptic pregnancies [17, 18]. Actually, there is conflicting evidence whether high levels of TNF-a shown in maternal circulation are of maternal or placental origin. Up to now, it is thought that increased expression of TNF-a in the placental compartment could lead to the activation of the circulating leukocytes, linking the oxidative stress of the placenta with a widespread increased oxidative stress and endothelial dysfunction in the mother [19]. Hung et al.[20] showed that in vitro hypoxia/reoxygenation leads to increased expression of TNF-a in placental cells, suggesting that the organ is the main source of circulating cytokine in preeclampsia. Parallel, we recently showed an increase of levels of H2O2 in both maternal circulation and placenta of women with preeclampsia [8], suggesting a close link between H2O2 and TNF-a. Considering the above, the objective of this study was to investigate the effects of H2O2 on the regulation of placental synthesis of TNF-a and its receptors. Materials and methods Human subjects and collection of placental samples Placentas were obtained from women with normal pregnancy and delivery, and neonates were of appropriate size for gestational age. All gave their written and informed consent to take part in the study, which was approved by the Ethics Committee of the CHUS (Centre Hospitalier Universitaire de Sherbrooke). Immediately after delivery of the placenta, samples were collected, and the total length of processing time was less than 15 min. As the placenta is a large heterogeneous organ and sampling bias is a potential hazard [21], three biopsies (2 cm2) of the placenta were taken (centre, mid and edge of placenta) avoiding areas of infarcts and thrombosis. Each sample was briefly rinsed 3 times in cold saline, blotted dry and flash-frozen in individual bags in liquid nitrogen. Cytotrophoblasts isolation and culture Human cytotrophoblasts were purified from five placentas of uncomplicated term pregnancies immediately after delivery, with the method previously described by Kliman et al.[22], with minor modifications. Approximately 35 to 40 g of placental villous tissue from the maternal side was digested three times for 30 min with 0.15% trypsin (Sigma, St. Louis, MO, USA) and 0.02% deoxyribonuclease I (Roche Diagnostics, Laval, QC, Canada) in Hank s' balanced salt solution (HBSS; Invitrogen Canada Inc, Burlington, ON, Canada) containing 25 raM HEPES (Sigma) in a water bath at 37°C. For the last two digestions, tissue fragments were allowed to settle and the supernatant was filtered through a 200 uM pore size nylon mesh and no more than 45 ml of cell suspension was layered over 5 ml aliquots of fetal calf serum (Sigma) and centrifuged at 1000xg for 46 lOmin at room temperature. The pelleted cells were then resuspended in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma) at room temperature. All of the resultant cell suspensions were pooled, centrifuged at lOOOxg, and resuspended in 5 ml of DMEM. The cells were layered over a discontinuous Percoll (Amersham Biosciences, Baie d'Urfe, QC, Canada) density gradient (5-70% in 5% steps of 3 ml each in HBSS) and centrifuged at 400 x g at room temperature for 20 min. The band of cells between 50 and 40% was collected, washed once, and cultured in DMEM containing 10% fetal bovine serum (FBS; Wisent Inc, St-Bruno, QC, Canada), 1% PSN (penicillinstreptomycin-neomycin) antibiotic mixture (Invitrogen, Canada Inc.), 2 mM glutamine (Wisent) and 44 mM sodium bicarbonate (Fisher Scientific Ltd, Nepean, ON, Canada), pH 7.2, in a humidified 5% C02/95% air incubator at 37°C in a poly-1-lysine (Sigma) treated six wells plate (Becton Dickinson Labware, Franklin Lakes, NJ, USA) at 1 x 10 cells per well with 2 ml of DMEM. For all experiments, the purity of isolated cytotrophoblasts was 98% as assessed by cytokeratin (positive) and vementin (negative) staining according to Guilbert et al.[23]. Exposure of cytotrophoblasts to H2O2 Cells were allowed to attach overnight and then washed with DMEM. To study the effects of oxidative stress in these cells, they were thereafter exposed for 48 h to different concentrations of H202: 0, 25, 50 and 100 uM of H2O2 (Fisher Scientific) in complete medium, according to our previous study [24]. Following incubation, media were collected and preserved at - 20°C until determination of TNF-a and its receptors. Cytotrophoblasts were then detached with trypsin 0.125% 2 min, inhibited by complete medium and preserved in lysis buffer for mRNA extraction (Stratagene, Cedar Creek, TX, USA). To establish the specificity of H2O2 effect on TNFR2 synthesis, cells were pretreated with neutralizing monoclonal antibody against TNFR2 (5 ug/ml, R&D system, Minneapolis, MN, USA) for 1 h at 37 °C, and then H2O2 was added and incubated for 48 h. Cell viability was determined by a Trypan Blue staining. Assessment of soluble TNF-a and its receptors Soluble concentrations of TNF-a, TNFR1 and TNFR2 were determined in the supernatants using an Enzyme Linked Immunosorbent Assay (ELISA) method, based on commercial kits and according to instructions of manufacturers (BD Biosciences, Mississauga, ON. Canada). All samples were assayed in duplicates and read with the spectrophotometer at 450 nm. Assessment of TNF-a mRNA RNA extraction Forty milligrams of tissue from -80°C stored biopsies of placenta were used for RNA extractions. Cultured cytotrophoblasts isolated from placenta were stored in lysis buffer at - 80°C until extractions. Total RNA was extracted by using the Absolutely RNA Miniprep Kit (Stratagene, La Jolla, CA, USA), following the standard manufacturer's protocol including DNase step. Total RNA was eluted from the columns in 50 ul of elution buffer 2 times. Total RNA concentration was determined by absorbance at 260 nm (Ultrospec 2100pro UV/Visible Spectrophotometer, Biochrom, England). Protein contamination was monitored by A260/A280 ratio. 48 Real-time RT-PCR First strand cDNA was synthesized using the iScriptTM cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA, USA) according to manufacturer's instructions. The maximum amount of total RNA was run in 20 ul reactions. The reactions were performed in a TGradient Thermoblock (Biometra, Germany). Human TNF-a and P-actin primers were purchased from R&D Systems, Inc. (Minneapolis, MN, USA). For TNF-a, the forward and reverse sequences were GTGACAAGCCTGTAGCCCA and ACTCGGCAAAGTCGAGATAG, respectively. Using these primers, the RT-PCR fragment was 414 bp, whereas amplification of genomic DNA would have yielded a 714 bp fragment. For P-actin, the forward and reverse sequences were CTACAATGAGCTGCGTGTGG and AAGGAAGGCTGGAAGAGTGC, respectively. Using these primers, the RT-PCR fragment was 528 bp, whereas amplification of genomic DNA would have yielded a 969 bp fragment. The iTaqTM SYBR Green Supermix with ROX (Bio-Rad Laboratories, Hercules, CA, USA) was used for the PCR reactions according to manufacturer's instructions. PCR master mixes were prepared containing either the TNF-a or p-actin primer pairs at 225 nM (kept on ice). Uracil-N-glycosylase (Roche, Penzberg, Germany) was added to the master mixes at 0.02 U/u.1 to prevent carry-over. Two microlitres of the RT reaction was added to 48 ul of the PCR master mix. The mixtures were gently mixed, quickly centrifuged and separated in 2 0.1 ml tubes (Corbett Research, Sydney, Australia) for 25 ul duplicate. For each assay, a no template control and 2 positive controls, one for each primer pair, were done. The PCR reaction was performed in a Rotor Gene RG-3000 (Corbett Research, Sydney, Australia). The PCR program was setup as follow: a first hold at 50°C for 2 min, followed by a second hold at 95°C for 2.5 min. Each cycle consisted of 30 sec at 95°C, 45 seconds at 55°C and 45 seconds at 72°C and was repeated 60 times. Then a standard melt step was performed to verify the specificity of the reactions. The analysis of the data was done according to the mathematical model for relative quantification described by M. W. Pfaffl [25]. For each primer, the efficiency (E) of the reactions was determined with realtime PCR of serial dilutions of a reverse transcripted sample and the equation: E=101/slope. The values of A Ct (A CP) determined by the Rotor Gene 6 software were used to calculate the relative expression ratio of a sample (exposed cytotrophoblasts) versus a control (nonexposed cytotrophoblasts). Immunofluorescence and cytometric analysis Expression of membranous receptors of TNF-a (mTNFRl or mTNFR2) in cytotrophoblasts was determined by indirect immunofluorescence and cytometry. Cover slips were placed into a 6-well plate. A total of 1x10 cells were added into each well and treated with different concentration of H2O2 as described above for 48 hours. Cytotrophoblasts were fixed with 4% paraformaldehyde for 30 min, washed 3 times with PBS, and permeabilized with triton X-100 ( 1 % in PBS) for 20 min. After a PBS rinse, cytotrophoblasts were incubated with a monoclonal mouse anti-human TNFR1 or TNFR2 antibody ( R & D System, Minneapolis, MN, USA) in 1% BSA/PBS for 2 hours. After washing with PBS, the cells were incubated with secondary antibody goat anti-mouse Alexa Fluor 594 for 1 hour. Nuclei were identified by 4',6'-diamidino-2-phenyllindole staining for 15 min at room temperature. PBS was used to replace first antibody and served as a negative control. After being washed 2 times with PBS, the cytotrophoblasts were mounted on slides and observed under an Olympus microscope (Olympus BX60 microscope) and photomicrographs were taken with corporate computer. Moreover, immuno- and Pi-stained cells were scanned with iCys imaging cytometer (Compucyte, Cambridge, MA) using Argon ion excitation laser (488 nm) with green (530nm/30nm, immunostaining detection of TNFR1 or TNFR2), long red (650nm/LP, DNA staining detection of cytotrophoblasts) and scatter detectors. Scanning was performed at 0.5 urn x 0.25 urn pixel size resolution. Sample focusing was adjusted with scatter detector. Event selection was performed using a virtual channel obtained by adding green and long red signals to insure detection of fused cells. Cell types were identified based on DNA content one genomic copy for cytotrophoblasts. An experimented scorer selected the scoring thresholds for immunostaining intensity. All cytotrophoblasts selections were confirmed by visualizing a gallery of at least 25 representative cells. Statistical analysis A non-parametric statistical test such as the KrusKal Wallis test was used. All analyses were carried out using the Statistical Analysis System (SAS Institute, Inc., Cary, NC). A P value less than 0.05 was considered statistically significant. 51 Results To test the effects of hydrogen peroxide on TNF-a mRNA expression in freshly isolated cytotrophoblasts, these cells were treated with increasing concentrations of H2O2 (0, 25, 50 and 100 uM) for 48-h. Viability was estimated at > 95% by Trypan Blue exclusion in unexposed cytotrophoblasts. This viability was decreased at 75% and 50% after the exposure of cytotrophoblasts to 50 and 100 uM of H2O2, respectively. As shown in Figure 1, our first observations were the dose-dependent increase of TNF-a mRNA expression, which was statistically significant with concentrations of H2O2 > 50 |iM. Moreover, released TNF-a protein was also increased following treatment of cytotrophoblasts with 50 uM H2O2 (see Figure 2). On the other hand, levels of soluble TNFR2 (sTNFR2) were increased after exposure cytotrophoblasts to 50 uM H2O2, whereas levels of soluble TNFR1 (sTNFRl) did not differ between treated and untreated cytotrophoblasts (see Figure 2). Using an immunofluorescence analysis, our results showed that membranous expression of TNFR2 (mTNFR2), but not membranous TNFR1 (mTNFRl), was increased after the exposure of cytotrophoblasts to 50 uM H2O2 (see Figure 3). Interestingly, the amount of detected soluble TNF-a (sTNF-a) has exceeded twice after pretreatment with neutralizing anti-TNFR2 antibody (see Figure 4). 52 0-" 1 0 1 1 1 1— 5 25 50 H202 exposure 100 Figure 1. Production of TNF-a mRNA by cytotrophoblasts exposed to H2O2. Cytotrophoblasts were exposed to increased concentrations of H2O2 (0, 25, 50 and 100 uM) for 48h. TNF-a mRNA was assessed as relative expression ratio. Central bars represent median values; boxes represent interquartile ranges; and whiskers represent the 90th and 10th percentiles. P < .005 and .001 for exposure to 50 and 100 uM H 2 0 2 , respectively, compared to unexposed controls (H2O2 = 0). P-values were determined by a non-parametric statistical test (Kruskal Wallis test). Five placentas were studied. 53 I 8. P<.001 Q. EZI Unexposed • H202 (50 fjM/L) C « 4H P<.001 © 2- go- CZI en s> / <? «> «> #* ^ Figure 2. Effects of H202 on released soluble TNF-a and its receptors by cytotrophoblasts. Cells were exposed to 50 uM H2O2 for 48h. Then soluble proteins of TNF-a, TNFR1 and TNFR2 were assessed in pg/mL. White and black bars represent unexposed and exposed cells to 50 uM H2O2, respectively. A control column for each of the protein was used. P < .001, for both TNF-a and TNFR2, but not TNFR1. Five placentas were studied. 54 IA • ; $ & & $ - >_ .-, * r. , •" ,** H i 3B P«£,WH U FigiDir© 3. Effects of H2O2 on membranous receptors of TNF-a expression. Cytotropfaoblasts exposed to 50 pM H202 and examined for expression of membranous receptors of TNF-a. Figure 3 showed immunofluorescence of mTNFR2 (3 A), including negative control in the absence of primary antibody (A), unexposed cells (B) and exposed cells (C). A graphical illustration of mTNFR2 immunostainiog (3B) was also showed. Data are presented as mean ± SEM. P < .001 for niTNFR2 after exposure to 50 |iM H2O2, compared to unexposed controls. Five placentas were studied. 55 P<.0001 E I 6P < .001 4_0) A O CO 201 r <> # Figure 4. Effects of anti-TNFR2 Ab on cytotrophoblasts exposed to H2O2. Neutralizing monoclonal antibody against TNFR2 (5 ng/ml) was pretreated to cells for 1 h at 37°C, and then 50 uM of H 2 0 2 was added for 48h. soluble TNF-a (sTNF-a) was assessed in pg/mL. Central bars represent median values; boxes represent interquartile ranges; and whiskers represent the 90th and 10l percentiles. Detected levels of sTNF-a were double after TNFR2 neutralizing, P < .001, compared to cytotrophoblasts exposed to only H2O2 and unexposed controls. Five placentas were studied. 56 Discussion Our results showed that cytotrophoblasts exposed to H2O2 express high levels of TNF-a at levels of protein and mRNA, confirming the direct activation of TNF-a by H2O2 and providing a convincing model of the induction of inflammation by oxidative stress, a phenomenon now called inflammatory stress. TNF-a and H2O2 share common signalling pathways that involve the activation of N F - K B (nuclear factor-KB), even if this is done according to different mechanisms [26]. Indeed, the phosphorylation of p65 at serine 529 has been shown to be required for the TNF-induced transcriptional activity of N F - K B [27], whereas H2O2 induces N F - K B activation, not through serine phosphorylation or degradation of IicBa, but through Syk-mediated tyrosine phosphorylation of IicBa [26]. On the other hand, as the activation of N F - K B can lead to the activation of TNF-a [28, 29], it is conceivable that H2O2 can induce TNF-a by activating N F - K B . Additionally, there are reports showing that H2O2 activates N F - K B [30-32]. The second evidence from our study is the common pattern in the response of TNF-a and TNFR2. High synthesis of membranous TNFR2 (mTNFR2) and soluble TNFR2 (sTNFR2) was the response of cytotrophoblasts to increased levels of H202 and TNF-a in surrounding environment. Both mTNFR2 and sTNFR2 could strongly bind to sTNF-a, preventing its diffusion. This becomes evident after neutralizing TNFR2 by an antibody treatment, resulting in more detectable sTNF-a. The consequences of this double binding of TNF-a would be to limit its effects on two targets: 1) on the same cells (autocrine effect) when TNF-a is bound to mTNFR2; and 2) on the surrounding cells (paracrine effect) when TNF-a is bound to sTNFR2. This is consistent with our previous study on the expression of TNF-a and its receptors in situ in placental villi, showing that expression of TNF-a and TNFR2 are inter-adaptative and follow the same pattern, whereas expression of TNFR1 is constitutive and independent from TNF-a [35]. On the other hand, our results showed that TNFR1 was not affected by H2O2. H2O2exposed cytotrophoblasts express similar levels of TNFR1 compared to unexposed cytotrophoblasts. This would suggest that TNFR1 is not influenced by the variation of significant oxidative stress surrounding cytotrophoblasts. Furthermore, we previously showed that the expression of TNFR1 was constitutive and present in all cells of normal placental villi, independently of TNF-a expression. Thus, these results suggest distinctive roles of TNF-a receptors against oxidative stress. Outside these cells, TNFR1 is expressed rather constitutively on a broad spectrum of different cell types and has been shown to mediate most of the commonly known biological effects of TNF-a [36, 37]. In contrast, expression of the TNFR2 seems to be regulated and there are only a few cellular responses which can be attributed exclusively to signalling via the TNFR2 (i.e. proliferation of thymocytes and mature T cells, NK and B cells, and GM-CSF secretion of T lymphocytes) [38, 39]. Recently a function for TNFR2 has also been proposed in expansion and activation of regulatory T cells [40]. Moreover, TNFR2 has been shown to be preferentially activated by membrane-bound TNF-a [41]. Excessive production of ROS and inflammatory factors may occur at certain windows in placental development and in pathologic pregnancies, such as those complicated by preeclampsia and/or IUGR, overpowering antioxidant defenses with deleterious outcome. In the first trimester, establishment of blood flow into the intervillous space is associated with a burst of oxidative stress. The inability to mount an effective antioxidant defense against this results in early pregnancy loss [42]. In late gestation increased oxidative stress is seen in pregnancies complicated by PE in association with increased trophoblast apoptosis and deportation and altered placental vascular reactivity [42]. Thus, the relevance of our findings is due to the fact that they mimic in vitro an oxidative environment that we usually find in placenta-mediated diseases such as preeclampsia and IUGR [7]. In conclusion, our findings emphasize that cytotrophoblasts respond to oxidative stress by enhancing synthesis of TNF-a, which appear to play a relevant role in both oxidative stress and preeclampsia, and secondly by enhancing TNFR2 which may attenuate the excessive levels of TNF-a by reducing its autocrine and paracrine effects. These findings also seem to provide an explanation of why we often fail to detect soluble TNF-a in biological media. Also, our findings appear to pave the way for considering the use of recombinant sTNFR2 in neutralizing the inflammatory effects of oxidative stress. Additional studies must be undertaken to explore the potential protective effect of TNFR2 in animal and human placentas. Eventually, this could provide novel anti-inflammatory prevention against several conditions and diseases which are characterized by high placental inflammatory stress such as gestational hypertension-preeclampsia. 59 Acknowledgments This study was supported by funding provided by the Fondation des etoiles (FE), the Fonds de Recherche en Sante du Quebec (FRSQ), the Canadian Instituts of Health Research (CIHR) and the Centre de Recherche Clinique (CRC)-Etienne-Lebel du CHUS. The authors wish to thank Danielle Beaulieu for recruitment of patients, samples collection and providing clinical data and Eric Bouchard for microscopy analysis. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. 60 References [1] Rejniak KA, Kliman HJ, Fauci LJ. A computational model of the mechanics of growth of the villous trophoblast bilayer. Bull Math Biol 2004; 66: 199-232. [2] Roberts JM, Cooper DW. Pathogenesis and genetics of pre-eclampsia. Lancet 2001; 357: 53-56. [3] Roberts JM, Lain KY. Recent Insights into the pathogenesis of pre-eclampsia. Placenta 2002; 23: 359-372. [4] Kharfi A, Giguere Y, Sapin V, Masse J, Dastugue B, Forest JC. Trophoblastic remodeling in normal and preeclamptic pregnancies: implication of cytokines. Clin Biochem 2003; 36: 323-331. [5] Lyall F, Myatt L. The role of the placenta in pre-eclampsia~a workshop report. Placenta 2002; 23 Suppl A: S142-145. [6] Granger JP, Alexander BT, Bennett WA, Khalil RA. Pathophysiology of pregnancyinduced hypertension. Am J Hypertens 2001; 14: 178S-185S. [7] Kharfi A, Giguere Y, De Grandpre P, Moutquin JM, Forest JC. Human chorionic gonadotropin (hCG) may be a marker of systemic oxidative stress in normotensive and preeclamptic term pregnancies. Clin Biochem 2005; 38: 717-721. [8] Aris A, Benali S, Ouellet A, Moutquin JM, Leblanc S. Potential Biomarkers of Preeclampsia: Inverse Correlation between Hydrogen Peroxide and Nitric Oxide Early in Maternal Circulation and at Term in Placenta of Women with Preeclampsia. Placenta 2009; 30: 342-347. [9] Roberts JM, Hubel CA. Is oxidative stress the link in the two-stage model of preeclampsia? Lancet 1999; 354: 788-789. 61 [10] Hubel CA. Oxidative stress in the pathogenesis of preeclampsia. Proc Soc Exp Biol Med 1999;222:222-235. [11] Khalil RA, Granger JP. Vascular mechanisms of increased arterial pressure in preeclampsia: lessons from animal models. Am J Physiol Regul Integr Comp Physiol 2002; 283: R29-45. [12] Heazell AE, Crocker IP. Live and let die - regulation of villous trophoblast apoptosis in normal and abnormal pregnancies. Placenta 2008; 29: 772-783. [13] Allaire AD, Ballenger KA, Wells SR, McMahon MJ, Lessey BA. Placental apoptosis in preeclampsia. Obstet Gynecol 2000; 96: 271-276. [14] Levy R, Nelson DM. To be, or not to be, that is the question. Apoptosis in human trophoblast. Placenta 2000; 21: 1-13. [15] Chen HL, Yang YP, Hu XL, Yelavarthi KK, Fishback JL, Hunt JS. Tumor necrosis factor alpha mRNA and protein are present in human placental and uterine cells at early and late stages of gestation. Am J Pathol 1991; 139: 327-335. [16] Shu HB, Takeuchi M, Goeddel DV. The tumor necrosis factor receptor 2 signal transducers TRAF2 and c-IAPl are components of the tumor necrosis factor receptor 1 signaling complex. Proc Natl Acad Sci U S A 1996; 93: 13973-13978. [17] Visser W, Beckmann I, Bremer HA, Lim HL, Wallenburg HC. Bioactive tumour necrosis factor alpha in pre-eclamptic patients with and without the HELLP syndrome. Br J Obstet Gynaecol 1994; 101: 1081-1082. [18] Vince GS, Starkey PM, Austgulen R, Kwiatkowski D, Redman CW. Interleukin-6, tumour necrosis factor and soluble tumour necrosis factor receptors in women with preeclampsia. Br J Obstet Gynaecol 1995; 102: 20-25. 62 [19] Walsh SW. Maternal-placental interactions of oxidative stress and antioxidants in preeclampsia. Semin Reprod Endocrinol 1998; 16: 93-104. [20] Hung TH, Charnock-Jones DS, Skepper JN, Burton GJ. Secretion of tumor necrosis factor-alpha from human placental tissues induced by hypoxia-reoxygenation causes endothelial cell activation in vitro: a potential mediator of the inflammatory response in preeclampsia. Am J Pathol 2004; 164: 1049-1061. [21] Mayhew TM, Burton GJ. Methodological problems in placental morphometry: apologia for the use of stereology based on sound sampling practice. Placenta 1988; 9: 565-581. [22] Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss JF, 3rd. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 1986; 118: 1567-1582. [23] Guilbert LJ, Winkler-Lowen B. Placental alkaline phosphatase (PLAP) staining and human chorionic gonadotropin (hCG) production in cultures of fresh and cryopreserved cytotrophob lasts isolated by CD9/MHC class I/MHC class II immunoelimination. Placenta 2007; 28: 348-349. [24] Kharfi Aris A, Leblanc S, Ouellet A, Moutquin JM. Dual action of H202 on placental hCG secretion: implications for oxidative stress in preeclampsia. Clin Biochem 2007; 40: 94-97. [25] Pfaffl MW. A new mathematical model for relative quantification in real-time RTPCR. Nucleic Acids Res 2001; 29: e45. [26] Takada Y, Mukhopadhyay A, Kundu GC, Mahabeleshwar GH, Singh S, Aggarwal BB. Hydrogen peroxide activates NF-kappa B through tyrosine phosphorylation of I 63 kappa B alpha and serine phosphorylation of p65: evidence for the involvement of I kappa B alpha kinase and Syk protein-tyrosine kinase. J Biol Chem 2003; 278: 24233-24241. [27] Wang D, Baldwin AS, Jr. Activation of nuclear factor-kappaB-dependent transcription by tumor necrosis factor-alpha is mediated through phosphorylation of RelA/p65 on serine 529. J Biol Chem 1998; 273: 29411-29416. [28] Shakhov AN, Collart MA, Vassalli P, Nedospasov SA, Jongeneel CV. Kappa B-type enhancers are involved in lipopolysaccharide-mediated transcriptional activation of the tumor necrosis factor alpha gene in primary macrophages. J Exp Med 1990; 171: 35-47. [29] Collart MA, Baeuerle P, Vassalli P. Regulation of tumor necrosis factor alpha transcription in macrophages: involvement of four kappa B-like motifs and of constitutive and inducible forms of NF-kappa B. Mol Cell Biol 1990; 10: 1498-1506. [30] Kretz-Remy C, Mehlen P, Mirault ME, Arrigo AP. Inhibition of I kappa B-alpha phosphorylation and degradation and subsequent NF-kappa B activation by glutathione peroxidase overexpression. J Cell Biol 1996; 133: 1083-1093. [31] Schreck R, Rieber P, Baeuerle PA. Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-kappa B transcription factor and HIVl.EmboJ 1991; 10:2247-2258. [32] Meyer M, Schreck R, Baeuerle PA. H202 and antioxidants have opposite effects on activation of NF-kappa B and AP-1 in intact cells: AP-1 as secondary antioxidantresponsive factor. Embo J 1993; 12: 2005-2015. [33] Kharfi A, Bureau M, Giguere Y, Moutquin JM, Forest JC. Dissociation between increased apoptosis and expression of the tumor necrosis factor-alpha system in term placental villi with preeclampsia. Clin Biochem 2006; 39: 646-651. 64 [34] Vercammen D, Vandenabeele P, Declercq W, Van de Craen M, Grooten J, Fiers W. Cytotoxicity in L929 murine fibrosarcoma cells after triggering of transfected human p75 tumour necrosis factor (TNF) receptor is mediated by endogenous murine TNF. Cytokine 1995;7:463-470. [35] Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell Death Differ 2003; 10:45-65. [36] Vandenabeele P, Declercq W, Vercammen D, Van de Craen M, Grooten J, Loetscher H, Brockhaus M, Lesslauer W, Fiers W. Functional characterization of the human tumor necrosis factor receptor p75 in a transfected rat/mouse T cell hybridoma. J Exp Med 1992; 176:1015-1024. [37] Grell M, Becke FM, Wajant H, Mannel DN, Scheurich P. TNF receptor type 2 mediates thymocyte proliferation independently of TNF receptor type 1. Eur J Immunol 1998; 28: 257-263. [38] Scumpia PO, Delano MJ, Kelly KM, O'Malley KA, Efron PA, McAuliffe PF, Brusko T, Ungaro R, Barker T, Wynn JL, Atkinson MA, Reeves WH, Salzler MJ, Moldawer LL. Increased natural CD4+CD25+ regulatory T cells and their suppressor activity do not contribute to mortality in murine polymicrobial sepsis. J Immunol 2006; 177: 7943-7949. [39] Seitz C, Mannel DN, Hehlgans T. Isolation and functional characterization of the mouse p75 TNF receptor promoter. Genomics 1998; 48: 111-116. [40] Myatt L, Cui X. Oxidative stress in the placenta. Histochem Cell Biol 2004; 122: 369-382. 65 2.2 Article 2: Dual action of H2O2 on placental hCG secretion : Implications for oxidative stress in preeclampsia 2.2.1Auteurs : Aziz Kharfi Aris, Samuel Leblanc, Annie Ouellet, Jean-Marie Moutquin 2.2.2 Role joue par le candidat dans cette etude : Design experimental, realisation des experiences, compilation des donnees, analyse statistique des donnees. Article publie dans CLINICAL BIOCHEMISTRY, 2007 JAN; 40(l-2):94-7. 66 2.2.3 Resume Nos resultats prealablement publies ont demontre que les niveaux circulants d't^Ch etaient augmentes et correles avec de hauts niveaux de 1'hCG chez les femmes souffrant de preeclampsie, suggerant que le stress oxydatif pourrait moduler la synthese de cette hormone placentaire. L'objectif de cette etude etait d'etudier in vitro les effets de TH202 sur la secretion placentaire de 1'hCG. Des cytotrophoblastes in vitro on ete traites avec des concentrations croissantes d'l^Ch et la secretion de 1'hCG de novo a ete mesuree. Le traitement avec de faibles concentrations d't^C^ (1-50 uM) augmentent la secretion de 1'hCG par les cytotrophoblastes, alors que des concentrations superieures a 50 uM reduisent la secretion de 1'hCG d'une maniere dose dependante. Nos resultats soulignent done que : (1) PH2O2 pourrait avoir une double action sur l'activite placentaire et agir non seulement comme un agent cytotoxique, mais aussi comme une molecule de signalisation pouvant induire la secretion de 1'hCG ; (2) 1'hCG pourrait etre un agent antioxydant protecteur relache par le placenta pour contrer un stress oxydatif leger. 67 2.2.4 Article 2 Published in Clinical Biochemistry, 2007 Jan; 40(l-2):94-7 Dual action of H2O2 on placental hCG secretion: implications for oxidative stress in preeclampsia Short Title: effects of oxidative stress on trophoblastic hCG secretion Aziz Kharfi Aris ab *, Samuel Leblanc, Annie Ouelleta, Jean-Marie Moutquina,b a Departement d'Obstetrique-Gynecologie, CHUS, FMSS, Universitaire de Sherbrooke. b Centre de Recherche Clinique Etienne Lebel, Axe Mere-Enfant, Faculte de Medecine, Universite de Sherbrooke, Quebec, Canada. Abstract Objective: Our previous published findings showed that circulating levels of H2O2 were increased and correlated with high levels of hCG in women with preeclampsia, suggesting that oxidative stress modulates placental hormone synthesis. The aim of this study was to investigate in vitro the effects of H2O2 on placental secretion of hCG. Design and Methods: in vitro trophoblasts were stimulated with increasing concentrations of t^Chand the de novo hCG secretion was assayed. Results: Stimulation with low concentrations of H2O2 (1-50 uM) enhance cytotrophoblastic hCG secretion, whereas concentrations of H2O2 > 50 uM reduce hCG secretion in a dose-dependent manner. Conclusions: our findings emphasize that: 1) H2O2 may have dual action on placental activity and acts not only as a cytotoxic mediator, but also as a signaling molecule able to induce hCG secretion; 2) hCG may be a protective antioxidant released by the placenta to counter low oxidative stress challenge. Key-words: hCG, H2O2, placenta, trophoblasts, oxidative stress, preeclampsia 69 Introduction Preeclampsia is a vasoconstrictive disorder, characterized by blood pressure > 140/90 mmHg in a previously normotensive women after the 20th week of pregnancy in the presence of significant proteinuria (> 300 mg/24 h urine collection)[l, 2]. The etiology of preeclampsia is still open to debate; it is now believed to result from a combination of immunologic, environmental and genetic factors that leads to the failure of normal trophoblastic invasion and remodeling of the uterine spiral arteries [3, 4]. This disorder causes underperfusion, ischemia and hypoxia in placenta [5, 6], which is then thought to release in maternal circulation a variety of mediators including hormones such as human chorionic gonadotropin (hCG), cytokines such as tumor necrosis-alpha (TNF-a), interleukins (IL-1 and IL-6), interferon gamma (INF-y), and reactive oxygen species (ROS) such as nitric oxide (NO), superoxide anion (O2-) and hydrogen peroxide (H2O2). These mediators thought to cause endothelial dysfunction and permanent systemic vasoconstriction leading to preeclampsia [7, 8]. Since trophoblastic abnormalities play a central role in the pathophysiology of preeclampsia and precede its clinical manifestations [9], it is conceivable that some placental hormone profiles such as an increased secretion of human chorionic gonadotropin (hCG) are modified in the maternal circulation, pointing out to the defect of placental function [9] and [10]. Increased production of hCG by preeclamptic placentas has been shown to be associated with strong hCG immunostaining of the syncytiotrophoblast [11]. Recently, we showed a correlation between the increased secretion of hCG and oxidative stress as measured by circulating hydrogen peroxide 70 (H2O2) in vivo in preeclampsia, suggesting that increased level of H2O2 in pregnant women can interfere with placental hormone synthesis [12]. H2O2 is a member of the ROS family, a product of produced by many enzymecatalyzed redox reactions and its determination is of fundamental importance in biological fields, particularly in the assessment of the oxidative state. Under normal conditions, the cells use their antioxidant defenses, which convert H2O2 to oxygen and water, thereby keeping the production of the ROS system under control. Although H2O2 is more known for its toxicological effects such as its ability to harm microorganisms and to damage cells, its association with a high placental secretion of hCG in vivo in preeclampsia allows us to think that H2O2 can act as a signaling molecule. Some believe that hCG secretion may be increased as a consequence of abnormal placental invasion, placental immaturity [13] or placental hypoxia with the development of a hypersecretory state [10]. However, the exact mechanisms interacting between H2O2 and placental hCG secretion are not clear. Their understanding would help in best preventing and treating pathologies implying oxidative stress and placenta such as preeclampsia. 71 Material and Methods Human subjects All placentas were obtained from women with normal pregnancy and delivery, and all neonates were of appropriate size for gestational age. All gave their written and informed consent to take part in the study, which was approved by the Ethic Committee of the CHUS. Collection of placental samples Immediately after delivery of the placenta, samples were collected, and the total length of processing time was less than 15 min. As the placenta is a large heterogeneous organ, and sampling bias is a potential hazard [14], a square grid of 16 holes in a 4 x 4 design within a circular template was constructed to overlay the placenta. Templates of various sizes were prepared to best fit placentas of different diameters. After blotting the maternal surface of blood, a sterile nickel cork borer was inserted into each of the 16 holes (~1 cm) to obtain a full thickness biopsy (~0.5 g tissue, which excluded the chorionic plate). Each sample was briefly rinsed 3 times in cold saline, blotted dry, and flash-frozen in individual bags in liquid nitrogen. Cytotrophoblast isolation and culture Human cytotrophoblasts were purified from five placentas of uncomplicated term pregnancies immediately after delivery, with the method previously described by Kliman et al.[15], with minor modifications. 35 to 40 grams of placental villous tissue from the maternal side were digested 3 times for 30 minutes with 0.15 % trypsin (Sigma, St Louis, MO, USA) and 0.02 % desoxyribonuclease I (Roche Diagnostics, 72 Laval, Qc, Can) in Hanks' balanced salt solution (HBSS) (Invitrogen Canada Inc, Burlington, ON, Can) containing 25 mM HEPES (Sigma) in a water bath at 37°C. For the two last digestions, tissue fragments were allowed to settle and the supernatant was filtered through a 200 uM pore size nylon mesh and no more than 45 ml of cell suspension was layered over 5 ml aliquots of fetal calf serum (Sigma) and centrifuged at 1000 x g for 10 min at room temperature. The pelleted cells were then resuspended in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma) at room temperature. All of the resultant cell suspensions were pooled, centrifuged at 1000 x g, and resuspended in 5 ml of DMEM. The cells were layered over a discontinuous Percoll (Amersham Bioscinces, Baie d'Urfe, QC, Can) density gradient (5-70% in 5% steps of 3 ml each in HBSS) and centrifuged at 400 g at room temperature for 20 minutes. The band of cells between 50 an 40 % were collected, washed once, and cultured in DMEM containing 10 per cent fetal bovine serum (FBS) (Wisent Inc, St-Bruno, QC, Can), 1 per cent PSN (Invitrogen Canada Inc.), 2 mM glutamine (Wisent) and 44 mM sodium bicarbonate (Fisher Scientific Ltd, Nepean, ON, Can) (pH 7.2) in a humidified 5% C0 2 /95% air incubator at 37°C in a poly-L-lysine (Sigma) treated six well plate (Becton Dickinson Labware, Franklin Lakes, NJ, USA) at 1 x 106 cells per well with 2 mL of DMEM. hCG-cytotrophoblasts stimulation Cells were allowed to attach overnight and then washed with DMEM. To study the effects of oxidative stress in these cells, they were thereafter exposed for 48 h to (0, 1, 5, 10, 50, 60, 70, 200 and 500 uM H 2 0 2 (Fisher Scientific) in complete medium. Following incubation, the media were collected and preserved at -20°C until 73 determination of hCG. Cytotrophoblasts were then detach with trypsin 0.125 % 2 minutes, inhibited by complete medium and preserved in lysis buffer for mRNA extraction (Stratagene, Cedar Creek, TX, USA). Determination of hCG production hCG was measured using a standard automated immunoassay which has a detection limit of 0.1 IU/L (Elecsys HCG+P; Roche Diagnostics Corporation, Indianapolis, IN). Statistical analysis Data are presented as the mean concentration and mean percent change from baseline, as determined from control cultures in unconditioned media, ± SD. Statistical analysis was performed by analysis of variance (ANOVA) and post hoc analysis (Dunnet test) for comparison to control. All analyses were carried out using the Statistical Analysis System (SAS Institue, Inc., Cary, NC). A P value less than 0.05 was considered statistically significant. 74 Results To test the effects of hydrogen peroxide on hCG secretion by trophoblast cells in vitro, these cells were stimulated with increasing concentrations of H2O2 (0, 1, 5, 10, 50, 60, 70, 200 and 500 uM). Using an hCG standard automated immunoassay, the hCG content of spent medium from control (unstimulated cells) and stimulated trophoblasts was measured following 48-h hydrogen peroxide exposure. Our first observations were the high individual variation of hCG secretion and the second observations were the existence of two distinctive responses of hCG secretion depending of the concentrations of H2O2. Thus, cultures stimulated with concentrations under 50 uM showed high secretion of hCG relative to unstimulated controls, reaching a maximal secretion with 5 uM H2O2 (Fig.lA). Ten, cultures stimulated with concentrations of H2O2 above 50 uM (i.e. 60, 70, 200 and 500 uM) exhibited a marked dose-dependent decrease in hCG secretion relative to their unstimulated controls (Fig. IB). We recapitulated these tendencies in Fig. 2, showing the essential points of these two windows of hydrogen peroxide effects on human trophoblast hCG secretion. 75 120 .0 e-110 *5 • 4i—- » u 8 100 w «4— O o o 90H 80- 1 5 10 50 i i H 2 0 2 (uM) 120-. § 3 90 g 8 60 CO O O M- o 30H i 50 60 70 200 500 H 2 0 2 (uM) Fig. 1. Representative illustration of dual action of H2O2 on hCG production by trophoblast cells in vitro. Concentrations of H2O2 from 1 to 50 uM produce more secretion of hCG than unstimulated controls (0 uM), a maximal secretion was reached with 5 uM H2O2 (A). In (B), cultures stimulated with concentrations of H2O2 > 50 uM exhibited a marked dose-dependent decrease in hCG secretion. 76 150-. P < 0.01 c ^r 125- .2 2 0 c 100- b o 75- o o 50- ° 250 p < 0.001 J Control 5 uM 50 uM 500 uM Peroxide concentration (uM) Fig. 2. Effects of hydrogen peroxide on the hCG production of trophoblast cells in vitro (n =5). For 48 h, the trophoblasts were stimulated by increasing concentrations of H2O2 (5, 50 and 500 uM), unstimulated controls (0 uM H2O2). Secretion of hCG was assessed by a standard automated immunoassay. Note the statistically significant increase of hCG secretion following exposure to 5 uM H2O2. Because hCG secretion of individual preparations varied, controls were arbitrarily set at 100 % in each experiment. *P < 0.01 or **P < 0.001 compared to unstimulated controls. P values were determinate by ANOVA and post hoc analysis (Dunnet test). 77 Discussion It is well established that oxidative stress is involved in human placental pathologies such as preeclampsia and fetal growth restriction (FGR) [16]. Our previous study showed that women with preeclampsia exhibit higher serum hCG and H2O2 than normotensive pregnant women at term [12]. Thus, the main objective of the present study was to examine in vitro the hCG production/secretion as response of trophoblastic cells to an oxidative stress, in occurrence hydrogen peroxide. As already highlight in our in vivo findings showing a positive correlation between increased circulating hCG and H2O2 in preeclampsia, our present in vitro study on placental function have revealed interactions between trophoblastic cells and oxidative stress generated by H2O2. Both in vivo and in vitro observations suggest that H2O2 triggers the placental hormone synthesis. Our results showed that the basal secretion of hCG was higher after 48 h culture than 24h. This is may be explained by the high differentiation of cytotrophoblasts to syncytiotrophoblasts, placental cells producing pregnancy hormones (i.e. hCG). Incubation of trophoblasts with increasing concentrations of H 2 0 2 (0, 1, 5, 10, 50, 60, 70, 200 and 500 uM) for 24 h as well as 48 h result in a dual action of hydrogen peroxide. Low concentrations of H2O2 (1-50 uM) enhance secretion of hCG, with a maximal production around 5 uM and moderate with about 50 uM and below. Theses findings outline that hydrogen peroxide may act not only as cytotoxic mediator, but also as a signaling molecule able to active cellular synthesis. Although, the exact mechanisms of activation of H2O2 are not yet clear it is possible that the secretion of hCG is a consequence of the increase in the differentiation of cytotrophoblasts into syncytiotrophoblasts. 78 Many agents producing intracellular oxidative stress, including hydrogen peroxide and steroid hormones, have also been found to induce the production of antioxidant factors [17]. This action may be essential for the survival of terminally differentiated cells subjected to oxidative stress, such as syncytiotrophoblasts, placental cells producing hCG and forming the maternal-fetal barrier. Changes in the localization of antioxidant factor expression were apparent after a low oxidative stress challenge [17]. In general, to counteract stress caused by H2O2, organisms have evolved a wide variety of defence mechanisms. For instance, the reactions leading to H2O2 are minimized by sequestering the metal ions that would otherwise act as catalysts into proteins. Ferritin, transferrin, hemosiderin and heme are examples of proteins that enclose iron and thus play a role in protecting the cell against oxidative damage [18]. Other tools in the combat against oxidative stress are enzymes that are employed to rapidly dismutate H2O2 to water. Superoxide dismutases (SOD), catalases, peroxidases (especially glutathione peroxidases), and thioredoxin-linked systems are examples of such enzymes [19], [20]. Many other molecules serve as antioxidants including vitamins (e.g. vitamin E, vitamin C, provitamin A, also called beta-carotine), hormones (e.g. melatonin) and cofactors such as coenzyme Q [21]. Thus, it is probable that hCG may have an antioxidant role against low oxidative stress allowing the survival of placental cells. Future investigations must aim to better eluciding these aspects. On the other hand, the fall of hCG production after incubation with concentration of H2O2 above 50 JJM could be explained by the toxological effect of hydrogen peroxide on trophoblasts. Just as H2O2 has the ability to harm microorganisms, it also has the ability to destroy our body's cells. H2O2 per se causes only mild damage. The damaging power comes from the transition to highly reactive hydroxyl radicals that 79 indiscriminately react with a wide variety of organic substrates causing peroxidation of lipids, cross-linking and inactivation of proteins, and mutations in DNA [22, 23]. Hydroxyl radicals form when H2O2 is exposed to ultraviolet light or when it comes in contact with a range of transition metal ions, of which the most important is probably iron. It is widely accepted that H2O2 is a toxin in vivo at concentrations of about 50 uM and above. In conclusion, our present in vitro findings confirm our previous in vivo study suggesting that there are interactions between oxidative stress and placental function. Hydrogen peroxide acts not only as a cytotoxic mediator, but also as a signaling molecule able to induce secretion of hCG by trophoblasts. This dual action of H2O2 depends on its concentrations. Its low concentrations (1-50 uM) result in increase of placental hCG secretion, whereas concentrations above 50 uM decrease this secretion. On the other hand, our findings also suggest that hCG may be an antioxidant factor produced by the placenta in order to counter low oxidative stress. Other investigation in this direction will help to better understand these phenomena. 80 Acknowledgments This work was supported by the Fonds de la Recherche en Sante du Quebec (FRSQ) and La Fondation de la Recherche sur les Maladies Infantiles (FRMI). We thank Helene Marceau, Jennifer Saint-Laurent, Christine Brown and Cynthia Rodrigue for technical assistance. 81 References [1] Roberts JM, Lain KY. Recent Insights into the pathogenesis of pre-eclampsia. Placenta 002; 23: 359-372. [2] Sibai BM. Diagnosis and management of gestational hypertension and preeclampsia. Obstet Gynecol 2003; 102: 181-192. [3] Roberts JM, Cooper DW. Pathogenesis and genetics of pre-eclampsia. Lancet 2001; 357: 3-56. [4] Aris Kharfi A, Giguere Y, Sapin V, Masse J, Dastugue B, Forest JC. Trophoblastic remodeling in normal and preeclamptic pregnancies: implication of cytokines. Clin Biochem 2003; 36: 323-331. [5] Lyall F, Myatt L. The role of the placenta in pre-eclampsia~a workshop report. Placenta 2002; 23 Suppl A: SI42-145. [6] Granger JP, Alexander BT, Llinas MT, Bennett WA, Khalil RA. Pathophysiology of reeclampsia: linking placental ischemia/hypoxia with microvascular dysfunction. Microcirculation 2002; 9: 147-160. [7] Hubel CA. Oxidative stress in the pathogenesis of preeclampsia. Proc Soc Exp Biol Med 1999;222:222-235. [8] Khalil RA, Granger JP. Vascular mechanisms of increased arterial pressure in preeclampsia: lessons from animal models. Am J Physiol Regul Integr Comp Physiol 2002; 283: R29-45. [9] Vaitukaitis JL, Ebersole ER. Evidence for altered synthesis of human chorionic gonadotropin in gestational trophoblastic tumors. J Clin Endocrinol Metab 1976; 42: 1048-1055. 82 [10] Hsu CD, Chan DW, Iriye B, Johnson TR, Hong SF, Repke JT. Elevated serum human chorionic gonadotropin as evidence of secretory response in severe preeclampsia. Am J Obstet Gynecol 1994; 170: 1135-1138. [ll]Barros JS, Baptista MG, Bairos VA. Human chorionic gonadotropin in human placentas from normal and preeclamptic pregnancies. Arch Gynecol Obstet 2002; 266:67-71. [12] Aris Kharfi A, Giguere Y, De grandpre P, Moutquin JM, Forest JC. Human chorionic gonadotropin (hCG) may be a marker of systemiv oxidative stress in nomotensive and preeclamptic. Clin Biochem 2005 Aug;38(8):717-21. [13] Lieppman RE, Williams MA, Cheng EY, Resta R, Zingheim R, Hickok DE, Luthy DA. An association between elevated levels of human chorionic gonadotropin in the midtrimester and adverse pregnancy outcome. Am J Obstet Gynecol 1993; 168: 18521856; discussion 1856-1857. [14] Mayhew TM, Burton GJ. Methodological problems in placental morphometry: apologia for the use of stereology based on sound sampling practice. Placenta 1988; 9: 565-581. [15]Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss JF, 3rd. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 1986; 118: 1567-1582. [16] Myatt L, Cui X. Oxidative stress in the placenta. Histochem Cell Biol 2004; 122: 369-382. [17] McAleer MF, Tuan RS. Metallothionein protects against severe oxidative stressinduced apoptosis of human trophoblastic cells. In Vitr Mol Toxicol 2001; 14: 219-231. 83 [18]Nappi AJ, Vass E. Iron, metalloenzymes and cytotoxic reactions. Cell Mol Biol (Noisy-le-grand) 2000; 46: 637-647. [19] Droge W. Free radicals in the physiological control of cell function. Physiol Rev 2002; 82: 47-95. [20] Arner ES, Holmgren A. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 2000; 267: 6102-6109. [21] Reiter RJ, Melchiorri D, Sewerynek E, Poeggeler B, Barlow-Walden L, Chuang J, Ortiz GG, Acuna-Castroviejo D. A review of the evidence supporting melatonin's role as an antioxidant. J Pineal Res 1995; 18: 1-11. [22] Liu D, Wen J, Liu J, Li L. The roles of free radicals in amyotrophic lateral sclerosis: reactive oxygen species and elevated oxidation of protein, DNA, and membrane phospholipids. Faseb J 1999; 13: 2318-2328. [23] Abe T, Konishi T, Hirano T, Kasai H, Shimizu K, Kashimura M, Higashi K. Possible correlation between DNA damage induced by hydrogen peroxide and translocation of heat shock 70 protein into the nucleus. Biochem Biophys Res Commun 1995;206:548-555. 84 2.3 Article 3: Potential biomarkers of preeclampsia : inverse correlation between hydrogen peroxide and nitric oxide early in maternal circulation and at term in placenta of women with preeclampsia 2.3.1 Auteurs : Aziz Aris, Sebastien Benali, Annie Ouellet, Jean-Marie Moutquin, Samuel Leblanc 2.3.2R61e joue par le candidat dans cette etude : Design experimental, realisation des experiences, compilation des donnees, analyse statistique des donnees. Article publie dans PLACENTA, 2009 APR;30(4):342-7. 85 2.3.3 Resume La preeclampsie (PE) est un desordre specifique a la grossesse qui a ete associe a des maladies cardiovasculaires ulterieures, aussi bien chez la mere que chez l'enfant. L'etiologie de la PE est inconnue, mais le stress oxydatif semble y jouer un role preponderant dans la dysfonction endothelial et la vasoconstriction systemique permanente qui la caracterise. Le peroxyde d'hydrogene (H2O2), un metabolite terminal dans la cascade du stress oxydatif cellulaire, est aussi un composant du stress oxydatif lors de Pischemie/reperfusion du placenta. Nous avons ete les premiers a demontrer Pelevation des niveaux d'EL^ dans le serum de femmes preeclamptiques a terme. L'H202 est deja connu pour son role dans la reduction de la production d'oxyde nitrique (NO) en augmentant le metabolisme des arginases. L'objectif de cette etude etait d'etudier une possible correlation entre le NO, un vasodilatateur puissant, et PH2O2 au cours de la grossesse normale et preeclamptique. Nous avons done simultanement dose les niveaux du NO et de P H2O2 dans le serum de femmes normales et preeclamptiques de 10 a 15 et de 37 a 40 semaines de grossesse, et dans leurs placentas a terme. Nos resultats revelent une correlation inverse entre les niveaux de PH2O2 et du NO, aussi bien au niveau sanguin qu'au niveau placentaire. Cette correlation est confirmee par nos experiences in vitro, demontrant que PH2O2 inhibe la synthese de NO par les cytotrophoblastes. Pour conclure, la correlation inverse entre PH2O2 et le NO observee en debut de grossesse dans la circulation maternelle et dans le placenta des femmes preeclamptiques ouvre la voie pour 86 de futures etudes sur l'usage potentiel du NO et de PH2O2 comme biomarqueurs de prediction de la preeclampsie. 87 2.3.4 Article 3 Published in Placenta Apr;30(4):342-7 Potential Biomarkers of Preeclampsia: Inverse Correlation Between Hydrogen Peroxide and Nitric Oxide Early in Maternal Circulation and at Term in Placenta of Women with Preeclampsia A. Arisa'b'*, S. Benalib, A. Ouelleta, J.M. Moutquin3, S. Leblancb department of Obstetrics-Gynecology, bLaboratory of Research in Reproductive and Gestational Health. Short Title: Potential use of hydrogen peroxide and nitric oxide as predictive biomarkers of preeclampsia Abstract Preeclampsia (PE) is a pregnancy-specific disease that has been associated with future cardiovascular disease for the mother and her child. The etiology of PE is unclear but oxidative stress seems to play a major role in endothelial dysfunction and permanent systemic vasoconstriction shown in PE. Hydrogen peroxide (H2O2), a terminal metabolite of the cellular oxidative stress cascade, is also revealed as a component of oxidative ischemia/reperfusion stress in placenta. We were the first to show an increase in the levels of H2O2 in the serum of preeclamptic women at term. H2O2 is already known to reduce the production of NO by increasing the metabolism of arginases. The objective of this study was to investigate a possible correlation between nitric oxide (NO), a potent vasodilator, and H2O2 throughout pregnancy. Thus, we simultaneously assessed the levels of NO and H2O2 in the serum of normal and preeclamptic women at 10-15 and 37-40 weeks of pregnancy, and in placentas at delivery. Our findings showed an inverse correlation between increased levels of H202 and decreased levels of NO early in maternal circulation and at term in placenta. This relationship is confirmed by our in vitro experiments which demonstrate that H2O2 inhibits NO synthesis of cytotrophoblasts. In conclusion, our findings highlight an inverse correlation between H2O2 and NO early in maternal circulation and placenta of women with preeclampsia, paving the way for further studies examining the potential use of NO and H2O2 as biomarkers in the prediction of preeclampsia. Keywords: preeclampsia; placenta; oxidative stress; hydrogen peroxide (H202); nitric oxide (NO); biomarkers 89 1. Introduction Preeclampsia is a vasoconstrictive disorder, characterized by blood pressure = 140/90 mm Hg in previously normotensive women after the 20th week of pregnancy in the presence of significant proteinuria (>300 mg/24 h urine collection) [1] and [2]. The etiology of preeclampsia is still open to debate; it is now believed to result from a combination of immunologic, environmental and genetic factors that leads to the failure of normal trophoblastic invasion and remodeling of the uterine spiral arteries [3] and [4]. This disorder causes underperfusion, ischemia and hypoxia in the placenta [5] and [6], which is then thought to release in maternal circulation a variety of mediators including hormones such as human chorionic gonadotropin (hCG), cytokines such as tumor necrosis-alpha (TNF-a), interleukins (IL-1 and IL-6), interferon gamma (INF-y) and reactive oxygen species (ROS) such as nitric oxide (NO), superoxide anion (O2 ) and hydrogen peroxide (H2O2). These mediators are thought to cause endothelial dysfunction and permanent systemic vasoconstriction leading to preeclampsia [7] and [8]. H2O2 is a member of the ROS family and a metabolite produced by many enzyme-catalyzed redox reactions. Its determination is of fundamental importance in many biological fields, particularly in the assessment of the oxidative state. In a previous study, we demonstrated that its circulating levels are increased in preeclampsia [9] as well as its correlation with human chorionic gonadotropin (hCG) [9], suggesting a direct effect of oxidative stress on placental function. This hypothesis was confirmed by our in vitro study 90 showing that H2O2 modulates directly the endocrine activity of placenta [10]. Thus, the vasoconstriction seen in preeclampsia may be the result of the combination of increased actions of many vasoconstrictor factors [11] including oxidants such as H2O2 [12] and of the reduction in levels of vasodilators such as nitric oxide (NO) [13]. NO production involves NO synthase (NOS) which, in the presence of sufficient cofactor tetrahydrobiopterin (BH4), receives electron from NAD(P)H to breakdown the cosubstrates 1-arginine and molecular oxygen into NO and 1citruline. There are three isoforms of NO synthase, neuronal NOS (nNOS), inducible NOS (iNOS) and endothelial NOS (eNOS), which are expressed in neurons, white cells, endothelium and placenta On the other hand, arginase plays a key role in NO metabolism by catabolising 1-arginine to urea and 1ornithine. There are two isoforms of arginase both of which reduce NO synthesis: arginase I, a cytosolic enzyme, and arginase II, a mitochondrial enzyme [15] and [16]. These two arginases have been shown to be expressed in cytotrophoblasts in the first trimester and at term in the human placenta [17]. NO has many key physiological effects such as maintaining systemic vascular health, mediating inflammatory response and promoting antithrombotic properties of the endothelium. The implication of NO in preeclampsia is widely referred despite the controversy surrounding its kinetics and the mystery around its regulation. Indeed, some studies found increased levels of NO in maternal circulation [18] and [19] and placenta [20] while 91 others found decreased levels in the maternal serum [21] and placenta [22] or even normal levels [23]. It is possible that H2O2 interacts with NO, providing two interdependent mediators. Indeed, H2O2 has been shown to reduce the production of NO by increasing arginases metabolism [15]. Thus, the purpose of this study is to investigate a possible correlation between H2O2 and NO production in maternal circulation and placenta throughout pregnancy with and without preeclampsia. 92 2. Materials and methods 2.1. Participants and blood sampling The data for this study were derived from a prospective assessment of risk for preeclampsia by a combination of maternal serum biomarkers. Among a cohort of healthy nulliparous pregnant women recruited at Centre Hospitalier Universitaire de Sherbrooke (CHUS), we examined twenty women with preeclampsia (preeclamptic group) which were matched for age, gestational age, and parity with twenty normotensive pregnant women (control group) (Table 1). Preeclampsia was diagnosed when the systolic blood pressure was greater than 140 mm Hg and/or diastolic blood pressure greater than or equal to 90 mm Hg and proteinuria greater than 0.3 g/24 h on a 24-h urine collection. Eligible participants had 25.3 ± 5.1 for age and were not known for cigarette or illicit drug use nor suffering from any medical conditions (i.e. diabetes, hypertension or metabolic disease). All of the examined patients had Cesarean delivery and did not have any adverse perinatal outcomes. The common indications for Cesarean section among the control group included labor dystocia, nonreassuring fetal heart testing, and malpresentation such as breech. All subjects in our study reported taking standard prenatal multivitamins with anti-oxidants such as vitamin C and E during pregnancy (typically 100 mg of vitamin C and 30 IU of vitamin E). Blood sampling was serially performed from each participant at 10-15 weeks of gestation and at delivery where routine blood sampling was performed as part of the preoperative assessment before 93 Table 1 Characteristics of the normotensive and preeclamptic nulliparous pregnant subjects and their serum and placental H2O2 and NO concentrations Controls (n = 20) Preeclampsia (n = 20) 3 P value Age - Mean ± SD 26.5 ± 5.8 25.3 ±5.4 NS Gestational age - Mean ± SD 38.5 ± 2.3 38.3 ±1.2 NS Birth weight -MeaniSD NS H 2 0 2 (mean ± SD, uM) - Serum • 10-15 weeks of gestation P<0.01 • 37-40 weeks of gestation P<0.01 - Placenta • Homogenates P<0.01 NO (mean ± SD, uM) - Serum • 10-15 weeks of gestation P<0.01 •37-40 weeks of gestation P < 0.001 - Placenta • Homogenates P < 0.001 3357 ± 345 g 3165 ± 4 3 8 30.45 ±6.31 41.75 ±13.36 33.15 ±8.09 46.20 ±14.63 7.88 ±1.94 10.99 ±3.48 25.81 ± 10.62 16.95 ± 6.23 29.49 ±6.86 19.82 ±6.51 4.54 ± 1.06 3.05 ± 1.00 H202 = hydrogen peroxide, NO = nitric oxide. Data are expressed as mean ± SD. NS = not significant. a P values were determined by Mann-Whitney test 94 Cesarean section. We did not recruit pregnant women before the 10 week of gestation; since they usually consult after this date. The study was approved by the CHUS Ethics Human Research Committee on Clinical Research. All participants gave written consent. Blood samples were centrifuged at 2000 rpm for 15 minutes within 1 h of collection. Serum was stored at -20°C until assayed for H202 and NO levels. 2.2. Placental samples collection Immediately after delivery of the placenta, samples for PE and or a selected following normotensive control were collected. The total length of processing time was less than 15 min. As the placenta is a large heterogeneous organ, and sampling bias is a potential hazard [24], three biopsies (2cm2) of the placenta were taken (centre, mid and edge of placenta) avoiding areas of infarcts and thrombosis. Almost all cases of proteinuric or severe PE deliver early in the day which allowed us to use the fresh specimens we needed for our lab processing. Collected placentas were homogenized according to a previous protocol [25] and assessed for spontaneous production of H202 and NO. For our in vitro study, a part of placental from normal pregnancy was immediately processed to isolate cytotrophoblasts. These cells were obtained by enzymatic dispersion and density gradient centrifugation as described previously by Kliman et al. [26], and placed in culture for 24h. 95 2.3. In vitro exposure of cytotrophoblasts to H2O2 Isolated cytotrophoblasts were allowed to attach overnight and then washed with DMEM, as we previously described [10]. To study the effects of oxidative stress in these cells, they were exposed thereafter for 48 h to 0, 1, 5, 10, 25, 50, 100, 200 and 500 uM H 2 0 2 (Fisher Scientific) in complete medium. The viability of trophoblasts was assessed for all experiments (n=5) and was not affected by a concentration of 500 uM H202. Indeed, trophoblasts have been shown to resist to a concentration of 1000 uM H2O2 [27]. Following incubation, the media were collected and preserved at — 20°C until determination of NO. 2.4. H202 determination Levels of H2O2 from serum and/or supernatant were measured using an electrochemical technique adapted in our laboratory [9]. Briefly, H202 is measured with a three-electrode system which consists of a platinum working electrode, an Ag/AgCI reference electrode, and a Pt auxiliary electrode. The potential at the working electrode is held at 650 mV vs the reference electrode. H202 is oxidized on the microelectrode at this potential and the current generated is recorded with an electrochemical detector. All measurements are performed at ambient temperature (20-25°C). H202 stock solutions (10 mM or 100 uM) are prepared by diluting the H2O2 (30%) in double-distilled water. Phosphate buffer (50 mM, pH 7.4) is made from K2HPO4 (0.84 g/1) and NaH2P04 (7.5 g/l). For the calibration curve, the 96 potential at the working electrode is highly linear from 0 to at least 200 \xM concentration. These levels are about 34 uM (33.57 uM)9 and these measurements are consistent with other studies claiming levels up to ^35 uM in human blood plasma [28-31]. Levels of H202 at or below about 20-50 uM seem to have limited cytotoxicity to many cell types [28]. Among the ROS family members, H202 is the more stable and easy to measure under assay conditions in which its removal is prevented. 2.5. NO determination Because NO is stoichiometrically converted to nitrate and nitrite [32], the stable metabolites of NO, levels of NO were determined by assaying for nitrate and nitrite. The nitrate is first converted to nitrite by enzymatic reduction catalyzed by nitrate reductase, and the nitrite is quantitated by a colorimetric assay using the Griess Reagent [31]. NO production was assessed by measuring levels of total nitrite and nitrate (NOx), the stable metabolites of NO, in deproteinized serum. Nitrate was reduced to nitrite by using the Griess method [33]. 97 3. Statistical analysis Statistical analyses were carried out by the Mann-Whitney test for group differences and by analysis of variance (ANOVA) and post hoc analysis (Dunnett test) for comparison to control. For correlations, the Pearson and Spearman rank correlation test were used. Data are presented as the mean concentration ± SD All analyses were carried out using the Statistical Analysis System (SAS Institute, Inc., Cary, NC). A P value less than 0.05 was considered statistically significant. 98 4. Results As expected, there were no significant differences in age, parity, gestational age and birth weight among the two matched groups (Table 1). Maternal serum concentrations of H2O2 were significantly greater in preeclamptic women in comparison to normotensive pregnant women at 10-15 weeks (41.75 ± 13.36 vs 30.45 ± 6.31, P < 0.01) and delivery (46.20 ± 14.63 vs 33.15 ± 8.09, P < 0.01) (Table 1, Figs. 1A,B). Levels of H2O2 were also higher in preeclamptic placentas than normotensive placentas (10.99 ± 3.48 vs 7.88 ± 1.94, P < 0.01) (Table 1, Fig. 2A). The concentration of serum NO was significantly lower in preeclamptic women compared to the controls at 10-15 gestational weeks (16.95 ± 6.23 vs 25.81 ± 10.62, P < 0.01) and at term (19.82 ± 6.51 vs 29.49 ± 6.86, P < 0.001) (Table 1, Figs. 1C,D). Also, concentrations of NO were low in placental homogenates from preeclamptic group in comparison to normotensive women (3.05 ± 1.00 vs 4.54 ± 1.06, P < 0.001) (Table 1, Fig. 2B). Finally, there was an inverse correlation between NO and H202 in maternal serum and placentas among all studied subjects (n = 40). Spearman correlation coefficients were r = -0.58, P = 0.0001 for serum at 10-15 gestational weeks, r = -0.60, P < 0.0001 for serum at 37-40 gestational weeks and r = -0.59, P < 0.0001 for placental homogenates (Figs. 3A,B and C). For the in vitro study, viability of freshly isolated cytotrophoblasts was estimated at > 95% by Tryptan Blue exclusion. To study the effects of H202 on NO production of trophoblasts cells in vitro, the number of of experiments 99 was 5 (n=5) (Fig. 4). Because NO production in individual preparations varied, controls were arbitrary set at 100% in each experiment. Thus, 50 uM of H202 reduced levels of NO by 30%, P < 0.01 and 100 uM of H202 reduced levels of NO by 40%, P < 0.01. P values were determinate by ANOVA and post hoc analysis (Dunnett test). 100 I A B 100-. 100n 40-1 Control ? 80- f 60- •£ 40- oo 0(P n OO Preeclampsia Preeclampsia Groups Groups C D 100-i 2- 1 3 50-i 80 3-40- 60- 3 30- tn I 40 o Z 20- A A S 20- A * > AA Isle*- iio0- 0- Control Preeclampsia Groups Control Preeclampsia Groups Fig.l. Serum levels of hydrogen peroxide (H202) (A and B) and nitric oxide (NO) (C and D) in pregnant women with (n = 20) or without preeclampsia (n = 20), at 10-15 gestational weeks (A and C) and at delivery (B and C). Maternal serum H202 and NO were measured by an electrochemical method and a colorimetric assay, respectively. The horizontal line refers to median concentration. Note the significant differences in the levels of H202 and NO among the studied groups (P < 0.01 for panel A, P < 0.01 for panel B, P = 0.01 for panel C and P < 0.001 for panel D). 101 A _l i 20-i Z2 *—^ CM O CM P < 0.01 15- X X— O CO 10- 5 0) S c 5- 0 o 0- m 1 Controls Preeclamptic Groups B 8-i 3 o 6P < 0.001 o 4- $c CD 2- o JO Q. 0Controls Preeclamptic Groups Fig.2. Placenta] levels of hydrogen peroxide (H202) (A) and nitric oxide (NO) (B) in pregnant women with (n = 20) or without preeclampsia (n = 20), at delivery (17-40 gestational weeks). Placental homogenates H202 and NO were measured by an electrochemical method and a colorimetric assay, respectively. Note the significant differences in the levels of H202 and NO between controls and preeclamptic subjects (A: P < 0.01 for and B: P < 0.001 for panel D). 102 A 5040g 303 O 20Z 1000 20 40 60 H202 (uM/L) 0 25 50 75 H202 (uM/L) 100 0 6 12 18 H202 (ulWL) 24 B 5040§ 303 O 20- z 100- c 10-1 8- IO 420- Fig.3. Relationship between levels of H202 and NO in serum at 10-15 gestational weeks (A), at term (37-40 weeks) (B) and in placentas at delivery (C) from normotensive and preeclamptic pregnant women. There are significant negative correlations between H202 and NO production among all studied subjects (n = 40) in serum at 10-15 gestational weeks (o) and 37-40 weeks (•) , and in placentas at delivery (*). Sepearman correlation coefficients r = -0.58, P = 0.0001, r = -0.60, P < 0.0001 and r = -0.59, P < 0.0001, respectively. 103 100Z ~ P < 0.001 1 80-| -1- ^ e g 60.2 o | P< 0.001 — —I— \ I o 40H 2 ~ 20H o-" £ <P* *J» X 5$> # ^ H202 exposure Fig.4. Effects of hydrogen peroxide on the production of NO by trophoblast cells (n=5). For 48 h, isolated primary cytotrophoblasts were treated by increasing concentrations of H202 (25, 50 and 100 uM), unstimulated controls (0 uM H202). Production of NO was assessed by a colorimetric assay. Note the significant decrease of NO production after exposure to 50 and 100 uM of H202. Because NO production of individual preparations varied, controls were arbitrarily set at 100% in each experiment. P < 0.001 for 50 uM H202 exposure and P < 0.01 for 100 uM H202 exposure compared to unstimulated controls. P values were determinate by ANOVA and post hoc analysis (Dunnett test). 104 Discussion It is well established that oxidative stress plays an important role in preeclampsia [34]. Thus, our findings highlight an increase of H202 levels in maternal circulation early at 10-15 weeks of gestation. Moreover, we noted that placentas from preeclamptic women exhibit more H202 than normotensive women. This suggests that oxidative stress seen in preeclampsia affects both maternal circulation and placenta. Under normal conditions, cells use their antioxidant defenses, which convert H202 to oxygen and water, thereby keeping the production of the ROS system under control. Increased H202 can result from overproduction of ROS and/or decreased antioxidant capacity. Although the origin of increased levels of H202 in preeclampsia remains unclear, our findings suggest that preeclampsia may be triggered by an oxidative stress state. On the other hand, multiple studies were interested in assessing levels of NO in preeclampsia. However, these studies were contradictory since some have shown increased levels of NO in maternal circulation [18,19] and placenta [20] while others found decreased levels in maternal serum [21] and placenta [22] or even normal levels [23]. The present work is the first study that simultaneously determines levels of circulating NO early and at term of pregnancy and at term in the placenta. Thus, our findings demonstrate that levels of NO are decreased at 10-15 weeks of gestation and at delivery in maternal serum, and at term in placenta. With regards to the process of vasoconstriction which characterizes preeclampsia, the reduction of a 105 powerful vasodilator such as nitric oxide in the first trimester may precede the generalised hypertension found in preeclamptic women which appears around the 20 th week of gestation. Our results also demonstrate a significant inverse correlation between serum H2O2 levels and NO production among all studied groups in serum at both 1015 gestational weeks and at term, and in placentas at delivery. This suggests that in maternal circulation as well as in placenta, the increase in H202 coincides with a decline in the production of NO, which may trigger the vasoconstriction. The majority of pregnant women take moderate vitamin supplements in Canada and this appeared not influence the incidence of preeclampsia. On potential interference of this vitamins supplementation on NO and H202 production, our analyses showed that multivitamins taken at normal concentrations did not affect production of NO, nor those of H202 in both normal and preeclamptic pregnancies. On the other hand, our in vitro experiments showed that H202 induces decreased production of NO by cytotrophoblasts and confirm our hypothesis that H202 is an inhibitor of NO synthesis. Moreover, Noris et al. showed increased levels of arginase II in preeclampsia [35]. Arginases may compete with eNOS for L-Arginine and metabolise the latter into urea and L-ornithine resulting in a decreased production of NO. In the presence of reduced levels of L-arginine, the activation of eNOS leads to overproduction of a large amount of superoxide anion (O ") via the uncoupling of eNOS on L-Arginine 106 [36]. O2" reacts with NO to form peroxynitrite (ONOO) [37], thereby decreasing the availability of NO. The hypothesis that levels of NO are decreased via inactivation of L-arginine is supported by the study of Altun et al [38], which demonstrated that L-Arginine supplementation in preeclamptic rats decreased clinical manifestations of preeclampsia. Thus, it is possible that H202 mediates decreased NO via increasing arginase I and II as their functions may overlap [15]. Indeed, Thengchaisri et al. [39], showed that upregulation of arginase by H202 impairs the NO-mediated endothelium dilatation. It is important to mention that our in vitro studies on isolated cytotrophoblasts-exposed H202 were performed using normal placentas and further studies using preeclamptic placentas are needed to observe a potential difference in responses. On the other hand, levels of both H2O2 and NO tend to increase, but are not statistically significant, from the beginning of pregnancy to its end, in control and preeclamptic groups. This may be explained by the increase of metabolic interactions due to the increase of fetal growth and advancement of pregnancy. Moreover, to our knowledge, there is no study that assessed levels of H2O2 and NO at 10-15 and 37-40 weeks of pregnancy, allowing comparison. In conclusion, our study highlights an increase in H202 levels in maternal serum early at 10-15 weeks, at term and in placenta of women with preeclampsia. The high circulating H202 has been shown to be inversely correlated with a lowered NO production. These findings provide a rationale 107 for considering these mediators as potential biomarkers in predicting preeclampsia. 108 5. Acknowledgments This study was supported by funding provided by the Fonds de Recherche en Sante du Quebec (FRSQ), the Fondation de Recherche sur les Maladies Infantiles (FRMI), the Canadian Institutes of Health Research (CIHR) and the Centre de Recherche Clinique (CRC)-Etienne-Lebel du CHUS. The authors wish to thank Christine Brown and Danielle Beaulieu for recruitment of patients, sample collection and providing clinical data; and Christine Lebel for laboratory and technical assistance. 109 6. References [1] Roberts JM, Lain KY. Recent insights into the pathogenesis of pre-eclampsia. Placenta. 2002;23:359-72. [2] Sibai BM, Diagnosis and management of gestational hypertension and preeclampsia. Obstet Gynecol. 2003;102:181-92. [3] Roberts JM, Cooper DW, Pathogenesis and genetics of pre-eclampsia. Lancet. 2001;357:53-6. [4] Kharfi A, Giguere Y, Sapin V, Masse J, Dastugue B, Forest JC, Trophoblastic remodeling in normal and preeclamptic pregnancies: implication of cytokines. Clin Biochem. 2003;36:323-31. [5] Lyall F, Myatt L. The role of the placenta in pre-eclampsia—A workshop report. Placenta. 2002;23:S 142-5. [6] Granger JP, Alexander BT, Llinas MT, Bennett WA, Khalil RA, Pathophysiology of preeclampsia: linking placental ischemia/hypoxia with microvascular dysfunction. Microcirculation. 2002;9:147-60. [7] Hubel CA, Oxidative stress in the pathogenesis of preeclampsia, Proc Soc Exp Biol Med. 1999;222:222-35. [8] Khalil RA, Granger JP, Vascular mechanisms of increased arterial pressure in preeclampsia: lessons from animal models. Am J Physiol Regul Integr Comp Physiol. 2002;283:R29-45. [9] Kharfi A, Giguere Y, De Grandpre P, Moutquin JM, Forest JC. Human chorionic gonadotropin (hCG) may be a marker of systemic oxidative stress in 110 normotensive and preeclamptic term pregnancies. Clin Biochem. 2005;38(8):717-21. [10] Kharfi Aris A, Leblanc S, Ouellet A, Moutquin JM. Dual action of H2O2 on placental hCG secretion: implications for oxidative stress in preeclampsia. Clin Biochem. 2007; 40: 94-7. [11] Ariza AC, Bobadilla NA, Halhali A. Endothelin 1 and angiotensin II in preeclampsia. Rev Invest Clin. 2007;59(l):48-56. [12] Agarwal A, Gupta S, Sharma RK. Role of oxidative stress in female reproduction. Reprod Biol Endocrinol. 2005;3:28. [13] Tsukimori K, Komatsu H, Fukushima K, Kaku T,Nakano H, Wake N. Inhibition of nitric oxide synthetase at mid-gestation in rats is associated with increases in arterial pressure, serum tumor necrosis factor-alpha, and placental apoptosis. Am J Hypertens. 2008;21(4):477-81. [14] Schiessl B, Mylonas I, Hantschmann P, Kuhn C, Schulze S, Kunze S, Friese K, Jeschke U. Expression of endothelial NO synthase, inducible NO synthase, and estrogen receptors alpha and beta in placental tissue of normal, preeclamptic, and intrauterine growth-restricted pregnancies. J Histochem Cytochem. 2005; 53(12):1441-9. [15] Cederbaum SD, Yu H, Grody WW, Kern RM, Yoo P, Iyer RK. Arginases I and II: do their functions overlap? Mol Genet Metab. 2004;81 :S38-44. [16] Morris SM Jr. Regulation of enzymes of the urea cycle and arginine metabolism. Annu Rev Nutr. 2002;22:87-105. Ill [17] Ishakawa T, Harada T, Koi H, Kubota T, Azuma H, Aso T. Identification of arginase in human placental villi. Placenta. 2007;28(2-3): 133-8. [18] Bhatnagar S, Bhattacharjee J, Vaid M, Madan T, Trivedi SS, Sarma PU. Inducible nitric oxide synthase (iNOS) gene polymorphism in pre-eclampsia: a pilot study in North India. Aust N Z J Obstet Gynaecol. 2007;47(6):477-82. [19] Vural P. Nitric oxide/endothelin-1 in preeclampsia. Clin Chim Acta. 2002;317(l-2):65-70. [20] Shaamash AH, Elsnosy ED, Makhlouf AM, Zakhari MM, Ibrahim OA, El Dien HM. Maternal and fetal serum nitric oxide (NO) concentrations in normal pregnancy, pre-eclampsia and eclampsia. Int J Gynaecol Obstet. 2000;68:20714. [21] Sandrim VC, Palei AC, Metzger IF, Gomes VA, Cavalli RC, Tanus-Santos JE. Nitric oxide formation is inversely related to serum levels of antiangiogenic factors soluble fms-like tyrosine kinase-1 and soluble endogline in preeclampsia. Hypertension. 2008;52(2):402-7. [22] Schonfelder G, Fuhr N, Hadzidiakos D, John M, Hopp H, Paul M. Preeclampsia is associated with loss of neuronal nitric oxide synthase expression in vascular smooth muscle cells of the human umbilical cord. Histopathology. 2004;44:116-28. [23] Orange SJ, Painter D, Horvath J, Yu B, Trent R, Hennessy A. Placental endothelial nitric oxide synthase localization and expression in normal human pregnancy and pre-eclampsia. Clin Exp Pharmacol Physiol. 2003;30(5-6):37681. 112 [24] Mayhew TM, Burton GJ. Methodological problems in placental morphometry: apologia for the use of stereology based on sound sampling practice. Placenta. 1988;9:565-81. [25] Huang X, Huang H, Dong M, Yao Q, Wang H. Serum and placental interleukin-18 are elevated in preeclampsia. J Reprod Immunol. 2005;65(1):7787. [26] Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss JF, 3rd. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology. 1986;118:1567-82. [27] Hallmann A, Klimek J, Masaoka M, Kaminski M, Kedzior J, Majczak A, Niemczyk E, Wozniak M, Trzonkowski P, Wakabayashi T. Partial characterization of human choriocarcinoma cell line JAR cells in regard to oxidative stress. Acta Biochim Pol. 2004;51:1023-38. [28] Halliwell B, Clement MV, Long LH. Hydrogen peroxide in the human body. FEBS Lett. 2000;486(1):10-13. [29] Varma SD, Devamanoharan PS. Hydrogen peroxide in human blood. Free Radic Res Commun. 1991;14:125-31. [30] Lacy F, O'Connor DT, Schmid-Schonbein GW. Plasma hydrogen peroxide production in hypertensives and normotensive subjects at genetic risk of hypertension. J Hypertens. 1998;16:291-303. [31] Deskur E, Przywarska I, Dylewicz P, Szczesniak L, Rychlewski T, Wilk M, Wysocki H. Exercise-induced increase in hydrogen peroxide plasma levels is 113 diminished by endurance training after myocardial infarction. Int J Cardiol. 1998;67:219-24. [32] Baylis C, Vallance P. Measurement of nitrite and nitrate levels in plasma and urine—what does this measure tell us about the activity of the endogenous nitric oxide system? Curr Opin Nephrol Hypertens. 1998;7:59-62. [33] Kleinbongard P, Rassaf T, Dejam A, Kerber S, Kelm M. Griess method for nitrite measurement of aqueous and protein-containing samples. Methods Enzymol. 2002;359:158-68. [34] Myatt L, Cui X. Oxidative stress in the placenta. Histochem cell biol 2004;122:369-382. [35] Noris M, Todeschini M, Cassis P, Pasta F, Cappellini A, Bonazzola S, Macconi D, Maucci R, Porrati F, Benigni A, Picciolo C, Remuzzi G.L-Arginine Depletion in Preeclampsia Orients Nitric Oxide Synthase Toward Oxidant Species. Hypertension. 2004;43(3):614-22. [36] Culcasi, M, Lafon-Cazal M, Pietri, S, Bockaert J. Glutamate receptors induce a burst of superoxide via activation of nitric oxide synthase in arginine-depleted neurons. J Biol Chem. 1994; 269(17):12589-93. [37] Beckman JS, Koppenol WH(1996). Nitric oxide, superoxide, and peroxynitrite : The good, the bad, and ugly. Am J Physiol. 1996; 271(5, Part 1):C 1424-37. [38] Altun ZS, Uysal S, Guner G, Yilmaz O, Posaci C. Effects of oral L-arginine supplementation on blood pressure and asymmetric dimethylarginine in stressinduced preeclamptic rats. Cell Biochem Funct. 2008;26(5):648-53. 114 [39] Thengchaisri N, Hein TW, Wang W, Xu X, Li Z, Fossum TW, Kuo L. Upregulation of arginase by H2O2 impairs endothelium-dependent nitric oxidemediated dilation of coronary arterioles. Arterioscler Thromb Vase Biol. 2006;26(9):2035-42. 3 DISCUSSION II est d'ores et deja etabli que le stress oxydatif est implique dans les pathologies de grossesse humaines comme la preeclampsie (MYATT et CUI, 2004). Comme nous avions constate dans des etudes precedentes une concentration serique plus elevee d't^C^ et d'hCG chez des femmes atteintes de preeclampsie, par rapport a des femmes normotensives (KHARFI et al, 2005), nous voulions verifier ses effets sur le placenta, plus precisement sur les cytotrophoblastes in vitro. D'abord, comme nous l'avions souligne dans nos resultats in vivo, une correlation positive de l'augmentation des niveaux circulants d't^C^ et d'hCG presente chez les femmes preeclamptiques, nos resultats in vitro sur la fonction placentaire revelent des interactions entre les cellules trophoblastiques et le stress oxydatif represente par PH2O2. Nos observations in vivo et in vitro suggerent que PH2O2 entraine la synthese d'hCG. En effet, l'exposition a des concentrations croissantes d't^C^ durant 24 et 48 heures resultent en une double action du peroxyde d'hydrogene. De plus faibles concentrations (1-50 uM) entrainent une augmentation de la secretion d'hCG, avec une production maximale a 5 uM, et plus moderee jusqu'a 50 uM. Ceci suggere done que le peroxyde d'hydrogene n'agit pas seulement comme un mediateur cytotoxique, mais aussi comme une molecule de signalisation pouvant activer la production d'autres molecules. Comme les mecanismes d'activation par PH2O2 sont encore mal connus, il est possible que la secretion de 1'hCG soit la consequence de l'augmentation de la differentiation des cytotrophoblastes en syncytiotrophoblastes, ces derniers constituent la veritable glande endocrine du placenta. 116 Plusieurs agents produisant un stress oxydatif intracellulaire, incluant le peroxyde d'hydrogene et les hormones stero'idiennes, induisent la production de facteurs antioxydants (MCALEER et TUAN, 2001). Cette action peut etre essentielle aussi bien aux cytotrophoblastes qu'aux syncytiotrophoblastes exposes au stress oxydatif. Des changements dans la localisation de facteurs antioxydants sont apparents, et ce, meme apres une exposition a un leger stress oxydatif (MCALEER et TUAN, 2001). Generalement, pour contrer ce stress entraine par PH2O2, de nombreux mecanismes de defense existent. Par exemple, les reactions menant a la formation d'l-bCh sont attenuees par la sequestration par des proteines d'ions metalliques, dont la ferritine, la transferrine et Theme, qui pourraient agir comme catalyseurs (NAPPI et VASS.,2000). D'autres facteurs de defense comme les superoxyde dismutases, les catalases, peroxydases et la thioredoxine sont des enzymes qui permettent de combattre le stress oxydatif en reagissant avec PH2O2 pour le transformer en eau (DROGE, 2002, ARNER et HOLMREN, 2000). D'autres molecules comme les vitamines C, E et A, des hormones comme la melatonine et des cofacteurs comme le coenzyme Q peuvent aussi servir d'antioxydants (REITER et a/., 1995). II est done probable que 1'hCG puisse avoir un certain role antioxydant pour contrer un stress oxydatif permettant la survie des cellules placentaires. D'un autre cote, la chute de la secretion de 1'hCG apres un traitement de plus de 50 uM d'H202 peut etre expliquee par les effets cytotoxiques du peroxyde d'hydrogene sur les cytotrophoblastes. Comme PH2O2 a la capacite de detruire des microorganismes, il est aussi habilite a endommager les cytotrophoblastes. Les 117 dommages causes par PH2O2 sont toutefois moderes, en comparaison avec des especes plus actives comme le radical hydroxyle qui peut reagir avec une grande variete de substrats organiques, causant la peroxydation des lipides, Pinactivation de proteines et des dommages a l'ADN (LIU et al, 1999, ABE et al., 1995). Le radical hydroxyle peut etre forme quand PH2O2 est expose aux rayons ultraviolets ou lorsqu'il entre en contact avec certains ions de metaux comme le fer. II est aussi largement accepte que PH2O2 agit comme molecules cytotoxique in vivo a une concentration superieure a 50 ^M. Nos resultats aussi ont demontre que les cytotrophoblastes exposes a FH2O2 expriment plus de TNF-a, tant au niveau de la proteine que de l'ARNm, confirmant l'activation directe du TNF-a par PH2O2 et fournissant un modele de Pinduction de Pinflammation par le stress oxydatif, appele stress inflammatoire. Le TNF-a et PH2O2 partagent une voie de signalisation qui implique l'activation de N F - K B (facteur nucleaire KB), et ce, meme si cela se produit par des mecanismes differents (TAKADA et al, 2003). II a ete demontre que la phosphorylation de la p65 a la serine 529 est requise pour induire Pactivite transcriptionnelle du TNF-a (WANG et BALDWIN, 1998), alors que PH 2 0 2 induit l'activation de N F - K B par la phosphorylation de la tyrosine via Syk de bcBa (TAKADA et al, 2003). II est done possible que, comme l'activation de N F - K B peut mener a la transcription du TNF-a ( SHAKHOV et al, 1990, COLLART et al, 1990), PH 2 0 2 puisse induire la transcription de TNF-a en activant N F - K B . Cette possibility est renforcee par d'autres etudes demontrant que PH 2 0 2 active N F - K B (KRETZ-REMY et al, 1996, SCHRECK et al, 1991, MEYER et al, 1993). 118 Cette etude a aussi permis d'entrevoir une reduction de la detection du TNF-a soluble dans le surnageant quand une forte secretion du TNFR2 se produit. Comme nous l'avons constate, une forte synthese de TNFR2 membranaire et de sa forme soluble augmenterait la liaison du TNF-a a ce recepteur et entraverait sa detection complete. Ceci a done ete confirme par la neutralisation du recepteur par un anticorps, rendant le TNF-a davantage detectable. En consequence, 1'augmentation de l'expression de ce recepteur pourrait limiter les effets du TNF-a, d'abord sur la cellule elle-meme (effet autocrine), quand cette cytokine se lie au recepteur membranaire, puis en liant le recepteur soluble, limitant ses effets sur les cellules environnantes (effet paracrine). L'F^Ch n'induit done pas seulement l'activation du TNF-a, mais aussi celle de son recepteur TNFR2. Cela concorde avec nos travaux anterieurs portant sur l'expression du TNF-a et de ses recepteurs in situ dans les cytotrophoblastes, demontrant que l'expression du TNFR2 parait etre adaptative avec celle du TNF-a et suit le meme profil (KHARFI et ah, 2006). Les cytotrophoblastes exposes a FH2O2 produisent done simultanement du TNF-a et le TNFR2 et relachent leur forme soluble dans le surnageant, et ce dernier peut Her le TNF-a, prevenant sa detection. D'un autre cote, nos resultats ont demontre que le TNFR1 ne parait pas etre influence par la variation des concentrations d'FfcC^, les cytotrophoblastes n'exprimant pas plus ou moins de recepteur selon diverses concentrations. Ceci suggererait que le TNFR1 n'est pas influence par la variation des metabolites oxydants environnants. De plus, nous avions precedemment demontre que l'expression du TNFR1 etait constitutive et presente dans toutes les cellules 119 placentaires, independamment de l'expression du TNF-a (KHARFI et al, 2006). Inversement au TNFR2, comme le TNFR1 apparait etre plus stable contre le stress oxydatif, cela suggere qu'ils ont un role different chez les cytotrophoblastes. En dehors de ces cellules, le TNFR1 est en effet exprime de maniere plutot constitutive sur un large eventail de cellules et est connu pour medier les effets les plus connus du TNF-a. (VERCAMMEN et al, 1995, WAJANT et al, 2003). L'expression du TNFR2 semble etre regulee et quelques reponses cellulaires peuvent etre attributes exclusivement par la signalisation via ce recepteur, comme la proliferation des thymocytes et des cellules T matures, des cellules NK et B, ainsi que la secretion de GM-CSF par les lymphocytes T (VANDENABEELE et al, 1992, GRELL et al, 1998). Recemment, il a ete demontre que le TNFR2 est aussi implique dans l'expansion et l'activation des cellules T regulatrices (SCUMPIA et al, 2006). De nombreuses etudes ont aussi portees sur les niveaux du NO dans la preeclampsie. Toutefois, ces etudes sont souvent contradictoires puisque quelques unes ont constate une augmentation du NO dans la circulation maternelle (BHATNAGAR et al, 2007, VURAL, 2002), et dans le placenta (SHAAMASH et al., 2000), alors que d'autres ont constate une diminution dans le serum maternel (SANDRIM et al, 2008) et le placenta (SCHONFELDER et al, 2004), ou encore des niveaux normaux (ORANGE et al, 2003). Nous avons done ete les premiers a determiner simultanement les niveaux circulants de NO en debut et en fin de grossesse et placentaires en fin de grossesse. Nous avons done demontre que les niveaux de NO sont diminues entre 10 et 15 semaines de grossesse dans le serum, a terme dans le serum, et a terme dans le placenta des femmes preeclamptiques. Tout 120 en tenant compte de l'etat de vasoconstriction qui caracterise la preeclampsie, la reduction d'un vasodilatateur important comme le NO dans le premier trimestre peut etre un precedent dans l'hypertension generalised presente chez les femmes preeclamptiques diagnostiquees apres la 20 e semaine de grossesse. Nos resultats demontrent aussi une correlation inverse significative entre les niveaux seriques d ' L L ^ et de NO, et ce, dans toutes les conditions etudiees. Ceci suggere done que dans la circulation maternelle, aussi bien que dans le placenta, l'augmentation de PH2O2 coincide avec le declin de la production de NO, qui pourrait entrainer la vasoconstriction. La majorite des femmes enceintes prennent moderement des supplements de vitamines au Canada, et ceci ne semble pas avoir d'influence sur 1'incidence de preeclampsie. L'interference de cette supplementation sur le NO et sur PH2O2, selon nos analyses, ne semble pas etre importante et n'affecte pas la production de NO, pas plus que celle de TH202, tant chez les femmes normales que preeclamptiques (ARIS et al, 2009). D'un autre cote, nos experiences in vitro demontrent que PH2O2 diminue la production de NO par les cytotrophoblastes et confirment notre hypothese que PH2O2 agit comme inhibiteur de la synthese de NO. De plus, Noris et al ont constate une augmentation des niveaux d'arginase II dans la preeclampsie (NORIS et al, 2004). Les arginases peuvent agir en competition avec eNOS pour la Larginine et la metaboliser en uree et en L-ornithine, resultant en une diminution de la production de NO. En la presence diminuee de L-arginine, l'activation de la eNOS mene a une surproduction de l'anion superoxyde (O2) via la non liaison 121 entre la eNOS et la L-arginine (CULCASI et al, 1994). L'anion superoxyde reagit aussi avec le NO pour former le peroxynitrite (ONOO) (BECKMAN et KOPPENOL, 1996), diminuant la disponibilite du NO. L'hypothese que les niveaux de NO sont diminues par la diminution de la biodisponibilite de la L-arginase est supportee par une etude de Altun et al qui demontre qu'une supplementation en Larginine chez des rats preeclamptiques diminue les manifestations cliniques de la preeclampsie (ALTUN et al, 2008). II est aussi possible que PH2O2 entraine une diminution du NO via 1'augmentation de l'activite des arginases I et II (CEDERBAUM et al, 2004). II a aussi ete demontre que l'augmentation de la regulation des arginases par PH202 interfere avec la vasodilatation de l'endothelium entralnee par le NO (THENGCHAISRJ et al.,2006). De plus, nous avons constate une augmentation de l'activite des arginases dans les milieux de culture de cytotrophoblastes stimules par PH2O2 (resultats non publies), proposant que la diminution de NO pourrait bel et bien etre due a la diminution de l'arginine disponible plutot qu'a une apoptose des cytotrophoblastes. Comme nos experiences in vitro ont ete effectuees sur des cytotrophoblastes provenant de placentas normaux, une difference pourrait etre observee si l'on repetait les traitements avec PH2O2 sur des cytotrophoblastes provenant de patientes preeclamptiques. Les niveaux de NO et d ' P L ^ semblent aussi tous deux augmenter durant la grossesse, mais de maniere non significative, tant chez les femmes preeclamptiques que chez les femmes normales. Ceci peut etre explique par l'augmentation des interactions metaboliques entrainees par la croissance foetale et l'avancement de la grossesse. De plus, selon nos connaissances, aucune autre etude n'a mesure les 122 niveaux d'HiCh et de NO en debut et en fin de grossesse, nous limitant dans nos comparaisons. 123 4 CONCLUSION ET PERSPECTIVES Pour conclure, les differents travaux effectues nous ont permis de confirmer le role important qu'occupe le stress oxydatif, et plus precisement PH2O2 dans la preeclampsie. D'abord, nos resultats nous ont permis de confirmer nos observations anterieures indiquant une correlation in vivo entre le stress oxydatif et la fonction placentaire. I/H2O2 n'agit done pas seulement comme un mediateur cytotoxique, mais aussi comme une molecule capable d'induire la secretion de 1'hCG par les cytotrophoblastes, et ce, selon sa concentration. Nos resultats suggerent aussi un role possible de 1'hCG comme facteur antioxydant produit par le placenta pour contrer un stress oxydatif leger, ouvrant la voie a d'autres etudes dans ce sens. Le stress oxydatif a aussi eu des effets sur la secretion du TNF-a et du recepteur TNFR2 par les cytotrophoblastes, ayant comme consequence possible de reduire les effets de cette cytokine, autant autocrines que paracrines. Nos resultats ont aussi permis d'expliquer pourquoi le TNF-a est parfois difficile a detecter dans des milieux biologiques. De plus, l'usage du TNFR2 soluble recombinant semble etre envisageable pour reduire les effets inflammatoires du stress oxydatif. D'autres etudes pourraient etre entreprises pour explorer l'effet protecteur potentiel du TNFR2 chez l'humain ou l'animal, tant pour des maladies de grossesse caracterisees par un stress inflammation comme la preeclampsie que pour d'autres maladies inflammatoires. De plus, nos travaux ont permis de constater une augmentation du peroxyde d'hydrogene, tant entre la 10e et la 15e semaine de grossesse qu'a terme, chez les femmes preeclamptiques par rapport aux femmes normales. Cette augmentation a 124 ete inversement correlee avec les niveaux de NO, qui se trouvent diminues avec 1'augmentation d'tfeC^, ce qui rend ces deux mediateurs comme etant de possibles biomarqueurs de prediction la preeclampsie. D'autres travaux sont aussi en cours pour determiner l'effet du peroxyde d'hydrogene sur les arginases cytotrophoblastiques in vitro. 125 5 REMERCIEMENTS Je tiens tout d'abord a remercier mon directeur de maitrise, Dr Aziz Aris, qui m'a accueilli dans son laboratoire. En effet, c'est grace a ses conseils, ses connaissances scientifiques, sa disponibilite et sa patience que ce travail a pu etre effectue. Merci infiniment Dr Aris. Merci aussi a tous ceux que j ' a i cotoye de pres ou de loin durant mon sejour, tant au gens du laboratoire meme, ou aux membres du departement de Genetique, pour leur aide, que ce soit au point de vue de mes nombreuses questions techniques ou autre. J'aimerais aussi remercier les infirmieres du departement d'Obstetrique- Gynecologie du CHUS pour leur disponibilite et leur gentillesse, sans lesquelles mon recrutement aurait ete beaucoup plus ardu. Finalement, merci a mes parents, Philippe et Madeleine, mes freres et sceurs et amis, pour leur support constant, leurs encouragements dans les moments les plus difficiles comme les plus faciles. 126 6 REFERENCES Abe T, Konishi T, Hirano T, Kasai H, Shimizu K, Kashimura M, Higashi K. Possible correlation between DNA damage induced by hydrogen peroxide and translocation of heat shock 70 protein into the nucleus. Biochem Biophys Res Commun. 1995 Jan 17;206(2):548-55. Aderka D, Engelmann H, Maor Y, Brakebusch C, Wallach D. Stabilization of the bioactivity of tumor necrosis factor by its soluble receptors. J Exp Med. 1992 Febl;175(2):323-9. Agarwal A, Saleh RA, Bedaiwy MA. Role of reactive oxygen species in the pathophysiology of human reproduction. Fertil Steril. 2003 Apr;79(4):829-43. Allaire AD, Ballenger KA, Wells SR, McMahon MJ, Lessey BA. Placental apoptosis in preeclampsia. Obstet Gynecol. 2000 Aug;96(2):271-6. Al-Mulhim AA, Abu-Heija A, Al-Jamma F, El-Harith eA. Pre-eclampsia: Maternal risk factors and perinatal outcome. Fetal Diagn Ther. 2003 Jul-Aug;18(4):27580. Alsat E, Tarrade A, Merviel P, Evain-Brion D. Le cytotrophoblaste humain, un casse-tete pour le biologiste. M/S. Novembre 1999;11(15):1236-43. Aris A, Benali S, Ouellet A, Moutquin JM, Leblanc S. Potential biomarkers of preeclampsia : inverse correlation between oxygen peroxide and nitric oxide early in maternal circulation and at term in placenta of women with preecclampsia. Placenta. 2009 Apr;30(4):342-7. Arner ES, Holmgren A. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem. 2000 Oct;267(20):6102-9. 127 Askie LM, Duley L, Henderson-Smart DJ, Stewart LA, PARIS Collaborative Group. Antiplatelet agents for prevention of pre-eclampsia: A meta-analysis of individual patient data. Lancet. 2007 May 26;369(9575):1791-8. Baeten JM, Bukusi EA, Lambe M. Pregnancy complications and outcomes among overweight and obese nulliparous women. Am J Public Health. 2001 Mar;91(3):436-40. Barros JS, Baptista MG, Bairos VA. Human chorionic gonadotropin in human placentas from normal and preeclamptic pregnancies. Arch Gynecol Obstet. 2002Apr;266(2):67-71. Barton JR, Sibai BM. Prediction and prevention of recurrent preeclampsia. Obstet Gynecol. 2008 Aug;l 12(2 Pt l):359-72. Baumann MU, Deborde S, Illsley NP. Placental glucose transfer and fetal growth. Endocrine. 2002 Oct; 19(1): 13-22. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci U S A . 1990 Feb;87(4): 1620-4. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: The good, the bad, and ugly. Am J Physiol. 1996 Nov;271(5 Pt l):C1424-37. Benirschke, K, Kaufmann, P. Pathology of the human placenta. Fourth edition ed. New York: Springer-Verlag; 2000. 947 p. Benyo DF, Smarason A, Redman CW, Sims C, Conrad KP. Expression of inflammatory cytokines in placentas from women with preeclampsia. J Clin Endocrinol Metab. 2001 Jun;86(6):2505-12. 128 Beutler B, Cerami A. The biology of cachectin/TNF--a primary mediator of the host response. Annu Rev Immunol. 1989;7:625-55. Beutler B, Cerami A. The history, properties, and biological effects of cachectin. Biochemistry. 1988 Oct 4;27(20):7575-82. Beyaert R, Fiers W. Molecular mechanisms of tumor necrosis factor-induced cytotoxicity, what we do understand and what we do not. FEBS Lett. 1994 Feb 28;340(l-2):9-16. Bhatnagar S, Bhattacharjee J, Vaid M, Madan T, Trivedi SS, Sarma PU. Inducible nitric oxide synthase (iNOS) gene polymorphism in pre-eclampsia: A pilot study in north india. Aust N Z J Obstet Gynaecol. 2007 Dec;47(6):477-82. Bischof P, Irminger-Finger I. The human cytotrophoblastic cell, a mononuclear chameleon. Int J Biochem Cell Biol. 2005 Jan;37(l):l-16. Brigelius-Flohe R, Aumann KD, Blocker H, Gross G, Kiess M, Kloppel KD, Maiorino M, Roveri A, Schuckelt R, Usani F. Phospholipid-hydroperoxide glutathione peroxidase, genomic DNA, cDNA, and deduced amino acid sequence. J Biol Chem. 1994 Mar ll;269(10):7342-8. Burton GJ, Jauniaux E, Watson AL. Maternal arterial connections to the placental intervillous space during the first trimester of human pregnancy: The boyd collection revisited. Am J Obstet Gynecol. 1999 Sep;l81 (3):718-24. Burton GW, Joyce A, Ingold KU. First proof that vitamin E is major lipid-soluble, chain-breaking antioxidant in human blood plasma. Lancet. 1982 Aug 7;2(8293):327. 129 Catalano PM. Management of obesity in pregnancy. Obstet Gynecol. 2007 Feb;109(2Pt l):419-33. Cederbaum SD, Yu H, Grody WW, Kern RM, Yoo P, Iyer RK. Arginases I and II: Do their functions overlap? Mol Genet Metab. 2004 Apr;81 Suppl l:S38-44. Chesley LC. Eclampsia: The remote prognosis. Semin Perinatol. 1978 Jan;2(l):99111. Chesley LC. Vascular reactivity in normal and toxemic pregnancy. Clin Obstet Gynecol. 1966 Dec;9(4):871-81. Collart MA, Baeuerle P, Vassalli P. Regulation of tumor necrosis factor alpha transcription in macrophages: Involvement of four kappa B-like motifs and of constitutive and inducible forms of NF-kappa B. Mol Cell Biol. 1990 Apr; 10(4): 1498-506. Conde-Agudelo A, Althabe F, Belizan JM, Kafury-Goeta AC. Cigarette smoking during pregnancy and risk of preeclampsia: A systematic review. Am J Obstet Gynecol. 1999 Oct;181(4):1026-35. Conner EM, Grisham MB. Inflammation, free radicals, and antioxidants. Nutrition. 1996 Apr; 12(4):274-7. Conrad KP, Miles TM, Benyo DF. Circulating levels of immunoreactive cytokines in women with preeclampsia. Am J Reprod Immunol. 1998 Aug;40(2): 102-11. Culcasi M, Lafon-Cazal M, Pietri S, Bockaert J. Glutamate receptors induce a burst of superoxide via activation of nitric oxide synthase in arginine-depleted neurons. J Biol Chem. 1994 Apr 29;269(17):12589-93. 130 Dantzer V, Leach L, Leiser R. Angiogenesis and placental vasculature-^ workshop report. Placenta. 2000 Mar-Apr;21 Suppl A:S69-70. Droge W. Free radicals in the physiological control of cell function. Physiol Rev. 2002 Jan;82(l):47-95. Evain-Brion D, Malassine A. Human placenta as an endocrine organ. Growth Horm IGF Res. 2003 Aug;13 Suppl A:S34-7. Fontaine V, Mohand-Said S, Hanoteau N, Fuchs C, Pfizenmaier K, Eisel U. Neurodegenerative and neuroprotective effects of tumor necrosis factor (TNF) in retinal ischemia: Opposite roles of TNF receptor 1 and TNF receptor 2. J Neurosci. 2002 Apr 1;22(7):RC216. Grell M, Becke FM, Wajant H, Mannel DN, Scheurich P. TNF receptor type 2 mediates thymocyte proliferation independently of TNF receptor type 1. Eur J Immunol. 1998 Jan;28(l):257-63. Gude NM, Roberts CT, Kalionis B, King RG. Growth and function of the normal human placenta. Thromb Res. 2004; 114(5-6):397-407. Halliwell B, Gutteridge JM, Cross CE. Free radicals, antioxidants, and human disease: Where are we now? J Lab Clin Med. 1992 Jun;l 19(6):598-620. Hiby SE, Walker JJ, O'shaughnessy KM, Redman CW, Carrington M, Trowsdale J, Moffett A. Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success. J Exp Med. 2004 Oct 18;200(8):957-65. 131 Higuchi M, Aggarwal BB. Differential roles of two types of the TNF receptor in TNF-induced cytotoxicity, DNA fragmentation, and differentiation. J Immunol. 1994 Apr 15;152(8):4017-25. Higuchi M, Aggarwal BB. TNF induces internalization of the p60 receptor and shedding of the p80 receptor. J Immunol. 1994 Apr l;152(7):3550-8. Hsu CD, Lucas RB, Johnson TR, Hong SF, Chan DW. Elevated urine thrombomodulin/creatinine ratio in severely preeclamptic pregnancies. Am J Obstet Gynecol. 1994 Sep;171(3):854-6. Hubel CA, Kozlov AV, Kagan VE, Evans RW, Davidge ST, McLaughlin MK, Roberts JM. Decreased transferrin and increased transferrin saturation in sera of women with preeclampsia: Implications for oxidative stress. Am J Obstet Gynecol. 1996 Sep;175(3 Pt l):692-700. Hung TH, Skepper JN, Charnock-Jones DS, Burton GJ. Hypoxia-reoxygenation: A potent inducer of apoptotic changes in the human placenta and possible etiological factor in preeclampsia. Circ Res. 2002 Jun 28;90(12):1274-81. Hunt JS, Chen HL, Miller L. Tumor necrosis factors: Pivotal components of pregnancy? Biol Reprod. 1996 Mar;54(3):554-62. Idriss HT, Naismith JH. TNF alpha and the TNF receptor superfamily: Structurefunction relationship(s). Microsc Res Tech. 2000 Aug 1;50(3): 184-95. Ishikawa T, Harada T, Koi H, Kubota T, Azuma H, Aso T. Identification of arginase in human placental villi. Placenta. 2007 Feb-Mar;28(2-3): 133-8. 132 Jauniaux E, Watson AL, Hempstock J, Bao YP, Skepper JN, Burton GJ. Onset of maternal arterial blood flow and placental oxidative stress. A possible factor in human early pregnancy failure. Am J Pathol. 2000 Dec;157(6):2111-22. Keith M, Norwich KH, Wong W, Jeejeebhoy KN. The tissue distribution of tumor necrosis factor-alpha in rats: A compartmental model. Metabolism. 2000 Oct;49(10):1309-17. Kharfi A, Bureau M, Giguere Y, Moutquin JM, Forest JC. Dissociation between increased apoptosis and expression of the tumor necrosis factor-alpha system in term placental villi with preeclampsia. Clin Biochem. 2006 Jun;39(6):646-51. Kharfi A, Giguere Y, De Grandpre P, Moutquin JM, Forest JC. Human chorionic gonadotropin (hCG) may be a marker of systemic oxidative stress in normotensive and preeclamptic term pregnancies. Clin Biochem. 2005 Aug;38(8):717-21. King A, Burrows TD, Hiby SE, Bowen JM, Joseph S, Verma S, Lim PB, Gardner L, Le Bouteiller P, Ziegler A, Uchanska-Ziegler B, Loke YW. Surface expression of HLA-C antigen by human extravillous trophoblast. Placenta. 2000 May;21(4):376-87. Knight M, Redman CW, Linton EA, Sargent IL. Shedding of syncytiotrophoblast microvilli into the maternal circulation in pre-eclamptic pregnancies. Br J Obstet Gynaecol. 1998 Jun;105(6):632-40. Kretz-Remy C, Mehlen P, Mirault ME, Arrigo AP. Inhibition of I kappa B-alpha phosphorylation and degradation and subsequent NF-kappa B activation by glutathione peroxidase overexpression. J Cell Biol. 1996 Jun;133(5):1083-93. 133 Li Y, Matsuzaki N, Masuhiro K, Kameda T, Taniguchi T, Saji F, Yone K, Tanizawa O. Trophoblast-derived tumor necrosis factor-alpha induces release of human chorionic gonadotropin using interleukin-6 (IL-6) and IL-6-receptordependent system in the normal human trophoblasts. J Clin Endocrinol Metab. 1992 Jan;74(l): 184-91. Liu D, Wen J, Liu J, Li L. The roles of free radicals in amyotrophic lateral sclerosis: Reactive oxygen species and elevated oxidation of protein, DNA, and membrane phospholipids. FASEB J. 1999 Dec; 13(15):2318-28. Lotz M, Setareh M, von Kempis J, Schwarz H. The nerve growth factor/tumor necrosis factor receptor family. J Leukoc Biol. 1996 Jul;60(l):l-7. Madazli R, Benian A, Aydin S, Uzun H, Tolun N. The plasma and placental levels of malondialdehyde, glutathione and superoxide dismutase in pre-eclampsia. J Obstet Gynaecol. 2002 Sep;22(5):477-80. Magee LA, von Dadelszen P, Bohun CM, Rey E, El-Zibdeh M, Stalker S, Ross S, Hewson S, Logan AG, Ohlsson A, Naeem T, Thornton JG, Abdalla M, Walkinshaw S, Brown M, Davis G, Hannah ME. Serious perinatal complications of non-proteinuric hypertension: An international, multicentre, retrospective cohort study. J Obstet Gynaecol Can. 2003 May;25(5):372-82. Marin JJ, Macias RI, Serrano MA. The hepatobiliary-like excretory function of the placenta. A review. Placenta. 2003 May;24(5):431-8. McAleer MF, Tuan RS. Metallothionein protects against severe oxidative stressinduced apoptosis of human trophoblastic cells. In Vitr Mol Toxicol. 2001 Fall; 14(3):219-31. 134 McCartney CH. The acute hypertensive disorders of pregnancy, classified by renal histology. Gynaecologia. 1969; 167(4):214-20. McWhirter SM, Fitzgerald KA, Rosains J, Rowe DC, Golenbock DT, Maniatis T. IFN-regulatory factor 3-dependent gene expression is defective in Tbkldeficient mouse embryonic fibroblasts. Proc Natl Acad Sci U S A . 2004 Jan 6;101(l):233-8. Mello Filho AC, Hoffmann ME, Meneghini R. Cell killing and DNA damage by hydrogen peroxide are mediated by intracellular iron. Biochem J. 1984 Feb 15;218(l):273-5. Meyer M, Schreck R, Baeuerle PA. H202 and antioxidants have opposite effects on activation of NF-kappa B and AP-1 in intact cells: AP-1 as secondary antioxidant-responsive factor. EMBO J. 1993 May;12(5):2005-15. Mikhail MS, Anyaegbunam A, Garfinkel D, Palan PR, Basu J, Romney SL. Preeclampsia and antioxidant nutrients: Decreased plasma levels of reduced ascorbic acid, alpha-tocopherol, and beta-carotene in women with preeclampsia. Am J Obstet Gynecol. 1994 Jul; 171(1): 150-7. Miller AF. Superoxide dismutases: Active sites that save, but a protein that kills. Curr Opin Chem Biol. 2004 Apr;8(2): 162-8. Moffett A, Loke YW. The immunological paradox of pregnancy: A reappraisal. Placenta. 2004 Jan;25(l):l-8. Morris SM,Jr. Regulation of enzymes of the urea cycle and arginine metabolism. Annu Rev Nutr. 2002;22:87-105. 135 Myatt L, Cui X. Oxidative stress in the placenta. Histochem Cell Biol. 2004 Oct;122(4):369-82. Nappi AJ, Vass E. Iron, metalloenzymes and cytotoxic reactions. Cell Mol Biol (Noisy-le-grand). 2000 May;46(3):637-47. Noris M, Todeschini M, Cassis P, Pasta F, Cappellini A, Bonazzola S, Macconi D, Maucci R, Porrati F, Benigni A, Picciolo C, Remuzzi G. L-arginine depletion in preeclampsia orients nitric oxide synthase toward oxidant species. Hypertension. 2004 Mar;43(3):614-22. Nose K. Role of reactive oxygen species in the regulation of physiological functions. Biol Pharm Bull. 2000 Aug;23(8):897-903. Orange SJ, Painter D, Horvath J, Yu B, Trent R, Hennessy A. Placental endothelial nitric oxide synthase localization and expression in normal human pregnancy and pre-eclampsia. Clin Exp Pharmacol Physiol. 2003 May-Jun;30(5-6):376-81. Page IH. A method for producing persistent hypertension by cellophane. Science. 1939 Mar 24;89(2308):273-4. Pasanen M. The expression and regulation of drug metabolism in human placenta. Adv Drug Deliv Rev. 1999 Jun 14;38(l):81-97. Perler L, Pfister H, Schweizer M, Peterhans E, Jungi TW. A bioassay for interferon type I based on inhibition of sendai virus growth. J Immunol Methods. 1999 Jan l;222(l-2):189-96. Pierce JD, Cackler AB, Arnett MG. Why should you care about free radicals? RN. 2004 Jan;67(l):38,42; quiz 43. 136 Pijnenborg R, Bland JM, Robertson WB, Brosens I. Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta. 1983 0ct-Dec;4(4):397-413. Pijnenborg R, Bland JM, Robertson WB, Dixon G, Brosens I. The pattern of interstitial trophoblastic invasion of the myometrium in early human pregnancy. Placenta. 1981 Oct-Dec;2(4):303-16. Poston L, Briley AL, Seed PT, Kelly FJ, Shennan AH, Vitamins in Pre-eclampsia (VIP) Trial Consortium. Vitamin C and vitamin E in pregnant women at risk for pre-eclampsia (VIP trial): Randomised placebo-controlled trial. Lancet. 2006 Apr 8;367(9517): 1145-54. Rabineau, D, Dupont, J, Plateaux, P. Embryologie humaine. [Internet]. France: SFRS Service du Film de Recherche Scientifique; 2003. http://cvirtuel.cochin.univ-paris5.fr/Embryologie/AnimEntre/AnimEntrel.html Ramsey, EM, Donner, MW. Placental Vasculature and Circulation. Anatomy, Physiology, Radiology, Clinical Aspects. Atlas and Textbook. Philadelphie: W.B.Saunders Company; 1980. 101 p. Reddy SA, Chaturvedi MM, Darnay BG, Chan H, Higuchi M, Aggarwal BB. Reconstitution of nuclear factor kappa B activation induced by tumor necrosis factor requires membrane-associated components, comparison with pathway activated by ceramide. J Biol Chem. 1994 Oct 14;269(41):25369-72. Reiter RJ, Melchiorri D, Sewerynek E, Poeggeler B, Barlow-Walden L, Chuang J, Ortiz GG, Acuna-Castroviejo D. A review of the evidence supporting melatonin's role as an antioxidant. J Pineal Res. 1995 Jan; 18(1): 1-11. 137 Report of the national high blood pressure education program, working group on high blood pressure in pregnancy. Am J Obstet Gynecol. 2000; 183 :S 1-22. Riley JC, Behrman HR. Oxygen radicals and reactive oxygen species in reproduction. Proc Soc Exp Biol Med. 1991 Dec; 198(3):781-91. Roberts JM, Gammill HS. Preeclampsia: Recent insights. Hypertension. 2005 Dec;46(6):1243-9. Roberts JM, Lain KY. Recent insights into the pathogenesis of pre-eclampsia. Placenta. 2002 May;23(5):359-72. Roberts JM, Redman CW. Pre-eclampsia: More than pregnancy-induced hypertension. Lancet. 1993 Jun 5;341(8858):1447-51. Sandrim VC, Palei AC, Metzger IF, Gomes VA, Cavalli RC, Tanus-Santos JE. Nitric oxide formation is inversely related to serum levels of antiangiogenic factors soluble fms-like tyrosine kinase-1 and soluble endogline in preeclampsia. Hypertension. 2008 Aug;52(2):402-7. Schiessl B, Mylonas I, Hantschmann P, Kuhn C, Schulze S, Kunze S, Friese K, Jeschke U. Expression of endothelial NO synthase, inducible NO synthase, and estrogen receptors alpha and beta in placental tissue of normal, preeclamptic, and intrauterine growth-restricted pregnancies. J Histochem Cytochem. 2005 Dec;53(12):1441-9. Schonfelder G, Fuhr N, Hadzidiakos D, John M, Hopp H, Paul M. Preeclampsia is associated with loss of neuronal nitric oxide synthase expression in vascular smooth muscle cells of the human umbilical cord. Histopathology. 2004 Feb;44(2): 116-28. 138 Schreck R, Rieber P, Baeuerle PA. Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-kappa B transcription factor and HIV-1. EMBO J. 1991 Aug;10(8):2247-58. Scumpia PO, Delano MJ, Kelly KM, O'Malley KA, Efron PA, McAuliffe PF, Brusko T, Ungaro R, Barker T, Wynn JL, Atkinson MA, Reeves WH, Salzler MJ, Moldawer LL. Increased natural CD4+CD25+ regulatory T cells and their suppressor activity do not contribute to mortality in murine polymicrobial sepsis. J Immunol. 2006 Dec 1;177(11):7943-9. Shaamash AH, Elsnosy ED, Makhlouf AM, Zakhari MM, Ibrahim OA, EL-dien HM. Maternal and fetal serum nitric oxide (NO) concentrations in normal pregnancy, pre-eclampsia and eclampsia. Int J Gynaecol Obstet. 2000 Mar;68(3):207-14. Shakhov AN, Collart MA, Vassalli P, Nedospasov SA, Jongeneel CV. Kappa Btype enhancers are involved in lipopolysaccharide-mediated transcriptional activation of the tumor necrosis factor alpha gene in primary macrophages. J Exp Med. 1990 Jan l;171(l):35-47. Shu HB, Takeuchi M, Goeddel DV. The tumor necrosis factor receptor 2 signal transducers TRAF2 and c-IAPl are components of the tumor necrosis factor receptor 1 signaling complex. Proc Natl Acad Sci U S A . 1996 Nov 26;93(24):13973-8. Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet. 2005 Feb 26-Mar 4;365(9461):785-99. 139 Sibai BM. Diagnosis and management of gestational hypertension and preeclampsia. Obstet Gynecol. 2003 Jul; 102(1): 181-92. Stone JR, Yang S. Hydrogen peroxide: A signaling messenger. Antioxid Redox Signal. 2006 Mar-Apr;8(3-4):243-70. Stone JR, Yang S. Hydrogen peroxide: A signaling messenger. Antioxid Redox Signal. 2006 Mar-Apr;8(3-4):243-70. Stulc J. Placental transfer of inorganic ions and water. Physiol Rev. 1997 Jul;77(3):805-36. Takada Y, Mukhopadhyay A, Kundu GC, Mahabeleshwar GH, Singh S, Aggarwal BB. Hydrogen peroxide activates NF-kappa B through tyrosine phosphorylation of I kappa B alpha and serine phosphorylation of p65: Evidence for the involvement of I kappa B alpha kinase and syk protein-tyrosine kinase. J Biol Chem. 2003 Jun 27;278(26):24233-41. Thengchaisri N, Hein TW, Wang W, Xu X, Li Z, Fossum TW, Kuo L. Upregulation of arginase by H202 impairs endothelium-dependent nitric oxide-mediated dilation of coronary arterioles. Arterioscler Thromb Vase Biol. 2006 Sep;26(9):2035-42. Thommesen L, Laegreid A. Distinct differences between TNF receptor 1- and TNF receptor 2-mediated activation of NFkappaB. J Biochem Mol Biol. 2005 May 31;38(3):281-9. Toth P, Lukacs H, Gimes G, Sebestyen A, Pasztor N, Paulin F, Rao CV. Clinical importance of vascular LH/hCG receptors—a review. Reprod Biol. 2001 Nov;l(2):5-ll. 140 Tracey KJ, Cerami A. Tumor necrosis factor: An updated review of its biology. Crit Care Med. 1993 Oct;21(10 Suppl):S415-22. Tracey KJ, Cerami A. Tumor necrosis factor: An updated review of its biology. Crit Care Med. 1993 Oct;21(10 Suppl):S415-22. Tuffnell DJ, Jankowicz D, Lindow SW, Lyons G, Mason GC, Russell IF, Walker JJ, Yorkshire Obstetric Critical Care Group. Outcomes of severe preeclampsia/eclampsia in yorkshire 1999/2003. BJOG. 2005 Jul;112(7):875-80. Vandenabeele P, Declercq W, Beyaert R, Fiers W. Two tumour necrosis factor receptors: Structure and function. Trends Cell Biol. 1995 Oct;5(10):392-9. Vandenabeele P, Declercq W, Vercammen D, Van de Craen M, Grooten J, Loetscher H, Brockhaus M, Lesslauer W, Fiers W. Functional characterization of the human tumor necrosis factor receptor p75 in a transfected rat/mouse T cell hybridoma. J Exp Med. 1992 Oct 1;176(4):1015-24. Veal EA, Day AM, Morgan BA. Hydrogen peroxide sensing and signaling. Mol Cell. 2007 Apr 13;26(1):1-14. Vercammen D, Vandenabeele P, Declercq W, Van de Craen M, Grooten J, Fiers W. Cytotoxicity in L929 murine fibrosarcoma cells after triggering of transfected human p75 tumour necrosis factor (TNF) receptor is mediated by endogenous murine TNF. Cytokine. 1995 Jul;7(5):463-70. Villamor E, Cnattingius S. Interpregnancy weight change and risk of adverse pregnancy outcomes: A population-based study. Lancet. 2006 Sep 30;368(9542): 1164-70. 141 Vural P. Nitric oxide/endothelin-1 in preeclampsia. Clin Chim Acta. 2002 Mar;317(l-2):65-70. Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell Death Differ. 2003 Jan;10(l):45-65. Walsh SW. Maternal-placental interactions of oxidative stress and antioxidants in preeclampsia. Semin Reprod Endocrinol. 1998;16(1):93-104. Wang D, Baldwin transcription by AS,Jr. Activation of nuclear factor-kappaB-dependent necrosis factor-alpha is tumor mediated through phosphorylation of RelA/p65 on serine 529. J Biol Chem. 1998 Nov 6;273(45):29411-6. Wang Y, Walsh SW. Antioxidant activities and mRNA expression of superoxide dismutase, catalase, and glutathione peroxidase in normal and preeclamptic placentas. J Soc Gynecol Investig. 1996 Jul-Aug;3(4): 179-84. Ward NC, Croft KD. Hypertension and oxidative stress. Clin Exp Pharmacol Physiol. 2006 Sep;33(9):872-6. Wickens D, Wilkins MH, Lunec J, Ball G, Dormandy TL. Free radical oxidation (peroxidation) products in plasma in normal and abnormal pregnancy. Ann Clin Biochem. 1981 May; 18(Pt 3): 158-62. Witting LA, Horwitt MK. Effect of degree of fatty acid unsaturation in tocopherol deficiency-induced creatinuria. J Nutr. 1964 Jan;82:19-33. Yu BP. Cellular defenses against damage from reactive oxygen species. Physiol Rev. 1994Jan;74(l):139-62. 142 Yucesoy G, Ozkan S, Bodur H, Tan T, Caliskan E, Vural B, Corakci A. Maternal and perinatal outcome in pregnancies complicated with hypertensive disorder of pregnancy: A seven year experience of a tertiary care center. Arch Gynecol Obstet. 2005 Nov;273(l):43-9. Yui J, Garcia-Lloret M, Wegmann TG, Guilbert LJ. Cytotoxicity of tumour necrosis factor-alpha and gamma-interferon against primary human placental trophoblasts. Placenta. 1994 Dec;15(8):819-35. Zygmunt M, Hahn D, Munstedt K, Bischof P, Lang U. Invasion of cytotrophoblastic JEG-3 cells is stimulated by hCG in vitro. Placenta. 1998 Nov;19(8):587-93. 143