Centre de recherche en Neurobiologie-Neurophysiologie de Marseille
Faculté de Médecine Nord
Université de la Méditerranée Aix - Marseille II Ecole Doctorale des Sciences de la Vie et de la Santé
Thèse présentée pour obtenir le grade de
Docteur de l’Université de la Méditerranée II Discipline Neurosciences
Par Frédéric CASTINETTI
Soutenue publiquement Le 11 Octobre 2010
FACTEURS DE TRANSCRIPTION A HOMEODOMAINE : DU MODELE MURIN A L’HYPOPITUITARISME HUMAIN
MEMBRES DU JURY
Pr Alain Enjalbert Président du jury Pr Thierry Brue Directeur de Thèse Pr Serge Amselem Rapporteur externe Pr Philippe Chanson Rapporteur externe Pr Sally Camper Examinateur
! "! REMERCIEMENTS
Pour leurs rôles majeurs dans la genèse de cette thèse, Thierry Brue pour sa présence et son soutien constants depuis le début de mon Internat ; Sally Camper et les membres du Camper’s lab, plus particulièrement Michelle, Shannon et Amanda, pour leur accueil, leur aide et leur soutien
Pour avoir accepté d’être président et rapporteurs de ce jury Alain Enjalbert, Philippe Chanson et Serge Amselem
Pour leur accueil chaleureux et leur aide précieuse lors de mon Master 2 Rachel, Marie-Hélène (en particulier lors des manips de radioactivité…) , Jean-Paul, Anne, Alex, Nicolas, Jean-Louis, Anne-Marie et Denis, et plus globalement tous les membres du laboratoire de la Faculté Nord
Pour tout… et pour le reste… Ma femme et mes parents
Pour leur support lors de mon séjour à Ann Arbor, Les laboratoires Novo-Nordisk GH, Novartis et IPSEN, l’ADEREM et la Société Française d’Endocrinologie
Pour leur bonne humeur (et pour ne pas m’avoir oublié pendant mon séjour Américain) Tous les membres du service d’endocrinologie de l’Hôpital de la Timone
! #! PLAN
INTRODUCTION
ETAT DES CONNAISSANCES 1. Le développement hypophysaire chez la souris a. De la placode hypophysaire à la poche de Rathke b. De la poche de Rathke à l’hypophyse différenciée c. La voie Wnt/ß-caténine d. Les facteurs de transcription hypophysaire 2. Les facteurs de transcription à homéodomaine de type paired a. Au cours du développement hypophysaire murin i. HESX1 ii. PROP1 iii. PITX1 iv. PITX2 v. OTX2 b. En pathologie humaine i. HESX1 ii. PROP1 iii. PITX2 iv. OTX2 3. POU1F1 (Pit1), facteur de transcription à homéodomaine POU a. Au cours du développement hypophysaire b. En pathologie humaine 4. Les facteurs de transcription à homéodomaine de type LIM a. Au cours du développement hypophysaire murin i. LHX3 ii. LHX4 iii. ISL1 b. En pathologie humaine i. LHX3 ii. LHX4
OBJECTIFS
RESULTATS 1. DEVELOPPEMENT HYPOPHYSAIRE: VERS UNE MEILLEURE COMPREHENSION DES INTERACTIONS ENTRE PROP1 et HESX1 « CO-REPRESSORS TLE1 AND TLE3 INTERACT WITH HESX1 AND PROP1. » LR. Carvalho*, ML. Brinkmeier*, F. Castinetti *, BS. Ellsworth, SA. Camper Molecular endocrinology, in press *co-premiers auteurs
2. MISE EN EVIDENCE D’UNE NOUVELLE MUTATION DE LHX4, RESPONSABLE D’HYPOPITUITARISME CONGENITAL “A NOVEL DYSFUNCTIONAL LHX4 MUTATION WITH HIGH PHENOTYPIC VARIABILITY IN PATIENTS WITH HYPOPITUITARISM” F. Castinetti *, A. Saveanu*, R. Reynaud, MH. Quentien, A. Buffin, R. Brauner, N.
! $! Kaffel, F. Albarel, AM. Guedj, M. El Kholy, M. Amin, A. Enjalbert, A. Barlier, T. Brue Journal of Clinical Endocrinology and Metabolism 2008 Jul;93(7):2790-9 *co-premiers auteurs
3. IMPLICATIONS POTENTIELLES DU FACTEURS DE TRANSCRIPTION PITX2 DANS LE DEVELOPPEMENT DE L’AXE THYREOTROPE F. Castinetti*, M.L. Brinkmeier*, D.F. Gordon, K. R. Vella, J.M. Kerr, A. N. Hollenberg, T. Brue, E.C. Ridgway, S.A. Camper Soumis à Molecular Endocrinology
4. EXPRESSION ET ROLES DU FACTEUR DE TRANSCRIPTION ISL-1 AU COURS DU DEVELOPPEMENT HYPOPHYSAIRE 1F. Castinetti*, M.L. Brinkmeier*, D.F. Gordon, K. R. Vella, J.M. Kerr, A. N. Hollenberg, T. Brue, E.C. Ridgway, S.A. Camper En préparation
PERSPECTIVES 1. Comment expliquer les ressemblances et différences entre les modèles humain et murin ? 2. Comment identifier de nouveaux facteurs de transcription impliqués dans le développement hypophysaire ? 3. Comment améliorer le traitement des déficits hypophysaires: les cellules souches hypophysaires ?
CONCLUSION
REFERENCES BIBLIOGRAPHIQUES
! %! ANNEXES: Articles de revue publiés sur le sujet
DEFICIT HYPOPHYSAIRE COMBINE MULTIPLE: ASPECTS CLINIQUES ET GENETIQUES F. Castinetti , R. Reynaud, A. Saveanu, M.-H. Quentien, F. Albarel, A. Barlier, A. Enjalbert, T. Brue , Annales d’Endocrinologie, 69 (2008) 7–17
CONGENITAL PITUITARY HORMONE DEFICIENCIES: ROLES OF LHX3/LHX4 GENES F. Castinetti , R. Reynaud*, A. Saveanu, M.-H. Quentien, F. Albarel, A. Enjalbert, A. Barlier, T. Brue Expert Reviews in Endocrinology and Metabolism, 2009 * Co-premiers auteurs
MOLECULAR MECHANISMS OF PITUITARY ORGANOGENESIS: IN SEARCH OF NOVEL REGULATORY GENES S.W. Davis , F. Castinetti , L.R. Carvalho , B.S. Ellsworth, M.A. Potok, R.H. Lyons, M.L. Brinkmeier, L.T. Raetzman, P. Carninci, A.H. Mortensen, Y. Hayashizaki, I.J.P. Arnhold, B.B. Mendonca, T. Brue, S.A. Camper Molecular and Cellular Endocrinology, in press
PITUITARY STEM CELLS UPDATE AND IMPLICATIONS FOR TREATING HYPOPITUITARISM F. Castinetti , S.W. Davis, T. Brue, S.A. Camper Endocrine reviews, soumis
! &! RESUME
L’hypopituitarisme se définit par le déficit d’une ou plusieurs hormones hypophysaires. L’hypopituitarisme congénital est lié à des mutations de facteurs de transcription impliqués dans le développement hypophysaire. Identifier les mécanismes et étiologies d’hypopituitarisme congénital doit permettre d’améliorer les traitements des patients. Dans cette optique, ce travail a porté sur 3 aspects : Clarifier les mécanismes permettant la différenciation des lignées hypophysaires. Au cours du développement hypophysaire chez la souris, il existe un phénomène complexe d’interaction entre 2 facteurs de transcription à homéodomaine paired (Prop1 et Hesx1), la voie Wnt-ßcaténine et les co-répresseurs de la famille Groucho/TLE. Ces interactions sont nécessaires à l’expression d’un autre facteur de transcription hypophysaire, Pit-1 (Pou1f1), impliqué dans la différentiation des lignées hypophysaires somato-lactotropes et thyréotropes. Nous avons démontré in vitro, que les co-répresseurs de la famille TLE jouaient un rôle inhibiteur direct sur l’activation de l’early enhancer de POU1F1 à e12- e13, indépendamment de l’action de HESX1. Nos modèles de souris transgéniques avec expression permanente de HESX1 et TLE3 permettent de mettre en évidence le rôle inhibiteur majeur de HESX1, et le rôle accessoire de TLE3. Les mutations de PROP1 étant à l’origine d’une expression persistante de HESX1 et TLE3, il est probable qu’ils jouent un rôle dans le déficit en sous-unité alpha observé chez les patients déficitaires en PROP1. Identifier et analyser la signification fonctionnelle de nouveaux variants alléliques du gène d’un facteur de transcription à homéodomaine LIM, LHX4 . La mutation T99fs de LHX4 est à l’origine d’un phénotype hypophysaire très variable au sein d’une même famille, en termes de déficits et de morphologie hypophysaires, et d’anomalies extra- hypophysaires associées. Les études fonctionnelles ont montré que cette mutation était responsable d’un phénomène d’haplo-insuffisance. Cette nouvelle mutation permet d’enrichir le spectre phénotypique des patients chez lesquels doit être effectué un séquençage du gène LHX4 à la recherche d’étiologie de déficit hypophysaire combiné multiple. Identifier des mécanismes nécessaires au développement de l’axe thyréotrope. Des souris exprimant une nouvelle recombinase Cre sous contrôle du promoteur de la Tshß ont été croisées avec des souris transgéniques pour lesquelles les gènes de Pitx2 ou d’ Isl1 (2 facteurs de transcription impliqués dans le développement hypophysaire) étaient encadrés de séquences flox. Les modèles permettaient ainsi l’inactivation de Pitx2 et Isl1 au sein des cellules thyréotropes au cours de l’embryogenèse. L’étude phénotypique retrouve un déficit de croissance compatible avec un déficit thyréotrope partiel en cas d’inactivation de Pitx2 : ce phénotype est probablement lié à un mécanisme compensateur assuré par PITX1, un facteur de transcription à homéodomaine bicoïde possédant le même homéodomaine et domaine C terminal que PITX2. A l’inverse, l’inactivation de Isl& se traduit par un déficit thyréotrope complet. Le fait que les transcrits de l’ensemble des facteurs de transcription nécessaires au développement de l’axe thyréotrope soient diminués dans ce modèle souligne le rôle majeur de ISL1 dans la fonction et la maintenance de l’axe thyréotrope. Nos résultats permettent de mieux appréhender certains des nombreux mécanismes et facteurs impliqués dans le développement hypophysaire chez la souris, et dans la pathologie hypophysaire chez l’homme.
! '!
INTRODUCTION
! (! L’hypophyse est le chef d’orchestre des fonctions endocriniennes chez l’Homme. Si sa taille ne laisse pas envisager un tel rôle, sa complexité (en termes de développement, organisation, fonctionnement) suggère fortement son importance, mais aussi les difficultés inhérentes à son étude. Les modèles murins nous ont fourni des indications précieuses sur cet organe : étapes nécessaires à son développement, interactions aux niveaux protéique et moléculaire, anomalies à l’origine de pathologies parfois sévères… Malgré cela, de nombreuses publications apportent fréquemment des nouvelles informations, complétant ou modifiant les théories précédentes. Ce dernier point prouve à quel point les études sur l’hypophyse doivent être poursuivies, car il est vraisemblable que nous ne connaissons qu’une infime partie de ce que ce précieux organe nous cache.
L’interface entre recherche fondamentale et médecine clinique est un pan important de l’étude de l’hypophyse. Les mutations de facteurs de transcription hypophysaires sont à l’origine de déficits hypophysaires, ou hypopituitarisme congénital, générateurs de morbidité voire de mortalité. Mais les anomalies hypophysaires peuvent aussi être à l’origine d’une hyperactivité (proliférante et/ou sécrétante) avec des phénomènes tumoraux, les adénomes hypophysaires. Chacune de ces 2 branches nécessite une thérapeutique adaptée, performante et bien tolérée. Je me suis plus particulièrement intéressé aux déficits hypophysaires, et j’ai pu travailler sur le modèle (murin) et l’original (humain) : cette dichotomie a été possible en travaillant en partie dans le laboratoire de Neuroendocrinologie de la faculté de médecine Nord dans le cadre du Centre de références DEFHY et grâce au réseau collaboratif GENHYPOPIT (hypopituitarisme humain) et dans le laboratoire du Dr Sally Camper, à Ann Arbor, Michigan (modèle murin). Chacun de ces laboratoires avait une riche expérience dans le développement hypophysaire, et m’a permis de mieux appréhender le fonctionnement de cet organe.
Ce mémoire sera axé autour de ce qui me semble être un aperçu des différentes possibilités d’études de l’hypophyse - le niveau protéique et moléculaire (interactions entre facteurs de transcription) - le modèle murin (étude phénotypique après inactivation de facteurs de transcription) - l’hypopituitarisme humain (étude phénotypique et fonctionnelle à partir d’une mutation d’un facteur de transcription)
Lors de la présentation de l’état des connaissances, je vais particulièrement développer les données concernant 2 familles de facteurs de transcription hypophysaires, et leurs rôles au cours du développement hypophysaire (principalement murin) et en pathologie humaine. Les résultats seront alors présentés par les publications obtenues, suivies d’une brève discussion pour chaque publication. Une discussion plus globale sous forme de perspectives de recherche sera enfin développée.
! )!
ETAT DES CONNAISSANCES
! *! 1. LE DEVELOPPEMENT HYPOPHYSAIRE MURIN
Le développement hypophysaire chez l’homme est proche du développement hypophysaire murin. Il est donc admis que les modèles murins représentent un bon modèle pour étudier le développement hypophysaire et comprendre la pathologie des hypopituitarismes congénitaux chez l’Homme (25; 95). Le développement de la poche de Rathke chez la souris est complexe, basé sur une régulation fine de facteurs de transcription et voies de signalisation, aboutissant à la formation d’une hypophyse différenciée.
a. De la placode hypophysaire à la poche de Rathke L’ontogenèse hypophysaire débute très tôt au cours de la neurogenèse cérébrale. Au 7ème jour embryonnaire (e7,5) est visualisée la placode hypophysaire. Une placode est un feuillet ectodermique qui va s’individualiser au contact du neurectoderme au niveau du tube neural antérieur. Dès e7,5, la dualité hypophysaire est ainsi déjà présente : ectoderme oral qui donnera naissance aux lobes antérieur et intermédiaire, en contact avec le neurectoderme qui donnera naissance au lobe postérieur hypophysaire. La placode hypophysaire se développe en partie médiane du tube neural antérieur. Elle migre progressivement au-dessus de la cavité orale suite au processus de neurulation. Au 9 e jour du développement embryonnaire (e9), sous le contrôle de molécules de signalisation morphogènes sécrétées par l’infundibulum diencéphalique, (Bone Morphogenetic Protein 4 (Bmp4) et Fibroblast Growth Factor 8 (Fgf8)), une invagination dorsale du toit du stomodeum se produit formant une poche de Rathke primitive. L’invagination se poursuit jusqu'à séparation de la poche de Rathke de l’ectoderme oral formant ainsi la poche de Rathke définitive (e10,5). En parallèle, l’infundibulum se développe à partir d’une région du diencéphale, pour donner progressivement naissance au lobe postérieur et à la tige pituitaire. Il existe ainsi des interactions permanentes entre ectoderme oral et neurectoderme, puis poche de Rathke et infundibulum : ces interactions sont indispensables au développement des 2 structures. La poche de Rathke définitive est observée à e11,5 (167).
b. De la poche de Rathke à l’hypophyse différenciée Au sein de la poche de Rathke primordiale puis définitive, des progéniteurs principalement situés autour de la lumière (entre les lobes antérieur et intermédiaire) vont progressivement migrer dans l’hypophyse en développement et se différencier sous l’influence de plusieurs facteurs (Sox2, Sox9, nestine, Isl1) (55; 72; 64). Le rôle précis de ces facteurs et les premières étapes des mécanismes de différenciation restent à l’heure actuelle sujets de controverse. A e11,5 le premier marqueur de différenciation, la sous- unité alpha, apparait dans les cellules de l’aileron rostral (qui va être à l’origine de cellules thyréotropes non fonctionnelles après la naissance chez la souris) (90). Les marqueurs des différentes lignées cellulaires hypophysaires sont détectés progressivement : Acth à e12,5, Tshß à e14,5, Pomc dans le lobe intermédiaire à e14,5, Gh et prolactine à e15,5 (104). Les derniers marqueurs à apparaître sont la Lhß à e16,5, et Fshß à e17,5. Là encore, les mécanismes précis de cette migration, et les facteurs déclenchants de différenciation sont imparfaitement compris. Au vu des données portant sur les réseaux fonctionnels de cellules à prolactine et Gh dans l’hypophyse, il est vraisemblable que des mécanismes de régulation précis entrent en jeu au cours du développement hypophysaire pour permettre à ces cellules de communiquer entre elles. Cette différenciation et cette régulation temporo-spatiale font appel à
! "+! - différentes voies de signalisation morphogènes (Bmp4, Bmp2, Fgf8, 10 et 18, que nous ne détaillerons pas dans cet exposé, et la voie Wnt/ß-caténine), interagissant avec des facteurs de transcription - différents facteurs de transcription, d’expression très précoce impliqués dans la morphogenèse de la poche de Rathke (Isl1, Lhx3 et Lhx4, Pitx1 et Pitx2, Hesx1), ou plus tardive, impliqués dans la différenciation hypophysaire (principalement Prop1 et Pou1f1). C’est le gradient d’expression temporo-spatiale de ces différents acteurs au sein de la poche de Rathke qui permettra son développement et sa morphologie finale.
c. La voie Wnt/ß-caténine Les facteurs de croissance de la famille Wnt sont impliqués dans les phénomènes de prolifération, différenciation, polarisation et contrôle des mouvements cellulaires au cours de l’embryogenèse (86). Les membres de cette famille (au moins 18 chez les mammifères) vont se fixer sur un complexe formé d’un récepteur à 7 domaines transmembranaires, dénommé Frizzled, et des protéines de type LRP (LDL-related receptor). Trois voies de signalisation différentes peuvent être activées (85) :
! La voie Wnt/ß-caténine ou voie canonique aboutit à la stabilisation d’une protéine intracellulaire multifonctionnelle dénommée la ß-caténine. En absence de stimulation, la ß-caténine est phosphorylée, ce qui entraîne sa dégradation par le protéasome. Lors de la fixation du ligand Wnt au complexe Frizzled/LRP, la β- caténine stabilisée peut passer la barrière nucléaire et interagir avec d’autres protéines dont des facteurs de transcription (en particulier TCF/LEF) pour activer la transcription de gènes cibles (99). Un exemple sera détaillé dans la suite de cet exposé, portant sur l’interaction avec Prop1 et Hesx1.
! La voie de polarisation cellulaire induit une activation de 2 enzymes, RhoA et Cdc42, par le facteur Dsh. Ces 2 enzymes vont phosphoryler des protéines qui agissent sur le cytosquelette et activer la transcription de gènes cibles.
! La voie calcium dépendante active la voie des phospho-inositols. L’augmentation de la concentration de calcium intra cytoplasmique active 3 enzymes, la calcineurine et 2 kinases (Protein kinase C et Calmododulin Dependant Kinase II), induisant l’activation du facteur de transcription NF-AT.
Dans la poche de Rathke, Wnt4 est exprimé à partir de e9,5. Les souris présentant une inactivation homozygote de Wnt4 ( Wnt4 -/-) présentent une hypoplasie hypophysaire et une diminution de l’expression de la sous-unité alpha, ou une diminution de l’expression de Pou1f1 au sein de cette hypophyse hypoplasique (les résultats des 2 études publiées sur ce point sont contradictoires) (207; 152). Tcf4, dont l’expression est stimulée par Wnt, est également exprimé dans la poche de Rathke : de façon surprenante, son inactivation induit une hyperplasie hypophysaire, soulignant que les mécanismes d’interaction sont complexes et impliquent vraisemblablement d’autres acteurs (19; 20).
d. Les facteurs de transcription hypophysaires Les facteurs de transcription sont des modulateurs de la transcription qui vont agir avec la région promotrice de transcription après une fixation à distance sur l’ADN. Ces facteurs de transcription présentent une grande conservation entre les espèces et ont plusieurs caractéristiques : ! un site de liaison spécifique à l’ADN : l’ homéodomaine.
! ""! ;"%
'()#8% '+!% (3!*% !"#$% '()*'()*%% !"#&% ,)#8% "03#*% '+,'*%%% )3"% ;-)-8%
!": 93"%
)6#*<%
-.)"%
+-)"1023%+ -5)0+(,+%!,60%- /(990+05)(-)0/%/ '!-.,/0% ',4."%' '()4()-+7%' .0!!%.
3=>?@A%BC@DECF?%GHB%IA=JHKLB%GH%JLAMB=LCDNOM%C@DECPK?B%GAMB%EH%G?QHEODDH@HMJ%>RDOD>RBACLH% ! un site d’activation du promoteur : le domaine de transactivation ! des signaux de localisation nucléaire (Nuclear Localisation Site (NLS)) ! des domaines de liaison permettant des interactions moléculaires indépendantes de leur liaison à l’ADN ! une fixation fréquemment dimérique sous forme d’homo ou hétérodimère
Nous nous intéresserons plus particulièrement à deux familles de facteurs de transcription impliqués dans le développement hypophysaire , la famille des facteurs de transcription à homéodomaine de type PAIRED: nous détaillerons particulièrement Prop1, Hesx1, Pitx2, et à un degré moindre Pitx1 et Otx2 , la famille des facteurs de transcription à homéodomaine de type LIM: Lhx3, Lhx4 et Isl1.
2. LES FACTEURS DE TRANSCRIPTION A HOMEODOMAINE DE TYPE PAIRED
Les facteurs de transcription à homéodomaine paired reconnaissent une séquence palindromique TAAT__ATTA séparée par 2 à 3 paires de bases. La fixation sur ce domaine de liaison peut être homo ou hétérodimèrique. Ainsi, Prop1 et Hesx1 sont capables de se fixer sous formes d’homodimères, ou d’hétérodimères Prop1-Hesx1 (92; 42; 155; 133). Il faut noter que cette séquence est une séquence théorique, et plusieurs études récentes ont souligné que cette séquence pouvait subir des modifications et être toujours efficace en termes de liaison de PROP1 (92; 88; 133). Nous détaillerons successivement Hesx1, Prop1, Pitx1, Pitx2 et Otx2 et leurs rôles respectifs dans le développement hypophysaire murin, et l’hypopituitarisme congénital chez l’homme.
a. Au cours du développement hypophysaire murin
i. Hesx1 L’expression de Hesx1 est très précoce puisque déjà retrouvée dans la placode hypophysaire. Elle est ensuite restreinte à la poche de Rathke à partir de e8.5-e9, puis décroît à partir de e13 pour devenir indétectable à e14-15,5. A e13, l’atténuation de l’expression d’Hesx1 est nécessaire au développement des lignées hypophysaires dépendant de Prop1, et donc à la poursuite de la différenciation hypophysaire (226; 206; 47; 121). Le modèle actuellement admis est basé sur l’interaction de Hesx1 et Prop1 via la formation d’un hétérodimère sur l’enhancer proximal de Pou1f1 (42; 140). Hesx1 a ainsi un rôle principalement répresseur : A e11, l’interaction du corépresseur Tle1 (transducing like enhancer of split-1, orthologue de la protéine répressive Groucho) avec Hesx1 (via son domaine eh1) participe à l’inhibition de Prop1 (140). L’action répressive d’Hesx1 est également facilitée par la fixation d’un autre corépresseur NCoR au niveau de l’homéodomaine. A e13, c’est l’interaction de Prop1 avec la ß-caténine qui va permettre l’inhibition de Hesx1, l’expression de Pou1f1 et la poursuite du développement hypophysaire, et plus particulièrement du processus de différenciation des lignées somato-lactotropes et thyréotropes (140). Ces mécanismes seront détaillés dans le premier article présenté dans cette thèse. Les souris avec invalidation homozygote de Hesx1 (Hesx1 -/-) présentent un phénotype variable, le plus souvent sévère, avec des anomalies majeures de l’encéphale et de la ligne médiane (absence de corps calleux ou de septum pellucidum, posthypophyse ectopique) et des malformations ophtalmiques (microphtalmie, anophtalmie) (42; 7). Une anomalie de la morphologie hypophysaire est également souvent retrouvée :
! "#! l’hypophyse peut être aplasique ou hypoplasique, ou divisée en de multiples ilôts. Les souris Hesx1 -/- meurent entre e10.5 et e12.5. A l’inverse, l’absence d’inhibition d’ Hesx1 après e12,5 entraîne un phénotype identique à celui de souris présentant une inactivation homozygote de Prop1 (42). Cela est vraisemblablement dû au maintien de l’action répressive de Hesx1 sur Prop1. Deux modèles murins de mutation homozygote de Hesx1 ( R160C et I26T ) ont été rapportés (15; 173). Le modèle porteur de la mutation R160C à l’état homozygote est proche du modèle porteur de l’invalidation homozygote de Hesx1 : il existe de sévères anomalies ophtalmiques et télencéphaliques ; l’hypophyse dysmorphique peut être localisée de façon ectopique au niveau du toit de la cavité naso-pharyngée, ou formée de multiples ilots individualisés pouvant faire évoquer plusieurs petites poches de Rathke primitives. Dans ce modèle, l’hypophyse devient rapidement hypoplasique ou aplasique après la naissance. Le modèle homozygote I26T est moins sévère : les anomalies ophtalmiques et hypophysaires sont identiques, mais les souris ne présentent pas d’anomalies encéphaliques. La mutation I26T génére une protéine incapable d’interagir avec le corépresseur TLE1, suggérant le rôle majeur de cette interaction lors du développement hypophysaire (173), mais aussi un phénotype différent selon le domaine fonctionnel d’Hesx1 atteint. L’expression de Hesx1 est également sous le contrôle de facteurs de transcription de type LIM comme Lhx1 et Lhx3 (33). Une interaction avec Six3, un autre facteur de transcription, pourrait également être nécessaire au cours du développement hypophysaire. Les souris double hétérozygotes Hesx1 Cre/+ (« équivalent » de Hesx1 +/-) Six3 +/- présentent en effet le même phénotype hypophysaire que les souris avec inactivation homozygote de Hesx1 (65).
ii. Prop1 L’expression de Prop1 est exclusivement hypophysaire. Prop1 est exprimé à partir de e10 dans la poche de Rathke. Son expression augmente jusqu’à e12, puis diminue progressivement pour disparaître à e15,5 (191). Une étude récente basée sur le développement d’un anticorps anti-Prop1 chez le rat a cependant retrouvé une expression de Prop1 persistant à e16,5, puis très réduite à e18,5 et au premier jour après la naissance (234). L’expression de Prop1 précède et est nécessaire à celle d’un autre facteur de transcription hypophysaire, Pit-1 (Pou1f1) , lui-même impliqué dans la différenciation des lignées somato-lactotropes et thyréotropes (88). Cependant, peu de cellules co-expriment Prop1 et Pou1f1, suggérant que Prop1 est nécessaire principalement lors de l’activation de Pou1f1 (aucune cellule Prop1+ n’exprimait d’hormones hypophysaires). Une autre possibilité est l’existence d’autres facteurs intermédiaires entre Prop1 et Pou1f1. Cette hypothèse est suggérée par les modèles murins qui ne retrouvent pas une parfaite symétrie entre les zones d’expression de Prop1 et de Pou1f1 au cours de l’embryogenèse ; de plus le début d’activation de Pou1f1 est retardé d’environ 2 jours par rapport au pic d’expression de Prop1 (48). Le délai entre le pic d’expression de Prop1 (e12,5) et l’activation de Pou1f1 (e13,5-e14,5) suggère également que d’autres facteurs ou étapes sont présents entre Prop1 et pou1f1. Le caractère transitoire de l’expression de Prop1 est important, puisque le maintien de son expression chez des souris conduit à un retard de maturation gonadotrope (un modèle murin surexprimant Prop1 sous contrôle du promoteur de la sous-unité alpha présente un retard d’expression de Fshß) et semble favoriser la genèse de tumeurs hypophysaires (41; 215). De même, une expression précoce de Prop1 (à partir de e9.0 dans un modèle de souris transgénique exprimant Prop1 sous contrôle d’un promoteur Pitx1) entraine une absence d’hypophyse (42) . Enfin, comme expliqué précédemment, il existe un phénomène complexe d’interactions entre Prop1 et
! "$! Hesx1 nécessaires à la différenciation hypophysaire. Ce modèle sera détaillé dans le premier article présenté dans cette thèse. Yoshida et al. ont observé une co-expression de Sox2 dans la majorité des cellules exprimant Prop1 à e12,5 et e13,5, et dans une moindre proportion à e16,5 et P5. Sox2 étant impliqué dans la genèse et le développement des cellules souches hypophysaires, il est vraisemblable que Prop1 joue un rôle dans la transition prolifération/différentiation des cellules souches (55; 64; 234). Notch pourrait jouer un rôle dans le maintien d’un état indifférencié des progéniteurs (en évitant leur différenciation corticotrope), et en permettant leur différentiation plus tardive en lignées Pou1f1 dépendantes (237; 100). Les modèles murins d’inactivation de Prop1 ont permis de clarifier son rôle au cours du développement hypophysaire. Les souris de type Ames présentent une mutation homozygote spontanée (S83P ) dans la zone codant pour l’homéodomaine de Prop1 : le phénotype est identique à celui présenté par des souris Prop1-/-. Il existe un déficit somato-lactotrope, thyréotrope et gonadotrope (215); la vascularisation hypophysaire est anormale (224); l’hypophyse présente une hypoplasie sévère chez la souris adulte (158; 134), et une pseudo-hyperplasie (en fait un élargissement de la lumière) au cours de l’embryogenèse (223). Le déficit gonadotrope présenté par les souris est également surprenant : il pourrait être exacerbé ou démasqué par le déficit en GH et/ou TSH ; en effet, une supplémentation en T4/GH permet d’obtenir une fertilité chez la plupart des mâles (9). Un autre facteur de transcription SF1, surexprimé en absence de Prop1, et impliqué dans la différenciation gonadotrope physiologique, pourrait également jouer un rôle dans ce déficit gonadotrope (223). Il est intéressant de noter qu’un déficit corticotrope n’a jamais été observé chez les souris Ames. Enfin, les souris Ames présentent vraisemblablement une anomalie de la migration des progéniteurs qui sont bloqués dans le lobe intermédiaire. Le volume des lobes hypophysaires est normal au cours de l’embryogenèse, et l’apoptose de ces progéniteurs après la naissance entraîne une hypoplasie secondaire. Cette anomalie de migration serait liée à une absence d’expression de Notch2 par ces progéniteurs (157). La sortie du cycle cellulaire (permettant la différenciation), liée à une diminution d’expression de la cycline D2, s’effectuerait alors que les progéniteurs seraient toujours dans le lobe intermédiaire (223). Une autre étude a rapporté au contraire que Notch2 était nécessaire au maintien de l’expression de Prop1 (100). Les mécanismes exacts liés à l’hyperplasie hypophysaire restent donc incompris.
iii. Pitx1 Pitx1 est un facteur de transcription de type paired à homéodomaine bicoïde (105). Sa structure et sa séquence sont très proches de celles de Pitx2 (52; 208). Pitx1 joue vraisemblablement un rôle précoce dans le développement hypophysaire (106; 195). Son expression est initialement retrouvée dans l’ectoderme oral à e8,0, puis dans la poche de Rathke à partir de e9,5. Son expression persiste jusqu’à l’âge adulte, où elle prédomine au niveau hypophysaire dans les cellules thyréotropes et gonadotropes, et à un degré moindre dans les cellules corticotropes (107). L’inactivation homozygote de Pitx1 ( pitx1 -/-) entraine une mort fœtale tardive ou néonatale précoce. De façon intéressante cependant, la morphologie hypophysaire est normale, avec une différenciation habituelle des 5 lignées cellulaires, mais une diminution notable des lignées thyréotrope et gonadotrope (29).
iv. Pitx2 L’expression de Pitx2 est retrouvée au niveau du stomodeum à partir de e8, puis de la poche de Rathke (qui dérive du stomodeum) à partir de e10,5 (112; 82). A e12,5, Pitx2 est exprimé dans les lobes antérieur et intermédiaire, mais pas dans le lobe postérieur
! "%! L’ARN de Pitx2 a été isolé dans l’hypophyse adulte et dans des lignées cellulaires hypophysaires pré-thyréotropes, pré-gonadotropes et somatotropes. A l’âge adulte, la majorité (80 à 90%) des cellules thyréotropes et gonadotropes expriment Pitx2, au contraire des cellules somatotropes, lactotrope et corticotropes (29). Cependant, alors qu’une forte expression d’ARN de Pitx2 est retrouvée au sein des cellules gonadotropes, il existe au contraire une faible expression protéique de Pitx2, suggérant un mécanisme de régulation post-transcriptionnel (208). Pitx2 est aussi exprimé dans le cerveau adulte, l’œil, le rein, les poumons, les testicules et la langue (61; 82). Les souris présentant une inactivation homozygote de Pitx2 ( Pitx2 -/-) (61; 63; 112) présentent une forte mortalité précoce (plus de 35% de mortalité avant e10, 100% de mortalité à e14,5) : on observe des anomalies de la position du cœur et de la différenciation pulmonaire, et une anomalie de fermeture de la paroi thoraco-abdominale avec extériorisation des organes à e12. L’hypophyse est hypoplasique dès e10. Cette hypoplasie est vraisemblablement liée à une augmentation de l’apoptose, alors que l’index de prolifération reste identique (29). Le développement hypophysaire est arrêté à partir de e12,5 avec une absence d’expression de Hesx1, Pit1, Tshß, Lhx4, et un très faible niveau d’expression de Gata2 et Prop1. L’expression de alpha-GSU est cependant retrouvée à e11,5, et les niveaux d’expression de Lhx3 sont inchangés (112). Pour étudier les effets de Pitx2 sur l’ensemble des processus de développement hypophysaire, Gage et al. ont développé un allèle hypomorphe de Pitx2 ( Pitx2 neo , avec une cassette de résistance à la néomycine inséré au sein de l’intron 3) (63; 192). L’hypophyse des souris Pitx2 neo/neo semble de morphologie comparable à celle des souris sauvages. En termes de volume hypophysaire, les souris Pitx2 neo/- ont un phénotype intermédiaire (en comparaison avec les souris Pitx2 -/- et les souris Pitx2 neo/neo ). Cependant, il semble que la différenciation cellulaire s’effectue différemment : au 1 er jour post-natal, l’expression de Lhß et Fshß est quasi absente dans les souris Pitx2 neo/neo , tout comme l’expression du récepteur de la GnRH. Le nombre de cellules somatotropes et thyréotropes est modérément réduit en comparaison avec les souris sauvages. Le niveau d’expression de POMC est inchangé. L’expression de certains facteurs de transcription est également modifiée dans les souris Pitx2 neo/neo : ainsi Gata2 est exprimé normalement à e12,5 mais n’est pas retrouvé à P1. De même, Egr1 et Sf1 (impliqués dans la différenciation gonadotrope) sont très faiblement exprimés à P1. Pit1 est également diminué, au contraire de Prop1 (63; 192). A l’inverse, la surexpression de Pitx2 ne modifie pas la morphologie hypophysaire à e18,5 ; par contre, une augmentation de l’expression de Lhß et Fshß est observée, avec une extension dorsale du domaine des cellules gonadotropes. Aucune modification des cellules thyréotropes, somatotropes ou corticotropes n’a été observée. L’augmentation d’expression de Sf1 (marqueur de cellules gonadotropes) ne se traduit pas par une co- localisation avec Pit1, excluant un changement de différenciation d’un type cellulaire (en l’occurrence de cellules Pit1+ en cellules gonadotropes) (63; 192). Les études in vitro indiquent que Pitx2 joue un rôle dans l’activation de la plupart des promoteurs des gènes codant pour les hormones hypophysaires. Pitx2 code pour 3 isoformes, PITX2a, PITX2b et PITX2c (61). D’autres isoformes ont été récemment décrites (Pitx2b2, Pitx2Cß) sans que leurs rôles in vivo aient été identifiés (102). Ces isoformes pourraient avoir des activités différentes(39). Pitx2 est capable d’activer les promoteurs de Gata2, Lhß, Fshß, Tshß, Gh, Prolactine et POMc (52; 153; 154). Il peut interagir de façon synergique avec Lhx3 (promoteur de la sous unité alpha (112)), Pit1 (promoteurs de la prolactine, Gh et Tshß ), NeuroD1 (promoteur POMc ), Sf1 et Egr-1 (promoteur Lhß (208; 183)), et smad3 (promoteur Fshß (194; 193)). Pitx2 est également capable de contrôler des gènes codant pour des protéines qui régulent le cycle cellulaire, comme la cycline D1 et la cycline D2.
! "&! Enfin, la voie Wnt/ß-caténine joue également un rôle pivot dans l’activation et la régulation de l’expression de Pitx2, qui joue également sur les composants de cette voie : dans l’œil, DKK2, antagoniste de la voie Wnt, est une cible de Pitx2 (62) ; l’action synergique de Pitx2 et de la ß-caténine active le promoteur de Lef1, qui est lui-même capable d’interagir avec l’homéodomaine de Pitx2 (99; 213; 4; 5). Enfin, la voie Wnt stimule l’effet activateur de Pitx2 sur la prolifération, en particulier au cours du développement hypophysaire (99).
v. Otx2 Otx2 est exprimé très précocement et joue un rôle majeur dans le développement de l’encéphale. L’inactivation homozygote de Otx2 (Otx2 -/-) entraîne des malformations sévères des structures antérieures. L’inactivation hétérozygote de Otx2 (Otx2 +/-) conduit à des phénotypes hypophysaires variables, allant d’une dysmorphie à une aplasie hypophysaire (2). Otx2 est retrouvé dans le diencéphale ventral de e10,5 à e14,5, où il pourrait interagir avec Hesx1 (Mortensen et Camper, données non publiées). Otx2 pourrait également avoir un rôle dans le développement des neurones à GnRH : les études fonctionnelles in vitro ont ainsi démontré que Otx2 était capable de lier et activer le promoteur GnRH (108). Dans cette optique, il pourrait interagir avec un membre de la famille des protéines Groucho, Grg4. Une expression d’Otx2 est retrouvée au sein de la poche de Rathke à e10,5, et n’est pas retrouvée à partir de e12,5. Otx2 est également réprimé par un autre co-répresseur, Tle4, qui est exprimé au cours du développement hypophysaire (Mortensen et Camper, données non publiées).
b. En pathologie humaine
i. HESX1 Treize mutations d’ HESX1 responsables d’hypopituitarisme congénital ont été rapportées à ce jour (47; 15; 205; 129; 24; 36; 197; 190; 189; 40; 38) (table 1). Les mutations homozygotes (5/13) ont une pénétrance complète, et un phénotype généralement plus sévère que les mutations hétérozygotes (8/13). La plupart des mutations sont à l’origine d’une diminution de liaison sur la séquence ADN avec diminution d’activité sur les promoteurs cibles de HESX1, à l’exception d’une mutation qui entraîne une augmentation de la liaison à l’ADN, et vraisemblablement une répression complète de PROP1 (36). Le phénotype hypophysaire est très variable : le déficit somatotrope est constant, les autres déficits sont présents dans environ 50% des cas. L’hypophyse est hypoplasique dans 11/13 cas (et normale dans les 2 autres mutations). La post-hypophyse est ectopique dans plus de 50% des cas, non visualisable dans 10% des cas, et normalement localisée et visualisée dans 40% des cas. Une anomalie des nerfs optiques est observée dans 30% des cas. Des anomalies cérébrales sont rapportées dans de rares cas : agénésie du corps calleux dans 3/13 cas, corps calleux fin dans 1 cas, hydrocéphalie dans 1 cas, selle turcique peu développée dans 1 cas. En termes de mécanismes physiopathologiques, Carvalho et al. ont rapporté la première et seule mutation homozygote d’ HESX1 ( I26T ) responsable d’hypopituitarisme par une anomalie d’interaction avec le co-répresseur TLE1 (24). La patiente présentait un hypopituitarisme avec hypoplasie hypophysaire, tige pituitaire fine, post-hypophyse ectopique, mais sans d’anomalie cérébrale ou des nerfs optiques. Gat- Yablonski et al. ont récemment corrélé l’existence d’une variation allélique hétérozygote de Hesx1 (Asn125Ser ) à un phénotype d’hypopituitarisme congénital chez 3 patients présentant une hypoplasie hypophysaire avec post-hypophyse ectopique (2 patients sur
! "'! D# !"# ED# EG# HE# DFI# Mutation c.306_307 p.Q117P c.357 Alu insert Alu insert p.E149K c.449_450 p.R160C p.K176T g.1684 insAG +2T>C Frère delCA delG GH D D D D D D D D D D TSH D D D D D N D D D N ACTH D D D D D N D D D N LH FSH D D N D D N N D N N Hypoph. Hypo Hypo Hypo Hypo Hypo Hypo Hypo Hypo N Hypo Lobe post. Ect Ect. N N N Ect. N Ect. Ect. Abs Nerfs Hypo N N Colobome N N N Hypo N Hypo optiques nerf optique droit Corps N N N N N N Epaissi Agénésie N Agénésie calleux
Autres Hypoplasie Coarctation Doigts HydroC de la selle aortique surnum. Chiari I Encephalo- Hernie malacie diaphragm. Transm. Dom Dom Rec Rec Rec Dom Rec Rec Dom Dom Tajima, Coya, Sobrier Sobrier, Sobrier, McNay, Sobrier, Dattani, Coya, Cohen, JCEM, 2003 JPEM, JCEM, HumMut, HumMut, JCEM, JCEM, 2006 NatGenet, JPEM, JCEM, 197 2007 2006 2005 2005 2007 189 1998 2007 2003 40 189 190 190 125 47 40 36 $%&'(#)*#+,-%./01#2(#!345)#6%77/6-8(1#9#:(#;/,6<#=(7681(0-%./0#1:>8?%.@,(#2(1#?,-%./01#A!"B#>/?8/2/?%C0(D# ' )(#4%$%4( '(9#'7# )8#97##$'"!8#'A97#!(#%0#)(#%$')2'(##!#"2)0$#(#3 #)#2##%#$)4%#3' !@#C4%$%@7#3$!2"# 4%$%4( '8#C4%$7#4%$%!( 8#97#3$!2"##$'"!@#D$#%$()@7#!$#%$()' 2'8#E)7#)$% &28#97#!$! (0$###$'"!@#9'(#$%0&2(9#97# #$'"!8#C4%$7#4%$%!( @#H2)'(9#C4%$%!#(!!7#4%$%!( #(!! '8#C4'$I7#4'$%! @#'#("9#'$"7#$" ##)8#K7#'(( @# D# !"# ED# EG# HE# DFI#
Mutation p.Q6H p.Q6H p.I26T p.S170L p.S170L p.T181A Cas index Frère GH D D D D D D TSH D D D N N N ACTH N D D N N N LH FSH N D D N N N Hypoph. Hypo Hypo Hypo N N Hypo Lobe post. Ect. Ect Ect. Ect. Ect. Abs Nerfs optiques N N N Hypo N N
Corps calleux N N N N N N
Autres Mandibular hypoplasia, Nose and digits anomalies Transm. Dom Rec Dom Dom Dom Thomas, Corneli, JEI, 2008 Carvalho, Thomas, Thomas, Thomas, HumMolGen, 2001 38 JCI, 2003 HumMolGen, HumMolGen, HumMolGen, 205 24 2001 2001 2001 205 205 205 $%&'(#)*#+,-%./01#2(#!345)#6%77/6-8(1#9#:(#;/,6<#=(7681(0-%./0#1:>8?%.@,(#2(1#?,-%./01#A!"B#>/?8/2/?%C0(D# * )(#4%$%4( '(9#*7# )8#:7##$'"!8#*A:7#!(#%0#)(#%$')2'(##!#"2)0$#(#3 #)#2##%#$)4%#3' !@#D4%$%@7#3$!2"# 4%$%4( '8#D4%$7#4%$%!( 8#:7#3$!2"##$'"!@#E$#%$()@7#!$#%$()' 2'8#F)7#)$% &28#:7#!$! (0$###$'"!@#:'(#$%0&2(9#:7# #$'"!8#D4%$7#4%$%!( @#I2)'(9#D4%$%!#(!!7#4%$%!( #(!! '8#D4'$J7#4'$%! @#'#("9#*$"7#$" ##)8#L7#'(( @# 3), un déficit en GH (avec croissance normale sans traitement ?), TSH (2 patients sur 3), ACTH et LH/FSH, et des éléments dysmorphiques (doigt surnuméraire, oreilles larges, anomalies de la dentition…). Aucune étude fonctionnelle n’a été effectuée. Ce « polymorphisme » est retrouvé à une fréquence importante dans la population sub- saharienne, ce qui rend peu probable l’implication exclusive de ce polymorphisme dans le phénotype. Seuls les gènes codant pour POU1F1, PROP1 et LHX4 ont été séquencés par ailleurs. L’hypothèse des auteurs est que la population sub-saharienne disposerait d’un allèle protecteur sur un des co-facteurs de HESX1 permettant d’éviter la survenue du phénotype déficitaire. Cette hypothèse est plausible, mais un séquençage des gènes d’autres facteurs de transcription impliqués dans les hypopituitarismes congénitaux devra être effectué en première intention (67). A noter cependant que le phénotype de déficit en GH et doigt surnuméraire a déjà été rapporté pour la mutation hétérozygote E149K de HESX1 (125). Une expression persistante de HESX1 a été observée chez l’homme à l’âge adulte sans que son rôle précis ne puisse être déterminé (120). HESX1 pourrait être nécessaire dans les phases de différenciation de progéniteurs en cas de lésion hypophysaire (235). Le rôle des mutations de HESX1 dans la dysplasie septo-optique est toujours discuté (164; 125; 94). L’incidence de cette anomalie est faible, estimée à 1/10000 naissances. Le diagnostic de dysplasie septo-optique est clinique, basé sur la présence d’au moins 2 des 3 critères précédents : hypoplasie des nerfs optiques, hypopituitarisme (le plus fréquent étant un déficit en GH), anomalies de la ligne médiane (agénésie du septum pellucidum ou du corps calleux, par exemple) (93). La majorité des patients porteurs de mutations de HESX1 présente donc une dysplasie septo-optique selon cette définition (225). HESX1 n’est pas le seul facteur de transcription impliqué dans la dysplasie septo- optique : dix mutations de SOX2 ont également été associées à des anomalies ophtalmiques sévères (anophtalmie, microphtalmie), des anomalies du corps calleux et une hypoplasie hypophysaire (225). Cependant, un facteur génétique (incluant les mutations de HESX1 et SOX2 ) n’a été identifié que dans environ 1% des cas de dysplasie septo-optique (225). Il est donc vraisemblable que d’autres facteurs (génétiques, et environnementaux par exemple) jouant un rôle précoce au cours du développement, sont impliqués dans la genèse de la dysplasie septo-optique.
ii. PROP1 A ce jour, 24 mutations distinctes de PROP1 responsables d’hypopituitarisme congénital ont été rapportées dans la littérature (34; 53; 58; 229; 50; 169; 3; 141; 148; 214; 8; 144; 218; 14; 163; 203; 220; 109; 162; 1; 110; 138; 217; 238; 96) (table 2). Les mutations de PROP1 représentent la plus fréquente des étiologies identifiées d’hypopituitarisme congénital (76; 216; 95). Toutes ces mutations sont de transmission récessive (parents porteurs sains hétérozygotes). Le phénotype hypophysaire classique associe un déficit somato-lactotrope, thyréotrope et gonadotrope. L’âge de survenue est variable (57). Le déficit gonadotrope, bien que constamment présent, est de présentation variable : il peut être diagnostiqué dès la naissance (micropénis et testicules non descendus) ou se révéler par un retard pubertaire, soulignant que PROP1 joue un rôle dans l’espèce humaine dans le développement et le maintien des cellules gonadotropes. La présentation est parfois plus atypique : ainsi, dans le cas d’une mutation concernant le domaine de transactivation (W194X ) rapportée par notre équipe (162), le phénotype bien que présentant l’habituel déficit somato-thyréo-gonadotrope, était particulier par l’ordre de survenue de ces déficits; 2 des 3 patients ont en effet présenté un déficit gonadotrope avant de présenter un déficit somatotrope. Cette mutation est une des 2 seules publiées concernant le domaine de transactivation. De façon surprenante, l’autre mutation située
! "(! !"#!!"" !"# $%$# Mutation c.2T>C c.109+1G>T c.112_124 c.C149_150 c.150delA c.157delA p.R71C* p.R71H* p.R73C p.R73H del13 delAG* GH D D D D D D D D D D TSH D D D D D D D D D D ACTH D D N/D - D/N N N N D/N N LH FSH D D D D D D - - D D Familial/ F F F F S F F F F/S F/S Sporadique Hypophyse Hypo Hyper/Hypo Hyper/ - Hyper/Hypo Hypo Hypo Hypo Hyper/ Hypo Hypo Hypo Transm. Rec Rec Rec Rec Rec Rec Rec Rec Rec Rec Lemos, Bottner, Agarwal, Fofanova, Riepe, Tatsumi, Paracchni, Paracchni, Dela- Vallette, ClinEnd JCEM, 2004 JCEM, JCEM, JCEM, ClinEndo, ClinGen, ClinGen, doey, JCEM, 2006 14 2000 1998 2001 2004 2003 2003 JCEM, 2001 110 3 58 166 203 144 144 1999 214 50 Duque- snoy, FEBS, 1998 53 $%&'(#)*#+,-%./01#2(#34536#7%88/7-9(1#:#;(#9?%.@,(#2(1#?,-%./01#A!"B#>/?9/2/?%C0(D# &'()*+,#-./0/-.,1*23,4#&5#6'()*+7#85#902:1;7#&<85#;3,#/1=39+,#/02+3>2,#63#;1#:>+1=09,#1?1*39+#>9#/-'90+./3#?12*1@;3A#B1:*;*1;
Mutation p.F117I p.R120C p.R125W* c.467insT p.W194X c.629delC* Complete deletion GH D D D D D D D TSH N N D D D/N D D ACTH D D N/D N N D/N N LH FSH F F D D D D D Familial/ F F F F F S F Sporadique Hypophyse Hypo Hypo Hypo Hypo Hypo Hypo Hyper/Hypo Transm. Rec Rec Rec Rec Rec Rec Rec Wu, NatGenet, Fluck, JCEM, Kelbermann, Nose, JPEM, Reynaud, Reynaud, Abrao, 1998 1998 ClinEndo, 2006 JCEM, 2005 JCEM, 2006 ClinEndo, 229 57 2009 138 162 164 2006 96 1 !"#&1(&)$!"" &&12310&"! &8&&$)&& !('(3$#$3'-&'9(!7((8(47("#&!- 8(!B47( '($-)"('($#&(1&'(( -(!1(-)#"'(-2-"((1"($"#(3$(2-&- @(>-! - B?$#&-%19(>7( -! - 8(?7('$#&-%1@(B3$#$3'7(2# 1!(3$#$3'-&9(B3$#7(3$#$ -'8(3$&7(3$&$ -'@(C&-"'!7((&-"'!''#"8(D7(&''2@(C7( !1(-)#"(#'&2(-"'( (-&(A1"(#1 ((+#)@( dans le même domaine ( S156InsT ) est à l’origine d’un phénotype classique de déficit hypophysaire(138). En termes d’études fonctionnelles, notre équipe avait observé une différence de liaison sur la séquence consensus Prdq9 pour le mutant W194X (162); ce résultat n’a pas été observé par une autre équipe qui a rapporté une liaison identique à celle observée pour PROP1 sauvage (88). Les auteurs ont également comparé la liaison de l’autre mutant situé dans le domaine de transactivation ( S156InsT ) et n’ont pas observé de différence en termes de liaison à l’ADN. Par contre, de façon surprenante, en utilisant comme séquence de liaison une séquence identifiée sur le promoteur de POU1F1 (séquence proche de Prdq9, bien que différente, TAAT___ATAA), le mutant S156InsT perdait sa capacité de liaison en gel retard (aucune différence n’était par contre observée pour W194X) (182). La différence en termes de résultats de liaison sur la séquence consensus Prdq9 est difficile à expliquer, mais la variation de liaison en fonction de la séquence cible implique la nécessité d’utiliser différentes séquences cibles avant de pouvoir affirmer de façon formelle qu’il n’existe pas de variation de la liaison ADN. Le déficit corticotrope est de façon surprenante (en comparaison avec le modèle murin), présent dans 10/24 mutations : il survient en général de façon retardée, et il est inconstamment présent chez les patients porteurs d’une même mutation, y compris au sein d’une même famille (103; 163). Kelbermann et al. ont ainsi rapporté une mutation double hétérozygote c.310delC/p.R125W à l’origine d’un bilan corticotrope normal et d’une hyperplasie hypophysaire chez un frère, et d’un déficit corticotrope avec hypoplasie hypophysaire chez sa sœur (96). Le déficit corticotrope pouvant survenir de façon retardée (parfois diagnostiqués après 40 ans), il n’est cependant pas possible d’affirmer que les 2 phénotypes seront différents au cours du suivi. Pour ces mêmes raisons, l’évaluation précise de l’incidence du déficit corticotrope chez les patients porteurs de mutations de PROP1 est difficile à déterminer : la plupart des bilans ayant été réalisés lors de l’enfance, avec des bilans de suivi à l’adolescence, il est probable que l’incidence réelle du déficit corticotrope est sous-estimée. L’imagerie hypophysaire peut révéler une hypophyse hyperplasique, d’aspect normal ou hypoplasique. L’étude des mutations de Prop1 dans le modèle murin, ainsi que certaines études de suivi de patients porteurs de mutations de PROP1 suggérent que l’hypoplasie hypophysaire pourrait succéder à une phase initiale d’hyperplasie hypophysaire (127; 59; 166; 219; 217; 221). Chez la souris, cette hyperplasie pourrait être liée à un défaut de migration des progéniteurs à l’origine des cellules hypophysaires différenciées ; ces progéniteurs cesseraient leur migration au sein du lobe intermédiaire, l’hyperplasie de ce lobe intermédiaire étant à l’origine d’un aspect d’hyperplasie hypophysaire globale (81). L’hypoplasie et les déficits hypophysaires seraient liés à une apoptose de ces progéniteurs. Une étude anatomo-pathologique récente basée sur l’analyse de 2 spécimens hypophysaires provenant de patients porteurs de mutations de Prop1 opérés pour hyperplasie hypophysaire semble confirmer cette hypothèse : l’analyse évoque une hyperplasie kystique d’un présumé vestige de lobe intermédiaire apoptotique (239). Cependant, une étude a décrit la survenue d’une hyperplasie « tertiaire » chez 2 patients ayant présenté une hyperplasie puis une hypoplasie, ce qui remet en cause l’hypothèse « hyperplasie puis hypoplasie définitive » (221).
iii. PITX1 Aucune mutation responsable d’hypopituitarisme congénital n’a été rapportée chez l’homme.
iv. PITX2
! ")! Le syndrome d’Axenfeld-Rieger est caractérisé par des anomalies du segment antérieur de l’œil (hypoplasie de l’iris, malposition de la pupille, anomalies cornéennes, anomalies de fermeture de l’angle irido-cornéen) et des anomalies systémiques (dysmorphie cranio-faciale, anomalies dentaires, anomalies morphologiques ombilicales (178; 209). A ce jour, 41 patients présentant un syndrome d’Axenfeld-Rieger lié à une anomalie de PITX2 (mutations ponctuelles, délétions ou translocations) ont été rapportés. La plupart des mutations concernent l’homéodomaine. A noter que, de façon surprenante, le phénotype observé ne présentait pas de différence majeure selon que la mutation entraînait une protéine hypofonctionnelle, et celles à l’origine d’une protéine avec gain de fonction ( p.Trp133X, p.Val83Leu, c.366DelC ) (209). A ce jour, seulement 3 cas ont été rapportés chez des patients porteurs d’un syndrome d’Axenfeld-Rieger associé à une hypoplasie hypophysaire et un déficit en GH (56; 171; 119). La forme de la selle turcique peut également être modifiée chez les patients porteurs d’un syndrome d’Axenfeld Rieger lié à une mutation de PITX2 (selle turcique allongée) (128). Le syndrome d’Axenfeld-Rieger a également été associé à des mutations de FOXC1 (Forkhead box C1), un facteur de transcription qui joue un rôle important au cours de l’embryogenèse (209).
v. OTX2 A ce jour, 19 mutations de OTX2 ont été rapportées. Sept seulement concernent des patients porteurs d’hypopituitarisme (les autres concernent des patients porteurs d’ophtalmopathies sans qu’il ne soit fait mention de déficit hypophysaire associé) (46; 44; 51; 78; 198) (Table 3). Il est difficile de savoir si ces mutations sont toujours associées à un déficit hypophysaire et/ou un déficit ophtalmique car ces caractéristiques ne sont pas systématiquement recherchées dans les études publiées à ce jour. L’étude des caractéristiques cliniques des patients porteurs de ces mutations permet de souligner l’extrême variabilité phénotypique, aussi bien en termes de déficits hypophysaire et cérébral (hypoplasie ou hypophyse de taille normale, post-hypophyse ectopique ou en position physiologique, anomalies cérébrales inconstantes à type de malformation de Chiari). Toutes les mutations sont hétérozygotes, et surviennent dans un contexte de novo (aucun cas familial, mutation non retrouvée chez les parents). La plupart des mutations sont à l’origine d’une protéine tronquée, avec un codon stop prématuré. L’homéodomaine est rarement modifié dans de son intégralité, permettant une liaison sur la séquence cible ADN en études fonctionnelles (gel retard). Les études fonctionnelles retrouvent pour l’ensemble des protéines une diminution de l’activation des promoteurs HESX1, POU1F1 et GnRH1. L’effet est le plus souvent lié à une haplo- insuffisance : une seule mutation a été rapportée comme étant responsable d’un effet dominant négatif.
3. POU1F1 (PIT1), FACTEUR DE TRANSCRIPTION A HOMEODOMAINE POU
a. Au cours du développement hypophysaire murin
Les facteurs de transcription à homéodomaine POU sont caractérisés par 2 domaines très conservés, à domaine POU spécifique, un domaine POU de liaison ADN, et un domaine de transactivation à l’extrémité N-terminale (184). Chez la souris, le gène de Pit1 est localisé sur le chromosome 16 (111). Deux isoformes ont été décrites, une isoforme majoritaire (Pit1) et une seconde isoforme, Pit1ß issue d’un épissage alternatif
! "*! 3-$# !"# !"# $%"# $"$BC# Mutation K74fsX103 G188X Whole gene S136fsX178 S135fsX136 N233S* S138X del GH D D D D D D D TSH N D N D N D N ACTH N N N D N D N LHFSH N D N D N D N Familial/ Sporadique ? Sporadique Sporadique Sporadique Sporadique Sporadique Sporadique Hypophyse Hypoplasie Hypoplasie Hypoplasie Hypoplasie Normale Hypoplasie Normale Lobe Ectopique Ectopique Normal Ectopique Normal Ectopique Normal postérieur Cerebral Chiari malf. Pathologie Oui Oui Oui ? Oui ? Oui oculaire Transmission Dominant Dominant Dominant Dominant Dominant Dominant Dominant Dateki, Dateki, Dateki, Tajima, JCEM, Dateki, JCEM, Diaczok, Henderson, JCEM, 2010 JCEM, 2010 JCEM, 2010 2009 2008 JCEM, 2008 MolVision, 44 44 44 198 44 51 2009 78 $%&'(#)*#+,-%./01#2(#3$45#6%77/6-8(1#9#:(#;/,6<#=(7681(0-%./0#1:>8?%.@,(#2(1#?,-%./01#A!"B#>/?8/2/?%C0(D# &'()*+,#-./0/-.,1*23,4#&5#6'()*+7#89## :;#<=23>1+36#/1?3=+,#23/02+36#@*+-#+-3#,1A3#A<+1?0=5#1=6#,1A3#/-3=0+./3# dans le 2 e intron et comprenant une insertion de 26 aminoacides dans le domaine de trans-activation. Le rôle de Pit1ß n’est pas connu. Pit1 est impliqué dans les étapes finales de différenciation des lignées somato- lactotropes et thyréotropes. Chez la souris, l’expression de Pit1 est retrouvée à partir de e13.5 dans le lobe antérieur. Son expression est maximale à e16,5, et persiste à l’âge adulte. Il existe 2 modèles murins d’inactivation spontanée de Pit1 (111). Les souris Snell présentent une mutation spontanée de Pit1 (W261C) à l’origine d’une anomalie de liaison sur sa séquence cible ; les souris Jackson présentent un réarrangement chromosomique à l’origine d’un défaut d’expression du facteur de transcription. Dans ces 2 modèles, il existe un déficit somato-lactotrope et thyréotrope post-natal, associé à une hypoplasie hypophysaire (6). Le fait que ce déficit survienne après la naissance souligne que Pit1 est probablement utile pour l’expansion de ces contingents cellulaires, mais pas nécessairement pour leur genèse. L’expression de Prop1 précède et est nécessaire à l’expression de Pit1 (165). Un complexe Prop1-ßcaténine se forme permettant la libération de Hesx1 et TLE du early enhancer de Pit1 . Pit1 est ensuite capable d’auto-régulation. Pit1 stimule le promoteur de la prolactine (en synergie avec Lhx3) (185), de la GH et de la TSHß (6). Pit1 est également capable de recruter des co-activateurs (230) comme l’élément de réponse à l’AMPc ou le CBP pour promouvoir l’expression de ses gènes cibles. Pit1 est enfin capable d’interagir de façon synergique avec Gata2 pour promouvoir l’expansion de l’axe thyréotrope, et inhibe l’action promotrice de ce même facteur de transcription sur l’axe gonadotrope (43).
b. En pathologie humaine
Chez l’homme, le gène de POU1F1 , orthologue humain de Pit1 , a été cloné sur le chromosome 3p11 (161). Il comprend 6 exons, et l’ADN complémentaire de 1050 pb code pour une protéine de 291 acides aminés. Les mutations de POU1F1 ont constitué la 1ere cause génétique identifiée de déficit hypophysaire combine multiple, en 1992. A ce jour, 28 mutations ont été rapportées (table 4), la majorité (23/28, soit 82% des cas) de transmission récessive (139; 156; 202; 89; 146; 180; 22; 60; 147; 35; 79; 228; 68; 77; 118; 124; 174; 66; 159; 210; 23). Le phénotype est stéréotypé associant un déficit somato-lactotrope et thyréotrope. L’âge au diagnostic du déficit en GH est en général précoce, alors que celui du déficit en TSH est plus variable. Turton et al. ont rapporté le seul patient à ce jour ne présentant pas de déficit en TSH à l’âge de 20 ans, malgré son génotype porteur de la mutation E230K de POU1F1 . Point intéressant, un autre patient, porteur de la même mutation, avait présenté un déficit thyréotrope peu de temps après la naissance, éliminant l’implication de la mutation dans la variabilité phénotypique (210). L’IRM retrouve une hypophyse normale ou hypoplasique, et l’absence d’anomalie extra-hypophysaire. Les mutations de POU1F1 surviennent plus fréquemment dans le cadre de déficits hypophysaires familiaux (25% contre moins de 5% dans les cas sporadiques) (95).
4. LES FACTEURS DE TRANSCRIPTION A HOMEODOMAINE DE TYPE LIM
Ces facteurs de transcription sont caractérisés par un homéodomaine localisé en position centrale et 2 domaines LIM à l’extrémité N terminale, régulant d’une part la fixation à l’ADN et d’autre part l’activité transcriptionnelle via des protéines partenaires. La séquence consensus reconnue par les facteurs de transcription à homéodomaine de
! #+! <".,/#)&LGX&<".,/#) LG 2".Y"C".,/#)&2".Y"C".,,/#) '& '6R& 'EF& 6'R& 6ZP& 6E'& !523('( !"'A( ,6('( 0'1&A( !#A( )AH( !"#$%& '& ()*+& %,-)./0&12314()50&6778&9& EQH( )AEH<( 4,:;$%& '& <".+& 4"=,#">,0&?.2@ABC1)#)D0&'EEF&9& PD(
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
%!(H@(!523('1( !( !"#$%$ (0))(02"!1((!($(50A( AY*,#0&.Y*,#/5.)&C)&-,&.:D,["#\&!"#$%0&#"#&5)#5]&4,:;$%0&=,:;&5)#5+&TN,#5.0&DN,#5./55/"#]&()*0&NY*)55/>)]&<".0& C"./#,#D)+&?C,JDY&C)&^)-_)N.,##0&B#C"*N/#)&()>/)V50&677E& !85610( !$!0D( ,910( 304)D( #%D( 2D6PXS( !"#$%&' (' )*+,' -"./'0123/'4554'6' UW( =SYPA`C( 27899":*' (' )*+,' 3+;*<<"8?2<@A/'455B'6' PQS( =QER=04C( 27899":*' C<.DA<'(' )*+,' ="D>AE":8<2<@A/'455J'6' PTX( 2DQRX( !"#$%&' K' )*+,' 1*D<"9*N/'0=23/'FIIO'6' PSV( !DVQTFVQU"#(CC( GP+">":*' K' )*+,' -"./'0123/'455J'6' UU( !DVSV"#(C( GP+">":*' K' )*+,' Q*<@D8+R9/'0=23/'455F'6' VX( 2D,QTI( ?A<%&' K' )*+,' CD8*/'2<@A+D0/'FIIJ'6' WX( !DVVW&04C( GP+">":*' K' )*+,' S#D.A (#(SC(!856104("#( !"#$%$ (322135$#4((!#('183D( 3P+""'E#."UA a. Au cours du développement hypophysaire murin i. Lhx3 Lhx3 est exprimé à partir de e9,5 dans la poche de Rathke en développement (180). Son expression augmente dans l’ensemble de la poche jusqu’à e12,5 puis se stabilise sans diminution après la naissance et au cours de la vie adulte (131). L’invalidation homozygote de lhx3 ( lhx3 -/-) entraine une mort néonatale précoce. L’hypophyse est aplasique, maintenue à un stade rudimentaire (e12,5), et ne présente pas de lobe antérieur ni de lobe intermédiaire. Il n’existe pas de différenciation des lignées hypophysaires (181; 180). Le lobe postérieur a une apparence normale. L’invalidation de lhx3 (lhx3 -/-) aboutit à un dérèglement de l’organisation spatio-temporelle du développement hypophysaire, avec en particulier une migration anormale (dorsale) des progéniteurs nécessaires au développement du lobe antérieur. L’hypoplasie hypophysaire est probablement liée à une apoptose massive de ces progéniteurs (la prolifération cellulaire ne semble pas affectée) (236; 54). L’invalidation hétérozygote de lhx3 ( lhx3 +/-) n’entraîne pas d’anomalie du phénotype hypophysaire (54). Lhx3 code pour 3 isoformes : Lhx3a, Lhx3b et M2-Lhx3 (187). Le rôle précis de chacune de ces isoformes reste mal défini. In vitro, Lhx3b a une activité plus faible que Lhx3a et M2-Lhx3 (186; 233). Cependant, in vivo, une surexpression de Lhx3a ou Lhx3b (souris transgéniques avec expression des 2 isoformes Lhx3a et Lhx3b sous contrôle du promoteur de la sous-unité alpha) est à l’origine d’un hypogonadisme hypogonadotrope, alors qu’une surexpression de M2-Lhx3 n’entraîne pas d’anomalie phénotypique (176). L’activation de Lhx3 semble multifactorielle. Outre le rôle de lhx4 (détaillé dans le paragraphe suivant), d’autres gènes d’expression plus ou moins précoce pourraient être impliqués : fgf8, sox2, pitx1 et pitx2 (29), (160), isl1 voire prop1 (11). D’autres protéines (Sp1, Nuclear factor 1) semblent nécessaires à l’expression de lhx3, via une interaction directe sur son promoteur (232). Inversement, l’expression de lhx3 est nécessaire à l’expression physiologique de Hesx1 (33), foxl2, Notch2 (l’absence d’expression de Notch2 chez les souris lhx3 -/- pourrait expliquer le dérèglement du gradient dorso-ventral au cours du développement hypophysaire), SF1, le récepteur de GnRH et FSHß (impliqués dans la différenciation gonadotrope) (227; 123; 75), et tbx19 (dont la diminution d’expression dans le lobe intermédiaire explique en partie l’absence de développement de la lignée corticotrope). Lhx3 agit enfin en synergie avec Pit1 (Pou1f1) lors de l’activation du promoteur des gènes de la prolactine , Tshß et Pit1 ( Pou1f1 ) (71). Le rôle de lhx3 ne se limite pas au développement de l’hypophyse. Lhx3 joue ainsi un rôle majeur lors du développement des motoneurones médullaires (179; 204) (expliquant au moins en partie le phénotype de limitation de rotation cervicale observé chez les patients avec mutation de LHX3 (188)), et de l’oreille interne (87; 84). ii. Lhx4 Lhx4 est exprimé à partir de e9,5 dans la poche de Rathke en développement. Son expression est restreinte à la zone correspondant au futur lobe antérieur à partir de e12,5. L’expression de lhx4 diminue fortement à partir de e15,5 même si une expression résiduelle est retrouvée à la naissance et chez la souris adulte (180; 158). L’invalidation homozygote de lhx4 ( lhx4 -/-) chez la souris entraine une mort néonatale précoce liée à des troubles respiratoires (180). Au niveau hypophysaire, il existe une spécification correcte des 5 lignées cellulaires, mais une réduction majeure de leur prolifération, à l’origine d’une hypoplasie hypophysaire sévère. A e12,5, le lobe antérieur ! #"! n’est pas clairement individualisé au sein de cette hypophyse hypotrophique. A e14,5, une ébauche de lobe antérieur est visualisable, les 5 types cellulaires sont présents, mais le lobe antérieur est hypotrophique par rapport aux souris de phénotype « sauvage ». Le développement hypophysaire ne se poursuit pas après e14,5. A la naissance, on observe ainsi une hypoplasie hypophysaire sévère, malgré la présence des 5 types cellulaires (180). Cette hypoplasie est probablement due à une augmentation de l’apoptose, plus qu’à une diminution de la prolifération (158). A l’inverse, les souris hétérozygotes pour l’invalidation de lhx4 ( lhx4 +/-) ne présentent pas d’anomalie hypophysaire (180). Ce point est d’une importance majeure pour établir une comparaison avec le phénotype humain induit par les mutations de LHX4 , comme nous le détaillerons par la suite. Il existe de fortes interactions entre lhx4 et d’autres facteurs de transcription hypophysaires lors du développement de la poche de Rathke (37). L’expression de lhx4 est nécessaire à l’expression de lhx3 : à e12,5, on ne retrouve en effet pas d’expression de lhx3 chez les souris présentant une invalidation homozygote de lhx4 ( lhx4 -/-) ; le niveau d’expression de lhx3 est à nouveau normal à e14,5, suggérant que des facteurs autres que Lhx4 sont impliqués après e12,5, permettant une expression normale de lhx3 à e14,5 en absence de lhx4. Un candidat potentiel à cette activation retardée est Prop1. Ainsi, les souris présentant une double invalidation de lhx4 et prop1 ( lhx4 -/- prop 1-/-) ne présentent pas d’expression de lhx3 à e14,5 (158). Plusieurs études ont été menées pour déterminer l’importance relative de lhx3 et lhx4 au cours du développement hypophysaire : le développement de souris avec inactivation de l’un ou des 2 gènes a permis d’observer , qu’un seul des gènes était suffisant pour que la poche de Rathke soit formée , que l’absence des 2 gènes entrainait un arrêt de progression de la poche rudimentaire après e9,5 , que la différenciation des 5 lignées hypophysaires nécessitait une expression normale de lhx3 (et pas de lhx4) (158) Lhx3 et lhx4 semblent donc avoir des rôles redondants (en ce point comparables par exemple à Pitx1 et Pitx2) mais sur une durée limitée de temps. Enfin, lhx4 joue vraisemblablement un rôle dans le développement des cellules souches hypophysaires : Chen et al. ont montré que le retrait de Leukemia Inhibiting Factor (un facteur inhibant la différenciation des cellules souches) en culture de cellules souches hypophysaires abolissait l’expression de Lhx4, suggérant que ce facteur de transcription joue un rôle dans le maintien d’un état indifférentié (31). iii. Isl1 Les données sur le rôle précis d’Isl1 lors de la formation de l’hypophyse sont rares. Isl1 semble jouer un rôle primordial dans les premières étapes du développement hypophysaire. Il est exprimé dès e8,5 dans l’ectoderme oral. Son expression est restreinte à la poche de Rathke à partir de e9,5, puis limitée à la zone ventrale de la poche de Rathke à partir de e10,5 (54). Ces cellules ventrales exprimeront TSHß et la sous-unité alpha. L’expression de Isl1 persiste après la naissance (données non publiées, voir résultats). L’invalidation homozygote de Isl1 ne bloque pas les premières étapes de la formation de la poche de Rathke. Cette invalidation est à l’origine d’une mort embryonnaire à e10 ,5 du fait de malformations cardiaques sévères. Aucune autre donnée n’a été publiée sur le rôle précis de Isl1 au cours des étapes suivantes du développement hypophysaire. Isl1 interagit avec plusieurs facteurs de transcription hypophysaire incluant Lhx3 (12; 122), Sf1(75), ou certaines voies de signalisation (113). Ainsi, l’inactivation de lhx3 aboutit à une expression ectopique de Isl1 (54). ! ##! Isl1 joue un rôle majeur dans le développement et la prolifération des cellules souches de nombreux organes dont le cœur et le pancréas (70; 168). Son expression dans les présumées cellules souches hypophysaires est inconstante. Cependant, le fait que son invalidation homozygote entraine un arrêt précoce du développement hypophysaire est un argument pour son rôle dans la prolifération et la différenciation des progéniteurs hypophysaires. b. En pathologie humaine i. LHX3 A ce jour, 10 mutations de LHX3 ont été rapportées comme étant responsables de déficit hypophysaire congénital chez l’homme (135; 83; 187; 10; 150; 175; 101) (Table 5). La transmission est autosomique récessive pour l’ensemble des patients. A l’exception d’une mutation ( p.K50X ), l’ensemble des cas sont familiaux. Toutes les mutations sont exoniques, à l’exception d’une mutation intronique récemment publiée : cette mutation abolit un site d’épissage, entrainant un codon stop prématuré au niveau de l’acide aminé 186. La protéine théorique ne possède pas d’homéodomaine, ni de domaine C-terminal (101). Le phénotype hypophysaire est assez constant, avec un déficit somatotrope, thyréotrope et gonadotrope présent chez tous les patients. Le déficit corticotrope est variable : seulement 5 mutations induisent un phénotype hypophysaire présentant ce déficit, et dans 3/5, le déficit est inconstant au sein des membres de la même famille porteurs de la mutation. L’IRM retrouve une hypophyse aplasique dans 10% des cas, hypoplasique dans 50% des cas, hyperplasique dans 30% des cas, et normale dans 10% des cas. Dans un cas, une image hypointense évocatrice de microadénome a également été observée. Le phénotype hypophysaire est souvent complété par des anomalies de la rotation du cou (70% des cas), des anomalies vertébrales (50% des cas) et une atteinte auditive plus ou moins sévère (50% des cas). Les anomalies de rotation du cou sont probablement multifactorielles, liées à l’implication de LHX3 dans le développement des motoneurones spinaux, mais aussi à des anomalies osseuses de la charnière atlanto- occipitale (101). ii. LHX4 Il n’existe qu’un faible nombre de mutations de LHX4 rapportées à ce jour (45; 117; 126; 196; 26; 149; 199) (Table 6). Elles se caractérisent principalement par leur grande variabilité phénotypique en termes de déficits hypophysaires et d’anomalies morphologiques cérébrales. Leurs caractéristiques et les phénotypes induits sont détaillés dans la discussion des résultats de du deuxième article présenté dans le présent mémoire. iii. ISL1 Aucune mutation d’ ISL1 responsable d’hypopituitarisme congénital n’a été rapportée à ce jour. ! #$! !"#$ !"#$ %&$ !"# $%# $&# %'"# %'(# )%&# Mutation Loss Exon 2 to 5 Loss Exon 2 to 5 Loss Exon 2 to 5 p.K50X c455-2>AG soeur soeur frère GH D D D D D TSH D D D D D ACTH D D D D N LH FSH D D D ? D Familial/ Familial Familial Familial Familial Familial Sporadique Hypoph. Hypo Hypo Hypo Hypo Hypo kyste Rotation cou Limitée Limitée Limitée Limitée Limitée Colonne cervicale Cyphose thoraco- Cyphose thoraco- Cyphose thoraco- Cyphose thoracique Perte de la lordose lombaire lombaire lombaire Perte auditive Modérée Modérée Modérée Sévère Modérée à sévère Retard mental Oui Oui Non ? Transm. Rec Rec Rec Rec Rec Rajab, HumMolGenet Rajab, HumMolGenet Rajab, HumMolGenet Rajab, Kristrom, JCEM, 2009 160 160 160 HumMolGenet 101 160 '()*+$,-$#./(0123$4+$!%56$7(8817/9+3$:$;+$<1.7=$>+8793+2/(012$3;?9@(0A.+$4+3$@./(0123$B%&C$?1@9141@(D2+E$ *+,-./0#123431205.6708#*9#:+,-./;#<9#=46>5?;#*@<9#?70#35A7=/0#346/7B60#:7#?5#>B/5A4=0#5C5.7=/#B=#31+=4/237#C56.5D?7E#F23431E9#C4?B>7# 123431205.67;#F2349#12343?50.7;#F23769#123763?50.7E#G65=0>8#H7-9#6+-700.IE# Mutation p.Y116C p.Y116C p.Y116C Del.23pb g.159delT GH D D D D D TSH D D D D D ACTH N N N N N LH FSH D D ? D D Familial/ Familial Familial Familial Familial Familial Sporadique Hypoph. Hypo Hypo Hypo Hyperplasie tardive Hyper avec aspect de microadénome Rotation cou Limitée Limitée Limitée Limitée Limitée Colonne cervicale Normale Normale Normale Normale Normale Perte auditive Modérée ? Modérée Sévère ? Retard mental ? ? ? Oui Oui Transm. Rec Rec Rec Rec Rec Netchine, Netchine, NatGenet, Netchine, NatGenet, Netchine, NatGenet, Banghoo, JCEM, NatGenet, 2000 2000 2000 2000 2006 135 135 135 135 10 !"#$%&'(&)*+",-./&0%&1234&5"66-5+7%/&8&9%&:-*5;&<%657/%.+",-.&/9=7>",?*%&0%/&>*+",-./&@2AB&=->7-0->"C.%D& !$$#()! !)#-"#4(!2($$3(42( "-3(!642(#(!-%$#(! "$'"#((-('$-% #(-(-$('(! $)!((-"-<5(>)! !52(( '( )! !)#-"3(>)! 2()! !-#3(>)!"2()!"!-#5(?"-#4(@$2("$##5( Mutation p.A210V p.A210V p.E173X Complete p.W224X p.W224X p.W224X Frère Soeur LHX3 deletion Frère Soeur Soeur GH D D D D D D ? TSH D D D D D D ? ACTH D N D N N N ? LH FSH D D D ? D D ? Familial/ Familial Familial Familial Familial Familial Familial Familial Sporadique Hypoph. Hyperplasie Hyperplasie Hypo Aplasie Normal Normal ? hyperintense hyperintense Rotation cou Limitée Limitée Limitée Limitée Normale Normale Normale Lobe posterieur ? ? ? Normal ? Colonne Perte de la Perte de la Normale Perte de la Normale Normale ? cervicale lordose lordose lordose Perte auditive ? ? ? ? ? ? ? Retard mental Non Non Non Oui Non Non ? Transm. Rec Rec Rec Rec Rec Rec Rec Pfaeffle, Pfaeffle, Pfaeffle, Pfaeffle, Pfaeffle, Pfaeffle, Pfaeffle, JCEM, 2007 JCEM, 2007 JCEM, 2007 JCEM, 2007 JCEM, 2007 JCEM, 2007 JCEM, 2007 150 150 150 150 150 150 150 !"#$%&'(&)*+",-./&0%&1234&5"66-5+7%/&8&9%&:-*5;&<%657/%.+",-.&/9=7>",?*%&0%/&>*+",-./&@2AB&=->7-0->"C.%D& !$$#()! !)#-"#4(!2($$3(42( "-3(!642(#(!-%$#(! "$'"#((-('$-% #(-(-$('(! $)!((-"-<5(>)! !52(( '( )! !)#-"3(>)! 2()! !-#3(>)!"2()!"!-#5(?"-#4(@$2("$##5( OBJECTIFS L’hypophyse est un organe clé contrôlant le fonctionnement des glandes endocrines de l’organisme. Une anomalie de développement provoqué par une anomalie d’un facteur de transcription ou d’un co-répresseur peut être à l’origine d’un dysfonctionnement hypophysaire diagnostiqué à la naissance ou de façon plus tardive. Ce dysfonctionnement justifiera le plus souvent un traitement substitutif prolongé, une surveillance adaptée et une éducation du patient. La plupart des étiologies d’hypopituitarisme congénital sont liées à des mutations de facteurs de transcription hypophysaires. Leur étude est cependant délicate : malgré une augmentation constante des connaissances au cours de 15 dernières années en termes de nouveaux facteurs de transcription et de mise en évidence de nouveaux mécanismes physiopathologiques, seulement un faible pourcentage d’étiologies d’hypopituitarisme congénital est identifié. Identifier les étiologies des hypopituitarismes congénitaux impose donc d’identifier ou de préciser des mécanismes physiologiques et physiopathologiques survenant au cours du développement hypophysaire. Le modèle murin est modèle théorique de référence pour étudier le développement hypophysaire et le rôle respectif des facteurs de transcription. Les travaux présentés dans cette thèse ont donc porté sur 4 aspects , Clarifier des mécanismes physiologiques : les interactions Prop1/Hesx1/Co- répresseurs Tle au cours du développement hypophysaire , Identifier de nouveaux mécanismes physiologiques : rôles potentiels du facteur de transcription hypophysaire Pitx2 dans le développement de l’axe thyréotrope , Identifier de nouveaux facteurs de transcription impliqués dans le développement hypophysaire : expression et rôle du facteur de transcription à homéodomaine LIM, Isl-1 , Identifier et étudier la signification de nouvelles mutations responsables d’hypopituitarisme congénital : analyse de variants alléliques de LHX4 chez des patients porteurs de déficit hypophysaire multiple congénital ! #%! RESULTATS 1. DEVELOPPEMENT HYPOPHYSAIRE: VERS UNE MEILLEURE COMPREHENSION DES INTERACTIONS ENTRE 2 DES PRINCIPAUX FACTEURS DE TRANSCRIPTION A HOMEODOMAINE DE TYPE PAIRED, PROP1 ET HESX1 CO-REPRESSORS TLE1 AND TLE3 INTERACT WITH HESX1 AND PROP1. LR. Carvalho*, ML. Brinkmeier*, F. Castinetti *, BS. Ellsworth, SA. Camper Molecular endocrinology, in press *co-premiers auteurs ! #&! INTRODUCTION Dans l’hypophyse en formation, Olson et al. ont souligné le rôle majeur de la voie Wnt/ß- caténine (85; 140). Deux modèles de souris transgéniques ont été utilisés : Pitx1 Cre où l’expression d’une recombinase Cre est ubiquitaire dans les cellules de la poche de Rathke à partir de e9.0 ; Pit1 Cre où l’expression d’une recombinasse Cre est limitée aux lignées Pit1 dépendantes à partir de e13,5 (140). Ces 2 modèles de souris Cre étaient croisés avec des souris porteuses d’un gène de la ß-caténine encadré de séquences Lox, permettant son inactivation en présence de la recombinase. Une inactivation précoce de la ß-caténine ( Pitx1 Cre ) entraine un développement hypophysaire initialement très proche de celui observé pour des souris de génotype sauvage (les facteurs de transcription hypophysaires d’expression précoce sont normalement exprimés), mais l’absence d’expression de Pit1 (et des lignées dépendantes de Pit1, c’est à dire somato- lactotropes et thyréotropes). A l’inverse, l’expression de Prop1 est normale. Une inactivation plus tardive de la ß-caténine ( Pit1 Cre ) ne provoque pas d’anomalie de l’expression de Pit1, ni de la différentiation terminale des lignées hypophysaires. Ces résultats suggèrent que la ß-caténine est nécessaire à l’initiation mais pas au maintien de l’expression de pit1 (140). La famille de co-répresseurs Groucho/Tle est composée de 7 membres dont 3 sont spécifiquement exprimés dans l’hypophyse au cours du développement : Tle1, Tle3 et Aes. Tle2 est retrouvé dans le lobe postérieur. Tle 4 et Tle5 n’ont pas d’expression hypophysaire (42; 19). Il n’existe pas de donnée spécifique sur l’expression hypophysaire de Tle6, qui réprime Tle1. Ces protéines ne lient pas directement l’ADN, mais sont recrutés par divers facteurs de transcription. Les mécanismes précis aboutissant à la répression de la transcription restent mal définis (91). Dans l’hypophyse en développement, un phénomène complexe d’interactions est nécessaire à la différenciation des lignées Pit1 dépendantes. Au cours du développement hypophysaire murin, Hesx1 et Tle1 ont des profils d’expression identiques entre e9,5 et e12,5 (et une diminution rapide d’expression après e13,5). Dasen et al. ont montré dès 2001 que Hesx1 interagissait avec Tle1 pour moduler l’expression de Prop1 (42). L’interaction entre les protéines est assurée via 2 domaines, un domaine eh1 fortement conservé sur la région N-terminale de Hesx1, et un domaine WD40 sur Tle1. D’autres co-répresseurs ou transférases (NCoR, la famille des Histones déacétylases, Brg1, DNMT1 (172)) sont également capables d’interagir avec l’homéodomaine de Hesx1, mais ils ne sont pas nécessaires Error! Bookmark not defined. à l’interaction Hesx1/Tle1. La famille Groucho/Tle compte 6 co-répresseurs (Tle1 à Tle5 dont la séquence et la structure sont très proches, et Aes) dont la plupart ont une expression hypophysaire lors de l’embryogenèse (42). Carvalho et al. ont souligné l’importance de l’interaction HESX1/TLE1 chez l’homme, en rapportant le cas d’un patient porteur d’une mutation de HESX1 situé dans la zone codant pour ce domaine eh1 (mutation I26T ) responsable d’hypopituitarisme congénital. A noter que cette mutation avait été également rapportée chez la souris comme responsable d’un phénotype mineur par rapport à d’autres mutations spontanées de Hesx1 (24). Le modèle proposé par Olson et al. souligne le rôle primordial des interactions Hesx1/Prop1/Tle au cours du développement hypophysaire (140). L’enhancer proximal de pit1 contient un site de liaison pour les facteurs de transcription à homéodomaine de type paired. Le domaine C-terminal de Prop1 interagit avec la ß-caténine via un domaine de répétitions de type armadillo. En culture cellulaire, il existe une activation ! #'! synergique du promoteur de Prop1 par Prop1 et la ß-caténine. Le modèle proposé par les auteurs est essentiellement basé sur les interactions entre Prop1, la ß-caténine et Hesx1 sur l’enhancer proximal de pit1. A e11,5, Hesx1 et le co-répresseur Tle1 sont fixés sur l’enhancer proximal de pit1. Hesx1 ayant un effet répressif, il n’existe pas d’activation de pit1. A e12,5, l’enhancer proximal est occupé par prop1 et un faible niveau de co-répresseur Tle1 et Hesx1. A e13,5 pit1 est activé, l’enhancer proximal est occupé par Prop1 et la ß-caténine. La voie Wnt/ß-caténine étant activée à partir de e12,5, il existe vraisemblablement une action synergique de Prop1 et la ß-caténine permettant l’activation de pit1 à e13,5. En parallèle, Prop1 et la ß-caténine se fixent à partir de e12,5 sur des séquences régulatrices de Hesx1 pour inhiber son activation. Ces résultats sont confirmés par le phénotype des souris transgéniques avec inactivation précoce de la ß-caténine : on observe une expression persistante de Hesx1 à e14,5 (et une hypophyse de morphologie similaire à celle observée pour les souris Prop1 -/-) (140). Nos objectifs étaient donc d’évaluer plus précisément le rôle des co-répresseurs Tle1 et Tle3 au cours du développement hypophysaire - A l’aide de souris transgéniques ayant une expression ectopique de Tle3 - via des études fonctionnelles in vitro pour déterminer si les co-répresseurs de la famille Groucho/Tle pouvaient avoir un effet répresseur propre sur Prop1. ! #(! Molecular Endocrinology. First published ahead of print February 24, 2010 as doi:10.1210/me.2008-0359 ORIGINAL RESEARCH Corepressors TLE1 and TLE3 Interact with HESX1 and PROP1 Luciani R. Carvalho,* Michelle L. Brinkmeier,* Frederic Castinetti,* Buffy S. Ellsworth, and Sally A. Camper Department of Human Genetics (L.R.C., M.L.B., F.C., S.A.C.), University of Michigan, Ann Arbor, Michigan 48109; and Department of Physiology (B.S.E.), Southern Illinois University School of Medicine, Carbondale, Illinois 62901 Pituitary hormone deficiency causes short stature in one in 4000 children born and can be caused by mutations in transcription factor genes, including HESX1 , PROP1 , and POU1F1 . HESX1 interacts with a member of the groucho-related gene family, TLE1, through an engrailed homology domain and represses PROP1 activity. Mice with Prop1 deficiency exhibit failed differentiation of the POU1F1 lineage, resulting in lack of TSH, GH, and prolactin. In addition, these mutants exhibit profound pituitary dysmorphology and excess Hesx1 and Tle3 expression. The ability of HESX1 to interact with TLE3 has not been explored previously. We tested the ability of TLE3 to enhance HESX1-mediated repression of PROP1 in cell culture. Both TLE3 and TLE1 repress PROP1 in con- junction with HESX1 with similar efficiencies. TLE1 and TLE3 can each repress PROP1 in the absence of HESX1 via a protein-protein interaction. We tested the functional consequences of ectopic TLE3 and HESX1 expression in transgenic mice by driving constitutive expression in pituitary thyro- trophs and gonadotrophs. Terminal differentiation of these cells was suppressed by HESX1 alone and by TLE3 and HESX1 together but not by TLE3 alone. In summary, we present evidence that HESX1 is a strong repressor that can be augmented by the corepressors TLE1 and TLE3. Our in vitro studies suggest that TLE1 and TLE3 might also play roles independent of HESX1 by interacting with other transcription factors like PROP1. (Molecular Endocrinology 24: 0000–0000, 2010) he pituitary gland is derived from two different layers There are many transcription factors involved in the Tduring embryogenesis. The neuroectoderm, originat- commitment and differentiation of cell types in the ante- ing from the floor of the diencephalon, will give rise to the rior lobe of the pituitary gland. Spontaneous mutants and posterior pituitary lobe, which stores and releases oxyto- genetically engineered murine models have demonstrated cin and antidiuretic hormone. The oral ectoderm, derived a role for many of these molecules in the etiology of pitu- from the roof of the mouth, will give rise to Rathke’s itary hormone deficiency. These include the transcription pouch, the primordium for the anterior and intermediate factors HESX1, PROP1, POU1F1 (PIT1), LHX3, LHX4, pituitary lobes (1). The anterior lobe will give rise to five OTX2, TBX19 (TPIT), SOX2, and SOX3 (reviewed in different cell types known as somatotrophs, lactotrophs, Ref. 2). In humans, the phenotypes produced by lesions in gonadotrophs, thyrotrophs, and corticotrophs, which these genes range from isolated hormone deficiency to produce and release six different hormones: GH, prolac- more complex disorders such as septo-optic dysplasia and tin, LH and FSH, TSH, and ACTH, respectively. These holoprosencephaly associated with combined pituitary hormones have important roles in growth, lactation, hormone deficiency (CPHD). Mutations in PROP1 are fertility, metabolism, and stress response in humans the most prevalent known cause of CPHD in humans. and mice. This is likely due to a hot spot in the sequence that is ISSN Print 0888-8809 ISSN Online 1944-9917 Abbreviations: CGA, Chromogranin A; CPHD, combined pituitary hormone deficiency; Printed in U.S.A. e9.0, embryonic d 9.0; ␣GSU, glycoprotein hormone ␣-subunit promoter; WT, wild type. Copyright © 2010 by The Endocrine Society doi: 10.1210/me.2008-0359 Received September 23, 2008. Accepted January 15, 2010. * L.R.C., M.L.B., and F.C. are equally contributing authors. Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 1 Copyright (C) 2010 by The Endocrine Society 2 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 vulnerable to a 2-bp deletion (3), but the etiology of many coimmunoprecipitation studies (9). A patient with short cases of CPHD remains unknown. Thus, it is clear that stature due to GH deficiency and evolving pituitary hor- other genes remain to be identified, and the characteriza- mone deficiency harbored a mutation in the eh1 domain tion of these will elucidate the pathogenesis of these com- (HESX1I26T), confirming the importance of the eh1 do- plex conditions and also shed light on normal pituitary main in vivo (19). development. TLE1 and TLE3 a have similar protein structure, and One of the first genes to be expressed in murine Rathke’s their patterns of expression during pituitary development pouch is Hesx1 at embryonic d 9.0 (e9.0), and its expres- are overlapping, but distinct (9, 14, 16). TLE1 is ex- sion wanes significantly after e12.5 (4). The Prop1 gene is pressed throughout the pituitary at e12.5 and e14.5, expressed in the developing pituitary gland at e10.5, whereas TLE3 is initially expressed in the ventral dien- peaks at e12.5, and decreases after e14.5, when Pou1f1 cephalon at e12.5 and then becomes restricted to the cells expression is initiated (5). Mouse mutants with lesions in dorsal to the lumen of Rathke’s pouch at e14.5. This these three genes, Hesx1 , Prop1 , and Pou1f1 , model hu- restricted expression is aberrant in the Prop1 mutant mice man CPHD and have contributed to understanding the in which TLE3 is expressed extensively in the developing mechanism of action of each gene (4–11). Mechanistic pituitary gland (14). analysis of known genes is a strategy for identifying genes The current study advances our understanding of the that could explain unknown genetic etiologies for CPHD. activity of TLE1, TLE3, PROP1, and HESX1 on gene For example, Prop1 is necessary for the inhibition of expression in cell culture and in pituitary development. Hesx1 transcription (11) and activation of Pou1f1 ex- Our cell culture results reveal that TLE3 functions simi- pression, and each of these genes is mutated in some cases larly to TLE1 in enhancing HESX1 repression of PROP1 of CPHD in humans (4, 12, 13). In mice, Prop1 affects the activity. Surprisingly, TLE1 and TLE3 can repress PROP1 in expression of many other genes, including restriction of the absence of HESX1 through protein interaction. We Tle3 expression to the dorsal part of the anterior pituitary tested the ability of ectopic TLE3 expression to interfere (14) and the activation of Notch2 (7), but these genes are with pituitary development, as a single transgene or in unlikely to be mutated in humans with CPHD due to their conjunction with HESX1. The results demonstrate that involvement in development of many organs critical for HESX1 has a strong inhibitory effect on differentiation viability. that is modestly enhanced by TLE3, but TLE3 has no Tle3 (transducin-like enhancer of split 3), and several effect on its own in this ectopic context. These results other members of the Gro/TLE family of groucho-related suggest that persistent, ectopic expression of HESX1 genes are expressed during pituitary development (9, 14, alone is sufficient to inhibit cell differentiation in trans- 15). The Gro/TLE family is comprised of six different genic mice and that HESX1 repression may be mediated genes Tle1 -4, Aes , and Tle6 (16, 17). TLE1-4 each has by TLE1 and/or TLE3. Taken together, these studies in cells and animals advance our understanding of pituitary glutamine repeats (Q region) at the N terminus and a development at the molecular level. WDR (WD-40 repeat) region for protein-protein interac- tion at the C terminus. AES has the Q region but lacks the WDR region, whereas TLE6 contains a WDR region but Results no Q region. The TLE genes act as corepressors by bind- ing to transcription factors such as HESX1 and TCF/LEF TLE3 enhances HESX1 repression of PROP1 family members that have the engrailed homology do- Previous studies have shown that TLE1 and HESX1 main (eh1) and SIX family members that contain short interact to repress PROP1 activity (9, 19, 20). To test the peptide sequences WRPW. potential of TLE1 and TLE3 to act as corepressors with PROP1 and HESX1 contain paired-type homeodo- HESX1 in regulating PROP1 activity, transient transfec- mains, and they bind to DNA as homodimers or het- tion studies were performed in cell culture using 293T erodimers at a paired homeodomain binding site, P3 cells (human embryonic kidney). We used expression vec- (TAAT TGA ATTA , core sequence is in bold) (5, 9, 18). tors for human TLE1, TLE3, PROP1, HESX1 wild type PROP1 acts as an activator of this promoter in transfec- (WT), and HESX1( ⌬1–50), which is missing the N-termi- tion, and its activity is repressed by HESX1. HESX1 re- nal 50 amino acids including the eh1 domain. As ex- cruits TLE1 through an eh1 domain creating a complex pected, PROP1 activated the POU1F1 promoter when that is a stronger repressor of PROP1 activity than transfected alone, and HESX1 WT, HESX1( ⌬50-185), HESX1 alone. Mutations in the eh1 domain of HESX1 TLE1, or TLE3 had no activating effect over the reporter are associated with impaired PROP1 repression in tissue alone (data not shown). HESX1 WT repressed PROP1 ac- culture studies and also impaired recruitment of TLE1 in tivation by 46%. HESX1 together with TLE1 or TLE3 en- Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 3 FIG. 2. Gro/TLE family members repress PROP1 activity in the absence of HESX1. 293T cells were transiently transfected with POU1F1 promoter-luciferase reporter construct with 50 ng PROP1 in the absence of HESX1. Increasing amounts of TLE1 or TLE3 were transfected: 50 ng (1X), 150 ng (2X), and 250 ng (3X). Maximal repressive effects were observed for 200 and 250 ng (plateau phase, data not shown). The luciferase activity produced from PROP1 alone and from PROP1 and TLE1 or TLE3 together was statistically different for the three amounts of DNA transfected ( P Ͻ 0.001). FIG. 1. Gro/TLE family members exert effects on HESX1-mediated repression of PROP1 activation. 293T cells were transiently transfected with expression vectors for PROP1, HESX1 WT, HESX1( ⌬1-50), TLE1, or creasing amounts of TLE1 or TLE3 strengthens the re- TLE3 with the POU1F1 promoter driving the luciferase reporter. The pressive effects, with a maximal effect observed for 200 normalized results are expressed as fold change relative to the vector and 250 ng (plateau phase). The maximal effect of TLE backbone. The experiments were performed in triplicate in at least three with PROP1 in the absence of HESX1 is the same as that independent experiments, and typical results are shown. **, P Ͻ 0.001. observed with HESX1 and PROP1 independent of TLE. hanced PROP1 repression up to 66 and 79%, respectively These data suggest that TLE1 and TLE3 can repress (Fig. 1). TLE1 and TLE3 consistently produced significant PROP1 activation of a minimal promoter in the absence repressive effects. HESX1( ⌬1–50), which harbors a de- of HESX1. The same results were obtained in ␣T3-1 cells, letion of the eh1 domain, was used to test TLE1 and a gonadotrope-like cell line, and with the PRDQ3 and TLE3 corepression and recruitment through the eh1 do- P3 6E4 reporters constructs (data not shown). main. Repression was significantly impaired in the ab- sence of the eh1 domain ( P Ͻ 0.001). The same results TLE repression of PROP1 likely involves were obtained with the multimerized consensus sequence protein-protein interaction reporter plasmids P3 6E4 and PRDQ3 (data not shown). To determine the mechanism by which TLE factors repress PROP1 activity, we performed EMSA to ascertain PROP1 repression by TLE1 and TLE3 occurs whether the TLE factors could bind the PROP1 consensus independently of HESX1 recognition sequence (PRDQ DNA element). We did not To evaluate the requirement of HESX1 for recruitment observe any interaction between this DNA element and of the TLE/Gro family proteins (TLE1 and TLE3), TLE1 or TLE3, whereas PROP1 bound the element as PROP1 activity was analyzed in the absence of HESX1 expected (Fig. 3). To determine whether protein interac- and the presence of TLE1 or TLE3 individually. PROP1 tions might be involved, we transfected expression vec- expression vector was cotransfected with vectors express- tors for PROP1-myc alone, TLE1 alone, and PROP-myc ing TLE1 or TLE3 in 293T cells with the three previously together with TLE1, prepared extracts, immunoprecipi- described reporters coupled to luciferase and varying tated with anti-myc antibody, and probed Western blots amounts of empty vector to keep the total amount of with the anti-myc antibody followed by the anti-TLE an- transfected DNA constant. With the POU1F1 promoter, tibody (Fig. 3). We observed a band corresponding to the addition of 50 ng TLE1 or TLE3 expression vectors TLE1 that coimmunoprecipitated with PROP1, suggest- repressed PROP1 activation by 27 and 37%, respectively ing that there could be an interaction between PROP1 and (Fig. 2). Dose-response curves show that transfecting in- TLE factors. This interaction may be weak and/or tran- 4 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 only Tg (Cga -Tle3 ). The level of transgene expression in Tg (Cga -Tle3 ), Tg (Cga -Hesx1 ) double-transgenic embryos at e18.5 was analyzed by in situ hybridization using a transgene-specific Prm1 probe (Fig. 4, A–C). Strong Prm1 expression was detected in the ventral cells of double-transgenic embryos, whereas no expression was present in nontransgenic controls. HESX1 and TLE3 both contributed substantially to the overall level of transgene FIG. 3. TLE interaction with PROP1 DNA. A, Radiolabeled PRDQ expression. Transgenics have detectable Hesx1 mRNA and oligonucleotide was incubated with TLE1, TLE3, and/or PROP1 proteins produced in reticulocyte lysates. Reticulocyte lysates with empty vector TLE3 protein in the ventral cells of the anterior pituitary (EV) served as a negative control. The samples were separated by gel gland, whereas nontransgenic littermate controls lack Hesx1 electrophoresis and autoradiographed. PROP1 bound the probe in this mRNA and restrict TLE3 protein to the dorsal region of the EMSA ( bar ). B, TLE1 and/or PROP1-myc were transfected in 293T cells (5 g each in the cotransfection of TLE1 and PROP1, 10 g for pituitary gland (Fig. 4, D–I). Six fetuses expressed the transfection with each individual vector). Cell lysates were prepared transgenes of 24 double transgenics. and subjected to immunoprecipitation with anti-myc epitope The consequence of inappropriate TLE3 and HESX1 antibodies, followed by SDS-PAGE. Lysates, One eighth of each input lysate, collected before incubation with antibodies, was subjected to expression in the gonadotrophs and thyrotrophs was an- gel electrophoresis. After transfer to nitrocellulose, Western blotting alyzed in the highest expressing double transgenics. The (WB) was performed with either anti-myc or anti-Gro/TLE (pan-TLE) differentiation of gonadotrophs and thyrotrophs was antibodies. Immuno-precipitates, Coimmunoprecipitation was blocked in Tg (Cga -Tle3 ), Tg (Cga -Hesx1 ) double-trans- performed with the anti-myc antibody. Western blot was performed with myc antibody ( left ) and pan-TLE antibody. PROP1 migrates at 25 genic embryos. Immunohistochemistry using antibodies kDa, TLE at 95 kDa, and the heavy chain (HC) at 66 kDa. Endogenous against TSH, LH, and FSH readily detected positive cells TLE is detectable in lysates from cells not transfected with TLE1. in nontransgenic controls; however, no immunopositive Positions of migration of the IgG heavy chain (HC) are indicated. Positions of size standards are indicated in kilodaltons. cells were detected in sections from four of four express- ing, double-transgenic embryos (Fig. 4, J–R), suggesting that transgene expression blocked final differentiation of sient given the intensity of the TLE band relative to the amount of DNA transfected. both thyrotrophs and gonadotrophs. Somatotrophs, lac- totrophs, and corticotrophs were appropriately repre- Ectopic TLE3 and HESX1 expression represses sented in double-transgenic embryos (data not shown). differentiation in transgenic mice The presence of the earliest gonadotroph marker, the nu- The functional consequences of ectopic expression of clear hormone receptor SF1 (NR5A1) in Tg (Cga -Tle3 ), the repressor Hesx1 and a corepressor Tle3 in pituitary Tg (Cga -Hesx1 ) double-transgenic embryos, together with development were determined through temporal and spa- the deficiency in LH and FSH expression, suggests that the tial misexpression in transient transgenic mice. Transgene gonadotroph cell lineage was determined but not fully constructs were designed by placing either mouse TLE3 differentiated (Fig. 4, S–U). Interestingly, chorionic go- or HESX1 coding sequences in frame downstream of 4.6 nadotropin alpha (CGA, also called ␣GSU) protein was kb of the mouse pituitary glycoprotein hormone ␣-sub- dramatically reduced in double-transgenic embryos com- unit promoter ( ␣GSU), officially named Cga , followed by pared with nontransgenic controls, suggesting the repres- the mouse protamine 1 ( Prm1 ) 3 Ј-untranslated region, sor/corepressor complex can negatively regulate endoge- splice, polyadenylylation, and termination sequences. The nous ␣GSU expression (Fig. 4, V–X), but not the well-characterized Cga promoter drives transient expression transgene itself. The most plausible explanation for this is in Rathke’s pouch and persistent, strong coexpression with the presence of an important, negative, regulatory se- Cga in thyrotrophs and gonadotrophs during development quence in or around the endogenous gene that is absent in and throughout adulthood (21). Tg (Cga -Tle3 ) and Tg (Cga - the 4.6 kb of Cga 5Ј-flanking sequence in the transgene. Hesx1 ) transgene constructs were both coinjected and singly Injection of only the Tg (Cga -Tle3 ) construct produced injected into fertilized eggs, and fetuses were harvested at 22 embryos at e18.5, 14% of which contained the trans- e18.5 and e14.5. The presence of each transgene was iden- gene, whereas injection of the Tg (Cga -Hesx1 ) construct tified in genomic DNA generated from either embryonic tail produced 66 embryos at e18.5, 8% of which contained or limb biopsies using PCR amplification with transgene- the transgene. Transgenic embryos expressing Tg (Cga - specific primers. Tle3 ) alone had appropriate gonadotroph and thyrotroph Coinjection of the transgene constructs produced a differentiation (Fig. 5, A–E). In contrast, transgenic em- total of 481 embryos with 8.5% containing both transgenes, bryos expressing Tg (Cga -Hesx1 ) alone exhibit a dramatic 2% containing only Tg (Cga -Hesx1 ), and 1% containing reduction in TSH- and LH-positive cells (Fig. 5, F–J). How- Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 5 trophs and thyrotrophs, although TLE3 may be required for inhibition of en- dogenous Cga . To determine whether the differen- tiation of gonadotrophs was delayed by expression of HESX1 and/or TLE3 transgenes, transient transgenics were analyzed at the normal timing of onset of SF1 protein expression, e14.5. High levels of ectopic TLE3 protein expres- sion were detected in the ventral cells of the developing anterior pituitary gland in three Tg (Cga -Tle3 ), Tg (Cga -Hesx1 ) double transgenics (Fig. 6, A–D). The presence of SF1 in double-transgenic e14.5 embryos demonstrates that an early differentiation step of the gona- dotroph cell lineage is not delayed (Fig. 6, E–H). A reduction in endogenous CGA protein levels was visualized in the highest expressing e14.5 double- transgenic embryos, consistent with data obtained from e18.5 double- transgenic embryos (Fig. 6, I–L). Other markers of cell differentiation, ISL1 and PITX2, were not changed between nontransgenic and double-transgenic littermates (Fig. 6, M–T). In vivo data support the role of HESX1 as a strong repressor that can be enhanced by the presence of at least two of the TLE/ groucho family members. Discussion Transcriptional repressors, activators, and cofactors work together to orches- FIG. 4. Ectopic, constitutive expression of HESX1 and TLE3 in transgenic embryos affects trate pituitary gland organogenesis. gonadotroph and thyrotroph differentiation. Levels of transgene expression were examined in This likely involves interplay between coronal sections of normal littermates and double-transgenic embryos at e18.5: Tg (Cga -Tle3 ), Tg (Cga -Hesx1 ). The first column (A, D, G, J, M, P, S, and V) display a representative PROP1, HESX1, SIX3, and other tran- nontransgenic littermate, whereas the second column (B, E, H, K, N, Q, T, and W) and the scription factors with coregulators third column (C, F, I, L, O, R, U, and X) demonstrate two independent double-transgenic from the TLE/groucho family: TLE1 embryos designated as ␣TLE3, ␣HESX1. Mouse protamine 1 (Prm1 ) mRNA expression was used to select transgenics with robust expression levels for further analysis (A–C), whereas the and TLE3 (22). We demonstrate func- expression of the individual transgene constructs was visualized by Hesx1 in situ hybridization tional interactions between a subset (D–F) and TLE3 immunohistochemistry (G–I). The arrows in H and I depict the ectopic of these proteins in tissue culture and expression of TLE3 in the ventral portion of the pituitary gland. The differentiation of thyrotrophs and gonadotrophs was analyzed using immunohistochemistry with antibodies also through misexpression of HESX1 directed against TSH  (J–L), LH  (M–O), FSH  (P–R), SF1 (S–U), and ␣GSU or CGA (V–X). and TLE3 together and singly in the con- text of pituitary development. ever, the expression of endogenous Cga was not affected. HESX1 is a strong repressor This suggests that the ectopic expression of HESX1 is HESX1 and TLE1 are coexpressed at e12.5 in the same sufficient to block terminal differentiation of gonado- region of the developing pituitary gland (Fig. 7) (9). TLE1 6 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 domain of HESX1 (HESX1I26T) exhibited evolving hy- popituitarism and no central nervous malformation. This mutation has no effect on DNA binding but impairs HESX1-mediated repression because of failure to recruit TLE1 (19). The HESX1-TLE1 complex is proposed to suppress PROP1-mediated transcription of Pou1f1 until -catenin displaces TLE1 (9). Are other genes natural targets of the HESX1-TLE1 repressor complex? These genes would be expected to be overexpressed in Hesx1 mutant mice. Foxd3 , Pax3 , Dmbx1 , and Sp5 expression and Wnt sig- naling are expanded and/or elevated in the anterior neural ridge of Hesx1 mutants (24, 25). Additional investigation of pituitary gene expression in Hesx1 mutants may be necessary to uncover the direct targets of HESX1-medi- ated repression. Mechanism of TLE-PROP1 action TLE1 and TLE3 have highly conserved protein struc- tures. The overall amino acid identity is 80% with 86% identity in the glutamine-rich region (Q) and 95% in the FIG. 5. Ectopic expression of HESX1 alone, but not TLE3, affects gonadotroph and thyrotroph differentiation. Levels of transgene WD repeat region (16). This conservation suggests a sim- expression were determined by TLE3 immunohistochemistry for Tg (Cga - ilar mechanism of action of TLE1 and TLE3. There are Tle3 ) or ␣GSU-TLE3 transgenic embryos and nontransgenic littermates at areas of overlapping expression in pituitary development e18.5 (A and B) or Hesx1 in situ hybridization for Tg (Cga -Hesx1 ) or ␣GSU- HESX1 transgenic embryos and nontransgenic littermates (nontg) at e18.5 (Fig. 7), and we observe similar effects of TLE1 and TLE3 (F and G). TSH  and LH  immunohistochemistry is used to screen for in transfection studies and transgenic mice. TLE3 and differentiated gonadotrophs and thyrotrophs in Tg (Cga -Tle3 ) embryos TLE1 repress PROP1 activation of the POU1F1 promoter (C and D) and Tg (Cga -Hesx1 ) embryos (H and I). CGA or ␣GSU protein is present in both Tg (Cga -Tle3 ) and Tg (Cga -Hesx1 ) single-transgenic independent of HESX1. The mechanism of this TLE-me- embryos (E and J). diated repression likely involves protein-protein interac- tion because TLE does not bind the paired homeodomain binds HESX1 and enhances its repressor activity in cell recognition sequence in EMSA, but it coimmunoprecipi- culture and in transgenic mice (9, 19, 20). We confirmed tates with PROP1 when coexpressed in 293T cells. Al- that HESX1 inhibits PROP1-mediated transactivation in though another study did not detect coimmunoprecipita- cell culture and that TLE1 acts as a corepressor, enhanc- tion, the protein interaction may be weak or transient (9). ing this inhibition. The effects we observed are of the same However, the interaction of PROP1 and TLE is consistent magnitude as other reports. We discovered that ectopic with chromatin immunoprecipitation studies that show HESX1 alone is a strong repressor of differentiation in occupancy of the Pou1f1 early enhancer by TLE and transgenic mice, blocking expression of Tshb and Lhb PROP1 at e12.5 after HESX1 is released (20). This sup- almost completely and modestly repressing Cga expres- ports the idea that PROP1 and TLE interaction could sion. This contrasts with the idea that ectopic TLE1 ex- have biological significance. pression is necessary for robust repression of differentia- TLE factors interact with transcription factors that tion by ectopic HESX1 transgene expression (9). contain WRPW or eh1 motifs (17, 22). Analysis of the The ability of HESX1 to function as a strong repressor PROP1 peptide sequence identified only weak alignment in vitro and in vivo is supported by mutations identified in with the WRPW and eh1 motifs, suggesting if there is a human patients. Two mutations that lead to hypopitu- direct interaction between TLE and PROP1, then other itarism alter HESX1 interactions with TLE factors. A pa- unknown sequences might be involved in mediating the tient with septo-optic dysplasia and hypopituitarism is interaction (26). Alternatively, the protein-protein inter- heterozygous for 1684delG in HESX1 that creates in- action may be indirect. Although the 293T cells do not creased DNA binding and more potent repression of express endogenous HESX1, we cannot rule out the pos- PROP1-mediated gene transcription (23). Transgenic sibility that another protein is present in the coimmuno- mice that misexpress HESX1 have similar features to this precipitated complex or involved in the TLE-mediated patient. A patient with a mutation in the eh1 repression repression of PROP1 activation. Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 7 FIG. 6. Initiation of gonadotroph differentiation in double-transgenic embryos. The three highest expressing Tg (Cga -Tle3 ), Tg (Cga -Hesx1 ) double- transgenic embryos (designated ␣TLE3, ␣HESX1 double tgs) were selected based on ectopic immunohistochemical staining for TLE3 at e14.5 in the ventral cells of Rathke’s pouch in sagittal sections ( columns beginning with B, C, and D; gradation bar indicates decreasing transgene expression levels) compared with a nontransgenic littermate ( column beginning with A). SF1 (NR5A1) immunohistochemical staining detected both mature and pre-gonadotrophs (E–H, arrows ). Immunohistochemistry using antibodies against ␣GSU (CGA) (I–L) demonstrated a decrease in expression correlating with level of transgene expression (area of expression outlined ). There was no difference in protein levels of ISL1 (M–P) or PITX2 (Q–T) in transgenics and nontransgenic littermates. Role for ectopic TLE3 and HESX1 expression in suppressed by HESX1 in the presence or absence of TLE3, Prop1 mutants? but TLE3 alone has no effect in this context. Endogenous Prop1 mutants express both Tle3 and Hesx1 ectopi- CGA expression is suppressed by TLE3 and HESX1 to- cally, in overlapping areas, which could contribute to the gether but not by HESX1 alone. These results suggest that pituitary developmental abnormalities characteristic of ectopic HESX1 and TLE3 expression in Prop1 mutants Prop1 mutants (Fig. 7). This idea is supported by our could contribute to failed differentiation. The failed thy- demonstration that TLE3 can enhance the repression me- rotroph differentiation in Prop1 mutants is attributable diated by HESX1 in cell culture and transgenic mice. Ter- to the requirement of PROP1 for Pou1f1 expression, minal differentiation of thyrotrophs and gonadotrophs is which in turn is essential for caudomedial expression of 8 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 FIG. 7. Expression domains of transcription factors, corepressors, and hormone subunits in developing pituitary gland and surrounding tissues. Rathke’s pouch and the overlying neural ectoderm are illustrated diagrammatically in the sagittal plane in normal mice at e12.5 and e14.5 and in Prop1 mutants at e14.5. The diagonal lines mark domains that are defined by differences in gene expression. Tle3 and Tcf7l2 (TCF4E) are expressed in the lower domain of the ventral diencephalon. Rathke’s pouch grows significantly between e12.5 and e14.5 and becomes dysmorphic in Prop1 mutants, probably because of an inability to produce a Notch2-positive transitional zone intermediate between the dorsal proliferating cells and the ventral differentiating cells. Cga, The ␣-subunit of the glycoprotein hormones LH, FSH, and TSH, is expressed in the differentiated cells of Rathke’s pouch that expand beyond the rostral tip at e12.5 to include more caudomedial cells at e14.5 in normal mice but not Prop1 mutants ( gray ). The diagram illustrates that 1) HESX1 and TLE1 could repress PROP1 activity at e12.5, 2) ectopic HESX1 and TLE3 expression might contribute to the Prop1 mutant phenotype, and 3) TLE1 and TLE3 likely have interacting partners besides PROP1 and HESX1. Tshb . However, the gonadotropin deficiency in humans 32). This suggests that TLE3 may be involved in regu- with PROP1 mutations is poorly understood, and it is one lating the differentiation of cells in the dorsal aspect of of the clinical hallmarks used to distinguish patients with Rathke’s pouch and/or regulating proliferation. Defi- PROP1 and POU1F1 mutations. Our results suggest that ciency of either HES1 or TBX19 (TPIT) permits prema- ectopic TLE3 and HESX1 expression may contribute to ture differentiation of these cells into GH- or LH-express- the gonadotropin deficiency in patients with PROP1 mu- ing cells, respectively, so they clearly have the capacity for tations, although other explanations, such as depletion of differentiation into a variety of cell types unless appropri- precursors, and bystander effects causing cell death, could ately regulated (33–37). Immunohistochemistry and in contribute. situ hybridization studies reveal that Six6 and Nkx3.1 have in a similar temporal and spatial expression pattern What is the normal function of TLE3? to Prop1 (8, 38) (our unpublished observations). In addi- Tle3 is expressed in the prospective intermediate lobe tion, Six3 , Tcf3 , and Hes1 have also been identified in the from e14.5–e16.5 around the time that Hesx1 and Prop1 developing anterior pituitary gland at e14.5 (20, 34, 35, transcripts have normally waned (Fig. 7). The transgenic 39, 40). Any of the genes encoding interactors with TLE1 expression studies were aimed to assess the consequences and TLE3 are candidate genes for screening patients with of ectopic Tle3 and Hesx1 expression in Prop1 mutants. pituitary abnormalities. The neutral effect of TLE3 on differentiation in Tg (Cga - Tle3 ) single transgenics implies that there were no endog- enous repressors that could interact with TLE3 in the Materials and Methods ventral aspect of Rathke’s pouch to suppress differentia- tion in this context. The endogenous partners for TLE3, Generation of transient transgenics however, are likely coexpressed dorsally in the dividing The open reading frame of mouse Tle3 was amplified with forward primer 5 Ј-gccaccatggactacaaggacgacgatgacaagatg- cells. tatccgcaaggcag-3 Ј and reverse primer 5 Ј-tcagtagatgacctcgtaaactgt- The Nkx , Hes , Runx , Lef/Tcf , and Six gene families ggccttctt-3 Ј using an e14.5 Rathke’s pouch cDNA library as the encode proteins with a WRPW or eh1 motif that can template (29, 30). The Tle3 open reading frame was cloned interact with the Gro/TLE family (27, 28). We identified into the TOPO-TA cloning vector (Invitrogen, Carlsbad, CA) members of some of these families in cDNA libraries of using the manufacturer’s protocol, and its identity was con- microdissected Rathke’s pouch at e12.5 and e14.5 (29, firmed by DNA sequence analysis. Tle3 was placed downstream of the previously characterized mouse ␣GSU promoter ( Cga ) 30). These include Nkx3.1 , Nkx2.4 , Runx1 , Hes6 , and and upstream of a mouse Protamine 1 (Prm1 ) 3 Ј-untranslated Otx2 and three members of the Six gene family . Six3 is and 3 Ј-flanking sequences, which contain an intron and poly- coexpressed with Tle3 and regulates proliferation (22, 31, adenylation sequences, using the Hin dIII and Eco RV sites (21). Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 9 The identity of the Cga -Tle3 transgene construct was confirmed Immunohistochemistry by DNA sequence analysis. Before microinjection, the transgene Sagittal (e14.5) and coronal (e18.5) sections of paraffin- insert was purified from the vector sequences after digestion embedded embryos were used in the immunohistochemistry exper- with the restriction enzymes Asp 718 and Not I. iments. Slides were boiled for 10 min in 10 m M citric acid if destined The open reading frame of mouse Hesx1 was amplified with for analysis with PITX2, ISL1, TLE3, or NR5A1 antibodies. All forward primer 5 Ј-gccaccatggactacaaggacgacgatgacaaggtc- slides were blocked with a 1:1 mixture of methanol/3% hydrogen tcccagccttcgggaaggtgctcag-3 Ј and reverse primer 5 Ј-ctatttcagaa- peroxide to inactivate endogenous peroxidases. Rabbit anti-PITX2 gatctgggttgaagggttttttcgccattagaaa-3 Ј using cDNA from an e14.5 (1:400; Dr. Tord Hjalt, Lund University, Sweden), mouse anti- Prop1 df/df Rathke’s pouch library as a template (29, 30). The ISL1 (1:600; Developmental Studies Hybridoma Band, University Hesx1 open reading frame was cloned in the pGEM T-easy cloning of Iowa), and rabbit anti-TLE3 (1:150; Millipore, Billerica, MA) vector (Promega, Madison, WI), and the integrity was confirmed were incubated overnight at 4 C. The following day, the sections by sequence analysis. The Hesx1 open reading frame was excised were washed and then incubated with either a biotinylated anti- with Eco RI and subcloned into the pBluescript SK ϩ (Stratagene, mouse IgG (1:200, Vectastain Mouse on Mouse kit; Vector Labo- La Jolla, CA). The completed transgene construct was generated by ratories, Burlingame, CA) or a biotinylated antirabbit IgG (1:200; cloning the Ϫ4.6-kb mouse ␣GSU promoter and enhancer se- Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). quences into the Kpn I and Hin dIII site upstream of Hesx1 , and the The antibodies were detected using the reagents and protocols pro- mouse Prm1 sequences described above were cloned downstream vided in the tyramide signal amplification fluorescein kit into the Bam HI/ Xba I sites. The identity of the mouse Cga -Hesx1 (PerkinElmer, Boston, MA). Mouse anti-NR5A1 (1:1500; Dr. transgene construct was confirmed by DNA sequence analysis. The Ken-ichirou Morohashi, National Institute for Basic Biology, Oka- transgene construct was released from the vector backbone using zaki, Japan), monkey anti-GH [1:1000; National Hormone Pitu- Bss HII. Purified DNA from both transgene constructs was either itary Program (NHPP), Torrance, CA], guinea pig anti-LH (1:500; comicroinjected or singly microinjected into fertilized eggs by The National Hormone Pituitary Program, NHPP), rabbit anti-TSH University of Michigan Transgenic Animal Model Core. Fertilized (1:1000; NHPP), rabbit anti-proopiomeanocortin (1:1000; eggs were obtained from mating (C57BL/6 ϫ SJL) F1 female and NHPP), rabbit anti-FSH (1:400; NHPP), and rabbit anti- ␣GSU male mice. Pronuclear microinjection was performed, and injected (1:200; NHPP) were also incubated for 2 h at room temperature. eggs were implanted into CD-1 surrogate mothers at the two-cell Biotinylated secondary antibodies were used in conjunction with avidin and biotinylated peroxidase (Vectastain Mouse on Mouse stage. kit, Vectastain Rabbit Elite kit, Vectastain Human Elite kit, and Vectastain Guinea Pig kit; Vector). With this experiment, diami- Embryo genotyping nobenzidine was used as the chromogen (Sigma, St. Louis, MO). Embryos were collected at d 14.5 and 18.5 of gestation All immunostaining experiments were done in duplicate in any one (e14.5 and e18.5). Embryonic d 0.5 was designated as the day experiment, and experiments were repeated to sample regions throughout the pituitary gland. the microinjected fertilized eggs were surgically transferred to pseudo-pregnant foster females. Genomic DNA extracted from either tail or limb biopsies was used to genotype the embryos by Plasmid constructs for transfection PCR amplification with transgene-specific primers. The transgenes Human PROP1 cDNA cloned into pCDNA3.1 was a gift from were amplified using a forward primer within the mouse Cga pro- Dr. Simon Rhodes (Indiana University School of Medicine, India- moter, 5 Ј-tcaactttcaggatgttttgtgtaa-3 Ј, and reverse primers specific napolis, IN) (41). Human HESX1 cDNA was generated from to either Tle3 , 5 Ј-cccgatgatggcgttcaa-3 Ј, or Hesx1 , 5 Ј- clones encoding GAL4-HESX1 fusion proteins (42), released with taggggtgggttgccacc-3 Ј. Products of 740 bp for the Cga -Tle3 trans- Eco RI and Bam HI and subcloned into pCDNA3.1. GAL4-HESX1 ⌬ gene and 558 bp for the Cga -Hesx1 transgene were amplified at 94 WT, GAL4-HESX1( 1-50), POU1F1 promoter coupled to lucif- C for 4 min followed by 30 cycles of 94 C for 30 sec, 55 C for 30 erase and (P3) 6E4 luciferase plasmids were gifts from Dr. Mehul sec, and 72 C for 30 sec and ending in 72 C for 10 min and Dattani (Biochemistry, Endocrinology, and Metabolism Unit, In- visualized with ethidium bromide on a 1% agarose gel. stitute of Child Health, London, UK) (19). Human pCMV2- FLAG-GroTLE1 and TLE3 cloned into the pCIneo mammalian expression vector were gifts from Dr. Stefano Stifani (Center for In situ hybridization Neuronal Survival, Montreal Neurological Institute, McGill Uni- Embryos were fixed in buffered 4% paraformaldehyde for versity, Montreal, Canada) (43). pCDNA3.1 was used as a nega- 90 min (e14.5) or overnight (e18.5). After fixation, the embryos tive control vector (Invitrogen). pRL-TK plasmid DNA (thymidine were washed in PBS and dehydrated through a series of ethanol kinase promoter-driven renilla luciferase; Promega) was used as an solutions. The embryos were embedded in paraffin, sectioned to internal control for the transfection. 6 m thickness, and processed as described below. In situ hy- bridization was performed using riboprobes labeled with Cell culture, transient transfection, and dual digoxigenin (Roche, Indianapolis, IN) (15). Tge 394-bp Hesx1 luciferase reporter assay cDNA clone was a generous gift from Dr. Paul Thomas (Bris- The ␣T3-1 (mouse pituitary pre-gonadotroph) (45) was a bane, Australia). The Bam HI linearized fragment was tran- gift from Dr. Pamela Mellon (University of San Diego, La Jolla, scribed with T3 polymerase and hybridized at a 1:100 dilution CA). These and 293T cells (human embryonic kidney) were overnight at 53 C. A 511-bp Prm1 probe was generated from the maintained at 37 C, 5% CO 2 in DMEM (Invitrogen) supple- 3Ј end of the transgene construct. The Bam HI linearized frag- mented with 10% heat-inactivated fetal bovine serum (Hyclone, ment was transcribed with T3 polymerase, diluted 1:100, and Logan, UT). Cells were plated onto 24-well plastic plates (Fisher hybridized to tissue sections overnight at 53 C. Scientific, Fair Lawn, NJ) at a density of 3 ϫ 10 4 (293T) and 6 ϫ 10 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 10 4 (␣T3-1) such that the cells were 50–80% confluent on the Acknowledgments day of transfection. Three reporter constructs coupled to lucif- erase were used for transfections: a 15-kb mouse Pou1f1 pro- We acknowledge Wanda Filipiak, D.V.M.; Dr. Maggie Van moter and two artificial promoters constructed with three Keuren; and Galina Gavrilina, M.S., for the preparation of (PRDQ3) or six (P3 6E4) paired homeodomain transcription fac- transgenic mice and Dr. Thomas L. Saunders of the Transgenic tor binding sites. A DNA cocktail consisting of a combination of Animal Model Core in the University of Michigan’s Biomedical the following was transfected: 100 ng of the reporter construct, Research Core Facilities. We thank Mary Anne Potok for assis- 10 ng pRL-TK renilla as an internal control, 50 ng PROP1, 50 tance with analysis of transgenic animals. We thank the following ng HESX1, 50 ng TLE1, 50 ng TLE3, and pcDNA 3.1 empty individuals and programs for providing reagents: Drs. Mehul expression vector ranging from 50–200 ng to normalize up to Dattani, Tord Hjalt, Pamela Mellon, Ken-ichirou Morohashi, 310 ng/well. Optimal DNA concentrations were determined by Takashi Okamoto, Simon Rhodes, Paul Thomas, Stefano Sti- titrating increasing amounts of HESX1 (25, 50, 75, and 100 ng) and constant Prop1 at 50 ng (data not shown). A dose-response fani, and the National Hormone Pituitary Program. We thank assay was carried out by transfecting variable quantities of Drs. Bob Lyons and the University of Michigan DNA Sequencing TLE1 or TLE3 individually with 50 ng PROP1 expression vec- Core, Stefano Stifani of McGill University, Alexander Augusto tor. The range of corepressor expression vector amounts were Lima Jorge (Clinicas’ Hospital-FMUSP, Brazil), Christopher Krebs 50, 100, 150, 200, and 250 ng of TLE1 or TLE3 cotransfected (University of Michigan, Department of Human Genetics), and with PROP1, and pcDNA 3.1 empty expression vector ranging Travis Maure (University of Michigan, Department of Molecular from 0–200 ng to generate a total of 310 ng/well. DNA was and Integrative Physiology) for scientific contributions and valu- transfected in 500 l serum-free DMEM (Invitrogen) into cul- able discussions. ture cells using Fugene 6 (Roche) at a ratio of 3:1 according to the manufacturer’s protocol. For each transfection experiment, the Address all correspondence and requests for reprints to: Sally pcDNA 3.1 empty expression vector was used as a negative con- A. Camper, 4909 Buhl Building, 1241 Catherine Street, Ann trol, and measurement of the basal level of luciferase activity. A Arbor, Michigan 48109-5618. E-mail: [email protected]. total of 300 ng DNA was transfected to keep the total amount of This work was supported by National Institutes of Health DNA transfected consistent. Forty-eight hours after transfection, (NIH) R37HD30428 and R01HD34283 (S.A.C.); The Endo- dual-luciferase assays (Promega) were performed according to the crine Society International Scholars Program (L.R.C.); Novo- manufacturer’s protocol and quantified using the Lmax Micro- Nordisk, Societe Franc¸aise d’endocrinologie (Novartis-Ipsen), plate Illuminometer (Molecular Devices, Sunnyvale, CA) with the and ADEREM (F.C.); F32 HD046300 (B.S.E.); NIH Center Grants SOFT max Pro software (Molecular Devices). All assays were done CA46592, AR20557, DK34933, DK34933, and P30AG013283 in triplicate, and the results were repeated at least three times. (University of Michigan Transgenic Animal Model Core); and Results were averaged and expressed as fold change of the percent Grant 085P1000815 (University of Michigan Center for Orga- activity of the experimental plasmids over basal. nogenesis, the Michigan Economic Development Corporation and the Michigan Technology Tri-Corridor). Transgenic animal analysis was done by M.L.B. Cell culture EMSA and coimmunoprecipitation studies studies with the (P3) 6E4 promoter were done by L.R.C. and Gel mobility shift assays were performed to search for TLE repeated by B.S.E., whereas the PRDQ3 and Pou1f1 promoter DNA interactions on a radiolabeled annealed synthetic PROP1 transfections were done by F.C. The EMSA and coimmunopre- recognition element (PRDQ3 oligonucleotide). PRDQ3 oligo- cipitation were by F.C. The paper was written by M.L.B., F.C., 32 nucleotide was labeled with [ P]deoxy-CTP. PROP1, TLE1, L.R.C., and S.A.C. and TLE3 were transcribed and translated using the TNT-cou- Current address for L.R.C.: Unidade de Endocrinologia do pled reticulocyte lysate system with T7 polymerase (Promega). Desenvolvimento, Laborato´rio de Hormoˆnios e Gene´tica Mo- The protein-DNA complexes were analyzed by electrophoresis lecular LIM/42, Disciplina de Endocrinologia, Hospital das through a 5% polyacrylamide gel containing 2.5% glycerol in a Clínicas da Faculdade de Medicina da Universidade de Sa˜o ϫ 0.5 Tris-borate buffer at 4 C. Paulo, Brasil. For the coimmunoprecipitation study, human HEK 293 cells Disclosure Summary: None of the authors have conflicts of were transfected with PROP1 and/or TLE1 (10 g total DNA, interests to disclose. with 10 g PROP1 alone, 10 g TLE1 alone, or 5 g PROP1 and 5 g TLE1). Whole-cell lysates were prepared and subjected to immunoprecipitation as described previously (43), followed References by Western blotting with antibodies against the myc epitope (Invitrogen) or the Gro/TLE proteins (pan-TLE antibodies, a 1. Takuma N, Sheng HZ, Furuta Y, Ward JM, Sharma K, Hogan BL, generous gift from Dr. Stefano Stifani) (44). The blot was visu- Pfaff SL, Westphal H, Kimura S, Mahon KA 1998 Formation of alized with the Odyssey infrared imaging system (LI-COR Bio- Rathke’s pouch requires dual induction from the diencephalon. sciences, Lincoln, NE). Development 125:4835–4840 2. Kelberman D, Dattani MT 2007 Hypothalamic and pituitary de- velopment: novel insights into the aetiology. Eur J Endocrinol Statistical analysis 157(Suppl 1):S3–S14 3. Deladoe¨y J, Flu¨ck C, Bu¨yu¨kgebiz A, Kuhlmann BV, Eble´ A, SigmaStat statistical software version 2.0 (SPSS Inc., Chicago, Hindmarsh PC, Wu W, Mullis PE 1999 “Hot spot” in the PROP1 gene IL) was used for data analysis. Statistically significant differences responsible for combined pituitary hormone deficiency. J Clin Endo- between the input groups were considered as P Ͻ 0.001 in the t test crinol Metab 84:1645–1650 and one-way ANOVA. 4. Dattani MT, Martinez-Barbera JP, Thomas PQ, Brickman JM, Gupta Mol Endocrinol, April 2010, 24(4):0000–0000 mend.endojournals.org 11 R, Mårtensson IL, Toresson H, Fox M, Wales JK, Hindmarsh PC, Enhancer-mediated high level expression of mouse pituitary gly- Krauss S, Beddington RS, Robinson IC 1998 Mutations in the ho- coprotein hormone ␣-subunit transgene in thyrotropes, gonado- meobox gene HESX1/Hesx1 associated with septo-optic dysplasia in tropes, and developing pituitary gland. Mol Endocrinol 8:1420 – human and mouse. Nat Genet 19:125–133 1433 5. Sornson MW, Wu W, Dasen JS, Flynn SE, Norman DJ, O’Connell SM, 22. Gaston-Massuet C, Andoniadou CL, Signore M, Sajedi E, Bird S, Gukovsky I, Carrie`re C, Ryan AK, Miller AP, Zuo L, Gleiberman AS, Turner JM, Martinez-Barbera JP 2008 Genetic interaction between Andersen B, Beamer WG, Rosenfeld MG 1996 Pituitary lineage deter- the homeobox transcription factors HESX1 and SIX3 is required mination by the Prophet of Pit-1 homeodomain factor defective in for normal pituitary development. Dev Biol 324:322–333 Ames dwarfism. Nature 384:327–333 23. Cohen RN, Cohen LE, Botero D, Yu C, Sagar A, Jurkiewicz M, 6. Gage PJ, Roller ML, Saunders TL, Scarlett LM, Camper SA 1996 Radovick S 2003 Enhanced repression by HESX1 as a cause of Anterior pituitary cells defective in the cell-autonomous factor, df, hypopituitarism and septooptic dysplasia. J Clin Endocrinol Metab undergo cell lineage specification but not expansion. Development 88:4832–4839 122:151–160 24. Andoniadou CL, Signore M, Sajedi E, Gaston-Massuet C, Kelberman 7. Raetzman LT, Ross SA, Cook S, Dunwoodie SL, Camper SA, D, Burns AJ, Itasaki N, Dattani M, Martinez-Barbera JP 2007 Lack of Thomas PQ 2004 Developmental regulation of Notch signaling the murine homeobox gene Hesx1 leads to a posterior transformation genes in the embryonic pituitary: Prop1 deficiency affects Notch2 of the anterior forebrain. Development 134:1499–1508 expression. Dev Biol 265:329–340 25. Sajedi E, Gaston-Massuet C, Signore M, Andoniadou CL, Kelberman 8. Li S, Crenshaw 3rd EB, Rawson EJ, Simmons DM, Swanson LW, D, Castro S, Etchevers HC, Gerrelli D, Dattani MT, Martinez-Barbera Rosenfeld MG 1990 Dwarf locus mutants lacking three pituitary JP 2008 Analysis of mouse models carrying the I26T and R160C cell types result from mutations in the POU-domain gene Pit-1. substitutions in the transcriptional repressor HESX1 as models Nature 347:528–533 for septo-optic dysplasia and hypopituitarism. Dis Model Mech 9. Dasen JS, Barbera JP, Herman TS, Connell SO, Olson L, Ju B, 1:241–254 Tollkuhn J, Baek SH, Rose DW, Rosenfeld MG 2001 Temporal 26. Jennings BH, Pickles LM, Wainwright SM, Roe SM, Pearl LH, regulation of a paired-like homeodomain repressor/TLE corepres- Ish-Horowicz D 2006 Molecular recognition of transcriptional re- sor complex and a related activator is required for pituitary orga- pressor motifs by the WD domain of the Groucho/TLE corepressor. nogenesis. Genes Dev 15:3193–3207 Mol Cell 22:645–655 10. Gage PJ, Lossie AC, Scarlett LM, Lloyd RV, Camper SA 1995 Ames 27. Goldstein RE, Cook O, Dinur T, Pisante´A, Karandikar UC, Bidwai dwarf mice exhibit somatotrope commitment but lack growth hormone- A, Paroush Z 2005 An eh1-like motif in odd-skipped mediates releasing factor response. Endocrinology 136:1161–1167 recruitment of Groucho and repression in vivo. Mol Cell Biol 25: 11. Gage PJ, Brinkmeier ML, Scarlett LM, Knapp LT, Camper SA, 10711–10720 Mahon KA 1996 The Ames dwarf gene, df, is required early in 28. Kobayashi M, Goldstein RE, Fujioka M, Paroush Z, Jaynes JB 2001 pituitary ontogeny for the extinction of Rpx transcription and ini- Groucho augments the repression of multiple Even skipped target tiation of lineage-specific cell proliferation. Mol Endocrinol 10: genes in establishing parasegment boundaries. Development 128: 1570–1581 1805–1815 12. Wu W, Cogan JD, Pfa¨ffle RW, Dasen JS, Frisch H, O’Connell SM, 29. Carninci P, Waki K, Shiraki T, Konno H, Shibata K, Itoh M, Flynn SE, Brown MR, Mullis PE, Parks JS, Phillips 3rd JA, Rosenfeld Aizawa K, Arakawa T, Ishii Y, Sasaki D, Bono H, Kondo S, Sugahara MG 1998 Mutations in PROP1 cause familial combined pituitary Y, Saito R, Osato N, Fukuda S, Sato K, Watahiki A, Hirozane- hormone deficiency. Nat Genet 18:147–149 Kishikawa T, Nakamura M, Shibata Y, Yasunishi A, Kikuchi N, 13. Tatsumi K, Miyai K, Notomi T, Kaibe K, Amino N, Mizuno Y, Yoshiki A, Kusakabe M, Gustincich S, Beisel K, Pavan W, Aidinis Kohno H 1992 Cretinism with combined hormone deficiency V, Nakagawara A, Held WA, Iwata H, Kono T, Nakauchi H, Lyons caused by a mutation in the PIT1 gene. Nat Genet 1:56–58 P, Wells C, Hume DA, Fagiolini M, Hensch TK, Brinkmeier M, 14. Brinkmeier ML, Potok MA, Cha KB, Gridley T, Stifani S, Meeldijk Camper S, Hirota J, Mombaerts P, Muramatsu M, Okazaki Y, J, Clevers H, Camper SA 2003 TCF and Groucho-related genes Kawai J, Hayashizaki Y 2003 Targeting a complex transcriptome: influence pituitary growth and development. Mol Endocrinol 17: the construction of the mouse full-length cDNA encyclopedia. 2152–2161 Genome Res 13:1273–1289 15. Douglas KR, Brinkmeier ML, Kennell JA, Eswara P, Harrison TA, 30. Brinkmeier ML, Davis SW, Carninci P, MacDonald JW, Kawai J, Patrianakos AI, Sprecher BS, Potok MA, Lyons Jr RH, MacDougald Ghosh D, Hayashizaki Y, Lyons RH, Camper SA 2009 Discovery of OA, Camper SA 2001 Identification of members of the Wnt signal- transcriptional regulators and signaling pathways in the developing ing pathway in the embryonic pituitary gland. Mamm Genome pituitary gland by bioinformatic and genomic approaches. Genomics 12:843–851 93:449–460 16. Chen G, Courey AJ 2000 Groucho/TLE family proteins and tran- 31. Zhu CC, Dyer MA, Uchikawa M, Kondoh H, Lagutin OV, Oliver scriptional repression. Gene 249:1–16 G 2002 Six3-mediated auto repression and eye development re- 17. Buscarlet M, Stifani S 2007 The ‘Marx’ of Groucho on develop- quires its interaction with members of the Groucho-related family ment and disease. Trends Cell Biol 17:353–361 of co-repressors. Development 129:2835–2849 18. Wilson D, Sheng G, Lecuit T, Dostatni N, Desplan C 1993 Coop- 32. Li X, Oghi KA, Zhang J, Krones A, Bush KT, Glass CK, Nigam SK, erative dimerization of paired class homeo domains on DNA. Genes Aggarwal AK, Maas R, Rose DW, Rosenfeld MG 2003 Eya protein Dev 7:2120–2134 phosphatase activity regulates Six1-Dach-Eya transcriptional ef- 19. Carvalho LR, Woods KS, Mendonca BB, Marcal N, Zamparini AL, fects in mammalian organogenesis. Nature 426:247–254 Stifani S, Brickman JM, Arnhold IJ, Dattani MT 2003 A homozy- 33. Pulichino AM, Vallette-Kasic S, Tsai JP, Couture C, Gauthier Y, gous mutation in HESX1 is associated with evolving hypopituitar- Drouin J 2003 Tpit determines alternate fates during pituitary cell ism due to impaired repressor-corepressor interaction. J Clin Invest differentiation. Genes Dev 17:738–747 112:1192–1201 34. Raetzman LT, Cai JX, Camper SA 2007 Hes1 is required for pitu- 20. Olson LE, Tollkuhn J, Scafoglio C, Krones A, Zhang J, Ohgi KA, Wu itary growth and melanotrope specification. Dev Biol 304:455–466 W, Taketo MM, Kemler R, Grosschedl R, Rose D, Li X, Rosenfeld 35. Zhu X, Zhang J, Tollkuhn J, Ohsawa R, Bresnick EH, Guillemot F, MG 2006 Homeodomain-mediated -catenin-dependent switching Kageyama R, Rosenfeld MG 2006 Sustained Notch signaling in events dictate cell-lineage determination. Cell 125:593–605 progenitors is required for sequential emergence of distinct cell 21. Kendall SK, Gordon DF, Birkmeier TS, Petrey D, Sarapura VD, lineages during organogenesis. Genes Dev 20:2739–2753 O’Shea KS, Wood WM, Lloyd RV, Ridgway EC, Camper SA 1994 36. Kita A, Imayoshi I, Hojo M, Kitagawa M, Kokubu H, Ohsawa R, 12 Carvalho et al. Hesx1 and Tle3 Block Pituitary Differentiation Mol Endocrinol, April 2010, 24(4):0000–0000 Ohtsuka T, Kageyama R, Hashimoto N 2007 Hes1 and Hes5 con- 41. Sloop KW, Dwyer CJ, Rhodes SJ 2001 An isoform-specific inhibi- trol the progenitor pool, intermediate lobe specification, and pos- tory domain regulates the LHX3 LIM homeodomain factor holo- terior lobe formation in the pituitary development. Mol Endocrinol protein and the production of a functional alternate translation 21:1458–1466 form. J Biol Chem 276:36311–36319 37. Monahan P, Rybak S, Raetzman LT 2009 The notch target gene 42. Brickman JM, Clements M, Tyrell R, McNay D, Woods K, Warner HES1 regulates cell cycle inhibitor expression in the developing J, Stewart A, Beddington RS, Dattani M 2001 Molecular effects of pituitary. Endocrinology 150:4386–4394 novel mutations in Hesx1/HESX1 associated with human pituitary 38. Jean D, Bernier G, Gruss P 1999 Six6 (Optx2) is a novel murine Six3- disorders. Development 128:5189–5199 related homeobox gene that demarcates the presumptive pituitary/hypo- 43. Nuthall HN, Joachim K, Stifani S 2004 Phosphorylation of serine thalamic axis and the ventral optic stalk. Mech Dev 84:31–40 239 of Groucho/TLE1 by protein kinase CK2 is important for in- 39. Brinkmeier ML, Potok MA, Davis SW, Camper SA 2007 TCF4 hibition of neuronal differentiation. Mol Cell Biol 24:8395–8407 deficiency expands ventral diencephalon signaling and increases 44. Stifani S, Blaumueller CM, Redhead NJ, Hill RE, Artavanis- induction of pituitary progenitors. Dev Biol 311:396–407 Tsakonas S 1992 Human homologs of a Drosophila enhancer of 40. Oliver G, Mailhos A, Wehr R, Copeland NG, Jenkins NA, Gruss P, split gene product define a novel family of nuclear proteins. Nat Li X, Perissi V, Liu F, Rose DW, Rosenfeld MG 1995 Six3, a murine Genet 2:343 homologue of the sine oculis gene, demarcates the most anterior 45. Windle JJ, Weiner RI, Mellon PL 1990 Cell lines of the pituitary border of the developing neural plate and is expressed during eye gonadotrape lineage derived by targeted oncogenesis in transgenic development. Development 121:4045–4055 mice. Mol Endocrinol 4:597–603 DISCUSSION : FACTEURS DE TRANSCRIPTION A HOMEODOMAINE DE TYPE PAIRED, COREPRESSEURS ET ACTIVATION DE POU1F1 Nos résultats permettent de détailler et modifier le modèle d’interaction entre 2 des principaux facteurs de transcription à homéodomaine de type paired, Prop1 et Hesx1, au cours du développement hypophysaire murin. Le modèle admis pour le rôle répresseur de Hesx1 au cours du développement hypophysaire repose sur l’interaction nécessaire avec les co-répresseurs de la famille TLE. L’action propre de Hesx1 est considérée insuffisante pour exercer une répression sur les promoteurs de certains facteurs de transcription hypophysaires (42; 140). Nos données provenant de modèles murins transgéniques (avec expression ectopique de Tle3 et/ou Hesx1, sous contrôle du promoteur de la sous-unité alpha, permettant le maintien de leur expression dans les cellules thyréotropes et gonadotropes après leur période physiologique d’extinction) soulignent au contraire le rôle mineur des co- répresseurs de la famille Groucho lorsqu’ils interagissent avec Hesx1. En effet, le développement des lignées thyréotrope et gonadotrope est bloqué chez les souris double-transgéniques (expression ectopique de Hesx1 et Tle3). Les souris transgéniques pour Hesx1 ne présentent pas de différenciation thyréotrope et gonadotrope, mais l’expression de la sous-unité alpha est conservée. A l’inverse, les souris avec expression ectopique de Tle3 présentent un développement hypophysaire normal. Nos résultats suggèrent donc que Hesx1 est suffisant pour exercer une forte action répressive, et que le rôle des co-répresseurs de la famille Groucho/TLE renforce cette répression, bien que leur action propre (sans Hesx1) soit négligeable. A l’inverse, nos études en culture cellulaire suggèrent que les co-répresseurs de la famille Groucho/Tle ont une action répressive propre importante sur un autre facteur de transcription à homéodomaine de type paired, Prop1. Nous avons observé in vitro que Tle1 et Tle3 avaient une action répressive propre sur l’action stimulatrice de Prop1 sur le promoteur de pou1f1 : la dose courbe réponse a ainsi retrouvé un niveau de répression similaire à celui observé pour Hesx1. Les gels retard (effectués avec la séquence consensus de liaison de Prop1) n’ont pas mis en évidence de liaison directe de Tle1 et Tle3 sur cette séquence consensus (excluant une liaison directe sur l’ADN). Par contre, une interaction protéine-protéine entre Tle1/Tle3 et Prop1 a été mise en évidence par co-immunoprécipitation. L’analyse de la séquence protéique de Prop1 retrouve des séquences potentielles d’interaction avec les protéines de la famille Tle, mais celles-ci ne respectent pas intégralement les séquences consensus reconnues comme responsables d’interaction avec Tle. Il existe vraisemblablement d’autres sites d’interaction de Prop1 avec les co-répresseurs de la famille Groucho (à explorer par mutagenèse dirigée par exemple). Chacune des méthodes proposées expose à des biais : les souris transgéniques présentent une expression ectopique de Hesx1 et Tle3, ce qui, par définition, n’est pas une expression physiologique. Les interprétations proposées sont à nuancer puisque l’expression de Hesx1 n’est plus retrouvée à partir de e14-e15,5 (et que nos effets détectés le sont entre e16 et e18,5). De plus, l’absence d’expression de Tle3 lors de cette expression ectopique de Hesx1 pourrait entraîner un recrutement plus important d’autres co-répresseurs (NCoR, HDAC…) via l’homéodomaine de Hesx1. Le modèle permet seulement d’envisager que les interactions Hesx1/Tle ne semblent pas aussi cruciales que ce qui était envisagé jusqu’à présent. ! #)! Les études en culture cellulaires sont également un modèle artificiel pour lequel les interactions potentielles entre différentes protéines peuvent être biaisées, et les doses utilisées en transfection ne représentent pas nécessairement les niveaux d’expression retrouvés au cours du développement. Cependant, ces méthodes donnent des éclaircissements sur le modèle proposé permettant l’activation de Pou1f1 (qui aboutira à la différenciation des lignées somato-lactotropes et thyréotropes). En particulier, s’il semble qu’ils ne jouent pas un rôle primordial sur Hesx1, les co-répresseurs Tle pourraient au contraire jouer un rôle majeur en inhibant Prop1. Olson et al. ont retrouvé un faible niveau de Tle1 lié à Prop1 sur l’enhancer proximal de Pou1f1 à e12,5 par immunoprécipitation de chromatine (140). Ce résultat est en contradiction avec celui publié par la même équipe en 2001, qui n’était pas parvenu à mettre en évidence d’interaction protéine-protéine entre Prop1 et Tle1 par co-immunoprécipitation (42). Nos études in vitro ont mis en évidence une interaction faible ou plus probablement transitoire entre Prop1 et Tle1 (une quantité importante de plasmide a du être transfectée pour visualiser une interaction protéine-protéine). Ces résultats ne modifient donc pas radicalement le modèle proposé par Olson et al. Ils soulignent que l’absence d’activation de pou1f1 à e12,5 est vraisemblablement lié à l’action répressive directe (sans Hesx1) de Tle1 sur Prop1. Le rôle de la ß-caténine est certainement compétitif par rapport aux co-répresseurs Tle. A e13,5, la ß-caténine remplace ainsi Tle1 en se liant à Prop1, et permet l’activation de pou1f1 (Figures 1 et 2). Carvalho et al. ont rapporté la première mutation de HESX1 responsable d’hypopituitarisme congénital via l’absence d’interaction entre HESX1 et TLE1 (mutation I26T codant pour un acide aminé du domaine eh1 de HESX1) (24). Même si l’action des co-répresseurs de la famille Groucho/TLE est ubiquitaire, le rôle majeur de TLE1 dans l’inhibition de PROP1 peut suggérer une mutation de ce co-répresseur dans certains cas d’hypopituitarisme congénital. Le phénotype qui pourrait en découler est difficile à déterminer : une expression précoce de PROP1 (e9.0) entraine une absence d’hypophyse. Mais l’action de TLE1 est effective à partir de e11,5. L’absence d’inhibition de PROP1 pourrait entrainer une anomalie de la lignée gonadotrope (telle que présentée par les souris surexprimant Prop1 sous contrôle du promoteur de la sous- unité alpha) et peut être une diminution d’expression plus précoce de Hesx1 avec hyperplasie hypophysaire (Le complexe Prop1-ßcaténine inhibe l’expression de Hesx1 à partir de e12,5, mais il n’est pas évident que TLE1, qui est lié à PROP1 sur les séquences régulatrices de Hesx1 à cette période exerce un rôle répresseur aussi important que sur l’enhancer proximal de Pou1f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“A NOVEL DYSFUNCTIONAL LHX4 MUTATION WITH HIGH PHENOTYPIC VARIABILITY IN PATIENTS WITH HYPOPITUITARISM” F. Castinetti *, A. Saveanu*, R. Reynaud, MH. Quentien, A. Buffin, R. Brauner, N. Kaffel, F. Albarel, AM. Guedj, M. El Kholy, M. Amin, A. Enjalbert, A. Barlier, T. Brue Journal of Clinical Endocrinology and Metabolism 2008 Jul;93(7):2790-9 *co-premiers auteurs ! $+! INTRODUCTION Lhx4 est un facteur de transcription à homéodomaine de type LIM, impliqué dans le développement hypophysaire. Le phénotype des souris présentant une invalidation homozygote de lhx4 (lhx4 -/-) confirme ce rôle: l’hypophyse est hypoplasique, la différenciation en 5 lignées cellulaires est correcte, mais le nombre de cellules différenciées est diminué de façon drastique par rapport à l’hypophyse des souris de génotype « sauvage ». Les souris Lhx4 -/- meurent peu après la naissance. A l’inverse, les souris hétérozygotes ( Lhx4 +/-) ont un phénotype hypophysaire normal (180; 158). Chez l’homme, seulement 4 mutations hétérozygotes de LHX4 ont été rapportées comme étant responsables d’hypopituitarisme congénital. La première mutation a été publiée en 2001 (mutation intronique c.607-1G>C), chez 2 enfants porteurs d’hypopituitarisme et d’une malformation de Chiari (117). Les 3 mutations hétérozygotes rapportées récemment (p.R84C, p.A210P, p.L190R ) ont souligné que le phénotype hypophysaire pouvait être extrêmement variable, avec en particulier la présence inconstante d’anomalies extra-hypophysaires (149). Ces observations rendent difficile la définition de critères permettant de déterminer quels patients doivent bénéficier d’un séquençage du gène LHX4 . Nous rapportons ici les résultats du séquençage du gène LHX4 dans une cohorte de 136 patients porteurs d’hypopituitarisme congénital, parfois associés à des malformations extra-hypophysaires, recrutés dans le cadre du réseau GENHYPOPIT. Notre étude a permis de révéler trois nouveaux variants alléliques de LHX4 . Les études fonctionnelles ont souligné le rôle d’au moins 1 de ces variants dans la symptomatologie d’hypopituitarisme congénital. ! $"! ORIGINAL ARTICLE Endocrine Research A Novel Dysfunctional LHX4 Mutation with High Phenotypical Variability in Patients with Hypopituitarism F. Castinetti,* A. Saveanu,* R. Reynaud, M. H. Quentien, A. Buffin, R. Brauner, N. Kaffel, F. Albarel, A. M. Guedj, M. El Kholy, M. Amin, A. Enjalbert, A. Barlier, and T. Brue Centre de Recherche en neurobiologie et neurophysiologie de Marseille (CRN2M) (F.C., A.S., R.R., M.H.Q., F.A., A.E., A.Ba., T.B.), Unite´ Mixte de Recherche 6231, Faculte´de Me´decine Nord, Centre National de la Recherche Scientifique, Universite´de la Me´diterrane´e and Centre de Re´fe´rence des de´ficits hypophysaires, Hoˆpital de la Timone, Assistance Publique Hoˆpitaux de Marseille, 13385 Marseille, France; Laboratoire de Biochimie-Biologie Mole´culaire (A.S., A.E., A.Ba.), Hoˆpital Conception, 13005 Marseille, France; Service de Pe´diatrie (A.Bu.), Centre Hospitalier Chambe´ry, 73000 Chambe´ry France; Universite´Paris-Descartes et Assistance Publique–Hoˆpitaux de Paris (R.B.), Hoˆpital Biceˆtre, 94270 Le Kremlin Biceˆtre, France; Service d’endocrinologie (N.K.), Centre Hospitalier Universitaire Hedi- Chaker, route El-Ain, 3029 Sfax, Tunisie; Service des Maladies Me´taboliques et Endocriniennes (A.M.G.), Centre Hospitalier Universitaire de Nîmes, 30000 NıˆmesFrance; Department of Paediatrics (M.E.K.), Ain Shams University, 11566 Cairo, Egypt; Department of Paediatrics (M.A.), Cairo University, 11566 Cairo, Egypt Context: LHX4 is a LIM homeodomain transcription factor involved in pituitary ontogenesis. Only a few heterozygous LHX4 mutations have been reported to be responsible for congenital pituitary hormone deficiency. Subjects and Methods: A total of 136 patients with congenital hypopituitarism associated with malformations of brain structures, pituitary stalk, or posterior pituitary gland was screened for LHX4 mutations. Results: Three novel allelic variants that cause predicted changes in the protein sequence of LHX4 (2.3%) were found ( p.Thr99fs , p.Thr90Met , and p.Gly370Ser ). On the basis of functional studies, p.Thr99fs mutation was responsible for the patients’ phenotype, whereas p.Thr90Met and p.Gly370Ser were likely polymorphisms. Patients bearing the heterozygous p.Thr99fs mutation had variable phe- notypes: two brothers presented somato-lactotroph and thyrotroph deficiencies, with pituitary hyp- oplasia and poorly developed sella turcica; the youngest brother (propositus) also had corpus callosum hypoplasia and ectopic neurohypophysis; their father only had somatotroph deficiency and delayed puberty with pituitary hyperplasia. Functional studies showed that the mutation induced a complete loss of transcriptional activity on POU1F1 promoter and a lack of DNA binding. Cotransfection of p.Thr99fs mutant and wild-type LHX4 failed to evidence any dominant negative effect, suggesting a mechanism of haploinsufficiency. We also identified prolactin and GH promoters as potential target genes of LHX4 and found that the p.Thr99fs mutant was also unable to transactivate these promoters. Conclusions: The present report describes three new exonic LHX4 allelic variants with at least one being responsible for congenital hypopituitarism. It also extends the phenotypical heterogeneity associated with LHX4 mutations, which includes variable anterior pituitary hormone deficits, as well as pituitary and extrapituitary abnormalities. (J Clin Endocrinol Metab 93: 2790–2799, 2008) ombined pituitary hormone deficiency (CPHD) is defined as mutations of transcription factors ( POU1F1 , PROP1 , HESX1 , C the presence of hormone deficits affecting at least two an- LHX3 , LHX4 , etc .) (1–4). terior pituitary hormone lineages. Congenital forms are due to LHX4 is a LIM homeodomain transcription factor crucial for 0021-972X/08/$15.00/0 Abbreviations: CPHD, Combined pituitary hormone deficiency; MRI, magnetic resonance Printed in U.S.A. imaging; PRL, prolactin. Copyright © 2008 by The Endocrine Society doi: 10.1210/jc.2007-2389 Received October 26, 2007. Accepted April 21, 2008. First Published Online April 29, 2008 * F.C. and A.S. contributed equally to this work. 2790 jcem.endojournals.org J Clin Endocrinol Metab. July 2008, 93(7):2790–2799 Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 jcem.endojournals.org 2791 the genesis and development of Rathke’s pouch (5–7). In mice, Subjects and Methods homozygous lhx4 invalidation by homologous recombination induces an abnormal pituitary phenotype and early, death but Subjects heterozygous animals display no apparent phenotypical modi- The GENHYPOPIT network was launched as a multicentric study in both national (France) and international pediatric and adult endocri- fication (8, 9). In humans, only four functionally defective mu- nology centers (Argentina, Belgium, Egypt, Lebanon, Switzerland, Tu- tations of LHX4 have previously been described. The first one nisia, and Turkey). Screening for LHX4 mutations was performed in 136 consists of a germline intronic splice-site mutation responsible patients (from 133 pedigrees). A total of 39 of the patients of this cohort for a short stature syndrome, due to GH deficiency. Pituitary had been included in a previous report, and the LHX4 allelic variant that was included in the latter study was found in the family reported by phenotype also included TSH and ACTH deficiencies, whereas Machinis et al . (10), referred to as “pedigree D” in the present paper. In gonadotroph axis evaluation was not available due to the young this study, patients bearing congenital hypopituitarism associated with age of the patients. Pituitary magnetic resonance imaging (MRI) malformations of the brain, pituitary stalk, or pituitary posterior gland revealed hindbrain defects in combination with abnormalities of were selected for LHX4 mutation screening on the basis of the first report by Machinis et al . (10), and murine models (14). Seven of the unrelated the sella turcica and of the central skull base (10). Functional patients had other members of the family presenting with CPHD phe- studies showed an abolished cooperative interaction between notype (familial cases). All patients or parents of minors gave their writ- LHX4 and another transcription factor, POU1F1, whose prox- ten informed consent to participate in this study, which was approved by imal promoter has a recognition site for LIM domain transcrip- our institutional ethics committee. Hormonal studies and intracranial imaging were performed in all tion factors (11). Recently, another group reported three new patients in each referring medical center, as previously described (15). On LHX4 mutations with variable pituitary phenotypes (including MRI, malformations were systematically sought and recorded in all pa- nonconstant GH, TSH, ACTH and LH and FSH deficiencies) tients. Patients with a known postnatal cause of acquired hypopituitar- and hypoplasic pituitary on MRI with nonconstant ectopic pos- ism were excluded. We also report the phenotypical evolution of the two patients pre- terior lobe and the lack of other brain malformations. Two of the senting with the first published mutation of LHX4 (intron 4 c.607– three mutants had impaired DNA binding, and the third had 1G ϾC). Due to their young age at diagnosis, gonadotroph axis evalu- reduced activity on ␣GSU, POU1F1, and TSH  promoters (12). ation was not available at the initial report. After their family moved, A striking feature was that all of these four mutations were they were referred to our center for further follow-up, with pituitary hormone evaluation, including gonadotroph axis, and cerebral MRI. heterozygous and likely to act through a mechanism of haplo- These patients are referred to in the text as “Pedigree D.” insufficiency (11). Another group reported a heterozygous exonic missense Screening for LHX4 mutations LHX4 allelic variant in a young baby presenting with panhypo- LHX4 sequence was amplified from peripheral blood DNA. pituitarism, Chiari syndrome, and ectopic neurohypophysis Genomic analysis of LHX4 was performed by direct sequencing. The six (13). However, no functional study was performed, and, thus, it coding exons of LHX4 were amplified from genomic DNA using exon- is not sure that this allelic variant was responsible for the flanking primers (published as supplemental data on The Endocrine So- ciety’s Journals Online web site at http://jcem.endojournals.org). The phenotype. same primers were used for sequencing, using CEQ 8000 sequencer Within a multicenter study of genetic determinants of pitu- (Beckman Coulter, Fullerton, CA). According to the previously de- itary deficiencies (GENHYPOPIT Network), 136 patients (from scribed genotyping algorithm, other candidate genes [ PROP1 , POU1F1 , 133 pedigrees) were screened for LHX4 mutations because of HESX1 , and LHX3 (15)] had previously been sequenced, and no alter- ation had been found in coding regions. In parallel, 37 normal subjects congenital hypopituitarism associated with malformations of were screened as controls for LHX4 mutations and polymorphisms. the brain, pituitary stalk, or posterior pituitary gland. Our pri- mary objective was to search for novel LHX4 mutations in this Plasmid constructs and genomic analyses extended cohort, and our secondary objective was to identify LHX4 in plasmid pCMV-XL5 was purchased from Origene Tech- new potential target genes. In this paper we present three new nologies, Inc. (Rockville, MD). In vitro site-directed mutagenesis was allelic variants of LHX4 that cause predicted changes in protein achieved using the Quick-Change kit (Stratagene cloning system; Strat- sequence: Thr99fs , Thr90Met , and Gly370Ser . Functional stud- agene, La Jolla, CA) according to the manufacturer’s instructions. In brief, Pfu DNA polymerase was used to react 50-ng template DNA with ies showed that only the frameshift mutation was defective in its the mutant sense primer and mutant antisense primer. This reaction transcriptional activity on the POU1F1 promoter, probably due involved 30-sec denaturation at 94 C and 12 cycles consisting of 30-sec to the lack of DNA binding. We found a striking variability in denaturation at 94 C, 1-min annealing at 55 C, and 2-min extension at phenotypical presentation in terms of associated pituitary hor- 72 C. After DNA purification and amplification (QIAGEN maxi kit; QIAGEN, Chatsworth, CA), the correct sequence was confirmed by mone deficits and morphological abnormalities within the same DNA sequencing. Mutagenesis primers used for p.Thr99fs , p.Thr90Met , pedigree. We also identified GH and prolactin (PRL) promoters and p.Gly370Ser allelic variants are given in the supplemental data. as potential in vitro target genes of LHX4 and found that the p.Thr99fs mutation was also defective in its ability to transac- EMSA tivate both of these promoters. Thus, the present study allowed Gel mobility shift assays were performed to assess the DNA binding us to extend the pattern of phenotypes associated with LHX4 properties of the LHX4 mutants. Wild-type and mutant LHX4 were transcribed and translated using the trinitrotoluene-coupled reticulocyte mutations: pituitary hyperplasia and corpus callosum hypopla- lysate system with T7 polymerase (Promega Corp., Madison, WI). An- sia may be associated with variable patterns of anterior pituitary nealed synthetic oligonucleotides with ␣GSU promoter LIM consensus hormone deficits. sequence, as previously described (12, 16), were labeled with [ 32 P]deoxy- Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 2792 Castinetti et al . LHX4 Mutations J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 TABLE 1. Allelic variant frequency in LHX4 gene sequences reported in our cohort of 133 unrelated patients Control Location population (NM_033343)_ Allelic allelic (NP_203129) Sequence variation frequencies frequencies AA change Exon 3 a p.Thr90Met c.269C ϾT 0.4% (one patient) 0% Exon 3 a p.Thr99AsnfsX53 c.293_294insC 0.4% (one patient) 0% Exon 6 a p.Gly370Ser c.1108A ϾG 0.4% (one patient) 0% Exon 6 p.Asn328Ser (rs7536561) c.983A ϾG 49.7% 47.3% No AA change Exon 3 p.Asn128Asn (29) c.384C ϾT 1.9% 2.7% Exon 3 p.Asn150Asn (rs16855642) c.450C ϾT 2.3% 0% Exon 6 p.Gly283Gly (29) c.849A ϾC 0.4% (one patient) 0% Intronic Intron 2 c.248 ϩ 65 A ϾG (rs35619850) 1.5% 2% Intron 3 c.452–5 T ϾC (rs2764449 ) 4.5% 4% Intron 5 c.778 ϩ 14 G ϾT (rs3806302) 3.7% 3.5% The p.Thr99AsnfsX53 (c.293InsC ) was considered a deleterious mutation on the basis of cosegregation of the CPHD phenotype with this heterozygous trait, and of functional studies (see Fig. 3). The p.Thr90Met and p.Gly370Ser newly described allelic variants were considered polymorphisms on the basis of functional studies. The other allelic variants had previously been described as frequent ( pAsn328Ser ) or rare polymorphisms ͓Ensembl single nucleotide polymorphism database and Melo et al . (29) ͔. a Other new rare allelic variants (exonic or intronic). CTP. Immunodepletion (supershift assay) was performed with 0.5 g/ Human heterologous HeLa cells were cultured and transfected as well rabbit anti-LHX4 affinity purified polyclonal antibody (CHEMI- follows: 0.3–0.6 ng reporter constructs (PRL-250, GH, or POU1F1 pro- CON International, Inc. Temecula, CA). The protein-DNA complexes moter) and 0.6 ng effector construct (pCMV-XL5 empty vector or wild- were analyzed by electrophoresis through a 8% polyacrylamide gel con- type or mutant LHX4) per well were cotransfected using the liposome taining 2.5% glycerol in a 0.5 ϫ Tris borate buffer at 4 C. technique (Polyfect transfection reagent; QIAGEN, Hilden, Germany). Total DNA was kept constant with pCMV-XL5 empty vector, which Protein translation study also acted as a control. Transfection efficiency was determined using Wild-type and mutant LHX4 were transcribed and translated using 0.01 ng pCMV-Renilla, and luciferase firefly values were normalized to the same reticulocyte lysate system but with nonradioactive amino acid it. Firefly and Renilla luciferase activity was measured 48 h after trans- fection. All assay points were performed in triplicate. Results are ex- mixture minus methionine and [ 35 S]methionine. The protein expression pressed in relative units (fold increase vs . empty vector). was analyzed by electrophoresis on a 12% sodium dodecyl sulfate poly- acrylamide gel and then detected by autoradiography. Cell culture and cotransfection Results Reporter constructs containing different gene-regulatory regions with putative LIM factor binding sites were fused to a firefly luciferase Screening for LHX4 mutations gene. These constructs included the proximal promoter regions of the human PRL gene (PRL-250) (a gift of J. A. Martial, Liege, Belgium) (17), Endocrine and neuroradiological phenotypes GH or the proximal promoter of the human gene (Pa3 Ghp-Luc) (a gift LHX4 screening was performed in 136 patients, from 133 of N. L. Eberhardt, Rochester, MN) (18), or the positive autoregulatory site of the human POU1F1 promoter gene (a gift of M. Delhase, San pedigrees. Seven of the 133 unrelated patients screened had a Diego, CA) (19). All nucleotide numbering was relative to the transcrip- history of familial CPHD (5.3%). Briefly, all but one patient tion start site. presented with GH deficiency, 65% had TSH deficiency, and TABLE 2. Hormone profiles of patients with p.Thr99fs LHX4 mutation GH1 GH2 FSH Cortisol PRL Patient IGF-I base/peak base/peak LH base/peak Testosterone base/peak TSH T4 base/peak no. (g/liter) (mU/liter) (mU/liter) base/peak (UI/liter) (nmol/liter) (nmol/liter) (mU/liter) (pmol/liter) (ng/ml) III1 Ͻ33 1.1/5 1/1.9 NE NE NE 876 1.5 8 2/2.5 III2 104 1.5/6.2 6.2/19 NE NE NE 362/748 2 9 0.8/1.5 II2 148 Ͻ0.5/0.8 Ͻ0.5/0.5 4.6/5 7.8/8.2 17.5 299/600 2.4 14 16/60 Complete GH deficiency was defined as GH response after stimulation below 20 mU/liter. To convert human GH: 1 ng/liter ϫ 3.0 ϭ 1 mUI/liter. Corticotroph deficiency was defined as plasma cortisol value below 500 nmol/liter after insulin test stimulation (test was not performed in patient III1 who had basal cortisol value above 500 nmol/liter). To convert cortisol: 1 nmol ϫ 0.362 ϭ 1 g. Gonadotroph axis was investigated only in patients of postpubertal age, i.e . older than 15 yr for female and 17 yr in male subjects. FSH-LH deficiency was diagnosed on the basis of delayed or absent pubertal development with low serum testosterone or estradiol levels and blunted LH/FSH response to a GnRH stimulation test. Testosterone normal values were 14–24 nmol/liter. LH and FSH normal values were 3–30 UI/liter. Patient II2 had delayed puberty and nonstimulated LH and FSH levels after the GnRH stimulation test, with normal basal testosterone levels. LH Ϫ FSH values are indicated with baseline and GnRH peak values. To convert testosterone: 1 nmol ϫ 0.288 ϭ 1 g. TSH normal values were 2–5 mUI/liter. T 4 normal values were 13–20 pmol/liter. Thyrotroph deficiency was defined as low or normal basal TSH levels associated with low T 4 levels. To convert T 4: 1 pmol ϫ 0.788 ϭ 1 ng. Cortisol, Plasma cortisol value before and after insulin test stimulation (insulin test not performed in patient III1); GH1, GH value before (base) and after GHRH stimulation (peak); GH2, GH value before (base) and after glucagon-propanolol stimulation (peak); NE, gonadotroph axis was not evaluated because of the young age of the patients. Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 jcem.endojournals.org 2793 48% had ACTH deficiency. Of the 74 patients that could be Their father (patient II2) had presented with delayed puberty evaluated in terms of pubertal age, 60 presented with FSH/LH with normal final stature (final height: 168 cm). He had a history deficiencies (81%). Of the patients, 76% had at least three pi- of severe obesity from the age of 6 yr to adulthood. Hormonal tuitary deficiencies. The most frequent encephalic MRI abnor- data showed somatotroph deficiency and blunted response to malities were pituitary hypoplasia in 70% of cases, abnormal GnRH stimulation with normal basal testosterone level, with pituitary stalk in 63% of cases (nine patients with a very thin irregular moderately hyperplastic pituitary on MRI. There were pituitary stalk, and 65 with a pituitary stalk interruption), and no pituitary stalk or neurohypophysis abnormalities (Fig 1B, ectopic neurohypophysis in 31% of cases. Associated cerebral images C1 and C2). abnormalities were less frequent, with 8% of patients presenting The mother of the propositus had no hormonal abnormalities. with Chiari malformation, and 5% with corpus callosum hyp- No other members of the family were available for assessment. oplasia or median line abnormalities. Only 3% of screened pa- tients had septooptic dysplasia. One patient presented cerebellar Genotype dysplasia; another had craniostenosis. Genetic analyses showed that the propositus, his brother, and his father carried the same heterozygous p.Thr99fs LHX4 allelic Genomic analyses of the 136 patients variant, with cytosine insertion in the third exon of LHX4 Three of our unrelated patients had unpublished LHX4 allelic (c.293_294 InsC ). At protein level, this predicts a frameshift variants (2.3%): p.Thr99fs , (pedigree A), p.Thr90Met (pedigree (p.Thr99 frameshift ) in the second LIM region with a stop codon B), and p.Gly370Ser (pedigree C). LHX4 polymorphisms were occurring 53 amino acids after codon 99 ( p.Thr99AsnfsX53 ). found in 54% of our patients; 49.5% of them were carrying the The mother did not carry the mutation. No other family member previously described p.Asn328Ser polymorphism (Table 1). was available for genetic studies (Fig. 1). None of the subjects from the control population bore any of the three unpublished LHX4 allelic variants. The frequency of each Pedigree B: p.Thr90Met LHX4 variant of the other polymorphisms was similar in our control popula- Phenotype tion and in previously published reports. The propositus was referred to an endocrinologist at the age of 50 yr. This man had been diagnosed at the age of 10 yr with Individual data of patients bearing LHX4 allelic variants somatotroph and corticotroph deficiencies. Delayed puberty Pedigree A: p.Thr99fs LHX4 variant (Fig. 1, Table 2) also confirmed at the age of 16 yr the diagnosis of gonadotroph Phenotype deficiency. He received hydrocortisone, testosterone, and GH The propositus (patient III1) was referred to a pediatrician for replacement therapy. At the age of 50 yr, pituitary MRI disclosed short stature and micropenis at the age of 9 months. He was then ectopic neurohypophysis (at the higher part of the stalk), a thin Ϫ4 SD for height and weight with a normal head circumference. pituitary stalk, and an empty sella syndrome. Hormonal data showed somatotroph and thyrotroph deficien- His father did not have any baseline hormonal abnormalities cies. MRI showed poorly developed sella turcica, marked pitu- (data not shown). No other members of the family were available itary hypoplasia, thin pituitary stalk, lack of visible neurohy- for assessment. pophysis, and corpus callosum hypoplasia (Fig. 1B, images A1 and A2). His brother (patient III2) was 4 yr old, with a short Genotype stature ( Ϫ2 SD ). Hormonal data showed somatotroph and thy- Genetic analyses revealed that the propositus was bearing the rotroph deficiencies. MRI showed poorly developed sella tur- allelic variant p.Thr90Met in a heterozygous state. Only his fa- cica, pituitary hypoplasia, and normal pituitary stalk and neu- ther could be tested by sequence analyses; he was not bearing this rohypophysis (Fig. 1B, images B1 and B2). As depicted in Fig. 2, allelic variant. The propositus and his clinically unaffected father A and B, both affected brothers had a poorly developed shallow, both bore the c.778 ϩ 14 G ϾT polymorphism in a heterozygous sella turcica without normal concavity of the sellar floor. state. TABLE 3. Clinical, endocrine, and MRI phenotype in patients bearing the five currently published LHX4 mutations Mutation c.607–1G >C PR84C pL190R pA210P pThr99fs Pituitary phenotype GH D D D D/N D TSH D N D D/N D/N ACTH D N D D/N N LH FSH D/N D N D/N D Familial/sporadic Familial Sporadic Sporadic Familial Familial Pituitary and cerebral MRI Pituitary size Hypoplasia Hypoplasia Hypoplasia Hypoplasia Hypoplasia or hyperplasia Posterior lobe Ectopic Ectopic Ectopic Normal Normal Intracranial imaging Chiari syndrome Normal Normal Normal Corpus callosum hypoplasia Sellar development Poor Normal Normal Normal Poor D, Deficiency in the evaluated pituitary axis; N, normal evaluation of the pituitary axis. Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 2794 Castinetti et al . LHX4 Mutations J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 FIG. 2. A, Schematic representation of LHX4 (true length of each domain was not considered); the three allelic variants described in the present report are indicated. B, S35 in vitro translation of wild-type and the three LHX4 mutants. C, EMSA was performed with P 32 -␣ GSU promoter response element, wild-type LHX4 (WT), and the three mutants. Immunodepletion was performed with polyclonal anti-LHX4 antibody. Lanes marked with an asterisk (*) indicate addition of polyclonal LHX4 antibody. Arrow indicates position of the expected LHX4 DNA binding. 1, pGly370Ser mutant; 2, pThr90Met mutant; 3, pThr99fs mutant; EV, empty vector; Fp, free probe. mother did not have any hormonal abnormalities. No other members of the family were evaluated. Genotype FIG. 1. A, Pedigree of family A, including the propositus (patient III1) with CPHD diagnosed at age 9 months ( arrow ), his clinically affected Genetic analyses showed that the propositus was bearing the brother III2 (age 4 yr), and their affected father (patient II2). Black dots allelic variant p.Gly370Ser in a heterozygous state. Only her indicate patients heterozygous for the p.Thr99fs mutation. Circles mother was available for sequencing analyses; she was not bear- represent females and squares , males. Note that subject II1 (mother of the index case) was found to have a wild-type LHX4 genotype, whereas ing this allelic variant. The propositus and her clinically unaf- members of the prior generation were not available for genotyping. B, fected mother were also bearing the previously described poly- Family A. The pituitary MRI of the propositus (patient III1, images A1 and morphism p.Asn328Ser in a heterozygous state. A2), his brother (patient III2, images B1 and B2), and their father (patient II2, images C1 and C2). T1-weighted section MRI scans after gadolinium of the brain of the indicated patients. A1, B1, and C1 are coronal; A2, B2, Phenotypical update of a previously described family and C2 are sagittal. 1, Corpus callosum hypoplasia. 2, Pituitary stalk. 3, (Pedigree D) carrying an intronic LHX4 mutation (intron Sella turcica. 4 c.607–1 G >C) Machinis et al . (10) previously described the same siblings (a Pedigree C: p.Gly370Ser LHX4 variant brother and sister) carrying an intronic splice site allelic variant Phenotype (G to C substitution in the intron preceding exon 5), responsible The propositus was referred to a pediatrician at the age of 18 for somatotroph, lactotroph, and corticotroph deficiencies. At yr. She had somatotroph, corticotroph, and thyrotroph deficien- the first publication, both children were too young for their go- cies. Pituitary MRI displayed pituitary hypoplasia with pituitary nadal axis to be evaluated, and an update was necessary regard- stalk interruption, but no neurohypophysis abnormalities. Her ing the outcome of the endocrine phenotype. The brother was Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 jcem.endojournals.org 2795 now aged 18 yr, and he presented with complete gonadotroph deficiency. Under appropriate GH and testosterone replacement therapy, he had reached normal height. Cerebral MRI performed at the age of 17 yr confirmed previous findings with a small sella C turcica and a hypoplastic anterior hypophysis associated with > ectopic posterior hypophysis. The sister was now aged 15 yr. She mutation responsible of congenital hypopituitarism began spontaneous puberty at the age of 14 yr and still receives pituitary stalk Previously published as a somatotroph, corticotroph, and thyrotroph substitutive thera- Ectopic, middle of pies. She was evaluated at Tanner stage IV and had not yet had menarche. Pituitary MRI was unchanged (Table 4). Previously published In vitro translation (Fig. 2B) Pedigree D c.607–1G As expected, a truncated protein was observed with the Brother Sister pThr99fs LHX4 mutant. In contrast, expression of pThr90Met mutation responsible of congenital hypopituitarism chiasm and pGly370Ser was detected with the same molecular weight as Chiari syndrome Chiari syndrome wild-type LHX4. Gel shift mobility assay (Fig. 2C) DNA binding of the LHX4 wild-type and the three LHX4 mutants was tested on the ␣GSU promoter LIM factor response Pedigree C element. Wild-type LHX4 had a strong detectable DNA binding p.Gly370Ser that was displaced by anti-LHX4 antibody (supershift). No de- Normal Ectopic, near the optic tectable DNA binding was observed with the pThr99fs mutant under the chosen experimental conditions, whereas pGly370Ser and had an equivalent DNA binding as wild-type LHX4. Inter- estingly, pThr90Met seemed to have a stronger DNA binding allelic variants Pedigree B than wild-type LHX4. p.Thr90Met LHX4 the stalk Functional studies Likely polymorphism Likely polymorphism Previously published as a As expected (11), cotransfection of wild-type LHX4 with POU1F1 promoter resulted in a strong stimulation of the lucif- erase reporter gene relative to the empty vector. No activation was observed with the p.Thr99fs mutant. Cotransfection of mu- tant p.Thr99fs and equivalent amounts of wild-type LHX4 did not modify the effect of the latter on POU1F1 promoter, exclud- congenital hypopituitarism ing dominant negative effect. Mutation responsible of In contrast, cotransfection of p.Thr90Met or p.Gly370Ser Pedigrees from our cohort LHX4 allelic variants with the POU1F1 promoter resulted in equal luciferase activity stimulation relative to wild-type LHX4. Equimolar cotransfection of either p.Thr90Met or p.Gly370Ser p.Thr99fs LHX4 plasmids and wild-type LHX4 plasmids showed similar results to double dose of individual plasmids on POU1F1 pro- congenital hypopituitarism moter, excluding dominant negative effect of these allelic vari- Pedigree A Mutation responsible of ants. These functional results showed that despite amino acid change, p.Thr90Met and p.Gly370Ser were probably nonfunc- tional polymorphisms (Fig. 3A). Identification of potential new target genes Based on the phenotypes of patients presenting mutations of Propositus Brother Father congenital hypopituitarism LIM homeodomain transcription factors (namely in terms of hypoplasia Mutation responsible of ThinPoorly developedCorpus callosum Poorly developed Normal Normal Normal Normal Normal Thin Poorly developed Stalk interruption Poorly developed Thin Thin DNDNE D N D NE D N N D DDD DDD NDD D N D D D D N somato-lactotroph deficiencies) (10, 20, 21), we decided to eval- HypoplasiaNot visualized Hypoplasia Normal Hyperplasia Normal Empty sella Ectopic, higher part of Hypoplasia Hypoplasia Hypoplasia uate the effects of wild-type LHX4 on GH and PRL promoters. Cotransfection of wild-type LHX4 with both promoters resulted Variable clinical, endocrine, and MRI phenotype in patients bearing four distinct in a strong stimulation of the luciferase reporter gene (Fig. 3, B and C). This stimulatory effect was observed in a dose-dependent manner (data not shown). Somatotroph Corticotroph Thyrotroph Gonadotroph Pituitary stalk Sella turcica Other signs Pituitary posterior lobe Pituitary Functional studies Clinical presentation Cerebral MRI TABLE 4. To evaluate the functional significance of our three new D, Deficiency in the evaluated pituitary axis; N, normal evaluation of the pituitary axis. NE, gonadotroph axis was not evaluated due to the young age of the patients. Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 2796 Castinetti et al . LHX4 Mutations J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 equivalent amounts of wild-type LHX4 did not modify the stim- 1n WT A ulatory effect of the latter on PRL or GH promoters. In contrast, 2n WT results obtained after cotransfection with p.Thr90Met or 1n T99fs 1n T90M p.Gly370Ser were identical to wild-type LHX4, confirming that 1n G370S these allelic variants were probably polymorphisms (Fig. 3, B 1n T99fs + WT and C). 1n T90M + WT 1n G370S + WT Control Discussion 0 10 20 Our screening of 136 patients (from 133 pedigrees) confirmed 1n WT B 2n WT the rarity of LHX4 allelic variants associated with amino acid change, with a prevalence of less than 3% in this large cohort of 1n T99fs selected patients. Four LHX4 mutations with impaired func- 1n T90M 1n G370S tional activity were previously reported (10, 12). A case of pan- 1n T99fs + WT hypopituitarism associated with another LHX4 allelic variant 1n T90M + WT was also recently reported (13). However, functional studies were not performed, making it difficult to draw an unambiguous 1n G370S +WT Control link between the phenotype and this allelic variant. In the present study, we report three novel allelic variants, but functional stud- ies confirmed that only one of them, the p.Thr99fs LHX4 mu- 1n WT C tation, was clearly responsible for the CPHD phenotype. The 2n WT functional alteration induced by this new mutation could be due 1n T99fs to nonsense-mediated RNA decay (22); alternatively, the lack of 1n T90M DNA binding could be due to the lack of homeodomain in the 1n G370S predicted in vitro truncated protein, as demonstrated in our 1n T99fs + WT 1n T90M + WT translation experiments, even if in vitro translation of the pro- 1n G370S +WT teins does not always mean that they have a similar expression Control level or stability as wild-type LHX4 proteins in the gene activa- tion assays (Fig. 4). The scarcity of CPHD due to LHX4 muta- tions may be due to several factors. First, because all previously Rela ve luciferase units (fold versus EV) published mutations like the one reported here are heterozygous FIG. 3. A, Expression vectors for wild-type (WT) and the three allelic (10, 12), and because lhx4 knockout mice die at birth (8), it is variants with POU1F1 promoter. Proteins were transiently cotransfected into heterologous human HeLa cells with a luciferase reporter gene under likely that homozygous LHX4 mutations are lethal or that they the control of the POU1F1 promoter. Promoter activity was assayed by increase human perinatal mortality. Second, in this and previous measuring luciferase activity 48 h after transfection. Negative controls studies, none of the identified allelic variants is identical (10, 12), (control) received equivalent amounts of empty expression vector plasmid. Results are expressed in fold-increased levels of luciferase suggesting a lack of a mutational hot spot in the LHX4 gene. stimulation compared with control. For each effector, 1n ϭ 300 ng. B, Third, the low incidence of LHX4 mutations could also be the Expression vectors for wild-type, and the three allelic variants with PRL result of inappropriate patient selection for screening. Based on promoter. Proteins were transiently cotransfected into heterologous human HeLa cells with a luciferase reporter gene under the control of PRL the phenotype of the patients bearing the first LHX4 published (Fig. 5A) or GH promoter (Fig. 5B). Promoter activity was assayed by mutation (10) and of murine models (14), we decided to screen measuring luciferase activity 48 h after transfection. Negative controls patients presenting with congenital hypopituitarism (including (control) received equivalent amounts of empty expression vector plasmid. Results are expressed in fold-increased levels of luciferase GH deficiency in all but one of them) associated with pituitary stimulation compared with control. For each effector, 1n ϭ 300 ng. C, abnormalities, including stalk interruption syndrome and ec- Expression vectors for wild-type, and the three allelic variants with GH topic posterior pituitary lobe, or with intracranial abnormalities, promoter. Proteins were transiently cotransfected into heterologous human HeLa cells with a luciferase reporter gene under the control of PRL including Chiari syndrome, septooptic dysplasia, or other rare (Fig. 5A) or GH promoter (Fig. 5B). Promoter activity was assayed by malformations ( e.g . corpus callosum hypoplasia). However, measuring luciferase activity 48 h after transfection. Negative controls three new LHX4 mutations were recently reported, and none of (control) received equivalent amounts of empty expression vector plasmid. Results are expressed in fold-increased levels of luciferase the patients presented brain abnormalities (12). Only further stimulation compared with control. For each effector, 1n ϭ 300 ng. EV, large-scale studies will help determine which patients should be Empty vector. screened for LHX4 mutations. Clinical and MRI phenotypes of patients bearing mutations LHX4 allelic variants on PRL and GH promoters, we cotrans- of LHX4 can be highly variable both within and between families fected mutant LHX4 allelic variants with each of the promoters. as shown in Tables 3 and 4. For instance, endocrine testing and No activation was found with the p.Thr99fs mutant, confirming cerebral MRI revealed a wide variability in the phenotypes of the the deleterious effect of this frameshift mutation. As shown for patients carrying the novel p.Thr99fs mutation, suggesting a the POU1F1 promoter, cotransfection of mutant p.Thr99fs and variable phenotype of this mutation. Despite identical genotype, Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 jcem.endojournals.org 2797 the father of the propositus presented few phenotypical signs manner. The lack of dominant negative effect, as shown by co- compared with his children; he only presented delayed puberty, transfection with wild-type LHX4, suggests a probable mecha- with a lack of thyrotroph deficiency, normal pituitary stalk, and nism of haploinsufficiency, as previously reported by Machinis sella turcica, and had reached normal final height. This could be and Amselem (11). Our genetic screening also identified two attributed to the delayed appearance of somatotroph deficiency previously unpublished nonsynonymous allelic variants or to obesity. Because sequencing of PROP1 , POU1F1 , HESX1 , (p.Thr90Met and p.Gly370Ser ). We considered them polymor- and LHX3 in the affected members of this family were all neg- phisms because DNA binding of the mutants was not impaired, ative, other genes might interfere with pituitary development in and our functional studies showed the lack of transactivation this family. Variability in the endocrine phenotypes was also alterations on POU1F1, PRL, and GH promoters. However, it present in patients bearing the first previously published intronic cannot be excluded that these variants have functional signifi- mutation of LHX4 (10) because we found only one of the pro- cance because functional studies are based on transfections done bands to have gonadotroph deficiency at pubertal age. To sum- in an in vitro system, in heterologous cells, and on a limited marize, high phenotypical variability was observed between ped- number of target genes, thus making it difficult to state that this igrees A and D in terms of endocrine profiles (inconstant result is entirely applicable to humans in a pathophysiological corticotroph, thyrotroph and gonadotroph deficiencies), pitu- setting. The fact that pThr90Met seemed to have an increased itary morphology (hypoplastic or hyperplastic), location of pi- DNA binding compared with wild-type LHX4 could be a point tuitary posterior lobe (normal or ectopic), pituitary stalk (thin or in favor of a mutation rather than a polymorphism; however, normal), or cerebral malformations (Chiari syndrome in two these data are not sufficient to support this hypothesis because patients, corpus callosum hypoplasia in one patient, lack of mal- EMSAs cannot be strictly considered as a quantitative approach, formation in two patients). Patients bearing the three recently and because functional studies performed on three different pro- published LHX4 mutations also presented extremely variable moters did not find any significant difference between both mu- pituitary intrafamily or interfamily phenotypes (12). The high tants and wild-type LHX4. Moreover, EMSAs were performed variability of endocrine phenotype could thus be accounted for under specific conditions that cannot be entirely transposed to in by different degrees of expressivity, or by the implication of other vivo experiments, making difficult to determine a potential func- epigenetic and/or environmental factors (23). The present study tional consequence of an increased in vitro DNA binding. At last, also allowed us to define better the MRI profile of patients bear- even though the Thr90Met and Gly370Ser show normal activity ing LHX4 mutations as shown in Table 4. First, one important on the tested candidate target genes, is it possible that they are phenotypical feature that was present in all four young patients nevertheless impaired in activation (or repression) of yet-uni- with LHX4 mutations is poorly developed sella turcica. This dentified genes that are critical to anterior pituitary develop- finding suggests the possible interaction of LHX4 with other ment? Finally, our study also confirms the high frequency of the transcription factors involved in the development of sella turcica. polymorphism p.Asn328ser (more than 50% patients had this Moreover, the fact that all four patients bearing LHX4 mutation polymorphism) and the rarity of the other previously unpub- (intron 4 c.607–1 G ϾC or p.Thr99fs ) presented with poorly lished variants that were found in our functional studies to likely developed sella turcica makes it an important point in the patient be polymorphisms. In a recent study based on 62 patients bearing selection method for LHX4 screening. It is noteworthy that this hypopituitarism and ectopic posterior pituitary lobe, a similar is not a constant feature because none of the recently described patients with novel LHX4 mutations had poorly developed sella frequency of the p.Asn328Ser polymorphism was reported (57% turcica (12). Moreover, it is difficult to ascertain whether pitu- of the patients), and no LHX4 mutation had been found (29). itary hypoplasia was a direct consequence of defective LHX4 A secondary objective of this study was the identification of function or may be secondarily associated with a poorly devel- some potential new target genes of LHX4 that might account for oped sella turcica. Second, pituitary hypoplasia was not always the hormonal phenotype of LHX4 mutations. Previous studies present, and a slightly hyperplastic gland was even observed in reported LHX4 activation of pituitary target genes such as ␣-gly- one case; in contrast to previously published data on PROP1, the coprotein, FSH  and TSH  (12, 30–32). We report a moderate older patient had the more hyperplastic pituitary. Pituitary hy- stimulatory effect of wild-type LHX4 on GH promoter and a poplasia was not unexpected because LHX4 is necessary for the strong stimulatory effect on PRL promoter that could be ex- survival of antehypophysis precursor cells (9). In contrast, the plained by the presence of LIM domain consensus sequences on father of the propositus (patient III1) is the first patient described both promoters. However, no direct binding experiment was with LHX4 mutation and hyperplastic pituitary on MRI. Fol- performed on these promoters. The high frequency of soma- low-up of the patients presenting the first-published mutation of totroph deficiency in patients presenting with LHX4 mutations LHX4 did not reveal any pituitary volume increase between the could be explained by the lack of stimulatory effects of mutant age of 5 and 18 yr. Pituitary hyperplasia has been described at an LHX4 ( intron 4 c.607–1 G ϾC or p.Thr99fs ) relative to wild- initial stage in association with PROP1 mutations (24–27) as type LHX4 on GH promoter and indirectly on POU1F1 pro- well as in cases of LHX3 mutations (21, 28). moter. Further studies particularly on the DNA binding prop- An interesting fact concerning the novel p.Thr99fs mutation erties of LHX4 on these LIM homeodomain transcription factors is that patients were heterozygous, as in the previously described consensus sequences, and identification of additional LHX4 tar- mutations of LHX4 (10, 28). This finding confirms that CPHD get genes will allow us to understand better the physiological due to LHX4 mutations is inherited in an autosomal dominant relevance of these findings in humans. The patients’ phenotypes Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 2798 Castinetti et al . LHX4 Mutations J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 could at least partly be accounted for by interactions of LHX4 References with these new target genes. To conclude, pathogenic LHX4 mutations are rare, with a 1. Kelberman D, Dattani MT 2006 The role of transcription factors implicated in anterior pituitary development in the aetiology of congenital hypopituitar- prevalence of less than 1% in a cohort of 136 patients with ism. Ann Med 38:560–577 CPHD and various pituitary or extrapituitary abnormalities. We 2. Reynaud R, Saveanu A, Barlier A, Enjalbert A, Brue T 2004 Pituitary hormone deficiencies due to transcription factor gene alterations. Growth Horm IGF Res identified a novel mutation, p.Thr99fs , responsible for variable 14:442–448 anterior pituitary hormone deficiencies and MRI abnormalities. 3. Li S, Crenshaw 3rd EB, Rawson EJ, Simmons DM, Swanson LW, Rosenfeld The high variability of the phenotypes of the patients within this MG 1990 Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1. Nature 347:528–533 new pedigree extends the pattern of phenotypes associated with 4. Sornson MW, Wu W, Dasen JS, Flynn SE, Norman DJ, O’Connell SM, Guk- LHX4 mutations, which includes not only multiple pituitary ovsky I, Carriere C, Ryan AK, Miller AP, Zuo L, Gleiberman AS, Andersen B, hormone deficiencies associated with pituitary hypoplasia, ec- Beamer WG, Rosenfeld MG 1996 Pituitary lineage determination by the Prophet of Pit-1 homeodomain factor defective in Ames dwarfism. Nature topic neurohypophysis, and Chiari syndrome, but also hyper- 384:327–333 plastic pituitary, poorly developed sella turcica, and inconstant 5. Yamashita T, Moriyama K, Sheng HZ, Westphal H 1997 Lhx4, a LIM ho- meobox gene. Genomics 44:144–146 brain malformations. This study provides additional evidence 6. Liu Y, Fan M, Yu S, Zhou Y, Wang J, Yuan J, Qiang B 2002 cDNA cloning, that the dominance of LHX4 mutations would be due to hap- chromosomal localization and expression pattern analysis of human LIM- loinsufficiency rather than a dominant negative effect of the al- homeobox gene LHX4. Brain Res 928:147–155 7. Mullen RD, Colvin SC, Hunter CS, Savage JJ, Walvoord EC, Bhangoo AP, Ten tered proteins over normal LHX4. S, Weigel J, Pfaffle RW, Rhodes SJ 2007 Roles of the LHX3 and LHX4 LIM- homeodomain factors in pituitary development. Mol Cell Endocrinol 265–266 8. Li H, Witte DP, Branford WW, Aronow BJ, Weinstein M, Kaur S, Wert S, Singh G, Schreiner CM, Whitsett JA, Scott WJ, Pottter SS 1994 Gsh-4 encodes a LIM-type homeodomain, is expressed in the developing central nervous sys- Acknowledgments tem and is required for early postnatal survival. EMBO J 13:2876–2885 9. Raetzman LT, Ward R, Camper SA 2002 Lhx4 and Prop1 are required for cell We thank Jean-Louis Franc and Jean-Paul Herman for fruitful team survival and expansion of the pituitary primordia. Development 129:4229– discussions. We also thank Nicole Peyrol for sequencing experiments. 4239 10. Machinis K, Pantel J, Netchine I, Leger J, Camand OJ, Sobrier ML, Dastot-Le We thank all the following clinicians who sent us samplings of their Moal F, Duquesnoy P, Abitbol M, Czernichow P, Amselem S 2001 Syndromic patients for genetic screening in the GENHYPOPIT network: Dr. P. short stature in patients with a germline mutation in the LIM homeobox Adiceam (Aix-en Provence, France); Professor A. Beckers (Liege, Bel- LHX4. Am J Hum Genet 69:961–968 gium); Dr. C. Bellesme (Paris, France); Dr. P. Berlier (Lyon, France); Dr. 11. Machinis K, Amselem S 2005 Functional relationship between LHX4 and H. Bony-Trifunovic (Amiens, France); Professor Ph. Bouchard (Paris, POU1F1 in light of the LHX4 mutation identified in patients with pituitary France); Professor P. Bougneres (Paris, France); Dr. E. Briand (Clamart, defects. J Clin Endocrinol Metab 90:5456–5462 France); Dr. C. Brue-Fabre (Marseille, France); Professor O. Bruno (Bue- 12. Pfaeffle RW, Hunter CS, Savage JJ, Duran-Prado M, Mullen RD, Neeb ZP, Eiholzer U, Hesse V, Haddad NG, Stobbe HM, Blum WF, Weigel JF, Rhodes nos-Aires, Argentina); Professor J. C. Carel (Paris, France); Professor Ph. SJ 2008 Three novel missense mutations within the LHX4 gene are associated Caron (Toulouse, France); Dr. A. Cartault (Toulouse, France); Professor with variable pituitary hormone deficiencies. J Clin Endocrinol Metab 93: O. Chabre (Grenoble, France); Dr. M. Colle (Bordeaux, France); Dr. Ch. 1062–1071 Cortet-Rudelli (Lille, France); Professor S. Christin-Maitre (Paris, 13. Tajima T, Hattori T, Nakajima T, Okuhara K, Tsubaki J, Fujieda K 2007 A France); Dr. F. Dallavale (Palavas, France); Professor M. David (Lyon, novel missense mutation (P366T) of the LHX4 gene causes severe combined France); Dr. R. Desailloud (Amiens, France); Professor F. Duron (Paris, pituitary hormone deficiency with pituitary hypoplasia, ectopic posterior lobe France); Dr. T. Edouard (Toulouse, France); Dr. O. Evliyaoglu (Ankara, and a poorly developed sella turcica. Endocr J 54:637–641 14. Sheng HZ, Zhadanov AB, Mosinger Jr B, Fujii T, Bertuzzi S, Grinberg A, Lee Turkey); Dr. Ch. Fedou (Montpellier, France); Professor R. Gaillard EJ, Huang SP, Mahon KA, Westphal H 1996 Specification of pituitary cell (Lausanne, Switzerland); Professor Ph. Garnier (Grenoble, France); Pro- lineages by the LIM homeobox gene Lhx3. Science 272:1004–1007 fessor G. Halaby (Beyrouth, Lebanon); Dr. B. Hamon (Chambery, 15. Reynaud R, Gueydan M, Saveanu A, Vallette-Kasic S, Enjalbert A, Brue T, France); Dr. C. Jeandel (Montpellier, France); Professor V. Kerlan (Brest, Barlier A 2006 Genetic screening of combined pituitary hormone deficiency: France); Professor P. Lecomte (Tours, France); Professor J. Leger (Paris, experience in 195 patients. J Clin Endocrinol Metab 91:3329–3336 France); Professor A. Lienhardt (Limoges, France); Dr. G. Loeuille (Be- 16. Sloop KW, Parker GE, Hanna KR, Wright HA, Rhodes SJ 2001 LHX3 tran- scription factor mutations associated with combined pituitary hormone defi- thune, France); Professor J. Mahoudeau (Caen, France); Dr. M. Mana- ciency impair the activation of pituitary target genes. Gene 265:61–69 vela (Buenos-Aires, Argentina); Professor M. Pugeat (Lyon, France); Pro- 17. Caccavelli L, Manfroid I, Martial JA, Muller M 1998 Transcription factor AP1 fessor Y. Reznik (Caen, France); Dr. L. Rocher (Marseille, France); is involved in basal and okadaic acid-stimulated activity of the human PRL Professor V. Rohmer (Angers, France); Dr. B. Ronci-Chaix (Bordeaux, promoter. Mol Endocrinol 12:1215–1227 France); Dr. N. Salah (Cairo, Egypt); Professor J. L. Sadoul (Nice, 18. Shepard AR, Zhang W, Eberhardt NL 1994 Two CGTCA motifs and a GHF1/ France); Dr. G. Simonin (Marseille, France); Dr. Ch. Stuckens (Lille, Pit1 binding site mediate cAMP-dependent protein kinase A regulation of France); Professor A. Tabarin (Bordeaux, France); Professor M. Tauber human growth hormone gene expression in rat anterior pituitary GC cells. J Biol Chem 269:1804–1814 (Toulouse, France); Dr. C. Teinturier (Paris, France); Professor M. P. 19. Delhase M, Castrillo JL, de la Hoya M, Rajas F, Hooghe-Peters EL 1996 AP-1 Teissier (Limoges, France); Dr. C. Thuillier (Limoges, France); Dr. Z. and Oct-1 transcription factors down-regulate the expression of the human Turki (Tunis, Tunisia); Dr. I. Vanpottelbergh (Ghent, Belgium); Dr. PIT1/GHF1 gene. J Biol Chem 271:32349–32358 M. C. Vantyghem (Lille, France); Dr. D. Vezzosi (Villejuif, France); Dr. 20. Pfaeffle RW, Savage JJ, Hunter CS, Palme C, Ahlmann M, Kumar P, Bellone J. Weil (Lille, France); and Dr. F. Wibaux (Bethune, France). J, Schoenau E, Korsch E, Bramswig JH, Stobbe HM, Blum WF, Rhodes SJ 2007 Four novel mutations of the LHX3 gene cause combined pituitary hormone Address all correspondence and requests for reprints to: Professor deficiencies with or without limited neck rotation. J Clin Endocrinol Metab 92:1909–1919 T. Brue, Department of Endocrinology, Hoˆpital de la Timone, 264 21. Netchine I, Sobrier ML, Krude H, Schnabel D, Maghnie M, Marcos E, Duriez rue St Pierre, cedex 5, 13385 Marseille, France. E-mail: thierry. B, Cacheux V, Moers A, Goossens M, Gruters A, Amselem S 2000 Mutations [email protected]. in LHX3 result in a new syndrome revealed by combined pituitary hormone Disclosure Statement: The authors have nothing to disclose. deficiency. Nat Genet 25:182–186 Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 J Clin Endocrinol Metab, July 2008, 93(7):2790–2799 jcem.endojournals.org 2799 22. Frischmeyer PA, Dietz HC 1999 Nonsense-mediated mRNA decay in health 28. Bhangoo AP, Hunter CS, Savage JJ, Anhalt H, Pavlakis S, Walvoord EC, Ten and disease. Hum Mol Genet 8:1893–1900 S, Rhodes SJ 2006 Clinical case seminar: a novel LHX3 mutation presenting 23. Peltonen L, McKusick VA 2001 Genomics and medicine. Dissecting human as combined pituitary hormonal deficiency. J Clin Endocrinol Metab 91:747– disease in the postgenomic era. Science 291:1224–1229 753 24. Reynaud R, Barlier A, Vallette-Kasic S, Saveanu A, Guillet MP, Simonin G, 29. Melo ME, Marui S, Carvalho LR, Arnhold IJ, Leite CC, Mendonca BB, Kno- Enjalbert A, Valensi P, Brue T 2005 An uncommon phenotype with familial epfelmacher M 2007 Hormonal, pituitary magnetic resonance, LHX4 and central hypogonadism caused by a novel PROP1 gene mutant truncated in the HESX1 evaluation in patients with hypopituitarism and ectopic posterior pi- transactivation domain. J Clin Endocrinol Metab 90:4880–4887 tuitary lobe. Clin Endocrinol (Oxf) 66:95–102 25. Riepe FG, Partsch CJ, Blankenstein O, Monig H, Pfaffle RW, Sippell WG 2001 30. Kawamata N, Sakajiri S, Sugimoto KJ, Isobe Y, Kobayashi H, Oshimi K 2002 Longitudinal imaging reveals pituitary enlargement preceding hypoplasia in A novel chromosomal translocation t(1;14)(q25;q32) in pre-B acute lympho- two brothers with combined pituitary hormone deficiency attributable to blastic leukemia involves the LIM homeodomain protein gene, Lhx4. Onco- PROP1 mutation. J Clin Endocrinol Metab 86:4353–4357 gene 21:4983–4991 26. Voutetakis A, Argyropoulou M, Sertedaki A, Livadas S, Xekouki P, Maniati- 31. Sloop KW, Dwyer CJ, Rhodes SJ 2001 An isoform-specific inhibitory domain Christidi M, Bossis I, Thalassinos N, Patronas N, Dacou-Voutetakis C 2004 regulates the LHX3 LIM homeodomain factor holoprotein and the production Pituitary magnetic resonance imaging in 15 patients with Prop1 gene muta- of a functional alternate translation form. J Biol Chem 276:36311–36319 tions: pituitary enlargement may originate from the intermediate lobe. J Clin 32. West BE, Parker GE, Savage JJ, Kiratipranon P, Toomey KS, Beach LR, Colvin Endocrinol Metab 89:2200–2206 SC, Sloop KW, Rhodes SJ 2004 Regulation of the follicle-stimulating hormone 27. Ward RD, Raetzman LT, Suh H, Stone BM, Nasonkin IO, Camper SA 2005  gene by the LHX3 LIM-homeodomain transcription factor. Endocrinology Role of PROP1 in pituitary gland growth. Mol Endocrinol 19:698–710 145:4866–4879 Downloaded from jcem.endojournals.org at Univ Of Mich Library on May 21, 2010 DISCUSSION : MUTATIONS DE LHX4 ET VARIABILITE PHENOTYPIQUE A ce jour, 7 mutations de LHX4 clairement associées à un phénotype d’hypopituitarisme ont été rapportées (deux nouvelle mutations ont été rapportées depuis la publication de l’article) (45; 117; 196; 26; 149; 199) (Table 6). La moitié de ces mutations est sporadique, l’autre moitié survient dans un contexte familial. Le tableau clinique présenté par les patients est très variable en termes de déficits hypophysaires et d’anomalies morphologiques cérébrale et/ou hypophysaire. La majorité des patients présente un déficit complet en hormone de croissance (11/13 soit 84%). Deux patients cependant, porteurs respectivement de la mutation A210P et T99fs , ont une taille finale normale au moment du diagnostic, à l’âge adulte: le premier a une réponse limite de la GH sous stimulation, le 2ème a un déficit en GH complet (26; 149). Ce cas peut indiquer un déficit survenant de façon retardée, ce qui n’a pas été rapporté précédemment dans la littérature. Une patiente porteuse de la mutation A210P , diagnostiquée dans l’enfance, a un déficit partiel en GH (149). Malgré ces quelques variations, l’axe somatotrope est celui qui est le plus fréquemment atteint chez les patients porteurs de mutations de LHX4 . Les atteintes des autres axes sont plus inconstantes, avec respectivement 73%, 46% et 50% de déficits thyréotrope, corticotrope et gonadotrope. Pour ce dernier axe, seulement 10 patients étaient évaluables, et un patient, porteur de la mutation T99fs , présentait un déficit gonadotrope diagnostiqué à l’âge adulte (mais avait eu 2 enfants porteurs de la mutation), évoquant un déficit d’apparition tardive (26) L’un des points intéressants dans l’étude des phénotypes est leur extrême variabilité, entre les différentes mutations, mais également entre les membres d’une même famille porteuse de la même mutation. Ainsi dans la famille rapportée dans notre étude porteuse de la mutation T99fs (26), les 2 enfants présentaient le même phénotype hypophysaire avec un déficit somatotrope et thyréotrope, alors que le père, porteur de la mutation, présentait un déficit somatotrope, mais une taille normale à l’âge adulte, et un déficit gonadotrope, évocateur de déficits de survenue retardée. De même, au sein de la famille porteuse de la mutation A210P , le cas index présente un panhypopituitarisme, son frère un déficit partiel en GH et ACTH (avec une réponse insuffisante sous stimulation), et le père un déficit isolé partiel en ACTH. L’aspect IRM est également variable (149). L’hypophyse est hypoplasique chez tous les patients, sauf un : le père de la famille porteuse de la mutation T99fs présente une hyperplasie hypophysaire (26). La post-hypophyse est ectopique dans un tiers des cas (5 patients sur 15). Quatre patients présentent également une selle turcique peu développée. Les mutations de LHX4 sont également associées à une malformation cérébrale dans 20% des cas, soit un syndrome de Chiari (2 cas), soit une hypoplasie du corps calleux (1 cas). Devant cette grande variabilité phénotypique, il est important de déterminer quels patients doivent bénéficier d’une recherche de mutations de LHX4 . L’étude réalisée dans le laboratoire sur la recherche de mutations de facteurs de transcription hypophysaires chez des patients porteurs de déficits hypophysaires multiples a proposé un algorithme permettant de définir quels gènes de facteurs de transcription rechercher en fonction du phénotype du patient (164). En 2005, date de publication de cette étude, une seule mutation de LHX4 avait été rapportée, chez des patients porteurs d’un panhypopituitarisme associé à un syndrome de Chiari (117). Depuis lors, avec la publication de 5 nouvelles mutations, les données ont pu être précisées. Cependant, au vu de la grande variabilité phénotypique présentée par les patients porteurs de mutations, il semble difficile de proposer un profil type : pour l’heure, afin d’éviter un ! $#! dépistage génétique de tous les patients porteurs de déficit hypophysaire multiple, il nous semble judicieux de rechercher une mutation de LHX4 chez tous les patients porteurs d’au moins 2 déficits hypophysaires (incluant un déficit en GH) associés à une hypoplasie hypophysaire et une post-hypophyse ectopique ou une malformation cérébrale ou un développement anormal de la selle turcique. Il est probable que cette présélection ne permettra pas d’identifier tous les patients porteurs d’une mutation de LHX4 . Cependant, l’étude des mutations actuelles répertoriées dans la littérature confirme que l’ensemble des patients auraient pu être identifiés, à l’exception de la famille porteuse de la mutation A210P , qui ne présentait aucune anomalie morphologique à l’IRM à l’exception d’une hypoplasie hypophysaire (149). Un autre critère dans la sélection des patients à dépister pourrait être l’existence d’une grande variabilité phénotypique en termes de déficits hypophysaires, au sein d’une même famille. Cette question de la sélection des patients se pose également au vu des 2 variants alléliques qui ont été identifiés dans notre population (26). Les études fonctionnelles n’ont pas retrouvé de différence par rapport à la protéine « sauvage ». Il n’existe pas d’effet dominant négatif, pas d’anomalie de liaison à l’ADN en gel retard. Des arguments pourraient néanmoins laisser penser qu’au moins un de ces 2 variants, T90M , pourrait être une véritable mutation pathogène : son absence dans notre population contrôle et le fait que la mutation T90M soit située dans la zone codant pour un domaine LIM. Ce dernier point est particulièrement intéressant car cette mutation pourrait être à l’origine d’une interaction protéique anormale avec un certain nombre de facteurs ou co- facteurs : Ldb1 (un des principaux co-facteurs régulant l’activité des facteurs de transcription à homéodomaine LIM (122)), Sp1 (Sp family factor specific protein 1, qui régule le promoteur de Lhx4 (115)), Isl1 (qui possède un site d’interaction avec les autres facteurs de type LIM) (122), un facteur de transcription à homéodomaine POU (par exemple Pou4f3 interagit avec Lhx3 dans l’oreille interne (80)), CLIM (cofacteur) ou RLIM (régulateur impliqué dans la dégradation protéasomique) (142). Cette mutation pourrait également être à l’origine d’un défaut de phosphorylation : il a été montré pour Lhx3 qu’une mutation d’une thréonine située dans le 1 er domaine LIM réduisait la capacité d’activer les promoteurs de gènes cibles via une phosphorylation anormale (145). Le génotypage des parents n’a pu être effectuée, ne permettant pas de confirmer ou infirmer le caractère de novo de ce variant (qui aurait été un argument supplémentaire en faveur d’une « mutation »). Le phénotype observé est en tout cas très proche de celui observé pour les patients présentant une mutation de LHX4 . Nous avons finalement considéré au moins de manière provisoire que ces 2 variants n’étaient pas responsables du phénotype. De futures études permettront de connaître les partenaires de LHX4 au cours du développement hypophysaire. Il faudra alors déterminer si la mutation T90M ne pourrait pas être responsable du phénotype du fait d’une anomalie d’interaction protéine/protéine. Un caractère pathogène semble moins probable pour la mutation G370S qui n’est théoriquement pas dans un domaine fonctionnel. A noter qu’une équipe japonaise a rapporté un « variant allélique » en position 366 de la séquence de LHX4 chez un patient porteur d’hypopituitarisme congénital avec hypoplasie hypophysaire, post-hypophyse ectopique et selle turcique peu développée (196). Nous n’avons pas considéré ce « variant » comme une mutation responsable du phénotype du fait de l’absence d’étude fonctionnelle, et du fait que cette variation allélique est dans une partie de la séquence ne codant pas pour un domaine fonctionnel (comme pour notre variant G370S). Cependant, les mutations du domaine C-terminal pourraient par exemple affecter la structure 3D de la protéine et être à l’origine d’une anomalie d’activité par ce biais. ! $$! !"#$ !"#$ %&$ !"# $%# $$# &'"#&''# (&'# Mutation p.R84C p.V101A p.L190R Microdeletion 522,009 bp GH D D D D TSH D D D D ACTH N D D N LH FSH D ? ? D Hypophyse Hypoplasie Hypoplasie Hypoplasie Hypoplasie, petite lésion kystique Lobe postérieur Ectopique Ectopique Ectopique Ectopique IRM cérébrale Normale Syndrome de Chiari Normale Normale Selle turcique Normals Normale Normale Peu développée Transmission Dominant Dominant Dominant Dominant Pfaeffle, JCEM, 2008 Tajima, ExpClinEndo, Pfaeffle, JCEM, 2008 Dateki, JCEM, 2010 149 2009 149 45 199 '()*+$,-$#./(0123$3415(678.+3$$6+$!%9:$5(4415/;+3$<$=+$>1.5?$@+45;3+2/(012$3=A;B(08.+$6+3$B./(0123$C%&D$A1B;161B(72+E$ )*+,-./#0123201/4-56/7#)8#9*+,-.:#;8#<35=4>:#)?;8#>6/#24@6<./#235.6A5/#96#>4#=A.4@36D# 3E(& 3E(&3 4B& 74" @8" @@" 594"599" 659" Mutation p.T99fs p.T99fs p.T99fs p.A210P p.A210P p.A210P c.607-1G>C c.607-1G>C Cas index Frère Père Cas Index Soeur Père Frère Soeur GH D D D* D D D D D TSH D D N D D N D D ACTH N N N D N N D D LH FSH NE NE D* D N N D N Hypophyse Hypoplasie Hypoplasie Hyperplasie Hypoplasie Hypoplasie + Normal Hypoplasie Hypoplasie + kyste kyste Lobe Non visualisé Normal Normal Normal Normal Normal Ectopique Ectopique postérieur IRM Hypoplasie Normale Normale Normale Normale Normale Syndrome de Syndrome de cérébrale du corps Chiari Chiari calleux Selle turcique Peu Peu Normale Normale Normale Normale Peu Peu développée développée développée développée Transmission Dominant Dominant Dominant Dominant Dominant Dominant Dominant Dominant Castinetti, Castinetti, Castinetti, Pfaeffle, Pfaeffle, Pfaeffle, Machinis, Machinis, JCEM, 2008 JCEM, 2008 JCEM, 2008 JCEM, 2008 JCEM, 2008 JCEM, AmJHum AmJHum 26 26 26 149 149 2008 Genet, 2001 Genet, 2001 149 117 117 !"#$%&' )"2 ()*"+,-.&/"01$1"$%.&2%&3456&7"88,7*9%.&:&;%&<,)7=&>%879.%-*"+,-&.;?90"+@)%&2%.&0)*"+,-.&A4BC&?,09,2,0"1-%D& $%&' "$$/1)"$'"3%&' ("5'"/0"2$%&' "/3 1"31//!" /3#1" 3. EVALUATION DU ROLE DU FACTEUR DE TRANSCRIPTION PITX2 DANS LE DEVELOPPEMENT DE L’AXE THYREOTROPE PITX2 AND PITX1 ARE NECESSARY FOR THYROTROPH FUNCTION AND RESPONSE TO HYPOTHYROIDISM 1F. Castinetti*, 1M.L. Brinkmeier*, 2D.F. Gordon, 3 K. R. Vella, 2J.M. Kerr, 3A. N. Hollenberg, 4T. Brue, 2E.C. Ridgway, 1S.A. Camper Soumis à Molecular Endocrinology ! $%! INTRODUCTION Le développement de l’axe thyréotrope au cours de l’ontogenèse hypophysaire est encore très imparfaitement élucidé. Chez la souris, 2 jours embryonnaires peuvent être soulignés : e10,5 correspondant à la première expression détectable de la sous-unité alpha, et e15,5 correspondant à la première expression détectable de Tshß. Il existe 2 contingents de cellules thyréotropes chez la souris : un premier localisé au niveau de l’extrémité rostrale, apparaissant à e12, mais disparaissant à la naissance ; un 2ème apparaissant dans la portion médio-caudale de la poche de Rathke vers e15, qui sera à l’origine des cellules thyréotropes fonctionnelles après la naissance (114). Le rôle des cellules rostrales reste mal déterminé. Il n’existe que peu de données sur les facteurs de transcription impliqués dans la différenciation des cellules hypophysaires progénitrices aboutissant à l’obtention de cellules thyréotropes. L’inactivation du gène de la sous-unité alpha (alpha Gsu) chez la souris a permis d’observer un modèle d’hypothyroïdie profonde, utile pour comparer les effets de certains facteurs de transcription hypophysaires (en gardant à l’esprit que ce modèle est également à l’origine d’un déficit gonadotrope) : A la naissance, les cellules thyréotropes sont normales. A 3 semaines, une hyperplasie des cellules à Tsh est observée; à 4,5 semaines, leur poids est inférieur de moitié à celui des souris contrôles (cassure de la courbe à 2-3 semaines). Les taux de T4 sont indétectables, la thyroïde n’est pas développée. L’hypophyse présente un marquage massif de Tshß correspondant à une hyperplasie des cellules thyréotropes (non sécrétantes du fait de l’absence de sous-unité alpha). A 6 mois, il existe une hyperplasie hypophysaire due à l’hyperplasie thyréotrope majeure. De façon intéressante, le nombre de cellules somatotropes et lactotrope est fortement réduit, soulignant l’intrication entre les différents réseaux de cellules hypophysaires (97). De nombreux modèles murins et l’étude phénotypique des patients porteurs de mutations de facteurs de transcription responsables d’hypopituitarisme congénital ont permis d’identifier 5 facteurs de transcription impliqués à des degrés variables dans ces étapes. LHX3, LHX4, PROP1 et POU1F1 ayant été détaillés précédemment, nous nous limiterons ici au rôle de GATA2. Gata2 est un facteur de transcription à homéodomaine de type « doigts de zinc ». Son expression est retrouvée dans l’hypophyse à partir de e10,5 jusqu’à l’âge adulte (dans les cellules thyréotropes et gonadotropes) (170). Les études en culture cellulaire ont démontré que Gata2 était capable de se lier et activer le promoteur de la sous-unité alpha , et d’agir de façon synergique avec Pit1 sur le promoteur de la Tshß (73; 43; 74). Cependant, le rôle de Gata2 semble plus important pour la différenciation des cellules gonadotropes que thyréotropes : ainsi, la surexpression de Gata2 entraîne une modification de la différenciation cellulaire hypophysaire avec une augmentation des cellules gonadotropes aux dépens des cellules pit1 dépendantes (28). L’inactivation conditionnelle de Gata2 au sein des cellules thyréotropes (via un système de recombinaison lox-Cre) a confirmé le rôle de ce facteur de transcription dans le développement des cellules thyréotropes. Les souris Gata2 flox/flox AlphaGsu-Cre présentent un retard de croissance chez les males (10% de poids en moins par rapport aux contrôles) mais qui s’atténue puis disparaît avec l’âge. Aucune différence n’était observée chez les femelles. En fait, il semble que Gata2 soit surtout important dans les premières semaines de vie : ainsi, en post-natal, il existe une différence majeure en termes de nombre de cellules thyréotropes et de volume thyroïdien entre les souris avec inactivation de Gata2 et les souris sauvages. Cependant, cette différence n’est plus présente à l’âge adulte. Une radio-ablation de la thyroïde entraînait par contre une faible augmentation de la Tshß chez les souris avec inactivation de Gata2 par rapport aux ! $&! contrôles, soulignant que Gata2 était probablement nécessaire pour la régénération des cellules thyréotropes. La délétion de Gata2 peut être compensée par un autre facteur, ce qui expliquerait la récupération progressive avec l’âge des souris ayant une inactivation de Gata2. Un de ces facteurs pourrait être Gata3, dont l’expression était retrouvée augmentée chez les souris avec inactivation de Gata2 (28). Aucune mutation de GATA2 responsable d’hypopituitarisme congénital n’a été observée chez l’homme. D’autres facteurs de transcription sont assurément nécessaires pour permettre la différenciation et l’expansion du contingent de cellules thyréotropes. Nous nous sommes intéressés plus particulièrement au facteur de transcription Pitx2. Les souris Pitx2 -/- (61; 63; 112) présentent une forte mortalité précoce (plus de 35% de mortalité avant e10, 100% de mortalité à e14,5) : l’hypophyse est hypoplasique dès e10, et son développement est arrêté à partir de e12,5 avec une absence d’expression de Hesx1, Pit1, Tshß, Lhx4, et un très faible niveau d’expression de Gata2 et Prop1. L’hypophyse des souris homozygotes pour un allèle hypomorphe de Pitx2 ( Pitx2 neo/neo ) semble de morphologie comparable à celle des souris sauvages (63; 192). Le nombre de cellules thyréotropes est modérément réduit en comparaison avec les souris sauvages. L’expression de certains facteurs de transcription est également modifiée : ainsi Gata2 est exprimé normalement à e12,5 mais n’est pas retrouvé à P1. Pit1 est également diminué, au contraire de Prop1 (63; 192). In vitro, Pitx2 et Lhx3 agissent de façon synergique sur le promoteur de la sous unité alpha (112). Pitx2 (comme Pitx1) est capable d’activer les promoteurs de la Tshß et Gata2 (52; 153; 154), et interagit de façon synergique avec Pit1, sur le promoteur de la Tshß . Pitx2 semble donc nécessaire au développement de l’axe thyréotrope. Cependant, il existe une grande similarité de séquence entre Pitx1 et Pitx2 : l’homéodomaine de PITX2 est identique à 97% à celui de PITX1 (seulement 2 substitutions d’AA). Le domaine C-terminal est identique à 67%, au contraire du domaine N-terminal (30% d’homologie) (63). Pitx1 et Pitx2 ont un profil d’expression proche au cours de l’embryogenèse: ils sont exprimés à partir de e8 dans le stomodeum, et maintenus dans toutes les lignées qui en dérivent, et en particulier la poche de Rathke. Au cours du développement hypophysaire, Pitx1 est exprimé dans la poche de Rathke entre e10,5 et e15 (192). A e14,5, Pitx1 est cependant exprimé dans l’ensemble de la poche, alors que Pitx2 n’est plus retrouvé dans le lobe intermédiaire. Ce profil est retrouvé à e18,5 (29). A l’âge adulte, les niveaux d’expression de Pitx1 et Pitx2 en fonction du type cellulaire sont sensiblement les mêmes, avec 80 à 90% des cellules thyréotropes et gonadotropes exprimant Pitx1. Cependant, les souris Pitx1 -/- présentent une poche de Rathke normale à l’initiation de son développement (au contraire des souris Pitx2 -/-). Cela suggère que Pitx2 joue un rôle plus précoce dans le développement de la poche de Rathke (192). Le phénotype hypophysaire des souris Pitx1 -/- est plus proche de celui des souris Pitx2 neo/neo (avec plus de cellules corticotropes, et moins de gonadotropes et thyréotropes). Dans les doubles mutants Pitx1 -/- Pitx2 neo/neo , la poche de Rathke ne se développe pas après e12,5, ce qui suggère que le développement de la poche nécessite l’expression des 2 facteurs de transcription. Il est donc probable que Pitx2 et Pitx1 ont des rôles complémentaires lors du développement hypophysaire. Pitx2 a probablement une action plus précoce, mais Pitx1 est nécessaire pour l’expression de certains facteurs de transcription comme lhx3, qui jouent un rôle majeur dans le développement hypophysaire. Ce rôle proche de Pitx2 et Pitx1 explique vraisemblablement que la délétion de Pitx2 au sein des cellules gonadotropes (Pitx2 flox/- LhßCre ) n’a pas mis en évidence de retard pubertaire, ou de troubles de la fertilité (27). ! $'! Objectifs du travail L’objectif de ce travail était donc de vérifier l’hypothèse selon laquelle Pitx2 est nécessaire au développement de la lignée thyréotrope. Plusieurs arguments suggèrent en effet un rôle de Pitx2 dans le développement de l’axe thyréotrope - Son expression à l’âge adulte dans plus de 80% des cellules thyréotropes - Sa possibilité en culture cellulaire d’activer plusieurs promoteurs de gènes impliqués dans le développement thyréotrope : Gata2, Lhx3, Pou1f1, Tshß - La diminution d’expression de facteurs de transcription impliqués dans le développement thyréotrope dans le modèle de souris Pitx2 -/- : Prop1, Gata2, Pou1f1 ! $(! "! PITX2 AND PITX1 ARE NECESSARY FOR THYROTROPH FUNCTION AND #! RESPONSE TO HYPOTHYROIDISM $! %! 1F. Castinetti*, 1M.L. Brinkmeier*, 2D.F. Gordon, 3 K. R. Vella, 2J.M. Kerr, 3A. N. &! Hollenberg, 4T. Brue, 2E.C. Ridgway, 1S.A. Camper '! (! 1Human Genetics, University of Michigan, Ann Arbor, MI, United States, 48109-5618 )! 2Department of Medicine, University of Colorado, Aurora, CO, United States, 80045. *! 3Beth Israel Deaconess Medical Center, Harvard University, NJ, United States "+! 4Department of Endocrinology, La Timone Hospital, Marseille, France ""! "#! *Both authors contributed equally to this work "$! "%! Short title: PITX2 and PITX1 in thyrotrophs "&! "'! "(! Corresponding author ")! "*! Sally A. Camper, Ph.D ., #+! Dept. Human Genetics #"! University of Michigan ##! 4909 Buhl Bldg. #$! 1241 Catherine St. #%! Ann Arbor, MI 48109-5618 #&! Telephone: 734-763-0682 #'! Fax: 734-763-5831 #(! email: [email protected] #)! #*! Keywords : Pitx2 , Pitx1 , pituitary, Gata2 , TSH, T4 challenge, thyrotroph-cre transgene $+! $"! The authors have nothing to disclose. $#! $$! Financial support came from Novo-Nordisk, Societe Francaise d’endocrinologie, $%! Novartis, Ipsen, ADEREM, and the Center for Genetics in Health and Medicine, $&! University of Michigan (all for FC), and the National Institutes of Health (R37HD30428, $'! R01HD34283 to SAC). $(! $)! ! "! #! ABSTRACT (235 WORDS) "! $! Pitx2 is a homeodomain transcription factor required in a dose dependent manner for the %! development of multiple organs. Pitx2 null homozygotes ( Pitx2 -/-) have severe pituitary &! hypoplasia, while mice with reduced function alleles ( Pitx2 neo/neo ) exhibit modest '! hypoplasia and reduction in the developing gonadotroph and Pou1f1 lineages. PITX2 is (! expressed broadly in Rathke’s pouch and the fetal pituitary gland. It predominates in )! adult thyrotrophs and gonadotrophs, although it is not necessary for gonadotroph *! function. To test the role of PITX2 in thyrotroph function we developed thyrotroph- #+! specific cre transgenic mice, Tg(Tshb-cre ) with a recombineered Tshb BAC that ablates ##! floxed genes in differentiated pituitary thyrotrophs. We used the best Tg(Tshb-Cre) #"! strain to generate thyrotroph-specific Pitx2 deficient offspring, Pitx2 flox/-;Tg(Tshb-cre) . #$! Double immunohistochemistry confirmed efficient, specific Pitx2 deletion. Pitx2 flox/- #%! ;Tg(Tshb-cre) mice have a modest weight decrease. The thyroid glands are #&! proportionately smaller, and circulating T4 and TSH levels are in the normal range. The #'! pituitary levels of Pitx1 transcripts are significantly increased, suggesting a #(! compensatory mechanism. Hypothyroidism induced by low iodine diet and oral propyl- #)! thio-uracil revealed a blunted TSH response in Pitx2 flox/-; Tg(Tshb-cre) mice. Pitx1 #*! transcripts increased significantly in control mice with induced hypothyroidism, but they "+! remained unchanged in Pitx2 flox/-; Tg(Tshb-cre) mice, possibly because Pitx1 levels were "#! already maximally elevated in untreated mutants. These results suggest that both ""! PITX2 and PITX1 have overlapping roles in thyrotroph function and response to "$! hypothyroidism. The novel cre transgene that we report will be useful for studying the "%! function of other genes in thyrotrophs. "&! "'! "(! ")! ! "! #! INTRODUCTION $! "! The pituitary gland controls development and function of several important target %! glands. In mice, the anterior and intermediate lobes of the pituitary gland are derived &! from the oral ectoderm that invaginates to form Rathke’s pouch, which gives rise to five '! pituitary specific cell lineages that produce growth hormone, luteinizing hormone and (! follicle-stimulating hormone, thyroid stimulating hormone, prolactin, and )! adrenocorticotropin (1). Pituitary organogenesis is a complex multi-step process under *! the control of at least a dozen known transcription factors and likely several that remain #+! unknown (2, 3). Mutations in the genes coding for these transcription factors lead to ##! multiple pituitary hormone deficiencies. For instance, recessive or dominant mutations #$! of POU1F1 , a pituitary specific POU homeodomain transcription factor, also known as #"! PIT1 , generally cause congenital somatotroph, lactotroph, and thyrotroph deficiencies #%! (2). Mutations in the LIM domain transcription factor LHX4 cause somatotroph #&! deficiency and variable thyrotroph, gonadotroph and corticotroph insufficiencies (4). #'! Despite the known transcription factor mutations, most cases of congenital #(! hypopituitarism remain of unknown etiology. Conditional knockout mice represent an #)! ideal model to improve the knowledge of hypopituitarism. Indeed, most of the genes #*! known to cause hypopituitarism were predicted from mouse models. $+! $#! Mutations in LHX3 , LHX4 , PROP1 and POU1F1 have been described in patients $$! with combined pituitary hormone deficiency, which includes thyrotroph deficiency in the $"! majority of cases (2). The mechanisms of thyrotroph cell specification remain $%! incompletely understood. Roles for pituitary transcription factors in thyrotroph $&! ontogenesis and expansion have been proposed. Early acting transcription factors like $'! LIM domain transcription factors LHX3 and LHX4 are crucial because mice with a ! "! #! homozygous inactivation of these transcription factors present with pituitary hypoplasia $! and decreased or absent thyrotroph cells (5-7). These LIM genes are not specific for %! thyrotrophs, however, as affected humans and mice have additional hormone "! deficiencies. Some later-acting transcription factors are necessary for thyrotroph &! development. Ames and Snell dwarves lack thyrotrophs due to spontaneous inactivation '! of the paired transcription factor PROP1 and the POU homeodomain transcription factor (! POU1F1 (or PIT1), respectively (8-10). )! *! Gata2 was proposed to be critical for thyrotroph specification, but loss of function #+! studies in mice suggest that it has a more modest role (11). Mice homozygous for ##! pituitary specific inactivation of Gata2 have transient growth insufficiency and #$! hypogonadism. They are unable to respond to induced hypothyroidism with a robust #%! elevation of TSH levels, but elevated Gata3 expression suggests that other members of #"! the same gene family may compensate for the loss of Gata2 (11). #&! #'! PITX2 is a homeodomain transcription factor involved in the early steps of #(! pituitary organogenesis (12-14). PITX2 expression is first observed at e8 in the #)! stomodeum and at e10 in Rathke’s pouch. It is still expressed in the adult pituitary gland #*! (13, 15, 16). Mice with homozygous inactivation of Pitx2 ( Pitx2 -/-) present with severe $+! pituitary hypoplasia due to reduced proliferation and enhanced cell death (13, 15, 16). It $#! is difficult to evaluate the role of PITX2 in developing specific pituitary cell lineages $$! because embyros die at e12.5 due to severe heart defects. Mice homozygous for a $%! hypomorphic, or reduced function, allele of Pitx2 ( Pitx2 neo ) survive until birth. Their $"! pituitaries lack SF1 and gonadotropins and have reduced differentiation of the POU1F1 $&! lineage, resulting in fewer somatotrophs and thyrotrophs (15). To characterize the role $'! of PITX2 in specific pituitary lineages, it must be deleted specifically in that cell type. ! "! #! Deletion of PITX2 in gonadotrophs using an Lhb-cre transgene did not alter puberty or $! fertility, suggesting it is dispensable in differentiated gonadotrophs (17). %! &! Several lines of evidence support the hypothesis that PITX2 plays a role in "! thyrotroph development, maintenance, and/or function. First, during embryogenesis, the '! expression of several transcription factors, including PROP1, POU1F1, and LHX4, is (! decreased in the pituitaries of mice with homozygous inactivation of Pitx2 ( Pitx2 -/-). This )! suggests that PITX2 is a master regulator of these downstream genes (15, 16). Second, *! PITX2 is able to transactivate Pou1f1 and Tshb promoters in cell culture, which is #+! consistent with a direct role for PITX2 in the committed thyrotroph (18, 19). Third, more ##! than 80% of adult thyrotroph cells express PITX2, implying a role in adult thyrotroph #$! function and/or maintenance (15). Fourth, Pitx2 neo/neo mice have reduced expression of #%! POU1F1 and fewer thyrotroph cells, implicating PITX2 as a dosage sensitive regulator of #&! commitment or expansion of the thyrotroph lineage (15, 16). #"! #'! Multiple lines of evidence support the idea of overlapping roles of PITX1 and #(! PITX2 in the developing pituitary gland. PITX1 and PITX2 have nearly identical #)! homeodomains and highly homologous carboxy termini. Both transcription factors can #*! bind the same sites in cell culture, have similar trans-activation activity in cell culture (19), $+! and are present in the majority of thyrotroph cells in adults (16). Double heterozygotes $#! (Pitx1 +/-, Pitx2 +/-) have poor viability, but the double mutant ( Pitx1 -/-, Pitx2 -/-) has arrested $$! development of the pituitary primordium and lacks LHX3 expression, which is unaffected $%! in the single mutants. The pituitaries of Pitx1 -/- embryos are not obviously altered in size, $&! and ( Pitx2 neo/neo mice have a modest reduction in pituitary volume, but these double $"! mutants Pitx1 -/-, Pitx2 neo/neo ) have minute pituitary primordia similar to Pitx2 -/- embryos, $'! indicating overlapping function in growth of the primordium. This is consistent with the ! "! #! idea of overall gene dosage for genes with similar function (20). The LIM genes Lhx3 $! and Lhx4 also have overlapping function in establishing the pituitary primordium (21). %! &! To test our hypothesis that PITX2 has a role in thyrotroph maintenance and '! function, we generated a new cre recombinase driven by recombineering a Tshb "! bacterial artificial chromosome (BAC) (22). Previous studies with smaller portions of the (! Tshb gene were unsuccessful in directing thyrotroph specific expression in transgenic )! mice (23). Pitx2 flox/-;Tg(Tshb-cre) mice delete Pitx2 specifically in thyrotrophs once they *! are specified and begin to produce TSH. The Pitx2 flox/-;Tg(Tshb-cre) mice produce TSH #+! with striking elevation of Pitx1 expression, and they have slightly reduced growth. The ##! affected mice have a blunted pituitary response to hypothyroidism, suggesting that #$! PITX2 is important for thyrotroph function, yet it can be partly compensated for by #%! increased Pitx1 transcription. #&! #'! #"! ! "! #! RESULTS $! %! Tshb-cre transgene is specific for pituitary thyrotroph cells &! '! The Tshb-cre transgene construct was generated by using recombineering to (! insert a cre recombinase cassette at the transcription start site of a BAC from a "! C57BL6/J library that contains -144kb to +58kb of the Tshb gene . Four BAC transgenic )! founders contained the entire Tshb-cre transgene construct and were bred to a cre *! reporter strain (floxed LacZ) to characterize the specificity of cre activity for each line #+! (17). Three of the Tg(Tshb-cre) lines had moderate levels of LacZ expression in the ##! pituitary gland, but it was not sufficiently specific to the thyrotrophs. One Tg(Tshb-cre) #$! line exhibited higher levels of LacZ expression in the pituitary and excellent specificity for #%! thyrotrophs. This line was selected for further analysis of developmental regulation and #&! tissue specificity. #'! #(! Transgene expression was analyzed at e15.5, when endogenous TSHß is first #"! detectable in the caudo-medial aspect of the pituitary. As expected, X-gal staining is #)! detectable in this area and in the POU1F1 independent, rostral tip thyrotrophs in #*! transgenics and not in controls ( Figure 1, panel A ). X-gal staining is strong in the $+! anterior lobe of the adult pituitary gland and co-localizes with the majority of the $#! thyrotrophs. A few somatotroph cells stained with X-gal. These may correspond to GH- $$! TSH double positive cells (24). No X-gal staining was observed in gonadotroph, $%! lactotroph, or corticotroph cells identified by LH, PRL and ACTH immunostaining, $&! respectively (Figure 1, panel B ). This indicates efficient, developmentally regulated and $'! cell-specific expression of the Tshb-cre transgene. $(! ! "! #! Adult tissues were examined from Tg(Tshb-cre); floxed-LacZ mice to ascertain $! the degree of transgene leakiness in non-pituitary tissues. There was no evidence of %! transgene activity in the brain or the testis. Transgene activity was detected at very low &! levels in a few cells in the heart, the kidney, the liver, and at moderate levels in the lung '! (Figure 1, panel C ). The faint X-gal staining detected in the thyroid of Tg(Tshb- (! cre);floxed-LacZ mice was indistinguishable from X-gal staining in the nontransgenic )! controls, consistent with a low background level of endogenous ß-galactosidase activity "! in the thyroid gland ( Figure 1, panel C ). We confirmed that there is no PITX2 *! expression in the thyroid glands of control mice (data not shown) (25). Thus, this Tshb- #+! cre line is appropriate for examining gene function in pituitary thyrotrophs. ##! #$! Pitx2 flox/-;Tg(Tshb-cre) mice are smaller than wild type littermates #%! The Tg(Tshb-cre) strain was mated with Pitx2 +/- mice to create a stock, Pitx2 +/- #&! ;Tg(Tshb-cre ) for mating with Pitx2 flox/flox mice. We obtained Pitx2 flox/-;Tg(Tshb-cre) mice #'! in the expected Mendelian ratio. Of 162 progeny 23% were Pitx2 flox/-;Tg(Tshb-cre), 19% #(! were Pitx2 flox/ +;Tg(Tshb-cre) , and among those without the Tshb-cre transgene, 32% #)! were Pitx2 flox/ + and 19% were Pitx2 flox/- mice (p=0.304). Co-immunostaining with PITX2 #"! and TSH antibodies confirmed that the majority of thyrotroph cells expressed PITX2 in #*! controls, whereas there was no expression of PITX2 in most of transgenic thyrotroph $+! cells ( Figure 2 ). $#! Male and female Pitx2 flox/-;Tg(Tshb-cre) mice weighed less than their control $$! littermates. The difference in weights were most significant in females at 5 weeks $%! (p=0.001), but the mutants were smaller at other ages post weaning (p<0.01 at 4 and 6 $&! weeks, p<0.05 at 7 and 8 weeks. Male mutants were smaller at 5 weeks (p<0.05). $'! (Figure 3 ). This growth insufficiency is modest compared to animals with severe $(! hypothyroidism, such as the alpha-subunit knockout mice, Cga tmSac , which are unable to ! "! #! produce biologically active TSH (26). To test for growth hormone deficiency, we $! measured Igf1 mRNA levels in Pitx2 flox/-;Tg(Tshb-cre) and control females (n=3 vs. 2 %! controls). There was no significant difference between these groups (data not shown). &! (27) '! (! Pitx2 flox/-;Tg(Tshb-cre) mice present subtle thyroid changes compared to controls )! To assess the number and size of thyrotroph cells in the pituitaries of Pitx2 flox/- *! ;Tg(Tshb-cre) mice, we performed immunohistochemistry with an antibody against TSH. "! At 8 weeks of age, we did not observe any significant difference in the number or size of #+! the thyrotrophs in affected mice and controls ( Figure 4, panel A ). The levels of total T4, ##! Tshb mRNA, and serum TSH were also normal in adults ( Figure 4, panel B-D). The #$! thyroid volume was smaller, although it is close to proportional (p=0.067 after correction #%! for weight), and the follicular size and height of the follicular epithelia were #&! indistinguishable ( Figure 4, panel B ). #'! #(! After birth the hypothalamic-pituitary-thyroid axis becomes responsive to #)! feedback regulation (28). To determine whether thyroid hormone feedback #*! progressively corrected an earlier thyrotroph deficiency we examined newborns and 4 #"! wk old mice. Immediately after birth, the number of thyrotroph cells was not obviously $+! reduced ( Figure 4, panel E ). At 4 weeks of age: thyroid volumes were smaller, but the $#! other parameters including thyrotroph cell size and number, Tshb mRNA levels, and T4 $$! levels were normal (data not shown). $%! $&! PITX2 and PITX1 are 2 closely related homeodomain transcription factors: both $'! are expressed in adult thyrotroph cells, and they are able to transactivate Pou1f1 and $(! Tshb promoters (19). Interestingly, at 8 weeks of age, the levels of Pitx1 mRNA were ! "! "! significantly increased in transgenics compared to controls (24 fold increase, p<0.001) $! (Figure 4, panel D ). This suggests that PITX1 might compensate for the deficiency of %! PITX2 in transgenic mice, allowing them to achieve a nearly euthyroid phenotype with a &! modest growth defect, smaller thyroid volume, but hormone levels in the normal range. '! (! Pitx2 flox/-;Tg(Tshb-cre) mice have a blunted TSH response to hypothyroidism )! We hypothesized that mice lacking Pitx2 in thyrotrophs would have a blunted *! TSH response to low thyroid hormone levels, similar to that observed in the pituitary +! specific Gata2 knockout mice (11). We challenged transgenic mice and controls (n=6 "#! vs. 8 controls) by inducing hypothyroidism with a low iodine diet enriched in a ""! goiterogen, 0.15% PTU, for 4 weeks. As expected, control mice exhibited decreased "$! circulating T4 levels (p<0.001), causing a 65-fold increase in pituitary Tshb mRNA "%! (p<0.001), and an 80-fold increase in serum TSH levels (p<0.001) ( Figure 5, panels A, "&! B and C ). Transgenic mice had a 32-fold increase in Tshb mRNA (p<0.001) and 48-fold "'! increase in serum TSH levels (p<0.001) at a comparable level of total T4. These "(! increases are inferior to those observed in controls. The hypothyroidism-induced ")! pituitary response in serum TSHß levels is significantly different in controls and "*! transgenic mice: 80+/-23 vs. 48+/-15 pg/ul respectively (p=0.04). "+! $#! To explore the mechanism of the blunted TSH response to hypothyroid $"! challenge, we measured Pitx1 and Pitx2 mRNA levels. The hypothyroid challenge did $$! not modify Pitx2 mRNA levels in controls (data not shown), but Pitx1 mRNA levels were $%! increased 20 fold relative to untreated controls (p<0.001). Pitx1 mRNA levels were the $&! same in challenged and unchallenged transgenics; the apparent 1.6 fold elevation is not $'! significant ( Figure 5, panel B ). The transgenics may have been unable to increase $(! Pitx1 transcription in response to the challenge because the levels were already ! "#! "! elevated, presumably to compensate for the lack of Pitx2 and develop a euthyroid state. #! These results suggest that PITX1 is a major factor in the pituitary's response to $! hypothyroidism with increased TSH transcription and secretion. This also implies that %! PITX2 is necessary to obtain an optimal response. &! ! ""! "! DISCUSSION #! $! Tshb-cre transgenic mice provide an effective tool for genetic engineering in %! pituitary thyrotrophs &! Previous attempts to utilize the Tshb promoter, 5' flanking region and large intron '! to drive transgene expression in thyrotrophs were not successful (23). We reasoned that (! essential regulatory sequences must be located at a distance from the gene and used a )! large Tshb BAC to drive expression. The cre recombinase is active at the time Tshb *! transcription is initiated, e15.5. This makes the transgenic line a useful tool for genetic "+! modifications in mature thyrotrophs. It cannot provide information about the process of ""! thyrotroph development prior to the initiation of Tshb transcription. One in four Tshb-cre "#! transgenic lines yielded developmentally regulated, efficient, thyrotroph-specific "$! expression, with little or no activity in inappropriate sites. This suggests that the "%! essential regulatory sequences lie within the BAC, which could be engineered to drive "&! expression of other genes in thyrotrophs. "'! "(! Pitx2 deletion in thyrotrophs leads to mild growth reduction and blunted TSH ")! response to hypothyroidism challenge "*! Specific inactivation of Pitx2 in thyrotrophs leads to mild phenotypic changes in #+! physiologic conditions: Pitx2 flox/-;Tg(Tshb-cre) mice have a modest growth insufficiency, #"! but TSH and thyroid hormone levels are apparently in the normal range. It is possible ##! that the TSH and T4 measurements missed small, but biologically relevant variations in #$! hormone levels or a period of reduced hormone production, but it is clear that PITX2 is #%! not essential for TSH production or survival of thyrotroph cells. The reduced growth is #&! more obvious at 5 wks than 8 wks, suggesting that pituitary changes in gene expression, #'! such as increased Pitx1 transcription, might progressively compensate for the lack of ! "#! "! Pitx2 . We considered the possibility that Pitx2 flox/-;Tg(Tshb-cre) mice might have $! reduced growth hormone production because a few somatotroph cells express the #! recombinase and growth hormone transcription is regulated by thyroid hormone (29). %! Although we cannot completely rule out an effect on growth hormone production, it is &! unlikely to be a contributor because most GH cells do not express PITX2 or cre '! recombinase and liver Igf1 transcription is unaltered. (! Hypothyroidism induced by low iodine diet enriched in PTU, an anti-thyroid drug, )! leads to dramatically increased TSH levels. Pitx2 flox/-;Tg(Tshb-cre) mice had a blunted *! TSH response, suggesting that PITX2 deficient thyrotrophs are only partially able to "+! respond to low thyroid hormone levels. This observation supports the idea that PITX2 ""! deficient thyrotrophs function less efficiently than normal thyrotrophs, possibly "$! contributing to the modest growth deficiency. The phenotypic anomalies of Pitx2 flox/- "#! ;Tg(Tshb-cre) mice are remarkably similar to those observed in the pituitary specific "%! knockout of Gata2 (11). The Gata2 Pitko males have a transient growth deficiency around "&! 5 wks of age, and newborns have reduced TSH immunostaining that is quickly "'! corrected. Challenging Gata2 knockout mice with radioactive iodine-mediated thyroid "(! ablation also produced a blunted TSH response, and Gata3 transcription was ")! dramatically increased. These data support the idea that Gata2 and Pitx2 are important "*! transcription factors for thyrotroph function which can be compensated for by related $+! genes, Gata3 and Pitx1, respectively. $"! $$! PITX1 partially compensates for PITX2 $#! PITX1 and PITX2 have nearly identical homeodomains and conserved C termini, $%! they are expressed in the majority of adult thyrotrophs, and they are able to transactivate $&! the same promoters in cell culture (15, 16, 18, 19). PITX1 appears to be a major $'! contributor to pituitary-thyroid axis homeostasis because normal mice respond to ! "#! "! hypothyroidism with elevated Pitx1 transcript levels and there is no change in PItx2 $! transcription. Pitx2 flox/-;Tg(Tshb-cre) mice have dramatically increased Pitx1 transcripts %! under basal conditions, suggesting that Pitx2 deficient thyrotrophs have reduced function #! and compensate by elevating Pitx1 transcription. The level of Pitx1 transcripts in normal &! mice with induced hypothyroidism equals the basal levels of Pitx1 in transgenics. This '! implies that Pitx1 is maximally induced by either hypothyroidism or PITX2 deficiency, (! and that the low iodine diet is unable to exert any further elevation of Pitx1 transcription )! in PITX2 deficient thyrotrophs, resulting in the blunted response. Taken together, these *! results are consistent with the idea that the modest growth defect of Pitx2 flox/-;Tg(Tshb- "+! cre) mice under basal conditions are caused by the loss of PItx2 in thyrotrophs. ""! We expected that PITX2 would be important for gonadotroph function because "$! Pitx2 neo/neo mice lack gonadotrophs and have decreased expression of the gonadotroph "%! transcription factors Gata2 , Nr5a1 and Egr1 . Moreover, PITX2 is expressed in the "#! majority of adult gonadotrophs. We ablated Pitx2 in gonadotrophs using a specific Lhb- "&! cre recombinase and found that PITX2 is dispensable for gonadotroph function after "'! birth (See supplemental data) (17). PITX1 is expressed in adult gonadotrophs and likely "(! compensates for PITX2 deficiency in those cells as well as thyrotrophs. ")! Pitx1 mutant pituitaries are minimally affected just before birth, while Pitx2 null "*! pituitaries are extremely underdeveloped. Thus, PITX2 is more important than PITX1 in $+! pituitary development (30). Our data suggests that PITX1 is an important contributor to $"! adult pituitary function, especially in response to hypothyroidism, and that the normal $$! combined dosage of PItx1 and Pitx2 are necessary for pituitary regulation of $%! homeostasis. $#! $&! PITX2 mutations are rare in patients with combined pituitary hormone deficiencies ! "#! "! A variety of PITX2 mutations have been reported in patients with Axenfeld-Rieger $! syndrome. This is a genetically heterogenous condition that consists in under- %! development of the anterior segment of the eye and associated tooth and craniofacial &! anomalies (31-34). A few patients have morphological anomalies of the sella turcica #! (35). Only 3 patients with Axenfeld Rieger syndrome have been reported with pituitary '! hypoplasia and GH deficiency, without thyrotroph deficiency. Our study of genetically (! modified mice suggests that PITX2 deficient patients may compensate by inducing )! elevated expression of PITX1. The stature of Axenfeld Rieger patients is usually normal *! or rarely slightly reduced (33). Administering a TRH stimulation test for Axenfeld-Rieger "+! patients could test this idea, but it is not clinically justifiable because these patients ""! would only present partial thyrotroph deficiency if hypothyroid, which could be easily "$! treated with thyroid hormone replacement. It is possible that some patients with isolated "%! TSH deficiency could have mutations in PITX1 or PITX2 but the mutations would likely "&! be in regulatory regions specific for thyrotrophs because systemic gene deletion in mice "#! causes limb and cardiac anomalies in Pitx1 and Pitx2 deficient mice, respectively. "'! The overlapping roles of PITX1 and PITX2 in the pituitary gland probably explain "(! the very low frequency of pituitary deficiencies in patients with Axenfeld-Rieger ")! syndrome. The fact that doubly heterozygous mice have extremely poor viability points "*! to the likelihood of overlapping roles in other critical organs like the head and the heart $+! (30). Doubly heterozygous mutations of PITX1 and PITX2 could exist in patients with $"! pituitary deficiencies and effects on a constellation of other organs. The phenomenon of $$! digenic inheritance has been reported for patients with hypogonadotrophic $%! hypogonadism and mutations in FGFR1 , GnRH receptor and NELF (36). Although $&! humans doubly heterozygous for mutations in PITX1 and PITX2 might be not viable, the $#! threshold for dosage sensitive defects is often different in humans and mice (37). $'! ! "#! "! In conclusion, we developed and characterized a new effective cre recombinase $! transgenic line that is specific for committed thyrotroph cells. This tool allowed us to %! determine that PITX2 and PITX1 cooperate to maintain thyrotroph function. Mice with a &! conditional inactivation of Pitx2 in thyrotrophs are able to maintain TSH production with '! elevation of Pitx1 expression, and do not undergo thyrotroph hypertrophy or hyperplasia. #! The affected mice have a moderate growth deficiency, and a blunted TSH response to (! hypothyroidism challenge, which favors the idea of a slight thyrotroph deficiency. The )! action of PITX2 is partially compensated by PITX1, and these 2 transcription factors *! have overlapping roles. These results also suggest that patients with Axenfeld-Rieger "+! syndrome induced by PITX2 mutations could present a partial thyrotroph deficiency. ""! "$! "%! ! "#! "! MATERIALS AND METHODS $! %! &! Generation of Tshb-cre transgene construct '! The Tshb-cre transgene was generated by recombineering a mouse Tshb BAC (! from a male C57BL6 library (RP24-230F23, Roswell Park Institute) (22). The Tshb- #! containing BAC included: 144 kb of 5’ flanking region, all 5 exons, with intervening )! introns, and 40 kb of 3’ flanking region. The first three exons of the murine Tshb gene *! are non-coding, and contain an alternate splice site, whereas exons 4 and 5 encode the "+! complete protein. In the recombineered BAC, the coding region of Tshb was substituted ""! for NLS-cre-β-actin, using an NLS-cre-β-actin cassette, flanked by ~50 bp 5’ and 3’ "$! homology arms of Tshb immediately upstream and downstream of the initiation and "%! termination codons, respectively. The final Tshb cre BAC construct was confirmed by "&! amplification and DNA sequencing of NLS-cre-β-actin , using primers outside of the "'! recombination site (exons 4 and 5 of Tshb ). The primers 5’-tgttcttgttattggctgtg-3’ and 5’- "(! gcttcgtaagctctctaatg-3’ were used with the following conditions: 94 ºC for 3 min, followed "#! by 35 cycles of 98º C for 20 s, 57º C for 30 s, and 72º C for 90 s, and a final 72º C ")! extension step for 10 min. In addition, the transgene-vector (pTARBAC) splice junctions "*! were confirmed using T7 and SP6 primer sites in the vector. $+! $"! Mice $$! All mice were maintained at the University of Michigan under the guidelines of the $%! Unit for Laboratory Animal Medicine and the University Committee for Care and Use of $&! Animals. $'! The transgenic founders carrying the Tg(Tshb-cre) BAC were generated by $(! microinjecting (C57BL/6 x SJL)F1xF1 fertilized eggs and transferring to surrogate ! "#! "! mothers, as previously described (23). Founders were identified by amplification of $! genomic DNA from tail biopsies with primers designed to amplify the vector-genomic %! DNA junctions at the 5’ and 3’ ends of the vector and at the Tshb-cre fusion site in exons &! 4 and 5. The universal cre primers 5’-gcataaccagtgaaacagcattgctg-3’ and 5’- '! ggacatgttcagggatcgccaggcg-3’ were used with the following conditions: 94°C for 3 min, (! followed by 32 cycles of 94°C for 30 s, 60°C for 60 s, and 72°C for 90 s, and a final 10- )! min extension at 72°C. #! Tg(Tshb-cre) mice were maintained by crossing transgenic founders with *! C57BL/6J mice from The Jackson Laboratory. B6;129-Gt(ROSA)26Sortm2Sho/J cre- "+! reporter mice were obtained from The Jackson Laboratory and maintained as ""! homozygotes. These mice, referred to here as floxed-LacZ mice, were genotyped by "$! PCR using primers 5’-ggcttaaaggctaacctgatgtg-3’, 5’-gcgaagagtttgtcctcaacc-3’, and 5’- "%! ggagcgggagaaatggatatg-3’ under the following conditions: 94°C for 3 min, followed by 35 "&! cycles of 94°C for 30 s, 64°C for 60 s, and 72°C for 60 s, and a final 10-min extension at "'! 72°C. The B6;129-Gt(ROSA)26-Sortm2Sho/J band is 1,146 bp and wild-type band is "(! 374 bp. ")! For transgene analysis, experimental animals carried one allele of the cre "#! transgene and one allele of the reporter gene, whereas controls were negative for the "*! cre transgene but positive for a reporter gene. $+! Pitx2 flox/-;Tg(Tshb-cre) mice were generated by mating B6 -Pitx2 +/- mice with $"! Tg(Tshb-cre) positive mice. The Pitx2 +/-;Tg(Tshb-cre) offspring were mated to B6- $$! Pitx2 flox/flox mice, and genotyping was performed as previously described (15). $%! $&! $'! Tissue Preparation and Histology $(! ! "#! "! X-gal staining $! X-gal staining was performed as previously described (38). Pituitaries were fixed %! in 4% formaldehyde overnight, rinsed in PBS, dehydrated and embedded in a Citadel &! 1000 (Thermo Electric, Chesire, England) paraffin-embedding machine, and sectioned '! coronally at 5 µm thickness. Immunohistochemical detection of the pituitary hormones (! was performed as previously described (39). Embryos and adult organs were dissected, )! frozen on dry ice, and stored at -80°C. Embryos were embedded in OCT (Sakura *! Finetek, Torrance, CA) and cryosectioned at 16 µm thickness. After X-gal staining, #! sections were counterstained for 2 min with 1% neutral red stain plus 4% sodium "+! acetate:glacial acetic acid. Sections were dehydrated and mounted with ""! xylene:permount 1:2 (Fisher) mounting media. Whole mount capture of freshly "$! dissected Tg(Tshb-cre Sac1 x B6;129-Gt(ROSA)26Sortm2Sho/J progeny was achieved "%! using a Leica MZFL III stereo/dissecting fluorescent microscope. "&! "'! Pituitaries "(! Adult pituitaries were collected at 4 and 8 wks of age and fixed for 1 h in 4% ")! formaldehyde in phosphate buffered saline (PBS). Newborn heads (day 2 after birth) "*! were also collected and fixed overnight in 4% formaldehyde in PBS. Pituitaries and "#! heads were rinsed in PBS, dehydrated and embedded in a Citadel 1000 (Thermo $+! Electric, Chesire, England) paraffin-embedding machine, and sectioned coronally at 5 $"! µm thickness. $$! Immunohistochemistry for PITX2 and TSHß was performed on pituitary sections $%! from 4 and 8 wk old mice. Epitopes were unmasked by boiling in citric acid (10mM) for 5 $&! minutes. After 10 minutes recovery at room temperature, endogenous peroxidases were $'! quenched in 3% H 202 for 20 minutes. After 1-hour of TSA blocking (Perkin Elmer, MA, $(! USA), TSHß antibody (from the National Hormone Pituitary Program) was diluted 1:500, ! "#! $! and 100 µl was placed on each slide overnight at 4°C. Slides were washed three times "! for 5 min using 0.5% Triton-X100 in 1X PBS. Washing with 0.5% Triton-X100 in PBS %! was followed by a 1-h incubation with biotinylated anti-rabbit secondary antibody and &! 0.5% Triton-X100/PBS washes. Streptavidin-HRP (Perkin Elmer, MA, USA) was added '! for 1 h at 1:200 dilution, followed by 3 washes in 1X PBS. TSA-TRITC (Perkin Elmer, (! MA, USA) was diluted 1:50, and 100 µl was placed on each slide for 10 minutes. Two )! blockings were then performed: Anti-rabbit IgG diluted 1:100 was placed on each slide *! for 3 hours. After 3 washing steps in 0.5% Triton-X100 in PBS, another blocking step +! was performed with 5% normal goat serum, 10% avidin, 10% biotin, in TSA block for 1 $#! hour. Rabbit-anti-PITX2 antibody, a gift from J. Drouin (Montreal, Canada), was diluted $$! 1:300 in the same block described earlier, and then 100 µl was placed on each slide $"! overnight at 4°C. Secondary detection was performed as described earlier using $%! biotinylated anti-rabbit antibody (Vector Laboratories, CA, USA) and then washed in $&! PBS/Triton. Streptavidin-HRP was added for 1 h at 1:200 dilution, followed by 3 washes $'! in 1X PBS. TSA-FITC (Perkin Elmer, MA, USA) was diluted 1:50, and 100 µl was placed $(! on each slide for 10 minutes. DAPI was diluted 1:600 in PBS and placed on each slide $)! for 5 minutes. After 3 washes in PBS, slides were mounted with fluorescent mounting $*! media, and images were captured using a Leica DMRB fluorescent microscope. $+! TSHß staining was performed identically in newborns and 4-8 week old mice. "#! After TSA-TRITC, nuclear staining with DAPI was performed, and slides were mounted "$! and captured using a fluorescent microscope. Staining was performed systematically ""! every 5 slides in at least 3 controls and 3 transgenic pituitaries. "%! "&! Thyroid volumetrics "'! Adult thyroids were collected at 4 and 8 wks of age or later and fixed for 1 h in "(! 4% formaldehyde in PBS. Hematoxylin and eosin staining was carried out for 20 ! "#! #! seconds each on 5 µm sections. Thyroid area was measured every 10 sections with "! Image J software, and the whole thyroid volume (mm 3) was determined by multiplying by $! the section thickness (0.005 mm). PITX2 staining was performed as previously %! described in the thyroids of 8 wk old control and Pitx2 flox/-;Tg(Tshb-cre) mice. &! '! Hypothyroidism challenge (! Low iodine diet enriched in propyl-thio-uracil (PTU) (0.15%) (Harlan lab, Madison, )! WI) was given to 4-8 wk old Pitx2 flox/-;Tg(Tshb-cre) and control mice for 4 weeks (40). *! #+! Hormone and mRNA evaluations ##! Blood was collected from 4-8 wk old mice fed regular chow or low iodine diet by #"! cardiac puncture after the mice were euthanized and their heart was still beating. After #$! collection, the blood clotted at 4°C for 24 hours and was centrifuged at 8000g for 10 min. #%! The serum was analyzed for total T4 (5 µl) concentration (MP Biomedicals, Ohio, USA), #&! and TSH levels (Millipore, Massachusetts, USA) (25 ul of serum diluted in Serum Assay #'! Buffer included in the Millipore TSH Multiplex kit. Serum from euthyroid animals was #(! diluted 1:5. Serum from mice placed on low iodine diet and PTU was diluted 1:50). Each #)! T4 measure was performed in triplicate, whereas TSH level was measured in single #*! samples. "+! Tshb, Pitx1, and Pitx2 mRNA levels were evaluated by real time PCR (Applied "#! Biosystems, CA, USA) in cDNA prepared from pituitaries of 4-8 week old transgenic and ""! control mice with or without low iodine diet enriched in PTU. GAPDH levels were used "$! to normalize the results, and Cga mRNA levels were used as a positive control. Igf1 "%! mRNA levels were evaluated by real time PCR in liver cDNA from 8 week old transgenic "&! and control mice. ! "#! #! "! Statistical analysis $! Data are given as mean+/-SD. Student’s T Test or ANOVA (for continuous data) %! were used for statistical comparisons. Data were analyzed with SPSS version 17.0. &! p<0.001 was considered significant. '! (! )! ACKNOWLEDGMENTS *! We would like to thank Amanda Vesper-Mortensen, Dr. Jun Z. Li and his lab, and #+! Dr. James Harper and the Miller Lab for their help with experimental design. ##! Recombinogenic bacterial strains and DNA plasmids were provided by Neal Copeland, #"! National Cancer Institute. NLS-cre-β-actin cassette provided as pML78 by Mark #$! Lewandoski. #%! Financial support came from Novo-Nordisk, Societe Francaise d’endocrinologie, #&! Novartis, Ipsen, ADEREM, and the Center for Genetics in Health and Medicine, #'! University of Michigan (all for FC), and the National Institutes of Health (R37HD30428, #(! R01HD34283 to SAC). #)! ! ""! $! "! FIGURE LEGENDS #! %! Figure 1. Tshb-cre transgene is thyrotroph-specific. &! Panel A: Transgene expression was analyzed at e15.5. Lac Z staining was detected in '! the caudomedial aspect of the anterior lobe (A) and in the rostral tip (R) in Tg(Tshb-cre) (! mice but not in controls. )! Panel B: X-gal staining in the adult pituitary was followed by hormone-specific antibody *! staining. Robust X-gal staining was detected in the anterior pituitary gland, in the $+! majority of thyrotrophs, and in rare somatotrophs, but not in the other types of pituitary $$! cells. $"! Panel C: Adult non-pituitary tissues were stained with X-gal to detect ectopic transgene $#! activity. There was no evidence of transgene activity in the brain or testis. There is a $%! low level in the heart, kidney, liver and lungs of Tg(Tshb-cre) mice. A low level of $&! background X-gal staining was present in the thyroids of transgenics and controls. $'! $(! Figure 2. Efficient deletion of Pitx2 in Pitx2 flox/-;Tg(Tshb-cre) mice. $)! Co-immunostaining PITX2 and TSH in 8 week old Pitx2 flox/-;Tg(Tshb-cre) mice compared $*! to controls. The majority of thyrotroph cells express PITX2 in controls (green arrows), "+! although there are rare thyrotrophs where PITX2 is not detectable (red arrow) (panel A). "$! In contrast, the majority of thyrotroph cells in Pitx2 flox/-;Tg(Tshb-cre) mice does not ""! express PITX2 (red arrow). PITX2, green staining; TSH, red staining; DAPI (nuclear): "#! blue staining (panel B). "%! "&! Figure 3. Pitx2 flox/-;Tg(Tshb-cre) mice present a moderate growth deficiency . "'! Growth curves of Pitx2 flox/-;Tg(TSHb-cre) (transgenics) and controls males and females "(! between 3 to 8 weeks of age. Statistical significance: ***, p<0.001; **, p<0.01; *, p<0.05 ! "#! $! "! Figure 4. Subtle phenotypic changes in Pitx2 flox/-;Tg(Tshb-cre) mice, which %! present dramatically increased Pitx1 transcripts. #! A: There is no obvious difference in TSH staining at in 8 week old in Pitx2 flox/-;Tg(Tshb- &! cre) (transgenics) mice and controls. TSH staining was performed in 3 transgenics and '! 3 controls, in 5 pituitary sections at regular intervals for each mouse. B: Thyroid volume (! is smaller in Pitx2 flox/-;Tg(Tshb-cre) mice (p=0.067 after correction for weight). Left, )! Hematoxylin eosin thyroid staining in a control and a Pitx2 flox/-;Tg(Tshb-cre) mice (blue *! line depicts the thyroid). Right, Average thyroid volume in 4 controls and 4 Pitx2 flox/- $+! ;Tg(Tshb-cre) mice. C: T4 levels are not significantly different between Pitx2 flox/-;Tg(Tshb- $$! cre) and control mice. D: Tshb transcript levels are identical, but Pitx1 transcripts are $"! dramatically increased in Pitx2 flox/-;Tg(Tshb-cre) mice compared to controls. Note the $%! logarithmic scale, presenting the fold change in transcripts in Pitx2 flox/-;Tg(Tshb-cre) mice $#! compared to controls. *** indicates p<0.001. E: There is no obvious difference in TSH $&! staining in Pitx2 flox/-;Tg(Tshb-cre) newborns (A) and controls (B). $'! $(! Figure 5. Blunted TSH response to hypothyroidism in Pitx2 flox/-;Tg(Tshb-cre) mice. $)! Pitx2 flox/-;Tg(Tshb-cre) (transgenics) mice and controls were treated for 4 weeks with a $*! low iodine diet enriched in propyl-thio-uracil (0.15%). "+! Panel A: Total T4 levels were significantly decreased at the end of the treatment. Dark "$! grey, transgenics; clear grey, controls. *** indicates p<0.001. ""! Panel B: Tshb and Pitx1 transcripts were less elevated following the thyroid challenge in "%! transgenics than in controls. Pitx1 transcripts were unchanged in challenged "#! transgenics compared to transgenics without treatment. There was a significant "&! increase in the challenged controls compared to controls without treatment. Note the "'! logarithmic scale comparing the fold change in transcripts levels between transgenics ! "#! $! with treatment vs. without treatment, and controls with treatment vs. without treatment. "! *** indicates p<0.001; ** indicates p<0.01. %! Panel C: TSH levels in serum increased less in transgenics than in controls. Note the &! logarithmic scale. * indicates p<0.05 #! '! Figure 6. Model for thyrotroph function and response to hypothyroidism in the (! presence or absence of PITX2. )! In the presence of PITX2 mice are euthyroid and have normal amounts of TSH. *! Hypothyroidism challenge induces an increase in TSH, made possible by the elevated $+! expression of PITX1. In the absence of PITX2, PITX1 plays a compensatory role, $$! leading to a close to euthyroid state in mice under basal conditions. Hypothyroidism $"! challenge induces a blunted TSH response in PITX2 deficient mice, as additional $%! increase of PITX1 is not achievable. $&! $#! $'! ! "#! $! REFERENCES "! %! 1. Burrows HL, Douglas KR, Seasholtz AF, Camper SA 1999 Genealogy of the &! Anterior Pituitary Gland: Tracing a Family Tree. Trends Endocrinol Metab 10:343- '! 352 #! 2. Kelberman D, Rizzoti K, Lovell-Badge R, Robinson IC, Dattani MT 2009 Genetic (! regulation of pituitary gland development in human and mouse. Endocr Rev )! 30:790-829 *! 3. Davis SW, Castinetti F, Carvalho LR, Ellsworth BS, Potok MA, Lyons RH, $+! Brinkmeier ML, Raetzman LT, Carninci P, Mortensen AH, Hayashizaki Y, Arnhold $$! IJ, Mendonca BB, Brue T, Camper SA 2010 Molecular mechanisms of pituitary $"! organogenesis: In search of novel regulatory genes. Mol Cell Endocrinol 323:4- $%! 19 $&! 4. Castinetti F, Saveanu A, Reynaud R, Quentien MH, Buffin A, Brauner R, Kaffel $'! N, Albarel F, Guedj AM, El Kholy M, Amin M, Enjalbert A, Barlier A, Brue T 2008 $#! A novel dysfunctional LHX4 mutation with high phenotypical variability in patients $(! with hypopituitarism. J Clin Endocrinol Metab 93:2790-2799 $)! 5. Ellsworth BS, Butts DL, Camper SA 2008 Mechanisms underlying pituitary $*! hypoplasia and failed cell specification in Lhx3-deficient mice. Dev Biol 313:118- "+! 129 "$! 6. Raetzman LT, Ward R, Camper SA 2002 Lhx4 and Prop1 are required for cell ""! survival and expansion of the pituitary primordia. Development 129:4229-4239 "%! 7. Sheng HZ, Zhadanov AB, Mosinger B, Jr., Fujii T, Bertuzzi S, Grinberg A, Lee "&! EJ, Huang SP, Mahon KA, Westphal H 1996 Specification of pituitary cell "'! lineages by the LIM homeobox gene Lhx3. Science 272:1004-1007 "#! 8. Li S, Crenshaw EB, 3rd, Rawson EJ, Simmons DM, Swanson LW, Rosenfeld MG "(! 1990 Dwarf locus mutants lacking three pituitary cell types result from mutations ")! in the POU-domain gene pit-1. Nature 347:528-533 "*! 9. Sornson MW, Wu W, Dasen JS, Flynn SE, Norman DJ, O'Connell SM, Gukovsky %+! I, Carriere C, Ryan AK, Miller AP, Zuo L, Gleiberman AS, Andersen B, Beamer %$! WG, Rosenfeld MG 1996 Pituitary lineage determination by the Prophet of Pit-1 %"! homeodomain factor defective in Ames dwarfism. Nature 384:327-333 %%! 10. Camper SA, Saunders TL, Katz RW, Reeves RH 1990 The Pit-1 transcription %&! factor gene is a candidate for the murine Snell dwarf mutation. Genomics 8:586- %'! 590 %#! 11. Charles MA, Saunders TL, Wood WM, Owens K, Parlow AF, Camper SA, %(! Ridgway EC, Gordon DF 2006 Pituitary-specific Gata2 knockout: effects on %)! gonadotrope and thyrotrope function. Mol Endocrinol 20:1366-1377 %*! 12. Gage PJ, Camper SA 1997 Pituitary homeobox 2, a novel member of the bicoid- &+! related family of homeobox genes, is a potential regulator of anterior structure &$! formation. Hum Mol Genet 6:457-464 &"! 13. Lin CR, Kioussi C, O'Connell S, Briata P, Szeto D, Liu F, Izpisua-Belmonte JC, &%! Rosenfeld MG 1999 Pitx2 regulates lung asymmetry, cardiac positioning and &&! pituitary and tooth morphogenesis. Nature 401:279-282 &'! 14. Semina EV, Reiter R, Leysens NJ, Alward WL, Small KW, Datson NA, Siegel- &#! Bartelt J, Bierke-Nelson D, Bitoun P, Zabel BU, Carey JC, Murray JC 1996 &(! Cloning and characterization of a novel bicoid-related homeobox transcription &)! factor gene, RIEG, involved in Rieger syndrome. Nat Genet 14:392-399 &*! 15. Gage PJ, Suh H, Camper SA 1999 Dosage requirement of Pitx2 for development '+! of multiple organs. Development 126:4643-4651 ! "#! $! 16. Suh H, Gage PJ, Drouin J, Camper SA 2002 Pitx2 is required at multiple stages "! of pituitary organogenesis: pituitary primordium formation and cell specification. %! Development 129:329-337 &! 17. Charles MA, Mortensen AH, Potok MA, Camper SA 2008 Pitx2 deletion in '! pituitary gonadotropes is compatible with gonadal development, puberty, and (! fertility. Genesis 46:507-514 #! 18. Quentien MH, Pitoia F, Gunz G, Guillet MP, Enjalbert A, Pellegrini I 2002 )! Regulation of prolactin, GH, and Pit-1 gene expression in anterior pituitary by *! Pitx2: An approach using Pitx2 mutants. Endocrinology 143:2839-2851 $+! 19. Tremblay JJ, Goodyer CG, Drouin J 2000 Transcriptional properties of Ptx1 and $$! Ptx2 isoforms. Neuroendocrinology 71:277-286 $"! 20. McIntyre DC, Rakshit S, Yallowitz AR, Loken L, Jeannotte L, Capecchi MR, $%! Wellik DM 2007 Hox patterning of the vertebrate rib cage. Development $&! 134:2981-2989 $'! 21. Mullen RD, Colvin SC, Hunter CS, Savage JJ, Walvoord EC, Bhangoo AP, Ten $(! S, Weigel J, Pfaffle RW, Rhodes SJ 2007 Roles of the LHX3 and LHX4 LIM- $#! homeodomain factors in pituitary development. Mol Cell Endocrinol 265-266:190- $)! 195 $*! 22. Warming S, Costantino N, Court DL, Jenkins NA, Copeland NG 2005 Simple and "+! highly efficient BAC recombineering using galK selection. Nucleic Acids Res "$! 33:e36 ""! 23. Camper SA, Saunders TL, Kendall SK, Keri RA, Seasholtz AF, Gordon DF, "%! Birkmeier TS, Keegan CE, Karolyi IJ, Roller ML, et al. 1995 Implementing "&! transgenic and embryonic stem cell technology to study gene expression, cell- "'! cell interactions and gene function. Biol Reprod 52:246-257 "(! 24. Westlund KN, Chmielowiec S, Childs GV 1983 Somatostatin fibers and their "#! relationship to specific cell types (GH and TSH) in the rat anterior pituitary. ")! Peptides 4:557-562 "*! 25. Huang Y, Guigon CJ, Fan J, Cheng SY, Zhu GZ Pituitary homeobox 2 (PITX2) %+! promotes thyroid carcinogenesis by activation of cyclin D2. Cell Cycle 9 %$! 26. Kendall SK, Samuelson LC, Saunders TL, Wood RI, Camper SA 1995 Targeted %"! disruption of the pituitary glycoprotein hormone alpha-subunit produces %%! hypogonadal and hypothyroid mice. Genes Dev 9:2007-2019 %&! 27. Iida K, Del Rincon JP, Kim DS, Itoh E, Nass R, Coschigano KT, Kopchick JJ, %'! Thorner MO 2004 Tissue-specific regulation of growth hormone (GH) receptor %(! and insulin-like growth factor-I gene expression in the pituitary and liver of GH- %#! deficient (lit/lit) mice and transgenic mice that overexpress bovine GH (bGH) or a %)! bGH antagonist. Endocrinology 145:1564-1570 %*! 28. Yamada M, Satoh T, Mori M 2003 Mice lacking the thyrotropin-releasing &+! hormone gene: what do they tell us? Thyroid 13:1111-1121 &$! 29. Jones PM, Burrin JM, Ghatei MA, O'Halloran DJ, Legon S, Bloom SR 1990 The &"! influence of thyroid hormone status on the hypothalamo-hypophyseal growth &%! hormone axis. Endocrinology 126:1374-1379 &&! 30. Charles MA, Suh H, Hjalt TA, Drouin J, Camper SA, Gage PJ 2005 PITX genes &'! are required for cell survival and Lhx3 activation. Mol Endocrinol 19:1893-1903 &(! 31. Phillips JC 2002 Four novel mutations in the PITX2 gene in patients with &#! Axenfeld-Rieger syndrome. Ophthalmic Res 34:324-326 &)! 32. Saadi I, Toro R, Kuburas A, Semina E, Murray JC, Russo AF 2006 An unusual &*! class of PITX2 mutations in Axenfeld-Rieger syndrome. Birth Defects Res A Clin '+! Mol Teratol 76:175-181 ! "#! $! 33. Tumer Z, Bach-Holm D 2009 Axenfeld-Rieger syndrome and spectrum of PITX2 "! and FOXC1 mutations. Eur J Hum Genet 17:1527-1539 %! 34. Weisschuh N, Dressler P, Schuettauf F, Wolf C, Wissinger B, Gramer E 2006 &! Novel mutations of FOXC1 and PITX2 in patients with Axenfeld-Rieger '! malformations. Invest Ophthalmol Vis Sci 47:3846-3852 (! 35. Meyer-Marcotty P, Weisschuh N, Dressler P, Hartmann J, Stellzig-Eisenhauer A )! 2008 Morphology of the sella turcica in Axenfeld-Rieger syndrome with PITX2 #! mutation. J Oral Pathol Med 37:504-510 *! 36. Pitteloud N, Quinton R, Pearce S, Raivio T, Acierno J, Dwyer A, Plummer L, $+! Hughes V, Seminara S, Cheng YZ, Li WP, Maccoll G, Eliseenkova AV, Olsen $$! SK, Ibrahimi OA, Hayes FJ, Boepple P, Hall JE, Bouloux P, Mohammadi M, $"! Crowley W 2007 Digenic mutations account for variable phenotypes in idiopathic $%! hypogonadotropic hypogonadism. J Clin Invest 117:457-463 $&! 37. Strachan T, Read AP 1994 PAX genes. Curr Opin Genet Dev 4:427-438 $'! 38. Brinkmeier ML, Gordon DF, Dowding JM, Saunders TL, Kendall SK, Sarapura $(! VD, Wood WM, Ridgway EC, Camper SA 1998 Cell-specific expression of the $)! mouse glycoprotein hormone alpha-subunit gene requires multiple interacting $#! DNA elements in transgenic mice and cultured cells. Mol Endocrinol 12:622-633 $*! 39. Kendall SK, Saunders TL, Jin L, Lloyd RV, Glode LM, Nett TM, Keri RA, Nilson "+! JH, Camper SA 1991 Targeted ablation of pituitary gonadotropes in transgenic "$! mice. Mol Endocrinol 5:2025-2036 ""! 40. Weiss RE, Murata Y, Cua K, Hayashi Y, Seo H, Refetoff S 1998 Thyroid "%! hormone action on liver, heart, and energy expenditure in thyroid hormone "&! receptor beta-deficient mice. Endocrinology 139:4945-4952 "'! "(! ! "#! :120<8"=0,-4/>" &!" %#" %!" ?" ?" ?" +,-./01.23"4-51/" ??" $#" ???" 67.8,75"4-51/" ??" +,-./01.23"914-51/" $!" 67.8,75"914-51/" #" !" &" '" #" (" )" *" :11;/" @20A,1"&" 1" 2"B;!"!" 2>C,)4@" 9)-D70"E77AF"F" $;!" A;!" !.(/&.0" !.(/&.0"!.(/&&.0" #;!" ';!" !;!" ()*+,)-1""()*+,)-1"" 2,/*130*451" %&'()*+(,-" %&'()*+(,-" 22)+/-"2$")+/-"2$" 8)-@"5>/*30" !" -090-1"-090-1"*3:7-"*3:7-" $" '%"'%" 2,/*130*451:5)*+,)-1" '$"'$" '!!" '#"'#" ###" '!"'!" &"&" '!" %%"" $"$" '" ##"" !"!" ()*+,)-" 2,/*130*45"2,/*130*45" ()*+,)-" 2,/*130*45"2,/*130*45" !;'" ./-01" ./-01" 607/-01" 807/-01"" <4+='" 21>?" (3/" 3" !.(/&.0" 4,*5&+"6" E" J" A" B%"" $"" %)""C"5$4#"6" H"5&4#"6" 5#(E" 777" 777" AE" "1"(6"" 77" A@@@@@@" 7"" 777" AC" A@@" A@@@@@" AB" A@@@@" A@" F" A@@@" A@" E" ( " A@@" 5!)3A" C" A" A@" B" @" A" A" (%"')#$)" 1%2"!%!$" (%"')#$)" 1%2"!%!$"!)" !)"$"5" $"5" B!0'"D" !"#$%&'(')&*!*'+,&-&'+! 12%"'+%2!-'!#$"'-#$1',&%3! ;$',9, "&-./!&%!2."12%%2,! !(')%#&!, %&$#"&, "&-.0! !!! "&-./! -%#! -.3,-.3 -%#! -#$1'-1'"#! -.2,0, *2((%! "&-:&-81$! ('5!-6!(272(%! "&-./! "&-.0! !!! "&-.0!4! -%#! -%#! "&-./!&%!89%2+-! -.3,-.3, =#,!#%,"%&, 4 ("',9, ",-.2, %&$#"&, '%"&%$'&, ?(%,4, DISCUSSION Nous avons généré une nouvelle recombinase sous contrôle du promoteur de la TSHb. Sa spécificité (hypophysaire, et plus spécialement thyréotrope) en fait un outil utile pour invalider un facteur de transcription dans les cellules thyréotropes. Il est important de tenir compte du fait que la recombinase n’est exprimée qu’après expression de la TSHb (environ au 15ème jour embryonnaire), donc ne donne pas d’informations sur le développement de l’axe thyréotrope, mais sur le fonctionnement des cellules thyréotropes matures. Cette recombinase nous a permis d’invalider spécifiquement Pitx2 dans les cellules thyréotropes. Les souris Pitx2 flox/- ;Tg(TSHb-cre) présentent un retard de croissance modéré, mais pas de signe biologique évident d’hypothyroïdie en conditions basales (taux de T4 et TSH non significativement différents). Elles présentent une forte augmentation des transcrits de PITX1, un facteur de transcription de la même famille (homéodomaine et domaine C terminal identiques), ce qui suggère un mécanisme compensateur permettant une euthyroïdie. Après induction d’une hypothyroïdie (4 semaines de traitement par régime pauvre en iode, enrichi en PTU), les souris transgéniques ont une réponse plus faible de la TSH par rapport aux souris contrôles. Dans les souris contrôles, il existe une forte augmentation des transcrits de Pitx1, alors qu’ils restent inchangés chez les souris transgéniques. Les souris avec inactivation de Pitx2 présentent donc un déficit thyréotrope partiel. Le mécanisme compensateur assuré par PITX1 en conditions basales est insuffisant pour avoir une réponse de la TSH adaptée en situation de stress (hypothyroïdie). Le retard de croissance est surprenant, compte tenu de l’euthyroïdie des souris transgéniques en conditions basales. Trois explications peuvent être évoquées : , L’existence d’une hypothyroïdie précoce, non recherchée. Par exemple, il pourrait exister une hypothyroïdie entre la naissance et 4 semaines d’âge. Nous avons vérifié par immunohistochimie la présence de cellules thyréotropes à la naissance. Il faudrait pouvoir mesurer les taux de T4 et TSH à ce stade. Le mécanisme compensateur permettrait l’obtention finale d’une euthyroïdie entre 4 et 8 semaines d’âge. , L’existence d’un déficit somatotrope associé. La recombinase est exprimé dans quelques cellules somatotropes (probablement des doubles positives pour GH et TSH). Les souris transgéniques pourraient présenter une diminution de ces cellules, entrainant un déficit somatotrope. Nous avons mesuré les ARNm d’IGF1 à 4 et 8 semaines, et n’avons pas observé de différence. Il faudra mesurer les taux d’IGF1 plasmatiques pour éliminer de façon définitive cette hypothèse. , Une sensibilité insuffisante de nos kits de dosage, n’ayant pas permis de mettre en évidence une discrète diminution de la T4 chez nos souris transgéniques. En pathologie humaine, les patients avec mutations de PITX2 présentent rarement une symptomatologie hypophysaire. Seulement quelques patients ont une hypoplasie hypophysaire et un déficit en GH. Même si nos résultats reflètent uniquement l’action de PITX2 au sein des cellules matures (et pas lors du développement), il est logique d’envisager que le rôle compensateur de PITX1 mis en évidence dans ce travail explique, au moins en partie, l’absence de déficit hypophysaire, en particulier thyréotrope, chez les patients présentant un syndrome d’Axenfeld-Rieger. Il est possible cependant que ces patients présentent un déficit thyréotrope partiel, sans que cela n’ait de véritable retentissement clinique. ! $)! %-! EXPRESSION ET ROLES DE ISL1 DANS L’AXE THYREOTROPE HYPOTHYROIDISM INDUCED BY ISL1 INACTIVATION IN THYROTROPHS 1Castinetti F., 1Brinkmeier M.L., 2Vella K., 2Hollenberg A., 3Gan L., 1Camper S.A. Manuscrit en préparation ! $*! INTRODUCTION Isl1 est un facteur de transcription à homéodomaine LIM, qui joue un rôle majeur dans le développement et la prolifération des cellules souches de nombreux organes dont le cœur et le pancréas (70; 168). Peu de données détaillées sont cependant disponibles sur son rôle précis dans le développement hypophysaire. Isl1 est exprimé dès e8,5 dans l’ectoderme oral. Son expression est restreinte à la poche de Rathke à partir de e9,5, puis limitée à la zone ventrale de la poche de Rathke à partir de e10,5 (54). Ces cellules ventrales exprimeront TSHß et la sous-unité alpha. L’invalidation homozygote de Isl1 ne bloque pas les premières étapes de la formation de la poche de Rathke, mais les souris présentent une mort embryonnaire précoce (e10,5). En outre, une expression (inconstante) de Isl1 a été observée dans des contingents potentiels de cellules souches hypophysaires. Isl1 interagit avec plusieurs facteurs de transcription hypophysaire incluant Lhx3 (12; 122), Sf1(75), ou certaines voies de signalisation (113). Ainsi, l’inactivation de lhx3 aboutit à une expression ectopique de Isl1 (54) Isl1 semble jouer un rôle dans le développement et la différenciation thyréotrope : Kerr et al. ont ainsi montré que l’expression de isl1 augmentait in vitro entre la phase de cellule pré-thyréotrope et celle de cellule thyréotrope différenciée (98). Brinkmeier et al. ont caractérisé l’hypophyse des souris présentant une inactivation conditionnelle de la sous-unité alpha : ces souris ne sécrètent pas de Tsh, et ont une hyperplasie des cellules thyréotropes (du fait de l’absence de rétrocontrôle négatif par la T3 et la T4) ; l’expression de isl1 est retrouvée à un niveau très élevé au sein de ces cellules thyréotropes hyperplasiques en comparaison avec des cellules thyréotropes « normales » (18; 21). De plus, dans la rétine, il a été montré que Isl1 était capable d’interagir avec Pou4f2, un facteur de transcription à homéodomaine POU, de la même famille que Pou1f1, impliqué dans le développement hypophysaire, et en particulier la différenciation de la lignée thyréotrope (130; 143). Notre objectif était donc de déterminer le(s) rôle(s) de Isl1 dans l’hypophyse au cours de l’embryogenèse, et plus particulièrement son rôle dans le développement de l’axe thyréotrope. ! %+! HYPOTHYROIDISM INDUCED BY ISL1 INACTIVATION IN THYROTROPHS 1Castinetti F., 1Brinkmeier M.L., 2Vella K., 2Hollenberg A., 3Gan L., 1Camper S.A. 1Human Genetics, University of Michigan, Ann Arbor, MI, United States, 48109-5618 2Beth Israel Deaconess Medical Center, Harvard University, NJ, United States 3University of Rochester School of Medicine and Dentistry, Rochester, NY 14642 United States Corresponding author Sally A. Camper, Ph.D ., Dept. Human Genetics University of Michigan 4909 Buhl Bldg. 1241 Catherine St. Ann Arbor, MI 48109-5618 Telephone: 734-763-0682 Fax: 734-763-5831 email: [email protected] Keywords : pituitary, Islet-1, isl-1, TSH, T4, thyrotrophs, gonadotrophs, growth insufficiency, gonadotrophs, transcription factor ! "! ABSTRACT (248 words) ISL1 is a LIM homeodomain transcription factor necessary for development of the heart, pancreas, and retina. Isl1 deficient mice (Isl1 -/-) die early during embryogenesis (e10.5) due to heart defects, and at that time, they have an undersized pituitary primordium. ISL1 exhibits dynamic expression in the pituitary from early embryogenesis to adulthood. Here we report the cell specific expression of ISL1 and assessment of its role in differentiated gonadotrophs and thyrotrophs. We used a Tshb BAC transgene recombineered with cre to generate mice with Isl1 deleted in thyrotrophs: Isl1 flox/flox ;Tg(Tshb-cre). These mice have growth insufficiency, and TSH levels are in the normal range. Circulating T4 levels are decreased, and their thyroid glands are significantly smaller. The presence of low T4 levels without increased pituitary TSH production is consistent with decreased thyrotroph function. The expression of several transcription factors with roles in early pituitary development, including Pitx2, Pitx1, Lhx3 and Lhx4, are significantly decreased in Isl1 flox/flox ;Tg(Tshb-cre) pituitary glands, suggesting that Isl1 is a master regulator . To challenge the thyrotroph function of Isl1 flox/flox ;Tg(Tshb-cre) mice, we treated them for 4 weeks with a low iodine diet enriched the anti-thyroid drug, propyl-thio-uracyle. This induces low T4 levels and elevated TSH response in normal mice, but the Isl1 flox/flo x;Tg(Tshb-cre) mice have a severely blunted TSH response. In contrast, deletion of Isl1 in gonadotrophs with an Lhb-cre transgene has no obvious effect on gonadotroph function or fertility. These results show that ISL1 is necessary for thyrotroph function, in addition to its role in development of Rathke's pouch. ! "! INTRODUCTION The pituitary gland is a master organ that controls the endocrine organs of the body. The anterior lobe secretes six hormones that regulate various processes like growth, lactation, stress response and reproduction, whereas the posterior lobe regulates water intake (1). Pituitary organogenesis is a complex process requiring a temporal-spatial regulation of several signaling pathways and transcription factors. Among them, the LIM homeodomain transcription factors LHX3 and LHX4 are crucial in the early steps of pituitary ontogenesis and have dosage sensitive, overlapping functions (REF SHENG)(2, 3). LHX3 and LHX4 expression are first observed throughout Rathke's pouch in the developing mouse pituitary at e9.5. LHX4 expression becomes restricted to the future anterior lobe of the pituitary gland at e12.5, whereas higher LHX3 protein levels are found in the dorsal aspect of Rathke's pouch. Homozygous inactivation of Lhx4 or Lhx3 induces death at birth: the pituitary is hypoplasic and undifferentiated (Lhx3 -/-) or presents a decreased number of all five hormone-producing cells ( Lhx4 -/-). Two other LIM domain transcription factors are expressed in the pituitary, LHX2, which seems to be important for pituitary posterior lobe development (4), and islet-1 (ISL1). ISL1 was identified 20 years ago from a cDNA library derived from insulin producing cells (5). Since then, roles for ISL1 are reported in the development, proliferation, function and maintenance of specialized cells in the heart, pancreas and retina. For instance, cardiovascular progenitors expressing ISL1 give rise to the cardiomyocyte, pacemaker, smooth muscle and endothelial cell lineages (6, 7). In the pancreas, precursors with Isl1 inactivation are unable to differentiate into functional islet cells (8, 9). In the eye, ISL1 is necessary for retinal ganglion cell differentiation (10, 11). ISL1 is also expressed in various regions of the hypothalamus, where it seems to be necessary for estrogen receptor alpha expression (12). In all these organs, ISL1 is expressed in the early steps of organogenesis, and it is crucial for the differentiation process leading to a mature and functional organ. In contrast, the potential role of ISL1 during pituitary development remains relatively unexplored. ISL1 is a potential marker of pituitary stem cells, coinciding with its expression timing (13)(Castinetti, EndoReviews). In mouse, ISL1 is expressed very early (e8.5) in the oral ectoderm, which will give rise to Rathke’s pouch. At e9.5, ISL1 is expressed in the anterior portion of Rathke’s pouch that will give birth to rostral thyrotrophs and alpha- subunit expressing cells (14). Homozygous inactivation of Isl1 leads to early embryonic death at e10.5 due to heart defects; at this time point, mice present a small rudimentary pouch (15). Studies in sheep and chick suggest that ISL1 is expressed in thyrotrophs and gonadotrophs (16, 17), which implies a potential role in these cell types. The role of ISL1 in later stages of pituitary growth and cell functioning has not been assessed. Several pieces of evidence suggest that ISL1 might be important in the pituitary, at least for thyrotroph differentiation. Differential gene expression studies show that Isl1 expression is increased in a model of transition between hyperplasic and normal thyrotrophs (microarray analysis of Isl1 expression inTtT97 thyrotropic tumor cells induced by peripheral hypothyroidism vs. thyrotroph cells in euthyroidism) (18). We investigated ISL1 expression in hypothyroid alpha-knockout mice that develop thyrotroph hypertrophy and hyperplasia (19)(+REF) and we found a dramatic increase in Isl1 expression (data not shown). These 2 points favor a role of ISL1 in the function and/or expansion of thyrotrophs. ! "! Our objective was to determine the cell specific expression of ISL1 in rodents and assess whether ISL1 is important in pituitary cell maintenance and function. We observed ISL1 expression only in thyrotrophs, gonadotrophs and alpha subunit expressing cells after birth in mice. We inactivated Isl1 in thyrotrophs and gonadotrophs with the specific cre-recombinases Tg (Tshb-cre) and Tg(Lhb-cre ), respectively, and Isl1 loxP/loxP mice (11, 20, 21)(Castinetti et al., submitted). Our phenotypic studies show that ISL1 is necessary for thyrotroph maintenance and function, as Isl1 loxP/loxP ;Tg(Tshb- cre) mice present growth insufficiency, mild hypothyroidism, and poor pituitary TSH response to hypothyroidism. Mice with ISL1 deficient thyrotrophs have normal fertility. In sum, these results reveal an important role for ISL1 in thyrotroph function after birth. ! "! MATERIAL AND METHODS Mice All mice were maintained at the University of Michigan under the guidelines of the Unit for Laboratory Animal Medicine and the University Committee for Care and Use of Animals. Tg(Tshb-cre) mice were generated using a BAC from a C57BL6 library (RP24-230F23) containing a sequence including the promoter of the mouse Tshb gene from -144kb to +58 kb and cre coding sequences introduced by recombineering (Castinetti et al., submitted). Tg(Tshb-cre) mice were identified by PCR amplification of genomic DNA with primers 5’-GCATAACCAGTGAAACAGCATTGCTG-3’ and 5’- GGACATGTTCAGGGATCGCCAGGCG-3’ under the following conditions: 94°C for 3 min, followed by 32 cycles of 94°C for 30 s, 60°C for 60 s, and 72°C for 90 s, and a final 10-min extension at 72°C. The generation of Isl1 loxP/+ mice in a 129S6 and C57BL/6J mixed background have been reported previously (11). Isl1 loxP/loxP ;Tg(Tshb-cre) mice were generated by mating Isl1 loxP/loxP mice with Tg(Tshb-cre) positive mice. The Isl1 loxP/+ ;Tg(Tshb-cre) offspring were mated to Isl1 loxP/loxP mice, and genotyping was performed as previously described (11). Briefly, the Isl1 conditional allele was identified by the following primers: 5'-GGTGCTTAGCGGTGATTTCCT-3' and 5'- GCACTTTGGGATGGTAATTGGAG-3'. Tissue Preparation and Histology Pituitaries Adult pituitaries were collected at 8 weeks of age and fixed for 1 h in 4% paraformaldehyde in PBS. Pituitaries were rinsed in PBS, dehydrated and embedded in a Citadel 1000 (Thermo Electric, Chesire, England) paraffin-embedding machine, and sectioned coronally at 5 µm thickness. Immunohistochemistry for ISL1 and TSH β was performed on pituitary sections from newborn mice. Epitopes were unmasked by boiling with citric acid (10 mM) for 10 minutes. After 3 washes in PBS, endogenous peroxidases were quenched in 3% H 202 for 20 minutes. After a mouse on mouse IgG block for 1 hour (Vector Lab, ME, USA), TSHß antibody was diluted 1:500, and ISL1 antibody was diluted 1:600 in TSA block (Perkin Elmer, MA, USA), and 100 µl was placed on each slide overnight at 4°C. After 3 washes in PBS, a 1 h incubation with biotinylated anti-mouse secondary antibody was performed (Vector Lab, ME, USA). After 3 washes in PBS, streptavidin-HRP (Perkin Elmer, MA, USA) was added for 1 h at 1:200 dilution. After 3 washes in PBS, TSA- TRITC (Perkin Elmer, MA, USA) was diluted 1:50, and 100 µl was placed on each slide for 10 minutes. After quenching endogenous peroxidases for 20 minutes, a blocking step with 5% normal goat serum, 10% avidin, 10% biotin, in TSA block was performed for 1 hour. Secondary detection was performed as described earlier using biotinylated anti-rabbit antibody (Jackson Lab, ME, USA) and then washed in PBS/Triton. Streptavidin-HRP was added for 1 h at 1:200 dilution, followed by 3 washes in 1X PBS. TSA-FITC (Perkin Elmer, MA, USA) was diluted 1:50, and 100 µl was placed on each slide for 10 minutes. After 2 washes in 1X PBS, DAPI was finally diluted 1:600 in PBS and placed on each slide for 5 minutes. After 3 washes in PBS, slides were mounted with fluorescent mounting media, and images were captured using a Leica DMRB fluorescent microscope. The same procedure was performed for co-immunostaining with hormones: ACTH (primary antibody diluted 1:1000, secondary anti-rabbit biotin antibody 1:100 for 1 hour), LHß (Primary antibody diluted 1:500, secondary anti-guinea pig biotin antibody 1:100 for 1 hour), GH (primary antibody diluted 1:100, secondary anti-human biotin antibody 1:200 for 1 hour), and the alpha subunit (primary antibody diluted 1:300, ! "! secondary anti-rabbit biotin antibody 1:100 for 1 hour). All of the primary antibodies were provided by the National Hormone Pituitary Program (NHPP, USA). All the secondary antibodies were provided by The Jackson Lab, ME, except the anti-human secondary antibody (Vector Lab, CA, USA). TSHß staining alone was performed identically on pituitary sections from 8 week old Isl1loxP/loxP ; Tg(Tshb-cre) mice and controls. After TSA-TRITC, nuclear staining with DAPI was performed, and slides were mounted and captured using a fluorescent microscope. Staining was performed systematically every 5 slides in at least 3 controls and 3 transgenic pituitaries. Thyroids Adult thyroids were collected at 8 weeks of age and fixed for 1 h in 4% paraformaldehyde in PBS. Hematoxylin and eosin staining (20 seconds each) was performed on 5 µm sections. Thyroid area was measured every 10 sections with Image J software, and the whole thyroid volume (mm 3) was determined by multiplying by the section thickness (0.005 mm). Hormone and mRNA evaluations Blood was collected by cardiac puncture after the mice were euthanized, and their heart was still beating. After collection, the blood clotted at 4°C for 24 hours and was centrifuged at 8000g for 10 min. After centrifugation, the serum (5 µl for T4, 20 µl for TSH) was analyzed for the total T4 concentration (MP Biomedicals, Ohio, USA), and TSH levels (Millipore, Massachusetts, USA). Each T4 measure was performed in triplicate, whereas the TSH level was measured singly. Tshb, Lhx3 , Lhx4 , Prop1 , Pou1f1 , Gata2, Pitx1 and Pitx2 mRNA levels were evaluated by real time PCR (Applied Biosystems, CA, USA) in pituitaries from 8 week old transgenic and control mice (methods previously described in (22)). GAPDH levels were used to normalize the results, and Cga mRNA levels were used as a positive control. Hypothyroidism challenge Low iodine diet enriched in propyl-thio-uracyle (PTU) (0.15%) (Harlan lab, Madison, WI) was given to 3-8 wk old Isl1 loxP/loxP ;Tg(Tshb-Cre) and control mice for 4 weeks (23). Blood was collected by cardiac puncture after the mice were euthanized: Blood total T4 (MP biomedicals, Ohio, USA) and TSH levels (Millipore, Massachusetts, USA) were evaluated, and compared to transgenics and control mice without treatment. Pituitaries were collected, and Tshß transcripts levels were determined by real time PCR (Applied Biosystems, CA, USA). Statistical analysis Data are given as mean+/-SD. Student’s t test or ANOVA (for continuous data) were used for statistical comparisons. Data were analyzed with SPSS version 17.0. ! "! RESULTS ISL1 is expressed during pituitary development and in thyrotrophs after birth We evaluated the expression of ISL1 in the mouse pituitary during development and at birth by ISL1 immunostaining. ISL1 immunoreactivity is detected in cells scattered throughout the whole Rathke’s pouch at e12.5, in the anterior lobe at e14.5, e18.5 and at day 1 after birth. There is a very low level of expression in the intermediate lobe ( Figure 1, panel A ), and no obvious dorsal ventral gradient in the anterior lobe. We determined which pituitary cells express ISL1 after birth. ISL1 is expressed in the majority of the gonadotrophs and in about 50% of the thyrotrophs. Almost all ISL1 positive cells also express alpha subunit, suggesting that ISL1 is exclusively expressed in thyrotrophs and gonadotrophs. Consistent with this, somatotrophs and corticotrophs did not express significant amounts of ISL1 ( Figure 1, panel B; Figure 2, panel A ). To understand the specific roles of ISL1 in the thyrotrophs and gonadotrophs after birth, we inactivated Isl1 by using a thyrotrophs specific cre recombinase transgenic line, Tg (Tshb-cre ) (Castinetti et al., submitted), and a gonadotroph specific cre line: Tg( Lhb- cre ) (REF). Isl1 loxP/loxP ;Tg(Tshb-cre) mice are smaller than wild type littermates Mice with a floxed Isl1 allele ( Isl1 loxP/loxP ) mated with mice carrying cre recombinase under the control of the Tshb promoter, Tg(Tshb-Cre) , to produce Isl1 loxP/+ ;Tg(Tshb-cre) mice. We crossed Isl1 loxP/+ ;Tg(Tshb-Cre) mice with Isl1 loxP/loxP mice to obtain Isl1 loxP/loxP ;Tg(Tshb-cre) mice. The expected Mendelian ratio was observed. Out of 150 mice we had 31.3% Isl1 loxP/ +;Tg(Tshb-cre) mice, 24.7% Isl1 loxP/loxP ;Tg(Tshb- cre) mice, 26.6% Isl1 loxP/+ ; nontransgenic and 17.4% Isl1 loxP/loxP ; nontransgenic mice (p=0.106). TSH β and ISL1 co-immunostaining confirmed that the majority of thyrotroph cells expressed ISL1 in controls, whereas there is no expression of ISL1 in most of Isl1 loxP/loxP ;Tg(Tshb-cre) thyrotroph cells ( Figure 2, panel A ). This demonstrates highly penetrant excision of Isl1 by cre recombinase and survival of the Isl1 deficient thyrotroph cell. We also checked whether cre transgene inappropriately modified ISL1 expression in adult thyroid glands. No obvious difference in Isl1 expression is observed between Isl1 loxP/loxP ;Tg(Tshb-cre) and controls ( Figure 2, panel B ) We found that male and female Isl1 loxP/loxP ;Tg(Tshb-cre) mice have significant growth delay relative to their WT littermates. The weight differences are statistically significant at every week of age from 4 to 7 weeks in females and 4-6 weeks of age in males (p<0.001) (p<0.05 at 7 weeks in males, and at 8 weeks in females) ( Figure 3 ). This growth deficiency is consistent expectations for moderate hypothyroidism. Isl1 loxP/loxP ;Tg(Tshb-Cre) mice present with TSH deficiency Isl1 loxP/loxP ;Tg(Tshb-cre) mice have lower total T4 levels (1.5 to 1.8 fold decrease, p<0.001) ( Figure 4, panel A ) and serum TSH levels in the normal range ( Figure 5 ). Affected mice have decreased pituitary Tshb transcript levels (from 1.5 to 7 fold less, n=6 vs. 4 controls, p<0.05) ( Figure 4, panel B ). We observed no difference in TSH β staining in transgenics vs. controls ( Figure 4, panel C ). The thyroid volume of Isl1 loxP/loxP ;Tg(TSHb-cre) mice is also lower than in controls (p=0.011) ( Figure 4, panel D). This biological phenotype, comprised of low thyroid volume, low T4 levels, and lack of increased TSH levels, is expected for a thyrotroph deficiency. We hypothesized that central hypothyroidism induced by inactivation of Isl1 in thyrotrophs could be associated with decreased expression of transcription factors involved in thyrotrophs development. By real time PCR, Pitx1 , Pitx2, Lhx3 and Lhx4 ! "! transcripts were decreased (p<0.05) in Isl1 loxP/loxP ;Tg(Tshb-cre) mice compared to controls, whereas Gata2 , Prop1 and Pou1f1 transcripts were unchanged ( Figure 4, panel B ). This suggests that Isl1 is upstream of Pitx1, Pitx2, Lhx3 and Lhx4, but not Gata2, Prop1 or Pou1f1. We hypothesized that mice lacking ISL1 in thyrotrophs would have a blunted TSH response to low thyroid hormone levels. We challenged transgenic mice and controls (n=6 vs. 8 controls) by inducing hypothyroidism with a low iodine diet enriched in 0.15% PTU during 4 weeks. As expected, the control mice exhibited dramatically increased serum TSH levels and Tshb mRNA levels (700 fold increase for serum TSH, and 65 fold increase for the transcripts) and decreased T4 levels in response to the diet (p<0.001) ( Figure 5 ). Although the low iodine diet provoked dramatically elevated serum TSH levels in the transgenics (p<0.001), the increase was 7 fold less than in controls (TSH levels 165+/-94 pg/microliter in controls vs. 27+/-13 in transgenics, p<0.01) ( Figure 5 ). The increase in Tshb transcripts was not significantly different between transgenics and controls (data not shown). Taken together, these data show that Isl1 is required for a normal ability of thyrotrophs to respond to hypothyroidism and maintain homeostasis. ISL1 is dispensable for gonadotroph maintenance and function ISL1 is expressed in gonadotrophs after birth. To test its importance in gonadotroph maintenance and function, we crossed Isl1 loxP/loxP mice with the previously described gonadotroph specific Tg(Lhb-cre) mice (20). The Isl1 loxP/+ ;Tg(Lhb-cre) progeny were crossed with Isl1 loxP/loxP mice to obtain Isl1 loxP/loxP ;Tg(Lhb-cre) mice with a conditional inactivation of Isl1 in gonadotrophs. We used co-immunostaining to demonstrate that Isl1 is efficiently deleted in the majority of gonadotrophs. Isl1 loxP/loxP ;Tg(Lhb-cre) male mice do not present any growth insufficiency, suggesting normal gonadal steroid effects on GH pulsatility. The Isl1 loxP/loxP ;Tg(Lhb-cre female mice have normal puberty as evaluated by vaginal openings, and both males and females have normal fertility (data not shown). This suggests that ISL1 is dispensable for mature gonadotroph function. ! "! DISCUSSION ISL1 is required for maintenance and function of thyrotrophs Mice with inactivation of Isl1 in thyrotrophs present with hypothyroidism characterized by growth insufficiency, smaller thyroid volume and lower T4 levels. The thyrotrophs are present in the pituitaries of affected mice, and they produce TSH, but their response to hypothyroidism is impaired. The thyrotroph cells of Isl1 loxP/loxP ;Tg(Tshb- cre) mice do not respond to lower T4 levels with increased TSH levels, indicating a basal deficiency in thyrotroph function. However, the degree of hypothyroidism, and particularly growth insufficiency in Isl1 loxP/loxP ;Tg(Tshb-cre) mice is more modest than the severe dwarfism of Cga (chorionic gonadotrophin alpha) knockout mice with inability to produce any bioactive TSH and an ancillary GH deficiency (19). In contrast, the Isl thyrotroph knockout mice have a more severe growth defect than the one observed in the pituitary specific knockout of Gata2 (24), a transcription factor involved in the final steps of thyrotroph differentiation. Mice with Gata2 deficiency induced during pituitary development present with a modest and transient growth defect in males but not in females. They have no thyrotroph cells at birth, but the cells eventually develop and hormone levels are normal by adulthood. This is probably due to a compensatory mechanism induced by GATA3, a close transcription factor. Interestingly, these mice have a blunted TSH response to induced hypothyroidism, suggesting that they present a partial thyrotroph deficiency+ Consistent with a thyrotroph deficiency, after inducting profound hypothyroidism by low iodine diet and exposure to PTU, Isl1 loxP/loxP ;Tg(Tshb-cre ) mice exhibit a partial response by increasing TSH production, but the response is 7 fold less than similarly treated controls, providing additional evidence that Isl1 is important for thyrotroph function. Central hypothyroidism observed in Isl1 loxP/loxP ;Tg(TSHb-cre) mice is also concordant with 2 previously published studies: these studies suggested that ISL1 was involved in the expansion and proliferation of mature thyrotroph cells: ISL1 expression was indeed increased in hyperplasic thyrotroph cells of the alpha knockout mice (19), and in mature vs. premature thyrotroph cells in vitro (18). The precise mechanisms of action of ISL1 in thyrotrophs remain to be determined Isl1 loxP/loxP ;Tg(Tshb-cre) mice have decreased Pitx2 transcripts compared to controls. This suggests that Pitx2 might be a downstream target of ISL1. Consistent with this, we found in silico 3 potential binding sites of ISL1 on PITX2 promoter (data not shown). We previously reported the phenotype of mice with a conditional inactivation of Pitx2 in thyrotrophs: Pitx2 flox/-;Tg(Tshb-cre) mice have a less severe phenotype than Isl1 loxP/loxP ;Tg(Tshb-cre) mice, with a less important growth insufficiency, unchanged hormone levels in basal conditions, and a 2-fold less important TSH increase after hypothyroidism challenge compared to controls (Castinetti et al., submitted). The interesting point was that these mice had a partial compensatory mechanism induced by an increase in Pitx1 transcription in response to Pitx2 deficiency. This compensatory mechanism was obviously not possible in Isl1 loxP/loxP ;Tg(Tshb-cre) mice as these latter also had decreased Pitx1 transcripts. This less severe phenotype suggests that ISL1 probably interacts with or regulates other factors in thyrotrophs. In the ganglion cells of the retina, ISL1 interacts synergistically with a POU homeodomain transcription factor, BRN3B, to promote cell differentiation (21). Because POU1F1 is the prototype POU homeodomain transcription factor expressed in the pituitary, and it is necessary for thyrotroph differentiation, we considered the possibility that ISL1 could interact with POU1F1 during pituitary development. However, our preliminary results do not confirm this hypothesis: ISL1 is not able to interact ! "! synergistically with POU1F1 to increase transcription of either the Pou1f1 or Tshb promoters in Hela cells (data not shown). Future work will try to determine in vitro the downstream targets of ISL1 during pituitary development. Thyrotrophs development and the LIM code Homozygous inactivation of Lhx3 and Lhx4 , 2 LIM domain transcription factors, leads to severe pituitary hypoplasia with either a dramatically decreased number ( Lhx4 -/-) or a lack ( Lhx3 -/-) of thyrotroph cells at birth (2, 25). In comparison, the precise role of ISL1 is difficult to predict: homozygous inactivation of Isl1 leads to an early death during embryogenesis (e10.5) due to heart defects, before anterior pituitary differentiation. However, during development, interactions between LIM domain transcription factors are necessary for a proper pituitary proliferation and differentiation, and suggest a role for ISL1 in thyrotroph differentiation. These interactions are defined as the LIM code. ISL1 and LHX3 expression overlaps early in pituitary development (e9.5–e10.5) but becomes mutually exclusive after that (14). Lhx3 null embryos have a transient lack of ISL1 expression at e12.5, a time point that is critical for ISL1 being restricted to the prospective anterior lobe (3, 26). However, in Lhx3 null mice, ISL1 expression is normal at later time points (3). This suggests that LHX3 expression is necessary for ISL1 expression at the specific e12.5 time point. We hypothesize that delayed ISL1 expression is probably one of the reasons why these mice lack thyrotrophs. As In silico analysis found potential binding sites of ISL1 in mouse promoters of Lhx3 and Lhx4 , it is likely that ISL1 also interacts or regulates LHX3 and LHX4 during pituitary development. Our model of inactivation of Isl1 in thyrotrophs does not give any information about the roles of ISL1 in the specification or expansion of Pou1f1 -dependent thyrotrophs between e10.5 and e15.5 because the cre recombinase becomes activated under the control of the Tshb promoter in caudo-medial thyrotrophs at ~e15.5. In mature thyrotrophs, our real time PCR data show that in the absence of ISL1, there is a significant decrease of Lhx3 and Lhx4 transcripts. This implies that interactions between LIM domain transcription factors are also necessary in mature cells. As ISL1 is also a marker of pituitary stem cells, its absence might lead to a lack of differentiation of precursors to a thyrotroph fate, which would induce a decreased expression of transcription factors (LHX4, LHX3 but also PITX2 and PITX1) involved in the following TSH differentiation steps. This suggests a master role of ISL1 in thyrotrophs function and maintenance. ISL1 mutations and congenital TSH deficiency in humans In humans there are several known genetic causes of TSH deficiency, including mutations of genes coding for transcription factors (POU1F1, LHX3, LHX4, PROP1, HESX1…), leptin receptor, TRH receptor or TSH β subunit (27). Because the ISL1 defect in mature mouse thyrotrophs induces central hypothyroidism, mutations of ISL1 could be responsible for congenital human TSH deficiency. On one hand, the fact that ISL1 is involved in the development of several mouse organs (heart, pancreas, retina), and homozygous inactivation in mice is responsible for early death during embryogenesis, the probability of having homozygous mutations of ISL1 in humans is highly unlikely, as these individuals would probably be not viable. On the other hand, correspondence between mouse and human phenotype is not always perfect, and heterozygous ISL1 mutations (for instance with a dominant negative effect, or as part of a digenic inheritance) could be identified, possibly in patients with associated heart defects or diabetes. To our knowledge there is only one study that searched for exonic ISL1 mutations in 3 patients presenting with pancreatic hypoplasia, neonatal diabetes mellitus, congenital heart defect and developmental delay, and none were found (28). ! "#! ISL1 is not required for maintenance and function of gonadotrophs Mice with Isl1 deleted in gonadotrophs had normal puberty and fertility, suggesting that ISL1 is dispensable for mature gonadotroph function. This is surprising as ISL1 is known to act synergistically with LHX3 and SF1 to promote gonadotroph differentiation, and ISL1 is necessary for the expression of the GnRH receptor (29). It is possible that ISL1 has a role that can be compensated for by other LIM homeodomain transcription factors. ISL1 might also be necessary for gonadotrophs development, between e10.5 and e16, but not for mature gonadotroph function. As the Lhb-cre recombinase is activated after the cells begin to express LHb, our experiments can’t give any clues about gonadotroph development. To conclude, our study shows for the first time the role of ISL1 in mature thyrotrophs. Inactivation of ISL1 leads to a thyrotroph deficiency with an inappropriate TSH response to low thyroid hormone levels. The downstream targets of ISL1 remain to be determined: our preliminary results suggest that these might not be direct targets (like Tshb or Pou1f1 promoters). Ongoing work will aim to determine ISL1 interacting factors in the pituitary. This might explain why ISL1 is necessary for mature thyrotrophs but not mature gonadotrophs, whereas it seems necessary for the development of both axes. ACKNOWLEDGEMENTS We would like to thank the Dr. Jun Z. Li and his lab members, and Dr. Richard Miller, James Harper and members of the Miller Lab for their help. Financial support came from the Center for Genetics in Health and Medicine, University of Michigan (FC), and the National Institutes of Health (R37HD30428, R01HD34283 to SAC). Experimental design and manuscript preparation was by FC and SAC. MLB established and characterized the Tshb-cre transgenic line and LG provided the floxed Isl1 allele. KV and AH established and carried out the TSH assays. All other experiments were planned and carried out by FC. ! ""! LEGEND TO FIGURES Figure 1. Isl1 is expressed in Rathke’s pouch during embryogenesis and in alpha- subunit secreting cells and gonadotrophs after birth. Panel A: ISL1 immunostaining in the developing pituitary during embryogenesis and after birth. ISL1 expression was observed at e12.5, e14.5, e18.5 and after birth. Panel B: Co-immunostaining ISL1 and pituitary hormones. The majority of cells expressing the alpha-subunit (CGA) and LH β are expressing ISL1, whereas no somatotroph, lactotroph or corticotroph cells express ISL1. Figure 2. Efficient deletion of Isl1 in thyrotrophs by Tshb-cre BAC transgene . Panel A: Co-immunostaining of ISL1 and TSH β (upper panel) and ISL1 and LHß (lower panel) in newborn control and Isl1 loxP/loxP ; transgenic mice. Upper panel. About 50% of thyrotroph cells express ISL1 in controls (green arrows), whereas ISL1 was not detected in other thyrotrophs (red arrow). In Isl1 loxP/loxP ;Tg(Tshb-cre) mice the majority of thyrotroph cells do not express ISL1 (red arrow). ISL1, green staining; TSH β, red staining. Lower panel. The majority of gonadotroph cells express ISL1 in controls (green arrows); it is undetectable in only a few, rare gonadotrophs (red arrow). In ISL1 loxP/loxP ;Tg(LHb-cre) mice, the majority of gonadotroph cells do not express ISL1 (red arrow). ISL1, green staining; LH β, red staining Panel B: There is no obvious change in ISL1 expression in the thyroid of Isl1 loxP/loxP ;Tg(Tshb-cre) mice, suggesting that the phenotype we observed has no peripheral etiology. Figure 3. Isl1 loxP/loxP ;Tg(Tshb-cre) mice present a significant growth insufficiency . Growth curves of Isl1 loxP/loxP ;Tg(Tshb-cre) (transgenics) and controls males and females between 3 to 8 weeks of age. Statistical significance: ***, p<0.001; **, p<0.01; *, p<0.05 Figure 4. Isl1 loxP/loxP ;Tg(Tshb-cre) mice have impaired thyrotroph function. Panel A: T4 levels are significantly decreased in Isl1 loxP/loxP ;Tg(Tshb-cre) vs. controls ***, p<0.001. Panel B: Decreased Tshb , Lhx3 , Lhx4 , Pitx1 and Pitx2 transcripts in Isl1 loxP/loxP ;Tg(Tshb-cre) mice compared to controls. *, p<0.05. Scale, fold decrease in transgenics vs. controls. Panel C: Thyroid volume is decreased in Isl1 loxP/loxP ;Tg(Tshb- cre) mice vs. controls (*, p<0.05 after correction for weight). Panel D: There is no obvious difference in TSH β staining in Isl1 loxP/loxP ;Tg(Tshb-cre) at 8 weeks old. TSH β staining was performed in 3 transgenics and 3 controls, in 5 pituitary sections at regular interval for each mouse. Figure 5. Blunted TSH response to hypothyroidism in Isl1 loxP/loxP ;Tg(Tshb-cre) mice. Isl1 loxP/loxP ;Tg(Tshb-cre) (transgenics) mice and controls were treated for 4 weeks with a low iodine diet enriched in propyl-thio-uracil (0.15%). Panel A: Total T4 levels were significantly decreased at the end of the treatment in both transgenics and controls. ***, p<0.001; **, p<0.01. Panel B: TSH β levels were not significantly different in basal conditions between transgenics and controls. After hypothyroidism challenge, there was a significantly reduced increase in transgenics compared to controls. Note the logarithmic scale. ***, p<0.001; **, p<0.01 ! "#! REFERENCES XT! 17(1C8'9!-R!5QQ@E5!R!@G177U7'031!R!@(59D@9! ! X`! `dT! 1&11!R!09)07D@39!R!E0H&CE!7R!8)7F9)! R!0DD077! !`hhe!0BF4C0809E! 1@C!!3@80@'@I!2090!D7`!49!8@E@C!90FC@9!2090C&E4@9!C0G0&7D!&!8@E@C! 90FC@9U)0A09)09E!DE0A!49!49E0C90FC@9!)4110C09E4&E4@9T!7077!gcSbYhUbaY! `eT! 4F!R!N0!$R!"&92!#R!%3092!#R!7F4!!aYYd!S0G07@A809E&7!(3&920D!@1!D70EU`!&9)! 4ED!(@U7@(&74Q&E4@9!H4E3!A4EF4E&CP!3@C8@90D!49!E30!A4EF4E&CP!27&9)!@1!(34(6! 08'CP@!'P!488F9@34DE@(3084DECPT!7077! 4DDF0!0D!baaSafhUagf! `fT! 4F!R!4F!%R!$4!R!7F4!!aYYd!D70EU`!0IAC0DD4@9!&9)!4ED!(@7@(&74Q&E4@9!H4E3! 7FE0494D492!3@C8@90R!E3PC@4)UDE48F7&E492!3@C8@90!&9)!@0DEC@209!C0(0AE@C! &7A3&!49!E30!)0G07@A492!A4EF4E&CP!27&9)!@1!E30!D300A!1@0EFDT!!0FC@09)@(C49@7! `fSffbUfgY! `gT! 0CC!R![@C)@9!S\R!"@@)8&9D00!""R!&C&AFC&!!SR!4)2H&P!87R!"@@)!"! aYYd![C@HE3!&CC0DE!@1!E3PC@EC@A4(!EF8@CD!'P!E3PC@4)!3@C8@90!4D!(@CC07&E0)! H4E3!9@G07!(3&920D!49!"9EU`Y^T!@7!7077!89)@(C49@7!abgSdfUef! `hT! 09)&77!R!&8F07D@9!7R!&F9)0CD! R!"@@)!R!7&8A0C!^!`hhd! &C20E0)! )4DCFAE4@9!@1!E30!A4EF4E&CP!27P(@AC@E049!3@C8@90!&7A3&UDF'F94E!AC@)F(0D! 3PA@2@9&)&7!&9)!3PA@E3PC@4)!84(0T![090D!S0G!hSaYYfUaY`h! aYT! 73&C70D!^R!@CE09D09!^NR!@E@6!^R!7&8A0C!^!aYYg!4EIa!)070E4@9!49! A4EF4E&CP!2@9&)@EC@A0D!4D!(@8A&E4'70!H4E3!2@9&)&7!)0G07@A809ER!AF'0CEPR! &9)!10CE474EPT![090D4D!ceSdYfUd`c! a`T! &9!R!S092!R!#40!#R![&9!!aYYg!`!&9)!_b_!(@UC02F7&E0!E30! )4110C09E4&E4@9!@1!8FC490!C0E49&7!2&9274@9!(077DT!S0G07@A809E!`bdS`hg`U`hhY! aaT! !0DA0C!^NR!&0EQ8&9! R!7&8A0C!^!aYYe!@70!@1!AC@A30E!@1!4E`!V`W!49! 2@9&)@EC@A0!)4110C09E4&E4@9!&9)!AF'0CEPT!89)@(C49@7@2P!`cfS`edcU`eeb! abT! "04DD!8R!FC&E&!$R!7F&!R!N&P&D34!$R!0@!NR!010E@11!!`hhg! 3PC@4)! 3@C8@90!&(E4@9!@9!74G0CR!30&CER!&9)!090C2P!0IA09)4EFC0!49!E3PC@4)!3@C8@90! C0(0AE@C!'0E&U)014(409E!84(0T!89)@(C49@7@2P!`bhSchcdUchda! acT! 73&C70D!^R!&F9)0CD! R!"@@)!"R!H09D!R!&C7@H!^\R!7&8A0C!^R! 4)2H&P!87R![@C)@9!S\!aYYe!4EF4E&CPUDA0(414(![&E&a!69@(6@FES!0110(ED!@9! 2@9&)@EC@A0!&9)!E3PC@EC@A0!1F9(E4@9T!@7!89)@(C49@7!aYS`beeU`bff! adT! 3092!N%R!%3&)&9@G!^_R!@D4920C!_R!CTR!\F544! R!_0CEFQQ4!R![C49'0C2!^R!00!8R! NF&92!R!&3@9!^R!"0DEA3&7!N!`hhe!A0(414(&E4@9!@1!A4EF4E&CP!(077!7490&20D! 'P!E30!!3@80@'@I!2090!3IbT!(409(0!afaS`YYcU`YYf! aeT! %3&@!$R!@C&70D!S7R!N0C80DQ!8R!00!"R!1&11!R!"0DEA3&7!N!aYYe!0)F(0)! 0IAC0DD4@9!@1!E30!U3@80@'@I!2090!3Ib!48A&4CD!2C@HE3!&9)! )4110C09E4&E4@9!@1!&E360XD!A@F(3!&9)!49(C0&D0D!(077!&A@AE@D4D!)FC492!8@FD0! A4EF4E&CP!)0G07@A809ET!0(3!S0G!`abSeYdUe`b! afT! $&8&)&!R!@C4!!aYYg!0(3&94D8D!C07&E0)!E@!E30!A&E3@A3PD4@7@2P!&9)! 8&9&20809E!@1!(09EC&7!3PA@E3PC@4)4D8T!&E!7749!C&(E!89)@(C49@7!0E&'! cSegbUehc! agT! _&7&DF'C&8&94&9!R!3407)!R!^(0C494!7R!"&760C!R!877&C)!R!&C(3&9)!R! @7&6!R!!&I477&4C0!R!7C@77&!^R!_F9P&9!SR!&(6&P!SR! 08A70!!aY`Y! &9(C0&E4(!3PA@A7&D4&!AC0D09E492!H4E3!90@9&E&7!)4&'0E0D!80774EFD!49! &DD@(4&E4@9!H4E3!(@92094E&7!30&CE!)010(E!&9)!)0G07@A809E&7!)07&PT!^8!!0)! [090E!^!`da^SbcYUbce! ahT! [C&920C!^R!_70FI!7R!@EE70C!R!3@)0D!R!7@F94D!R!&G0CC40C0!!aYYe! 30! U3@80@)@8&49!AC@E049D!D7U`!&9)!3Ib!&(E!H4E3!DE0C@4)@2094(!1&(E@C!`!E@! ! `c! $%&'%($!)*%'+*,-*.$/0.$(121(!'(,131,4!*2!,&$!)*%'+*,-*.1%/-$5$'01%)! &*-6*%$!-$($.,*-!)$%$!.-*6*,$-7!8*5!9%+*(-1%*5!:;<:;=>/:";?! ! ! "#! !" -./01$ -.201$ -.301$ *.$ #" '&$ *+%$'&$'&$ ,$ ($ (!)&$ !"#$ !"#$ %&$ #" $%&" C T '&" !" :120;8"<0,-4/=" &!" ?" %#" ???" ???" %!" ???" +,-./01.23"4-51/" ?" 67.8,75"4-51/" $#" ???" ???" +,-./01.23"914-51/" 67.8,75"914-51/" ???" $!" ???" #" !" &" '" #" (" )" *" :11>/" @20A,1"&" !"!" G" :9-;"7<345."64"" *+,!"-./.-0" =913-"=(" 123405.46708794129-0" %" )" #%>)" ?3-.0" :.*3-.0" %" #%>(" #)" #%>'" /01% !"#$% &#'(% &#')% *+,'-% *+,'.% *234-% #%>&" 71,1.% #(" I" *35-6-% #$" I" #$>)" #'" I" #$>(" I" #&" #$>'" III" III" #$>&" #$%" I" B" H" = :65D2."(" EEE" ?56-/"?&"@=>A;/B" EEE" ?D2"@4>A;/B" +" #!!!!!!" EE" *" EE" #!!!!!" )" (" #!!!!" '" #!!!" &" %" #!!" $" #!" #" !" #" ,-.-/"/010/." 2345673859:9.;" ,-.-/"/010/." 2345673859:9.;" <7-//0=>0" <7-//0=>0" ?8-=.>0=9<." CC5=685/."5=685/." F9>G80"'" DISCUSSION Les souris avec inactivation d’Isl1 présentent un déficit thyréotrope complet. Du fait même du mode d’expression de notre recombinase (après l’expression de TSHb, donc après e15 au cours du développement murin), il ne nous est pas possible d’affirmer de façon formelle que ISL1 est nécessaire au développement de l’axe thyréotrope. Par contre, nos résultats soulignent clairement son rôle majeur dans les cellules thyréotropes matures. En particulier, ces souris transgéniques ont une réponse très faible de la TSH en situation d’hypothyroïdie induite. Point très intéressant, en tenant compte de la période d’expression très précoce de ISL1 au cours du développement hypophysaire (en comparaison avec d’autres facteurs de transcription), l’expression de la majorité des facteurs de transcription nécessaires au développement et/ou à la fonction des cellules thyréotropes est retrouvée abaissée chez les souris avec inactivation d’Isl1. Ces résultats suggèrent donc que ISL1 joue un rôle majeur et précoce dans les cellules thyréotropes. Les résultats obtenus dans le système rétinien, avec une action synergique de Pou4f3 et Isl1 sur le promoteur de Pou4f3, laissaient envisager une action similaire dans l’hypophyse dans l’hypophyse : Isl1 aurait pu agir en synergie avec Pou1f1 pour activer le promoteur de Pou1f1 et/ou de Tshb. Nos résultats préliminaires in vitro n’ont pas permis de confirmer cette hypothèse. Il est probable que ISL1 puisse interagir avec de nombreux autres facteurs de transcription : en effet, les transcrits de ces facteurs sont diminués dans les souris avec inactivation d’Isl1 (comme expliqué précédemment), et il existe in silico des sites potentiels de liaison d’ISL1 sur les promoteurs des gènes de facteurs de transcription hypophysaires (Lhx3, lhx4, pitx2…). Les futurs axes de recherche devront chercher à déterminer les partenaires d’ISL1 au niveau thyréotrope. De façon surprenante, alors qu’il existait des arguments en faveur d’un rôle de ISL1 au sein de l’axe gonadotrope (en particulier l’expression de ISL1 dans la majorité des cellules gonadotropes à l’âge adulte), les souris avec inactivation de Isl1 spécifiquement dans les cellules gonadotropes (en utilisant une recombinase sous contrôle du promoteur de Lhb) ne présentent pas de défit statural ou de retard pubertaire. Cela pourrait suggérer que ISL1 n’est pas nécessaire dans les cellules gonadotropes matures, mais plutôt au cours du développement de l’axe (hypothèse que nous ne pouvons pas tester avec cette recombinase). ISL1 pourrait-il avoir un rôle dans les insuffisances thyréotropes congénitales observées chez l’homme ? Le principal écueil à cette hypothèse est l’implication multi-organes de ISL1 : ce facteur de transcription est en effet impliqué dans le développement du cœur, du pancréas, de la rétine. De plus, les souris avec inactivation homozygote de Isl1 meurent de façon très précoce au cours de l’embryogenèse. Il est donc probable qu’une mutation homozygote ne serait pas viable chez l’Homme. Il reste la possibilité d’une mutation hétérozygote, avec effet dominant négatif ou haplo-insuffisance, qui se manifesterait par un déficit thyréotrope, potentiellement associée à des anomalies d’autres organes. Un séquençage du gène chez ce type de patients pourrait donc être proposé dans une optique de recherche de gènes candidats. ! %"! PERSPECTIVES ! %#! La plupart des données et des algorithmes décisionnels d’aide au diagnostic moléculaire des hypopituitarismes congénitaux sont basés sur les phénotypes des patients déjà rapportés dans la littérature, et, en amont, sur les données obtenues à partir des modèles murins d’inactivation de gènes. Il existe cependant des différences entre les modèles murin et humain. Comprendre les causes de ces différences devrait augmenter le pourcentage d’étiologies identifiées. Il existe cependant d’autres possibilités d’identification des facteurs en cause dans les hypopituitarismes congénitaux : transcriptome, séquençage exome, gènes candidats… L’objectif final est d’améliorer le traitement des patients atteints d’hypopituitarisme congénital : dans cette optique, identifier les gènes impliqués dans le développement hypophysaire pourrait permettre de faciliter l’identification puis la différenciation de cellule souches hypophysaires dans une optique thérapeutique. Ces 3 points spécifiques vont être détaillés dans la suite de cette discussion. 1. COMMENT EXPLIQUER LES DIFFERENCES ENTRE LES MODELES MURIN ET HUMAIN ? Le modèle murin reste le modèle de choix pour améliorer nos connaissances sur le développement hypophysaire humain. De nombreux gènes codant pour des facteurs de transcription impliqués dans l’hypopituitarisme congénital (Prop1, Pou1f1, Hesx1, Lhx3, Lhx4…) ont été rapportés chez la souris, avant de faire l’objet de séquençage chez l’homme et d’identifier des mutations responsables d’hypopituitarisme congénital (95). Les modèles murins ont également permis de clarifier les mécanismes d’interaction et les mécanismes physiopathologiques engendrés par ces mutations. Le décalage entre l’identification du gène chez la souris et la corrélation avec la pathologie humaine peut être plus ou moins long, ce qui explique en partie pourquoi certains gènes, bien que importants dans le développement hypophysaire murin, n’ont pas encore été identifiés en pathologie humaine (voir la table 6 pour les différences phénotypiques entre les modèles murin et humain). Ainsi, le rôle de lhx4 dans le développement hypophysaire murin a été identifié chez la souris en 1997 (180), la première mutation humaine responsable d’hypopituitarisme a été identifiée en 2001 (117), et les 5 suivantes entre 2008 et 2010 (26; 149; 199). Un tel délai peut s’expliquer par la variabilité du phénotype engendré par les mutations chez l’homme, à l’origine d’une probable sélection inappropriée des patients à séquencer. Le premier patient porteur d’une mutation de LHX4 présentait une anomalie cérébrale (117); le séquençage de LHX4 a ainsi logiquement été effectué majoritairement sur les patients porteurs d’hypopituitarisme associé à une anomalie cérébrale entre 2001 et 2008. On sait désormais que le phénotype des patients porteurs de mutations de LHX4 peut être extrêmement variable, avec en particulier des patients sans aucune anomalie extra-hypophysaire (149). Le délai peut aussi s’expliquer par une différence de phénotype entre le modèle murin et l’homme : ainsi les souris hétérozygotes pour une inactivation de lhx4 ( Lhx4 +/-) ne sont pas affectées, alors que les souris présentant une inactivation homozygote de (lhx4 -/-) présentent un phénotype hypophysaire (et une atteinte systémique avec une mort néonatale précoce) (180; 158). Chez l’homme, toutes les mutations de LHX4 décrites à ce jour sont hétérozygotes et à l’origine d’un phénomène d’haplo-insuffisance. Il est probable que les mutations homozygotes de LHX4 sont létales pour l’homme au cours du développement embryonnaire. Cependant, à l’exception de l’atteinte systémique, le phénotype hypophysaire homozygote murin est finalement très proche du phénotype hétérozygote ! %$! humain, suggérant que les mutations de lhx4 sont moins sévères chez la souris que chez l’homme. De même, il existe des différences importantes entre le modèle murin de mutation spontanée de Prop1 (souris Ames) et la pathologie humaine liée aux mutations de PROP1 (191; 95). Les déficits hypophysaires de même que la morphologie hypophysaire (voir plus loin) sont différents : à l’exception du classique déficit somato- lactotrope et thyréotrope (présent dans les 2 espèces), le déficit corticotrope est présent de façon inconstante chez l’homme alors qu’il n’est jamais retrouvé chez la souris (95). De même, le déficit gonadotrope (présent systématiquement chez les patients porteurs de mutations de PROP1 bien que d’âge et de présentation variables au diagnostic) semble être fonctionnel chez la souris puisque la substitution en hormones thyroïdiennes et GH permet un retour à une fertilité normale; chez l’homme, le déficit gonadotrope est organique avec une infertilité dans 100% des cas. Les différences murin/humain ne peuvent pas s’expliquer par une simple différence inter-espèces d’activité de la protéine Prop1 sur son promoteur : les mutations murines et humaines sont homozygotes, et le modèle de souris Ames, lié à une mutation spontanée d’un acide aminé, est identique à celui observé pour les souris Prop1 -/- (en culture cellulaire, Prop1-S83P n’a aucune activité de stimulation résiduelle du promoteur de Pou1f1). L’hyperplasie hypophysaire est un bon exemple de la différence entre les modèles murin et humain. Ainsi, une hyperplasie hypophysaire a été observée chez des patients porteurs de mutations de PROP1 , LHX3 et LHX4 (voir table 7 pour les mutations de facteurs de transcription responsables d’une hyperplasie hypophysaire chez l’homme) sans que le modèle murin ne permette d’expliquer ce phénomène : les souris Ames présentent au cours de l’embryogenèse une pseudo-hyperplasie hypophysaire, avec un volume normal du lobe antérieur, et une lumière élargie ; après la naissance, une hypoplasie est systématiquement observée (191; 158). De façon intéressante, l’hypoplasie hypophysaire survient après l’extinction de l’expression de Prop1, ce qui est en accord avec l’hypothèse que cette hypoplasie pourrait être liée à une apoptose de progéniteurs localisés en situation ectopique (223). Il a été suggéré par une étude histologique que l’hyperplasie hypophysaire chez l’homme pourrait être due à la masse de ces progéniteurs incapables de se différencier (239). Le mécanisme physiopathologique est cependant complexe, et doit faire intervenir d’autres facteurs (facteurs de transcription, voie de signalisation, facteurs exogènes environnementaux) puisque des patients avec la même mutation (y compris au sein de la même famille) ne présentent pas nécessairement la même morphologie hypophysaire au même âge. Ce dernier point et le fait que des patients très jeunes ont été rapportés comme ayant une hypoplasie hypophysaire, exclut la possibilité d’une hyperplasie systématique précédant l’hypoplasie. La fréquence du déficit corticotrope est un autre point marquant dans le phénotype des patients avec hyperplasie hypophysaire et mutations de PROP1, : 80% des patients avec hyperplasie présentent un déficit corticotrope, en comparaison avec environ 40% de la population générale de patients avec mutations de PROP1 . Cela est d’autant plus surprenant que le déficit corticotrope est en général retardé, alors que l’hyperplasie hypophysaire survient chez des patients plus jeunes. Ce dernier point est également en faveur d’une anomalie de la différenciation et de la prolifération des progéniteurs. Mieux comprendre l’hyperplasie hypophysaire pourrait permettre de mieux comprendre la survenue d’un déficit corticotrope (inconstant) chez les patients porteurs de mutations de PROP1 . Enfin, de la même façon, les souris avec inactivation homozygote de lhx3 et lhx4 présentent une hypoplasie hypophysaire au cours de ! %%! !"#$%&'!()*+,-$)./0/1) !"#$%&'!()"+*2-$).*+&2&-%$31) &,2$3*-33-%$) !,%!4) !"%(8'()Q')Q'/R'"%(8'/)V3)' !"%(8'()Q')Q'/R'!"%(8'/)V3)S'& ' " 77(C '72B6(7' 8' #219211 )R'47' '"%(8' 1'=>)' C6() R'!"%(8'=>)'(1217818' '200 ' )E424'E7 S''E424)7( ' )E424'E7 S''E4 64)7( '8617(82(6 '(1217818 ' !%+454) !"%(8'()Q')Q'/' !"%(8'()Q')Q'/' " 77(C '72B6(7R' )E424'E7 S''E424)7( ' )E424'E7 S''E424)7( ' 6" 77(C V20(118' '200 ' "(364) '"128E4 'C6() Q'120)( 7' ')U 1"4') ' !"%(87''E424'E7(6 7'C6() 7' " 77(#'' F')'72B6(7R' 8' ')')(&1 '0"(1 Q'0)#2609217' )E424'E7 ''E424)7(5B ' !20(118'2B'6" 77(#' 24'8)0(5B 7' 278W'E424'E7 ' 824(5B '2B'1260) Q' ' F')U'200 ' )E424'E7 S''E4 64)7( '4B(7''E424)7( Q'2B' 'E424)7( ' 7'1 6#7'2495B 7' 8''E424)7( 'B' )2)(7921' 824(5B Q'2B'0B)94) 7'()387' 2647')) BD'(12178187T'' !E74)7( '7 482W2495B T' %&67) P)#2609217'7"C!6 7' 7'786B8B6 7' !"%(87''E424'E7(6 7'C6() 7'X"%(8'(72)"' " 77(#'72B6(7' 8' 18"6( B6 7' 1'()''41'E424(8B(86(70 Y' '200 ' !E70264'( ''E424'E7(6 '7('(19C921' )E424'E7 S''E424)7( '2B'1260) ' '"8"62FE&28 ' 278W'E424'E7 ' 824(5B '2B'1260) ' =120)( 7'"6"6) 7V24'8)02)2&(5B 7' (1217818 7' 8"69) =4)7( ''E424'E7(6 Q'717')2 '18"6( B6'1(' !"%(8'()Q')Q'/)V3)R'"%(8'=>)' " 77(C '72B6(7' 8' (18 60"((6 ' (1217818' '200 ' )E424'E7 S''E424)7( '2B''E4 64)7( ' =120)( 7' ')'628921'B'2BQ'A (18 ' B(9C ' 8"6:) )E424'E7 ''E424)7(5B Q'($"6 1(921' !"%(87''E424'E7(6 7' 8'120)( 7' " 77(C '' F')' 1260) ' 7'`')(&1" 7' 0264'2)2&(5B 7''E424'E7(6 7' 8' D86W 72B6(7Q'20(118'' F' 'E424'E7(6 7'C6() 7'' )U'200 ' !;<=>BC)GHCB;IFBC)B=J;B)ABC)?KF=GJL?BC)MN;<=)BJ)KNM@<=)<=DN !"#$%&'()*+),$-#+".()*+)#.$'(-./0%&')$((&-/1+()2)"'+)340+.05$(/+)340&034($/.+6) !!#"4()2(!,""3(12(2##5(6'&$2(;!(26#"(# !"(6( ! #"( &!(;#!(&(( #"5( l’embryogenèse (181; 180), en contraste avec l’hyperplasie hypophysaire observée chez certains patients, sans qu’un explication plausible n’ait été envisagée jusqu’alors. Les différences modèles murin/humain pourraient donc s’expliquer par - des interactions différentes avec d’autres protéines/facteurs de transcription impliqués dans le développement hypophysaire, - des périodes d’expression spatiales ou temporelles différentes des facteurs de transcription au cours de l’embryogenèse murine et humaine - des différences d’activités résiduelles (perte partielle ou complète d’activité) des protéines codées par chaque mutant sur les promoteurs cibles (y compris des promoteurs cibles non explorés à ce jour en pathologie) - des rôles plus ou moins important de protéines/facteurs de transcription de la même famille au cours du développement hypophysaire. Par exemple, les souris présentant une inactivation de Gata2 dans les cellules thyréotropes ont une augmentation d’expression de Gata3 (28). Les souris avec inactivation de Pitx2 dans les cellules gonadotropes ont un phénotype normal car il est probable que Pitx1 supplée Pitx2 dans cette situation (27). Il est donc tout à fait possible que les rôles respectifs de chaque protéine d’une même famille soient différents dans les 2 espèces, aboutissant à des différences phénotypiques. Cela pourrait expliquer que l’inactivation de Pitx2 dans les cellules thyréotropes aboutisse à un phénotype de déficit thyréotrope modéré. - Des rôles différents de certaines voies de signalisation impliquées dans l’embryogénèse (Wnt, Notch), et interagissant avec les facteurs de transcription hypophysaires. - Des facteurs environnementaux ou de terrain génétique sous-jacent. Ainsi, la souche d’environnement génétique dont sont issues les souris Prop1 conditionne leur survie néonatale en termes de sévérité du déficit thyréotrope. ! %&! 2. COMMENT AMELIORER L’IDENTIFICATION DES FACTEURS DE TRANSCRIPTION IMPLIQUES DANS LE DEVELOPPEMENT ET LA PATHOLOGIE HYPOPHYSAIRES ? Le séquençage basé sur une approche « gènes candidats » est l’approche utilisée dans le laboratoire pour identifier les étiologies d’hypopituitarisme congénital. Cette approche a été d’une grande utilité car elle a permis d’identifier la plupart des mutations de facteurs de transcription hypophysaire identifiées à ce jour, et responsables d’hypopituitarisme congénital. Elle reste d’une importance majeure en première intention dans l’étude des étiologies de déficits hypophysaires congénitaux. Cependant, cette approche est considérée comme à risque pour plusieurs raisons : - la méthode de sélection des gènes est basée sur le modèle murin. Comme expliqué précédemment, le modèle d’inactivation d’un gène murin ne reproduit pas intégralement le modèle de pathologie humaine. Cela peut donc conduire à une sélection inappropriée de gènes candidats à partir d’un phénotype. - L’algorithme de sélection des gènes à séquencer est basé sur les phénotypes de patients porteurs de mutations déjà publiées. Cela explique peut être en partie pourquoi certains gènes ne sont retrouvés que rarement comme étant responsables d’hypopituitarisme congénital (voir l’exemple de LHX4 décrit précédemment). - La méthode ne permet pas de mettre en évidence les remaniements de grande taille suite à une large délétion ou une large insertion, les anomalies introniques situées à distance des bornes mais modifiant les sites d’épissage, voire, à un niveau moindre (car n’a jamais été rapporté comme étant responsable de pathologie hypophysaire chez l’homme) les anomalies au niveau des extrémités 5’UTR et 3’UTR. D’autres méthodes pourraient être également utilisés pour améliorer le pourcentage d’étiologies identifiées ou la sélection des gènes candidats. a. Identifier des gènes candidats dans le modèle murin Davis et al. ont développé un transcriptome d’ADN complémentaires d’hypophyses de souris de génotype sauvage et de souris Ames à e12,5 et e14,5 (48; 49). L’objectif était de déterminer les gènes surexprimés et sous-exprimés dans l’hypophyse de souris Ames par rapport à des souris de génotype sauvage. Ils ont ainsi rapporté l’implication de gènes de « nouveaux » facteurs de transcription à type de domaines en doigts de zinc (Zhfx3 qui est capable de se lier au promoteur Pou1f1 et semble nécessaire à l’expression de Pou1f1, TSHß et GH ; Zeb2 dont le rôle dans l’hypophyse n’avait pas encore été mis en évidence), de facteurs de transcription à homéodomaine LIM (lhx2 dont le rôle dans le développement hypophysaire n’avait pas été mis en évidence), de facteurs de transcription à homéodomaine POU (Adnp qui semble nécessaire à l’expression de NeuroD1, impliqué dans la différenciation corticotrope), et une protéine à homéodomaine régulant la migration et différenciation de progéniteurs (Emx2). Au total, 45 nouveaux facteurs, appartenant à 15 familles différentes, potentiellement impliqués dans l’embryogenèse hypophysaire, ont pu être identifiés (17). La première étape ultérieure consistera à évaluer leur profil d’expression au cours du développement hypophysaire (immunohistochimie ou hybridation in situ). En fonction des résultats, et d’éventuelles études associées en cultures cellulaires (activation de promoteurs d’hormones hypophysaires, ou de facteurs de transcription impliqués dans la différenciation hypophysaire), un sous-groupe de gènes pourra être sélectionné pour envisager une inactivation dans le modèle murin. Cette approche a déjà été utilisée pour Isl1. Nos résultats d’expression et d’inactivation de Isl1 au cours du développement ! %'! hypophysaire suggèrent que le séquençage de Isl1 pourrait être effectué chez les patients présentant un déficit thyréotrope congénital isolé. b. Utiliser un autre modèle animal pour tester des gènes candidats : le poisson zèbre (zebrafish) Le zebrafish (Danio rerio ) est un poisson tropical qui présente de nombreux avantages pour l’étude du développement : l’ensemble de son génome a été décodé ; son développement est rapide (3 jours pour le passage des œufs aux larves, même si la période globale de développement est identique à celle de la souris) ; surtout, les embryons sont transparents et se développent en dehors de la mère, ce qui permet une observation en temps réel des étapes majeures de l’embryogenèse ; les drogues éventuelles peuvent être ajoutées directement dans l’aquarium. L’inactivation d’un gène est possible via l’injection d’oligonucléotides antisens de type morpholino, qui vont se lier sur la séquence complémentaire d’ARN et ainsi réduire l’expression du gène d’intérêt. Le principal problème théorique est la difficulté d’inhiber complètement l’expression d’un gène du fait de la duplication du génome (2 gènes paralogues) au cours de l’évolution du zebrafish. Le modèle du zebrafish pourrait être une étape de présélection des gènes candidats du fait de la rapidité du développement, avant de passer à l’étape d’inactivation du gène chez la souris (151). c. Utiliser le profil d’expression de gènes dans des cellules souches en différenciation La différenciation de cellules souches embryonnaires en cellules hypophysaires différenciées a été rapportée à 2 reprises (avec obtention de cellules somato-lactotropes et thyréotropes dans le 1 er cas, et de cellules gonadotropes dans le 2 ème cas) (235; 222). Evaluer le profil d’expression successif des gènes impliqués dans cette différenciation pourrait permettre d’identifier de nouveaux gènes potentiellement impliqués dans les hypopituitarismes congénitaux. Il devrait également être possible de définir les partenaires potentiels des principaux facteurs de transcription (co-répresseurs, co- activateurs) dont des anomalies pourraient être en cause dans les hypopituitarismes congénitaux. Les limites actuelles reposent sur la difficulté de répliquer ces expériences, et de comprendre les différences entre les 2 expériences (ou entre les 2 souches différentes de cellules embryonnaires) qui sont à l’origine d’une différentiation différente (Pit1 dépendante ou non) : il n’existe ainsi pas de différence formelle en termes de milieu ou de conditions de culture entre la publication ayant décrit l’obtention de cellules exprimant Pit-1 et la publication ayant décrit l’obtention de cellules gonadotropes (235; 222). d. Utiliser le séquençage de l’exome d’un individu atteint d’hypopituitarisme congénital Le séquençage de l’ensemble du génome d’un individu en routine ne peut pas être effectué au moins pour des raisons évidentes de coût et de temps. Une possibilité est de se limiter à séquencer l’exome, c’est à dire l’ensemble des régions codantes du génome d’un individu (136; 137). L’efficacité de la technique a été évaluée en recherchant le gène (MYH3, déjà identifié) responsable du syndrome Freeman-Sheldon chez 4 patients porteurs de la pathologie, en comparaison avec 8 sujets contrôles (137). La présence de sujets contrôles est nécessaire pour éliminer les variants alléliques les plus fréquents ; idéalement les parents de l’individu doivent aussi être séquencés pour éliminer les variations intrafamiliales. Cette technique pourrait être particulièrement utile pour les patients présentant un hypopituitarisme congénital, pour lesquels les gènes des facteurs de transcription classiques ont déjà été séquencés. ! %(! 3. COMMENT AMELIORER LE TRAITEMENT DES DEFICITS HYPOPHYSAIRES: UN ROLE POUR LES CELLULES SOUCHES ? (voir Annexe : Castinetti et al., soumis à Endocrine Reviews) La prise en charge des déficits hypophysaires est délicate aussi bien en termes de voies d’administration (injection pour l’hormone de croissance, multiprise quotidienne pour l’hydrocortisone) qu’en termes de concentrations de la substance dans l’organisme par rapport aux concentrations physiologiques (cela est en particulier vrai pour l’hydrocortisone et la difficulté de reproduire le rythme nycthéméral physiologique). Les hormones hypophysaires ne sont pas substituées, et la substitution des hormones périphériques est parfois difficile à surveiller (dans les déficits thyréotropes, il n’existe pas de marqueur biologique fiable d’évaluation de l’efficacité du traitement par les hormones thyroïdiennes). De plus ces traitements ont un coût non négligeable (donnés à vie pour la plupart), et peuvent entrainer des effets secondaires. Les cellules souches hypophysaires seraient théoriquement capables de réplication illimitée, et pourraient donner naissance aux 5 lignées de cellules hypophysaires. Elles représenteraient donc un outil idéal pour le traitement des déficits hypophysaires. a. Caractériser les cellules souches hypophysaires Une cellule souche se définit par l’expression de marqueurs de cellules non différenciées, une capacité de réplication et prolifération non limitée, et une capacité de régénération de tissu après une perte tissulaire en se différenciant en plusieurs types cellulaires impliqués dans le tissu (multipotence). Le terme progéniteur est réservé aux cellules issues de cellules souches en cas de division « asymétrique » : un progéniteur est une cellule ayant une capacité de réplication limitée, qui donnera naissance à une cellule fille (cellules d’amplification transitoire), qui va progressivement se différencier (132). Plusieurs travaux récents ont renforcé l’hypothèse selon laquelle l’hypophyse contient des cellules souches, malgré son statut d’organe à « faible taux de prolifération ». La population de cellules souches ou progéniteurs est cependant difficile à caractériser du fait de différentes périodes d’évaluation (embryogenèse, après la naissance, âge adulte) et de différents marqueurs cellulaires évalués dans les études. Les données sur les cellules souches au cours de l’embryogenèse doivent être séparées de celles portant sur les cellules souches après la naissance, même s’il n’existe pas de certitude sur le fait que ces cellules souches représentent des contingents cellulaires différents. - Cellules souches au cours de l’embryogenèse Seulement 2 études ont évalué le contingent de cellule souches hypophysaires au cours de l’embryogenèse. Fauquier et al. ont caractérisé les cellules exprimant Sox2, un membre de la famille des facteurs de transcription à homéodomaine de type HMG, requis pour la formation de cellules souches dans plusieurs tissus (55). Dans la poche de Rathke, les auteurs ont observé une expression précoce de Sox2 dès e11,5, autour de la lumière (zone marginale) et dans de petites zones réparties dans l’hypophyse. Ces cellules en exprimant progressivement Sox9 entre e12,5 et e18,5, entrent dans une phase de prolifération plus importante (cellules d’amplification transitoire). Le co- marquage Sox2 Sox9 n’est pas retrouvé dans les cellules différenciées. Enfin, les cellules Sox2+ sont capables de former des pituispheres, qui en culture, et après expression de Sox9 et de divers facteurs de transcription hypophysaires, vont se différencier pour donner naissance à tous les types cellulaires hypophysaires, en faveur de caractéristiques multipotentes. Les cellules exprimant Sox2 au cours de l’embryogenèse sont donc de parfaits candidats en tant que cellules souches hypophysaires, ou au moins en tant que progéniteurs (les auteurs ne sont pas parvenus ! %*! à obtenir plus de 2 passages de pituispheres, ce qui pourrait signifier une difficulté technique, ou un statut de progéniteur à capacité de réplication limitée plutôt que de réelles cellules souches) (55). L’expression de nestine a également été retrouvée à cette période d’embryogenèse (72). Plusieurs types de cellules souches impliqués dans la neurogenèse au cours de l’embryogenèse et à l’âge adulte ont une expression de nestine. Les auteurs ont utilisé un modèle de souris transgéniques où GFP était exprimé sous contrôle du promoteur et d’éléments régulateurs de nestine ; les résultats peuvent porter à discussion, essentiellement du fait d’une possibilité d’expression ectopique du transgène (177). De plus, une autre étude a montré que la nestine était également présente dans les progéniteurs des cellules endothéliales, rendant ce marqueur probablement insuffisamment spécifique pour tirer des conclusions formelles (69). Les auteurs ont observé une expression précoce de nestine, à e11,5 dans la poche de Rathke. Cependant, l’étude du devenir de ces cellules semble montrer qu’elles restent quiescentes pendant toute l’embryogenèse ; ces cellules entrent en réplication quelques jours après la naissance et pourraient jouer un rôle à l’âge adulte plutôt qu’en période embryonnaire. A l’heure actuelle, il semble donc qu’il existe une population de cellules souches hypophysaires embryonnaires, évoluant d’un profil de positivité pour Sox2, à une progressive différenciation, acquérant une positivité pour Sox9, puis exprimant certains des facteurs de transcription hypophysaire, pour se différencier en différents contingents sécrétoires hypophysaires. - Cellules souches à l’âge adulte Plusieurs études ont évalué d’éventuelles populations de cellules souches hypophysaires à l’âge adulte. Fauquier et al. ont montré que la majorité des cellules exprimant Sox2 à l’âge adulte exprimaient également Sox9 ; ces cellules n’expriment pas de marqueurs hormonaux. Il existe également une faible proportion de cellules exprimant uniquement Sox2. Il est envisageable que les cellules Sox2+ Sox9- soient des cellules souches quiescentes, alors que les cellules Sox2+ Sox9+, cellules d’amplification transitoire, sont déjà impliquées dans les 1eres phases de différenciation (55). Il faut noter que l’index de prolifération est particulièrement élevé après la naissance, ce qui pourrait correspondre à la prolifération avant la différenciation de cellules nécessaires à assurer la progressive complexité des fonctions hypophysaires. Les cellules exprimant la nestine connaissent également une phase de réplication majeure quelques jours après la naissance. Environ 50% de ces cellules expriment Sox2 (aucune évaluation de l’expression de Sox9 n’a été effectuée). Les 2 populations (celles décrites par Fauquier et Gleiberman) sont probablement identiques (72). Enfin, les 50% de cellules n’exprimant pas SOX2 sont vraisemblablement des progéniteurs endothéliaux. D’autres marqueurs ont également été évalués : - GFRa2 (GDNF Receptor alpha 2) est exprimé dans les cellules souches gonadiques. Son expression a été retrouvée au sein d’une population de cellules localisées dans la zone marginale (représentant 1% de l’ensemble de la population de cellules hypophysaires). Plus de 90% de ces cellules expriment Sox2 et Sox9. Elles expriment également Prop1, ce qui est en faveur d’un état prédifférencié, en accord avec l’expression de Sox9. Ces cellules sont capables de former des pituispheres, qui se différencieront en cellules des 6 lignées hypophysaires. GFRa2 doit donc être un autre marqueur de la population de cellules Sox2+ Sox9+ Nestine+ décrite précédemment (64). - Sca1 est un marqueur de cellules souches dans divers organes. Une équipe a identifié par cytomètrie de flux (FACS) une population de cellules hypophysaires (« side ! &+! population ») dont 40% des cellules ont des propriétés de cellules souches. Ces cellules qui ont une faible expression de Sca1, expriment Sox2 et Sox9 dans environ 50% des cas, ainsi que de nombreux facteurs de transcription hypophysaires. Les auteurs soulignent le caractère très hétérogène de cette population qui comprend vraisemblablement des cellules souches, et des progéniteurs à un stade plus ou moins différencié. Ces cellules sont capables de former des pituispheres en culture (32; 30; 31). Il est donc vraisemblable qu’il existe à l’âge adulte une population de progéniteurs ou cellules d’amplification transitoire, exprimant Sox2, Sox9, Nestine GFRa2 et Sca1, qui vont progressivement se différencier. Un autre contingent de cellules, en faible quantité, exprime seulement Sox2, et représente vraisemblablement des cellules souches, maintenues dans une phase quiescente. Extrapoler par rapport aux données de l’embryogenèse est délicat, car la plupart de ces marqueurs n’ont pas été évalués avant la naissance. b. Cellules souches comme éventuel traitement des hypopituitarismes congénitaux : des avantages et des inconvénients potentiels Utiliser des cellules souches comme traitement implique de savoir réduire leur différenciation, les greffer et contrôler leur prolifération - Différencier des cellules souches in vitro Trois études ont rapporté la possibilité d’obtenir des cellules hormonales hypophysaires différenciées à partir de cellules souches embryonnaires (211; 235; 222): U et al. ont obtenu des cellules sécrétant de la GH et de la prolactine à partir de cellules souches neurales de rat, mises en culture dans un milieu conditionné par des cellules de type GH3. Une équipe japonaise a obtenu des cellules à LHß et FSHß à partir de cellules souches embryonnaires de souris. Une autre étude basée sur des cellules souches embryonnaires murines de type D3 a observé leur différenciation en cellules hypophysaires somato-lactotropes et inconstamment thyréotropes. - Greffer des cellules souches in vivo Une greffe de cellules souches neurales de rat dans la région hypophysaire chez des rats (cellules marquées par la GFP) a permis d’obtenir environ 10% de cellules exprimant Pit1, puis GH et prolactine. Ces cellules ont survécu environ 4 semaines (212). Cependant, il est important de noter que les rats receveurs avaient une hypophyse normale (différente par définition des hypophyses hypo ou aplasiques des patients présentant un hypopituitarisme congénital). Ces cellules ont peut être pu se différencier parce qu’elles étaient dans un environnement hypophysaire normal. L’existence de réseaux de cellules sécrétrices (hormone de croissance, prolactine) pose également la question de la possibilité pour les cellules greffées de s’intégrer à ce réseau, ou en tout cas d’être fonctionnelles après greffe(13). Enfin, il n’existe pas de données sur la tolérance de l’organisme à la greffe de ces cellules. Le moyen d’obtenir des cellules souches immuno-compatibles ou la nécessité de donner des traitements immunosuppresseurs pourrait constituer de sérieuses difficultés à l’utilisation de cellules souches en pratique courante. Cet écueil pourrait être comblé par l’utilisation de cellules souches du patients ou l’utilisation de traitements immunosuppresseurs. L’avenir est probablement basé sur le développement de la technique d’induction de cellules souches multipotentes, ou comment dédifférencier in vitro des cellules de l’individu (par exemple des fibroblastes) en leur faisant exprimer des facteurs de transcription impliqués dans la différenciation de cellules souches en cellules hypophysaires, puis les greffer pour traiter un déficit (200; 201). - Contrôler la prolifération des cellules souches ! &"! Les résultats de certaines études ayant observé des cellules exprimant des marqueurs de cellules souches au sein d’adénomes hypophysaires pose la question du risque d’une prolifération anarchique et excessive de cellules issues de ces cellules souches (231). La question de l’intégration dans un réseau fonctionnel est à cet égard fondamentale. Il est logique d’imaginer qu’une cellule somatotrope non connectée au réseau de cellules à GH agisse de façon anarchique et puisse aboutir à un adénome somatotrope (même si de nombreux autres facteurs déclencheurs sont probablement nécessaires) (13). Les études futures devront également déterminer si les hyperplasies hypophysaires peuvent être dues à un excès de progéniteurs, donc de cellules non différenciées : les patients présentant initialement une hyperplasie hypophysaire (par exemple dans le cas de certaines mutations de PROP1 ) pourraient ne pas bénéficier d’une greffe de cellules souches non différenciées (si on considère qu’il existe une anomalie des signaux de différenciation expliquant le blocage du cycle cellulaire de ces progéniteurs). Il faudrait dans ce cas plutôt chercher comment activer les voies de différenciation, ou greffer des cellules déjà différenciées. ! ! &#! ! CONCLUSION ! &$! Le développement hypophysaire est un phénomène complexe nécessitant l’interaction entre des facteurs de transcription et des voies de signalisation multiples. La complexité réside aussi dans la régulation temporo-spatiale de l’expression de chacun de ces facteurs de transcription. Le phénomène de boucle de rétrocontrôle positif ou négatif, bien connu dans la physiologie et la pathologie endocrinienne à l’échelle hormonale est déjà présent à l’échelle protéique/moléculaire et joue un rôle important dans cette régulation: certains facteurs de transcription sont ainsi capables d’inhiber ou de stimuler le promoteur du gène codant pour le facteur de transcription qui les a activés. Rechercher les causes des hypopituitarismes congénitaux implique donc de façon directe, de rechercher les mutations de ces facteurs de transcription (comme pour la mutation T99fs de LHX4 ), mais aussi de façon plus indirecte de rechercher des altérations de cette boucle de régulation : dans cette optique, la mise en évidence d’une interaction des co-répresseurs de la famille Tle sur Prop1 (sans la présence de HESX1) doit faire évoquer la possibilité d’un phénotype de déficit hypophysaire lié à une anomalie de cette interaction (au niveau de PROP1 ou de TLE1/3), et à l’origine d’un phénotype semblable à celui observé pour les mutations de PROP1 . Ce mécanisme a déjà été mis en évidence pour un autre facteur de transcription hypophysaire à homéodomaine de type paired, HESX1 . L’augmentation constante du nombre de mutations identifiées dans les cas d’hypopituitarismes congénitaux ne doit pas masquer le faible pourcentage de causes identifiées (actuellement de l’ordre de 15%). Les raisons de ce faible score d’identification d’anomalies géniques en sont multiples, et il est difficile de déterminer si l’une prédomine par rapport aux autres : facteurs étiologiques non encore identifiés, sélection inappropriée des patients pour le séquençage, algorithme de séquençage basé sur un modèle (murin) qui n’est pas entièrement comparable au modèle humain, anomalies géniques non détectables par les techniques courantes (délétions, anomalies de promoteurs…). Il est vraisemblable que toutes ces raisons sont à mettre en cause dans le faible ratio étiologies identifiées/étiologies recherchées. Il semble difficile cependant d’améliorer sensiblement ce ratio, essentiellement du fait du coût du séquençage : idéalement, pour améliorer l’algorithme de séquençage, tous les facteurs de transcription identifiés devraient être séquencés pour chaque patient, sans tenir compte des phénotypes précédents, ce qui est impossible en pratique pour des raisons coût-efficacité. Pour augmenter le nombre d’étiologies identifiées, il faudra également identifier de nouveaux facteurs de transcription et/ou de nouveaux mécanismes. En se basant sur le modèle murin comme première étape, les différences avec le modèle humain ne sont pas prises en compte. Cependant, le modèle murin doit permettre de tracer de grandes lignes sur les implications prédominantes de certains facteurs ou voies de signalisation par rapport à d’autres. Dans cette optique, identifier précisément les éléments nécessaires au développement de chaque lignée hypophysaire est nécessaire : à ce jour, les mécanismes précis du développement de l’axe thyréotrope (ou du/des facteurs indispensables à son développement) restent imparfaitement élucidés. L’exemple de Pitx2 (avec la probable implication de facteurs redondants ou substitutifs, au premier rang desquels Pitx1) souligne la complexité des mécanismes (et en particulier des mécanismes de remplacement) qui vont permettre le développement d’une lignée indispensable à l’homéostasie du corps humain. La rareté des phénotypes hypophysaires dans les mutations de PITX2 chez l’homme montre clairement que pour certains facteurs de transcription, il existe des relais qui vont pouvoir remplacer le facteur déficient. Cette duplication fonctionnelle est moins observée pour d’autres facteurs (par exemple LHX4 et LHX3), ce qui aboutit à un phénotype hypophysaire déficitaire y compris pour des mutations hétérozygotes. L’exemple de l’hyperplasie ! &%! hypophysaire est symptomatique du fait que le moindre dérèglement dans la régulation temporo-spatiale de ces facteurs peut être à l’origine d’anomalie hypophysaire via des anomalies de migration, de prolifération, ou d’apoptose… Ces mécanismes sont complexes : à ce jour, nul ne sait pourquoi certains seulement certains patients avec mutations de PROP1, LHX3 ou LHX4 présentent une hyperplasie hypophysaire, ni quel en est le mécanisme… Identifier les mécanismes à l’origine des hypopituitarismes congénitaux doit secondairement bénéficier au patient au moins à 2 niveaux : - prévoir le phénotype pour l’individu et sa famille. Ce point est surtout applicable en cas d’apparition tardif de déficits. Une fois le gène identifié, un profil phénotypique classique peut être envisagé. Ainsi, la survenue d’un déficit corticotrope est en général retardée chez les patients porteurs de mutations de PROP1 . La mutation T99fs de LHX4 est à l’origine d’un déficit gonadotrope et somatotrope de survenue très retardée chez 1 patient porteur. Cela justifie une surveillance régulière de ces fonctions hypophysaires. - traiter au mieux le patient. Le traitement actuel est basé sur un remplacement des hormones périphériques, parfois difficile à adapter, souvent contraignant pour le patient, et ne permettant pas de reproduire un schéma physiologique de sécrétion basale et sa régulation. Identifier les facteurs en cause dans les hypopituitarismes congénitaux pourrait permettre (à plus long terme) de traiter la cause en « remplaçant » le gène défectueux par des approches de type thérapie génique. Plus probablement, identifier les facteurs de transcription en cause doit permettre d’améliorer la compréhension des mécanismes de différenciation nécessaires à l’obtention d’une cellule différenciée hypophysaire à partir d’une cellule souche multipotente. Le développement hypophysaire est donc effectivement un phénomène complexe nécessitant l’interaction entre des facteurs de transcription et des voies de signalisation. L’augmentation de nos connaissances sur le sujet devrait permettre de proposer un traitement adapté à chaque étiologie. Cela nécessite la poursuite de l’étude – irremplaçable – des modèles pathologiques humains spontanés, mais aussi de modèles murins ou d’autres modèles animaux. Comme nous l’avons vu, le chemin est encore long pour la mise au point de thérapies ciblées fondées sur le recours aux cellules souches. ! &&! REFERENCES BIBLIOGRAPHIQUES ! &'! 1. Abrao, M. G., Leite, M. V., Carvalho, L. R., Billerbeck, A. E., Nishi, M. Y., Barbosa, A. S., Martin, R. M., Arnhold, I. J. and Mendonca, B. B. (2006). "Combined pituitary hormone deficiency (CPHD) due to a complete PROP1 deletion." Clin Endocrinol (Oxf) 65(3): 294-300. 2. Acampora, D., Gulisano, M. and Simeone, A. (2000). "Genetic and molecular roles of Otx homeodomain proteins in head development." Gene 246 (1-2): 23-35. 3. Agarwal, G., Bhatia, V., Cook, S. and Thomas, P. Q. (2000). "Adrenocorticotropin deficiency in combined pituitary hormone deficiency patients homozygous for a novel PROP1 deletion." J Clin Endocrinol Metab 85 (12): 4556-61. 4. Ai, D., Wang, J., Amen, M., Lu, M. F., Amendt, B. A. and Martin, J. F. (2007). "Nuclear factor 1 and T-cell factor/LEF recognition elements regulate Pitx2 transcription in pituitary development." Mol Cell Biol 27 (16): 5765-75. 5. Amen, M., Liu, X., Vadlamudi, U., Elizondo, G., Diamond, E., Engelhardt, J. F. and Amendt, B. A. (2007). "PITX2 and beta-catenin interactions regulate Lef-1 isoform expression." Mol Cell Biol 27 (21): 7560-73. 6. Andersen, B. and Rosenfeld, M. G. (1994). "Pit-1 determines cell types during development of the anterior pituitary gland. A model for transcriptional regulation of cell phenotypes in mammalian organogenesis." J Biol Chem 269 (47): 29335-8. 7. Andoniadou, C. L., Signore, M., Sajedi, E., Gaston-Massuet, C., Kelberman, D., Burns, A. J., Itasaki, N., Dattani, M. and Martinez-Barbera, J. P. (2007). "Lack of the murine homeobox gene Hesx1 leads to a posterior transformation of the anterior forebrain." Development 134 (8): 1499-508. 8. Arroyo, A., Pernasetti, F., Vasilyev, V. V., Amato, P., Yen, S. S. and Mellon, P. L. (2002). "A unique case of combined pituitary hormone deficiency caused by a PROP1 gene mutation (R120C) associated with normal height and absent puberty." Clin Endocrinol (Oxf) 57 (2): 283-91. 9. Bartke, A. (1965). "Influence of Luteotrophin on Fertility of Dwarf Mice." J Reprod Fertil 10 : 93-103. 10. Bhangoo, A. P., Hunter, C. S., Savage, J. J., Anhalt, H., Pavlakis, S., Walvoord, E. C., Ten, S. and Rhodes, S. J. (2006). "Clinical case seminar: a novel LHX3 mutation presenting as combined pituitary hormonal deficiency." J Clin Endocrinol Metab 91 (3): 747-53. 11. Bhati, M., Lee, C., Nancarrow, A. L., Lee, M., Craig, V. J., Bach, I., Guss, J. M., Mackay, J. P. and Matthews, J. M. (2008). "Implementing the LIM code: the structural basis for cell type-specific assembly of LIM-homeodomain complexes." EMBO J 27 (14): 2018-29. 12. Bhati, M., Lee, M., Nancarrow, A. L., Bach, I., Guss, J. M. and Matthews, J. M. (2008). "Crystallization of an Lhx3-Isl1 complex." Acta Crystallogr Sect F Struct Biol Cryst Commun 64 (Pt 4): 297-9. 13. Bonnefont, X., Lacampagne, A., Sanchez-Hormigo, A., Fino, E., Creff, A., Mathieu, M. N., Smallwood, S., Carmignac, D., Fontanaud, P., Travo, P., Alonso, G., Courtois-Coutry, N., Pincus, S. M., Robinson, I. C. and Mollard, P. (2005). "Revealing the large-scale network organization of growth hormone-secreting cells." Proc Natl Acad Sci U S A 102 (46): 16880-5. 14. Bottner, A., Keller, E., Kratzsch, J., Stobbe, H., Weigel, J. F., Keller, A., Hirsch, W., Kiess, W., Blum, W. F. and Pfaffle, R. W. (2004). "PROP1 mutations cause progressive deterioration of anterior pituitary function including adrenal insufficiency: a longitudinal analysis." J Clin Endocrinol Metab 89 (10): 5256-65. 15. Brickman, J. M., Clements, M., Tyrell, R., McNay, D., Woods, K., Warner, J., Stewart, A., Beddington, R. S. and Dattani, M. (2001). "Molecular effects of novel ! &(! mutations in Hesx1/HESX1 associated with human pituitary disorders." Development 128 (24): 5189-99. 16. Bridwell, J. A., Price, J. R., Parker, G. E., McCutchan Schiller, A., Sloop, K. W. and Rhodes, S. J. (2001). "Role of the LIM domains in DNA recognition by the Lhx3 neuroendocrine transcription factor." Gene 277 (1-2): 239-50. 17. Brinkmeier, M. L., Davis, S. W., Carninci, P., MacDonald, J. W., Kawai, J., Ghosh, D., Hayashizaki, Y., Lyons, R. H. and Camper, S. A. (2009). "Discovery of transcriptional regulators and signaling pathways in the developing pituitary gland by bioinformatic and genomic approaches." Genomics 93 (5): 449-60. 18. Brinkmeier, M. L., Gordon, D. F., Dowding, J. M., Saunders, T. L., Kendall, S. K., Sarapura, V. D., Wood, W. M., Ridgway, E. C. and Camper, S. A. (1998). "Cell- specific expression of the mouse glycoprotein hormone alpha-subunit gene requires multiple interacting DNA elements in transgenic mice and cultured cells." Mol Endocrinol 12 (5): 622-33. 19. Brinkmeier, M. L., Potok, M. A., Cha, K. B., Gridley, T., Stifani, S., Meeldijk, J., Clevers, H. and Camper, S. A. (2003). "TCF and Groucho-related genes influence pituitary growth and development." Mol Endocrinol 17 (11): 2152-61. 20. Brinkmeier, M. L., Potok, M. A., Davis, S. W. and Camper, S. A. (2007). "TCF4 deficiency expands ventral diencephalon signaling and increases induction of pituitary progenitors." Dev Biol 311 (2): 396-407. 21. Brinkmeier, M. L., Stahl, J. H., Gordon, D. F., Ross, B. D., Sarapura, V. D., Dowding, J. M., Kendall, S. K., Lloyd, R. V., Ridgway, E. C. and Camper, S. A. (2001). "Thyroid hormone-responsive pituitary hyperplasia independent of somatostatin receptor 2." Mol Endocrinol 15 (12): 2129-36. 22. Brown, M. R., Parks, J. S., Adess, M. E., Rich, B. H., Rosenthal, I. M., Voss, T. C., VanderHeyden, T. C. and Hurley, D. L. (1998). "Central hypothyroidism reveals compound heterozygous mutations in the Pit-1 gene." Horm Res 49 (2): 98-102. 23. Carlomagno, Y., Salerno, M., Vivenza, D., Capalbo, D., Godi, M., Mellone, S., Tiradani, L., Corneli, G., Momigliano-Richiardi, P., Bona, G. and Giordano, M. (2009). "A novel recessive splicing mutation in the POU1F1 gene causing combined pituitary hormone deficiency." J Endocrinol Invest 32 (8): 653-8. 24. Carvalho, L. R., Woods, K. S., Mendonca, B. B., Marcal, N., Zamparini, A. L., Stifani, S., Brickman, J. M., Arnhold, I. J. and Dattani, M. T. (2003). "A homozygous mutation in HESX1 is associated with evolving hypopituitarism due to impaired repressor-corepressor interaction." J Clin Invest 112 (8): 1192-201. 25. Castinetti, F., Reynaud, R., Saveanu, A., Quentien, M. H., Albarel, F., Barlier, A., Enjalbert, A. and Brue, T. (2008). "[Clinical and genetic aspects of combined pituitary hormone deficiencies]." Ann Endocrinol (Paris) 69 (1): 7-17. 26. Castinetti, F., Saveanu, A., Reynaud, R., Quentien, M. H., Buffin, A., Brauner, R., Kaffel, N., Albarel, F., Guedj, A. M., El Kholy, M., Amin, M., Enjalbert, A., Barlier, A. and Brue, T. (2008). "A novel dysfunctional LHX4 mutation with high phenotypical variability in patients with hypopituitarism." J Clin Endocrinol Metab 93 (7): 2790-9. 27. Charles, M. A., Mortensen, A. H., Potok, M. A. and Camper, S. A. (2008). "Pitx2 deletion in pituitary gonadotropes is compatible with gonadal development, puberty, and fertility." Genesis 46 (10): 507-14. 28. Charles, M. A., Saunders, T. L., Wood, W. M., Owens, K., Parlow, A. F., Camper, S. A., Ridgway, E. C. and Gordon, D. F. (2006). "Pituitary-specific Gata2 knockout: effects on gonadotrope and thyrotrope function." Mol Endocrinol 20 (6): 1366-77. ! &)! 29. Charles, M. A., Suh, H., Hjalt, T. A., Drouin, J., Camper, S. A. and Gage, P. J. (2005). "PITX genes are required for cell survival and Lhx3 activation." Mol Endocrinol 19 (7): 1893-903. 30. Chen, J., Crabbe, A., Van Duppen, V. and Vankelecom, H. (2006). "The notch signaling system is present in the postnatal pituitary: marked expression and regulatory activity in the newly discovered side population." Mol Endocrinol 20 (12): 3293-307. 31. Chen, J., Gremeaux, L., Fu, Q., Liekens, D., Van Laere, S. and Vankelecom, H. (2009). "Pituitary progenitor cells tracked down by side population dissection." Stem Cells 27 (5): 1182-95. 32. Chen, J., Hersmus, N., Van Duppen, V., Caesens, P., Denef, C. and Vankelecom, H. (2005). "The adult pituitary contains a cell population displaying stem/progenitor cell and early embryonic characteristics." Endocrinology 146 (9): 3985-98. 33. Chou, S. J., Hermesz, E., Hatta, T., Feltner, D., El-Hodiri, H. M., Jamrich, M. and Mahon, K. (2006). "Conserved regulatory elements establish the dynamic expression of Rpx/HesxI in early vertebrate development." Dev Biol 292 (2): 533- 45. 34. Cogan, J. D., Wu, W., Phillips, J. A., 3rd, Arnhold, I. J., Agapito, A., Fofanova, O. V., Osorio, M. G., Bircan, I., Moreno, A. and Mendonca, B. B. (1998). "The PROP1 2-base pair deletion is a common cause of combined pituitary hormone deficiency." J Clin Endocrinol Metab 83 (9): 3346-9. 35. Cohen, L. E., Zanger, K., Brue, T., Wondisford, F. E. and Radovick, S. (1999). "Defective retinoic acid regulation of the Pit-1 gene enhancer: a novel mechanism of combined pituitary hormone deficiency." Mol Endocrinol 13 (3): 476-84. 36. Cohen, R. N., Cohen, L. E., Botero, D., Yu, C., Sagar, A., Jurkiewicz, M. and Radovick, S. (2003). "Enhanced repression by HESX1 as a cause of hypopituitarism and septooptic dysplasia." J Clin Endocrinol Metab 88 (10): 4832- 9. 37. Colvin, S. C., Mullen, R. D., Pfaeffle, R. W. and Rhodes, S. J. (2009). "LHX3 and LHX4 transcription factors in pituitary development and disease." Pediatr Endocrinol Rev 6 Suppl 2 : 283-90. 38. Corneli, G., Vivenza, D., Prodam, F., Di Dio, G., Vottero, A., Rapa, A., Bellone, S., Bernasconi, S. and Bona, G. (2008). "Heterozygous mutation of HESX1 causing hypopituitarism and multiple anatomical malformations without features of septo-optic dysplasia." J Endocrinol Invest 31 (8): 689-93. 39. Cox, C. J., Espinoza, H. M., McWilliams, B., Chappell, K., Morton, L., Hjalt, T. A., Semina, E. V. and Amendt, B. A. (2002). "Differential regulation of gene expression by PITX2 isoforms." J Biol Chem 277 (28): 25001-10. 40. Coya, R., Vela, A., Perez de Nanclares, G., Rica, I., Castano, L., Busturia, M. A. and Martul, P. (2007). "Panhypopituitarism: genetic versus acquired etiological factors." J Pediatr Endocrinol Metab 20 (1): 27-36. 41. Cushman, L. J., Watkins-Chow, D. E., Brinkmeier, M. L., Raetzman, L. T., Radak, A. L., Lloyd, R. V. and Camper, S. A. (2001). "Persistent Prop1 expression delays gonadotrope differentiation and enhances pituitary tumor susceptibility." Hum Mol Genet 10 (11): 1141-53. 42. Dasen, J. S., Barbera, J. P., Herman, T. S., Connell, S. O., Olson, L., Ju, B., Tollkuhn, J., Baek, S. H., Rose, D. W. and Rosenfeld, M. G. (2001). "Temporal regulation of a paired-like homeodomain repressor/TLE corepressor complex and ! &*! a related activator is required for pituitary organogenesis." Genes Dev 15 (23): 3193-207. 43. Dasen, J. S., O'Connell, S. M., Flynn, S. E., Treier, M., Gleiberman, A. S., Szeto, D. P., Hooshmand, F., Aggarwal, A. K. and Rosenfeld, M. G. (1999). "Reciprocal interactions of Pit1 and GATA2 mediate signaling gradient-induced determination of pituitary cell types." Cell 97 (5): 587-98. 44. Dateki, S., Fukami, M., Sato, N., Muroya, K., Adachi, M. and Ogata, T. (2008). "OTX2 mutation in a patient with anophthalmia, short stature, and partial growth hormone deficiency: functional studies using the IRBP, HESX1, and POU1F1 promoters." J Clin Endocrinol Metab 93 (10): 3697-702. 45. Dateki, S., Fukami, M., Uematsu, A., Kaji, M., Iso, M., Ono, M., Mizota, M., Yokoya, S., Motomura, K., Kinoshita, E., Moriuchi, H. and Ogata, T. "Mutation and gene copy number analyses of six pituitary transcription factor genes in 71 patients with combined pituitary hormone deficiency: identification of a single patient with LHX4 deletion." J Clin Endocrinol Metab 95 (8): 4043-7. 46. Dateki, S., Kosaka, K., Hasegawa, K., Tanaka, H., Azuma, N., Yokoya, S., Muroya, K., Adachi, M., Tajima, T., Motomura, K., Kinoshita, E., Moriuchi, H., Sato, N., Fukami, M. and Ogata, T. "Heterozygous orthodenticle homeobox 2 mutations are associated with variable pituitary phenotype." J Clin Endocrinol Metab 95 (2): 756-64. 47. Dattani, M. T., Martinez-Barbera, J. P., Thomas, P. Q., Brickman, J. M., Gupta, R., Martensson, I. L., Toresson, H., Fox, M., Wales, J. K., Hindmarsh, P. C., Krauss, S., Beddington, R. S. and Robinson, I. C. (1998). "Mutations in the homeobox gene HESX1/Hesx1 associated with septo-optic dysplasia in human and mouse." Nat Genet 19 (2): 125-33. 48. Davis, S. W., Castinetti, F., Carvalho, L. R., Ellsworth, B. S., Potok, M. A., Lyons, R. H., Brinkmeier, M. L., Raetzman, L. T., Carninci, P., Mortensen, A. H., Hayashizaki, Y., Arnhold, I. J., Mendonca, B. B., Brue, T. and Camper, S. A. (2009). "Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes." Mol Cell Endocrinol . 49. Davis, S. W., Potok, M. A., Brinkmeier, M. L., Carninci, P., Lyons, R. H., MacDonald, J. W., Fleming, M. T., Mortensen, A. H., Egashira, N., Ghosh, D., Steel, K. P., Osamura, R. Y., Hayashizaki, Y. and Camper, S. A. (2009). "Genetics, gene expression and bioinformatics of the pituitary gland." Horm Res 71 Suppl 2 : 101-15. 50. Deladoey, J., Fluck, C., Buyukgebiz, A., Kuhlmann, B. V., Eble, A., Hindmarsh, P. C., Wu, W. and Mullis, P. E. (1999). ""Hot spot" in the PROP1 gene responsible for combined pituitary hormone deficiency." J Clin Endocrinol Metab 84 (5): 1645-50. 51. Diaczok, D., Romero, C., Zunich, J., Marshall, I. and Radovick, S. (2008). "A novel dominant negative mutation of OTX2 associated with combined pituitary hormone deficiency." J Clin Endocrinol Metab 93 (11): 4351-9. 52. Drouin, J., Lamolet, B., Lamonerie, T., Lanctot, C. and Tremblay, J. J. (1998). "The PTX family of homeodomain transcription factors during pituitary developments." Mol Cell Endocrinol 140 (1-2): 31-6. 53. Duquesnoy, P., Roy, A., Dastot, F., Ghali, I., Teinturier, C., Netchine, I., Cacheux, V., Hafez, M., Salah, N., Chaussain, J. L., Goossens, M., Bougneres, P. and Amselem, S. (1998). "Human Prop-1: cloning, mapping, genomic structure. Mutations in familial combined pituitary hormone deficiency." FEBS Lett 437 (3): 216-20. ! '+! 54. Ellsworth, B. S., Butts, D. L. and Camper, S. A. (2008). "Mechanisms underlying pituitary hypoplasia and failed cell specification in Lhx3-deficient mice." Dev Biol 313 (1): 118-29. 55. Fauquier, T., Rizzoti, K., Dattani, M., Lovell-Badge, R. and Robinson, I. C. (2008). "SOX2-expressing progenitor cells generate all of the major cell types in the adult mouse pituitary gland." Proc Natl Acad Sci U S A 105 (8): 2907-12. 56. Feingold, M., Shiere, F., Fogels, H. R. and Donaldson, D. (1969). "Rieger's syndrome." Pediatrics 44 (4): 564-9. 57. Fluck, C., Deladoey, J., Rutishauser, K., Eble, A., Marti, U., Wu, W. and Mullis, P. E. (1998). "Phenotypic variability in familial combined pituitary hormone deficiency caused by a PROP1 gene mutation resulting in the substitution of Arg- ->Cys at codon 120 (R120C)." J Clin Endocrinol Metab 83 (10): 3727-34. 58. Fofanova, O., Takamura, N., Kinoshita, E., Parks, J. S., Brown, M. R., Peterkova, V. A., Evgrafov, O. V., Goncharov, N. P., Bulatov, A. A., Dedov, II and Yamashita, S. (1998). "Compound heterozygous deletion of the PROP-1 gene in children with combined pituitary hormone deficiency." J Clin Endocrinol Metab 83 (7): 2601-4. 59. Fofanova, O., Takamura, N., Kinoshita, E., Vorontsov, A., Vladimirova, V., Dedov, I., Peterkova, V. and Yamashita, S. (2000). "MR imaging of the pituitary gland in children and young adults with congenital combined pituitary hormone deficiency associated with PROP1 mutations." AJR Am J Roentgenol 174 (2): 555-9. 60. Fofanova, O. V., Takamura, N., Kinoshita, E., Yoshimoto, M., Tsuji, Y., Peterkova, V. A., Evgrafov, O. V., Dedov, II, Goncharov, N. P. and Yamashita, S. (1998). "Rarity of PIT1 involvement in children from Russia with combined pituitary hormone deficiency." Am J Med Genet 77 (5): 360-5. 61. Gage, P. J. and Camper, S. A. (1997). "Pituitary homeobox 2, a novel member of the bicoid-related family of homeobox genes, is a potential regulator of anterior structure formation." Hum Mol Genet 6(3): 457-64. 62. Gage, P. J., Qian, M., Wu, D. and Rosenberg, K. I. (2008). "The canonical Wnt signaling antagonist DKK2 is an essential effector of PITX2 function during normal eye development." Dev Biol 317 (1): 310-24. 63. Gage, P. J., Suh, H. and Camper, S. A. (1999). "Dosage requirement of Pitx2 for development of multiple organs." Development 126 (20): 4643-51. 64. Garcia-Lavandeira, M., Quereda, V., Flores, I., Saez, C., Diaz-Rodriguez, E., Japon, M. A., Ryan, A. K., Blasco, M. A., Dieguez, C., Malumbres, M. and Alvarez, C. V. (2009). "A GRFa2/Prop1/stem (GPS) cell niche in the pituitary." PLoS One 4(3): e4815. 65. Gaston-Massuet, C., Andoniadou, C. L., Signore, M., Sajedi, E., Bird, S., Turner, J. M. and Martinez-Barbera, J. P. (2008). "Genetic interaction between the homeobox transcription factors HESX1 and SIX3 is required for normal pituitary development." Dev Biol 324 (2): 322-33. 66. Gat-Yablonski, G., Klar, A., Hirsch, D., Eliakim, A., Lazar, L., Hurvitz, H. and Phillip, M. (2005). "Three novel mutations in POU1F1 in Israeli patients with combined pituitary hormone deficiency." J Pediatr Endocrinol Metab 18 (4): 385- 93. 67. Gat-Yablonski, G., Lazar, L., Bar, M., de Vries, L., Weintrob, N. and Phillip, M. (2009). "Growth without growth hormone and similar dysmorphic features in three patients with sporadic combined pituitary hormone deficiencies." Horm Res 72 (5): 302-9. ! '"! 68. Gat-Yablonski, G., Lazar, L., Pertzelan, A. and Phillip, M. (2002). "A novel mutation in PIT-1: phenotypic variability in familial combined pituitary hormone deficiencies." J Pediatr Endocrinol Metab 15 (3): 325-30. 69. Gautron, L., De-Smedt, V. and Laye, S. (2009). "Age-related changes in nestin immunoreactivity in the rat pituitary gland." Neuroendocrinology 90 (1): 19-30. 70. Genead, R., Danielsson, C., Andersson, A. B., Corbascio, M., Franco-Cereceda, A., Sylven, C. and Grinnemo, K. H. "Islet-1 cells are cardiac progenitors present during the entire lifespan: from the embryonic stage to adulthood." Stem Cells Dev . 71. Girardin, S. E., Benjannet, S., Barale, J. C., Chretien, M. and Seidah, N. G. (1998). "The LIM homeobox protein mLIM3/Lhx3 induces expression of the prolactin gene by a Pit-1/GHF-1-independent pathway in corticotroph AtT20 cells." FEBS Lett 431 (3): 333-8. 72. Gleiberman, A. S., Michurina, T., Encinas, J. M., Roig, J. L., Krasnov, P., Balordi, F., Fishell, G., Rosenfeld, M. G. and Enikolopov, G. (2008). "Genetic approaches identify adult pituitary stem cells." Proc Natl Acad Sci U S A 105 (17): 6332-7. 73. Gordon, D. F., Lewis, S. R., Haugen, B. R., James, R. A., McDermott, M. T., Wood, W. M. and Ridgway, E. C. (1997). "Pit-1 and GATA-2 interact and functionally cooperate to activate the thyrotropin beta-subunit promoter." J Biol Chem 272 (39): 24339-47. 74. Gordon, D. F., Woodmansee, W. W., Black, J. N., Dowding, J. M., Bendrick- Peart, J., Wood, W. M. and Ridgway, E. C. (2002). "Domains of Pit-1 required for transcriptional synergy with GATA-2 on the TSH beta gene." Mol Cell Endocrinol 196 (1-2): 53-66. 75. Granger, A., Bleux, C., Kottler, M. L., Rhodes, S. J., Counis, R. and Laverriere, J. N. (2006). "The LIM-homeodomain proteins Isl-1 and Lhx3 act with steroidogenic factor 1 to enhance gonadotrope-specific activity of the gonadotropin-releasing hormone receptor gene promoter." Mol Endocrinol 20 (9): 2093-108. 76. Halasz, Z., Toke, J., Patocs, A., Bertalan, R., Tombol, Z., Sallai, A., Hosszu, E., Muzsnai, A., Kovacs, L., Solyom, J., Fekete, G. and Racz, K. (2006). "High prevalence of PROP1 gene mutations in Hungarian patients with childhood-onset combined anterior pituitary hormone deficiency." Endocrine 30 (3): 255-60. 77. Hashimoto, Y., Cisternino, M. and Cohen, L. E. (2003). "A novel nonsense mutation in the Pit-1 gene: evidence for a gene dosage effect." J Clin Endocrinol Metab 88 (3): 1241-7. 78. Henderson, R. H., Williamson, K. A., Kennedy, J. S., Webster, A. R., Holder, G. E., Robson, A. G., FitzPatrick, D. R., van Heyningen, V. and Moore, A. T. (2009). "A rare de novo nonsense mutation in OTX2 causes early onset retinal dystrophy and pituitary dysfunction." Mol Vis 15 : 2442-7. 79. Hendriks-Stegeman, B. I., Augustijn, K. D., Bakker, B., Holthuizen, P., van der Vliet, P. C. and Jansen, M. (2001). "Combined pituitary hormone deficiency caused by compound heterozygosity for two novel mutations in the POU domain of the Pit1/POU1F1 gene." J Clin Endocrinol Metab 86 (4): 1545-50. 80. Hertzano, R., Dror, A. A., Montcouquiol, M., Ahmed, Z. M., Ellsworth, B., Camper, S., Friedman, T. B., Kelley, M. W. and Avraham, K. B. (2007). "Lhx3, a LIM domain transcription factor, is regulated by Pou4f3 in the auditory but not in the vestibular system." Eur J Neurosci 25 (4): 999-1005. 81. Himes, A. D. and Raetzman, L. T. (2009). "Premature differentiation and aberrant movement of pituitary cells lacking both Hes1 and Prop1." Dev Biol 325 (1): 151- 61. ! '#! 82. Hjalt, T. A., Semina, E. V., Amendt, B. A. and Murray, J. C. (2000). "The Pitx2 protein in mouse development." Dev Dyn 218 (1): 195-200. 83. Howard, P. W. and Maurer, R. A. (2001). "A point mutation in the LIM domain of Lhx3 reduces activation of the glycoprotein hormone alpha-subunit promoter." J Biol Chem 276 (22): 19020-6. 84. Huang, M., Sage, C., Li, H., Xiang, M., Heller, S. and Chen, Z. Y. (2008). "Diverse expression patterns of LIM-homeodomain transcription factors (LIM- HDs) in mammalian inner ear development." Dev Dyn 237 (11): 3305-12. 85. Huelsken, J. and Behrens, J. (2002). "The Wnt signalling pathway." J Cell Sci 115 (Pt 21): 3977-8. 86. Huelsken, J. and Birchmeier, W. (2001). "New aspects of Wnt signaling pathways in higher vertebrates." Curr Opin Genet Dev 11 (5): 547-53. 87. Hume, C. R., Bratt, D. L. and Oesterle, E. C. (2007). "Expression of LHX3 and SOX2 during mouse inner ear development." Gene Expr Patterns 7(7): 798-807. 88. Ikeshita, N., Kawagishi, M., Shibahara, H., Toda, K., Yamashita, T., Yamamoto, D., Sugiyama, Y., Iguchi, G., Iida, K., Takahashi, Y., Kaji, H., Chihara, K. and Okimura, Y. (2008). "Identification and analysis of prophet of Pit-1-binding sites in human Pit-1 gene." Endocrinology 149 (11): 5491-9. 89. Irie, Y., Tatsumi, K., Ogawa, M., Kamijo, T., Preeyasombat, C., Suprasongsin, C. and Amino, N. (1995). "A novel E250X mutation of the PIT1 gene in a patient with combined pituitary hormone deficiency." Endocr J 42 (3): 351-4. 90. Japon, M. A., Rubinstein, M. and Low, M. J. (1994). "In situ hybridization analysis of anterior pituitary hormone gene expression during fetal mouse development." J Histochem Cytochem 42 (8): 1117-25. 91. Jennings, B. H. and Ish-Horowicz, D. (2008). "The Groucho/TLE/Grg family of transcriptional co-repressors." Genome Biol 9(1): 205. 92. Kato, Y., Kimoto, F., Susa, T., Nakayama, M., Ishikawa, A. and Kato, T. "Pituitary homeodomain transcription factors HESX1 and PROP1 form a heterodimer on the inverted TAAT motif." Mol Cell Endocrinol 315 (1-2): 168-73. 93. Kelberman, D. and Dattani, M. T. (2007). "Genetics of septo-optic dysplasia." Pituitary 10 (4): 393-407. 94. Kelberman, D. and Dattani, M. T. (2008). "Septo-optic dysplasia - novel insights into the aetiology." Horm Res 69 (5): 257-65. 95. Kelberman, D., Rizzoti, K., Lovell-Badge, R., Robinson, I. C. and Dattani, M. T. (2009). "Genetic regulation of pituitary gland development in human and mouse." Endocr Rev 30 (7): 790-829. 96. Kelberman, D., Turton, J. P., Woods, K. S., Mehta, A., Al-Khawari, M., Greening, J., Swift, P. G., Otonkoski, T., Rhodes, S. J. and Dattani, M. T. (2009). "Molecular analysis of novel PROP1 mutations associated with combined pituitary hormone deficiency (CPHD)." Clin Endocrinol (Oxf) 70 (1): 96-103. 97. Kendall, S. K., Samuelson, L. C., Saunders, T. L., Wood, R. I. and Camper, S. A. (1995). "Targeted disruption of the pituitary glycoprotein hormone alpha-subunit produces hypogonadal and hypothyroid mice." Genes Dev 9(16): 2007-19. 98. Kerr, J. M., Gordon, D. F., Woodmansee, W. W., Sarapura, V. D., Ridgway, E. C. and Wood, W. M. (2005). "Growth arrest of thyrotropic tumors by thyroid hormone is correlated with novel changes in Wnt-10A." Mol Cell Endocrinol 238 (1-2): 57-67. 99. Kioussi, C., Briata, P., Baek, S. H., Rose, D. W., Hamblet, N. S., Herman, T., Ohgi, K. A., Lin, C., Gleiberman, A., Wang, J., Brault, V., Ruiz-Lozano, P., Nguyen, H. D., Kemler, R., Glass, C. K., Wynshaw-Boris, A. and Rosenfeld, M. ! '$! G. (2002). "Identification of a Wnt/Dvl/beta-Catenin --> Pitx2 pathway mediating cell-type-specific proliferation during development." Cell 111 (5): 673-85. 100. Kita, A., Imayoshi, I., Hojo, M., Kitagawa, M., Kokubu, H., Ohsawa, R., Ohtsuka, T., Kageyama, R. and Hashimoto, N. (2007). "Hes1 and Hes5 control the progenitor pool, intermediate lobe specification, and posterior lobe formation in the pituitary development." Mol Endocrinol 21 (6): 1458-66. 101. Kristrom, B., Zdunek, A. M., Rydh, A., Jonsson, H., Sehlin, P. and Escher, S. A. (2009). "A novel mutation in the LIM homeobox 3 gene is responsible for combined pituitary hormone deficiency, hearing impairment, and vertebral malformations." J Clin Endocrinol Metab 94 (4): 1154-61. 102. Lamba, P., Hjalt, T. A. and Bernard, D. J. (2008). "Novel forms of Paired-like homeodomain transcription factor 2 (PITX2): generation by alternative translation initiation and mRNA splicing." BMC Mol Biol 9: 31. 103. Lamesch, C., Neumann, S., Pfaffle, R., Kiess, W. and Paschke, R. (2002). "Adrenocorticotrope deficiency with clinical evidence for late onset in combined pituitary hormone deficiency caused by a homozygous 301-302delAG mutation of the PROP1 gene." Pituitary 5(3): 163-8. 104. Lamolet, B., Pulichino, A. M., Lamonerie, T., Gauthier, Y., Brue, T., Enjalbert, A. and Drouin, J. (2001). "A pituitary cell-restricted T box factor, Tpit, activates POMC transcription in cooperation with Pitx homeoproteins." Cell 104 (6): 849-59. 105. Lamonerie, T., Tremblay, J. J., Lanctot, C., Therrien, M., Gauthier, Y. and Drouin, J. (1996). "Ptx1, a bicoid-related homeo box transcription factor involved in transcription of the pro-opiomelanocortin gene." Genes Dev 10 (10): 1284-95. 106. Lanctot, C., Gauthier, Y. and Drouin, J. (1999). "Pituitary homeobox 1 (Ptx1) is differentially expressed during pituitary development." Endocrinology 140 (3): 1416-22. 107. Lanctot, C., Lamolet, B. and Drouin, J. (1997). "The bicoid-related homeoprotein Ptx1 defines the most anterior domain of the embryo and differentiates posterior from anterior lateral mesoderm." Development 124 (14): 2807-17. 108. Larder, R. and Mellon, P. L. (2009). "Otx2 induction of the gonadotropin-releasing hormone promoter is modulated by direct interactions with Grg co-repressors." J Biol Chem 284 (25): 16966-78. 109. Lebl, J., Vosahlo, J., Pfaeffle, R. W., Stobbe, H., Cerna, J., Novotna, D., Zapletalova, J., Kalvachova, B., Hana, V., Weiss, V. and Blum, W. F. (2005). "Auxological and endocrine phenotype in a population-based cohort of patients with PROP1 gene defects." Eur J Endocrinol 153 (3): 389-96. 110. Lemos, M. C., Gomes, L., Bastos, M., Leite, V., Limbert, E., Carvalho, D., Bacelar, C., Monteiro, M., Fonseca, F., Agapito, A., Castro, J. J., Regateiro, F. J. and Carvalheiro, M. (2006). "PROP1 gene analysis in Portuguese patients with combined pituitary hormone deficiency." Clin Endocrinol (Oxf) 65 (4): 479-85. 111. Li, S., Crenshaw, E. B., 3rd, Rawson, E. J., Simmons, D. M., Swanson, L. W. and Rosenfeld, M. G. (1990). "Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1." Nature 347 (6293): 528-33. 112. Lin, C. R., Kioussi, C., O'Connell, S., Briata, P., Szeto, D., Liu, F., Izpisua- Belmonte, J. C. and Rosenfeld, M. G. (1999). "Pitx2 regulates lung asymmetry, cardiac positioning and pituitary and tooth morphogenesis." Nature 401 (6750): 279-82. 113. Lin, L., Cui, L., Zhou, W., Dufort, D., Zhang, X., Cai, C. L., Bu, L., Yang, L., Martin, J., Kemler, R., Rosenfeld, M. G., Chen, J. and Evans, S. M. (2007). "Beta-catenin directly regulates Islet1 expression in cardiovascular progenitors ! '%! and is required for multiple aspects of cardiogenesis." Proc Natl Acad Sci U S A 104 (22): 9313-8. 114. Lin, S. C., Li, S., Drolet, D. W. and Rosenfeld, M. G. (1994). "Pituitary ontogeny of the Snell dwarf mouse reveals Pit-1-independent and Pit-1-dependent origins of the thyrotrope." Development 120 (3): 515-22. 115. Liu, S., Luo, H., Liu, J., McNeilly, A. S. and Cui, S. (2008). "Specificity protein 1 (Sp1) plays role in regulating LIM homeodomain transcription factor Lhx4 gene expression." Biochem Biophys Res Commun 366 (1): 36-41. 116. Machinis, K. and Amselem, S. (2005). "Functional relationship between LHX4 and POU1F1 in light of the LHX4 mutation identified in patients with pituitary defects." J Clin Endocrinol Metab 90 (9): 5456-62. 117. Machinis, K., Pantel, J., Netchine, I., Leger, J., Camand, O. J., Sobrier, M. L., Dastot-Le Moal, F., Duquesnoy, P., Abitbol, M., Czernichow, P. and Amselem, S. (2001). "Syndromic short stature in patients with a germline mutation in the LIM homeobox LHX4." Am J Hum Genet 69 (5): 961-8. 118. Malvagia, S., Poggi, G. M., Pasquini, E., Donati, M. A., Pela, I., Morrone, A. and Zammarchi, E. (2003). "The de novo Q167K mutation in the POU1F1 gene leads to combined pituitary hormone deficiency in an Italian patient." Pediatr Res 54 (5): 635-40. 119. Mammi, I., De Giorgio, P., Clementi, M. and Tenconi, R. (1998). "Cardiovascular anomaly in Rieger Syndrome: heterogeneity or contiguity?" Acta Ophthalmol Scand 76 (4): 509-12. 120. Mantovani, G., Asteria, C., Pellegrini, C., Bosari, S., Alberti, L., Bondioni, S., Peverelli, E., Spada, A. and Beck-Peccoz, P. (2006). "HESX1 expression in human normal pituitaries and pituitary adenomas." Mol Cell Endocrinol 247 (1-2): 135-9. 121. Martinez-Barbera, J. P., Rodriguez, T. A. and Beddington, R. S. (2000). "The homeobox gene Hesx1 is required in the anterior neural ectoderm for normal forebrain formation." Dev Biol 223 (2): 422-30. 122. Matthews, J. M., Bhati, M., Craig, V. J., Deane, J. E., Jeffries, C., Lee, C., Nancarrow, A. L., Ryan, D. P. and Sunde, M. (2008). "Competition between LIM- binding domains." Biochem Soc Trans 36 (Pt 6): 1393-7. 123. McGillivray, S. M., Bailey, J. S., Ramezani, R., Kirkwood, B. J. and Mellon, P. L. (2005). "Mouse GnRH receptor gene expression is mediated by the LHX3 homeodomain protein." Endocrinology 146 (5): 2180-5. 124. McLennan, K., Jeske, Y., Cotterill, A., Cowley, D., Penfold, J., Jones, T., Howard, N., Thomsett, M. and Choong, C. (2003). "Combined pituitary hormone deficiency in Australian children: clinical and genetic correlates." Clin Endocrinol (Oxf) 58 (6): 785-94. 125. McNay, D. E., Turton, J. P., Kelberman, D., Woods, K. S., Brauner, R., Papadimitriou, A., Keller, E., Keller, A., Haufs, N., Krude, H., Shalet, S. M. and Dattani, M. T. (2007). "HESX1 mutations are an uncommon cause of septooptic dysplasia and hypopituitarism." J Clin Endocrinol Metab 92 (2): 691-7. 126. Melo, M. E., Marui, S., Carvalho, L. R., Arnhold, I. J., Leite, C. C., Mendonca, B. B. and Knoepfelmacher, M. (2007). "Hormonal, pituitary magnetic resonance, LHX4 and HESX1 evaluation in patients with hypopituitarism and ectopic posterior pituitary lobe." Clin Endocrinol (Oxf) 66 (1): 95-102. 127. Mendonca, B. B., Osorio, M. G., Latronico, A. C., Estefan, V., Lo, L. S. and Arnhold, I. J. (1999). "Longitudinal hormonal and pituitary imaging changes in two females with combined pituitary hormone deficiency due to deletion of A301,G302 in the PROP1 gene." J Clin Endocrinol Metab 84 (3): 942-5. ! '&! 128. Meyer-Marcotty, P., Weisschuh, N., Dressler, P., Hartmann, J. and Stellzig- Eisenhauer, A. (2008). "Morphology of the sella turcica in Axenfeld-Rieger syndrome with PITX2 mutation." J Oral Pathol Med 37 (8): 504-10. 129. Mitchell, L. A., Thomas, P. Q., Zacharin, M. R. and Scheffer, I. E. (2002). "Ectopic posterior pituitary lobe and periventricular heterotopia: cerebral malformations with the same underlying mechanism?" AJNR Am J Neuroradiol 23 (9): 1475-81. 130. Mu, X., Fu, X., Beremand, P. D., Thomas, T. L. and Klein, W. H. (2008). "Gene regulation logic in retinal ganglion cell development: Isl1 defines a critical branch distinct from but overlapping with Pou4f2." Proc Natl Acad Sci U S A 105 (19): 6942-7. 131. Mullen, R. D., Colvin, S. C., Hunter, C. S., Savage, J. J., Walvoord, E. C., Bhangoo, A. P., Ten, S., Weigel, J., Pfaffle, R. W. and Rhodes, S. J. (2007). "Roles of the LHX3 and LHX4 LIM-homeodomain factors in pituitary development." Mol Cell Endocrinol 265-266 : 190-5. 132. Murry, C. E. and Keller, G. (2008). "Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development." Cell 132 (4): 661-80. 133. Nakayama, M., Kato, T., Susa, T., Sano, A., Kitahara, K. and Kato, Y. (2009). "Dimeric PROP1 binding to diverse palindromic TAAT sequences promotes its transcriptional activity." Mol Cell Endocrinol 307 (1-2): 36-42. 134. Nasonkin, I. O., Ward, R. D., Raetzman, L. T., Seasholtz, A. F., Saunders, T. L., Gillespie, P. J. and Camper, S. A. (2004). "Pituitary hypoplasia and respiratory distress syndrome in Prop1 knockout mice." Hum Mol Genet 13 (22): 2727-35. 135. Netchine, I., Sobrier, M. L., Krude, H., Schnabel, D., Maghnie, M., Marcos, E., Duriez, B., Cacheux, V., Moers, A., Goossens, M., Gruters, A. and Amselem, S. (2000). "Mutations in LHX3 result in a new syndrome revealed by combined pituitary hormone deficiency." Nat Genet 25 (2): 182-6. 136. Ng, S. B., Buckingham, K. J., Lee, C., Bigham, A. W., Tabor, H. K., Dent, K. M., Huff, C. D., Shannon, P. T., Jabs, E. W., Nickerson, D. A., Shendure, J. and Bamshad, M. J. "Exome sequencing identifies the cause of a mendelian disorder." Nat Genet 42 (1): 30-5. 137. Ng, S. B., Turner, E. H., Robertson, P. D., Flygare, S. D., Bigham, A. W., Lee, C., Shaffer, T., Wong, M., Bhattacharjee, A., Eichler, E. E., Bamshad, M., Nickerson, D. A. and Shendure, J. (2009). "Targeted capture and massively parallel sequencing of 12 human exomes." Nature 461 (7261): 272-6. 138. Nose, O., Tatsumi, K., Nakano, Y. and Amino, N. (2006). "Congenital combined pituitary hormone deficiency attributable to a novel PROP1 mutation (467insT)." J Pediatr Endocrinol Metab 19 (4): 491-8. 139. Ohta, K., Nobukuni, Y., Mitsubuchi, H., Fujimoto, S., Matsuo, N., Inagaki, H., Endo, F. and Matsuda, I. (1992). "Mutations in the Pit-1 gene in children with combined pituitary hormone deficiency." Biochem Biophys Res Commun 189 (2): 851-5. 140. Olson, L. E., Tollkuhn, J., Scafoglio, C., Krones, A., Zhang, J., Ohgi, K. A., Wu, W., Taketo, M. M., Kemler, R., Grosschedl, R., Rose, D., Li, X. and Rosenfeld, M. G. (2006). "Homeodomain-mediated beta-catenin-dependent switching events dictate cell-lineage determination." Cell 125 (3): 593-605. 141. Osorio, M. G., Kopp, P., Marui, S., Latronico, A. C., Mendonca, B. B. and Arnhold, I. J. (2000). "Combined pituitary hormone deficiency caused by a novel mutation of a highly conserved residue (F88S) in the homeodomain of PROP-1." J Clin Endocrinol Metab 85 (8): 2779-85. ! ''! 142. Ostendorff, H. P., Tursun, B., Cornils, K., Schluter, A., Drung, A., Gungor, C. and Bach, I. (2006). "Dynamic expression of LIM cofactors in the developing mouse neural tube." Dev Dyn 235 (3): 786-91. 143. Pan, L., Deng, M., Xie, X. and Gan, L. (2008). "ISL1 and BRN3B co-regulate the differentiation of murine retinal ganglion cells." Development 135 (11): 1981-90. 144. Paracchini, R., Giordano, M., Corrias, A., Mellone, S., Matarazzo, P., Bellone, J., Momigliano-Richiardi, P. and Bona, G. (2003). "Two new PROP1 gene mutations responsible for compound pituitary hormone deficiency." Clin Genet 64 (2): 142-7. 145. Parker, G. E., West, B. E., Witzmann, F. A. and Rhodes, S. J. (2005). "Serine/threonine/tyrosine phosphorylation of the LHX3 LIM-homeodomain transcription factor." J Cell Biochem 94 (1): 67-80. 146. Pellegrini-Bouiller, I., Belicar, P., Barlier, A., Gunz, G., Charvet, J. P., Jaquet, P., Brue, T., Vialettes, B. and Enjalbert, A. (1996). "A new mutation of the gene encoding the transcription factor Pit-1 is responsible for combined pituitary hormone deficiency." J Clin Endocrinol Metab 81 (8): 2790-6. 147. Pernasetti, F., Milner, R. D., al Ashwal, A. A., de Zegher, F., Chavez, V. M., Muller, M. and Martial, J. A. (1998). "Pro239Ser: a novel recessive mutation of the Pit-1 gene in seven Middle Eastern children with growth hormone, prolactin, and thyrotropin deficiency." J Clin Endocrinol Metab 83 (6): 2079-83. 148. Pernasetti, F., Toledo, S. P., Vasilyev, V. V., Hayashida, C. Y., Cogan, J. D., Ferrari, C., Lourenco, D. M., Jr. and Mellon, P. L. (2000). "Impaired adrenocorticotropin-adrenal axis in combined pituitary hormone deficiency caused by a two-base pair deletion (301-302delAG) in the prophet of Pit-1 gene." J Clin Endocrinol Metab 85 (1): 390-7. 149. Pfaeffle, R. W., Hunter, C. S., Savage, J. J., Duran-Prado, M., Mullen, R. D., Neeb, Z. P., Eiholzer, U., Hesse, V., Haddad, N. G., Stobbe, H. M., Blum, W. F., Weigel, J. F. and Rhodes, S. J. (2008). "Three novel missense mutations within the LHX4 gene are associated with variable pituitary hormone deficiencies." J Clin Endocrinol Metab 93 (3): 1062-71. 150. Pfaeffle, R. W., Savage, J. J., Hunter, C. S., Palme, C., Ahlmann, M., Kumar, P., Bellone, J., Schoenau, E., Korsch, E., Bramswig, J. H., Stobbe, H. M., Blum, W. F. and Rhodes, S. J. (2007). "Four novel mutations of the LHX3 gene cause combined pituitary hormone deficiencies with or without limited neck rotation." J Clin Endocrinol Metab 92 (5): 1909-19. 151. Pogoda, H. M. and Hammerschmidt, M. (2009). "How to make a teleost adenohypophysis: molecular pathways of pituitary development in zebrafish." Mol Cell Endocrinol 312 (1-2): 2-13. 152. Potok, M. A., Cha, K. B., Hunt, A., Brinkmeier, M. L., Leitges, M., Kispert, A. and Camper, S. A. (2008). "WNT signaling affects gene expression in the ventral diencephalon and pituitary gland growth." Dev Dyn 237 (4): 1006-20. 153. Quentien, M. H., Manfroid, I., Moncet, D., Gunz, G., Muller, M., Grino, M., Enjalbert, A. and Pellegrini, I. (2002). "Pitx factors are involved in basal and hormone-regulated activity of the human prolactin promoter." J Biol Chem 277 (46): 44408-16. 154. Quentien, M. H., Pitoia, F., Gunz, G., Guillet, M. P., Enjalbert, A. and Pellegrini, I. (2002). "Regulation of prolactin, GH, and Pit-1 gene expression in anterior pituitary by Pitx2: An approach using Pitx2 mutants." Endocrinology 143 (8): 2839-51. 155. Quirk, J. and Brown, P. (2002). "Hesx1 homeodomain protein represses transcription as a monomer and antagonises transactivation of specific sites as a homodimer." J Mol Endocrinol 28 (3): 193-205. ! '(! 156. Radovick, S., Nations, M., Du, Y., Berg, L. A., Weintraub, B. D. and Wondisford, F. E. (1992). "A mutation in the POU-homeodomain of Pit-1 responsible for combined pituitary hormone deficiency." Science 257 (5073): 1115-8. 157. Raetzman, L. T., Ross, S. A., Cook, S., Dunwoodie, S. L., Camper, S. A. and Thomas, P. Q. (2004). "Developmental regulation of Notch signaling genes in the embryonic pituitary: Prop1 deficiency affects Notch2 expression." Dev Biol 265 (2): 329-40. 158. Raetzman, L. T., Ward, R. and Camper, S. A. (2002). "Lhx4 and Prop1 are required for cell survival and expansion of the pituitary primordia." Development 129 (18): 4229-39. 159. Rainbow, L. A., Rees, S. A., Shaikh, M. G., Shaw, N. J., Cole, T., Barrett, T. G. and Kirk, J. M. (2005). "Mutation analysis of POUF-1, PROP-1 and HESX-1 show low frequency of mutations in children with sporadic forms of combined pituitary hormone deficiency and septo-optic dysplasia." Clin Endocrinol (Oxf) 62 (2): 163- 8. 160. Rajab, A., Kelberman, D., de Castro, S. C., Biebermann, H., Shaikh, H., Pearce, K., Hall, C. M., Shaikh, G., Gerrelli, D., Grueters, A., Krude, H. and Dattani, M. T. (2008). "Novel mutations in LHX3 are associated with hypopituitarism and sensorineural hearing loss." Hum Mol Genet 17 (14): 2150-9. 161. Raskin, S., Cogan, J. D., Summar, M. L., Moreno, A., Krishnamani, M. R. and Phillips, J. A., 3rd (1996). "Genetic mapping of the human pituitary-specific transcriptional factor gene and its analysis in familial panhypopituitary dwarfism." Hum Genet 98 (6): 703-5. 162. Reynaud, R., Barlier, A., Vallette-Kasic, S., Saveanu, A., Guillet, M. P., Simonin, G., Enjalbert, A., Valensi, P. and Brue, T. (2005). "An uncommon phenotype with familial central hypogonadism caused by a novel PROP1 gene mutant truncated in the transactivation domain." J Clin Endocrinol Metab 90 (8): 4880-7. 163. Reynaud, R., Chadli-Chaieb, M., Vallette-Kasic, S., Barlier, A., Sarles, J., Pellegrini-Bouiller, I., Enjalbert, A., Chaieb, L. and Brue, T. (2004). "A familial form of congenital hypopituitarism due to a PROP1 mutation in a large kindred: phenotypic and in vitro functional studies." J Clin Endocrinol Metab 89 (11): 5779- 86. 164. Reynaud, R., Gueydan, M., Saveanu, A., Vallette-Kasic, S., Enjalbert, A., Brue, T. and Barlier, A. (2006). "Genetic screening of combined pituitary hormone deficiency: experience in 195 patients." J Clin Endocrinol Metab 91 (9): 3329-36. 165. Rhodes, S. J., Chen, R., DiMattia, G. E., Scully, K. M., Kalla, K. A., Lin, S. C., Yu, V. C. and Rosenfeld, M. G. (1993). "A tissue-specific enhancer confers Pit-1- dependent morphogen inducibility and autoregulation on the pit-1 gene." Genes Dev 7(6): 913-32. 166. Riepe, F. G., Partsch, C. J., Blankenstein, O., Monig, H., Pfaffle, R. W. and Sippell, W. G. (2001). "Longitudinal imaging reveals pituitary enlargement preceding hypoplasia in two brothers with combined pituitary hormone deficiency attributable to PROP1 mutation." J Clin Endocrinol Metab 86 (9): 4353-7. 167. Rizzoti, K. and Lovell-Badge, R. (2005). "Early development of the pituitary gland: induction and shaping of Rathke's pouch." Rev Endocr Metab Disord 6(3): 161-72. 168. Rolletschek, A., Schroeder, I. S., Schulz, H., Hummel, O., Huebner, N. and Wobus, A. M. "Characterization of mouse embryonic stem cell differentiation into the pancreatic lineage in vitro by transcriptional profiling, quantitative RT-PCR and immunocytochemistry." Int J Dev Biol 54(1): 41-54. ! ')! 169. Rosenbloom, A. L., Almonte, A. S., Brown, M. R., Fisher, D. A., Baumbach, L. and Parks, J. S. (1999). "Clinical and biochemical phenotype of familial anterior hypopituitarism from mutation of the PROP1 gene." J Clin Endocrinol Metab 84 (1): 50-7. 170. Rosenfeld, M. G., Briata, P., Dasen, J., Gleiberman, A. S., Kioussi, C., Lin, C., O'Connell, S. M., Ryan, A., Szeto, D. P. and Treier, M. (2000). "Multistep signaling and transcriptional requirements for pituitary organogenesis in vivo." Recent Prog Horm Res 55 : 1-13; discussion -4. 171. Sadeghi-Nejad, A. and Senior, B. (1974). "A familial syndrome of isolated "aplasia" of the anterior pituitary. Diagnostic studies and treatment in the neonatal period." J Pediatr 84 (1): 79-84. 172. Sajedi, E., Gaston-Massuet, C., Andoniadou, C. L., Signore, M., Hurd, P. J., Dattani, M. and Martinez-Barbera, J. P. (2008). "DNMT1 interacts with the developmental transcriptional repressor HESX1." Biochim Biophys Acta 1783 (1): 131-43. 173. Sajedi, E., Gaston-Massuet, C., Signore, M., Andoniadou, C. L., Kelberman, D., Castro, S., Etchevers, H. C., Gerrelli, D., Dattani, M. T. and Martinez-Barbera, J. P. (2008). "Analysis of mouse models carrying the I26T and R160C substitutions in the transcriptional repressor HESX1 as models for septo-optic dysplasia and hypopituitarism." Dis Model Mech 1(4-5): 241-54. 174. Salemi, S., Besson, A., Eble, A., Gallati, S., Pfaffle, R. W. and Mullis, P. E. (2003). "New N-terminal located mutation (Q4ter) within the POU1F1-gene (PIT- 1) causes recessive combined pituitary hormone deficiency and variable phenotype." Growth Horm IGF Res 13 (5): 264-8. 175. Savage, J. J., Hunter, C. S., Clark-Sturm, S. L., Jacob, T. M., Pfaeffle, R. W. and Rhodes, S. J. (2007). "Mutations in the LHX3 gene cause dysregulation of pituitary and neural target genes that reflect patient phenotypes." Gene 400 (1-2): 44-51. 176. Savage, J. J., Mullen, R. D., Sloop, K. W., Colvin, S. C., Camper, S. A., Franklin, C. L. and Rhodes, S. J. (2007). "Transgenic mice expressing LHX3 transcription factor isoforms in the pituitary: effects on the gonadotrope axis and sex-specific reproductive disease." J Cell Physiol 212 (1): 105-17. 177. Sclafani, A. M., Skidmore, J. M., Ramaprakash, H., Trumpp, A., Gage, P. J. and Martin, D. M. (2006). "Nestin-Cre mediated deletion of Pitx2 in the mouse." Genesis 44 (7): 336-44. 178. Semina, E. V., Reiter, R., Leysens, N. J., Alward, W. L., Small, K. W., Datson, N. A., Siegel-Bartelt, J., Bierke-Nelson, D., Bitoun, P., Zabel, B. U., Carey, J. C. and Murray, J. C. (1996). "Cloning and characterization of a novel bicoid-related homeobox transcription factor gene, RIEG, involved in Rieger syndrome." Nat Genet 14 (4): 392-9. 179. Sharma, K., Sheng, H. Z., Lettieri, K., Li, H., Karavanov, A., Potter, S., Westphal, H. and Pfaff, S. L. (1998). "LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities for motor neurons." Cell 95 (6): 817-28. 180. Sheng, H. Z., Moriyama, K., Yamashita, T., Li, H., Potter, S. S., Mahon, K. A. and Westphal, H. (1997). "Multistep control of pituitary organogenesis." Science 278 (5344): 1809-12. 181. Sheng, H. Z., Zhadanov, A. B., Mosinger, B., Jr., Fujii, T., Bertuzzi, S., Grinberg, A., Lee, E. J., Huang, S. P., Mahon, K. A. and Westphal, H. (1996). "Specification of pituitary cell lineages by the LIM homeobox gene Lhx3." Science 272 (5264): 1004-7. ! '*! 182. Shibahara, H., Ikeshita, N., Sugiyama, Y., Toda, K., Yamamoto, D., Herningtyas, E. H., Maki, T., Kubota, E., Iguchi, G., Iida, K., Takahashi, Y., Kaji, H., Chihara, K. and Okimura, Y. "W194XProp1 and S156insTProp1, both of which have intact DNA-binding domain, show a different DNA-binding activity to the Prop1-binding element in human Pit-1 gene." Mol Cell Endocrinol 323 (2): 167-71. 183. Shima, Y., Zubair, M., Komatsu, T., Oka, S., Yokoyama, C., Tachibana, T., Hjalt, T. A., Drouin, J. and Morohashi, K. (2008). "Pituitary homeobox 2 regulates adrenal4 binding protein/steroidogenic factor-1 gene transcription in the pituitary gonadotrope through interaction with the intronic enhancer." Mol Endocrinol 22 (7): 1633-46. 184. Simmons, D. M., Voss, J. W., Ingraham, H. A., Holloway, J. M., Broide, R. S., Rosenfeld, M. G. and Swanson, L. W. (1990). "Pituitary cell phenotypes involve cell-specific Pit-1 mRNA translation and synergistic interactions with other classes of transcription factors." Genes Dev 4(5): 695-711. 185. Sloop, K. W., Dwyer, C. J. and Rhodes, S. J. (2001). "An isoform-specific inhibitory domain regulates the LHX3 LIM homeodomain factor holoprotein and the production of a functional alternate translation form." J Biol Chem 276 (39): 36311-9. 186. Sloop, K. W., Meier, B. C., Bridwell, J. L., Parker, G. E., Schiller, A. M. and Rhodes, S. J. (1999). "Differential activation of pituitary hormone genes by human Lhx3 isoforms with distinct DNA binding properties." Mol Endocrinol 13 (12): 2212-25. 187. Sloop, K. W., Parker, G. E., Hanna, K. R., Wright, H. A. and Rhodes, S. J. (2001). "LHX3 transcription factor mutations associated with combined pituitary hormone deficiency impair the activation of pituitary target genes." Gene 265 (1- 2): 61-9. 188. Sobrier, M. L., Attie-Bitach, T., Netchine, I., Encha-Razavi, F., Vekemans, M. and Amselem, S. (2004). "Pathophysiology of syndromic combined pituitary hormone deficiency due to a LHX3 defect in light of LHX3 and LHX4 expression during early human development." Gene Expr Patterns 5(2): 279-84. 189. Sobrier, M. L., Maghnie, M., Vie-Luton, M. P., Secco, A., di Iorgi, N., Lorini, R. and Amselem, S. (2006). "Novel HESX1 mutations associated with a life- threatening neonatal phenotype, pituitary aplasia, but normally located posterior pituitary and no optic nerve abnormalities." J Clin Endocrinol Metab 91 (11): 4528- 36. 190. Sobrier, M. L., Netchine, I., Heinrichs, C., Thibaud, N., Vie-Luton, M. P., Van Vliet, G. and Amselem, S. (2005). "Alu-element insertion in the homeodomain of HESX1 and aplasia of the anterior pituitary." Hum Mutat 25 (5): 503. 191. Sornson, M. W., Wu, W., Dasen, J. S., Flynn, S. E., Norman, D. J., O'Connell, S. M., Gukovsky, I., Carriere, C., Ryan, A. K., Miller, A. P., Zuo, L., Gleiberman, A. S., Andersen, B., Beamer, W. G. and Rosenfeld, M. G. (1996). "Pituitary lineage determination by the Prophet of Pit-1 homeodomain factor defective in Ames dwarfism." Nature 384 (6607): 327-33. 192. Suh, H., Gage, P. J., Drouin, J. and Camper, S. A. (2002). "Pitx2 is required at multiple stages of pituitary organogenesis: pituitary primordium formation and cell specification." Development 129 (2): 329-37. 193. Suszko, M. I., Antenos, M., Balkin, D. M. and Woodruff, T. K. (2008). "Smad3 and Pitx2 cooperate in stimulation of FSHbeta gene transcription." Mol Cell Endocrinol 281 (1-2): 27-36. ! (+! 194. Suszko, M. I., Lo, D. J., Suh, H., Camper, S. A. and Woodruff, T. K. (2003). "Regulation of the rat follicle-stimulating hormone beta-subunit promoter by activin." Mol Endocrinol 17 (3): 318-32. 195. Szeto, D. P., Rodriguez-Esteban, C., Ryan, A. K., O'Connell, S. M., Liu, F., Kioussi, C., Gleiberman, A. S., Izpisua-Belmonte, J. C. and Rosenfeld, M. G. (1999). "Role of the Bicoid-related homeodomain factor Pitx1 in specifying hindlimb morphogenesis and pituitary development." Genes Dev 13 (4): 484-94. 196. Tajima, T., Hattori, T., Nakajima, T., Okuhara, K., Tsubaki, J. and Fujieda, K. (2007). "A novel missense mutation (P366T) of the LHX4 gene causes severe combined pituitary hormone deficiency with pituitary hypoplasia, ectopic posterior lobe and a poorly developed sella turcica." Endocr J 54 (4): 637-41. 197. Tajima, T., Hattorri, T., Nakajima, T., Okuhara, K., Sato, K., Abe, S., Nakae, J. and Fujieda, K. (2003). "Sporadic heterozygous frameshift mutation of HESX1 causing pituitary and optic nerve hypoplasia and combined pituitary hormone deficiency in a Japanese patient." J Clin Endocrinol Metab 88 (1): 45-50. 198. Tajima, T., Ohtake, A., Hoshino, M., Amemiya, S., Sasaki, N., Ishizu, K. and Fujieda, K. (2009). "OTX2 loss of function mutation causes anophthalmia and combined pituitary hormone deficiency with a small anterior and ectopic posterior pituitary." J Clin Endocrinol Metab 94 (1): 314-9. 199. Tajima, T., Yorifuji, T., Ishizu, K. and Fujieda, K. (2009). "A Novel Mutation (V101A) of the LHX4 Gene in a Japanese Patient with Combined Pituitary Hormone Deficiency." Exp Clin Endocrinol Diabetes . 200. Takahashi, K., Okita, K., Nakagawa, M. and Yamanaka, S. (2007). "Induction of pluripotent stem cells from fibroblast cultures." Nat Protoc 2(12): 3081-9. 201. Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K. and Yamanaka, S. (2007). "Induction of pluripotent stem cells from adult human fibroblasts by defined factors." Cell 131 (5): 861-72. 202. Tatsumi, K., Miyai, K., Notomi, T., Kaibe, K., Amino, N., Mizuno, Y. and Kohno, H. (1992). "Cretinism with combined hormone deficiency caused by a mutation in the PIT1 gene." Nat Genet 1(1): 56-8. 203. Tatsumi, K. I., Kikuchi, K., Tsumura, K. and Amino, N. (2004). "A novel PROP1 gene mutation (157delA) in Japanese siblings with combined anterior pituitary hormone deficiency." Clin Endocrinol (Oxf) 61 (5): 635-40. 204. Thaler, J. P., Lee, S. K., Jurata, L. W., Gill, G. N. and Pfaff, S. L. (2002). "LIM factor Lhx3 contributes to the specification of motor neuron and interneuron identity through cell-type-specific protein-protein interactions." Cell 110 (2): 237- 49. 205. Thomas, P. Q., Dattani, M. T., Brickman, J. M., McNay, D., Warne, G., Zacharin, M., Cameron, F., Hurst, J., Woods, K., Dunger, D., Stanhope, R., Forrest, S., Robinson, I. C. and Beddington, R. S. (2001). "Heterozygous HESX1 mutations associated with isolated congenital pituitary hypoplasia and septo-optic dysplasia." Hum Mol Genet 10 (1): 39-45. 206. Thomas, P. Q., Johnson, B. V., Rathjen, J. and Rathjen, P. D. (1995). "Sequence, genomic organization, and expression of the novel homeobox gene Hesx1." J Biol Chem 270 (8): 3869-75. 207. Treier, M., Gleiberman, A. S., O'Connell, S. M., Szeto, D. P., McMahon, J. A., McMahon, A. P. and Rosenfeld, M. G. (1998). "Multistep signaling requirements for pituitary organogenesis in vivo." Genes Dev 12 (11): 1691-704. 208. Tremblay, J. J., Goodyer, C. G. and Drouin, J. (2000). "Transcriptional properties of Ptx1 and Ptx2 isoforms." Neuroendocrinology 71 (5): 277-86. ! ("! 209. Tumer, Z. and Bach-Holm, D. (2009). "Axenfeld-Rieger syndrome and spectrum of PITX2 and FOXC1 mutations." Eur J Hum Genet 17 (12): 1527-39. 210. Turton, J. P., Reynaud, R., Mehta, A., Torpiano, J., Saveanu, A., Woods, K. S., Tiulpakov, A., Zdravkovic, V., Hamilton, J., Attard-Montalto, S., Parascandalo, R., Vella, C., Clayton, P. E., Shalet, S., Barton, J., Brue, T. and Dattani, M. T. (2005). "Novel mutations within the POU1F1 gene associated with variable combined pituitary hormone deficiency." J Clin Endocrinol Metab 90 (8): 4762-70. 211. U, H. S., Alilain, W. and Saljooque, F. (2002). "Fetal brain progenitor cells transdifferentiate to fates outside the nervous system." Mol Endocrinol 16 (11): 2645-56. 212. U, H. S., Wu, B., Wilkes, N., Ho, A. and Saljooque, F. (2007). "Brain stem cells adopt a pituitary fate after implantation into the adult rodent pituitary gland." Neuroendocrinology 86 (1): 58-68. 213. Vadlamudi, U., Espinoza, H. M., Ganga, M., Martin, D. M., Liu, X., Engelhardt, J. F. and Amendt, B. A. (2005). "PITX2, beta-catenin and LEF-1 interact to synergistically regulate the LEF-1 promoter." J Cell Sci 118(Pt 6): 1129-37. 214. Vallette-Kasic, S., Barlier, A., Teinturier, C., Diaz, A., Manavela, M., Berthezene, F., Bouchard, P., Chaussain, J. L., Brauner, R., Pellegrini-Bouiller, I., Jaquet, P., Enjalbert, A. and Brue, T. (2001). "PROP1 gene screening in patients with multiple pituitary hormone deficiency reveals two sites of hypermutability and a high incidence of corticotroph deficiency." J Clin Endocrinol Metab 86 (9): 4529- 35. 215. Vesper, A. H., Raetzman, L. T. and Camper, S. A. (2006). "Role of prophet of Pit1 (PROP1) in gonadotrope differentiation and puberty." Endocrinology 147 (4): 1654-63. 216. Vieira, T. C., Boldarine, V. T. and Abucham, J. (2007). "Molecular analysis of PROP1, PIT1, HESX1, LHX3, and LHX4 shows high frequency of PROP1 mutations in patients with familial forms of combined pituitary hormone deficiency." Arq Bras Endocrinol Metabol 51 (7): 1097-103. 217. Vieira, T. C., da Silva, M. R. and Abucham, J. (2006). "The natural history of the R120C PROP1 mutation reveals a wide phenotypic variability in two untreated adult brothers with combined pituitary hormone deficiency." Endocrine 30 (3): 365-9. 218. Vieira, T. C., Dias da Silva, M. R., Cerutti, J. M., Brunner, E., Borges, M., Arnaldi, L. T., Kopp, P. and Abucham, J. (2003). "Familial combined pituitary hormone deficiency due to a novel mutation R99Q in the hot spot region of Prophet of Pit-1 presenting as constitutional growth delay." J Clin Endocrinol Metab 88 (1): 38-44. 219. Voutetakis, A., Argyropoulou, M., Sertedaki, A., Livadas, S., Xekouki, P., Maniati- Christidi, M., Bossis, I., Thalassinos, N., Patronas, N. and Dacou-Voutetakis, C. (2004). "Pituitary magnetic resonance imaging in 15 patients with Prop1 gene mutations: pituitary enlargement may originate from the intermediate lobe." J Clin Endocrinol Metab 89 (5): 2200-6. 220. Voutetakis, A., Maniati-Christidi, M., Kanaka-Gantenbein, C., Dracopoulou, M., Argyropoulou, M., Livadas, S., Dacou-Voutetakis, C. and Sertedaki, A. (2004). "Prolonged jaundice and hypothyroidism as the presenting symptoms in a neonate with a novel Prop1 gene mutation (Q83X)." Eur J Endocrinol 150 (3): 257-64. 221. Voutetakis, A., Sertedaki, A., Livadas, S., Xekouki, P., Bossis, I., Dacou- Voutetakis, C. and Argyropoulou, M. I. (2006). "Pituitary size fluctuation in long- term MR studies of PROP1 deficient patients: A persistent pathophysiological mechanism?" J Endocrinol Invest 29 (5): 462-6. ! (#! 222. Wagner, J., Lepore, D. and Thomas, P. (2007). "Differentiation of mouse embryonic stem cells into growth hormone and prolactin expressing cells in vitro." Mol Cell Endocrinol 273 (1-2): 68-74. 223. Ward, R. D., Raetzman, L. T., Suh, H., Stone, B. M., Nasonkin, I. O. and Camper, S. A. (2005). "Role of PROP1 in pituitary gland growth." Mol Endocrinol 19 (3): 698-710. 224. Ward, R. D., Stone, B. M., Raetzman, L. T. and Camper, S. A. (2006). "Cell proliferation and vascularization in mouse models of pituitary hormone deficiency." Mol Endocrinol 20 (6): 1378-90. 225. Webb, E. A. and Dattani, M. T. "Septo-optic dysplasia." Eur J Hum Genet 18 (4): 393-7. 226. Webb, G. C., Thomas, P. Q., Ford, J. H. and Rathjen, P. D. (1993). "Hesx1, a homeobox gene expressed by murine embryonic stem cells, maps to mouse chromosome 14, bands A3-B." Genomics 18 (2): 464-6. 227. West, B. E., Parker, G. E., Savage, J. J., Kiratipranon, P., Toomey, K. S., Beach, L. R., Colvin, S. C., Sloop, K. W. and Rhodes, S. J. (2004). "Regulation of the follicle-stimulating hormone beta gene by the LHX3 LIM-homeodomain transcription factor." Endocrinology 145 (11): 4866-79. 228. Wit, J. M., Vulsma, T. and de Vijlder, J. J. (2001). "[From gene to disease; POU1F1- and PROP1-mutations in pituitary hormone deficiency]." Ned Tijdschr Geneeskd 145 (50): 2425-7. 229. Wu, W., Cogan, J. D., Pfaffle, R. W., Dasen, J. S., Frisch, H., O'Connell, S. M., Flynn, S. E., Brown, M. R., Mullis, P. E., Parks, J. S., Phillips, J. A., 3rd and Rosenfeld, M. G. (1998). "Mutations in PROP1 cause familial combined pituitary hormone deficiency." Nat Genet 18 (2): 147-9. 230. Xu, L., Lavinsky, R. M., Dasen, J. S., Flynn, S. E., McInerney, E. M., Mullen, T. M., Heinzel, T., Szeto, D., Korzus, E., Kurokawa, R., Aggarwal, A. K., Rose, D. W., Glass, C. K. and Rosenfeld, M. G. (1998). "Signal-specific co-activator domain requirements for Pit-1 activation." Nature 395 (6699): 301-6. 231. Xu, Q., Yuan, X., Tunici, P., Liu, G., Fan, X., Xu, M., Hu, J., Hwang, J. Y., Farkas, D. L., Black, K. L. and Yu, J. S. (2009). "Isolation of tumour stem-like cells from benign tumours." Br J Cancer 101 (2): 303-11. 232. Yaden, B. C., Garcia, M., 3rd, Smith, T. P. and Rhodes, S. J. (2006). "Two promoters mediate transcription from the human LHX3 gene: involvement of nuclear factor I and specificity protein 1." Endocrinology 147 (1): 324-37. 233. Yaden, B. C., Savage, J. J., Hunter, C. S. and Rhodes, S. J. (2005). "DNA recognition properties of the LHX3b LIM homeodomain transcription factor." Mol Biol Rep 32 (1): 1-6. 234. Yoshida, S., Kato, T., Susa, T., Cai, L. Y., Nakayama, M. and Kato, Y. (2009). "PROP1 coexists with SOX2 and induces PIT1-commitment cells." Biochem Biophys Res Commun 385 (1): 11-5. 235. Zhao, X., Teng, R., Asanuma, K., Okouchi, Y., Johkura, K., Ogiwara, N. and Sasaki, K. (2005). "Differentiation of mouse embryonic stem cells into gonadotrope-like cells in vitro." J Soc Gynecol Investig 12 (4): 257-62. 236. Zhao, Y., Morales, D. C., Hermesz, E., Lee, W. K., Pfaff, S. L. and Westphal, H. (2006). "Reduced expression of the LIM-homeobox gene Lhx3 impairs growth and differentiation of Rathke's pouch and increases cell apoptosis during mouse pituitary development." Mech Dev 123 (8): 605-13. 237. Zhu, X., Zhang, J., Tollkuhn, J., Ohsawa, R., Bresnick, E. H., Guillemot, F., Kageyama, R. and Rosenfeld, M. G. (2006). "Sustained Notch signaling in ! ($! progenitors is required for sequential emergence of distinct cell lineages during organogenesis." Genes Dev 20 (19): 2739-53. 238. Zimmermann, A., Schenk, J. P., Grigorescu Sido, P., Pfaffle, R., Lazea, C., Zimmermann, T., Heinrich, U., Weber, M. M. and Bettendorf, M. (2007). "MRI findings and genotype analysis in patients with childhood onset growth hormone deficiency--correlation with severity of hypopituitarism." J Pediatr Endocrinol Metab 20 (5): 587-96. 239. Zygmunt-Gorska, A., Starzyk, J., Adamek, D., Radwanska, E., Sucharski, P., Herman-Sucharska, I. and Pietrzyk, J. J. (2009). "Pituitary enlargement in patients with PROP1 gene inactivating mutation represents cystic hyperplasia of the intermediate pituitary lobe. Histopathology and over 10 years follow-up of two patients." J Pediatr Endocrinol Metab 22 (7): 653-60. ! (%! TABLES ET FIGURES 11-12 : Schéma simplifié des facteurs de transcription impliqués dans le développement hypophysaire 16-17 : Tableaux récapitulatifs des mutations de HESX1 17-18 : Tableaux récapitulatifs des mutations de PROP1 19-20 : Tableaux récapitulatifs des mutations de OTX2 20-21 : Tableaux récapitulatifs des mutations de POU1F1 23-24 : Tableaux récapitulatifs des mutations de LHX3 29-30 : Schéma récapitulatif et modèle de développement hypophysaire de l’activation de POU1F1 33-34 : Tableaux récapitulatifs des mutations de LHX4 44-45 : Principales différences entre modèles murin et humain 44-45 : Mutations des principaux facteurs de transcription associés à une hyperplasie hypophysaire. ANNEXES ! (&! Disponible en ligne sur www.sciencedirect.com Annales d’Endocrinologie 69 (2008) 7–17 Mise au point Déficit hypophysaire combiné multiple : aspects cliniques et génétiques ଝ Clinical and genetic aspects of combined pituitary hormone deficiencies F. Castinetti a,c,e, R. Reynaud b,c,e, A. Saveanu c,d,e, M.-H. Quentien c,e, F. Albarel a,c,e, A. Barlier c,d,e, A. Enjalbert c,d,e, T. Brue a,c,e,∗ a Service d’endocrinologie, diabète et maladies métaboliques, hôpital de la Timone, 13385 Marseille, cedex 5, France b Service de pédiatrie multidisciplinaire, hôpital de la Timone, 13385 Marseille cedex 5, France c Laboratoire ICNE, UMR 6544, faculté de médecine du Nord, institut fédératif Jean-Roche, 13014 Marseille, France d Laboratoire de biologie moléculaire, hôpital de la Conception, 13005 Marseille, France e Centre de référence des maladies rares d’origine hypophysaire, hôpital de la Timone, 13385 Marseille cedex 5, France Disponible sur Internet le 4 mars 2008 Résumé Définition clinique. – Les insuffisances antéhypophysaires d’origine génétique sont caractérisées par l’association de déficits hormonaux de plusieurs des lignées antéhypophysaires : somatotrope (GH), thyréotrope (TSH), lactotrope (PRL), corticotrope (ACTH), gonadotrope (LH et FSH). Ces déficits sont liés à des mutations de facteurs de transcription impliqués dans l’ontogénèse hypophysaire. Epidémiologie. – En comparaison avec les causes classiques (secondaires à un processus expansif intracrânien ou iatrogènes après chirurgie ou radiothérapie cérébrale), l’incidence des hypopituitarismes congénitaux est faible. Elle est estimée à une pour 3000 ou 4000 naissances, même si ces valeurs sont probablement surestimées car certains déficits sont transitoires. Clinique. – La présentation clinique varie en fonction des lignées hormonales concernées ainsi que de la précocité et de l’intensité de l’atteinte. En l’absence de traitement, les principaux symptômes sont le retard de croissance aboutissant à un nanisme (en cas de déficit en GH), un retard psychomoteur (en cas de déficit en TSH), des anomalies de la puberté (en cas de déficit en gonadotrophines). Diagnostic. – Le diagnostic de déficit antéhypophysaire est clinique et biologique. Le diagnostic de déficit somatotrope ou corticotrope nécessite l’utilisation de tests dynamiques de stimulation. Les causes classiques doivent toujours être éliminées par la réalisation systématique d’une IRM cérébrale et hypothalamohypophysaire. Le diagnostic génétique repose sur le séquenc¸age direct des zones codantes des gènes des facteurs de transcription impliqués. La sélection du ou des facteurs à séquencer est fondée sur les données clinicobiologiques et radiologiques. Étiologie. – Les hypopituitarismes congénitaux sont dus à des mutations des gènes codant pour des facteurs de transcription impliqués dans les premières étapes du développement hypophysaire et qui sont associées à divers phénotypes. Les facteurs de transcription les plus fréquemment impliqués sont PROP1 (déficits somatolactotrope, thyréotrope et gonadotrope, parfois associé à un déficit corticotrope ; hyper puis hypoplasie hypophysaire), POU1F1 (déficits somatolactotrope et thyréotrope, hypoplasie hypophysaire), HESX1 (déficits hypophysaires variables, dysplasie septo-optique) et à un degré moindre LHX3 (déficits somatolactotrope, thyréotrope et gonadotrope ; limitation de la rotation de la tête et du cou) et LHX4 (déficits hypophysaires variables, posthypophyse ectopique, malformations cérébrales). Prise en charge. – Elle consiste à substituer chacun des déficits antéhypophysaires observés et à éduquer le patient sur la nécessité de ces traitements au long cours. La surveillance porte sur l’adaptation de ces traitements et le dépistage de la survenue de nouveaux déficits. Conseil génétique/diagnostic anténatal. – Le type de transmission varie en fonction du facteur de transcription impliqué et de la mutation (récessif pour PROP1 et LHX3 , dominant pour LHX4 , récessif ou dominant selon les mutations pour POU1F1 et HESX1 ). Le conseil génétique est donc nécessaire afin de dépister les nouveau-nés à risque et de pouvoir adapter la surveillance. Pronostic. – Le pronostic est bon, avec un devenir identique à celui d’un patient non déficitaire si le traitement substitutif est pris dès le diagnostic posé et adapté correctement, avec un suivi par un médecin spécialisé. © 2008 Publi e´ par Elsevier Masson SAS. ଝ Cet article est publié en partenariat avec Orphanet et disponible sur le site www.orpha.net . © 2007 Orphanet. Publié par Elsevier Masson SAS. Tous droits réservés. Abbreviations: GH:, growth hormone; IGF1:, insulin like growth Factor 1; LH:, luteinizing hormone; FSH:, follicle stimulating hormone; ACTH:, adrenocorticotropin hormone; TSH:, thyreotropin stimulating hormone; PRL:, prolactine. ∗ Auteur correspondant. Adresse e-mail : [email protected] (T. Brue). 0003-4266/$ – see front matter © 2008 Publi e´ par Elsevier Masson SAS. doi: 10.1016/j.ando.2008.01.001 8 F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 Abstract Definition. – Congenital hypopituitarism is characterized by multiple pituitary hormone deficiency, including somatotroph, thyrotroph, lactotroph, corticotroph or gonadotroph deficiencies, due to mutations of pituitary transcription factors involved in pituitary ontogenesis. Incidence. – Congenital hypopituitarism is rare compared with the high incidence of hypopituitarism induced by pituitary adenomas, transsphenoidal surgery or radiotherapy. The incidence of congenital hypopituitarism is estimated to be between 1:3000 and 1:4000 births. Clinical signs. – Clinical presentation is variable, depending on the type and severity of deficiencies and on the age at diagnosis. If untreated, main symptoms include short stature, cognitive alterations or delayed puberty. Diagnosis. – A diagnosis of combined pituitary hormone deficiency (CPHD) must be suspected when evident causes of hypopituitarism (sellar tumor, postsurgical or radioinduced hypopituitarism . . . ) have been ruled out. Clinical, biological and radiological work-up is very important to better determine which transcription factor should be screened. Confirmation is provided by direct sequencing of the transcription factor genes. Aetiology. – Congenital hypopituitarism is due to mutations of several genes encoding pituitary transcription factors. Phenotype varies with the factor involved: PROP1 (somatolactotroph, thyrotroph, gonadotroph and sometimes corticotroph deficiencies; pituitary hyper and hypoplasia), POU1F1 (somatolactotroph and thyrotroph deficiencies, pituitary hypoplasia), HESX1 (variable pituitary deficiencies, septo-optic dysplasia), and less frequently LHX3 (somatolactotroph, thyrotroph and gonadotroph deficiencies, limited head and neck rotation) and LHX4 (variable pituitary deficiencies, ectopic neurohypophysis, cerebral abnormalities). Management. – An appropriate replacement of hormone deficiencies is required. Strict follow-up is necessary because patients develop new deficiencies (for example late onset corticotroph deficiency in patients with PROP1 mutations). Genetic counselling. – Type of transmission varies with the factor and the mutation involved (recessive transmission for PROP1 and LHX3 , dominant for LHX4 , autosomal or recessive for POU1F1 and HESX1 ). Prognosis. – It is equivalent to patients without pituitary deficiencies if treatment is started immediately when diagnosis is confirmed, and if a specialized follow-up is performed. © 2008 Publi e´ par Elsevier Masson SAS. Mots clés : Déficits hypophysaires combinés multiples ; Hypopituitarismes congénitaux ; Hypophyse ; Facteurs de transcription hypophysaires ; Hormone de croissance Keywords: Combined pituitary hormone deficiency (CPHD); Somatotroph deficiency; Pituitary; Congenital hypopituitarism; Growth hormone; Pituitary transcription factors 1. Définition même avec un phénotype compatible (déficits somatolactotrope et thyréotrope, sans anomalie IRM), les mutations de POU1F1 Les déficits hypophysaires ou hypopituitarismes se défi- sont rares au sein des patients atteints de déficits hypophysaires nissent par une insuffisance de synthèse ou de sécrétion d’une congénitaux combinés surtout en l’absence d’antécédents fami- ou plusieurs hormones antéhypophysaires. Ces déficits peuvent liaux (1–3 % des cas sporadiques, 10–30 % des cas familiaux être secondaires à des causes tumorales (adénome hypophysaire selon les études) [9,25] . Cependant, la différence d’incidence compressif, craniopharyngiome . . . ) ou iatrogènes (postchirurgie des mutations entre facteurs de transcription semble également ou postradiothérapie cérébrale). En l’absence de cause lésion- liée à l’origine des patients : une étude japonaise et une étude nelle, des formes plus rares d’origine congénitale peuvent être australienne ont ainsi trouvé un taux très faible de mutations mises en évidence. de PROP1 dans la population de leur pays où les mutations Les déficits hypophysaires multiples se définissent par le de POU1F1 sont moins rares [17,35] . Les mutations d’ HESX1 déficit d’au moins deux lignées hypophysaires, s’opposant restent exceptionnelles, malgré leur recherche dans de larges aux déficits isolés portant sur un seul axe hypophysaire études et sont classiquement rapportées dans des cas familiaux (comme le déficit isolé en hormone de croissance par muta- de dysplasie septo-optique [13,15,25] . Les mutations des autres tion du gène GH1 , déficit corticotrope isolé par mutation de facteurs de transcription hypophysaire ( LHX3 , LHX4 . . . ) font TPIT . . . ). Les déficits hypophysaires multiples associent le plus essentiellement l’objet de publications de cas isolés, souvent souvent un déficit somatotrope à un autre déficit antéhypophy- dans un contexte familial. La prévalence des déficits hypo- saire (le plus fréquemment gonadotrope et thyréotrope), mais physaires congénitaux est estimée à une pour 3000 ou 4000 toutes les associations de déficits hypophysaires sont envisa- naissances ; cependant, ce chiffre est probablement surestimé geables. car il tient compte de déficits en GH parfois transitoires. De Les déficits hypophysaires multiples congénitaux sont liés plus, la prévalence varie fortement en fonction des critères diag- aux mutations de facteurs de transcription dont le rôle est essen- nostiques utilisés. tiel dans l’ontogenèse hypophysaire, dans le développement et Outre les déficits antéhypophysaires, la mutation de ces dans la croissance des lignées cellulaires antéhypophysaires facteurs de transcription s’associe parfois à des anomalies (PROP1, POU1F1, HESX1 . . . ). Leur incidence est faible et cérébrales ou hypophysaires : hypoplasie hypophysaire, posthy- variable selon le facteur de transcription impliqué : ainsi, les pophyse ectopique, dysplasie septo-optique, malformation de mutations du facteur de transcription hypophysaire PROP1 Chiari, atrophie du corps calleux . . . éléments qui devront être constituent actuellement la première cause humaine identifiée de pris en compte lors de la démarche diagnostique et thérapeu- déficit hypophysaire multiple congénital [7,8,45] ; en revanche, tique. F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 9 2. Description clinique/critères diagnostiques décompensation aiguë, des douleurs abdominales avec déshy- dratation et tachycardie. La pâleur retrouvée lors de l’examen Le diagnostic est suspecté devant l’apparition progressive de contraste avec la classique mélanodermie de l’insuffisant déficits hypophysaires, survenant peu après la naissance ou de surrénalien périphérique. un retentissement statural peut éga- fac¸on retardée, après avoir éliminé l’ensemble des causes, en lement être observe ; particulier tumorales, pouvant être à l’origine d’un hypopitui- • chez l’adulte, le tableau est dominé par l’asthénie, tarisme. Les critères diagnostiques, qui vont orienter vers la l’hypotension artérielle, la pâleur et l’amaigrissement. recherche de la mutation d’un facteur de transcription spéci- fique, sont cliniques, hormonaux et morphologiques : type de Le diagnostic de déficit corticotrope doit être recherché dès déficits présents, âge d’apparition des déficits, malformations la moindre suspicion clinique, car il peut être à l’origine d’une éventuellement associées . . . [30] . On distingue ainsi différents insuffisance surrénalienne aiguë, à retentissement sévère, voire tableaux cliniques évocateurs d’anomalies de plusieurs facteurs fatal. Les signes évocateurs d’insuffisance surrénalienne aiguë de transcription. associent une asthénie majeure, une hypotension artérielle, des douleurs abdominales inconstantes, des signes de déshydrata- 2.1. Description clinique en fonction du déficit hormonal tion avec nausées et vomissements, une tachycardie. La prise en suspecté charge est une urgence thérapeutique (cf. traitement et prise en charge). 2.1.1. Déficit somatotrope (déficit en GH) La présentation clinique varie en fonction de l’âge au diag- 2.1.3. Déficit thyréotrope (déficit en TSH) nostic : Il a un retentissement variable en fonction de l’âge : • en période néonatale, le déficit somatotrope peut être sus- • pecté devant des hypoglycémies récidivantes avec sueurs et en période néonatale, le tableau classique est identique à un ictère néonatal persistant ; celui de l’hypothyroïdie congénitale : petite taille, œdème • chez l’enfant, le déficit somatotrope va se traduire par une généralisé, ictère néonatal prolongé, altération des capacités cassure de la courbe de croissance à partir de l’âge de quatre mentales. Il est nécessaire de rappeler que le déficit thyréo- ans (âge à partir duquel la sécrétion d’hormone de crois- trope n’est pas diagnostiqué par le test de dépistage néonatal sance influe sur la croissance) avec un retentissement statural de l’hypothyroïdie (car la TSH n’est pas augmentée contrai- précédant un éventuel retentissement pondéral. En cas de rement à l’hypothyroïdie périphérique) ; • déficit sévère, on observe classiquement un front bombé, un chez l’enfant, le déficit thyréotrope se traduit par une asthé- faciès poupin avec ensellure nasale large. Des signes fonction- nie, une constipation avec une prise de poids et une cassure nels associant asthénie et sensations d’hypoglycémie pourront de la courbe staturale. L’âge osseux est très retardé car également être retrouvés ; les hormones thyroïdiennes sont nécessaires à la maturation • chez l’adulte, les signes sont aspécifiques avec asthénie, osseuse ; • baisse de l’activité physique, faiblesse musculaire et épisodes chez l’adulte, le déficit thyréotrope se traduit par une asthénie, d’hypoglycémie. une constipation avec une prise de poids, parfois des troubles cognitifs voire une pseudodémence (en cas d’hypothyroïdie Les signes de déficit en hormone de croissance sont peu spé- prolongée chez un sujet d’âge avancé). cifiques, à l’exception de la cassure staturale survenant chez l’enfant. Chez l’adulte, le diagnostic de déficit somatotrope sera 2.1.4. Déficit gonadotrope (déficit en LH/FSH) donc le plus souvent effectué lors d’un dépistage de déficit Il a un retentissement chez l’enfant essentiellement en période hypophysaire multiple, évoqué devant d’autres déficits symp- pubertaire et chez l’adulte. tomatiques (déficit corticotrope ou thyréotrope, par exemple). • chez l’enfant, le déficit gonadotrope va être à l’origine d’un 2.1.2. Déficit corticotrope (déficit en ACTH) retard pubertaire (avec aménorrhée chez la fille et absence La présentation clinique varie en fonction de l’âge au diag- de développement des caractères sexuels secondaires dans nostic : les deux sexes). Comme les hormones sexuelles amplifient la fréquence et l’amplitude des pics d’hormone de croissance, • en période néonatale, c’est devant des hypoglycémies sévères, un déficit fonctionnel en hormone de croissance va également répétées avec retentissement fonctionnel majeur et un traite- être associé, le tout aboutissant à un retard de croissance sta- ment symptomatique peu efficace, que devra être évoqué le turale, puis pondérale, avec absence de développement des diagnostic de déficit corticotrope. Un ictère néonatal prolongé caractères sexuels secondaires. On peut aussi être alerté dès peut également être présent. Le tableau peut également asso- la naissance ou dans l’enfance par la présence d’un micropénis cier une hypotension artérielle sévère avec syndrome de perte ou de cryptorchidie ; de sel ; • chez l’adulte, le déficit gonadotrope va se traduire par une • chez l’enfant, le tableau clinique associe une asthénie avec diminution de la libido, avec une asthénie et une diminu- hypotension artérielle, des hypoglycémies et, en cas de tion de la masse musculaire associées à une dysérection 10 F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 chez l’homme et une aménorrhée chez la femme. Ce tableau première mutation de PROP1 décrite dans le domaine de tran- s’associe à une infertilité d’origine centrale. sactivation (mutation W194X) [27] . Cette présentation clinique doit conduire à réaliser une recherche de mutation de PROP1 2.1.5. Déficit lactotrope (déficit en prolactine) chez tout patient présentant un hypogonadisme hypogonado- Il n’a aucun retentissement clinique chez l’homme. Chez la trope isolé sans anosmie. L’hypophyse peut être hyperplasique, femme, il se traduit par l’absence de montée de lait après la normale ou hypoplasique [41] : l’équipe de Camper a récem- grossesse. Les autres effets de la prolactine sont encore mal ment démontré chez la souris que cette modification d’aspect connus et ne permettent pas de définir d’autres impacts cliniques hypophysaire pourrait être due à des anomalies de migration de de cette carence. cellules progénitrices antéhypophysaires, bloquées dans le lobe intermédiaire (hyperplasie initiale), avec dégénérescence tardive 2.1.6. Déficits multiples (hypoplasie) [42,43] . Plusieurs déficits peuvent être associés, pouvant aboutir à un tableau de panhypopituitarisme. Toutes les associations de 2.2.3. HESX1 déficits sont envisageables. Les patients présentent une asthé- Association classique : déficit somatotrope, dysplasie septo- nie marquée, une pâleur, une peau fine et atrophique, avec des optique, posthypophyse ectopique. Le tableau phénotypique cheveux fins et peu de sourcils. peut comporter une dysplasie septo-optique, avec parfois une posthypophyse ectopique [4,6,32,36] . Une étude récente a 2.2. Exemples de tableau clinique en cas de mutation de cependant décrit une mutation de HESX1 avec aplasie hypo- certains facteurs de transcription hypophysaire physaire et une posthypophyse non ectopique [31] . Les déficits hypophysaires sont variables liés à une pénétrance incomplète, 2.2.1. POU1F1 étagés du déficit isolé en GH au panhypopituitarisme avec dia- Association classique : déficits somatolactotrope et thyréotrope. bète insipide. Il faut toutefois préciser que des mutations de Le phénotype du patient porteur d’une mutation de POU1F1 est HESX1 ne sont trouvées que dans moins de 5 % des cas de variable, principalement en terme d’âge d’apparition du défi- dysplasie septo-optique [18] . cit thyréotrope. L’âge au diagnostic s’échelonne en effet de la naissance à l’âge de 25 ans, mais il est le plus souvent 3. Étiologie/aspects génétiques précoce, avant l’âge de deux ans. Le déficit somatolactotrope est classiquement complet ; le déficit thyréotrope peut être 3.1. Mutations de POU1F1 complet dès la naissance ou s’aggraver progressivement avec l’âge [9,11,22,24,37] . Un seul cas a été décrit avec absence POU1F1 est un facteur de transcription hypophysaire appar- de déficit thyréotrope à l’âge de 20 ans [37] . Les fonctions tenant à la famille des facteurs de transcription à homéodomaine gonadotrope et corticotrope sont préservées. L’hypophyse est POU. Il a été identifié pour la première fois en 1988 par les classiquement normale ou hypoplasique, sans anomalies de la groupes de Karin et Rosenfeld [12] . Chez la souris, Pit-1 (ortho- posthypophyse ni section de tige. logue murin de POU1F1 ) est exprimé à partir de e14,5 (jour 14,5 de la vie embryonnaire de la souris), avec une expression 2.2.2. PROP1 persistant à l’âge adulte au niveau du tissu hypophysaire. Pit-1 Association classique : déficits somatolactotrope, thyréotrope, est nécessaire au développement des lignées somatotrope, lacto- gonadotrope et parfois corticotrope retardé. Le phénotype du trope et thyréotrope de l’antéhypophyse : ce point a été démontré patient porteur d’anomalies de PROP1 est variable, aussi bien dans les modèles murin Snell et Jackson, porteurs respective- en termes de type de déficit que d’âge d’apparition des défi- ment d’une mutation W261C ou d’un remaniement du gène de cits. Aucune corrélation génotype/phénotype n’a pu être établie Pit-1 et présentant un phénotype déficitaire somatotrope, lacto- y compris entre patients porteurs de la même mutation au sein trope et thyréotrope [3] . Des éléments de liaison de Pit-1 sont de la même famille [28] . On observe classiquement un déficit présents sur les séquences des promoteurs des gènes codant pour somatolactotrope précoce (vers l’âge de huit ans), un déficit thy- la GH, la prolactine, la ß-TSH, les récepteurs de la GHRH et réotrope (vers l’âge de neuf ans) puis un déficit gonadotrope. Ce de la ß-TSH et sur son propre promoteur (autorégulation) [5] . déficit gonadotrope est d’expression variable en fonction des Une étude récente a également retrouvé un effet antiapoptotique individus porteurs de la mutation et n’est pas retrouvé chez de Pit-1 favorisant la croissance des cellules antéhypophysaires la souris [39] . Quelques cas de déficit corticotrope ont égale- [23] . ment été décrits, parfois de survenue très retardée (jusqu’à l’âge Le gène POU1F1 codant pour une protéine de 291 acides de 40 ans) sans que le mécanisme physiopathologique sous- aminés a été cloné chez l’homme en 1996 [26] . Il est localisé jacent ait pu être mis en évidence [38] . Enfin, si ce schéma sur le bras court du chromosome 3 et comporte six exons et cinq stéréotypé d’apparition des déficits est le plus souvent observé, introns. La première mutation humaine de POU1F1 a été décrite il faut également souligner la présentation phénotypique inha- en 1992 chez un enfant présentant un déficit triple somatolac- bituelle décrite chez trois frères dont le diagnostic initial était totrope et thyréotrope et porteur d’une mutation non-sens [34] . celui d’hypogonadisme hypogonadotrope isolé. Malgré la taille Même si la majorité des mutations décrites de POU1F1 sont normale atteinte sans traitement, il existait à l’âge adulte un défi- de transmission récessive (en particulier la mutation R271W, cit thyréotrope et somatotrope. Ce phénotype était associé à la située au niveau du codon 271 considéré comme une zone de F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 11 forte mutabilité chez l’homme [24] ), quelques mutations hété- identifié en 1998 [6] . À ce jour, 12 mutations de HESX1 ont été rozygotes, donc à transmission autosomique dominante, ont été rapportées, six de transmission autosomique dominante et six décrites. À ce jour, 27 mutations de POU1F1 ont été décrites, 22 de transmission autosomique récessive. de transmission récessive et cinq de transmission autosomique dominante. 3.4. Mutations d’autres facteurs de transcription 3.2. Mutations de PROP1 hypophysaire LHX3 PROP1 est un facteur de transcription à homéodomaine est un facteur de transcription exprimé à partir du neu- de type paired [33] . Chez la souris, PROP1 , qui est exprimé vième jour embryonnaire chez la souris dans la poche de Rathke LHX3 précocement du dixième au quinzième jour embryonnaire, est [19] . Les mutations de entraînent un hypopituitarisme nécessaire au développement des lignées somatolactrotropes, antérieur épargnant la lignée corticotrope, à transmission auto- thyréotrope et gonadotrope [33] . L’expression de PROP1 pré- somique récessive. Cliniquement, les patients atteints présentent cède ainsi celle de Pit-1 [1] . Il semble que PROP1 soit dans la majorité des cas une limitation de la rotation du cou. nécessaire à l’expression de Pit-1 [21] . Le caractère transitoire L’hypophyse est hypoplasique, hyperplasique ou pseudoglobu- de l’expression de PROP1 est également important, puisque le leuse avec un aspect évocateur de microadénome. À ce jour, LHX3 maintien de son expression chez des souris conduit à un retard de quatre mutations de ont été décrites [2,20] . LHX4 maturation gonadotrope et semble favoriser la genèse de tumeurs est un facteur de transcription à domaine LIM, dont hypophysaires. La première mutation de PROP1 chez la souris l’expression apparaît chez la souris au neuvième jour embryon- (souris Ames) a été mise en évidence par Sornson en 1996 (muta- naire au niveau de la poche de Rathke, pour s’éteindre au tion S83P) : elle est à l’origine d’un déficit somatolactotrope, quinzième jour [19] . Une seule mutation intronique a été décrite thyréotrope et gonadotrope de degré variable [10,33] . dans une famille dont trois membres présentaient un déficit Le gène PROP1 codant pour une protéine de 226 acides somatotrope, thyréotrope et corticotrope avec des anomalies aminés a été identifié chez l’homme pour la première fois extrahypophysaires (interruption de tige, anomalies des amyg- en 1998 [8,33,45] . Il est localisé sur le chromosome 3 et dales cérébelleuses et de la selle turcique) et une hypophyse comporte trois exons. La protéine comporte trois hélices alpha, hypoplasique [16,46] . La transmission est dominante probable- un domaine de transactivation en C-terminal et un homéodo- ment par mécanisme d’haploinsuffisance. PITX2 maine central permettant l’interaction d’un homodimère PROP1 est un facteur de transcription à homéodomaine sur ses séquences ADN cible [8] . Ces séquences cibles ne sont « bicoïde » exprimé depuis l’embryogenèse jusqu’à l’âge toujours pas connues, mais on utilise pour les études fonction- adulte. Les mutations, de transmission autosomique dominante, nelles une séquence consensus palindromique baptisée Prdq9, concernent classiquement l’homéodomaine et sont associées à commune aux facteurs de transcription à homéodomaine de type un phénotype de déficit hypophysaire variable (déficits combi- paired (séquence 5 ′-AC TAAT TGA ATTA GC-3 ′). PROP1 est nés en GH et TSH, GH et ACTH) avec présence d’un syndrome de Rieger : anomalies de la chambre antérieure de l’œil, hypo- nécessaire à l’expression des lignées somatolactotrope et thy- . . . réotrope [33] . Une étude récente a retrouvé l’implication d’une plasie dentaire, retard mental [40] . SOX3 interaction entre PROP1 et la ß-caténine au niveau du promo- Des anomalies du gène de , facteur de transcription teur proximal de POU1F1 dans l’expression de ces trois lignées impliqué dans la différenciation sexuelle, ont été mises en évi- [21] , mais ces mécanismes restent à préciser chez l’homme. dence chez des garc¸ons présentant un déficit hypophysaire isolé La première mutation humaine de PROP1 a été identifiée en (déficit somatotrope) ou multiple avec retard psychomoteur. Il 1998 [8] . Depuis, 24 mutations ont été décrites, de transmis- s’agit de mutations, délétions ou duplications, de transmission sion autosomique récessive. Toutes ces mutations concernent liée à l’X. L’hypophyse était hypoplasique avec une posthypo- l’homéodomaine, à l’exception d’une décrite dans notre labora- physe ectopique et des anomalies du corps calleux [14,44] . toire, qui concerne le domaine de transactivation [27] . 4. Diagnostic 3.3. Mutations de HESX1 4.1. Critères diagnostiques HESX1 est également un facteur de transcription hypophy- saire à homéodomaine de type paired ; il est exprimé au niveau Après la suspicion clinique de déficits hypophysaires (cf. de la poche de Rathke chez la souris entre le neuvième et le Diagnostic), le diagnostic est confirmé sur le plan biologique treizième jour embryonnaire [6,36] . Son inhibition est néces- par des dosages hormonaux statiques et des tests dynamiques saire à l’expression du facteur de transcription PROP1 [21] . (cf C-diagnostic biologique). Le diagnostic de déficit d’origine Les souris invalidées pour HESX1 présentent un phénotype génétique ne peut être affirmé qu’après avoir éliminé une cause proche de celui d’une dysplasie septo-optique associant hypo- organique : l’IRM hypothalamohypophysaire est donc indispen- plasie des nerfs optiques, hypoplasie hypophysaire et anomalies sable pour éliminer un processus expansif intracrânien (adénome de la ligne médiane (absence de corps calleux ou de septum hypophysaire, méningiome, craniopharyngiome . . . ), un proces- pellucidum, posthypophyse ectopique). Chez l’homme, le gène sus infiltratif (sarcoïdose, histiocytose, hémochromatose . . . ) ou HESX1 codant pour une protéine de 185 acides aminés a été auto-immun (hypophysite). De la même fac¸on, des antécédents 12 F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 de chirurgie intracrânienne ou de radiothérapie cérébrale font • l’existence de malformations cérébrales oriente vers une évoquer en premier lieu une cause iatrogène. mutation de HESX1 (dysplasie septo-optique), LHX4 (mal- Les arguments suivants doivent faire évoquer une cause géné- formation de Chiari) ou LHX3 (anomalie de rotation de la tique : tête et du cou). Cependant, dans ces deux derniers cas, le nombre de mutations rapportées est insuffisant pour définir • élimination d’autres étiologies ; un profil type de patients à dépister. Pour ces trois facteurs de • diagnostic chez le nouveau-né ou dans la petite enfance ; transcription, les déficits hypophysaires sont variables et ne • caractère familial des déficits hypophysaires multiples ; permettent pas de guider le choix décisionnel. • augmentation du nombre de déficits avec l’âge, avec tableau stéréotypé d’apparition des déficits (cf Présentation clinique) Ces critères phénotypiques ne sont cependant qu’un résumé pour certaines mutations de POU1F1 , PROP1 et HESX1 ; des tableaux cliniques possibles et, en pratique, le choix des • présence de malformations cérébrales associées : par facteurs de transcription à séquencer est souvent plus difficile. exemple, dysplasie septo-optique et mutations de HESX1 , À l’inverse, au sein des mutations de chaque facteur de trans- limitation de rotation de la tête et mutations de LHX3 , mal- cription hypophysaire, il ne semble pas exister de corrélation formation de Chiari et mutations de LHX4 . entre le type de mutation et le tableau clinique présenté. Le Tableau 1 présente un récapitulatif des données cliniques Cependant, des déficits hypophysaires congénitaux spora- en fonction des facteurs de transcription hypophysaires impli- diques à l’âge adulte peuvent être découverts chez des patients qués. ne présentant pas de malformation cérébrale associée. C’est donc un faisceau d’arguments (cliniques, biologiques et radiolo- 4.3. Diagnostic biologique giques) qui va orienter le clinicien vers la recherche d’une cause génétique à l’origine des déficits hypophysaires multiples. En 4.3.1. Déficit somatotrope résumé, tout déficit hypophysaire multiple sans cause organique Le diagnostic est suspecté devant un dosage bas d’IGF1 évidente doit faire rechercher une étiologie génétique. corrélé au sexe et à l’âge. Le diagnostic de confirmation nécessite la mise en évidence d’une absence de stimulation 4.2. Corrélation clinique–génotype de l’axe somatotrope (déterminée par la valeur absolue du pic de GH sous stimulation) lors de tests dynamiques : hypo- Elle est définie par le profil de déficits hypophysaires et par glycémie insulinique, test à l’ornithine ou à l’arginine, test l’existence de malformations cérébrales associées : à la GHRH . . . Classiquement, le diagnostic de déficit en GH nécessite l’utilisation d’un test couplé chez l’enfant : test • classiquement, l’absence de malformations cérébrales (post- glucagon-propanolol, arginine-insuline . . . Malgré la reproduc- hypophyse en place, tige pituitaire normale) oriente vers une tibilité imparfaite de ces tests et le caractère arbitraire des mutation de PROP1 ou POU1F1 . Il n’est donc pas nécessaire seuils, le déficit en GH est dit complet si le pic est inférieur de rechercher des anomalies de ces gènes en cas de syndrome à 15 mUI/l (5 ng/ml avec les facteurs de conversion actuels) d’interruption de la tige pituitaire. Un déficit gonadotrope est et partiel si le pic est entre 15 et 30 mUI/l (5 à 10 ng/ml). souvent présent en cas de mutation de PROP1 , jamais en cas Le diagnostic est infirmé si le pic est supérieur à 30 mUI/l de mutation de POU1F1 ; (10 ng/ml). Tableau 1 Phénotypes cliniques, mode de transmission et aspects IRM en fonction des facteurs de transcription hypophysaires impliqués dans le déficit hypophysaire Table 1 Clinical phenotypes, mode of inheritance and MRI presentation in combined pituitary hormone deficiencies involving gene alterations affecting each particular transcription factor Facteur de Mode de Déficit Déficit Déficit Déficit Déficit Hypophyse Posthypophyse Autres transcription transmission GH TSH LH–FSH ACTH PRL POU1F1 Récessif ou Oui Oui Non Non Oui Hypoplasie Normale dominant PROP1 Récessif Oui Oui Oui Variable Oui Hyperplasie puis Normale hypoplasie HESX1 Récessif ou Oui Variable Variable Variable Variable Hypoplasie Ectopique Dysplasie dominant septo-optique LHX3 Récessif Oui Oui Oui Non Oui Variable Normale Anomalie de la rotation de la tête LHX4 Dominant Oui Oui Variable Variable Non Hypoplasie Variable Malformations cérébrales SOX3 Lié à l’X Oui Variable Variable Variable Variable Hypoplasie Ectopique Retard psychomoteur, anomalies du corps calleux F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 13 4.3.2. Déficit corticotrope faite en pratique de fac¸on rétrospective devant un déclenchement Le bilan biologique est évocateur en présence d’une hypo- retardé de la puberté dans le deuxième cas). natrémie (hyponatrémie de dilution à natriurèse conservée). La kaliémie est en général normale (pas de retentissement de la 4.3.5. Déficit lactotrope carence en ACTH sur la sécrétion d’aldostérone). Le diagnostic Le diagnostic est affirmé par une valeur basse de la prolactine est affirmé par un test dynamique de stimulation de l’axe cor- plasmatique et éventuellement par l’absence de réponse après ticotrope : en absence de stimulation de l’axe corticotrope par test de stimulation (métoclopramide par exemple). l’hypoglycémie insulinique (cortisol inférieur à 500 nmol/l en cas d’hypoglycémie inférieure à 0,3 g/l), le diagnostic de déficit 4.4. Diagnostic génétique corticotrope est affirmé. Les dosages biologiques de base pour- ront également retrouver un dosage de cortisol libre urinaire Le diagnostic de certitude est fait après séquenc¸age direct effondré et des taux d’ACTH et cortisol bas à huit heures du des zones codantes des gènes des facteurs de transcription matin : le diagnostic de déficit corticotrope est très probable en hypophysaire. L’étape de sélection des facteurs de transcription cas de taux de cortisol plasmatique inférieur à 300 nmol/l à huit potentiellement impliqués est essentielle et doit être faite à par- heures du matin. tir des données cliniques, biologiques et morphologiques. Notre équipe a ainsi récemment proposé un algorithme d’aide déci- 4.3.3. Déficit thyréotrope sionnelle pour le choix des facteurs de transcription à évaluer Le diagnostic est affirmé devant un taux de TSH normal ou en fonction du tableau clinique présenté par le patient [29] . Une bas, inadapté par rapport à des taux de T4 et T3 libres bas. Aucun représentation modifiée de cet algorithme est fournie en Fig. 1 . test de stimulation n’est nécessaire. Cette stratégie de diagnostic génétique est évidemment appelée à évoluer en fonction des avancées prévisibles au fur et à mesure 4.3.4. Déficit gonadotrope de l’évolution des connaissances. Le diagnostic est affirmé devant un dosage de gonadotro- Il faut insister sur le fait que la réalisation d’un diagnostic phines (LH et FSH) normales ou basses, en tout cas inadaptées génétique peut apporter un bénéfice direct au sujet concerné : par rapport à un taux de testostérone ou estradiol bas. Chez il permet notamment d’anticiper le devenir pubertaire (pas de l’enfant avec un retard pubertaire, un test de stimulation par puberté spontanée en cas de mutation de PROP1 mais puberté LHRH confirme le diagnostic en l’absence de stimulation des normale en cas d’anomalie de POU1F1 ), de prévoir et sur- gonadotrophines. Cependant, il n’est pas possible avant l’âge veiller le risque d’apparition ultérieure, parfois très retardée, de de 16 à 18 ans de différencier le déficit gonadotrope du retard déficits associés (en particulier corticotrope en cas d’anomalie pubertaire simple (la différence entre les deux diagnostics étant de PROP1 ) ; il aide à l’identification d’un syndrome de masse Fig. 1. Algorithme décisionnel simplifié pour déterminer les facteurs de transcription hypophysaire à séquencer. Fig. 1. Simplified algorithm for transcription factor gene genotyping according to phenotype. 14 F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 en cas d’hyperplasie hypophysaire et rend possible le conseil 6.1.2. Chez l’adulte génétique. L’objectif est l’amélioration de la qualité de vie et la nor- malisation de la composition corporelle. Le traitement est 5. Diagnostic différentiel effectué selon les mêmes modalités, mais à des doses beau- coup plus faibles que chez l’enfant et peut être poursuivi de Le clinicien devra, avant d’évoquer une cause congéni- fac¸on prolongée. Les modalités de la thérapeutique pendant la tale, éliminer toutes les causes secondaires d’hypopituitarisme, période de transition de l’adolescence à l’âge adulte ne sont en particulier l’existence d’une tumeur de la région hypo- pas actuellement bien codifiées. La mise en route du traite- thalamohypophysaire compressive (craniopharyngiome . . . ) ou ment peut nécessiter une adaptation des doses de L-thyroxine et cérébrale. Une IRM cérébrale et hypophysaire devra donc d’hydrocortisone. L’administration d’estrogènes par voie orale être systématiquement réalisée pour éliminer ces diagnos- nécessite souvent l’augmentation des doses de GH. tics différentiels ; l’IRM permettra également de préciser les caractéristiques hypophysaires (hypoplasie, hyperplasie . . . ) et 6.2. Déficit corticotrope l’existence d’anomalies de la ligne médiane (posthypophyse ectopique, interruption de tige, anomalies du corps calleux . . . ) Le traitement nécessite une prise d’hydrocortisone à vie. La qui orienteront vers la recherche de mutations d’un facteur de posologie moyenne est de 15 à 25 mg/j. En théorie, un traitement transcription spécifique. complémentaire par fludrocortisone n’est pas nécessaire. La distinction entre retard pubertaire simple et déficit gona- L’essentiel est l’éducation du patient : le traitement ne doit dotrope est difficile devant un retard pubertaire avec des taux bas jamais être arrêté et un régime normosodé doit être systémati- de gonadotrophines et d’hormones sexuelles. Le diagnostic peut quement suivi (contre-indication à un régime sans sel). Les doses être orienté par la présence d’autres déficits hypophysaires (en d’hydrocortisone doivent être doublées en cas de stress majeur faveur d’un déficit hypophysaire) ou de puberté retardée chez ou d’infections. En cas d’impossibilité de prendre ses compri- . . . les ascendants directs (en faveur d’un retard pubertaire simple). més (vomissements ), le patient doit consulter en urgence pour Le diagnostic sera le plus souvent confirmé de fac¸on rétrospec- que l’hydrocortisone lui soit injectée et éviter ainsi une insuf- tive après 18 ans ou plus tôt après induction par de faibles doses fisance surrénalienne aiguë. Le patient doit être porteur d’une d’androgènes chez le garc¸on : en cas de démarrage pubertaire il carte spécifiant qu’il est insuffisant surrénalien, ainsi que le nom s’agira d’un retard pubertaire simple ; en l’absence de démarrage de son médecin référent. spontané, il s’agira d’un déficit gonadotrope à traiter. Un déficit fonctionnel en GH peut être observé en cas de 6.3. Déficit thyréotrope déficit thyréotrope ou de déficit gonadotrope. Ces axes doivent être substitués avant de conclure de fac¸on formelle au diagnostic Le traitement nécessite une substitution par L-thyroxine à de déficit en GH. Ainsi, en cas de retard pubertaire simple, les vie, initiée en fonction du poids. Le traitement est primordial en tests de stimulation de la GH devront être répétés après un court période néonatale pour éviter un déficit intellectuel sévère. Dans traitement par faibles doses d’estradiol ou de testostérone pour l’enfance, la substitution en hormones thyroïdiennes évitera la affirmer de fac¸on définitive le déficit somatotrope. survenue d’une petite taille avec retard osseux sévère. 6.4. Déficit gonadotrope 6. Prise en charge/traitement 6.4.1. Dans l’enfance et en période pubertaire Il n’existe aucun traitement étiologique des déficits hypophy- Le micropénis ou la cryptorchidie peuvent être traités effi- saires multiples congénitaux. Le traitement princeps est fondé cacement par testostérone ou hCG. Le traitement fait ensuite sur l’hormonothérapie substitutive adaptée en fonction des défi- appel aux hormones sexuelles périphériques (testostérone ou cits présentés par le patient. Au vu des différents tableaux composé estroprogestatif) pour permettre l’apparition des carac- existants et surtout de l’évolution chronologique des déficits, tères sexuels secondaires. La décision de traiter doit prendre en une surveillance de l’ensemble des lignées hypophysaires à la compte le vécu de la petite taille et du retard pubertaire par le recherche d’un déficit de survenue tardive devra être effectuée patient, ainsi que le pronostic de taille finale (risque de sou- à intervalles réguliers et de fac¸on prolongée. dure prématurée des cartilages de conjugaison en cas de déficit somatotrope associé non traité). 6.1. Déficit somatotrope 6.4.2. Chez l’adulte 6.1.1. Chez l’enfant Le traitement a pour but de permettre une vie sexuelle L’objectif est l’obtention d’une taille finale proche de la taille satisfaisante et d’éviter le retentissement osseux des carences cible attendue. Le traitement doit être débuté le plus précocement en stéroïdes sexuels (visible également chez l’homme par possible. Il nécessite un traitement par GH humaine biosynthé- absence d’aromatisation de la testostérone). Il fait appel aux tique à doses moyennes de 0,025 à 0,035 mg/kg par semaine, en hormones sexuelles périphériques. En cas de projet parental, un injections sous-cutanées. Le traitement peut être arrêté quand traitement par gonadotrophines devra être proposé (en remplace- l’âge osseux dépasse 15 à 18 ans, selon la taille finale atteinte. ment du traitement par hormones périphériques qui bloqueront F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 15 l’axe gonadotrope, donc la spermatogenèse ou l’ovulation) : par Seules les mutations des facteurs de transcription PROP1, exemple, par injections d’HCG (ou LH) et FSH recombinante, LHX3 et LHX4 ont à ce jour un mode de transmission unique : dans le cadre d’une prise en charge par une équipe spécialisée. récessif pour PROP1 et LHX3 , dominant avec un mécanisme Il faudra souvent plusieurs années sous ce traitement combiné d’haplo-insuffisance pour LHX4 . Dans ce cas, l’enquête géné- pour obtenir une spermatogenèse satisfaisante. tique est plus facile pour déterminer les sujets à risque. Il faut cependant garder à l’esprit que les familles porteuses de ces 6.5. Surveillance mutations sont souvent issues de mariage consanguin. Les muta- tions de POU1F1 et HESX1 peuvent être transmises selon un Elle porte sur deux aspects. mode autosomique dominant ou récessif. Les principaux modes de transmission en fonction des facteurs de transcription impli- qués sont fournis dans le Tableau 1 . 6.5.1. Surveillance de l’apparition de nouveaux déficits. Un bilan annuel doit être réalisé à la recherche de la survenue 8. Pronostic d’un ou plusieurs nouveaux déficits hypophysaires comprenant selon les déficits déjà présents : IGF1, TSH, T4, T3, Prolac- Le pronostic est bon et la qualité de vie est sensiblement tine, LH, FSH, estradiol ou testostérone, cortisol libre urinaire identique à celle d’un sujet non déficitaire si le traitement sub- des 24 heures et ACTH cortisol plasmatiques à huit heures du stitutif est instauré dès le diagnostic posé et adapté correctement. matin. La surveillance doit être prolongée (bien que non codi- En cas de déficits survenant pendant l’enfance ou de retard fiée). Des patients porteurs de mutations de PROP1 ont ainsi pubertaire, le pronostic de taille finale est bon, généralement présenté des déficits corticotropes retardés à l’âge de 40 ans légèrement inférieur à la taille cible prévue. Le pronostic de fer- avec un diagnostic de déficit des autres lignées hypophysaires tilité est variable mais, en général, une grossesse est possible dans l’enfance. après stimulation par les gonadotrophines. Il faut insister sur l’éducation du patient et sur la nécessité pour celui-ci d’être 6.5.2. Surveillance de l’adaptation correcte du traitement suivi par un médecin spécialisé qui assurera une surveillance Axe somatotrope : la surveillance est clinique à la recherche adaptée. de signes de surdosage (gonflement ou paresthésies des extré- . . . mités, sueurs ) ou de sous-dosage (persistance des signes 9. Questions non résolues évocateurs de déficit somatotrope) ; l’objectif biologique est l’obtention d’un taux d’IGF1 dans les normes. Ce taux doit • déterminer l’implication des facteurs de transcription hypo- être contrôlé régulièrement à l’instauration du traitement, puis physaire dans les tumeurs hypophysaires développées chez la surveillance peut être espacée. l’enfant et l’adulte et, plus généralement, déterminer le rôle Axe corticotrope : la surveillance est uniquement clinique à de ces facteurs de transcription hypophysaires chez l’adulte, la recherche de signes de surdosage (prise de poids, vergetures, la plupart continuant à être exprimés après la naissance ; hirsutisme, hypertension artérielle) ou de sous-dosage (signes • déterminer les mécanismes physiopathologiques à l’origine évocateurs d’insuffisance surrénalienne). Aucun dosage biolo- des différences de tableau phénotypique chez l’homme et la gique n’est nécessaire. souris en cas de mutation de PROP1 (déficits gonadotrope et Axe thyréotrope : la surveillance est clinique à la recherche de corticotrope) ; signes de surdosage (signes d’hyperthyroïdie avec tachycardie, • rechercher d’autres facteurs génétiques ou environnementaux . . . sueurs, diarrhée, tremblements des extrémités ) ou de sous- à l’origine des dysplasies septo-optiques (les mutations de . . . dosage (prise de poids, constipation, troubles de la mémoire ) ; HESX1 ne sont retrouvées que dans 5 % des cas) ; la surveillance biologique est basée sur une normalisation des • d’une fac¸on globale, améliorer la compréhension des méca- taux de T3 et T4, le taux de TSH étant non interprétable. nismes d’adressage nucléaire et la physiologie des facteurs de Axe gonadotrope : la surveillance est essentiellement clinique transcription hypophysaire ; par la survenue d’hémorragies de privation (avec un traitement • les facteurs de transcription hypophysaire décrits ne repré- substitutif estroprogestatif) ou d’érections avec rapports sexuels sentent qu’une faible partie des causes d’hypopituitarismes satisfaisants. Les dosages de testostérone ou d’estradiol en fin congénitaux. D’autres facteurs ou voies de synthèse incon- de dose de traitement substitutif ont un intérêt limité. nus à ce jour, doivent donc être impliqués dans ces déficits hypophysaires. 7. Conseil génétique Références L’enquête génétique doit être faite dans l’ensemble de la fra- trie dans un but de dépistage puis de traitement en cas de maladie [1] Andersen B, Pearse 2nd RV, Jenne K, Sornson M, Lin SC, Bartke A, à transmission récessive. Le conseil génétique peut être effectué et al. The ames dwarf gene is required for Pit-1 gene activation. Dev Biol lorsque le diagnostic génétique est confirmé. En cas de trans- 1995;172:495–503. [2] Bhangoo AP, Hunter CS, Savage JJ, Anhalt H, Pavlakis S, Walvoord EC, mission autosomique dominante, le dépistage devra également et al. Clinical case seminar: a novel LHX3 mutation presenting as com- être fait chez les ascendants du cas index, en particulier pour les bined pituitary hormonal deficiency. J Clin Endocrinol Metab 2006;91: déficits hypophysaires à pénétrance variable. 747–53. 16 F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 [3] Camper SA, Saunders TL, Katz RW, Reeves RH. The Pit-1 transcription somatolactotrope cell growth and death: an approach using dominant- factor gene is a candidate for the murine Snell dwarf mutation. Genomics negative Pit-1 mutants. Mol Endocrinol 2006;20:3212–27. 1990;8:586–90. [24] Radovick S, Nations M, Du Y, Berg LA, Weintraub BD, Wondisford FE. [4] Carvalho LR, Woods KS, Mendonca BB, Marcal N, Zamparini AL, A mutation in the pou-homeodomain of Pit-1 responsible for combined Stifani S, et al. A homozygous mutation in HESX1 is associated with evol- pituitary hormone deficiency. Science 1992;257:1115–8. ving hypopituitarism due to impaired repressor–corepressor interaction. J [25] Rainbow LA, Rees SA, Shaikh MG, Shaw NJ, Cole T, Barrett TG, Clin Invest 2003;112:1192–201. et al. Mutation analysis of pouf-1 , prop-1 and hesx-1 show low fre- [5] Chen RP, Ingraham HA, Treacy MN, Albert VR, Wilson L, Rosenfeld MG. quency of mutations in children with sporadic forms of combined pituitary Autoregulation of Pit-1 gene expression mediated by two cis-active pro- hormone deficiency and septo-optic dysplasia. Clin Endocrinol (Oxf) moter elements. Nature 1990;346:583–6. 2005;62:163–8. [6] Dattani MT, Martinez-Barbera JP, Thomas PQ, Brickman JM, Gupta R, [26] Raskin S, Cogan JD, Summar ML, Moreno A, Krishnamani MR, Phillips Martensson IL, et al. Mutations in the homeobox gene HESX1 /HESX1 3rd JA. Genetic mapping of the human pituitary-specific transcriptional associated with septo-optic dysplasia in human and mouse. Nat Genet factor gene and its analysis in familial panhypopituitary dwarfism. Hum 1998;19:125–33. Genet 1996;98:703–5. [7] Deladoey J, Fluck C, Buyukgebiz A, Kuhlmann BV,Eble A, Hindmarsh PC, [27] Reynaud R, Barlier A, Vallette-KasicS, Saveanu A, Guillet MP,Simonin G, et al. “Hot spot” in the PROP1 gene responsible for combined pituitary et al. An uncommon phenotype with familial central hypogonadism caused hormone deficiency. J Clin Endocrinol Metab 1999;84:1645–50. by a novel PROP1 gene mutant truncated in the transactivation domain. [8] Duquesnoy P, Roy A, Dastot F, Ghali I, Teinturier C, Netchine I, et al. J Clin Endocrinol Metab 2005;90:4880–7. Human prop-1 : cloning, mapping, genomic structure. Mutations in familial [28] Reynaud R, Chadli-Chaieb M, Vallette-Kasic S, Barlier A, Sarles J, combined pituitary hormone deficiency. FEBS Lett 1998;437:216–20. Pellegrini-Bouiller I, et al. A familial form of congenital hypopituitarism [9] Fofanova OV, Takamura N, Kinoshita E, Yoshimoto M, Tsuji Y, Peter- due to a PROP1 mutation in a large kindred: Phenotypic and in vitro kova VA, et al. Rarity of Pit-1 involvement in children from Russia with functional studies. J Clin Endocrinol Metab 2004;89:5779–86. combined pituitary hormone deficiency. Am J Med Genet 1998;77:360–5. [29] Reynaud R, Gueydan M, Saveanu A, Vallette-Kasic S, Enjalbert A, [10] Gage PJ, Brinkmeier ML, Scarlett LM, Knapp LT, Camper SA, Mahon KA. Brue T, et al. Genetic screening of combined pituitary hormone defi- The ames dwarf gene, df, is required early in pituitary ontogeny for the ciency: Experience in 195 patients. J Clin Endocrinol Metab 2006;91: extinction of rpx transcription and initiation of lineage-specific cell proli- 3329–36. feration. Mol Endocrinol 1996;10:1570–81. [30] Reynaud R, Saveanu A, Barlier A, Enjalbert A, Brue T. Pituitary hormone [11] Gat-Yablonski G, Klar A, Hirsch D, Eliakim A, Lazar L, Hurvitz H, deficiencies due to transcription factor gene alterations. Growth Horm IGF et al. Three novel mutations in POU1F1 in Israeli patients with combined Res 2004;14:442–8. pituitary hormone deficiency. J Pediatr Endocrinol Metab 2005;18:385–93. [31] Sobrier ML, Maghnie M, Vie-Luton MP, Secco A, Di Iorgi N, Lorini R, [12] Ingraham HA, Chen RP, Mangalam HJ, Elsholtz HP, Flynn SE, Lin CR, et al. Novel HESX1 mutations associated with a life-threatening neona- et al. A tissue-specific transcription factor containing a homeodomain spe- tal phenotype, pituitary aplasia, but normally located posterior pituitary cifies a pituitary phenotype. Cell 1988;55:519–29. and no optic nerve abnormalities. J Clin Endocrinol Metab 2006;91: [13] Kim SS, Kim Y, Shin YL, Kim GH, Kim TU, Yoo HW. Clinical charac- 4528–36. teristics and molecular analysis of Pit-1 , PROP1 , LHX3 , and HESX1 in [32] Sobrier ML, Netchine I, Heinrichs C, Thibaud N, Vie-Luton MP, Van Vliet combined pituitary hormone deficiency patients with abnormal pituitary G, et al. Alu-element insertion in the homeodomain of HESX1 and aplasia mr imaging. Horm Res 2003;60:277–83. of the anterior pituitary. Hum Mutat 2005;25:503. [14] Laumonnier F, Ronce N, Hamel BC, Thomas P, Lespinasse J, [33] Sornson MW, Wu W, Dasen JS, Flynn SE, Norman DJ, O’connell SM, Raynaud M, et al. Transcription factor sox3 is involved in x-linked mental et al. Pituitary lineage determination by the prophet of Pit-1 homeodomain retardation with growth hormone deficiency. Am J Hum Genet 2002;71: factor defective in ames dwarfism. Nature 1996;384:327–33. 1450–5. [34] Tatsumi K, Miyai K, Notomi T, Kaibe K, Amino N, Mizuno Y, et al. Creti- [15] Lebl J, Vosahlo J, Pfaeffle RW, Stobbe H, Cerna J, Novotna D, et al. Auxo- nism with combined hormone deficiency caused by a mutation in the Pit-1 logical and endocrine phenotype in a population-based cohort of patients gene. Nat Genet 1992;1:56–8. with PROP1 gene defects. Eur J Endocrinol 2005;153:389–96. [35] Tatsumi KI, Kikuchi K, Tsumura K, Amino N. A novel PROP1 gene [16] Machinis K, Pantel J, Netchine I, Leger J, Camand OJ, Sobrier ML, et al. mutation (157dela) in Japanese siblings with combined anterior pituitary Syndromic short stature in patients with a germline mutation in the lim hormone deficiency. Clin Endocrinol (Oxf) 2004;61:635–40. homeobox LHX4 . Am J Hum Genet 2001;69:961–8. [36] Thomas PQ, Dattani MT, Brickman JM, Mcnay D, Warne G, Zacha- [17] Mclennan K, Jeske Y, Cotterill A, Cowley D, Penfold J, Jones T, et al. rin M, et al. Heterozygous HESX1 mutations associated with isolated Combined pituitary hormone deficiency in Australian children: clinical congenital pituitary hypoplasia and septo-optic dysplasia. Hum Mol Genet and genetic correlates. Clin Endocrinol (Oxf) 2003;58:785–94. 2001;10:39–45. [18] Mcnay DE, Turton JP, Kelberman D, Woods KS, Brauner R, [37] Turton JP, Reynaud R, Mehta A, Torpiano J, Saveanu A, Woods KS, Papadimitriou A, et al. HESX1 mutations are an uncommon cause of et al. Novel mutations within the POU1F1 gene associated with variable septo-optic dysplasia and hypopituitarism. J Clin Endocrinol Metab 2006. combined pituitary hormone deficiency (cphd). J Clin Endocrinol Metab [19] Mullen RD, Colvin SC, Hunter CS, Savage JJ, Walvoord EC, Bhangoo AP, 2005. et al. Roles of the LHX3 and LHX4 lim-homeodomain factors in pituitary [38] Vallette-Kasic S, Barlier A, Teinturier C, Diaz A, Manavela M, development. Mol Cell Endocrinol 2007. Berthezene F, et al. PROP1 gene screening in patients with multiple [20] Netchine I, Sobrier ML, Krude H, Schnabel D, Maghnie M, Marcos E, pituitary hormone deficiency reveals two sites of hypermutability and et al. Mutations in LHX3 result in a new syndrome revealed by combined a high incidence of corticotroph deficiency. J Clin Endocrinol Metab pituitary hormone deficiency. Nat Genet 2000;25:182–6. 2001;86:4529–35. [21] Olson LE, Tollkuhn J, Scafoglio C, Krones A, Zhang J, Ohgi KA, et al. [39] Vesper AH, Raetzman LT, Camper SA. Role of prophet of Pit-1 Homeodomain-mediated bêta-catenin-dependent switching events dictate (PROP1 ) in gonadotrope differentiation and puberty. Endocrinology cell-lineage determination. Cell 2006;125:593–605. 2006;147:1654–63. [22] Pellegrini-Bouiller I, Belicar P, Barlier A, Gunz G, Charvet JP, Jaquet P, [40] Vieira V, David G, Roche O, De La Houssaye G, Boutboul S, Arbogast L, et al. A new mutation of the gene encoding the transcription factor Pit-1 is et al. Identification of four new PITX2 gene mutations in patients with responsible for combined pituitary hormone deficiency. J Clin Endocrinol axenfeld-rieger syndrome. Mol Vis 2006;12:1448–60. Metab 1996;81:2790–6. [41] Voutetakis A, Argyropoulou M, Sertedaki A, Livadas S, Xekouki P, [23] Pellegrini I, Roche C, Quentien MH, Ferrand M, Gunz G, Thirion S, Maniati-Christidi M, et al. Pituitary magnetic resonance imaging in 15 et al. Involvement of the pituitary-specific transcription factor Pit-1 in patients with PROP1 gene mutations: Pituitary enlargement may ori- F. Castinetti et al. / Annales d’Endocrinologie 69 (2008) 7–17 17 ginate from the intermediate lobe. J Clin Endocrinol Metab 2004;89: [44] Woods KS, Cundall M, Turton J, Rizotti K, Mehta A, Palmer R, et al. 2200–6. Over- and underdosage of sox3 is associated with infundibular hypoplasia [42] Ward RD, Raetzman LT, Suh H, Stone BM, Nasonkin IO, Camper SA. and hypopituitarism. Am J Hum Genet 2005;76:833–49. Role of PROP1 in pituitary gland growth. Mol Endocrinol 2005;19: [45] Wu W, Cogan JD, Pfaffle RW, Dasen JS, Frisch H, O’connell SM, et al. 698–710. Mutations in PROP1 cause familial combined pituitary hormone deficiency. [43] Ward RD, Stone BM, Raetzman LT, Camper SA. Cell proliferation and Nat Genet 1998;18:147–9. vascularization in mouse models of pituitary hormone deficiency. Mol [46] Yamashita T, Moriyama K, Sheng HZ, Westphal H. LHX4 , a lim homeobox Endocrinol 2006;20:1378–90. gene. Genomics 1997;44:144–6. Review Congenital pituitary hormone deficiencies: role of LHX3 /LHX4 genes Expert Rev. Endocrinol. Metab. 3(6), xxx–xxx (2008) F Castinetti *, LHX3 and LHX4 are LIM domain transcription factors involved in the early steps of pituitary R Reynaud *, organogenesis. They are necessary to the proper differentiation of Rathke’s pouch that gives A Saveanu, rise to the anterior pituitary lobe. Mutations of these transcription factors are involved in MH Quentient, congenital hypopituitarism: to date, nine mutations of LHX3 have been reported, responsible for variable pituitary hormone deficiencies, and extra pituitary manifestations including limited F Albarel, A Enjalbert, neck rotation. By contrast, only five LHX4 mutations have been reported, responsible for variable † A Barlier and T Brue hormone deficiencies, and pituitary/intracranial abnormalities. Future investigations will aim to †Author for correspondence better understand human pituitary organogenesis, and to shed light on the interspecies Centre de Recherche en differences in the roles of these transcription factors. neurobiologie et neurophysiologie de Marseille, KEYWORDS : COMBINED PITUITARY HORMONE DElCIENCIES s CONGENITAL HYPOPITUITARISM s HAPLOINSUFlCIENCY s ,(8 Faculté de Médecine Nord, s ,(8 s PITUITARY TRANSCRIPTION FACTORS Centre National de la Recherche Scientifique, Université de la The pituitary gland is a key organ, as it con- LHX4 pituitary transcription factor Méditerranée and Centre de trols several important target glands. In mice, LHX4 is a LIM homeodomain transcription Référence des déficits it is composed of an anterior, a posterior and factor involved in pituitary organogenesis, hypophysaires, Hôpital de la Timone, Assistance Publique an intermediate lobe. The anterior and inter- and is crucial for the genesis and develop- LHX4 Hôpitaux de Marseille, 13385 mediate lobes are derived from the oral ecto- ment of Rathke’s pouch. The human Marseille, France derm that invaginates to form Rathke’s pouch, gene is located on chromosome 1q24: it con- Tel.: which gives rise to six pituitary specific cell sists of six exons, translated into a protein of Fax: lineages that produce growth hormone (GH), 390 amino-acids, as shown in F 1. The [email protected] luteinizing hormone (LH), follicle-stimulating protein contains a central homeodomain with hormone (FSH), thyroid-stimulating hormone DNA-binding activity (encoded by exons 4 *Both authors contributed (TSH), prolactin (PRL) and adrenocorticotropin and 5) and two repeated amino-terminal LIM equally to this work (ACTH). Humans only have anterior and poste- domains (encoded by exons 2 and 3) involved rior pituitary lobes, and a rudimentary structure in protein–protein interactions, which modu- derived from the side of the pouch leading to the late transcriptional activity [7,8] . As a LIM Authorintermediate lobe in mice and other species.Proof homeodomain transcription factor, LHX4 Pituitary organogenesis is a complex multi- shows significant structure similarity with step process, mainly under the control of several LHX3, suggesting a possible overlap between transcription factors [1] . LHX3 and LHX4 are each factor during Rathke’s pouch forma- two LIM domain transcription factors that are tion. This assumption was confirmed by the expressed earlier than the majority of other tran- observation of similar activities of LHX4 and scription factors, such as POU1F1 or PROP1 [2,3] ; LHX3 in assays using pituitary hormone pro- they play a crucial role in pituitary organogen- moter genes, as well as by the role of each factor esis. Four novel mutations of LHX3 and LHX4 in ventral motor neuron differentiation [2,9] . were recently reported as a cause of combined Similar to LHX3 mRNA, LHX4 mRNA has pituitary hormone deficiencies (CPHD) [4–6] , an internal methionine codon in the second drawing attention to the role of these transcrip- LIM domain that could induce a truncated tion factors in the pathophysiology of congenital LHX4 protein, lacking part of the second LIM hypopituitarism. This review will focus on the domain coding region. However, no particu- genetics, molecular targets and the human role lar function of this predicted protein has been of these transcription factors in CPHD. found to date [10] . www.expert-reviews.com 10.1586/17446651.3.6.xxx © 2008 Expert Reviews Ltd ISSN 1744-6651 1 Review Castinetti, Reynaud, Saveanu et al. of these two factors in mouse development [12] . The lhx4 gene is also necessary for lhx3 expression, which is not observed in lhx4 -/- pituitary mice at e12.5; however, the normal pattern of lhx3 expression is established at e 14.5, despite the lack of lhx4 expression [11,12] , suggesting the involvement of other transcrip- tion factors. Prop1 might play a major role in this delayed activa- tion of lhx3 expression by acting independently or upstream of pitx1 and pitx2 [12] . Molecular targets of LHX4 Several studies reported transcriptional targets of LHX4. For instance, a putative LIM homeodomain binding site is located in the proximal human POU1F1 (human gene name for mouse pit-1 gene) upstream regulatory sequence (this binding site is spe- cific to humans, i.e., not present in the mouse promoter of pit-1). Functional studies confirmed that LHX4 was able to bind to the POU1F1 promoter and activate it [13] . Many other promoters are LHX4 targets: we recently reported that LHX4 was able to Figure 1. LHX4 mutations published to date. stimulate both GH and prolactin regulated promoters [6] , whereas other studies reported activation of pituitary target genes, such as Physiological mechanisms of lhx4 in pituitary -glycoprotein, FSH ! and TSH ! [5,14,15] . However, these potential organogenesis in mice target genes only explain some parts of the wide range of pheno- The human LHX4 gene shares high (>95%) identity with types reported in association with LHX4 mutations, suggesting orthologs in other mammals, allowing us to study mouse devel- that there are many more molecular targets of LHX4, unknown opment in a mouse model to better understand human pituitary to date. For instance, a recent study reported the interactions of development. In mice, the lhx4 gene is expressed in the develop- LHX4 with another potential activating transcription factor, sp-1, ing hindbrain, cerebral cortex, pituitary gland and spinal cord. are necessary for mouse early development [16] . The gene is expressed throughout the invaginating pouch at e9.5 and becomes restricted to the future anterior lobe of the pituitary Other physiological implications of LHX4 gland at e12.5, and expression eventually diminishes at e14.5; a Lhx4 is expressed in the motor neurons whose axons emerge low level of expression is maintained at birth and in adulthood ventrally from the neural tube. It seems that lhx3 and lhx4 act in the pituitary [8,11] . tandemly to specify the trajectory of motor axons from the neural Homozygous lhx4 invalidation by homologous recombination tube [9] . In mice however, in contrast with lhx3 , lhx4 expression (lhx4 -/- ) induces an abnormal pituitary phenotype and early death is only transient and mainly localized at the caudal extremity of from lung defects, whereas heterozygous animals ( lhx4 +/- ) seem the spinal cord [17] . To our knowledge, this finding does not seem unaffected [11] . At pituitary level, lhx4-deficient pituitaries ( lhx4 - to have functional significance in humans, in contrast with the /- ) exhibit correct specification of all five hormone-producing cell limited head and neck rotation observed with LHX3 mutations. types, but the expansion of each specialized cell is dramatically LHX4 might also be somehow involved in leukemia patho- reduced. This can be followed during mouse development: at genesis, since LHX4 seems to be deregulated in leukemic cells e12.5, the lhx4 -/- Rathke’s pouchesAuthor are substantially smaller than [18] and inProof the development of human synovial sarcomas, as syn- in the wild type, and there is no evidence of anterior lobe for- ovial sarcoma associated proteins have been found to interact mation. At e14.5, the pouch has progressed to be larger and a with LHX4 [19] . However, precise pathophysiological mechanisms small anterior lobe is discernible. Interestingly, specification of and interacting factors remain unknown to date. each pituitary cell type is present at this time of development. After e14.5, the development of the gland becomes limited and Congenital pituitary hormone deficiency & LHX4 no further expansion is observed [12] . This development in lhx4 - mutations in humans /- mice explains that, at birth, each cell type is present but there Genotypes are fewer differentiated cells than in normal mice. Interestingly, To date, as shown in T 1, only five heterozygous unequivo- the small size of the gland is probably much more attributable to cally functionally defective LHX4 mutations have been reported: an increased rate of apoptosis of precursor cells than to a lower four exonic, including three missense (p.R84C, p.L190R and rate of proliferation [12] . p.A210P) and one frameshift (p.T99fs) mutations [5,6] , and Lhx4 also shares close interactions with other transcription fac- one intronic mutation (intron 4, G-C substitution) [20] . The tors: indeed, specification of pituitary cell types is not observed first human LHX4 mutation, reported in 2001 by the group of in double mutants for lhx4 and prop1 (particularly for gonado- Amselem et al. , consisted in a heterozygous intronic mutation; tropes and thyrotropes) at e14.5, suggesting an overlapping role this mutation affected the AG invariant dinucleotide in intron 2 Expert Rev. Endocrinol. Metab. 3(6), (2008) Congenital pituitary hormone deficiencies: role of LHX3/LHX4 genes Review [5] [5] [5] IV, resulting in two cryptic acceptor splice sites and the [6] [6] [6] [21] Ref. [6,20] izing generation of two types of transcripts, each of them miss- ing part of the homeodomain [20] . We recently reported a novel dysfunctional LHX4 mutation associated with high phenotypic variability: the p.T99fs LHX4 mutation induces a frameshift that predicts a truncated protein onally lacking the homeodomain and the second LIM domain, thus inducing a complete loss of activity on POU1F1 defective mutation polymorphism polymorphism defective mutation defective mutation defective mutation defective mutation studies promoter in functional studies. The functional altera- tion induced by this new mutation could also be due to sion Functional ygous Likely ygous Likely nonsense-mediated RNA decay [6] . The three remain- zygous Functionally ozygous Functionally ing mutations are missense mutations (with substitution terozygous Not performed of conserved amino acids), two of them lacking DNA binding [5] . Functional studies are of importance in LHX4 allelic variants; in a recent paper, we reported two previously l Heterozygous Functionally unpublished allelic variants, p.T90M and p.G370S; we r Heterozygous Functionally considered them as polymorphisms, as our in vitro stud- development ies did not show any functional consequences of these ; FSH: Follicle-stimulating hormone; GH: Growth hormone; LH: Lutein mutations. Even if functional studies are based on trans- fections made in an in vitro system, in heterologous cells and on a limited number of target genes, we considered these two mutations as silent polymorphisms [6] . Another missense allelic variant (p.P366T) was also reported in the literature, but functional studies were not performed, thus, making it difficult to definitely consider it a patho- genic mutation, despite a compatible phenotype. The fact Intracranial imaging Sellar that p.P366T mutation occurred de novo could also be an indication for a pathogenic mutation [21] . What seems remarkable about LHX4 mutations in humans is the fact that that all patients were heterozy- Posterior lobe Normalcallosum Corpus hyperplasia Poor Heterozygous Functi gous for the mutation, suggesting transmission as a dominant trait [5,6,20] . The phenotypes induced by mutations and allelic variants published to date. these heterozygous mutations could either be depen- dant on a dominant negative effect, or a mechanism Pituitary size hyperplasia LHX4 of haploinsufficiency. Machinis et al. were the first to report a mechanism of haploinsufficiency, rather than a eficient or variable phenotype in the patients presenting the mutation dominant negative effect, by studying the effects of an intronic mutation in POU1F1 promoter [13] . Since then, sporadic Author Proofthe two other published functional studies, including ours, confirmed the lack of dominant negative effects of each newly described mutation, a finding that indi- cates a mechanism of haploinsufficiency [5,6] . Indeed, presence of mutant LHX4 did not impair the capacity of normal LHX4 to activate the promoters’ regulatory sequence. Of note, LHX4 heterozygous mice did not seem affected [11,12] , illustrating the limitation of the similarity between humans and mice, although still GH TSH ACTH LH FSH Familial/ useful to better understand human pituitary develop- ment. As homozygous lhx4 invalidation was shown to be lethal in mice [12] , one cannot exclude that homozygous LHX4 mutations are also lethal for humans, which may explain why no homozygous LHX4 mutation has been reported to date as a cause of congenital hypopituitar- ACTH: Adrenocorticotrophic hormone; D: Deficient; D/N: Partially d hormone; N: Normal; TSH: Thyroid-stimulating hormone. p.P366Tp.T90M D D D N D D ? D Sporadic Sporadic Hypoplasia Ectopic Empty sellaEctopic syndrome Chiari Normal Poor He Normal Heteroz p.G370S D D D N Sporadic HypoplasiaNormal Normal Normal Heteroz p.T99fs D D/N N D Familial Hypo or p.A210P D/ND/N D/N D/N Familial Hypoplasia Normal Normal Norma p.L190R D D D ? Sporadic HypoplasiaEctopic Normal Normal Heter p.R84C D D N D Sporadic HypoplasiaEctopic Normal Normal Hetero Table 1. Phenotypic 1. Table and genotypic description of Mutation 607-1G>Cc. Pituitary phenotype D D D D/N Familial Hypoplasia Pituitary and cerebral MRI Ectopic syndrome Chiari Poo ism Transmis in humans. www.expert-reviews.com 3 Review Castinetti, Reynaud, Saveanu et al. Phenotypes in a less-stringent screening of all patients with congenital GH Currently, only five LHX4 mutations have been reported, with a deficiency associated with a least one other pituitary hormone wide range of phenotypes, both in terms of hypopituitarism and deficiency [5] . Other studies did not report any LHX4 muta- of pituitary/cerebral MRI morphology, as shown in T 1 [5,6,20] . tion in selected cohorts of patients with hypopituitarism [22,23] . Mutations were found either sporadic (for two of them) or familial The rarity of congenital hypopituitarism due to LHX4 muta- (for three of them). tions could be owing to the lack of a mutational hot spot in Growth hormone deficiency was present in the majority of the LHX4 gene, or to an inappropriate selection of patients to patients. However, the father of our propositus, carrying the screen. In light of these recent reports, our genetic screening p.Thr99fs mutation, had final normal height without GH treat- strategy should probably be modified. As shown in T 1, there ment, suggesting a delayed somatotroph deficiency [6] . All other are indeed slight differences between mutant phenotypes, as hormone deficiencies (thyrotroph, corticotroph and gonadotroph) endocrine and MRI phenotypes can be highly variable between were inconstant between each mutation, and even within each mutations and between members of a family carrying the same family carrying the same mutation. For example, our propositus mutation. To date, it is thus difficult to define the phenotype and his brother had somatotroph and thyrotroph deficiencies, of patients to screen; however, due to the possibility of delayed whereas their father presented with somatotroph and gonadotroph panhypopituitarism, we think that LHX4 screening could be deficiencies. We also reported the endocrine outcome of the two proposed in patients with familial forms of isolated congenital brothers carrying the first published mutation of LHX4 , show- hypopituitarism (in case of negative screening for mutations in ing that one of them presented gonadotroph deficiency and the classical transcription factors, including POU1F1 or PROP1 ), or other retained a normal gonadotroph axis [6,20] . These reports associated with pituitary or intracranial abnormalities (includ- draw attention on the wide intra- and interfamily variability of ing poorly developed sella turcica) in first line. LHX4 screen- phenotypes and, moreover, underline the risk for developing ing in sporadic cases should probably only be discussed in first delayed pituitary deficiencies, thus requiring prolonged follow- intention in case of congenital hypopituitarism associated with up of patients carrying LHX4 mutations. pituitary and intracranial abnormalities (including poorly devel- Pituitary MRI findings can also be highly variable: the first oped sella turcica). phenotypic description of a patient carrying a mutation of LHX4 included pituitary hypoplasia, ectopic posterior lobe and Chiari LHX3 pituitary transcription factor syndrome. This suggested that, until recently, LHX4 mutation The human LHX3 gene, also known as LIM-3 , pituitary LIM or screening had to be performed in patients carrying hypopi- P-LIM , is located on chromosome 9q34.3. It encodes three dif- tuitarism associated with cerebral abnormalities [20] , which ferent isoforms: LHX3a, LHX3b and M2 LHX3. As shown in represented the main criteria of selection for LHX4 screen- ing. Since then, four other mutations were reported [5,6] , and no pituitary or intracranial abnormalities were systematically associated with LHX4 mutations. For instance, we reported the first case of pituitary hyperplasia associated with LHX4 mutations in the father of the propositus carrying the p.T99fs LHX4 mutation, while all other patients carrying an LHX4 mutation presented with pituitary hypoplasia [6] . Ectopic poste- rior pituitary was observed in only 60% of patients with LHX4 mutations. Extrapituitary abnormalities were also inconstant: whereas one of our patients Authorhad corpus callosum hypoplasia, Proof none of the patients carrying one of the three recently published mutations had cerebral abnormalities [5] . An interesting point that was reported in some of our patients, as well as in the first published LHX4 mutation, was the presence of a poorly developed sella turcica, suggesting the interaction of LHX4 with factors implicated in sella ontogenesis [6,20] . Of note, none of the patients presenting with one of the three other LHX4 mutations harbored this sellar abnormality [5] . LHX4 signaling pathway could, thus, be useful, but not always necessary, for shaping the sphenoid bone. Our screening of 136 patients with hypopituitarism associated with malformations of the brain, pituitary stalk or pituitary pos- terior gland confirmed the rarity of LHX4 allelic variants with a prevalence of less than 3% [6] . In the same way, Pfaeffle et al. reported three mutations in a cohort of 253 patients (1.2%), Figure 2. LHX3 mutations published to date. 4 Expert Rev. Endocrinol. Metab. 3(6), (2008) Congenital pituitary hormone deficiencies: role of LHX3/LHX4 genes Review F 2, LHX3a and LHX3b contain two LIM domains (LIM1 ear development [17] . It persists in the anterior and intermediate and LIM2) and one DNA binding homeodomain but differ in lobes but not in the posterior lobe of the pituitary gland (weeks their amino acid termini. M2 LHX3 is a shorter isoform, which 9 to 19 of development) [17,31] . retains some gene regulatory activity despite the lack of LIM domains [10,14,24–26] . As a consequence, the DNA binding and Pituitary lhx3 transcriptional regulation & gene activating capacity of these isoforms differ significantly: molecular targets LHX3a has a higher level of activity on the target genes com- Several growth factors or transcription factors have been reported pared with the two other isoforms, suggesting different in vivo as being involved in the activation of LHX3. These genes are functions. LHX3a and LHX3b are expressed at the same level in expressed early during Rathke’s pouch formation, such as FGF8 ; mouse thyrotroph and gonadotroph cell lines, whereas LHX3b others ( LHX4 , PITX1 , PITX2 and SOX2 ) are expressed during is predominantly expressed in corticotroph cell lines, making initial cell proliferation [12,30–33] . LHX3a probably dispensable for corticotroph differentiation LHX3a can activate pituitary hormone genes, such as those [2] . Isoform expression in pituitary tumors was also found dif- encoding the subunit of glycoproteins ( GSU), TSH !, FSH ! ferent as somatotroph and lactotroph tumors equally express and GnRH receptor, but does not seem to affect LH ! gene expres- LHX3a and LHX3b, whereas corticotroph tumors mainly sion [15,24,34–36] . It was also reported as a critical factor for con- express LHX3b [27] . tinuous expression of HESX1 [37] , or as an upstream activator of FOXL2 and PIT-1 pituitary transcription factor genes [28] . LHX3 Physiological mechanisms and PIT1 act in a synergistic way in activating PRL , TSH ! and lhx3 in pituitary organogenesis in mice Pit-1 promoters [34] . In mice, lhx3 pituitary expression begins at e9.5, increases until e12.5 with dorsal Rathke’s pouch expression and remains expressed Other physiological implications of LHX3 throughout adulthood. This expression is not restrained to the LHX3 has an important role in the development of spinal motor pituitary but extends to the hindbrain, pineal gland and ventral neurons. In humans, LHX3 is expressed in the neural tube from spinal cord [11,28] . Homozygous lhx3 invalidation by homologous 5 to 19 weeks of development [17] . It plays a role in directing axon recombination ( lhx3 -/- ) induces a severe phenotype: lhx3 -/- mice projections of V2 spinal motor neurons to the ventral side of the die at birth or within 24 h after birth; by contrast, lhx3 +/- mice neural tube. This role is dependent of LIM domains and the appear normal [28] . Embryonic lhx3 -/- mice show normal rudi- homeodomain, via the formation of binding protein complexes mental Rathke’s pouch formation but lack pituitary development with LIM cofactors, NLI or ISL1 [9,38] . from e10.5. Dorsal proliferation stops, but the posterior lobe grows In the inner ear, LHX3 is expressed early during human embry- normally. Dorsal–ventral patterning is modified with a dorsal- onic development (up to 9 weeks), with a similar expression pat- ization of progenitors normally expressed in ventral cell types. tern as SOX2 [31] . In the vestibular epithelium, SOX2 is expressed At birth, lhx3 -/- mice lack the anterior and intermediate lobes of prior to the onset of LHX3 expression. As SOX2 is able to bind pituitary gland [29] . the proximal promoter of LHX3 , it is possible that common In hypomorphic mutant mice, where the pituitary has the molecular pathways of both factors are involved in pituitary and Cre/Cre most significant reduction in LHX3 expression ( Lhx3 ) inner ear development [31,39] . At early human embryonic stage, and in null mice, apoptosis is increased in the ventral portion SOX2 is also expressed in developing cochlear duct without any from e11.5; in contrast, cell proliferation is not affected [29,30] . LHX3 expression. Late human fetal LHX3 expression analysis All pituitary lineages, except corticotrophs, failed to develop in cochlear duct has not been performed yet, but during mouse or maintain. This is probably explained by the fact that initial development, lhx3 is also expressed in cochlear sensory epithelium corticotroph specificationAuthor is spared with tbx19 and neuroD1 fromProof late embryonic stage to postnatal, with overlapping temporal expression at e12.5 and e14.5, respectively; thereafter, increased expression with Sox2 [40] . It is also of importance to note that apoptosis, owing to deficient lhx3 expression, probably leads to SOX2 has also been shown to be expressed in the pituitary and the lack of corticotroph cells at birth. The fact that neonatal neural tube in regions overlapping LHX3 expression in addition adrenal glands are also hypoplastic highly suggests that corti- to the inner ear, confirming the straight interactions between cotroph deficiency is also present during mouse development [28] . these two factors during pituitary ontogenesis [41] . In the intermediate lobe, differentiation of melanotroph cells is also impaired [29] . In mouse development, lhx3 is therefore Congenital pituitary hormone deficiency & LHX3 critical for Rathke’s pouch formation, correct dorsal–ventral mutations in humans patterning, late lineage proliferation, differentiation and cell Genotypes survival. Human LHX3 mutations remain rare. Since the first reported case in 2000 to novel mutations reported in 2007, several teams tried lhx3 during early human pituitary development unsuccessfully to find any LHX3 mutation in large series [42–45] . LHX3 is expressed during Rathke’s pouch formation at weeks 5 to In our series of 109 patients without extrapituitary abnormali- 6 of development. This expression is consecutive to SOX2 expres- ties, no LHX3 gene defect was found, even with a neck rotation sion, another transcription factor involved in pituitary and inner deficit, whereas 20 patients had PROP1 mutations and one had a www.expert-reviews.com 5 Review Castinetti, Reynaud, Saveanu et al. [45] [4] [4] [4] [4] POU1F1 gene defect patients . Pfaeffle et [31] [31] [47] [46] [46] al. identified 1.2% LHX3 mutations among 302 pedigrees of CPHD and no mutation in 40 isolated GH deficiency pedigrees [4] . To date, only nine LHX3 mutations were reported, all inherited in an autosomal reces- main sive manner. cated protein Phenotypes Truncated protein Truncated lacking LIM2 and HD lacking LIM and HD changing from LIM2 domain TD LIM2 LIM1 Netchine et al. reported the first LHX3 Genotype Ref. mutations in two unrelated families. As shown in T 2, patients’ phenotypes were characterized by abnormal neck rotation ozygous protein from Truncated and CPHD with normal corticotroph axis. Transmission Altered domain Homozygous HD Homozygous protein Truncated Homozygous protein from Truncated MRI demonstrated pituitary hypoplasia in two patients, while another had an enlarged pituitary gland. Genetic ana lysis in the first family revealed a homozygous missense mutation in exon 3 (Y116C LHX3a), in ; FSH: follicle-stimulating hormone; GH: Growth hormone; HD: the second LIM domain. The second fam- Mental retardation Yes Homozygous Complete deletion ily was carrying a 23 bp deletion, leading to a severely truncated protein lacking the entire homeodomain [46] . Functional stud- loss Mildies confirmed Homozygous protein from Truncated the deleterious effect of these mutations [17,46] . At least 5 years later, three reports described novel mutations, as shown in T 2 and F !"# 2 [4,31,47] . Phenotypes associated with these new mutations added interesting data to the variable phenotypes of patients carrying LHX3 mutations; in Cervical spine Hearing lordosis lordosis kyphosis, spinal stenosis particular, mild-to-severe hearing impair- ments in the two recently described families: Rajab et al. described two LHX3 mutations mutations and allelic variants published to date. Neck rotation associated with hearing loss and analyzed LHX3 transcript expression in inner ear: LHX3 Phenotype they reported an early expression of LHX3 ion domain; TSH: Thyroid-stimulating hormone. in vestibular epithelium following SOX2 eficient or variable phenotype in the patients presenting the mutation Pituitary size expression and suggested a role of LHX3 in late cochlear sensory epithelium devel- Author Proofopment [31] . Comparison between the phe- Familial/ Familial/ sporadic notypes of the nine reported mutations revealed variable incidence of corticotroph deficiency that was present in 50% patients. FSH While the majority of the patients presented neonatal events, such as prolonged jaundice, respiratory distress or hypoglycemia, one with a W224X mutation was only diagnosed after perinatal period. Rigid cervical spine with limited neck rotation was present in all but one patient. Mechanisms underly- GH TSH ACTH LH D D N ? Familial Aplasia Limited Loss cervical of D D D Ding this Familial Hypoplasiasevere limitation Thoracolum-bar of cervical rotation are not clearly established. Abnormal steep- ness of cervical spine has been reported but no vertebral or spinal malformations were g.159delT D D N D Familial Hyperplasia Limited Yes Homozygous Trun Del.23bp D Dp.A210V N Dp.E173X D D D Familial D/N D Hyperplasia Limited D D Familial D Hyperplasia Limited Loss cervical of Familial Hypoplasia Limited Severe Yes Homozygous Table 2. PhenotypicTable and genotypic description of Mutation p.Y116C D D N D Familial Hypoplasia Limited Mild Homozygous LIM2 do p.W224X D D N D Familial Normal Normal Complete LHX3 deletion Loss Exon 2 5 to p.K50X D D D/N ? Sporadic Hypoplasia Thoracic kyphosis Severe Hom Homeodomain; LH: Luteinizing hormone; N: Normal; TD: Transactivat ACTH: Adrenocorticotrophic hormone; D: Deficient; D/N: Partially d visualized on cervical x-rays and spinal MRI 6 Expert Rev. Endocrinol. Metab. 3(6), (2008) Congenital pituitary hormone deficiencies: role of LHX3/LHX4 genes Review [4,46] . EMG was normal or suggested a cervical anterior horn cell be systematically performed to ascertain the link between a given disease process [46] . Spectrum of LHX3 mutations also includes allelic variant and the observed phenotype, keeping in mind that inconstant mental retardation, whereas no intracranial abnormali- such functional studies are made in a ‘virtual’ in vitro environ- ties have been reported to date. Limited head and neck rotation ment not entirely transposable to human pituitary development. is probably due to the involvement of LHX3 in motor neuron Thus, precautions have to be taken before considering an allelic development. variant as a defective mutation. Pituitary MRI findings can also be highly variable: approxi- Human LHX3 mutations also remain rare, with a recessive mately 50% of patients present with pituitary hypoplasia or inheritance. Hypopituitarism is frequently, but not always, hyperplasia. Interestingly, one patient presented a hypointense revealed by neonatal events. Pathophysiological mechanisms pituitary lesion consistent with a microadenoma, but no evident leading to vertebral malformations or inner ear disorders remain correlation between LHX3 and adenoma pathogenesis has been only partially understood. Most mutations are associated with reported to date. severe pheno types, including inconstant corticotroph deficit. Not In light of the recently described LHX3 mutations, as for LHX4 , surprisingly, mutations of the genes encoding such transcription it is hard to determine which patients should be screened. It seems factors involved in the early steps of pituitary ontogenesis and to make sense to screen at least all patients with CPHD associated expressed in several nonpituitary tissues, induce a wide range of with limited head and neck rotation and/or hearing impairment. endocrine and extrapituitary phenotypes. However, it is important to note that the degree of hearing loss reported in patients with LHX3 mutations was extremely vari- Five-year view able, and may be so mild that it may be overlooked in some cases. Future research should focus on the differences between mouse As cases can be either sporadic or familial, presence of several and human pituitary development. It is likely that other targets affected members in the same family does not seem to be a good will be determined, thus allowing us to better understand the screening criterion. wide range of phenotypes induced by LHX3 and LHX4 muta- tions. In the same way, there are still a number of cases of congeni- Expert commentary tal hypopituitarism for which screening for mutations in ‘classical’ Mutations of LHX4 appear as rare dominantly transmitted muta- pituitary transcription factors remained negative: in these cases, tions, leading to variable forms of congenital hypopituitarism it is likely that other transcription factors, signaling proteins or and extrapituitary malformations. However, due to the possible diffusible growth factors are involved and at least some of them delayed appearance of hormone deficiencies, mutations of this will most probably be found over the next few years. gene deserve screening. The wide range of phenotypes induced Another important part of research projects is to gain a better by these mutations makes it difficult to determine the best profile understanding of pituitary stem cells [48,49] . A study reported that of patients to screen: mutations are frequently associated with adult mouse anterior pituitary contained a side-cell population intracranial abnormalities or poorly developed sella turcica and with stem cell characteristics. Interestingly, these cells expressed these criteria should prompt screening. Functional studies should lhx3 and/or lhx4 transcription factors. However, the precise Key issues s LHX4 and LHX3 are LIM homeodomain transcription factors involved in pituitary organogenesis, necessary for specification and proliferation of anterior pituitary hormone cells lineages. Lack of expression of these transcription factors induces increased apoptosis of these cells, thus leading to combined pituitary hormone deficiencies (CPHD). s To date, only five LHX4 mutations have been reported as unequivocally responsible for CPHD, with a sporadic or familial setting. Nine homozygous LHX3 mutationsAuthor have been described in CPHD patients withProof vertebral anomalies and/or or hearing loss. s There is a wide range of phenotypes induced by LHX4 mutations, in terms of pituitary hormone deficiency profiles, as well as intracranial abnormalities. This variability is observed between mutations, but also between the members of a family carrying the same mutation. s LHX3 phenotype spectrum is also variable, including inconstant corticotroph deficiency, limited head and neck limited rotations and sometimes hearing loss or mental retardation. s Mutations of LHX4 responsible of CPHD are heterozygous; thus, transmitted as a dominant trait. The mechanism whereby these heterozygous mutations induce CPHD is probably haploinsufficiency rather than dominant negative effect. Mutations of LHX3 are homozygous; they include complete gene deletion, as well as missense mutations, which do not necessarily alter LIM domains or homeodomain. s Molecular targets of LHX3 and LHX4 are numerous, including POU1F1 , growth hormone, prolactin, thyroid-stimulating hormone- , follicle-stimulating hormone- regulated and !-subunit promoters. However, there are probably many yet unknown targets, able to explain the wide range of phenotypes induced by mutations of these transcription factors. s The main problem in screening for mutations in LHX3 and LHX4 is to determine which phenotypic profile should be screened: LHX4 screening should be performed systematically at least in patients with congenital hypopituitarism associated with intracranial abnormalities or poorly developed sella turcica, and LHX3 screening in patients with CPHD associated with limited neck rotation and/or hearing impairments. www.expert-reviews.com 7 Review Castinetti, Reynaud, Saveanu et al. physiological role of these transcription factors in these cells Financial & competing interests disclosure remains unclear [50] . Future studies will aim at better defining The GENHYPOPIT network for the study of genetic determinants of this stem cell population, which may prove to play a major role hypopituitarism, coordinated by T Brue ([email protected]), was in the therapy of hypopituitarism and may help to determine the funded by the Groupement d’Intérêt Scientifique Institut des Maladies Rares role of lhx4 and/or lhx3 in the development of these cells. (GISMR0201) and the Programe Hospitalier de Recherche Clinique (PHRC 25/2003, French Ministry of Health). The authors have no other relevant Acknowledgements affiliations or financial involvement with any organization or entity with The authors would like to thank participating patients and all clinicians a financial interest in or financial conflict with the subject matter or materi- who sent clinical data and samples for screening in the ‘Genhypopit’ als discussed in the manuscript apart from those disclosed. database. No writing assistance was utilized in the production of this manuscript. References J. Biol. Chem. 276, 36311–36319 (2001). interacts with the lim homeobox protein lhx4. Oncogene 27, 653–662 (2008). 1 Castinetti F, Reynaud R, Saveanu A et al. 11 Sheng HZ, Moriyama K, Yamashita T Clinical and genetic aspects of combined et al. Multistep control of pituitary 20 Machinis K, Pantel J, Netchine I et al. pituitary hormone deficiencies. Ann. organogenesis. Science 278, 1809–1812 Syndromic short stature in patients with a Endocrinol. (Paris) 69, 7–17 (2008). (1997). germline mutation in the LIM homeobox lhx4. Am. J. Hum. Genet. 69, 961–968 2 Mullen RD, Colvin SC, Hunter CS et al. 12 Raetzman LT, Ward R, Camper SA. Lhx4 (2001). Roles of the lhx3 and lhx4 lim- and prop1 are required for cell survival and homeodomain factors in pituitary expansion of the pituitary primordia. 21 Tajima T, Hattori T, Nakajima T, development. Mol. Cell. Endocrinol. Development 129, 4229–4239 (2002). Okuhara K, Tsubaki J, Fujieda K. A novel 265–266, 190–195 (2007). 13 Machinis K, Amselem S. Functional missense mutation (p366t) of the lhx4 gene causes severe combined pituitary hormone 3 Reynaud R, Saveanu A, Barlier A, Enjalbert relationship between lhx4 and pou1f1 in deficiency with pituitary hypoplasia, A, Brue T. Pituitary hormone deficiencies light of the lhx4 mutation identified in ectopic posterior lobe and a poorly due to transcription factor gene alterations. patients with pituitary defects. J. Clin. developed sella turcica. Endocr. J. 54, Growth Horm. IGF Res . 14, 442–448 Endocrinol. Metab. 90, 5456–5462 (2005). 637–641 (2007). (2004). 14 Sloop KW, Parker GE, Hanna KR, Wright 22 Melo ME, Marui S, Carvalho LR et al. 4 Pfaeffle RW, Savage JJ, Hunter CS et al. HA, Rhodes SJ. Lhx3 transcription factor Hormonal, pituitary magnetic resonance, Four novel mutations of the lhx3 gene mutations associated with combined lhx4 and hesx1 evaluation in patients with cause combined pituitary hormone pituitary hormone deficiency impair the hypopituitarism and ectopic posterior deficiencies with or without limited neck activation of pituitary target genes. Gene pituitary lobe. Clin. Endocrinol. (Oxf.) 66, rotation. J. Clin. Endocrinol. Metab. 92, 265, 61–69 (2001). 95–102 (2007). 1909–1919 (2007). 15 West BE, Parker GE, Savage JJ et al. 23 Vieira TC, Boldarine VT, Abucham J. 5 Pfaeffle RW, Hunter CS, Savage JJ et al. Regulation of the follicle-stimulating Molecular analysis of prop1, pit1, hesx1, Three novel missense mutations within the hormone gene by the lhx3 lhx3, and lhx4 shows high frequency of lhx4 gene are associated with variable LIM-homeodomain transcription factor. prop1 mutations in patients with familial pituitary hormone deficiencies. J. Clin. Endocrinology 145, 4866–4879 (2004). forms of combined pituitary hormone Endocrinol. Metab. 93, 1062–1071 (2008). 16 Liu S, Luo H, Liu J, McNeilly AS, Cui S. deficiency. Arq. Bras. Endocrinol. Metabol . Specificity protein 1 (sp1) plays role in 6 Castinetti F, Saveanu A, Reynaud R et al. 51, 1097–1103 (2007). A novel dysfunctional lhx4 mutation with regulating lim homeodomain transcription 24 Sloop KW, Meier BC, Bridwell JL, high phenotypic variability in patients with factor lhx4 gene expression. Biochem. Parker GE, Schiller AM, Rhodes SJ. hypopituitarism. J. Clin. Endocrinol. Biophys. Res. Commun. 366, 36–41 (2008). Differential activation of pituitary hormone Metab. 93, 2790–2799 (2008).Author17 Sobrier ML, Attie-Bitach T, NetchineProof I, genes by human lhx3 isoforms with Encha-Razavi F, Vekemans M, Amselem S. 7 Yamashita T, Moriyama K, Sheng HZ, distinct DNA binding properties. Mol. Pathophysiology of syndromic combined Westphal H. Lhx4 , a LIM homeobox gene. Endocrinol. 13, 2212–2225 (1999). Genomics 44, 144–146 (1997). pituitary hormone deficiency due to a lhx3 defect in light of lhx3 and lhx4 expression 25 Sloop KW, Showalter AD, Von Kap- 8 Liu Y, Fan M, Yu S et al. Cdna cloning, during early human development. Gene Herr C, Pettenati MJ, Rhodes SJ. Analysis chromosomal localization and expression Expr. Patterns 5, 279–284 (2004). of the human lhx3 neuroendocrine pattern analysis of human LIM-homeobox transcription factor gene and mapping to 18 Kawamata N, Sakajiri S, Sugimoto KJ, gene lhx4 . Brain Res. 928, 147–155 (2002). the subtelomeric region of chromosome 9. Isobe Y, Kobayashi H, Oshimi K. A novel 9 Sharma K, Sheng HZ, Lettieri K et al. LIM Gene 245, 237–243 (2000). chromosomal translocation t(1;14) homeodomain factors lhx3 and lhx4 assign (q25;q32) in pre-b acute lymphoblastic 26 Yaden BC, Savage JJ, Hunter CS, subtype identities for motor neurons. Cell leukemia involves the lim homeodomain Rhodes SJ. DNA recognition properties of 95, 817–828 (1998). protein gene, lhx4 . Oncogene 21, 4983– the lhx3b LIM homeodomain transcription 10 Sloop KW, Dwyer CJ, Rhodes SJ. An 4991 (2002). factor. Mol. Biol. Rep. 32, 1–6 (2005). isoform-specific inhibitory domain 19 de Bruijn DR, van Dijk AH, Willemse MP, 27 Schmitt S, Biason-Lauber A, Betts D, regulates the lhx3 LIM homeodomain van Kessel AG. The C terminus of the Schoenle EJ. Genomic structure, factor holoprotein and the production of a synovial sarcoma-associated ssx proteins chromosomal localization, and expression functional alternate translation form. pattern of the human LIM-homeobox3 8 Expert Rev. Endocrinol. Metab. 3(6), (2008) Congenital pituitary hormone deficiencies: role of LHX3/LHX4 genes Review (lhx3 ) gene. Biochem. Biophys. Res. 110, 237–249 (2002). USA 105, 6332–6337 (2008). Commun. 274, 49–56 (2000). 39 Yaden BC, Garcia M, 3rd, Smith TP, 50 Chen J, Hersmus N, Van Duppen V, 28 Sheng HZ, Zhadanov AB, Mosinger B Jr Rhodes SJ. Two promoters mediate Caesens P, Denef C, Vankelecom H. The et al. Specification of pituitary cell lineages transcription from the human lhx3 gene: adult pituitary contains a cell population by the lim homeobox gene lhx3. Science Involvement of nuclear factor i and displaying stem/progenitor cell and early 272, 1004–1007 (1996). specificity protein 1. Endocrinology 147, embryonic characteristics. Endocrinology 29 Ellsworth BS, Butts DL, Camper SA. 324–337 (2006). 146, 3985–3998 (2005). Mechanisms underlying pituitary 40 Hume CR, Bratt DL, Oesterle EC. hypoplasia and failed cell specification in Expression of lhx3 and sox2 during mouse Affiliations lhx3-deficient mice. Dev. Biol . 313, inner ear development. Gene Expr. Patterns 118–129 (2008). 7, 798–807 (2007). s F Castinetti 30 Charles MA, Suh H, Hjalt TA, Drouin J, 41 Kelberman D, de Castro SC, Huang S Centre de Recherche en neurobiologie et Camper SA, Gage PJ. Pitx genes are et al. Sox2 plays a critical role in the neurophysiologie de Marseille (CRN2M), required for cell survival and lhx3 pituitary, forebrain, and eye during human UMR6231, Faculté de Médecine Nord, activation. Mol. Endocrinol . 19, 1893–1903 embryonic development. J. Clin. Centre National de la Recherche (2005). Endocrinol. Metab. 93, 1865–1873 (2008). Scientifique, Université de la Méditerranée and Centre de Référence des déficits 31 Rajab A, Kelberman D, de Castro SC et al. 42 Sloop KW, Walvoord EC, Showalter AD, hypophysaires, Hôpital de la Timone, Novel mutations in lhx3 are associated Pescovitz OH, Rhodes SJ. Molecular Assistance Publique Hôpitaux de Marseille, with hypopituitarism and sensorineural analysis of lhx3 and prop-1 in pituitary 13385 Marseille, France hearing loss. Hum. Mol. Genet . (2008). hormone deficiency patients with posterior s R Reynaud 32 Takuma N, Sheng HZ, Furuta Y et al. pituitary ectopia. J. Clin. Endocrinol. Centre de Recherche en neurobiologie et Formation of Rathke’s pouch requires dual Metab. 85, 2701–2708 (2000). neurophysiologie de Marseille (CRN2M), induction from the diencephalon. 43 Kim SS, Kim Y, Shin YL, Kim GH, UMR6231, Faculté de Médecine Nord, Development 125, 4835–4840 (1998). Kim TU, Yoo HW. Clinical characteristics Centre National de la Recherche and molecular analysis of pit1, prop1, lhx3, 33 Tremblay JJ, Lanctot C, Drouin J. The Scientifique, Université de la Méditerranée and hesx1 in combined pituitary hormone pan-pituitary activator of transcription, and Centre de Référence des déficits deficiency patients with abnormal pituitary ptx1 (pituitary homeobox 1), acts in hypophysaires, Hôpital de la Timone, MR imaging. Horm. Res . 60, 277–283 synergy with sf-1 and pit1 and is an Assistance Publique Hôpitaux de Marseille, (2003). upstream regulator of the lim- 13385 Marseille, France homeodomain gene lim3/lhx3. Mol. 44 Arrigo T, Wasniewska M, De Luca F et al. s A Saveanu Endocrinol . 12, 428–441 (1998). Congenital adenohypophysis aplasia: Centre de Recherche en neurobiologie et Clinical features and analysis of the 34 Bach I, Rhodes SJ, Pearse RV 2nd et al. neurophysiologie de Marseille (CRN2M), transcriptional factors for embryonic P-LIM, a LIM homeodomain factor, is UMR6231, Faculté de Médecine Nord, pituitary development. J. Endocrinol. Invest . expressed during pituitary organ and cell Centre National de la Recherche 29, 208–213 (2006). commitment and synergizes with pit-1. Scientifique, Université de la Méditerranée Proc. Natl Acad. Sci. USA 92, 2720–2724 45 Reynaud R, Gueydan M, Saveanu A et al. and Centre de Référence des déficits (1995). Genetic screening of combined pituitary hypophysaires, Hôpital de la Timone, 35 Granger A, Bleux C, Kottler ML, Rhodes hormone deficiency: experience in 195 Assistance Publique Hôpitaux de Marseille, SJ, Counis R, Laverriere JN. The patients. J. Clin. Endocrinol. Metab. 91, 13385 Marseille, France LIM-homeodomain proteins isl-1 and lhx3 3329–3336 (2006). and act with steroidogenic factor 1 to enhance 46 Netchine I, Sobrier ML, Krude H et al. Laboratoire de Biochimie-Biologie gonadotrope-specific activity of the Mutations in lhx3 result in a new syndrome Moléculaire, Hôpital Conception, gonadotropin-releasingAuthor hormone receptor revealed by combined pituitaryProof hormone Marseille, France gene promoter. Mol. Endocrinol . 20, deficiency. Nat. Genet . 25, 182–186 s MH Quentient 2093–2108 (2006). (2000). Centre de Recherche en neurobiologie et 36 McGillivray SM, Bailey JS, Ramezani R, 47 Bhangoo AP, Hunter CS, Savage JJ et al. neurophysiologie de Marseille (CRN2M), Kirkwood BJ, Mellon PL. Mouse gnrh Clinical case seminar: a novel lhx3 UMR6231, Faculté de Médecine Nord, receptor gene expression is mediated by the mutation presenting as combined pituitary Centre National de la Recherche lhx3 homeodomain protein. Endocrinology hormonal deficiency. J. Clin. Endocrinol. Scientifique, Université de la Méditerranée 146, 2180–2185 (2005). Metab. 91, 747–753 (2006). and Centre de Référence des déficits hypophysaires, Hôpital de la Timone, 37 Ellsworth BS, Egashira N, Haller JL et al. 48 Fauquier T, Rizzoti K, Dattani M, Assistance Publique Hôpitaux de Marseille, Foxl2 in the pituitary: molecular, genetic, Lovell-Badge R, Robinson IC. Sox2- 13385 Marseille, France and developmental analysis. Mol. expressing progenitor cells generate all of Endocrinol . 20, 2796–2805 (2006). the major cell types in the adult mouse s F Albarel Centre de Recherche en neurobiologie et 38 Thaler JP, Lee SK, Jurata LW, Gill GN, pituitary gland. Proc. Natl Acad. Sci. USA neurophysiologie de Marseille (CRN2M), Pfaff SL. LIM factor lhx3 contributes to 105, 2907–2912 (2008). UMR6231, Faculté de Médecine Nord, the specification of motor neuron and 49 Gleiberman AS, Michurina T, Encinas JM Centre National de la Recherche interneuron identity through cell-type- et al. Genetic approaches identify adult Scientifique, Université de la Méditerranée specific protein–protein interactions. Cell pituitary stem cells. Proc. Natl Acad. Sci. and Centre de Référence des déficits www.expert-reviews.com 9 Review Castinetti, Reynaud, Saveanu et al. hypophysaires, Hôpital de la Timone, Marseille, France s T Brue Assistance Publique Hôpitaux de Marseille, s A Barlier Centre de Recherche en neurobiologie 13385 Marseille, France Centre de Recherche en neurobiologie et et neurophysiologie de Marseille s A Enjalbert neurophysiologie de Marseille (CRN2M), (CRN2M), UMR6231, Faculté de Centre de Recherche en neurobiologie et UMR6231, Faculté de Médecine Nord, Médecine Nord, Centre National de la neurophysiologie de Marseille (CRN2M), Centre National de la Recherche Recherche Scientifique, Université de la UMR6231, Faculté de Médecine Nord, Scientifique, Université de la Méditerranée Méditerranée and Centre de Référence des Centre National de la Recherche and Centre de Référence des déficits déficits hypophysaires, Hôpital de la Scientifique, Université de la Méditerranée hypophysaires, Hôpital de la Timone, Timone, Assistance Publique Hôpitaux de and Centre de Référence des déficits Assistance Publique Hôpitaux de Marseille, Marseille, 13385 Marseille, France hypophysaires, Hôpital de la Timone, 13385 Marseille, France Tel.: Assistance Publique Hôpitaux de Marseille, and Fax: 13385 Marseille, France Laboratoire de Biochimie-Biologie [email protected] and Moléculaire, Hôpital Conception, Laboratoire de Biochimie-Biologie Marseille, France Moléculaire, Hôpital Conception, Author Proof 10 Expert Rev. Endocrinol. Metab. 3(6), (2008) G Model MCE 7393 1–16 ARTICLE IN PRESS Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 1 Contents lists available at ScienceDirect Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce 1 Review 2 Molecular mechanisms of pituitary organogenesis: In search of novel regulatory 3 genes a a,b c d a a a 4 S.W. Davis , F. Castinetti , L.R. Carvalho , B.S. Ellsworth , M.A. Potok , R.H. Lyons , M.L. Brinkmeier , e f a f c c b 5 L.T. Raetzman , P. Carninci , A.H. Mortensen , Y. Hayashizaki , I.J.P. Arnhold , B.B. Mendonca , T. Brue , a,∗ 6 S.A. Camper 7 a University of Michigan Medical School, Ann Arbor, MI 41809-0618, USA 8 b Center for Research in Neurobiology and Neurophysiology of Marseille, Timone Hospital, Marseille, France 9 c Hormone and Molecular Genetics Laboratory, Endocrinology, Clinical Hospital, Faculty of Medicine of the University of São Paulo, Brazil 10 d Southern Illinois University School of Medicine, Department of Physiology, Carbondale, IL 62901, USA 11 e University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA 12 f RIKEN Genomic Sciences Center, Yokohama Institute, Japan 13 14 article info abstract 15 16 Keywords: Defects in pituitary gland organogenesis are sometimes associated with congenital anomalies that affect 17 Beta helix-loop-helix (bHLH) head development. Lesions in transcription factors and signaling pathways explain some of these devel- 18 High mobility group (HMG) opmental syndromes. Basic research studies, including the characterization of genetically engineered 19 T-box mice, provide a mechanistic framework for understanding how mutations create the clinical character- 20 Forkhead istics observed in patients. Defects in BMP, WNT, Notch, and FGF signaling pathways affect induction 21 Q1 Hypopituitarism and growth of the pituitary primordium and other organ systems partly by altering the balance between signaling pathways. The PITX and LHX transcription factor families influence pituitary and head devel- opment and are clinically relevant. A few later-acting transcription factors have pituitary-specific effects, including PROP1, POU1F1 (PIT1), and TPIT (TBX19), while others, such as NeuroD1 and NR5A1 (SF1), are syndromic, influencing development of other endocrine organs. We conducted a survey of genes tran- scribed in developing mouse pituitary to find candidates for cases of pituitary hormone deficiency of unknown etiology. We identified numerous transcription factors that are members of gene families with roles in syndromic or non-syndromic pituitary hormone deficiency. This collection is a rich source for future basic and clinical studies. © 2009 Published by Elsevier Ireland Ltd. 22 Contents 36 23 1. Introduction ...... 00 24 1.1. Human growth insufficiency ...... 00 25 1.2. Are mouse studies informative for clinical endocrinologists? ...... 00 26 1.3. Cell–cell signaling plays a critical role in pituitary organogenesis ...... 00 27 1.4. Transcription factor regulation of pituitary development ...... 00 28 2. Results and discussion ...... 00 29 2.1. Prioritizing genes for molecular studies in human patients ...... 00 30 2.2. Gene discovery to identify new candidate genes ...... 00 31 2.3. The PITX gene family ...... 00 32 2.4. The LHX family ...... 00 33 2.5. Forkhead factors are essentialUNCORRECTED for diverse developmental processes ...... PROOF ...... 00 34 2.6. Forkhead factors in pituitary development ...... 00 35 2.7. Basic helix-loop-helix family is highly represented in the developing pituitary gland ...... 00 ∗ Corresponding author at: Department Human Genetics, University of Michigan, 4909 Buhl Bldg., 1241 Catherine St., Ann Arbor, MI 48109-5618, USA. Tel.: +1 734 763 0682; fax: +1 734 763 5831. E-mail address: [email protected] (S.A. Camper). 0303-7207/$ – see front matter © 2009 Published by Elsevier Ireland Ltd. doi: 10.1016/j.mce.2009.12.012 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 2 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 2.8. High mobility group genes expressed in the developing pituitary ...... 00 2.9. Several T-box genes are expressed in the developing pituitary gland ...... 00 3. Future directions ...... 00 Acknowledgments ...... 00 References ...... 00 37 1. Introduction not always efficacious for some idiopathic short stature patients or 71 for other problems associated with syndromic pituitary hormone 72 38 1.1. Human growth insufficiency deficiency (i.e. septo-optic dysplasia and other severe craniofacial 73 abnormalities) ( Bryant et al., 2007; Hintz, 2005; Zucchini, 2008; 74 39 Height of 2 or more standard deviations (SD) below the mean Zucchini et al., 2008; Cohen et al., 2008 ). Thus, treatment of children 75 40 for age and sex is defined as short stature. Metabolic or endocrine with pituitary hormone deficiency can be challenging, as well as 76 41 disorders usually cause proportionate short stature, while skeletal expensive. 77 42 defects often cause disproportionate short stature ( Weedon and 43 Frayling, 2008; Rimoin et al., 2007; Cha et al., 2004a; Jorge et al., 1.2. Are mouse studies informative for clinical endocrinologists? 78 44 2007 ). The sitting height to standing height ratio can be used to 45 distinguish proportionate and disproportionate short stature in Studies in genetically engineered and mutant mice have 79 46 cases where the distinction is not immediately obvious. Skele- advanced the understanding of the mechanisms underlying pitu- 80 47 tal and hypothalamic–pituitary axis-based growth insufficiencies itary organogenesis defects that lead to short stature ( Zhu et al., 81 48 occur with similar frequencies. Genetic causes of growth hormone 2007; Kelberman and Dattani, 2007 ). In many cases, genes discov- 82 49 deficiency (GHD) are thought to occur in approximately 1/4000 to ered in the mouse led to the discovery of lesions in human patients 83 50 1/10,000 births ( Procter et al., 1998; Patel et al., 2006 ). These can be and have revealed the mechanism of action and genetic hierarchy of 84 51 syndromic, including pituitary and head defects as well as defects control of pituitary cell specification and growth ( Kelberman and 85 52 in the development of other organs, or non-syndromic, with pitu- Dattani, 2009 ). For example, the discovery of the etiology of the 86 53 itary gland-specific effects ( Table 1 ). Because the pituitary gland Snell and Ames dwarf mutations ( Pou1f1 and Prop1 , respectively), 87 54 is critical for the development and function of many other organs, and characterization of the phenotypes of genetically engineered 88 55 all defects in pituitary organogenesis cause secondary effects on mice with mutations in Tpit (officially Tbx19 ), Hesx1 , Lhx3 and Lhx4 , 89 56 target organs. Syndromic pituitary deficiencies include effects on paved the way for identification of the mutated human genes. Some 90 57 non-pituitary tissues that are not within the expectations for sec- genes necessary for normal growth in mice, i.e. Aes , have not yet 91 58 ondary effects on target organs. Genetic defects in the GH gene itself been reported to have lesions in human patients, but there can 92 59 and mutations in the growth hormone-releasing hormone recep- be a considerable lag between discovery in mice and identifica- 93 60 tor (GHRHR) cause isolated GHD (IGHD), but most cases of IGHD tion of rare human patients ( Brinkmeier et al., 2003; Wang et al., 94 61 are idiopathic (reviewed in: Hernandez et al., 2007 ). Patients with 2004 ). 95 62 mutations in the growth hormone receptor gene (Laron dwarfism, The ability of mouse mutants to predict the correct human 96 63 a growth hormone insensitivity syndrome) have a clinical presen- patient characteristics for screening is remarkable, as evidenced 97 64 tation similar to patients with IGHD, but they have elevated levels by Pou1f1, Prop1, Tpit , Hesx1 , Lhx3 and Lhx4 , although the corre- 98 65 of GH and are treated with insulin like growth factor because they spondence is imperfect. For example, LHX4 mutations cause similar 99 66 are unable to respond to GH therapy (reviewed in: Savage et al., hormone deficiencies in humans and mice, and while the mouse 100 67 2006 ). mutations are recessive and cause perinatal lethality, the human 101 68 The availability of recombinant growth hormone has generally mutations are haploinsufficient and viable ( Sheng et al., 1997; 102 69 led to effective correction of growth insufficiency in children with Castinetti et al., 2008; Pfaeffle et al., 2008; Rajab et al., 2008; 103 70 multiple or IGHD due to pituitary developmental defects, but it is Kristrom et al., 2009 ). PROP1 mutations are another example of 104 Table 1 Variety of transcription factor defects affect pituitary function. Gene DNA binding motif Clinical features, mouse phenotypes Syndromic: affecting pituitary development and other head structures PITX2 Paired/bicoid homeo Rieger syndrome: eyes, teeth, umbilical defects Rarely, isolated GH deficiency, haploinsufficient in humans but not obviously so in mice OTX2 POU homeo Anophthalmia, microphthalmia, hypopituitarism LHX3 LIM homeo GH, TSH, PRL, LH, FSH, ACTH, variable including rigid cervical spine, sensorineural deafness LHX4 LIM homeo GH, TSH, PRL, LH, FSH, ACTH, cerebellar and skull defects SOX2 HMG box Hypogonadotrophic hypogonadism, rare isolated GH deficiency SOX3 HMG box MPHD, mental retardation HESX1 Paired homeo Variable including septo-optic dysplasia and severe or mild pituitary hypoplasia or aplasia; UNCORRECTEDGH, TSH, PRL, LH, FSH, ACTH, PROOF or IGHD GLI2 Kruppel family Holoprosencephaly, cleft lip, central incisor, hypopituitarism Non-syndromic: affecting pituitary development PROP1 Paired homeo Progressive hypopituitarism, GH, TSH, PRL, LH, FSH, ACTH POU1F1 POU homeo GH, TSH, PRL TPIT T box ACTH OTX1 POU homeo No human mutations described, mice have delayed growth, puberty Syndromic: affecting pituitary development and other peripheral organs NR5A1 Nuclear receptor LH, FSH, 46, XY disorder of sexual development, hypogonadism, premature ovarian failure, adrenal failure Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 3 105 an imperfect correspondence between the mouse and human fea- Expression of noggin, an antagonist of BMP signaling, TCF7L2, an 169 106 tures. In mice, lesions in Prop1 cause pituitary hypoplasia and effector of canonical WNT signaling, and WNT5A, typically acting 170 107 congenital pituitary hormone deficiency, including GH, TSH, PRL, in the non-canonical pathway, are critical for maintaining the bal- 171 108 and gonadotropin deficiencies ( Bartke et al., 1977; Tang et al., ance of signaling pathways necessary for normal pituitary growth 172 109 1993 ). Both male and female mutant mice go through puberty and and morphology ( Brinkmeier et al., 2003; Potok et al., 2008; Cha 173 110 become fertile with GH, thyroid hormone and PRL supplements, et al., 2004b; Davis and Camper, 2007; Brinkmeier et al., 2007 ). 174 111 suggesting that the gonadotropin deficiency is secondary to the For example, excessive BMP signaling in noggin mutant mice is 175 112 lack of POU1F1 ( Buckwalter et al., 1991; Soares et al., 1984 ). In associated with reduced FGF10 expression, alteration in the SHH 176 113 contrast, humans have variable pituitary size and progressive hor- signaling domain, and multiple invaginations of Rathke’s pouch 177 114 mone deficiency, usually with failure to undergo puberty and the (Davis and Camper, 2007 ). TCF7L2 deficient mice exhibit expan- 178 115 additional involvement of evolving ACTH deficiency, which can be sion of the FGF10 and BMP signaling domains and an abnormally 179 116 fatal if untreated ( Pernasetti et al., 2000; Reynaud et al., 2006; large Rathke’s pouch and subsequently oversized anterior lobe 180 117 Bottner et al., 2004; Riepe et al., 2001 ). The missense mutation (Brinkmeier et al., 2007 ). Finally, WNT5A deficient mice also have 181 118 (S83P) in the spontaneous Ames dwarf mutant, Prop1 df/df , mini- expanded FGF and BMP signaling domains, and the pouch is dys- 182 119 mally transactivates an artificial paired homeodomain binding site morphic but not markedly oversized ( Potok et al., 2008 ). In each of 183 120 in cell transfection assays, while the most common human PROP1 these cases, disruption of one signaling pathway has pleiotropic 184 121 mutation creates a frame shift and likely complete loss of function effects on other signaling pathways. This paradigm is emerging 185 122 (Sornson et al., 1996; Deladoey et al., 1999 ). This does not account as a common theme for signaling pathway function in pituitary 186 123 for the differences in pituitary dysfunction between the species, as development. 187 124 mice homozygous for genetically engineered Prop1 loss of function Popular models suggest that signaling molecules influence the 188 125 alleles have features similar to the missense mutation, and the S83P spatial patterns of pituitary transcription factor expression, leading 189 126 mutant appears to have no activity in culture on the Pou1f1 early to the emergence of specialized cell types that produce pituitary 190 127 enhancer, which is considered a bona fide target ( Sornson et al., hormones, yet there is also compelling evidence that alterations 191 128 1996; Nasonkin et al., 2004; Olson et al., 2006 ). in signaling pathways affect the morphology and size of the organ 192 129 Dissimilarities in the features that characterize mouse mutants more than cell specification ( Brinkmeier et al., 2003, 2007; Ericson 193 130 and human patients may be attributable to differences in the et al., 1998; Potok et al., 2008; Cha et al., 2004b; Davis and Camper, 194 131 effect of the mutations (i.e. partial vs. complete loss of function), 2007; Treier et al., 1998 ). The noggin, WNT5A, and TCF7L2 mutants 195 132 species differences in temporal or spatial expression, overlapping are each able to generate the 5 major hormone-producing cells of 196 133 gene function amongst gene family members, and/or “genome the anterior lobe despite variations in size and shape of the organ. 197 134 variation.” Genome variation means different phenotypic manifes- Rizzoti and Lovell-Badge (2005) recently reviewed the effects of 198 135 tations of the same genetic defect due to the influence of other various genetic lesions on pituitary growth and shape. 199 136 genes in the genome that have functional variant alleles segregating The developing pituitary transcriptome contains many mem- 200 137 in the population. Analysis of mutant mice on different strain back- bers of the BMP, FGF, WNT, Notch and SHH signaling pathways 201 138 grounds can be exploited to uncover the influence of these modifier (Brinkmeier et al., 2009 ). Using Genomatix software we identified 202 139 genes that magnify or minimize the manifestations of reduced 61 additional genes in these pathways that are expressed at a time 203 140 function in other genes ( Nadeau, 2003 ). For example, genetic back- when they could influence pituitary development. 17 genes may be 204 141 ground has a profound effect on the survival of Prop1 mutant mice, involved in cross talk between the pathways. Gene ontology terms 205 142 ranging from neonatal lethal, juvenile lethal to completely viable revealed an additional 72 genes that could contribute to cell signal- 206 143 (Nasonkin et al., 2004 ). The utility of genetically engineered mice ing in the developing pituitary gland. RT-PCR surveys of WNT genes 207 144 and well defined inbred strains may make it possible to tease out expressed in and around the developing organ have identified many 208 145 the genetic risk factors that could cause some human mutations different candidates for regulation of -catenin activity, but little 209 146 to have mild effects in some individuals and more severe ones in is known about the functional significance of many of these genes 210 147 other patients with the same mutation ( Nadeau, 2003; Badano and (Olson et al., 2006; Potok et al., 2008 ). Several pituitary transcrip- 211 148 Katsanis, 2002 ). tion factors are regulated by -catenin, including the EGR1, NR5A1 212 complex, PITX2, and the HESX1, PROP1 complex ( Olson et al., 2006; 213 149 1.3. Cell–cell signaling plays a critical role in pituitary Salisbury et al., 2009, 2007; Garcia-Lavandeira et al., 2009; Kioussi 214 150 organogenesis et al., 2002 ). Because -catenin is regulated by G-protein coupled 215 receptors, some of the pituitary transcription factors that respond 216 151 Classic embryology experiments involving tissue transplanta- to -catenin could be independent of WNT molecules themselves, 217 152 tion and recombination reveal that diffusible molecules produced which is an area for future study ( Gardner et al., 2007 ). Many of the 218 153 by the neural tissue located dorsal to Rathke’s pouch, the primordia signaling pathways involved in pituitary development play impor- 219 154 for the intermediate and anterior lobes of the pituitary gland, are tant roles in ontogeny of other organs, leading to lethality in mice 220 155 essential for pouch induction and growth ( Couly and Le Douarin, homozygous for complete loss of function alleles. Thus, it seems 221 156 1985; ElAmraoui and Dubois, 1993; Hermesz et al., 2003; Ericson unlikely, but not impossible, that genes in these pathways will be 222 157 et al., 1998; Gleiberman et al., 1999; Takuma et al., 1998 ). Sub- responsible for hypopituitarism in humans. 223 158 sequently, members of the WNT, BMP, FGF, Notch, and hedgehog 159 pathways were discovered to haveUNCORRECTED profound effects on pituitary 1.4. Transcription PROOF factor regulation of pituitary development 224 160 development ( Olson et al., 2006; Ericson et al., 1998; Takuma et 161 al., 1998; Potok et al., 2008; Cha et al., 2004b; Davis and Camper, Many transcription factors play important roles in pituitary 225 162 2007; Brinkmeier et al., 2007; Raetzman et al., 2004, 2007, 2006; development and hormone production ( Table 1 ). The early-acting 226 163 Treier et al., 2001, 1998; Kita et al., 2007; Zhu et al., 2006; Ezzat et genes are not pituitary specific, and lesions in these genes cause 227 164 al., 2002; Ohuchi et al., 2000 ). Some essential signaling molecules defects in development of craniofacial or other structures. Some of 228 165 are expressed in the infundibulum, but there are some in the mes- these are homeobox genes with overlapping functions and multi- 229 166 enchyme surrounding the pituitary, i.e. Tgfbi , and some in the pouch ple roles during ontogeny, i.e. Pitx1 and Pitx2 , Lhx3 and Lhx4 (Sheng 230 167 itself ( Brinkmeier et al., 2009 ). This suggests that the regulation of et al., 1997, 1996; Charles et al., 2005; Suh et al., 2002; Gage et al., 231 168 pituitary development by signaling molecules is complex. 1999a,b; Ellsworth et al., 2008 ). Defects in some of these genes 232 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 4 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 233 cause apoptosis, reduced cell proliferation, or both, which ulti- Several of the known pituitary transcription factors were discov- 299 234 mately results in pituitary hypoplasia. The functions of genes like ered using the approach of defining the key cis-acting sequences 300 235 Nr5a1, Pitx2 , and Gata2 , a downstream target of Pitx2 , with broad in hormone genes and the trans-acting factors that bind to them 301 236 expression patterns and roles in the pituitary as well as other crit- (Ingraham et al., 1988; Bodner et al., 1988; Lamonerie et al., 1996; 302 237 ical organs, can be dissected by tissue-specific disruption in mice Lamolet et al., 2001; Gordon et al., 2002, 1997 ). Additional advances 303 238 (Zhao et al., 2001a,b ). Such studies reveal roles for Gata2 in thy- could be made by pursuing this strategy more extensively and/or 304 239 rotropin and gonadotropin production, and implicate Gata3 as a by identifying the regulatory sequences for some of the early-acting 305 240 gene with potential for compensatory activity ( Charles et al., 2006, pituitary-specific transcription factors and their binding factors, as 306 241 2008 ). well as the downstream targets of key transcription factors. For 307 242 Prop1 and Pou1f1 are examples of homeodomain transcription example, we used comparative genomics and bioinformatics to 308 243 factors critical for pituitary development, specifically. Mutations in identify regulatory sequences in Prop1 , and confirmed their rele- 309 244 the human ortholog of Prop1 are the most common known cause of vance in cell culture and transgenic mice ( Ward et al., 2007 ). A 310 245 multiple pituitary hormone deficiency in humans ( Kelberman and highly conserved intragenic enhancer that affects spatial expres- 311 246 Dattani, 2009; Deladoey et al., 1999; Mody et al., 2002; Cogan et sion of a Prop1 transgene is a target of Notch signaling ( Zhu et al., 312 247 al., 1998 ). There are dramatic differences in the effects of Prop1 2006; Ward et al., 2007 ). This suggests that screening for PROP1 313 248 and Pou1f1 mutations on fetal and neonatal pituitary develop- mutations in human patients should include a scan of the intronic 314 249 ment in mice. Prop1 mutants have poor pituitary vascularization enhancer that controls spatial expression of the gene in mice 315 250 and dysmorphology that appears to result, in part, from the fail- (Carvalho et al., 2007 ). 316 251 ure of progenitors to migrate away from the proliferative zone and Another approach is to identify gene expression differences in 317 252 undergo differentiation ( Ward et al., 2006, 2005 ). The defect may the pituitary glands of normal and mutant mice to identify poten- 318 253 result from failure to undergo epithelial to mesenchymal transition, tial downstream targets of Prop1 and Pou1f1 (Brinkmeier et al., 319 254 as Prop1 is required for normal N-cadherin expression, and changes 2009; Douglas et al., 2001; Douglas and Camper, 2000; Carninci 320 255 in cadherin gene expression are typically associated with epithelial et al., 2003; Davis et al., 2009 ). This gene discovery approach has 321 256 to mesenchymal transition ( Himes and Raetzman, 2009; Kikuchi et revealed new members of transcription factor families that are 322 257 al., 2006, 2007 ). In contrast, there are no obvious effects on pituitary exciting candidates for regulating pituitary development and the 323 258 vascularization or morphology in Pou1f1 mouse mutants. basis of human hormone deficiency disease. Here we report the dis- 324 259 Pou1f1 is generally accepted as a direct downstream target of covery of transcription factors expressed in the developing mouse 325 260 Prop1 . This is based on the ability of PROP1 to transactivate a DNA pituitary gland that are members of several important gene families 326 261 fragment of Pou1f1 that contains the early enhancer in cell culture including basic helix-loop-helix, high mobility group, and T-box. 327 262 and the occupancy of PROP1 at that site by chromatin immunopre- These genes are intriguing candidates for future functional stud- 328 263 cipitation in extracts of microdissected embryonic pituitary glands ies and evaluation in human patients. In addition, we present a 329 264 at e12.5 and e13.5 ( Sornson et al., 1996; Olson et al., 2006 ). Care- summary of the clinical features associated with hypopituitarism 330 265 ful review of the evidence suggests that the story may be more caused by known transcription factors, with the purpose of stream- 331 266 complicated. First, there is a profound temporal delay (approxi- lining molecular diagnostic studies. 332 267 mately 4 days) between activation of Prop1 and Pou1f1 expression 268 in mice, which is unusual for a direct downstream target ( Sornson 2. Results and discussion 333 269 et al., 1996 ). Second, human newborns with loss of function alle- 270 les in PROP1 have low but biologically significant levels of TSH, 2.1. Prioritizing genes for molecular studies in human patients 334 271 GH and PRL initially, suggesting that PROP1 is not required for 272 initial expression of POU1F1 in humans ( Bottner et al., 2004 ). Simi- There are approximately a dozen different transcription factor 335 273 larly, mice with Prop1 mutations express limited amounts of Pou1f1 genes that are mutated in cases of short stature and/or pituitary 336 274 and its targets Tshb, Gh , and Prl (Gage et al., 1995, 1996 ). More gland dysfunction ( Table 1 ). These are classified based on the type 337 275 work needs to be done to reconcile these apparently conflict- of pituitary defect that they produce as well as any other clini- 338 276 ing observations and clarify the role of PROP1 in humans and cal features. Several of these genes are expressed in the developing 339 277 mice. hypothalamus and are likely to affect anterior pituitary gland devel- 340 278 We hypothesize that the role of PROP1 is to generate precursor opment by disrupting the normal balance of signaling pathways 341 279 cells that are capable of becoming hormone-producing cells of the and inductive factors produced by the hypothalamus. For exam- 342 280 anterior lobe and promote the transition from proliferation to dif- ple, GLI2, SOX2, SOX3, and TCF7L2 are primarily expressed in the 343 281 ferentiation. It may also play a role in regulating the accessibility of neural ectoderm ( Brinkmeier et al., 2003, 2007; Gustavsson et al., 344 282 the POU1F1 regulatory elements. POU1F1 is activated in some of the 2006; Tziaferi et al., 2008; Woods et al., 2005; Weiss et al., 2003; 345 283 precursor cells to promote differentiation into somatotrophs, thy- Laumonnier et al., 2002; Rizzoti et al., 2004 ). Some of the pituitary 346 284 rotrophs and lactotrophs and to expand the proliferation of that transcription factor genes are large and can pose difficulties for DNA 347 285 lineage after birth. If this hypothesis is true, the progressive hor- sequence analysis because of high GC content. Thus, it is useful to 348 286 mone deficiency in humans with PROP1 mutations could arise by predict which of the many genes are most likely to be mutated 349 287 depletion of the progenitor pool, and the more severe, congenital given a set of clinical characteristics. 350 288 hormone deficiency in Prop1 mutant mice could be due to a stronger The primary sources of diagnostic information for a clinical 351 289 and/or earlier requirement of Prop1UNCORRECTEDfor establishing the precursor endocrinologist PROOF are the family history, evidence of growth retar- 352 290 pool in mice than humans. Investigation of genes expressed in the dation based on the longitudinal height and weight curve of 353 291 developing pituitary gland between peak Prop1 and Pou1f1 expres- population matched controls, basal and stimulated levels of cir- 354 292 sion may uncover direct targets of Prop1 that are intermediates culating hormones, and imaging to analyze the shape of the sella 355 293 between Prop1 and Pou1f1 . Neurod4 (also known as Math3 ) is a can- turcica, the size and position of the anterior pituitary, the pituitary 356 294 didate for an intermediate, as it is activated at e13.5 before Pou1f1 stalk and the posterior pituitary. Non-pituitary related syndromic 357 295 is generally detected, although maintenance of Neurod4 expression features such as Chiari malformation, craniofacial abnormalities, 358 296 requires Pou1f1 (Zhu et al., 2006 ). Novel genes expressed at these limited neck rotation, eye abnormalities, and hearing deficits can 359 297 early developmental times will be candidates to explain pituitary suggest a causative gene that could account for hypopituitarism and 360 298 deficiency diseases of unknown etiology. the syndromic features ( Fig. 1 ). Congenital hypopituitarism associ- 361 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 5 Fig. 1. A guide for planning genetic screening for hypopituitary patients based on clinical findings. The patient characteristics identified by hormone screening, imaging studies, and analyses of syndromic features are itemized for each candidate gene based on currently known patient mutations. Note that for most genes there are variable hormone deficiencies, pituitary size and placement, and variable syndromic features. 362 ated to dysmorphic features are generally linked to early-expressed tend to worsen progressively ( Pernasetti et al., 2000; Reynaud et 391 363 genes during pituitary development as GLI2, SOX2, SOX3, HESX1, al., 2006; Bottner et al., 2004; Deladoey et al., 1999 ). Other genes 392 364 LHX3, LHX4 , whereas non-syndromic hypopituitarism is generally that can be mutated in patients with MPHD, eutopic posterior lobe 393 365 due to defects in later-expressed, pituitary-specific genes such as and normal pituitary stalk are POU1F1 and LHX3 , although LHX3 394 366 PROP1 and POU1F1 (Fig. 1 ). There can be a large spectrum of clin- mutations can present with cervical stiffness ( Fig. 1 ). Despite the 395 367 ical features associated with any one gene. For example, HESX1 common occurrence of PROP1 mutations in multiple ethnic groups, 396 368 mutations can cause multiple pituitary hormone deficiencies or and the rare mutations reported in other transcription factor genes, 397 369 occasionally, isolated GH deficiency ( Dattani et al., 1998; Thomas most patients with hypopituitarism have no identifiable genetic 398 370 et al., 2001 ). Obviously, not all patients with mutations in the lesions ( Osorio et al., 2002 ). Ectopic posterior pituitary and stalk 399 371 same gene have identical clinical findings, but there are trends disruptions are common features of patients with unknown etiol- 400 372 that suggest prioritization of the molecular diagnostic tests. Distin- ogy ( Dattani et al., 1998 ). 401 373 guishing between hypothalamic and pituitary origins of hormone There have been impressive advances in identifying the molec- 402 374 deficiency is useful in planning the strategy for genetic analysis, ular basis for pituitary hormone deficiency over the past 20 years. 403 375 although this is not without controversy ( Hernandez et al., 2007; It is also appreciated that non-genetic traumatic events can cause 404 376 Mehta et al., 2003 ). There is a lack of data in the literature cor- hypopituitarism ( Osorio et al., 2002 ). Yet birth trauma does not 405 377 relating TRH stimulation pattern in patients with known genetic appear to be an obvious contributor to the majority of cases of 406 378 defects associated to hypothalamic or pituitary cells differentia- hypopituitarism without a molecular diagnosis. Thus, additional 407 379 tion. In a study of 43 patients with MPHD, 26% were compatible genes are likely to be involved, and alternative approaches must be 408 380 with pituitary hypopituitarism, and 70% had hormonal responses taken to identify these genes. 409 381 suggestive of hypothalamic hypopituitarismUNCORRECTED ( Osorio et al., 2002 ). PROOF 382 Ectopic posterior lobe was a characteristic of putative hypothalamic 2.2. Gene discovery to identify new candidate genes 410 383 hypopituitarism. 384 Mutations in HESX1, LHX3, LHX4, GLI2, SOX2 , and SOX3 are rare In an effort to identify novel candidate genes for hypopitu- 411 385 causes of congenital hypopituitarism, while PROP1 defects are a itarism, we undertook a project to discover genes expressed during 412 386 common cause of familial, congenital hypopituitarism ( Kelberman a critical period of mouse pituitary gland development ( Brinkmeier 413 387 and Dattani, 2007, 2006; Cogan et al., 1998; McNay et al., 2007 ). et al., 2009; Carninci et al., 2003 ). A brief description of library 414 388 Although PROP1 mutations have a variable effect the size of the preparation and analysis follows. We dissected pituitary glands 415 389 gland, all cases described to date have an intact stalk, eutopic (nor- from normal mice at e12.5 (CD1 strain) and from DF/B- Prop1 df/df 416 390 mally placed) posterior pituitary, and hormone deficiencies that mutants and their wild type littermates at e14.5 and genotyped 417 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 6 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 418 individual fetuses using previously described protocols ( Douglas Mice heterozygous for loss of function mutations typically have 482 419 and Camper, 2000 ). Appropriate samples were pooled based on no obvious abnormalities, and eye and tooth defects occur with 483 420 genotype, total RNA isolated, and cDNA prepared using the cap- extremely low penetrance. Multiple genetic backgrounds studied 484 421 trapper method to enhance the representation of full-length clones do not increase the penetrance (Camper and Gage, unpublished 485 422 and normalization to enrich for rare sequences ( Carninci et al., observations). This suggests that reduced levels of PITX2 are bet- 486 423 2003, 2000 ). Two subtracted libraries were prepared to enhance ter tolerated in mice than humans. Homozygous mutants die at 487 424 the representation of Prop1 targets ( Brinkmeier et al., 2009 ). One approximately e12.5 due to severe heart and abdominal body wall 488 425 subtracted library was enriched for genes activated at e14.5 but defects ( Gage et al., 1999a ). Multiple organs are severely affected, 489 426 not e12.5, and the other was enriched for genes expressed in nor- including the eyes, teeth, and lungs, which exhibit isomerization. 490 427 mal pituitaries but not in Prop1 mutants at e14.5. Single pass DNA The pituitary gland exhibits hypoplasia, partly caused by enhanced 491 428 sequencing was conducted on 56,716 clones using an algorithm to cell death at the ventral aspect that is suggestive of defective sonic 492 429 maximize gene discovery and minimize re-sequencing cDNAs rep- hedgehog and/or FGF signaling ( Charles et al., 2005 ). 493 430 resenting the same genes. The cDNA sequences were analyzed by Mice homozygous for a partial loss of function allele, Pitx2 neo , 494 431 comparison with publicly available databases using an expect value have milder pituitary defects, with little reduction in organ size, but 495 − 432 of 10 5 to favor putative identifications that could be validated by reduction in the Pou1f1 lineage, resulting in reduced expression of 496 433 more complete sequence analysis. We placed the matches of single both GH and TSH ( Suh et al., 2002 ). The reduced differentiation 497 434 pass sequence with public database entries into a local database of this lineage could explain the reduction in GH secretion that 498 435 that is searchable by gene name, RefSeq ID, Unigene ID, gene ontol- characterizes some patients with PITX2 lesions. Homozygotes for 499 436 ogy terms (GO), and DNA sequence and identified 12,009 different this hypomorphic allele lack gonadotrophs, as there is virtually no 500 437 expressed genes. To obtain database access contact Drs. Camper or detectable Nr5a1, Lhb or Fshb expression. Gata2 expression is pro- 501 438 Lyons at [email protected] or [email protected] . foundly reduced as well. There are no reports of hypogonadism 502 439 Analysis of the database revealed numerous homeobox genes in Rieger patients, but it is possible that feedback regulation from 503 440 not previously known to be expressed in the pituitary gland, mem- the ovaries and testes normalizes gonadotropin production after 504 441 bers of signaling pathways, and genes involved in cell proliferation, birth. Indeed, there are some animal models that exhibit transient 505 442 apoptosis, cell migration, and cell adhesion ( Brinkmeier et al., 2009; hypogonadism and delayed puberty but eventually become fertile 506 443 Davis et al., 2009 ). Here we highlight transcription factors identi- (Acampora et al., 1998; Cushman et al., 2001; Vesper et al., 2006 ). 507 444 fied by this approach that are members of the PITX, and LHX gene No PITX1 mutations have been described in humans to date. 508 445 families, as well as the forkhead group, families of basic helix-loop- Pitx1 loss of function is recessive lethal in mice ( Lanctot et al., 1999; 509 446 helix (bHLH), high mobility group (HMG), and the T-box family. Szeto et al., 1999 ). Mutants have multiple defects including reduc- 510 447 Members of each of these families have been implicated in pitu- tion in the mandible and hind limbs, but the pituitary is mildly 511 448 itary development and function, and there are several precedents affected, exhibiting a reduction in the number of gonadotropes at 512 449 for overlapping and critical functions of related genes within a sin- birth. Despite the fact that Pitx1 is dispensable for development of a 513 450 gle family for pituitary development ( Sheng et al., 1997; Charles et normally sized pituitary gland and differentiation of the major cell 514 451 al., 2005 ). types of the anterior lobe at birth, Pitx1 and Pitx2 have overlapping 515 452 Initial validation studies reveal that the single pass sequences functions early in pituitary development and are required for acti- 516 ′ 453 obtained from the 3 end are not as reliable as those from the vation of Lhx3 (Charles et al., 2005; Suh et al., 2002; Tremblay et al., 517 ′ 454 5 end, probably due to the difficulty of accurate amplification 1998 ). 518 455 through stretches of polyadenine. In rare cases we identified 456 chimeric clones or unexplained differences between the single 2.4. The LHX family 519 457 pass sequence and the validation by complete sequencing, but in 458 most cases the predictions are accurate ( Brinkmeier et al., 2009 ). Several LIM homeodomain transcription factors are expressed 520 459 Another important type of verification is demonstration that the in the pituitary gland: Lhx2, Lhx3, Lhx4 and Isl1 . The first to be 521 460 cDNA is expressed in the pituitary gland because the embryonic implicated in the pituitary gland was Lhx2 , which was identified 522 461 pituitary dissection includes some surrounding mesenchyme, oral by cloning the transcription factor that bound to a critical cis- 523 462 ectoderm, and neural ectoderm. Complete sequencing and expres- acting sequence in the gene that encodes the alpha subunit of 524 463 sion studies have not yet been done on all of the genes in the the pituitary glycoprotein hormones, Cga , in the gonadotrope- and 525 464 basic helix-loop-helix (bHLH) and high mobility group (HMG) fam- thyrotrope-like cell lines, ␣T3-1 and ␣TSH, respectively ( Roberson 526 465 ilies reported here or for another study based on gene ontology, et al., 1994 ). Additional studies demonstrated that LHX2 interacts 527 466 biological process terms ( Davis et al., 2009 ). Thus, this report with a LIM domain binding protein (LBD1) to regulate Cga expres- 528 467 represents a foundation for comprehensive validation and future sion, and the single-stranded DNA binding protein, SSBP3, regulates 529 468 studies. the abundance of this complex ( Cai et al., 2008 ). We identified Lhx2 530 expression in a cDNA library made by subtracting e12.5 Rathke’s 531 469 2.3. The PITX gene family pouch cDNA from e14.5, suggesting that Lhx2 could have a role 532 in pituitary development ( Brinkmeier et al., 2009 ). To explore the 533 470 The PITX gene family illustrates how multiple members within spatial expression of Lhx2 in early pituitary development we car- 534 471 the same gene family can have unique and overlapping functions ried out immunohistochemical staining ( Fig. 2 ). The majority of the 535 472 in organ development. PITX2 mutationsUNCORRECTED are one cause of Rieger stain appears PROOF in the neural ectoderm, dorsal to the prospective pos- 536 473 syndrome, a genetically heterogeneous, dominant disorder charac- terior lobe and signaling center for BMP and FGF. Staining is also 537 474 terized by defects in development of the eyes, teeth, and umbilical readily detectable in the caudal hindbrain, but little or no staining 538 475 cord ( Semina et al., 1996 ). Some individuals have additional abnor- is detected in the developing pituitary gland at these stages. These 539 476 malities including heart defects, and rarely, growth hormone immunohistochemistry results are consistent with the failure to 540 477 insufficiency. Molecular studies show that missense mutations detect Lhx2 transcripts in Rathke’s pouch or its derivatives at e10.5 541 478 can cause loss of function, gain of function, or dominant nega- and e14.5 ( Lu et al., 2000; Visel et al., 2004, 2007 ). This suggests 542 479 tive effects, but loss of function mutations appear to be the most that LHX2 regulation of Cga expression initiates later in develop- 543 480 common, consistent with a haploinsufficiency disorder ( Suh et al., ment or requires amounts of the protein that are difficult to detect. 544 481 2005 ). In some tissues Lhx2 is expressed after stem cells become commit- 545 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 7 Fig. 2. LHX2 expression in the neural ectoderm. LHX2 immunoreactivity (green) is detected in sagittal sections of developing mice at e10.5 through e13.5. DAPI (blue) Q2 counterstain reveals nuclei of individual cells (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.). 546 ted progenitors but before terminal differentiation occurs ( Rhee et LHX3 mutations generally lead to MPHD with variability in cor- 571 547 al., 2006 ). If Lhx2 plays a role in pituitary gland, it might explain ticotrope axis function, abnormal neck rotation, mild to severe 572 548 our difficulties in detecting immunoreactivity. Lhx2 mutants die hearing impairments, and/or mental retardation. The pituitary can 573 549 during gestation from defects in erythropoiesis and exhibit anoph- either be hypo- or hyperplastic, or even associated with a microade- 574 550 thalmia and brain defects ( Porter et al., 1997 ). Additional studies are noma. Only 9 LHX3 mutations have been reported, and all are 575 551 necessary to define the role of Lhx2 in pituitary gland development. inherited in an autosomal recessive manner ( Rajab et al., 2008; 576 552 LHX4 is a LIM homeodomain transcription factor involved in Sobrier et al., 2004; Netchine et al., 2000; Bhangoo et al., 2006; 577 553 pituitary organogenesis, and crucial for the genesis and devel- Pfaeffle et al., 2007 ). 578 554 opment of Rathke’s pouch. As a LIM homeodomain transcription Mice homozygous for an Lhx4 disruption induced by homol- 579 − − 555 factor, LHX4 shows significant structural similarity with LHX3, sug- ogous recombination ( Lhx4 / ) have an abnormal pituitary 580 556 gesting a possible overlap between each factor during Rathke’s phenotype and die soon after birth from lung defects, whereas 581 − 557 pouch formation ( Mullen et al., 2007 ). This point is suggested by heterozygous animals ( Lhx4 +/ ) seem unaffected ( Sheng et al., 582 558 the observation of similar activities of LHX4 and LHX3 in assays 1997; Raetzman et al., 2002 ). It is possible that homozygous LHX4 583 559 using pituitary hormone promoter genes, and by the role of each mutation causes lethality in humans. It is noteworthy that mice het- 584 560 factor in ventral motor neuron differentiation ( Sharma et al., 1998 ). erozygous for Lhx4 loss of function do not appear affected, just as 585 561 LHX4 mutations produce a wideUNCORRECTED intra- and inter-family range PITX2 haploinsufficiency PROOF is evident in humans but not mice. This 586 562 of phenotypes in humans, both in terms of hypopituitarism and of species difference in the tolerance for reduced LHX4 and PITX2 587 563 pituitary/cerebral MRI images of the morphology. The gland may levels illustrates a limitation of the comparison between human 588 564 exhibit hypo- or hyperplasia, variable ectopic posterior lobe, and and mice, though the accessibility of tissues in mice throughout 589 565 assorted intracranial abnormalities including Chiari syndrome and development is invaluable for understanding the mechanisms that 590 566 corpus callosum hypoplasia, and poorly developed sella turcica. underlie human pituitary developmental defects. 591 − − 567 To date, 5 heterozygous mutations, including 1 intronic lesion, are Mice homozygous for Lhx3 disruption ( Lhx3 / ) exhibit a severe 592 568 reported, suggesting that the mechanism underlying the functional phenotype with death within 24 h after birth ( Sheng et al., 1996 ). 593 − − − 569 defect is haplo-insufficiency ( Castinetti et al., 2008; Pfaeffle et al., In contrast, Lhx3 +/ mice appear normal. Embryonic Lhx3 / mice 594 570 2008; Machinis and Amselem, 2005; Machinis et al., 2001 ). show normal rudimentary Rathke’s pouch formation but lack 595 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 8 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 596 further pituitary development from e10.5 onward and undergo ovarian failure ( Crisponi et al., 2001; De Baere et al., 2001 ). Muta- 660 597 apoptosis ( Ellsworth et al., 2008; Zhao et al., 2006 ). Dorsal–ventral tions in the orthologous mouse genes cause similar phenotypes 661 598 patterning is modified with dorsal location of some progenitors (Kume et al., 2000, 1998; Hong et al., 1999; Kidson et al., 1999; 662 599 normally located in the ventral aspect of the gland ( Ellsworth et Winnier et al., 1999, 1997; Iida et al., 1997; Smith et al., 2000; Blixt 663 − − 600 al., 2008 ). At birth, Lhx3 / mice lack the anterior and intermediate et al., 2000; Brownell et al., 2000; Ottolenghi et al., 2005; Schmidt 664 601 lobes of pituitary gland. et al., 2004; Uda et al., 2004 ). 665 602 Lhx3 and Lhx4 are expressed throughout the pouch at e9.5 Secondly, many forkhead proteins are important for cell cycle 666 603 (Sheng et al., 1997 ). At e12.5 Lhx3 continues to be expressed regulation and act as tumor suppressors. Overexpression of the 667 604 throughout the pouch in a gradient with higher protein levels at forkhead transcription factors FoxO3a , AFX , or FoxO1a cause growth 668 605 the dorsal aspect of the pouch ( Raetzman et al., 2002 ), while Lhx4 suppression in a number of cell lines, including a Ras-transformed 669 606 expression becomes restricted to the developing anterior lobe. At cell line and a cell line lacking a known tumor suppressor ( Medema 670 607 e15.5, Lhx4 decreases, while Lhx3 continues to be expressed at high et al., 2000 ). Amplification of the forkhead gene, FoxA1 , occurs 671 608 levels. The overlap in gene expression suggests the possibility of in lung tumors and esophageal adenocarcinomas implicating this 672 609 functional overlap, which is borne out by analysis of double mutant gene in tumorigenesis ( Lin et al., 2002 ). 673 610 mice. Finally, approximately half of the known null mutations in fork- 674 − − − − 611 Lhx3 / and Lhx4 / single mutants form a definitive pouch head genes result in death before or shortly after birth ( Carlsson and 675 612 (Sheng et al., 1997, 1996 ). The pouch fails to expand in Lhx3 mutants Mahlapuu, 2002 ). This raises three important points: (1) this family 676 613 due to increased apoptosis resulting in severe pituitary hypoplasia of genes is very important for normal development, (2) members of 677 614 (Ellsworth et al., 2008; Zhao et al., 2006 ). Lhx3 nulls exhibit reduced this family generally do not exhibit functional redundancy and (3) 678 615 ACTH immunostaining and deficiency of all other hormones nor- mouse models of forkhead gene deficiency are good predictors of 679 − − 616 mally produced in the anterior lobe. The Lhx4 / phenotype is less the human phenotype. Null mouse models have been described for 680 617 severe. Specification of five hormone-producing cell types occurs, approximately 31 forkhead genes so far and all except for Foxo4 681 618 but expansion of these lineages is greatly reduced, and increased result in an abnormal phenotype. Moreover, a number of these 682 619 apoptosis is evident ( Raetzman et al., 2002 ). One wild type Lhx3 or knockouts involved members of the same subfamily. For example, 683 620 Lhx4 allele is sufficient for formation of a definitive pouch, as evi- Foxa1 knockout mice die postnatally with severe growth retarda- 684 − − − − − − 621 denced by analysis of Lhx3 / , Lhx4 +/ and Lhx3 +/ , Lhx4 / mutants. tion ( Kaestner et al., 1999 ), and Foxa2 knockout mice do not develop 685 622 Loss of all alleles for Lhx3 and Lhx4 results in formation of a rudi- beyond embryonic day 8.5 (e8.5) and lack the node, notochord and 686 623 mentary pouch, which fails to grow into a definitive pouch and foregut ( Ang and Rossant, 1994; Weinstein et al., 1994; Sund et 687 624 remains under the sphenoid cartilage ( Sheng et al., 1997 ). The fact al., 2001, 2000 ). These data suggest that FOXA1 and FOXA2 cannot 688 625 that a definitive pouch is formed in the absence of Lhx3, but only a compensate for each other, in contrast to Foxc1 and Foxc2 . Loss of 689 626 rudimentary pouch is formed when both Lhx3 and Lhx4 are deleted, Foxc1 results in multiple abnormalities including hydrocephalus, 690 627 suggests that Lhx4 can substitute for the function of Lhx3 to support skeletal, ocular, renal and cardiovascular defects, and Foxc2 defi- 691 628 the formation of a definitive pouch. The fact that deletion of Lhx3 ciency causes skeletal, cardiovascular and ocular defects, indicating 692 629 alone results in loss of most of the anterior lobe cell types suggests that each gene is required independently for development of sev- 693 630 that Lhx4 can not substitute for the function of Lhx3 to activate eral organ systems including the skeleton. Loss of both Foxc1 and 694 631 a pituitary-specific transcription program ( Sheng and Westphal, Foxc2 disrupts somitogenesis, revealing an early overlapping func- 695 632 1999 ). tion. 696 633 Isl1 expression is detectable throughout Rathke’s pouch at e9.5 634 and by e12.5 it is restricted to the ventral, differentiating cells that 2.6. Forkhead factors in pituitary development 697 635 express Cga and Foxl2 (Ellsworth et al., 2008; Raetzman et al., 2002 ). 636 Lhx3 and Lhx4 mutants have different effects on Isl1 expression in Foxl2 (Pfrk ) is expressed in the prospective anterior lobe of the 698 637 the pituitary gland, causing a gain and loss of expression, respec- developing pituitary gland starting at e11.5 and continuing into 699 638 tively ( Ellsworth et al., 2008 ). Isl1 is implicated as a lineage-specific adulthood in gonadotrope and thyrotrope cells of the anterior pitu- 700 639 transcription factor in cell fate choice in progenitors of the retina, itary ( Treier et al., 1998; Ellsworth et al., 2006 ). In addition, FOXL2 701 640 heart, forebrain, and motor neurons ( Cai et al., 2003; Elshatory and is part of a transcription complex that binds the gonadotropin- 702 641 Gan, 2008; Elshatory et al., 2007; Pan et al., 2008; Pfaff et al., 1996 ). releasing hormone receptor gene in gonadotrope cells ( Ellsworth 703 642 Isl1 deficiency causes an arrest in pituitary gland development at et al., 2003 ). Finally, Foxl2 stimulates expression of Cga (␣GSU) in 704 643 an early stage, and the embryos die at e11.5 ( Cai et al., 2003 ). Thus, cell culture studies and when overexpressed in transgenic mice 705 644 it is an essential regulator of the early steps, but its role, if any, in (Ellsworth et al., 2006 ). 706 645 later stages is unknown. Human patients heterozygous for FOXL2 mutations have dom- 707 inant blepharophimosis ptosis epicanthus inversus syndrome 708 646 2.5. Forkhead factors are essential for diverse developmental (BPES, eyelid abnormalities) and premature ovarian failure ( Loffler 709 647 processes et al., 2003 ). Mice homozygous for loss of function alleles are mostly 710 non-viable, but those that survive have craniofacial and ovarian 711 648 There are several common functional themes among forkhead abnormalities ( Schmidt et al., 2004 ). While the effects of this defi- 712 649 factors. First, they are responsible for numerous autosomal domi- ciency on the pituitaries of most mutants has not been reported, 713 650 nant human developmental disorders.UNCORRECTED For example, mutations in Foxl2 transgene PROOF expression is sufficient to drive expression of Cga 714 651 four different forkhead genes affect ocular development. Mutations at ectopic sites within the pituitary primordium, and it has perma- 715 652 in FOXC1 result in Axenfeld–Rieger anomaly that is characterized nent expression in thyrotropes and gonadotropes, suggesting a role 716 653 by facial, teeth and eye malformations ( Nishimura et al., 2001; in gonadotrope differentiation and function ( Ellsworth et al., 2006 ). 717 654 Mirzayans et al., 2000; Mears et al., 1998; Lehmann et al., 2000 ). Consistent with a role in gonadotrope differentiation, FOXL2 regu- 718 655 FOXC2 mutations result in lymphedema and distichiasis, a double lates the expression of Gnrhr , Fshb , and follistatin in gonadotropes 719 656 row of eyelashes ( Fang et al., 2000; Finegold et al., 2001 ). Mutations (Ellsworth et al., 2003; Blount et al., 2009; Lamba et al., 2009 ). Dur- 720 657 in FOXE3 result in malformations in the anterior segment of the eye ing pituitary development, FOXL2 protein is localized to quiescent 721 658 referred to as Peter’s anomaly ( Semina et al., 2001; Ormestad et al., cells, suggesting that FOXL2 may be important for inhibiting cell 722 659 2002 ). FOXL2 mutations cause eyelid malformations and premature proliferation ( Ellsworth et al., 2006 ). This idea is supported by the 723 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 9 724 discovery that mutations in FOXL2 are associated with aggressive ing pathway, along with the Hey class of bHLH transcription factors. 788 725 ovarian granulosa cell tumors in children ( Kalfa et al., 2007 ). Loss of Hes1 results in a cell fate switch such that intermediate lobe 789 726 Several other forkhead genes are expressed in the pituitary cells differentiate as GH hormone secreting somatotropes instead of 790 727 gland: Foxe1, Foxf1, Foxa1 , and Foxd1 . Foxe1 expression is first POMC expressing melanotropes ( Raetzman et al., 2007 ). Hes1 plays 791 728 detected at e9.5 in the ectoderm that will give rise to the anterior an additional role in maintaining anterior lobe precursor cells such 792 729 pituitary ( Zannini et al., 1997 ). Foxf1 is expressed in the mes- that they do not differentiate prematurely ( Kita et al., 2007; Zhu et 793 730 enchyme of the developing pituitary gland by e9.5 ( Kalinichenko al., 2006; Himes and Raetzman, 2009 ). Recent studies demonstrate 794 731 et al., 2003 ). FOXA1 (a.k.a. HNF-3 ␣) represses growth hormone that this is accomplished through control of cell cycle regulators 795 732 expression in mouse and human pituitary tissue by binding to (Monahan et al., 2009 ). 796 733 the P sequence element C of the human GH gene ( Norquay et The expression patterns of bHLH transcriptions factors Hey1 and 797 734 al., 2006 ). FOXD1 (or brain factor 2, BF2) is expressed in the Hes6 suggest possible roles in the developing pituitary gland, but 798 735 diencephalon, retina, and kidney. Mutations in Foxd1 affect the more studies are necessary to assess their functional significance. 799 736 retina, optic chiasm and kidney ( Herrera et al., 2004 ). The kid- The Hey1 expression domain overlaps with Hes1 in presumptive 800 737 neys are small with decreased ureteric branching, and the mice precursor cells in the pouch, suggesting that these genes may have 801 738 die within 24 h after birth due to renal failure. Ectopic BMP sig- overlapping functions in regulating the progression of cells from 802 739 naling is thought to be responsible for the dysmorphology and proliferation to differentiation ( Raetzman et al., 2007, 2006 ). In con- 803 740 loss of kidney function ( Hatini et al., 1996 ). Foxd1 expression is trast, Hes6 is expressed in the differentiating cells of the anterior 804 741 detectable in the pituitary gland after birth, and during devel- lobe, positioning it for maintaining quiescence and/or cell speci- 805 742 opment it is evident in the diencephalon and the mesenchyme fication ( Raetzman et al., 2004 ). The contributions of these bHLH 806 743 surrounding the pituitary at e10.5. Interestingly, these regions are transcription factors to pituitary gland development are still spec- 807 744 essential for the production of BMPs, which are required for nor- ulative. 808 745 mal pituitary development, and thus FOXD1 regulation of BMP We searched our cDNA encyclopedia for members of the bHLH 809 746 production in these tissues likely contributes to normal pituitary family and identified 33 genes including expected factors such as 810 747 development. Neurod1 , Hes6 and Hey1 as well as additional bHLH family members 811 whose embryonic pituitary expression was not previously known 812 748 2.7. Basic helix-loop-helix family is highly represented in the (Table 2 ). Among these new bHLH family members are the Id genes, 813 749 developing pituitary gland which act downstream of BMP signaling ( Brinkmeier et al., 2009 ). 814 Identification of Id3 in the library allowed for its use as a reporter 815 750 Members of the basic helix-loop-helix (bHLH) family of tran- of BMP signaling in the ventral diencephalon overlying Rathke’s 816 751 scriptions factors are found in all eukaryotes, and they bind DNA in pouch ( Davis and Camper, 2007 ). 817 752 complexes of homo- or heterodimers through a conserved helix- Interestingly, the bHLH member, aryl-hydrocarbon receptor 818 753 loop-helix domain ( Murre et al., 1989 ). These transcription factors (Ahr ), and the aryl-hydrocarbon interacting protein (Aip , a tetratrico 819 754 play diverse roles in many developmental pathways and tissues. peptide repeat containing protein) are contained in the develop- 820 755 Myod , Myog , and Myf5 are bHLH proteins that are instrumen- mental library. AIP is associated with increased risk of pituitary 821 756 tal in the differentiation of skeletal muscle (reviewed in Arnold adenomas that secrete GH in some populations; and the molecular 822 757 and Braun, 1996 ), while Ascl1 , Neurog1 , and Neurod1 participate mechanism appears to be loss of tumor suppression ( Vierimaa et al., 823 758 in neuronal differentiation ( Chu et al., 2001; Cho and Tsai, 2004; 2006; Leontiou et al., 2008 ). The overall contribution of mutations 824 759 Henke et al., 2009; Fratticci et al., 2007; Cai et al., 2000; Cau et in AIP to sporadic adenoma risk world wide appears to be low ( Daly 825 760 al., 1997; Mizuguchi et al., 2006 ). Several members of the bHLH et al., 2007 ), but the presence of this Ahr, Aip complex during the 826 761 family are known to be expressed in pituitary development: Ascl1, development of the pituitary suggests that an early developmen- 827 762 Neurod1 , Neurod4 , and Hes1 . Roles of these genes in corticotrope, tal mechanism for growth regulation may be recapitulated during 828 763 somatotrope, and melanotrope development will be reviewed adenoma formation. 829 764 below. 765 Neurod1 (also known as BETA2, BHF-1, bHLHa3, NeuroD) is 2.8. High mobility group genes expressed in the developing 830 766 expressed in the embryonic pituitary at e12.5, where it precedes pituitary 831 767 the appearance of POMC in the corticotrope lineage, and it is down 768 regulated at e15.5 ( Poulin et al., 2000, 1997 ). Neurod1 has a role in The high mobility group or HMG class of DNA binding proteins 832 769 corticotrope development and function, and recent evidence sug- bind DNA through a conserved domain that consists of three alpha 833 770 gests that it may affect gene expression in gonadotropes as well helices arranged in an “L-shape.” These proteins are known to bend 834 771 (Cherrington et al., 2008 ). Loss of Neurod1 delays the terminal dif- the DNA to which they are bound ( Thomas, 2001 ). The Sox genes, 835 772 ferentiation of corticotropes from e12.5 to e16.5, indicating that it is which are related to SRY, the mammalian male sex determina- 836 773 a critical factor for promoting corticotrope differentiation, although tion gene, are members of the HMG family, and they have received 837 774 it is not required ( Lamolet et al., 2001; Lavoie et al., 2008 ). NEU- increased attention recently because of their role in stem cell main- 838 775 ROD1 and TPIT (a T box gene officially named TBX19, discussed tenance. Sox2 is expressed in embryonic stem cells and stem cells 839 776 below) both bind to the POMC promoter to drive expression. How- from a variety of tissues, including a potential stem cell popula- 840 777 ever, neither Neurod1 nor Tpit is required for POMC expression, tion in the pituitary ( Masui et al., 2007; Fauquier et al., 2008 ). The 841 778 nor does loss of one prevent theUNCORRECTED binding of the other to the POMC presence ofPROOF a pituitary stem cell population that gives rise to all 842 779 promoter, suggesting that these transcription factors act indepen- the cell types of the anterior lobe is an exciting development as it 843 780 dently of each other to drive POMC expression ( Lamolet et al., 2004; has been a proposed mechanism for explaining the plasticity of the 844 781 Pulichino et al., 2003a ). pituitary gland, which can change its cellular make up in response 845 782 Neurod4 (also known as Math3, Atoh3, bHLHa4) is a downstream to changing physiological demands. 846 783 target of Pou1f1 that is necessary for maintenance of the soma- Sox2 and Sox3 are also critical factors in embryonic pituitary 847 784 totrope cells ( Zhu et al., 2006 ). development. Both are expressed in the ventral diencephalon over- 848 785 The bHLH factor, Hes1 , is necessary for POMC expression in lying Rathke’s pouch ( Rizzoti et al., 2004; Fauquier et al., 2008; 849 786 melanotropes in the intermediate lobe ( Raetzman et al., 2007 ). The Kelberman et al., 2006 ). Sox3 homozygous null mice have disrup- 850 787 Hes transcription factors transduce signals from the Notch signal- tions in the patterning of the ventral diencephalon such that both 851 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 10 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx Table 2 Developing pituitary transcriptome contains many transcription factors of the bHLH, HMG and Tbox families. Gene symbol Full gene name Gene symbol Full gene name bHLH = Basic helix-loop-helix Ahr Aryl-hydrocarbon receptor Mlx MAX-like protein X Arnt Aryl hydrocarbon receptor nuclear translocator Mnt Max binding protein Arntl Aryl hydrocarbon receptor nuclear translocator-like Msc Musculin Ascl1 Achaete-scute complex homolog-like 1 ( Mash1) Mxi1 Max interacting protein 1 Bhlhb9 Basic helix-loop-helix domain containing, class B9 Mycl1 v-Myc myelocytomatosis viral oncogene homolog 1, lung carcinoma derived Ebf2 Early B-cell factor 2 Mycn v-Myc myelocytomatosis viral related oncogene, neuroblastoma derived Ebf4 Early B-cell factor 4 Neurod1 Neurogenic differentiation 1 Hes6 Hairy and enhancer of split 6 Npas3 Neuronal PAS domain protein 3, transcript variant 2 Hey1 Hairy/enhancer-of-split related with YRPW motif 1 Srebf1 Sterol regulatory element binding factor 1 Hif1a Hypoxia inducible factor 1, alpha subunit Tcf23 Transcription factor 23 Hif3a Hypoxia inducible factor 3, alpha subunit Tcf25 Transcription factor 25 Id1 Inhibitor of DNA binding 1 Tcf4 Transcription factor 4 Id2 Inhibitor of DNA binding 2 Tcfe2a Transcription factor E2a Id3 Inhibitor of DNA binding 3 Tcfe3 Transcription factor E3 Id4 Inhibitor of DNA binding 4 Tcfeb Transcription factor EB Max Max protein Usf2 Upstream transcription factor 2 Mesp2 Mesoderm posterior 2 HMG = High mobility group Bbx Bobby sox homolog Nsbp1 Nucleosome binding protein 1 Cic Capicua homolog Sox2 SRY-box containing gene 2 Hmga1 High mobility group AT-hook 1 Sox9 SRY-box containing gene 9 Hmga2 High mobility group AT-hook 2 Sox11 SRY-box containing gene 11 Hmgb1 High mobility group box 1 Sox12 SRY-box containing gene 12 Hmgb2 High mobility group box 2 Sox30 SRY-box containing gene 30 Hmgb3 High mobility group box 3 Ssrp1 Structure-specific recognition protein 1 Hmgn1 High mobility group nucleosomal binding domain 1 Taf1 TAF1 RNA polymerase II, TATA box binding protein (TBP)-associated factor Hmgn3 High mobility group nucleosomal binding domain 3 Tfam Transcription factor A, mitochondrial Mll3 Myeloid/lymphoid or mixed-lineage leukemia 3 T-box Tbx19 T-box19 ( Tpit ) Tbx3 T-box 3 Tbx2 T-box 2 852 Bmp4 and Fgf8 expression domains, which are critical for Rathke’s 2.9. Several T-box genes are expressed in the developing pituitary 882 853 pouch induction and proliferation, are expanded, resulting in a gland 883 854 dysmorphic Rathke’s pouch. Sox3 null mice also have hypopitu- 855 itarism with decreased levels of GH, LH, FSH, and TSH ( Woods et al., The T-box genes are a family of transcription factors that bind 884 856 2005 ). Sox2 heterozygous mice have a dysmorphic Rathke’s pouch, DNA through the ∼200 amino acid T-box domain. They are highly 885 857 which likely results from a similar mechanism as Sox3 in the ven- conserved, found in all metazoans and every known vertebrate 886 858 tral diencephalon. Unlike Sox3 , Sox2 is also expressed in Rathke’s genome. T-box genes can act as transcriptional regulators, either 887 859 pouch so that the hypopituitarism observed in Sox2 heterozygous as activators or repressors, in a context-dependent manner. T-box 888 860 mice may result from a direct affect of Sox2 in the anterior lobe genes are involved in early embryogenesis, extra-embryonic tissue 889 861 (Kelberman and Dattani, 2006 ). Both SOX2 and SOX3 are associated survival, cell fate decisions, embryonic patterning and organogen- 890 862 with human disorders; SOX3 causes X-linked panhypopituitarism esis (reviewed in Naiche et al., 2005 ). The first T-box gene to be 891 863 (Rizzoti et al., 2004 ), while SOX2 mutations cause anterior pitu- identified in the pituitary was Tbx19 (Tpit ) ( Lamolet et al., 2001 ). In 892 864 itary hypoplasia and hypogonadotropic hypogonadism ( Kelberman the absence of TPIT, the POMC lineages, including intermediate lobe 893 865 et al., 2006 ). melanotropes and anterior lobe corticotropes, fail to differentiate 894 866 NUPR1 (p8) is a high mobility group protein that was discov- fully ( Pulichino et al., 2003b ). In addition, TPIT inactivation results 895 867 ered as a differentially expressed gene in cell lines representing in a cell fate change, permitting prospective melanotropes to dif- 896 868 different stages of gonadotrope development ( Quirk et al., 2003 ). ferentiate into gonadotropes and Pou1f1 -independent thyrotropes. 897 869 It is expressed during late gestation in mice, concomitant with the The phenotype of mutant mice predicted the clinical characteristics 898 870 activation of the gonadotropin beta subunit genes. It is essential of human patients with TPIT mutations ( Pulichino et al., 2003a ). 899 871 for timely activation of gonadotropin expression ( Million Passe et We identified Tpit, Tbx2 and Tbx3 as genes expressed in the 900 872 al., 2008 ). In addition, it is implicated in pituitary tumorigenesis developing pituitary gland between e12.5 and e14.5 using our 901 873 (Brannon et al., 2007; MohammadUNCORRECTED et al., 2004 ). cDNA encyclopedia. PROOFTbx2 and Tbx3 expression was detected by RT- 902 874 Given the importance of HMG genes in pituitary development PCR in cDNA prepared from dissected pituitary glands at e12.5, 903 875 we screened our developmental library for expected and novel e14.5, e18.5, and in Prop1 df/df cDNA at e14.5. Localization of Tbx2 904 876 HMG family members expressed in the pituitary ( Table 2 ). We iden- and Tbx3 transcripts by in situ hybridization suggests that there is 905 877 tified 19 different HMG genes in our libraries. Five members of the little or no overlap of either gene expression pattern with that of 906 878 Sox group were identified, including Sox2. Mixed-lineage leukemia Tpit (Fig. 3 ) and ( Pontecorvi et al., 2008 ). Both genes are strongly 907 879 3 ( Mll3 ) is in this set. Myeloid leukemias have deletions in MLL3 expressed in the developing ventral diencephalon and the posterior 908 880 (Ruault et al., 2002 ). Further analysis of these HMG genes will enrich lobe of the pituitary gland. Prominent expression of Tbx3 overlaps 909 881 our understanding of pituitary gland development, and perhaps the area of the neural ectoderm where factors that induce pitu- 910 adenoma formation. itary growth, BMP and FGF, are expressed. There are no obvious 911 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 11 Fig. 3. Tbx3 expression in the ventral diencephalon and prospective posterior lobe of the pituitary gland during development. Tbx3 transcripts are readily detectable in mid-sagittal sections of developing normal and Prop1 df/df mouse embryos at e12.5 and e14.5 by in situ hybridization. 912 differences in Tbx3 expression in normal and Prop1 df/df mice ( Fig. 3 ). Sox2 and Sox3 , and Tpit . Expression studies suggest that Tbx2, 3 941 913 Tbx3 is expressed in the rostral tip, where Pou1f1- independent thy- have distinct functions from Tpit , and Lhx2 may act differently than 942 914 rotropes are located ( Pontecorvi et al., 2008 ). These transcriptional the Lhx3, 4 genes. The sheer number of genes in these families 943 915 repressors, Tbx2 and Tbx3 , might have functional overlap in regu- that are expressed at a time when they could have an important 944 916 lating posterior lobe development because the expression patterns impact suggests that the ideal strategy for identifying the genes 945 917 overlap, but neither gene has overlapping expression with Tpit . with essential functions is a high throughput screen. Given the 946 918 The function of these genes in pituitary development could best be strong parallels between the function of orthologous developmen- 947 919 determined using organ-specific inducible loss of function models tal regulators in fish and mammalian pituitary gland, it is possible 948 920 because of the embryonic lethality and ancillary organ defects char- that zebrafish could provide the basis for such a screen (reviewed 949 921 acteristic of mice homozygous for systemic null alleles ( Harrelson in Pogoda and Hammerschmidt, 2007 ). Alternatively, embryonic 950 922 et al., 2004; Davenport et al., 2003 ). stem cells have recently been coaxed to differentiate into pituitary 951 923 Several other T-box family members are expressed in and hormone-producing cells, suggesting embryonic stem cells might 952 924 around the developing pituitary gland, but they have not been be adaptable for screening studies ( Wagner et al., 2007; Zhao et 953 925 studied extensively and were not detected in our cDNA encyclo- al., 2005 ). Success with such a high throughput screening approach 954 926 pedia. These include Tbx15, Tbx18 , and the T-box brain gene 1, Tbr1 would be invaluable for nominating candidates for human muta- 955 927 (www.genepaint.org ). While Tbx18 and Tbr1 do not display over- tion screening in cases of hypopituitarism of unknown etiology. 956 928 lapping expression patterns with Tbx2, 3 , or Tpit , they appear to be 929 expressed in a subset of anterior lobe cells at e14.5, consistent with Acknowledgments 957 930 expression of Pou1f1 in that region. Funding : NIH R37HD030428, R01HD034283 (SAC); University of 958 Michigan Center for Computational Medicine and Biology, Clinical 959 931 3. Future directions UNCORRECTEDTranslational PROOF Science Award (SAC), Reproductive Sciences Training 960 Grant (NIH T32 HD07048 (SWD & BSE), NIH NRSA F32-HD046300 961 932 The developing pituitary cDNA libraries we made and analyzed (BSE), Endocrine Society, International Scholar’s Program (LC), 962 933 reveal that the transcriptome has great depth at the time the cells Novo-Nordisk and University of Michigan Center for Genetics in 963 934 are differentiating and the organ is undergoing substantial growth. Health and Medicine (FC). 964 935 There are already several compelling precedents for functional 936 overlap of transcription factors within a particular gene family, i.e. 965 937 PITX and LHX families. Thus, the discovery of many members of References 938 the Forkhead, HMG, bHLH and Tbox families, suggested that there Acampora, D., Mazan, S., Tuorto, F., Avantaggiato, V., Tremblay, J.J., Lazzaro, D., di 966 939 may be genes with essential functions that overlap important tran- Carlo, A., Mariano, A., Macchia, P.E., Corte, G., Macchia, V., Drouin, J., Brulet, P., 967 940 scription factors in these families like Foxl2, Nupr1, NeuroD1, Hes1, Simeone, A., 1998. Transient dwarfism and hypogonadism in mice lacking Otx1 968 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 12 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 969 reveal prepubescent stage-specific control of pituitary levels of GH, FSH and LH. Cha, K.B., Karolyi, I.J., Hunt, A., Wenglikowski, A.M., Wilkinson, J.E., Dolan, D.F., Dootz, 1055 970 Development 125, 1229–1239. G., Finnegan, A.A., Seasholtz, A.F., Hankenson, K.D., Siracusa, L.D., Camper, S.A., 1056 971 Ang, S.L., Rossant, J., 1994. HNF-3 beta is essential for node and notochord formation 2004a. Skeletal dysplasia and male infertility locus on mouse chromosome 9. 1057 972 in mouse development. Cell 78, 561–574. Genomics 83, 951–960. 1058 973 Arnold, H.H., Braun, T., 1996. Targeted inactivation of myogenic factor genes reveals Cha, K.B., Douglas, K.R., Potok, M.A., Liang, H., Jones, S.N., Camper, S.A., 2004b. WNT5A 1059 974 their role during mouse myogenesis: a review. Int. J. Dev. Biol. 40, 345–353. signaling affects pituitary gland shape. Mech. Dev. 121, 183–194. 1060 975 Badano, J.L., Katsanis, N., 2002. Beyond Mendel: an evolving view of human genetic Charles, M.A., Suh, H., Hjalt, T.A., Drouin, J., Camper, S.A., Gage, P.J., 2005. PITX 1061 976 disease transmission. Nat. Rev. Genet. 3, 779–789. genes are required for cell survival and Lhx3 activation. Mol. Endocrinol. 19, 1062 977 Bartke, A., Goldman, B.D., Bex, F., Dalterio, S., 1977. Effects of prolactin (PRL) 1893–1903. 1063 978 on pituitary and testicular function in mice with hereditary PRL deficiency. Charles, M.A., Saunders, T.L., Wood, W.M., Owens, K., Parlow, A.F., Camper, S.A., 1064 979 Endocrinology 101, 1760–1766. Ridgway, E.C., Gordon, D.F., 2006. Pituitary-specific Gata2 knockout: effects on 1065 980 Bhangoo, A.P., Hunter, C.S., Savage, J.J., Anhalt, H., Pavlakis, S., Walvoord, E.C., Ten, S., gonadotrope and thyrotrope function. Mol. Endocrinol. 20, 1366–1377. 1066 981 Rhodes, S.J., 2006. Clinical case seminar: a novel LHX3 mutation presenting as Charles, M.A., Mortensen, A.H., Potok, M.A., Camper, S.A., 2008. Pitx2 deletion in 1067 982 combined pituitary hormonal deficiency. J. Clin. Endocrinol. Metab. 91, 747–753. pituitary gonadotropes is compatible with gonadal development, puberty, and 1068 983 Blixt, A., Mahlapuu, M., Aitola, M., Pelto-Huikko, M., Enerback, S., Carlsson, P., 2000. fertility. Genesis 46, 507–514. 1069 984 A forkhead gene, FoxE3, is essential for lens epithelial proliferation and closure Cherrington, B.D., Bailey, J.S., Diaz, A.L., Mellon, P.L., 2008. NeuroD1 and Mash1 1070 985 of the lens vesicle. Genes Dev. 14, 245–254. temporally regulate GnRH receptor gene expression in immortalized mouse 1071 986 Blount, A.L., Schmidt, K., Justice, N.J., Vale, W.W., Fischer, W.H., Bilezikjian, L.M., gonadotrope cells. Mol. Cell. Endocrinol. 295, 106–114. 1072 987 2009. FoxL2 and Smad3 coordinately regulate follistatin gene transcription. J. Cho, J.H., Tsai, M.J., 2004. The role of BETA2/NeuroD1 in the development of the 1073 988 Biol. Chem. 284, 7631–7645. nervous system. Mol. Neurobiol. 30, 35–47. 1074 989 Bodner, M., Castrillo, J.L., Theill, L.E., Deerinck, T., Ellisman, M., Karin, M., 1988. The Chu, K., Nemoz-Gaillard, E., Tsai, M.J., 2001. BETA2 and pancreatic islet development. 1075 990 pituitary-specific transcription factor GHF-1 is a homeobox-containing protein. Recent Prog. Horm. Res. 56, 23–46. 1076 991 Cell 55, 505–518. Cogan, J.D., Wu, W., Phillips 3rd, J.A., Arnhold, I.J., Agapito, A., Fofanova, O.V., Osorio, 1077 992 Bottner, A., Keller, E., Kratzsch, J., Stobbe, H., Weigel, J.F., Keller, A., Hirsch, W., M.G., Bircan, I., Moreno, A., Mendonca, B.B., 1998. The PROP1 2-base pair deletion 1078 993 Kiess, W., Blum, W.F., Pfaffle, R.W., 2004. PROP1 mutations cause progressive is a common cause of combined pituitary hormone deficiency. J. Clin. Endocrinol. 1079 994 deterioration of anterior pituitary function including adrenal insufficiency: a Metab. 83, 3346–3349. 1080 995 longitudinal analysis. J. Clin. Endocrinol. Metab. 89, 5256–5265. Cohen, P., Rogol, A.D., Deal, C.L., Saenger, P., Reiter, E.O., Ross, J.L., Chernausek, 1081 996 Brannon, K.M., Million Passe, C.M., White, C.R., Bade, N.A., King, M.W., Quirk, C.C., S.D., Savage, M.O., Wit, J.M., 2008. Consensus statement on the diagnosis and 1082 997 2007. Expression of the high mobility group A family member p8 is essential treatment of children with idiopathic short stature: a summary of the Growth 1083 998 to maintaining tumorigenic potential by promoting cell cycle dysregulation in Hormone Research Society, the Lawson Wilkins Pediatric Endocrine Society, and 1084 999 LbetaT2 cells. Cancer Lett. 254, 146–155. the European Society for Paediatric Endocrinology Workshop. J. Clin. Endocrinol. 1085 1000 Brinkmeier, M.L., Potok, M.A., Cha, K.B., Gridley, T., Stifani, S., Meeldijk, J., Clevers, H., Metab. 93, 4210–4217. 1086 1001 Camper, S.A., 2003. TCF and Groucho-related genes influence pituitary growth Couly, G.F., Le Douarin, N.M., 1985. Mapping of the early neural primordium in 1087 1002 and development. Mol. Endocrinol. 17, 2152–2161. quail-chick chimeras. I. Developmental relationships between placodes, facial 1088 1003 Brinkmeier, M.L., Potok, M.A., Davis, S.W., Camper, S.A., 2007. TCF4 deficiency ectoderm, and prosencephalon. Dev. Biol. 110, 422–439. 1089 1004 expands ventral diencephalon signaling and increases induction of pituitary Crisponi, L., Deiana, M., Loi, A., Chiappe, F., Uda, M., Amati, P., Bisceglia, L., Zelante, 1090 1005 progenitors. Dev. Biol. 311, 396–407. L., Nagaraja, R., Porcu, S., Ristaldi, M.S., Marzella, R., Rocchi, M., Nicolino, M., 1091 1006 Brinkmeier, M.L., Davis, S.W., Carninci, P., MacDonald, J.W., Kawai, J., Ghosh, D., Lienhardt-Roussie, A., Nivelon, A., Verloes, A., Schlessinger, D., Gasparini, P., 1092 1007 Hayashizaki, Y., Lyons, R.H., Camper, S.A., 2009. Discovery of transcriptional Bonneau, D., Cao, A., Pilia, G., 2001. The putative forkhead transcription factor 1093 1008 regulators and signaling pathways in the developing pituitary gland by bioin- FOXL2 is mutated in blepharophimosis/ptosis/epicanthus inversus syndrome. 1094 1009 formatic and genomic approaches. Genomics 93, 449–460. Nat. Genet. 27, 159–166. 1095 1010 Brownell, I., Dirksen, M., Jamrich, M., 2000. Forkhead Foxe3 maps to the dysgenetic Cushman, L.J., Watkins-Chow, D.E., Brinkmeier, M.L., Raetzman, L.T., Radak, A.L., 1096 1011 lens locus and is critical in lens development and differentiation. Genesis 27, Lloyd, R.V., Camper, S.A., 2001. Persistent Prop1 expression delays gonadotrope 1097 1012 81–93. differentiation and enhances pituitary tumor susceptibility. Hum. Mol. Genet. 1098 1013 Bryant, J., Baxter, L., Cave, C.B., Milne, R., 2007. Recombinant growth hormone for 10, 1141–1153. 1099 1014 idiopathic short stature in children and adolescents. Cochrane Database Syst. Daly, A.F., Vanbellinghen, J.F., Khoo, S.K., Jaffrain-Rea, M.L., Naves, L.A., Guitelman, 1100 1015 Rev., CD004440. M.A., Murat, A., Emy, P., Gimenez-Roqueplo, A.P., Tamburrano, G., Raverot, G., 1101 1016 Buckwalter, M.S., Katz, R.W., Camper, S.A., 1991. Localization of the panhypopi- Barlier, A., De Herder, W., Penfornis, A., Ciccarelli, E., Estour, B., Lecomte, P., Gatta, 1102 1017 tuitary dwarf mutation (df) on mouse chromosome 11 in an intersubspecific B., Chabre, O., Sabate, M.I., Bertagna, X., Garcia Basavilbaso, N., Stalldecker, G., 1103 1018 backcross. Genomics 10, 515–526. Colao, A., Ferolla, P., Wemeau, J.L., Caron, P., Sadoul, J.L., Oneto, A., Archambeaud, 1104 1019 Cai, L., Morrow, E.M., Cepko, C.L., 2000. Misexpression of basic helix-loop-helix genes F., Calender, A., Sinilnikova, O., Montanana, C.F., Cavagnini, F., Hana, V., Solano, 1105 1020 in the murine cerebral cortex affects cell fate choices and neuronal survival. A., Delettieres, D., Luccio-Camelo, D.C., Basso, A., Rohmer, V., Brue, T., Bours, V., 1106 1021 Development 127, 3021–3030. Teh, B.T., Beckers, A., 2007. Aryl hydrocarbon receptor-interacting protein gene 1107 1022 Cai, C.L., Liang, X., Shi, Y., Chu, P.H., Pfaff, S.L., Chen, J., Evans, S., 2003. Isl1 identifies mutations in familial isolated pituitary adenomas: analysis in 73 families. J. Clin. 1108 1023 a cardiac progenitor population that proliferates prior to differentiation and Endocrinol. Metab. 92, 1891–1896. 1109 1024 contributes a majority of cells to the heart. Dev. Cell 5, 877–889. Dattani, M.T., Martinez-Barbera, J.P., Thomas, P.Q., Brickman, J.M., Gupta, R., 1110 1025 Cai, Y., Xu, Z., Nagarajan, L., Brandt, S.J., 2008. Single-stranded DNA-binding proteins Martensson, I.L., Toresson, H., Fox, M., Wales, J.K., Hindmarsh, P.C., Krauss, 1111 1026 regulate the abundance and function of the LIM-homeodomain transcription S., Beddington, R.S., Robinson, I.C., 1998. Mutations in the homeobox gene 1112 1027 factor LHX2 in pituitary cells. Biochem. Biophys. Res. Commun. 373, 303–308. HESX1/Hesx1 associated with septo-optic dysplasia in human and mouse. Nat. 1113 1028 Carlsson, P., Mahlapuu, M., 2002. Forkhead transcription factors: key players in Genet. 19, 125–133. 1114 1029 development and metabolism. Dev. Biol. 250, 1–23. Davenport, T.G., Jerome-Majewska, L.A., Papaioannou, V.E., 2003. Mammary gland, 1115 1030 Carninci, P., Shibata, Y., Hayatsu, N., Sugahara, Y., Shibata, K., Itoh, M., Konno, H., limb and yolk sac defects in mice lacking Tbx3, the gene mutated in human ulnar 1116 1031 Okazaki, Y., Muramatsu, M., Hayashizaki, Y., 2000. Normalization and subtrac- mammary syndrome. Development 130, 2263–2273. 1117 1032 tion of cap-trapper-selected cDNAs to prepare full-length cDNA libraries for Davis, S.W., Camper, S.A., 2007. Noggin regulates Bmp4 activity during pituitary 1118 1033 rapid discovery of new genes. Genome Res. 10, 1617–1630. induction. Dev. Biol. 305, 145–160. 1119 1034 Carninci, P., Waki, K., Shiraki, T., Konno, H., Shibata, K., Itoh, M., Aizawa, K., Arakawa, Davis, S.W., Potok, M.A., Brinkmeier, M.L., Carninci, P., Lyons, R.H., MacDonald, J.W., 1120 1035 T., Ishii, Y., Sasaki, D., Bono, H., Kondo, S., Sugahara, Y., Saito, R., Osato, N., Fleming, M.T., Mortensen, A.H., Egashira, N., Ghosh, D., Steel, K.P., Osamura, R.Y., 1121 1036 Fukuda, S., Sato, K., Watahiki, A., Hirozane-Kishikawa, T., Nakamura, M., Shi- Hayashizaki, Y., Camper, S.A., 2009. Genetics, gene expression and bioinformat- 1122 1037 bata, Y., Yasunishi, A., Kikuchi, N., Yoshiki, A., Kusakabe, M., Gustincich, S., ics of the pituitary gland. Horm. Res. 71 (Suppl. 2), 101–115. 1123 1038 Beisel, K., Pavan, W., Aidinis, V., Nakagawara, A., Held, W.A., Iwata, H., Kono, De Baere, E., Dixon, M.J., Small, K.W., Jabs, E.W., Leroy, B.P., Devriendt, K., Gillerot, 1124 1039 T., Nakauchi, H., Lyons, P., Wells, C., Hume, D.A., Fagiolini, M., Hensch, T.K., Y., Mortier, G., Meire, F., Van Maldergem, L., Courtens, W., Hjalgrim, H., 1125 1040 Brinkmeier, M., Camper, S., Hirota, J., Mombaerts, P., Muramatsu, M., Okazaki, Huang, S., Liebaers, I., Van Regemorter, N., Touraine, P., Praphanphoj, V., 1126 1041 Y., Kawai, J., Hayashizaki, Y., 2003. Targeting a complex transcriptome: the Verloes, A., Udar, N., Yellore, V., Chalukya, M., Yelchits, S., De Paepe, A., 1127 1042 construction of the mouse full-length cDNA encyclopedia. Genome Res. 13, Kuttenn, F., Fellous, M., Veitia, R., Messiaen, L., 2001. Spectrum of FOXL2 1128 1043 1273–1289. UNCORRECTEDgene mutations PROOF in blepharophimosis-ptosis-epicanthus inversus (BPES) fam- 1129 1044 Carvalho, L.R., Ward, R.D., Davis, S.W., Nishi, M.Y., Cogan, J.D., Donohoue, P.A., ilies demonstrates a genotype–phenotype correlation. Hum. Mol. Genet. 10, 1130 1045 Rhodes, S., Walvoord, E.C., Lyons, R.H., Phillips, J.A., Arnhold, I.J.P., Mendonca, 1591–1600. 1131 1046 B.B., Camper, S.A., 2007. Analysis of PROP1 transcriptional regulatory sequences Deladoey, J., Fluck, C., Buyukgebiz, A., Kuhlmann, B.V., Eble, A., Hindmarsh, P.C., Wu, 1132 1047 in patients with hypopituitarism. In: 89th Endocrine Society Annual Meeting, W., Mullis, P.E., 1999. “Hot spot” in the PROP1 gene responsible for combined 1133 1048 Toronto. pituitary hormone deficiency. J. Clin. Endocrinol. Metab. 84, 1645–1650. 1134 1049 Castinetti, F., Saveanu, A., Reynaud, R., Quentien, M.H., Buffin, A., Brauner, R., Kaffel, Douglas, K.R., Camper, S.A., 2000. Partial transcriptome of the developing pituitary 1135 1050 N., Albarel, F., Guedj, A.M., El Kholy, M., Amin, M., Enjalbert, A., Barlier, A., Brue, T., gland. Genomics 70, 335–346. 1136 1051 2008. A novel dysfunctional LHX4 mutation with high phenotypical variability Douglas, K.R., Brinkmeier, M.L., Kennell, J.A., Eswara, P., Harrison, T.A., Patrianakos, 1137 1052 in patients with hypopituitarism. J. Clin. Endocrinol. Metab. 93, 2790–2799. A.I., Sprecher, B.S., Potok, M.A., Lyons Jr., R.H., MacDougald, O.A., Camper, S.A., 1138 1053 Cau, E., Gradwohl, G., Fode, C., Guillemot, F., 1997. Mash1 activates a cascade of bHLH 2001. Identification of members of the Wnt signaling pathway in the embryonic 1139 1054 regulators in olfactory neuron progenitors. Development 124, 1611–1621. pituitary gland. Mamm. Genome 12, 843–851. 1140 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 13 1141 ElAmraoui, A., Dubois, P.M., 1993. Experimental evidence for the early commitment Herrera, E., Marcus, R., Li, S., Williams, S.E., Erskine, L., Lai, E., Mason, C., 2004. 1227 1142 of the presumptive adenohypophysis. Neuroendocrinology 58, 609–615. Foxd1 is required for proper formation of the optic chiasm. Development 131, 1228 1143 Ellsworth, B.S., Burns, A.T., Escudero, K.W., Duval, D.L., Nelson, S.E., Clay, C.M., 2003. 5727–5739. 1229 1144 The gonadotropin releasing hormone (GnRH) receptor activating sequence Himes, A.D., Raetzman, L.T., 2009. Premature differentiation and aberrant movement 1230 1145 (GRAS) is a composite regulatory element that interacts with multiple classes of of pituitary cells lacking both Hes1 and Prop1. Dev. Biol. 325, 151–161. 1231 1146 transcription factors including Smads, AP-1 and a forkhead DNA binding protein. Hintz, R.L., 2005. Growth hormone treatment of idiopathic short stature: clinical 1232 1147 Mol. Cell. Endocrinol. 206, 93–111. studies. Growth Horm. IGF Res. 15 (Suppl. A), S6–8. 1233 1148 Ellsworth, B.S., Egashira, N., Haller, J.L., Butts, D.L., Cocquet, J., Clay, C.M., Osamura, Hong, H.K., Lass, J.H., Chakravarti, A., 1999. Pleiotropic skeletal and ocular phe- 1234 1149 R.Y., Camper, S.A., 2006. FOXL2 in the pituitary: molecular, genetic, and devel- notypes of the mouse mutation congenital hydrocephalus (ch/Mf1) arise 1235 1150 opmental analysis. Mol. Endocrinol. 20, 2796–2805. from a winged helix/forkhead transcriptionfactor gene. Hum. Mol. Genet. 8, 1236 1151 Ellsworth, B.S., Butts, D.L., Camper, S.A., 2008. Mechanisms underlying pituitary 625–637. 1237 1152 hypoplasia and failed cell specification in Lhx3-deficient mice. Dev. Biol. 313, Iida, K., Koseki, H., Kakinuma, H., Kato, N., Mizutani-Koseki, Y., Ohuchi, H., Yosh- 1238 1153 118–129. ioka, H., Noji, S., Kawamura, K., Kataoka, Y., Ueno, F., Taniguchi, M., Yoshida, N., 1239 1154 Elshatory, Y., Gan, L., 2008. The LIM-homeobox gene Islet-1 is required for the devel- Sugiyama, T., Miura, N., 1997. Essential roles of the winged helix transcription 1240 1155 opment of restricted forebrain cholinergic neurons. J. Neurosci. 28, 3291–3297. factor MFH-1 in aortic arch patterning and skeletogenesis. Development 124, 1241 1156 Elshatory, Y., Everhart, D., Deng, M., Xie, X., Barlow, R.B., Gan, L., 2007. Islet-1 controls 4627–4638. 1242 1157 the differentiation of retinal bipolar and cholinergic amacrine cells. J. Neurosci. Ingraham, H.A., Chen, R.P., Mangalam, H.J., Elsholtz, H.P., Flynn, S.E., Lin, C.R., Sim- 1243 1158 27, 12707–12720. mons, D.M., Swanson, L., Rosenfeld, M.G., 1988. A tissue-specific transcription 1244 1159 Ericson, J., Norlin, S., Jessell, T.M., Edlund, T., 1998. Integrated FGF and BMP signaling factor containing a homeodomain specifies a pituitary phenotype. Cell 55, 1245 1160 controls the progression of progenitor cell differentiation and the emergence of 519–529. 1246 1161 pattern in the embryonic anterior pituitary. Development 125, 1005–1015. Jorge, A.A., Souza, S.C., Nishi, M.Y., Billerbeck, A.E., Liborio, D.C., Kim, C.A., Arnhold, 1247 1162 Ezzat, S., Zheng, L., Zhu, X.F., Wu, G.E., Asa, S.L., 2002. Targeted expression of a I.J., Mendonca, B.B., 2007. SHOX mutations in idiopathic short stature and Leri- 1248 1163 human pituitary tumor-derived isoform of FGF receptor-4 recapitulates pitu- Weill dyschondrosteosis: frequency and phenotypic variability. Clin. Endocrinol. 1249 1164 itary tumorigenesis. J. Clin. Invest. 109, 69–78. (Oxf.) 66, 130–135. 1250 1165 Fang, J., Dagenais, S.L., Erickson, R.P., Arlt, M.F., Glynn, M.W., Gorski, J.L., Seaver, L.H., Kaestner, K.H., Katz, J., Liu, Y., Drucker, D.J., Schutz, G., 1999. Inactivation of the 1251 1166 Glover, T.W., 2000. Mutations in FOXC2 (MFH-1), a forkhead family transcription winged helix transcription factor HNF3alpha affects glucose homeostasis and 1252 1167 factor, are responsible for the hereditary lymphedema-distichiasis syndrome. islet glucagon gene expression in vivo. Genes Dev. 13, 495–504. 1253 1168 Am. J. Hum. Genet. 67, 1382–1388. Kalfa, N., Philibert, P., Patte, C., Ecochard, A., Duvillard, P., Baldet, P., Jaubert, F., Fel- 1254 1169 Fauquier, T., Rizzoti, K., Dattani, M., Lovell-Badge, R., Robinson, I.C., 2008. SOX2- lous, M., Sultan, C., 2007. Extinction of FOXL2 expression in aggressive ovarian 1255 1170 expressing progenitor cells generate all of the major cell types in the adult mouse granulosa cell tumors in children. Fertil. Steril. 87, 896–901. 1256 1171 pituitary gland. Proc. Natl. Acad. Sci. U.S.A. 105, 2907–2912. Kalinichenko, V.V., Gusarova, G.A., Shin, B., Costa, R.H., 2003. The forkhead box F1 1257 1172 Finegold, D.N., Kimak, M.A., Lawrence, E.C., Levinson, K.L., Cherniske, E.M., Pober, B.R., transcription factor is expressed in brain and head mesenchyme during mouse 1258 1173 Dunlap, J.W., Ferrell, R.E., 2001. Truncating mutations in FOXC2 cause multiple embryonic development. Gene Expr. Patterns 3, 153–158. 1259 1174 lymphedema syndromes. Hum. Mol. Genet. 10, 1185–1189. Kelberman, D., Dattani, M.T., 2006. The role of transcription factors implicated in 1260 1175 Fratticci, A., Grieco, F.A., Spilioti, C., Giangaspero, F., Ventura, L., Esposito, V., Piccirilli, anterior pituitary development in the aetiology of congenital hypopituitarism. 1261 1176 M., Santoro, A., Gulino, A., Cantore, G., Alesse, E., Jaffrain-Rea, M.L., 2007. Differ- Ann. Med. 38, 560–577. 1262 1177 ential expression of neurogenins and NeuroD1 in human pituitary tumours. J. Kelberman, D., Dattani, M.T., 2007. Hypothalamic and pituitary development: novel 1263 1178 Endocrinol. 194, 475–484. insights into the aetiology. Eur. J. Endocrinol. 157 (Suppl. 1), S3–14. 1264 1179 Gage, P.J., Lossie, A.C., Scarlett, L.M., Lloyd, R.V., Camper, S.A., 1995. Ames dwarf mice Kelberman, D., Dattani, M.T., 2009. Role of transcription factors in midline central 1265 1180 exhibit somatotrope commitment but lack growth hormone-releasing factor nervous system and pituitary defects. Endocr. Dev. 14, 67–82. 1266 1181 response. Endocrinology 136, 1161–1167. Kelberman, D., Rizzoti, K., Avilion, A., Bitner-Glindzicz, M., Cianfarani, S., Collins, J., 1267 1182 Gage, P.J., Roller, M.L., Saunders, T.L., Scarlett, L.M., Camper, S.A., 1996. Anterior Chong, W.K., Kirk, J.M., Achermann, J.C., Ross, R., Carmignac, D., Lovell-Badge, 1268 1183 pituitary cells defective in the cell-autonomous factor, df, undergo cell lineage R., Robinson, I.C., Dattani, M.T., 2006. Mutations within Sox2/SOX2 are associ- 1269 1184 specification but not expansion. Development 122, 151–160. ated with abnormalities in the hypothalamo-pituitary–gonadal axis in mice and 1270 1185 Gage, P.J., Suh, H., Camper, S.A., 1999a. Dosage requirement of Pitx2 for development humans. J. Clin. Invest. 116, 2442–2455. 1271 1186 of multiple organs. Development 126, 4643–4651. Kidson, S.H., Kume, T., Deng, K., Winfrey, V., Hogan, B.L., 1999. The forkhead/winged- 1272 1187 Gage, P.J., Suh, H., Camper, S.A., 1999b. The bicoid-related Pitx gene family in devel- helix gene, Mf1, is necessary for the normal development of the cornea 1273 1188 opment. Mamm. Genome 10, 197–200. and formation of the anterior chamber in the mouse eye. Dev. Biol. 211, 1274 1189 Garcia-Lavandeira, M., Quereda, V., Flores, I., Saez, C., Diaz-Rodriguez, E., Japon, 306–322. 1275 1190 M.A., Ryan, A.K., Blasco, M.A., Dieguez, C., Malumbres, M., Alvarez, C.V., 2009. Kikuchi, M., Yatabe, M., Fujiwara, K., Takigami, S., Sakamoto, A., Soji, T., Yashiro, 1276 1191 A GRFa2/Prop1/stem (GPS) cell niche in the pituitary. PLoS ONE 4, e4815. T., 2006. Distinctive localization of N- and E-cadherins in rat anterior pituitary 1277 1192 Gardner, S., Maudsley, S., Millar, R.P., Pawson, A.J., 2007. Nuclear stabilization of beta- gland. Anat. Rec. A Discov. Mol. Cell. Evol. Biol. 288, 1183–1189. 1278 1193 catenin and inactivation of glycogen synthase kinase-3beta by gonadotropin- Kikuchi, M., Yatabe, M., Kouki, T., Fujiwara, K., Takigami, S., Sakamoto, A., Yashiro, 1279 1194 releasing hormone: targeting Wnt signaling in the pituitary gonadotrope. Mol. T., 2007. Changes in E- and N-cadherin expression in developing rat adenohy- 1280 1195 Endocrinol 21, 3028–3038. pophysis. Anat. Rec. (Hoboken) 290, 486–490. 1281 1196 Gleiberman, A.S., Fedtsova, N.G., Rosenfeld, M.G., 1999. Tissue interactions in the Kioussi, C., Briata, P., Baek, S.H., Rose, D.W., Hamblet, N.S., Herman, T., Ohgi, K.A., 1282 1197 induction of anterior pituitary: role of the ventral diencephalon, mesenchyme, Lin, C., Gleiberman, A., Wang, J., Brault, V., Ruiz-Lozano, P., Nguyen, H.D., Kem- 1283 1198 and notochord. Dev. Biol. 213, 340–353. ler, R., Glass, C.K., Wynshaw-Boris, A., Rosenfeld, M.G., 2002. Identification of a 1284 1199 Gordon, D.F., Lewis, S.R., Haugen, B.R., James, R.A., McDermott, M.T., Wood, Wnt/Dvl/beta-Catenin → Pitx2 pathway mediating cell-type-specific prolifera- 1285 1200 W.M., Ridgway, E.C., 1997. Pit-1 and GATA-2 interact and functionally coop- tion during development. Cell 111, 673–685. 1286 1201 erate to activate the thyrotropin beta-subunit promoter. J. Biol. Chem. 272, Kita, A., Imayoshi, I., Hojo, M., Kitagawa, M., Kokubu, H., Ohsawa, R., Ohtsuka, T., 1287 1202 24339–24347. Kageyama, R., Hashimoto, N., 2007. Hes1 and Hes5 control the progenitor pool, 1288 1203 Gordon, D.F., Woodmansee, W.W., Black, J.N., Dowding, J.M., Bendrick-Peart, J., intermediate lobe specification, and posterior lobe formation in the pituitary 1289 1204 Wood, W.M., Ridgway, E.C., 2002. Domains of Pit-1 required for transcrip- development. Mol. Endocrinol. 21, 1458–1466. 1290 1205 tional synergy with GATA-2 on the TSH beta gene. Mol. Cell. Endocrinol. 196, Kristrom, B., Zdunek, A.M., Rydh, A., Jonsson, H., Sehlin, P., Escher, S.A., 2009. A novel 1291 1206 53–66. mutation in the LIM homeobox 3 gene is responsible for combined pituitary 1292 1207 Gustavsson, P., Schoumans, J., Staaf, J., Jonsson, G., Carlsson, F., Kristoffersson, U., hormone deficiency, hearing impairment, and vertebral malformations. J. Clin. 1293 1208 Borg, A., Nordenskjold, M., Dahl, N., 2006. Hemizygosity for chromosome 2q14.2- Endocrinol. Metab. 94, 1154–1161. 1294 1209 q22.1 spanning the GLI2 and PROC genes associated with growth hormone Kume, T., Deng, K.Y., Winfrey, V., Gould, D.B., Walter, M.A., Hogan, B.L., 1998. The 1295 1210 deficiency, polydactyly, deep vein thrombosis and urogenital abnormalities. forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation 1296 1211 Clin. Genet. 69, 441–443. congenital hydrocephalus. Cell 93, 985–996. 1297 1212 Harrelson, Z., Kelly, R.G., Goldin, S.N., Gibson-Brown, J.J., Bollag, R.J., Silver, L.M., Kume, T., Deng, K., Hogan, B.L., 2000. Murine forkhead/winged helix genes Foxc1 1298 1213 Papaioannou, V.E., 2004. Tbx2 is essential for patterning the atrioventricular (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney 1299 1214 canal and for morphogenesis of the outflow tract during heart development. and urinary tract. Development 127, 1387–1395. 1300 1215 Development 131, 5041–5052. UNCORRECTEDLamba, P., Fortin, PROOF J., Tran, S., Wang, Y., Bernard, D.J., 2009. A novel role for the forkhead 1301 1216 Hatini, V., Huh, S.O., Herzlinger, D., Soares, V.C., Lai, E., 1996. Essential role of stro- transcription factor FOXL2 in activin A-regulated follicle-stimulating hormone 1302 1217 mal mesenchyme in kidney morphogenesis revealed by targeted disruption of beta subunit transcription. Mol. Endocrinol. 23, 1001–1013. 1303 1218 Winged Helix transcription factor BF-2. Genes Dev. 10, 1467–1478. Lamolet, B., Pulichino, A.M., Lamonerie, T., Gauthier, Y., Brue, T., Enjalbert, A., Drouin, 1304 1219 Henke, R.M., Meredith, D.M., Borromeo, M.D., Savage, T.K., Johnson, J.E., 2009. Ascl1 J., 2001. A pituitary cell-restricted T box factor, Tpit, activates POMC transcrip- 1305 1220 and Neurog2 form novel complexes and regulate Delta-like3 (Dll3) expression tion in cooperation with Pitx homeoproteins. Cell 104, 849–859. 1306 1221 in the neural tube. Dev. Biol. 328, 529–540. Lamolet, B., Poulin, G., Chu, K., Guillemot, F., Tsai, M.J., Drouin, J., 2004. 1307 1222 Hermesz, E., Williams-Simons, L., Mahon, K.A., 2003. A novel inducible element, Tpit-independent function of NeuroD1(BETA2) in pituitary corticotroph differ- 1308 1223 activated by contact with Rathke’s pouch, is present in the regulatory region of entiation. Mol. Endocrinol. 18, 995–1003. 1309 1224 the Rpx/Hesx1 homeobox gene. Dev. Biol. 260, 68–78. Lamonerie, T., Tremblay, J.J., Lanctot, C., Therrien, M., Gauthier, Y., Drouin, J., 1996. 1310 1225 Hernandez, L.M., Lee, P.D., Camacho-Hubner, C., 2007. Isolated growth hormone Ptx1, a bicoid-related homeo box transcription factor involved in transcription 1311 1226 deficiency. Pituitary 10, 351–357. of the pro-opiomelanocortin gene. Genes Dev. 10, 1284–1295. 1312 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 14 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 1313 Lanctot, C., Moreau, A., Chamberland, M., Tremblay, M.L., Drouin, J., 1999. Hindlimb Nadeau, J.H., 2003. Modifier genes and protective alleles in humans and mice. Curr. 1399 1314 patterning and mandible development require the Ptx1 gene. Development 126, Opin. Genet. Dev. 13, 290–295. 1400 1315 1805–1810. Naiche, L.A., Harrelson, Z., Kelly, R.G., Papaioannou, V.E., 2005. T-box genes in ver- 1401 1316 Laumonnier, F., Ronce, N., Hamel, B.C., Thomas, P., Lespinasse, J., Raynaud, M., tebrate development. Annu. Rev. Genet. 39, 219–239. 1402 1317 Paringaux, C., Van Bokhoven, H., Kalscheuer, V., Fryns, J.P., Chelly, J., Moraine, Nasonkin, I.O., Ward, R.D., Raetzman, L.T., Seasholtz, A.F., Saunders, T.L., Gillespie, 1403 1318 C., Briault, S., 2002. Transcription factor SOX3 is involved in X-linked mental P.J., Camper, S.A., 2004. Pituitary hypoplasia and respiratory distress syndrome 1404 1319 retardation with growth hormone deficiency. Am. J. Hum. Genet. 71, 1450–1455. in Prop1 knockout mice. Hum. Mol. Genet. 13, 2727–2735. 1405 1320 Lavoie, P.L., Budry, L., Balsalobre, A., Drouin, J., 2008. Developmental dependence Netchine, I., Sobrier, M.L., Krude, H., Schnabel, D., Maghnie, M., Marcos, E., Duriez, B., 1406 1321 on NurRE and EboxNeuro for expression of pituitary proopiomelanocortin. Mol. Cacheux, V., Moers, A., Goossens, M., Gruters, A., Amselem, S., 2000. Mutations 1407 1322 Endocrinol. 22, 1647–1657. in LHX3 result in a new syndrome revealed by combined pituitary hormone 1408 1323 Lehmann, O.J., Ebenezer, N.D., Jordan, T., Fox, M., Ocaka, L., Payne, A., Leroy, B.P., Clark, deficiency. Nat. Genet. 25, 182–186. 1409 1324 B.J., Hitchings, R.A., Povey, S., Khaw, P.T., Bhattacharya, S.S., 2000. Chromosomal Nishimura, D.Y., Searby, C.C., Alward, W.L., Walton, D., Craig, J.E., Mackey, D.A., 1410 1325 duplication involving the forkhead transcription factor gene FOXC1 causes iris Kawase, K., Kanis, A.B., Patil, S.R., Stone, E.M., Sheffield, V.C., 2001. A spectrum 1411 1326 hypoplasia and glaucoma. Am. J. Hum. Genet. 67, 1129–1135. of FOXC1 mutations suggests gene dosage as a mechanism for developmental 1412 1327 Leontiou, C.A., Gueorguiev, M., van der Spuy, J., Quinton, R., Lolli, F., Hassan, S., Chahal, defects of the anterior chamber of the eye. Am. J. Hum. Genet. 68, 364–372. 1413 1328 H.S., Igreja, S.C., Jordan, S., Rowe, J., Stolbrink, M., Christian, H.C., Wray, J., Bishop- Norquay, L.D., Yang, X., Jin, Y., Detillieux, K.A., Cattini, P.A., 2006. Hepatocyte nuclear 1414 1329 Bailey, D., Berney, D.M., Wass, J.A., Popovic, V., Ribeiro-Oliveira Jr., A., Gadelha, factor-3alpha binding at P sequences of the human growth hormone locus is 1415 1330 M.R., Monson, J.P., Akker, S.A., Davis, J.R., Clayton, R.N., Yoshimoto, K., Iwata, T., associated with pituitary repressor function. Mol. Endocrinol. 20, 598–607. 1416 1331 Matsuno, A., Eguchi, K., Musat, M., Flanagan, D., Peters, G., Bolger, G.B., Chap- Ohuchi, H., Hori, Y., Yamasaki, M., Harada, H., Sekine, K., Kato, S., Itoh, N., 2000. FGF10 1417 1332 ple, J.P., Frohman, L.A., Grossman, A.B., Korbonits, M., 2008. The role of the aryl acts as a major ligand for FGF receptor 2 IIIb in mouse multi-organ development. 1418 1333 hydrocarbon receptor-interacting protein gene in familial and sporadic pituitary Biochem. Biophys. Res. Commun. 277, 643–649. 1419 1334 adenomas. J. Clin. Endocrinol. Metab. 93, 2390–2401. Olson, L.E., Tollkuhn, J., Scafoglio, C., Krones, A., Zhang, J., Ohgi, K.A., Wu, W., 1420 1335 Lin, L., Miller, C.T., Contreras, J.I., Prescott, M.S., Dagenais, S.L., Wu, R., Yee, J., Orringer, Taketo, M.M., Kemler, R., Grosschedl, R., Rose, D., Li, X., Rosenfeld, M.G., 1421 1336 M.B., Misek, D.E., Hanash, S.M., Glover, T.W., Beer, D.G., 2002. The hepatocyte 2006. Homeodomain-mediated beta-catenin-dependent switching events dic- 1422 1337 nuclear factor 3 alpha gene, HNF3alpha (FOXA1), on chromosome band 14q13 tate cell-lineage determination. Cell 125, 593–605. 1423 1338 is amplified and overexpressed in esophageal and lung adenocarcinomas. Cancer Ormestad, M., Blixt, A., Churchill, A., Martinsson, T., Enerback, S., Carlsson, P., 2002. 1424 1339 Res. 62, 5273–5279. Foxe3 haploinsufficiency in mice: a model for Peters’ anomaly. Invest. Ophthal- 1425 1340 Loffler, K.A., Zarkower, D., Koopman, P., 2003. Etiology of ovarian failure in ble- mol. Vis. Sci. 43, 1350–1357. 1426 1341 pharophimosis ptosis epicanthus inversus syndrome: FOXL2 is a conserved, Osorio, M.G., Marui, S., Jorge, A.A., Latronico, A.C., Lo, L.S., Leite, C.C., Estefan, 1427 1342 early-acting gene in vertebrate ovarian development. Endocrinology 144, V., Mendonca, B.B., Arnhold, I.J., 2002. Pituitary magnetic resonance imaging 1428 1343 3237–3243. and function in patients with growth hormone deficiency with and without 1429 1344 Lu, C.H., Rincon-Limas, D.E., Botas, J., 2000. Conserved overlapping and reciprocal mutations in GHRH-R, GH-1, or PROP-1 genes. J. Clin. Endocrinol. Metab. 87, 1430 1345 expression of msh/Msx1 and apterous/Lhx2 in Drosophila and mice. Mech. Dev. 5076–5084. 1431 1346 99, 177–181. Ottolenghi, C., Omari, S., Garcia-Ortiz, J.E., Uda, M., Crisponi, L., Forabosco, A., Pilia, G., 1432 1347 Machinis, K., Amselem, S., 2005. Functional relationship between LHX4 and POU1F1 Schlessinger, D., 2005. Foxl2 is required for commitment to ovary differentiation. 1433 1348 in light of the LHX4 mutation identified in patients with pituitary defects. J. Clin. Hum. Mol. Genet. 14, 2053–2062. 1434 1349 Endocrinol. Metab. 90, 5456–5462. Pan, L., Deng, M., Xie, X., Gan, L., 2008. ISL1 and BRN3B co-regulate the differentiation 1435 1350 Machinis, K., Pantel, J., Netchine, I., Leger, J., Camand, O.J., Sobrier, M.L., Dastot-Le of murine retinal ganglion cells. Development 135, 1981–1990. 1436 1351 Moal, F., Duquesnoy, P., Abitbol, M., Czernichow, P., Amselem, S., 2001. Syn- Patel, L., McNally, R.J., Harrison, E., Lloyd, I.C., Clayton, P.E., 2006. Geographical 1437 1352 dromic short stature in patients with a germline mutation in the LIM homeobox distribution of optic nerve hypoplasia and septo-optic dysplasia in Northwest 1438 1353 LHX4. Am. J. Hum. Genet. 69, 961–968. England. J. Pediatr. 148, 85–88. 1439 1354 Masui, S., Nakatake, Y., Toyooka, Y., Shimosato, D., Yagi, R., Takahashi, K., Okochi, Pernasetti, F., Toledo, S.P., Vasilyev, V.V., Hayashida, C.Y., Cogan, J.D., Ferrari, C., 1440 1355 H., Okuda, A., Matoba, R., Sharov, A.A., Ko, M.S., Niwa, H., 2007. Pluripotency Lourenco, D.M., Mellon, P.L., 2000. Impaired adrenocorticotropin-adrenal axis 1441 1356 governed by Sox2 via regulation of Oct3/4 expression in mouse embryonic stem in combined pituitary hormone deficiency caused by a two-base pair deletion 1442 1357 cells. Nat. Cell Biol. 9, 625–635. (301-302delAG) in the prophet of Pit-1 gene. J. Clin. Endocrinol. Metab. 85, 1443 1358 McNay, D.E., Turton, J.P., Kelberman, D., Woods, K.S., Brauner, R., Papadimitriou, A., 390–397. 1444 1359 Keller, E., Keller, A., Haufs, N., Krude, H., Shalet, S.M., Dattani, M.T., 2007. HESX1 Pfaeffle, R.W., Savage, J.J., Hunter, C.S., Palme, C., Ahlmann, M., Kumar, P., Bellone, 1445 1360 mutations are an uncommon cause of septooptic dysplasia and hypopituitarism. J., Schoenau, E., Korsch, E., Bramswig, J.H., Stobbe, H.M., Blum, W.F., Rhodes, S.J., 1446 1361 J. Clin. Endocrinol. Metab. 92, 691–697. 2007. Four novel mutations of the LHX3 gene cause combined pituitary hormone 1447 1362 Mears, A.J., Jordan, T., Mirzayans, F., Dubois, S., Kume, T., Parlee, M., Ritch, R., Koop, B., deficiencies with or without limited neck rotation. J. Clin. Endocrinol. Metab. 92, 1448 1363 Kuo, W.L., Collins, C., Marshall, J., Gould, D.B., Pearce, W., Carlsson, P., Enerback, 1909–1919. 1449 1364 S., Morissette, J., Bhattacharya, S., Hogan, B., Raymond, V., Walter, M.A., 1998. Pfaeffle, R.W., Hunter, C.S., Savage, J.J., Duran-Prado, M., Mullen, R.D., Neeb, Z.P., 1450 1365 Mutations of the forkhead/winged-helix gene, FKHL7, in patients with Axenfeld- Eiholzer, U., Hesse, V., Haddad, N.G., Stobbe, H.M., Blum, W.F., Weigel, J.F., 1451 1366 Rieger anomaly. Am. J. Hum. Genet. 63, 1316–1328. Rhodes, S.J., 2008. Three novel missense mutations within the LHX4 gene are 1452 1367 Medema, R.H., Kops, G.J., Bos, J.L., Burgering, B.M., 2000. AFX-like Forkhead tran- associated with variable pituitary hormone deficiencies. J. Clin. Endocrinol. 1453 1368 scription factors mediate cell-cycle regulation by Ras and PKB through p27kip1. Metab. 93, 1062–1071. 1454 1369 Nature 404, 782–787. Pfaff, S.L., Mendelsohn, M., Stewart, C.L., Edlund, T., Jessell, T.M., 1996. Requirement 1455 1370 Mehta, A., Hindmarsh, P.C., Stanhope, R.G., Brain, C.E., Preece, M.A., Dattani, M.T., for LIM homeobox gene Isl1 in motor neuron generation reveals a motor neuron- 1456 1371 2003. Is the thyrotropin-releasing hormone test necessary in the diagnosis of dependent step in interneuron differentiation. Cell 84, 309–320. 1457 1372 central hypothyroidism in children. J. Clin. Endocrinol. Metab. 88, 5696–5703. Pogoda, H.M., Hammerschmidt, M., 2007. Molecular genetics of pituitary develop- 1458 1373 Million Passe, C.M., White, C.R., King, M.W., Quirk, P.L., Iovanna, J.L., Quirk, C.C., 2008. ment in zebrafish. Semin. Cell Dev. Biol. 18, 543–558. 1459 1374 Loss of the protein NUPR1 (p8) leads to delayed LHB expression, delayed ovar- Pontecorvi, M., Goding, C.R., Richardson, W.D., Kessaris, N., 2008. Expression of Tbx2 1460 1375 ian maturation, and testicular development of a sertoli-cell-only syndrome-like and Tbx3 in the developing hypothalamic-pituitary axis. Gene Expr. Patterns 8, 1461 1376 phenotype in mice. Biol. Reprod. 79, 598–607. 411–417. 1462 1377 Mirzayans, F., Gould, D.B., Heon, E., Billingsley, G.D., Cheung, J.C., Mears, A.J., Walter, Porter, F.D., Drago, J., Xu, Y., Cheema, S.S., Wassif, C., Huang, S.P., Lee, E., Grinberg, 1463 1378 M.A., 2000. Axenfeld-Rieger syndrome resulting from mutation of the FKHL7 A., Massalas, J.S., Bodine, D., Alt, F., Westphal, H., 1997. Lhx2, a LIM homeobox 1464 1379 gene on chromosome 6p25. Eur. J. Hum. Genet. 8, 71–74. gene, is required for eye, forebrain, and definitive erythrocyte development. 1465 1380 Mizuguchi, R., Kriks, S., Cordes, R., Gossler, A., Ma, Q., Goulding, M., 2006. Ascl1 and Development 124, 2935–2944. 1466 1381 Gsh1/2 control inhibitory and excitatory cell fate in spinal sensory interneurons. Potok, M.A., Cha, K.B., Hunt, A., Brinkmeier, M.L., Leitges, M., Kispert, A., Camper, 1467 1382 Nat. Neurosci. 9, 770–778. S.A., 2008. WNT signaling affects gene expression in the ventral diencephalon 1468 1383 Mody, S., Brown, M.R., Parks, J.S., 2002. The spectrum of hypopituitarism caused by and pituitary gland growth. Dev. Dyn. 237, 1006–1020. 1469 1384 PROP1 mutations. Best Pract. Res. Clin. Endocrinol. Metab. 16, 421–431. Poulin, G., Turgeon, B., Drouin, J., 1997. NeuroD1/beta2 contributes to cell-specific 1470 1385 Mohammad, H.P., Seachrist, D.D., Quirk, C.C., Nilson, J.H., 2004. Reexpression of p8 transcription of the proopiomelanocortin gene. Mol. Cell. Biol. 17, 6673–6682. 1471 1386 contributes to tumorigenic properties of pituitary cells and appears in a subset of Poulin, G., Lebel, M., Chamberland, M., Paradis, F.W., Drouin, J., 2000. Specific pro- 1472 1387 prolactinomas in transgenic mice thatUNCORRECTED hypersecrete luteinizing hormone. Mol. tein–protein PROOF interaction between basic helix-loop-helix transcription factors 1473 1388 Endocrinol. 18, 2583–2593. and homeoproteins of the Pitx family. Mol. Cell. Biol. 20, 4826–4837. 1474 1389 Monahan, P., Rybak, S., Raetzman, L.T., 2009. The Notch Target gene Hes1 regulates Procter, A.M., Phillips 3rd, J.A., Cooper, D.N., 1998. The molecular genetics of growth 1475 1390 cell cycle inhibitor expression in the developing pituitary. Endocrinology. hormone deficiency. Hum. Genet. 103, 255–272. 1476 1391 Mullen, R.D., Colvin, S.C., Hunter, C.S., Savage, J.J., Walvoord, E.C., Bhangoo, A.P., Ten, Pulichino, A.M., Vallette-Kasic, S., Couture, C., Gauthier, Y., Brue, T., David, M., 1477 1392 S., Weigel, J., Pfaffle, R.W., Rhodes, S.J., 2007. Roles of the LHX3 and LHX4 LIM- Malpuech, G., Deal, C., Van Vliet, G., De Vroede, M., Riepe, F.G., Partsch, C.J., 1478 1393 homeodomain factors in pituitary development. Mol. Cell. Endocrinol. 265-266, Sippell, W.G., Berberoglu, M., Atasay, B., Drouin, J., 2003a. Human and mouse 1479 1394 190–195. TPIT gene mutations cause early onset pituitary ACTH deficiency. Genes Dev. 1480 1395 Murre, C., McCaw, P.S., Vaessin, H., Caudy, M., Jan, L.Y., Jan, Y.N., Cabrera, C.V., Buskin, 17, 711–716. 1481 1396 J.N., Hauschka, S.D., Lassar, A.B., et al., 1989. Interactions between heterologous Pulichino, A.M., Vallette-Kasic, S., Tsai, J.P., Couture, C., Gauthier, Y., Drouin, J., 2003b. 1482 1397 helix-loop-helix proteins generate complexes that bind specifically to a common Tpit determines alternate fates during pituitary cell differentiation. Genes Dev. 1483 1398 DNA sequence. Cell 58, 537–544. 17, 738–747. 1484 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 15 1485 Quirk, C.C., Seachrist, D.D., Nilson, J.H., 2003. Embryonic expression of the luteinizing the Prophet of Pit-1 homeodomain factor defective in Ames dwarfism. Nature 1571 1486 hormone beta gene appears to be coupled to the transient appearance of p8, a 384, 327–333. 1572 1487 high mobility group-related transcription factor. J. Biol. Chem. 278, 1680–1685. Suh, H., Gage, P.J., Drouin, J., Camper, S.A., 2002. Pitx2 is required at multiple stages of 1573 1488 Raetzman, L.T., Ward, R., Camper, S.A., 2002. Lhx4 and Prop1 are required for cell sur- pituitary organogenesis: pituitary primordium formation and cell specification. 1574 1489 vival and expansion of the pituitary primordia. Development 129, 4229–4239. Development 129, 329–337. 1575 1490 Raetzman, L.T., Ross, S.A., Cook, S., Dunwoodie, S.L., Camper, S.A., Thomas, P.Q., 2004. Suh, H., Martin, D.M., Charles, M.A., Nasonkin, I.O., Gage, P.J., Camper, S.A., 2005. Role 1576 1491 Developmental regulation of Notch signaling genes in the embryonic pituitary: of PITX2 in the pituitary gland. In: Amendt, B.A. (Ed.), The Molecular Mechanism 1577 1492 Prop1 deficiency affects Notch2 expression. Dev. Biol. 265, 329–340. of Axenfeld–Rieger Syndrome. Springer, pp. 54–64. 1578 1493 Raetzman, L.T., Wheeler, B.S., Ross, S.A., Thomas, P.Q., Camper, S.A., 2006. Persistent Sund, N.J., Ang, S.L., Sackett, S.D., Shen, W., Daigle, N., Magnuson, M.A., Kaestner, 1579 1494 expression of Notch2 delays gonadotrope differentiation. Mol. Endocrinol. 20, K.H., 2000. Hepatocyte nuclear factor 3beta (Foxa2) is dispensable for main- 1580 1495 2898–2908. taining the differentiated state of the adult hepatocyte. Mol. Cell. Biol. 20, 1581 1496 Raetzman, L.T., Cai, J.X., Camper, S.A., 2007. Hes1 is required for pituitary growth 5175–5183. 1582 1497 and melanotrope specification. Dev. Biol. 304, 455–466. Sund, N.J., Vatamaniuk, M.Z., Casey, M., Ang, S.L., Magnuson, M.A., Stoffers, D.A., 1583 1498 Rajab, A., Kelberman, D., de Castro, S.C., Biebermann, H., Shaikh, H., Pearce, K., Hall, Matschinsky, F.M., Kaestner, K.H., 2001. Tissue-specific deletion of Foxa2 in 1584 1499 C.M., Shaikh, G., Gerrelli, D., Grueters, A., Krude, H., Dattani, M.T., 2008. Novel pancreatic beta cells results in hyperinsulinemic hypoglycemia. Genes Dev. 15, 1585 1500 mutations in LHX3 are associated with hypopituitarism and sensorineural hear- 1706–1715. 1586 1501 ing loss. Hum. Mol. Genet. 17, 2150–2159. Szeto, D.P., Rodriguez-Esteban, C., Ryan, A.K., O’Connell, S.M., Liu, F., Kioussi, C., 1587 1502 Reynaud, R., Gueydan, M., Saveanu, A., Vallette-Kasic, S., Enjalbert, A., Brue, T., Barlier, Gleiberman, A.S., Izpisua-Belmonte, J.C., Rosenfeld, M.G., 1999. Role of the 1588 1503 A., 2006. Genetic screening of combined pituitary hormone deficiency: experi- Bicoid-related homeodomain factor Pitx1 in specifying hindlimb morphogenesis 1589 1504 ence in 195 patients. J. Clin. Endocrinol. Metab. 91, 3329–3336. and pituitary development. Genes Dev. 13, 484–494. 1590 1505 Rhee, H., Polak, L., Fuchs, E., 2006. Lhx2 maintains stem cell character in hair follicles. Takuma, N., Sheng, H.Z., Furuta, Y., Ward, J.M., Sharma, K., Hogan, B.L., Pfaff, 1591 1506 Science 312, 1946–1949. S.L., Westphal, H., Kimura, S., Mahon, K.A., 1998. Formation of Rathke’s 1592 1507 Riepe, F.G., Partsch, C.J., Blankenstein, O., Monig, H., Pfaffle, R.W., Sippell, W.G., 2001. pouch requires dual induction from the diencephalon. Development 125, 1593 1508 Longitudinal imaging reveals pituitary enlargement preceding hypoplasia in two 4835–4840. 1594 1509 brothers with combined pituitary hormone deficiency attributable to PROP1 Tang, K., Bartke, A., Gardiner, C.S., Wagner, T.E., Yun, J.S., 1993. Gonadotropin secre- 1595 1510 mutation. J. Clin. Endocrinol. Metab. 86, 4353–4357. tion, synthesis, and gene expression in human growth hormone transgenic mice 1596 1511 Rimoin, D.L., Cohn, D., Krakow, D., Wilcox, W., Lachman, R.S., Alanay, Y., 2007. The and in Ames dwarf mice. Endocrinology 132, 2518–2524. 1597 1512 skeletal dysplasias: clinical–molecular correlations. Ann. N.Y. Acad. Sci. 1117, Thomas, J.O., 2001. HMG1 and 2: architectural DNA-binding proteins. Biochem. Soc. 1598 1513 302–309. Trans. 29, 395–401. 1599 1514 Rizzoti, K., Lovell-Badge, R., 2005. Early development of the pituitary gland: induc- Thomas, P.Q., Dattani, M.T., Brickman, J.M., McNay, D., Warne, G., Zacharin, M., 1600 1515 tion and shaping of Rathke’s pouch. Rev. Endocr. Metab. Disord. 6, 161–172. Cameron, F., Hurst, J., Woods, K., Dunger, D., Stanhope, R., Forrest, S., Robin- 1601 1516 Rizzoti, K., Brunelli, S., Carmignac, D., Thomas, P.Q., Robinson, I.C., Lovell-Badge, R., son, I.C., Beddington, R.S., 2001. Heterozygous HESX1 mutations associated with 1602 1517 2004. SOX3 is required during the formation of the hypothalamo-pituitary axis. isolated congenital pituitary hypoplasia and septo-optic dysplasia. Hum. Mol. 1603 1518 Nat. Genet. 36, 247–255. Genet. 10, 39–45. 1604 1519 Roberson, M.S., Schoderbek, W.E., Tremml, G., Maurer, R.A., 1994. Activation of the Treier, M., Gleiberman, A.S., O’Connell, S.M., Szeto, D.P., McMahon, J.A., McMa- 1605 1520 glycoprotein hormone alpha-subunit promoter by a LIM-homeodomain tran- hon, A.P., Rosenfeld, M.G., 1998. Multistep signaling requirements for pituitary 1606 1521 scription factor. Mol. Cell. Biol. 14, 2985–2993. organogenesis in vivo. Genes Dev. 12, 1691–1704. 1607 1522 Ruault, M., Brun, M.E., Ventura, M., Roizes, G., De Sario, A., 2002. MLL3, a new human Treier, M., O’Connell, S., Gleiberman, A., Price, J., Szeto, D.P., Burgess, R., Chuang, 1608 1523 member of the TRX/MLL gene family, maps to 7q36, a chromosome region fre- P.T., McMahon, A.P., Rosenfeld, M.G., 2001. Hedgehog signaling is required for 1609 1524 quently deleted in myeloid leukaemia. Gene 284, 73–81. pituitary gland development. Development 128, 377–386. 1610 1525 Salisbury, T.B., Binder, A.K., Grammer, J.C., Nilson, J.H., 2007. Maximal activity of the Tremblay, J.J., Lanctot, C., Drouin, J., 1998. The pan-pituitary activator of transcrip- 1611 1526 luteinizing hormone beta-subunit gene requires beta-catenin. Mol. Endocrinol. tion, Ptx1 (pituitary homeobox 1), acts in synergy with SF-1 and Pit1 and is an 1612 1527 21, 963–971. upstream regulator of the Lim-homeodomain gene Lim3/Lhx3. Mol. Endocrinol. 1613 1528 Salisbury, T.B., Binder, A.K., Grammer, J.C., Nilson, J.H., 2009. GnRH-regulated 12, 428–441. 1614 1529 expression of Jun and JUN target genes in gonadotropes requires a functional Tziaferi, V., Kelberman, D., Dattani, M.T., 2008. The role of SOX2 in hypogonadotropic 1615 1530 interaction between TCF/LEF family members and beta-catenin. Mol. Endocrinol. hypogonadism. Sex Dev. 2, 194–199. 1616 1531 23, 402–411. Uda, M., Ottolenghi, C., Crisponi, L., Garcia, J.E., Deiana, M., Kimber, W., Forabosco, 1617 1532 Savage, M.O., Attie, K.M., David, A., Metherell, L.A., Clark, A.J., Camacho-Hubner, A., Cao, A., Schlessinger, D., Pilia, G., 2004. Foxl2 disruption causes mouse ovar- 1618 1533 C., 2006. Endocrine assessment, molecular characterization and treatment of ian failure by pervasive blockage of follicle development. Hum. Mol. Genet. 13, 1619 1534 growth hormone insensitivity disorders. Nat. Clin. Pract. Endocrinol. Metab. 2, 1171–1181. 1620 1535 395–407. Vesper, A.H., Raetzman, L.T., Camper, S.A., 2006. Role of prophet of Pit1 (PROP1) in 1621 1536 Schmidt, D., Ovitt, C.E., Anlag, K., Fehsenfeld, S., Gredsted, L., Treier, A.C., Treier, M., gonadotrope differentiation and puberty. Endocrinology 147, 1654–1663. 1622 1537 2004. The murine winged-helix transcription factor Foxl2 is required for gran- Vierimaa, O., Georgitsi, M., Lehtonen, R., Vahteristo, P., Kokko, A., Raitila, A., Tup- 1623 1538 ulosa cell differentiation and ovary maintenance. Development 131, 933–942. purainen, K., Ebeling, T.M., Salmela, P.I., Paschke, R., Gundogdu, S., De Menis, 1624 1539 Semina, E.V., Reiter, R., Leysens, N.J., Alward, W.L., Small, K.W., Datson, N.A., Siegel- E., Makinen, M.J., Launonen, V., Karhu, A., Aaltonen, L.A., 2006. Pituitary ade- 1625 1540 Bartelt, J., Bierke-Nelson, D., Bitoun, P., Zabel, B.U., Carey, J.C., Murray, J.C., 1996. noma predisposition caused by germline mutations in the AIP gene. Science 1626 1541 Cloning and characterization of a novel bicoid-related homeobox transcription 312, 1228–1230. 1627 1542 factor gene, RIEG, involved in Rieger syndrome. Nat. Genet. 14, 392–399. Visel, A., Thaller, C., Eichele, G., 2004. GenePaint.org: an atlas of gene expression 1628 1543 Semina, E.V., Brownell, I., Mintz-Hittner, H.A., Murray, J.C., Jamrich, M., 2001. Muta- patterns in the mouse embryo. Nucleic Acids Res. 32, D552–D556. 1629 1544 tions in the human forkhead transcription factor FOXE3 associated with anterior Visel, A., Carson, J., Oldekamp, J., Warnecke, M., Jakubcakova, V., Zhou, X., Shaw, 1630 1545 segment ocular dysgenesis and cataracts. Hum. Mol. Genet. 10, 231–236. C.A., Alvarez-Bolado, G., Eichele, G., 2007. Regulatory pathway analysis by high- 1631 1546 Sharma, K., Sheng, H.Z., Lettieri, K., Li, H., Karavanov, A., Potter, S., Westphal, H., Pfaff, throughput in situ hybridization. PLoS Genet. 3, 1867–1883. 1632 1547 S.L., 1998. LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities Wagner, J., Lepore, D., Thomas, P., 2007. Differentiation of mouse embryonic stem 1633 1548 for motor neurons. Cell 95, 817–828. cells into growth hormone and prolactin expressing cells in vitro. Mol. Cell. 1634 1549 Sheng, H.Z., Westphal, H., 1999. Early steps in pituitary organogenesis. Trends Genet. Endocrinol. 273, 68–74. 1635 1550 15, 236–240. Wang, W., Wang, Y.G., Reginato, A.M., Glotzer, D.J., Fukai, N., Plotkina, S., Karsenty, 1636 1551 Sheng, H.Z., Zhadanov, A.B., Mosinger Jr., B., Fujii, T., Bertuzzi, S., Grinberg, A., Lee, G., Olsen, B.R., 2004. Groucho homologue Grg5 interacts with the transcription 1637 1552 E.J., Huang, S.P., Mahon, K.A., Westphal, H., 1996. Specification of pituitary cell factor Runx2-Cbfa1 and modulates its activity during postnatal growth in mice. 1638 1553 lineages by the LIM homeobox gene Lhx3. Science 272, 1004–1007. Dev. Biol. 270, 364–381. 1639 1554 Sheng, H.Z., Moriyama, K., Yamashita, T., Li, H., Potter, S.S., Mahon, K.A., West- Ward, R.D., Raetzman, L.T., Suh, H., Stone, B.M., Nasonkin, I.O., Camper, S.A., 2005. 1640 1555 phal, H., 1997. Multistep control of pituitary organogenesis. Science 278, Role of PROP1 in pituitary gland growth. Mol. Endocrinol. 19, 698–710. 1641 1556 1809–1812. Ward, R.D., Stone, B.M., Raetzman, L.T., Camper, S.A., 2006. Cell proliferation and vas- 1642 1557 Smith, R.S., Zabaleta, A., Kume, T., Savinova, O.V., Kidson, S.H., Martin, J.E., Nishimura, cularization in mouse models of pituitary hormone deficiency. Mol. Endocrinol. 1643 1558 D.Y., Alward, W.L., Hogan, B.L., John, S.W., 2000. Haploinsufficiency of the tran- 20, 1378–1390. 1644 1559 scription factors FOXC1 and FOXC2UNCORRECTED results in aberrant ocular development. Ward, R.D., Davis,PROOF S.W., Cho, M., Esposito, C., Lyons, R.H., Cheng, J.F., Rubin, 1645 1560 Hum. Mol. Genet. 9, 1021–1032. E.M., Rhodes, S.J., Raetzman, L.T., Smith, T.P., Camper, S.A., 2007. Comparative 1646 1561 Soares, M.J., Bartke, A., Colosi, P., Talamantes, F., 1984. Identification of a placental genomics reveals functional transcriptional control sequences in the Prop1 gene. 1647 1562 lactogen in pregnant Snell and Ames dwarf mice. Proc. Soc. Exp. Biol. Med. 175, Mamm. Genome. 1648 1563 106–108. Weedon, M.N., Frayling, T.M., 2008. Reaching new heights: insights into the genetics 1649 1564 Sobrier, M.L., Attie-Bitach, T., Netchine, I., Encha-Razavi, F., Vekemans, M., Amselem, of human stature. Trends Genet. 24, 595–603. 1650 1565 S., 2004. Pathophysiology of syndromic combined pituitary hormone deficiency Weinstein, D.C., Ruiz i Altaba, A., Chen, W.S., Hoodless, P., Prezioso, V.R., Jessell, 1651 1566 due to a LHX3 defect in light of LHX3 and LHX4 expression during early human T.M., Darnell Jr., J.E., 1994. The winged-helix transcription factor HNF-3 beta is 1652 1567 development. Gene Expr. Patterns 5, 279–284. required for notochord development in the mouse embryo. Cell 78, 575–588. 1653 1568 Sornson, M.W., Wu, W., Dasen, J.S., Flynn, S.E., Norman, D.J., O’Connell, S.M., Weiss, J., Meeks, J.J., Hurley, L., Raverot, G., Frassetto, A., Jameson, J.L., 2003. Sox3 is 1654 1569 Gukovsky, I., Carriere, C., Ryan, A.K., Miller, A.P., Zuo, L., Gleiberman, A.S., Ander- required for gonadal function, but not sex determination, in males and females. 1655 1570 sen, B., Beamer, W.G., Rosenfeld, M.G., 1996. Pituitary lineage determination by Mol. Cell. Biol. 23, 8084–8091. 1656 Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 G Model MCE 7393 1–16 ARTICLE IN PRESS 16 S.W. Davis et al. / Molecular and Cellular Endocrinology xxx (2009) xxx–xxx 1657 Winnier, G.E., Hargett, L., Hogan, B.L., 1997. The winged helix transcription factor Zhao, X., Teng, R., Asanuma, K., Okouchi, Y., Johkura, K., Ogiwara, N., Sasaki, K., 2005. 1678 1658 MFH1 is required for proliferation and patterning of paraxial mesoderm in the Differentiation of mouse embryonic stem cells into gonadotrope-like cells in 1679 1659 mouse embryo. Genes Dev. 11, 926–940. vitro. J. Soc. Gynecol. Investig. 12, 257–262. 1680 1660 Winnier, G.E., Kume, T., Deng, K., Rogers, R., Bundy, J., Raines, C., Walter, M.A., Hogan, Zhao, Y., Morales, D.C., Hermesz, E., Lee, W.K., Pfaff, S.L., Westphal, H., 2006. Reduced 1681 1661 B.L., Conway, S.J., 1999. Roles for the winged helix transcription factors MF1 and expression of the LIM-homeobox gene Lhx3 impairs growth and differentiation 1682 1662 MFH1 in cardiovascular development revealed by nonallelic noncomplementa- of Rathke’s pouch and increases cell apoptosis during mouse pituitary develop- 1683 1663 tion of null alleles. Dev. Biol. 213, 418–431. ment. Mech. Dev. 123, 605–613. 1684 1664 Woods, K.S., Cundall, M., Turton, J., Rizotti, K., Mehta, A., Palmer, R., Wong, J., Chong, Zhu, X., Zhang, J., Tollkuhn, J., Ohsawa, R., Bresnick, E.H., Guillemot, F., Kageyama, R., 1685 1665 W.K., Al-Zyoud, M., El-Ali, M., Otonkoski, T., Martinez-Barbera, J.P., Thomas, P.Q., Rosenfeld, M.G., 2006. Sustained Notch signaling in progenitors is required for 1686 1666 Robinson, I.C., Lovell-Badge, R., Woodward, K.J., Dattani, M.T., 2005. Over- and sequential emergence of distinct cell lineages during organogenesis. Genes Dev. 1687 1667 underdosage of SOX3 is associated with infundibular hypoplasia and hypopitu- 20, 2739–2753. 1688 1668 itarism. Am. J. Hum. Genet. 76, 833–849. Zhu, X., Gleiberman, A.S., Rosenfeld, M.G., 2007. Molecular physiology of pitu- 1689 1669 Zannini, M., Avantaggiato, V., Biffali, E., Arnone, M.I., Sato, K., Pischetola, M., Taylor, itary development: signaling and transcriptional networks. Physiol. Rev. 87, 1690 1670 B.A., Phillips, S.J., Simeone, A., Di Lauro, R., 1997. TTF-2, a new forkhead protein, 933–963. 1691 1671 shows a temporal expression in the developing thyroid which is consistent with Zucchini, S., 2008. Growth hormone use in the treatment of idiopathic short stature. 1692 1672 a role in controlling the onset of differentiation. EMBO J. 16, 3185–3197. Curr. Opin. Investig. Drugs 9, 396–401. 1693 1673 Zhao, L., Bakke, M., Krimkevich, Y., Cushman, L.J., Parlow, A.F., Camper, S.A., Parker, Zucchini, S., Wasniewska, M., Cisternino, M., Salerno, M., Iughetti, L., Maghnie, M., 1694 1674 K.L., 2001a. Hypomorphic phenotype in mice with pituitary-specific knockout Street, M.E., Caruso-Nicoletti, M., Cianfarani, S., 2008. Adult height in children 1695 1675 of steroidogenic factor 1. Genesis 30, 65–69. with short stature and idiopathic delayed puberty after different management. 1696 1676 Zhao, L., Bakke, M., Krimkevich, Y., Cushman, L.J., Parlow, A.F., Camper, S.A., Parker, Eur. J. Pediatr. 167, 677–681. 1697 1677 K.L., 2001b. Steroidogenic factor 1 (SF1) is essential for pituitary gonadotrope function. Development 128, 147–154. UNCORRECTED PROOF Please cite this article in press as: Davis, S.W., et al., Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Mol. Cell. Endocrinol. (2009), doi: 10.1016/j.mce.2009.12.012 PITUITARY STEM CELL UPDATE AND POTENTIAL IMPLICATIONS FOR TREATING HYPOPITUITARISM 1,2 Frederic Castinetti, 1Shannon W. Davis, 2Thierry Brue, 1Sally A. Camper 1University of Michigan Medical School, Ann Arbor, MI 41809-0618, USA 2Centre de Recherche en neurobiologie et neurophysiologie de Marseille (CRN2M), UMR6231, Faculté de Médecine Nord, Centre National de la Recherche Scientifique, Université de la Méditerranée and Centre de Référence des déficits hypophysaires, Hôpital de la Timone, Assistance Publique Hôpitaux de Marseille, 13385 Marseille, France Corresponding author: Sally A. Camper, Ph.D ., Dept. Human Genetics University of Michigan 4909 Buhl Bldg. 1241 Catherine St. Ann Arbor, MI 48109-5618 Telephone: 734-763-0682 Fax: 734-763-5831 email: [email protected] Keywords: SOX2, SOX9, nestin, GFRa2, side population, Sca1, S100ß, pituitary tumors, folliculo- stellate cells, cell cycle, proliferation, differentiation, CPHD, hypopituitarism Plan INTRODUCTION 1. PITUITARY GROWTH AND REGENERATION 2. POTENTIAL PITUITARY STEM CELL POPULATIONS a. The SOX2+ cell population b. The GFRa2+ cell population c. The “side population” cells d. The Nestin+ cell population e. The folliculo-stellate cells 3. PITUITARY STEM CELLS: FROM DIFFERENTIATION TO PATHOGENESIS 4. PITUITARY STEM CELLS AS POTENTIAL THERAPEUTIC TOOLS CONCLUSION !" " ABSTRACT Stem cells have been identified in organs with both low and high cell turnover rates. They are characterized by the expression of key marker genes for undifferentiated cells, the ability to self renew, and the ability to regenerate tissue after cell loss. Several recent reports present evidence for the presence of pituitary stem cells. Here we offer a critical review of the field and suggest additional studies that could resolve points of debate. Recent reports have relied on different markers, including SOX2, nestin, GFRa2, and SCA1, to identify pituitary stem cells and progenitors. Future studies will be needed to resolve the relationships between cells expressing these markers. Members of the Sox family of transcription factors are likely involved in the earliest steps of pituitary stem cell proliferation and the earliest transitions to differentiation. The transcription factor PROP1 and the NOTCH signaling pathway may regulate the transition to differentiation. Identification of the stem cell niche is an important step in understanding organ development. The niche may be the marginal zone around the lumen of Rathke's pouch, between the anterior and intermediate lobes of mouse pituitary, because cells in this region apparently give birth to all 6 pituitary hormone cell lineages. Stem cells have been shown to play a role in recurrent malignancies in some tissues, and their role in pituitary hyperplasia, pituitary adenomas and tumors is an important area of future investigation. From a therapeutic viewpoint, the ability to cultivate and grow stem cells in a pituitary pre-differentiation state might also be helpful for the long-term treatment of pituitary deficiencies. !" " INTRODUCTION specialized cells that produce polypeptide hormones that regulate many body processes. Embryonic stem cells have the These cells include somatotrophs that produce theoretical potential to give rise to all of the cell growth hormone (GH), lactotrophs producing types in the human body, raising the prospect of prolactin (PRL), gonadotrophs which secrete advances in the treatment of congenital and both luteinizing hormone (LH) and follicle acquired deficiencies. Stem cells exist in various stimulating hormone (FSH), thyrotrophs adult organs, including heart, brain, lungs, producing thyroid stimulating hormone (TSH), gonads, and many others. In all these organs, and corticotrophs that express pro- stem cells are characterized by two main criteria opimelanocortin (POMC) and cleave it to (1): produce adrenocorticotropin (ACTH). In early development the oral ectoderm is closely - The ability to self renew: stem cells juxtaposed with the neural ectoderm. In response undergo symmetric division to produce 2 to signals from the neural ectoderm, the oral stem cells . Cells can also divide ectoderm invaginates to produce Rathke's pouch, asymmetrically to give birth to 1 stem cell which later develops into the anterior lobe and and 1 committed progenitor cell. The latter intermediate lobe in the rodent. The mature gives rise to a daughter cell that proliferates intermediate lobe contains melanotrophs that and forms a population of expanding transit- express POMC, which is cleaved to produce amplifying cells before final differentiation. melanocyte stimulating hormone (MSH) and endorphin. The neural ectoderm evaginates to - The ability to regenerate tissue after cell produce the pituitary stalk or infundibulum and loss by completing the differentiation develops into the posterior lobe of the pituitary programs for multiple cell fates. gland, which differentiates and contains the terminals for the neurons that secrete oxytocin The existence and the nature of pituitary and vasopressin. All of the hormone-producing stem cells remains a matter of debate because the cell types differentiate by birth in the rodent, key criteria remain to be proven. The variety of although the sizes of the various cell populations markers and approaches used to identify change after birth as the gland grows in response pituitary progenitors and stem cells makes it to physiological needs. difficult to compare results and integrate the findings without further analyses. In this review Growth and proliferation levels in we summarize the literature and evaluate it Rathke’s pouch suggest two major time points critically, proposing experiments that could be for stem cell or progenitor cell activity: carried out to clarify the nature of pituitary embryogenesis and early post-natal days. progenitors and stem cells in rodents. In During mouse embryogenesis, between e11.5 addition, we will discuss the potential for human and e18.5, cells in the pituitary evolve from pituitary stem cells to contribute to malignancies primarily proliferating to mostly differentiating and to be harnessed for human therapeutic use. cells. At e13.5, there is a clear division in cell- cycle state between the dorsal side of the anterior lobe, which contains proliferating cells, and the 1. PITUITARY GROWTH AND ventral side where the first differentiated cells REGENERATION appear (3). At e14.5, proliferating cells are highly enriched around the lumen of Rathke's Fate mapping studies demonstrate that pouch, which is the zone described as the the vertebrate pituitary gland originates from the marginal zone or the niche for pituitary stem most anterior aspect of the anterior neural ridge cells ( Figure 1 ). The number of proliferating (2). The cells that become the hypothalamus are cells decreases progressively to adulthood, while located in the midline just posterior to the cells the number of differentiated cells increases (3). that become the anterior lobe, which contains Interestingly, terminally differentiated cells can !" " re-enter the cell cycle a few days after birth, recruited to fuel these population expansions are thereby increasing the population size for each not known. The low cell turnover rate in the cell type so that the neonate can function pituitary gland has been used to argue against the independently (4-6). existence of pituitary stem cells, but stem cells clearly exist in other organs with a low turn-over Yoshimura et al. described the first rate like heart and lungs (1). purported progenitor cells in the rat pituitary gland in 1969 (7). Transplantation of anterior All of this evidence makes the pituitary chromophobe cells into the possibility of pituitary stem cells very likely. hypothalamus of hypophysectomized rats led to Over the past 5 years, several groups have used viable, differentiated pituitary cells. innovative approaches to define new populations Chromophobe cells were defined as non- of multipotent pituitary progenitor cells. It will secreting cells, in contrast to acidophils, which be interesting to apply this new knowledge of secrete GH and/or PRL, and basophils secreting markers for stem cells and progenitors to better LH, FSH, TSH, and to a lesser extent ACTH. understand how the pituitary gland responds to The distinction between different cell types was these physiological challenges. not clear in this study, precluding any firm conclusions. Later studies suggested that the chromophobes were progenitors within the 2. POTENTIAL PITUITARY STEM CELL pituitary gland that could be induced to POPULATIONS differentiate further in response to hypothalamic releasing factors (8). Recent studies have reported potential populations of stem cells in the pituitary. Each Initial indirect evidence for pituitary study was focused on a different marker (SOX2, multipotent progenitors was based on the GFRa2, SCA1, nestin, S100ß… see Table 1 ) responsiveness of the pituitary to physiological expressed in stem cells of other organs, and or pathological conditions. This response could characterized the population of cells positive for employ a variety of mechanisms including these markers in the pituitary. pituitary stem cells, proliferation of committed progenitors, transdifferentiation of another type a. THE SOX2 + CELL POPULATION of secreting cell, or expansion of limited SOX2 is a member of the family of high potential precursors (9). Examples of the mobility group (HMG) box transcription factors. potential for pituitary expansion and It is required for the maintenance of several stem regeneration are manifold. During pregnancy cell populations in humans, for instance during there is a 3-fold increase in lactotroph cells, and central nervous system development (14). The the pituitary regenerates after tissue loss either best evidence for pituitary stem cells comes from from surgery or from immune diseases such as a study by Fauquier, Robinson and colleagues. hypophysitis (10). Transgenic mouse studies They discovered a SOX2 + cell population provide clear evidence for regeneration of GH located around the lumen in the marginal zone cells following massive cell loss: mice carrying between the anterior and the intermediate lobes the herpes virus thymidine kinase gene under the and scattered in small clusters inside the control of GH promoter exhibited ablation of pituitary gland of mice ( Figure 2 ). The lumen is more than 95% of the somatotroph cells after created when oral ectoderm cells that face the administration of an anti-herpes drug; GH cells oral cavity invaginate to produce Rathke's pouch were able to regenerate after 8 weeks following (5). These SOX2 + cells are initially present drug withdrawal (11). Acquired or congenital throughout Rathke’s pouch (e11.5), and they hypothyroidism can result in massive expansion become progressively restricted to the marginal of the thyrotrope cell population and thyrotroph zone around both sides of the lumen. In adult hyperplasia (12). Finally, the number of mice, SOX2 + cells represent 3-5% of the whole gonadotropes increases substantially during population of the anterior lobe, which is puberty (13). The type of progenitors that are consistent with the theoretical limited number of !" " stem cells in adult organs. For instance, less maintains pancreatic progenitors by stimulating than 1% of cells of mature adult organs their proliferation, survival and persistence in an (pancreas, skin, mammary gland) are known to undifferentiated state. Because SOX2 be able to form spheres that evolve into expression was not evaluated in the pancreas, it differentiated cells (15-17). is difficult to draw an analogy to pituitary SOX2+, SOX9+ cells. In contrast to the role of There is a stepwise progression of stem SOX9 in the pancreas, SOX9 seems to promote cell marker expression during embryogenesis: differentiation in the intestinal epithelium, as its cells only express SOX2 at e12.5, whereas they inactivation leads to increased proliferation of express SOX2 and SOX9 at e18.5. During progenitors, and differentiation into Paneth cells embryogenesis the SOX2+, SOX9- cells are (20, 21). These apparently different roles of highly proliferating, which is consistent with a SOX9 might be correlated with different levels role of progenitors involved in pituitary of expression in the progenitors or stem cells. In organogenesis. The location of these cells in the support of this idea, Sox9 transgenes with low developing pituitary is consistent with studies of expression support proliferative capacity, cell cycle markers that indicate a regional whereas high SOX9 expression suppresses separation of cells transitioning from proliferation and induces differentiation (22). In proliferation to differentiation (3, 18). the embryonic pituitary, SOX9 is probably a Interestingly, in adult pituitaries, the majority of marker of pituitary progenitor/transit amplifying SOX2+ cells also express SOX9, while only rare cells, representing one step after initial SOX2+ cells are SOX2+, SOX9-. Although the role of stem cells. In other words, stem cells would lose adult SOX2+ cells remains to be determined, it part of their differentiation and self-renewal is plausible that these cells are a reserve of potentials by acquiring SOX9 expression. quiescent, multipotent cells for organ SOX2+ SOX9- and SOX2+ SOX9+ cells might maintenance. These SOX2+ cells are indeed be called upon in case of tissue loss or in slowly dividing, a feature observed for some response to physiological demands. stem cells. The SOX2+, SOX9+ cells are more rapidly dividing, which is consistent with the The methodology of sphere formation idea that they represent transit amplifying cells. follows the progression of single cells into a Shortly after birth the level of proliferation of heterogeneous population of differentiated cells. SOX2+, SOX9+ cells is increased, in agreement Pituispheres cultured from dispersed adult with the wave of cell proliferation necessary to anterior pituitaries exhibited the ability to self accommodate pituitary growth in the first week renew and differentiate into hormone producing of life (19). We hypothesize that the progression cells. The pituispheres displayed a pattern of from SOX2+, SOX9- cells to SOX2+, SOX9+ marker expression evolving from SOX2, SCA1 cells observed during embryogenesis also occurs and E-cadherin positivity in the first few days to in adulthood. About half of the pituitary SOX2 + cells expressing SOX2, SOX9, and then S100 β cells also express epithelial markers like E- after 6-7 days (5). SCA1 expression was not cadherin, suggesting that an epithelial to detectable at the later time point, which is mesenchymal transition might be necessary consistent with another study focused on a side before these cells become rapidly dividing, population cells identified by cell sorting (23). differentiating cells. The delayed expression of S100ß in pituispheres favors a transient state of S100ß expression The role of SOX9+ cells in the pituitary during progenitor cell differentiation. is unclear. SOX9 has opposite roles in Pituispheres derived from adult pituitaries were development of the pancreas and intestinal able to form secondary spheres, demonstrating epithelium suggesting tissue-specific differences the ability to self-renew. In different cell culture in function. Seymour et al. demonstrated that conditions (mainly with Matrigel), these SOX9 is necessary in a mitotically active, Notch pituispheres differentiated into the five anterior responsive, subset of progenitors during pituitary cell lineages, demonstrating their pancreas development. SOX9 expression multipotent status. Some of the cells in the !" " pituispheres exhibited co-expression of SOX2 mice rescued the phenotype, but the essential and hormones, a feature not observed in vivo . cell types were not identified (25). The reason for this discrepancy is not clear. It will be important to assess the potential of b. THE GFRa2 + CELL POPULATION purified cell types to form pituispheres and self Garcia-Lavandeira, Alvarez and renew, i.e. fluorescence activated cell sorting colleagues recently used the glial cell-line (FACS) purified SOX2+, and SOX2+, SOX9+ derived neurotrophic factor (GDNF) receptor cells. This will establish the potential of each alpha 2 (GFRa2), as a marker for pituitary stem progenitor for self-renewal and differentiation. cells (26). This receptor is expressed in putative stem cells of the testis and ovaries (27). It The pituispheres derived from adult belongs to a family of glycosyl-phosphatidyl pituitary cells could be passaged for only two receptors that modulate signaling pathways generations, which falls short of expectations for induced by their ligands, the main one being stem cells, but it implies that the adult pituitary GDNF. In developing neurons, GDNF was does contain progenitors. The classical originally characterized as a growth factor definition of stem cells requires at least 5 promoting the survival of ventral midbrain passages as spheres (24). The limitation in dopaminergic neurons. Later, it was found to passage capacity might be due to the fact that have a potent survival effect on motor neurons these SOX2+ cells are already progenitors rather and other neuronal subpopulations in the central than stem cells. Alternatively, the apparent and peripheral nervous systems. In addition to limitation could be due to technical issues, such its role as a survival factor, GDNF is also as the lack of appropriate culture conditions. essential for proliferation, migration and Further studies may uncover growth factors or differentiation of neuronal cells (28). contacts with neighboring cells that simulate the niche and permit additional passages. A population of GFRa2 + cells exists in adulthood that are restricted to a single layer in To summarize , Fauquier and colleagues the marginal zone of the mouse pituitary gland have proven that SOX2+ cells have (Figure 2 ). In adults the population of GFRa2 + characteristics of progenitors and stem cells as cells represented less than 1% of the whole they normally do not express pituitary hormone pituitary cell population. Interestingly, more markers, are capable of at least limited self- than 90% of these cells expressed SOX2 and renewal, and are able to differentiate into the five SOX9, which would suggest that these cells are anterior pituitary cell types. SOX2 expression is the progenitors/transit amplifying cells described detectable during embryogenesis and at by Fauquier and colleagues (5). Other points of adulthood, consistent with the idea that the adult concordance are the expression of the epithelial pituitary gland has progenitors or stem cells that marker E-cadherin in the majority of cells, and can regenerate all the cell types in the organ, and S100 β in about 50% of the cells. Interestingly, that the process for adult cell replenishment may these GFRa2+ cells expressed PROP1, a be similar to the process that occurs in pituitary transcription factor important for the development. While it is exciting that SOX2+ POU1F1 lineage, comprised of somato- cells gave birth to all five anterior pituitary cell lactotroph and thyrotroph cell types (29). This lineages, it will be important to assess the might imply that the GFRa2+ cells are in a pre- capacity for these cells to expand and regenerate differentiated state. Given the critical role of a functional organ after tissue loss. An PROP1 during pituitary organogenesis, one interesting experiment would be to graft wild might expect pituitary stem cells to progress type Sox2+ stem cells into Prop1 mutant mice through a PROP1+ stage in order to produce new and evaluate their ability to correct the hormone producing cells. PROP1 is also hypoplasia and hypopituitarism that characterize required for normal E-cadherin expression, these mice (4). A similar kind of experiment which is present in all GFRa2 + cells (30). conducted in the early 1960’s showed that Finally, GFRa2 + cells were able to form primary transplanting normal pituitary cells in dwarf and secondary pituispheres and to differentiate !" " into the five pituitary cell lineages. It is sensitive dye efflux capacity (31). Vankelecom significant that a comparison of GFRa2+ cells and colleagues were the first to apply this with GFRa2- cells clearly showed that the latter technology to the pituitary gland and proceeded were not able to form pituispheres, suggesting to identify cells in mouse, rat and chicken with a that GFRa2- cells are not multipotent. Only a “side-population” phenotype ( Figure 2 ). These limited number of passages of GFRa2+ cells cells clonally replicate as non-adherent spheres were possible, similar to the observations by and express candidate stem cell markers (8). Fauquier et al. for SOX2+ cells. With special The side population isolated from the anterior medium conditions spheres derived from pituitary of 3-8 wk old mice represents 1.7% of GFRa2 + cells are capable of differentiation into the whole anterior pituitary cell population. This tubulin-beta III expressing cells, a marker side population is composed of cells expressing characteristic of neurons (26). Thus, these SCA1 at a high level (SCA1 high , about 60% of GFRa2+ cells are clearly multipotent the side population cells) and at a lower level progenitors, but they may not be truly “stem (Non-SCA1 high , less than 40% of the cells). cells”. SCA1, a putative stem cell marker (32), has been observed as a transient marker in pituispheres The precise role of GFRa2 in these cells derived from SOX2+ cells (5). In contrast, the remains to be determined. GFRa2 is not main population of pituitary cells, which is detectable when the cells express pituitary mostly differentiated cells, contains only 5% of hormone markers, suggesting that it has a so-called “SCA1 low cells” and 2.5% of SCA1 high transient role during the differentiation process. cells (the majority of cells being SCA1 It will be particularly interesting to establish the negative). It would be valuable to have a more expression profile of this receptor during precise definition and/or additional markers to embryonic development to determine whether describe the SCA1 high , Non-SCA1 high and GFRa2 is only present in adult stem cells, or if it SCA1 low cells. also plays a role in pituitary development during embryogenesis. To confirm the idea that GFRa2 Vankelecom and colleagues compared characterizes progenitors rather than stem cells, SCA1 high and Non-SCA1 high side population cells the size of the pituispheres should be evaluated. in adult mice (8, 23). The SCA1 high cells express Pituispheres are larger if they are derived from several stem cell markers, including OCT4 and stem cells rather than progenitors with more Nanog mRNA, and they express higher levels of limited potential. nestin, Bmi-1, Notch1 and HES1 compared to the main population (other anterior pituitary To summarize , the study by Garcia- cells). Some of these SCA1 high cells also Lavandeira and coworkers made very important expressed S100 !, and the fraction, about 6%, is contributions by demonstrating that GFRa2+ similar to the proportion observed in the anterior cells have the characteristics of progenitor/stem pituitary as a whole. This suggests that the cells as 1) they expressed no pituitary hormone SCA1 high cell population might contain folliculo- markers, 2) were capable of at least limited self stellate cells. Microarray analysis revealed renewal (they form pituispheres when cultured prominent expression of angiogenesis related with specific media), and 3) are able to genes in the SCA1 high cells. As SCA1 is differentiate into the five pituitary cell types. As involved in endothelial cell development (33), with the previous study, it will be important to these SCA1 high cells may be endothelial explore the ability of GFRa2+ cells to drive progenitors rather than pituitary progenitors. In additional self renewal and to regenerate tissue. contrast, SOX2 and SOX9 were expressed in about 50% of Non-SCA1 high cells and absent c. THE “SIDE POPULATION” CELLS from the SCA1 high cells. OCT4, Bmi-1 and FACS has been used to identify a side nestin displayed similar levels of expression population enriched in stem cells in several between SCA1 high and Non-SCA1 high tissues (31). The initial description of these cells populations, suggesting that these markers are in bone marrow was based on verapamil- probably not specific to pituitary stem cells, or !" " that SCA1 level does not perfectly define the To summarize , Vankelecom and colleagues stem cell population. Most of the transcription have made an important contribution by using factors involved in pituitary development cell sorting to purify cell populations and (LHX3, LHX4, PITX2, ISL1, PITX1, OTX2) examine gene expression and pluripotency. (29) were expressed at increased levels (2.5 to They have shown that the Non-SCA1 high side 25 fold more) in Non-SCA1 high cells relative to population cells express no pituitary hormone SCA1 high cells. Interestingly, the Non-SCA1 high markers, are capable of at least limited self cells also expressed PROP1 and HESX1. These renewal because they form pituispheres when transcription factors are normally expressed at cultured in specific media, and are able to different developmental times and are essential differentiate into the five anterior pituitary cell for normal pituitary organogenesis (29). Thus, types. Future studies will need to address the the Non-SCA1 high population is likely to be very potential of Non-SCA1 high side population cells heterogeneous, possibly including both stem to self renew more extensively and to expand to cells and progenitors at various steps of regenerate tissue. Gene expression profiling of differentiation. this cell population has been useful for comparing the side population with other studies Vankelecom and colleagues derived using a variety of markers, yet discrepancies in pituispheres from Non-SCA1 high cells using a SOX2 expression need to be resolved. All culture medium supplemented with growth studies of pituitary progenitors could benefit by factors. Neither the SCA1 high cells nor the main implementing FACS in order to analyze gene population of sorted cells were able to produce expression in cell populations purified using pituispheres. In six-day old spheres the majority other markers and to assess the potential for of the cells expressed SOX2 and nestin. pituisphere formation. Because 50% of Non-SCA1 high cells express SOX2, the cells that formed pituispheres may d. THE NESTIN + CELL POPULATION have been cells initially expressing SOX2. The Nestin is an intermediate filament presence of SOX2 expression at day 6 is protein mostly expressed in nerve cells where it concordant with previous studies (5, 26). Again, is implicated in the radial growth of axons (34). it was difficult to maintain a non-differentiated Nestin is a marker of stem cells in several types state during extended serial passaging. Second of tissues, including adult and embryonic neural generation spheres cultured in the appropriate stem cells (34). Gleiberman, Rosenfeld, medium gave birth to the five pituitary cell Enikolopov and colleagues used transgenic mice lineages and ceased expression of SOX2, as expressing GFP driven by regulatory elements of expected. The authors agree with Fauquier et al. the nestin gene to characterize a potential (5) that the SOX2+ cells are around the lumen in progenitor population in the pituitary (6). They the presumptive stem cell niche and in clusters also used a nestin-cre transgene and cre scattered over the anterior pituitary. The responsive GFP reporter strain to carry out relationship of SOX2+ cells to non-SCA1 high lineage tracing of nestin cells. This approach is cells is ambiguous. A unique marker for Non- different from the previous ones, in which the SCA1 high cells is necessary for identifying the cell population of interest was isolated from the location of these cells in the gland. Co- pituitary using an endogenous marker and expression of SOX2 and pituitary hormone cultured to form pituispheres. The transgenic markers was observed in some adult pituitary marker approach carries the inherent risk of cells. This is surprising because other studies ectopic expression of the transgene, and did not detect SOX2 in fully differentiated cells conclusions from this study are diminished (5, 26). The authors hypothesize that SOX2 because the transgene was not proven to might be localized to the cytoplasm during the recapitulate endogenous nestin gene expression. differentiation steps and localized to the nucleus Indeed, there are examples of various nestin-cre during the stem cell state. Further studies are transgenes that do not mimic endogenous nestin needed to clarify this issue. gene expression (35). Despite this caveat, the authors did prove that a set of genetically !" " marked cells in Rathke's pouch can differentiate were able to differentiate into each kind of into all hormone producing cell types. pituitary cell. This approach, however, was not based on pituisphere formation. Gleiberman and Nestin-GFP expressing cells were first colleagues suggest that the nestin+ cells might detected at e11.5 in the dorsal part of Rathke’s be pituitary stem cells in adulthood, in contrast pouch: they represented about 2% of the whole to other stem cells that might be used during cells of the pouch. In adulthood, cells in this embryogenesis: the number of GFP cells indeed region or "niche" express SOX2 and epithelial remains unchanged during embryogenesis, and markers cytokeratin 8 and EpCAM. The co- increases only after birth. The lack of expression of SOX2 and nestin during characterization of the nestin-GFP and nestin-cre development remains to be demonstrated. These transgenes relative to endogenous nestin GFP expressing cells also expressed LHX3, a expression weakens the conclusions that can be LIM domain transcription factor involved in the drawn from this study. Nestin could be mainly early steps of pituitary development (29). To expressed during embryogenesis in endothelial trace the lineage of these cells, the authors progenitors, as suggested by Chen et al. (23), crossed a nestin-cre line with the ROSA26-loxP- and in pituitary progenitors after birth. A precise stop-loxP-GFP reporter line. Progeny carrying evaluation of nestin expression coupled to cell both the cre reporter and the cre transgenes will markers during embryogenesis and adulthood is reveal the differentiated cells derived from the necessary to clarify this point. cre expressing progenitors. The cre-expressing precursor cells undergo cre-mediated Interestingly, GFRa2 + cells are all recombination of the reporter gene in the negative for nestin (26), whereas the NonSca1 high genomic DNA, which permanently marks the side population is positive for nestin. Both precursors, and the differentiated cells derived populations are expected to be SOX2+, SOX9+. from them, with GFP expression. The number This suggests four possibilities: a technical of GFP expressing cells represented a maximum difficulty underlies some of the results (26); the of 20% of the whole population in adulthood (5 pituitary contains distinct populations of months of age), whereas it only represented 2% multipotent progenitor cells with unique sets of of the anterior pituitary cells immediately after expressed genes; there is only one population of birth. A small fraction of the GFP+ cells progenitors that expresses different markers at expressed POU1F1 (PIT1), a pituitary each time point; or the Nestin-GFP transgene has transcription factor essential for establishing the ectopic expression in a set of cells within fate of somato-lactotroph and thyrotroph Rathke's pouch that happens to mark the stem lineages (29). POU1F1 expression in GFP+ cell niche (23). As nestin+ cells were mainly cells is consistent with them having achieved a dividing after birth, the authors’ final hypothesis differentiated state. Interestingly, early post- was that these cells were probably quiescent natal days were the only period (in comparison progenitors necessary for the initial wave of with later postnatal weeks) during which these pituitary growth after birth, and pituitary cells stained for Ki67, a marker of proliferating maintenance function at adulthood, different cells. Ward et al. had previously shown that from the ones involved during embryonic POU1F1 cells had a very high proliferation rate development (pituitary organ formation). in the neonatal period (4). The concept of nestin as a pituitary stem In adulthood, nestin-GFP positive cells cell marker is intriguing, but this idea requires were observed mainly in the marginal zone more rigorous investigation. The nestin (Figure 1 ), consistent with the localization of expression profile in the developing rat pituitary pituitary progenitor cells in other studies (5, 6). gland is dynamic (36). Nestin expression GFP expression from the cre-reporter was increased progressively after birth, between P5 detected in the five types of terminally and P12, and decreased precipitously at P21, a differentiated pituitary cells at adulthood, time when much of the pituitary organ growth suggesting that nestin-cre expressing progenitors has been achieved. Levels of nestin expression !" " remained significant, however, in adulthood and e. THE FOLLICULO-STELLATE in 2 year-old rats, particularly in zones that CELLS might correspond to hyperplastic aberrant Initial work suggested that folliculo- nodular growth. Two different morphological stellate cells could be pituitary stem cells (38- nestin + cell populations were observed: first, a 40). Folliculo-stellate cells are non-hormone filamentous cell type, close to the lumen, and producing, agranular cells, in contrast to secondly, cells close to blood vessels, likely pituitary polypeptide hormone secreting cells, involved in vascular development. The which are filled with secretory granules. The increased expression of nestin in hyperplastic folliculo-stellate cells are located in the nodules relative to normal tissue could be linked parenchymal tissue of the anterior pituitary lobe, to increased vascularization of the adenoma. mainly around the lumen of large follicles The nestin-GFP + population described by scattered throughout the lobe, and they constitute Gleiberman and colleagues (6) could include about 5-10% of the whole population of cells both kinds of nestin + cells. A portion of these (38-40). They have a stellate morphology would be stem cells expressing SOX2 and comprised of long cytoplasmic projections LHX3, whereas other nestin + cells would be between the glandular cells. Folliculo-stellate supportive cells involved in vascular cells have several roles in the pituitary, including development, and would be negative for SOX2 acting as scavenger cells with phagocytic and LHX3 expression. These data are consistent activity, and as supportive cells through both with a study that reported nestin expressing mechanical structure and by production of pituitary cells to be very heterogeneous, cytokines and growth factors. They also including a subpopulation frequently associated facilitate interactions between the endocrine and with endothelial cells (37). This hypothesis is the immune system, between pituitary cells via also in agreement with the study reported by their own functional network, and as supportive Chen et al. (23), which showed that nestin was cells for the GH network (41). equally present in the Non-Sca1 high cells, which are presumably pituitary progenitors, and the Their main markers are S100 β protein Sca1 high cells of the side population that are and GFAP (Glial Fibrillary Acidic Protein). likely endothelial progenitors. Interestingly, S100 β is detectable only after birth, first in the marginal zone (40). Using To summarize , Gleiberman and S100 β as a folliculo-stellate marker is colleagues have used a lineage tracing to ambiguous as not all folliculo-stellate cells determine which hormone cell types arise from express this marker. Moreover, S100 ! may not nestin-GFP+ cells. The transgene expressing be specific to the folliculo-stellate cells (40). cells also express key markers of the Indirect evidence suggests the possibility of undifferentiated state, and they seem to give retro-differentiation of endocrine cells into birth to all five pituitary lineages. In addition, folliculo-stellate cells (42), or expansion of the transgene expressing cells are proliferative folliculo-stellate compartment in parallel with during the post-natal growth period, when the gonadotrophs after castration in rats (43). It will demand for new progenitor cells is high. Future be valuable to have more precise markers for studies testing the capability of nestin-GFP+ folliculo-stellate cells and to perform more direct cells to form pituispheres and differentiate into experiments testing their ability to self renew each cell type would provide a valuable (pituisphere approach). confirmation of the ability of these cells to self- renew and exhibit pluripotency. Other important Lepore, Thomas and colleagues corollary experiments are to test the ability of described a pituitary cell population able to form nestin negative cells to form colonies, and to colonies in adult mice (pituitary colony forming compare transgene expression with endogenous cells representing 0.2% of the whole anterior gene expression. pituitary cells) ( Figure 2 ). These cells are contained in a subpopulation of pituitary cells that import fluorescent ß-Ala-Lys-N epsilon- !"# # AMCA (Y-amino-4-methylcoumarin-3-acetic We hypothesize that there are two acid). They are stellate, with long cytoplasmic critical roles of stem cells: one in establishing processes. All of them express S100ß and the pituitary gland during development and the GFAP, whereas only 40% express SCA1 and other involved in maintenance of the mature angiotensin converting enzyme, and a few pituitary gland in response to physiological express GH. This expression of both S100ß and challenges and normal cell turnover. The GFAP is consistent with the idea that these hypothesis of two different populations of stem colony-forming cells are folliculo-stellate cells. cells, one involved in embryogenesis and one No other theoretical pituitary stem cell marker involved in maintenance function after birth has been evaluated. The ability to self renew remains highly controversial. The fundamental and to differentiate into the 5 pituitary lineages question of whether adult cell renewal follows has to be shown to confirm the progenitor or the same or a different pathway than initial stem cell status of these cells (44, 45). embryonic differentiation remains to be resolved. A recent study reported ongoing work on transgenic mice expressing GFP under a cell An important next step in the analysis of specific promoter of the S100 β protein (46). pituitary stem cells is to expand the data on This approach, if not compromised by ectopic embryonic marker profiles, including co- expression of the transgene, might allow for cell expression studies with attention to spatial and sorting, expression profiling, and development temporal location of each cell type. This would of reliable markers as well as determination of expand on the critical studies by Fauquier et al., the potential of these cells to differentiate into which is the only group that examined pituitary hormone secreting cells. embryonic time points revealing the progression from SOX2+ to SOX2+, SOX9+. Examination of GFRa2, SCA1 and other markers during f. PROSPECTIVE FOR FUTURE embryogenesis, will be invaluable for ANALYSES establishing the steps in normal pituitary Future studies will facilitate a direct development. A brief summary of the different comparison of the stem cell populations markers used to characterize pituitary identified in these four reports by comparing progenitors and differentiated cells is given in each of the markers during similar periods of Table 1 . observation. A suggested schema of pituitary progenitor cell marker expression during progression to differentiated cells is shown in 3. PITUITARY STEM CELLS: FROM Figure 3 . None of the studies demonstrated DIFFERENTIATION TO PATHOGENESIS more than two rounds of pituisphere self- renewal. The difference in outcome observed In the pituitary, Notch might play a with Matrigel compared to classical growth major role in proliferation and fate selection. factors is interesting, and it suggests the HES1, a downstream target of Notch, is possibility that a specific three-dimensional necessary for suppression of differentiation: loss structure is necessary to promote the renewal of HES1 results in a cell fate switch such that and differentiation of the cells. Thus, another intermediate lobe cells differentiate as GH important advance will be to identify conditions hormone producing somatotropes instead of that permit five consecutive generations of POMC expressing melanotropes (47). This pituisphere formation. This will firmly establish suggests that intermediate lobe cells, normally the existence of stem cells, instead of fated to become melanotropes, can differentiate multipotent progenitors, which is the current into different pituitary cell lineages depending state of the art. Culture conditions leading to on transcription factor interactions (48). differentiated cells are summarized in Figure 4 . Interestingly, Chen et al. have shown that the Notch signaling system is active in the side population cells of the post-natal pituitary (49). !!" " Activation of Notch signaling increased the results in extinguishing LHX4 and ISL1 (another number of side population cells, whereas down LIM domain transcription factor) expression regulation of Notch reduced the proportion of (41). This suggests that LIM domain side population cells. The impact of this study transcription factors are involved in proliferation was limited by its reliance on the entire side rather than differentiation of progenitors and/or population cells, which is heterogeneous, stem cells. Surprisingly, however, pituitary containing both the subpopulation of Sca1 high hyperplasia has also been observed in some cells, which were reported as endothelial human patients with loss of function mutations progenitors by the same group (23), and Non- of LIM domain transcription factors (56, 57). Sca1 high cells, which are the potential pituitary PROP1, a paired-like homeodomain stem cells. Further study is needed to determine transcription factor, is probably involved later in the role of Notch signaling in the subset of Non- the pathway, promoting transition from Sca1 high side population cells. The Notch proliferation to differentiation, thereby signaling pathway might also interact with generating precursor cells capable of becoming SOX9 in the proliferation and differentiation the hormone-producing cells of the anterior lobe steps of progenitors, as demonstrated in pancreas (4). Interestingly, a recent report suggested that development. SOX9 controls the maintenance PROP1 might interact with the transcription of pluripotent pancreatic progenitors by factor SOX2 to promote POU1F1 dependant stimulating their proliferation and survival. lineage differentiation (somato-lactotroph and Interestingly, Sox9 -deficient progenitors have thyrotroph) (58). reduced expression of the Notch target HES1, suggesting possible interactions between SOX9 The roles of stem cells in the pathology and the Notch signaling pathway in stem cell of pituitary hypoplasia and hyperplasia are less maintenance or fate selection (50). Moreover, clear. We have some clues about the transitions there is a feedback regulatory loop between between proliferation and differentiation from PROP1, which is involved in the transition from patients with hypopituitarism. A good example proliferating progenitors to quiescent is the pituitary morphology modification induced differentiating cells, and Notch2: PROP1 by PROP1 mutations in combined pituitary activates expression of Notch2, and Notch hormone deficiencies (CPHD). Some patients signaling feeds back to maintain and/or enhance present with pituitary hypoplasia, while others PROP1 expression (18, 30, 51, 52). frequently present a transient pituitary hyperplasia, leading to a secondary hypoplasia Transcription factors from the LIM (29, 59, 60). Based on analyses of Prop1 mutant family, including ISL1, LHX3 and LHX4, are mice, Ward et al. implied that the pituitary likely to be involved in the early steps of hyperplasia in humans might be due to migratory pituitary stem cell and/or progenitor and/or cell adhesion defects of progenitors, that differentiation, as they are known to be essential produces dysmorphology and apparent for expansion of the pituitary primordium and overgrowth, before being eliminated through differentiation of hormone producing cells apoptosis. Proliferating cells are retained in the during embryogenesis (29, 53). There are perilumenal zone of the Prop1 df/df pituitaries, contradictory reports on co-expression of these instead of colonizing the anterior lobe at e12.5- transcription factors and SOX2 or SOX9 (5, 6, e14.5. This suggests that cells are unable to 26). The hypoplasia in Lhx3 and Lhx4 mutants differentiate, and that the ultimate hypoplasia of results from reduced proliferation and increased Prop1 mutant pituitaries might not be due to a cell death (53-55). Lhx3 and Lhx4 have failure of early progenitors to proliferate, but overlapping functions consistent with dosage rather to a defect in transitioning to sensitive effects, and part of the Lhx4 mutant differentiation and in seeding the anterior lobe phenotype could be attributable to a delay in with cells able to re-enter the cell cycle at later Lhx3 activation (53). Chen et al. have reported stages (4). This idea could also explain the that withdrawal of LIF, which is necessary to progressive nature of the hormone deficiency in maintain stem cells in an undifferentiated state, human patients. Cell cycle regulation of stem !"# # cells is probably under the control of several aged (2 years-old) rats have at least a 2-fold cyclin dependent kinases and kinase inhibitors. increase in nestin expression in apparent Landmark studies by Drouin and colleagues hyperplastic nodules. These nodules also had have demonstrated that 2 members of the enriched vasculature, and it is thus not clear Cip/Kip families of cell cycle inhibitors, p27 and whether the nestin cells were stem cells involved p57, play a major role in the p57 dependent cell in proliferation or supportive cells involved in cycle exit and progression to differentiation vasculature development. Moreover, the precise during embryogenesis. Cyclins D1, D2 and E nature of these masses (Glial-like or are important actors in the different steps of the adenomatous-like) has not been determined (36). cell cycle (3). Notch and HES1 also play roles Benign pituitary adenomas contain cells capable in the control of cell cycle exit (18, 30). An of producing pituispheres that express markers important area of future research is investigation of stem cells in other tissues, namely Oct4, of the regulation of the cell cycle, including exit CD90 and nestin. No SOX2, SOX9, SCA1 or for differentiation and re-entry for expansion of GFRa2 staining was performed, however (62). cell populations. AIP , encoding aryl hydrocarbon interacting protein, is correlated with increased risk of The hypothesis that hypoplasia can familial pituitary adenomas (63-65), and result from defects in the transition from members of the AIP complex are expressed proliferation to differentiation is supported by during mouse pituitary development (66, 67). studies in CDK4 deficient mice. CDK4 is This suggests the possibility that an early necessary for stem cells to enter into a transit developmental mechanism for growth regulation amplifying state leading to differentiated cells of progenitors may be involved in adenoma (61). Mice deficient in CDK4 have hypoplastic formation. Clearly, more basic studies on pituitaries with a dramatic reduction in all pituitary progenitors are needed as a foundation hormone secretory cells in the anterior pituitary for exploring the role of progenitors in adenoma during post-natal life (26). Interestingly, the development, progression and recurrence. number of progenitors, based on the marker GFRa2, was increased in the pituitaries of these mice. This suggests that CDK4 deficient 4. PITUITARY STEM CELLS AS progenitors are able to proliferate, but are unable POTENTIAL THERAPEUTIC TOOLS? to undergo differentiation into hormone producing cells. This situation is reversed by re- The presence of pituitary stem cells that expression of CDK4 (26). The reason why these can give rise to all pituitary hormone cell types CDK4 deficient progenitors do not enter the implies that these critical endocrine cells can be differentiated phase is unknown. replaced after loss or damage. These stem cells could thus be of major interest in the treatment Are stem or progenitor cells involved in of congenital (CPHD) or acquired pituitary tumorigenesis? Few studies have hypopituitarism (induced by surgery, addressed this issue. Gleiberman et al. observed radiotherapy or traumatic injury) (68, 69). nestin-GFP+ cells in pituitary tumors of nestin- GFP, retinoblastoma, Rb +/- , mice. Rb CPHD is characterized by multiple heterozygotes that carry one functional allele of pituitary hormone deficiencies, including retinoblastoma Rb1 gene develop tumors in the somatotroph, thyrotroph, lactotroph, intermediate lobe of the pituitary (6). These corticotroph, and/or gonadotroph deficiencies. cells express SOX2 and LHX3 but no pituitary The condition occurs between 1:3000 and hormones. These cells might be an 1:4000 births. It is important to diagnose and undifferentiated cell compartment connected to treat CPHD in order to avoid morbidity and the initiation and growth of pituitary tumors (6). mortality and to maintain a high quality of life On the other hand, these nestin-GFP+ cells (70). CPHD can be due to mutations of several might be indicative of modified vasculature genes encoding pituitary transcription factors induced by the tumor. Interestingly, adult and involved in pituitary ontogenesis, leading to !"# # predominantly GH deficiency combined with GH3 pituitary tumor cell line and reported trans- variable loss of other hormones (29). differentiation into GH and prolactin expressing Management requires an appropriate cells (73). While this suggests that factors replacement of hormone deficiencies, and strict secreted from somatotrophs and/or lactotrophs follow-up as delayed deficiencies commonly may be sufficient to induce trans-differentiation appear for some individuals with PROP1 of neural stem cells, these studies were limited deficiency (59, 60). Substitutive treatment by the lack of genetically marked cells which is remains challenging for all hormones: pituitary necessary to prove that the rat neural stem cells hormones are indeed not substituted, and were not contaminated with rat GH3 cells. In a peripheral hormones, though efficient to follow up study U et al. (46) grafted GFP decrease morbidity and allow a normal daily life, expressing fetal rat central nervous system stem do not ideally mimic the physiological secretions cells into adult rodent pituitary glands. The of each endocrine organ. Moreover, these authors reported for the first time that about 10% treatments have other drawbacks including of implanted cells eventually expressed potential side effects, high expense, and POU1F1, and secondarily GH, Prolactin and, sometimes daily injection (for instance for unexpectedly, FSHß. In contrast, expression of growth-hormone substitution). Growth hormone TSHß and ACTH was rarely observed. These treatment in adulthood remains a matter of cells survived for at least 4 weeks, acquiring the debate as contradictory data have been published morphology of original pituitary cells. This on beneficial effects in terms of bone, result suggests that undifferentiated cells can metabolism, and quality of life, among other differentiate into a specific cell type provided physiological measures (71). they are in an appropriate environment. Exposure to the pituitary host cells was essential If pituitary stem cells are able to self because the same cells grafted in the renew and give rise to a population of expanding hippocampus did not express any pituitary transit amplifying cells before final markers. This experiment is of major differentiation leading to all five pituitary importance for potential therapeutic approaches lineages, then these cells would be able to in the future. There are, however, several replace deficient pituitary cells and contribute to caveats that must be explored. The main organ regeneration. A few studies have reported concern is that the recipient rats had a normal the ability to differentiate embryonic stem cells pituitary. If remaining normal pituitary tissue is in vitro to obtain pituitary hormone secreting essential, the application to CPHD might be cells. These studies were based on embryonic significantly limited. Replication of this stem cells, which are derived from the inner cell approach in various animal models of pituitary mass of blastocyst embryos (72). Embryonic deficiency is important. In most of the plasticity stem cells have been isolated from several studies, genetically marked cells from one organ species including mice and humans. They have of an adult mouse apparently give rise to cell unlimited self-renewal ability, and are type characteristics of other organs following pluripotent, capable of generating differentiated transplantation. A critical aspect of the cells from ectoderm, mesoderm and endoderm observation of adult stem cell plasticity is that in tissues (1, 72). Embryonic stem cells can order for plasticity to occur, cell injury is differentiate into a wide variety of cell types, necessary. This suggests that micro- although predictability of differentiation and environmental exposure to the products of efficiency remain largely unsolved problems. injured cells may play a key role in determining the differentiated expression of stem cells after a Embryonic stem cells and neural stem graft (74). More research is necessary to cells are non-pituitary cells that each have the determine the molecular mechanism of cell capacity to differentiate into pituitary hormone injury and the molecular components of the host producing cells in culture. U et al. cultured pituitary cells that are important. neural stem cells derived from the fetal rat brain in the presence of medium conditioned by the !"# # Mouse embryonic stem cells have been that grafted GH cells might not replicate the full differentiated in vitro to produce pituitary differentiation program rendering them hormone secreting cells by two different unresponsive to appropriate feedback regulation, laboratories. Mouse embryonic stem cells from which could translate into insufficient hormone the 129/sv strain were cultured to form embryoid production or excessive, unregulated GH bodies, from which LHß and FSHß producing secretion, hyperplasia and/or development of cells were identified (75). No POU1F1 mRNA somatotroph adenomas. The length of survival was detected in the cultures, indicating the of these cells and their immune tolerance is also failure of these embryonic stem cells to generate a matter of question. The question of potential POU1F1 lineages. Another study produced immunosuppressive agents (or in contrast the embryoid bodies from mouse D3 embryonic way to obtain immune-compatible stem cells) stem cells (D3 embryonic stem cells are also will need to be addressed if stem cells are derived from a 129/sv line, but due to the genetic suggested as an alternate treatment to pituitary variation among the 129 substrains of mice we deficiencies in the coming years. T he use of do not know how related these two strains of ES adult stem cells as opposed to human embryonic cell are (76)), and detected differentiation of the stem cells could overcome the problem of POU1F1 lineages (77). In the absence or immunological rejection by being isolated from presence of GH3 conditioned media, these patients (82). Lepore et al. transplanted enriched embryoid bodies produced GH, prolactin, and populations of pituitary colony forming cells occasionally TSHß, albeit inefficiently. PROP1 (PCFC) (44, 45) into an in vivo microchamber in and POU1F1 expression was detected after 9 SCID mice (83). Donor cells survived in days in culture, and pituitary hormone markers chambers and underwent division. After 6 after day 15. This suggests that embryonic stem weeks, GH cells were detected in grafts, cell differentiation might progress with a time suggesting that PCFCs have the capacity to course that mimics normal pituitary divide and differentiate into somatotroph cells in development. Although some embryonic stem vivo. At least 2 points remain to be addressed. cells are able to concentrate hormone proteins Why would stem cells only differentiate into GH from the culture medium (78, 79), the cells and not the other types of cells? How demonstration that the embryoid bodies would these cells be regulated properly if grafted activated PROP1 expression and contain into the groin region, far from any hypothalamic hormone transcripts, makes this scenario very or pituitary stimuli? At this time it is difficult to unlikely. Further studies are necessary to extrapolate from these results to a cure for determine the critical differences in the culturing human hypopituitarism. conditions that produce these unique pituitary lineages and to enhance the efficiency and Finally, the scarce data observed about predictability of the process. Additional studies pathophysiological mechanisms of pituitary of transdifferentiation are also warranted as this hypo- and hyperplasia lead to a yet unanswered may be a more feasible method for controlled though crucial question: do patients presenting production of new hormone producing cells (80). with pituitary hyperplasia have an excess of pituitary progenitors? If so, it might mean that A challenge for any type of therapeutic the signals necessary for progenitor transplantation is to not simply produce hormone differentiation are lacking. In this case, grafts of but to do it in a fashion that is capable of undifferentiated stem cells would probably lead physiological rescue without risk of harmful side to persistent undifferentiated cells, secondary effects. Somatotroph cells are interconnected apoptosis, and ineffective treatment. In this via a functional network in the pituitary; this specific case, being able to better determine and network likely coordinates the levels of GH modify the signals necessary for differentiation secretion (pulsatility for instance) (81). Grafted might lead to interesting therapies. GH secreting cells might be required to interconnect to this network to become The discussion is obviously different in physiologically functional. There is some risk patients with pituitary hypoplasia, in which !"# # progenitors may be completely lacking, and for of the genes of pituitary transcription factors. For whom stem cells grafts could be an interesting instance, Takahashi et al. used Sox2 as inducer therapeutic approach. However, if these graft in mouse embryonic and adult fibroblasts to data are confirmed, it is also clear that the next obtain pluripotent stem cells (86). As a proof of step will be to improve the ratio of differentiated concept, induced pluripotent stem cells cells to progenitors. Pituitary progenitors cells generated from autologous skin have been used may yield more differentiatied pituitary cells to treat Sickle cell anemia in mouse (87). The than progenitors collected from other tissue main advantage would be to use a somatic cell types. Garcia-Lavendeira et al. showed that from the patient, which would assure a perfect GFRa2+ cells are able to differentiate into immune compatibility. neurons or pituitary hormone cell types. Perhaps the majority of pituitary progenitors are already predetermined to give birth to differentiated CONCLUSION pituitary cells compared to CNS tissue, which There are several lines of evidence suggesting has more obstacles to overcome to form pituitary the presence of stem cells in the pituitary, even if cells (which probably explains the 10% ratio of there are unresolved differences between the pituitary differentiation of CNS stem cells) (26). reports. These cells might play a major role during embryogenesis as they give birth to all 5 Replacing deficient cells by pre- pituitary lineages. The number of multipotent determined stem cells has already been progenitors in adulthood seems decreased. Their performed efficiently with bone marrow role is less evident. Our hypothesis is that the transplantation for instance. During the majority of them are already pre-differentiated, treatment of hematological malignancies, probably useful in case of tissue loss. Future patient’s cancerous cells are first destroyed by studies will need to optimize the culturing chemo/radiotherapy and replaced with conditions necessary to differentiate pituitary hematopoietic stem cells transplant from a stem cells into the specific cell types. Our human leukocyte antigen (HLA)-matched donor knowledge of the factors necessary for pituitary (84). Another option is to collect hematopoietic organogenesis will be useful in guiding these stem cells prior to the treatment and re-infuse experiments. Such work will increase our them into the patients after the course of the understanding of the mechanisms of aggressive chemotherapy, or to reconstitute differentiation and probably of pituitary immune cells as a treatment for autoimmune adenomas genesis, and will further the potential disorders (85). In this latter case, this has to be utility of pituitary stem as cells as therapeutic performed early during disease development to tools for pituitary deficiencies. avoid a destruction of stem cells contingent. We can hopefully hypothesize that future work on pituitary deficiency treatment will aim to ACKNOWLEDGEMENTS evaluate the possibility of collecting adult We thank Dr. Deborah Gumucio for critical pituitary stem cells from the patient (or from an reading of the manuscript. Financial support immune-compatible patient) and injecting them came from Novo-Nordisk, Societe Francaise as a substitutive treatment. A solution for the d’endocrinologie, Novartis, Ipsen, ADEREM, immuno-compatibility issue might be the use of and the Center for Genetics in Health and induced pluripotent stem cells: pituitary stem Medicine, University of Michigan (all for FC), cells could theoretically be obtained from adult and the National Institutes of Health somatic cell, by inducing a “forced” expression (R37HD30428, R01HD34283 to SAC). !"# # LEGEND TO FIGURES AND TABLE Figure 1: Schematic representation of the marginal zone, the niche of presumed pituitary stem cells. P, posterior lobe; I, intermediate lobe; A, anterior lobe; L, lumen. Figure 2: Different markers of potential pituitary stem cells a,b: Sox9 is expressed in the embryonic pituitary and defines two populations of Sox2+ cells. At e12.5 (a) the majority of Sox2+ cells are Sox9 negative (RP, Rathke’s pouch; A, anterior; P, posterior). At e18.5 (b) and at adult age, the majority of Sox2+ are Sox9+ cells; these cells are probably transit amplifying cells (5) (AL, anterior lobe; IL, intermediate lobe). c: The marginal zone of human adult pituitary contains a niche of progenitors expressing GFRa2 (26) (AP, anterior pituitary; MZ, marginal zone, NP, neuro- pituitary, posterior lobe). d: Colony forming pituitary cells in low-density culture colonies at day 8. Phase contrast low-power view showing stellate-shaped cells with long cytoplasmic processes (arrow a) and round refractile cells (arrow b). These cells express S100ß and GFAP, which are folliculo-stellate markers (44, 45). e: Nestin-GFP cells in mouse pituitary at P0. The cells are seen almost exclusively in the perilumenal area of pituitary (6) (AL, anterior lobe; IL, intermediate lobe, PL, posterior lobe). f: The adult mouse anterior pituitary contains a side population divided in Sca1 high (presumably endothelial progenitors) and non Sca1 high (presumably pituitary stem cells) expressing cells. Dual wave length FACS analysis reveals the presence of typical side population cells (1.7% of total living cells) in the adult anterior pituitary that are Hoechst low (A) and can be blocked by verapamil (B) (8, 23) (SP, side population; MP, main population). Figure 3: Stem cells during embryogenesis and adulthood Only one study (5) reported the outcome of a population of pituitary stem cells during embryogenesis to differentiate and form the terminal pituitary lineages. These cells evolve from Sox2 to Sox9 positivity. No other marker has been evaluated during embryogenesis. Nestin does not seem to play a role at this time point as nestin+ cells begin to proliferate after birth (6). In adults, rare cells express only Sox2. The majority of adult “progenitors” are Sox2+ Sox9+ (5). It is likely that the same cells express GFRa2 (26), AMCA and Angiotensin converting enzyme (ACE) (44), and have a low level of Sca1 expression (as side population cells). These cells are probably transit amplifying cells and are able of limited self renewal. Nestin may be expressed in a subset of these cells although it is not a specific marker of pituitary cells (6). Final differentiation requires a spatio-temporal regulation by several transcription factors and signaling pathways (review in (29)). Figure 4: Different ways to isolate and differentiate pituitary stem cells 3a (6). Lineage tracing involves isolation of nestin expressing cells based on a GFP transgene. Cells are cultured on gelatin. After 6-8 days culture in FGF2 medium and cholera toxin, colonies can be visualized. They can be handpicked individually, dispersed into single cells and cultured to evaluate their self-renewal capacity. After multilayer aggregates formation, differentiated cells can be observed. 3b (26). Anterior pituitary lobe cells are dissociated into single cells and sorted by FACS based on a stem cell marker (AMCA). Cells can be cultured on gelatin with mouse embryonic fibroblasts. To induce differentiation, growth medium can be replaced by a specific differentiation medium. Cells are coated on Collagen IV or poly-L-lysine culture slides. 3c . Anterior pituitary lobe cells are dissociated into single cells and sorted by FACS based on a specific characteristic (side population cells display verapamil-sensitive Hoechst dye efflux capacity) and/or a stem cell marker (Sca1, GFRa2). Cells can be seeded in dishes with medium enriched in growth factors (DMEM/Ham’s F12 supplemented with growth factors) (5, 23, 26). Whole anterior pituitary cells can also be used as a feeder layer (23). After 6-8 days, pituispheres appear (6). They can be handpicked individually, dispersed into single cells and cultured to evaluate their self-renewal capacity (generation of secondary spheres). To induce differentiation, growth medium can be replaced by a specific differentiation medium. Cells are coated on growth-factor reduced Matrigel (5, 23). !"# # Table 1: Summary of some markers used to characterize or potentially involved in pituitary stem cells. !"# # REFERENCES 1. Murry CE, Keller G 2008 Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell 132:661-680 2. Kawamura K, Kouki T, Kawahara G, Kikuyama S 2002 Hypophyseal development in vertebrates from amphibians to mammals. General and comparative endocrinology 126:130-135 3. Bilodeau S, Roussel-Gervais A, Drouin J 2009 Distinct developmental roles of cell cycle inhibitors p57Kip2 and p27Kip1 distinguish pituitary progenitor cell cycle exit from cell cycle reentry of differentiated cells. Molecular and cellular biology 29:1895-1908 4. Ward RD, Raetzman LT, Suh H, Stone BM, Nasonkin IO, Camper SA 2005 Role of PROP1 in pituitary gland growth. Molecular endocrinology (Baltimore, Md 19:698-710 5. Fauquier T, Rizzoti K, Dattani M, Lovell-Badge R, Robinson IC 2008 SOX2-expressing progenitor cells generate all of the major cell types in the adult mouse pituitary gland. Proceedings of the National Academy of Sciences of the United States of America 105:2907-2912 6. Gleiberman AS, Michurina T, Encinas JM, Roig JL, Krasnov P, Balordi F, Fishell G, Rosenfeld MG, Enikolopov G 2008 Genetic approaches identify adult pituitary stem cells. Proceedings of the National Academy of Sciences of the United States of America 105:6332-6337 7. Yoshimura F, Harumiya K, Ishikawa H, Otsuka Y 1969 Differentiation of isolated chromophobes into acidophils or basophils when transplanted into the hypophysiotrophic area of hypothalamus. Endocrinologia japonica 16:531-540 8. Chen J, Hersmus N, Van Duppen V, Caesens P, Denef C, Vankelecom H 2005 The adult pituitary contains a cell population displaying stem/progenitor cell and early embryonic characteristics. Endocrinology 146:3985-3998 9. Vidal S, Horvath E, Kovacs K, Cohen SM, Lloyd RV, Scheithauer BW 2000 Transdifferentiation of somatotrophs to thyrotrophs in the pituitary of patients with protracted primary hypothyroidism. Virchows Arch 436:43-51 10. Thodou E, Asa SL, Kontogeorgos G, Kovacs K, Horvath E, Ezzat S 1995 Clinical case seminar: lymphocytic hypophysitis: clinicopathological findings. The Journal of clinical endocrinology and metabolism 80:2302-2311 11. Borrelli E, Heyman RA, Arias C, Sawchenko PE, Evans RM 1989 Transgenic mice with inducible dwarfism. Nature 339:538-541 12. Alkhani AM, Cusimano M, Kovacs K, Bilbao JM, Horvath E, Singer W 1999 Cytology of pituitary thyrotroph hyperplasia in protracted primary hypothyroidism. Pituitary 1:291-295 13. Ebling FJ 2005 The neuroendocrine timing of puberty. Reproduction (Cambridge, England) 129:675-683 14. Episkopou V 2005 SOX2 functions in adult neural stem cells. Trends in neurosciences 28:219- 221 15. Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, Wicha MS 2003 In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & development 17:1253-1270 16. Seaberg RM, Smukler SR, Kieffer TJ, Enikolopov G, Asghar Z, Wheeler MB, Korbutt G, van der Kooy D 2004 Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages. Nature biotechnology 22:1115-1124 17. Fernandes KJ, McKenzie IA, Mill P, Smith KM, Akhavan M, Barnabe-Heider F, Biernaskie J, Junek A, Kobayashi NR, Toma JG, Kaplan DR, Labosky PA, Rafuse V, Hui CC, Miller FD 2004 A dermal niche for multipotent adult skin-derived precursor cells. Nature cell biology 6:1082-1093 18. Monahan P, Rybak S, Raetzman LT 2009 The Notch Target gene Hes1 regulates cell cycle inhibitor expression in the developing pituitary. Endocrinology 19. Carbajo-Perez E, Watanabe YG 1990 Cellular proliferation in the anterior pituitary of the rat during the postnatal period. Cell and tissue research 261:333-338 !"# # 20. Bastide P, Darido C, Pannequin J, Kist R, Robine S, Marty-Double C, Bibeau F, Scherer G, Joubert D, Hollande F, Blache P, Jay P 2007 Sox9 regulates cell proliferation and is required for Paneth cell differentiation in the intestinal epithelium. The Journal of cell biology 178:635-648 21. Mori-Akiyama Y, van den Born M, van Es JH, Hamilton SR, Adams HP, Zhang J, Clevers H, de Crombrugghe B 2007 SOX9 is required for the differentiation of paneth cells in the intestinal epithelium. Gastroenterology 133:539-546 22. Formeister EJ, Sionas AL, Lorance DK, Barkley CL, Lee GH, Magness ST 2009 Distinct SOX9 levels differentially mark stem/progenitor populations and enteroendocrine cells of the small intestine epithelium. Am J Physiol Gastrointest Liver Physiol 296:G1108-1118 23. Chen J, Gremeaux L, Fu Q, Liekens D, Van Laere S, Vankelecom H 2009 Pituitary progenitor cells tracked down by side population dissection. Stem cells (Dayton, Ohio) 27:1182-1195 24. Vankelecom H, Gremeaux L 2009 Stem cells in the pituitary gland: A burgeoning field. General and comparative endocrinology 25. Carsner RL, Rennels EG 1960 Primary site of gene action in anterior pituitary dwarf mice. Science (New York, NY 131:829 26. Garcia-Lavandeira M, Quereda V, Flores I, Saez C, Diaz-Rodriguez E, Japon MA, Ryan AK, Blasco MA, Dieguez C, Malumbres M, Alvarez CV 2009 A GRFa2/Prop1/stem (GPS) cell niche in the pituitary. PloS one 4:e4815 27. Hofmann MC 2008 Gdnf signaling pathways within the mammalian spermatogonial stem cell niche. Molecular and cellular endocrinology 288:95-103 28. Paratcha G, Ledda F 2008 GDNF and GFRalpha: a versatile molecular complex for developing neurons. Trends in neurosciences 31:384-391 29. Kelberman D, Rizzoti K, Lovell-Badge R, Robinson IC, Dattani MT 2009 Genetic regulation of pituitary gland development in human and mouse. Endocrine reviews 30:790-829 30. Himes AD, Raetzman LT 2009 Premature differentiation and aberrant movement of pituitary cells lacking both Hes1 and Prop1. Developmental biology 325:151-161 31. Challen GA, Little MH 2006 A side order of stem cells: the SP phenotype. Stem cells (Dayton, Ohio) 24:3-12 32. Asakura A, Seale P, Girgis-Gabardo A, Rudnicki MA 2002 Myogenic specification of side population cells in skeletal muscle. The Journal of cell biology 159:123-134 33. Cherqui S, Kurian SM, Schussler O, Hewel JA, Yates JR, 3rd, Salomon DR 2006 Isolation and angiogenesis by endothelial progenitors in the fetal liver. Stem cells (Dayton, Ohio) 24:44-54 34. Gilyarov AV 2008 Nestin in central nervous system cells. Neuroscience and behavioral physiology 38:165-169 35. Sclafani AM, Skidmore JM, Ramaprakash H, Trumpp A, Gage PJ, Martin DM 2006 Nestin- Cre mediated deletion of Pitx2 in the mouse. Genesis 44:336-344 36. Gautron L, De-Smedt V, Laye S 2009 Age-Related Changes in Nestin Immunoreactivity in the Rat Pituitary Gland. Neuroendocrinology 37. Krylyshkina O, Chen J, Mebis L, Denef C, Vankelecom H 2005 Nestin-immunoreactive cells in rat pituitary are neither hormonal nor typical folliculo-stellate cells. Endocrinology 146:2376- 2387 38. Vankelecom H 2007 Stem cells in the postnatal pituitary? Neuroendocrinology 85:110-130 39. Vankelecom H 2007 Non-hormonal cell types in the pituitary candidating for stem cell. Seminars in cell & developmental biology 18:559-570 40. Devnath S, Inoue K 2008 An insight to pituitary folliculo-stellate cells. Journal of neuroendocrinology 20:687-691 41. Fauquier T, Guerineau NC, McKinney RA, Bauer K, Mollard P 2001 Folliculostellate cell network: a route for long-distance communication in the anterior pituitary. Proceedings of the National Academy of Sciences of the United States of America 98:8891-8896 42. Horvath E, Kovacs K 2002 Folliculo-stellate cells of the human pituitary: a type of adult stem cell? Ultrastructural pathology 26:219-228 !"# # 43. Shirasawa N, Kihara H, Yamaguchi S, Yoshimura F 1983 Pituitary folliculo-stellate cells immunostained with S-100 protein antiserum in postnatal, castrated and thyroidectomized rats. Cell and tissue research 231:235-249 44. Lepore DA, Jokubaitis VJ, Simmons PJ, Roeszler KN, Rossi R, Bauer K, Thomas PQ 2006 A role for angiotensin-converting enzyme in the characterization, enrichment, and proliferation potential of adult murine pituitary colony-forming cells. Stem cells (Dayton, Ohio) 24:2382-2390 45. Lepore DA, Roeszler K, Wagner J, Ross SA, Bauer K, Thomas PQ 2005 Identification and enrichment of colony-forming cells from the adult murine pituitary. Experimental cell research 308:166-176 46. U HS, Wu B, Wilkes N, Ho A, Saljooque F 2007 Brain stem cells adopt a pituitary fate after implantation into the adult rodent pituitary gland. Neuroendocrinology 86:58-68 47. Raetzman LT, Cai JX, Camper SA 2007 Hes1 is required for pituitary growth and melanotrope specification. Developmental biology 304:455-466 48. Pulichino AM, Vallette-Kasic S, Tsai JP, Couture C, Gauthier Y, Drouin J 2003 Tpit determines alternate fates during pituitary cell differentiation. Genes & development 17:738-747 49. Chen J, Crabbe A, Van Duppen V, Vankelecom H 2006 The notch signaling system is present in the postnatal pituitary: marked expression and regulatory activity in the newly discovered side population. Molecular endocrinology (Baltimore, Md 20:3293-3307 50. Seymour PA, Freude KK, Tran MN, Mayes EE, Jensen J, Kist R, Scherer G, Sander M 2007 SOX9 is required for maintenance of the pancreatic progenitor cell pool. Proceedings of the National Academy of Sciences of the United States of America 104:1865-1870 51. Zhu X, Zhang J, Tollkuhn J, Ohsawa R, Bresnick EH, Guillemot F, Kageyama R, Rosenfeld MG 2006 Sustained Notch signaling in progenitors is required for sequential emergence of distinct cell lineages during organogenesis. Genes & development 20:2739-2753 52. Kita A, Imayoshi I, Hojo M, Kitagawa M, Kokubu H, Ohsawa R, Ohtsuka T, Kageyama R, Hashimoto N 2007 Hes1 and Hes5 control the progenitor pool, intermediate lobe specification, and posterior lobe formation in the pituitary development. Molecular endocrinology (Baltimore, Md 21:1458-1466 53. Raetzman LT, Ward R, Camper SA 2002 Lhx4 and Prop1 are required for cell survival and expansion of the pituitary primordia. Development (Cambridge, England) 129:4229-4239 54. Ellsworth BS, Butts DL, Camper SA 2008 Mechanisms underlying pituitary hypoplasia and failed cell specification in Lhx3-deficient mice. Developmental biology 313:118-129 55. Sheng HZ, Moriyama K, Yamashita T, Li H, Potter SS, Mahon KA, Westphal H 1997 Multistep control of pituitary organogenesis. Science (New York, NY 278:1809-1812 56. Castinetti F, Saveanu A, Reynaud R, Quentien MH, Buffin A, Brauner R, Kaffel N, Albarel F, Guedj AM, El Kholy M, Amin M, Enjalbert A, Barlier A, Brue T 2008 A novel dysfunctional LHX4 mutation with high phenotypical variability in patients with hypopituitarism. The Journal of clinical endocrinology and metabolism 93:2790-2799 57. Netchine I, Sobrier ML, Krude H, Schnabel D, Maghnie M, Marcos E, Duriez B, Cacheux V, Moers A, Goossens M, Gruters A, Amselem S 2000 Mutations in LHX3 result in a new syndrome revealed by combined pituitary hormone deficiency. Nature genetics 25:182-186 58. Yoshida S, Kato T, Susa T, Cai LY, Nakayama M, Kato Y 2009 PROP1 coexists with SOX2 and induces PIT1-commitment cells. Biochemical and biophysical research communications 385:11-15 59. Reynaud R, Barlier A, Vallette-Kasic S, Saveanu A, Guillet MP, Simonin G, Enjalbert A, Valensi P, Brue T 2005 An uncommon phenotype with familial central hypogonadism caused by a novel PROP1 gene mutant truncated in the transactivation domain. The Journal of clinical endocrinology and metabolism 90:4880-4887 60. Reynaud R, Chadli-Chaieb M, Vallette-Kasic S, Barlier A, Sarles J, Pellegrini-Bouiller I, Enjalbert A, Chaieb L, Brue T 2004 A familial form of congenital hypopituitarism due to a !"# # PROP1 mutation in a large kindred: phenotypic and in vitro functional studies. The Journal of clinical endocrinology and metabolism 89:5779-5786 61. Kuwahara R, Kofman AV, Landis CS, Swenson ES, Barendswaard E, Theise ND 2008 The hepatic stem cell niche: identification by label-retaining cell assay. Hepatology (Baltimore, Md 47:1994-2002 62. Xu Q, Yuan X, Tunici P, Liu G, Fan X, Xu M, Hu J, Hwang JY, Farkas DL, Black KL, Yu JS 2009 Isolation of tumour stem-like cells from benign tumours. British journal of cancer 101:303-311 63. Vierimaa O, Georgitsi M, Lehtonen R, Vahteristo P, Kokko A, Raitila A, Tuppurainen K, Ebeling TM, Salmela PI, Paschke R, Gundogdu S, De Menis E, Makinen MJ, Launonen V, Karhu A, Aaltonen LA 2006 Pituitary adenoma predisposition caused by germline mutations in the AIP gene. Science (New York, NY 312:1228-1230 64. Barlier A, Vanbellinghen JF, Daly AF, Silvy M, Jaffrain-Rea ML, Trouillas J, Tamagno G, Cazabat L, Bours V, Brue T, Enjalbert A, Beckers A 2007 Mutations in the aryl hydrocarbon receptor interacting protein gene are not highly prevalent among subjects with sporadic pituitary adenomas. The Journal of clinical endocrinology and metabolism 92:1952-1955 65. Daly AF, Vanbellinghen JF, Khoo SK, Jaffrain-Rea ML, Naves LA, Guitelman MA, Murat A, Emy P, Gimenez-Roqueplo AP, Tamburrano G, Raverot G, Barlier A, De Herder W, Penfornis A, Ciccarelli E, Estour B, Lecomte P, Gatta B, Chabre O, Sabate MI, Bertagna X, Garcia Basavilbaso N, Stalldecker G, Colao A, Ferolla P, Wemeau JL, Caron P, Sadoul JL, Oneto A, Archambeaud F, Calender A, Sinilnikova O, Montanana CF, Cavagnini F, Hana V, Solano A, Delettieres D, Luccio-Camelo DC, Basso A, Rohmer V, Brue T, Bours V, Teh BT, Beckers A 2007 Aryl hydrocarbon receptor-interacting protein gene mutations in familial isolated pituitary adenomas: analysis in 73 families. The Journal of clinical endocrinology and metabolism 92:1891-1896 66. Brinkmeier ML, Davis SW, Carninci P, MacDonald JW, Kawai J, Ghosh D, Hayashizaki Y, Lyons RH, Camper SA 2009 Discovery of transcriptional regulators and signaling pathways in the developing pituitary gland by bioinformatic and genomic approaches. Genomics 93:449-460 67. Davis SW, Potok MA, Brinkmeier ML, Carninci P, Lyons RH, MacDonald JW, Fleming MT, Mortensen AH, Egashira N, Ghosh D, Steel KP, Osamura RY, Hayashizaki Y, Camper SA 2009 Genetics, gene expression and bioinformatics of the pituitary gland. Hormone research 71 Suppl 2:101-115 68. Castinetti F, Regis J, Dufour H, Brue T Role of stereotactic radiosurgery in the management of pituitary adenomas. Nat Rev Endocrinol 69. Castinetti F, Morange I, Dubois N, Albarel F, Conte-Devolx B, Dufour H, Brue T 2009 Does first-line surgery still have its place in the treatment of acromegaly? Ann Endocrinol (Paris) 70:107-112 70. Davis SW, Castinetti F, Carvalho LR, Ellsworth BS, Potok MA, Lyons RH, Brinkmeier ML, Raetzman LT, Carninci P, Mortensen AH, Hayashizaki Y, Arnhold IJ, Mendonca BB, Brue T, Camper SA 2009 Molecular mechanisms of pituitary organogenesis: In search of novel regulatory genes. Molecular and cellular endocrinology 71. Molitch ME, Clemmons DR, Malozowski S, Merriam GR, Shalet SM, Vance ML, Stephens PA 2006 Evaluation and treatment of adult growth hormone deficiency: an Endocrine Society Clinical Practice Guideline. The Journal of clinical endocrinology and metabolism 91:1621-1634 72. Trounson A 2006 The production and directed differentiation of human embryonic stem cells. Endocrine reviews 27:208-219 73. U HS, Alilain W, Saljooque F 2002 Fetal brain progenitor cells transdifferentiate to fates outside the nervous system. Molecular endocrinology (Baltimore, Md 16:2645-2656 74. Smith AG 2001 Embryo-derived stem cells: of mice and men. Annual review of cell and developmental biology 17:435-462 !!" " 75. Zhao X, Teng R, Asanuma K, Okouchi Y, Johkura K, Ogiwara N, Sasaki K 2005 Differentiation of mouse embryonic stem cells into gonadotrope-like cells in vitro. Journal of the Society for Gynecologic Investigation 12:257-262 76. Simpson EM, Linder CC, Sargent EE, Davisson MT, Mobraaten LE, Sharp JJ 1997 Genetic variation among 129 substrains and its importance for targeted mutagenesis in mice. Nature genetics 16:19-27 77. Wagner J, Lepore D, Thomas P 2007 Differentiation of mouse embryonic stem cells into growth hormone and prolactin expressing cells in vitro. Molecular and cellular endocrinology 273:68-74 78. Lumelsky N, Blondel O, Laeng P, Velasco I, Ravin R, McKay R 2001 Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science (New York, NY 292:1389-1394 79. Rajagopal J, Anderson WJ, Kume S, Martinez OI, Melton DA 2003 Insulin staining of ES cell progeny from insulin uptake. Science (New York, NY 299:363 80. Collombat P, Xu X, Ravassard P, Sosa-Pineda B, Dussaud S, Billestrup N, Madsen OD, Serup P, Heimberg H, Mansouri A 2009 The ectopic expression of Pax4 in the mouse pancreas converts progenitor cells into alpha and subsequently beta cells. Cell 138:449-462 81. Bonnefont X, Lacampagne A, Sanchez-Hormigo A, Fino E, Creff A, Mathieu MN, Smallwood S, Carmignac D, Fontanaud P, Travo P, Alonso G, Courtois-Coutry N, Pincus SM, Robinson IC, Mollard P 2005 Revealing the large-scale network organization of growth hormone-secreting cells. Proceedings of the National Academy of Sciences of the United States of America 102:16880-16885 82. Hombach-Klonisch S, Panigrahi S, Rashedi I, Seifert A, Alberti E, Pocar P, Kurpisz M, Schulze-Osthoff K, Mackiewicz A, Los M 2008 Adult stem cells and their trans-differentiation potential--perspectives and therapeutic applications. Journal of molecular medicine (Berlin, Germany) 86:1301-1314 83. Lepore DA, Thomas GP, Knight KR, Hussey AJ, Callahan T, Wagner J, Morrison WA, Thomas PQ 2007 Survival and differentiation of pituitary colony-forming cells in vivo. Stem cells (Dayton, Ohio) 25:1730-1736 84. Gratwohl A, Baldomero H 2009 Trends of hematopoietic stem cell transplantation in the third millennium. Curr Opin Hematol 16:420-426 85. Snowden JA, Martin-Rendon E, Watt SM 2009 Clinical stem cell therapies for severe autoimmune diseases. Transfus Med 19:223-234 86. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S 2007 Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131:861- 872 87. Hanna J, Wernig M, Markoulaki S, Sun CW, Meissner A, Cassady JP, Beard C, Brambrink T, Wu LC, Townes TM, Jaenisch R 2007 Treatment of sickle cell anemia mouse model with iPS cells generated from autologous skin. Science (New York, NY 318:1920-1923 !"# # Gene name Protein encoded Main characteristic Transcription factors Sox2 SRY related HMG box transcription factor Expressed in pituitary marginal zone, stem cell marker in several tissues Sox9 SRY related HMG box transcription factor Might play a role in the transition between pituitary stem cells/progenitors and transit amplifying cells Prop1 Paired Homeodomain transcription factor Expressed in pituitary marginal zone prior to emergence of most differentiated cell types Tpit (Tbx19) Pituitary T box transcription factor Promotes corticotroph differentiation Oct4 (Pou5f1) POU Homeodomain transcription factor Expressed in several tissue stem cells Pou1f1 (Pit1) POU Homeodomain transcription factor Involved in differentiation of somato-lactotroph and thyrotroph lineages Nanog Homeobox Transcription factor Involved in maintaining stem cell pluripotency Hes1 Basic Helix loop helix transcription factor Notch downstream target, repressor of cell cycle inhibitors. Expressed in S/G2/M/G1, not in G0 Cell cycle regulators Bmi-1 ! BMI1 polycomb ring finger oncogene Regulates cell cycle inhibitor genes Cyclin D1 G1/S specific cyclinD1 Involved in G1/S cell cycle transition Cyclin D2 G1/S specific cyclinD2 Involved in G1/S cell cycle transition Cyclin E Cyclin E Allows progression to S phase Cdk4 Cyclin dependent kinase 4 Involved in cell cycle G1 phase progression Ki67 Antigen KI-67 Nuclear protein associated with cell proliferation. Marks all steps except G0 Intermediate filament proteins Nestin Type VI intermediate filament protein Expressed in pituitary marginal zone Gfap Glial fibrillary acidic protein, (intermediate Expressed in folliculo-stellate cells filament protein) Cytokeratin 8 Keratin containing intermediate filament Expressed in pituitary marginal zone in protein adulthood Adhesion and cell surface proteins, receptors, and others E-cadherin Calcium dependent adhesion molecule Might be involved in transition from (type 1 transmembrane protein) proliferation to differentiation EpCAM Epithelial cell adhesion molecule Pan-epithelial differentiation antigen expressed (CD326) in carcinomas CD90 Thy1 or CD90 (cell surface protein) Marker of a variety of stem cells GFRa2 Glial cell-line derived neurotrophic factor Stem cell marker in testis and ovary (GDNF) receptor alpha 2 Expressed in pituitary marginal zone S100ß S100 calcium binding protein B Expressed in folliculo-stellate cells Sca1(Ataxin1) Stem cell antigen 1 Expressed in the side population cells and pericytes c-Kit (CD117) Cytokine receptor CD117 Expressed in hematopoietic stem cells Notch1 Notch 1 (transmembrane receptor) Involved in progenitor differentiation in CNS Rb retinoblastoma Tumor suppressor gene. Haploinsufficiency increases risk of retinoblastoma in humans and intermediate lobe adenomas in mice. AIP Aryl hydrocarbon receptor interacting Involved in familial pituitary adenomas protein