Progress in molecular biology and translational science, volume 143

Table 1 Mouse Models That Have Contributed to Our Knowledge of Gonadotrope Differentiation.Signaling Pathways Fgf8 Hypomorphic allele, Neo insertion, global deletion, Fgf8Neo/ Variable,

P ROGRESS IN

MOLECULAR BIOLOGY AND TRANSLATIONAL

SCIENCE

Gonadotropins: From Bench Side

to Bedside

VOLUME ONE HUNDRED AND FORTY THREE

P ROGRESS IN

MOLECULAR BIOLOGY AND TRANSLATIONAL

CO, United States

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO

SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

Academic Press is an imprint of Elsevier

525 B Street, Suite 1800, San Diego, CA 92101-4495, United States

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom First edition 2016

Copyright © 2016 Elsevier Inc All Rights Reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing As new research and experience broaden our understanding, changes in research methods, profes- sional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein In using such information or methods they should

be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors,

or editors, assume any liability for any injury and/or damage to persons or property

as a matter of products liability, negligence or otherwise, or from any use or operation

of any methods, products, instructions, or ideas contained in the material herein ISBN: 978-0-12-801058-7

ISSN: 1877-1173

For information on all Academic Press publications

visit our website at https://www.elsevier.com/

Publisher: Zoe Kruze

Acquisition Editor: Alex White

Editorial Project Manager: Helene Kabes

Production Project Manager: Magesh Kumar Mahalingam

Designer: Maria Ines Cruz

Typeset by Thomson Digital

S.L Asa

Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto,

ON, Canada; Department of Pathology, Laboratory Medicine Program, University Health Network, Toronto, ON, Canada

L Casarini

Unit of Endocrinology, Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Modena, Italy; Center for Genomic Research, University of Modena and Reggio Emilia, Modena, Italy

B.S Ellsworth

Department of Physiology, School of Medicine, Southern Illinois University, Carbondale,

IL, United States

S Ezzat

Department of Medicine, University of Toronto, Endocrine Oncology, Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada

J Kapali

Department of Physiology, School of Medicine, Southern Illinois University, Carbondale,

IL, United States

D Santi

Unit of Endocrinology, Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Modena, Italy; Center for Genomic Research, University of Modena and Reggio Emilia, Modena, Italy; Azienda USL of Modena, Modena, Italy

M Simoni

Unit of Endocrinology, Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Modena, Italy; Center for Genomic Research, University of Modena and Reggio Emilia, Modena, Italy; Azienda USL of Modena, Modena, Italy

C.E Stallings

Department of Physiology, School of Medicine, Southern Illinois University, Carbondale,

IL, United States

PREFACE

Knowing is not enough; we must apply.Willing is not enough; we must do

GoetheBasic and clinical research on pituitary gonadotropins started nearly

100 years ago In the beginning, hypophysectomy, a surgical feat revealedthe importance of pituitary hormones in many physiological systems includ-ing reproduction Later, most of the focus was on whether two gonadotro-pins existed, shared a common alpha subunit that was linked to thehormone-specific beta subunit Having realized that they did indeed existand were heterodimers, the next goal was to develop specific and sensitivebioassays and immunoassays to measure them in circulation and pituitaryextracts, and localize them within gonadotropes under a variety of physio-logical conditions Along the way came the immunoneutralizationapproaches, which identified the specific need for LH and FSH in gonadalfunction The localization of cell-surface receptors on gonads and theirpurification from gonadal cell membranes provided new insights into gonad-otropin action Subsequent structure-function studies laid the foundationfor future three-dimensional modeling research The above mentioned basicscience discoveries slowly began to impact clinical research Clinicians begantesting the human urinary gonadotropins, albeit not entirely pure, onpatients

The advent of molecular biology and cloning of the subunit-encodinggenes heralded a new era in gonadotropin gene regulation, and led to theproduction of pure, safe, and efficacious recombinant gonadotropic hor-mones for clinical use Then came the major breakthrough It was possible

to achieve gene manipulation and understanding the genetics and physiology

of gonadotropins at the whole organism level This led to modeling humanreproductive diseases (infertility and pituitary and gonadal tumors), in mice,and integrating the human patient data on polymorphisms and mutations ingonadotropins/their cognate receptors These developments resulted in bet-ter diagnosis and designing treatment options for gonadotropin-dependent

fertility disorders Two major surprises came recently The discovery of related FSH glycoforms and extragonadal FSH receptors We must furtherexplore the functional significance of these two controversial observations,because they have tremendous clinical significance, particularly, in ARTprotocols and menopause research Unraveling the mysteries surroundingthese two novel issues, may be a future goal in many research laboratories.Volume 143 of the Progress in Molecular Biology and Translational Sciences(PMBTS) is devoted to Gonadotropins: From Bench side to the Bedside Expertsfrom all over the world have contributed chapters on Mouse Models forGonadotrope Development (Chapter 1), Mouse Models for the Study ofSynthesis, Secretion and Actions of Pituitary Gonadotropins (Chapter 2),Clinical Applications of Gonadotropins in the Female (Chapter 3),Clinical Applications of Gonadotropins in the Male (Chapter 4), BeyondReproduction: Pituitary Hormone Actions on Bone (Chapter 5), andGonadotrope Tumors (Chapter 6) I thank all the contributing authors for

age-an excellent job of amalgamating the up to date knowledge on age-animalmodels and human conditions related to gonadotropins These Chaptersclearly illustrate how the bench side research work could benefit patients

at the clinic Undoubtedly, much remains to be done at both the frontiers—bench side and bedside on gonadotropin research Certainly, there is a needand scope to further updating, including additional chapters, and bringing anew expanded volume in the future

I thank Professor P Michael Conn, PMBTS Series Editor, for inviting

me to edit this state-of-the-art volume on gonadotropins His constantsupport and genuine encouragement are truly inspiring Finally, I owe mysincere thanks to Ms Helene Kabes and her Production Team members

at the Elsevier Press, for their patience, and rendering marvelous guidanceand support throughout the journey To all the Readers—enjoy the PMBTS,Volume 143, Gonadotropins: From Bench side to Bedside

T RAJENDRAKUMAR, PhD

Editor

CHAPTER ONE

Development

C.E.Stallings,J.Kapali,B.S.Ellsworth1

Department of Physiology, School of Medicine, Southern Illinois University, Carbondale, IL, United States

1 Corresponding author E-mail address: bells@siu.edu

Contents

3.13 Additional Genes Known to Contribute to Gonadotrope Development 28

Abstract

The pituitary gonadotrope is central to reproductive function Gonadotropes develop

in a systematic process dependent on signaling factors secreted from surrounding

Progress in Molecular Biology andTranslational Science, Volume 143

1

tissues and those produced within the pituitary gland itself These signaling pathways are important for stimulating specific transcription factors that ultimately regulate the expression of genes and define gonadotrope identity Proper gonadotrope develop- ment and ultimately gonadotrope function are essential for normal sexual maturation and fertility Understanding the mechanisms governing differentiation programs of gonadotropes is important to improve treatment and molecular diagnoses for patients with gonadotrope abnormalities Much of what is known about gonadotrope development has been elucidated from mouse models in which important factors contributing to gonadotrope development and function have been deleted, ectop- ically expressed, or modified This chapter will focus on many of these mouse models and their contribution to our current understanding of gonadotrope development.

CentraltoreproductivefunctionisthehypothalamicpituitarygonadalaxisinwhichhypothalamicGnRHactivatesspecificreceptorsonthesurfaceofpituitarygonadotropes.ActivationofGnRHsignalingstimulatesexpressionofthegonadotropinsubunitsandtheGnRHreceptor,Gnrhr.1–16Thepituitarygonadotropinsaredimericglycoproteinhormoneswithacommonα-subunit(Cga)anduniqueβ-subunits(LhbandFshb)thatgivethemtheiruniquefunc-tions.17Gonadotropins are essential for gonadal function inboth malesandfemales.18Thus,thepituitarygonadotropeisvitalforreproductivefunction.Theanteriorlobeofthepituitarygland,togetherwiththeintermediatelobe, is derived from a structure referred to as Rathke’s pouch Rathke’spouchoriginatesfromoral ectoderm whilethe posterior lobe formsfromneuralectoderm.DuringgestationmostproliferatingcellsofRathke’spouchborder the luminal area These cells then cease proliferating and migrateventrallyviaanEMT-liketransitiontoexpandtheanteriorlobe.Theante-riorlobehasveryfewproliferatingcellsrelativetotheperiluminalarea.19–22

In vivo data suggest that pituitary cell specification occurs betweenembryonic day (e)10.5 and e12.5, while most pituitary cell types do notbeginterminaldifferentiationuntilapproximatelye15.5.23Davisetal.usedbirth-datingstudiestoshowthatallanteriorlobecelltypesexitthecellcycleandbegin thedifferentiation processbetweene11.5ande13.5,suggestingthatspecialized celltypes arenot groupedtogether basedon birthdate.24

At birth, the pituitary cell types are roughly organized into layers withgonadotropesbeingthemostventral Byadulthoodspatialorganization ofthecell types appears more random, although recent studies demonstratethatthecelltypesformnetworksthatareattachedbyadherensjunctions.25

Thelayering ofpituitarycell typesat birthmaybe dueto cellmovementrequiredtoestablishnetworksofspecificcelltypes,ratherthanarelationshipwiththetimingofcellcycleexit.26

Pituitarycelltypesexpresstheirsignaturehormonesinadistincttemporalpattern.Hormoneexpressionisdependent,inpart,onregulationbyspecifictranscriptionfactors.Theforkheadtranscriptionfactor,Foxl2,iscoexpressedwithCga,thefirsthormone-encodingtranscripttobedetectedinitiatingatapproximatelye10.5.CGAprotein ispresentbye11.5.27,28 Thefirstgona-dotrope-specific markers are Nr5a1and Gnrhr at approximately e13.5.29,30Birth-datingstudiessuggestthatgonadotropes,whichoccupyamorerostrallocationduringdevelopmentthanotheranteriorlobecelltypes,exitthecellcycleandarespecifiedinhighestnumbersate11.5.24Althoughgonadotropespecification occurs early in the pituitary development, the gonadotropesterminally differentiatelate indevelopment withLhbtranscriptsdetectable

byapproximatelye16.5andFshbshortlythereafter.31Gonadotropesaretheleast abundant of six hormone-producing cell types (gonadotropes, thyro-tropes, somatotropes,lactotropes, corticotropes,and folliculostellatecells) intheanteriorpituitaryglandrepresenting5–10%oftheanteriorpituitarycells.17Thereisincreasingevidencethatgonadotropesdevelopandpersistasaheterogeneouspopulation.Colabelingstudiesdemonstratethepresenceoftwodistinctgonadotropesubtypesatthebeginningofgonadotropediffer-entiation:(1)LHB/GnRHR-positivecellsand(2)FSHB/TSHB-positive,GnRHR-negativecells.TheFSHB/TSHB-positivecellsarethoughttobethe precursors of gonadotropes and thyrotropes The FSHB-positivegonadotropes begin to express Gnrhr bye18.75.55 By postnatal day (P)7,three distinctpopulations of gonadotropesexist: FSH-onlygonadotropes,LH-onlygonadotropes,andbihormonalgonadotropeswithbothFSHand

LH (Fig 1) While nearly all LHB-positive gonadotropes also containNR5A1, only some FSHB-positive gonadotropes contain NR5A1.33Mucheffort hasgoneintounderstandinghowundifferentiatedprogenitorcellsbecomefullyfunctionaldifferentiatedgonadotropes.Inthischapterwewilldiscussmanyofthemousemodelsthathavecontributedtoourunder-standingofgonadotropedevelopment(Table1)

mesenchymeregulatepituitarygrowthandmorphogenesisandappeartobeimportant for promoting differentiation of several pituitary cell types,includinggonadotropes.Sonichedgehog (SHH)andbonemorphogeneticprotein(BMP)2aresecretedfromtheventraljuxta–pituitarymesenchymeanddiffuseintothesurroundingtissue,includingRathke’spouch.Fibroblastgrowth factor (FGF)8 and BMP4 are secreted from the infundibulumcreatingagradientofsignalingfactorsinthedevelopingpituitarygland.96

Fgf8is first detected in the infundibulum at e10.5.96 Studies of pituitaryexplantsshowthatFGF8isimportantforthemaintenanceofLhx3expres-sionandrepressionoflsl1expressioninthedorsalaspectofRathke’spouch,suggestingthatFGF8signalingfromtheinfundibulumisrequiredtoestablishproper patterning of LIM homeobox gene expression during pituitary

Embryonic day 10

Uncommitted progenitors

divide rapidly.

Cells committed to the gonadotrope lineage express Nr5a1.

By e16.75, two populations of differentiating gonadotropes appear:

1) Express Lhb,

2) Express Fshb and Tshb.

Ultimately, three populations of mature gonadotropes exist:

Terminal differentiation begins around e16.5 with the expression of Lhb, dark blue cells Studies by Wen et al show that by e16.75 an occasional cell is positive for FSHB and TSHB, but not LHB (yellow cells).55Ultimately, gonadotropes exist as a heterogeneous population with FSH-only gonadotropes (green), LH-only gonadotropes (dark blue), and bihormonal gonadotropes (orange).33P7, Postnatal day 7.

Table 1 Mouse Models That Have Contributed to Our Knowledge of Gonadotrope Differentiation.

Signaling Pathways

Fgf8 Hypomorphic allele, Neo

insertion, global deletion, Fgf8Neo/

Variable, loss of anterior lobe to normal morphology with loss of LH

[34]

Fgf8 Hypomorphic allele, Neo

insertion, Fgf8neo/neo

Lack GnRH neurons at birth [35,36]

Bmp4 Transgenic, Cga-Bmp4 Absence of Cga, Tshb, Gh, Prl at e17.0, dorsal expansion

of expression domain for Gata2, Isl1, and Msx1 at e17.0

[28]

Bmpr1a Conditional deletion, Bmpr1a£ox/
;

Cga-cre

Hypoplastic Rathke’s pouch, loss of Isl1 expression [37,38]

Bmpr1a Conditional deletion, Bmpr1a£ox/
;

Gnrhr+/GRIC

Fertile with normal gonadotropin levels in adulthood [32,39]

Notch2 Transgenic, Cga-Notch2 Delayed gonadotrope differentiation [40]

Notch2 Transgenic, Pou1f1-Notch2 Normal Lhb expression at e17.5 [41]

Notch2 Conditional deletion, Notch2£ox/

Hes1 Transgenic, Cga-Hes1 Fewer gonadotropes and thyrotropes at birth [45]

Shh Global deletion, Shh
/
Pituitary gland is absent [46]

Shh Transgenic, Cga-Shh Increased BMP2, expanded thyrotropes and gonadotropes [46]

Hip Transgenic, Pitx1-Hip Inhibition of BMP signaling, severe reduction in size of

Genes Description Gonadotrope Phenotype References

Gli2 Global deletion of Zn-finger

Reduced LH and subfertility in males, females are fertile [32,49,50]

Ctnnb1 Conditional constitutive

activation, Ctnnb1£ox(ex3)/+; Pitx1-cre or Pou1f1-cre

Normal gonadotrope development [51]

Gnrh Spontaneous, Gnrhhpg /hpg Gonadotropes are present, but do not express Lhb or

Fshb in the absence of exogenous GnRH

Pitx1 Global deletion, Pitx1
/
Reduced gonadotrope population [56]

Pitx1, Pitx2 Global deletion, Pitx1+/

;Pitx2+/

Reduced gonadotrope population [57]

Pitx2 Transgenic, Cga-Pitx2 Expansion of gonadotrope population [58]

Pitx2 Hypomorphic allele, Pitx2neo/neo Absence of gonadotropes [59]

Pitx2 Conditional deletion, Pitx2£ox^;

Lhb-cre

Normal gonadotrope function [60]

Lhx3 Global deletion, Lhx3
/
Reduced NR5A1 in the ventral pituitary and ectopic

NR5A1 in the dorsal pituitary, dorsal pregonadotropes

Isl1 Global deletion, Isl1
/
Hypoplastic, undifferentiated Rathke’s pouch, embryonic

Gata2 Transgenic, Gh-Gata2 Normal gonadotrope differentiation [23]

Gata2 Transgenic, Cga-dnGata2 Inhibition of terminal differentiation of gonadotropes

Prop1 Spontaneous, Prop1df/df Reduced gonadotrope population, persistent Hesx1

expression, absence of Notch2 expression

[65,69]

Prop1 Global deletion, Prop1
/
Reduced gonadotrope population, persistent Hesx1

expression, absence of Notch2 expression

[69]

Prop1 Transgenic, Cga-Prop1 Delayed gonadotrope differentiation resulting in delayed

puberty and hypogonadism, normal NR0B1, NR5A1, Gata2 and Egr1

[70]

Prop1 Prop1df/df;Lhx3
/
Severe reduction of anterior lobe, expansion of NR5A1

domain

[65]

Hesx1 Global deletion, Hesx1
/
Variable, hypoplastic anterior lobe, animals with mildest

phenotype were viable and fertile

Genes Description Gonadotrope Phenotype References

Hesx1 Compound heterozygote, Six+/
;

Hesx1+/cre

Delayed gonadotrope differentiation, but normal specification

[72]

Otx1 Global deletion, Otx
/
Serum LH and FSH reduced 70–80% in adults,

expression of Gnrhr and Gnrh is normal

Pax6 Spontaneous, Pax6Sey/Sey Dorsally expanded domains of Cga, Gata2, Prop1 and Isl1

at e12.5 and Nr5a1, Gata2 and Foxl2 at e15.5, at birth expression of Cga is increased, Lhb is downregulated possibly due to lack of GnRH stimulation

[76,77]

Pax6 Global deletion, Pax6
/
Similar phenotype as Pax6Sey/Sey [76,78]

Pax6 Deletion of transactivation

domain, Pax6Neu/Neu

Similar phenotype as Pax6Sey/Sey [76,79,80]

Egr1 Global deletion, Egr1
/
Gonadotrope lineage is specified normally, but fail to

produce LHB, FSHB is produced normally

Tbx19 Transgenic, Cga-Tbx19 Severe reduction in the gonadotrope population [84]

Nr5a1 Global deletion, Nr5a1
/
Gonadotropes are specified normally, express Lhb and

Nr0b1 (Dax1) Global deletion, Nr0b1£ox/£ox;

CMV-cre

Normal number of LHB- and FSHB-positive cells [87]

Foxl2 Transgenic, Cga-Foxl2 Stimulates ectopic expression of Cga [27]

Foxl2 Global deletion, Foxl2
/
Reduction in Fshb, Cga, Gnrhr, and Fst, normal levels

of Lhb, normal gonadotrope differentiation, increased pituitary cell density

Reduced population of gonadotropes at e17.5 [51,93]

Dicer Conditional deletion, Dicer£ox/

development.96 Human mutations in FGF8 cause holoprosencephaly,diabetesinsipidus,hypopituitarism, and,insomecases,hypogonadotropichypogonadism.35Kallmannsyndrome,characterizedbyanabsenceoffunc-tionalGnRH neurons,can becaused bymutationsinFGF8andFGFR1,the main receptor for mediating FGF8 signaling.35,97,98 McCabe et al.employed a mouse model containing one hypomorphic allele of Fgf8(Fgf8neo)andonenullallele(Fgf8
)todelineatetheroleofFGF8inregu-latingpituitarydevelopment.34,36Fgf8neo/
mice haveavariablephenotypewith approximately one-third exhibiting a severe reduction of anteriorpituitaryand an absence of posterior pituitary at e17.5.34 Approximatelytwo-thirdsofFgf8neo/
mice haveamilder phenotypewitha morphologi-callynormalpituitary,butareductionorabsenceofLhbexpression.Otherhormone-producingcelltypeswere present.34Fgf8neo/
mice dieimmedi-atelyafterbirth,thusitisdifficulttodeterminewhethertheabsenceofLhbexpressionisduetolossofGnRHstimulation,orwhethergonadotropesfail

todifferentiate

Micehomozygousfor thehypomorphicallele(Fgf8neo/neo)lackGnRHneuronsatbirth.35,36Thesemicehavea55%reductioninfunctionalFGF8proteinlevelsanddiewithin1dayofbirth.36Micethatareheterozygousforthehypomorphic allele(Fgf8+/neo) haveapproximately half thenumberofGnRHneuronsastheirwildtypelittermatesat120daysofage,althoughLHand GnRH peptide levels are normal Females have delayed puberty andmaleshave normal testicular development.99 Establishing a mouse modelwithpituitary-specificdeletionofFGFR1wouldgreatlyaidourunderstand-ingofthespecificroleofFGF8inrelationtogonadotropedevelopmentanddifferentiation

BMP signaling is critical for pituitary organogenesis.28,37,96 Bmp4 isexpressedintheembryonicinfundibulumandisrequiredfortheinduction

ofRathke’spouch.67,96Bmp2is expressedinthemesenchymeadjacenttoRathke’spouchate10.5–e12.5andthroughoutRathke’spouchate12.5.37Overexpression of Bmp in presumptive gonadotrope precursors results indorsal expansion of the expression domain of the transcription factors,Gata2,Isl1, and Msx1,at e17.0 suggesting thatexpression ofthese factors

in presumptive gonadotrope precursors is induced by BMP signaling.23Expression of Cga,Tshb, Gh, and Prl are almost entirely absent in Cga-Bmp4miceate17.0.28

byactivatingthetypeIreceptor,BMPR1A.100,101Davisetal.conditionallydeletedBmpr1ainallpituitarycelltypesusingCga-cre.37,38,102,103Ate10.5,Rathke’s pouch is thin and underdeveloped in Bmpr1a£ox/
;Cga-creembryos.37 PITX1 and LHX3 are present inRathke’s pouch, suggestingthat pituitaryorganogenesis has initiated However, ISL1is absent Thus,induction of Isl1 expression during pituitary organogenesis is dependentupon signaling through BMPR1A.37 Bmpr1a£ox/
;Cga-cre exhibit earlyembryonic lethality at e12.5, likely due to cre expression and deletion ofBmpr1aincardiacmusclecausingheartdefects.37

Surprisingly, deletion of Bmpr1a specifically in gonadotropes doesnot affect gonadotrope development or function.32,39,102,103 Bmpr1a£ox/
;GnrhrGRIC/+micearefertilewithnormalserumgonadotropinsandnormalgonadotropintranscriptlevels.39Theembryonicphenotype ofthesemicewas notexamined Theapparent inconsistency betweenthesetwomousemodels is likely due to the fact that Cga-cre causes recombination in thepituitaryprimordiumandallcelltypesveryearlyindevelopmentwithinitialexpressionate9.5,whereasGnrhrGRIC/+causesstimulationonlyingonado-tropesandrecombinationoccurslaterate12.75.32ThesestudiessuggestthatBMP signaling is notrequired for gonadotrope function However, addi-tional studies are required to determine if BMP signaling is importantforveryearlystagesofgonadotropedevelopment

The Notch signaling pathway is key for the development of numeroustissues and has been implicated in several human developmental disor-ders.104TheNotchreceptorgene,Notch2,isexpressedinRathke’spouchfrome12.5-e14.5intheproliferatingcellssurroundingthelumen,butnot

in differentiated cells expressing Cga.105 As pituitary development ceeds, Notch2 expression decreases, consistent with a role in preventingcell differentiation Notch2expressionis absentin micelackingthepitui-tary-specifictranscriptionfactor,Prop1.105OverexpressionoftheNOTCHtarget, Hes1, inhibits gonadotrope and thyrotrope differentiation.45Consistent with these findingsoverexpression of theconstitutivelyactiveNotch2 intracellular domain driven by the Cga promoter (Cga-Notch2)delaysgonadotropedifferentiationwithalmostnofullydifferentiatedgona-dotropes at birth Differentiated gonadotropes do eventually develop,although these cells do not express the Notch2 transgene.105 In a similar

studytheNotch2intracellular domain wasoverexpressed undercontrol ofthePou1f1promoter.InthisscenarionoeffectwasobservedonLhbexpres-sionate17.5.41ThedifferenceinthesestudiesmaybeduetothefactthatCgastimulatesexpressionapproximately3daysearlierandinamoreventraldomain thanPou1f1.Deletion ofNotch2specifically inthe pituitarygland(Notch2£/£;Foxg1+/cre)increasesNr5a1andLhbexpressionatbirth,althoughthe number of gonadotropes is not statistically different from wild typelittermate controls.42–44 Similarly, conditional deletion of the NOTCHmediator, Rbpj, does not affect gonadotrope commitment, althoughcorticotropes differentiate prematurely.41 Together these data suggestthat persistent NOTCH signaling inhibits gonadotrope differentiation,and although gonadotropin expression may be increased with loss ofNOTCHsignaling,gonadotropedevelopmentisunaffected.

Themorphogen,Shh,isexpressedintheventraldiencephalonandtheoralectoderm, but its expression is extinguished from Rathke’s pouch as itbeginsto form.46 Treier etal.studied theroleof SHHin pituitaryglanddifferentiation.Global deletion of Shh (Shh
/
) results in micein whichthepituitaryglandisnotdetectable.Thismaybeduetodirector indirecteffectsofSHH action.46

Usingatransgenicapproach,9kbofregulatoryupstreamsequencefromthePitx1genewaslinkedtotheDNAsequenceencodingHip(Pitx1-Hip).HIPisaninhibitorofallmammalianhedgehogfamilymembers.106Ate11.5Rathke’s pouch is hypoplastic and dysmorphic in Pitx1-Hip transgenicsalthough Lhx3 expression can be readily detected By e13.0, the arrest

of pituitary development is even more apparent Pitx1-Hip mice display

apituitaryphenotypesimilartothatofLhx3
/
miceleadingtospeculationthat Lhx3 expression may be regulated by SHH in addition to FGFsignaling.46

Complementarytoloss-of-functionstudies,Treieretal.generatedmice

inwhichexpressionofthefull-lengthcodingregionofShhareregulatedbytheCgapromoter,targetingexpressiontoRathke’spouch.Cga-Shhtrans-genic mouse embryos exhibit significant pituitary hyperplasia at e17.5.ExpressionofLhx3isslightlyincreasedintheventralaspectofthepituitarygland, again suggesting that Lhx3 expression is stimulated by SHH TheincreaseinpituitaryglandsizeisduetoanincreaseinthenumberofCga-expressing cells, consisting of both thyrotropes and gonadotropes, with

Ctnnb1 (previouslyreferred to as Catnb) isa geneinvolved in thecanonicalWnt signaling proven to be active inpituitary development.51 The proteinproductofCtnnb1,CTNNB1(alsoknownascateninbeta-1orβ-catenin),isnormallysequesteredinthecytoplasmbyadestructioncomplex,whichpro-motesdegradationofthemolecule.57When variousWNTligandsbindtotheircorrespondingreceptors,thepathwayisstimulatedandCTNNB1translocatesinto the nucleus to induce transcription Chromatin immunoprecipitationstudiesperformedonwildtypemousepituitaryglandrevealCTNNB1asso-ciatedwithLef/Tcf-bindingregionsonthe Axin2promoterate12.5,indicating

a role in early pituitary gland development.51A numberof mouse modelswith mutations in Ctnnb1 expression havebeen developed and investigatedforpituitaryeffects,primarilyCtnnb1£ox(ex2^6)/£ox(ex2^6)andCtnnb1£ox(ex3)/+

Analleleforconditional deletionofCtnnb1 (Ctnnb1£ox(ex2^6)/£ox(ex2^6)),wasgeneratedbyflankingCtnnb1exons2–6withloxPsites.48ResearchershaveusedthismodelcrossedwithGnrhr+/GRICmiceto inducedisruptionlocalized to gonadotropes and some neurons Ctnnb1£ox(ex2^6)/£ox(ex2^6);Gnrhr +/GRIC mice have no observable phenotypic effects or changes infertilitydespiteverificationoftherecombinationevents.32,49

A complementary model with a dominant stable mutation wasdeveloped by flanking only exon 3 with loxP sites and is referred to asCtnnb1£ox(ex3)/£ox(ex3)but was originally published as Catnblox(ex3)/lox(ex3).50Upon CRE-mediated recombination the resultant offspring produce anin-frame mutated CTNNB1 truncated to eliminate thephosphorylationsites involved in ubiquitination and subsequent degradation, resulting inconstitutive activation of CTNNB1 signaling.50 Studies of heterozygousCtnnb1£ox(ex3)/+micecrossedwiththeaforementionedGnrhr +/GRICmicerevealed significant effects on male offspring but not females.32,50 Malesanalyzed at 8–12weeks of age were subfertile with a slight decrease incirculatingLHlevelsand reducedLhbmRNAexpressioninthepituitarygland.49Additionally,pituitaryFshbmRNAwassignificantlydecreased.Byusing primaryculturesofpituitarycells fromCtnnb1£ox(ex3)/+;Gnrhr+/GRICmice,researcherswereabletostimulateFshbmRNAproductionbytreat-mentwithactivin.Furtherinvestigation revealedthatprimaryculturesofpituitarycellsfromCtnnb1£ox(ex3)/+;Gnrhr+/GRICmicearemoresensitiveto

inhibinstimulationthanwildtypecontrols.Overall,theauthorsconcludedthat Ctnnb1 may contribute to male-specificFshb mRNA production aspartoftheactivin/inhibin/follistatinsystem.50Whetherthisisduetoearlydevelopmental effects on gonadotrope specification or simply postnatalgonadotropefunction isunknown.

StudiesusingcellcultureandadditionalmousemodelshaveimplicatedCtnnb1asanactivatorofvarioustranscriptionaleventsrelatedtogonadotropedevelopment Pitx2+/
mice injected with knownWnt pathway agonistlithium chloride exhibit increased Pitx2 expression, again connectingCTNNB1 to early pituitary gland formation.57 Cell culture experimentshaveshownCTNNB1bindstoNR5A1asanecessarystructureforfurtherNR5A1/EGR1bindingandrobustLhbpromoteractivation.107Thesetwoprocesses areat opposite temporalends of pituitary development; PITX2regulatesinitial gonadotropedevelopment (first detectedat e8.5) and Lhbexpression precedes terminal gonadotrope differentiation Therefore,CTNNB1maybeamultifunctionalfactor inbothgonadotropespecifica-tionandgonadotropinproduction

Additional research has shown CTNNB1 to bea necessary factor inearly Pou1f1 activation using the Ctnnb1£ox(ex2^6)/£ox(ex2^6) mouse modelmated to Pitx1-cre or Pou1f1-cre; however, surprisingly these mutants didnothave significantalterations ingonadotrope developmentor function.TheCtnnb1£ox(ex2^6)/£ox(ex2^6);Pitx1-cremiceexhibitnoPou1f1expressionintheanteriorlobepaste14.5;however,thisisaneventnotcurrentlylinked

togonadotropes.51Mostrecentlya modelusingWnt1-cretodeleteCtnnb1from the developing embryo demonstrates its crucial role in regulatingsignalingfromthepituitaryorganizertoinfluence“pituitaryglandgrowth,development, and vascularization”.108 While the data from these studiesarenotgonadotrope-specific, theydohighlightupstreamsignaling eventsnecessary for proper pituitary gland formation Overall, mouse modelswith mutations in the Ctnnb1 gene have provided data implicatingCTNNB1 both in early pituitary developmental events upstream ofPitx1andPitx2anddirectlyprecedingterminaldifferentiation.Datafrommurine models show constitutive Ctnnb1 expression is associated withdecreasedfertilityandgonadotropin levelsspecificallyinmalemice

GnRHsecretedfrom parvocellular neurons of thehypothalamus intothepituitaryportalvesselsystemregulatesgonadotropeproductionofgonado-tropinsviabindingtoaseventransmembraneG-proteincoupledreceptor,

at approximately e13.0, coincident with Nr5a1, making themthe earliestgonadotrope-specificgenes.29,113NR5A1stimulatesexpressionofGnrhrvia

acomplexenhancer.114,115ExpressionofGnrhrisadditionallystimulatedbyits own ligand, GnRH, via protein kinase C activation of an activatorprotein-1 (AP-1) element.14–16,116 In vitro studies demonstrate severalHboxsitesthatcontributetobasalandGnRHactivationoftheGnrhrgene.Activationatthesesitesappearstoinvolve,inpart,bindingofthehomeoboxtranscriptionfactor,OCT1.117,118ThepituitaryglandisabletosecreteLH

inresponsetoGnRHasearlyase16.55WhileGnRHisessentialfornormalexpressionofgonadotropinsandGnrhr,itdoesnotappeartoberequiredfordifferentiation ofgonadotropes.Thisconclusionisbasedon studiesofhpgmice,whichdonotproduceGnRHduetoadeletionintheGnrhgene.4,52GnRHtreatmentstimulatesgonadotropinproductioninhpgmicesuggest-ingthat gonadotropedifferentiation occursin theabsenceof endogenousGnRH.52–54 Interestingly, ablation of Gnrhr-expressing cells appears todiminishtheabilityofFSHB/TSHB-positivepregonadotropestodifferen-tiateintomaturegonadotropes.Theauthorssuggestthatthisisduetolossof

LH production by the earlier arising LH-only gonadotropes Therefore,GnRH signaling in late gestation mayhave subtle and indirecteffects ongonadotropesubpopulations.55

Signalingfactorsplayaninfluentialroleinregulatingtheexpressionofvarioustranscriptionfactorsinthedevelopingpituitarygland.96,119Manyofthese transcription factors are important for gonadotrope development.Some transcriptionfactors are necessaryfor theearliest stages ofpituitarydevelopment Loss of these factors blocks not only development of thepituitarygland,butalsoterminaldifferentiationofoneor morehormone-producing cell types.Mutations in transcriptionfactors thatare expressedspecifically ingonadotropescan resultingonadotropindeficiencywithoutanyeffectsontheotherpituitarycelltypes

Twooftheearliesttranscriptionfactorsexpressedinthedevelopingpituitaryglandarethebicoidhomeodomainfactors,Pitx1andPitx2.Initialexpression

of these factors is detected at e8.5 coincident with initiation of Rathke’s

pouch formation.120,121 At this stage of development Pitx1 and Pitx2arecoexpressedintheuncommittedprogenitorcellsthatmakeupthedevelop-ingRathke’spouch.120,121LaterinadulthoodPitx1andPitx2areexpressed

inallanteriorpituitarycelltypes.120,122

As differentiatedcell typesbegin toappear in thedevelopingpituitarygland,Pitx1(alsoknownasPtx1andP-Otx)continuestobeexpressedinallhormone-producingpituitarycelltypes.122,123InvitrostudiesdemonstratethatPITX1canstimulatetheexpressionofLhbandLhx3.122,124,125GlobaldeletionofPitx1inmice(Pitx1
/
)resultsinanincreasednumberofcorti-cotropesandadecreasednumberofthyrotropesandgonadotropesatbirth,suggesting that PITX1 is important for the balance of pituitary cell-typespecification.56 Interestingly,double heterozygous Pitx1 and Pitx2 mutantembryosdemonstratereductionofallthehormone-producingpituitarycelltypesate18.5.57

PITX2(PTX2,RIEG1)ispresentindifferentiatedthyrotropes,tropes,lactotropes,andgonadotropes,butnotcorticotropes.120Mutationsinthe human PITX2 gene cause Rieger Syndrome This is an autosomaldominantdisordercharacterizedbyocularabnormalities,dentalhypoplasia,craniofacialabnormalities, and,insomecases,decreasedgrowthhormonelevels.121,126 Several mouse models have been established to analyze thedosageeffectsofPITX2.GlobaldeletionofPitx2(Pitx2
/
)causespituitaryhypoplasia and lethality at e14.5 due to severe heart defects.127–130Reduction in Pitx2 expression in mice homozygous for a hypomorphicallele of Pitx2 (Pitx2neo/neo) results in an absence of Lhb, Fshb, and Gnrhr.Othergonadotrope markers, includingEgr1 and Nr5a1, are also absent inthese mice.59,127 Overexpression of Pitx2 in transgenic mice (Cga-Pitx2)resultsintheexpansionofthegonadotropepopulation,asevidencedbyanincreased number of LH-and FSH-positive cells, which are present in adorsallyexpandeddomainate18.5.58TheexpressiondomainforNR5A1,whichis expressed ingonadotropes but no otherpituitary cell types, wassimilarlyexpandedinCga-Pitx2mice.Nosignificantdifferencewasobserved

somato-inthe numberof somatotropes, thyrotropes, or corticotropes.58 Togetherthesestudies showthatgonadotrope differentiation is exquisitelysensitive

toPITX2dosage

Interestingly, mice with gonadotrope-specific deletion of Pitx2(Pitx2£ox/
;Lhb-cre) have normal expression of LH and are fertile.60,127These data suggest that PITX2 is not required for gonadotropemainte-nance orfor regulated productionof gonadotropes.60Many studies haveshown that PITX2 transactivates the genes encoding the gonadotropin

subunits:Cga,Lhb,andFshb.131,132Onepossibleexplanationisthatintheabsence of PITX2, PITX1 may compensate to maintain gonadotropefunction Together thesestudies imply that PITX2is more important inearly gonadotropedevelopment, while PITX1 isthemost imperative forgonadotropemaintenance afterbirth.60Gonadotrope-specificdeletion ofboth Pitx1and Pitx2 would confirmtherole of thesefactors ingonado-tropemaintenance.

TheLIMhomeodomainisnamedafterthethreemembersofthisfamilyoftranscriptionfactors:LIN1,ISL1,andMEC3.133Inhumans, mutationsinthegeneencodingtheLIMprotein,LHX3,areassociatedwithcombinedpituitary hormone deficiency,whichfeatures GH,TSH, FSH, LH, PRL,and sometimes ACTH insufficiency.These patients have compound syn-dromesdisplayingdwarfism,hypogonadism,andhypothyroidismandoftenhave rigid cervical spines, deafness, developmental delay, and intellectualdisabilities.134

Sheng etal studied mice with a globalloss of Lhx3(Lhx3
/
) MiceheterozygousforthisLhx3mutationareviableandfertilebuthomozygousembryos die shortly after birth In the mutant embryos, induction ofRathke’s pouch formation initiatess but its growth is arrested Analysis

ofpituitary-specificlineagemarkersrevealsthatLHX3isrequiredfor genitor cells to commit to thyrotrope, gonadotrope, somatotrope, andlactotropelineages.NoLH-positivecellsarepresentine18.5mutantpitu-itary.61,62Consistentwiththesefindings,Lhx3nullmiceexhibitareduction

pro-in the number NR5A1-positive cells in the ventral portion of Rathke’spouchate14.5andadditionalectopicNR5A1-positivecellsintheextremedorsalaspectofthepituitarybye18.5.However,thosecellsdonotstainfor

LH β-subunit, indicating that they are unable to terminally differentiatedespitetheirexpressionofNr5a1.63

Lhx4 is expressed throughout the invaginating pouch at e9.5, becomesrestricted to the future anterior lobe of the pituitary gland at e12.5, anddiminishesbye15.5.64MicewithhomozygousLhx4genedisruptionhaveaseverelyreduced populationofCGA-positivecells ate14.5andatbirthandfewtonoLHβ-positivecells.ThesameistrueforGnrhratbothe15.5ande18.5.64 Thus, LHX3 is required for the cells of pituitary primordium tocommit to gonadotrope lineage, as well as other pituitary cell types andLHX4maysupport,butisnotrequiredfor,specificationofgonadotropecells

TheLIMhomeodomainfactor,Isl1, iscoexpressedwithLhx3earlyinpituitary development, but their expression domains become distinct bye12.5.96 Specifically, Isl1 is expressed throughout the oral ectoderm ate8.5,maintainedinRathke’spouchate9.5,andventrallyrestrictedbetweene10.5ande11.5.96Bye12.5itisexpressedonlyintheventral,differentiatingcellsthatexpressCgaand Foxl2.63,65Global deletionofIsl1inmice causesarrested developmentsoon after e9.5.135 ISL1and LHX3 are involved inearlystagesofpituitaryontogenesis.Together,thesetwofactorsillustratethecloserelationship betweenthemolecular mechanismsinvolvedincelldif-ferentiationandthoseinvolvedintheexpression ofmarkergenesdefiningmaturecelltypes.136

The GATA-binding family of transcription factors contains zinc fingerswithin their DNA-binding domains Gata2 is the most abundantlyexpressed GATA family member in the pituitary gland.68 Gata2 isexpressed in the ventral pituitary during the closure of Rathke’s pouch

at e10.5 It is expressed at highest levels ventrally throughout pituitarydevelopment.23 In adults, GATA2 is presentin thyrotropes and gonado-tropes.68 Globaldeletion ofGata2is earlyembryoniclethal bye11.5 due

to severe anemia.68 In mice with pituitary-specific deletion of Gata2(Gata2£ox/£ox;Cga-cre) the number of FSH-positive cells appear fewer atbirth, although gonadotropes appear to have differentiated based on thepresenceofLHβ-subunitimmunoreactivityandNR5A1 expression.38,68

In Gata2£ox/£ox;Cga-cre adult mice, the number of FSHB-positive cells isnormal, although serum FSH levels are reduced Thus, gonadotropesare specified normally in the absence of GATA2, although GATA2 isimportantfornormalgonadotropefunction.68

OverexpressionofGata2(Pou1f1-Gata2)causesareductioninthesion domainof dorsalpituitarycell typesand an expansionofventral celltypesat e16.5.This phenotype persists at birth with an expansionof thegonadotrope lineage.23 These data suggest that GATA2 is sufficient forshiftingdifferentiation of precursorcells toward the gonadotrope lineage.While GATA2 is sufficient for specification of the gonadotrope lineage,gonadotropesarespecifiednormallyinGata2£ox/£ox;Cga-cremice,suggestingthatGATA2 is not required for thisprocess.68This may be due to othertranscription factors, such as GATA3 that may partially compensate forGATA2 during pituitary development.68 This possibility is supported by

expres-studiesoftransgenicmiceoverexpressingadominantnegativeformofGata2(Cga-dnGata2).23In thesemice, expression ofGATA2 targets is repressed,effectivelyeliminatingcompensationbyotherfactors.Cga-dnGata2embryosexhibit a severe reduction in Lhb,Tshb, Cga,and Nr5a1, but an expandedexpression domain for Pou1f1.23 These data suggest thatGATA2 or othermembersoftheGATA-bindingproteinfamilyarenecessaryforgonadotropedifferentiation.

The balance between expression of Gata2 and Pou1f1 appears to beimportant for appropriate gonadotrope specification.23 POU1F1 (alsoknownasPIT1)isrequiredfordifferentiationofthePOU1F1lineage,which

iscomprisedofthyrotropes,somatotropes,andlactotropes.137Misexpression

ofPou1f1(Cga-Pou1f1,Table1)intheventralaspectofthepituitaryresultsinexpansion of thyrotropes and loss ofgonadotropes at e18.5.23 These datasuggestthatectopicPOU1F1stimulatesprogenitorcellstoswitchtheirfatedecisionanddifferentiateintothyrotropesinsteadofgonadotropes(Fig.1).Consistentwiththisfinding,lossofPOU1F1function,asseeninPou1f1dw/dwmice results in increased expression and dorsally expanded domains forLhb and Nr5a1 at birth.23,137 In vitro studies demonstrate that POU1F1can bind to GATA2via protein–proteininteractionsand prevent GATA2transactivationofNr5a1.23Interestingly,POU1F1doesnotinhibitbinding

ofGATA2totheTshbpromoternordoesitpreventtransactivationoftheTshbpromoter.In fact, theyact synergistically.23,138 Thismayexplain howthebalance betweenGATA2 andPOU1F1 regulatesgonadotrope andthyro-trope specification This mechanism is supported bystudies of mice thatoverexpressaformofPOU1F1incapableofbindingDNA,butretainingtheabilitytointeractwithGATA2(Cga-Pou1f1DBmut).23ThesemicehavenormalTshbandseverelyreducedLhbandNr5a1ate18.5.23MutationsofresiduesinPOU1F1thatarenecessaryforinteractionwithGATA2eliminatetheability

ofPOU1F1to reducegonadotropespecification.23Together thesestudiesconfirmthesuggestionthatPOU1F1inhibitsgonadotropespecificationinaDNA-bindingindependentmanner

Prop1 encodes a paired-like homeodomain transcription factor present inmurinepituitaryfromaboute10.5–e16.5.Thisgenehasbeenthesubjectofintensive study, as humans with combined pituitary hormone deficiencyfrequentlyhavemutationsintheProp1locus.139Thenameindicatesitsrole

astheProphetofPIT1anditwasoriginallylinkedtotheemergenceofthe

Pit-1lineageofanteriorlobecells.40Recentlyithasbeenproventhatpositivecells giveriseto everyotherhormone-producingcell typeintheanterior lobe of the pituitary gland.140 This was effectively demonstratedusingaProp1-creandlineagetracingexperimentswiththeRosa26tdTomatoEGFPmousereporterinwhichcellscontainredfluorescentdtTOMATOprotein

Prop1-intheabsence of CREactivityor green fluorescent eGFP protein inthepresenceofCREactivity.140,141

ThemostextensivelystudiedmousemodelforProp1functioninonicdevelopment is theAmes dwarf mouse Prop1df/df, firstdiscoveredas aspontaneousmutationcausingdwarfisminacolonylocatedinAmes,Iowa,USA.ThesemicehavemissensemutationSer83ProinthehomeodomainofProp1.65,69 Although the mice exhibit no morphological changes untile12.5–e14.5,theyeventually have anterior lobedefects, including ventralhypoplasiaanddorsaldysmorphology.20Prop1
/
micehaveessentiallythesame phenotype, although they occasionally die shortly after birth fromrespiratory distress.69 The gonadotrope phenotype of these mice appearsnormalat8weeksof age, butreducedFSH β-andno LHβ-subunits aredetectedate18.5.69 TheAmes dwarfs exhibitreducedgonadotropinpro-duction,persistentHesx1expression,alessdefinedregionofNr5a1andTle3,reduced Lef1 at e13.5–e14.5 in the anterior lobe, and no detectableNotch2.40,51 Thelattercauses severepituitary glanddysmorphology and isdiscussedfurtherintheNotch2sectionofthischapter.40Thesedatahaveled

embry-totheassociationofnumeroussignalingandtranscriptionfactorswithProp1andultimatelypituitarydevelopment

Another model of study providing valuable data regarding therole ofProp1ingonadotropedevelopment istheCga-Prop1transgenicmouse,themostrecentiterationbeingtheTgN(Cga-Prop1)D65acstrain.70Undercontrol

oftheCgapromotera2.2-kbfragmentofProp1isexpressedstartingate9.5intheanteriorlobeofthepituitarygland.139MultiplemiceweregeneratedandthemodelwiththehighestProp1expression,as detectedbyinsituhybrid-ization,waschosenforstudy.ItwasalsodeterminedthatmutantmiceonaC57BL/6backgroundwere themostviable.70,139 ThesePROP1gain-of-functionmice undergo delayed puberty and present withhypogonadism,mostlikelytheresultofa delayingonadotropedifferentiation,asLhb andFshb are not detected before birth.70 Extensive study of this model haselucidateda novel mechanism mediating Prop1regulation of gonadotropedevelopment as mutant animals had comparable NR0B1, NR5A1,POU1F1,Gata2,andEgr1atembryonictimepoints.70Microarrayanalysisidentified Prss28 and Klrg1 as putative PROP1targets responsible for the

delayed puberty in gain-of-function animals Unfortunately, no furtherstudy involving thesegenes has been published since 2006 Crossing thePROP1 gain-of-function mice with the Ames dwarf rescues the dwarfphenotype discussed earlier Although this corrected the GH deficiency,the offspring maintained other defects including hypothyroidism.139Therefore,Prop1mustbedownregulatedaftere14.5for normalcelldiffer-entiationtooccurinthepituitarygland,andconstitutiveexpressionofthesameresultsindelayedspecificationofgonadotropes.

Informationongonadotropedevelopmenthasalsobeengatheredfrombreeding the Ames dwarf with Lhx4
/
animals to obtain the Prop1df/df;Lhx4
/
double knockouts These mutants have almost no anterior lobeand at e12.5 no LHX3-positive cells.65 Single mutants of Prop1 or Lhx4maintainnormalandreducedLHX3levels,respectively,therebyindicatingthatProp1playssomeroleinLhx3expressioninthepituitarygland.Cgaisalso not detected during embryogenesis in the double knockoutmodel.There areno detectablehormone-expressingcell typesatP0 withthe exception of a reduced corticotrope population.65 Interestingly,the NR5A1 domain is larger in the double-mutant pituitaries than intheLhx4mutantwhere almostnoNR5A1-positivecellsexist.InProp1df/dfmice there is an expanded population of NR5A1-positive cells.65Therefore,PROP1andLHX4have antagonisticrolesinregulatingNr5a1,withPROP1repressingandLHX4actingtoupregulateexpression

HESX1is ahomeoboxprotein expressedintheventraldiencephalonandRathke’spouchofmiceate9.0.Itsexpressionisextinguishedbye15.5.142Mutations inthe humanHESX1 genecause septooptic dysplasiacharac-terizedbyoptic nervehypoplasia, pituitarygland hypoplasia,andmidlineabnormalities of the brain.143–146 Pituitary hypoplasia may manifest asisolatedgrowthhormonedeficiencyorcombinedpituitaryhormonedefi-ciencyconsistingofreductioninGH,TSH,LH,andFSH.147,148Pituitarydeficiencymaybe secondarytohypothalamicdefects.148HESX1actsasatranscriptionalrepressorandabnormalitieswithHESX1mutationsmaybedueto theinabilityofHESX1to interactwithits corepressorsGroucho/Transducin-likeenhancerofsplit-1.149,150

GlobaldeletionofHesx1resultsinavariablephenotype.71Theanteriorpituitary is hypoplastic and less associated with the posterior lobe in allHesx1
/
mice In micewith themoresevereclass I phenotype,embryos

lacka Rathke’s pouchstructureat e12.5.71Themilderclass IIphenotypeincludesaberrant morphogenesisof Rathke’s pouch in whichthe pouchcontainsbifurcationsoneachendate12.5.71Approximately1%ofhetero-zygousHesx1mutantsexhibittheclassIIphenotype.Whileanumberofclass

IImicedieperinatally,somesurviveweaning.Thesemicewerefertileandgaverisetoviableoffspring,suggesting thattheyhave functionalgonado-tropes.Theremaybe geneinteractionsthatcausethevariablephenotype.Overall, this study suggests that HESX1 is not essential for gonadotropedevelopmentorfunctionincertaingeneticcontexts.71

SIX3 is a homeobox transcription factor that is required for normalforebrainformation.151TheexpressiondomainsforSix3andHesx1partiallyoverlap within Rathke’s pouch, suggesting they may both contribute

to pituitary development Hesx1cre/+ mice were engineered by replacingtheentire Hesx1-codingsequence with a cre cassette, thus creating a nullallelefor Hesx1.72Thesemicewere matedtoSix3heterozygous nullmice

to generate offspring that are doubly heterozygous for Six3 and Hesx1(Six3+/
;Hesx1cre/+).72,151Six3+/
;Hesx1cre/+miceexhibit twodistinctphe-notypesate17.5.Two-thirdsofSix3+/
;Hesx1cre/+embryoshavehypoplasticpituitaryglandswithareducednumberofLhb-positivecells.Inone-thirdofthemutantembryos,thepituitaryglandisabsentfromthesellaandisinsteadectopically located in the nasopharyngeal cavity Lhb expression is barelydetectable at e17.5 suggesting a delay in gonadotrope differentiation.GonadotopespecificationisnormalasevidencedbytheexpectedexpressionpatternsofNr5a1.72ThesestudiessuggestthatHESX1andSIX3contribute

tonormalLhbexpressionandpossiblygonadotropedifferentiation

Otx1encodesahomeoboxtranscriptionfactordetectedinmicefrombirthuntilapproximately4monthsofage.73TheOtx1geneisrelatedtothePtxgenefamilyandincellculturehasbeenshowntobindthePitx1recognitionsequencetoactivate Gh, Cga,Lhb,and Fshbpromoters, aswellas interactwith SMAD proteins.73,152 Mice with Otx1 exons 1 and 2 replaced

by lacZ via homologous recombination were classified as Otx1
/
.153Heterozygousmutantsshownodiscernablephenotypethereforehomozy-gous nullswere analyzed.Researchers used a hybrid B6/D2 backgroundover the highly lethal 129/Sv, although the former still demonstrates30% mortality in the first 30days postbirth and a higher incidence ofepilepticseizures.73,153Otx1
/
micepresentwithtransientdwarfismfrom

approximately P7to4monthsofage, withthemostseveresize difference

atP30.Thisisaccompaniedbya 70–80%reductionincirculatingLHandFSHduringtheaffectedage Testesandovariesinmutantmice aresignif-icantlyreducedinsizeduringthedwarfperiodandmalesexhibitadecrease

in mating frequency as compared to controls.73 Pituitary gonadotropinexpression as detected by immunohistochemistry shows a “strongly”reduced gonadotrope population at P25, which is restored to wild-typelevels by 4months of age.73 Analysis of hypothalamic GnRH reveal nodifferences in protein levels compared to wild type mice, therefore anyaberrantgonadotropedevelopmentis likelynotthedirectresultofalteredreleasinghormonestimulation.73 Histologicalanalysis ofGnrhrexpressionandanapoptosisassayofthepituitaryglandindicatenormallevelsofGnrhrand imply no increase in programmed cell death.73 The researchers didnot treat mice with GnRH to verify competent hypothalamic signaling.However, by proving comparable numbers of Gnrhr-positive cells in thepituitary gland and GnRHexpression in thehypothalamus,as wellas noincrease in cell death in the pituitary, they suggest that the gonadotropepopulation is intact in the mutant animals without normal LH and FSHexpression.73 Therefore, this model purports Otx1 transiently regulatespostnatalhormone productioninthepituitaryas opposedtogonadotropedevelopment,potentiallyinconcertwithSMADproteins

PatientswithmutationsinOTX2exhibitaspectrumofphenotypesthatinclude, in some instances, combined pituitary hormone deficiency andhypogonadotropichypogonadism.34GlobaldeletionofOtx2inmiceresults

in embryonic lethality due to absenceof forebrain, midbrain, and rostralhindbrain.154ConditionaldeletionofOtx2usingFoxg1cre/+miceresultedin

an array of pituitary dysmorphology phenotypes, but did not affect cellspecification based on CGA and POU1F1 immunohistochemistry.44,74,75Similarly,Otx2£ox/£ox;Pitx2+/cremice exhibitedpropercellspecificationandpituitarymorphologywastypical.74,155Thesestudiessuggestthatgonado-tropespecificationisnotdependentonOTX2.Gonadotropephenotypesinhumanpatientsmaybeduetolossofhypothalamicfunctionoraninabilitytomaintaingonadotropefunction

Pax6isa transcriptionfactor thatisfirst detectedate9.0withtheproteinevident by e11.5 It is primarily localized to the dorsal portion of thedeveloping pituitarygland.76Expression is observed inthe diencephalon

bye9.5,a structurereferredtoas thepituitaryorganizer.78Variousmousemodelshave greatlyaidedthe studyofPax6 originatingwiththesponta-neouslymutatedSmalleyeorSeymice(Pax6Sey/Sey)andarecapitulationofthesamedesignatedas Pax6
/
.76–78Pax6Sey/Sey micehave apointmuta-tion in chromosome 2 truncating thetranslated protein and resulting infunctionallyinactivePAX6.79,156AnumberofmodelssimilartoPax6Sey/Seyhavebeen developedincludingPax6
/
andPax6Neu/Neu(thelatter main-tains the homeobox domain but lack the transactivation domain) andbecause of comparable pituitary phenotypes these three models will bediscussedasawhole.

AbsenceoffunctionalPAX6resultsinmice withnoreactive neurons during development.80 Thus, anygonadotrope develop-ment would be independent of GnRH stimulation Pituitary glands ofPAX6-deficientmice exhibit ventralization,a shift in theidentityof cellswheredorsalpituitarycellsadoptventralfates.76Inwildtypemicegonado-tropesarelocatedprimarilyintheventralaspectofthepituitarygland,andanincrease ofthis cell type is observed in Pax6
/
mutants as ventralizationoccurs.Additionally,noLHB-positivecellsintheCGA-positivepopulationwereobservedreaffirmingpreviousstudiesconnectingPax6tohypothalamicGnRH stimulation.76 Overall, this model emphasizesthe effect of spatialdevelopmentsignalsoncontrollingthespecificationof thehormone-pro-ducing cell typemake-up of thepituitary gland, and the preciseregionalcontrol Pax6hasover thegonadotropelineage Whether thealterationingonadotropefunctionis entirelydueto theabsenceofGnRHstimulationfromthehypothalamusorisalsoeffectedbyventralizationisunclear.Studies

GnRH-immuno-ofPax6knockoutmicewiththemutationlocalizedspecificallytoeitherthepituitaryglandor thehypothalamus wouldbe valuable to understandthisphenomenon

EGR1,alsoknownasKROX24,NGFIA,andZIF268,isanimmediateearlygene containing a DNA-binding domain composed of three zinc-fingermotifs.157,158EGR1isexpressedintheendothelium,thymus,muscle,car-tilageandbone,andthecentralandperipheralnervoussystems.159,160EGR1

isimportant for numerous physiologicalprocessesincluding synapticticity, woundrepair, inflammation, and differentiation.161 Invitro studiessuggestthatEGR1synergizeswithPITX1andNR5A1tostimulateexpres-sionof Lhbinresponse toGnRH.125,162,163 Inthepituitarygland,Egr1 is

plas-expressedintheposteriorandanteriorlobes,aswellasinthemesenchymesurrounding the pituitary gland during development In the adult, Egr1expression is absent from the posterior pituitary, but strongly expressedthroughouttheanteriorlobe.81GlobaldeletionofEgr1(Egr1
/
)resultsinreducedgrowthrateofmice.Bothmaleandfemalemutantsareinfertilewithpituitaryandovariandefects.Specifically,pituitarysomatotropesarereduced

innumberandserumGHlevelsarelowat12weeksofage.GonadotropesarepresentinexpectedquantitiesbasedonthenumberofFSHβsubunit–positivecells.However,LHβ-subunitisabsentandserumLHisundetect-able.81Ovariectomyof Egr1
/
mice stimulatesFshb, but notLhb expres-sion.162 Interestingly, expression of the gene encoding the LH receptor,Lhcgr,isabsentintheovary.81BasedonthesestudiesitappearsthatERG1

is requiredfor normalexpressionofLhb,but notfordifferentiation ofthegonadotropelineage

Msx1, originally known as Hox7, is a homeobox gene with similarity tomuscle segment homeobox (msh) from Drosophila.164 Msx1 is expressed infacialprocesses ofe10.5–e13.5 mouseembryos, as wellas theventral aspect

ofthedevelopingpituitarygland.165,166Msx1isfirstexpressedinthepituitarygland at e10.5 becoming localized to ventral aspect of Rathke’s pouch bye12.5.166,167The ventral patternof Msx1 expression together with the factthatitisexpressedingonadotrope-derivedcelllinessuggeststhatMSX1mayplaya roleingonadotropedevelopmentorfunction.167AmousemodelforglobaldeletionofMsx1wasgeneratedbyinsertionofalacZcassetteintothehomeobox domain of Msx1.82,83 Msx1
/
mice exhibit perinatal lethalitywithdefectsincraniofacialdevelopment,cleftpalate,andtoothagenesis.168,169ThesemiceexhibitincreasedGnrhrandCgaexpressionate18.5.167LhbandFshb are not significantly increased, but show a trend upward.167 This isconsistentwithinvitrostudiesdemonstratingtheabilityofMSX1torepressGnrhrandCgapromoteractivity.167MSX1may beimportantforregulatingtiming of gonadotrope differentiation.167 Inhibition of BMP signaling byoverexpressingadominantnegativeversionofBmpr1a(Cga-ΔBMP)stimulatesrobust upregulation of Msx1 expression throughout the pituitary gland ate17.0 Surprisingly, gonadotropes are specified despite increased expression

ofMsx1.28StudiesthatexaminethetimingofgonadotropedifferentiationinMsx
/
micewillbeimportanttodeterminewhetherdeletionofMsx1causesprematuredifferentiationofgonadotropes

3.10 TBX19

TheT-boxtranscriptionfactor,TBX19(alsoknownasTPIT),wasidentified

asacell-specificfactorthatstimulatesexpressionofthepituitaryPomcgene.TBX19isfirstdetectedinnucleiofpituitarycorticotropesate12.5.Ate13.5,allcorticotropescontainTBX19.170Inmicewithaglobaldeletionof Tbx19

inwhichmost of theTbox–coding sequences are deleted(Tbx19
/
), thepituitaryexhibitsapparently normalhistologyat e14.5but thenumberofPOMC-positivecells is greatly reduced.84These cells may be blocked intheirdifferentiationprocessormaydifferentiateintoothercelltypes.IntheabsenceofTBX19,alargenumberofNR5A1-positivecellsareectopicallylocatedintheintermediatelobesof Tbx19
/
miceate16.5,suggestingthatprogenitorcellsoftheintermediatelobehaveerroneouslydifferentiatedintogonadotropes.84 Colocalization experiments show that intermediate lobecellsofTbx19
/
mice expressPomcorCga,neverboth, andthatCga andLhbexpressioniscolocalizedinmanycells.84AfewPOU1F1-independentthyrotropesarealsopresentintheintermediatelobe.ThusintheabsenceofTBX19,theintermediatelobecontainsgonadotropes,POU1F1-indepen-dentthyrotropes,aswellastheexpectedcelltype,melanotropes.Thesethreecelltypesdonotappeartohaveamixedorabnormalcellidentity.84Cellfatechangesalsooccur intheanteriorlobe.Thecaudalregionoftheanteriorlobeof Tbx19
/
micecontainNR5A1ate16.5.Atthisdevelopmentaltimepoint,thiscaudalpartofthepituitarydoesnotnormallyhaveNR5A1-orCGA-expressing cells, suggesting a cell fate switchfrom the corticotropelineagetothegonadotropelineage.84

Gain-of-function experiments using theCga promoter to driveTbx19expressioninthegonadotropelineagedemonstrate thatpituitaries ofsuchtransgenicmice havemorePOMC-positivecellsand fewerCGA-positivecells in the anterior lobe while TSHB-positive cells are unaltered LHβ-subunit is undetectable and FSH β-subunit is significantly decreased.Also,veryfewNR5A1-positivecellsarepresent.84ThesedatasuggestthatTBX19isanegativeregulatorofgonadotropedifferentiation

Nr5a1(alsoknownasSf1orFtz-F1)isexpressedinpituitarygonadotropes,aswell as hypothalamus and steroidogenic tissues.171 The onset of pituitaryexpressionofNr5a1occursatapproximatelye13.5.30Nr5a1regulatesmany

of thegenes thatconstitute thegonadotrope phenotype including Gnrhr,Lhb,andFshb.114,162,172

Micewithhomozygous disruption of theNr5a1gene (Nr5a1
/
) lackdetectable LH, FSH, and GnRHR However, GnRH treatment restoresgonadotropinexpression.30 Similarly,mice withpituitary-specificdeletion

ofNr5a1(Nr5a1£ox/
;Cga-cre)haveseverelyreducedexpressionoftropinsandGnRHtreatmentrestoresproductionofLH.86Together thesestudiessuggestaroleforNR5A1innormalgonadotropefunction,butnotdifferentiation

gonado-Anotherorphannuclearreceptor implicated ingonadotropementand function is NR0B1.Humanmales withmutations inthegeneNR0B1 (also known as DAX1 and AHCH) exhibit hypogonadotropichypogonadism.87 In an attempt to recapitulate this phenotype, a mousemodel was created by crossing Nr0b1£ox/£ox mice with CMV-cre mice.87ThepituitaryglandsisolatedfromthesemicehadnochangeinLHβ-andFSH β-subunit staining as compared to controls.87 Additional testing ofthesemutantsonamixedB6/129Svgeneticbackgroundrevealedaberrantgonadaldifferentiation.However,pituitary-specific deletionofNr0b1wasnot performed nor was Nr5a1 measured.173 Molecular studies implicateNr0b1 and Nr5a1 as negative regulators of each other, but no in vivoexperiments have yet revealed a specific role for Nr0b1 in gonadotropedevelopmentorgonadotropinproduction.107,174

Forkhead transcription factors are classified basedon a conservedwingedhelix/forkhead DNA-binding domain consisting of three β-sheets, threeα-helices, and two loops forming the “wings”.175 FOXL2, is present inthepituitary,ovary,anddevelopingeye andistheforkheadfactor thathasbeen studied most extensively in reproduction.27,88–91,176–186 Foxl2 tran-scriptisfirstdetectedinthedevelopingpituitaryate10.5andFOXL2protein

is detectableaday laterate11.5.27,28 FOXL2 appearsto serveas a marker

of gonadotrope and thyrotrope commitment.28 In adultpituitary FOXL2

ispresentingonadotropesandthyrotropesandduringpregnancy,FOXL2ispresentinasmallfractionoflactotropes.27FOXL2interactswithSMADstostimulateexpressionofgonadotropegenes.116,177,178,183

Overexpressionof Foxl2 in pituitary tissue (Cga-Foxl2) is sufficient tostimulateproductionofCGA.Consistentwiththis,FOXL2proteinisfirstdetectableate11.5intherostraltipofthedevelopingpituitary,coincidentwith thespatialand temporal patternsof CGA protein.27 Interestingly, inCga-Foxl2 embryos ectopic FOXL2 was also detected in discrete regions

borderingtheposteriorlobeandwassufficienttodriveectopicexpressionofCGA inthese regions as well.27 Invitro studiessuggest that FOXL2 alsostimulatesexpressionofGnrhr.116PituitaryglandsfrommicewithaglobaldeletionofFoxl2(Foxl2
/
)haveverylowlevelsofFshbandCgaandreducedlevelsofGnrhrand Fst.89 Activinis unabletostimulateFshbtranscription,suggesting thatFOXL2 mediates activin stimulationof Fshb expression.88Gonadotrope differentiation does occur as evidenced by the presence of

LH.88 Pituitary glands from mice with gonadotrope-specific deletion ofFoxl2(Foxl2£ox/£ox;Gnrhr+/GRIC)areofnormalsizewithreducedexpression

ofFshbandFst,althoughexpressionofCga,Lhb,andGnrhristypical.32,90,91ThisdifferenceinthesemousemodelsmaybeduetothedistincttimingofFoxl2deletioninthetwomodelsortotheabsenceofFoxl2innonpituitarytissuesinFoxl2
/
mice

The forkhead transcription factor, FOXD1 (also known as BF2), isrequired for normalkidney development.187,188 Foxd1is expressed inthemesenchymesurroundingthepituitarygland,butnotinthepituitaryglanditselfduringdevelopment.92Likemanyforkheadfactors,FOXD1isasingle-exongene,thusnearlytheentirecodingsequencewasreplacedwithalacZcassettetogenerateanullallele.187Foxd1
/
micedieshortlyafterbirthdue

to renal failure.187 Foxd1
/
mice have a dysmorphic pituitary gland thatextendsthroughthecartilageplatefrome14.5throughe18.5.92ExpressionlevelsofGh,Tshb,Pomc,Fshb,andCgaarenormalate18.5.However,inspite

ofthefactthatFoxd1isnotexpressedinthepituitaryduringdevelopment,Lhb expressionis significantly reducedin embryoslacking Foxd1.92 ThesefindingsindicatethatFOXD1isrequiredfornormalgonadotropefunctionduringdevelopment.Basedonitsexpressioninthemesenchymesurround-ingthepituitary gland, FOXD1 likelyacts indirectly viasignaling factorssecretedfromthemesenchyme

Development

Whilemuch isknownaboutregulationofthegenesinvolved intrope development by classical transcription factors, little information iscurrentlyavailableregardingnonclassicalmethodscontrollinggonadotropedevelopmentandfunction.Epigeneticregulationinvolvesstable,heritablealterations of the genomethat do not affect DNA sequence Epigeneticregulationis essentialto normaldevelopmentand differentiation.189 Twobasic mechanisms of epigenetic regulation are DNA methylation and

gonado-histone modification Lysine-specific demethylase 1 (LSD1) is a histonelysinedemethylase.190,191GlobaldeletionofLsd1resultsinearlyembryoniclethalityat e7.5.93 Pituitaryspecificdeletionof Lsd1(Lsd1£ox/£ox;Pitx1-cre)reveals a role for LSD1 in terminal differentiation of pituitary celltypes.51,93 Lsd1£ox/£ox;Pitx1-cre embryos exhibited a severe reduction inmarkersforterminaldifferentiationofsomatotropes,thyrotropes,gonado-tropes and corticotropes Lhbexpressionwas barely detectableand Nr5a1expressionwasdiminishedate17.5.Cgaexpressionwasapparent,althoughveryfewCGA-positivecellscouldbedetectedbyimmunohistochemistry,suggesting possibleposttranscriptionalregulation.93Thus,epigeneticreg-ulationisanimportant componentofgonadotropedevelopment.

Theamountofproteinproduceddependsnotonlyontherateofscription, but also on transcript stability and the rate of translation intoprotein.InvitrostudiessuggestthatGnRHregulatesstabilityofLhb,Cga,and Gnrhr transcripts.192–198 Translation of transcripts can be inhibitedthrough binding of short, ∼22 nucleotide RNAs called microRNAs.These microRNAsare processed from primarytranscriptsvia a multistepprocess involving a cytoplasmic double-stranded, RNA-specific RNasecalledDicer.199ToevaluatetheroleofmicroRNAsingonadotropes,Wang

tran-etal.deletedDicerspecificallyingonadotropes(Dicer£ox/£ox;Lhb-cre).60,94,95Mice lacking Dicer in gonadotropes had significantly reduced expression

of Fshb,Lhb,andCga.Also, stimulationofLH productionbyGnRHwasabolishedintheabsenceofDicer.94Thesestudiessuggestthatposttranscrip-tional regulation is an important mechanism controlling gonadotropinproduction.UseofanearlieractingCREwouldallowresearcherstodeter-mine whether epigenetic regulation is equally vital during gonadotropedifferentiation

Cre-loxPstrategy allows for disruption of genes without embryoniclethalitycausedbyglobalknockoutsandeliminatessecondaryeffectsduetodeletionofgenesinothertissues.Specificpromoterstoregulatethetemporalandspatialexpressionofcrearekeytothisprocess.SeveralCREmouselineshavebeendevelopedthattargetgonadotropecells(Table2)

ThegeneencodingthereceptorforGnRH,Gnrhr,isoneoftheearliestgonadotrope-specificgenestobeexpressed,making itanexcellentchoicefor gonadotrope-specific expression studies in embryos and adults.29

CRE Line Generation Targeted Expression References

GRIC Targeted mutation, IRES and cre inserted

in frame downstream of exon 3 of Gnrhr

Recombination in gonadotropes and gonadotrope precursors, as well as GnRH neurons,

recombination detected by e12.75

Cga-icre Transgenic, 4.6 kb of proximal promoter

from mouse Cga fused to cre with reduced susceptibility to epigenetic silencing

CRE activity in adult gonadotropes and thyrotropes with high penetrance, no ectopic CRE activity, recombination in differentiating pituitary cells at e12.5

[203]

MGI ID: 5547616

Bα/CreTeR Transgenic, –313 to +48 of bovine Cga

driving tetracycline-inducible expression

Lhb-cre Transgenic, 776 bp of bovine Lhb promoter

fused to cre with engineered KOZAK

Gonadotrope-specific, recombination by e16.5 [60]

MGI ID: 3822455

Fshb-icre Transgenic, 4.7 kb of proximal promoter

from ovine Fshb fused to cre with reduced susceptibility to epigenetic silencing

CRE activity beginning embryonically, detected in 99–100% of adult gonadotropes, no

recombination in hypothalamus, testis, or ovary

[209]

MGI ID: NA

Nr5a1-cre Transgenic, BAC containing –23 kb to

+88 kb of mouse Nr5a1 with cre inserted

at ATG initiating methionine

Recombination in pituitary by e13.5, recombination also in somatic cells of gonads, adrenal cortex, spleen, ventromedial

hypothalamic nucleus, choroid plexus, hippocampus, and cerebral cortex

tamoxifen-inducible form of cre replaced the coding exon for Foxl2, creates a null allele of Foxl2

single-Recombination in pituitary not examined, recombination detected in ovary

[212]

MGI ID: 4940388

Foxl2tm1.1(cre/ ERT2)

Kliu Targeted mutation, IRES and

tamoxifen-inducible cre were inserted downstream

of the Foxl2 gene

Recombination in pituitary not examined, recombination detected in ovary

[213]

MGI ID: 5548141

Foxl2tm4(cre)Tre Targeted mutation, cre replaced the

translational start site of Foxl2, creates a null allele of

ELHX3-cre Transgenic, 7.9-kb enhancer from human

LHX3 and Hsp68 minimal promoter fused to cre

Recombination in caudomedial pituitary gland

at e14.5, recombination in CGA-, TSHB-, FSHB-, and LHB-positive cells at birth

[33]

MGI ID: 5576209

Isl1-cre Targeted mutation, cre inserted into Isl1

locus replacing endogenous ATG

Recombination in pituitary not examined, recombination detected in cardiac and hindlimb progenitors

[219]

MGI ID: 3623159

Isl1-IRES-cre Targeted mutation, IRES and cre inserted

downstream of the exon containing the second LIM domain of Isl1

Recombination in pituitary not examined, recombination detected in motor neurons and dorsal root ganglia at e12.5

Gnrhr^internal ribosome entry site–cre (GRIC) mice contain an internalribosome entry site (IRES) and cre-coding sequence inserted in-framedownstreamof exon3oftheGnrhrgeneallowingforcoexpressionoftheGnrhrgeneandcre GRICmiceareviable,fertile,andpresentinexpectedMendelian ratios.32 GRIC mice were crossed with Rosa26-YFP reportermice to label cells in which CRE activity was present.32,200 FSHB andLHB are present 99.9% of gonadotropin-containing cells labeled withYFP.YFP-positivecells areonly presentin 1.8%of cells lackinggonado-tropins Recombination occurs as early as e12.75.55 Recombination alsooccursectopicallyinhypothalamicGnRHneuronsandtestis.201

Thegonadotropinsubunit,Cgaisinitiallyexpressedintheoralectoderm,which will form the pituitary gland Transgenic Cga-cre mice have beencreatedinwhichexpressionofcreisdrivenby4.6kbofproximalpromoterfromthemousegenecodingforthecommongonadotropinα-subunit,Cga

In thesemice, recombination occurs inpituitary primordiumand all celltypesoftheadultanteriorpituitary.38Ectopicexpressionispresentincardiacandskeletalmuscle.38Littleornoexpressionisdetectedingonads,adrenalglands,brain,ventromedialhypothalamus,orkidneys.38,86Additionalstud-iesusingCga-cremice indicatethatinitialcreexpressionoccursatapproxi-matelye9.5.37A similarCga-crelinewasdevelopedusing4.5kbof mouseCgapromotertodrivecreexpressioninthepituitarygland.Recombinationoccurs in pituitary, but was not tested in specific pituitary cell types.Recombinationwasnotdetectedinhypothalamus.202

Another transgenic CRE line, Cga-icre, has been developed in whichCpG content of cre is reduced as compared to the prokaryotic codingsequence to diminish the chance of epigenetic silencing in mammals.203Thecre-coding sequencewas inserted downstream of the ATG encodingtheinitiatingmethionineforCga.CREactivityisdetectedindifferentiatingpituitarycellsate12.5.InadultmiceCREactivityispresentingonadotropesand thyrotropes withhigh penetrance Little or no activity is detectedinskeletalorcardiacmuscle,brain,kidneys,lungs,testis,ovary,andliver.203Tetracycline-inducibleCga-cretransgenicmice(Bα/CreTeR)havebeengeneratedin whichthe –313 to +48 region ofthe bovine Cga promotertargetsexpressiontogonadotropes,butnotthyrotropes.204Thisfragmentofthe bovine Cga promoter was subcloned into the CreTeR vector, whichcontaineda 1.7-kb fragmentwith the cre-coding sequence and a nuclearlocalizationsignalinserteddownstreamoftheTetOsequence.205Tocharac-terize Bα/CreTeR mice, they were crossed with a Rosa26lacZ reporterline.206InadultBα/CreTeRmice,creexpressionisinducedwith7daysof

doxycycline treatment CRE activity was apparent in all gonadotropes.Othercelltypeswerenotevaluated.205

Othergonadotropinsubunits have alsobeen used to developcrelines.Lhb,isexpressed specificallyingonadotropecellsstarting atapproximatelye16.5.31Lhb-cremiceweregeneratedusing776bpofbovineLhbpromoter

todriveexpressionofacre-codingsequencewithanengineeredconsensusKOZAKsequenceandanuclearlocalizationsignal,polyadenylationsignaland intron from the rabbit β-globin gene.60 Lhb-cre mice crossed with aRosa26-GFP reporter line or a Rosa26-lacZ reporter line to evaluate cellspecific CRE activity reveal that recombination occurs in approximately80% of LHB-positive cells and not in LHB-negative cells of adultmice.60,207,208Embryonically,oneortwopituitarycellspersectionexhibitrecombinationate16.5,whilemanycellshaveundergonerecombinationate17.5 and e18.5, consistent with the expected pattern of expression forLHB.60 CRE activity is absent from hypothalamus and adrenal gland.60Limited CRE activity is present in thebrain and kidney.60 Low levels ofrecombinationoccurinafewovarianfolliclesandcorporaluteaoffemalemiceandinsomeseminiferoustubulesofmalemice.60

Expressionofthegonadotropinsubunit,Fshb,isdetectableshortlybeforebirthinthemouse.31Transgenicmice inwhich4.7kb ofovineFshbpro-moterdrivesexpressionoficrearereferredtoasFshb-icremice.209InFshb-icremicerecombinationisfirstdetectedintheanteriorpituitaryduringembry-onicdevelopment.Intheadultmouse,recombinationoccursin97–100%ofgonadotropes and a small percentage of other pituitary cell types.Recombinationwasnotdetectedinthehypothalamus,testis,orovary.209

Inadditiontocrelinesusinghormoneorreceptorregulatorysequences

to target recombination togonadotropes, manytranscription factor geneshave been utilized for this purpose,including Nr5a1.Nr5a1 codes for theorphannuclear receptor,steroidogenicfactor-1,andishighlyexpressed insteroidogenictissuesincludinggonadsandadrenals,aswellasnonsteroido-genictissues,suchashypothalamusandpituitary.30,210 LowlevelsofNr5a1expression are initially detected in the developing pituitary gland ate13.5.30Transgenicmiceweregeneratedwithabacterialartificialchromo-some containing 23kb of sequence 50 to the start site of transcriptionand 88kb of themouse Nr5a1gene and 30 region A cassette containingthecre-codingsequenceand thebovineGhpolyadenylationsequencewasinsertedintotheNr5a1locusattheATGinitiator methionine.211Nr5a1-cremice were crossed with Rosa26-lacZ reporter mice to characterize CREactivity.206Nr5a1-cremiceexhibitrecombinationinthedevelopingpituitary