Role of Retinal Pigment Epithelium in Myopia …...based, corneal reshaping therapy and atropine, a...

RoleofRetinalPigmentEpitheliuminMyopiaDevelopmentandControl

By

YanZhang

Adissertati tionofthe

onsubmittedinpartialsatisfac

requ eof

irementforthedegre

DoctorofPhilosophy

in

VisionScience

inthe

GraduateDivision

ofthe

UniversityofCal ornia,Berkeleyif

Committeeincharge:

Profess ,Chair

o trChristineF.WildsoeProfessorMarlaFellerProfessorKunxinLuo

Spring2013


©2013

By

YanZhang

UniversityofCalifornia,Berkeley

Abstract


by

YanZhang

DoctorofPhilosophyinVisionScience

UniversityofCalifornia,Berkeley

ProfessorChristineF.Wildsoet,Chair

Myopia(near‐sightedness)isoneofthemostcommonoculardisordersinhumans.Duetothe dramatic increases in prevalence of myopia worldwide, especially in children andyoungadults,myopiahasalsobecomeasignificantpublichealthproblem,bothsociallyandeconomically.Whiletheprevalenceandseverityofmyopiacontinueto increase,effectivetherapeuticinterventionsformyopiaremainlimited.Currently,managementofmyopiaislargelylimitedtotraditionalopticalcorrections‐spectacles,contactlenses,andrefractivesurgery ‐which correct distance visionbuthavenoeffect onmyopiaprogression.Whileslowedmyopia progression has been reported in clinical studies using the contact lens‐based,cornealreshapingtherapyandatropine,apharmaceuticalagent,theseapproachescomewith limitations and in the latter case, significant ocular side‐effects. Uncontrolledprogression may lead to high “degenerative” myopia, for which posterior scleralreinforcement surgery remains the only treatment option and a last resort directed atpreserving vision. Thus there is a clear need for new myopia control treatments.nderstandingmoreaboutthemolecularandcellularmechanismsunderlyingmyopiceyeU

1

growthhasthepotentialtouncovernoveltreatmentoptions.This dissertation presents results from three investigations into the role of the retinalpigment epithelium (RPE) in eye growth regulation, focusing on molecular and cellularmechanisms,andusingboth invivoanimalmodelsand invitrocellculturemodels. Inthefirstinvivostudy(Chapter2),weinvestigatedexpressionofcandidategenesinchickRPEof imposing short‐term optical defocus. Specifically, gene expression levels of the threebonemorphogeneticproteins(BMP‐2,4&7)wereexaminedafter2and48hoftreatment,withnegativeandpositive lensesused to imposedefocusofoppositesign.ThesegrowthfactorswereobservedtobedifferentiallyandbidirectionallyexpressedinRPE,expressiongenerally increasing with imposed myopia, which is associated with ocular growthinhibition.Becauseeyeshadlittlechancetochangetheirdimensionswithsuchshort‐termlens treatments, these genes are assumed toplay important roles in the onset and earlyphase of defocus‐induced ocular growth changes. For this reason, these genes representpotentialtargetsformolecular‐basedmyopiatreatments.Inthesecondstudy(Chapter3),high‐through gene expression profiling was employed to examine changes in gene

2

expressioninchickRPEwithlong‐termimposedhyperopicdefocus,whichresultedineyesbeing longer than normal and highly myopic. This DNA microarray screening revealedchanges in the expression of many genes, including BMPs, noggin (NOG), dopaminereceptorD4(DRD4),retinoicacidreceptor,beta(RARB),andretinalpigmentepithelium‐derived rhodopsin homolog (RRH). Some of these genes showed increased expressionwhile others showed decreased expression. It is plausible that somemay be linked theocular pathological complications seen inmyopia, while othersmay be linked to oculargrowth regulation, the imposedvisual conditions resulting in sustained, increasedoculargrowth.Thethirdandfinalstudy(Chapter4),addressedthepossibilityofanRPEsiteforthe anti‐myopia actionof apomorphine (APO), adopamine receptor agonist, observed inanimal studies. We further investigated the possibility that TGF‐β secretion from RPEmediatesthisinhibitorygrowtheffect.APOappliedtoculturedhumanfetalRPEcellswasfoundtoalterthesecretionofbothTGFβ1andTGFβ2,whichwasbiasedtowardsthebasal(choroidal)side.Thesegrowthfactorsalsoexhibitedconstitutivepolarizedsecretion,albeitbiased in theoppositedirection toAPO‐inducedparacrinesecretion.Theresults forAPOreconsistentwithitsobservedinhibitory(anti‐myopia)effects invivoandoffertheRPEaasapossiblesiteofaction.Insummary,theresearchreportedinthisdissertationprovidesevidencethatRPEplaysanimportantroleinpostnataleyegrowthregulation,(includingmyopicgrowth),asaconduitfor relaying growth modulatory retinal signals to choroid/sclera. Genes and moleculesidentifiedinthesestudiesofferpotentialdirectionsfornovelanti‐myopiatreatments,withtheRPEbeingapotentialtargetforthesame.

TABLEOFCONTENTS

i

TABLEOFCONTENTS

i

LISTOFFIGURES iiiLISTOFTABLES vLA

ISTOFABBREVIATIONS vii

CKNOWLEDGEMENTS x1

CHAPTER1:INTRODUCTION‐MYOPIADEVELOPMENTANDCONTROL

ABSTRACTOP A

11.1 MY IONIA R

REV

NDEYEGROWTH EGULAT

1.1.1 FICANCE2

P ETIOLOGY

ALENCEANDSIGNI

1.1.2 ALO L

OFMYOPIA1.1.3 CA

SCLE

1

EYE GROWTHREGULATION1.1.4 RE RALSIGNALCASCADESTINO‐

1..1.4.1 RETINA

1.1 .1.4.2 RPE

oroi1

.4 Ch

.3 d

.1 4.4 Sclera1.1.5 ANTI‐MYOPIATREATMENT

2334410131415

1.2 GENERALEXPERIMENTALAPPROACHESANDMETHODS 171.3 DISSERTATIONOUTLINE 22 CHAPTER2:BIDIRECTIONAL,OPTICALSIGN‐DEPENDENTREGULATIONOFBMPGENEXPRESSIONINCHICKRPEE

23

ABSTRACT 232.1 INTRODUCTION 252.2 MATER ALI ND

L

SA METHODS2.2.1

27ANIMA SANDLENSTREATMENTS

2.2.2 PROTEINSTUDIES SUESAMPLECOLLECTIONFORRNAAND2.2.3 UANTIFICATION

TIS NAR EVERSET

PURIFICATIONANDQ2.2.4 R ANSCR RIPTION2.2.5 NDVALIDATION IMERDE IGN

2.2.6 PR S A

R ESTERNBLOT

EAL‐TIMEPCR2.2.7 W

IMMUN

2.2.8 YOHISTOCHEMISTR

2.2.9 STATISTICALANALYSISULT

2.3.1

272728282829293031

2.3 RES S N

31R

AYIELD&QUALITY2.3.2 INNORMALCHICKS RNAEXPRESSIONOFBM DB PRECEPTORS2.3.3

M PSAN M

ROTEIN XPRESSIONOF IN ORMAL 2.3.4

P E BMP2 N CHICKS ROTEIN XPRESSIONOFBM LCHI

2.3.5 P E P4INNORMA CKS

PROTEINEXPRESSIONOFBMP7INNORMALOCULARTISSUES2.3.6 PROTEINLOCALIZATIONOFBMP2,BMP4,ANDBMP7INNORMALCHICKS

313133333637

2.3.7 CULAR IO D SMENSIONALCHANGESAFTERLEN TMENT2.3.8 7INRPE

TREAD OCUS‐INDUCED ENEEXPRE HANGESOFBMP2,BMP AND MP

2.3.9 YEF G SSIONC 4, BKINGEFFECTS FLE MEN SONBMPGEN EXPRESSIONINRPE

2.3.10 FTERLENSTREATMENTSO O NSTREAT T E

GENEEXPRESSIONCHANGESOFBMPRECEPTORSINRPEA2.3.11 VALIDATIONOFUSINGGAPDHASAHOUSEKEEPINGGENE

3739414141

2.4 DISCUSSION 44 CHAPTER3:LONG‐TERMIMPOSEDHYPEROPICDEFOCUSINDUCEDGENEEXPRESSIONHANGES CHICKRPE:AMICROARRAYSTUDYC

IN

50

ABSTRACT 503.1 INTRODUCTION 513.2 MATER DIALSAN ODS

ENSTRE TMENTS3.2.2 XTRACTION

METH

3.2.1 52

ANIMALSANDL A

R ISOLATION&

PE RNAE3.2.3 LYSE M A S

ICROARRAY NA

3.2.4 REAL‐TIMEPCR3.2.5 TATISTICALANALYSISS

UL

3

5252535354

3.3 RES TS U

55.3.1 OC CT

LAREFFE SOFLENSTREATMENT3.3.2 &CRNAQUANTIFICATIONR QUALITYANNA ALYSIS

3.3.3 LYSESMICROARRAYANA3.3.4 REAL‐TIMEPCR

5555568855

3.4 DISCUSSION

67CHAPTER4:APOMORPHINEREGULATESTGF‐Β1ANDTGF‐Β2SECRETIONINHUMANFETALRETINALPIGMENTEPITHELIALCELLS ABSTRACT 674.1 INTRO UCTIOD N

AL

684.2 MATERI

F RPE CSANDMETHODS

4.2.1 HUMAN ETAL ELL ULTUREHARACTERI TIONOF RPERECEPTORS

4.2.3 NE(APO)ONTGF‐Β1ANDTGF‐Β2SECRETIONFROMRPE

C 4.2.2 C ZA

EFFECTOFAPOMORPHI4.2.4 S ANALYSISTATISTICAL

S4.

6969697171

4.3 RESULTHARACTERIZA IO

723.1 REDH RPEC T OP FNOFD AMINEANDTG E EPTO TU

4.3.2 ‐ΒR C RSON CUL F

EXPRESSIONANDCONSTITUTIVESECRETIONOFTGF‐Β1ANDTGF‐Β24.3.3 APOMORPHINE‐INDUCEDALTERATIONSINTGF‐Β1ANDTGF‐Β2SECRETION

72747564.4 DISCUSSION 7

CHAPTER5:DISSERTATIONSUMMARYANDFUTUREDIRECTIONS 79 5.1 DISSERTATIONSUMMARY 795.2 FUTUREDIRECTIONS 80 BIBLIOGRAPHY 83

ii

LISTOFFIGUFigure1‐1.

RES

Schematicdiagramillustratingtheprincipalanatomicaldifferencebetweenanormalemmetropichumaneyeandamyopiceye,whichtypicalhasalongeraxiallength,largelyattributabletoalongervitreouschamber.

gramillustratingapossiblelocalretino‐scleralsignalpathwayopicgrowthchanges.

2

Figure1‐2. Schematicdiamediatingmy

Chickretina.

Emaybe

4

Figure1‐3. 5Figure1‐4. SchematicdiagramshowingpotentialmechanismsbywhichRP

involvedineyegrowthregulation.

Histologicalcross‐sectionoftheposterioreyewallofthechick.

ssertationresearchandthepatternsofmmetropic)chickeyes.

12

Figure1‐5. 14Figure1‐6. Monocularlenstreatmentsusedindi

opticaldefocusimposedonnormal(e

A‐scanultrasonographyofchickeye.

18

Figure1‐7.Figure2‐1.

Proteinsequencealignmentforhuman,mouse,chickBMP2,BMP4,&BMP7.

PE

1930

Figure2‐2.

Figure2‐3.

Resultsofelectrophoresisusinga1.2%agarosegelandEBstainingfor8RRNAsamplescheckedforRNAintegrity.

mRNAexpressionofBMP2,BMP4,BMP7,andBMPtypeIandIIreceptors

31

32(BMPR1A,BMPR1B,BMPR2)innormal(untreated)chickretina,RPE,andchoroid.

P2forbothnon‐reducingandFigure2‐4. WesternblotsshowingproteinexpressionofBMreducingconditions.

DiagramofBMPproproteinandmatureprotein.

34

Figure2‐5. 34

Figure2‐6. WesternblotsshowingproteinexpressionprofilesforBMP4inretina,RPE,andchoroidfromadolescentchicks,preparedunderreducingconditions.

P7inchickretina,

35

Figure2‐7. WesternblotsshowingproteinexpressionprofilesforBMRPE,andchoroid.

ImmunohistochestryforBMP2inchickposterioreyecup.

36

Figure2‐8.


3838Figure2‐9.


Effectsof+10Dand‐10Dlenstreatmentsonaxiallength(AL),vitreousf

3839


chamberdepth(VCD),andchoroidalthickness(CT)following2hand48holenswear,shownasinteroculardifferences.

andFigure2‐12. DifferentialexpressionofBMP2,BMP4,andBMP7mRNAinRPEafter248hofimposeddefocus.

40

Figure2‐13. BMP2,BMP4,andBMP7mRNAlevelsinRPEafter+10Dand‐10Dlenstreatmentsappliedfor2or48h.

lenstreatments

42

Figure2‐14. BMPreceptormRNAexpressioninRPEafter+10and‐10Dandineyesofuntreatedbirds.

ExpressionofGAPDHinRPEnormalizedtototalRNA(µg).

43

Figure2‐15.

.

43Figure2‐16

Figure3‐1.

CartoonsummarydiagramshowingBMPgeneexpressionchangesinRPEwithlenstreatments.

Interoculardifferencesinaxiallength,vitreouschamberdepth,andchoroidal

47

56

iii

thicknessafter18and38daysofcontinuous‐15Dlenstreatment.

essmentofsamplesusedinmicroarrayanalysis.Figure3‐2. Qualityass

57Figure3‐3. PlotofM(log2fold‐change)asafunctionofmeanexpressionlevelA(log2

intensity).

nes

57

Figure3‐4. Heatmapfrommicroarrayanalysisshowingresultsfor55candidategewithplausiblerolesinregulatingoculargrowthand/orRPEfunctions.

Comparisonofmicroarrayandreal‐timePCRgeneexpressionresults.

hfRPEwereculturedonECM‐coatedinsertsoftranswellsthatallowed

61


6269

isolationofmediumbathingapicalandbasalsidesofcells,andthustheirseparatesamplingandanalysis.

mRNAlevelsfordopaminereceptorsandTGF‐βreceptorsinculturedhfRPE.

tein.


ExpressioninculturedhfRPEoftheD2receptorproteinandTGFBR1pro

GFBR1.

7273

Figure4‐4.

ImmunocytochemistryofculturedhfRPEforD2receptorsandT

73

Figure4‐5.

TGF‐β1and‐β2mRNAexpressioninuntreatedculturedhfRPE.

74

Figure4‐6. ConstitutivepolarizedsecretionofTGF‐β1andTGF‐β2fromuntreatedculturedhfRPE.

74

Figure4‐7.

Figure4‐8.

Dose‐dependenteffectsofapically‐appliedAPOonsecretionbyculturedhfRPEofTGF‐β1andTGF‐β2.

amillustratingamodelforretinaldopamine‐regulatedeyegrowth.

76

78Diagr

iv

LISTOFTA

BLES

licatedinTable1‐1. Retinalcells,neurotransmittersandothermoleculesandgenesimpeitherorbotheyegrowthregulationandmyopia.

titativeestimatesofgeneexpression.

Equationsusedinsemi‐quan

5

Table1‐2.

Table2‐1.

MembersoftheBMPfamily.

21

26

Table2‐2.

PrimerinformationforBMPsandBMPreceptors.

lyieldpereyeforretina,RPE,and

28Table2‐3. RNAconcentration,A260/A280ratio,andtota

choroidsamples.

mRNAlevelsforBMPsandBMPreceptors.

ofBMP2,BMP4andBMP7fordifferent

31

Table2‐4. 33Table2‐5. Massescorrespondingtodifferentforms

species.

Primerinformationforcandidategenes.

Listofgenes,showingsignificanttreatment‐relateddifferentialexpressionchangesinRPEinmicroarrayanalyses,andwithplausiblelinkstoeyegrowth

35

Table3‐1.

Table3‐2.

54

59

regulationand/ormyopia.

Resultsofreal‐timePCRvalidationexperimentsforninegenesshowingdifferentiallyexpressioninmicroarrays;negativevaluessignifydown‐

gulation.

Table3‐3.

regulation&positvevalues,up‐re

TaqManqPCRassayinformation.

Dose‐dependenteffectsofapically‐appliedAPOonTGF‐βsecretionbyculturedhfRPE(24htreatment)

62

Table4‐1.able4‐2.

7075T

v

LISTOFABBRE

vi

VIATIONSaa

C

aminoacidACh

A2acetylcholine

ralpha2Cnase‐2

ADR adrenoceptoAHD2

A

aldehydedehydroge

ianceAL axiallength

rANOV analysisofvaAPO

apomorphine

AQP4 aquaporin4bFGF

P

basicfibroblastgrowthfactorhogeneticprotein

oteinreceptorBM bonemorpBMPR bonemorphogeneticprbp

1G

basepairs

a1GsubunitBSA bovineserumalbuminCACNA

Acalciumchannel,voltage‐dependent,Ttype,alph

entarydeoxyribonucleicacidpressingamacrinecell

cDN complemCGAC

conventionalglucagon‐ex

feraseCh cholineChAT cholineacetyltransCHO choroid

sitiveChT choroidthinknessCLCN

chloridechannel,voltage‐senmrotein

CNSABPA

centralnervoussystedingp

CR retinoicacidbincRN complementaryRNActF

thresholdcyclelthicknesstissuegrowthfactor

CT choroidaCTG connectiveD dioptersDA dopamine

ationandintegrateddiscoveryDAPI

D4',6‐diamidino‐2‐phenylindole

n,visualizDAVI databaseforannotatioDOPAC

3,4‐dihydroxyphenylaceticacid

DRD dopaminereceptorDDMEM

Dulbecco’smodifiedEaglemediumeicacid

DNATP

deoxyribonucldN deoxyribonucleotidetriphosphate

mplificationDTT dithiothreitolE efficiencyofprimeraEB ethidiumbromide

iniumionECM extracellularmatrixECMA

ethylcholinemustardaziridEGR1

B2

earlygrowthresponse1EDNR

endothelinreceptortypeBwthfactorreceptor2

arendothelialgrowthfactorD)FGFR fibroblastgroFIGF c‐fosinducedgrowthfactor(vascul

amacrinecellFN1GACGAG

fibronectin1glucagon‐expressingglycosaminoglycan

vii

GPCRsH

Gprotein‐coupledreceptorsgamacrinecellshosphatedehydrogenase

GACsPD

glucagon‐expressinaldehyde‐3‐p

layerGA glycerGCL ganglioncellh hourhf

humanfetaln

ltransferase1HH HamburgerandHamiltoHPRT1

Phypoxanthinephosphoribosy

peroxidasewthfactor‐1

HR horseradishIGF‐1

A

insulin‐likegroIL

B

interleukinINH inhibin,betaA

INL innernuclearlayer

innerpIPL lexiformlayerIOP intraocularpressure

eIIIreceptortyrosinekinase)IVT invitrotranscriptionKDR

Rs

kinaseinsertdomainreceptor(atypon‐expressingneuron

eceptorsLGEN largeglucagmACh muscarinicacetylcholinermm millimetre

MMP

matrixmetalloproteinasedexpression

onucleicacidMNE

Ameannormalize

mRN messengerribNOG

noggin

derrorNTS

Eneurotensin

caledstandarperature

NUS normalizedunsOCT optimalcuttingtemOD opticaldensity

erONL

‐k

outernuclearlayOPL outerplexiformlayerortho orthokeratologyPBS phosphatebufferedsalinePCR

A

polymerasechainreactionlphapolypeptideeceptor,alphapolypeptide

PDGFAFR

platelet‐derivedgrowthfactoradgrowthfactorr

torPDG platelet‐derivePDGF platelet‐derivedgrowthfacPM perfectmatchPNS peripheralnervoussystem

laction

QC qualitycontroqPCR

2

quantitativepolymerasechainre

se2RA retinoicacidRALDH

retinaldehydedehydrogenaRAR

B

retinoicacidreceptorptor,beta

agingRAR retinoicacidreceRMA robustmultiarrayaverRE refractiveerror

mberRIN RNAintegritynuRLERNARRH

relativelogexpressionribonucleicacidretinalpigmentepithelium‐derivedrhodopsinhomolog

RXR RetinoidXreceptorhelium

rRPE

retinalpigmentepit SCL‐C scleracartilaginouslaye

SCL‐F

sclerafibrouslayer

fmeandium

SEM standarderroromede

SFM

serum‐freeSP

C

signalpeptiSPAR osteonectin

hismSPP1 osteopontinSNP single‐nucleotidepolymorpSST somatostatinTBP TATAboxbindingprotein

cedproteinig‐h3TGF‐β

BItransforminggrowthfactor,beta

factor‐beta‐indufactorreceptor

TGF transforminggrowthhTGFBR

transforminggrowtTH

1

tyrosinehydroxylasepondin1

alloproteaseTHBS thrombosTIMP tissueinhibitorofmatrixmetTNC

tenascin

t‐PA tissueplasminogenactivatorepthlpolypeptide

VCDVIPZENK

vitreouschamberdvasoactiveintestinazincfingerprotein

viii

ix

ACKNOWLEDGEMENTS

Iwould like to thankmymentor,ProfessorWildsoet forherguidance that leadsme intothis interestingandexcitingmyopiaresearch,andforhercontinuousandstrongsupportformydissertationresearchandcareerdevelopmentoverthepastsixyears.IamgratefulformyPh.D.dissertationcommitteemembers,ProfessorsChristineF.Wildsoet,ProfessorMarlaFeller,andProfessorKunxinLuofortheiradvice,guidanceandlong‐lastingsupport.Iwould like to thankmyqualifyingcommitteemembers,ProfessorXiaohuaGongandLuChenfortheirguidanceandinspiringsuggestionsforresearchprojects.Iwouldalsoliketothankallofmyresearchcollaborators,Dr.SheldonS.Miller(NIH),ArvydasA.Maminishkis(NIH), Rong Li (NIH), Don Yuen (UC Berkeley), and Tan Truong (UC Berkeley), forproviding research insights, advice, assistance, research materials and sharingexperimental instruments. I would like to thank all members of the Wildsoet Lab forworkingtogetherasateamandcreatinganinvaluabletrainingexperiencethroughoutmygraduate studies. I would like to thank for the Vision Science Program at School ofOptometry, which has provided me with unique learning experiences through courses,seminars, and teaching experience. The mentoring and teaching experience withOptometrystudentsandundergraduatestudents filledmy lifewithsurprisesand leftmeithlovelymemories.Finally,Iwouldliketothankmywholefamilyfortheirunconditionaloveandsupport.wl

1

Chapter1

IntroductionMyopiaDevelopmentandControlAbstract

Myopia,ornear‐sightedness,isoneofthemostcommonoculardisordersinhuman.Duetothe increased prevalence and severity of myopia worldwide, especially in children andyoung adults, myopia has become a significant public health problem, both socially andeconomically. The ocular change underlying myopia is accelerated eye growth, whichresultsinamismatchbetweentheaxiallengthofeyeanditsrefractivepower,andinturn,blurredretinalimageswithoutopticalcorrection.Inhighmyopia,generallycategorizedas‐6.0Dorhigher,theassociatedstretchingofinternalocularstructureswiththeincreaseinaxial length may lead to blinding complications such as retinal degeneration, retinaldetachment, myopic maculopathy, choroidal neovascularization and glaucoma. Bothgeneticandenvironmentalfactorsarethoughttoplayrolesinthedevelopmentofhumanmyopia. Studies using animalmodels have provided convincing evidence for the role ofvisual environmental factors, withmanipulations of the visual input, eitherwith opticaldefocusinglensesordiffusers,producingconsistentabnormalitiesintheemmetropizationprocess.Sincelocalizedmanipulationofretinalimagesinducelocalizedeyeelongation,andoptic nerve section does not prevent the development ofmyopia, it has been concludedthatpostnataleyegrowthiscontrolledlocallywithintheeyeitself.TheRPEisamonolayerofcellspositionedbetweentheneuronalretinaandthechoroid. It isacomponentof theblood‐retinabarrier,withalreadyestablishedrolesinmaintainingnormalretinalfunction.However,basedonitslocation,itisplacedtoalsoplayacriticalroleintheregulationofeyegrowth. Identificationof genes showingdifferential regulation inRPEduring altered eyegrowthwouldprovideindirectevidenceforsucharoleandcouldalsoprovidenewinsightsintothemolecularandcellularmechanismsunderlyingmyopiadevelopment.Furthermore,evidence linkingspecificgeneswithgrowthmodulationwouldopenup thepossibilityofdevelopingnovelanti‐myopiatreatmentstargetingsuchgenesintheRPE,forwhichthereisprecedenceintheformofRPE‐targetedtreatmentsofsomeretinaldiseases.

2

1.1 MyopiaandEyeGrowthRegulation1.1.1 PrevalenceandsignificanceMyopia,ornear‐sightedness,describestheconditioninwhichimagesofdistantobjectsarefocused in frontof theretina, resulting inblurredvision.Myopiareflects themismatchesbetween the refracting power of the eye, to which the cornea and crystalline lenscontribute, and its optical axial length, which composes anterior chamber depth, lensthickness,andvitrealchamberdepth.Mostmyopiaisaxialratherthanrefractiveinnature,the product of excessive elongation of the vitreous chamber (Figure 1‐1).1,2 Babies aretypically born with refractive errors, which are corrected during early developmentthroughaprocessof coordinatedoculargrowthknownasemmetropization.3,4However,myopiamorecommonlyoccursinchildhoodasafailureofemmetropization,whentheeyecontinuestoelongateafteremmetropiaisachieved.Excessiveocularelongationresultsinhigh myopia (classically defined as spherical equivalent refractive errors equal to orgreaterthan‐6.0D),whichcarriesahighriskofsight‐threateningcomplicationssuchasretinal degeneration, retinal detachment, choroidal neovascularization, myopicmaculopathy,cataract,andglaucoma.5‐7

Figure1‐1.Schematicdiagramillustratingtheprincipalanatomicaldifferencebetweenanormalemmetropic human eye and a myopic eye, which typical has a longer axial length, largelyattributabletoalongervitreouschamber.

3

Myopiaisoneofthemostcommonrefractiveerrors,whichareoneoftheworld’sleadingcauses of functional blindness due to lack of access to optical corrections; they are alsosignificantcontributors to theglobalburdenofeyedisease.8 Ina recentpublishedstudy,theoverallprevalenceofmyopiaintheUSwasgivenas41.6%forpersonsaged12to54years.9 However, in many Asia countries and populations, myopia has now reachedepidemic levels.10,11Due to the increased prevalence and severity ofmyopiaworldwide,especially in children and young adults, myopia has become a significant public healthproblem,bothsociallyandeconomically.8,9,12‐15Theincreasingprevalenceandseverityofmyopia has stimulated increased research into the underlyingmechanisms, an essentialstepindevelopingeffectivetherapies.1.1.2 AetiologyofmyopiaItisnowgenerallyacceptedthatbothgeneticandenvironmentalfactorsplayrolesinthedevelopment ofmyopia.7,16‐18 Genetic studies ofmyopia using linkage and genome‐wideassociationapproacheshave identifiedmany loci (MYP1–MYP17)andgenes althoughnouniquegeneforthemostcommonformofjuvenilemyopiahasemerged.19‐21Human epidemiological studies and studies using animal models have provided strongevidence for the environmental contributions to myopia development. Near work,22educational levels,23 life‐styles,24 and outdoor activities11, 25 are among the factorsidentified to influence refractive errors in human clinical studies. Animal studies usingchickens, guinea pigs, tree shrews, and monkeys have provided the most convincingevidenceforvisualenvironmentalinfluencesonpostnataleyegrowthregulationandthusmyopiadevelopment.16,26‐30Currentlytherearetwodifferentexperimentalparadigmsforinducing myopia in animal studies, utilizing form depriving diffusers and negativedefocusinglenses,respectively.Bothparadigmshavebeenshowntoelicitrobustresponsesinmostanimalmodels. Inall cases, the inducedmyopia ischaracterizedbyan increasedrateofaxialelongation,attributabletoexpansionoftheposteriorvitreouschamberoftheeye,andreflectingalteredgrowthoftheouterchoroidalandscleralsupportlayers.In the case of human myopia, myopia likely reflects interactions between genetic andenvironmental factors; how each factor contributes to the development and progress ofmyopiaisthesubjectofmuchon‐goingdebate.16,17Itisalsopossiblethatalargenumberofgenes, each contributing a small effect, may prove important in determining thedevelopmentofmyopia,whichisaetiologicallyheterogeneousinhumanpopulations.31‐331.1.3 LocaleyegrowthregulationMyopia development is characterized by accelerated eye growth, which can be inducedexperimentally by visualmanipulations such as form‐deprivation and optical defocus, asalreadynoted.Thepreservationofthesealteredgrowthpatternsineyesisolatedfromthebrain by optic nerve section points to local ocular growth regulation.34‐36 That localizedmanipulation of retinal images induces localized ocular changes, confined to the areaunderlyingtheaffectedretina,isinterpretedasfurtherevidencethatpostnataleyegrowth

4

is controlled locally.37,38 Such local regulatorymechanismsmust necessarily involve theretina, choroid, and sclera (retino‐sclera signal cascades, Figure 1‐2). Identifying genesshowingdifferentialregulationinretina,choroidandscleraduringalteredeyegrowthhasbeen one commonly used approach to investigate the ocular growth modulatorypathways.16,26,39‐41Intheresearchdescribedinthisdissertation,wehavetargetedtheRPEforsimilarstudiesonthebasisofitscriticallocationbetweentheretinaandchoroid.

Figure1‐2.Schematicdiagramillustratingapossiblelocalretino‐scleralsignalpathwaymediatingmyopicgrowthchanges;inthemodelshown,thesignalisinitiatedintheretinaandtransducedinturnintoanRPEsignal,choroidalsignalandscleralsignal.

1.1.4 Retino‐scleralsignalcascades1.1.4.1 RetinaTheexperimental animalmodelsofmyopia inducedby retinal imagedegradation (form‐deprivation and optical defocus), combined with evidence of local control point to theretinaasasourceofmyopia‐generatingsignals.16Apartfrominputfromthecentralretina,recentresearchpointstoanimportantrolefortheperipheralretinainregulatingpostnataleyegrowth.Forexample,studiesinmonkeyhaveshownthatthefoveaisnotessentialforeither normal refractive development (emmetropization), or the development of form‐deprivation‐inducedmyopia,42andinchicks,myopicdefocusimposedonperipheralretinausing2‐zonelensesresultsinslowedaxialmyopiaprogression.43,44Interestingly,arecentstudyofretinalfunctioninmyopichumaneyesusingmultifocalelectroretinography(ERG)documented impaired function in the paracentral and peripheral retina and not in thecentral retina.45At least for inducingmyopia experimentally in chicks, high visual acuitydoesnotappeartobeessential,asRitcheyetal.,haveshownrecentlythatexcessiveocularelongationcanbeinducedbybothform‐deprivationandimposedhyperopicdefocususing

5

genetically mutant chickens with poor visual acuity.46 Nonetheless, changes in theresponses to such experimental manipulations have been reported with retinalneurotoxins, providing furtherevidence for the roleof the retinaas the sourceof oculargrowthregulatingsignals.47,48Since emmetropization and eye growth regulation is visually guided and thus the retinamust necessarily play a key role, it has also been extensively investigated using visualmanipulations combined with molecular and cellular approaches. The following sectionfocusesonwhat isknownabouttheroleofretinalamacrinecells,neurotransmittersandothermolecules,andgeneandproteinexpressioninmyopiadevelopment.As a part of the central nervous system, the retinaprimarilyconsistsofthreelayersofnervecellbodies(outernuclearlayer[ONL],innernuclearlayer[INL],and ganglion cell layer [GCL]) and two layers ofsynapses (inner plexiform layer [IPL] and outerplexiformlayer[OPL];Figure1‐3).Theouternuclearlayercontainscellbodiesoftherodsandcones(thephotoreceptors),theinnernuclearlayercontainscellbodiesof thebipolar,horizontalandamacrine cells,and the ganglion cell layer contains cell bodies ofganglion cells and displaced amacrine cells. Theouter segments of thephotoreceptors lie outermostin the retina,against the retinalpigmentepithelium(RPE). The retina contains numerousneurotransmitters, including glutamate, dopamine,acetylcholine, and serotonin,whichmediate normalretinalfunctionsandmayalsohaverolesinpostnataleyegrowthregulation(Table1‐1).

Table1‐1.Retinalcells,neurotransmittersandothermoleculesandgenesimplicatedineitherorbotheyegrowthregulationandmyopia

Cells NeurotransmittersandMolecules

GeneandProteinsExpression

Conventionalglucagon‐expressingamacrinecells(CGACs)Largeglucagon‐expressingneurons(LGENs):bullwhipandmini‐bullwhipPhotoreceptors

AcetylcholinebFGFDopamineGlucagonInsulinandIGF‐1RetinoicacidTGF‐βVIP

AHD2BMP2CTGFEDNRBIL‐18NOGNTSTGF‐βVIPZENK

Figure 1‐3. Chick retina. Layers of retinaare shown and labeled as GCL, IPL, INL,OPL,andONL.Theoutersegmentsofthephotoreceptors lie outermost in theretina,againsttheRPE.

6

1.1.4.1.1 Amacrinecells,bullwhipcells,andmini‐bullwhipcellsAmacrinecellsareahighlyvariedretinalcellclass,whichincludesatleast24differentcelltypesinthemonkeyandhumanretina.49,50Theyareclassifiedintodifferenttypesbasedonmorphologicalcharacteristicsofdendritic treesize(small‐,medium‐,and large‐field)andthestratificationoftheirdendritesintheIPL.Theymayalsobeclassifiedaccordingtotheirneurotransmitter and protein‐expression profiles; thus many cell types have beendescribed, including glucagon‐expressing amacrine cells (GACs), and amacrine cellsshowing immunoreactivity to retinoic acid binding protein (CRABP), cholineacetyltransferase (CHAT), somatostatin (SST), tyrosine hydroxylase (TH), and vasoactiveintestinal polypeptide (VIP), etc.51 Although the roles of many amacrine cells in retinalphysiologyremainpoorlyunderstood,ofrelevancetothisdissertationresearcharestudiessuggestingtheirinvolvementintheretinalprocessingofcontrastinformationandgrowth‐modulatingvisualstimuli.47,52Conventional glucagon‐expressing amacrine cells (CGACs) have been most extensivelystudiedinthecontextofeyegrowthregulation.Thesecellsarereportedtoshowchangesinthe transcription factor zinc finger protein, ZENK (otherwise known as early growthresponse 1, EGR1), in response to form‐deprivation and lens‐induced defocus in youngchicks.51 Specifically, Fischer et al., reported that two hours of form‐deprivationsignificantlyreducedZENKproteinexpressioninglucagon‐expressingamacrinecells.Withlens treatments that imposed optical defocus, the number of ZENK‐positive glucagonamacrinecellswasfoundtobeincreasedwithpositivelenses,afteraslittleas30minutesof wear, and decreased after 2 hours of negative lens wear. Bitzer and Schaeffel laterundertook a closely related study, and found thenumberof ZENK‐expressing cells tobeincreased with positive lenses and reduced with negative lenses, after as little as 40minutes of exposure to the lenses,with the changes lasting at least 2 hours.53 The sametrends in termsof the total number of glucagon amacrine cells expressing ZENK in eyestreatedwithpositive andnegative lenseswereobservedunder reduced illuminanceandmonochromatic light.53 A related immunohistochemistry study showed the CGACs to beasymmetrically distributed across the retina, with highest density in a central zonebetweenthenasalandtemporaledgesoftheretina.54Otherfollow‐upstudieshavefurtherinvestigatedtheroleofglucagon inmyopiadevelopmentandrelatedchangesZENKgeneexpression using pharmacological manipulations. We will discuss these studies in aseparatesectionundertheneurotransmittersubheading,glucagon.Anothertwotypesofamacrinecellsintheavianretina,labeledbullwhipandmini‐bullwhipcells, are also glucagon‐expressing neurons (LGENs). Their roles in ocular growthregulation have been the subject of more recent investigations.54 Bullwhip cells sharephenotypic features with both amacrine and ganglion cells, including morphologicalfeaturesandproteinexpression.Inthechickretina,bullwhipcellsarefoundonlyinventraland mid‐peripheral regions while mini‐bullwhip cells are found only in dorsal and far‐peripheral regions. Both bullwhip and mini‐bullwhip cells has been implicated in theregulation of eye size and shape in chick studies.55 For example, while eliminating themajorityofbullwhipandmini‐bullwhipcellsbyintravitrealinjectionofcolchicineintothevitreous chambers of young chick eyes did not change the axial length, it resulted in

7

expansionofequatorialdiameter,and the lattereffectcouldbepreventedby intravitrealinjectionsofglucagon.Visualmanipulationwithform‐deprivationfor24hoursalsocausedasignificantreductioninthepercentageofbullwhipcellsexpressingZENK(Egr1),while2hoursofunrestrictedvisionfollowing5daysofform‐deprivationleadtoanincreaseinthenumber of ZENK‐positive bullwhip cells as compared to control eyes. Together theseresults suggest that bullwhip and mini‐bullwhip cells may play important roles inregulatingtheequatorialdimensionsofeyes,withthepeptide,glucagon,playingakeyroleasagrowthinhibitor.1.1.4.1.2 Neurotransmitter:glucagonGlucagon,whichisa29‐amino‐acidlongpeptide,isoneoftheneuropeptidesreleasedfromZENK‐expressing retina amacrine cells in the chick and so likely represents a keymessengerforoculargrowthmodulationalthoughthedownstreamelementsofthisZENKpathwayarestillnotverywellunderstood.53,56In studies aimed at understanding the role of glucagon in eye growth regulation,Feldkaemper etal., using Northern blot analysis, reported that proglucagon mRNA wasincreasedinchickretinaaftertreatmentwithpositivelenses.57Theretinalgeneexpressionlevels of glucagon and its receptor have also been investigated using semi‐quantitativereal‐timePCRusingthechickmodel,forform‐deprivationaswellaslens‐inducedmyopiaandhyperopiaparadigms.58,59Form‐deprivationandnegativelensesinducedasignificantdown‐regulation of glucagonmRNA expression comparing to contralateral control eyes,whilepositive lens resulted inup‐regulationof glucagonmRNAexpression, although thelatter change did not reach statistical significance. Glucagon receptor mRNA expressionwas also not different between lens‐treated and contralateral control eyes, although atransient increase in glucagon receptor mRNA expression was observed in lens‐treatedeyes.Afollow‐upstudyfromFeldkaemperandSchaeffelshowedretinalglucagonproteinlevels tobedecreasedwithnegative lenses,whilepositive lenseshadnoeffectonretinalglucagon protein levels, even though the latter treatment reliably inhibits eye growth.60Currently, there is no direct evidence linking the changes of glucagon mRNA geneexpressionwithproteinlevelsinretina.Theonlysmallpopulationofretinalneuronsthatsynthesize and release glucagon, the possibility that there are other sources of glucagonsynthesis and secretion (e.g. RPE), and the complicated metabolism of glucagon, allrepresentobstaclestodesigningconclusivestudies.54,58,60,61Pharmacological studies using glucagon analogs represent a more direct approach tostudying its role in eye growth regulation. In one such study, both glucagonor glucagonreceptoragonistswereobservedto inhibitmyopiadevelopment, inducedbyeither form‐deprivation or negative lenses.60, 62, 63 Certain concentrations of intravitreally‐injectedglucagon also slowed the growth of otherwise untreated eyes.63 Surprisingly, for eyestreatedwith positive lens, both glucagon and a glucagon antagonist inhibited hyperopiadevelopment.60,62,63 These apparently paradoxical results for glucagon injected, positivelens‐treated eyes may reflect interactions between endogenous glucagon levels, theinjectedglucagonconcentrations,and/orthenumberandsensitivityofglucagonreceptors.These studies also focused on changes in axial ocular dimensions and refractive errors.

8

Thusitisperhapsalsoofrelevancethatinalaterstudy,Fischeretal.,foundthatexogenousglucagon primarily decreased equatorial eye growth, and conversely, glucagon receptorantagonists caused excessive equatorial growth. In summary, based on reportedexperimentalevidence,glucagonappearstoserveasastopsignalforeyegrowth,althoughtheunderlyingsignalpathwayandmechanismsappearcomplicated.1.1.4.1.3 Neurotransmitter:dopamine

Dopamine(DA)isamonoamineneurotransmitter,whichissynthesizedfromL‐tyrosinebya subset of amacrine cells. One of its principalmetabolites is 3,4‐dihydroxyphenylaceticacid (DOPAC).DA receptorshavebeen identifiedonbothneural cellswithin the retina64and RPE.65‐68 There are five subtypes of DA receptors (D1‐D5), and based on theirbiochemicalandpharmacologicalpropertiestheyhavebeencategorizedintoD1‐like(D1,D5),andD2‐like(D2‐D4)subfamilies.69,70In the retina, dopamine plays major roles in light adaptive processes and contrastsensitivity.71 In myopia‐related research, retinal dopamine has also been shown to beassociatedwitheyegrowthregulation.Specifically,animalstudieshavedemonstratedthatretinal DA concentration, retinal DA synthesis rate, and concentrations of retinal andvitreal DOPAC, the latter serving as indices of retinal dopamine release, show changesrelated to altered eye growth, induced with form‐deprivation, negative or positive lenstreatmentsorevenchangesintheilluminanceusedinrearing.72‐77Nonetheless, inoneoffew studies to challenge the role of DA in eye growth regulation, Luft et al., reportedchanges intheretinaldopaminergicsysteminresponsetoalteredlight intensity,butnotspatialfrequency,withshort‐term(2hours)treatment.78DirectevidenceforaroleofDAineyegrowthregulationisprovidedbyobservationsthatocular administration of apomorphine, a nonselective DA receptor agonist, inhibits bothform‐deprivation and lens‐induced myopia, in both chick and monkey models.74, 79‐84Furthermore, co‐administration of a DA receptor selective antagonist along withapomorphine,preventsthelatterinhibitoryeffectsoneyeelongationinchicks,74,79andD2‐selective antagonists alone enhance form‐deprivation myopia in chicks.85 These datatogether provide convincing evidence for a role of retinal dopamine in eye growthregulation.1.1.4.1.4 Neurotransmitter:acetylcholineAcetylcholine (Ach), another classic neurotransmitter, is widely distributed throughoutboth the peripheral nervous system (PNS) and central nervous system (CNS).86 In theretina, cholinergic amacrine cells have been identified in both INL and GCL.87‐91 Achactivates two types of integral membrane receptors, ionotropic nicotinic acetylcholinereceptors (nAChR) and metabotropic muscarinic acetylcholine receptors (mAChRs).92‐97ThereareatleastfivesubtypesofmAChRs(M1‐M5),whicharewidelydistributedthroughoutoculartissues;thusinchicksandtreeshrews,mAChRsubtypeshavebeenlocalizedtoretina, RPE, choroid, and ciliary body.98‐102 Retinal cholinergic mechanisms have been

9

implicatedinmanyimportantaspectsofvisualprocessing,includingmotiondetectionandlightadaptation.86,103The observed potent inhibitory effects on eye elongation of muscarinic receptorantagonistssuchasatropineareamongevidenceimplicatingacetylcholineinthecontrolofocular growth.104‐106 Intravtireal administration of selective mAChR antagonists furthersuggest that suchanti‐myopia effects aremediatedbyM1and/orM4 receptor subtypes.However,theexactocularsiteofactionofmAChRantagonistsremainsunresolvedduetothewidedistributionofmAChRsandthevariableeffectsofdifferentmuscarinicreceptorantagonists onmyopia development.105,107‐109 Arguing against a retinal site of action areresults from investigations of form‐deprived chick and tree shrew eyes, which did notreveal any treatment‐related differences in the retinal content of Ach or choline, itsmetabolite, or choline acetyltransferase (ChAT), the Ach synthesizing enzyme.110, 111Furthermore,thedepletionofthemajorityofcholinergicamacrinecellsinthechickretinausingECMA,aneurotoxin,didnotpreventthedevelopmentofmyopiainresponsetoform‐deprivation.112 These results thus call into question the role of retinal Ach and retinalcholinergiccellsinmyopiadevelopmentandraisethepossibilitythatmAChRantagonistsmay act through other ocular tissue sites such as RPE, choroid, and sclera rather thanretinaorthatothernoncholinergicmechanismsmaybeinvolved.40,98,101,113,1141.1.4.1.5 OthermoleculesandgeneexpressionregulationRetinoic acid (RA) is known to be a potent regulator of cell growth and differentiation,acting via nuclear receptors (RARs and RXRs), to activate or repress the expression ofgenes.115,116RAalsohasbeenputforwardasapotentialeyegrowthregulator.117,118Thereare extensive data linking RA with ocular growth regulation in a number of differentspecies, although there are also subtle species‐relateddifferences. In chicks,RA levels inbothretinaandchoroidexhibitbidirectionalchangeswithalteredeyegrowth.Inthecaseofbothform‐deprivationornegativelenstreatment,retinalRAlevelsareincreased.117,119‐122Inaddition,negativelenstreatmentsup‐regulateretinalmRNAexpressionofaldehydedehydrogenase‐2(AHD2),oneoftheenzymesinvolvedinRAsynthesis.123Inmonkey,therateofretinalRAsynthesiswasfoundtobepositivelycorrelatedwitheyeelonagetion.124Otherneurotransmittersormoleculesthathavebeensuggestedtoplayrolesineyegrowthregulation and/or myopia development include vasoactive intestinal polypeptide,125‐127somatostatin, insulin,63, 128, 129 insulin‐like growth factor‐1 (IGF‐1),63 basic fibroblastgrowthfactor(bFGF),130transforminggrowthfactor‐β(TGF‐β).130ZENK is one of themost extensively studied genes in the context ofmyopia. As alreadynoted(sectiononretinalamacrinecells),earlystudies foundZENKproteinexpression inchick retina to be altered by visual manipulations known to affect eye growth.51,53,131Interestingly,morerecentstudieshavereportedfortwodrugsknowntoinhibitmyopia,amuscariniccholinergicantagonistandadopamineagonist,rapidincreasesinZENKmRNAexpression in the retinas of form‐deprived chicks, although previously identified ZENKexpression glucagonergic amacrine cells only comprised a small portion of retinal cells

10

affected,132andanotherresearcherreporteddisparateresults inrelationto theeffectsofmuscarinicantagonistsonZENKexpressioninthechickretina.132Recent studies combining high‐throughput microarray and semi‐quantitative PCR haveidentifiedanumberofretinalorretina/RPEgeneswithpotentiallyimportantrolesineyegrowth regulation.127, 133, 134 Listed genes include BMP2, endothelin receptor type B(EDNRB), interleukin‐18 (IL‐18), connective tissue growth factor (CTGF), neurotensin(NTS),andnoggin(NOG).Thesesresultsprovideadditionalinsightintopossiblemoleculesinvolvedintheregulationofoculargrowth,andalsoreinforcethecomplexnatureofoculargrowthregulation.1.1.4.2 RPETheretinalpigmentepithelium(RPE)isauniquetissue,locatedbetweentheneuroretinaand choroid/sclera. Arrayed in a monolayer of cells, linked by tight junctions, the RPErepresentsacriticalcomponentoftheretina‐bloodbarrier,preventingthefreeexchangeofmolecules and ions between the retina and choroid. Thus, theRPE serves to protect thehealthand integrityof theneural retinaandmaintainofphysiologicalhomoestasis.135‐137Byvirtueofitsanatomicallocation,theRPEisalsoideallysuitedtorelaypresumedgrowthmodulatorysignalsthatoriginateintheretina,tothechoroidandsclera,whereremodelingandothergrowthprocessesoccurtoachievechangesineyesize.16,40TheRPEpossessesarich array of receptors, including ones for neurotransmitters already implicated in eyegrowthregulation.40,98,138,139TheRPEisalsoknowntobeanimportantsourceofgrowthfactors that may have autocrine and/or paracrine functions.135, 140 Of relevance to theresearchreportedinthisdissertation,differentialgeneexpressionaswellasmorphologicalchangesinRPEhasbeendocumentedduringthedevelopmentofmyopiaandhyperopia.141‐1451.1.4.2.1 NeurotransmitterreceptorsandretinoicacidreceptorsGlucagonreceptors:GlucagonreceptorsbelongtotheGprotein‐coupledfamilyofreceptors(GPCRs).146 Stimulation of the GPCR results in activation of adenylate cyclase and theregulationofintracellularcAMPlevels.147RPEisbothapossiblesourceofglucagonandatissue site of action as it also expresses glucagon receptors.58 Studies have shown theapplication of glucagon to embryonic chickRPE to increase intracellular cyclic AMP andadenylatecyclaseactivity.148Dopaminereceptors:ThereisevidencelinkingD2receptorstoinhibitoryeyegrowtheffectsofdopamine,74,79andinchicks,dopamineD2/3receptorshavebeenidentifiedonthebasalsideoftheRPE,withtheirpresenceontheapicalsurfaceofRPEcellsbeingleftunresolvedduetotheheavypigmentationinthisregion.65mRNAexpressionalsohasbeendetectedforbothD2andD4receptorsinchickRPE.65,143D2receptorshavealsobeendemonstratedonthe RPE ofmammals, including cat and cow,67,68while their presence on human RPE iscontroversial.149,150Nonetheless,culturedhumanfetalRPE(hfRPE)arereportedtopossesD2receptors.151Invitroelectrophysiologicalstudiesusingchickretina‐RPE‐choroidtissuepreparations suggest the presence of DA receptors on both apical and basolateral

http://en.wikipedia.org/wiki/Adenylate_cyclase�

http://en.wikipedia.org/wiki/Intracellular�

http://en.wikipedia.org/wiki/Cyclic_adenosine_monophosphate�

11

membranes, which result in altered RPE membrane potentials when pharmacologicallymanipulated.152 At least some of these DA effects on RPE can be abolished using a Cl‐channel blocker, suggested thatDAmay act through a signalingpathway involvingbasalmembraneCl‐ channels. Intriguingly, application of dopamine to an invitro co‐culture ofRPE with scleral chondrocytes showed an effect on the proliferation of scleralchondrocytesthatwasnotseenwhendopaminewasappliedtochondrocytesculturedinisolation.153Muscarinicacetylcholine receptors (mAChRs):Since mAChR antagonists such as atropineandpirenzepinehavebeenshowntobeamongthemostpotentanti‐myopiaagents,muchresearch efforts havebeen focused on the characterization ofmAChR in different oculartissues.154 However, as noted previously, the target tissues for the above mAChRantagonistsremainunresolved,atleastinpartduetothewideexpressionofmAChRintheeye.98,102ExpressionofmAChRhasbeendemonstratedintheRPEofanumberofdifferentspecies,includingchick,forwhichM2,M3,andM4proteinshavebeenidentifiedinRPE.98ActivationofmAChreceptorsinculturedhumanRPEleadstoincreasedphosphoinositideturnover and increases in intracellular calcium.155 Other pharmacological studies haveimplicated mAChRs in RPE metabolism in humans and rats.156, 157Retinoicacidreceptor:RetinalandchoroidalRAconcentrationsshowbidirectionalchangesunderconditionscorrespondingtoacceleratedorslowedeyegrowthrespectively.117,118,120SinceRPEis interposedbetweentheretinaandchoroid, itmaypotentiallybeaffectedbyeitherorbothchangesinretinalRAandchoroidalRA,makinginterpretationofstudiesintothe role of the RPE in these events very difficult.135, 158, 159 Nonetheless, the RPE doespossessRAreceptorsanddown‐regulationofRAR‐βmRNAinRPEhasbeenreportedwithlens‐inducedmyopia.143In addition to the neurotransmitters and receptors already discussed above, RPE is alsoreportedtopossessanumberofothertypesofreceptorsincludingthoseforinsulin,IGF‐I,160,161 and VIP,148 with application of VIP to cultured RPE cells reported to stimulatepolarizedsecretionofmacromolecules.1621.1.4.2.2 IonchannelsandfluidtransportAs noted above, RPE cells are interconnected by tight junctions,which prevent the free,transepithelialmovementofionsandwaterbetweentheretinaandchoroid.ThusionandwaterchannelsonRPEplayimportantrolesinfluidtransportacrossRPEandmaintaininghomeostasis for adjacent tissues, the retina and choroid.40 The RPE is also polarized,reflecting the asymmetric arrangements of these channels on the apical and basalmembranesofRPEcells.Inthecontextofmyopiaresearch,therolesofRPEonionandfluidchannelsarenotwellunderstood. Potassiumand chloride channels are known to regulate transepithelial fluidtransportwhilevoltage‐dependentcalciumchannelsactasregulatorsofsecretoryactivity.It is possible that some neurotransmitters, including dopamine, modulate eye growththrough effects on these ion channels.40,136 The observed rapid thickness changes in the

12

choroidofthechickinresponsetomyopia‐andhyperoia‐inducingstimulimayalsoreflect,atleastinpart,transepithelialfluidtransportbetweenthesubretinalspaceandchoroid.16,35Thevolumeandthicknesschangesinthechoroidarediscussedinmoredetailinthenextsection.1.1.4.2.3 GrowthfactorsandcytokinesTheRPErepresentsamajorsourceofgrowthfactorsandcytokines,includingIGF‐I,163,164TGF‐βs,165,166 FGFs,167,168 MMPs,169 and TIMP.170 Synthesized locally and subsequentlysecreted, they have been attributed roles in the maintenances of the structure andhomeostasisofretinaandchoroid,135andsomehavealsobeenimplicatedinpostnataleyegrowth regulation. Depending on the direction of their secretion, i.e., towards the retinaand/or choroid, the growth factors synthesized by RPE have potential to affect themorphologicalandfunctionalchangesofretinaand/orchoroidandsclera.Excepting the research reported in thisdissertation, to‐date, therehasbeenonly limitedstudy of the roles of RPE and RPE‐derived growth factors in postnatal eye growthregulation,140, 141, 143, 151, 171 and in studies involving the chick model, gene expressionstudieshavebeenlimitedtoretina/RPEcombinedtissuesratherthanRPEinisolation.133,134 This dissertation researchwas dedicated to the investigation of role of RPE itself inmyopia development, with studies into the roles of RPE‐derived growth factors andcytokines in the retino‐scleral signaling cascades, usingboth invivo chick animalmodelsand in vitro human fetal RPE culture models. A schematic diagram showing potentialmechanismsbywhichtheRPEmightbeinvolvedintheeyegrowthregulationisprovidedinFigure1‐4.

Figure1‐4.SchematicdiagramshowingpotentialmechanismsbywhichRPEmaybeinvolvedineyegrowthregulation.

(ModifiedbasedonRymerandWildsoet.VisNeurosci.2005.22;251–261.)

13

1.1.4.3 ChoroidBetween the RPE and sclera lies the choroid, which is a richly vascular structure, withresidentmelanocytes, fibroblasts and immunocompetent cells, supported by collagenousandelasticelements.41 Itsmajor functions includesupplyingoxygenandnutrients to theouter retina, light absorption (pigmented choroid), thermoregulation, andmodulation ofintraocular pressure (IOP). More recently, it has also been recognized that the choroidplaysacriticalroleinemmetropization,throughearlypostnataleyegrowthregulation.41Histoloigcally, the choroid is divided into five layers starting from the inner retinal side:Bruch'smembrane, choroicapillaris, two vascular layers (Haller's and Sattler's), and thesuprachoroidea.41 In birds, the suprachoroid contains large, endothelium‐lined spaces(lacunae),whichresemblelymphaticvessels.172,173In experimental myopia studies involving the chickmyopia, it has been shown that theproteincontentofthesuprachoroidalfluiddecreasesinform‐deprivedeyesandincreasesafter vision is restored.174 Bidirectional changes in the permeability of the choroidalvasculature also have been reported with form‐deprivation and recovery from form‐deprivationmyopia in the chick.174‐176 These changes are linked to thickness changes ofchick choroid – thinning in eyes with form‐deprivation and lens‐induced myopia, andthickening in eyes recovering from form‐deprivation myopia and with lens‐inducedhyperopia.16,35,140,177Thesechangesinthicknesscanbeverydramatic.Forexample,undernormalvisualconditions,thechoroidofyoungchicksisabout250µmthickinthecentral,axialregionand100µmintheperiphery.177Withimposedmyopicdefocus,thechoroidcanincrease its thickness by asmuch as 1mm (>17 D), the effect being to push the retinatowardsthealteredimageplane,therebycompensatingfortheimposedfocusingerror.177Incontrast,withimposedhyperopicdefocus,thechoroidthinstopulltheretinabackwardsinthedirectionofthescleratowardsthealteredimageplane.Likewise,form‐deprivationalso induceschoroidal thinning.Thesechanges inchoroidal thicknessoccurveryrapidly,beingdetectiblewithhighfrequencyultrasonographyamatterof1‐2hours.178‐180In addition to serving as a focusing mechanism as accomplished by moving the retinaforward and backward to match the image focal plane, the choroid may also play animportantroleinregulatingscleralgrowthandremodelling.41Ofpotentialrelevancehereareobservationsthatthechoroidexpressesandsynthesizesavarietyofgrowthfactorsandenzymes,includingbFGF,181MMPs,182tissueplasminogenactivator(t‐PA),183andTGF‐β.131,184,185 For example, during thedevelopment ofmyopia, TGF‐β gene expressionhas beenshowndifferentiallyexpressed in the choroid.131,185Thechoroidalsoexpressesglucagonand its receptor,58 and choroidal glucagon content is reported to increasewith positivelens treatments, although is not affected by negative lens treatments, at least up to oneday.60Exogenousglucagonaswellasinsulinalsoaffectsthethicknessofchoroid.63Thereisalsostrongevidenceforaroleofchoroidalretinoicacid(RA)inoculargrowthregulation‐RA shows bidirectional changes in response to visual manipulations that slow (positivelens and removal of diffusers) or accelerate (negative lens or diffuser) eye growth.118Choroidal expression of the RA‐synthesizing enzyme, retinaldehyde dehydrogenase 2(RALDH2) is also increased after12 and24hours of recovery from form‐deprivation.186

14

Finally, dopamine also has been linked to choroidal thickness changes in myopiadevelopmentandinhibitioninanumberofanimalmodels,althoughitisnotclearwhetherthesiteofactionislocaltotheretinainthesecases.41,187,1881.1.4.4 ScleraThe sclera represents the outer supportive coat of the eye, ultimately determining theshapeandsizeoftheglobe.Thescleraisconstantlyundergoingremodelingand/orgrowthchanges, especially during early development, and responds rapidly to visualmanipulationswithchangesinremodeling,witheffectsonbotheyesizeandrefractions.Inmammalsandprimates,thescleracomprisesadensefibrousconnectivetissue,whichisprimarilycomposedofextracellularmatrix(ECM)andfibroblasts.26 In theECM,collagen,proteoglycans, and elastic fibers are major components. In mammals, the scleral tissuecontainsapproximately90%collagen(typeI,III,IV,V,VI,VIII,XIIandXIII)byweight.Therelativeproportionsof these various subtypesof collagendetermine the size of collagenfibrils, which are irregularly organized into lamellae. Located between the collagenlamellae are scleral fibroblasts, whose functions include the production and turnover ofECM; thus they play an important role in the regulating eye elongation. Proteoglycansincluding glycosaminoglycan (GAGs) are one of the most important non‐collagenouscomponents of the ECM (approximately 0.7–0.9% of the total dryweight of the sclera).Knownfunctionsofproteoglycans include themodulationofcollagen fibrilassemblyandarrangement.26Thescleraalsoincludeselasticfibers,whichaccountforapproximately2%of the dry weight of the ECM,189 are also synthesized by scleral fibroblasts and aredistributedhomogenouslythroughoutthesclerastroma.During the development of myopia, morphologicaland histological changes of sclera has beendocumented, including overall enlargement of thesclera surface, sclera thinning, reduceddiameterandderangement of collagen fibrils, with these changesbecoming more exaggerated with time.190, 191 Withshorttermmyopiainducingtreatments,collagenfibrildiametersmayremainnormalbutscleraldryweightdecreased.192 Changes in ECM synthesis anddegradation, and related molecules have also beenobserved during the development of myopia andrecovery inmammalianmodels.190,193,194Specifically,decreases in the amount and/or gene expression oftypeIcollagen,195GAGs,194MMP2,196MMP4,197TIMP3,TGFB1,TGFB2,thrombospondin1(THBS1), tenascin(TNC), osteonectin (SPARC), and osteopontin(SPP1)198 in sclera have been reported during thedevelopment of myopia, while mRNA expression isreported to be up‐regulated for TGFBR3, TGFBI,MMP14, FGFR‐1.198, 199 These molecular changes go

Figure1‐5.Histologicalcross‐sectionoftheposterioreyewallofthechick;SCL‐C&SCL‐Findicatescleralcartilageandfibrouslayersrespectively.

15

hand‐in‐hand with changes in the biomechanical properties of sclera, which shows anincreasein“creeprate”duringthedevelopmentofmyopia.200,201Theabovediscussion is relevant tomammalianandprimate sclera.Thereare importantdifferences between these scleras and the sclera of chicks, which is one of the mostcommonusedmyopiamodels in research. Specifically, in addition to a fibrous layer, theaviansclerahasanadditionalinnerlayerofcartilage(Figure1‐5),andinmyopiceyes,thecartilage layerbecomes thicker thannormalwhile the fibrous layer thins.202,203Both thescleraldryweightandproteinsynthesisarealsoincreasedinthemyopiceyesofchicks.204,205Growth factors likely play important roles in regulating scleral remodeling during thedevelopmentofmyopia,apossibilitysupportedbyresultsofpharmacologicalstudies.Forexample, in chicks subconjunctival and intravitreal delivery of bFGF and TGF‐β1 arereportedtoinfluencethedevelopmentofform‐deprivationmyopia,andbFGF,TGF‐β,andBMPshavealsobeenshowntostimulatetheproliferationofbothscleralchondrocytesandfibroblastsinvitro.130,206‐208Asretinoicacid(RA),isknowntoregulateECMremodelingina variety of connective tissues, it is plausible that it may also regulate sclera ECMremodeling,giventhat italsoshowsbidirectionalregulation in thenearbychoroid.118,120,209Finally,thepresenceofmAChRsonscleralcellsopensupthepossibilityofcholinergicregulationofscleralgrowth,althoughthesourceofcholinergicinputisyettoberesolved.Nonetheless,thescleraisconsideredaplausiblesiteofactionfortheanti‐myopiaeffectsofmAChRantagonists,suchasatropine.98,104,113,2101.1.5 Anti‐MyopiaTreatmentsWhile the prevalence and severity ofmyopia continues to increase, effective therapeuticinterventions for myopia remain limited.211 Most myopia develops during the early tomiddle childhood years, extending into late teenage years and early adulthood insusceptible individuals.11 As already indicated, most myopia results from excessivelengtheningofthevitreouschambersuchthatimagesofdistantobjectsarefocusedinfrontoftheretinaandthusblurredintheabsenceofcorrectinglenses.212Human‐based epidemiological research points to a role of prolonged near work,particularly excessive reading, in the development ofmyopia.213,214 Interestingly, recentstudiespointtooutdooractivityasbeingprotectiveagainstthedevelopmentofmyopia.25,215Amongthefactorslikelytobecontributingtothelatteroutdooreffectarethereductionin near work activities and the properties of the outdoor visual environment, whichincludeshigher light levels compared to indoor environments.25 In termsof anti‐myopiatreatments, theseobservationsargueforchanges in lifestyle, toreducethetimespent inintensive,continuousnearwork,andtoincreasetheoutdooractivitiesandduration.Currently, the management of myopia remains largely limited to traditional opticalcorrections – single vision spectacles and contact lenses, and refractive surgery. Thesetraditionalcorrectionsareintendedtocorrectthemismatchbetweentheopticalpowerof

16

the eye and its length and so to correct distance vision, with no effect on myopiaprogression.However,novelsoftcontactlensdesignsintendedtoimposemyopicdefocuson periphery retina are currently under trial with early results showing promisingevidence of slowed myopia progression.43, 216‐218 Studies of orthokeratology (ortho‐k),whichusesrigidcontactlensestoreshapeandsoreducethepowerofthecentralcornea,have also reported promising results of slowed myopia progression,219 with imposedperipheral myopic defocus resulting from corneal reshaping being proposed as theunderlyingmechanism.Clinicallytested,pharmacologicalinterventionsforslowingorpreventingthedevelopmentof myopia are limited in most countries to two mAChR antagonists, atropine andpirenzapine, which have both been shown to retard myopia progression, although themechanismsunderlyingtheirinhibitoryeffectsremainpoorlywellunderstood,asalreadynoted.220, 221 Most studies involving atropine have used concentrations intended forophthalmic diagnostic applications, with their use accompanied by many visuallydebilitatingocularside‐effects,includingglareandlossofaccommodation,andtheefficacyofsuchatropinetreatmentsisalsoreportedtoattenuateovertime.Thesefindingsunderlietheonly limiteduseof atropine formyopia control in theUS.However, thispicturemaychange with a recent report showing significant reductions of myopia progression withmuchlowerconcentrations(100‐foldless),withfarfewerocularside‐effects.222Forhigh,sometimereferredtoas “degenerative”myopia,posteriorscleralreinforcementsurgery is currently the only clinical treatment option, a last resort aimed at preservingvisioninaneyethathasbecomeexcessivelylargeandmechanicallyunstable.223However,recent and still on‐going studies investigating the application of biomaterials and slowrelease drug delivery systems to reinforce the sclera and/or manipulate the scleraremodeling have shown interesting results.224 Apart from such treatments aimed atstabilizingthemyopiceye,specificpathologicalcomplicationsofhighmyopiaaretargetsofothertreatments,includingforchoroidalneovascularization,intravitrealinjectionsofanti‐VEGF‐A,whichhaveyieldedeffectiveandsustainedbenefits.225In summary, current treatment options for controlling myopia progression and itscomplicationsarelimited.Forthedevelopmentofneweffectiveclinicalinterventions,itisimportanttobetterunderstandthemolecularmechanismsunderlyingthedevelopmentofmyopiaandassociatedcomplications.TheresearchreportedinthisdissertationsearchedforthemolecularandcellularsignalingpathwaysandmechanismsinvolvingtheRPEthatcouldpotentiallybeexploitedformyopiacontrol.

17

1.2 GeneralExperimentalApproachesandMethodsThefollowingsectionssummarizethegeneralexperimentalapproachesandmethodsusedin the studies reported in this dissertation, including animal models (chick) and visualmanipulationsused toperturbeyegrowth (defocusing lens treatments),ocularbiometry(formeasurement of axial ocular dimensions), RPE gene expression assessments (oculartissuecollection,RNApurificationandquantification,reversetranscription,real‐timePCR),and protein assays (western blot and immunohistochemistry). Details of materials andmethodsthatarespecifictoindividualstudiesarepresentedinrelevantchapters.1.2.1 AnimalmodelChicks were used as the animal model for this dissertation research. A White‐Leghornstrainwasused,alsoforconsistencywiththatusedmostlycommonlyinmyopiaresearch.Thechickisthemostwidelyusedanimalmodelofmyopiaandalsothemostpracticalandreliablemodel.Youngchicksrespondveryrapidlytoform‐deprivationandimposedopticaldefocus. The refractive error and axial length changes in response to various visualmanipulations have been characterized in detail for the chick.35, 59, 226, 227 For example,form‐deprivation,achievedbycoveringeyeswithdiffusers,inducesrefractiveerrorsupto‐30diopters(D),within~10daysinyoungchicks,andincreasedocularelongationresultsinalmostcompletecompensationfortheopticaldefocusimposedby‐10Dlensesafter~7days.59Theseaxialgrowthchangesoccurmoreslowlyinolderbirds.227Apartfromchangesto the overall axial ocular dimensions, changes in the thickness of the choroid are alsoobserved with both form‐deprivation and imposed optical defocus.35, 179, 180 Choroidalthickening is detectablewith just 10minutes of positive lens treatments, and significantthinningofthechoroidisdetectibleafteran1hofnegativelenswear.16,1791.2.2 LenstreatmentTo manipulate ocular growth, monocular defocusing lenses were used, negative lenses(imposedhyperopia)toinduceacceleratedeyegrowthandsomyopiaandpositivelenses(imposed myopia) to induce slowed eye growth and hyperopia. The magnitude of theoculargrowthchangesdirectlyreflectthepowerofthedefocusinglensanddurationofthetreatmentperiod,beingonlysmallforveryshorttreatmentdurations,irrespectiveoflenspower.Threetreatmentdurations,2h,48h,and38days,wereemployedinthisresearch,to investigate the gene and protein expression at different stages of altered eye growth(Figure1‐6).Specifically,toidentifygenesimportantforinitiatingchangesineyegrowth,short‐term(2h) treatmentwasused tominimizeocular growthchangesandpotentiallyconfounding effects of the latter. Genes showing differential expression after 48 h oftreatmentareassumedtobecriticalformaintaininginducedalteredgrowthpatterns,asbythistimealteredgrowthisdetectable.Thelonger‐termnegativelenstreatment(38days)wasemployedtoinvestigatethegenesinvolvedinmaintainingalreadyestablishedmyopiaandocular(retinal)complicationsofthesame.

18

Figure1‐6.Monocular lens treatmentsused indissertation research and thepatternsof optical defocusimposedonnormal(emmetropic)chickeyes.Fordistantobjects,positivelensesmovetheimageplaneinfront the retina, imposing myopic defocus while negative lenses move the image plane backwards,imposinghyperopicdefocus.

19

1.2.3 OcularbiometrymeasurementHigh frequencyA‐scanultrasonographywasused tomeasure,bothbeforeandafter lenstreatments,axialoculardimensions, includinganteriorchamberdepth,vitreouschamberdepth,andchoroidalthickness.AtypicalA‐scanultrasonographytraceisshowninFigure1‐7.TheresolutionofourcustomizedA‐scanultrasonographyset‐upis10µm.Axiallengthisrepresentedbythesumofanteriorchamberdepth(Ftopeak1),lensthickness(peak1to 2), vitreous chamber depth (peak 2 to 3), retinal thickness (peak 3 to 4), choroidalthickness(peak4to5),andscleralthickness(peak5to6).

Figure1‐7.A‐scanultrasonographyofchickeye.

1.2.4 OculartissuecollectionInallexperiments,chicksweresacrificed,eyesquicklyenucleated,andtissuesseparatelycollectedovericeformolecularand/orbiochemicalanalysis.First,theanteriorsegmentoftheeye iscutaway; theremainingposterioreyecup is immersed incoldRinger’sbuffer,the retina peeled off from the RPE with forceps, and then the RPE collected by gentlyrinsing cells off the choroid with buffer. In experiments requiring retina, piecescontaminatedwithRPEwerediscarded.Choroidwascollectedlast,bypeelingawayitfromtheadjacent sclera.ForRNAsamples, allocular tissueswere lysedwithRLTbuffer fromRNeasy Mini kits (Qiagen, Valencia, CA). Ocular protein samples were lysed with RIPAbuffer (Sigma‐Aldrich, St. Louis, MO) containing a protease inhibitor cocktail (Sigma‐Aldrich).Allsampleswerestoredat‐80°Cimmediatelyforlateruse.

20

1.2.5 RNApurificationandquantificationTotalRNA fromretinaandRPEsampleswaspurifiedusingRNeasyMinikits,while totalRNA fromchoroidwaspurifiedusingRNeasyFibrousTissueMiniKits (Qiagen,Valencia,CA), with on‐column DNase digestion, according to the manufacturer’s protocol. RNAconcentration and A260/A280 optical density ratiosweremeasured for quantification andquality controlwithaspectrophotometer (NanoDrop2000,NanoDropTechnologies, Inc.,Wilmington, DE). RNA quality was also examined by gel electrophoresis, using a 1.2%agarosegel,withethidiumbromidestaining.1.2.6 ReversetranscriptionTotal RNAwas reverse transcribed to cDNA using SuperScript III First‐Strand SynthesisSystemforRT‐PCR(Invitrogen,Carlsbad,CA).TheamountofcDNAusedineachreal‐timePCR reaction varied across tissues and between genes, reflecting an adjustment fordifferences in expression levels. The possibility of genomic DNA contamination wasexaminedusingRNAsampleswithoutRTenzymes.1.2.7 Real‐timePCRPrimers for these studieswere designed using Primer Express 3.0 (Applied Biosystems,FosterCity,CA).Ten‐foldserialdilutionsofcDNAwereusedforgeneratingstandardcurvesfor each pair of primers. Melt curves were performed for all genes examined to detectpotential non‐specific amplification of primers. The efficiency (E) of each primer wascalculatedusingequation(1)inTable1‐2.QuantiTectSYBRGreenPCRKits(Qiagen,Valencia,CA)andaStepOnePlusReal‐TimePCRSystem(AppliedBiosystems,FosterCity,CA)wereusedforgeneexpressionquantification.The equations using in analyzing these data are given in Table 1‐2. Mean normalizedexpression(MNE)valueswereusedtocomparegeneexpression levels inRPEfromlens‐treatedeyes,theirfellow(control)eyes,andnormaleyesfromuntreatedbirds.MNEvalueswere derived as equation (2). Mean mRNA expression levels were calculated usingequations(3).Lenstreatment‐inducedchangeswereexpressedasFold‐changes,calculatedusingequation(4).1.2.8 WesternblotProteinexpressionprofileswereestablished for chickocular tissuesusingWesternblotsandsamplesfromuntreatedbirds.TotalproteinconcentrationwasmeasuredusingaBCAassay (Pierce Biotechnology, Rockford, IL). For Western blots, protein samples wereprepared inNuPAGELDS samplebuffer (Invitrogen),withorwithoutDTTasa reducingagent,andheatedat95°Cfor5minutes.Proteinsampleswerethenelectrophoresedundernon‐reducingandreducingconditionson4%‐12%gradientgels(NuPAGE4‐12%Bis‐TrisGel,Invitrogen),beforebeingtransferredtonitrocellulosemembranes(iBlotGelTransferStacks, Invitrogen). Membranes were blocked (StartingBlock T20 [TBS]; PierceBiotechnology), then incubated with primary antibodies, and finally labeled with HRP‐

21

conjugated secondary antibodies. Immunoreactive bands were detected withchemiluminescence (Supersignal Pico ECL; Pierce Biotechnology), and images developedusingabioimagingsystem(FluorChemQ,AlphaInnotech;SanLeandro,CA).Thechoiceofantibodywasbasedon results of database searches.Bothpositive andnegative controlswere included. All assays involved three independent biological samples and triplicaterepeats.

1.2.9 ImmunohistochemistryPosterior eyecups were prepared from enucleated eyes, immersed in Optimal CuttingTemperature (OCT) compound and stored at ‐ 80 °C immediately for later use. Seven‐micron cryostat sections were cut and dried at room temperature, fixed with acetone,washedwithPBS,andthenblockedwith10%normalseruminPBScontaining2%bovineserumalbumin(BSA). Immunostainingmadeuseofaprimaryantibodyandfluorescenceconjugated secondary antibody; as a negative control, some sections were incubated inisotypeIgG.Afterlabeling,sectionsweremountedonglassslideswithmediumcontainingthenuclearstain,DAPI(VectorLabs,Burlingame,CA),andphotodocumentedwithaZeissAxioplan2microscope(CarlZeiss,Inc.,Germany).

Table1‐2.Equationsusedinsemi‐quantitativeestimatesofgeneexpression Equations Number

(1)

(2)

(3)

or

(4)

22

1.3 DissertationOutlineThis dissertation describes investigations into the role of RPE in myopia development,focusingonthemolecularandcellularmechanismssuchasdifferentialgeneexpressioninRPE underlying the altered eye growth. The ultimate goal of these studies is to developnovelapproachesformyopiacontroltargetingRPEandthesesignalingpathways.Chapters2,3,and4ofthisdissertationaredividedandorganizedaccordingtothespecificaimsinthefollowingorder:

1. Toinvestigatethedifferentialgeneexpressionofbonemorphogeneticprotein(BMP)2,BMP4,andBMP7inchickRPEwithshort‐term(2and48hours)imposedopticaldefocus. As part of this study, the expression of related proteins, as well as thelocalization of BMPs in chick posterior ocular tissues,was also investigated. Ourinterestwas inwhether theexpressionofoneofmoreof thesegenes,whichhavebeen linked togrowthmodulation inother tissues,mightbemodulatedbyopticaldefocus,consistentwithrolesininitiationandmaintainingearlystageofalteredeyegrowth. The results of this study were published in the journals of InvestigativeOphthalmologyandVisualScienceandExperimentalEyeResearch.

2. Toestablishusinghigh‐throughputDNAmicroarrayaglobalgeneexpressionprofile

inchickRPEafter long‐term(38days) imposedhyperopicdefocus,which inducesocular elongation andmyopia.Myopia carries a high risk of retinal complicationsthat compromise vision, some of which could plausibly reflect altered RPE geneexpression and thus RPE function. Genes showing altered expression in RPEmayalsoplayimportantrolesinmaintainingmyopia.

3. To investigate the effects of apomorphine (APO) on TGF‐βs secretion in cultured

hfRPEcells.Dopamine(DA)receptoragonistsareknowntoinhibitmyopicgrowth,althoughtheirsiteofactionandotherdetailsofthesignalpathwayinvolvedarenotwellunderstood.AstheRPEisknowntohaveDAreceptors,itisalsoaplausiblesiteofactionforthiseffect.

Inthelastchapter,wesummarizethefindingsfromthisdissertationstudy,anddiscussthefuturedirectionsanddevelopmentbasedonthecurrentresults.

23

Chapter2Bidirectional,OpticalSign‐DependentRegulationofBMPGeneExpressioninChickRPEAbstractThis study explored the role ofBoneMorphogenetic Proteins (BMPs), BMP2, BMP4, andBMP7indefocus‐inducedoculargrowthchangesusinggeneexpressionchangesinretinalpigmentepithelium(RPE)asasurrogate.YoungWhite‐Leghornchickenswereusedinthisstudy. Normal gene expression of BMP2, BMP4, BMP7 and their receptors was alsoexaminedinretina,RPE,andchoroid,andtheexpressionprofilesofallthreeBMPproteinswereassessedinthesametissuesusingWesternblotsandimmunohistochemistry.Semi‐quantitative PCR (qPCR)was used to assess the effects of short‐term exposure (2 or 48hours)tomonocular+10and+10diopter(D)lenses,onRPEgeneexpressionofBMPsandtheir receptors. Ocular growth was measured using high frequency A‐scanultrasonography. In theeyesofuntreated(normal)chickens,expressionofall threeBMPmRNAswashighinRPEcomparedtoretinaandchoroidandallthreetissuesexpressedtherelatedBMPproteins.Atthelevelofgeneexpression,allthreereceptorsalsoweredetectedin these tissues, with BMPR2 showing the highest, and BMPR1B, the lowest expression.BMP2, 4, and 7 were up‐regulated in the RPE from eyes wearing +10 D lenses, whichexhibitedshorterthannormalvitreouschamberdepth(VCD)andthickenedchoroid,whileBMP gene expression was down‐regulated in the RPE from eyes wearing ‐10 D lenses,whichdevelopedenlargedVCDand thinned choroid. In contrast, theBMP receptorsdidnotshowdifferentialexpressionchangesinRPEinresponsetothesedefocustreatments.ThatmRNAexpressionofBMPsinchickRPEshowsbidirectional,defocussign‐dependentchanges is suggestive of a role for BMPs in eye growth regulation, although the diffuseocularexpressionofBMPsandtheirreceptorssuggestscomplexgrowth‐modulatorysignalpathways.Chapter2wasreproducedwithmodificationfromthefollowingpublishedpapers:

ZhangY,LiuY,WildsoetCF.2012.Bidirectional,opticalsign‐dependentregulationofBMP2gene expression in chick retinal pigment epithelium. Invest Ophthalmol Vis Sci. 2012.53:6072‐6080.

24

YanZhang,YueLiu,CarolHo,ChristineF.Wildsoet.EffectsofimposeddefocusofoppositesignontemporalgeneexpressionpatternsofBMP4andBMP7inchickRPE.ExperimentalEyeResearch.2013;109:98‐106.

25

2.1 IntroductionUncorrected refractive errors are one of the world’s leading causes of blindness andsignificantcontributorstotheglobalburdenofeyedisease.6,8,9,228Ocularrefractiveerrorsreflect the imbalance between the refracting power of the eye, towhich the cornea andcrystalline lens contribute, and its optical axial length,which defines the position of theretina relative to the latter optical elements.Mismatches between these parameters canresultineithermyopia,wheretheeyeistoolonginrelativeterms,orhyperopia,wheretheeye is too short. Babies are typically born with refractive errors, which are correctedduring early development through a process of coordinated ocular growth known asemmetropization.3,4,229‐231 However,myopiamay also occur in childhood as a failure ofemmetropization,whentheeyecontinuestoelongateafteremmetropiaisachieved.9,232Studiesusinganimalmodelshaveprovidedconvincingevidencefortheroleofvisualinputin the emmetropization process and its abnormalities.16,233,234 For example, both spatialform‐deprivation and negative defocusing lenses accelerate the rate of eye elongationwhilepositivedefocusinglensessloweyeelongation.Thenetresultsinrefractivetermsareinducedmyopiaandhyperopiarespectively.Avarietyofstudies,includingneurallesioningones,supportamodeloflocalregulationofeyegrowth,withtheretinabeingthepresumedoriginofgrowthmodulatorysignals,linkedviaoneormorelocalsignalcascadesdirectedatthetwoouterlayersoftheeyewall–thechoroidandsclera,whichultimatelydetermineeyesize.34‐37Althoughthenatureoftheseregulatingpathwaysremainspoorlyunderstood,oneinvestigationalapproachhasbeentolookforgenesshowingdifferentialregulationinone or more of these key tissues during altered eye growth.26, 39‐41 Becauseemmetropization is bidirectional, at least in chicks, bidirectional, optical defocus sign‐dependent regulation of genes has been interpreted as evidence of their roles inemmetropization.16Exceptingthedatapublishedfromthisdissertation,onlyexpressionofthe ZENK protein in a subset of retinal amacrine cells exhibits this profile (i.e., opticaldefocussign‐dependence).16,51,53The retinal pigment epithelium (RPE) is a unique tissue, lying between the retina andchoroid,andcomprisingasinglelayerofpolarizedcellsinterconnectedbytightjunctions.Itservesnotonlytoabsorbstraylightwithintheeyebuttotightlyregulatetheexchangeofmolecules,includingionsandwater,betweentheretinaandchoroid.ThustheRPEhostsavarietyofreceptorsandtransporters.135,137OurinterestintheRPEisasalikelyconduitforgrowthregulatorysignalsoriginatingintheretina.ByexamininggeneexpressionpatternsintheRPEfromeyesundergoingalteredgrowth,wehopetoobtaininsightintohowsuchretinalsignalsarerelayedtothechoroid/scleracomplex,withthepossibilityofidentifyingkeygrowthregulatorymoleculesunderlyingmyopiceyegrowth.40,135BMPs represent a large family of multi‐functional growth factors that belong to thetransforming growth factor‐β superfamily, with important roles in embryogenesis andosteogenesis.235‐239 So far, at least 15 members of the BMP family have been reported(Table2‐1).239,240Basedonaminoacidsequencehomology,BMP2andBMP4belongtoonesubgroup, being very similar to each other, while BMP7 belongs to another.239 Of this

26

family,BMP2hasalreadybeenlinkedtooculardevelopmentandgrowthregulation.133,241,242 Importantly, BMP2 gene expression in chick retina/RPE is down‐regulated in form‐deprivationmyopia.133BMP2hasalsobeenreportedtoinhibitserum‐inducedhumanRPEcell proliferation, consistentwith the profile of a negative growth regulator,243 althoughBMP2 is reported to have the opposite action in vitro on human scleral fibroblasts, ofstimulatingtheproliferationanddifferentiation.208PreviousstudieshavealsolinkedbothBMP4 and BMP7 to embryonic eye morphogenesis as well as diseases of the adultretina.243‐247 For example, a genetics study using a loss‐of functionmutation of BMP7 inmousereportedseveredefectsinthedevelopingeye,248andarecenthumangeneticstudyproposedBMP4asapossiblecandidategeneformyopia.Bothstudiesprovidedadditionalmotivationforthecurrentstudy.249To investigate the roles of BMP2, BMP4, and BMP7 in defocus‐mediated eye growthregulation, we used short exposures to both positive and negative lenses, to limit themagnitude of induced ocular dimensional changes, which may themselves affect geneexpression in one or more tissues. We observed defocus sign‐dependent, bidirectionalregulationofBMP2,BMP4,andBMP7geneexpressioninRPEbutnotofitsRPEreceptors,althoughtheexpressionofallthreeBMPreceptorswasconfirmedinRPE,aswellasretinaandchoroid,alongwithallthreeBMPs.

Table2‐1.MembersoftheBMPfamily

BMPs Synonyms

BMP1 PCP(procollagenC‐endopeptidase)

BMP2 BMP2A

BMP3 Osteogenin

BMP4 BMP2B

BMP5

BMP6 Vgr‐1

BMP7 OP‐1(osteogenicprotein1)

BMP8 OP‐2

BMP9 Gdf‐2(growth/differentiationfactor2)

BMP10

BMP11 Gdf‐11

BMP12 Gdf‐7

BMP13 Gdf‐6

BMP14 Gdf‐5

BMP15 Gdf‐9B

27

2.2 MaterialsandMethods2.2.1 AnimalsandLensTreatmentsWhite‐Leghornchickenswereobtainedashatchlingsfromacommercialhatchery(Privett,Portales, NM) and raised under a 12 h‐light/12‐h dark cycle. To induce myopic andhyperopicgrowthpatterns,19day‐old(adolescent)chickensworemonocular‐10and+10Dlensesrespectively,foreither2or48h.Tocharacterizetheeffectsofthelenstreatmentsoneyegrowth,theaxialoculardimensionsofbotheyesofindividualbirdsweremeasuredunder isoflurane anesthesia (1.5% inoxygen), at both thebeginning and endof the lenstreatment periods, using high‐frequency A‐scan ultrasonography (n = 53). Only data forocular parameters showing significant change are reported, i.e., vitreous chamber depth(VCD),choroidalthickness(CT)andaxiallength(AL;thedistancebetweenoftheanteriorcornealandposteriorscleralsurfaces).Thesametreatmentswereappliedtoaseparatesetof chickens for use in gene expression studies, to avoid the potentially confoundinginfluence of anesthesia on gene expression. In this case, each of the 4 treatment groupscomprisedatotalof4‐6birds,madeupfrom3independentrepetitionsoftheexperiment(n=16for2hof‐10Dlenstreatmentgroup;n=14forallthreeothertreatmentgroups);age‐matcheduntreatedbirds,i.e.,nolenstreatment,werealsoincluded(n=24). ExperimentswereconductedaccordingtotheARVOStatementfortheUseofAnimalsinOphthalmicandVisionResearchandapprovedbytheAnimalCareandUseCommittee(ACUC)atUniversityofCalifornia,Berkeley.2.2.2 TissueSampleCollectionforRNAandProteinStudiesRetina,RPE,andchoroidsamplescollected from19‐dayand21‐dayolduntreatedchickswereusedtostudynormalBMP(BMP2,‐4,‐7)andBMPreceptor(BMPR1A,‐1B,‐2)geneexpression. BMP protein expression in the same three tissues was characterized usingWestern blots. Additional posterior eyecups were collected from untreated chicks andprocessedforimmunohistochemistry,tolocalizetheproteinsinthetissuesmakinguptheposterioreyewall–retina,RPE,choroidandsclera.OnlyRPEwascollectedfromthelensexperiments,intheinterestofobtainingsamplesinminimaltime.TheexpressionofBMP2,‐4,‐7andtheirreceptorswasexaminedinsamplesfromlens‐treatedanduntreatedfelloweyesofexperimentalsubjects,afterlens‐wearingperiodsofboth2and48h.Themethodofsamplecollectionwasasdescribedindetails inChapter1.Briefly,chicksweresacrificed,eyesenucleatedandthenretina,RPE,andchoroidisolatedandcollectedoverice.ForRNAsamples,allthreeoculartissueswerelysedwithRLTbufferfromRNeasyMinikits(Qiagen,Valencia,CA).OcularproteinsampleswerelysedwithRIPAbuffer(Sigma‐Aldrich,St. Louis,MO) containing a protease inhibitor cocktail (Sigma‐Aldrich). All sampleswerestoredat‐80°Cimmediatelyforlateruse.

28

2.2.3 RNAPurificationandQuantificationTotalRNA fromretinaandRPEsampleswaspurifiedusingRNeasyMinikits,while totalRNA fromchoroidwaspurifiedusingRNeasyFibrousTissueMiniKits (Qiagen,Valencia,CA), with on‐column DNase digestion, according to the manufacturer’s protocol. RNAconcentration and A260/A280 optical density ratio were measured for quantification andquality controlwithaspectrophotometer (NanoDrop2000,NanoDropTechnologies, Inc.,Wilmington, DE). RNA quality was also examined by gel electrophoresis, using a 1.2%agarosegel,withethidiumbromide(EB)staining.2.2.4 ReverseTranscriptionTotal RNAwas first reverse transcribed to cDNA (SuperScript III First‐Strand SynthesisSystemforRT‐PCR,Invitrogen,Carlsbad,CA).GenomicDNAcontaminationwasexaminedusing RNA samples without RT enzymes. The amount cDNA used in each PCR reactionvariedacrosstissuesandbetweengenes,accordingtoexpressionlevels.2.2.5 PrimerDesignandValidationPrimers for these studies were designed using Primer Express 3.0 (Table 2‐2; AppliedBiosystems, Foster City, CA). Ten‐fold serial dilutions of cDNAwere used for generatingstandardcurvesforeachpairofprimers.Theefficiency(E)ofeachprimerwascalculatedusing the following equation, E=10(‐1/slope). Melt curves were performed for all genesexamined; all PCR tests yielded single peak products. Amplification of each gene wasperformedintriplicate.

Table2‐2.PrimerinformationforBMPsandBMPreceptorsGene NCBIAccess

NumberSequences(5’‐3’) Efficiency Amplicon

BMP2 NM_204358 Forward:5’‐AGCTTCCACCACGAAGAAGTTT‐3’Reverse:5’‐CTCATTAGGGATGGAAGTTAAATTAAAGA‐3’

93.6% 96bp

BMP4 NM_205237.2 Forward:5’‐CCAGCAAATCAGCCGTCAT‐3’Reverse:5’‐CGGACTGGAGCCGGTAGA‐3’

97.5% 57bp

BMP7 XM_417496.3 Forward:5’‐GGTGGCAGGACTGGATCATC‐3’Reverse:5’‐GCGCATTCTCCTTCACAGTAATAC‐3’

100% 64bp

BMPR1A NM_205357.1 Forward:5’‐TGTCACAGGAGGTATTGTTGAAGAG‐3’Reverse:5’‐AAGATGGATCATTTGGCACCAT‐3’

93.8% 68bp

BMPR1B NM_205132.1 Forward:5’‐GGGAGATAGCCAGGAGATGTGT‐3’Reverse:5’‐GGTCGTGATATGGGAGCTGGTA‐3’

105% 66bp

BMPR2 NM_001001465.1 Forward:5’‐GCTACCTCGAGGAGACCATTACA‐3’Reverse:5’‐CATTGCGGCTGTTCAAGTCA‐3’

100% 62bp

GAPDH NM_204305.1 Forward:5’‐AGATGCAGGTGCTGAGTATGTTG‐3’Reverse:5’‐GATGAGCCCCAGCCTTCTC‐3’

95.6% 71bp

29

2.2.6 Real‐timePCRQuantiTectSYBRGreenPCRKits(Qiagen,Valencia,CA)andaStepOnePlusReal‐TimePCRSystem(AppliedBiosystems,FosterCity,CA)wereusedforgeneexpressionquantification.Meannormalizedexpression(MNE)valueswereusedtoexpressmRNAlevelsinRPEfromlens‐treated eyes, their fellow (control) eyes, and normal eyes from untreated birds.CalculationforMNEvalues,mRNAlevelsandfoldchangeshasbeendescribedinChapter1(Table1‐2). Chick glyceraldehyde3‐phosphatedehydrogenase (GAPDH)wasused as thehousekeepinggene.2.2.7 WesternBlotNormal BMP2, BMP4, and BMP7 protein expression profiles were established for chickretina,RPEandchoroidusingWesternblotsandsamplesfromuntreatedbirds.Theoculartissueswere collected and lysed at 4 °CwithRIPAbuffer (Sigma‐Aldrich, St. Louis,MO),containing a protease inhibitor cocktail (Sigma‐Aldrich, St. Louis, MO). Total proteinconcentration was measured using a BCA assay (Pierce Biotechnology, Rockford, IL).ProteinsampleswerepreparedinNuPAGELDSsamplebuffer(Invitrogen),withorwithoutDTTasareducingagent,andheatedat95°Cfor5minutes.Proteinsamples(20μg)werethen electrophoresed under non‐reducing and reducing conditions on 4%‐12% gradientgels (NuPAGE 4‐12%Bis‐Tris Gel, Invitrogen), before being transferred to nitrocellulosemembranes (iBlot Gel Transfer Stacks, Invitrogen). Membranes were blocked(StartingBlock T20 [TBS]; Pierce Biotechnology), then incubated with anti‐human BMPantibody(Abcam,SanFrancisco,CA,#ab6285forBMP2,#ab93939forBMP4,#ab93636forBMP7),andfinallylabeledwithHRP‐conjugatedsecondaryantibody(Pierce,Rockford,IL, # 31430 or Millipore, MA, # AP132P). Immunoreactive bands were detected withchemiluminescence (Supersignal Pico ECL; Pierce Biotechnology) and imaged with abioimagingsystem(FluorChemQ,AlphaInnotech;SanLeandro,CA).The choiceof antibodywasbasedon resultsofdatabase searches;maturehumanBMP2andchickenBMP2have96.5%identityandhumanandmouseBMP2have100%identity.For BMP4, there is 95.7% similarity between chick and mouse, and chick and humansequences. For BMP7, the equivalent figures are 99.3% and 98.6% respectively. Proteinsequence alignment for human,mouse, and chickBMP2, BMP4, andBMP7 are shown inFigure2‐1.ThespecificityoftheBMP2primaryantibodywasverifiedusingcommercialBMP2protein(Abcam, # ab87065).Mouse brain lysateswere used as positive controls. As a negativecontrolthecommercialBMP2proteinwasalsousedasapre‐absorbedblockingpeptideforthe BMP2 antibody. Commercial mature BMP4 protein (# ab87063, Abcam) andmousebrain lysateswereusedaspositivecontrols for testing theBMP4antibody,whilehumankidneylysates(#ab30203,Abcam)andmousebrainlysateswereusedaspositivecontrolsfor BMP7 antibody. Assays involved three independent biological samples and triplicaterepeats.

30

Figure2‐1.Proteinsequencealignmentforhuman,mouse,chickBMP2(A),BMP4(B),&BMP7(C).

2.2.8 ImmunohistochemistryPosterioreyecupswerepreparedfromenucleatedeyes, immersedinOCT(TedPellaInc.,Redding,CA)andstoredat‐80°Cimmediatelyforlateruse.Seven‐microncryostatsectionsweredriedatroomtemperature, fixedwithacetone,washedwithPBS,andthenblockedwith 10% normal goat serum in PBS containing 2% bovine serum albumin (BSA).Immunostaining used the BMP2, BMP4, and BMP7 primary antibodies, as described inprevious section covering Western blots, and fluorescence‐conjugated secondaryantibodies.Isotypecontrol(Invitrogen)wasalsoincluded.Sectionswerelabeledandthenmounted on glass slides with medium containing the nuclear stain, DAPI (Vector Labs,Burlingame,CA),andphotodocumentedwithaZeissAxioplan2microscope(CarlZeiss,Inc.,Germany).

31

2.2.9 StatisticalAnalysisDatawereexpressedasmean±SEM.PairedStudent’st‐testswereusedtocomparelens‐treated eyes with their fellow (contralateral) control eyes; one‐way ANOVAs combinedwith post‐hoc analysis (Fisher's Least SignificantDifference)were used for comparisonsinvolvingmorethan2groups.Inanalyzinggeneexpressiondata,comparisonsweremadebothbetweenthetwoeyesoftreatedbirds,asameasureoftheprimarytreatmenteffects,andalsobetweenthefelloweyesoftreatedbirdsandeyesofuntreatedbirds,to lookforeffectsonthefellowuntreatedeyesthatwouldimplyinterocularyokinginfluences.2.3 Results2.3.1 RNAYield&QualityMeanRNAconcentrationsandA260/A280opticaldensityratiosforretina,RPEandchoroidare shown in Table 2‐3. Retinal samples had highest RNA yield (16.9 ± 0.8 μg/eye),followedbychoroidalsamples(8.7±1.1μg/eye),withRPEsamplesgivingthelowestyield(1.9±0.1μg/eye).A260/A280ratiosforallthreetissueswereabout2.0.GelelectrophoresisconfirmedtheintegrityofRNAinthesamples(Fig.2‐2).

Table2‐3.RNAconcentration,A260/A280ratio,andtotalyieldpereyeforretina,RPE,andchoroidsamples

RNAconc.(ng/ul) A260/A280 Totalyield(ng)/eye

Retina 338.8±17.0 2.0±0.003 16942±848

RPE 38.7±1.79 2.0±0.009 1935±90

Choroid 173.1±22.7 2.0±0.008 8655±1135

Figure 2‐2. Results of electrophoresis using a 1.2% agarose gel and EB staining for 8 RPE RNA samplescheckedforRNAintegrity.LaneM,marker;samplesinlanes1‐8.2.3.2 mRNAExpressionofBMPsandBMPReceptorsinNormalChicksBMP2,BMP4,BMP7 and all threeBMP receptor subtypes examined ‐ BMPR1A,BMPR1BandBMPR2,wereexpressedinall threetissuetypesexamined‐retina,RPE,andchoroid(Fig. 2‐3). The expression levels of each gene relative to housekeeping gene GAPDH aresummarized in Table 2‐4. In relative terms, BMP2, ‐4, and ‐7 appeared more highlyexpressed in RPE compared to retina and choroid. Of the BMP receptors, BMPR2 andBMPR1A showed much higher expression than BMPR1B across all 3 tissues. Note thatdifferences in baseline expression of GAPDH between these ocular tissues were also

32

evidentwhenCt valueswerenormalized against totalRNA amount. The ratio ofGAPDHexpression–retina:RPE:choroidwas15:3:1.

A

B

C

Figure2‐3.mRNAexpressionofBMP2,BMP4,BMP7,andBMPtypeIandIIreceptors(BMPR1A,BMPR1B,BMPR2)innormal(untreated)chickretina(A),RPE(B),andchoroid(C).GAPDHusedasthehousekeepinggene.DataareexpressedasmeanMNE±SEM.

33

Table2‐4.mRNAlevelsforBMPsandBMPreceptors.

BMP2 BMP4 BMP7 BMPR1A BMPR1B BMPR2Retina 0.0007±

0.00008(n=20)

0.0002±0.00002(n=20)

0.005±0.0007(n=12)

0.002±0.0001(n=14)

0.0004±0.00003(n=12)

0.005±0.0004(n=12)

RPE 0.21±0.02

(n=48)

0.10±0.01

(n=38)

2.20±0.37

(n=32)

0.013±0.002(n=18)

0.0002±0.00003(n=20)

0.019±0.003(n=26)

Choroid 0.017±0.003(n=16)

0.024±0.002(n=16)

0.014±0.001(n=8)

0.0309±0.002(n=16)

0.0005±0.00009(n=16)

0.025±0.004(n=8)

2.3.3 ProteinExpressionofBMP2inNormalChicksWesternblotsindicatedthepresenceofBMP2proteininchickretina,RPEandchoroid(Fig.2‐4).Tounderstandthecomplexbandingpatternsobservedunderbothnon‐reducingandreducingconditions,itisimportanttonotethatbothmatureandproproteinofBMP2havebeen reported for the chick (http://www.uniprot.org/uniprot/Q90751), aswell as otheranimals.236,250Thematureproteinhas114aminoacids(aa;13kDa),whiletheproproteinismuch larger(353aa,40.3kDa),withsomeglycosylationsitesatwhich furtherproteinmodificationmayoccur.251,252Thepresenceoftheaminoacid,cysteine,alsoallowsdimersto form frommonomersviadisulfidebonds(Fig.2‐5).250,253,254 Indescribingour results,we have made tentative assignments to observed bands, based on this backgroundknowledge.Undernon‐reducingconditions(Fig.2‐4A),theretinalsample(lane2)showed4 strong bands corresponding to the dimer of the proprotein (~ 80 kDa), a modified(glycosylated)monomeroftheproprotein(~50kDa),amonomeroftheproprotein(~40kDa),andadimerofthematureprotein(~28kDa).ThedimerofthematureBMP2(~28kDa)wasnotdetected ineitherRPE(lane3)or choroid (lane4). Interestingly, in lane4(choroid), therewasanadditionalweakbandat~39kDa,whichmay representeitheratrimerofthematureproteinorotherformsofBMP2.255Inlane5,towhichBMP2protein(0.02μg)wasaddedasacontrol,abandat~13kDawasdetected.Comparedtothenon‐reducingconditions,thereducingconditions(Fig.2‐4B),generatedstrongerbandsandinsomecases,additionalbands(e.g.lane4,choroid),presumablyreflectingimprovedbindingoftheantibody,althoughtheresultsforthetwoconditionsweregenerallysimilar.Inbothcases,nomatureproteinofBMP2wasdetectedinRPE.Noobviousbandswerevisibleinthe negative control test, for which the BMP2 primary antibody was first neutralized,implyingverylownonspecificbinding.Themassesforhuman,mouse,andchickproproteinandmatureproteinofBMP2arepresentedinTable2‐5.2.3.4 ProteinExpressionofBMP4inNormalChicksBMP4 protein was detected inWestern blots prepared under reducing conditions fromchickretina,RPEandchoroidsamples(Fig.2‐6).Lanes1and2wereloadedwithpositivecontrols, while lanes 3 ‐ 6 were loaded with chick tissue samples. To understand thecomplicatedpatternsofBMP4proteinexpression, it is important tonote thatmonomers

34

anddimersoftheBMP4proproteinsandmatureproteins,aswellasotherformshavebeenreported in published literature, and databases (in the public domain athttp://www.uniprot.org/uniprot/Q90752).255‐258 Masses for different forms of BMP4proteinarealsopresentedinTable2‐5.

Figure2‐4.WesternblotsshowingproteinexpressionofBMP2forbothnon‐reducing(A)andreducing(B)conditions.Inbothcases,laneMwasloadedwithmarker,lane1withmousebrainlysates(positivecontrol),andlanes2,3,and4,withchickretina,RPEandchoroidrespectively,lane5withcommercialBMP2protein.Differences in BMP2 expression, both between tissues and between conditions, were evident, with thechoroid showing the highest expression and multiple forms. Molecular weights of main mature andproproteinofBMP2are13.0and40.3kDarespectively.NegativecontrolusingBMP2peptidepreabsorbedprimaryantibodywasincluded(C).Primaryantibodyconcentration1:500.MB,mousebrain;CB,chickbrain;Ret,retina;Cho,choroid.

Figure 2‐5.Diagramof BMPproprotein andmature protein. BMPs are synthesized as dimeric proproteinsincludinganN‐terminalsignalpeptide(SP),a largeprodomainandaC‐terminalmaturecomponentwithacharacteristiccystine‐knotmotif.ProteolyticprocessingbyfurinproteasesoccursattheRXXRmotif.Thereisalsoanintermoleculardisulfidebondlinkingthematurecomponentsofthedimer(seeasterisk).

http://www.uniprot.org/uniprot/Q90752�

35

Table2‐5.MassescorrespondingtodifferentformsofBMP2,BMP4andBMP7fordifferentspecies.

Human Mouse Chick

Proprotein Mature Proprotein Mature Proprotein Mature

BMP2 44.7(396aa)

12.9(114aa)

44.5(394aa)

12.9(114aa)

40.3(353aa)

13.0(114aa)

BMP4 46.6(408aa)

13.1(116aa)

46.5(408aa)

13.2(116aa)

46.5(405aa)

12.8(114aa)

BMP7 49.3(431aa)

15.7(139aa)

49.2(430aa)

15.6(139aa)

49.5(435aa)

15.7(139aa)

ForBMP4, theretina(lane4,Fig.2‐6)producedthemostcomplexbandingpattern.Twobands correspond to the dimer and monomer of the proprotein (~ 90 and ~ 46 kDa,respectively),andanotherbandat~35kDarepresentsthedimerofmatureprotein.Therealsoaretwoadditionalbandsat~60and~70kDa,possiblyrepresentingtheglycosylatedproprotein.Thechoroid(lane6)yieldedonly2bands,correspondingtothedimeroftheproprotein(~90kDa)andthemonomerofmatureBMP4(~13kDa,as in lane1,whichwasloadedwithcommercialBMP4protein,andshowssinglebandat~13kDa).TheRPE(lane 5) yielded the simplest pattern, with only one band, at ~ 40 kDa, which mayrepresentanuclearvariantofBMP4ormatureBMP4,basedonitsmolecularweight.255,257,259ThepatternofBMP4expressionshowedvariabilitybothbetweenindividualchicksandbetween the 3 ocular tissues analyzed, with one bird showing negligible choroidalexpression of BMP4. Samples prepared frommouse and chick brain show similar, albeitcomplex patterns. No obvious bands were detected in the negative control (data notshown).

Figure2‐6.WesternblotsshowingproteinexpressionprofilesforBMP4inretina,RPE,andchoroidfromadolescentchicks,preparedunder reducingconditions.LaneMwas loadedwithproteinmarker.Lane1was loadedwithdenatured, reducedBMP4protein (0.02μg), lanes2and3withmousebrainandchickbrainrespectively,andlanes4to6withchickretina,RPE,andchoroid,respectively.MB,mousebrain;CB,chickbrain;Ret,retina;Cho,choroid.

36

2.3.5 ProteinExpressionofBMP7inNormalOcularTissuesIn chick retina, RPE and choroid samples, BMP7 protein was detected inWestern blotsprepared under reducing conditions (Fig. 2‐7). Lane 1 and 2 were loadedwith positivecontrols,while lanes3‐6wereloadedwithchicktissuesamples.Database(inthepublicdomain at http://www.uniprot.org/uniprot/F1NUT2) and literature searches260 showedtheproproteinandmatureprotein information forchickBMP7.MassescorrespondingtothedifferentformsofBMP7indifferentspeciesarelistedinTable2‐5.TheBMP7proteinwasdetectedinchickretina,RPEandchoroidunderreducingconditions(Fig.2‐7).Theretina(lane4),RPE(lane5),andchoroid(lane6)allshowedbandsat~30kDa, corresponding to the dimer of the mature protein. The RPE (lane 5) yielded anadditionalweakbandat~49kDa,whichrepresentstheproprotein.Thechoroid(lane6)showed anotherweak band at 15 kDa, corresponding to themonomer ofmatureBMP7.Humankidneylysates(lane1),whichservedasapositivecontrol,yieldedasinglebandat~50kDa.Mouseandchickbrainsamplesshowedsimilarpatterns(lane2and3);ineachcase, there are two visible bands at ~ 30 and 15 kDa, corresponding to the dimer andmonomerofmatureBMP7,respectively.Noobviousbandsweredetected inthenegativecontrol(datanotshown).

Figure2‐7.WesternblotsshowingproteinexpressionprofilesforBMP7inchickretina,RPE,andchoroid.LaneMwasloadedwithproteinmarker.Lane1wasloadedwithhumankidneylysates,andlanes2and3wereloadedwithmouseandchickbrain,respectively.Lanes4to6wereloadedwithchickretina,RPE,andchoroid, respectively. MB, mouse brain; CB, chick brain; Ret, retina; Cho, choroid. HK, human kidneylysates.

37

2.3.6 ProteinLocalizationofBMP2,BMP4,andBMP7inNormalChicksBMP2, BMP4, and BMP7 labelingwere observed in all layersmaking up thewall of theposterioreyecup,withsimilarlocalizationpatterns(Fig.2‐8,2‐9,2‐10).Intheretina,thereisintenseBMPlabelinginregionscorrespondingtotheinnerplexiformlayer(IPL),outerplexiformlayer(OPL),andphotoreceptoroutersegments.Intheganglioncelllayer(GCL),theinnernuclearlayer(INL),andtheouternuclearlayer(ONL),BMPstainingseemstobedistributedthroughoutthecytoplasmofcells,whileintheRPE,labelingappearslimitedtothe basal (choroid) side of RPE, although this may represent an artifact of the heavypigmentationelsewhere inthis layer.Thechoroidshowsdiffuse labelingwhile thesclerashowsmorevariablestaining;thereisintenselabelingatthechoroid‐scleraboundaryandthroughout the outer fibrous component, but more localized labeling in the innercartilaginous layer, confined to the chondrocytes. The negative control, which waspreparedbyincubatingsectionsinbothmouseisotypeIgGandsecondaryantibody,showsonly very low background labeling for all three BMPs (Fig. 2‐8 E, 2‐9 E, 2‐10 D). Theseimmunostainingdataare consistentwith theaboveWesternblot results,whichdetectedtheBMP2,BMP4,andBMP7proteinsinretina,RPEandchoroid.2.3.7 OcularDimensionalChangesafterLensTreatmentWith the +10D lenses, VCDwas significantly decreased in treated eyes relative to theirfellowsafteronly2hofwear(p<0.001,n=6.Fig.2‐11A).Thelongerexposureperiodof48hyieldedasimilarresponsepatternalthoughthechangeinVCDwassignificantlylargerthanthatrecordedwiththeshorter,2hperiodoflenswearandCTwasnowsignificantlyincreased(p<0.001inbothcases,n=6).Althoughthislenstreatmentisexpectedtoslowaxialelongation,nostatisticallysignificantinteroculardifferencesinALwereseenovertheshorttreatmentdurationsusedinthisstudy.The‐10Dlenstreatmentalsoinducedchangesinoculardimensions(Fig.2‐11B),althoughtheyreachedstatisticalsignificanceonlyafterthelonger,48hperiodoflenswear.Atthistime, lens‐treatedeyeshad longerVCDsandALsandthinnerchoroidscomparedto theirfellows, with interocular differences reaching statistical significance in all 3 cases (p<0.001,n=18).TheinterocularVCDandALdifferencedatafor2(n=11)and48hwerealsosignificantlydifferentfromeachother(p<0.001).The two eyes of normal, untreated birds typically had similar dimensions and thus asexpected,nosignificantinteroculardifferencesinVCD,CT,andALwereobserved(datanotshown,n=12).Chicksusedforthecollectionofocularbiometrydatawerenotusedingeneexpression experiments, to avoid potentially confounding effects from themeasurementprocedure,whichincludedbriefexposuretoisofluraneandunobstructedvision.

38

Figure 2‐8. Representative sections from the posteriorwall of the eyecup labeled for BMP2 (in red),withnucleistainedwithDAPI(inblue)(A), labeledforBMP2alone(B),orstainedwithDAPIalone(C);double‐labeledsectionoverlaidonimageofunstainedsection(D);negativeisotypecontrolimagedinblueandredchannels (E). GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outerplexiform layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium; CHO, choroid; SCL‐C, scleracartilaginouslayer;SCL‐F,sclerafibrouslayer.*BasalsideofRPE,▼ Innerboundarybetweenchoroidandsclera,↓Borderbetweencartilaginousandfibrouslayersofsclera.Scalebar,200μm.

Figure 2‐9. Representative sections from the posteriorwall of the eyecup labeled for BMP4 (in red),withnucleistainedwithDAPI(inblue)(A), labeledforBMP4alone(B),orstainedwithDAPIalone(C);double‐labeledsectionoverlaidonimageofunstainedsection(D).Negativecontrolimagedinblueandredchannels(E).GCL,ganglioncelllayer;IPL,innerplexiformlayer;INL,innernuclearlayer;OPL,outerplexiformlayer;ONL, outer nuclear layer;RPE, retinal pigment epithelium; CHO, choroid; SCL‐C, sclera cartilaginous layer;SCL‐F, sclera fibrous layer. * Basal side of RPE,▼ Inner boundary between choroid and sclera, ↓Borderbetweencartilaginousandfibrouslayersofsclera.Scalebar,50μM.

Figure2‐10.Representativesections fromtheposteriorwallof theeyecup labeled forBMP7(inred),withnucleistainedwithDAPI(inblue)(A),labeledforBMP7alone(B),orstainedwithDAPIalone(C).NegativecontrolforBMP7(D)wasimagedinblueandredchannels.Scalebar,50μM.

39

Figure2‐11.Effectsof+10D(A)and‐10D(B)lenstreatmentsonaxiallength(AL),vitreouschamberdepth(VCD), and choroidal thickness (CT) following 2 h (n= 6, 11 resp.) and 48 h (n=6, 18 resp.) of lenswear,shown as interocular differences (treated‐control eyes) (mean ± SEM). Shown also are treatment‐inducedstatisticallysignificantchangesfrombaseline(*ontopofbars)andsignificantdifferencesbetween2and48htreatmenteffects(*&connectinglines).***p<0.001.2.3.8 Defocus‐InducedGeneExpressionChangesofBMP2,BMP4,andBMP7inRPEThe+10D lens treatment inducedanup‐regulationofBMP2 intheRPE,with the largestincreasebeingrecordedafteronly2h(Fig.2‐12A).ExpressionofBMP2wasincreasedby7.2‐and4.1‐foldintreatedeyescomparedtotheirfellow(control)eyes,with2and48hoftreatmentrespectively(p<0.001,n=14forbothcases).Theoppositetrendwasobservedwith the ‐10D lens treatment,which inducedadown‐regulationofBMP2,andherealso,thechangerecordedwith the2h treatmentwas larger(Fig.2‐12B).ExpressionofBMP2wasdecreasedby13.3‐and3.7‐foldintreatedeyescomparedtotheirfelloweyes,with2and 48 h of treatment respectively (p< 0.001, n = 16; p< 0.01, n = 14). No significantinterocular difference in BMP2 gene expressionwas observed in RPE from eyes of age‐matcheduntreatedbirds(datanotshown).Boththe+10and‐10DlenstreatmentsalteredBMP4geneexpression,withthedirectionof change being again defocus sign‐dependent and consistent in direction for bothtreatmentdurations (Figs. 2‐13), andwith thepatterns described forBMP2. Specifically,with the+10D lenses,BMP4geneexpressionwasup‐regulatedbysimilaramountsafterboth2and48hof treatment,by3.6‐and4.4‐fold, respectively (p<0.01,n=18 inbothcases;Fig.2‐13A).Withthe‐10Dlenses,BMP4geneexpressionwasdown‐regulated,albeitslightlymorewiththeshorter2hthanthelonger48htreatmentperiod,by3.8‐and1.4‐fold, respectively (p< 0.001; p< 0.01, n = 19 for both cases; Fig. 2‐13B). No significantinterocular difference in BMP4 gene expressionwas observed in RPE from eyes of age‐matcheduntreatedbirds(datanotshown).WhileBMP7alsoshowedbidirectionalchanges ingeneexpression inresponse toopticaldefocus,thepatternsdifferedintheirtemporalprofilesfromthosedescribedforBMP2andBMP4.Specifically,the+10Dlensinduced2.9‐foldup‐regulationofBMP7onlyafter48hoftreatment(p<0.05,n=18.Fig.2‐14A);ashorter2hexposuretothis lensdidnotaffect

40

gene expression (p > 0.05, n = 18). In contrast, gene expression of BMP7 was down‐regulatedwithboth2and48hof‐10Dlenstreatment,by1.37‐and1.38‐foldrespectively(p<0.01,n=20;p<0.01,n=17;Fig.2‐14B).Asexpected,RPEfromeyesofuntreatedage‐matchedbirdsshowednosignificantinteroculardifferencesinexpression(right&lefteyescompared;datanotshown).

Figure2‐12.Differential expressionofBMP2,BMP4, andBMP7mRNA inRPEafter2and48hof imposeddefocus (+10 D, A, C, E; ‐10 D, B, D, F). Comparison between treated and fellow control eyes; data areexpressedasmean±SEM.GAPDHwasusedasthehousekeepinggene.**p<0.01,***p<0.001.

41

2.3.9 YokingEffectsofLensTreatmentsonBMPGeneExpressioninRPEFor all three BMPs, gene expression in RPE from the contralateral fellow eyes of lens‐treatedbirdsfromeyesofuntreatedbirdswerecomparedtolookforindirectevidenceofinterocular yoking effects, which cannot be detected by the within‐bird interocularcomparisonsreportedabove.Withthe longer48h, ‐10Dlenstreatment,bothBMP4andBMP7expressionlevelsinRPEfromthefellowsofthelens‐treatedeyesshowedsignificantdown‐regulationrelativetovaluesforRPEofeyesofuntreatedchicks(p<0.01forBMP4andBMP7,Fig.2‐13D,F).ForBMP2,geneexpressionlevelsinbothtreatedeyesandtheirfellowsappearedreducedrelativetolevelsintheeyesofuntreatedbirds,hintingatyoking;however, the difference between fellow and untreated eyes did not reach statisticalsignificance (p=0.077, Fig. 2‐13B).No equivalent trendswere apparentwith the +10Dlenstreatment(Fig.2‐13A,C,E).2.3.10 GeneExpressionChangesofBMPReceptorsinRPEafterLensTreatmentsNoneof the threegenes ‐BMPR1A,BMPR1B,BMPR2 ‐ showeddifferences inexpressionbetween treated eyes and their fellows, for either of the lens treatments, irrespective oftheir duration (Fig. 2‐14A).However,when gene expression (MNEs) in the eyes of lens‐treatedbirdswascomparedwithequivalentdataforuntreatedbirds,BMPR2wasfoundtobesignificantlydown‐regulatedinbothtreatedandfelloweyeswith‐10Dlenstreatment,for both the 2 and 48 h treatment durations, implying yoked down‐regulation of thisreceptor(p<0.01,Fig.2‐14B).2.3.11 ValidationofUsingGAPDHasaHousekeepingGeneThestabilityofGAPDHexpressionacrossdifferenttreatmentconditionswasassessedbycomparingtheexpressionofGAPDH/totalRNA(μg)inRPEfromuntreated,treated,andfelloweyes.Itsexpressionwasnotsignificantlyaffectedbythelenstreatmentconditions(Fig.2‐15).

42

Figure 2‐13. BMP2, BMP4, and BMP7mRNA levels in RPE after +10 D (A, C, E) and ‐10 D (B, D, F) lenstreatments applied for 2 or 48 h. The fellows to eyeswearing ‐10D lenses showed similar, albeit smallerdown‐regulation of BMP gene expression after 48 h of treatment compared to untreated eyes (onlydifferencesforBMP4andBMP7reachedstatisticalsignificance,**p<0.01;BMP2,p=0.077).Nosuchyokingisevidentinthe+10Dlensdata.Dataareexpressedasmean±SEM.GAPDHwashousekeepinggene.

43

Figure 2‐14. BMP receptor mRNA expression in RPE after +10 and ‐10 D lens treatments and in eyes ofuntreatedbirds.Nodifferences ingeneexpressionbetween lens‐treatedand felloweyes, orbetween rightand lefteyesofuntreatedbirdswereobserved(A,p>0.05).mRNAexpressionofBMPR2wassignificantlydown‐regulatedinbothtreatedandfelloweyescomparedtountreatedeyesafter‐10Dlenstreatmentfor2and48h(p<0.01,**).

Figure 2‐15. Expression of GAPDH in RPE, normalized to total RNA (μg); data plotted as the ratio ofexpression levels in treatedand felloweyes fortreatedbirds,andforuntreatedbirds, theratioof levels inrightandlefteyes.Dottedlineindicatesaratioof1.0,representingnodifference.

44

2.4 DiscussionInthepresentstudy,wedemonstratedinnormalchickensthegeneandproteinexpressionofBMP2,BMP4, andBMP7 aswell as the gene expression of threedifferent subtypes ofBMPreceptorsinthreeposterioroculartissues–retina,RPEandchoroid,andforthefirsttime, the optical defocus‐sign‐dependent, bidirectional regulation of BMP2, BMP4, andBMP7 gene expression in chickRPE. The latter results are consistentwith, althoughnotdefinitive evidence for a role of these threeBMPs in defocus‐inducedmodulation of eyegrowth.The initial discovery that BMPs could induce ectopic bone and cartilage formationunderlies the name of this family of growth factors.261, 262 However, they are betterdescribed as multifunctional regulators, with influences on cell proliferation,differentiation, apoptosis and extracellular matrix accumulation, as well as on thedevelopmentofmanyorgans.235,236,238,239,263‐265BMP2,BMP4,andBMP7are threeof themostwidelystudiedgrowthfactorsintheBMPfamily.Whileseveralstudieshavefocusedon the roles of BMPs in embryonic eye development, investigations into their roles inpostnataloculardevelopmentand function inadulteyesarevery limited.133,143,242,243,247,263,266Inearlyembryonicchickeyes(E3eyesatHamburgerandHamiltonstages15–18),BMP2 mRNA was not detected in the eye. However, BMP2 expression was detectablethroughoutmostoftheretinawithstrongerexpressionindorsalregionsinE8retina.Thedorso‐ventralasymmetryexpressionofBMP2maintainedtothelatedevelopmentalstage(E18).241BMP4mRNAexpressionwasfoundtobeinitiallyrestrictedtothedorsalretina,with the expression pattern becoming less regionally localized and alsoweaker in laterembryonicdevelopment.Moremature retinasalsoexpressBMP4althoughdorso‐ventralasymmetriesinexpressionarenolongerapparent.241Apartfromstudiesinchicks,ocularstudies of BMP4 expression are very limited in number. In the human embryonic eye,BMP4 expression has been observed in the optic vesicle, developing optic cup anddevelopinglens.249,263ScleralexpressionofBMP4hasalsobeendescribedforbothguineapigandadulthumaneyes.208,267Intheadultmouse,allcellsintheretinawereshownbyinsituhybridizationtoexpressBMP4andtheRPE,tohighlyexpressedit.243Thedistributionof BMP7 in early embryonic chick eyes was strong in the ventral RPE, and later onthroughouttheretina,withstrongexpressioninthedorsalregion.Intheearlyembryonicstage, BMP7 had a more laminar distribution, initially confined to the presumptive INL(pINL),pGCL,pONL,andrestrictedtotheINLandGCLatamorematurestage.241BMP7hasalsobeenreportedtoenhancechickphotoreceptoroutersegmentdevelopmentinvitro.268Inanotherstudy,BMP7proteinexpressionwasobservedinallhumansretinallayers,andasimilarly diffuse pattern of BMP7 expression was observed in rat retina.269 Overall, theobservationsfortheretinaofotherspeciesaresimilartotheresultsreportedhereforthechickretina.Thecurrentstudyrepresentsthemostcomprehensivestudyto‐dateofBMP2,BMP4,BMP7andBMPreceptorgeneandproteinexpressionintheposterioroculartissuesofadolescentchickens, complementing and expanding on an earlier investigation of retina/RPEBMP2expression in 7 day‐old chicken.133While the expression of threeBMPswas found to be

45

onlylowintheretina,theywereallhighlyexpressedintheRPE,consistentwithRPEbeingamajorocularsourceofthisgrowthfactor.Furthermore,thegeneexpressionprofilesfortheBMPreceptorssuggestthattheyactatmultiplesiteswithintheposteriorlayersoftheeye,withpotentiallymultiplefunctions.Specifically,wewereabletoconfirmthepresenceinallthreeoculartissuesofthereceptorsinvolvedindownstreamsignalingofBMPs,i.e.,theheterodimerizedtypeI(eitherBMPR1AorBMPR1B)andtypeII(BMPR2).235Althoughthere were receptor‐related differences in gene expression patterns, nonetheless theimpliedbroadoculardistributionofthesereceptorsiscompatiblewithbothparacrineandautocrinesignaling.BecauseBMPsexistinmultipleforms,includingdimersandmonomersofproproteinandmatureprotein,andfurther,thatgeneexpressionlevelsdonotreliablypredicttranslationinto protein,we also examined for these three BMPs, both their protein expression andtheir localizationinposterioroculartissues.TheWesternblotsdetectedvariousformsofBMPs in all three tissues ‐ retina, RPE and choroid, and also showed tissue‐relateddifferencesinproteinexpressionprofiles.OfnotewasthedetectionofboththebiologicallyactiveformsandproproteinofBMPsinRPE.ThusitseemslikelythattheRPEservesasastoragesiteforBMPs.Indeed,thehighlevelofBMPgeneexpressionintheRPEcomparedtothetwoneighboringtissues(retina&choroid),isconsistentwithitbeingamajorsourceofBMPs for these tissues,whichnonethelessalsoappearhave thecapacity tosynthesizeandsecreteBMPslocally,basedonourgeneexpressiondata.Theimmunohistochemistrydata in our study lend further support for this interpretation; BMP labeling was foundthroughouttheretina,choroidandadjacentsclera.Theseprofilesarealsoconsistentwithsecretion of BMPs by RPE as part of a paracrine signaling pathway, modulating as yetunknownocularfunctions.Inthecontextofoculargrowthregulationandmyopia,themostsignificantfindingsfromour study were the optical defocus, sign‐dependent bidirectional changes in geneexpressioninchickRPE.Specifically,BMP2,BMP4,andBMP7geneexpressionallshowedsignificantup‐regulation inresponse to imposedmyopicdefocus (+10D lens treatment),whileexpressionofthesegeneswasdown‐regulatedwithimposedhyperopicdefocus(‐10D lens treatment). These patterns of up‐ and down‐regulation of BMP expressioncorrespond to slowed and accelerated ocular elongation, respectively. In the case of allthreeBMPs(exceptforBMP7with+10Dtreatment),changesingeneexpressionwereseensoonafter treatmentswere initiated (i.e., at2h).Therapidonsetof thegeneexpressionchangesarecompatiblewithrolesforBMP2,BMP4andBMP7ininitiatingdefocus‐driveneye growth changes, and also tend to rule out the possibility that the changes were abyproductofalteredoculargrowth,whichwouldhavebeenminimalatthistime.Withbothlenstreatments,theearlychangesinexpressionofthesethreeBMPspersistedoutto48hoftreatment.Thatthesamepatternsofdifferentialgeneexpressionwerestillevidentafter48hofnegative lenswear,wheneyesweregrowing faster thannormal,asevident fromultrasonography data, suggest a further role for BMP2, BMP4, andBMP7 inmaintainingthis altered growth pattern. These prolonged changes in gene expression tend to argueagainstthembeingsimplyresponsestothealteredvisual(retinaldefocus)conditions.Notethattheapparentreductioninthemagnitudeofthechangeafter48hcomparedto2hoftreatmentisatleastpartlyaproductofyokedchangesinthefelloweyeatthelattertime

46

point (see Figure 2‐13). However, the mechanisms underlying the regulation of BMPexpressioninRPEremainlargelyunknownanditsroleinpostnataleyegrowthregulationisyettobedirectlydemonstrated.WhilehighbasallevelofBMPexpressionintheRPE,asobserved,isnotanecessarypre‐requisiteforbidirectionalchangesinexpression,itcouldplausibly extend the range of response, although this point has not been emphasized inrelevantpreviousgeneexpressionstudies.Of the few other molecules known to be bidirectionally regulated in the eye by opticaldefocus,ZENK,animmediateearlygene,hasbeenshowntoundergoopticallymodulatedexpressionchanges in retina.51,58,117,118,120 For example, thenumberofZENK‐expressingglucagonamacrinecellswasfoundtobeincreasedwithpositivelenses,afteraslittleas30minutesofwear,anddecreasedafter2hoursofnegativelenswearinchicks.Itremainstobe determined whether or not these cells are part of a signal pathway mediating theobserved changes in BMP expression in the RPE. It is possible that BMP expression isregulatedbyanindependent,yet‐to‐beidentifiedretinalcellpopulation.Itisalsonoteworthythatretinoicacid(RA),whichhasbeenputforwardasapotentialeyegrowth regulator, has also been linked to the regulation of BMP expression in otherstudies.270‐273 The data tying RA with eye growth regulation in chicks also exhibitbidirectionality; retinal RA levels are increased in eyes wearing negative lenses anddiffusersandlevelsaredecreasedineyeswearingpositivelenses,withtheoppositetrendsbeingtrueforchoroidalRAlevels.117‐120WhileRAcouldbeactingup‐streamfromBMPsontheretinaland/orchoroidalsideofRPE‐thesepossibilitiesarenotdistinguishablebasedoncurrentlyavailabledata‐asignalpathwaylinkedtoeyegrowthregulationwouldarguefor anupstream retina‐RA,RPE‐BMPassociation.Nonetheless, retina, choroid and scleraareallplausiblesitesofactionofBMPs,basedonourimmunohistochemistryandWesternblotresults.Our results raise the possibility that BMPs are themselves negative growthmodulators,acting on the choroid and/or sclera. Roles for BMP2 and BMP4 as growth inhibitors, assuggestedby thesegeneexpressionprofiles,are in linewith theresultsofanotherstudydescribing BMP2 and BMP4 as negative growth regulators.243 In two separate studiesinvolving cultured human scleral fibroblasts, BMP2 was shown to promote cellproliferationanddifferentiationandalso alter the expressionof keyextracellularmatrix(ECM)genesandproteins.207,208TheadditionofexogenousBMP4isreportedtoaltereyeshapeandreduceeyesizeinthemouseembryo.266TherehasbeennoequivalentstudyinchicksalthoughintraocularinjectionofBMP4,givenbeforetreatmentwithaneurontoxin,is reported tosuppress theproliferationofMüllergliaand thusreduceretinadamage inyoung chicks.244Assuming thatdifferentially expressedBMPgenes inRPEare translatedintoprotein,andsubsequentlysecretedintothechoroid,theseBMPsmayaffectchangesinoculardimensions throughchanges inECMremodeling ineitherorboth thechoroidandsclera.239,274Local tissue‐specificmanipulationsofBMP levelsmayberequired todissectthisapparentlyverycomplexsignalingcascade(Figure2‐20).255

47

Figure2‐16.CartoonsummarydiagramshowingBMPgeneexpressionchangesinRPEwithlenstreatments.WhileBMP2,BMP4andBMP7arecloselyrelatedstructurally,andallthreeBMPsactivatethesamesubtypesofBMPreceptors, studiesofembryoniceyedevelopment in thechickdescribespatiallyandtemporallydistinctexpressionpatternsfortheseBMPs.235,239,241,270,275,276Nonetheless, the latter observations are consistentwith general behavior of thesegrowth factors; their effects on developmental regulation, cell proliferation anddifferentiationcanbesimilarordifferent,dependingonthedevelopmentalstageandcelltype.208,243,277TheclosesimilarityoftheRPEgeneexpressionprofilesofBMP2,BMP4andBMP7fortheopticaldefocusconditions imposedinourstudiessuggeststhatthesethreeBMPs are co‐regulated and/ormayplay similar roles in refractive error andeye growthregulation.Interestingly, BMP2 gene expression was reported to be down‐regulated in chickretina/RPEwithform‐deprivationmyopia,133inthesamedirectionasthatinducedbyour

48

negativelenstreatment,whichalsoinducesmyopia.Whilethereisaccumulatingevidencethatthemechanismsunderlyingthesetwotypesofmyopiaaredifferent,104,278ourresultsaddtootherdatasuggestthatsomecomponentsoftheregulatorypathwaysareshared.33,51Astheeffectsonretinal imagequalityof thesetreatmentsaregenerallyquitedifferentandthuslikelytoelicitdifferentretinalresponses,wespeculatethattheRPEwasthesiteofBMP2 gene expression changes in the previous form‐deprivation study, with the RPEservingasaconduitorpointofconvergenceofdifferentretinalsignalpathways.Inthecurrentstudy,wefoundnoevidenceofdefocus‐dependentdifferentialregulationofBMP receptor expression, althoughwe did observe yoked down‐regulation of BMPR2 inresponse to the negative lens treatment. The experimental myopia literature containsmanyexamplesofinterocularyokinginvolvinggeneexpression,aswellasrefractiveerrorandoculargrowthratechangesinresponsetolensesanddrugtreatments.35,51,58,197,279‐281FurtherexamplesareprovidedbyourfindingsforBMP4andBMP7,thatgeneexpressionwassignificantlydown‐regulatedinboth‐10Dlens‐treatedeyesandtheirfellowsafter48h of lenswear. A hint of similar yoking for BMP2with ‐10 D lenseswas also observed,although the effect was not statistically significant (p = 0.077). This interocular yokingphenomenon was not evident with either the shorter 2 h exposure for the ‐10 D lenstreatment or with either the 2 or 48 h + 10 D lens treatments. While the mechanismsunderlyingsuchinterocularyokingremaintobeelucidated,thereappeartobetreatment‐and gene‐dependent differences that need to be accounted for. For example, for BMP2,BMP4andBMP7, theyokingeffectsonRPEgeneexpressionwas limited toournegativelenstreatment,whileyokingofZENKproteinexpressioninretinahasbeenreportedwithbothpositiveandnegativelenstreatments,aswellaswithrecoveryfromform‐deprivationmyopia.51,53 In a previous study, we found that sectioning of the optic nerve in chickseliminated the inhibitory effect ofmonocular atropine on lens‐inducedmyopia in felloweyes.282Whileneuralfeedbackloopsmayalsobeinvolvedintheseyokedgeneexpressionchanges,ahumoralmechanismisalsoplausible.51Although it is now generally accepted that both genetic and environmental factors arelikelytocontributetothedevelopmentofhumanmyopia,thereappearslittleoverlapinthegenes implicated in humanmyopia and those linked to eye growth regulation in animalmodelsofmyopia.11OurfindingsimplicatingBMP2andBMP4ineyegrowthregulationinthe chick represents a departure from this trend and a potentially important break‐through, with both BMP2 and BMP4 being recently put forward as candidate genes forhumanmyopia.249,283 The recentmeta‐analyses of human genetic studies using genome‐widesingle‐nucleotidepolymorphism(SNP)identifiedBMP2asanewlocusforrefractiveerror and myopia.283 The latter study also reported a mutation in BMP4 in a range ofsubjectspresentingwithanophthalmia,microphthalmia,ormyopiaalthoughacausallinkisyettobeestablished.Intermsofanimalstudies,alsoofpotentialrelevanceisthefindingin zebrafish of severemyopia linked to amutation of low density lipoprotein receptor‐related protein 2 (lrp2),which serves as an endocytic receptor for BMP4, in addition tootherbioactivemolecules.284Theseobservations lendweight toourdata suggesting thatBMP4playsanimportantroleineyegrowthregulation.

49

In summary,we demonstrated the expression of genes for BMP2, BMP4, andBMP7 andtheirreceptors,aswellasofthethreeBMPproteinsthroughoutthetissuesmakinguptheposterior ocular wall, implying that they serve important regulatory functions in thesetissues.KeyoculargrowthmodulatoryrolesforthesegrowthfactorsandakeyroleoftheRPEasasignal transducerorrelay intherelatedretino‐scleralcascadearesuggestedbythe further observations of sign‐dependent, optical defocus‐induced changes in geneexpressionintheRPEforallthreeBMPs–BMP2,BMP4andBMP7.Thelatterresultsalsoraise the possibility of targeting these BMP signaling pathways as a novel therapeuticinterventionformyopia.

50

Chapter3Long‐TermImposedHyperopicDefocusInducedGeneExpressionChangesinChickRPE:aMicroarrayStudyAbstractThisstudyexaminedtheeffectongeneexpressionintheretinalpigmentepithelium(RPE)of long‐term, lens‐induced myopia. To induce myopia, 5 White‐Leghorn chicks woremonocular‐15Dlensesfrom10‐daysofagefor38days.Retinoscopyandhigh‐frequencyA‐scan ultrasonography were used to monitor refractive errors (RE) and axial oculardimensionsofbothtreatedanduntreatedfelloweyes,whichservedascontrols.RPEwasisolatedfromenucleatedeyesandtotalRNApurified.AftertheverificationofthequalityofRNA,itwassubjectedtocDNAsynthesis,invitrotranscription,labeling,andhybridizationwithAffymetrixGeneChipChickenGenomeArrays.MicroarraydatawereanalyzedusingBioConductor package and bioinformatic databases. Seventeen candidate genes werevalidatedwith real‐time PCR. Expression patterns for RPE frommyopic and fellow eyeswere compared. Significant myopia, axial length (AL) increases, and choroidal thinningwererecordedintreatedeyescomparedtotheirfellowsafter38daysoftreatment.Eighthundred and fifty‐two transcripts were up‐ or down‐regulated in myopic compared tofellow eyes, by at least 1.5‐fold (p<0.05). Real‐time PCR confirmed the differentialexpressionofmanygenes, including theup‐regulationof FIGF (p = 0.053),NOG,PDGFA,and down‐regulation of BMP2, 4, 7, DRD4, RARB, and RRH. Of the genes exhibitingdifferential expression,many could plausibly be involved in ocular growth regulation inRPE, based on their known actions in other tissues, although functional changes in RPEsecondarytomyopicgrowthmayunderliethechangesinsomeaffectedgenes.Some of the data reported in this chapter were previously published in the followingconferenceabstract:

ZhangY,LiuY,XuJ,NimriN,WildsoetCF.MicroarrayanalysisofRPEgeneexpressioninchicksduring long‐termimposedhyperopicdefocus.Invest.Ophthalmol.Vis.Sci.2010;51:E‐Abstract3680.

51

3.1 IntroductionMyopiadescribesthecondition inwhich imagesofdistantobjectsare focused in frontoftheretina,duetoamismatchbetweentheopticalpowerandaxiallengthoftheeye.Mostmyopia is caused by excessive ocular elongation,285 leading in high myopia to sight‐threatening complications such as retinal degeneration, retinal detachment, choroidalneovascularization, myopic maculopathy, cataract, and glaucoma.6, 286, 287 In a recentlypublishedstudy, theoverallprevalenceofmyopia in theUSwasgivenas41.6%,andtheprevalence of high myopia (≤ ‐7.9 D) as 1.6% for persons aged 12 to 54 years.9 Thesefigures represent significant increases over previous figures,9, 288 mirroring trends inequivalent statistics fromEastAsia.232,289‐291 Because of these altered prevalence figuresand the complications of highmyopia, it has become a significantpublic health problemworld‐wide. For effective clinical intervention, it is thus imperative to elucidate themechanismsunderlyingthedevelopmentofmyopiaandassociatedcomplications.Studiesusinganimalmodels,suchaschickens,guineapigs,treeshrews,andmonkeys,haveprovidedsignificantinsightsintopostnataleyegrowthregulation,andspecifically,theroleoftheretinaasasourceofmyopia‐generatingstimuli.16,26,292Myopiacanbegeneratedintwo different ways, using diffusers to impose form‐deprivation or negative lenses toimpose hyperopic defocus. In both cases, the result is an increase in the rate of axialelongation.Oftheaboveexperimentalmodels,chickensareboththemostwidelyusedandmost reliable, responding rapidly to imposedhyperopicdefocus,with choroidal thinningbeing detectible within 1 hour of the initiation of lens treatment, ahead of detectiblechangesinoveralleyelength.179,180Withimposedopticaldefocus,thegrowthchangesarecompensatory,withemmetropiabeingachievedwiththelensesinplace,providedthesizeofthedefocusstimulusisnottoolargeortheanimalstooold.Inyoungchickens,around9daysold,nearcompletecompensation(80%)isseenafterjustoneweekoftreatmentwith–10Dlenses.226Compensationoccursmoreslowlyinolderbirds.35,227Inallcases,treatedeyes will continue to grow at a faster rate with the lenses in place, to maintaincompensationafterithasbeenachieved.Since localizedmanipulation of retinal images induces localized changes in the adjacentsclera, and optic nerve section, which prevents communication between the retina andbrain,doesnotpreventtheinductionofexperimentalmyopia,itisgenerallyacceptedthatlocal ocular circuits underliemyopic growth, and that the retina is the origin of growthmodulatingsignals.36,37Becausetheretinaisseparatedfromthetwotargettissues,thechoroidandsclera,bytheretinalpigmentepithelium(RPE),itislikelytoplayanimportantroleasasignalrelay.RPEcells express many different types of receptors, for both neurotransmitters and growthfactors,andalsosecretesavarietyofgrowth factorsandcytokines,manyofwhichcouldplausiblybecomponentsofagrowthmodulatorysignaltransductionpathway,and/orbeaffected by excessive ocular enlargement.40,65,66,135 As the RPE is increasingly stretchedandthinnedduringthisenlargementprocess,itisalsoplausiblethatitisadverselyaffected

52

inthisprocess,andsocontributestosomeoftheocularcomplicationsofmyopia,suchaschoroidalneovascularizationandmyopicmaculopathy.293‐295DNAmicroarraytechnologymakesitpossibletoscreenalargenumberofgenesinasingleexperiment,tolookforeffectsofappliedtreatments.296To‐date,theuseofthistechnologyinmyopiaresearchhasbeenlargelylimitedtoretinaorretina/RPE,withrelevantstudiesinmice,chickens,andmonkeys, involvingretinal imagecontrastmanipulation(e.g. form‐deprivation), as well as optical defocus.127, 133, 161, 297‐299 In these case, RPE was notseparatedfromtheadjacentretinaorchoroid.In the study described here, chickens wore monocular ‐15 D lenses over an extendedperiod, i.e.38days,to induce long‐termhighmyopia,afterwhichRPEcellswere isolatedforgeneexpressionstudies.Withthisexperimentalprotocol,changesingeneexpressioninthe RPEmay occur as a part of the signal cascade underlyingmyopic “growth”, or as aconsequence of changes in RPE secondary to the excessive expansion of the vitreouschamber. 3.2 MaterialsandMethods3.2.1 Animals&LensTreatmentWhite‐Leghornchickenswereobtainedashatchlingsfromacommercialhatchery(Privett,Portales,NM),andwererearedina12h:12hlight‐darkcycle.Five(5)birdsintotalwereused for this study. Four birds (1, 2, 3, 4)were used for themicroarray screening. RNAfrom an additional bird (5) was included in follow‐up qPCR validation experiments. Toinducemyopia, chickswore ‐15D lensesover their righteyes from10‐dayofage for38days. Their contralateral (fellow) untreated eyes were used as controls. Lenses werecleaned twice a day. Both eyes of each bird had refractive errors (RE) and axial oculardimensions, including vitreous chamber depth (VCD) and choroidal thickness (CT),measured under isoflurane anesthesia (1.5% in oxygen) with retinoscopy and high‐frequencyA‐scanultrasonographyrespectively,after18and38daysoflenswear.3.2.2 RPEIsolation&RNAExtractionTheRPEwascollectedfrombotheyesat least2hafter the lastbiometrymeasurements.Birdswereeuthanizedwithanoverdoseofpentobarbitalinjection,aftereyesimmediatelyenucleated, anterior segments removed, RPE isolated and lysed with RLT buffer fromRNeasyMinikits(Qiagen,Valencia,CA),andthenimmediatelystoredat‐80°Cforlateruse.RNA was extracted from RPE cells and purified using RNeasy Mini kits and RNeasyMinElute Cleanup kits (Qiagen, Valencia, CA), according to the manufacturer’s protocol.RNAconcentrationandopticaldensity ratioofA260/A280weremeasuredusingaDU650Spectrophotometer (Beckman Coulter, Fullerton, CA). The quality of extracted RNAwasfurther analyzed in theFunctionalGenomicLab at theUniversity of California,Berkeley,usinganAgilent2100bioanalyzerincombinationwithanRNA6000PicoChipkit(AgilentTechnologies,Inc.,SantaClara,CA).300‐302

53

3.2.3 MicroarrayAnalysescDNA synthesis, in vitro transcription, and microarray hybridization: AffymetrixGeneChip ChickenGenomeArrays (Affymetrix, Santa Clara, CA)were used in this study.First, cDNAwas synthesized from1.4 μg samples of RNA from each eye of the 4 chicks,using SuperScript One‐Cycle cDNA Kit (Invitrogen, Carlsbad, CA), according to themanufacture’sprotocol.Thesynthesizeddouble‐strandedcDNAwas thencleanedupandused as a template to synthesize biotinylated complementary RNA (cRNA) in an invitrotranscription (IVT) reaction.The synthesized cRNAwas then cleanedup, quantified, andfragmentedinto35‐200basefragments.Finally,30μgsamplesoffragmentedcRNAwerehybridizedontothechips,followedbywashing,stainingandscanning.Microarray data analysis:Raw microarray data (.CEL files) were analyzed using bothBioconductorprojectRandAffypackages.303‐305Robustmultiarrayaveraging (RMA)wasused as a normalizationmethod.306,307 TheAffy package includes a suite for probe‐levelquality control checks of signal intensity and variance. Boxplots of perfect match (PM)probedataprovidetheinformationoftheoveralldistributionofprobeintensitiesoneacharray. Three quality control (QC) analyses were performed to check array qualities.RelativeLogExpression(RLE)representsthelogratiooftheexpressionofaprobesetonan individualarray to themedianexpressionof thisprobesetacrossallarrays.TheRLEprovides information about the overall distribution of probe intensity. NormalizedUnscaledStandardError(NUSE)representstheratioofexpressionofthestandarderrorofanindividualprobesetandmedianstandarderrorforthisprobesetacrossallarrays.TheNUSEprovidesinformationabouttheoveralldistributionofprobevariance.RLEandNUSEdata for each eyeare showngraphically asboxplots, as isRNAdegradationdata,whichshowsmean expression of all probe sets as a function of relative position (5’ end to 3’end).302,308IdentificationofdifferentiallyexpressedgeneswasbasedonanMAplot,whichshows M (log2 fold‐change) plotted as a function of mean expression level A (log2intensity).309Foreachgene,thelog2foldchangeinexpressionwasderivedfromtheratiooftreatedtocontroleyedata.Thecriteriaappliedinselectingcandidategenesforexportwere fold≥1.5 andp ≤0.05.Up‐regulationby≥1.5 fold is equivalent toM≥0.585anddown‐regulationby≤‐1.5foldisequivalenttoM≤‐0.585.Identifieddifferentiallyexpressed(candidate)geneswereanalyzedusingNetAffx,310withfunctional annotations obtained from the Database for Annotation, Visualization andIntegrated Discovery (DAVID), and NCBI databases (GenBank, PubMed).311 Canonicalpathway information was obtained using Ingenuity Pathways Analysis software(Ingenuity® Systems, www.ingenuity.com).133, 299 Heatmaps were prepared for 55candidategenesusingMultiExperimentViewer(http://www.tm4.org/mev.html).312,3133.2.4 Real‐TimePCRFromthecandidategenesidentifiedinthemicroarrayanalyses,17geneswereselectedforfurther real‐time PCR validation study. Using extracted RNA from RPE, reversetranscription was performed to synthesize cDNA (SuperScript III First‐Strand SynthesisSystem for RT‐PCR, Invitrogen, Carlsbad, CA). Custom‐designed primers and QuantiTect

http://www.tm4.org/mev.html�

54

SYBRGreenPCRKits(Qiagen,Valencia,CA)wereusedincombinationwithaStepOnePlusReal‐TimePCRSystem(AppliedBiosystems,FosterCity,CA).PrimersweredesignedusingPrimer Express 3.0 (Applied Biosystems, Foster City, CA). Chick glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as the housekeeping gene. Gene symbols,NCBI (National Center for Biotechnology Information, Bethesda, MD) access numbers,sequences, and sizes of amplicons of candidate genes are listed in Table 3‐1. The PCRcycling conditions included an initial denaturation for 10 min at 95 °C, followed by 40cycles of denaturation for 15 s at 95 °C, annealing for 1 min at 60 °C. Ten‐fold serialdilutions of cDNA were used for generating standard curves for each pair of primers.Amplificationofeachgenewasperformedintriplicate.Meannormalizedexpression(MNE)valueswere used to compare gene expression levels in RPE from lens‐treated eyes andtheirfellowcontroleyes.3.2.5 StatisticalAnalysisPairedStudent’st‐testwereusedtocomparelens‐treatedeyeswiththeirfellows,bothinterms of biometric data (axial length, vitreous chamber depth, choroidal thickness,refractiveerror),aswellasgenearrayandreal‐timePCRdata.

Table3‐1.Primerinformationforcandidategenes

Gene NCBIAccessNumber Sequences(5’‐3’)

Amplicon

AQP4 NM_001004765.1 Forward:5’‐TGGGAGTCACTGCGGTACAC‐3’Reverse:5’‐TGAATCACAGCTGGCAAAAATAG‐3’

107bp

BMP2 NM_204358 Forward:5’‐AGCTTCCACCACGAAGAAGTTT‐3’Reverse:5’‐CTCATTAGGGATGGAAGTTAAATTAAAGA‐3’

96bp

BMP3 NM_001034819.1 Forward:5’‐GGAATGAGCCACGGTATTGTG‐3’Reverse:5’‐GCCAATGTCAGCAAAATCCA‐3’

59bp

BMP4 NM_205237.1 Forward:5’‐GCACAGACTCATCAGGGCAAA‐3’Reverse:5’‐GCCGTGCCCCTGAGGTA‐3’

60bp

BMP6 XM_418956 Forward:5’‐TCATGGTGGCATTCTTCAAAGT‐3’Reverse:5’‐TGCCGCTGACCTCGTAGTC‐3’

59bp

BMP7 XM_417496.2 Forward:5’‐CGGGAATTTGGAAATCAGTCA‐3’Reverse:5’‐GCCATCTGGTCTGGATTTGG‐3’

66bp

DRD4 NM_001142849.1 Forward:5’‐GCTCAAGACCACCACCAACTATT‐3’Reverse:5’‐GGAGGGCGAGCAGAAGGT‐3’

65bp

FGF1 NM_205180.1 Forward:5’‐TCCTGTATGGCTCGCAGCTA‐3’Reverse:5’‐TGGAGATGTATGTGTTGTAATGGTTCT‐3’

84bp

FGFR2 NM_205319.1 Forward:5’‐ACACGTAGAGAGGAATGGCAGTAA‐3’Reverse:5’‐AACACCGGCAGCCTTTAAAA‐3’

76bp

FIGF NM_204568.1 Forward:5’‐CTCATGGGTCACAAAGCAAAGA‐3’Reverse:5’‐GCAGTGGATTTCTGGAGAACTGA‐3’

65bp

GAPDH NM_204305.1 Forward:5’‐AGATGCAGGTGCTGAGTATGTTG‐3’ 71bp

55

Reverse:5’‐GATGAGCCCCAGCCTTCTC‐3’

KDR NM_001004368.1 Forward:5’‐TCTGGCGCTCACCAATACC‐3’Reverse:5’‐ACACCACCTCCCCCACCTAT‐3’

67bp

NOG NM_204123.1 Forward:5’‐CAGAAGGCATGGTCTGCAAA‐3’Reverse:5’‐CGCCACCTCAGGATCGTTA‐3’

58bp

PDGFA NM_204306.1 Forward:5’‐GTGCACTAGACCGCGATGAG‐3’Reverse:5’‐GGACAGGTAGCCAAAACCTATCA‐3’

63bp

PDGFRA NM_204749.1 Forward:5’‐GCCTGAGAATGAAAGGGAGAGA‐3’Reverse:5’‐CAACGCTGAGATGCTCATAAAGTAC‐3’

85bp

RARB NM_205326.1 Forward:5’‐CCACCCCCGCGTGTTTA‐3’Reverse:5’‐AAAGCCCTTACATCCCTCACAA‐3’

96bp

RRH NM_001079759 Forward:5’‐GGGCGTCCATGCCTACTGTA‐3’Reverse:5’‐AGTTGCTCCAGTCGGATCTGA‐3’

59bp

TGFB2 NM_001031045.1 Forward:5’‐CGCCTGCAGAACTCAAAGG‐3’Reverse:5’‐TTCAGAACCTGGTACAGCTCTATCC‐3’

62bp

3.3 Results3.3.1 OcularEffectsofLensTreatmentMonocular‐15Dlenses,wornforanextendedperiodoftime,inducedchoroidalthinningandexcessiveocularelongationthatwaslargelylimitedtothevitreouschamber(Figure3‐1),andasaconsequenceoftheincreaseinocularlength,treatedeyeswerehighlymyopic.Attheendofthelenstreatmentperiodof38days,treatedeyeshadameanrefractiveerrorof‐14.1±0.60D,comparedto+2.5±0.29D(mean±SEM)fortheirfellows,thisdifferencebeinghighlysignificant(p<0.01,pairedStudent’st‐test).Instatisticalterms,treatedeyeshad significantly longer anterior chamber depths ([distance from anterior cornea toanteriorlens]2.49±0.10mmvs.2.03±0.03mm,p<0.01),vitreouschamberdepths(8.63± 0.19 mm vs. 6.91 ± 0.14 mm, p < 0.001 [interocular difference is 1.72mm]), thinnerchoroids(0.18±0.02mmvs.0.24±0.01mm,p<0.05),and,longeroveralllengths(axiallength:14.50±0.24mmvs.12.56±0.17mm,p<0.001,pairedStudent’st‐test[interoculardifferenceis1.94mm]).

3.3.2 RNAQualityAnalysis&cRNAQuantificationTheyieldoftotalRNAfromRPEwasapproximately1.5‐2.8μg/eye,andtheopticaldensityratio of A260/A280 was 1.76 ± 0.02. RNA integrity number (RIN), which measures theintegrityoftotalRNAusingtheelectrophoretictraceoftheRNAsampleonanRNA6000PicoChip,was8.27±0.10.TheconcentrationofcRNAwas2.47±0.11μg/μl.Thesedataareconsistentwithhighyieldandhighintegrity.

56

Age of Birds

Figure3‐1. Interoculardifferences(treated‐controleyes), inaxial length,vitreouschamberdepth(leftaxis),andchoroidalthickness(rightaxis),after18and38daysofcontinuous‐15Dlenstreatment(mean±SEM,n=7).

3.3.3 MicroarrayAnalysesQuality control analyses of microarray data: In the boxplot showing Relative LogExpression(RLE) foreacharray(representing4treatedeyesandtheir fellows;Figure3‐2A),allboxesarecenteredcloseto0,anddistributionsofprobeintensityofallarraysaresimilartoeachother.IntheboxplotofNormalizedUnscaledStandardError(NUSE)(Figure3‐2B), themedian standard error lies close to 1 for all arrays. In RNA degradation plot(Figure3‐2C),which represents the averagedegradation for the completeprobe set, thesignal intensity ratio for the housekeeping gene 3' to 5' probe sets was ≤ 3.0. Theconsistency across all 8 arrays in their RLE and NUSE profiles, as evidenced from theboxplots, and the similarity of their RNA degradation profiles validates the comparisonssubsequentlymadewithingenesacrossthearrays.305,308,314Differentiallyexpressedgenes:TheMA plot (Figure 3‐3) shows a broad spread in thedata,withmanypoints showingagreater than1.5 fold change inexpression inabsoluteterms,implyingthatmanygeneshadbeensignificantlyup‐ordown‐regulatedbythelenstreatment.Moregeneswereup‐regulatedthandown‐regulated.Specifically,atotalof851transcripts out of 32,773 chick transcripts or 670 geneswere up‐ or down‐regulated bymore than 1.5 fold by the lens treatment (p ≤ 0.05), and of the differentially expressedgenes,520wereup‐regulatedand150weredown‐regulated.

57

Figure 3‐2. Quality assessment of samplesusedinmicroarrayanalysis:A)RLEplot,B)NUSEplot,C)RNAdegradationplot.InRLEplot, all boxes are centered close to 0. InNUSE plot, the median standard erroracross all arrays is close to 1. In RNAdegradation plot, signal intensity ratio ofthehousekeepinggene3'to5'probesetsis≤3.0. InFigs2A&B,X‐axis labels refer tosamples frompairsofeyes (right,R& left,L) from individual birds, identified by thenumbercode.

Figure3‐3.PlotofM(log2fold‐change)asafunctionofmeanexpressionlevelA(log2intensity).Datafrom8arraysrepresenting4treatedand4controleyesareshown.Foldchangerepresentstheratioofexpressionintreated and control eyes. The superimposed redlines indicate a 1.5 fold change in expression; genes lyingoutsidethecentralzonedefinedbytheredlineswereconsideredtobesignificantlyup‐ordown‐regulated.309

58

In further analyses of the 851differentially expressed transcripts, usingNetAffx, DAVID,and NCBI databases to classify them, particular attention was paid to genes that hadalreadybeen linked toeithermyopiaoreyegrowth regulation inother studies,or couldplausiblyplaya role inoculargrowth regulationormyopiapathogenesis,basedon theirknownfunctions.SelectedgenesarelistedinTable3‐2,clusteredintogroupsaccordingtothe gene ontology terms for molecular function. Among the up‐regulated genes were anumber of growth factors (BMP3, BMP6, FIGF, INHBA, PDGFA, TGFB2), a variety ofreceptors (EDNRA, FGFR2, KDR, PDGFRA), protein/lipid binding molecules (FN1, NOG),andwater/ionchannels (AQP4,CACNA1G).Amongthedown‐regulatedgeneswereothergrowth factors (BMP2, BMP4, BMP7), receptors (ADRA2C, DRD4, RARB, RRH), and ionchannels(CLCN5,CLCN7).A heatmap of 55 candidate genes (Table 3‐2) that could plausibly regulate RPE andchoroidal/scleralfunctionandoculargrowthisshowninFigure3‐4.3.3.4 Real‐TimePCRTreatment‐induced changes in expression aswell as results of statistical analyses of thePCRdatacollectedforninegenesaresummarizedinTable3‐3.Theseninegenesrepresentasubsetof theseventeencandidategeneswhosemicroarrayresultswerevalidatedwithreal‐time PCR (Figure 3‐5). A comparison of gene expression changes from both themicroarrayandreal‐timePCRexperimentsareshowninFigure3‐6.Allninegenesshowedsimilarexpressionpatternswiththetwotechniques.Specifically,withreal‐timePCR,FIGF,NOG,andPDGFAonceagainshowedup‐regulationintreatedrelativetocontroleyes,andBMP2,BMP4,BMP7,DRD4,RARB,andRRHallshoweddown‐regulation.

3.4 Discussion

In this study, we found in eyes made highly myopic 851 transcripts in RPE to bedifferentiallyexpressedusingmicroarraysandconfirmedtheobservedtrendsforasubsetofninegenesusingreal‐timePCR.Belowwediscussthepossiblegrowthregulatoryrolesofthe lattergenes,basedonalreadyestablished links toeyegrowthregulationorplausiblelinks,basedontheirfunction.Becauseoftherelativelylongtreatmentperiodusedinthisstudy,andrelativelyhighmyopicerrorsinduced,wealsoconsiderthepossibilitythatthechanges ingeneexpressioncouldbeaproductof excessiveeyeelongation, secondary tostretchingandthinningofRPE,andthusperhapslinkedtotheretinalpathologydescribedinmyopiceyes.One family of genes selected for further investigation were the bone morphogeneticproteins (BMPs),which belong to the transforming growth factor β (TGF‐β) superfamilythathasbeenlinkedtoeyegrowthregulationinotherstudies.TheBMPsrepresentmulti‐functional growth factors, first linked to cell proliferation and extracellular matrixsynthesis inboneandcartilage formation,236,261,262,315,316andnowknown toplaycrucialrolesinembryonicmorphogenesis,aswellaspostnataldevelopmentandcellularfunctionsinadultanimals.235,236,246,248,317,318

59

Table3‐2.Listofgenes,showingsignificanttreatment‐relateddifferentialexpressionchangesinRPEinmicroarrayanalyses,andwithplausiblelinkstoeyegrowthregulationand/ormyopia.

ProbeSetID

GeneSymbol

GeneTitle

Fold

PValue

GO:0015267~channelactivity Gga.11374.1.S1_at AQP4 aquaporin4 1.56 0.022

GgaAffx.4763.4.S1_s_at

CACNA1G

calciumchannel,voltage‐dependent,Ttype,alpha1Gsubunit

1.76

0.043

Gga.1805.1.S1_at

CLCN5

chloridechannel5(nephrolithiasis2,X‐linked,Dentdisease)

‐1.56

0.004

GgaAffx.292.1.S1_s_at CLCN7 chloridechannel7 ‐1.80 0.019GO:0008083~growthfactoractivity Gga.3950.1.S2_at BMP2 bonemorphogeneticprotein2 ‐6.47 0.005

GgaAffx.6892.1.S1_at BMP3 bonemorphogeneticprotein3(osteogenic) 2.41 0.049

Gga.686.1.S1_at BMP4 bonemorphogeneticprotein4 ‐3.15 0.002

GgaAffx.23995.1.S1_at BMP6 bonemorphogeneticprotein6 3.05 0.05

Gga.6770.2.S1_s_at BMP7 bonemorphogeneticprotein7 ‐1.97 0.006

Gga.3219.1.S1_at

FIGF

c‐fosinducedgrowthfactor(vascularendothelialgrowthfactorD)

2.83

0.012

Gga.648.1.S2_at FGF1 fibroblastgrowthfactor1(acidic) 1.91 0.032

Gga.3982.1.S1_at INHBA inhibin,betaA 1.96 0.045

Gga.3899.3.S1_a_at

PDGFA

platelet‐derivedgrowthfactoralphapolypeptide

1.87

0.041

GgaAffx.22982.1.S1_at TGFB2 transforminggrowthfactor,beta2 1.86 0.023GO:0005515~proteinbinding Gga.4285.1.S1_at CEBPB CCAAT/enhancerbindingprotein(C/EBP),beta ‐1.82 0.007

Gga.4941.2.S1_a_at,Gga.4941.1.S1_at

CDH11

cadherin11,type2,OB‐cadherin(osteoblast)

2.35

0.0006

Gga.1917.1.S1_at,GgaAffx.21844.1.S1_s_at

CDH2

cadherin2,type1,N‐cadherin(neuronal)

2.01

0.045

GgaAffx.6053.1.S1_at COL1A2 collagen,typeI,alpha2 2.87 0.031

Gga.2592.1.S1_at COL5A1 collagen,typeV,alpha1 2.10 0.015

GgaAffx.21771.1.S1_at COL6A1 collagen,typeVI,alpha1 2.49 0.011

Gga.4257.1.S1_at COL6A2 collagen,typeVI,alpha2 2.13 0.008

Gga.4965.2.S1_at COL12A1 collagen,typeXII,alpha1 2.06 0.03

Gga.17860.1.S1_at DOK7 dockingprotein7 1.92 0.015

Gga.9293.1.S1_at FN1 fibronectin1 3.12 0.048

Gga.5710.1.S1_a_at HTRA3 HtrAserinepeptidase3 3.29 0.013

Gga.8434.1.S1_at IGFBP4 insulin‐likegrowthfactorbindingprotein4 1.66 0.043

Gga.19049.1.S1_at

IL18

interleukin18(interferon‐gamma‐inducingfactor) 1.68

0.037

Gga.3199.1.S1_at

MMP2

matrixmetallopeptidase2(gelatinaseA,72kDagelatinase,72kDatypeIVcollagenase) 2.73

0.009

Gga.449.1.S1_at NOG noggin 4.28 0.036

GgaAffx.25896.1.S1_at,SLC1A3 solutecarrierfamily1(glialhighaffinity

4.37

0.039

60

GgaAffx.25896.1.S1_s_at glutamatetransporter),member3GO:0004872~receptoractivity Gga.8371.1.S1_s_at ADIPOR2 adiponectinreceptor2 ‐1.56 0.004

GgaAffx.9949.1.S1_at ADRA2C adrenergic,alpha‐2C‐,receptor ‐2.09 0.05

Gga.2733.1.S1_at CNR1 cannabinoidreceptor1(brain) 2.05 0.008

GgaAffx.3186.1.S1_at DRD4 dopaminereceptorD4 ‐1.50 0.003

Gga.137.1.S1_at EDNRA endothelinreceptortypeA 2.21 0.046

Gga.694.1.S1_at EPHB1 EPHreceptorB1 1.57 0.011

Gga.633.1.S1_at EPHB6 EPHreceptorB6 1.82 0.049

Gga.11368.1.S1_at F2RL2 coagulationfactorII(thrombin)receptor‐like2 ‐2.10 0.003

Gga.14338.1.S1_at FGFR2 fibroblastgrowthfactorreceptor2 1.58 0.043

Gga.3907.1.S1_at LOC395603 fibroblastgrowthfactorreceptorFREK ‐1.69 0.001

Gga.8771.1.S1_at

FGFRL1

fibroblastgrowthfactorreceptor‐like1;similartofibroblastgrowthfactorreceptor‐likeprotein

1.72

0.012

Gga.2908.1.S1_at

LOC395407

fibroblastgrowthfactorreceptor‐likeembryonickinase

‐1.70

0.03

Gga.1941.1.S1_at FZD4 frizzledhomolog4(Drosophila) 1.87 0.009Gga.2690.1.S1_at FZD6 frizzledhomolog6(Drosophila) 1.53 0.005Gga.311.1.S1_at FZ‐8 frizzled‐8 2.91 0.04Gga.2529.1.S1_a_at IL1RL1 interleukin1receptor‐like1 3.52 0.017

Gga.7339.1.S1_at

KDR

kinaseinsertdomainreceptor(atypeIIIreceptortyrosinekinase)

2.18

0.007

Gga.4738.1.S1_at NEO1 neogenin 1.67 0.001Gga.14911.1.S1_at OXTR oxytocinreceptor ‐2.03 0.03

Gga.274.1.S1_at

PDGFRA

platelet‐derivedgrowthfactorreceptor,alphapolypeptide

2.18

0.043

GgaAffx.3811.1.S1_at PLXNA1 plexinA1 ‐1.69 0.008Gga.8211.1.S1_at PLXNB2 plexinB2 2.12 0.018Gga.2668.4.S1_a_at RARB retinoicacidreceptor,beta ‐1.84 0.029

GgaAffx.7723.3.S1_at

RRH

retinalpigmentepithelium‐derivedrhodopsinhomolog

‐1.60

0.037

Gga.679.1.S1_at VLDLR verylowdensitylipoproteinreceptor 2.28 0.031

61

Figure3‐4.Heatmapfrommicroarrayanalysisshowingresultsfor55candidategeneswithplausiblerolesinregulatingoculargrowthand/orRPEfunctions.EachcolumnshowsthefoldchangeofRPEgeneexpressionforeachof4birds,i.e.,ratioofexpressionintreatedandcontroleyes(n=4).Down‐regulatedgenesarerepresentedingreen,andup‐regulatedgenesareinred.Interoculardifferenceofvitreouschamberdepth(mm)isshownforeachbirdontopoftherelevantcolumn.

62

Figure 3‐5. Comparison of microarray and real‐time PCR gene expression results (treated vs. fellowcontrol eye);microarray findingsof up‐regulation forFIGF,NOG, andPDGFA, anddown‐regulation forBMP2,BMP4,BMP7,DRD4,RARB,andRRHwereconfirmedwithreal‐timePCR(mean±SEM,n=4).

Table3‐3.Resultsofreal‐timePCRvalidationexperimentsforninegenesshowingdifferentially

expressioninmicroarrays;negativevaluessignifydown‐regulation&positivevalues,up‐regulation.

GeneSymbol

GeneName Foldchange

pvalue

BMP2 bonemorphogeneticprotein2 ‐6.47 0.011BMP4 bonemorphogeneticprotein4 ‐3.66 0.015BMP7 bonemorphogeneticprotein7 ‐2.40 0.007DRD4 dopaminereceptor4 ‐2.19 0.005FIGF* c‐fosinducedgrowthfactor

(vascularendothelialgrowthfactorD)2.21 0.053

NOG noggin 3.11 0.022PDGFA platelet‐derivedgrowthfactoralphapolypeptide 1.86 0.049RARB retinoicacidreceptor,beta ‐1.55 0.001RRH retinalpigmentepithelium‐derivedrhodopsinhomolog ‐2.34 0.015*pvalueforFIGFis0.053

Inthecurrentstudy,weobserveddefocus‐induceddown‐regulationinRPEofthreeBMPs,includingBMP2,whichwasalsoreportedtobedown‐regulatedinretina/RPEwith6hand3daysof form‐deprivation in aprevious study involving young chicks.133Wealso foundBMP2,BMP4,andBMP7geneexpressiontobedown‐regulatedwithshort‐termexposureto negative lenses, aswell up‐regulation in response to short‐term exposure to positivelenses.140 The consistency of the pattern of altered gene expressionwith very short andlongertermmyopia‐inducingtreatmentssuggestsanon‐goingroleforBMP2inmyopiceyegrowth.Furthermore,whilethisfamilyofgrowthfactorshasbeenimplicatedinembryonicretinalpatterninganddifferentiationinthechick,241nonetheless,aRPEsitefortheinducedchanges in BMP2 expression would explain why two different myopia‐inducing stimuliwithlikelydifferentretinalsignalpathways,hadthesameeffectonRPEgeneexpression,

63

assuming that distinct retinal signal pathways converge at the RPE. A study that hint atsuch a role for BMP2 as a growth inhibitor in eye growth regulation found the BMP2protein to be localized in human sclera, and observations of increased human scleralfibroblastsproliferation,increasedsecretionofTIMP2anddecreasedsecretionofMMP2inresponsetotreatmentwithrecombinanthumanBMP2areconsistentwithreducedscleralremodelingandreducedaxialelongation,opposite thatexpected for reducedBMP2geneexpressionandproteinsecretion.207Nevertheless,todefinitivelyestablishBMP2tobepartof a myopia‐inducing signal cascade, it will be necessary to establish that these samemyopia‐inducingstimulidecreasethesecretionoftheBMP2proteinfrombasal(choroid)oftheRPE.OftheothertwoBMPsshowingalteredgeneexpression,BMP4hasbeenplacedinthesamesubgroupasBMP2,basedonthesimilarityoftheiraminoacidsequences.Ithasalsobeenlinked to eye growth and scleral fibroblast function. Specifically,mutations in the BMP4genemayleadtodevelopmentalocularabnormalitiesincludingmicrophthalmia,249andinculturedhumanscleralfibroblasts,mechanicalstresscausesthedown‐regulationofBMP4expression.267 This growth factor also appears to influence wound healing and tissuerepair.Of particular note, transgenicmice that over expressBMP4 in theRPE show lesssevere choroidal neovascularization (CNV); this action appears to be through themodulationofTNF‐αinducedMMP9expression.319Extrapolationfromthisobservationtoassume a role for BMP4 in the maintenance of choroidal structure and function, it isplausible that sustained down‐regulation in RPE in myopic eyes may contribute to theassociatedpathologicalcomplicationsinvolvingthechoroid.BMP7 (osteogenic protein‐1, or OP‐1) belongs to another subgroup of BMPs alongwithBMP5andBMP6.320Comparedtoothermembersofthisgroup,includingBMP2,4,and6,BMP7 exhibits stronger pro‐anabolic activities as well as anti‐catabolic activities incartilage repair.316 While the effect on BMP7 protein secretion from the RPE was notstudied, the predicted effect of reduced secretionwould tend to rule out the sclera as atargetasreducedproteoglycansynthesis isoppositetotheresultsexpectedforthechickeye,whosescleraaddstoitscartilagecomponentduringeyeenlargement.Thechoroidisaplausible alternative target, as the observed thinning during myopic growth isaccompaniedbyreducedglycosaminoglycan(GAG)synthesis.124IncontrasttotheconsistentpatternsofdownregulationfortheBMPgenes, thegeneforNoggin(NOG),aBMPbindingprotein,wasup‐regulated.236,321‐323InbindingtoBMPs,NOGalsoinactivatesthem.Itappearstoplayacriticalroleinboneandcartilagedevelopment.For example, over‐expression ofNoggin inmice impairs bone formation,324,325while theNOGnullmutation results inexcess cartilage, a consequenceof elevatedBMPactivity.326Sincethechicksclera includesacartilage layer, it isplausibletarget forNOG.Intermsofeyegrowthregulation,NOGmaymodulatetheactivityofsecretedBMPssinceitisknownto bind to all three of the BMPs whose genes showed differential expression with ourmyopia‐inducinglenstreatment.Theup‐regulationoftheNOGgenewouldservetoamplifythe effects of the reduced expression of the BMP genes observed, assuming comparabletrends in protein secretion in all cases. More direct evidence of NOG’s involvement inmyopiceyegrowthwouldbeconfirmationthatneutralizingantibodiesinhibitorattenuate

64

myopic eyegrowth. Sucha resultwouldbeofpotential therapeutic interest, assuming itcanbegeneralizedtothemammalianeye.In this study, gene expression for c‐fos induced growth factor (FIGF), also known asvascularendothelialgrowthfactorD(VEGF‐D)wasup‐regulated2.21‐foldintreatedeyes.FIGF is amember of the PDGF/VEGF superfamily, whosemembers play critical roles inembryonic vessel development, angiogenesis and lymphogenesis under physiologicalconditions,andtheyhavealsobeen implicated,eitherdirectlyor indirectly, inarangeofocularpathologiesincludingage‐relatedmaculardegeneration(AMD),diabeticretinopathyandpathologicmyopia.287,327‐334TargetreceptorsforFIGFincludeVEGFR‐2and‐3,whoseactivationleadstoangiogenesisandlymphangiogenesisintumors.330,335BothVEGFR‐2and‐3havebeenlocalizedtothechoriocapillarisinmonkeyandhumaneyes295,336andVEGFR‐2mRNAshasalsobeendetectedinmonkeychoroid‐retinalpigmentepithelialcomplex.337It isnoteworthy that in the chick, thedefocus‐inducedchanges in choroid thicknessarelargelylimitedtosuprachoroidallayer,wherelymphatic‐likevesselsorlacunaeexpandorflatten to achieve thickening or thinning respectively.175, 177, 338‐340 Changes in thepermeabilityofthechoroidalvasculaturehavealsobeenreportedwithform‐deprivationinthechick.174TheFIGFmayplayaroleinthechoroidalcomponentoftheemmetropizationprocesses, although choroidal thickness changes tend to occur early,within the first fewdaysofdefocustreatment(unpublisheddata).Thesustainedchoroidalthinningobservedin this study, after 38 days of treatment, is likely to partly reflect the stretching of thechoroid,asthescleralcupexpanded.Whileneitherneovascularizationnorlymphogenesishave been reported with shorter term myopia‐inducing treatment regimens in chicks,intraocular neovascularization has been observed with long term treatments, extendingout to12monthsormore, raising thequestionofwhether theobserved change inFIFGgene expression could be a stimulus for such pathology.328 Nonetheless, three furtherobservations are compatible with a role for FIGF in ocular growth regulation. FIGFexpression is regulated,at least in fibroblasts,byc‐fos, amemberof the immediate‐earlygene family,341 and the expression of FIGF in RPE is also reported to be regulated by c‐fos.342Finally,expressionofc‐fosinretinalamacrinecellsisknowntoberesponsivetolightandtovisualstimuliknowntoaltereyegrowth.51,343PDGF‐A was included among the genes selected for further analysis based on the largeamount of data localizing members of the platelet‐derived growth factor (PDGF) andreceptor proteins to RPE, and their apparent roles in both development and ocularpathology, including proliferative diabetic retinopathy as well as corneal and choroidalneovascularization.344‐346Both freshly isolatedhumanRPEandculturedhumanRPEcellsare reported to express all isoforms of PDGF mRNA, and secret PDGF into culturemedium.347‐350 Native human RPE and cultured RPE cells also express PDGF receptormRNAaswellasPDGFR‐αand‐βprotein.348,349Thispatternofco‐expressionofPDGFanditsreceptorssuggestsautocrineregulationofPDGFsecretionbyRPE.348,349Intheadjacentchoroid, both PDGFR‐α and‐β have been detected in the normal human eyes; culturedchoroidal fibroblasts also express PDGFR‐αmRNA,346,351 and human recombinant PDGFstimulates their proliferation andmigration.351 Further studies arewarranted to explorethe possible roles of this growth factor in the signal pathway underlying myopic eyegrowthand/orrelatedpathology.

65

The finding that gene expression for dopamine D4 receptors was down‐regulated isperhapsnotsurprising,giventhemanystudies linkingretinaldopaminewitheyegrowthregulation. The D4 receptor is one of five subtypes of DA receptors, which have beencategorized based on their properties into two subfamiliesD1‐like (D1, D5) andD2‐like(D2,D3,D4).70Reduce levelsofdopamine (DA)and itsmetabolite in the retinaof form‐deprived eyes is a consistent finding across relevant studies in chicks, as is thecomplementary finding that apomorphine, a nonselective DA agonist, inhibits both formdeprivation‐ anddefocus‐inducedmyopic eyegrowth.74,79,104Results fromanothermorerecentpharmacologicalstudy,alsoinchicks,pointedtoinvolvementofD2‐likedopaminereceptors,whichwefindtohavethehighestexpression inhuman fetalRPEcells.80,151 Inrelatedstudies,appliedapomorphinewasshowntostimulatethesecretionofTGFβ‐1and‐2 from cultured human fetal RPE cells, offering a plausible signaling pathway for oculargrowthmodulation, given that the latter growth factors have been implicated in scleralextracellular matrix remodeling.151 Nonetheless, our finding of down‐regulation ratherthanup‐regulationofD4receptorgeneexpressionwouldseemintheoppositedirectiontothat expected if retinalDA releasehad remaineddepressed, although such changeshaveonlybeenobserved inoneofanumberofstudies involvingdefocus‐inducedmyopia.77,85Species differences also do not explain changes were limited to D4 rather than D2receptorsinourstudy,althoughtheyaremembersofthesamesubfamilyanditispossiblethatthereissomeoverlapintheirroles.Retinoic acid (RA) is known to be a potent regulator of cell growth and differentiation,actingvianuclearreceptors(RARsandRXRs),toactivateorrepresstheexpressionofothergenes.115,116 Interestingly,many of the growth factors showing differential expression inthecurrentstudy,includingBMP‐2,‐4,‐7,FIGF,PDGFA,havebeenshowntoberegulatedbyRAinavarietyofcellsandtissues,eitherinvivoorinvitro,anddirectlyorindirectly.270,271,352‐358Therearealreadyextensivedata linkingRAwithoculargrowth regulation inanumber of different species, although there are also subtle species‐related differences.Specificallyinthechick,negativelenstreatmentsup‐regulateretinalmRNAexpressionofaldehyde dehydrogenase‐2 (AHD2), one of the enzymes involved in RA synthesis, andretinalRAlevelsareincreasedbybothform‐deprivationornegativelenstreatment.117,119,120, 123 In contrast, choroidal RA synthesis is decreased significantly by the lattertreatments. The opposite trends have been observed in eyes recovery from form‐deprivation and ones wearing positive lenses in chick.118 Mammals and primates showsimilar retinal trends to thosedescribed for chicks but opposite trends for choroidalRAsynthesis.16,118,120,124 Since the RPE is interposed between the retina and choroid, it islikelythatitisaffectedbybothretinalandchoroidalRAlevels.Thusourfindingofdown‐regulationofRAR‐βmRNAinRPEismoreconsistentwitharesponsetotreatment‐inducedelevationofretinaRAlevels,assumingRAsynthesisremainshighwithextendednegativelens treatments. However, it is also acknowledged that the regulation of expression ofretinoic acid receptors in RPE and thus effects ofmyopia‐inducing stimulimay bemorecomplicated than themechanisms underlying expression changes inmyopic chick retinaandsclera,asreportedbyotherstudies.122,123,135,158,159,359‐361ThelastoftheninegenestargetedforadditionalinvestigationwithrealtimePCRwasRRHorperopsin.Inmice,thisvisualpigment‐likeproteinisexpressedexclusivelyinRPE,and

66

prominently in the microvilli surrounding photoreceptor outer segments in mice.362 Inchick,RRH isexpressed inboth theretinaandpinealgland,363although itsexpression inRPEhasnotbeenconfirmed,perhapsdue to theheavypigmentationof this layer. Ithasbeenassignedvariousroles,includingprovidinganindexoftheconcentrationofretinoidsand/orotherphotoreceptor‐derivedcompounds inRPE,andactingasaphotoisomerase,facilitating the conversion of all‐trans‐retinal to 11‐cis‐retinal in response to light.362,364The down‐regulation of RRH gene expression observed in our studymay signify alteredlevels of retinoids in the RPE or altered photoreceptor functions. Both are plausibleconsequences of long‐term negative lens treatment, given the altered anatomy of themyopiceye,andpreviousreportsofalteredphotoreceptormorphologyanddisksheddinginform‐deprivedchickeyes.365,366Onthetechnicalside,itisimportanttoacknowledgetwopotentiallimitationsofourdata.The first limitationrelates theRPE isolationproceduresused tocollectoursamples.Thetipsofphotoreceptorareanunavoidable contaminant, as theyareembedded in theRPEmicrovilli.Whiledarkadaptationpriortosacrificeissometimesusedinretinalresearch,tomorecleanlyseparatetheRPEfromtheretina,thistechniquewasnotusedinthecurrentstudy,becauseofthelikelyimpactofdarkadaptationongeneexpression.Forexample,theexpressionofretinalZENK,agenelinkedtoeyegrowthregulation,isknowntobeaffectedby thenumberof hours of light exposure after lights‐on.51Thuswe cannot rule out thatchangesinretinalgeneexpressioncontributedtoourresults.Thesecondlimitationrelatesto our use of the fellow eyes as controls or references. Interocular yoking has beenobserved ingeneexpressionchanges involving the retinaandscleraof thechick,53,280aswellasintheRPEforsomeBMPgenes(seeChapter2).Whiletheoriginsoftheseyokingeffectsarenotfullyunderstood,nonetheless,wecannotruleoutsimilarintraocularyokingforsomeofthegenesexaminedinthisstudy.However,asmostreportedyokingeffectsaremanifest as a similar butweaker trend in the control eye, thereby reducing rather thanexaggeratingtreatmenteffects,ouruseof felloweyesascontrolsstrengthensratherthanweakensthecasestyingchangesingeneexpressiontotheappliedtreatments.In summary, long‐term,myopia‐inducing lens (hyperopic defocus) treatments applied tochicksresult in thedifferentialexpressionofavarietyofgenes in theRPE,amongwhicharegrowthfactors,agrowthfactorantagonistprotein,aneurotransmitterreceptor,andaretinoic acid receptor. Plausible explanations linking them with either ocular growthregulation, including myopic growth, and pathological complications of myopia werepresented.Follow‐upstudiesareneededtodirectly investigate thesepossibilities,andtounderstandtheinter‐relationshipbetweenthem.

67

Chapter4ApomorphineRegulatesTGF‐β1andTGF‐β2SecretioninHumanFetalRetinalPigmentEpithelialCells AbstractThis study investigated dopaminergic regulation of TGF‐β secretion from human retinalpigmentepithelium(RPE)asapossiblemechanismbywhichinhibitoryeyegrowthsignalsmayberelayedfromretinatochoroid/sclera.Usingprimaryculturesofhumanfetal(hf)RPEasamodel,RPEgeneexpressionfordopamine(DA)andTGF‐βreceptorsandTGF‐βisoformswas examined using real‐time PCR, receptor protein expressionwithWesternsblots, and receptor localizationwith immunocytochemistry.Cultureswerealso subjectedtop treatmentwith the DA agonist, apomorphine (APO; 0–20 μM), applied to the apical(retinal)sideofcells,andthesecretionofTGF‐βisoformsintothemediummeasuredafter24hwithELISAs.DopamineD1,D2,D5aswellasTGFBR1,2,3receptors,andTGF‐β1,‐β2isoformswerealldetected inhfRPE,withD2andTGFBR2receptorsshowing thehighestexpression of receptors. D2 and TGFBR1 receptorswere also detected byWestern blotsandimmunocytochemistry.Inuntreatedcells,themRNAexpressionofTGF‐β1was4.6‐foldhigherthanTGF‐β2andbothTGF‐β1andTGF‐β2wereconstitutivelysecreted,mainlyfromthe apical side. In contrast, apically appliedAPO (0.2–2.0 μM) increased the secretion ofboth isoforms,mainly from thebasolateral (choroidal) side.Thepolarizednatureof thispharmacologicaleffectprovidesaplausibleexplanationforhowactivationofDAreceptorsontheapicalmembraneofhumanRPE,i.e.facingtheretina,couldinitiatesignalsthatactonthechoroid/scleratoinhibiteyegrowth.Some of the data reported in this chapter were previously published in the followingconferenceabstract:

Zhang Y, Maminishkis A, Zhi C, Li R, Agarwal R, Miller SS, Wildsoet CF. ApomorphineRegulatesTGF‐β1andTGF‐β2ExpressioninHumanFetalRetinalPigmentEpithelialCells.Invest.Ophthalmol.Vis.Sci.2009;50:E‐Abstract3845.

68

4.1 Introduction Myopia, or near‐sightedness, is one of the most common ocular disorders in humans,affecting41.6%ofU.S.populationaged12to54years.9Theprevalenceandtheseverityofmyopia have risen worldwide during the past several decades, stimulating increasedresearch into the underlying mechanisms, an essential step in developing effectivetherapies.Animalstudiesprovideconvincingevidencefortheimportanceofretinalsignalingineyegrowthregulation.Currentlytherearetwodifferentexperimentalparadigmsforinducingmyopia, utilizing form‐depriving diffusers and negative defocusing lenses, respectively.Bothparadigmshavebeenshowntoelicitrobustresponsesinchickens,guineapigs,treeshrewsandmonkeys.16 Inall cases, the inducedmyopia ischaracterizedbyexpansionoftheposteriorvitreouschamberoftheeye,reflectingtheremodelingoftheouterchoroidaland scleral support layers. The preservation of these response patterns in eyes isolatedfromthebrainbyopticnervesectionpointstolocalgrowthregulation.34,35TheRPEservestoprotect thehealth and integrity of theneural retina and is known tobe an importantsource of growth factors thatmay have autocrine or paracrine functions.40,135,367‐369 Byvirtueof itsanatomical location, it isalso ideally suited to relay retinalparacrinesignalsregulatingthegrowthofthechoroidandsclera.Previousstudiesofexperimentalmyopiainchickshaveprovidedstrongevidenceforaroleforretinaldopamine(DA) ineyegrowthregulation.74,75,79‐81Directevidence foraroleofDA in eye growth regulation is provided by observations that ocular administration ofapomorphine,anonselectiveDAreceptoragonist,slowsbothformdeprivation‐andlens‐inducedmyopia.DAhasbeen localized to subsetsof retinalamacrineand interplexiformcellsandDAreceptorshavealsobeenidentifiedonbothneuralcellswithintheretinaandRPE.64‐68TherearefivesubtypesofDAreceptors(D1‐D5),andbasedontheirbiochemicalandpharmacologicalpropertiestheyhavebeencategorizedintoD1‐like(D1,D5),andD2‐like (D2‐D4)subfamilies.69,70 In chicks,D2/3receptorshavebeen identifiedon thebasalsideoftheRPE,withtheirpresenceontheapicalsurfaceofRPEcellsbeingleftunresolvedduetotheheavypigmentationinthisregion.65D2receptorshavealsobeendemonstratedontheRPEofothermammals,includingcatandcow,whiletheirpresenceonhumanRPEiscontroversial.149,150The transforming growth factor‐β (TGF‐β) superfamily encompasses structurally‐relatedproteins that function as regulatory cytokines. They are involved in a wide variety offunctions, including the modulation of chemotaxis, wound healing, angiogenesis,stimulationorinhibitionofcellproliferation,andextracellularmatrix(ECM)remodeling.239TGF‐βisknowntomodulatefibroblastproliferation,aswellasECMproteinexpressionanddegradation,allofwhichhavebeentiedtoscleralgrowthmodulation.185,239,280,370‐372Non‐polarizedeffluxofTGF‐βfromRPEhasbeenmeasuredinseveralpreparations,165,166,373‐375andwhile the effects of dopamine analogs on TGF‐β secretion from RPE have not beenpreviously studied, both dopamine and DA agonists have been shown to induce dose‐

69

dependent increases in TGF‐β1 secretion and TGF‐β1 mRNA expression coupled toinhibitedproliferationinpituitarylactotropiccells.376,377Inthepresentstudy,wedeterminedtheexpressionandlocalizationofdopamineandTGF‐β receptors on primary hfRPEmonolayers in culture andmeasured the constitutive andapomorphineinducedpolarizedsecretionsofTGF‐β1and‐β2.Wefoundthatactivationofdopaminergic receptors on the retina‐facing, apicalmembranes of humanRPE results inincreased TGF‐β secretion, predominately from the basolateral (choroidal) side of theepithelium.Theseresultssuggestthatretina‐deriveddopaminergicsignalsmayberelayedto the choroid and sclera via modulated growth factor secretion from the RPE for themodulationofeyegrowth.4.2 MaterialsandMethods4.2.1 HumanFetalRPECellCultureRetinal pigment epithelial cells were obtained from 16‐18 weeks old human fetal eyes(Advanced Bioscience Resources, Alameda, CA) and cultured as described previously.367Briefly,hfRPEcellsfromdonoreyeswereseededintotissuecultureflaskswitha15%FBS‐containingculturemedium(P0).Theculturemediumwasreplacedthenextdaywith5%FBS‐containingmedium,whichwas changed every2 to3days thereafter.After the cellsbecame confluent and pigmented, they were trypsinized and passaged onto cell culturetranswellinsertscoatedwithhumanextracellularmatrix(ECM,10μg/well,BDBiosciences,Bedford,MA)atadensityof2.0×105perwell(P1).Foruseinthestudiesreportedhere,weusedconfluentP1cellswithtotaltissueresistanceof>200Ω•cm2.Thecellcultureset‐upusedinthisstudyisshownschematicallyinFigure4‐1.

Figure4‐1.hfRPEwereculturedonECM‐coatedinsertsoftranswellsthatallowedisolation

ofmediumbathingapicalandbasalsidesofcells,andthustheirseparatesamplingandanalysis.4.2.2 CharacterizationofRPEReceptorsReal‐timePCR:Real‐timePCRtargetinghumandopaminereceptors(D1,D2andD5),TGF‐β receptors (TGFBR1, 2, 3), and TGF‐β isoforms (TGF‐β1, TGF‐β2), was performed on

70

culturedhfRPEafterincubatinginserum‐freemedium(SFM)for24h.TotalRNAfromsixcultureinsertswasextractedusingRNeasyMiniKit(Qiagen,Valencia,CA),withon‐columnDNase treatment, before being reverse transcribed to cDNA (SuperScript III First‐StrandSynthesis System for RT‐PCR, Invitrogen, Carlsbad, CA). Real‐time PCR assays wereperformed according to the manufacturer’s instructions (TaqMan & ABI SequenceDetectionSystem7900;AppliedBiosystems,FosterCity,CA).TaqManassayIDsarelistedin Table 4‐1. Standard curves were generated by 10‐fold serial dilutions of cDNA. ThemRNAconcentrationofeachsamplewasnormalizedagainsthumanGAPDHlevels.Testingforeachgenewasperformedusingsixsamples,withtwotechnicalrepeats.Becausesomeof the qPCR primers used were not designed to flank exon‐exon junctions, controls forgenomiccontaminationwereincluded.

Table4‐1.TaqManqPCRassayinformation

Assay IDDRD1 Hs00377719_g1DRD2 Hs01024460_m1DRD5 Hs00361234_s1TGFBR1 Hs00610318_m1TGFBR2 Hs00234253_m1TGFBR3 Hs01114253_m1TGFB1 Hs00998130_m1TGFB2 Hs00234244_m1GAPDH Hs999999905_m1

WesternBlots:Westernblotswereused toconfirmthepresenceof theD2receptorandTGFBR1proteinsinhfRPEcells.Whilestillattachedtotheinserts,confluentmonolayersofhfRPE were lysed at 4°C with RIPA buffer (Sigma‐Aldrich, St. Louis, MO) containing aproteaseinhibitorcocktail(RocheDiagnostics,Indianapolis,IN),andthetotalproteinwasmeasured (BCA;PierceBiotechnology,Rockford, IL).Westernblot samples (30μg)weredenaturedandthenelectrophoresedona4%‐12%gradientgel (NuPAGE4‐12%Bis‐TrisGel,Invitrogen),beforebeingtransferredtonitrocellulosemembranes(iBlotGelTransferStacks, Invitrogen). Membranes were blocked (StartingBlock T20 [TBS]; PierceBiotechnology) and then incubated with either goat anti‐human polyclonal antibodyagainsttheD2receptor(SantaCruzBiotechnology,SantaCruz,CA,#sc‐7522),orgoatanti‐humanpolyclonal antibodyagainstTGFBR1(R&DSystems,Minneapolis,MN,#AF3025),and subsequently labeled with HRP‐conjugated rabbit anti‐goat IgG (HRP; PierceBiotechnology). Immunoreactive bands were detected with chemiluminescence(Supersignal Pico ECL; Pierce Biotechnology) and images developed using a bioimagingsystem (Autochemie; UVP, Upland, CA).348,369 Assays involved three separate biologicalsamples and triplicate repeats. Human brain and heart lysates andmouse brain lysateswereusedaspositivecontrolsforbothreceptors.Immunocytochemistry:hfRPEcellswerefixedwith4%formaldehyde,permeabilizedwith0.2% Triton X‐100, blocked with a signal enhancer (Image‐iT FX; Invitrogen). RPEmonolayers were incubated with antibodies against human D2 receptors (Santa CruzBiotechnology,#sc‐7522)andTGFBR1(R&DSystems,#AF3025),andZO‐1(Zymed,South

71

SanFrancisco,CA,#339100),prelabeledwithdifferentfluorophoresusingZenonLabelingTechnology (Invitrogen), followingmanufacturer’s instructions. Normal goat serum pre‐labeled with fluorophore was used as a negative control. Cells were then stained withHoechst33342,mountedonglassslideswithantifadereagent(ProlongGold;Invitrogen),andimagedwithaZeissAxioplan2microscopewithapotomeandAxiovision3.4software(CarlZeissAG,Germany).4.2.3 EffectofApomorphine(APO)onTGF‐β1andTGF‐β2SecretionfromRPEDose‐responsestudy:Confluentmonolayers of hfRPE cells on insertswere incubated inserum‐free medium (SFM) for 24 h prior to drug treatment. One of 4 differentconcentrations,0,0.2,2,20μMofAPO(Sigma,St.Louis,MO),wasthenaddedtotheSFMmediumbathingtheapicalsideofthehfRPEcells.After24h,culturemediumwascollectedfrombothapicalandbasalchambersandassayedforTGF‐β1and‐β2(pg/ml)usingELISAkits. The total amounts (pg) of TGF‐β1 and ‐β2 secreted into apical and basal culturemediumwere calculated formmeasured concentrations, accounting for the difference inthe volumes of culture medium in the apical and basal chambers, i.e. 0.5 and 1.5 mlrespectively.TheeffectofeachdoseofAPOwasstudiedintriplicate. ELISA:SamplesofmediawereclarifiedbycentrifugationandsubjectedtoimmunoassaysforTGF‐β1and‐β2,carriedoutaccordingtothemanufacturer’sinstructions(R&DSystems,Minneapolis, MN). In brief, acid‐activated samples were added to a 96‐well microplatecoatedwithmonoclonal antibodies specific to eitherTGF‐β1or ‐β2, incubated for2h atroom temperature, and then reacted with horseradish peroxidase (HRP)‐conjugatedpolyclonal antibodies specific for TGF‐β1 and ‐β2. Finally, a color reagent mixture wasadded to thewells and optical densities at 450 nmmeasuredwith amicroplate reader,usingawavelengthcorrectionsettingof540nm.4.2.4 StatisticalAnalysisData reported in both the text and figures are expressed as mean ± SEM. StatisticalcomparisonsmadeuseofStudent’st‐tests (unpaired, two‐tailed),whentwogroupswereinvolved,andone‐wayANOVA,whenmorethan2groupswereinvolved.ThedescribedresearchfollowedthetenetsoftheDeclarationofHelsinkiandtheNationalInstitutesofHealthInstitutionalReviewBoard.

72

4.3 Results4.3.1 CharacterizationofDopamineandTGF‐βReceptorsonCulturedhfRPEGene/proteinexpressionfordopamineandTGF‐βreceptorsReal‐timePCRresultsforsixofthegenescodingfordopamineandTGF‐βreceptorsareshowninFigure4‐2.WhileD1,D2 andD5 receptorswere all expressed at relatively low levels, D2 receptor expressionwasapproximately22‐foldhigherthanD1receptorexpression,and2.5‐foldhigherthanofD5 receptor expression (Fig. 4‐2A,p<0.001 forboth cases). In contrast, all threeTGF‐βreceptorswere strongly expressed, althoughTGFBR2expressionwas approximately4.4‐foldhigherthanTGFBR1expression(Fig.4‐2B,p<0.01)and2.3‐foldhigherthanTGFBR3expression (p< 0.05). Western blots confirmed the presence of both D2 receptor andTGFBR1proteins(Fig.4‐3).Figure4‐3AshowsresultswithaSantaCruzantibody(sc‐7522)usedto labeltheD2receptorandincludespositivecontrolsofbrain(humanandmouse)and heart (human). In one blot this antibody showed three bands, all of which werereproduced with different antibodies from other sources (Chemicon: AB5084, Abcam:ab21218, ab32349 and ab30743). The band at 51 kDa, which is best described in theliterature,was detected in all of our hfRPE samples (lanes 1‐3) and confirmedwith thepositivecontrolsinhumanheart(lane5)andmousebrain(lane6).378Thetwoadditionalbands,at68and90kDa,havealsobeendescribedintheliterature.378Thespecificityofthe68 kDa for DRD2 has been confirmed using pre‐immune serum and peptide‐blockedimmune serum.378 This larger band represents the receptor in a different state ofglycosylation and was the major band in mouse brain lysates revealed with anotherantibody(Chemicon:AB5084,unpublisheddata).Wealsodetecteda90kDaband,whichlikelyrepresentsaDRD2dimerprotein(http://www.scbt.com/datasheet‐7522‐d2dr‐n‐19‐antibody.html). Both the Abcam ab32349 and Santa Cruz antibodies to the D2 receptorproducedanadditionalbandat37kDa.ForTGFBR1,therewasjustonebandat50kDa(Fig.4‐3B),whichwasevidentinallthreehfRPEsamples(lanes1‐3)aswellasinthepositivecontrol(mousebrain,lane4).

Figure4‐2.mRNAlevelsfor(A)dopaminereceptorsand(B)TGF‐βreceptorsinculturedhfRPEnormalizedtoGAPDH.Dataareexpressedasmean±SEM(n=6). * p < 0.05, ** p < 0.01, *** p < 0.001.

73

Figure4‐3.ExpressioninculturedhfRPEoftheD2receptorprotein(A),detectedat90,68,and51kDa,andTGFBR1protein(B),detectedat50kDa.Inbothcases,lanes1,2,and3wereloadedwithhfRPEcelllysatesfrom3differentdonors.Lanes4,5,and6in(A)includepositivecontrolsfortheD2receptorfromhumanbrainandheartlysates,andmousebrainlysates,respectively.In(B)mousebrainlysateswereusedasapositivecontrolforTGFBR1(lane4).

LocalizationofdopamineandTGF‐βreceptorsImmunocytochemistrylabelingwasusedto localize D2 and TGFBR1 receptors on hfRPE grown on transwells. Additional ZO‐1labeling of tight junctions allowed the apical and basolateral membrane regions to bedistinguished.SampleresultsareshowninFigure4‐4;theupperpartofeachpanelshowsacross sectional view of the epithelium. Both D2 and TGFBR1 receptors were largelyconfinedtotheapicalmembranes.

Figure4‐4.ImmunocytochemistryofculturedhfRPEforD2receptors(A)andTGFBR1(B),showninred,ZO‐1 shown in green, and nuclei shown in blue. The upper part of each panel shows a cross section of themonolayer;thedottedwhitelineindicatesthebasalsurfaceofthecellsandthewhitearrowsindicatetheirapical surface. The lower part of each panel shows an optical section obtained from a z‐stack. Negativecontrolsarealsoshown,forbothDRD2(C)andTGFBR1(D).

74

4.3.2 ExpressionandconstitutivesecretionofTGF‐β1andTGF‐β2Real‐timePCRresultsforuntreatedculturedhfRPEshowedTGF‐β1mRNAexpressiontobe4.6‐foldhigher thanTGF‐β2 expression (datanormalized toGAPDH;Fig. 4‐5,p<0.001).Consistentwithconstitutivesecretionof thesecytokines,bothTGF‐β1and ‐β2werealsodetected in the culturemedium (Fig. 4‐6), and in each case, the secretionwaspolarized,predominantly fromtheapical sideofcells,whethermeasured in termsofconcentration(pg/ml,Fig.4‐6A)ortotalamount(pg,Fig.4‐6B).However,whilegeneexpressiondatawasbiased in favorofTGF‐β1,TGF‐β2was thedominant isoform in thesecretedproduct, itsconcentrationintheapicalchamberbeingapproximately5.5‐foldhigherthanthatofTGF‐β1(Fig.4‐6A).Similartrendsareevidentinvolume‐adjustedtotalproteindata(Fig.4‐6B).

Figure 4‐5. TGF‐β1 and ‐β2mRNA expression in untreated cultured hfRPE normalized to GAPDH. Dataexpressedasmean±SEM(n=6).

Figure 4‐6. Constitutive polarized secretion of TGF‐β1 and TGF‐β2 from untreated cultured hfRPE,determinedbyELISAsandexpressedin(A)asconcentrations(pg/ml)andin(B)astotalamount(pg).Foreachchamber,theamount(pg)ofsecretedproductwascalculatedasconcentration(pg/ml)xbathvolume(ml), using0.5 and1.5ml for apical andbasal chambers respectively.TGF‐β2was thedominant isoformsecreted,andmainlyintotheapicalbath(n=3).*p<0.05,**p<0.01,***p<0.001.

75

4.3.3 Apomorphine‐InducedAlterationsinTGF‐β1andTGF‐β2SecretionApomorphine(APO)increasedthesecretionofbothTGF‐β1and‐β2inapolarized,dose‐dependant fashion (Table 4‐2, Fig. 4‐7). Specifically, APO stimulation of apical dopaminereceptors significantly increased the basal (but not the apical) secretion of both TGF‐β1andTGF‐β2.Thetotalproteindata(Figs.4‐7B&D)showthatAPO,atboth0.2and2μMconcentrations, reversed the direction of polarized secretion from the apical side, in theabsenceofAPO,tothebasalsideofthemonolayer.Thischangefrompredominantlyapicalconstitutive secretion to predominantly basal secretion with the addition of APO is notevidentintheproteinconcentrationdata(Figs.4‐7A&C),whichdoesnotaccountforthedifferencesinthevolumesofapicalandbasalchambers.In the absence of APO, the total secretion of TGF‐β1 (apical+basal chambers) was 196pg/well.ThemaineffectofapicallyappliedAPOwastoincreasesecretionofTGF‐β1fromthebasalsideofcells.Secretionwasverylowincontrol(0APO)and20μMAPO(25&33pg/wellrespectively),butwassignificantly increasedwith0.2and2μMAPO,by68‐and44‐fold over the control value (1700 & 1101 pg/well, respectively; p < 0.01 & 0.05,respectively).Incontrast,thechangeinsecretionofTGF‐β1fromtheapicalsideofcellsdidnotreachstatisticalsignificanceforanydoseofAPO.In the absence of APO, the total secretion of TGF‐β2 was 1099 pg/well. As with APO‐inducedchangesinTGF‐β1secretion,themaineffectofapicallyappliedAPOwasincreasedTGF‐β2 secretion from thebasal sideof the epithelium.Basal secretionwas significantlyincreasedwithboth0.2and2μMAPOtreatments(1878&2040pg/well,respectively;p<0.001 for both cases). These increases correspond to 11.4‐ and 12.4‐fold increasescomparedtothecontrolvalue.AswithTGF‐β1,secretionofTGF‐β2wasminimallyaffectedby20μMAPO.Table4‐2.Dose‐dependenteffectsofapically‐appliedAPOonTGF‐βsecretionbyculturedhfRPE

(24htreatment). TGF‐β1(pg) TGF‐β2(pg)

APO(μM) Apical Basal Total* Apical Basal Total*

0 171±30 25±13 196±17 934±93 165±104 1099±72

0.2 312±89 1700±500 2012±423 776±95 1878±50 2654±54

2.0 365±177 1101±356 1466±187 921±46 2040±340 2962±353

20 193±65 33±16 225±51 786±78 108±78 893±155

*TotalsecretionrepresentscombinedsecretionsofTGF‐βintoapicalandbasalchambers(0.5&1.5mlrespectively).Dataexpressedasmean±SEM(n=3)

76

Figure4‐7.Dose‐dependenteffectsofapically‐appliedAPOonsecretionbyculturedhfRPEofTGF‐β1(A&B),andTGF‐β2(C&D).TGFsecretion isexpressedaseitherconcentration(pg/ml)oramount(pg).TheAPO‐inducedchangesinTGF‐β1and‐β2secretionweregreaterfromthebasalcomparedtoapicalsidesofcells.*p<0.05,**p<0.01;***p<0.001.4.4 DiscussionItisknownthatdopaminereceptoragonistssuchasapomorphine(APO)caninhibitform‐deprivation and lens‐induced myopia in chicks and monkeys.74, 79‐81, 84 APO is anonselective dopamine receptor agonist that activates D2‐like receptors at nanomolarconcentrations and D1‐like receptors at micromolar concentrations.69, 82, 379, 380 Theobservationthatsulpiride,aD2receptorantagonist,enhancesform‐deprivationmyopiainchickrepresentsfurtherevidenceimplicatingD2receptorsineyegrowthregulation.85ThepresentstudyidentifiedD2receptorsonapicalmembraneofhfRPE,offeringaplausibleasite of action for APO and relay mechanism for visually‐regulated dopaminergic signalsfromtheretina. The hfRPE constitutively secretes a wide range of cytokines, chemokines and growthfactors, with presumed roles in maintaining the homeostatic environment of the distalretina and the choroid.368,381,382 In the present study,we demonstrated the presence ofTGF‐βreceptorsontheapicalmembraneofhfRPEandshowedforthefirsttime,polarized(apical) constitutive secretion of TGF‐β from hfRPE. Localization of TGFBR1 to the RPEapicalmembraneisconsistentwithautocrineregulationofTGF‐βsecretionintotheretina,which is thought to help protect the retina from extreme variations in TGF‐β levels andplaysanimportantroleinmaintainingnormalretinalphysiology.375,383,384

77

The apical bias in the constitutive secretion of both TGF‐β isoforms was reversed withapicallyappliedAPO(0.2and2μM),whichincreasedthesecretionofTGF‐βintothebasal(choroidal) chamber. The dose‐response curve for APO‐induced TGF‐β secretion had aninvertedU‐shape,withintermediateconcentrations,0.2and2μMAPO,resultinginhighestsecretion.Interestingly,20μMAPOhadminimaleffectonthesecretionofbothTGF‐β1and‐β2, consistent with a previous in vivo observation.81The reduced efficacy of this highconcentration of APO ismost likely due to receptor desensitization or secondary effectsmediatedbybindingtootherreceptors,suchasD5receptors(alsodetectedinhfRPE),andβ‐adrenoceptors.385GiventhatAPOisknowntoinhibitthedevelopmentofmyopiainbothchickandmonkeymodels,74, 75, 81, 84 our observation of APO‐induced secretion of TGF‐β into the basalchamber opens the possibility that this cytokine may act as an inhibitor of ocularelongation. Figure 4‐8 is a schematic diagram illustrating APO “rescue of myopia”. AparallelisdrawnbetweenthelowestconcentrationofapicalAPOtestedinourhumanRPEmodel (Fig. 4‐8A), and the decreased retinal dopamine level recorded in experimentalmyopia(Fig.4‐8E).AfurtherparallelisdrawnbetweentheintermediateconcentrationsofAPO(Fig.4‐8B),andtheretinalconditionsencounteredinuntreatednormaleyes(Fig.4‐8D),andAPO‐treatedform‐deprivedmyopiceyes(Fig.4‐8F).ThemodelfordopaminemediatedTGF‐βsecretionfromRPEshowninFigure4‐8indicatesthe sclera to be the target of TGF‐β, although the choroid represents another possibletarget.41Nonetheless, in tree shrews, which have a primate‐like fibrous sclera, mRNAexpressionsofTGF‐β1, ‐β2,and‐β3wereall foundtobedown‐regulatedfollowingform‐deprivation,consistentwitharoleofTGF‐βasanoculargrowthinhibitor.371OtherinvitrostudiesinvolvingtreeshrewscleraarealsoconsistentwithascleralsiteofactionforTGF‐βandan inhibitoryroleoneyeelongation.371,386Ourstudydidnotdistinguishbetweentheactive and latent forms of TGF‐β1 and ‐β2. However, other studies indicate that bothisoforms were secreted in their latent forms from the hfRPE.374,387 Activation of TGF‐βisoforms by plasmin, matrix metalloprotease‐2 and ‐9, thrombospondin, integrins andotherfactorsprovidemechanismsfortheircontrolledreleaseintargettissues(choroidandsclera).239,388‐391Inaddition todopamine,otherparacrinesignalsmayalsocontribute to theregulationofTGF‐β2secretionfromRPE.Forexample,arecentstudydescribedcholinergicregulationofTGF‐β2secretion fromhumanRPE.392‐394TheroleofcholinergicreceptorsonRPE ineyegrowthregulationisstrengthenedbytheidentificationinrecentgenome‐wideassociationstudies of susceptibility loci formyopiaon chromosome15,20,395which implicate severalcholinergic receptors, one of which (CHRNA3) is a highly expressed signature gene forhumanRPE.394Insummary,keyfindingsfromthisstudy‐theapicallocalizationofD2receptorsonhfRPEcells and the polar nature of the APO‐induced secretion of TGF‐β, from the basal side ‐supportaroleforRPEasarelayforretinaldopaminesignalsunderlyingvisually‐regulatedeyegrowth.

78

Figure4‐8.Diagramillustratingamodelforretinaldopamine‐regulatedeyegrowth.Thedecreasedlevelofretinaldopaminereportedinexperimentalmyopiamodels(E)ismimickedbythelowAPOcondition(<0.2µM;A),leadingtolowsecretionofTGF‐βfromthebasalsideofRPE.In(B),intermediatedosesofAPO(0.2µM≤[APO]≤2µM)increasebasalsecretionofTGF‐β,tolevelsinnormaluntreatedeyes(D);APOtreatmentof form‐deprived eyes (F), normalizes TGF‐β secretion to slow myopia progression. The dose‐responsecurveforAPOisU‐shaped;thushighdosesofAPO(>20µM,C),likeverylowdoses,arewithouteffectonTGF‐βsecretion.

79

Chapter5DissertationSummaryandFutureDirections 5.1 DissertationsummaryMyopia is one of the most common ocular disorders in human. The ocular changeunderlyingmyopia is accelerated eye growth, which results in amismatch between theaxial length of eye and its refractive power, and in turn, blurred retinal imageswithoutopticalcorrection.Inhighmyopia,generallycategorizedas‐6.0Dorhigher,theassociatedstretching of internal ocular structures with the increase in axial length may lead toblinding complications such as retinal degeneration, retinal detachment, myopicmaculopathy, choroidal neovascularization and glaucoma. Understanding of themechanism underlying myopic eye “growth” is an essential prerequisite to developingeffective anti‐myopia treatments. The research described in this dissertation focused onthe role of retinal pigment epithelium (RPE) in eye growth regulation, with specialattention to its potential role as a relay in retino‐sclera signal cascadesmodulating eyegrowth. An overarching motivation was the possibility that this line of research mightuncovernovelapproachestocontrollingmyopia.This dissertation research addressed two questions in relation to the role of RPE in eyegrowthregulation: (1)what is its role inaccelerated, i.e.myopic, eyeelongation,and (2)whatisitsroleinslowedeyeelongation,asrequiredtocontrolmyopia.ThefirstquestionwasinvestigatedinChapters2and3.ExperimentsusingnegativelensestoinducemyopiainvivoinyoungchicksrevealeddifferentialgeneexpressioninRPEwithbothshort‐term(2and48h)andlong‐term(38days)lenstreatments.ThesecondquestionwasinvestigatedinChapters2and4,usingbothourinvivo chickmodelandan invitrohfRPEcellculturemodel.Weidentifiedgenesthatarebidirectionallyregulated,showingalteredexpressionin opposite directions according towhether negative lens or positive lens treatment areimposed,thelatterresultinginslowedeyegrowth.Wealsodocumentedwithadopamineagonist known to inhibit myopia in animal studies, increased secretion from culturedhfRPEofagrowthfactoralreadylinkedtomyopiagrowth.Themajorkeyfindingsfromthisdissertationworkmaybesummarizedasfollows:

1. In investigations into the effects of imposed optical defocus on gene expressionchanges in chick RPE,we found that gene expression of the bonemorphogeneticproteins(BMPs),BMP2,BMP4,andBMP7tobebidirectionallyregulatedbyshort‐

80

term (2 and 48 h) imposed defocus of opposite sign. Because changes in geneexpression were elicited before significant changes in ocular dimensions hadoccurred, these genes are assumed to play important roles in initiation andmaintaining early responses to imposed optical defocus rather than being aconsequenceofalteredeyegrowth.Thesegenesrepresentpotential targets in thedevelopmentofmolecular‐basedanti‐myopiatreatments.

2. In investigating, the effects of long‐term (38 days), myopia generating, imposed

hyperopic defocus on gene expression in chick RPE using high‐throughput DNAmicroarray technology, we identified a group of genes that could plausibly beinvolvedinmaintainingthemyopicphenotypeand/orpathologicalcomplicationsofmyopia. This group included BMPs, NOG, dopamine receptor D4 (DRD4), retinoicacid receptor, beta (RARB), and retinal pigment epithelium‐derived rhodopsinhomolog(RRH).

3. In investigating the effects on TGF‐β secretion from cultured hfRPE cells of

apomorphine(APO),adopaminereceptoragonistwithknownanti‐myopiaaction,we documented a pattern of regulation, i.e., apical application of APO leading toincreasedTGF‐βssecretionfromthebasalsideofcells,consistentwiththemyopiainhibitoryeffectofAPOandofferingaplausiblemechanismforthesame.

Thisdissertationworkcanbeconsiderednovelandinnovativeinthefollowingrespects:(1)theWildsoetlabistheonlymyopiaresearchlabfocusedontheroleofRPEineyegrowthregulationandthefindingthatRPE‐derivedBMPsareup‐regulatedunderconditionsthatsloweyegrowthisnewtothemyopiafieldandopensupthepossibilitythattheymaybeused as novel anti‐myopia therapies; (2) the attempt to tie retinal DA to RPE‐TGF‐βregulation is also a new idea and approach used novel; previous studies have looked atneurotransmittersandgrowthfactorsinisolation,andhaveneverfocusedontheRPEasapotentialsiteforsuchregulation.Insummary,weinvestigatedtheroleofRPEinmyopiadevelopmentandcontrol,focusingon the molecular and cellular mechanisms underlying the altered eye growth, usingchanges in gene expression in RPE as a key signature. The findings reported hereinrepresent the first studies to implicate the RPE in postnatal eye growth regulation, as arelayorconduitforretina‐derivedgrowthmodulatorysignalsdirectedatthechoroidandsclera.Thegenesandmolecules identified inthesestudiesmay leadtonovel therapeuticapproachestocontrollingofmyopia.Thisstudyalsoopensupthepossibilityof targetingtheRPEwithsomeformofanti‐myopiagenetherapy.5.2 FuturedirectionsBasedonthefindingsreportedinthisdissertation,thereareanumberoflogicaldirectionsfor future research. For example, roles for BMPs in eye growth regulation are stronglysuggestedbyresultsfromgeneexpressionstudiesbuttheseresultsarenotdefinitive.Thusthereisaneedformoredirectevidence.Thearealsomanyunresolvedquestionsrelatedto

81

thepathwaysmediatingtheregulationofRPE‐BMPsandthepotentialrolesineyegrowthregulationand/ormyopia‐relatedpathologyofothergenesshowingalteredexpressionintheRPEofmyopiceyes.Someavenuesforfollow‐upstudiesareoutlinedbelow.

1. FurthercharacterizationoftheocularrolesofBMPs:(1) QuantificationofBMPsatprotein level indifferentmyopiaandhyperopia

animalmodels.(2) Application of BMPs and BMP signaling pathway inhibitors to directly

study the role of BMPs in eye growth regulation. BMPs (BMP2, BMP4,and/or BMP7 protein) and signaling pathway inhibitors such as LDN‐193189andBMPantibodiescanbeadministratedthroughintravitrealandsubconjunctivalinjectionsandtheireffectsoneyegrowthexamined.

(3) DirectmanipulationofBMPgeneexpressioninRPEtoinvestigatetherole

ofRPE‐BMPs in eye growth regulation, ocularmorphology and functions.Possibleapproaches includetheover‐expressionandknock‐downofBMPgenesinRPE,whichhasbeendemonstratedtobeanidealtargettissueforgenetransfer.

(4) DeploymentofagenetherapyapproachformyopiacontroltargetingRPE.

Since positive lens treatments up‐regulate BMP2, BMP4, and BMP7 geneexpression inRPE, and the opposite tends are seenwithnegative lenses,the over‐expression of these BMPs is predicted to slow eye elongation,whileknock‐downof thesegenesshouldaccelerateeyegrowth.Althoughgene therapy has never been evaluated in any myopia animal models,promisingresultshavebeenobtainedwithgenetherapy forsomehumanretinaldiseasesand this approachhasalsobeen testedexperimentally inthechick,asatreatmentforoneblindingretinalcondition.

(5) InvestigationoftheeffectsofRPE‐derivedBMPson(i)retinalremodeling

andneurogenesisand(ii)choroidal/scleralremodeling,usingboth invivoanimalandinvitrocellandtissueculturemodels.

2. CharacterizationoftheretinalmechanismsregulatingRPE‐BMPexpression,usinga

similar approach to that used forRPE‐TGF in this dissertation research, startingwith a study of the effects of dopamine receptor agonists onBMP secretion fromRPE usinginvivo animal and invitro hfRPE cell culturemodels. The lattermodelavoids potentially confounding influences of nearby ocular structures. The invivoapproachwouldbetoexaminetheeffectofapomorphine(APO),asanexampleofanonselectiveDAagonist, onBMPexpression inRPEof chicksundergoingmyopia‐inducingtreatments.TheinvitroapproachwouldbetocharacterizeAPO’seffectonBMPsecretionfromculturedhfRPE.

3. Identification of novel genes involved in eye growth regulation. Increasing our

understandingof thegenes involved inmyopiadevelopmentpotentiallyopensup

82

the possibility of new therapeutic targets; the microarray study reported in thisdissertation identified a number of genes showing altered expression in myopiceyesandthuswarrantingfollow‐upstudies.

Fromtheperspectivesofbasicscienceandclinicalpractice,thesuccessfulaccomplishmentofoneormoreof theaboveproposed linesofresearchshouldprovidenewinsights intothe molecular and cellular mechanisms underlying eye growth regulation and myopiadevelopment. Targeting specific genes in RPE as a therapeutic intervention for myopiarepresents a novel approach but one with potential long term benefits to myopes, ifsuccessfully accomplished. Given the now high prevalence of myopia world‐wide, manywouldalsostandtobenefit.

83

Bibliography1. McBrienNA,MillodotM.Abiometric investigationof lateonsetmyopiceyes.Acta

Ophthalmol(Copenh)1987;65:461‐468.2. Whitmore WG. Congenital and developmental myopia. Eye (Lond) 1992;6 ( Pt

4):361‐365.3. MehraKS,KhareBB,VaithilingamE.RefractioninFull‐TermBabies.BrJOphthalmol

1965;49:276‐277.4. ChenJ,XieA,HouL,SuY,LuF,ThornF.Cycloplegicandnoncycloplegicrefractions

ofChineseneonatalinfants.InvestOphthalmolVisSci2011;52:2456‐2461.5. JonesD,LuensmannD.Theprevalenceandimpactofhighmyopia.EyeContactLens

1999;38:188‐196.6. Saw SM, Gazzard G, Shih‐Yen EC, Chua WH. Myopia and associated pathological

complications.OphthalmicPhysiolOpt2005;25:381‐391.7. FlitcroftDI.Thecomplexinteractionsofretinal,opticalandenvironmentalfactorsin

myopiaaetiology.ProgRetinEyeRes2012;31:622‐660.8. Ono K, Hiratsuka Y, Murakami A. Global inequality in eye health: country‐level

analysis from the Global Burden of Disease Study. Am J Public Health2010;100:1784‐1788.

9. VitaleS,SperdutoRD,FerrisFL,3rd.IncreasedprevalenceofmyopiaintheUnitedStatesbetween1971‐1972and1999‐2004.ArchOphthalmol2009;127:1632‐1639.

10. Pan CW, Ramamurthy D, Saw SM. Worldwide prevalence and risk factors formyopia.OphthalmicPhysiolOpt2012;32:3‐16.

11. MorganIG,Ohno‐MatsuiK,SawSM.Myopia.Lancet2012;379:1739‐1748.12. SawS,YangA,ChanY,TeyF,NahG.The increase Inmyopiaprevalence inyoung

maleSingaporeadultsfrom1996‐1997to2009‐2010InvestOphthalmolVisSci;52:E‐Abstract24902011.

13. VitaleS,CotchMF,SperdutoR,EllweinL.CostsofrefractivecorrectionofdistancevisionimpairmentintheUnitedStates,1999‐2002.Ophthalmology2006;113:2163‐2170.

14. LimMC,GazzardG,SimEL,TongL,SawSM.DirectcostsofmyopiainSingapore.Eye(Lond)2009;23:1086‐1089.

15. BarDayanY,LevinA,MoradY,etal.Thechangingprevalenceofmyopia inyoungadults:a13‐yearseriesofpopulation‐basedprevalencesurveys.InvestOphthalmolVisSci2005;46:2760‐2765.

16. Wallman J, Winawer J. Homeostasis of eye growth and the question of myopia.Neuron2004;43:447‐468.

17. Wojciechowski R. Nature and nurture: the complex genetics of myopia andrefractiveerror.ClinGenet2011;79:301‐320.

18. MorganI,RoseK.Howgeneticisschoolmyopia?ProgRetinEyeRes2005;24:1‐38.19. Solouki AM, Verhoeven VJ, van Duijn CM, et al. A genome‐wide association study

identifiesasusceptibilitylocusforrefractiveerrorsandmyopiaat15q14.NatGenet2012;42:897‐901.

84

20. HysiPG,YoungTL,MackeyDA,etal.Agenome‐wideassociationstudyformyopiaand refractive error identifies a susceptibility locus at 15q25. Nat Genet2010;42:902‐905.

21. LiYJ,GohL,KhorCC,etal.Genome‐wideassociationstudiesrevealgeneticvariantsin CTNND2 for high myopia in Singapore Chinese.Ophthalmology 2011;118:368‐375.

22. IpJM,SawSM,RoseKA,etal.Roleofnearworkinmyopia:findingsinasampleofAustralianschoolchildren.InvestOphthalmolVisSci2008;49:2903‐2910.

23. WensorM,McCartyCA,TaylorHR.PrevalenceandriskfactorsofmyopiainVictoria,Australia.ArchOphthalmol1999;117:658‐663.

24. RoseKA,MorganIG,SmithW,BurlutskyG,MitchellP,SawSM.Myopia,lifestyle,andschoolinginstudentsofChineseethnicityinSingaporeandSydney.ArchOphthalmol2008;126:527‐530.

25. RoseKA,MorganIG,IpJ,etal.Outdooractivityreducestheprevalenceofmyopiainchildren.Ophthalmology2008;115:1279‐1285.

26. RadaJA,SheltonS,NortonTT.Thescleraandmyopia.ExpEyeRes2006;82:185‐200.27. Wallman J,Turkel J,Trachtman J. Extrememyopiaproducedbymodest change in

earlyvisualexperience.Science1978;201:1249‐1251.28. Wiesel TN, Raviola E. Myopia and eye enlargement after neonatal lid fusion in

monkeys.Nature1977;266:66‐68.29. Sherman SM, Norton TT, Casagrande VA. Myopia in the lid‐sutured tree shrew

(Tupaiaglis).BrainRes1977;124:154‐157.30. Howlett MH, McFadden SA. Form‐deprivation myopia in the guinea pig (Cavia

porcellus).VisionRes2006;46:267‐283.31. Visscher PM, Hill WG, Wray NR. Heritability in the genomics era‐‐concepts and

misconceptions.NatRevGenet2008;9:255‐266.32. DiraniM,ChamberlainM,ShekarSN,etal.Heritabilityofrefractiveerrorandocular

biometrics: the Genes in Myopia (GEM) twin study. Invest Ophthalmol Vis Sci2006;47:4756‐4761.

33. ChenYP,HockingPM,WangL,etal.Selectivebreedingforsusceptibilitytomyopiareveals a gene‐environment interaction. InvestOphthalmolVis Sci 2011;52:4003‐4011.

34. TroiloD,GottliebMD,Wallman J.Visualdeprivationcausesmyopia inchickswithopticnervesection.CurrEyeRes1987;6:993‐999.

35. Wildsoet C, Wallman J. Choroidal and scleral mechanisms of compensation forspectaclelensesinchicks.VisionRes1995;35:1175‐1194.

36. WildsoetC.Neuralpathwayssubservingnegativelens‐inducedemmetropizationinchicks‐‐insightsfromselectivelesionsoftheopticnerveandciliarynerve.CurrEyeRes2003;27:371‐385.

37. Wallman J, Gottlieb MD, Rajaram V, Fugate‐Wentzek LA. Local retinal regionscontrollocaleyegrowthandmyopia.Science1987;237:73‐77.

38. Diether S, Schaeffel F. Local changes in eye growth induced by imposed localrefractiveerrordespiteactiveaccommodation.VisionRes1997;37:659‐668.

39. StoneRA,KhuranaTS.Geneprofilinginexperimentalmodelsofeyegrowth:Cluestomyopiapathogenesis.VisionRes2010;50:2322‐2333.

85

40. Rymer J, Wildsoet CF. The role of the retinal pigment epithelium in eye growthregulationandmyopia:areview.VisNeurosci2005;22:251‐261.

41. NicklaDL,WallmanJ.Themultifunctionalchoroid.ProgRetinEyeRes2010;29:144‐168.

42. Smith EL, 3rd, Ramamirtham R, Qiao‐Grider Y, et al. Effects of foveal ablation onemmetropization and form‐deprivation myopia. Invest Ophthalmol Vis Sci2007;48:3914‐3922.

43. Liu Y, Wildsoet C. The effect of two‐zone concentric bifocal spectacle lenses onrefractiveerrordevelopmentandeyegrowthinyoungchicks.InvestOphthalmolVisSci2011;52:1078‐1086.

44. LiuY,WildsoetC.Theeffectiveaddinherentin2‐zonenegativelensesinhibitseyegrowthinmyopicyoungchicks.InvestOphthalmolVisSci2012;53:5085‐5093.

45. HoWC,NgYF,ChuPH,etal.Impairmentofretinaladaptivecircuitryinthemyopiceye.VisionRes2011;51:367‐375.

46. RitcheyER,ZelinkaC,TangJ,etal.Vision‐guidedoculargrowthinamutantchickenmodelwithdiminishedvisualacuity.ExpEyeRes2012;102:59‐69.

47. Fischer AJ, Morgan IG, Stell WK. Colchicine causes excessive ocular growth andmyopiainchicks.VisionRes1999;39:685‐697.

48. FischerAJ,SeltnerRL,StellWK.N‐methyl‐D‐aspartate‐inducedexcitotoxicitycausesmyopiainhatchedchicks.CanJOphthalmol1997;32:373‐377.

49. Mariani AP. Amacrine cells of the rhesus monkey retina. J Comp Neurol1990;301:382‐400.

50. KolbH,LinbergKA,FisherSK.Neuronsof thehumanretina:aGolgistudy. JCompNeurol1992;318:147‐187.

51. FischerAJ,McGuireJJ,SchaeffelF,StellWK.Light‐andfocus‐dependentexpressionofthetranscriptionfactorZENKinthechickretina.NatNeurosci1999;2:706‐712.

52. WallmanJ.Retinalcontrolofeyegrowthandrefraction.In:OsborneNN,ChaderGJ(eds),ProgressinRetinalResearch.NewYork:PergamonPress;1993:133‐153.

53. BitzerM, Schaeffel F.Defocus‐induced changes inZENKexpression in the chickenretina.InvestOphthalmolVisSci2002;43:246‐252.

54. Fischer AJ, Skorupa D, Schonberg DL, Walton NA. Characterization of glucagon‐expressingneuronsinthechickenretina.JCompNeurol2006;496:479‐494.

55. FischerAJ,RitcheyER,ScottMA,WynneA.Bullwhipneuronsintheretinaregulatethesizeandshapeoftheeye.DevBiol2008;317:196‐212.

56. Hasegawa S, TerazonoK,Nata K, TakadaT, YamamotoH, OkamotoH.Nucleotidesequence determination of chicken glucagon precursor cDNA. Chickenpreproglucagondoesnotcontainglucagon‐likepeptideII.FEBSLett1990;264:117‐120.

57. Feldkaemper MP, Wang HY, Schaeffel F. Changes in retinal and choroidal geneexpressionduringdevelopmentofrefractiveerrorsinchicks.InvestOphthalmolVisSci2000;41:1623‐1628.

58. BuckC,SchaeffelF,SimonP,FeldkaemperM.Effectsofpositiveandnegative lenstreatmentonretinalandchoroidalglucagonandglucagonreceptormRNAlevelsinthechicken.InvestOphthalmolVisSci2004;45:402‐409.

86

59. AshbyR,KozulinP,MegawPL,Morgan IG.Alterations inZENKandglucagonRNAtranscript expression during increased ocular growth in chickens. Mol Vis2010;16:639‐649.

60. FeldkaemperMP,SchaeffelF.Evidence forapotential roleofglucagonduringeyegrowthregulationinchicks.VisNeurosci2002;19:755‐766.

61. FeldkaemperMP,BurkhardtE,SchaeffelF.Localizationandregulationofglucagonreceptorsinthechickeyeandpreproglucagonandglucagonreceptorexpressioninthemouseeye.ExpEyeRes2004;79:321‐329.

62. Vessey KA, Lencses KA, Rushforth DA, Hruby VJ, Stell WK. Glucagon receptoragonistsandantagonistsaffectthegrowthofthechickeye:aroleforglucagonergicregulationofemmetropization?InvestOphthalmolVisSci2005;46:3922‐3931.

63. Zhu X, Wallman J. Opposite effects of glucagon and insulin on compensation forspectaclelensesinchicks.InvestOphthalmolVisSci2009;50:24‐36.

64. ReisRA,VenturaAL,KubruslyRC,deMelloMC,deMelloFG.Dopaminergicsignalinginthedevelopingretina.BrainResRev2007;54:181‐188.

65. RohrerB,StellWK.LocalizationofputativedopamineD2‐likereceptorsinthechickretina, using in situ hybridization and immunocytochemistry. Brain Res1995;695:110‐116.

66. Dearry A, Burnside B. Stimulation of distinct D2 dopaminergic and alpha 2‐adrenergic receptors induces light‐adaptive pigment dispersion in teleost retinalpigmentepithelium.JNeurochem1988;51:1516‐1523.

67. BruininkA,DawisS,NiemeyerG,LichtensteigerW.Catecholaminergicbindingsitesincatretina,pigmentepitheliumandchoroid.ExpEyeRes1986;43:147‐151.

68. Nguyen‐LegrosJ,Versaux‐BotteriC,VernierP.Dopaminereceptorlocalizationinthemammalianretina.MolNeurobiol1999;19:181‐204.

69. KebabianJW,CalneDB.Multiplereceptorsfordopamine.Nature1979;277:93‐96.70. Vallone D, Picetti R, Borrelli E. Structure and function of dopamine receptors.

NeurosciBiobehavRev2000;24:125‐132.71. Brandies R, Yehuda S. The possible role of retinal dopaminergic system in visual

performance.NeurosciBiobehavRev2008;32:611‐656.72. Cohen Y, Peleg E, Belkin M, Polat U, Solomon AS. Ambient illuminance, retinal

dopamine releaseand refractivedevelopment in chicks.ExpEyeRes 2012;103:33‐40.

73. Megaw P,Morgan I, BoelenM. Vitreal dihydroxyphenylacetic acid (DOPAC) as anindexofretinaldopaminerelease.JNeurochem2001;76:1636‐1644.

74. Stone RA, Lin T, Laties AM, Iuvone PM. Retinal dopamine and form‐deprivationmyopia.ProcNatlAcadSciUSA1989;86:704‐706.

75. MegawPL,Morgan IG, BoelenMK.Dopaminergic behaviour in chicken retina andtheeffectofformdeprivation.AustNZJOphthalmol1997;25Suppl1:S76‐78.

76. Ohngemach S, Hagel G, Schaeffel F. Concentrations of biogenic amines in fundallayers in chickenswithnormal visual experience,deprivation, andafter reserpineapplication.VisNeurosci1997;14:493‐505.

77. GuoSS, Sivak JG,CallenderMG,Diehl‐JonesB.Retinaldopamineand lens‐inducedrefractiveerrorsinchicks.CurrEyeRes1995;14:385‐389.

78. Luft WA, Iuvone PM, Stell WK. Spatial, temporal, and intensive determinants ofdopaminereleaseinthechickretina.VisNeurosci2004;21:627‐635.

87

79. Rohrer B, Spira AW, Stell WK. Apomorphine blocks form‐deprivation myopia inchickens by a dopamine D2‐receptor mechanism acting in retina or pigmentedepithelium.VisNeurosci1993;10:447‐453.

80. McCarthy CS, Megaw P, Devadas M, Morgan IG. Dopaminergic agents affect theability of brief periods of normal vision to prevent form‐deprivationmyopia.ExpEyeRes2007;84:100‐107.

81. SchmidKL,WildsoetCF. Inhibitoryeffectsofapomorphineandatropineand theircombinationonmyopiainchicks.OptomVisSci2004;81:137‐147.

82. Nickla DL, Totonelly K. Dopamine antagonists and brief vision distinguish lens‐induced‐andform‐deprivation‐inducedmyopia.ExpEyeRes2011;93:782‐785.

83. SchaeffelF,HagelG,BartmannM,KohlerK,ZrennerE.6‐Hydroxydopaminedoesnotaffectlens‐inducedrefractiveerrorsbutsuppressesdeprivationmyopia.VisionRes1994;34:143‐149.

84. Iuvone PM, TiggesM, Stone RA, Lambert S, Laties AM. Effects of apomorphine, adopamine receptor agonist, onocular refractionandaxial elongation in aprimatemodelofmyopia.InvestOphthalmolVisSci1991;32:1674‐1677.

85. Schaeffel F, Bartmann M, Hagel G, Zrenner E. Studies on the role of the retinaldopamine/melatonin system in experimental refractive errors in chickens. VisionRes1995;35:1247‐1264.

86. Hutchins JB. Acetylcholine as a neurotransmitter in the vertebrate retina.ExpEyeRes1987;45:1‐38.

87. EckensteinF,ThoenenH.Productionofspecificantiseraandmonoclonalantibodiesto choline acetyltransferase: characterization and use for identification ofcholinergicneurons.EmboJ1982;1:363‐368.

88. VoigtT.Cholinergicamacrinecellsintheratretina.JCompNeurol1986;248:19‐35.89. MillarT,IshimotoI,JohnsonCD,EpsteinML,ChubbIW,MorganIG.Cholinergicand

acetylcholinesterase‐containing neurons of the chicken retina. Neurosci Lett1985;61:311‐316.

90. SpiraAW,MillarTJ, Ishimoto I, et al. Localizationof choline acetyltransferase‐likeimmunoreactivityintheembryonicchickretina.JCompNeurol1987;260:526‐538.

91. BaughmanRW,BaderCR.Biochemicalcharacterizationandcellular localizationofthecholinergicsysteminthechickenretina.BrainRes1977;138:469‐485.

92. Fukuda K, Kubo T, Akiba I, Maeda A, Mishina M, Numa S. Molecular distinctionbetweenmuscarinicacetylcholinereceptorsubtypes.Nature1987;327:623‐625.

93. NathansonNM.Molecularpropertiesofthemuscarinicacetylcholinereceptor.AnnuRevNeurosci1987;10:195‐236.

94. Tietje KM, Nathanson NM. Embryonic chick heart expresses multiple muscarinicacetylcholinereceptorsubtypes.Isolationandcharacterizationofageneencodinganovelm2muscarinicacetylcholinereceptorwithhighaffinityforpirenzepine.JBiolChem1991;266:17382‐17387.

95. Tietje KM, Goldman PS, NathansonNM. Cloning and functional analysis of a geneencoding a novel muscarinic acetylcholine receptor expressed in chick heart andbrain.JBiolChem1990;265:2828‐2834.

96. GadbutAP,GalperJB.AnovelM3muscarinicacetylcholinereceptorisexpressedinchickatriumandventricle.JBiolChem1994;269:25823‐25829.

88

97. CreasonS,TietjeKM,NathansonNM.IsolationandfunctionalcharacterizationofthechickM5muscarinicacetylcholinereceptorgene.JNeurochem2000;74:882‐885.

98. FischerAJ,McKinnonLA,NathansonNM,StellWK.Identificationandlocalizationofmuscarinicacetylcholinereceptorsintheoculartissuesofthechick.JCompNeurol1998;392:273‐284.

99. McKinnonLA,NathansonNM.Tissue‐specificregulationofmuscarinicacetylcholinereceptor expressionduring embryonicdevelopment. JBiolChem 1995;270:20636‐20642.

100. Pilar G, Nunez R, McLennan IS, Meriney SD. Muscarinic and nicotinic synapticactivationofthedevelopingchickeniris.JNeurosci1987;7:3813‐3826.

101. Meriney SD, Pilar G. Cholinergic innervation of the smooth muscle cells in thechoroidcoatofthechickeyeanditsdevelopment.JNeurosci1987;7:3827‐3839.

102. McBrienNA,JoblingAI,TruongHT,CottriallCL,GentleA.Expressionofmuscarinicreceptor subtypes in tree shrew ocular tissues and their regulation during thedevelopmentofmyopia.MolVis2009;15:464‐475.

103. SchmidtM,HumphreyMF,WassleH.Actionandlocalizationofacetylcholineinthecatretina.JNeurophysiol1987;58:997‐1015.

104. Schmid K, Wildsoet C. Inhibitory effects of apomorphine and atropine and theircombinationonmyopiainchicks.OptomVisSci2004;81:137‐147.

105. Stone RA, Lin T, Laties AM. Muscarinic antagonist effects on experimental chickmyopia.ExpEyeRes1991;52:755‐758.

106. McBrien NA, Moghaddam HO, Reeder AP. Atropine reduces experimental myopiaandeyeenlargementviaanonaccommodativemechanism.InvestOphthalmolVisSci1993;34:205‐215.

107. CottriallCL,McBrienNA.TheM1muscarinicantagonistpirenzepinereducesmyopiaand eye enlargement in the tree shrew. InvestOphthalmolVisSci 1996;37:1368‐1379.

108. Cottriall CL, McBrien NA, Annies R, Leech EM. Prevention of form‐deprivationmyopia with pirenzepine: a study of drug delivery and distribution. OphthalmicPhysiolOpt1999;19:327‐335.

109. Luft WA, Ming Y, Stell WK. Variable effects of previously untested muscarinicreceptor antagonists on experimental myopia. Invest Ophthalmol Vis Sci2003;44:1330‐1338.

110. McBrien NA, Cottriall CL, Annies R. Retinal acetylcholine content in normal andmyopiceyes:aroleinoculargrowthcontrol?VisNeurosci2001;18:571‐580.

111. Pendrak K, Lin T, Stone RA. Ciliary ganglion choline acetyltransferase activity inavianmacrophthalmos.ExpEyeRes1995;60:237‐243.

112. Fischer AJ, Miethke P, Morgan IG, Stell WK. Cholinergic amacrine cells are notrequired for the progression and atropine‐mediated suppression of form‐deprivationmyopia.BrainRes1998;794:48‐60.

113. Lind GJ, Chew SJ, Marzani D, Wallman J. Muscarinic acetylcholine receptorantagonists inhibit chick scleral chondrocytes. Invest Ophthalmol Vis Sci1998;39:2217‐2231.

114. SchwahnHN,KaymakH,SchaeffelF.Effectsofatropineonrefractivedevelopment,dopamine release, and slow retinal potentials in the chick. Vis Neurosci2000;17:165‐176.

89

115. Means AL, Gudas LJ. The roles of retinoids in vertebrate development. AnnuRevBiochem1995;64:201‐233.

116. MangelsdorfDJ,ThummelC,BeatoM, et al.Thenuclear receptor superfamily: theseconddecade.Cell1995;83:835‐839.

117. SekoY,ShimizuM,TokoroT.Retinoicacidincreasesintheretinaofthechickwithformdeprivationmyopia.OphthalmicRes1998;30:361‐367.

118. MertzJR,WallmanJ.Choroidalretinoicacidsynthesis:apossiblemediatorbetweenrefractiveerrorandcompensatoryeyegrowth.ExpEyeRes2000;70:519‐527.

119. MertzJR,HowlettMHC,McFaddenSA,WallmanJ.Retinoicacidfromboththeretinaand choroid influences eye growth. Invest Ophthalmol Vis Sci (ARVO Suppl)1999;40:S849.

120. McFadden SA, HowlettMH,Mertz JR. Retinoic acid signals the direction of ocularelongationintheguineapigeye.VisionRes2004;44:643‐653.

121. MaoJF,LiuSZ,DouXQ.Retinoicacidmetabolicchangeinretinaandchoroidoftheguineapigwithlens‐inducedmyopia.IntJOphthalmol2012;5:670‐674.

122. SekoY,ShimokawaH,TokoroT.Invivoandinvitroassociationofretinoicacidwithform‐deprivationmyopiainthechick.ExpEyeRes1996;63:443‐452.

123. BitzerM, FeldkaemperM, Schaeffel F. Visually induced changes in components oftheretinoicacidsysteminfundallayersofthechick.ExpEyeRes2000;70:97‐106.

124. TroiloD,NicklaDL,Mertz JR, SummersRada JA. Change in the synthesis rates ofocular retinoic acid and scleral glycosaminoglycan during experimentally alteredeyegrowthinmarmosets.InvestOphthalmolVisSci2006;47:1768‐1777.

125. StoneRA,LatiesAM,RaviolaE,WieselTN. Increase inretinalvasoactive intestinalpolypeptide after eyelid fusion in primates.ProcNatlAcadSciUSA 1988;85:257‐260.

126. SeltnerRL,StellWK.Theeffectofvasoactive intestinalpeptideondevelopmentofformdeprivationmyopiainthechick:apharmacologicalandimmunocytochemicalstudy.VisionRes1995;35:1265‐1270.

127. TkatchenkoAV,WalshPA,TkatchenkoTV,GustincichS,RaviolaE.Formdeprivationmodulatesretinalneurogenesisinprimateexperimentalmyopia.ProcNatlAcadSciUSA2006;103:4681‐4686.

128. FeldkaemperMP,SchaeffelF.Insulininhibitscompensationofpluslensesinchicksand stimulatesmyopia development InvestOphthalmolVisSci2007;48: E‐Abstract5924.

129. Zhu X, Chandrasekar D, Wallman J. Insulin has effects on chick eyes opposite tothoseofglucagon.InvestOphthalmolVisSci2007;48:E‐Abstract5925.

130. RohrerB,StellWK.Basicfibroblastgrowthfactor(bFGF)andtransforminggrowthfactor beta (TGF‐beta) act as stop and go signals to modulate postnatal oculargrowthinthechick.ExpEyeRes1994;58:553‐561.

131. SimonP,FeldkaemperM,BitzerM,OhngemachS,SchaeffelF.EarlytranscriptionalchangesofretinalandchoroidalTGFbeta‐2,RALDH‐2,andZENKfollowingimposedpositiveandnegativedefocusinchickens.MolVis2004;10:588‐597.

132. AshbyR,McCarthyCS,MaleszkaR,MegawP,Morgan IG.Amuscarinic cholinergicantagonistandadopamineagonistrapidlyincreaseZENKmRNAexpressionintheform‐deprivedchickenretina.ExpEyeRes2007;85:15‐22.

90

133. McGlinn AM, Baldwin DA, Tobias JW, Budak MT, Khurana TS, Stone RA. Form‐deprivation myopia in chick induces limited changes in retinal gene expression.InvestOphthalmolVisSci2007;48:3430‐3436.

134. Stone RA, McGlinn AM, Baldwin DA, Tobias JW, Iuvone PM, Khurana TS. Imagedefocus and altered retinal gene expression in chick: clues to the pathogenesis ofametropia.InvestOphthalmolVisSci2011;52:5765‐5777.

135. Strauss O. The retinal pigment epithelium in visual function. Physiol Rev2005;85:845‐881.

136. Wimmers S, Karl MO, Strauss O. Ion channels in the RPE. Prog Retin Eye Res2007;26:263‐301.

137. MarmorM,Wolfensberger T.Theretinalpigmentepithelium: functionanddisease:OxfordUniversityPress;1998:745.

138. BuckC,SchaeffelF,SimonP,FeldkaemperM.Effectsofpositiveandnegative lenstreatmentonretinalandchoroidalglucagonandglucagonreceptormRNAlevelsinthechicken.InvestOphthalmolVisSci2004;45:402‐409.

139. Friedman Z, Hackett SF, Campochiaro PA. Human retinal pigment epithelial cellspossess muscarinic receptors coupled to calcium mobilization. Brain Res1988;446:11‐16.

140. Zhang Y, Liu Y, Wildsoet CF. Bidirectional, optical sign‐dependent regulation ofBMP2geneexpressioninchickretinalpigmentepithelium.InvestOphthalmolVisSci2012;53:6072‐6080.

141. ZhangY,LiuY,HoCarol,HammondD,CF.W.DifferentialexpressionofBMP7,TGF‐β2,andnoggininchickRPEafterimposedopticaldefocusInvestOphthalmolVisSci53,E‐Abstract34582012.

142. Lin T, Grimes PA, Stone RA. Expansion of the retinal pigment epithelium inexperimentalmyopia.VisionRes1993;33:1881‐1885.

143. Zhang Y, Liu Y, Xu J, Nimri N, Wildsoet CF. Microarray analysis of RPE geneexpression in chicks during long‐term imposed hyperopic defocus. InvestOphthalmolVisSci2010;51:E‐Abstract3680.

144. Fleming PA, Harman AM, Beazley LD. Changing topography of the RPE resultingfromexperimentallyinducedrapideyegrowth.VisNeurosci1997;14:449‐461.

145. Harman AM, Hoskins R, Beazley LD. Experimental eye enlargement in matureanimalschangestheretinalpigmentepithelium.VisNeurosci1999;16:619‐628.

146. Rodgers RL. Glucagon and cyclic AMP: time to turn the page? CurrDiabetesRev2012;8:362‐381.

147. Zhao P, Cladman W, Van Tol HH, Chidiac P. Fine‐tuning of GPCR signals byintracellularGproteinmodulators.ProgMolBiolTranslSci2013;115:421‐453.

148. Koh SW, Chader GJ. Elevation of intracellular cyclic AMP and stimulation ofadenylate cyclase activity by vasoactive intestinal peptide and glucagon in theretinalpigmentepithelium.JNeurochem1984;43:1522‐1526.

149. DongF,An JH,RenYP, et al. [Expressionofdopamine receptorD2andadenosinereceptor A2A in human retinal pigment epithelium]. Zhonghua Yan Ke Za Zhi2007;43:1110‐1113.

150. DearryA, FalardeauP, ShoresC,CaronMG.D2dopamine receptors in thehumanretina:cloningofcDNAand localizationofmRNA.CellMolNeurobiol1991;11:437‐453.

91

151. Zhang Y,Maminishkis A, Zhi C, et al. Apomorphine regulates TGF‐β1 and TGF‐β2expressioninhumanfetalretinalpigmentepithelialcells. InvestOphthalmolVisSci2009;50:E‐Abstract3845.

152. Gallemore RP, Steinberg RH. Effects of dopamine on the chick retinal pigmentepithelium.Membranepotentialsandlight‐evokedresponses.InvestOphthalmolVisSci1990;31:67‐80.

153. SekoY,TanakaY,TokoroT.Apomorphineinhibitsthegrowth‐stimulatingeffectofretinalpigmentepitheliumonscleralcellsinvitro.CellBiochemFunct1997;15:191‐196.

154. ChuaW‐H,BalakrishnanV,ChanY‐H,etal.Atropineforthetreatmentofchildhoodmyopia.Ophthalmology2006;113:2285‐2291.

155. Friedman Z, Hackett SF, Campochiaro PA. Human retinal pigment epithelial cellspossess muscarinic receptors coupled to calcium mobilization. Brain Res1988;446:11‐16.

156. OsborneNN,FitzGibbonF,SchwartzG.Muscarinicacetylcholinereceptor‐mediatedphosphoinositide turnover in cultured human retinal pigment epithelium cells.VisionRes1991;31:1119‐1127.

157. Salceda R. Muscarinic receptors binding in retinal pigment epithelium during ratdevelopment.NeurochemRes1994;19:1207‐1210.

158. de The H, Marchio A, Tiollais P, Dejean A. Differential expression and ligandregulationof the retinoic acid receptor alpha andbeta genes.EmboJ 1989;8:429‐433.

159. Balmer JE, Blomhoff R. Gene expression regulation by retinoic acid. J Lipid Res2002;43:1773‐1808.

160. Waldbillig RJ, Arnold DR, Fletcher RT, Chader GJ. Insulin and IGF‐I binding indeveloping chick neural retina and pigment epithelium: a characterization ofbindingandstructuraldifferences.ExpEyeRes1991;53:13‐22.

161. ZhangZ,GellerS, Su J,Flannery JG,WildsoetCF.Effectof imposeddefocuson theprofileofgeneexpressioninchickretinalpigmentepithelium.InvestOphthalmolVisSc2007;48:E‐Abstract4417.

162. Koh SW. VIP stimulation of polarized macromolecule secretion in cultured chickembryonicretinalpigmentepithelium.ExpCellRes1991;197:1‐7.

163. MartinDM,YeeD,FeldmanEL.Geneexpressionof the insulin‐likegrowth factorsandtheirreceptorsinculturedhumanretinalpigmentepithelialcells.BrainResMolBrainRes1992;12:181‐186.

164. Takagi H, Yoshimura N, Tanihara H, Honda Y. Insulin‐like growth factor‐relatedgenes,receptors,andbindingproteinsinculturedhumanretinalpigmentepithelialcells.InvestOphthalmolVisSci1994;35:916‐923.

165. Kvanta A. Expression and secretion of transforming growth factor‐beta intransformed and nontransformed retinal pigment epithelial cells. OphthalmicRes1994;26:361‐367.

166. TaniharaH,YoshidaM,MatsumotoM,YoshimuraN. Identificationof transforminggrowth factor‐beta expressed in cultured human retinal pigment epithelial cells.InvestOphthalmolVisSci1993;34:413‐419.

92

167. Jacquemin E, Halley C, Alterio J, LaurentM, Courtois Y, Jeanny JC. Localization ofacidicfibroblastgrowthfactor(aFGF)mRNAinmouseandbovineretinabyinsituhybridization.NeurosciLett1990;116:23‐28.

168. Sternfeld MD, Robertson JE, Shipley GD, Tsai J, Rosenbaum JT. Cultured humanretinal pigment epithelial cells express basic fibroblast growth factor and itsreceptor.CurrEyeRes1989;8:1029‐1037.

169. Alge‐Priglinger CS, Kreutzer T, Obholzer K, et al. Oxidative stress‐mediatedinduction of MMP‐1 and MMP‐3 in human RPE cells. Invest Ophthalmol Vis Sci2009;50:5495‐5503.

170. Alexander JP, Bradley JM, Gabourel JD, Acott TS. Expression of matrixmetalloproteinases and inhibitor by human retinal pigment epithelium. InvestOphthalmolVisSci1990;31:2520‐2528.

171. ZhangY,LiuY,HammondD,WildsoetCF.PossiblerolesofBMP2andBMP4inchickRPEinsignalingthesignofdefocus.InInvestOphthalmolVisSci2011;52:E‐Abstract6320.

172. De StefanoME,Mugnaini E. Fine structure of the choroidal coat of the avian eye.Lymphaticvessels.InvestOphthalmolVisSci1997;38:1241‐1260.

173. De StefanoME,Mugnaini E. Fine structure of the choroidal coat of the avian eye.Vascularization, supporting tissue and innervation. Anat Embryol (Berl)1997;195:393‐418.

174. Pendrak K, Papastergiou GI, Lin T, Laties AM, Stone RA. Choroidal vascularpermeabilityinvisuallyregulatedeyegrowth.ExpEyeRes2000;70:629‐637.

175. JunghansBM,CrewtherSG,LiangH,CrewtherDP.A role for choroidal lymphaticsduringrecoveryfromformdeprivationmyopia?OptomVisSci1999;76:796‐803.

176. Hirata A, Negi A. Morphological changes of choriocapillaris in experimentallyinducedchickmyopia.GraefesArchClinExpOphthalmol1998;236:132‐137.

177. Wallman J, Wildsoet C, Xu A, et al. Moving the retina: choroidal modulation ofrefractivestate.VisionRes1995;35:37‐50.

178. Nickla DL. Transient increases in choroidal thickness are consistently associatedwith brief daily visual stimuli that inhibit ocular growth in chicks. Exp EyeRes2007;84:951‐959.

179. ZhuX, ParkTW,Winawer J,Wallman J. In amatter ofminutes, the eye canknowwhichwaytogrow.InvestOphthalmolVisSci2005;46:2238‐2241.

180. KeeCS,MarzaniD,WallmanJ.Differencesintimecourseandvisualrequirementsofocularresponsestolensesanddiffusers.InvestOphthalmolVisSci2001;42:575‐583.

181. Hu W, Criswell MH, Fong SL, et al. Differences in the temporal expression ofregulatory growth factors during choroidal neovascular development.ExpEyeRes2009;88:79‐91.

182. Lambert V, Munaut C, Jost M, et al. Matrix metalloproteinase‐9 contributes tochoroidalneovascularization.AmJPathol2002;161:1247‐1253.

183. WangY,GilliesC,ConeRE,O'RourkeJ.Extravascularsecretionoft‐PAbytheintactsuperfusedchoroid.InvestOphthalmolVisSci1995;36:1625‐1632.

184. LuttyGA,MergesC,ThrelkeldAB,CroneS,McLeodDS.Heterogeneityinlocalizationof isoformsofTGF‐beta inhuman retina, vitreous, and choroid. InvestOphthalmolVisSci1993;34:477‐487.

93

185. JoblingAI,WanR,GentleA,BuiBV,McBrienNA.RetinalandchoroidalTGF‐betainthe tree shrew model of myopia: isoform expression, activation and effects onfunction.ExpEyeRes2009;88:458‐466.

186. RadaJA,HollawayLR,LamW,LiN,NapoliJL.IdentificationofRALDH2asavisuallyregulatedretinoicacidsynthesizingenzymeinthechickchoroid.InvestOphthalmolVisSci2012;53:1649‐1662.

187. DhillonB,NicklaD.Theoculargrowthinhibitioneffectedbydopamineagonistsandatropine is associated with transient increases in choroidal thickness in chicks.InvestOphthalmolVisSciE‐Abstract17322008.

188. NicklaDL,ChengAT.Atropineandapomorphineinemmetropization:evidenceforinhibitoryinteractionsonacommongrowthregulatorypathway.InvestOphthalmolVisSciE‐Abstract38482009.

189. MarshallGE.Humanscleralelastic system:an immunoelectronmicroscopic study.BrJOphthalmol1995;79:57‐64.

190. RadaJA,NicklaDL,TroiloD.Decreasedproteoglycansynthesisassociatedwithformdeprivation myopia in mature primate eyes. Invest Ophthalmol Vis Sci2000;41:2050‐2058.

191. KangRN,NortonTT.Alterationofscleralmorphologyintreeshrewswithinducedmyopia.InvestOphthalmolVisSci34,S12091993.

192. McBrien NA, Cornell LM, Gentle A. Structural and ultrastructural changes to thesclera in a mammalian model of high myopia. Invest Ophthalmol Vis Sci2001;42:2179‐2187.

193. Rada JA, Brenza HL. Increased latent gelatinase activity in the sclera of visuallydeprivedchicks.InvestOphthalmolVisSci1995;36:1555‐1565.

194. NortonTT,RadaJA.Reducedextracellularmatrixinmammaliansclerawithinducedmyopia.VisionRes1995;35:1271‐1281.

195. GentleA,LiuY,MartinJE,ContiGL,McBrienNA.Collagengeneexpressionandthealteredaccumulationof scleral collagenduring thedevelopmentofhighmyopia. JBiolChem2003;278:16587‐16594.

196. GuggenheimJA,McBrienNA.Form‐deprivationmyopiainducesactivationofscleralmatrixmetalloproteinase‐2 in tree shrew. InvestOphthalmolVisSci 1996;37:1380‐1395.

197. SiegwartJT,Jr.,NortonTT.ThetimecourseofchangesinmRNAlevelsintreeshrewsclera during induced myopia and recovery. Invest Ophthalmol Vis Sci2002;43:2067‐2075.

198. Gao H, Frost MR, Siegwart JT, Jr., Norton TT. Patterns of mRNA and proteinexpressionduringminus‐lenscompensationandrecoveryintreeshrewsclera.MolVis2011;17:903‐919.

199. Gentle A, McBrien NA. Retinoscleral control of scleral remodelling in refractivedevelopment:aroleforendogenousFGF‐2?Cytokine2002;18:344‐348.

200. PhillipsJR,KhalajM,McBrienNA.Inducedmyopiaassociatedwithincreasedscleralcreepinchickandtreeshreweyes.InvestOphthalmolVisSci2000;41:2028‐2034.

201. Siegwart JT, Jr.,NortonTT.Regulationof themechanicalproperties of tree shrewsclerabythevisualenvironment.VisionRes1999;39:387‐407.

202. GottliebMD, Joshi HB, Nickla DL. Scleral changes in chickswith form‐deprivationmyopia.CurrEyeRes1990;9:1157‐1165.

94

203. KusakariT,SatoT,TokoroT.Regionalscleralchangesin form‐deprivationmyopiainchicks.ExpEyeRes1997;64:465‐476.

204. ChristensenAM,WallmanJ.Evidencethatincreasedscleralgrowthunderliesvisualdeprivationmyopiainchicks.InvestOphthalmolVisSci1991;32:2143‐2150.

205. Rada JA, Thoft RA, Hassell JR. Increased aggrecan (cartilage proteoglycan)productioninthescleraofmyopicchicks.DevBiol1991;147:303‐312.

206. SekoY,TanakaY,TokoroT. Influenceof bFGFas apotent growth stimulator andTGF‐beta as a growth regulator on scleral chondrocytes and scleral fibroblasts invitro.OphthalmicRes1995;27:144‐152.

207. Hu J, Cui D, Yang X, et al. Bonemorphogenetic protein‐2: a potential regulator inscleralremodeling.MolVis2008;14:2373‐2380.

208. WangQ,ZhaoG,XingS,ZhangL,YangX.Roleofbonemorphogeneticproteins inform‐deprivationmyopiasclera.MolVis2011;17:647‐657.

209. Vincenti MP, White LA, Schroen DJ, Benbow U, Brinckerhoff CE. Regulatingexpression of the gene formatrixmetalloproteinase‐1 (collagenase):mechanismsthat control enzyme activity, transcription, andmRNA stability. CritRevEukaryotGeneExpr1996;6:391‐411.

210. LiuQ,WuJ,WangX,ZengJ.Changesinmuscarinicacetylcholinereceptorexpressioninformdeprivationmyopiainguineapigs.MolVis2007;13:1234‐1244.

211. GwiazdaJ.Treatmentoptionsformyopia.OptomVisSci2009;86:624‐628.212. TongL,WongEH,ChanYH,BalakrishnanV.Amultipleregressionapproachtostudy

opticalcomponentsofmyopiainSingaporeschoolchildren.OphthalmicPhysiolOpt2002;22:32‐37.

213. SawSM,CarkeetA,ChiaKS,StoneRA,TanDT.Componentdependentrisk factorsfor ocular parameters in Singapore Chinese children. Ophthalmology2002;109:2065‐2071.

214. MuttiDO,MitchellGL,MoeschbergerML,JonesLA,ZadnikK.Parentalmyopia,nearwork,schoolachievement,andchildren'srefractiveerror.InvestOphthalmolVisSci2002;43:3633‐3640.

215. JonesLA, Sinnott LT,MuttiDO,MitchellGL,MoeschbergerML, ZadnikK.Parentalhistory of myopia, sports and outdoor activities, and future myopia. InvestOphthalmolVisSci2007;48:3524‐3532.

216. TseDY,ToCH.Gradedcompetingregionalmyopicandhyperopicdefocusproducesummated emmetropization set points in chick. Invest Ophthalmol Vis Sci2011;52:8056‐8062.

217. SankaridurgP,HoldenB,SmithE,3rd,etal.Decreaseinrateofmyopiaprogressionwith a contact lens designed to reduce relative peripheral hyperopia: one‐yearresults.InvestOphthalmolVisSci2011;52:9362‐9367.

218. AnsticeNS,Phillips JR.Effectofdual‐focus soft contact lenswearonaxialmyopiaprogressioninchildren.Ophthalmology2011;118:1152‐1161.

219. ChoP,CheungSW.Retardationofmyopia inOrthokeratology(ROMIO)study:a2‐yearrandomizedclinicaltrial.InvestOphthalmolVisSci2012;53:7077‐7085.

220. SongYY,WangH,WangBS,QiH,RongZX,ChenHZ.Atropine inameliorating theprogressionofmyopiainchildrenwithmildtomoderatemyopia:ameta‐analysisofcontrolledclinicaltrials.JOculPharmacolTher2011;27:361‐368.

95

221. Lee JJ, Fang PC, Yang IH, et al. Prevention of myopia progression with 0.05%atropinesolution.JOculPharmacolTher2006;22:41‐46.

222. ChiaA,ChuaWH,CheungYB,etal.Atropineforthetreatmentofchildhoodmyopia:safetyandefficacyof0.5%,0.1%,and0.01%doses(AtropinefortheTreatmentofMyopia2).Ophthalmology2012;119:347‐354.

223. WardB,TaruttaEP,MayerMJ.Theefficacyandsafetyofposteriorpolebuckles inthe control of progressive high myopia. Eye 2009;Ahead of print. doi:10.1038/eye.2008.1433.

224. Su J, Wall ST, Healy KE, Wildsoet CF. Scleral reinforcement through host tissueintegration with biomimetic enzymatically degradable semi‐interpenetratingpolymernetwork.TissueEngPartA2010;16:905‐916.

225. GharbiyaM, Cruciani F, Parisi F, Cuozzo G, Altimari S, Abdolrahimzadeh S. Long‐term results of intravitreal bevacizumab for choroidal neovascularisation inpathologicalmyopia.BrJOphthalmol2012;96:1068‐1072.

226. IrvingEL,Sivak JG,CallenderMG.Refractiveplasticityof thedevelopingchickeye.OphthalmicPhysiolOpt1992;12:448‐456.

227. PapastergiouGI,SchmidGF,LatiesAM,PendrakK,LinT,StoneRA.Inductionofaxialeye elongation and myopic refractive shift in one‐year‐old chickens. Vision Res1998;38:1883‐1888.

228. Kocur I, Resnikoff S. New challenges for VISION 2020. Ophthalmic Epidemiol2005;12:291‐292.

229. GoldschmidtE.Refractioninthenewborn.ActaOphthalmol(Copenh)1969;47:570‐578.

230. GrahamMV,GrayOP.Refractionofprematurebabies'eyes.BrMedJ1963;1:1452‐1454.

231. DengL,GwiazdaJ.Birthseason,photoperiod,andinfancyrefraction.OptomVisSci2011;88:383‐387.

232. Saw SM, Tong L, Chua WH, et al. Incidence and progression of myopia inSingaporeanschoolchildren.InvestOphthalmolVisSci2005;46:51‐57.

233. McBrien NA, Norton TT. The development of experimental myopia and ocularcomponentdimensionsinmonocularly lid‐suturedtreeshrews(Tupaiabelangeri).VisionRes1992;32:843‐852.

234. Smith EL, 3rd, Hung LF. The role of optical defocus in regulating refractivedevelopmentininfantmonkeys.VisionRes1999;39:1415‐1435.

235. WagnerDO,SieberC,BhushanR,BorgermannJH,GrafD,KnausP.BMPs:frombonetobodymorphogeneticproteins.SciSignal2010;3:mr1.

236. Chen D, Zhao M, Mundy GR. Bone morphogenetic proteins. Growth Factors2004;22:233‐241.

237. PlasDT,DhandeOS,Lopez JE, etal.Bonemorphogeneticproteins, eyepatterning,andretinocollicularmapformationinthemouse.JNeurosci2008;28:7057‐7067.

238. HoganBL.Bmps:multifunctionalregulatorsofmammalianembryonicdevelopment.HarveyLect1996;92:83‐98.

239. DerynckR,MiyazonoK.TheTGF‐βfamily. ColdSpringHarbor,NewYork: : ColdSpringHarborLaboratoryPress2008:1‐149.

240. BandyopadhyayA,YadavPS,PrasharP.BMPsignalingindevelopmentanddiseases:Apharmacologicalperspective.BiochemPharmacol2013.

96

241. Belecky‐Adams T, Adler R. Developmental expression patterns of bonemorphogeneticproteins,receptors,andbindingproteinsinthechickretina.JCompNeurol2001;430:562‐572.

242. Sakuta H, Takahashi H, Shintani T, Etani K, Aoshima A, Noda M. Role of bonemorphogenicprotein2 in retinalpatterningandretinotectalprojection. JNeurosci2006;26:10868‐10878.

243. Mathura JR, Jr., Jafari N, Chang JT, et al. Bone morphogenetic proteins‐2 and ‐4:negativegrowthregulatorsinadultretinalpigmentedepithelium.InvestOphthalmolVisSci2000;41:592‐600.

244. FischerAJ, SchmidtM,OmarG,RehTA.BMP4andCNTFareneuroprotective andsuppressdamage‐inducedproliferationofMullergliaintheretina.MolCellNeurosci2004;27:531‐542.

245. Lyons KM, Hogan BL, Robertson EJ. Colocalization of BMP 7 and BMP 2 RNAssuggeststhatthesefactorscooperativelymediatetissueinteractionsduringmurinedevelopment.MechDev1995;50:71‐83.

246. Muller F, Rohrer H, Vogel‐Hopker A. Bone morphogenetic proteins specify theretinalpigmentepitheliuminthechickembryo.Development2007;134:3483‐3493.

247. Xu J, Zhu D, Sonoda S, et al. Over‐expression of BMP4 inhibits experimentalchoroidal neovascularization by modulating VEGF and MMP‐9. Angiogenesis2012;15:213‐227.

248. Dudley AT, Lyons KM, Robertson EJ. A requirement for bone morphogeneticprotein‐7 during development of the mammalian kidney and eye. Genes Dev1995;9:2795‐2807.

249. BakraniaP, EfthymiouM,Klein JC, et al.Mutations inBMP4 cause eye, brain, anddigitdevelopmentalanomalies:overlapbetweentheBMP4andhedgehogsignalingpathways.AmJHumGenet2008;82:304‐319.

250. HillgerF,HerrG,RudolphR,SchwarzE.BiophysicalcomparisonofBMP‐2,ProBMP‐2, and the free pro‐peptide reveals stabilization of thepro‐peptide by thematuregrowthfactor.JBiolChem2005;280:14974‐14980.

251. Israel DI, Nove J, Kerns KM, Moutsatsos IK, Kaufman RJ. Expression andcharacterization of bonemorphogenetic protein‐2 in Chinese hamster ovary cells.GrowthFactors1992;7:139‐150.

252. WangEA,RosenV,D'AlessandroJS,etal.Recombinanthumanbonemorphogeneticproteininducesboneformation.ProcNatlAcadSciUSA1990;87:2220‐2224.

253. Scheufler C, Sebald W, Hulsmeyer M. Crystal structure of human bonemorphogeneticprotein‐2at2.7Aresolution.JMolBiol1999;287:103‐115.

254. MuellerTD,Nickel J. Promiscuity and specificity inBMP receptor activation.FEBSLett2012;586:1846‐1859.

255. Felin JE,Mayo JL,LoosTJ, etal.Nuclearvariantsofbonemorphogeneticproteins.BMCCellBiol2010;11:20.

256. CuiY,HackenmillerR,BergL,etal.TheactivityandsignalingrangeofmatureBMP‐4 is regulated by sequential cleavage at two sites within the prodomain of theprecursor.GenesDev2001;15:2797‐2802.

257. Klosch B, FurstW, Kneidinger R, et al. Expression and purification of biologicallyactive rat bone morphogenetic protein‐4 produced as inclusion bodies inrecombinantEscherichiacoli.BiotechnolLett2005;27:1559‐1564.

97

258. NelsenSM,ChristianJL.Site‐specificcleavageofBMP4byfurin,PC6,andPC7.JBiolChem2009;284:27157‐27166.

259. Bessa PC, Cerqueira MT, Rada T, et al. Expression, purification and osteogenicbioactivity of recombinant human BMP‐4, ‐9, ‐10, ‐11 and ‐14. ProteinExprPurif2009;63:89‐94.

260. Monroe DG, Jin DF, Sanders MM. Estrogen opposes the apoptotic effects of bonemorphogeneticprotein7ontissueremodeling.MolCellBiol2000;20:4626‐4634.

261. UristMR.Bone:formationbyautoinduction.Science1965;150:893‐899.262. WozneyJM,RosenV,CelesteAJ,etal.Novelregulatorsofboneformation:molecular

clonesandactivities.Science1988;242:1528‐1534.263. WordingerRJ,ClarkAF.Bonemorphogeneticproteinsandtheirreceptorsintheeye.

ExpBiolMed(Maywood)2007;232:979‐992.264. Tominaga T, Abe H, Ueda O, et al. Activation of bone morphogenetic protein 4

signalingleadstoglomerulosclerosisthatmimicsdiabeticnephropathy.JBiolChem2011;286:20109‐20116.

265. Trousse F, Esteve P, Bovolenta P. Bmp4 mediates apoptotic cell death in thedevelopingchickeye.JNeurosci2001;21:1292‐1301.

266. Behesti H, Holt JK, Sowden JC. The level of BMP4 signaling is critical for theregulationofdistinctT‐boxgeneexpressiondomainsandgrowthalongthedorso‐ventralaxisoftheopticcup.BMCDevBiol2006;6:62.

267. CuiW,BryantMR,SweetPM,McDonnellPJ.Changesingeneexpressioninresponsetomechanicalstraininhumanscleralfibroblasts.ExpEyeRes2004;78:275‐284.

268. Sehgal R, Andres DJ, Adler R, Belecky‐Adams TL. Bone morphogenetic protein 7increases chick photoreceptor outer segment initiation. InvestOphthalmolVisSci2006;47:3625‐3634.

269. ShenW,FinneganS,LeinP,SullivanS,SlaughterM,HigginsD.Bonemorphogeneticproteins regulate ionotropic glutamate receptors in human retina. Eur JNeurosci2004;20:2031‐2037.

270. FrancisPH,RichardsonMK,BrickellPM,TickleC.Bonemorphogeneticproteinsanda signalling pathway that controls patterning in the developing chick limb.Development1994;120:209‐218.

271. RogersMB,RosenV,WozneyJM,GudasLJ.Bonemorphogeneticproteins‐2and‐4are involved in the retinoic acid‐induced differentiation of embryonal carcinomacells.MolBiolCell1992;3:189‐196.

272. Abrams KL, Xu J, Nativelle‐Serpentini C, Dabirshahsahebi S, Rogers MB. Anevolutionary and molecular analysis of Bmp2 expression. J Biol Chem2004;279:15916‐15928.

273. HelveringLM,SharpRL,OuX,GeiserAG.Regulationofthepromotersforthehumanbonemorphogeneticprotein2and4genes.Gene2000;256:123‐138.

274. ReddiAH.Morphogeneticmessages are in the extracellularmatrix: biotechnologyfrombenchtobedside.BiochemSocTrans2000;28:345‐349.

275. GoldmanDC,DonleyN,Christian JL.Genetic interactionbetweenBmp2andBmp4revealssharedfunctionsduringmultipleaspectsofmouseorganogenesis.MechDev2009;126:117‐127.

276. KawabataM, ImamuraT,MiyazonoK.Signal transductionbybonemorphogeneticproteins.CytokineGrowthFactorRev1998;9:49‐61.

98

277. Chalazonitis A,D'Autreaux F, GuhaU, et al. Bonemorphogenetic protein‐2 and ‐4limitthenumberofentericneuronsbutpromotedevelopmentofaTrkC‐expressingneurotrophin‐3‐dependentsubset.JNeurosci2004;24:4266‐4282.

278. BartmannM,SchaeffelF,HagelG,ZrennerE.Constantlightaffectsretinaldopaminelevels and blocks deprivation myopia but not lens‐induced refractive errors inchickens.VisNeurosci1994;11:199‐208.

279. OhngemachS,BuckC, SimonP, Schaeffel F, FeldkaemperM.Temporal changesofnovel transcripts in the chicken retina following imposed defocus. Mol Vis2004;10:1019‐1027.

280. SchippertR,BrandC,SchaeffelF,FeldkaemperMP.ChangesinscleralMMP‐2,TIMP‐2andTGFbeta‐2mRNAexpressionafterimposedmyopicandhyperopicdefocusinchickens.ExpEyeRes2006;82:710‐719.

281. Stone RA, Liu J, Sugimoto R, Capehart C, Zhu X, Pendrak K. GABA, experimentalmyopia,andoculargrowthinchick.InvestOphthalmolVisSci2003;44:3933‐3946.

282. XuJ,LiuY,WildsoetCF.Theeffectofopticnervesectionontheyokedinhibitionbyatropine of lens‐induced myopia in chicks InvestOphthalmolVis Sci 2010;51: E‐Abstract3677.

283. Verhoeven VJ, Hysi PG, Wojciechowski R, et al. Genome‐wide meta‐analyses ofmultiancestry cohorts identifymultiple new susceptibility loci for refractive errorandmyopia.NatGenet2013;45:314‐318.

284. VethKN,WillerJR,ColleryRF,etal.Mutationsinzebrafishlrp2resultinadult‐onsetocularpathogenesis thatmodelsmyopiaandotherriskfactors forglaucoma.PLoSGenet2011;7:e1001310.

285. Hornbeak DM, Young TL. Myopia genetics: a review of current research andemergingtrends.CurrOpinOphthalmol2009;20:356‐362.

286. JacobiFK,ZrennerE,BroghammerM,PuschCM.Ageneticperspectiveonmyopia.CellMolLifeSci2005;62:800‐808.

287. Cohen SY. Anti‐VEGF drugs as the 2009 first‐line therapy for choroidalneovascularizationinpathologicmyopia.Retina2009;29:1062‐1066.

288. IpJM,HuynhSC,RobaeiD,etal.Ethnicdifferencesintheimpactofparentalmyopia:findings from a population‐based study of 12‐year‐old Australian children. InvestOphthalmolVisSci2007;48:2520‐2528.

289. WooWW,LimKA,YangH,etal.RefractiveerrorsinmedicalstudentsinSingapore.SingaporeMedJ2004;45:470‐474.

290. Matsumura H, Hirai H. Prevalence of myopia and refractive changes in studentsfrom3to17yearsofage.SurvOphthalmol1999;44Suppl1:S109‐115.

291. SawSM,GazzardG,KohD, et al. Prevalence rates of refractive errors in Sumatra,Indonesia.InvestOphthalmolVisSci2002;43:3174‐3180.

292. WildsoetCF.Active emmetropization‐‐evidence for its existenceand ramificationsforclinicalpractice.OphthalmicPhysiolOpt1997;17:279‐290.

293. WatanabeD,TakagiH,SuzumaK,OhH,OhashiH,HondaY.Expressionofconnectivetissue growth factor and its potential role in choroidal neovascularization.Retina2005;25:911‐918.

294. McLeodDS, Grebe R, Bhutto I,Merges C, Baba T, Lutty GA. Relationship betweenRPEandchoriocapillarisinage‐relatedmaculardegeneration.InvestOphthalmolVisSci2009;50:4982‐4991.

99

295. Witmer AN, Vrensen GF, Van Noorden CJ, Schlingemann RO. Vascular endothelialgrowthfactorsandangiogenesisineyedisease.ProgRetinEyeRes2003;22:1‐29.

296. Lipshutz RJ, Fodor SP, Gingeras TR, Lockhart DJ. High density syntheticoligonucleotidearrays.NatGenet1999;21:20‐24.

297. BrandC,SchaeffelF,FeldkaemperMP.Amicroarrayanalysisofretinal transcriptsthatarecontrolledbyimagecontrastinmice.MolVis2007;13:920‐932.

298. Schippert R, Schaeffel F, Feldkaemper MP. Microarray analysis of retinal geneexpressioninchicksduringimposedmyopicdefocus.MolVis2008;14:1589‐1599.

299. SheltonL,TroiloD,LernerMR,GusevY,BrackettDJ,RadaJS.Microarrayanalysisofchoroid/RPE gene expression in marmoset eyes undergoing changes in oculargrowthandrefraction.MolVis2008;14:1465‐1479.

300. Becker C, Hammerle‐Fickinger A, Riedmaier I, Pfaffl MW. mRNA and microRNAqualitycontrolforRT‐qPCRanalysis.Methods2010;50:237‐243.

301. Schroeder A, Mueller O, Stocker S, et al. The RIN: an RNA integrity number forassigningintegrityvaluestoRNAmeasurements.BMCMolBiol2006;7:3.

302. RamanT,O'ConnorTP,HackettNR,etal.Qualitycontrolinmicroarrayassessmentofgeneexpressioninhumanairwayepithelium.BMCGenomics2009;10:493.

303. RDevelopmentCoreTeam.R:Alanguageandenvironmentforstatisticalcomputing.Vienna, Austria. <Error! Hyperlink reference not valid. R Foundation forStatisticalComputing;2009.

304. GentlemanRC,CareyVJ,BatesDM,etal.Bioconductor:opensoftwaredevelopmentforcomputationalbiologyandbioinformatics.GenomeBiol2004;5:R80.

305. Gautier L, Cope L, Bolstad BM, Irizarry RA. affy‐‐analysis of Affymetrix GeneChipdataattheprobelevel.Bioinformatics2004;20:307‐315.

306. IrizarryRA,HobbsB,CollinF, etal.Exploration,normalization, and summariesofhighdensityoligonucleotidearrayprobeleveldata.Biostatistics2003;4:249‐264.

307. Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP. Summaries ofAffymetrixGeneChipprobeleveldata.NucleicAcidsRes2003;31:e15.

308. KauffmannA,GentlemanR,HuberW.arrayQualityMetrics‐‐abioconductorpackageforqualityassessmentofmicroarraydata.Bioinformatics2009;25:415‐416.

309. Cope LM, Irizarry RA, Jaffee HA, Wu Z, Speed TP. A benchmark for AffymetrixGeneChipexpressionmeasures.Bioinformatics2004;20:323‐331.

310. LiuG, LoraineAE, ShigetaR, et al.NetAffx:Affymetrixprobesets and annotations.NucleicAcidsRes2003;31:82‐86.

311. HuangdaW,ShermanBT,LempickiRA.SystematicandintegrativeanalysisoflargegenelistsusingDAVIDbioinformaticsresources.NatProtoc2009;4:44‐57.

312. SaeedAI,BhagabatiNK,BraistedJC,etal.TM4microarraysoftwaresuite.MethodsEnzymol2006;411:134‐193.

313. SaeedAI, SharovV,White J, etal.TM4:a free,open‐sourcesystem formicroarraydatamanagementandanalysis.Biotechniques2003;34:374‐378.

314. Li M, Reilly C. Assessing the quality of hybridized RNA in Affymetrix GeneChipsusinglinearregression.JBiomolTech2008;19:122‐128.

315. LuytenFP,CunninghamNS,MaS,etal.Purificationandpartialaminoacidsequenceofosteogenin,aproteininitiatingbonedifferentiation.JBiolChem1989;264:13377‐13380.

100

316. Chubinskaya S, Hurtig M, Rueger DC. OP‐1/BMP‐7 in cartilage repair. IntOrthop2007;31:773‐781.

317. Christiansen JH, Coles EG, Wilkinson DG. Molecular control of neural crestformation,migrationanddifferentiation.CurrOpinCellBiol2000;12:719‐724.

318. Thomadakis G, Ramoshebi LN, Crooks J, Rueger DC, Ripamonti U.Immunolocalization of Bone Morphogenetic Protein‐2 and ‐3 and OsteogenicProtein‐1 during murine tooth root morphogenesis and in other craniofacialstructures.EurJOralSci1999;107:368‐377.

319. Xu J,HeS, SpeeC,RyanSJ,HintonDR.TNF‐{alpha} InducedMMP‐9Expression isRegulated by BMP‐4 in Choroidal Neovascularization Invest Ophthalmol Vis Sci2010;51:E‐Abstract4100.

320. Wozney JM.Thebonemorphogeneticprotein familyandosteogenesis.MolReprodDev1992;32:160‐167.

321. GroppeJ,GreenwaldJ,WiaterE,etal.StructuralbasisofBMPsignallinginhibitionbythecystineknotproteinNoggin.Nature2002;420:636‐642.

322. Adler R, Belecky‐Adams TL. The role of bone morphogenetic proteins in thedifferentiationoftheventralopticcup.Development2002;129:3161‐3171.

323. Zimmerman LB,De Jesus‐Escobar JM,HarlandRM. The Spemannorganizer signalnogginbindsandinactivatesbonemorphogeneticprotein4.Cell1996;86:599‐606.

324. WuXB,LiY,SchneiderA,etal.Impairedosteoblasticdifferentiation,reducedboneformation, and severe osteoporosis in noggin‐overexpressing mice. J Clin Invest2003;112:924‐934.

325. Devlin RD, Du Z, Pereira RC, et al. Skeletal overexpression of noggin results inosteopeniaandreducedboneformation.Endocrinology2003;144:1972‐1978.

326. BrunetLJ,McMahonJA,McMahonAP,HarlandRM.Noggin,cartilagemorphogenesis,andjointformationinthemammalianskeleton.Science1998;280:1455‐1457.

327. Yamada Y, Nezu J, Shimane M, Hirata Y. Molecular cloning of a novel vascularendothelialgrowthfactor,VEGF‐D.Genomics1997;42:483‐488.

328. Marconcini L,Marchio S,Morbidelli L, et al. c‐fos‐induced growth factor/vascularendothelialgrowthfactorDinducesangiogenesisinvivoandinvitro.ProcNatlAcadSciUSA1999;96:9671‐9676.

329. RisauW.Mechanismsofangiogenesis.Nature1997;386:671‐674.330. StackerSA,CaesarC,BaldwinME,etal.VEGF‐Dpromotesthemetastaticspreadof

tumorcellsviathelymphatics.NatMed2001;7:186‐191.331. Takahashi H, Shibuya M. The vascular endothelial growth factor (VEGF)/VEGF

receptorsystemanditsroleunderphysiologicalandpathologicalconditions.ClinSci(Lond)2005;109:227‐241.

332. Pe'er J, FolbergR, ItinA, GnessinH,Hemo I, Keshet E. Upregulated expression ofvascular endothelial growth factor in proliferative diabetic retinopathy. Br JOphthalmol1996;80:241‐245.

333. Hera R, Keramidas M, Peoc'h M, Mouillon M, Romanet JP, Feige JJ. Expression ofVEGF and angiopoietins in subfoveal membranes from patients with age‐relatedmaculardegeneration.AmJOphthalmol2005;139:589‐596.

334. Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z, Ferrara N. VEGF coupleshypertrophic cartilage remodeling, ossification and angiogenesis duringendochondralboneformation.NatMed1999;5:623‐628.

101

335. AchenMG,JeltschM,KukkE,etal.VascularendothelialgrowthfactorD(VEGF‐D)isaligandforthetyrosinekinasesVEGFreceptor2(Flk1)andVEGFreceptor3(Flt4).ProcNatlAcadSciUSA1998;95:548‐553.

336. Blaauwgeers HG, Holtkamp GM, Rutten H, et al. Polarized vascular endothelialgrowth factor secretion by human retinal pigment epithelium and localization ofvascularendothelialgrowthfactorreceptorsontheinnerchoriocapillaris.Evidenceforatrophicparacrinerelation.AmJPathol1999;155:421‐428.

337. Kim I, Ryan AM, Rohan R, et al. Constitutive expression of VEGF, VEGFR‐1, andVEGFR‐2innormaleyes.InvestOphthalmolVisSci1999;40:2115‐2121.

338. FitzgeraldME,WildsoetCF,ReinerA.Temporalrelationshipofchoroidalbloodflowandthicknesschangesduringrecoveryfromformdeprivationmyopiainchicks.ExpEyeRes2002;74:561‐570.

339. Troilo D, Nickla DL,Wildsoet CF. Choroidal thickness changes during altered eyegrowth and refractive state in a primate. InvestOphthalmolVisSci 2000;41:1249‐1258.

340. SiegwartJT, Jr.,NortonTT.Thesusceptibleperiodfordeprivation‐inducedmyopiaintreeshrew.VisionRes1998;38:3505‐3515.

341. OrlandiniM,Marconcini L, Ferruzzi R, Oliviero S. Identification of a c‐fos‐inducedgene that is related to the platelet‐derived growth factor/vascular endothelialgrowthfactorfamily.ProcNatlAcadSciUSA1996;93:11675‐11680.

342. ParralesA,Palma‐NicolasJP,LopezE,Lopez‐ColomeAM.ThrombinstimulatesRPEcell proliferation by promoting c‐Fos‐mediated cyclinD1 expression. JCellPhysiol2010;222:302‐312.

343. Sagar SM, Sharp FR. Light induces a Fos‐like nuclear antigen in retinal neurons.BrainResMolBrainRes1990;7:17‐21.

344. BetsholtzC.Biologyofplatelet‐derivedgrowthfactorsindevelopment.BirthDefectsResCEmbryoToday2003;69:272‐285.

345. AndraeJ,GalliniR,BetsholtzC.Roleofplatelet‐derivedgrowthfactorsinphysiologyandmedicine.GenesDev2008;22:1276‐1312.

346. RobbinsSG,MixonRN,WilsonDJ,etal.Platelet‐derivedgrowth factor ligandsandreceptors immunolocalized in proliferative retinal diseases. InvestOphthalmolVisSci1994;35:3649‐3663.

347. Yoshida M, Tanihara H, Yoshimura N. Platelet‐derived growth factor geneexpression inculturedhumanretinalpigmentepithelialcells.BiochemBiophysResCommun1992;189:66‐71.

348. Li R, Maminishkis A, Wang FE, Miller SS. PDGF‐C and ‐D inducedproliferation/migration of human RPE is abolished by inflammatory cytokines.InvestOphthalmolVisSci2007;48:5722‐5732.

349. CampochiaroPA,HackettSF,VinoresSA,etal.Platelet‐derivedgrowthfactorisanautocrinegrowthstimulatorinretinalpigmentedepithelialcells.JCellSci1994;107(Pt9):2459‐2469.

350. CampochiaroPA,SuggR,GrotendorstG,HjelmelandLM.Retinalpigmentepithelialcells produce PDGF‐like proteins and secrete them into their media. ExpEyeRes1989;49:217‐227.

102

351. NagineniCN,KuttyV,DetrickB,HooksJJ.ExpressionofPDGFandtheirreceptorsinhumanretinalpigmentepithelialcellsandfibroblasts:regulationbyTGF‐beta.JCellPhysiol2005;203:35‐43.

352. HellerLC,LiY,AbramsKL,RogersMB.TranscriptionalregulationoftheBmp2gene.Retinoic acid induction in F9 embryonal carcinoma cells and Saccharomycescerevisiae.JBiolChem1999;274:1394‐1400.

353. GolzS,LantinC,MeyJ.Retinoicacid‐dependentregulationofBMP4andTbx5intheembryonicchickretina.Neuroreport2004;15:2751‐2755.

354. VandenWijngaardA,PijpersMA, JoostenPH,Roelofs JM,VanzoelenEJ,OlijveW.Functional characterization of two promoters in the human bonemorphogeneticprotein‐4gene.JBoneMinerRes1999;14:1432‐1441.

355. Dorai H, Shepard A, Ozkaynak E, et al. The 5' flanking region of the human bonemorphogeneticprotein‐7gene.BiochemBiophysResCommun2001;282:823‐831.

356. GrimsrudCD,RosierRN,Puzas JE,etal.Bonemorphogeneticprotein‐7 ingrowth‐plate chondrocytes: regulation by retinoic acid is dependent on the stage ofchondrocytematuration.JOrthopRes1998;16:247‐255.

357. TrellesRD,LeonJR,KawakamiY,SimoesS, IzpisuaBelmonteJC.ExpressionofthechickvascularendothelialgrowthfactorDgeneduringlimbdevelopment.MechDev2002;116:239‐242.

358. Kaetzel DM. Transcription of the platelet‐derived growth factor A‐chain gene.CytokineGrowthFactorRev2003;14:427‐446.

359. MorganI,KucharskiR,KrongkaewN,FirthSI,MegawP,MaleszkaR.Screeningfordifferentialgeneexpressionduringthedevelopmentofform‐deprivationmyopiainthechicken.OptomVisSci2004;81:148‐155.

360. DolleP,RuberteE,LeroyP,Morriss‐KayG,ChambonP.Retinoicacidreceptorsandcellularretinoidbindingproteins.I.Asystematicstudyoftheirdifferentialpatternoftranscriptionduringmouseorganogenesis.Development1990;110:1133‐1151.

361. PavanB,DalpiazA,BiondiC, et al.AnRPE cell line as a useful in vitromodel forstudying retinoic acid receptor beta: expression and affinity. Biosci Rep2008;28:327‐334.

362. Sun H, Gilbert DJ, Copeland NG, Jenkins NA, Nathans J. Peropsin, a novel visualpigment‐like protein located in the apical microvilli of the retinal pigmentepithelium.ProcNatlAcadSciUSA1997;94:9893‐9898.

363. BaileyMJ,CassoneVM.Opsinphotoisomerasesinthechickretinaandpinealgland:characterization, localization, and circadian regulation. Invest Ophthalmol Vis Sci2004;45:769‐775.

364. Koyanagi M, Terakita A, Kubokawa K, Shichida Y. Amphioxus homologs of Go‐coupled rhodopsin and peropsin having 11‐cis‐ and all‐trans‐retinals as theirchromophores.FEBSLett2002;531:525‐528.

365. Liang H, Crewther DP, Crewther SG, Barila AM. A role for photoreceptor outersegmentsintheinductionofdeprivationmyopia.VisionRes1995;35:1217‐1225.

366. CrewtherDP.Theroleofphotoreceptorsinthecontrolofrefractivestate.ProgRetinEyeRes2000;19:421‐457.

367. Maminishkis A, Chen S, Jalickee S, et al. Confluentmonolayers of cultured humanfetalretinalpigmentepitheliumexhibitmorphologyandphysiologyofnativetissue.InvestOphthalmolVisSci2006;47:3612‐3624.

103

368. ShiG,MaminishkisA,BanzonT,etal.Controlofchemokinegradientsbytheretinalpigmentepithelium.InvestOphthalmolVisSci2008;49:4620‐4630.

369. Li R, Maminishkis A, Banzon T, et al. IFN{gamma} regulates retinal pigmentepithelialfluidtransport.AmJPhysiolCellPhysiol2009;297:C1452‐1465.

370. SekoY,ShimokawaH,TokoroT.ExpressionofbFGFandTGF‐beta2inexperimentalmyopiainchicks.InvestOphthalmolVisSci1995;36:1183‐1187.

371. Jobling AI, Nguyen M, Gentle A, McBrien NA. Isoform‐specific changes in scleraltransforminggrowthfactor‐betaexpressionandtheregulationofcollagensynthesisduringmyopiaprogression.JBiolChem2004;279:18121‐18126.

372. HondaS,FujiiS,SekiyaY,YamamotoM.Retinalcontrolontheaxiallengthmediatedby transforming growth factor‐beta in chick eye. Invest Ophthalmol Vis Sci1996;37:2519‐2526.

373. KishiH,KurodaE,MishimaHK,YamashitaU.RoleofTGF‐betaintheretinoicacid‐induced inhibition of proliferation andmelanin synthesis in chick retinal pigmentepithelialcellsinvitro.CellBiolInt2001;25:1125‐1129.

374. Nagineni CN, Cherukuri KS, Kutty V, Detrick B, Hooks JJ. Interferon‐gammadifferentially regulates TGF‐beta1 and TGF‐beta2 expression in human retinalpigment epithelial cells through JAK‐STAT pathway. JCellPhysiol 2007;210:192‐200.

375. Pfeffer BA, Flanders KC, Guerin CJ, Danielpour D, Anderson DH. Transforminggrowth factor beta 2 is the predominant isoform in the neural retina, retinalpigment epithelium‐choroid and vitreous of the monkey eye. Exp Eye Res1994;59:323‐333.

376. SarkarDK,ChaturvediK,Oomizu S,BoyadjievaNI, ChenCP.Dopamine,dopamineD2 receptor short isoform, transforminggrowth factor (TGF)‐beta1, andTGF‐betatypeIIreceptorinteracttoinhibitthegrowthofpituitarylactotropes.Endocrinology2005;146:4179‐4188.

377. PastorcicM,DeA,BoyadjievaN,ValeW,SarkarDK.Reductionintheexpressionandactionoftransforminggrowthfactorbeta1onlactotropesduringestrogen‐inducedtumorigenesisintheanteriorpituitary.CancerRes1995;55:4892‐4898.

378. Boundy VA, Luedtke RR, Artymyshyn RP, Filtz TM, Molinoff PB. Development ofpolyclonalanti‐D2dopaminereceptorantibodiesusingsequence‐specificpeptides.MolPharmacol1993;43:666‐676.

379. Kebabian JW, Petzold GL, Greengard P. Dopamine‐sensitive adenylate cyclase incaudatenucleusofratbrain,anditssimilaritytothe"dopaminereceptor".ProcNatlAcadSciUSA1972;69:2145‐2149.

380. CreeseI,SibleyDR,HamblinMW,LeffSE.Theclassificationofdopaminereceptors:relationshiptoradioligandbinding.AnnuRevNeurosci1983;6:43‐71.

381. Li R,WenR, BanzonT,MaminishkisA,Miller SS.Neurotrophic factors affect RPEphysiology.InvestOphthalmolVisSci2010;51:E‐Abstract470.

382. MaW,ZhaoL,FontainhasAM,FarissRN,WongWT.Microglia inthemouseretinaalter the structure and function of retinal pigmented epithelial cells: a potentialcellularinteractionrelevanttoAMD.PLoSOne2009;4:e7945.

383. BeierM,FrankeA,Paunel‐GorguluAN,ScheererN,DunkerN.Transforminggrowthfactorbetamediatesapoptosisintheganglioncelllayerduringallprogrammedcelldeathperiodsofthedevelopingmurineretina.NeurosciRes2006;56:193‐203.

104

384. Hocking JC,HehrCL,ChangRY, Johnston J,McFarlaneS.TGFbeta ligandspromotetheinitiationofretinalganglioncelldendritesinvitroandinvivo.MolCellNeurosci2008;37:247‐260.

385. FriedmanZ,Hackett SF,CampochiaroPA.Characterizationof adenylate cyclase inhumanretinalpigmentepithelialcellsinvitro.ExpEyeRes1987;44:471‐479.

386. JoblingAI,GentleA,MetlapallyR,McGowanBJ,McBrienNA.Regulation of scleralcellcontractionbytransforminggrowth factor‐betaandstress:competingroles inmyopiceyegrowth.JBiolChem2009;284:2072‐2079.

387. ConnorTB,Jr.,RobertsAB,SpornMB,etal.Correlationoffibrosisandtransforminggrowthfactor‐betatype2levelsintheeye.JClinInvest1989;83:1661‐1666.

388. Lyons RM, Keski‐Oja J, Moses HL. Proteolytic activation of latent transforminggrowthfactor‐betafromfibroblast‐conditionedmedium. JCellBiol1988;106:1659‐1665.

389. Yu Q, Stamenkovic I. Cell surface‐localized matrix metalloproteinase‐9proteolyticallyactivatesTGF‐betaandpromotes tumor invasionandangiogenesis.GenesDev2000;14:163‐176.

390. Schultz‐Cherry S, Murphy‐Ullrich JE. Thrombospondin causes activation of latenttransforming growth factor‐beta secreted by endothelial cells by a novelmechanism.JCellBiol1993;122:923‐932.

391. Munger JS, Huang X, Kawakatsu H, et al. The integrin alpha v beta 6 binds andactivates latent TGF beta 1: amechanism for regulating pulmonary inflammationandfibrosis.Cell1999;96:319‐328.

392. Tan J, Deng ZH, Liu SZ, Wang JT, Huang C. TGF‐beta2 in human retinal pigmentepithelial cells: expression and secretion regulated by cholinergic signals in vitro.CurrEyeRes2010;35:37‐44.

393. LiuNP,FitzgibbonF,NashM,OsborneNN.Epidermalgrowthfactorpotentiatesthetransmitter‐induced stimulation of C‐AMP and inositol phosphates in humanpigmentepithelialcellsinculture.ExpEyeRes1992;55:489‐497.

394. StrunnikovaNV,MaminishkisA,BarbJJ,etal.Transcriptomeanalysisandmolecularsignatureofhumanretinalpigmentepithelium.HumMolGenet2010;19:2468‐2486.

395. Solouki AM, Verhoeven VJ, van Duijn CM, et al. A genome‐wide association studyidentifiesasusceptibilitylocusforrefractiveerrorsandmyopiaat15q14.NatGenet2010;42:897‐901.

Role of Retinal Pigment Epithelium in Myopia …...based, corneal reshaping therapy and atropine, a...

Documents

Transcript of Role of Retinal Pigment Epithelium in Myopia …...based, corneal reshaping therapy and atropine, a...