Intracellular routing of β-catenin Hendriksen, J.V.R.B.

Intracellular routing of β-catenin

Hendriksen, J.V.R.B.

Citation

Hendriksen, J. V. R. B. (2008, June 19). Intracellular routing of β-catenin.

Retrieved from https://hdl.handle.net/1887/12965

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/12965

Note: To cite this publication please use the final published version (if

applicable).

Chapter 4

Plasma membrane recruitment of signaling-competent ß-catenin upon activation of the Wnt pathway

JCellSci.2008;121:1793-1802

4

Plasma membrane recruitment of signaling competent ß- catenin upon activation of the Wnt pathway

JolitaHendriksen

,MarnixJansen

,CarolynM.Brown

,

HellavanderVelde

,MarcovanHam

,NielsGaljart

,G.JohanOfferhaus

,

FrancoisFagotto

andMaartenFornerod

1Dept.ofTumorBiology,TheNetherlandsCancerInstitute,Plesmanlaan121,1066CXAmsterdam,TheNetherlands 2DepartmentofPathology,UniversityMedicalCenterUtrecht,3584ZXUtrecht,TheNetherlands

3Dept.ofCellBiology,ErasmusMC,Dr.Molewaterplein50,3015GERotterdam,TheNetherlands 4 Department of Biology, McGill University,1205 Dr. Penfield Ave., Montreal, QC H3A 1B1, Canada 5Theseauthorscontributedequallytothiswork

The standard model of Wnt signaling specifies that after receipt of a Wnt ligand at the mem- brane-associated receptor complex, downstream mediators inhibit a cytoplasmic destruc- tion complex, allowing ß-catenin to accumulate in the cytosol and nucleus and co-activate Wnt target genes. Unexpectedly, upon Wnt treatment, we detected the dephosphorylated form of ß-catenin at the plasma membrane, displaying a discontinuous punctate labeling. This pool of ß-catenin could only be detected in E-cadherin (-/-) cells, because in E-cadherin (+/+) cells Wnt-induced, membrane-associated ß-catenin was conceiled by a constitutive junctional pool. Wnt signaling-dependent dephosphorylated ß-catenin co-localized at the plasma mem- brane with members of the destruction complex APC and Axin and the activated Wnt co-re- ceptor LRP6. ß-Catenin induced through the Wnt receptor complex was transcriptionally sig- nificantly more competent than overexpressed ß-catenin, both in cultured cells and in early Xenopus embryos. Our data reveal an unappreciated step in processing of the Wnt signal and suggest multiple levels of regulation of signaling output beyond the level of protein accumulation.

TheWntpathwayisacriticaldeterminantofcell

proliferation during development and regenera- tiveprocessessuchasstemcellproliferationin

theadult(Clevers,2006).Aberrantactivationof

thepathwayhasbeenlinkedtooncogenesisin

multiple systems. A central player in the Wnt

pathwayisß-catenin.Mostofthecellularpoolof

ß-cateninistetheredtoE-cadherin(encodedby

theCdh1gene)asanadherensjunctioncompo- nentmediatingcell-celladhesion(McCreaetal.,

1991;Peiferetal.,1994a).Alessabundantpool

ofß-catenin,oftenreferredtoasthe‘free’pool

ofß-catenin,functionsincomplexwithTCF/Lef

transcription factors as a transcriptional co-ac- tivatorofWntsignalinginthenucleus(Cadigan

andNusse,1997).IntheabsenceofaWntsig- nal,thefreepoolofß-cateninistightlyregulated

through phosphorylation at specific N-terminal residues by a so-called ‘destruction complex’

consistingoftheserinekinasesCK1αandGSK3

and the tumor suppressors Adenomatous Pol- yposis Coli (APC) and Axin (Logan and Nusse,

2004). Phosphorylated ß-catenin is marked

for rapid ubiquitinization and degradation by

theproteasome.ReceiptofaWntligandatthe

membrane-associatedreceptorcomplexresults

in inhibition ofß-catenin breakdown, allowing

ß-catenintoaccumulate,enterthenucleusand

activateaWnttargetgeneprogram(Loganand

Nusse,2004).

However, our current understanding of Wnt

signal transduction andß-catenin processing

suffers from significant gaps. In particular, the make-upandsubcellularlocalizationofthema- ture destruction complex is unclear at present.

For this reason, the mechanism through which

thedestructioncomplexsensesligandengage- ment at the Frizzled/LRP receptor complex re- mains unidentified. Recently, increasing evidence suggestsimportantregulatorystepsintheturn- overofthedestructioncomplexmaytakeplace

attheplasmamembrane(forareviewseeCadi- ganandLiu,2006).EngagementoftheFrizzled

andLRP5/6co-receptorsonthecellsurfaceby

Wntligands,resultsinthephosphorylationofthe

intracellular domain of LRP5/6 by CK1γand/or

GSK3ß. Phosphorylated LRP5/6 presents a

dockingsiteforAxinthatisrecruitedtotheplas- ma membrane in response to Wnt stimulation

(Cliffeetal.,2003;Davidsonetal.,2005;Zenget

al.,2005)alongwithothercanonicalWntpath- waycomponentsincludingAxin,GSK3ßandFz8

(Bilic et al., 2007). The scaffold protein dishev- elled(Dvl)appearstoberequiredforthistrans-

4

location(Schwarz-Romondetal.,2007).Although

itremainsunproven,ithasbeenhypothesizedthat

cytoplasmicdestructionofß-cateninishaltedas

a result of Axin relocation, allowingß-catenin to

redistributetothenucleus.

Nuclear localization ofß-cateninisconsidereda

hallmark of Wnt activation, yet in many systems

itisonlyincidentallydetectedinthenucleus(An- derson et al., 2002; Kobayashi et al., 2000). The

nuclearleveloftheN-terminallydephosphorylat- ed (or ‘dephospho’) form ofß-catenin has been

showntocorrelatemuchbetterwithWntactivity

(Staaletal.,2002).Dephospho-ß-cateninhasbeen

suggested to reflect the de novo translatedformof

ß-catenin,whichisinvolvedinsignaltransduction

(Willertetal.,2002).Wesetouttooptimizeexperi- mentalconditionsforthedetectionofdephospho- ß-catenin in cultured mammalian cell lines. In a

seriesofcoloncarcinomacelllines,dephospho-ß- cateninoftenlocalizestotheplasmamembrane.

Although we find no correlation with either APC orß-catenin mutation status, the plasma mem- brane localization of dephospho-ß-catenin does

correlatewithE-cadherinexpression.Surprisingly,

stimulationofE-cadherin-/-cellswithWnt3Are- sultedintheappearanceofdephospho-ß-catenin

at the plasma membrane, where it co-localizes

withtheactivatedformofLRP6,APCandAxin.By

unmaskingthetranscriptionallycompetentpoolof

ß-catenin,weprovideevidenceforakeystepin

ß-cateninprocessingandWntsignaltransduction

attheplasmamembrane.

Results

Dephospho-ß-catenin is present in cadherin complexes

BecauseN-terminallydephosphorylatedß-catenin

representsabettermarkerforWntsignalingactiv- itythantotalß-catenin,weoptimizedexperimen- talconditionsallowingdetectionofdephospho-ß- catenin by theα-ABC (8E7) antibody(van Noort

etal.,2002)inculturedmammaliancelllines.This

antibody specifically reacts with an N-terminally unphosphorylated peptide (amino acids 36-44)

thatcontainstheGSK3ßtargetresiduesS37and

T41(vanNoortetal.,2007).Reproducibledetec- tionofdephospho-ß-cateninwiththeα-ABCanti- body was highly dependent on fixation and block- ingconditions,andfurtherimprovedafterantigen

retrieval (see Materials and Methods). Screening

ofasetofcoloncarcinomacelllinesusingthese

conditions showed both nuclear and plasma

membrane localization of dephospho-ß-catenin,

dependingonthecellline(Fig1A,Chapter3,this

thesis).Althoughwefoundnocorrelationwithei-

therAPCorß-cateninmutationstatus,theplasma

membrane localization of dephospho-ß-catenin

doescorrelatepositivelywithE-cadherinexpres- sion(Fig1,Chapter3,thisthesis).

Cadherin-independent plasma membrane lo- calization of dephospho-ß-catenin upon Wnt3A stimulation

In order to eliminate E-cadherin expression as

a confounding factor in our interpretation of en- dogenous Wnt-induced dephospho-ß-catenin

accumulation,weturnedtothemurinemammary

epithelial cell line Kep1, which does not express

E-cadherin due to Cre-mediated recombination

of both E-cadherin alleles, and compared it with

its E-cadherin (+/+) isogenic control counterpart

Kp6(Derksenetal.,2006).IntheabsenceofWnt

stimulation,thesecelllineslinedonotactivatea

Wnt-responsive luciferase reportergene, indicat- ingthatthesecelllinesdonotcarryWntpathway

activatingmutations(Fig1A).Earlierstudieshave

demonstratedthatinthepresenceofanintactde- structioncomplexthelossofE-cadherinisneutral

withrespecttoWntstimulation(vandeWetering

et al., 2001). No or very littleß-catenincouldbe

detected in unstimulated Kep1 cells. After 2.5 h

stimulation with Wnt3A protein, however, a clear

accumulation of the dephospho-ß-catenin form

wasseen(Fig1B).Incontrast,intheE-cadherin

+/+ Kp6 cells dephospho-ß-catenin was clearly

detectedbeforeWntstimulation(Fig1B).

Inviewofthetranscriptionalactivation,weantici- patedthatinWnt-stimulatedKep1cells,dephos- pho-ß-catenin would be mainly nuclear. Surpris- ingly,stimulationofE-cadherin-/-Kep1cellswith

Wnt3A resulted in the appearance of dephos- pho-ß-cateninattheplasmamembrane(Fig.1C).

NuclearstainingoftheABCantibodyisalsoob- served. Note that this is partly aspecific, as some nuclearstainingisalsoobservedinunstimulated

Kep1cellsandinNCI-H28ß-cateninknock-out

cells (data not shown), which is likely caused by

across-reactingprotein(inFig.1Bmarkedbyan

asterisk). As expected, a similar plasma mem- brane accumulation in response to Wnt stimula- tion in E-cadherin -/- Kep1 cells is confirmed with antibodies to totalß-catenin (Fig. 2C). The dis- continuous punctate plasma membrane labeling

ofdephospho-ß-cateninisstrikinglysimilartothe

plasma membrane-associated puncta described

for LRP6-Axin (Bilic et al., 2007) and dishevelled

(Dvl) (Schwarz-Romond et al., 2007) appearing

uponWnttreatment(seebelow).Inunstimulated

isogenicE-cadherin+/+Kp6cellsdephospho-ß- cateninisprominentattheplasmamembrane(Fig.

1C, right panel). This pool reflects transcriptionally

inactiveß-catenin,asnoreporteractivityisde- tected(Fig.1A).Wnt3Ainducesreporteractivity

inKp6(Fig.1A),andaminorincreaseinsignalis

indeed detected on Western blot (Fig. 1B), but

thisincreasedoesnottranslateinanynoticeable

increaseindephospho-ß-cateninstaininginsitu

Figure 1. Cadherin-independent plasma membrane localization of dephospho-ß-catenin upon Wnt3A stimula- tion. A.E-cadherinnegativecellsrespondnormallytoWnt3a.LuciferasereporterassayinKep1(E-cadherin-/-)and

Kp6(E-cadherin+/+)cellsusingtheTCFreporterTOP-TKandthecontrolFOP-TK,normalizedfortransfectionef- ficiency using pRL-CMV-Renilla. 24 Hours after transfection cells were stimulated overnight with Wnt3a conditioned orcontrolmediumandluciferaseactivitywasmeasured.B.(Dephospho)ß-cateninlevelsinKep1orKp6cells.Cells

wereinducedwithWnt3Aproteinorcontrolandanalyzed0.5or2.5hafterinductionbyWesternblottingusingan

antibodyrecognizingallformsofß-catenin(α-TCAT) or an antibody specific for the N-terminal dephospho form (α- ABC).*,cross-reactingepitope.C.Subcellularlocalizationofdephospho-ß-catenininE-Cadherinpositiveornegative

cellsuponWntstimulation.Kep1(E-cadherin-/-)orKp6(E-cadherin+/+)wereinducedwithWnt3Aproteinorcontrol

and analyzed 2.5 h after induction by immunolocalization with an antibody specific for the N-terminal dephospho form (α-ABC).DAPIwasusedasanuclearmarker.Notethatdephospho-ß-cateninlevelsinE-cad-/-cellsaremuchlower

thaninE-cad+/+cells,requiringunequalconfocalsettingstobeused.

Figure 1

* 0.5 2.5 0.5 2.5 0.5 2.5 0.5 2.5 h induction

E-Cadh +/+

E-Cadh -/-

Contr Wnt3A Contr Wnt3A

1 2 3 4 5 6 7 8

B

2 4 6 8 10 12 14

0

Relative luc activity (TOP/FOP)

Contr Wnt3A

Kep1(E-Cadh -/-) Kp6(E-Cadh +/+)

A

Wnt3A Vehicle

+DAPI

E-cadh-/- E-cadh+/+

+DAPI

C

Zoom

Vehicle Wnt3A Vehicle Wnt3A

4

(Figure 1C, right panel). This indicates that the

signaling-competent dephospho-ß-catenin in- ducedinresponsetoWntstimulationisrelatively

minor in comparison to the steady-state junc- tionalpoolofdephospho-ß-catenin.

The absence of classical cadherins from Kep1

cells was confirmed using a ‘pan-cadherin’ an- tibody, which recognizes E, N and P cadherin.

This antibody failed to show any membrane

staining in Kep1 cells, neither before nor after

Wnt3Astimulation(SupplementaryFigureS1A).

Wewantedtofurtherruleoutthepossibilitythat

the recruitment of dephospho-ß-catenin to the

plasmamembranecouldbetheresultofaWnt- induced upregulation of a cadherin or another

membrane protein acting as a cryptic docking

site.Therefore,westudiedthelocalizationofde- phospho-ß-catenininKep1cellsstimulatedwith

Wnt3A protein, in the presence or absence of

thetranscriptioninhibitoractinomycinD.Under

these conditions,ß-catenin was still stabilized

andrecruitedtotheplasmamembrane(Supple- mentaryFigureS1B),indicatingthatthisprocess

is independent of the induction of Wnt target

genes.

Dephospho-ß-catenin colocalizes with APC, Axin and LRP6 at the plasma membrane after Wnt3A stimulation

Earlier studies have suggested translocation

of members of the destruction complex to the

LRP6 co-receptor upon Wnt signaling (Cliffe et

al., 2003; Tolwinski et al., 2003). Phosphoryla- tion on threonine 1479 of LRP6 is required for

the recruitment of Axin and is carried out by

CK1γinresponsetoWntsignaling(Davidsonet

al., 2005). We therefore used a phospho-specific antibodyrecognizingthisresidueofLRP6inim- munofluorescence in Kep1 cells. We found that after stimulation with Wnt3A, phospho-LRP6

co-localized with dephospho-ß-catenin on the

plasmamembrane(Fig2A).Likewise,wefound

thatAxincolocalizedwithdephospho-ß-catenin

(Fig2B).TheseresultssuggestthattheN-termi- nallydephosphorylated,orsignaling-competent

form,ofß-cateninistranslocatedtotherecep- tor in a Wnt signaling-associated complex. An- other component of the destruction complex,

APC,hasbeenreportedtolocalizetotheplasma

membraneindifferentepithelialcelllines,includ- ing colon carcinoma cell lines (Miyashiro et al.,

1995).AsmanyavailableantibodiesagainstAPC

are not reliable for immunofluorescence (Bro- cardoetal.,2005),wedevelopedamonoclonal

ratantibody,3E7,thatdetectsendogenousAPC

in immunofluorescence studies and on Western blot (Supplementary Figure S2). APC detected

with this antibody also clearly colocalized with

ß-cateninonthemembraneofWnt3Astimulated

Kep1cells(Fig.2C).

To further characterize the involvement of the

Wnt receptor complex in recruitment of ß- catenin, we expressedΔN-LRP6, a dominant

active LRP6 receptor that mimics Wnt ligand

engagementatthereceptorcomplex.Wecould

express this protein in SK-BR-3 cells, a breast

cancercelllinewithahomozygousdeletionofE- cadherin(vandeWeteringetal.,2001).Expres- sionofthisconstructinKep1cellsfailed,sinceit

wasnotproperlypresentedattheplasmamem- braneinthesecells(datanotshown).Asshown

inFigure2D,expressionofΔN-LRP6resultedin

a prominent plasma membrane localization of

dephospho-ß-catenin, resembling the appear- ance of dephospho-ß-catenin in Kep1 cells at

theplasmamembraneafterWntstimulation(Fig.

2A,B).ThisindicatesthatactivationofLRP6is

involved in the E-cadherin-independent plasma

membranerecruitmentofdephospho-ß-catenin

inresponsetoWntsignaling.ExpressionofΔN- LRP6inSK-BR-3alsoresultedincolocalization

ofdephospho-ß-cateninwithAPC(Fig.2E).We

wereunabletodetectAxininSK-BR-3cells,pos- siblyduetoverylowexpressionlevels(datanot

shown). SK-BR-3 cells were found to be unre- sponsivetoWnt3Astimulation(datanotshown)

possiblyduetothefactthatthesecellslackthe

appropriateFrizzledreceptorforthisligand.We

concludethatactivationoftheWntpathwayby

eitherWnt3AordominantactiveLRP6leadsto

recruitment of Axin and/or APC and signaling

competentß-catenintotheplasmamembrane.

LRP6-initiated dephospho-ß-catenin is tran- scriptionally significantly more competent than ‘downstream-initiated’ dephospho-ß- catenin

OurdatasofarareconsistentwithamodelofWnt

signaltransductionwhere,uponWntstimulation,

de novosynthesizedß-cateninisattractedtothe

Wnt receptor complex together with members

of the destruction complex. As Wnt activation

resultsinco-activationofgenesbyß-cateninin

thenucleus,ß-cateninlikelyisreleasedfromthe

membranecomplexandisroutedtothenucleus.

Inordertotesttherelevanceofmembraneas- sociationofß-catenin,wecomparedtheactivity

ofß-catenin,eitherroutedornotroutedthrough

the Wnt receptor complex. To mimicß-catenin

accumulationduetoreceptoractivationweex- pressedΔN-LRP6inSK-BR-3cells.Toproduce

ß-cateninaccumulationwithoutreceptoractiva- tion, we over-expressed wild-typeß-catenin. If

thetransactivatingpotentialissimilarregardless

thesourceofdephospho-ß-catenin,theamount

ofluciferaseoutputisexpectedtocloselyparal- lel the amount of dephospho-ß-catenin gener- ated.AsshowninFig.3,expressionofΔN-LRP6

resulted in significant upregulation of a lucifease reporter gene (Fig. 3C). A similar degree of

TCF-reporter activation could be produced by

transfectionofwild-typeß-catenin,butthiswas

accompanied by much higher cellular levels of

dephospho-ß-catenin(Fig.3DandE).Also,there

wasnoenrichmentofdephospho-ß-cateninon

the plasma membrane under these conditions

(Fig. 3B), while a prominent plasma membrane

localization of the comparatively minor pool of

dephospho-ß-cateninwasinducedbyΔN-LRP6

(Fig. 3A). Thus, in spite of significantly lower cel- lular levels of dephospho-ß-catenin, a similar

degreeofTCFreporteroutputcanbeachieved

throughLRP6co-receptoractivation.

merge APC

total β -cat

H2B-RFP +ΔN-LRP H2B-RFP

H2B APC β-cat

10μm

merge Axin

DAPI Axin

Wnt3a Vehicle merge

phospho-LRP

DAPI pLRP deP-β-cat

10μm

Kep1 (E-Cadh -/-)

Figure 2

Wnt3a Vehicle merge

APC total β-cat

DAPI APC β-cat

10μm

Kep1 (E-Cadh -/-)

10μm 10μm

H2B-RFP +ΔN-LRP H2B-RFP

merge phospho-LRP

pLRP deP-β-cat

A

B

C

D

E ^H2B

H2B-RFP ΔN-LRP+ H2B-RFP

Wnt3a Vehicle

SKBR-3 (E-Cadh -/-)

SKBR-3 (E-Cadh -/-) Kep1 (E-Cadh -/-)

dephospho β-cat

dephospho β-cat

dephospho β-cat

deP-β-cat

Figure 2. Dephospho-ß-catenin colocalizes with p-LRP6, Axin and APC at the plasma membrane in a Wnt-de- pendent (Kep1) or ΔN-LRP6 dependent (SK-BR-3) manner. Kep1 cells were induced with purified Wnt3A protein or vehicle for 2.5 h (Kep1), or transfected with ΔN-LRP6-encoding plasmid (SK-BR-3) and analyzed for subcellular local- izationofdephospho-ß-cateninandphospho-LRP6(AandD),dephospho-ß-cateninandAxin(B),totalß-cateninand

APC(C),anddephospho-ß-cateninandAPC(E).Arrowsdenoteplasmamembranelocalizations.Mergedimagesare

alsoshownthatincludeDAPIasanuclearmarker(A-C)orH2B-mRFPasatransfectionmarker(DandE).

4

Supraphysiological levels of exogenous ß- catenin are required to mimic Wnt activity in Xenopus embryos

We sought to support these observations in a

secondmodel.Wntsignalingactivityintheearly

X. laevis embryo can be readily monitored by

theformationofanectopicbodyaxis.Wethus

compared the levels of exogenously expressed

ß-cateninrequiredtoinducesecondaryaxes,to

thelevelsproducedbytheendogenousdorsaliz- ingcenter,orbyamuchstrongeractivationofthe

pathwaybyectopicWntexpression(Fig.4G).

The endogenous Wnt pathway is active in the

blastula(stage8.5-9.5)andcanbedetectedas

Figure 3. LRP6-initiated dephospho-ß-catenin is transcriptionally more active than downstream-initiated dephos- pho-ß-catenin. Cadherin-deficient SK-BR-3 breast carcinoma cells were transiently transfected with 25 or 50 ng plasmid encoding wild-type ß-catenin or 160 ng of a plasmid encoding ΔN-LRP6 and in parallel analyzed by immunolocalization (A and B), TCF transcriptional activity (C) and Western blotting (D and E). A-B. Immunolocalization of dephospho-ß- catenin in cells exogenously expressing ß-catenin or ΔN-LRP6, identified by co-expression of mRFP-tagged histone H2B.C.TCF-dependenttranscriptionalactivityinSK-BR3cellstransfectedwithindicatedplasmids24hoursaftertrans- fection.TOP,TCF-reporterluciferaseactivity;FOP,mutatedTCF-reporteractivity.Valueswerenormalizedtoatransfec- tioncontrol(constitutiveRenillaluciferasereporter).D-E,WesternblotanalysisofcellsshowninA-C,detectingtotal(D)

ordephospho-ß-catenin(E)levels.4x,fourfoldloaded;0.5x,onehalfloaded.

10 15

FOPTOPFOPTOPFOPTOPFOPTOP

Relative luciferase activity (firefly/renilla)

control -cat -cat N-LRP 25 ng 50 ng 0

5

dephospho- -catenin H2B-mRFP dephospho- -catenin H2B-mRFP

N-LRP -cat (50 ng)

C

B

D A

20

E

Figure 3

* -cat25 ng

-cat50 ng N-LR P

conrol N-LR P

control -cat 25 ng

4x 0.5x

-cat50 ng -cat25 ng N-LR

P

1 2 3 4 5 6 7

1 2 3

actin

N-LRP

nuclear accumulation ofß-catenin strongest in

thedorsalside,butalsospreadthroughoutthe

prospectivemesoderm,whilethesignalremains

lowerintheectoderm(SchohlandFagotto,2002;

Schohl and Fagotto, 2003). With 50 pg Wnt8

mRNA,whichisinexcessoftheamountrequired

toinducecompletedorsalization,ß-cateninnu- clearlevelswerefoundtobeonlyslightlyhigher

thanlevelsinducedbytheendogenouspathway

(Fig. 4). However, the levels of exogenousß- catenincorrespondingtoinductionofasecond- ary axis were well beyond physiological levels,

bothinthecytoplasmandinthenucleus.These

observationsindicatethatexogenousß-catenin

is less effective at activating the pathway than

endogenousß-cateninregulatedbyWntsignals,

consistentwiththehypothesisthatWnt-induced

ß-cateninisqualitativelydifferent.

Quantitative differences in dephospho-ß- catenin plasma membrane labeling in Wnt re- sponding tissues in vivo

OurdataintheisogenicKep1andKp6celllines

demonstrate that dephospho-ß-catenin is re- cruited to the plasma membrane in response

to Wnt stimulation in an E-cadherin indepen- dentfashion.Suchanunbiasedanalysiswould

not be feasible in other in vitro model systems

such as the commonly used colon cancer cell

lines, where the Wnt pathway is constitutively

active and E-cadherin expression varies even

amongstsubclonesoflow-expressingcelllines

such as LS174T (unpublished data). We find that in our model system the increase in dephos- pho-ß-catenin is subtle and not detectable in situinanE-cadherin+/+background.However,

based on earlier data in the Drosophila system

from the Wieschaus lab (see below) we opti- mizedourstainingprotocolforthedetectionof

dephospho-ß-catenininothersystemsaswell.

As a first approach we double-stained Xenopus

embryo sections for total and dephospho-ß- catenin(Fig.5).Wedetectedaclearsignalfor

dephospho-ß-cateninattheplasmamembrane

ofmesodermalcells,butlessinectodermalcells

(Fig5B).Wntover-expressionledtomembrane

recruitment of dephospho-ß-catenin in ecto- dermal cells (Fig 5A) and increased membrane

stainingoftotalß-catenin.Despitethefactthat

thismodelsystemlackstheadvantageofanE- cadherinnegativebackground,theobservations

inaphysiologicalsettingofaquantitativediffer- encein(dephospho-)ß-cateninaccumulationat

theplasmamembraneinWntrespondingversus

Wntnon-respondingcellsareinagreementwith

theobservationsinourinitialmodelsystemnot

sufferingfromthisdrawback.Inordertounam- Figure 4. Supraphysiological levels of exogenous

ß-catenin are required to mimic Wnt activity in Xenopus embryos. ß-Catenin(1.0ng)orWnt8(50pg)

mRNAwasinjectedintotheventralsideof4-cellstage

embryos and total ß-catenin levels achieved at blas- tula stage were estimated by immunofluorescence on cryosections.A.Diagramofcrossectionofablastula

embryoindicatingtheareasusedtocompareß-catenin

staining.  B, C. Ventral-animal and dorsal regions of

uninjected embryos.  Arrows: nuclear ß-catenin.  Ar- rowheads:plasmamembranes.D.Ventralregionfrom

Wnt8-injected embryo.  Note the increased nuclear

signal.  E. Ventral region from ß-catenin-injected em- bryo.Notetheverystrongsignalthroughoutthecells.

F. Same field recorded with a 5 times shorter exposure time. G. Comparison of dorsalizing activity of Wnt

and ß-catenin in early Xenopus embryos. The degree

ofdorsalizationobtainedwasscoredattailbudstage.

The phenotypes were classified in five categories of in- creasing dorsalization: normal, partial duplicated axis,

complete duplicated axis, complete duplicated axis

withshortenaxisindicatingpartialglobaldorsalization

andcompletelydorsalized,i.e.globaldorsalizationwith

reducedornoaxisandradialheadstructures.pvalue

accordingtoχ2test.

4

Figure 4

G

none 25 pg Wnt

Dorsalized

Duplicated axis partially dors.

Completely duplicated axis Partial duplicated axis Normal

0.0 0.2 0.4 0.6 0.8 1.0

Fraction observed

n = 66 n = 40 n = 43 p = 1 x 10-12

Injection 1.0 ng

β-cat

biguously assign translocation of Wnt-induced

dephospho-ß-catenin as E-cadherin-indepen- dent however, an E-cadherin negative back- groundisrequired.Wepointoutthattheresults

ondephospho-ß-cateninplasmamembraneac- cumulationinresponsetoWntstimulationinthe

Xenopusmodelsystemparallelearlierdataob- tainedinDrosophila whereWgsignalingsimilarly

uninjected ventral uninjected dorsal Wnt-injected

DAPI total β-catenin de-P-β-catenin

merge

de-P-β-catenin

total β-catenin

uninjected ectoderm uninjected mesoderm

DAPI de-P-β-catenin

merge

Figure 5. Membrane localization of dephospho-ß-catenin in the early Xenopus embryo. Cryosectionsofstage9

embryoswerestainedfordephospho-ß-catenin(ABC)andtotalß–catenin(H102)andnucleiwerecounterstainedwith

DAPI. A. Selected fields of the ventral presumptive ectoderm and mesoderm (uninjected ventral), dorsal mesoderm (uninjecteddorsal),andventralectodermofanembryoinjectedwith50pgWnt8mRNA(Wnt-injected).Inuninjected

embryos,ABCstainstheoutlinesofmesodermalcells(me)(arrowheads)butnotectodermalcells(ec)(arrows).Mem- braneABCisdetectedinectodermalcellsofWnt-injectedembryos(arrowheads).Smallbrightcytoplasmicspots

seen all three color channels correspond to autofluorescent pigment granules. B. Higher magnification view of ABC stainingofectodermal(ec)andmesodermal(me)cellsofuninjectedcells.Thesetwoimageswereobtainedbycollec- tionofz-stacksfollowedby2D-nearestneighborsdiconvolutionandmergeof10imagesfromeachstack.

increasesplasmamembranelevelsofArmadillo

(the fly ß-cateninhomologue)inWg-responding

striperegions,inadditiontoelevatingcytosolic

,levels of the protein (Peifer et al., 1994b). This

suggests that an increase in Armadillo at the

plasmamembraneisimportantforendogenous

Wg signaling in the fruit fly as well.

We also tested Wnt-induced dephospho-ß- cateninplasmamembraneenrichmentinasec- ond in vivo model of Wnt signaling. The intes- tinal epithelium of the human small intestine is

organized into flask-shaped invaginations called crypts and finger-like projections termed villi.

Wntsignalinghasbeenshowntobeessentialfor

maintaining stem cell turn-over in the intestinal

crypt (Korinek et al., 1998). We stained normal

humansmallintestinalepitheliumfordephospho- ß-cateninandfoundthatsimilartothesituation

inXenopus,dephospho-ß-cateninisenrichedat

the plasma membrane of Wnt-responsive crypt

epithelialcellswhencomparedtodifferentiated

cellsonthevillusepithelium(Fig.6,Chapter3,

Fig3AandB).

Weemphasizethatitremainstobetestedwheth- erinthesetwoin vivoexamplesofWntsignaling,

the plasma membrane localization of (dephos- pho-)b-catenin,whichcoincideswithregionsof

known Wnt activity, represents the E-cadherin- independent signaling competent form, as this

formcouldstillpotentiallybemasked.

Discussion

The currently prevailing model of Wnt signal

transduction specifies that upon receipt of a Wnt ligand at the membrane-associated receptor

complex,ß-cateninproteolysisispreventedand

theproteinaccumulatesinthecytosol.Ithasre- mainedunclearhowcytosolicdestructioncom- plexessenseligandengagementattheplasma

membrane.RecentstudieshaveshownthatWnt

treatment induces the formation of LRP6 co- receptor aggregates at the plasma membrane

Figure 6. Dephospho-ß-catenin is enriched on the plasma membrane in human intestinal crypts. A.Low-power

photomicrographofanormalhumansmallintestinalcrypt-villusaxis.Dephospho-ß-cateninisenrichedonthecrypt

epithelialplasmamembranewhencomparedtotheplasmamembranelabelingondifferentiatedvilluscells.B.High- powerphotomicrographofcryptcompartimentshowninA.Cryptepithelialcellsincludingthepresumptiveintestinal

epithelialstemcellsorcryptbasecolumnar(CBC)cells(Barkeretal.,2007)showrobustplasmamembranelabel-

ling.

4

(Bilicetal.,2007).Thesereceptoraggregatesin

turn promote the recruitment of canonical Wnt

pathwaycomponentsincludingdishevelled,Axin

andGSK3ß(Bilicetal.,2007;Schwarz-Romond

etal.,2007).Theseobservationsparallelearlier

data obtained in the Drosophila system where

similarlyrecruitmentofpathwaycomponentsto

theplasmamembranehasbeenrecordedupon

activationofthepathway(Cliffeetal.,2003).How

theinhibitionofß-cateninproteolysistiesupwith

theformationoftheseLRP6signalosomesand

whetherß-cateninitselfmaytranslocatetothe

plasma membrane along with its canonical de- struction complex members is unknown. Here

we show using an in vitro model system, that

endogenous dephosphorylated ß-catenin in- deed appears on the plasma membrane upon

Wnt3A treatment. This translocation occurs in- dependentofE-cadherinanddephosphorylated

ß-cateninco-localizesattheplasmamembrane

with phospho-LRP6, Axin and APC. Together,

ourresultssuggeststhatWntsignaltransduction

may be regulated at multiple levels other than,

orinadditionto,theinhibitionofbreakdownand

that routing of de novo synthesizedß-catenin

through the Wnt receptor complex is required

foroptimaltranscriptionalactivityoftheprotein.

Thisstepintheprocessingofsignaling-compe- tentß-catenin may have remained difficult to de- tectsofarduetoplasmamembranemaskingby

thejunctionalpoolofß-catenin.Wenotethatthe

punctateplasmamembranelabelingobservedin

ourmodelsystembearsastrikingresemblance

to the plasma membrane labeling described in

the former studies (Bilic et al., 2007; Schwarz- Romond et al., 2007). In contrast to what has

been described in these studies however, our

plasmamembraneenrichmentappearedtooc- curatalatertime-point.Withregardsthistempo- raldifference,westressthatincontrasttothese

studiesusingoverexpressionassaystoachieve

stoichiometric amounts of destruction complex

members,wehavefocusedontheendogenous

fraction of Wnt-responsive dephosphorylated

ß-catenin only which, as we show, is relatively

minor.

We propose that under normal physiological

conditions,ß-cateninisactivatedattheplasma

membrane upon Wnt stimulation, generating a

signalingcompetentform.Inlinewiththis,Got- tardiandGumbinerhaveshownthatWntstimu- lationgeneratesamonomericformofß-catenin

that selectively binds TCF and not E-cadherin

(Gottardi and Gumbiner, 2004). Therefore, the

activation step may constitute this transition.

Plasma membrane activation is not absolutely

requiredforsignaltransduction,asincreasingß- catenintosupraphysiologicallevelsbyinterfering

withitsdegradationwillleadtotransactivationas

well.Itwillbeimportanttotestthishypothesisby

studyingtheroutingofWnt-inducedß-cateninat

endogenouslevels.

Previous work in X. laevis has generated evi- denceforthenotionthatß-cateninstabilityalone

may not explain Wnt signaling outcome (Guger

and Gumbiner, 2000; Nelson and Gumbiner,

1999). Later studies from the Wieschaus lab in

the fly embryo using hypomorphic Armadillo al- lelesshowthatmodulationofWgsignalingcan

occurinthepresenceofuniformlyhighlevelsof

Armadillo (Tolwinski et al., 2003; Tolwinski and

Wieschaus, 2001; Tolwinski and Wieschaus,

2004).Moreover,studiesinculturedmammalian

cell lines show that receptor-mediated signal

transductioneventssuchasWntstimulationor

secreted Frizzled related protein (sFRP) inhibi- tioncanimpingeonWntsignalingoutputevenin

thepresenceofdownstreammutationsprevent- ingß-cateninbreakdown(Heetal.,2005;Suzuki

etal.,2004).Wearecurrentlyintheprocessof

studying potential post-translational modifica- tionsonß-cateninemployingtheE-cadherin-/-

Kep1cellline.

Amodelofß-cateninactivationatthereceptor

complexwouldallowß-cateninoutputtobereg- ulated on a direct stoichiometic ‘per molecule’

basis,intheoryallowingoneWntmoleculetolib- erateapredeterminedquantaofsignaling-com- petentß-cateninmolecules.Ifcorrect,thisregu- lation would be considerably more efficient than thecurrentlyproposedmodelsinwhichWntsig- nalinginputistitratedagainsttheactivityofcyto- plasmicdegradationcomplexestoregulategene

expressioninthenucleus.Regulationatmultiple

levelsissimilarlyobservedintheHedgehogsig- nal transduction pathway where stabilization of

the transcriptional co-activator does not suffice forfullactivation(MethotandBasler,1999).Itis

currently not known what mediates this activa- tionstepinHedgehogsignaltransductionatthe

plasma membrane (Hooper and Scott, 2005).

Regulationatmultiplelevelsbeyondthemerein- hibitionofproteolysiswouldallowtheWntpath- waytojoinotherdevelopmentalpathwayssuch

astheHedgehogandNotchsignalingpathways,

wherethetranscriptionalco-activatorislicensed

forsignalingattheplasmamembrane.



Acknowledgements

We thank Jos Jonkers and Patrick Derksen for

cell lines Kep1 and Kp6, Roel Nusse for purified

Wnt3A protein, Christof Niehrs for antibodies,

Hans Clevers for discussions and plasmid re- agents, JH was supported by the Netherlands

CancerFundKWF.

Materials and Methods

Data analysis

StatisticalanalysiswasdoneusingtheRsoftware

package(RDevelopmentCoreTeam,2005).

Cell culture, transfection and reporter assays SK-BR-3,Kep1(E-cadherin-/-,p53-/-),Kp6(E- cadherin+/+,p53-/-),SW480,LS174T,Colo320,

HCT15, Colo205 and SW48 were cultured in

DMEMsupplementedwith10%fetalcalfserum

andpenicillin/streptomycin(Gibco-BRL)andwere

transfectedusingFugene-6(Roche)asinstruct- edbythesupplier.1x105cellsweretransfected

with300ngTOP/FOP-TK-luc,1.5ngpRL-CMV,

325ngΔNLRP6,50or100ngß-Catenin,10ng

Wnt1and50ngH2B-mRFP.Luciferasereporter

activity was measured 24 hours after transfec- tioninSK-BR-3.24hoursaftertransfectionwith

Top/Fop-TK-luc,Kep1cellswerestimulatedwith

Wnt3Afor7hours,afterwhichluciferaseactiv- itywasmeasuredusingtheDual-LuciferaseRe- porterAssaySystem(Promega).

APC antibodies

DomainsofmouseAPC,termedAPC-A(amino

acids 788-1038), APC-B (amino acids 2170- 2394), and APC-C (amino acids 2644-2845),

were fused to GST and purified in bacteria.

Rabbitpolyclonalantiserawerepreparedasde- scribed(Hoogenraadetal.,2000).Ratmonoclo- nal antibodies against APC were generated by

Absea (China) using the same GST fusion pro- teins. Hybridomas were first tested for specific recognition of the respective GST fusion pro- teins.Positiveclones(51hybridomasforAPC-A,

44 hybridomas for APC-B, and 58 hybridomas

for APC-C) were subsequently tested on West- ern blot for recognition of eCFP-tagged APC

domains and by immunofluorescence for detec- tionofendogenousAPC.Wescreened9rabbit

polyclonalsagainsttheA,BandCdomains,but

none of the rabbit polyclonal antibodies were

monospecific (data not shown). We subsequently screened51antibody-producingrathybridomas

for APC-A (amino acids 788-1038), 44 hybrid- omasforAPC-B(aminoacids2170-2394)and58

hybridomasforAPC-C(aminoacids2644-2845).

Two rat monoclonal antibodies (13F7, APC-A- derived,and3E7,APC-B-derived)detectedboth

overexpressed,eCFP-taggedAPC(Supplemen- tary Figure S2 C), as well as endogenous APC

onWesternblots(SupplementaryFigureS2D).

These antibodies also recognized GFP-tagged

fulllengthAPCintransfectedCOS-1cells(data

notshown)andendogenousAPCindifferentcell

lines(FigureS2Danddatanotshown).As3E7

recognizedclustersoffulllengthAPCinMDCK

cells,andnotinSW480cellsthatcontaintrun- catedAPC(SupplementaryFigureS2 E,F),we

conclude that 3E7 detects endogenous APC in

immunofluorescence studies.

Other antibodies and reagents

Furtherantibodiesusedwereagainstß-Catenin

(C19220)(TransductionLabs)andH-102(Santa

Cruz), activeß-Catenin (ABC 8E7), Actin (Ab- 1, Oncogene), N-Axin (Fagotto), E-cadherin

(C20820, Transduction Labs), pan-cadherin

(C3678,Sigma),Tp1479LRP6(Niehrs),M2mAb

FLAG (Sigma). Purified recombinant Wnt3A was akindgiftfromR.Nusse(Stanford,CA)orob- tained from R&D Systems. 4’,6-diamidino-2- phenylindole (DAPI) and Actinomycin D were

obtainedfromSigma.

Plasmids

Top/Fop-TK, pRL-CMV and pRK5SK-ß-catenin

weredescribedbefore(Hendriksenetal.,2005)

andΔN-LRP6 was a kind gift from H. Clevers

(Hubrechtlaboratory,Utrecht,TheNetherlands).

Western blotting

Proteins were analyzed by SDS-polyacrylamide

gelelectrophoresis(25μgperlane)andWestern

blotting using Immobilon-P transfer membrane

(Millipore). Aspecific sites were blocked with 5%

skim milk (Oxio, Hampshire, England) at room

temperatureforonehour.Notethatdetectionof

dephospho-ß-catenin with the αABC antibody

wasinhibitedbycertainbrands/lotsofskimmilk.

Primary antibodies were incubated in 1% skim

milkfor2hoursatroomtemperatureinthefol- lowing dilutions: E-cadherin 1:1500;ß-Catenin

mAbC192201:5000,ABC1:500;Actin1:5000;

mFLAGM21:500.Blotswerewashedwithphos- phate buffered saline (PBS)/0.05% Tween 20.

Enhancedchemiluminescence(Amersham)was

usedfordetectionofproteins.

Immunofluorescence and confocal micros- copy

For immunofluorescence, cells were grown on glass coverslips coated with fibronectin (Sigma) and fixed in 3.7% formalin in PBS for 10 min and permeabilizedfor5minin0.2%Triton/PBS.For

antigenretrieval,cellswereincubatedin10mM

citric acid buffer pH 6 at 95°C for 20 min and

blocked in 5% BSA/PBS at room temperature

4

for 10 min. Primary antibodies were incubated

for 2 hours in 1% purified BSA/PBS using the following dilutions; ABC 1:200; totalß-Catenin

C19220 1:250; totalß-Catenin H102 1:65; N- Axin1:50;N-APC1:100;APC3E71:100;p-LRP6

1:250; LRP6 1:300; pan-cadherin 1:5000. Cells

were shortly washed in PBS and incubated in

fluorescently conjugated secondary antibodies (MolecularProbes)andDAPIin1%BSA/PBSfor

30min,washedshortlyinPBSandmountedin

Mowiol.ImageswererecordedusingaLeicaNT,

SP2orSP2AOBSconfocalmicroscope.

Embryo injections and immunofluorescence 4-cellstageembryoswereinjectedinoneven- tralblastomerewith25or50pgWnt8mRNAor

1000pgmyc-taggedß-cateninmRNAasprevi- ouslydescribed(Fagottoetal.,1996).Stage9

embryos were fixed in 3-4% paraformaldeyde andsectionswerepreparedandstainedaspre- viously described (Schohl and Fagotto, 2002).

Antibodiesusedweretotalanti-ß-cateninH102

diluted 1:50, ABC 1:250,  and secondary goat

Alexa546/Alexa488 anti-rabbit/anti-mouse (Mo- lecular Probes).  Images were recorded with a

Leica microscope using a narrow Cy3 filter and a 20xoilimmersionobjective.

Immunohistochemistry

Sections (4 μm) were deparaffinized and antigen retrievalwascarriedoutby10minofboilingin10

mMTris/1mMEDTA(pH9).Subsequentlyslides

were immersed in 0.3% hydrogen peroxide in

methanol for 30 min and nonspecific binding was blockedwith5%normalgoatserumfor1hrat

roomtemperature.Thesectionswereincubated

for1hratroomtemperatureinprimaryantibod- ies against totalß-Catenin (C19220 Transduc- tionLabs)andactiveß-Catenin(ABC8E7).The

UltravisionantipolyvalentHRPdetectionsystem

(LabVisionCorp.,Fremont,CA,USA)wasused

to visualize antibody binding sites with 3,3’-di- aminobenzidineasachromogen.Sectionswere

counterstainedwithhematoxylin.

References

Anderson,C.B.,Neufeld,K.L.andWhite,R.L.(2002).

SubcellulardistributionofWntpathwayproteinsinnor- malandneoplasticcolon.ProcNatlAcadSciUSA99,

8683-8.

Barker,N.,vanEs,J.H.,Kuipers,J.,Kujala,P.,vanden

Born, M., Cozijnsen, M., Haegebarth, A., Korving, J.,

Begthel, H., Peters, P. J. et al. (2007). Identification of stemcellsinsmallintestineandcolonbymarkergene

Lgr5.Nature449,1003-7.

Bilic, J., Huang, Y. L., Davidson, G., Zimmermann, T.,

Cruciat, C. M., Bienz, M. and Niehrs, C. (2007). Wnt

induces LRP6 signalosomes and promotes dishev- elled-dependent LRP6 phosphorylation. Science 316,

1619-22.

Brocardo,M.,Nathke,I.S.andHenderson,B.R.(2005).

Redefining the subcellular location and transport of APC:newinsightsusingapanelofantibodies.EMBO

Rep6,184-90.

Cadigan,K.M.andLiu,Y.I.(2006).Wntsignaling:com- plexityatthesurface.JCellSci119,395-402.

Cadigan,K.M.andNusse,R.(1997).Wntsignaling:a

commonthemeinanimaldevelopment.GenesDev11,

3286-305.

Clevers,H.(2006).Wnt/beta-cateninsignalingindevel- opmentanddisease.Cell127,469-80.

Cliffe, A., Hamada, F. and Bienz, M. (2003). A role of

DishevelledinrelocatingAxintotheplasmamembrane

duringwinglesssignaling.CurrBiol13,960-6.

Davidson, G., Wu, W., Shen, J., Bilic, J., Fenger, U.,

Stannek,P.,Glinka,A.andNiehrs,C.(2005).Caseinki- nase1gammacouplesWntreceptoractivationtocyto- plasmicsignaltransduction.Nature438,867-72.

Derksen,P.W.,Liu,X.,Saridin,F.,vanderGulden,H.,

Zevenhoven, J., Evers, B., van Beijnum, J. R., Griffioen, A.W.,Vink,J.,Krimpenfort,P.etal.(2006).Somaticin- activationofE-cadherinandp53inmiceleadstometa- staticlobularmammarycarcinomathroughinductionof

anoikis resistance and angiogenesis. Cancer Cell 10,

437-49.

Fagotto,F.,Funayama,N.,Gluck,U.andGumbiner,B.

M. (1996). Binding to cadherins antagonizes the sig- nalingactivityofbeta-cateninduringaxisformationin

Xenopus.JCellBiol132,1105-14.

Gottardi,C.J.andGumbiner,B.M.(2004).Distinctmolecu- larformsofbeta-cateninaretargetedtoadhesiveortran- scriptionalcomplexes.JCellBiol167,339-49.

Guger,K.A.andGumbiner,B.M.(2000).Amodeofregula- tionofbeta-cateninsignalingactivityinXenopusembryos

independentofitslevels.DevBiol223,441-8.

He, B., Reguart, N., You, L., Mazieres, J., Xu, Z., Lee, A.

Y., Mikami, I., McCormick, F. and Jablons, D. M. (2005).

Blockade of Wnt-1 signaling induces apoptosis in human

colorectal cancer cells containing downstream mutations.

Oncogene24,3054-8.

Hoogenraad,C.C.,Akhmanova,A.,Grosveld,F.,DeZeeuw,

C.I.andGaljart,N.(2000).FunctionalanalysisofCLIP-115

anditsbindingtomicrotubules.JCellSci113(Pt12),2285- 97.

Hooper,J.E.andScott,M.P.(2005).Communicatingwith

Hedgehogs.NatRevMolCellBiol6,306-17.

Kobayashi,M.,Honma,T.,Matsuda,Y.,Suzuki,Y.,Narisa-

wa,R.,Ajioka,Y.andAsakura,H.(2000).Nucleartransloca- tion of beta-catenin in colorectal cancer. Br J Cancer 82,

1689-93.

Korinek,V.,Barker,N.,Moerer,P.,vanDonselaar,E.,Huls,

G.,Peters,P.J.andClevers,H.(1998).Depletionofepithe- lial stem-cell compartments in the small intestine of mice

lackingTcf-4.NatGenet19,379-83.

Logan,C.Y.andNusse,R.(2004).TheWntsignalingpath- way in development and disease. Annu Rev Cell Dev Biol

20,781-810.

McCrea,P.D.,Turck,C.W.andGumbiner,B.(1991).Aho- molog of the armadillo protein in Drosophila (plakoglobin)

associatedwithE-cadherin.Science254,1359-61.

Methot, N. and Basler, K. (1999). Hedgehog controls limb

development by regulating the activities of distinct tran- scriptional activator and repressor forms of Cubitus inter- ruptus.Cell96,819-31.

Miyashiro,I.,Senda,T.,Matsumine,A.,Baeg,G.H.,Kuroda,

T.,Shimano,T.,Miura,S.,Noda,T.,Kobayashi,S.,Monden,

M.etal.(1995).SubcellularlocalizationoftheAPCprotein:

immunoelectronmicroscopicstudyoftheassociationofthe

APCproteinwithcatenin.Oncogene11,89-96.

Nelson,R.W.andGumbiner,B.M.(1999).Acell-freeassay

system for beta-catenin signaling that recapitulates direct

inductiveeventsintheearlyxenopuslaevisembryo.JCell

Biol147,367-74.

Peifer,M.,Pai,L.M.andCasey,M.(1994a).Phosphoryla- tionoftheDrosophilaadherensjunctionproteinArmadillo:

rolesforwinglesssignalandzeste-white3kinase.DevBiol

166,543-56.

Peifer, M., Sweeton, D., Casey, M. and Wieschaus, E.

(1994b). wingless signal and Zeste-white 3 kinase trigger

opposingchangesintheintracellulardistributionofArma- dillo.Development120,369-80.

Schohl,A.andFagotto,F.(2002).Beta-catenin,MAPKand

SmadsignalingduringearlyXenopusdevelopment.Devel- opment129,37-52.

Schohl,A.andFagotto,F.(2003).Aroleformaternalbeta- catenin in early mesoderm induction in Xenopus. Embo J

22,3303-13.

Schwarz-Romond, T., Metcalfe, C. and Bienz, M. (2007).

DynamicrecruitmentofaxinbyDishevelledproteinassem- blies.JCellSci120,2402-12.

Staal,F.J.,NoortMv,M.,Strous,G.J.andClevers,H.C.

(2002).WntsignalsaretransmittedthroughN-terminallyde- phosphorylatedbeta-catenin.EMBORep3,63-8.

Suzuki,H.,Watkins,D.N.,Jair,K.W.,Schuebel,K.E.,Mar- kowitz,S.D.,Chen,W.D.,Pretlow,T.P.,Yang,B.,Akiyama,

Y.,VanEngeland,M.etal.(2004).Epigeneticinactivationof

SFRPgenesallowsconstitutiveWNTsignalingincolorectal

cancer.NatGenet36,417-22.

Tolwinski,N.S.,Wehrli,M.,Rives,A.,Erdeniz,N.,DiNardo,

S.andWieschaus,E.(2003).Wg/Wntsignalcanbetrans- mitted through arrow/LRP5,6 and Axin independently of

Zw3/Gsk3betaactivity.DevCell4,407-18.

Tolwinski,N.S.andWieschaus,E.(2001).Armadillonuclear

importisregulatedbycytoplasmicanchorAxinandnuclear

anchordTCF/Pan.Development128,2107-17.

Tolwinski,N.S.andWieschaus,E.(2004).Anuclearfunction

forarmadillo/beta-catenin.PLoSBiol2,E95.

vandeWetering,M.,Barker,N.,Harkes,I.C.,vanderHey- den,M.,Dijk,N.J.,Hollestelle,A.,Klijn,J.G.,Clevers,H.

and Schutte, M. (2001). Mutant E-cadherin breast cancer

cellsdonotdisplayconstitutiveWntsignaling.CancerRes

61,278-84.

vanNoort,M.,Meeldijk,J.,vanderZee,R.,Destree,O.and

Clevers,H.(2002).Wntsignalingcontrolsthephosphoryla- tionstatusofbeta-catenin.JBiolChem277,17901-5.

vanNoort,M.,Weerkamp,F.,Clevers,H.C.andStaal,F.J.

(2007). Wnt signaling and phosphorylation status of beta- catenin: importance of the correct antibody tools. Blood

110,2778-9.

Willert, J., Epping, M., Pollack, J. R., Brown, P. O. and

Nusse,R.(2002).AtranscriptionalresponsetoWntprotein

inhumanembryoniccarcinomacells.BMCDevBiol2,8.

Zeng,X.,Tamai,K.,Doble,B.,Li,S.,Huang,H.,Habas,R.,

Okamura,H.,Woodgett,J.andHe,X.(2005).Adual-kinase

mechanismforWntco-receptorphosphorylationandacti- vation.Nature438,873-7.

4

A B

Vehicle Vehicle

+Act D Wnt3A Wnt3A +Act D 0

10 20

n=57 n=35

n=1 n=4

30 p=2e-10

p=0.1

p=0.3

Wnt3a Vehicle

Cdh+/+

Wnt3a Cdh-/- Cdh-/-

pan-cadherin merge

Supplementary Figure 2

Figure S1A.Membranelocalizationofdephospho-ß-catenindoesnotcoincidewithmembranelocalizationofcad- herins. Cells were stimulated as in A and stained for dephospho-ß-catenin and cadherins using a pan-cadherin

antiserum.Asareference,E-cad+/+cellsareused.B.Plasmamembranerecruitmentofdephospho-ß-cateninis

independentofongoingtranscription.E-cadherin-/-cellswereanalyzedasinAinthepresenceorabsenceofthe

transcription inhibitor Actinomycin D at 4 μg/ml. No significant difference in plasma membrane localization of dephos- pho-ß-cateninwasobserved.PvaluesaccordingtoFisher’sexacttests.

Figure S2 Generation of anti-APC antisera.A.DomainstructureofAPC.Coiledcoilandbasicregionsareshown

asdarkgreyboxesattheN-andC-terminalendsofAPC,respectively.Armadillorepeats(lightblue),andß-catenin

(yellowandgreen)andaxin(red)bindingregionsarealsodepicted.Thedomains(aminoacidsindicated)usedtomake

GSTfusionproteinsareshownabovethesequence.B-D.Westernblotanalysis.In(B)ablotisshownofoverex- pressedeCFP,andeCFP-taggedAPC-A,-B,and-CinCOS-1cells.Proteinsweredetectedwithanti-GFPantibodies.

In(C)ablotisshownofeCFP-APC-A(leftpanel)oreCFP-APC-B(rightpanel),overexpressedinCOS-1cells,and

detectedwith5ratmonoclonalsagainstAPC-Aorwith5monoclonalsagainstAPC-B.Notethatonlysomeantibodies

recognizetheeCFP-taggedproteins.Stripsincubatedwith13F7and3E7areindicated.In(D)blotsareshownwith

indicatedcelllysates.Ofthe4celllinesusedonlyHCT116expressesfulllengthAPC(indicatedaswt).Theamino

acidsatwhichAPCistruncatedintheothercelllinesareindicated.Notethat3E7onlyrecognizesfulllengthAPC

(and degradation products), while 13F7 recognizes both full length and truncated forms of APC (E, F) on immunofluo- rescence. In (E) SW480 and MDCK cells were fixed and stained with the indicated antibodies (IN: polyclonal rabbit antibodydonatedbyInkeNathke).APCisfoundclusteredinperipheraldomains(indicatedbyarows).Notethat3E7

failstostainsuchclustersinSW480cells.In(F)adoublelabellingof3E7withanti-tubulinantibodieswasperformed.

APCisconcentratedattheendsofasubsetofmicrotubules(arrows).Thestainingpatternsobtainedwith3E7and

13F7arecomparabletothatobtainedwiththeINantibodyandresemblepublishedpatterns.

4