The Abl pathway bifurcates to balance Enabled and Rac signaling in axon patterning in Drosophila

49 Pages·2017·2.6 MB·English

by  Site License

©2017.PublishedbyTheCompanyofBiologistsLtd|Development(2017)144,487-498doi:10.1242/dev.143776
RESEARCHARTICLE
The Abl pathway bifurcates to balance Enabled and Rac signaling
in axon patterning in Drosophila
RamakrishnanKannan1,*,Jeong-KuenSong1,‡,TatianaKarpova2,AkanniClarke1,MadhuriShivalkar1,
BenjaminWang1,LyudmilaKotlyanskaya1,IrinaKuzina1,QunGu1andEdwardGiniger1,§
ABSTRACT Garbe et al., 2007; Gupton et al., 2012; Hsouna et al., 2003; Yu
et al., 2001). The mechanism of Abl signaling, therefore, is
The Abl tyrosine kinase signaling network controls cell migration,
fundamental for understanding growth cone guidance by each of
epithelial organization, axon patterning and other aspects of
these receptors,as wellastheintegration ofsignalsfrommultiple
development. Although individual components are known, the
receptors (Crowner et al., 2003; Kuzina et al., 2011; Wills et al.,
relationships among them remain unresolved. We now use FRET
1999a;Winbergetal.,1998).
measurementsofpathwayactivity,analysisofproteinlocalizationand
The Abl tyrosine kinase pathway was among the first of the
geneticepistasistodissectthestructureofthisnetworkinDrosophila.
conservedsignaltransductionpathwaystobeidentified,andmany
We find that the adaptor protein Disabled stimulates Abl kinase
of its components have long been known (Dai and Pendergast,
activity.Ablsuppressestheactin-regulatoryfactorEnabled,andwe
1995;Gertleretal.,1995,1996;Henkemeyeretal.,1987;Howell
findthatAblalsoactsthroughtheGEFTriotostimulatethesignaling
et al., 1997; Liebl et al., 2000; Luo, 2000; Wang and Baltimore,
activityofRacGTPase:Ablgatestheactivityofthespectrinrepeatsof
1983). A core set of pathway components cooperate with Abl in
Trio,allowingthemtorelieveintramolecularrepressionofTrioGEF
neurons,culturedfibroblastsandtransformedcelllines(Bradleyand
activity by the Trio N-terminal domain. Finally, we show that a key
Koleske, 2009; Hoffmann, 1991). Disabled (Dab) is an adaptor
targetofAblsignalinginaxonsistheWAVEcomplexthatpromotes
proteinthatassociatesdirectlywithreceptorsthatregulatemotility
the formation of branched actin networks. Thus, we show that Abl
andguidance(Howelletal.,1999a,b;LeGalletal.,2008).Trioisa
constitutesabifurcatingnetwork,suppressingEnaactivityinparallel
guanine nucleotide exchange factor (GEF) that activates both Rac
withstimulationofWAVE.Wesuggestthatthebalancingoflinearand
and Rho GTPase and is essential for neuronal morphogenesis
branchedactinnetworksbyAblislikelytobecentraltoitsregulation
(Awasakietal.,2000;Newsomeetal.,2000;SchmidtandDebant,
ofaxonpatterning.
2014; Steven et al., 1998). Abi is a component of the WAVE
KEYWORDS:Abl,Disabled,Rac,Trio,FRET,Axonguidance complex that activates Arp2/3 to extend branched actin networks
(Dai and Pendergast, 1995; Gautreau et al., 2004). Enabled (Ena)
INTRODUCTION extends and bundles linear actin filaments (Gertler et al., 1995,
During development of the nervous system, cell surface receptors 1996). Dab, Trio and Abi work synergistically with Abl in many
regulate cytoskeletal dynamics to direct axon growth (Dickson, contexts (Gertleret al., 1993; Luo, 2000), whereas Enaactivity is
2002;SongandPoo,2001,1999;Tessier-LavigneandGoodman, suppressed by Abl (Gates et al., 2007; Gertler et al., 1995;
1996). The transformation of extracellular cues into cytoskeletal Grevengoedetal.,2003).
dynamics is executed by cytoplasmic signaling networks. Among DespitethelonghistoryofstudiesofAbl,however,thefunctional
these,theAbltyrosinekinasesignalingpathwayplaysaleadingrole organization of the pathway has remained obscure. Nearly every
in regulation of the neuronal cytoskeleton (Bradley and Koleske, componenthasbeenlinked,genetically,biochemicallyorboth,to
2009;Krauseetal.,2003;LanierandGertler,2000;Morescoand nearly every other component (Forsthoefel et al., 2005; Gertler
Koleske, 2003). Components of the Abl pathway link to growth et al., 1993; Hill et al., 1995; Liebl et al., 2000; Sonoshita et al.,
cone guidance receptors and actin-regulatory proteins (Le Gall 2015).Manyexperimentshaveyieldedcontradictorypicturesofthe
etal.,2008;Liebletal.,2003;Rheeetal.,2002;Willsetal.,1999a, relationshipsamongAblsignalingcomponentsandoftheirrolesin
b). Indeed, nearly all of the common, phylogenetically conserved motility(Bashawetal.,2000;Bearetal.,2000;Forsthoefeletal.,
families of axon guidance receptors signal through Abl. This 2005; Hsouna et al., 2003; Juang and Hoffmann, 1999; Krause
includes receptorsforNetrin,Slit,Semaphorin,Ephrinandothers etal.,2002;Liebletal.,2003;Linetal.,2009;Songetal.,2010;
(Bashawetal.,2000;Deinhardtetal.,2011;Forsthoefeletal.,2005; Trichetetal.,2008;Willsetal.,1999a,b).Thus,asyetthereisno
clear and testable model of which components lie upstream or
downstream of which othercomponents, an essential requisite for
1NationalInstituteofNeurologicalDisordersandStroke,NationalInstitutesof us to interpret the extensive data linking Abl to growth cone
Health,Bethesda,MD20892,USA.2NationalCancerInstitute,NationalInstitutesof motilityandneuronalmigrationincultureandinvivo(Bradleyand T
Health,Bethesda,MD20892,USA. N
*Presentaddress:NeurobiologyResearchCenter(NRC),Departmentof Koleske,2009). E
Psychiatry,NationalInstituteofMentalHealthandNeurosciences,Bangalore40, We therefore set out to establish the functional relationships M
India.‡Presentaddress:L&JBiosciences,19FristfieldRd,Gaithersberg,
MD20878,USA. amongthecoreproteinsoftheAblpathwayinvivoinneurons,using P
direct measures of protein activity. We deployed fluorescence O
§Authorforcorrespondence([email protected]) resonanceenergytransfer(FRET)biosensorprobesthatallowedus L
E
E.G.,0000-0002-8340-6158 toassaydirectly,inlivingcells,theactivityofthetwokeyoutputsof V
theAblpathwayinaxons:RacGTPasesignaling(Itohetal.,2002; E
D
Received18August2016;Accepted15December2016 Wangetal.,2010)andAblkinaseactivity(Sterneetal.,2015;Ting
487
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
etal.,2001). Wefurther validatedtheresulting modelfortheAbl RESULTS
pathwaybothbyinvestigatinghoweachcomponentofthepathway DabregulatesAblkinaseactivity
regulatesthe subcellular localization of othercomponents, and by PreviousexperimentsshowedthatDabisacorecomponentofthe
usinggeneticepistasistoinvestigatethedependentrelationshipsof Abl pathway and suggested that Dab is functionally upstream of
mutationsingenesencodingtheseproteins. Abl:UAS-AblsuppressesthephenotypeofaDabmutantbutUAS-
We now reveal the functional relationships among the core Dab cannot compensate for lack of Abl (Song et al., 2010).
components of the Abl signaling network in Drosophila neurons. Moreover,DabcontrolsthesubcellularlocalizationofAbl,butnot
WeshowthattheadaptorproteinDabstimulatesthekinaseactivity viceversa(Songetal.,2010).Wewondered,however,whetherDab
ofAblaswellasregulatingitssubcellularlocalization.Consistent also controls Abl kinase activity. We therefore expressed a FRET
with this, Dab is found in a biochemical complex with Abl in probethatmeasuresAblkinaseactivityinvivo.
Drosophila embryos. The kinase activity of Abl stimulates the We expressed in Drosophila a unimolecular CFP-YFP FRET
signaling activityof Rac GTPase, primarily byacting through the biosensorforAblkinaseactivitythatwasoriginallyconstructedfor
GEFTrio.AbldoesnotactdirectlyontheRacGEFdomainofTrio; useinmammaliancells,employingsequencesfromthemammalian
rather,AblgatestheactionofthespectrinrepeatsofTrio,allowing CrkIIprotein(Sterneetal.,2015;Tingetal.,2001).Multiplelines
themtocounteractarepressiveintramolecularinteractionbetween of evidence demonstrate that FRET activity of this bioprobe
the Trio N-terminal domain (NTD) and GEF1 domain, thereby faithfully reports Abl kinase activity in Drosophila. (1) By
allowingTriotostimulateRac.InparallelwiththeinhibitionbyAbl ratiometric imaging, the reporter gives a FRET signal in cultured
of the actin polymerization factor Ena, the Abl-Trio-Rac pathway photoreceptor(PR)neuronsthatisdecreasedincellsthataremutant
activates the Abi/WAVE complex that promotes the extension of forAblandincreaseduponoverexpressionofwild-type(WT)Ablor
branched actin networks. We suggest that this combination of expressionofBcr-Abltransgenes(Fig.1A,B).(2)TheFRETsignal
effectsexecutestheAbl-dependentregulationofaxongrowthand isinhibitedacutelybytreatmentofPRswiththeAbl-specificdrug
guidance. STI-571(Gleevec),andthisinhibitionisreversibleuponwashoutof
T
N
Fig.1.ValidationoftheAblFRETbiosensor.ACFP-YFPunimolecularFRETbiosensorforAblkinaseactivitywasexpressedinDrosophilaphotoreceptor E
neuronsunderthecontrolofGMR-GAL4,andFRETwasimagedandquantifiedinculturedprimaryphotoreceptors.(A)Singleopticalsectionofoneortwo M
photoreceptorsshowingbiosensordistribution(YFP,top)andFRET(bottom)incellsfromlarvaleyediscsoftheindicatedgenotypes.Quantifiedcellisoutlinedin
P
red,andthepseudocolorscaleforFRETsignalisshownwithlimitingvaluesindicated,inthisandallotherFRETimages.NotethattheWTphotoreceptorinAis O
the30min(pre-drug)imagefromC.(B)MeanFRETratio(MFR)foreachgenotype.(C)SinglephotoreceptorshowingAblFRETprobedistributionand L
timecourseofFRETchannelupontreatmentfortheindicatedtimewiththeAblkinaseinhibitorSTI-571,anddrugwashout.(D)TimecourseofAblinhibitionby E
STI-571andwashout(n=32).MFRhasbeennormalizedto1attime0.(E)QuantificationofAblreporterFRETbyYFPphotobleachingfortheindicated V
genotypes.Errorbarsindicates.e.m.andthenumberofcellsimaged(n)isshown.ns,notstatisticallysignificant;***P<0.001;inallcasestwo-tailedt-test,with E
Bonferronicorrectionformultipletesting.Scalebar:10μminA,C. D
488
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
thedrug(Fig.1C,D).TheresidualFRETsignalinthepresenceof co-precipitated Dab (Fig. 2C). Co-precipitation was also observed
thedrugisslightlylessthanthatinAblmutantPRs,whichislikely fromextractsofculturedDrosophilaS2cells(notshown).Acontrol
tobebecauseofperduranceofmaternalAbl.Wecannotruleoutthe documenting immunoprecipitation of Abl byanti-Abl antibody is
possibilitythattheeffectofSTI-571isincompletesincetheFRET providedinFig.S1.ThesedatademonstratethatAblandDabare
signalintreatedAblmutantcellsappearstobeslightlylessthanthat associatedincommoncomplexesinvivoinDrosophila.
intreatedWTcells,butthedifferenceisnotstatisticallysignificant.
(3)PhotobleachingoftheYFPacceptorrevealsaFRETsignalthat AblkinaseregulatesthesignalingactivityofRacGTPase
is consistent with the FRET efficiency calculated by ratiometric Inneurons,AblcooperateswithRhofamilyGTPases,particularly
imaging (Fig. 1E). (4) A mutant of the reporter that lacks its Rac. However, it is not clear whether Rac and Abl are both
phosphorylatabletyrosine(YFmutant)doesnotgiveaFRETsignal downstreamofaregulatorthatactsonbothorwhetheroneofthese
(Fig.1E). proteins controlstheother. Wedeployed asecondFRETprobeto
WenextusedtheAblFRETprobetotestwhetherDabmodulates assay Rac activity in living cells. Raichu-Rac was developed as a
Ablkinaseactivity(Fig.2A,B).WeassayedPRsfromWTorDab FRETprobeofRacsignalinginmammaliancells,anditwasshown
mutant fliesthat expressthe Abl FRET probe and found that Abl also to function in the Drosophila ovary (Itoh et al., 2002; Wang
kinase activity was suppressed in Dab mutant neurons [mean etal.,2010),sowevalidatedtheprobeinDrosophilaPRsandS2
FRET ratio (MFR)=1.55±0.1 in WT versus 0.93±0.2 in Dabmz; cells. In photoreceptors, Rac biosensor activity decreased
mean±s.e.m.inthisandallsubsequentFRETvalues;P<0.01].We substantially in the background of mutations that inactivate the
also overexpressed Dab in photoreceptors bearing the probe and GEFTrio(MFR=0.8±0.12intriom89/trio1;0.9±0.1intriom89/trio8)
found that Abl kinase activity was stimulated (MFR=2.1±0.1 in or upon co-expression of a dominant-negative form of Rac
UAS-Dab;P<0.01).ControlwesternanalysisverifiedthattheDab (0.79±0.1 in UAS-Rac1N17) compared with WT controls
genotype does not alter the level of Abl protein (not shown). (1.68±0.08; P<0.001 in each comparison; Fig. 3A,B). We further
Together,thesedatashowthatDabenhancesAblkinaseactivity. validated these results in a gain-of-function background by co-
IthasbeensuggestedthattheAblandDaborthologsassociatein expressingthebiosensorwitheithertheGEF1domainortheGEF2
mammaliancells(Sonoshitaetal.,2015).WethereforetestedAbl/ domainofTrioinWTPRs.TheGEF1domainofTrioactivatesRac
Dabassociation inDrosophila byco-immunoprecipitation (co-IP) signaling, whereas the GEF2 domain does not (Newsome et al.,
from embryo extracts and found that immunoprecipitation of Abl 2000). Consistent with this, co-expression of GEF1 gave a
Fig.2.DabstimulatesAblkinaseactivity.
(A)Culturedphotoreceptorsoftheindicated
genotypesexpressingtheAblFRETbioprobe.The
leftpanelofeachpairshowstheprobedistribution
(YFP)andtherightpanelshowsFRET.Areaused
forquantificationisoutlinedandapseudocolored
activitykeyisshown.Scalebar:10μm.(B)Mean
AblFRETratioforeachgenotype.Errorbars
indicates.e.m.**P<0.01.(C)Anti-Dabwesternof
WTembryolysates.Dabrunsasafamilyoftwoor
threeproductsof∼250-280kDa(arrows).
Molecularweightmarkersareindicated.Similar
resultswereobtainedinsixindependent
experiments,andinextractsofS2cells(not
shown).
T
N
E
M
P
O
L
E
V
E
D
489
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
Fig.3.ValidationofRacFRET
probeandAbl/Racepistasis.
Abl-responsiveorRac-responsive
CFP-YFPunimolecularFRETprobes
wereexpressedinDrosophilaunder
thecontrolofGMR-GAL4and
ratiometricFRETwasimaged.
(A)Singleculturedphotoreceptorsof
theindicatedgenotypesexpressing
theRaichu-RacFRETprobe.Top
panelofeachpairshowsprobe
distribution(YFP)andbottompanel
showspseudocoloredFRETchannel
withcolorscale.Scalebar:10μm.
(B)MeanRacFRETratioforeach
geneticbackground.(C)MeanAbl
FRETratioingenotypesthataltertrio
orRac1activity.(D)MeanRacFRET
ratioingenotypesthatalterAbl
activity,orinWTwithandwithout
30mintreatmentwithSTI-571.
(E)Singleculturedphotoreceptors
expressingtheRaichu-RacFRET
bioprobe.(F)TimecourseofRaichu-
RacFRETsignalinphotoreceptors
expressingUAS-Ablupontreatment
withtheAblinhibitorSTI-571and
washout.Signalsarenormalizedto
FRETratioatstartofexperiment.
Errorbarsindicates.e.m.*P<0.05,
**P<0.01,***P<0.001.
substantial increase in the FRET signal of the Rac bioprobe with basal expression in controls (1.7±0.11). Acute inhibition of
(MFR=3.04±0.12)comparedwiththecontrol(1.68±0.08),whereas AblkinasebySTI-571alsosuppressedRacactivitywithinminutes,
co-expression ofGEF2 hadnoeffect (1.64±0.11;Fig.3B).These both in cultured S2 cells (Rac MFR=1.1±0.08 after 30min of
results verify the reliability of the Rac biosensor for reporting treatment with STI-571 versus 1.6±0.1 before treatment; P<0.05;
endogenousRacactivityinDrosophilaPRs. Fig. 3D) and in PRs (Rac MFR=1.1±0.16 after treatment versus
We next used the Abl and Rac FRET biosensors to perform 1.6±0.08 before treatment; P<0.05; Fig. 4A, Fig. 3F). Note that
parallel epistasis experiments, which revealed that Abl kinase absolute FRET ratio values measured in S2 cells should not be
regulatesRacsignalingactivitybutthatRacdoesnotregulateAbl compared directly with those in PRs. These data demonstrate that
kinase.WefirstassayedAblkinaseactivitywhilewemanipulated AblkinaseactivityisrequiredtostimulateRac.Moreover,theacute
Rac activity in PRs (Fig. 3C). In WT PRs, Abl FRET activity is suppressionofRacbyanAblinhibitorarguesagainstthehypothesis T
N
1.6±0.2,andthisremainedunalteredinthebackgroundofdifferent that suppression of Rac in an Abl mutant arises from indirect
E
trio mutant alleles (1.7±0.14 in triom89/trio8; 1.6±0.11 in triom89/ compensation for a chronic genetic manipulation, and rather M
trio1) or upon co-expression of dominant-negative UAS-Rac1N17 suggeststhat the effect is relatively direct. For reasonsthat we do P
(1.64±0.1)orUAS-trio(1.6±0.2).Bycontrast,alteringendogenous not understand, in a small fraction of experiments suppression of O
AblactivityexertedastronginfluenceonendogenousRac,bothin Rac signal by Abl inhibition was not reversed by washout of the L
E
PRs and S2 cells (Fig. 3D,E). First, Rac biosensor activity drug (Fig. 3F, but compare with Fig. 4B). This was observed V
diminishedinAblmutantphotoreceptors(1.17±0.11;P<0.01)and occasionallyinboththeWTbackgroundandinthecontextofAbl E
D
was hyperactivated by UAS-Abl (2.74±0.2; P<0.001) compared overexpression.
490
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
Fig.4.AblregulationofRacis
mediatedthroughTrio.(A)MeanRac
FRETratiofortheindicatedgenotypes,
withorwithout30mintreatmentwiththe
AblinhibitorSTI-571.*P<0.05.
(B)TimecourseofRaichu-RacFRET
signaloftheindicatedgenotypesupon
treatmentwithSTI-571andwashout.
FRETsignalsforthetimecoursesarenot
normalizedtothestartingvaluesinorder
toallowcomparisonoftriomutantto
control.(C)Nomarskiimagesshowing
threehemisegmentsofstage17
embryosoftheindicatedgenotypes,
immunostainedwithanti-Fas2andfilet-
mountedtorevealtheISNbmotonerve.
Blackarrowsindicatemuscle12
neuromuscularsynapses;redarrows
indicatepositionsofmissingsynapses
(‘stall’phenotype).Scalebar:10µm.
(D)QuantificationofISNbstall
phenotypes(failuretoinnervatemuscle
12).***P<0.001;χ2,withBonferroni
correction.Errorbarsindicates.e.m.
AblactsthroughTriotostimulateRacsignalinginneurons WeverifiedbygeneticepistasisthatAblactslinearlyupstreamof
There are a number of Rac GEFs in Drosophila, but one in trio. Abl and trio mutants each cause stalling of the ISNb motor
particular, Trio, behaves like a core component of the Abl nerveatthejunctionofmuscles6and13,withfailuretoinnervate
signaling pathway. trio mutants produce axonal phenotypes that muscle12,asreportedpreviously(Batemanetal.,2000;Songand
mimic Abl mutations, and trio interacts genetically with Abl: Giniger, 2011). We found, however, that overexpression of trio
heterozygous mutations of trio enhance the phenotype of Abland suppressedtheISNbphenotypeofAblmutants,butoverexpression
vice versa, and ena heterozygotes suppress the trio phenotype ofAbldidnotsuppresstrio(Fig.4C,D).Thus,intrio123.4/8therewas
(Awasaki et al., 2000; Bateman et al., 2000; Liebl et al., 2000; afailuretoinnervatemuscle12in65%ofhemisegments(n=290),
Newsome et al., 2000). Our FRET experiments now show that andthisfrequencywasnearlythesameuponoverexpressionofAbl
Abl-dependentregulationofRacinphotoreceptorsismediatedby (78%stall,n=220).Bycontrast,whereasAbl4/4mutantsshowedthe
Trio. Rac FRET activity is inhibited significantly either by ISNb stall phenotype in 80% of hemisegments (n=186),
inhibition of Abl or mutation of trio, as shown above. overexpression of trio reduced that to just 13% (n=288; P<0.001,
Combining both, however, does not further depress Rac activity χ2), which is not significantly different from the percentage
(MFR=0.89±0.1 in triom89/trio8 versus 0.86±0.2 with STI-571; innervationofcontrolembryosatthisstage(trio123.4/+,muscle12
and MFR=0.82±0.11 in triom89/trio1 versus 0.85±0.2 with STI- notyetinnervatedin16%ofhemisegments;n=194;P=0.37).These T
N
571; Fig. 4A). This is clearest in a timecourse, where addition of dataareconsistentwithasimple,lineardependencepathwaywith
E
STI-571 had no effect on the Rac FRET ratio in a trio mutant Ablupstreamoftrio,asimpliedbytheFRETresults. M
geneticbackground(Fig.4B).ThisdemonstratesthatAblandTrio P
actserially,inacommonpathway,andnotinparallel.Thefailure AblcontrolsderepressionoftheTrioGEF1domainviathe O
to observe an additive effect of trio and STI-571 is not due to a Triospectrinrepeats L
E
limitation of the dynamic range of the assay, since other We wondered whether Abl stimulates Trio GEF activity directly V
experimental paradigms do produce significantly lower Rac through an effect on the Rac-specific GEF1 domain or by some E
D
FRETratios (Fig. 5D, see below). other mechanism. Expression of Trio GEF1 by itself in WT
491
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
Fig.5.Structure/functionanalysisofTrioanditsregulationbyAbl.(A)MeanRacFRETratioinculturedphotoreceptorsfortheindicatedgenotypes.
(B)SchematicofDrosophilaTrioproteindomainstructureandthedeletionstestedhere.Deletedcodonsareindicated.SR,spectrinrepeats.(C)Anti-Triowestern
ofDrosophilaS2cellsexpressingtheindicatedTrioderivatives.EndogenousWTTrioisseeninalllanes(arrow).Deletedderivativesarevisibleinlanes4-7
(bracket);derivativesbearingpointmutations(lanes2and3)cannotbedistinguishedfromtheendogenousprotein.(D)MeanRacFRETratioassayedafterco-
transfectionofRaichu-RacreporterintoS2cellswiththeindicatedtransgenes.GraybarsshowFRETactivityafter30mintreatmentwithSTI-571;purplebars
showFRETactivityafter30minmocktreatment.WTreferstoresultsoftransfectionwithFRETreportertransgenepluscarrierDNA,butnotriotransgene.Note
thatthedataforcontrolwithandwithoutSTI-571(toptwobars)werealsopresentedinFig.3D.Errorbarsindicates.e.m.*P<0.05,**P<0.01,***P<0.001.
photoreceptors increased the Rac FRET ratio from 1.7±0.08 to thatNTD-dependentrepression(Fig.5D).Asdescribedabove,the
2.9±0.19 (Fig. 5A). Expression of Trio GEF1 in Abl mutant PRs, RacbioprobegaveabaselineFRETratioof1.6±0.1inS2cells,and
however,decreasedthisratioonlyslightly(to2.7±0.23),consistent addition of STI-571 suppressed this to 1.1±0.08 (Fig. 3D).
withAblsuppressingendogenousTriobutnottheexpressedTrio;if Expression of trio mGEF2 in cells bearing the Rac FRET
the expressed GEF1 also required Abl, then the reduction of the biosensorenhancedtheRacFRETratioto2.8±0.2intheabsence
FRET ratio would have been far greater (Fig. 5D). We therefore of drug, and addition of STI-571 suppressed that FRET signal to
inferred that Abl does not regulate Trio via a direct effect on the 1.5±0.11(P<0.001forthecomparisonwithandwithoutdrug).This
GEF1domain. establishesthedynamicrangeoftheassay.
To map the region of Trio that mediates the effect of Abl we Next,wetestedthefunctionsofvariousTriodomainsbydeletion
analyzed Trio derivatives by co-expressing them with the Rac (Fig.5B,D,TableS1).First,expressionoftrioΔSH3(inthemGEF2
bioprobeinS2cells.Toprovideanidealcontrol,eachtriomutation backbone) did not enhance the Rac FRET ratio (MFR=1.7±0.07;
wasintroducedinparallelintotwobackbones,onebearingapoint P>0.4 versus control with no trio transgene). This shows that the
mutation that inactivates the Rac-specific GEF1 domain (trio SH3domainisessentialforTrioactivity.Second,expressionoftrio
mGEF1)andtheotherwiththesameaminoacidchangebutinthe mGEF2ΔNTD (MFR=2.4±0.19) increased Rac activity nearly as
Rho-specificGEF2domain(triomGEF2).triomGEF1providesa much as trio mGEF2 itself (MFR=2.8±0.2; P>0.1), showing that
negative control backbone that should be unable to activate Rac the NTD domain is not needed for activation of Rac. Strikingly,
FRETactivityregardlessofothermutations;triomGEF2providesa however,additionofSTI-571 did notaffect theRacsignal oftrio
backbonethatisfullyactiveforRacFRETactivityandistherefore mGEF2ΔNTD(MFR=2.3±0.23;P>0.8forthecomparisonwithand
appropriate for assaying the effects of modifications to other withoutdrug),showingthatthisderivativenolongerrequiresAbl
domains, but bears the identical amino acid change in the GEF2 kinase for activity. A plausible explanation for this observation T
N
domain in order to control for nonspecific effects of the GEF came from analyzing the effects of deleting the spectrin repeats.
E
mutation through unrelated mechanisms (Shivalkar and Giniger, Expression of trio mGEF2Δspec, which deletes only the spectrin M
2012; Song and Giniger, 2011); it also rules out any background repeats,abolished GEFactivity(MFR=1.6±0.06;P>0.8versus no P
contributionsofGEF2totheRacFRETsignal. transgenecontrol).Bycontrast,expressingaderivativethatlacked O
RacFRETanalysisrevealed(1)thattheSH3domainisrequired both the spectrin repeats and the NTD, termed trio mGEF2Δ L
E
forallactivityoffull-lengthTrio,(2)thattheNTDofTriorepresses (NTD+spec), yielded as much activity (MFR=2.4±0.17) as trio V
theRacGEFactivityofTriointheabsenceofAblkinase,and(3) mGEF2Δ(NTD)itself(P>0.8).Thesimplesthypothesis,therefore, E
D
thatinthepresenceofAblkinase,theTriospectrinrepeatsrelieve isthatinthenativeproteintheNTDsuppressesGEFactivityinthe
492
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
absenceofAblkinase,andinthepresenceofAblthissuppressionis incontrol(n=177;P<0.001,χ2;Fig.7B,C,F).Thisissupportedby
counteracted by a mechanism that employs sequences within the analysisofembryosdoublyheterozygousforbothascarmutation
spectrin repeat region. Consistent with this hypothesis, trio and a mutation in Abl-interacting protein (Abi), which encodes
mGEF2Δ(NTD+spec) was also largely insensitive to Abl another core component of the WAVE complex; the doubly
(MFR=1.8±0.12 in the presence of STI-571, reflecting inhibition heterozygous embryos displayed the same axon stalling
of endogenous Trio but little effect on the expressed protein). phenotype in 79% of hemisegments (n=214) versus 16% stalling
Potentialmolecularmechanismsfortheseeffectsareconsideredin for scarΔ37/+ (n=210) and 31% stalling for AbiKO/+ (n=256;
the Discussion. In these experiments, all trio derivatives were P<0.001 compared with either heterozygous mutation alone, χ2;
expressedatsimilarlevels(Fig.5C). Fig. 7A,D,F). The high stall frequency in Abi heterozygotes was
unexpected, but it was observed with two independent Abi alleles
Trio/RacactsinparalleltoEna,downstreamofAbl and in multiple genetic backgrounds. Apparently, Abi is partly
Ena is a direct regulator of actin organization that functions haploinsufficient for ISNb extension. In ISNb extension, SCAR
downstream of Abl (Gates et al., 2007; Gertler et al., 1995; evidentlyinteractsintimatelywithTrioandRac,asembryosthatare
Grevengoed et al., 2003; Kannan et al., 2014), but it was not doubly heterozygous for mutations in trio and scar also show a
known whether Trio acts between Abl and Ena in a single linear stronglysynergisticISNbstallingphenotype[67%stall(n=86)for
pathwayorinparalleltoEna.Wehaveshownpreviouslythatlossof trio123.4/+; Df(scar)/+ and 51% stall (n=187) for trio123.4/+;
AblorDabcausesEnaproteintocoalesceinlargeaggregatesatthe scarΔ37/+ versus 16% stall (n=254) for trio123.4/+; in each case
mostbasalpartofthesomaofPRs(Kannanetal.,2014).Ifthisis P<0.001, χ2] (Fig. 7E,F). Together, these data show that the
mediatedviaTriothenatriomutantshouldshowthesamephenotype. functional interaction of Trio/Rac with the WAVE complex is
WethereforeexaminedEnalocalizationintriomutanteyediscsand essentialforacharacteristicAbl-dependentaxonguidancedecision.
foundthatEnaproteinremainsdistributedthroughoutthecytoplasm
ofthecellsoma,anddoesnotsettleattheaxonexitsiteofthePRsoma DISCUSSION
(Fig.6).ThisexcludesthehypothesisthatTrioisupstreamofEnaina WeshowherethatAbltyrosinekinaseanditsinteractingco-factors
simple, linear dependence pathway, and instead argues that the formabifurcatingproteinnetworkthatcontrolsaxonpatterningin
signalingnetworkbifurcatesdownstreamofAbl,withsuppressionof Drosophila (Fig. 7G). Dab, which associates with guidance and
EnaandactivationofTrio/Racformingparallelbranches. motilityreceptors,isanupstreamregulatorofAbllocalizationand
activity. Abl,in turn, regulatesthelocalization and suppressesthe
RacinteractswithWAVEtocontrolsISNbaxonpatterning activityoftheactinregulatorEna.AblalsostimulatesRacGTPase
Rachasmanybiochemicaltargets,butoneofitskeyeffectorsisthe signalingthroughregulationoftheRacGEFTrio.TrioGEFactivity
WAVEcomplex.IntheabsenceofRacactivity,theVCAmotifof is repressed by its own NTD in the absence of Abl, but in the
WAVEprotein (SCAR in Drosophila) issequestered (Chenet al., presence of Abl the spectrin repeats of Trio relieve that NTD-
2010;Gautreauetal.,2004).BindingofSCARtoGTP-Racreleases dependent repression. Finally, a key target of Abl-dependent Rac
that interaction, allowing recruitment of Arp2/3 and promotion of activity in axon patterning is a second actin-regulatory complex,
branched actin networks (Blanchoin et al., 2000; Machesky and WAVE: Trio/Rac interacts genetically with WAVE components,
Insall, 1998; Marchand et al., 2001). We therefore performed includingWAVE/SCARitselfandtheAbl-interactingproteinAbi,
genetic interaction tests and discovered that Trio/Rac interacts to promote Abl-dependent axon patterning. Together, these
functionally with WAVE components to control Drosophila axon interactions allow the Abl signaling network to control a wide
patterning. We knocked down scar expression by RNAi (Zallen varietyofaxonpatterningdecisionsthroughoutanimalphylogeny,
et al., 2002) in neurons and found that this produced the ISNb perhapsbycoordinatingtheEna-dependentdynamicsoflinearactin
stalling phenotype characteristic of Abl pathway mutants: ISNb bundles with the WAVE-dependent dynamics of branched actin
stallsin24%ofRNAi-expressinghemisegments(n=377)versus7% networks.
Fig.6.EnaandTrioactinparalleldownstreamofAbl.
Eyeimaginaldiscswereisolatedfromthirdinstarlarvaeof
theindicatedgenotypes,stainedwithanti-Ena(green)and
phalloidin(red)andimagedbyconfocalmicroscopy.
(A-C)Enachannel;(A′-C′)Enaandphalloidin.Arrows
highlightthedisc-likeappearanceofEnaimmunoreactivity
associatedwithcis-Golgi,distributedthroughoutthe
photoreceptorsomain(A,A′)WTand(C,C′)trio−.
(B,B′)EnaimmunoreactivityintheDabmutant
concentratesinasingle,largeblobatthemostbasalpoint
inthesomaofeachphotoreceptor.Scalebar:5µm.
T
N
E
M
P
O
L
E
V
E
D
493
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
Fig.7.GeneticinteractionofTrio/RacwithWAVEcomponentsandmodelfortheAblpathway.(A-E)Nomarskiimagesshowingthreehemisegmentsofstage
17embryosoftheindicatedgenotypes,immunostainedwithanti-Fas2andfilet-mountedtorevealtheISNbmotonerve.Blackarrowsindicatemuscle12
neuromuscularsynapses;redarrowsindicatepositionsofmissingsynapses(‘stall’phenotype).Scalebar:10µm.(F)QuantificationofISNbstallphenotypes
(failuretoinnervatemuscle12).***P<0.001,χ2withBonferronicorrection.(G)ModelforthefunctionalorganizationoftheAblsignalingpathway.Arrowindicatesa
stimulatoryinteraction,T-barindicatesaninhibitoryinteraction,dashedlineindicatesaphysicalassociation(co-IP)withoutspecifyingepistaticdirectionality.
The relationship of Dab to Abl has long been controversial indeed,DabandAblco-IPfrombothculturedDrosophilaS2cells
(Gertleret al., 1993; Howell et al., 1997; Liebl et al., 2003; Song and WT embryo lysates. We also found that Dab enhances the T
N
etal.,2010;Sonoshitaetal.,2015).Forsometime,itwasunclear kinase activityof Abl in vivo, as assayed by FRET, in addition to
E
whetherDabisevenacorecomponentoftheAblpathway,although controlling Abl subcellular localization. These data provide a M
morerecentlyweestablishedthecentralroleofDabinAblpathway potential biochemical basis for the functional interaction of these P
function(Liebletal.,2003;Songetal.,2010).Moreover,arecent proteins. O
reportsuggestedthatDabcanco-IPwithAblfrommammaliancells HerewedemonstratethatDabphysicallyassociateswithAbl,and L
E
(Sonoshita et al., 2015), whereas previous investigators only we have shown previously that Dab associates with Trio (Le Gall V
detectedbindingbyinvitropulldownassays(Gertleretal.,1993; etal.,2008).Themostparsimoniousinterpretation,therefore,isthat E
D
Howelletal.,1997).Wereinvestigatedthisquestionandfoundthat, Abl,DabandTrioformasingle,trimericcomplex.Wehaveasyet
494
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
beenunable,however,toshowdirectinteractionofAblwithTrioor componentofalargeproteincomplexalsocomprisingSra1,Nap1,
co-IP of Abl and Trio from tissue or cell lysates. Although we AbiandHSPC300(Chenetal.,2010;Gautreauetal.,2004).This
cannot exclude the possibility that Dab forms separate complexes complexpromotesbranchedactinnetworksbyrecruitingtheArp2/3
with Abl and with Trio, we think it more likely that this reflects complextothesideofanexistingactinfilamentandstimulatingits
technicallimitationsoftheexperiment.Itmightbethatthetrimeric abilitytonucleateadaughterfilamentata70°angletothemother
complexislabileinvitro,preventingusfromdetectingco-IPofAbl filament (Blanchoinetal.,2000;MacheskyandInsall,1998).The
withTrio. WAVE complex is initially inactive, and must be activated by
ItisunclearhowAblkinaseinducestheTriospectrinrepeatsto association with GTP-Rac (Eden et al., 2002). Once activated,
relieve repression of Trio GEF1 activity by the Trio NTD. A WAVEcomplexesextendbranchedactinnetworks,forexamplein
previousstudyreporteddirectbindingbetweenspecificAblandTrio lamellipodia.ThecontrastingactivitiesofEnaandRac/WAVEcan
domainsinvitro,asassayedusingpulldownswithproteinfragments beseenintheireffectsoncellmorphology,whereoverexpressionof
(Forsthoefeletal.,2005).Wewereunabletoconfirmthisbyco-IPof Enapromotesaspiky,filopodialmorphology,whereasactivationof
thefull-lengthproteinsfromcellorembryolysates,butthatcould Racproducesabroad,lamellarmorphology(InsallandMachesky,
reflectinsufficientsensitivityoftheassay,particularlyifassociation 2009;Lacayoetal.,2007;Rogersetal.,2003).
withspecificactivatedreceptorsisnecessaryforcomplexformation. Our data demonstrating that Abl simultaneously suppresses the
Previous studies have also failed to provide rigorous evidence of activityofEna,butstimulatesWAVE/Rac,thereforerevealthatthe
directphosphorylationofTriobyAbl(Forsthoefeletal.,2005).Our structureoftheAblnetworkintrinsicallyintroducesanantagonism
FRETresults,however,areconsistentwiththeideathatanadditional betweenthetwomajorclassesofactinstructuresinthecell(Burke
proteinmaybeinvolvedinthestimulationofTriobyAbl–onethat etal.,2014;Chenetal.,2014).Areceptorthatevokestheactivityof
can be titrated into inactive complexes by expression of non- the Abl pathway will automatically induce a default balance
inducibleTrioderivatives(Fig.5D).Itmightbe,forexample,that betweenthetwomajorclassesofactinstructuresinthegrowthcone,
thishypotheticalthirdcomponentisthedirecttargetofAblkinase, linear actin bundles and branched actin networks. Receptors that
andthenassociateswithTrioorsignalstoit.Testingthishypothesis modulate the two legs of the pathwayseparately, by contrast,will
awaitstheisolationofsuchaputativeco-factor.Inaddition,although modifythatbalance.ItisattractivetospeculatethattheabilityofAbl
ourdatashowastrongrequirementforAblkinaseactivityinTrio toautomaticallybalance,andrebalance,thedifferentkindsofactin
activation, we cannot rule out the possibility that Abl also has a structuresinthecellmightbewhysomanymotilityandguidance
scaffoldingrolethatisseparatefromitskinaseactivity(Hoffmann, receptorshaveevolvedtosignalthroughtheAblnetwork(Bradley
1991; Rogers et al., 2016). We could not use transgenic andKoleske,2009;LanierandGertler,2000).
overexpression of Abl derivatives to query possible non-kinase Formorethantwodecades,theAblsignalingcassettehasbeen
functionsofAbl,ortochallengethetriodependenceofactivationof oneofthekeysystemsusedtointerrogateaxongrowthandguidance
Rac FRET by Abl, since overexpression of Abl caused non- and neuronal migration (Goodman and Shatz, 1993; Hoffmann,
physiological activation of other Rac GEFs, including SOS (R.K. 1991;LanierandGertler,2000;Willsetal.,1999a).Itsexplanatory
andE.G.,unpublishedobservations). value,however,hasbeenseverelylimitedbythelackofamolecular
It is striking that conceptually analogous, but molecularly very schemaforinterpretingtheinteractionsamongitscomponents.The
different, interactions have now been observed between Abl and resultsthatwereporthereprovideadetailedmolecularmodelforthe
Trio in different contexts. In experiments here, we show that Abl Abl signaling network in neuronsthat can now be used to design
interactsfunctionallywithTriotostimulateRacsignaling,andthis and interpret cellularand molecular studies of the mechanisms of
requires Abl kinase activity. Analogously, in mammalian tumor neuronalmorphogenesisandmotility.
cells,AblphosphorylatestheC-terminalportionofTriotostimulate
itsRhoGEFactivity,andthispromotesmetastasis(Sonoshitaetal., MATERIALSANDMETHODS
2015).ThedomainofTriothatbecomesphosphorylatedbyAblin Drosophilastocks
the mouse tumor model, however, and by Fyn in mouse neurons DrosophilabearingP[UASRacFRET]wereobtainedfromD.Montell(UC
(DeGeeretal.,2013)doesnotexistinDrosophilaTrio.Moreover, SantaBarbara,CA,USA);P[UASBCR-p210]andP[UASBCR-p185]were
previousexperimentshaveshownthattheRhoGEFdomainofTrio fromM.Peifer(UNC-ChapelHill,NC,USA);AbimutantswerefromJ.-L.
isdispensableforaxonanddendritedevelopmentinDrosophilaand Juang(NHRI,Taipei,Taiwan);andP[UAS-SCARRNAi]wasfromJ.Zallen
C.elegans,andthatitistheRacGEFactivityofTriothatcontrols (MSKCC, New York, NY, USA). All other fly stocks were described
previously(Kannanetal.,2014;SongandGiniger,2011;Songetal.,2010)
neuronal morphogenesis(Iyeretal.,2012; Shivalkarand Giniger,
orobtainedfromtheBloomingtonDrosophilaStockCenter.
2012;SongandGiniger,2011;Stevenetal.,1998).Nonetheless,it
isremarkablethatinbothcontextsAblactsthroughstimulationof
TriotocontrolmotilityeventsthatdependonRhofamilyGTPases, ConstructionofAblFRETbiosensorandgenerationofDrosophila
albeitbydifferentmolecularmechanisms. transgenics
Plasmid for mammalian CFP-YFP Abl biosensor (GenBank accession
EnaandRac/WAVEaretwokeyoutputsoftheAblpathwayfor
numberAF440203;2334bp)wasobtainedfromDrRogerTsien(UCSD,
regulationoftheneuronalcytoskeleton.Enaregulatesactinviathree
San Diego, CA, USA). Site-directed mutagenesis was performed to
activities(Krauseetal.,2003;Trichetetal.,2008):itpromotesactin T
constructaY221Fmutationasanegativecontrol.cDNAofboththeWT N
polymerization,bundlesactinfilamentsthroughitsmultimerization
andY221FbiosensorswassubclonedintopUAS-TwithBamHI/XhoI,and E
domain, and prevents capping of the actin barbed end. Mutant transgenicflylinesweregeneratedbyBestGene. M
analysissuggeststhatallthreeactivitiescontributetoEna-dependent P
axonpatterning.Forexample,mutationofthedimerizationdomain O
DrosophilaS2cellgrowthandtransfection
generatesanullenaalleleinDrosophila,andmutationofcapping L
DrosophilaS2cells(E.Serpe,NIH,Bethesda,MD,USA)wereculturedin E
proteinβcansuppressenaphenotypes(Gatesetal.,2009).Together, Schneider’s medium (Gibco, Thermo Fisher) with 10% heat-inactivated V
thedifferentactivitiesofEnaconspiretoextendandbundleparallel fetal bovine serum (JRH Biosciences) and 1% penicillin-streptomycin E
D
actinfilaments,asarefoundinfilopodia.WAVE/SCARisthecore (1:100;Gibco).Mediumwasfiltersterilizedandcellsgrownat23-25°C.S2
495
RESEARCHARTICLE Development(2017)144,487-498doi:10.1242/dev.143776
cell transfection was bystandard methods using DDAB reagent (Sigma) (Kannan et al., 2014). z-sections were acquired using a Zeiss LSM 510
plus0.7-4µgplasmidDNA.Transgeneexpressionwasdrivenwithactin- confocalat63×magnificationanddeconvolutedusingAutoQuant(Media
GAL4 (pRK241). For FRETexperiments, measurements were performed Cybernetics).
48haftertransfection.
EmbryostainingforISNbmotoraxondefects
DrosophilaPRculture Embryofixation,staining,imagingandquantificationofISNbphenotypes
Culture of larval photoreceptors was performed by published methods wereasdescribedpreviously(SongandGiniger,2011;Songetal.,2010).
(Newsomeetal.,2000).Briefly,100-150eye-antennalimaginaldiscswere Abdominal hemisegments 2-7 of earlystage 17 embryos were scored by
dissected from third instar larvae in S2 cell culture media, and the eye Nomarskimicroscopyforpresenceorabsenceofaneuromuscularjunction
portion of each was isolated with a tungsten needle and transferred to a on muscle 12. Data from independent experiments were pooled for each
microcentrifugetubeusingasilanizedpipet.Discswereincubatedfor30 genotypetoderivethenumberofhemisegmentsinwhichISNbdid,ordid
minatroomtemperaturein400μlcollagenaseplus100μlliberaseI,then not,formthemuscle12neuromuscularjunction.Significancewasassessed
trituratedtohomogeneity.Cellswerepelletedfor5minat5000g,washed byχ2test.
threetimeswithculturemedium,suspendedin100μlculturemediumina
MatTekdishcoatedwithpolyL-lysineandconconavalin-A,andincubated Antibodiesandphalloidin
at25°Cinthedarkinahumidchamber. Mouseanti-Ena(5G2;1:50),ratanti-Elav(7E8A10;1:20),anti-Fasciclin2
(1D4 concentrate; 1:100), anti-Dab (P4D11 for immunoprecipitation;
RatioFRETimaginginS2cellsandPRs P6E11 for western analysis) and anti-Trio (9A; 1:50) were from the
RatiometricFRETimagingwascarriedoutbythemethodofKogaetal. Developmental Studies Hybridoma Bank (Iowa City, IA, USA). Rabbit
(2006).ImageswerecollectedusingaDeltaVisionmicroscopewith60×, anti-Dabandrabbitanti-Ablweredescribedpreviously(Songetal.,2010).
1.42NA lens, 2×zoom and2×2 binning (for PRs) ora Zeiss LSM 510 Mouseanti-DrosophilaAblpolyclonalserumwaspreparedbytheAntibody
confocalmicroscopeusing63×,1.4NA,2×zoom(forS2cells).Onthe DevelopmentLaboratoryoftheFredHutchinsonCancerResearchCenter,
DeltaVision,stacksof∼50-75z-sectionswereacquiredwith0.15µmstep usingthesameimmunogenasforrabbitanti-Abl.Specificitywasverified
sizesequentiallyinYFP,CFP,FRET,andDIC.OntheLSM510,a405nm bystainingofAblnullmutantimaginaldiscs.Rabbitanti-β-galactosidase
laser was used to excite and a stack of ∼14 z-sections was collected (55976; 1:10,000) was from Cappel. Secondary antibodies (1:250) were
simultaneouslyinCFPandFRET.Measurementswereperformedfromz- from Jackson ImmunoResearch. Alexa 488-conjugated phalloidin (Life
projections of the image stack. Regions of interest (ROIs) were drawn Technologies)wasusedat1:300.
manuallyinMetaMorph(MolecularDevices)andaverageintensityvalues
forthesameROIinCFP(I )andFRET(I )wereexportedintoExcel Biochemicalmethods
cfp fret
(Microsoft). Background average intensity values (I and I ) Co-IPexperimentswereperformedasdescribedpreviously(LeGalletal.,
bg-cfp bg-fret
were measured from the same images. The FRET ratio (R) was 2008).
f
calculated as R=(I −I )/(I −I ). CFP/FRETratio images were
f cfp bg-cfp fret bg-fret
presented for display purposes in the intensity modified display mode Westernblotting
(Wangetal.,2005).23-38cellswereimagedforeachvaluereported,except Western blotting was performed by standard methods. Signals were
forSTI-571timecourseexperiments(n=17-20). visualized using secondary antibodies coupled with IR-Dye 700 or 800
and scanned with the Odyssey infrared imaging system (Li-Cor
AcceptorphotobleachingforAblFRET Biosciences),orusingperoxidase-coupledsecondaryantibodiesandECL
FRET measurement by acceptor photobleaching was performed as reagents(Lumigen).
described(Karpovaetal.,2003).SingleplanesequentialimagesinCFP,
then YFP, were collected using a Zeiss LSM 780 with 63×, 1.46 NA Imagepreparation
objective,4×zoom.GaASPdetectorwasadjustedtoeliminatecross-talk Image adjustments, including brightness, contrast, gamma and color
and optimize dynamic range. Photobleaching of the whole cell was balance, were applied to entire figure panels, as necessary. DIC embryo
performedwitha514nmlaser.Wecollectedthreereferenceimagesbefore images were montaged from multiple focal planes and in some cases an
photobleachandfiveimagesafter.CircularROIsencompassingthewhole unsharpmaskwasappliedtothefinalimagetoclarifyanatomicalfeatures.
cellweredrawnmanuallyinbackground-subtractedCFPchannelimagesto
measureaverageintensityvalues(I )beforeandafterphotobleach.FRET
efficiency (E) was calculated as En=(I −I )×100/I , where I represents Statisticalmethods
f f 6 5 6 n Numerical data in all experiments were corrected for multiple testing by
averageintensityatthenthtimepoint.Asacontrol,weperformedmock
the Bonferroni method. For embryo axonal phenotypes, statistical
photobleach with the laser switched off without changing other FRET
significanceofplannedcomparisonsbetweengenotypeswasassessedby
imagingparameters.
atwo-class(innervatedversusnon-innervated)χ2testwithonedegreeof
freedom, with P-value calculated in Excel (CHITEST), followed by
STI-571preparationandtreatmentinPRandS2cellcultures
Bonferronicorrection.
A 100mg pill of STI-571 (Novartis, obtained from NIH Clinical Center
Pharmacy) was dissolved in 20ml PBS to provide a 10mM stock, and
aliquots were stored at −20°C. STI-571 was added to media at time of Acknowledgements
Wearegratefultoeveryonewhogaveusadvice,assistanceandreagentsforthese
imaging.Washoutwasinitiatedwithde-oxygenatedculturemediumusing
experiments.WeparticularlythankRogerTsienandAliceTingforthemammalian
anautomaticpump.
AblFRETprobe,andDeniseMontellandmembersofherlabforDNAandflies
bearingtheRacFRETprobe.Inadditiontothemembersofourlab,especially T
triomutantconstructs GingerHunterandJoshSpurrierforcommentsonthemanuscript,weare N
EquivalentdomaindeletionswereintroducedintoUAS-TriomGEF1and particularlygratefultoChi-HonLee,ElaSerpeandmembersoftheirlabsformany E
UAS-TriomGEF2backbonesbyPCRusingExpandhigh-fidelityPCRTaq helpfuldiscussionsandforinvaluableassistance,andtoStevenVogelandClare M
WatermanforadviceonFRET.WealsothankStephenWincovitchoftheNational
DNApolymerase(Roche)andverifiedbysequencing. P
HumanGenomeResearchInstitute(NHGRI)MicroscopyCoreformuchhelpwith O
imaging,andKoryJohnsonoftheNINDSBiostatisticsCoreforassistancewith
L
Immunostainingandantibodies statisticalanalysis. E
Larvaleyediscs V
Thirdinstarlarvaleyediscswithbrainlobesattachedweredissected,fixed, Competinginterests E
D
mounted in Vectashield (Vector Labs) and imaged by standard methods Theauthorsdeclarenocompetingorfinancialinterests.
496