EphrinB2 affects apical constriction in Xenopus embryos and is regulated by ADAM10 and flotillin-1

15 Pages·2014·2.66 MB·English

by  Yon Ju Ji

ARTICLE
Received7Nov2013|Accepted 26 Feb 2014|Published 24Mar2014 DOI:10.1038/ncomms4516
EphrinB2 affects apical constriction in Xenopus
embryos and is regulated by ADAM10 and flotillin-1
Yon Ju Ji1, Yoo-Seok Hwang1, Kathleen Mood1, Hee-Jun Cho1, Hyun-Shik Lee2, Emily Winterbottom1,
He´le`ne Cousin3 & Ira O. Daar1
The Eph/ephrin signalling pathways have a critical function in cell adhesion and repulsion,
andthusplaykeyrolesinvariousmorphogeneticeventsduringdevelopment.Hereweshow
thatadecreaseinephrinB2proteincausesneuraltubeclosuredefectsduringXenopuslaevis
embryogenesis.SuchadecreaseinephrinB2proteinlevelsisobservedonthelossofflotillin-1
scaffold protein, a newly identified ephrinB2-binding partner. This dramatic decline in
ephrinB2 protein levels on the absence of flotillin-1 expression is specific, and is partly the
result of an increased susceptibility to cleavage by the metalloprotease ADAM10. These
findings indicate that flotillin-1 regulates ephrinB2 protein levels through ADAM10, and is
required for appropriate neural tube morphogenesis in the Xenopus embryo.
1LaboratoryofCellandDevelopmentalSignaling,NationalCancerInstitute-Frederick,Frederick,Maryland21702,USA.2ABRC,CMRI,SchoolofLife
Sciences,CollegeofNaturalSciences,KyungpookNationalUniversity,Daegu702-701,SouthKorea.3DepartmentofVeterinaryandAnimalSciences,
UniversityofMassachusetts,Amherst,Massachusetts01003,USA.CorrespondenceandrequestsformaterialsshouldbeaddressedtoI.O.D.
(email:[email protected]).
NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications 1
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
T
he ephrins are all membrane-bound proteins subdivided contexts.Werecentlyidentifiedthelipidraftproteinflotillin-2as
into two classes, the A subclass being glycosylphosphati- anephrinB1interactorbymassspectrometricanalysisofimmune
dylinositol-linked to the membrane and the B subclass complexes from Xenopus embryos overexpressing ephrinB1. On
beingtransmembraneproteinswithashortcytoplasmicdomain. thebasisoftheseresults,wetestedwhetherephrinB2wasableto
The cognate Eph receptors are transmembrane receptor tyrosine interact with flotillin-1 and flotillin-2 since the intracellular
kinases,andarealsodividedintotwoclasses(AandB)basedon domains of ephrinB1 and ephrinB2 are 82% identical and
theirsequencesimilarityandtheirbindingspecificitytowardsthe flotillin-1andflotillin-2,similartotheephrinBs,areknowntobe
ephrin subclasses. Eph-ephrin contact dependent interactions expressed in the neural tissue in vivo15–17. Flag-tagged ephrinB2
between two cells result in bi-directional signalling. During and haemagglutinin (HA)-tagged flotillin-1a or -1b were
development, these Eph/ephrin interactions lead to cell sorting overexpressed in Xenopus oocytes, and co-immunoprecipitation
and boundary formation between receptor and ligand-bearing (co-IP) analysis was performed. Flotillin-1a-HA and flotillin-1b-
cells1.WhenmotilecellsexpressingeitherEphorephrincomein HA were detected in ephrinB2-Flag immune complexes, and
contactwithcellsexpressingthecognatepartner,theresponseis ephrinB2-Flag was detected in flotillin-1a- and -1b-HA immune
often adhesion or repulsion. complexes (Fig. 1a). Similarly, co-IP analysis of oocytes over-
The choice between cell adhesion/attraction or de-adhesion/ expressing flotillin-2 with ephrinB1 or ephrinB2 also showed an
repulsiondependsonthecelltypeandsignallingcontext2.Inthe association of both ephrinBs with flotillin-2 (Supplementary
latter case, Eph/ephrin-mediated adhesion can be released by Fig. 1a). In HT-29 colon carcinoma cells, we were able to co-
endocytosis of the Eph/ephrin complex into either Eph- or immunoprecipitateendogenousflotillin-1withephrinB2immune
ephrin-expressing cells, allowing the cells to move on to their complexes (Fig. 1b), confirming that ephrinB2 and flotillin-1
respective destination. This endocytosis can be accomplished by associate in vivo. In addition, we show by in situ hybridization
ephrinB or EphB transendocytosis3–5. The EphB/ephrinB thatexpressionofephrinB2,flotillin-1aandflotillin-1boverlapin
complex is endocytosed in an EphB kinase-dependent manner, the neural plate of neurula stage embryos (Supplementary
preferentially into cells with more adhesive contacts with the Fig. 1b), as has been shown by others17,18. Whole mount
substrate and a well-developed actin cytoskeleton3. Loss of cell immunocytochemistry analysis confirmed the expression of
adhesion initiated by EphB/ephrinB is observed during flotillin-1 in the neuroepithelium and ectoderm (Fig. 1c), where
developmental processes such as notochord formation where in ephrinB2 is known to be expressed (Supplementary Fig. 1b;
response to non-canonical Wnt signalling, phosphorylated (refs18,19),andalsoshowsexpressioninthesomites,notochord
EphB receptors make a ternary complex with the scaffold and the roof of the archenteron (Fig. 1c).
protein dishevelled2 and the formin homology protein Immunocytochemistry was also performed on embryos
Daam1, which is transported to the endocytic vesicles in a injected with antisense morpholino oligonucleotides (MOs)
dynamin-dependent manner. This removal of EphB molecules embryos against flotillin-1a and flotillin-1b (F1aMO and
from the cell surface results in loss of adhesion, leading to F1bMO). The F1aMO- and F1bMO-injected embryos show a
initiation of convergent extension cell movements during dramaticdecreaseinflotillin-1immunoreactivity,confirmingthe
notochord development6. specificity of the immunostaining as well as the ability of these
OnecriticalfactorwhenconsideringEph/ephrin-mediatedcell MOs to inhibit translation of the endogenous flotillin-1
repulsion and disengagement is that the interaction between paralogues (Fig. 1c). Moreover, exogenously expressed flotillin-
Eph receptors and ephrin proteins must first be terminated. 1b-HAandephrinB2-Flagarebothdetectedinthecellmembrane
While endocytosis certainly offers a long-term solution to this (Fig. 1d), suggesting that they are spatially co-localized. In
termination7, another efficient way to cease the adhesion is by addition, using live cell imaging in embryos co-expressing an
ectodomaincleavage.Adisintegrinandmetalloprotease(ADAM) ephrinB2-mCherry fusion construct along with a flotillin-1-
proteinsaretypeItransmembraneproteinswithanextracellular enhancedgreenfluorescentprotein(EGFP)fusionproteinshows
metalloproteinase domain and disintegrin and cysteine-rich thatephrinB2andflotillin-1canbefoundco-localizingonthecell
domains8,9. ADAMs have been shown to cleave ephrinA and surface (Fig. 1e; Supplementary Movie 1). Altogether, these data
ephrinB proteins7,10,11, and Eph receptors are also subject to suggest that ephrinB2 and flotillin-1 interact in vivo and may
cleavage by metalloproteases and g-secretase12,13. However, little function together during Xenopus development.
is known about the mechanisms that control the cleavage of the
ephrins and Ephs by these metalloproteases.
Here we show that loss of the ephrinB2 interactor, flotillin-1, Lossofflotillin-1leadstoareductioninephrinB2levels.Since
leads to a marked increase in ephrinB2 protein cleavage and ephrinB2isnormallylocalizedtothebasolateralregionofplasma
processing, which causes neural tube closure defects in Xenopus membranes,weexaminedwhetherflotillin-1influencesephrinB2
embryos and an associated disruption of cell shape and actin localizationorexpression(Fig.2a–c).EphrinB2RNAwasinjected
cytoskeleton. Moreover, we identify ADAM10 as the specific into both sides of an embryo, and the F1aMO or F1bMO along
metalloproteasethatcleavesephrinB2intheabsenceofflotillin-1. with GFP RNA was introduced into only one side. Surprisingly,
Thus, we show that ephrinB2 protein levels are sustained by a ephrinB2 protein levels were greatly reduced in the plasma
lipid raft protein (flotillin-1) that interacts with ephrinB2, and membraneofF1aMO-andF1bMO-bearingcells,asevidencedby
inhibits cleavage and processing by ADAM10. Moreover, this the dramatically reduced ephrinB2 fluorescence (red) found in
study provides a link between ephrinB2 regulation and the the F1MO-bearing cells (green) when compared with control
important developmental process of neural tube closure. MO (CoMO)-bearing cells (non-green; Fig. 2b,c). Western blot
analysis also showed that ephrinB2-HA protein expression was
greatly diminished when flotillin-1a was knocked down, while
Results exogenousexpressionofephrinB1andephrinB3wascomparable
EphrinB2 associates with flotillin-1. EphrinBs have several in the presence of F1aMO or CoMO (Fig. 2d). Moreover, the
interactingproteinsthatmediateorregulatetheirmorphogenetic introduction of MO-resistant flotillin-1a or -1b RNA along with
functions during development14. Identifying these regulators in the respective MO rescued ephrinB2 protein levels (Fig. 2e,f),
the context of various morphogenetic events and systems can indicatingthatflotillin-1specificallyinhibitsthelossofephrinB2
reveal regulatory networks that may function in other cellular protein expression.
2 NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
a
Flotillin-1-HA – a b a b b
EphrinB2-flag + – – + + E
Cph
IP flag 50 HFlAag IP ontrolrinB2
36
50 Flotillin-1
IP HA 36 Flag 36 EphrinB2
50 HA Lysate 50 Flotillin-1
36 EphrinB2
Lysate 36 Flag
50 HA
c
Enlarged image F1aMO
of boxed region +F1bMO
1
n-
otilli
Fl
d
Flotillin-1b-HA EphrinB2-flag Flotillin-1b-HA
ephrinB2-flag
e
0.0 31.6 63.2 94.7 126.3
Figure1|EphrinB2isassociatedandco-localizedwithflotillin-1.(a)Flotillin-1isassociatedwithephrinB2.Co-IPandwesternblotanalysisof
XenopusoocytesexogenouslyexpressingephrinB2-Flagwithflotillin-1a-HAor-1b-HA.(b)EndogenousephrinB2interactswithendogenousflotillin-1.
EndogenousephrinB2wasimmunoprecipitatedwithanti-goatephrinB2antibodyinHT-29coloncarcinomacells,andflotillin-1wasdetectedusinganti-
mouseflotillin-1antibodyinephrinB2immunecomplexesbywesternblotting.(c)Flotillin-1immunostainingofembryosections.Immunostainingof
endogenousflotillin-1instage15embryos,andamagnifiedviewoftheneuralplateregion.EmbryoswerealsoinjectedwithF1aMOandF1bMOas
indicated,andimmunostainedforflotillin-1.Scalebar,200mm.(d)Subcellularlocalizationofflotillin-1b-HAandephrinB2-Flag.Flotillin-1b-HAand
ephrinB2-FlagRNAwasinjectedintoXenopusembryosandstainedwithanti-HAoranti-Flagantibodies.Scalebar,20mm.(e)Co-localizationofephrinB2-
mCherryandflotillin-1-GFPattheplasmamembraneinembryos.Flotillin-1-GFP(green)andephrinB2-mCherry(red)RNAswereinjectedintoembryosand
localizationofproteinswasobservedbyconfocalmicroscopy.Themergedimageisvisibleasyellowstaining.Theenlargedimagesaresplitscreen
enlargementsoftheboxedareas(fromdifferentframesoftheSupplementaryMovie1)withflotillin-1-GFP(green)attopofsplitandephrinB2-mCherry
(red)atbottom.Bluearrowsindicatethesameareasofco-localizationinthemembraneduringmovement.Scalebar,10mm.
NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications 3
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
b c
a
EphrinB2 EphrinB2 EphrinB2-HA Ephrin-B2HA/GFP(F1aMO) EphrinB2-flag EphrinB2-flag/GFP(f1bMO)
+CoMO +F1MO(GFP)
GFP(F1aMO) EphrinB2-HA
GFP(F1bMO) EphrinB2-flag
d e f g h
EphrinBEM-rHHkOAA2 –– CoB1F1aCoB2F1aCoB3F1a 434050 F1EapΔhUriTnFRB1-2aEFf-MlHHralakOgAAg2 –––––+ ++–+++45440500 F1EbpΔhUriTnFRB1-2bEFf-MlrHHalakOgAAg2 –––––+ ++–+++ 4354405500 Non-EsppBhe1rcMMinifOOiB4c0 Co– +–F1a+F1b+F1a+F1b+B2+ 3E5rk2 Non-spBe2cMiMfiO34cO50 Co– –+F1a++B1 EErkp2hrinB
Figure2|Lossofflotillin-1leadstoareductioninephrinB2expression.(a)Injectionscheme.CoMOwithephrinB2RNAisinjectedintoonecell
oftwo-cell-stageembryos,andF1MOalongwithephrinB2RNAandGFPRNAwereinjectedintotheothercell.(b)EphrinB2expressionisreducedinthe
absenceofflotillin-1a.F1aMO,ephrinB2-HARNAandGFPRNAco-injectedcellsaregreen.CoMOandephrinB2-HARNA-co-injectedcellsarenotgreen.
EphrinB2-HAexpressionisvisualizedinred.CellscontainingF1aMO,butdisplayingreducedephrinB-HAexpressionareoutlinedinwhitedots.
Scalebar,20mm.(c)F1bMOdecreasesephrinB2expression.F1bMO,ephrinB2-FlagRNAandGFPRNAco-injectedcellsfluorescegreen.CoMOand
ephrinB2-FlagRNA-co-injectedcellsarenotgreen.EphrinB2-HAexpressionisvisualizedinred.ArrowheadsindicateGFP-expressingcellsthatharbourthe
F1bMO.Scalebar,20mm.(d)Flotillin-1knockdownspecificallycausesephrinB2loss.EmbryoswereinjectedwithHA-taggedephrinB1,-B2or-B3RNAs
alongwitheitherCoMOorF1aMO.Westernblotanalysiswasperformedwithanti-HAantibodyandanti-Erk2asaloadingcontrol.(e)EphrinB2lossdueto
theF1aMOisrescuedbyre-expressionofflotillin-1a.WesternblotanalysisofembryosinjectedwithephrinB2-HARNAandF1aMO,withorwithout
F1aDUTR-FlagRNA.(f)DecreaseinephrinB2duetoF1bMOisrescuedbyF1bDUTR-Flagexpression.Westernblotanalysisofembryosinjectedwith
ephrinB2-HARNAandF1bMO,withorwithoutF1bDUTR-FlagRNA.(g)WesternblotanalysisofendogenousephrinB2proteininneuralfoldsthathavebeen
injectedwiththeindicatedMOs,andblottedusingtheindicatedantibodies.Erk2isusedasaloadingcontrol.(h)Westernblotanalysisofendogenous
ephrinB1proteininneuralfoldsthathavebeeninjectedwiththeindicatedMOs,andblottedusingtheindicatedantibodies.Erk2isusedasaloadingcontrol.
WeexaminedtheendogenousephrinB1andephrinB2protein abundance of the ephrinB2D60 mutant, similar to the effect on
expression in the presumptive neural tube of F1aMO-injected wild-type ephrinB2 (Fig. 3a). In contrast, ephrinB2TM-Cyto pro-
embryos, since this region shows robust overlap in expression tein levels were unaffected by the F1aMO, indicating that the
oftheephrinBandflotillin-1genes(SupplementaryFig.1b).Since ephrinB2 ectodomain is an important determinant of ephrinB2
the ephrinB antibody has cross-reactivity to all three ephrinBs, loss in the absence of flotillin-1 expression (Fig. 3a).
the F1aMO-injected embryos were co-injected with either Immunoprecipitation analysis shows that flotillin-1 interacts
ephrinB1MO (B1MO) and/or ephrinB2 MO (B2MO) to detect with both ephrinB1 and ephrinB2, but possesses much less
the remaining ephrinB isoform. The dorsal axial region (neural affinity for ephrinB3(Fig. 3b).AlthoughephrinB1 and ephrinB2
fold) of these embryos was excised and subjected to western canbothinteractwithflotillin-1,onlyephrinB2proteinlevelsare
analysis (Fig. 2g). The analysis shows that injection of both compromised by flotillin-1 knockdown (Fig. 2d,g). Thus, we
B1MO and B2MO blocks almost all detectable ephrinB signal in hypothesized that the ephrinB1 ectodomain may confer resis-
the dorsal axial region, indicating that most of the detectable tancetotheeffectoftheF1aMO.Totestthishypothesis,wemade
ephrinB protein is due to ephrinB1 and ephrinB2 (Fig. 2g). As chimericproteinscomposedoftheephrinB1ectodomainandthe
expected, when B1MO-harbouring embryos are co-injected with ephrinB2 transmembrane and intracellular domains (B1B2-HA)
F1aMO or F1bMO, a dramatic loss of the endogenous ephrinB2 or the ephrinB2 ectodomain and the ephrinB1 transmembrane
proteinisobservedinthedorsalaxialregion(Fig.2g).Incontrast, and intracellular domains (B2B1-HA; Fig. 3c). Consistent with
co-injection of F1aMO in B2MO-bearing embryos shows only a ourprediction,theB1B2-HAchimericproteinisstablyexpressed
verymodesteffectontherelativelylowlevelofephrinB1protein in F1aMO-injected embryos, while B2B1-HA expression is
foundinthistissue(Fig.2h).Thesedataindicatethatendogenous reduced in the presence of the F1aMO, similar to wild-type
ephrinB2proteinispreferentiallydecreasedinneuraltissueinthe ephrinB2 (Fig. 3c). These data suggest that the reduction in
absence of flotillin-1. ephrinB2 protein levels in the absence of flotillin-1 depends on
the presence of the ephrinB2 ectodomain.
EphrinB2ectodomainiscriticalfordecreasedephrinB2levels.
EphrinBsareeachcomposedofanectodomain,atransmembrane ADAM10leadstolossofephrinB2levels.Ithaspreviouslybeen
domain and an intracellular domain; therefore, we analysed shown that the ectodomain of ephrins can be cleaved by metal-
whichephrinB2domainscontributetothelossofephrinB2inthe loproteases. For example, ephrinB1 is cleaved by matrix metal-
absence of flotillin-1. We generated an ephrinB2D60 mutant that loproteinase-8 (MMP-8), and ephrin-A2 and ephrin-A5 are
has 60 amino acids deleted from the intracellular domain. cleavedbyADAM10(refs7,10,21).Thus,wetestedthepossibility
Anothermutant,ephrinB2TM-Cyto,thatlackstheectodomainand that a metalloprotease cleaves ephrinB2 and causes its decreased
consists of only the transmembrane and intracellular domains expression in the absence of flotillin-1. We found that two
was also created, and was found to localize to the plasma mem- different broad-spectrum metalloprotease inhibitors, BB-94 and
brane, similar to full-length ephrinB2 (Supplementary Fig. 1c). GM6001,bothstabilizedephrinB2expressioninF1aMO-injected
Injection of the F1aMO led to a significant decrease in protein embryos(Fig.4a).Recently,itwasshownthatXenopusephrinB1
4 NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
a ADAM17 had no such effect (Fig. 4b,c). Reintroduction of
Ectodomain TM Cytoplasmic ADAM10 in the presence of the ADAM10 MO led to reduced
Δ60 ephrinB2 levels, supporting that ADAM10 specifically reduces
TM-Cyto ephrinB2proteinlevels(Fig.4b).Together,thesedatasuggestthat
flotillin-1inhibitsADAM10-mediatedephrinB2loss.Toconfirm
EphrinB2-HA WT Δ60 TM-Cyto that ADAM10 is the specific metalloprotease that reduces
F F F ephrinB2 protein expression, ADAM10 and ADAM17 were
MO – Co 1a – Co 1a Co 1a overexpressed with ephrinB2 (Fig. 4d). Indeed, ADAM10
40
overexpression significantly lowers ephrinB2 protein levels in a
15 HA
35 dose-dependentmanner,whileADAM17hasmoresubtleeffects
(Fig. 4d). To control for nonspecific competitive inhibition of
b
ephrinB2 translation, GFP was overexpressed using the same
Flotillin-1-HA – – – a b a b a b a b RNAconcentrations,andthishadnoeffectonephrinB2protein
EphrinB-flag B1 B2 B3 – – B1 B1 B2 B2B3 B3 levels (Supplementary Fig. 2).
40 Flag Since ADAM10 cleaves the juxtamembrane region of ephrin-
(ephrin) A2 (ref. 7), and MMP-8 cleaves the juxtamembrane region of
IP HA
HA ephrinB1(ref.20),wetestedwhetherthisregionoftheephrinB2
55
(flotillin-1) ectodomain is important for loss of ephrinB2 expression in the
absence of flotillin-1 (Fig. 4e). Indeed, juxtamembrane deletion
40
mutants were stably expressed in the presence of F1aMO
Flag
(Fig. 4e,f), and specifically deleting sequences in this region
Lysate
55 (aminoacids182–194and197–218)rendersephrinB2resistantto
HA
the ADAM10-mediated effect (Fig. 4g). These data suggest that
ADAM10 cleaves the juxtamembrane region of the ephrinB2
c
ectodomain.
B1B2-HA EphrinB1 ectodomain EphrinB2 tm-cyto
B2B1-HA EphrinB2 ectodomain EphrinB1 tm-cyto ADAM10 cleaves ephrinB2 and regulates ephrinB2 protein
levels. ADAM10 cleavage of ephrinB2 should result in two
fragments: a 25-kDa amino-terminal fragment and a 15-kDa
EphrinB-HA – B1wt B2wt B1B2 B2B1
carboxyl-terminal fragment (CTF). Without any treatment, the
F F F F
MO – Co 1a Co 1a Co 1a Co 1a B40-kDa full-length ephrinB2-HA protein and two ephrinB2
40 CTF bands of B15kDa were detected in embryonic lysates,
suggestingthatephrinB2iscleavedbyendogenousADAM10and
HA
further processed (Fig. 5a). Since it is known that MMP-8 can
cleave the ephrinB1 ectodomain and that subsequently g-secre-
40 tase cleaves the juxtamembrane region of the intracellular
ErK2
domain20, we speculated that g-secretase may also cleave the
Figure3|TheephrinB2ectodomainisimportantfordecreasedephrinB2 ephrinB2 intracellular domain. This could produce the two
expressionintheabsenceofflotillin-1.(a)DeletionoftheephrinB2 observed CTF bands: an upper band generated by ADAM10
ectodomainstabilizesitsexpressionintheabsenceofflotillin-1.A cleavage,andalowerbandresultingfrombothADAM10andg-
schematicdiagramofephrinB2andmutantsisdisplayed,alongwith secretase cleavage. When ADAM10 was overexpressed, full-
westernblotanalysisofembryosinjectedwithCoMOorF1aMOandwild- length ephrinB2 was reduced and the upper CTF band was
type(WT)ephrinB2,ephrinB2D60orephrinB2TM-CytoRNAs.(b)EphrinB1 increased,indicatingthatADAM10indeedcleavesephrinB2and
aswellasephrinB2areassociatedwithflotillin-1.Flotillin-1a-HAor reduces its expression (Fig. 5a). Conversely, when embryos were
flotillin-1b-HARNAwasinjectedwithorwithoutephrinB1-Flag,ephrinB2- treated with the ADAM10-specific inhibitor GI254023X, full-
FlagorephrinB3-FlagRNA.Flotillin-1wasimmunoprecipitatedwithanti-HA length ephrinB2 was slightly increased and the upper CTF band
antibody,andimmunoblottedwithFlagantibodytodetectephrinBs. wasreduced,withthegreatestdifferencedetectedintheembryos
EphrinB1andephrinB2wereco-immunoprecipitatedwithflotillin-1aand overexpressing ADAM10 (Fig. 5a). In addition, we used co-
flotillin-1b.Proteinexpressionwasconfirmedinwholelysatesbywestern expression and co-IP analysis in embryos to determine whether
blottingwiththeindicatedantibodies.(c)TheephrinB2ectodomainmakes ADAM10 associates with ephrinB2. We found that ephrinB2 is
ephrinB2proteinsusceptibletodecreasedexpressioninthepresence robustly present in ADAM10 immune complexes, but found at
oftheF1aMO.ThedomainstructureofB1B2-HAandB2B1-HAchimeric relatively low levels in ADAM17 immmune complexes,
proteinsisdisplayed,alongwithwesternblotanalysisofembryosinjected supporting the idea that ephrinB2 is specifically cleaved by the
withF1aMOorCoMOandtheindicatedchimericephrinBs.Erk2is ADAM10 metalloprotease (Fig. 5b). Moreover, although
usedasaloadingcontrol. ephrinB1 also associates with ADAM10 in co-IPs from Xenopus
oocytes co-expressing both proteins (Fig. 5c), only ephrinB2 is a
target of ADAM10 cleavage in a dose-dependent manner in
and ephrinB2 are cleaved by ADAM13 (ref. 11). Among embryos (Fig. 5d). In HT-29 cells, endogenous ADAM10 was
identified Xenopus orthologues of ADAM family proteins22, we co-IP’dinephrinB2immunecomplexes(Fig.5e),confirmingthat
selectedADAM10,ADAM13andADAM17totestwhetherthese ephrinB2 and ADAM10 associate either directly or indirectly
ectodomainsheddases22–25maytargetephrinB2forcleavageand in vivo.
lead to the loss of its expression in the absence of flotillin-1. HavingestablishedthatADAM10canassociatewithephrinB2,
Surprisingly, knockdown of ADAM10 prevented a considerable weconsideredwhetherflotillin-1may,inpart,preventADAM10
portion of the loss of ephrinB2 protein that results from the cleavagebyrestrictingtheinteractionbetweenthesetwoproteins.
introduction of the F1aMO, while knockdown of ADAM13 or To address this possibility, we co-expressed ephrinB2 and an
NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications 5
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
a b c d f
InhiRbMiNHtoOAAr ––DMCSoOEphBrFiBn1B-a92M4- HOGAM6001 AADDAAMMF111a00RMM-HNVOOAA5 –––– E–––ph+––rinB+––2-–+H+A+++ 40 ADAFM1aR-MMHNOOAA ––– –– +–Ep–h1r0i+nB2–-1H3A+ –17+ 40ADVA5HMA- ADAEMph1r0inB2A-DHAAM17 4305 EpF-h1HraiAnMBO2 –– –WT+ –Δjuxta+ –Δ168–177+ –Δ182–194+ –Δ197–218+
Erk2 3560 V5 3515000 Erk2 50 ErVk52 14000 HA 4305
Erk2 Erk2 40
e g
EphrinB2-HA – WT Δ168–177 Δ182–194 Δ197–218
ADAM10-V5 – – – – – –
40
HA
V5 100
70
50
Erk2
Figure4|ADAM10isresponsibleforthelossofephrinB2expressionintheabsenceofflotillin-1.(a)Metalloproteasesareresponsibleforthelossof
ephrinB2expressionintheabsenceofflotillin-1.WesternblotanalysisofephrinB2-HAexpressioninthepresenceofCoMOorF1aMOandincreasing
amountsofthebroad-spectrummetalloproteaseinhibitorsBB-94andGM6001,asindicated.(b)ADAM10reducesephrinB2expression,butisinhibitedby
thepresenceofendogenousflotillin-1.WesternblotanalysisofephrinB2-HAinembryosinjectedwithADAM10MOand/orF1aMO,andADAM10-V5
RNAasindicated.Erk2isusedasaloadingcontrol.(c)SpecificknockdownofADAM10rescuesephrinB2lossinthepresenceofF1aMO.Western
blotanalysisofephrinB2-HAinthepresenceofF1aMOaloneorwiththeindicatedADAMMO.Erk2isusedasaloadingcontrol.(d)ADAM10
overexpressionreducesephrinB2expressioninadose-dependentmanner.WesternblotanalysisofembryosexpressingephrinB2-HAwithincreasing
amountsofADAM10-V5orADAM17-V5.(e)EphrinB2amino-acidsequence.BlacklineindicatestheglobularregionoftheephrinB2ectodomainthatis
knowntobindEphreceptors.Theblackboxesindicateaminoacids168–177,182–194and197–218inthejuxtamembraneregionoftheephrinB2
ectodomain,thegreyboxdenotesthetransmembranedomainandthesixasterisksindicatethesixtyrosineresiduesintheintracellulardomainof
ephrinB2.(f)Aminoacids182–214ofephrinB2areimportantforthedecreaseinephrinB2mediatedbyF1MO.WesternblotanalysisofephrinB2mutants
lackingtheindicatedaminoacidsorjuxtamembranedomaininthepresenceorabsenceofF1aMO.(g)Aminoacids182–214ofephrinB2areimportant
forthedecreaseinephrinB2mediatedbyADAM10.WesternblotanalysisofephrinB2mutantslackingtheindicatedaminoacidsorjuxtamembrane
domaininthepresenceorabsenceofADAM10.
ADAM10 point mutant with compromised protease activity (to dynasore alone increased ephrinB2 levels, but to a lesser degree
prevent cleavage of ephrinB2) in Xenopus embryos in the than MG132 alone (Fig. 5h). Of note, the CTFs are similarly
presence of the proteasomal inhibitor MG132 and the F1MO or affectedbythesereagents(Fig.5h).Thesedataindicatethatinthe
CoMO.Co-IPanalysisshowsamarkedincreaseintheADAM10 absence of flotillin-1, full-length ephrinB2 and the ADAM10-
mutant detected in ephrinB2 immune complexes from the cleaved CTFs are targeted for destruction through the lysosomal
embryos with flotillin-1 knocked down when compared with and proteasomal pathways.
thosewithcontrolMO(Fig.5f).Thisresultsuggeststhatflotillin- We also examined whether the observed F1aMO-mediated
1, at least in part, inhibits the association between ephrinB2 and reduction in ephrinB2 protein was due to a decrease in the half-
ADAM10 in vivo. life of ephrinB2 (as would be expected if the protein is being
ToconfirmthatendogenousephrinB2istargetedbyADAM10, degraded), and whether cleavage by ADAM10 promotes this
embryoswereinjectedwithB1MOaloneorwithF1aMOineither activity. For this purpose, we carefully titrated the amount of
the absence or presence of the ADAM10-specific inhibitor. The injected ephrinB2 RNA along with CoMO or F1aMO to yield
dorsal axial region (neural fold) was excised and subjected to roughly equivalent ephrinB2 protein levels when embryos
western blot analysis using the ephrinB antibody. In the absence reached the neurula stage. The embryos were then injected with
of ephrinB1, ephrinB2 levels are detected in the dorsal axial cycloheximide(CHX)orwithbothCHXandADAM10inhibitor,
region with this antibody. As expected, the introduction of an and cultured in CHX to block further protein synthesis for 5h
F1aMOresultsinadramaticreductioninephrinB2proteinlevels, (Fig. 6a,b). Western blot analysis shows that ephrinB2 levels are
and this is partially rescued by treatment with the ADAM10- reduced by half within 153min (B2.5h) after the addition of
specific inhibitor (Fig. 5g). CHX to control MO-containing embryos. In the F1aMO-
SincetheF1MO-mediateddecreaseinephrinB2isquiterobust containing embryos, the half-life of ephrinB2 is reduced to
(Fig.5a,g),weexaminedwhethertherewasadditionalprocessing 57min (B1h; Fig. 6a,b). In contrast, the half-life of ephrinB2 is
via proteasomal or lysosomal degradation as well as the actually extended to 102min (B1.7h) in the F1aMO plus
ADAM10-mediated cleavage that would be expected to remove ADAM10 inhibitor-injected embryos (Fig. 6a,b). These data are
cell surface ephrinB2. Inhibitors of proteasomal degradation consistent with flotillin-1 hindering ephrinB2 degradation, in
(MG132), or dynamin-dependent endocytosis (dynasore) were part, by inhibiting its cleavage by ADAM10.
usedtodeterminetheeffectivepathwaysemployedinresponseto Since flotillin-1 is known to affect trafficking of some
F1MO-induced regulation of ephrinB2 protein levels. Xenopus transmembrane proteins26, we examined whether the reduction
embryos were cultured until the neurula stage after the intheendogenousephrinB2proteinintheabsenceofflotillin-1is
introduction of HA-tagged ephrinB2 RNA alone or along with similarly affected at the cell surface when protein degradation is
F1aMO. These embryos were cultured in the inhibitors for 5h, blocked.XenopusembryoswereinjectedwithB1MOandtreated
and then western blot analysis was performed. As expected, with MG132 to prevent ephrinB2 degradation. A group of these
flotillin-1 knockdown resulted in a dramatic loss of ephrinB2 embryoswasalsoinjectedwithF1aMOinthepresenceorabsence
protein in the embryonic lysates (Fig. 5h). Treatment with both oftheADAM10-specificinhibitorthatpreventscleavageandthus
MG132 and dynasore efficiently rescued ephrinB2 levels, while loss from the cell surface. The dorsal axial region was excised at
6 NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
a
b
GI254023X – – + – + ADAM-HA – – 101710 17
EphrinB2-HA – + + + + EphrinB2-flag – + – – + +
50
ADAM10-V5 – – – + + Flag
40 148
25 HA IP HA 98
15 CTFs 64 HA
100 V5
70 50
50
Erk2 Flag
Lysate 148
98 HA
c
64
ADAM10-V5 – – – + + +
EphrinB-HA – B1B2 – B1B2
100 V5 d
IP HA
EphrinB-HA – B1 B2
40 HA ADAM10-V5 – – –
40 HA HA 40
IP V5
100 V5
40 HA V5 100
Lysate
100 V5 Erk2 40
e f
E MO – Co F1a
IP Control phrinB2 ADAEMp1h0ri(nPBD2)--HVA5 –– –+ +– ++ +– –+ ++
100 V5
100 IP HA
ADAM10
HA
70
35
IP 55
EphrinB2 100 V5
100 Lysate HA
ADAM10 35
70
40 Erk2
Lysate 55
EphrinB2
h EphrinB2-HA
F1aMO
Inhibitor – – –M DM/D
40
g
25
Low exp
GI254023X – – – + –
15
B1MO – + + + + HA
MO Co – F1a F1a B2 Non-specific 4205 High exp
40
EphrinB
15
40 Erk2 40 Erk2
Figure5|ADAM10cleavesephrinB2andregulatesephrinB2proteinlevels.(a)ADAM10isresponsibleforsheddingoftheephrinB2ectodomain.
WesternblotanalysisofembryosexpressingephrinB2-HAwithADAM10-V5and/ortheADAM10-specificinhibitorGI254023X.Thefull-lengthephrinB2-
HAandtheC-terminalfragments(CTFs)ofephrinB2-HAareindicated.ADAM10-V5isindicated,asistheErk2loadingcontrol.(b)EphrinB2associates
withADAM10.ADAM10-HAimmunoprecipitationandwesternblotanalysisfromembryoniclysatesco-expressingephrinB2-FlagandADAM10-HA
orADAM17-HA.Directlysatesareprobedwithanti-FlagorHAasindicated.(c)ADAM10associateswithephrinB1aswellasephrinB2.Westernblot
analysisoftheindicatedIPsordirectlysatesfromoocytesexpressingephrinB2-HAorephrinB1-HAaloneorwithADAM10-V5.(d)ADAM10specifically
targetsephrinB2.WesternblotanalysiswithindicatedantibodiesoflysatesfromembryosexogenouslyexpressingincreasingamountsofADAM10-V5
alongwithephrinB1-HAorephrinB2-HA.(e)EndogenousADAM10associateswithephrinB2.EphrinB2wasimmunoprecipitatedfromHT-29cellsand
westernblotanalysiswasperformedusinganti-ephrinB2orADAM10antibodies.EphrinB2andADAM10expressionlevelsinHT-29celllysatesare
shown.(f)InthepresenceofMG132,F1aMOleadstoincreasedassociationbetweenexogenouslyexpressedephrinB2andADAM10.Westernblotanalysis
oftheHA(ephrinB2)IPsanddirectlysatesfromembryosco-expressingephrinB2-HAandanADAM10mutantwithcompromisedproteaseactivity
(ADAM10PD-V5)andinjectedwiththeindicatedMOs.(g)EndogenousephrinB2iscleavedanddegradedinthepresenceofF1aMO,butispartially
rescuedbyanADAM10-specificinhibitor.WesternblotanalysisoflysatesfromneuralfoldsofembryosinjectedwiththeindicatedMOs,andtheADAM10
inhibitorGI254023X(stage9injectionsintotheblastocoel).WesternblotswereprobedusingC-18(pan-ephrinB)andERK2antibodies.(h)EphrinB2is
degradedbyboththeproteosomalanddynamin-dependentdegradationpathwaysintheabsenceofflotillin-1.Westernblotanalysiswiththe
indicatedantibodiesofembryosinjectedwithephrinB2-HARNAaloneorwithF1aMO,andtreatedwithvehiclecontroldimethylsulphoxide((cid:2)),or
MG132(M),and/ordynamininhibitordynasore(D).Erk2isusedasaloadingcontrol.
NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications 7
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
b CoMO
a
CoMO+CHX
5450 F1aMO+CHX
35 F1aMO+CHX+GI254023X
HA
100
15
t =153±4 min
1/2
O 40 Erk2 %)
M n (
Co 345505 protei 10
HA 2
B
15 n t =102±8 min
hri 1/2
p
40 Erk2 E
t =57±8 min
55 1 1/2
40
35 0 0.5 1 2 3 4 5 h CHX
X O HA
H M c
C a
+ F1 15 GI254023X – – – – – + –
MG132 – – – + + + +
40 Erk2 B1MO – – + + + + +
55 MO Co – – F1a B2
40
40
35
IP biotin EphrinB
X HA
O+023 15 40 Non-specific
FaMGI254 40 Erk2 Lysate ephrinB
40 Erk2
0 0.51 2 3 4 5
No
biotin
Figure6|EphrinB2proteinhalf-lifeisdecreasedintheabsenceofflotillin-1andpartiallyrescuedbytheADAM10inhibitor.(a)Two-cell-stage
embryoswereinjectedwithcarefullytitratedephrinB2-HARNAalongwithcontrolMOorF1aMOtoyieldroughlyequivalentephrinB2proteinlevelswhen
embryosreachedtheearlyneurulastages.Agroupoftheinjectedembryoswassubjectedtoasecondaryinjectionintotheblastocoelatstage9with
ADAM10inhibitor(10nlof1mM),andthearchenteroncavity(stage14)withCHX(10nlof75mgml(cid:2)1),andexternallyincubatedinCHX(7.5mgml(cid:2)1)to
blockfurtherproteinsynthesisfortheindicatedtimes.WesternblotanalysiswasperformedontheembryoniclysatesusingHAantibodiesorErk2
antibodies(asaloadingcontrol).(b)AgraphofthemeanbandintensitiesasmeasuredbyImageJsoftwareshowstheapproximatehalf-livesinthe
presenceofCHXandtheindicatedMOsandADAM10inhibitor.TheephrinB2C-terminalfragments(shortarrow)andthefull-lengthprotein(longarrow)
areindicated.Thesedataaretheresultofthreeindependentexperimentsand±representss.d.(c)EndogenoussurfaceephrinB2levelsarereduced
byknockdownofflotillin-1,butprominentlyrescuedbytheADAM10inhibitor.EmbryoswereinjectedwiththeindicatedMOsandinhibitors.Neuralfolds
wereexcisedandleftnon-biotinylated(lane1)orbiotinylated.Lysateswerepreparedandcellsurfaceproteinsimmunoprecipitatedwithstreptavidin-
conjugatedsepharose.Westernblotanalysisofbiotin-labelledcellsurfaceproteinswasperformedusinganti-ephrinBantibody.Directwesternblot
analysisofneuralfoldlysateswereprobedforephrinB.Erk2expressionisshownasaloadingcontrol.
stage 16–18 and the cell surface proteins were biotinylated. MO-mediated knockdown of ephrinB1 did not disrupt neural
Immunoprecipitation of cell surface proteins from lysates of the tube closure (Fig. 7e). Further evidence of the specificity of the
explantswasperformedusingstreptavidin-conjugatedsepharose, B2MOandtherequirementforephrinB2inneuraltubeclosureis
and western blotting with ephrinB antibody. As expected, use of providedbytherescueofneuraltubedefects(59.6%closedneural
both B1MO and B2MO removes nearly all the detectable tubes) by re-expressing ephrinB2 using B211MT RNA, which is
ephrinBs in the dorsal axial region (Fig. 6c). In the presence of resistant to the B2MO (Fig. 7a,b).
MG132,theF1aMOcausesaprominentreductioninbiotinylated Having established that knockdown of flotillin-1 in embryos
(cell surface) ephrinB2 protein, and this is partially, but leadstoadramaticreductioninephrinB2proteinlevelsthatisat
considerably rescued in the presence of the ADAM10 inhibitor least partly dependent on cleavage via ADAM10, we decided to
(Fig. 6c). These data indicate that loss of flotillin-1 considerably test whether knockdown of flotillin-1 also affected neural tube
reduces endogenous cell surface ephrinB2 protein levels by closure. Both F1aMO and F1bMO caused neural tube closure
promoting ADAM10-mediated cleavage. defects that are similar to ephrinB2 morphant phenotypes
(Fig. 7a,b). Only 5.7 and 4.6% of F1aMO- or F1bMO-injected
stage18/19embryosdisplayedclosedneuraltubes,whileflotillin-
EphrinB2 and flotillin-1 morphants show neural tube defects. 2 morphants showed normal neural tube closure (Fig. 7e).
To assess whether there is a functional effect mediated by the Importantly, these defects were rescued byexogenous flotillin-1a
interaction between flotillin-1 and ephrinB2, we examined the or -1b expression (81.0% by F1aDUTR-Flag and 87.7% by
effect of knocking down ephrinB2 in Xenopus embryos, which F1bDUTR-Flag, respectively). To test whether there was redun-
expressephrinB2 atearly stages of neural tubeformation.When dancy between the a and b paralogues of flotillin-1, RNA
embryos were injected with the B2MO, only 10.2% displayed a encoding either paralogue was expressed along with exogenous
closed neural tube by stage 18/19 (Fig. 7a,b), while 96.5% of the ephrinB2 in F1aMO-injected embryos. Western analysis shows
CoMO-injected embryos had closed neural tubes (Fig. 7a,b). At that each paralogue can rescue ephrinB2 protein levels in the
laterstages,profounddelaysorfailureofneuraltubeclosurewere embryos injected with F1aMO (Supplementary Fig. 3a), and a
observed (Fig. 7c,d). This effect is specific to the B2MO, since partial, but sizeable rescue of neural tube closure is observed
8 NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
(SupplementaryFig.3b)andevidencedbyaclearreductioninthe neural tube. This process is driven by three overlapping events.
gap width between the neural folds (Supplementary Fig. 3c). First, convergence and extensionmovements elongatetheneural
These data suggest that the isoforms have at least partially plate28. Second, apical constriction of the neuroepithelial cells
overlapping functions. within the neural plate brings the sides of the plate closer
Neuraltubeformationisacomplexprocess27thattransformsa together. Third, the lateral edges of the plate elevate to make
flat oval sheet of thickened epithelium (the neural plate) into a neural folds. The neural folds move towards the dorsal midline,
where they meet and fuse to form a tubular structure, finalizing
the separation between the future epidermis and the neural
epithelium.
a
CoMO B2MO B2MO+B211MT
Flotillin-1 and ephrinB2 morphants fail to apically constrict.
CoMO-injectedembryosandB2MO-orF1MO-injectedembryos
appear to form normal neural plates until stage 14. Strikingly,
after this stage we observed that ephrinB2 and flotillin-1 mor-
CoMO F1aMO F1aMO+F1aΔUTR phants develop broad neural grooves and lack elevated neural
folds (Fig. 7a–d), indicating a possible problem in the process of
neural plate bending, which is driven by apical constriction29.
WhenonesideoftheembryoisinjectedwithB2MOorF1bMO,
it displays apical constriction defects of the neuroepithelium,
CoMO F1bMO F1bMO+F1bΔUTR whiletheuninjectedsideretainsnormalmorphologyandwedge-
shapedcells(Fig.8a).Celllengthsaresignificantlydecreasedand
cell widths are increased on the B2MO- or F1bMO-injected side
whencomparedwiththeuninjectedside.Theaverageratioofthe
length of the apical surface to the cell perimeter was calculated
using five different cells adjacent to the midline. The ratio
b increased for cells harbouring the B2MO or the F1bMO
120
compared with uninjected cells (Fig. 8b), indicating that loss of
e 100 n=140 n=113 n=127 ephrinB2 or flotillin-1b reduces apical constriction in the
sur neuroepithelium.
clo 80 n=92 Apical constriction results from the contraction of apically
e
ub 60 localizedfilamentousactin(F-actin),drivenbythemotorprotein
al t myosin28,29. Therefore, we examined F-actin intensities in
eur 40 neuroepithelium in the presence or absence of ephrinB2 or
N
% 20 n=112 n=104 n=113 flotillin-1. B2MO, F1aMO or F1bMO was injected along with
GFPRNA(asatracer)intoonesideoftheembryos,andTexas-
0 Red-conjugated phalloidin staining was performed to compare
CoMO B2MO B2MO F1aMO F1aMO F1bMO F1bMO
+B211MT +F1aΔUTR +F1bΔUTR F-actin levels on the uninjected control side with those on the
MO-injected side. Injection of each MO (B2MO, F1aMO or
c
CoMO B2MO F1aMO F1bMO F1bMO)causedabroadeningoftheneuralplateandasignificant
decreaseintotalactinintensityintheneuralfold(Fig.8c,d).The
a decrease in intensity within individual cells can also be observed
urulage (Fig. 8c, insets). The B2MO as well as the F1aMO or F1bMO
Nest notably reduced the total actin intensity when compared to the
uninjectedcontrolsideoftheneuralplate(Fig.8d).ThisB2MO-
e
adpolstage
T Figure7|EphrinB2andflotillin-1morphantsshowneuraltubeclosure
d 120 defects.(a)NeuraltubeclosuredefectsinephrinB2orflotillin-1morphants.
n=65n=61 Early stage Dorsalviewofstage18control,ephrinB2,flotillin-1aorflotillin-1b
ure 100 Late stage MO-injectedembryos,andthoseco-injectedwithmorpholino-resistant
os 80 RNA(B211MT-HAforB2MO,F1aDUTR-FlagforF1aMOandF1bDUTR-Flagfor
e cl F1bMO);leftisanteriorandrightisposterior.(b)Quantificationofneural
ub 60 n=25 tubeclosuredefectsinephrinB2orflotillin-1morphants.Embryos
ural t 40 n=42 n=54 showingasingledorsallinewerecountedashavingclosedneuraltubes.
Ne n=46 Atleastthreeindependentexperimentswereperformedanderrorbars
% 20 n=55 n=25 indicates.d.(c)B2MO,F1aMO,orF1bMOwereinjectedintoembryos,and
neuraltubeclosurewasexaminedatneurula(stage18/19)andswimming
0
CoMO B2MO F1aMO F1bMO tadpolestages.Notetheanteriorneuraltubedefects.(d)Percentageof
e neuraltubeclosureinB2MO-,F1aMO-,orF1bMO-injectedembryos.Note
CoMO B1MO F2MO
thatthepercentageofneuraltubeclosureissomewhatelevatedatlater
stagesinB2MO-,F1aMO-andF1bMO-injectedembryos,indicatingthat
someoftheembryosdisplayedaprofounddelayinneuraltubeclosure,
whileothersretainanopenneuraltube.Errorbarsindicates.d.(e)
EphrinB1-MO-orflotillin-2MO-injectedembryosshownormalneuraltube
100% (n=24) 100% (n=32) 100% (n=29)
closure.
NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications 9
&2014MacmillanPublishersLimited.Allrightsreserved.
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms4516
a b d Uninjected side Injected side
P=0.47
P<0.01 P>0.5 P<0.05 P>0.5
120 P<0.05
B2MO F1bMO
* ** * ** Apical surfaceper perimeter 00000.....43215 UInnjeincjteecdt esdid seide % Total actin intensityin neuroectoderm 10864200000
0 0
B2MO F1bMO B2MO B2MO F1bMO F1bMO F1aMO F1aMO
+B211MT +F1bΔUTR +F1aΔUTR
c
GFP/actin Actin Actin Un-injected side MO-Injected side
B
2
M
O
B
2
M
O
+
B
2
F
1
b
M
O
F
1
b
M
O
+
F
1
b
UΔ
TR
F
1
a
M
O
F
1
a
M
O
+
F
1
a
UΔ
TR
Figure8|Neuroepithelialcellsofflotillin-1andephrinB2morphantsfailtoapicallyconstrictinthepresumptiveneuraltube.(a)Apicalconstriction
defectsofephrinB2andflotillin-1morphants.B2MOorF1bMOwereinjectedintoonesideoftheembryosalongwithGFPRNAasatracer.Atthe
neurulastage,embryoswerestainedwithanti-tubulin(green)antibodytooutlinethecellsandwithanti-GFP(red)antibodytolabeltheinjectedside.Cell
shapesareoutlinedbelow.Blue,uninjectedcells;red,MO-injectedcells.Scalebar,50mm.(b)Theaverageratiosoftheapicalsurfacestotheperimeters
amongfivedifferentcellsoneachsideofthemidlinewerecalculatedforfivedifferentembryos.Bluebars,uninjectedside;redbars,MO-injectedside.
(c)DecreasedactinintensityinephrinB2orflotillin-1morphants.EachembryowasinjectedwiththeindicatedMOandGFPRNAwithorwithoutthe
appropriaterescueRNAasindicated.GFP(green)showstheinjectedsideandactinisstainedred(phalloidinstaining).Inthethirdcolumn,the
neuroepitheliumisoutlinedingreenontheMO-injectedsideandredontheuninjectedside.Redboxedregionsfromembryosinthesecondcolumnare
presentedasenlargedimagesinthefourthandfifthcolumns.Scalebar,300mm(horizontal);50mm(vertical).(d)Totalactinintensityisdecreased
inephrinB2orflotillin-1morphants.Percenttotalactinintensitieswerecalculatedfortheuninjectedorinjectedsideoftheembryousingthe
methodologydescribedintheMethodssection.Atwo-tailedt-testwasusedtogeneratethePvalue.Theseresultsrepresentthreeindependent
experiments.Errorbarsrepresents.d.
or F1MO-mediated decrease in F-actin intensity levels can be constriction in naive ectodermal explants as evidenced by actin
rescued by the introduction of B211MT-HA or F1DUTR-Flag staining or membrane-bound GFP visualization (Supplementary
RNAs, respectively. These data are reflective and consistent with Fig. 4). Taken together, these data indicate that ephrinB2 and
theexperimentsshowingthattheapicalsurfaceoftheneuralfold flotillin-1 are required for a normal actin cytoskeleton in
cells are increased (Fig. 8a,b). We found that neither ephrinB2 neuroepithelial cells, and the loss of either protein leads to
nor flotillin-1 overexpression was capable of inducing apical failure of apical constriction.
10 NATURECOMMUNICATIONS|5:3516|DOI:10.1038/ncomms4516|www.nature.com/naturecommunications
&2014MacmillanPublishersLimited.Allrightsreserved.