kermit 2/xgipc, an igf1 receptor interacting protein, is required … · (tan et al., 2001), that...

DEVELO

PMENT

3651RESEARCH ARTICLE

INTRODUCTIONGIPC (GAIP interacting protein, C terminus) is a PDZ-domain-containing protein initially identified by virtue of its interaction withRGS-GAIP (regulator of G protein signaling-GTPase activatingprotein for G�i) (De Vries et al., 1998b). The list of identifiedbinding partners for GIPC includes numerous membrane proteins(Katoh, 2002), such as the transmembrane semaphorin M-SemF(Wang et al., 1999), neuropilin 1 (Cai and Reed, 1999), the Wntreceptor frizzled 3 (Tan et al., 2001), the �1-adrenergic receptor (Huet al., 2003), the TGF�-type III receptor (Blobe et al., 2001),melanosomal membrane protein gp75 (tyrosinase-related protein 1)(Liu et al., 2001), the tyrosine kinase receptor TrkA (Lou et al.,2001), intergrin � subunits (El Mourabit et al., 2002; Tani andMercurio, 2001), insulin-like growth factor 1 receptor (IGF1R)(Booth et al., 2002; Ligensa et al., 2001), human lutropin receptor(Hirakawa et al., 2003), myosin VI (Dance et al., 2004; Hasson,2003), dopamine D2 and D3 receptors (Jeanneteau et al., 2004a;Jeanneteau et al., 2004b), a myeloid cell-surface marker CD93(Bohlson et al., 2005) and human papillomavirus type 18 E6 protein(Favre-Bonvin et al., 2005). Although the list of GIPC-bindingpartners is long, the endogenous function of GIPC and thephysiological significance of these associations are much lessstudied. Loss-of-function studies published so far have beenprimarily in ex vivo or in vitro systems, such as Xenopus ectodermalexplants (Tan et al., 2001) and cultured cell lines (Favre-Bonvin etal., 2005; Hirakawa et al., 2003); the in vivo role of GIPC inembryonic development has not yet been examined.

Previously, we identified a Xenopus homolog of GIPC, kermit(Tan et al., 2001), that is 72% identical to mammalian GIPC (DeVries et al., 1998b). Knockdown of kermit using antisense

morpholino oligonucleotides blocked neural crest induction byXenopus frizzled 3 in ectodermal (animal cap) explants (Tan et al.,2001), but did not inhibit neural crest formation in whole embryos.We reasoned that this negative result may indicate the presence of acompensating or redundant activity in whole embryos that is absentin animal cap explants. We therefore identified a second kermit gene(67% amino acid identity with kermit 1) in Xenopus early embryosto study the redundancy between kermit and kermit2. Unexpectedly,we discovered a novel function for kermit 2 in IGF signaling thatdoes not overlap with kermit 1.

Kermit 2 is identical to XGIPC, identified in a yeast two hybridscreen for IGF1 receptor binding proteins in Xenopus oocytes(Booth et al., 2002). XGIPC/kermit 2 binds to the cytoplasmicdomain of the Xenopus IGF1R and this interaction appears to requirethe PDZ domain of XGIPC. Overexpression of C-terminaltruncation mutants of XGIPC that retain the PDZ domain blocksinsulin-induced Xenopus MAP kinase activation and oocytematuration. Human GIPC was also identified as a binding partnerfor human IGF1R (Ligensa et al., 2001).

IGF signaling has been recently implicated in neural inductionin Xenopus and zebrafish embryos (Eivers et al., 2004; Pera et al.,2003; Pera et al., 2001; Richard-Parpaillon et al., 2002). Ectopicexpression of IGF in dorsal cells leads to the induction of ectopiceyes and eye expansion and, when expressed in ventral cells,induces secondary head-like structures (Pera et al., 2001; Richard-Parpaillon et al., 2002). Inhibition of IGF signaling by eitherdominant-negative IGF1R or IGF1R depletion reduces headstructures (Pera et al., 2001; Richard-Parpaillon et al., 2002). Askermit 2/XGIPC physically interacts with XIGF1R, we examinedthe involvement of kermit 2/XGIPC in IGF signaling in Xenopusembryonic development.

We report here that kermit2/XGIPC is expressed throughoutXenopus early embryonic development and is localized to theanterior region during neurula stage in a pattern highly similar toXIGF1R (Richard-Parpaillon et al., 2002). Knockdown of kermit 2leads to embryos with reduced anterior structures, specificallyreduction in the presumptive eye region. Furthermore, depletionof kermit 2 and expression of dominant-negative XIGF1R

Kermit 2/XGIPC, an IGF1 receptor interacting protein, isrequired for IGF signaling in Xenopus eye developmentJinling Wu1, Michael O’Donnell2, Aaron D. Gitler1 and Peter S. Klein1,2,*

GIPC is a PDZ-domain-containing protein identified in vertebrate and invertebrate organisms through its interaction with a varietyof binding partners including many membrane proteins. Despite the multiple reports identifying GIPC, its endogenous function andthe physiological significance of these interactions are much less studied. We have previously identified the Xenopus GIPC homologkermit as a frizzled 3 interacting protein that is required for frizzled 3 induction of neural crest in ectodermal explants. Weidentified a second Xenopus GIPC homolog, named kermit 2 (also recently described as an IGF receptor interacting protein andnamed XGIPC). Despite its high amino acid similarity with kermit, kermit 2/XGIPC has a distinct function in Xenopus embryos. Loss-of-function analysis indicates that kermit 2/XGIPC is specifically required for Xenopus eye development. Kermit 2/XGIPC functionsdownstream of IGF in eye formation and is required for maintaining IGF-induced AKT activation. A constitutively active PI3 kinasepartially rescues the Kermit 2/XGIPC loss-of-function phenotype. Our results provide the first in vivo loss of function analysis of GIPCin embryonic development and also indicate that kermit 2/XGIPC is a novel component of the IGF pathway, potentially functioningthrough modulation of the IGF1 receptor.

KEY WORDS: Kermit 2, GIPC, Insulin like growth factor (IGF), IGF1 receptor (IGF1R), Eye development, Xenopus

Development 133, 3651-3660 (2006) doi:10.1242/dev.02547

1Cell and Molecular Biology Graduate Group, University of Pennsylvania School ofMedicine, 364 Clinical Research Building, 415 Curie Blvd, Philadelphia, PA 19104,USA. 2Department of Medicine (Hematology-Oncology), University of PennsylvaniaSchool of Medicine, 364 Clinical Research Building, 415 Curie Blvd, Philadelphia,PA 19104, USA.

*Author for correspondence (e-mail: [email protected])

Accepted 18 July 2006

DEVELO

PMENT

3652

synergistically inhibit eye development. Kermit 2 is required forIGF1 induced eye formation in whole embryos and for the inductionof eye molecular markers in ectodermal explants. Finally, we presentevidence that kermit 2 is required to maintain IGF/PI3 kinase-dependent activation of AKT. These results indicate that kermit2/XGIPC is required for IGF signaling in Xenopus eye developmentlikely through its interaction with the IGF1 receptor.

MATERIALS AND METHODSCloning of kermit2, plasmid constructs and in vitro transcriptionkermit2 was identified through a Blast search of the Xenopus EST databaseusing the kermit sequence (Tan et al., 2001). The GenBank accessionnumber for kermit2 is BC089139. kermit2 5�UTR sequence was identifiedfrom a stage 10 Xenopus 5�RACE cDNA library. Capped mRNAs weresynthesized by in vitro transcription of plasmids encoding XIGF1R(Richard-Parpaillon et al., 2002), DN-IGFR (Pera et al., 2001), XIGF1(Pera et al., 2001), p110* (Carballada et al., 2001), kermit 2/pCS2, kermit2-GFP/pCS2 and 5�UTR-kermit 2-GFP/pCS2 (described below) using SP6mMessage mMachine kits (Ambion). Kermit 2/pCS2 was cloned byRT-PCR amplifying stage 1 Xenopus RNA using primers (forward,AGGGATCCATGCCTCTGGGATTGCGCGTA; reverse, AGTCTAGAT-TAAAAGCGTCCTTGTTTAGC) with engineered BamHI and XbaI sitesand cloned directionally into pCS2+. Kermit 2-GFP/pCS2 was generatedby PCR using kermit 2/pCS2 as the template and cloned intoBamHI/EcoRI sites of pCS2-EGFP. 5�UTR-kermit 2-GFP/pCS2 wascloned by PCR amplifying using primers (upstream, CGGGATC-CCAGGACTAAGGAACAGGAGGCAGG; downstream, AGGAA-TTCAAAGCGTCCTTGTTTAGCATC) with engineered BamHI andEcoRI sites and cloned directionally into pCS2-EGFP.

RT-PCRRNA isolation and RT-PCR methods are described elsewhere (Yang et al.,2002). Primers for EF-1� (Yang et al., 2002), ODC (Yang et al., 2002),Otx2 (Cox and Hemmati-Brivanlou, 1995), En2 (Hemmati-Brivanlou etal., 1994), muscle actin (Hemmati-Brivanlou and Melton, 1994), XAG(Blitz and Cho, 1995) and Xrx (Mathers et al., 1997) have been describedpreviously. Primers were designed to detect kermit2 (upstream, ATGC-CTCTGGGATTGCGCGTAAAG; downstream, TTTAACTTTGTCA-ATCGTGCTCTC), HoxD1 (upstream, CACTTCTTGCGGGGATGTTT;downstream, AGAGTCCTGTAGCTCAGCTG), Pax6 (by N. Hirsch andW. Harris; upstream, AGTGTCCTCATTCACATCG; downstream, AG-TACTGAGACATGTCAGG) and Vent1 (upstream, GCATCTCCTTG-GCATATTTGG; downstream, TTCCCTTCAGCATGGTTCAAC).

Lineage tracing and in situ hybridizationEmbryos were co-injected with RNA encoding �-galactosidase with anuclear localization motif and control morpholino or kermit 2 morpholino(5�-AGAGGCATCTTTCTTTCAGCGAAGG-3�). �-Galactosidase activitywas visualized in embryos with Red-Gal (Research organics). Whole-mountin situ hybridization was performed as described (Deardorff et al., 1998).Antisense probes detected chordin (Sasai et al., 1994), gooscoid (Blumberget al., 1991), Bf1 (Papalopulu and Kintner, 1996), Otx2 (Lamb et al., 1993),Pax6 (Hirsch and Harris, 1997) and Xrx (Mathers et al., 1997). For senseprobes, kermit 2/pCS2 was linearized by NotI and transcribed by Sp6polymerase. For antisense probes, kermit 2/pCS2 was linearized by BamHIand transcribed by T7 polymerase.

Embryos, microinjection, immunoprecipitation andimmunoblottingEggs were obtained from Xenopus females, in vitro fertilized andmicroinjected as described (Deardorff et al., 1998). Each injected blastomerereceived 10 nl of RNA or morpholino. For unilateral injections, morpholinoand/or mRNA were injected into one dorsal blastomere at the four-cell stage.Animal pole explants assays were performed as described previously(Deardorff et al., 2001). For immunoprecipitation, Xenopus embryos wereinjected in the animal pole at the one-cell stage, cultured to stage 10 andlysed in embryo lysis buffer [20 mM Tris (pH 7.5), 140 mM NaCl, 10%glycerol, 1 mM DTT, 2 mM sodium vanadate, 25 mM NaF, 1% Nonidet P-

40 and protease inhibitor cocktail for mammalian cells (Sigma)]. Anti-GFPmonoclonal antibody (1 �l) (Roche, 1 814 460) was incubated with 250 �lof cleared lysate overnight at 4°C, collected on 30 �l protein G agarosebeads (Invitrogen, 15920-010) for 1 hour, washed three times with PBS, andanalyzed by western blot using IGF1R antibody (Cell Signaling, 3022).AKT, P-AKT, MAPK, P-MAPK and �-tubulin antibodies are from Sigma(p2482), Cell Signaling (9271), Cell Signaling (9102), Sigma (M9692) andBD Biosciences (556321), respectively.

TUNEL assay and phosphorylated histone H3 stainingPhosphorylated histone H3 staining was performed as described (Bellmeyeret al., 2003). TUNEL staining protocol was from www.xenbase.org

RESULTSExpression of kermit2/XGIPC in Xenopus embryosTo characterize the temporal pattern of kermit2 expression, weextracted RNA from multiple stages of embryogenesis andperformed RT-PCR. kermit2 was detected maternally (four-cellembryos) and throughout early embryonic development (Fig. 1A).kermit2 is ubiquitously expressed in early and mid-neurula stageembryos (Fig. 1B and data not shown). As neurulation progresses,kermit2 expression is restricted to the anterior region (Fig. 1B,D),including the cement gland and neural plate border, adjacent toand/or overlapping the presumptive eye region (Fig. 1D). Theexpression pattern of kermit2 is very similar to XIGF1R at this stage(Richard-Parpaillon et al., 2002). By tailbud stages, kermit2expression becomes further restricted, with strong expression in thecement gland, otic vesicles (Fig. 1E), pronephros, branchial arches(Fig. 1F), and cranial nerves V, VII and IX (data not shown).

Kermit2/XGIPC is required for eye developmentTo study the endogenous function of kermit 2/XGIPC in Xenopusembryos, we depleted kermit 2 using morpholino antisenseoligonucleotides. The kermit 2 morpholino blocked translation ofkermit 2-GFP mRNA containing the endogenous 5�UTR but had noeffect on kermit 2-GFP mRNA lacking the 5�UTR (Fig. 2A).Unilateral injection (into one dorsal blastomere at the four-cell stage)of kermit 2 morpholino strongly inhibited anterior development,especially eye development, in a dose-dependent manner (Fig. 2B;data not shown). Although there was a distribution of phenotypes,60% (n=38) of embryos receiving 40 ng of kermit 2 morpholinodeveloped without any detectable eyes on the injected side and 32%had small eyes (Fig. 2B). Co-injection of kermit2-GFP mRNAlacking the morpholino target sequence restored eye formation in72% (n=43) of kermit2-depleted embryos, and of these, 28%appeared completely normal (Fig. 2B). kermit2-GFP mRNA alonedid not cause apparent developmental defects. In addition, bilateralinjection of kermit 2 morpholino caused absent or miniscule eyes in83% (n=73) of embryos, and this effect was reversed in 60% (n=85)of embryos co-injected with the kermit2-GFP rescuing mRNA,although bilateral injections were also associated with more severetrunk defects that were not completely rescued by kermit2-GFP(data not shown). These results suggest that the reduced eyeformation phenotype is specifically due to loss of kermit 2.

To characterize further the kermit 2 loss-of-function phenotype,we analyzed molecular markers at various stages by in situhybridization. Kermit 2 depletion disrupted anterior development,which could indicate direct inhibition of anterior neural developmentor could alternatively indicate mild ventralization. To examinewhether kermit 2 plays a role in dorsal development, we examinedthe expression of the dorsal organizer genes chordin and goosecoid.Kermit 2 or control morpholinos were injected into two dorsalanimal blastomeres of four-cell/eight-cell embryos, and chordin or

RESEARCH ARTICLE Development 133 (18)

DEVELO

PMENT

goosecoid expression was assessed at the early gastrula stage (Fig.3A-D). Expression of chordin and goosecoid was not affected bydepletion of kermit 2, suggesting that loss of kermit 2 does notdisrupt early dorsal specification. We then examined the expressionof the anterior neural markers Bf1 (a forebrain marker), Otx2(expressed in the forebrain, eyes and anterior midbrain) and Pax6 (aforebrain and eye marker) at the neurula stage (stage 20). To oursurprise, unilateral injection of kermit 2 morpholino intodorsoanimal blastomeres did not obviously affect Bf1 expression(80% no change, n=45, Fig. 3F). Furthermore Pax6 and Otx2

expression were reduced only in the presumptive eye domain (Pax6,93% eye reduction, n=75; Otx2, 81% eye reduction, n=32, Fig.3H,J). Pax6 expression in the spinal cord was not affected.Consistent with an apparently eye-specific phenotype, theexpression of the eye-specific markers Xrx (85% reduction, n=32,Fig. 3L) and Xath5 (data not shown) was also strongly reduced bykermit 2 depletion. This eye-specific phenotype is probably due toloss of kermit 2, as co-injection of kermit2 mRNA lacking the5�UTR partially recovered Xrx expression in 73% (n=26, Fig. 3S,compared with 3R and 3L) of kermit 2-depleted embryos. We alsoanalyzed the expression of Xrx and Pax6 at earlier stages ofdevelopment (early neurula stage, stage 14) and found that depletionof kermit 2 did not have an obvious effect on Xrx (74% no change,n=49, Fig. 3N) or Pax6 (84% no change, n=50, Fig. 3P) expressionat this stage, suggesting that kermit 2 is required for themaintenance, but not the initiation, of eye formation.

Interaction between kermit2/XGIPC and XIGF1R ineye developmentKermit 2, despite its high amino acid similarity with Kermit, doesnot bind to the Kermit interaction partner, frizzled 3 (unpublisheddata). However, kermit 2/XGIPC was identified as an XIGF1R-

3653RESEARCH ARTICLEKermit 2/XGIPC in IGF-dependent eye development

Fig. 1. Temporal and spatial pattern of kermit2 expression. (A) RT-PCR analysis of kermit2 temporal expression. Total RNA was isolatedfrom embryos at different stages and analyzed by RT-PCR. –RT, negativecontrol without reverse transcriptase. ODC was used as a loadingcontrol. (B) RT-PCR analysis of kermit2 spatial expression. Stage 16embryos were dissected into dorsal, ventral, anterior and posteriorpieces, as shown in the diagram. Stage 20 embryos were dissected intoanterior and posterior pieces. EF1� was used as a loading control.kermit2 is equally expressed in anterior, posterior, dorsal and ventralregions at mid-neurula stage (stage16), but is localized to the anteriorregion by stage 20. Anterior marker Otx2, posterior dorsal markerHoxD1, and posterior ventral marker Vent-1 were used as positivecontrols. (C-F) In situ hybridization of neurula and tadpole stageembryos. (C,D) Stage 20, anterior view, dorsal towards the top. (C)Sense control; (D) antisense probe. At stage 20, kermit2 is expressed inthe cement gland and neural plate border, which is adjacent to and/oroverlapping the presumptive eye region. (E,F) Antisense probe. Bytailbud stages, kermit2 expression becomes further restricted, stronglyexpressed in the cement gland and otic vesicles (ov: arrow) at stage 26(E), and is also expressed in the pronephros and branchial arches instage 33 tadpoles (F).

Fig. 2. Kermit 2 loss-of-function phenotype. (A) A kermit 2-directedantisense morpholino blocks translation of kermit 2-GFP mRNA (5�UTR-kermit 2-GFP, which contains the morpholino target sequence in the5�UTR), but not kermit 2-GFP mRNA lacking the 5�UTR. Kermit 2-GFPmRNA (1 ng) with or without the 5�UTR was injected into one-cellembryos together with kermit 2 morpholino (K2M; 20 ng). Embryoswere harvested at stage 10 and analyzed by western blot with GFPantibodies. �-Tubulin was used as a loading control. (B) Unilateralinjection of kermit 2 morpholino (40 ng) into one dorsal blastomere atthe four-cell stage completely blocked eye development in 60% ofembryos (lower left panel, n=38) and reduced eye formation in anadditional 32% (not shown). Co-expression of kermit2-GFP mRNA,which lacked the morpholino target sequence, restored eye formationin 72% of kermit2-depleted embryos (lower right panel, n=43).kermit2-GFP mRNA alone did not cause apparent embryonic defects(upper right panel).

DEVELO

PMENT

3654

binding protein by Booth et al. from a yeast two-hybrid screen(Booth et al., 2002). To study the role of kermit 2/XGIPC in the IGFsignaling pathway, we confirmed that kermit 2/XGIPC interacts withthe IGF1R in Xenopus embryos. Full-length XIGF1R co-immunoprecipitated with GFP-tagged kermit 2 expressed ingastrula-stage embryos (Fig. 4A), suggesting that kermit 2/XGIPCphysically interacts with the XIGF1R in Xenopus.

Inhibition of IGF signaling by a dominant-negative XIGF1R(DN-IGFR) reduces the size of anterior structures, including eyes,in Xenopus embryos (Pera et al., 2001). To test for a functionalinteraction between kermit 2 and IGF1R, we used doses of thekermit 2 morpholino and DN-IGFR mRNA that alone only mildlyreduce eye formation (Fig. 4B). When co-injected, kermit 2morpholino and DN-IGFR led to embryos with strongly reducedeyes (Fig. 4B), suggesting a synergistic effect between the two.Furthermore, although these levels of kermit 2 morpholino or DN-IGFR alone minimally affected Pax6 expression (72.5%, n=24;100%, n=27, respectively), co-injection strongly reduced Pax6expression in the presumptive eye in 82% (n=22) of the embryos(Fig. 4C). Taken together, these data indicate that kermit 2/XGIPCcould be involved in the IGF pathway, potentially functioning at thereceptor level.

Kermit 2/XGIPC is required for IGF1 induced eyeformationIn order to test the requirement of kermit 2/XGIPC for IGFsignaling, we examined whether depletion of kermit 2 inhibits IGFfunction in embryos. Dorsal overexpression of IGF1 mRNA hasbeen shown to induce ectopic eyes and overgrowth of endogenouseyes (Pera et al., 2001; Richard-Parpaillon et al., 2002). We foundsimilarly that dorsal injection of XIGF1 mRNA caused expandedeyes in 64% (n=22) of the embryos. Depletion of kermit 2dramatically inhibited the XIGF1 induced eye phenotype and led toembryos (73%, n=20) with small eyes or no eyes (Fig. 5A). Kermit2 depletion alone reduced eye formation (similar to Fig. 2). Dorsalinjection of IGF1 mRNA also expanded cement gland formation inembryos, but this phenotype was not affected by kermit 2 depletion.

Kermit 2 is also required for IGF induced expression of eyemarkers in animal cap explants. As reported (Pera et al., 2001;Richard-Parpaillon et al., 2002), expression of IGF1 alone in animalcaps induces anterior neural markers (Fig. 5B, lane 4), includingXAG (a cement gland marker), Otx2 (a marker for the forebrain andanterior midbrain), Pax6 (a marker for eyes and the forebrain) andXrx (an eye marker), but not the midbrain-hindbrain boundarymarker En2 or the mesodermal marker muscle actin. Kermit 2


Fig. 3. Depletion of kermit 2/XGIPCspecifically inhibits Xenopus eyedevelopment. (A-D) Depletion of kermit 2does not affect the expression of dorsalmesoderm markers chordin and goosecoid.Embryos are viewed from the vegetal side anddorsal is towards the top. Control morpholinoor kermit 2 morpholino (40 ng) was injectedinto two dorsal animal blastomeres of four-cellembryos. Embryos were fixed at the gastrulastage (stage 10) and whole-mount in situhybridization was performed for chordin (A,B)and goosecoid (C,D). (E-L) Depletion of kermit2 specifically reduces marker gene expressionwithin the presumptive eye field in stage 20embryos. Embryos are viewed from theanterior side with dorsal towards the top.Control or kermit 2 morpholino was injectedinto one dorsal animal blastomere of four-cell/eight-cell embryos with 500 pg of mRNAfor nuclear �-galactosidase. Embryos werefixed at stage 20 and �-galactosidase activitywas measured in situ (red) followed by whole-mount in situ hybridization for Bf1 (E,F), Otx2(G,H), Pax6 (I,J) and Xrx (K,L). Bf1 expressionwas not affected by depletion of kermit 2 (F).Expression of Otx2 (H) and Pax6 (J) werereduced only within the presumptive eye-forming region (red arrowheads) in embryosinjected with kermit 2 morpholino. Theexpression of the eye marker Xrx was alsoinhibited (L red arrowhead). (M-P) Depletion ofkermit 2 does not reduce Xrx (N) or Pax6 (P)expression in early neurula stage embryos(stage 14). (Q-S) Kermit 2 mRNA restored Xrxexpression in kermit 2-depleted embryos. Redarrowhead in R indicates strongly reduced Xrxexpression within presumptive eye domain inkermit 2-depleted embryo; red arrow in Sindicates recovered Xrx expression in embryoco-injected with kermit 2 morpholino andkermit 2 mRNA.

DEVELO

PMENT

depletion strongly reduced the induction of Xrx, mildly reducedPax6 induction, and did not affect the induction of the anterior neuralmarkers XAG and Otx2 (lane 6). Co-injection of control morpholinodid not affect any of the IGF1-induced markers (lane 5). Theseresults suggest that kermit 2/XGIPC is required for IGF1-inducedeye formation in Xenopus.

Kermit 2/XGIPC is required for long-termactivation of AKT by IGF1As shown above, kermit 2 is required for IGF induced embryonicphenotypes, but whether it is required for the activation ofdownstream IGF signaling is still unknown. The two mainintracellular pathways activated by IGF/IGFR are the PI3 kinase/AKT pathway and the MAP kinase pathway (Oldham and Hafen,2003). We first examined the requirement of kermit 2/XGIPC forIGF signaling in Xenopus oocytes. Stage VI oocytes have low levelsof background activation of AKT or MAP kinase, but respondrobustly to exogenous IGF1. We injected control morpholino orkermit 2 morpholino into oocytes and cultured these oocytes for 48hours to deplete endogenous kermit 2. We then treated these oocyteswith IGF1 protein for 30 minutes or overnight (18 hours). IGF1induced phosphorylation of AKT (Ser 473) and MAPK (e.g. Thr183and Tyr185 in ERK2) at sites associated with activation asdocumented previously (Alessi et al., 1996; Payne et al., 1991;Stephens et al., 1998; Stokoe et al., 1997; Yung et al., 1997). Kermit

2 depletion blocked phosphorylation/activation of AKT induced bylong-term exposure to IGF1, but not by short-term exposure (30minutes) (Fig. 6A). MAP kinase phosphorylation induced by IGF1protein was not affected by depletion of kermit 2 at any time points(Fig. 6A). The control morpholino did not affect IGF1-induced AKTor MAP kinase phosphorylation/activation. These data suggest thatkermit 2/XGIPC is required to maintain activation of AKT inducedby IGF1, but is not required for the initiation of signaling.

This analysis was extended to ectodermal (animal cap) explants.Compared with oocytes, animal caps have high levels of endogenousAKT and MAP kinase phosphorylation/activation. We first testedthe effect of kermit 2 morpholino on the endogenousphosphorylation of AKT and MAP kinase. Control morpholino orkermit 2 morpholino was injected into fertilized eggs. Animal capswere dissected at the late blastula stage (stage 9) and cultured for 2hours (stage 10/10.5) or overnight (stage 20). Similar to the resultin oocytes, kermit 2 depletion inhibited AKT phosphorylation/activation after overnight incubation, but not after 2 hoursincubation; kermit 2 morpholino had no effect on MAP kinasephosphorylation (Fig. 6B). To test whether kermit 2 is required forAKT phosphorylation/activation by exogenous IGF1, embryos wereinjected with IGF1 mRNA with or without kermit 2 morpholino,animal caps were explanted at the blastula stage and cultured untillate neurula stage (stage 20). We found that kermit 2 depletionblocked phosphorylation/activation of AKT by overexpressed IGF1


Fig. 4. Interaction between kermit2/XGIPCand XIGF1R in eye development. (A) Co-immunoprecipitation of kermit 2 and full-lengthXIGF1R. Embryos were injected at the one-cellstage with mRNAs encoding XIGF1R (2 ng) andGFP-tagged kermit 2 (1 ng), and cultured until thegastrula stage. XIGF1R/kermit 2 complexes wereimmunoprecipitated from embryo lysates withanti-GFP antibody and XIGF1R was visualized bywestern blotting. Mouse IgG was used as anegative control. (B) Kermit 2 morpholino andDN-IGFR synergistically inhibit eye formation inembryos. Kermit 2 morpholino (20 ng), DN-IGFRmRNA (500 pg), or both, were injected into twodorsal animal blastomeres of four-cell/eight-cellembryos. Embryos were cultured until tadpolestages to score phenotypes. The percentage ofembryos with either strong or mild reduction ineyes is tabulated in the panel on the right side.(C) Kermit 2 morpholino and DN-IGFRsynergistically reduce expression of Pax6 inpresumptive eye domain (arrowhead).Microinjections were performed as in Fig. 2B.Nuclear �-galactosidase mRNA was co-injected asa lineage tracer. Embryos were fixed at stage 20and �-galactosidase activity was measured in situ(red) followed by whole-mount in situhybridization for Pax6.

DEVELO

PMENT

3656

(Fig. 6C, compare lane 5 with lane 4). (Kermit 2 morpholino did notinterfere with translation of injected mRNAs, as GFP mRNA co-injected with IGF1 mRNA was expressed at equal levels with orwithout kermit 2 morpholino co-injection.) Taken together, theresults in oocytes and animal cap explants suggest that kermit2/XGIPC is required for maintaining IGF induced AKT activation,but not for MAPK activation.

Activated PI3K partially rescues the kermit2/XGIPC loss-of-function phenotypeAs kermit 2 depletion strongly inhibits the AKTphosphorylation/activation induced by IGF1, we examined thephysiological relevance of this inhibition by rescuing the kermit 2loss-of-function phenotype using a constitutively active p110subunit (p110*) of phosphatidylinositol-3� kinase (PI3 kinase)


Fig. 5. Kermit 2/XGIPC is required for IGF1 induced eye formationin whole embryos and in animal cap explants. (A) Kermit 2 isrequired for IGF1 induced eye formation in Xenopus embryos. XIGF1mRNA (1 ng), kermit 2 morpholino (40 ng), or both, were injected intoone dorsal animal blastomere of four-cell/eight-cell embryos. IGF1injection alone leads to embryos with expanded eyes and cementglands and co-injection of kermit 2 morpholino specifically inhibits theIGF1-induced eye phenotype. (B) Kermit 2 is required for IGF1-inducedeye marker expression in animal cap explants. XIGF1 mRNA alone (2ng), or with control morpholino (CM) or kermit 2 morpholino (K2M)was injected into four animal blastomeres of four-cell/eight-cellembryos. Animal caps were explanted from stage 9 blastulae, cultureduntil stage 20, and then harvested for RT-PCR. St. 20 WE indicates thewhole embryo control; uninjected represents animal caps fromuninjected embryos; –RT, without reverse transcriptase; EF1� is theloading control. Kermit 2 morpholino specifically reduces XIGF1-induced expression of the presumptive eye markers Xrx and Pax6.

Fig. 6. Kermit 2/XGIPC is required to maintain IGF1 inducedphosphorylation of AKT in oocytes and animal cap explants.(A) Kermit 2 is required for maintaining IGF1-induced AKTphosphorylation in oocytes. Oocytes were injected with 40 ng ofcontrol morpholino (CM) or kermit 2 morpholino (K2M), cultured for48 hours and then treated with recombinant human IGF1 protein (18ng/ml) for 30 minutes or overnight. Kermit 2 is required for IGF1-induced AKT phosphorylation/activation after overnight treatment, butnot for short-term phosphorylation of AKT or MAPK. Total AKT or totalMAPK was used as the loading control. (B) Effects of kermit 2morpholino on the endogenous activation of AKT and MAPK. Controlmorpholino or kermit 2 morpholino was injected into four animalblastomeres of four-cell/eight-cell embryos. Animal caps were explantedat stage 9 and harvested after 2 hours or overnight. Kermit 2morpholino reduces endogenous AKT phosphorylation after overnightincubation but not after 2 hours incubation and does not affect MAPKphosphorylation at either time point. (C) Kermit 2 is required for IGF1-induced AKT phosphorylation in stage 20 animal cap explants. 500 pgof XIGF1 mRNA and 400 pg of GFP mRNA were injected into one-cellembryos and 40 ng of kermit 2 morpholino was injected into fouranimal blastomeres of four-cell/eight-cell embryos. Animal caps weredissected at stage 9 and harvested at stage 20. GFP was used as theinjection control.

DEVELO

PMENT

(Carballada et al., 2001), a lipid kinase upstream of AKT anddownstream of IGF1R. Expression of p110* partially rescues thekermit 2 depletion phenotype (Fig. 7B). In a typical experiment,96% (n=24) of the embryos injected dorsally with the kermit 2morpholino alone had either miniscule or no apparent eyes, whereasonly 26% (n=20) of embryos co-injected with p110* showed thisphenotype, and 74% developed small but normal-appearing eyes.Dorsal expression of p110* itself did not cause apparentdevelopmental defects in embryos. p110* was confirmed to be activeas it induced strong AKT phosphorylation in gastrula-stage animalcaps (Fig. 7A). In addition to the morphological rescue, p110* alsopartially recovers the expression of Xrx, an eye marker, in kermit 2-depleted embryos. Compared with control embryos, most kermit 2-depleted embryos had strongly reduced Xrx expression (63%, n=39;Fig. 7D, middle panel) at the neurula stage (stage 19), whereas themajority of embryos co-injected with p110* showed only a mildreduction of Xrx expression (73%, n=53; Fig. 7D, right panel). Theserescue results suggest that the kermit 2 loss-of-function phenotypeis at least partially due to the inhibition of the PI3 kinase/AKTpathway.

Kermit 2/XGIPC is required for cell survival, butnot for cell proliferation in XenopusThe IGF pathway affects multiple cellular processes, including cellsurvival and cell proliferation. We examined whether kermit2/XGIPC is involved in these processes. We injected controlmorpholino or kermit 2 morpholino into the right side of embryos(leaving the left side as the control) and followed apoptosis byTUNEL analysis. Seventy-two percent (n=61) of kermit 2-depletedembryos demonstrate a substantial increase in the number ofTUNEL-positive nuclei on the injected side compared with thecontrol side (Fig. 8B). This effect was dose dependent (data notshown) and no difference was detected with control morpholino atequivalent doses (Fig. 8A). Expression of kermit2 mRNA withoutthe 5�UTR significantly reduced TUNEL-positive nuclei in 83%(n=66) of kermit 2-depleted embryos (Fig. 8C), which indicates thatthe elevated cell death was specifically due to loss of kermit 2.Expression of the same amount of n�-gal mRNA did not reduceapoptosis in kermit 2-depleted embryos (data not shown). We alsoassessed whether kermit 2 regulates cell proliferation. Embryosunilaterally injected with kermit 2 morpholino were immunostained


Fig. 7. Activation of PI3K partiallyrescues eye development in kermit2/XGIPC-depleted embryos. (A) p110*,a constitutively active subunit of PI3 kinase,induces AKT phosphorylation/activation instage 10 animal cap explants. mRNAencoding p110* was injected intofertilized eggs and animal caps wereexplanted at the gastrula stage andanalyzed by western blot with antibodiesrecognizing AKT phosphorylated at serine-473 (upper panel) or general anti-AKTantibodies (lower panel). (B) p110*partially rescues eye development in kermit2-depleted embryos. Kermit 2 morpholino(40 ng), 3 ng of p110* mRNA, or both,were injected into two dorsal animalblastomeres of four-cell/eight-cell embryos.No apparent eyes are present in embryosinjected with the kermit 2 morpholino,while small eyes are present in over 70%of embryos co-injected with p110* mRNAand morpholino. p110* injection alonedoes not affect embryo development.(C) The percentage of embryos withnormal, small or absent/miniscule eyes issummarized. (D) p110* partially recoversthe expression of Xrx in kermit 2-depletedembryos. One dorsal-animal blastomere offour-cell/eight-cell embryos was injectedwith kermit 2 morpholino with or withoutp110* mRNA (right side, indicated by GFPlineage tracer) and harvested at stage 20.Whole-mount in situ hybridization wasperformed for Xrx. Embryos are viewedfrom the anterior side with dorsal towardsthe top.

DEVELO

PMENT

3658

with an antibody against phosphorylated histone H3, whichspecifically recognizes mitotic chromosomes. We did not observe anobvious change in the number of mitotic cells on the injected sidecompared with the control side in early neurula stage (stage 14) andlate neurula stage (stage 22) embryos (Fig. 8E,F). These resultsindicate that kermit 2/XGIPC is required for cell survival, but do notsupport a requirement for kermit 2 in proliferation in Xenopusembryos. As kermit 2 is involved in IGF signaling, we further testedwhether inhibition of the IGF pathway increases cell death inXenopus embryos. Indeed, DN-IGFR, like kermit 2 depletion,markedly increased the number of TUNEL-positive nuclei (76%,n=29; Fig. 8D) when expressed in embryos. Activation of theMAPK branch of IGF signaling leads to transcriptional responsesassociated with cell proliferation, whereas activation of thePI3K/AKT branch regulates the activity of BCL2 family membersand anti-apoptotic responses (O’Connor, 2003; Oldham and Hafen,2003). Therefore, these data are consistent with a requirement forkermit 2 in IGF-induced AKT activation. Finally, to determinewhether the kermit 2 depletion phenotype is due to apoptosis of cellswithin the eye-field, we examined whether inhibition of apoptosisrestores eye formation in kermit 2-depleted embryos. Kermit 2morpholino injection alone strongly reduced Pax6 expression in thepresumptive eye in 75% (n=24, Fig. 8H) of embryos. Co-expressionof bcl2 rescued this phenotype and restored Pax6 expression in thepresumptive eye in almost 72% (n=25, Fig. 8I) of embryos, as wellas restoring eye development to normal or near normal morphologyin a majority of sibling embryos allowed to develop to later stages(data not shown). This result suggests that the kermit 2 phenotype isat least partially due to elevated cell death in the presumptive eyeregion.

DISCUSSIONWe report here the identification of the second Xenopus GIPChomolog, kermit 2. Despite its high amino acid similarity with thefirst Xenopus GIPC homolog Kermit, kermit 2 has a differentfunction in Xenopus embryos. Loss-of-function and molecularmarker analyses indicate that kermit 2 is specifically required for

Xenopus eye development. We have confirmed the physicalinteraction of kermit 2/XGIPC with XIGF1R in Xenopus and furtherestablished that kermit 2 is required for IGF-dependent eyedevelopment. In addition, kermit 2 is required for maintaining IGFinduced AKT phosphorylation/activation and a constitutively activePI3 kinase partially rescues kermit 2 loss of function. These resultsprovide the first demonstration of the in vivo role of GIPC inembryonic development and also show that kermit 2/XGIPC is anovel component of the IGF pathway, potentially acting through themodulation of IGF1R function.

kermit 2/XGIPC and IGF signaling in Xenopusembryonic developmentInhibition of IGF signaling in Xenopus by either dominant-negativeIGF receptor or IGF receptor depletion disrupts anterior neuraldevelopment broadly (Pera et al., 2001; Richard-Parpaillon et al.,2002), yet loss of kermit 2/XGIPC specifically inhibits eyeformation without apparent effect on other aspects of anterior neuraldevelopment. At least two mechanisms could explain the more-restricted requirement for kermit 2/XGPIC. First, depletion of kermit2 blocks the activation of AKT, but not of MAPK. It is likely that theIGF/PI3K/AKT branch is specifically involved in eye formation,while the IGF/MAPK branch regulates anterior neural developmentmore broadly. Indeed, IGF inhibits SMAD1 activity and inducesneural differentiation via MAPK (Pera et al., 2003; Kuroda et al.,2005). Second, eye development could have a distinct temporalrequirement for IGF signaling. For example, the specification ofanterior neural tissue in general may only require an early exposureto IGF, whereas proper eye formation may require continualactivation of IGF signaling. As kermit 2/XGIPC is required for themaintenance, but not the initiation of IGF signaling, depletion ofkermit 2/XGIPC may inhibit eye formation only at late neurulastages, without disrupting general neural specification.

We also found that depletion of kermit 2 dramatically increasescell death in embryos, but does not reduce cell proliferation. Thisresult is consistent with the data that kermit 2 depletion blocksIGF induced AKT activation, not MAPK activation. However, the


Fig. 8. Kermit 2/XGIPC is required for cell survival, butnot for cell proliferation in Xenopus. (A-F) Controlmorpholino (A), kermit 2 morpholino alone (B,E,F), kermit 2morpholino with 3 ng of kermit 2 mRNA (C) or 1 ng of DN-IGFR (D) was injected into the right side dorsal animalblastomere of four-cell/eight-cell embryos. The left sideserves as a control. TUNEL staining and phosphorylatedhistone H3 staining were performed on neurula stageembryos. Embryos are viewed from the dorsal side, anteriortowards the top. Compared with the control side (left side),the side injected with kermit 2 morpholino (right side) showsa substantial increase in the number of TUNEL-positive nuclei(B), without apparent change in the number of mitotic cells(E,F). The increased apoptosis in kermit 2-depleted embryoscan be rescued by co-expression of kermit 2 mRNA lackingthe 5�UTR (C). DN-IGFR also leads to elevated cell death onthe injected side, as shown in D. (G-I) Co-injection of Bcl2mRNA recovers Pax6 eye expression in kermit 2-depletedembryos. Microinjections were performed as above and n�-gal was used as the lineage tracer. Embryos were fixed atstage 20 and �-galactosidase activity was measured in situ(red) followed by whole-mount in situ hybridization for Pax6.Red arrowhead in H indicates strongly reduced eye region inkermit 2 morpholino-injected embryo and red arrow in Iindicates recovered eye expression domain by Bcl2 mRNA.

DEVELO

PMENT

increased cell death in kermit 2-depleted embryos is not restrictedto the presumptive eye field. Thus, the extent to which increasedcell death contributes to the eye phenotype remains to beestablished.

Regulation of the IGF1 receptor by kermit 2/XGIPCAs kermit 2/XGIPC functions downstream of the IGF ligand andupstream of PI3 kinase and physically associates with the XIGF1R,kermit 2 probably regulates IGF signaling through modulation of thereceptor. In principle, kermit 2 could regulate the stability, activityor subcellular localization of the receptor. However, kermit 2depletion did not obviously reduce the level of XIGF1R underconditions that reduced AKT activation (data not shown), arguingagainst kermit 2 regulation of IGF1R stability. Furthermore, kermit2 depletion did not significantly affect the level of IGF-inducedMAPK phosphorylation, arguing against a role for kermit 2 inregulating overall stability or activity of the IGF1R. Kermit 2depletion also did not interfere with the early (30 minute)phosphorylation of AKT in response to IGF, suggesting thatcoupling to downstream signaling, at least initially, is intact.

Alternatively, mammalian GIPC is involved in the regulation ofendocytic trafficking, and kermit 2/XGIPC may similarly regulatethe subcellular localization of the IGF receptor. A function for GIPCin endocytic trafficking was first proposed based on its localizationto endocytic vesicles (Dance et al., 2004; De Vries et al., 1998a; DeVries et al., 1998b; Jeanneteau et al., 2004a; Lou et al., 2002; Louet al., 2001). For example, GIPC is enriched at clathrin-richinvaginations and the endocytic compartments found betweenmicrovilli in proximal tubule kidney cells (Lou et al., 2002). A rolefor GIPC in endocytosis is also supported by functional studies incultured cells (Hirakawa et al., 2003; Jeanneteau et al., 2004a).Hirakawa et al. used RNAi to knockdown GIPC in 293 cells andfound that GIPC is partially responsible for maintaining a relativelyconstant level of the human lutropin receptor (LHR) at the cellsurface during ligand-induced internalization and for recycling ofthe ligand (CG) (Hirakawa et al., 2003). In a similar manner, kermit2 could be required for the recycling of the IGF1R to the plasmamembrane after ligand-dependent internalization. A failure torecycle internalized IGF1R could explain why depletion of kermit 2reduces AKT phosphorylation after prolonged exposure to IGF, butnot after short-term exposure. A defect in recycling of IGF1R wouldnot be expected to disrupt MAP kinase phosphorylation, as MAPkinase can be activated by receptors that have been internalized inassociation with �-arrestin and component kinases of the MAPKcascade (DeFea et al., 2000; Lefkowitz and Shenoy, 2005; Lin et al.,1998; Luttrell et al., 2001; Tohgo et al., 2003), while AKT and PI3kinase have not been reported to be activated by internalizedreceptors within endosomes. However, we have so far been unableto demonstrate an effect of kermit 2 depletion on IGF1R trafficking,and this will remain a focus of future studies.

In summary, we have shown that kermit 2/XGIPC is required foreye development in Xenopus embryos. Kermit 2/XGIPC physicallyand functionally interacts with the IGF1R and is required for IGFsignaling in anterior neural development specifically in eyeformation. Expression of molecular markers of eye development isinduced but not maintained in kermit 2-depleted embryos, andphosphorylation of AKT is similarly induced but not maintainedafter prolonged exposure to IGF when kermit 2 is depleted. Eyedevelopment can be partially rescued in kermit 2-depleted embryosby an active PI3 kinase. Based on these observations, we proposethat kermit 2/XGIPC, through the modulation of IGF1R, mediatesendogenous IGF signaling in Xenopus eye formation.

We thank Eddy M. De Robertis, Laurent Richard-Parpaillon, Patrick Lemaire,Monica Vetter, Vijayasaradhi Setaluri, Carole LaBonne, Dan Kessler and Jean-Pierre Saint-Jeannet for plasmids and reagents. We thank Dan Kessler, Jean-Pierre Saint-Jeannet, Mary Mullins, Jonathan Epstein, Tom Kadesch and JingYang for helpful discussions.

ReferencesAlessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P.

and Hemmings, B. A. (1996). Mechanism of activation of protein kinase B byinsulin and IGF-1. EMBO J. 15, 6541-6551.

Bellmeyer, A., Krase, J., Lindgren, J. and LaBonne, C. (2003). Theprotooncogene c-myc is an essential regulator of neural crest formation inxenopus. Dev. Cell 4, 827-839.

Blitz, I. and Cho, K. (1995). Anterior neurectoderm is progressively inducedduring gastrulation: the role of the Xenopus homeobox gene orthodenticle.Development 121, 993-1004.

Blobe, G. C., Liu, X., Fang, S. J., How, T. and Lodish, H. F. (2001). A novelmechanism for regulating transforming growth factor beta (TGF-beta) signaling.Functional modulation of type III TGF-beta receptor expression throughinteraction with the PDZ domain protein, GIPC. J. Biol. Chem. 276, 39608-39617.

Blumberg, B., Wright, C. V., De Robertis, E. M. and Cho, K. W. (1991).Organizer-specific homeobox genes in Xenopus laevis embryos. Science 253,194-196.

Bohlson, S. S., Zhang, M., Ortiz, C. E. and Tenner, A. J. (2005). CD93 interactswith the PDZ domain-containing adaptor protein GIPC: implications in themodulation of phagocytosis. J. Leukoc. Biol. 77, 80-89.

Booth, R. A., Cummings, C., Tiberi, M. and Liu, X. J. (2002). GIPC participatesin G protein signaling downstream of insulin-like growth factor 1 receptor. J.Biol. Chem. 277, 6719-6725.

Cai, H. and Reed, R. R. (1999). Cloning and characterization of neuropilin-1-interacting protein: a PSD- 95/Dlg/ZO-1 domain-containing protein that interactswith the cytoplasmic domain of neuropilin-1. J. Neurosci. 19, 6519-6527.

Carballada, R., Yasuo, H. and Lemaire, P. (2001). Phosphatidylinositol-3 kinaseacts in parallel to the ERK MAP kinase in the FGF pathway during Xenopusmesoderm induction. Development 128, 35-44.

Cox, W. G. and Hemmati-Brivanlou, A. (1995). Caudalization of neural fate bytissue recombination and bFGF. Development 121, 4349-4358.

Dance, A. L., Miller, M., Seragaki, S., Aryal, P., White, B., Aschenbrenner, L.and Hasson, T. (2004). Regulation of myosin-VI targeting to endocyticcompartments. Traffic 5, 798-813.

De Vries, L., Elenko, E., McCaffery, J. M., Fischer, T., Hubler, L., McQuistan, T.,Watson, N. and Farquhar, M. G. (1998a). RGS-GAIP, a GTPase-activatingprotein for Galphai heterotrimeric G proteins, is located on clathrin-coatedvesicles. Mol. Biol. Cell 9, 1123-1134.

De Vries, L., Lou, X., Zhao, G., Zheng, B. and Farquhar, M. G. (1998b). GIPC, aPDZ domain containing protein, interacts specifically with the C terminus ofRGS-GAIP. Proc. Natl. Acad. Sci. USA 95, 12340-12345.

Deardorff, M. A., Tan, C., Conrad, L. J. and Klein, P. S. (1998). Frizzled-8 isexpressed in the Spemann organizer and plays a role in early morphogenesis.Development 125, 2687-2700.

Deardorff, M. A., Tan, C., Saint-Jeannet, J. P. and Klein, P. S. (2001). A role forfrizzled 3 in neural crest development. Development 128, 3655-3663.

DeFea, K. A., Zalevsky, J., Thoma, M. S., Dery, O., Mullins, R. D. andBunnett, N. W. (2000). beta-arrestin-dependent endocytosis of proteinase-activated receptor 2 is required for intracellular targeting of activated ERK1/2. J.Cell Biol. 148, 1267-1281.

Eivers, E., McCarthy, K., Glynn, C., Nolan, C. M. and Byrnes, L. (2004). Insulin-like growth factor (IGF) signalling is required for early dorso-anteriordevelopment of the zebrafish embryo. Int. J. Dev. Biol. 48, 1131-1140.

El Mourabit, H., Poinat, P., Koster, J., Sondermann, H., Wixler, V., Wegener,E., Laplantine, E., Geerts, D., Georges-Labouesse, E., Sonnenberg, A. etal. (2002). The PDZ domain of TIP-2/GIPC interacts with the C-terminus of theintegrin alpha5 and alpha6 subunits. Matrix Biol. 21, 207-214.

Favre-Bonvin, A., Reynaud, C., Kretz-Remy, C. and Jalinot, P. (2005). Humanpapillomavirus type 18 E6 protein binds the cellular PDZ protein TIP-2/GIPC,which is involved in transforming growth factor beta signaling and triggers itsdegradation by the proteasome. J. Virol. 79, 4229-4237.

Hasson, T. (2003). Myosin VI: two distinct roles in endocytosis. J. Cell Sci. 116,3453-3461.

Hemmati-Brivanlou, A. and Melton, D. A. (1994). Inhibition of activin signallingpromotes neuralization in Xenopus laevis. Cell 77, 273-281.

Hemmati-Brivanlou, A., Kelley, O. G. and Melton, D. A. (1994). Follistatin, anantagonist of activin, is present in the Spemann organizer and displays directneuralizing activity. Cell 77, 283-295.

Hirakawa, T., Galet, C., Kishi, M. and Ascoli, M. (2003). GIPC binds to thehuman lutropin receptor (hLHR) through an unusual PDZ domain binding motif,and it regulates the sorting of the internalized human choriogonadotropin andthe density of cell surface hLHR. J. Biol. Chem. 278, 49348-49357.


DEVELO

PMENT

3660

Hirsch, N. and Harris, W. A. (1997). Xenopus Pax-6 and retinal development. J.Neurobiol. 32, 45-61.

Hu, L. A., Chen, W., Martin, N. P., Whalen, E. J., Premont, R. T. andLefkowitz, R. J. (2003). GIPC interacts with the beta1-adrenergic receptor andregulates beta1-adrenergic receptor-mediated ERK activation. J. Biol. Chem. 278,26295-26301.

Jeanneteau, F., Diaz, J., Sokoloff, P. and Griffon, N. (2004a). Interactions ofGIPC with dopamine D2, D3 but not D4 receptors define a novel mode ofregulation of G protein-coupled receptors. Mol. Biol. Cell 15, 696-705.

Jeanneteau, F., Guillin, O., Diaz, J., Griffon, N. and Sokoloff, P. (2004b). GIPCrecruits GAIP (RGS19) to attenuate dopamine D2 receptor signaling. Mol. Biol.Cell 15, 4926-4937.

Katoh, M. (2002). GIPC gene family (Review). Int. J. Mol. Med. 9, 585-589.Kuroda, H., Fuentealba, L., Ikeda, A., Reversade, B. and De Robertis, E. M.

(2005). Default neural induction: neuralization of dissociated Xenopus cells ismediated by Ras/MAPK activation. Genes Dev. 19, 1022-1027.

Lamb, T. M., Knecht, A. K., Smith, W. C., Stachel, S. E., Economides, A. N.,Stahl, N., Yancopolous, G. D. and Harland, R. M. (1993). Neural Induction bythe Secreted Polypeptide Noggin. Science 262, 713-718.

Lefkowitz, R. J. and Shenoy, S. K. (2005). Transduction of receptor signals bybeta-arrestins. Science 308, 512-517.

Ligensa, T., Krauss, S., Demuth, D., Schumacher, R., Camonis, J., Jaques, G.and Weidner, K. M. (2001). A PDZ domain protein interacts with the C-terminal tail of the insulin-like growth factor-1 receptor but not with the insulinreceptor. J. Biol. Chem. 276, 33419-33427.

Lin, F. T., Daaka, Y. and Lefkowitz, R. J. (1998). beta-arrestins regulatemitogenic signaling and clathrin-mediated endocytosis of the insulin-like growthfactor I receptor. J. Biol. Chem. 273, 31640-31643.

Liu, T. F., Kandala, G. and Setaluri, V. (2001). PDZ domain protein GIPC interactswith the cytoplasmic tail of melanosomal membrane protein gp75 (tyrosinase-related protein-1). J. Biol. Chem. 276, 35768-35777.

Lou, X., Yano, H., Lee, F., Chao, M. V. and Farquhar, M. G. (2001). GIPC andGAIP form a complex with TrkA: a putative link between G protein and receptortyrosine kinase pathways. Mol. Biol. Cell 12, 615-627.

Lou, X., McQuistan, T., Orlando, R. A. and Farquhar, M. G. (2002). GAIP, GIPCand Galphai3 are concentrated in endocytic compartments of proximal tubulecells: putative role in regulating megalin’s function. J. Am. Soc. Nephrol. 13,918-927.

Luttrell, L. M., Roudabush, F. L., Choy, E. W., Miller, W. E., Field, M. E., Pierce,K. L. and Lefkowitz, R. J. (2001). Activation and targeting of extracellularsignal-regulated kinases by beta-arrestin scaffolds. Proc. Natl. Acad. Sci. USA 98,2449-2454.

Mathers, P. H., Grinberg, A., Mahon, K. A. and Jamrich, M. (1997). The Rxhomeobox gene is essential for vertebrate eye development. Nature 387, 603-607.

O’Connor, R. (2003). Regulation of IGF-I receptor signaling in tumor cells. Horm.Metab. Res. 35, 771-777.

Oldham, S. and Hafen, E. (2003). Insulin/IGF and target of rapamycin signaling: aTOR de force in growth control. Trends Cell Biol. 13, 79-85.

Papalopulu, N. and Kintner, C. (1996). A posteriorising factor, retinoic acid,reveals that anteroposterior patterning controls the timing of neuronaldifferentiation in Xenopus neuroectoderm. Development 122, 3409-3418.

Payne, D. M., Rossomando, A. J., Martino, P., Erickson, A. K., Her, J. H.,Shabanowitz, J., Hunt, D. F., Weber, M. J. and Sturgill, T. W. (1991).Identification of the regulatory phosphorylation sites in pp42/mitogen-activatedprotein kinase (MAP kinase). EMBO J. 10, 885-892.

Pera, E. M., Wessely, O., Li, S. Y. and De Robertis, E. M. (2001). Neural andhead induction by insulin-like growth factor signals. Dev. Cell 1, 655-665.

Pera, E. M., Ikeda, A., Eivers, E. and De Robertis, E. M. (2003). Integration ofIGF, FGF, and anti-BMP signals via Smad1 phosphorylation in neural induction.Genes Dev. 17, 3023-3028.

Richard-Parpaillon, L., Heligon, C., Chesnel, F., Boujard, D. and Philpott, A.(2002). The IGF pathway regulates head formation by inhibiting Wnt signaling inXenopus. Dev. Biol. 244, 407-417.

Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L. K. and De Robertis, E.M. (1994). Xenopus chordin: a novel dorsalizing factor activated by organizer-specific homeobox genes. Cell 79, 779-790.

Stephens, L., Anderson, K., Stokoe, D., Erdjument-Bromage, H., Painter, G.F., Holmes, A. B., Gaffney, P. R., Reese, C. B., McCormick, F., Tempst, P. etal. (1998). Protein kinase B kinases that mediate phosphatidylinositol 3,4,5-trisphosphate-dependent activation of protein kinase B.[see comment]. Science279, 710-714.

Stokoe, D., Stephens, L. R., Copeland, T., Gaffney, P. R., Reese, C. B., Painter,G. F., Holmes, A. B., McCormick, F. and Hawkins, P. T. (1997). Dual role ofphosphatidylinositol-3,4,5-trisphosphate in the activation of protein kinaseB.[see comment]. Science 277, 567-570.

Tan, C., Deardorff, M. A., Saint-Jeannet, J. P., Yang, J., Arzoumanian, A.and Klein, P. S. (2001). Kermit, a frizzled interacting protein, regulatesfrizzled 3 signaling in neural crest development. Development 128, 3665-3674.

Tani, T. T. and Mercurio, A. M. (2001). PDZ interaction sites in integrin alphasubunits. T14853, TIP/GIPC binds to a type I recognition sequence in alpha6A/alpha 5 and a novel sequence in alpha 6B. J. Biol. Chem. 276, 36535-36542.

Tohgo, A., Choy, E. W., Gesty-Palmer, D., Pierce, K. L., Laporte, S., Oakley, R.H., Caron, M. G., Lefkowitz, R. J. and Luttrell, L. M. (2003). The stability ofthe G protein-coupled receptor-beta-arrestin interaction determines themechanism and functional consequence of ERK activation. J. Biol. Chem. 278,6258-6267.

Wang, L. H., Kalb, R. G. and Strittmatter, S. M. (1999). A PDZ protein regulatesthe distribution of the transmembrane semaphorin, M-SemF. J. Biol. Chem. 274,14137-14146.

Yang, J., Tan, C., Darken, R. S., Wilson, P. A. and Klein, P. S. (2002). Beta-catenin/Tcf-regulated transcription prior to the midblastula transition.Development 129, 5743-5752.

Yung, Y., Dolginov, Y., Yao, Z., Rubinfeld, H., Michael, D., Hanoch, T.,Roubini, E., Lando, Z., Zharhary, D. and Seger, R. (1997). Detection of ERKactivation by a novel monoclonal antibody. FEBS Lett. 408, 292-296.


kermit 2/xgipc, an igf1 receptor interacting protein, is required … · (tan et al., 2001), that...

Documents