Reciprocal regulation of Wnt and Gpr177/mouseWntless is required for embryonic axis formationJiang Fu1, Ming Jiang1, Anthony J. Mirando1, Hsiao-Man Ivy Yu, and Wei Hsu2

Department of Biomedical Genetics, Center for Oral Biology, James P Wilmot Cancer Center, University of Rochester Medical Center, 601 Elmwood Avenue,Box 611, Rochester, NY 14642

Edited by Kathryn V. Anderson, Sloan-Kettering Institute, New York, NY, and approved September 16, 2009 (received for review May 6, 2009)

Members of the Wnt family are secreted glycoproteins that triggercellular signals essential for proper development of organisms. Cel-lular signaling induced by Wnt proteins is involved in diverse devel-opmental processes and human diseases. Previous studies havegenerated an enormous wealth of knowledge on the events insignal-receiving cells. However, relatively little is known about themaking of Wnt in signal-producing cells. Here, we describe thatGpr177, the mouse orthologue of Drosophila Wls, is expressed duringformation of embryonic axes. Embryos with deficient Gpr177 exhibitdefects in establishment of the body axis, a phenotype highly remi-niscent to the loss of Wnt3. Although many different mammalian Wntproteins are required for a wide range of developmental processes,the Wnt3 ablation exhibits the earliest developmental abnormality.This suggests that the Gpr177-mediated Wnt production cannot besubstituted. As a direct target of Wnt, Gpr177 is activated by �-cateninand LEF/TCF-dependent transcription. This activation alters the cellu-lar distributions of Gpr177 which binds to Wnt proteins and assiststheir sorting and secretion in a feedback regulatory mechanism. Ourfindings demonstrate that the loss of Gpr177 affects Wnt productionin the signal-producing cells, leading to alterations of Wnt signalingin the signal-receiving cells. A reciprocal regulation of Wnt andGpr177 is essential for the patterning of the anterior-posterior axisduring mammalian development.

A-P axis � �-catenin � developmental deformities � primitive streak �Wnt production

Members of the Wnt family are secreted glycoproteins whichtrigger cellular signals essential for proper development of

organisms (1, 2). Aberrant regulation of an evolutionary conservedWnt signal transduction pathway has been implicated in a variety ofcancers and congenital diseases (3, 4). There is no question thatWnt signaling is intimately involved in human health and disease.Genetic analysis in mice has revealed the essential role of differentWnt proteins in a wide range of developmental processes (http://www.stanford.edu/�rnusse/wntwindow.html). Wnt3 deficiency ap-pears to cause the earliest abnormality during embryogenesis,suggesting the importance of Wnt signaling in axis determination(5). Three body axes develop sequentially to generate embryoorientations (6–8). At the egg cylinder stage, the dorsal-ventral axisis the first to form. The anterior-posterior (A-P) axis is establishedto form the primitive streak at the posterior end before gastrulation.Lastly, the left-right asymmetry is formed, followed by embryoturning. Wnt signaling is critical for initiation of the embryonic axes inearly development (9–11). Disruptions of key molecules necessary forWnt signaling regulation lead to defects in axial patterning (12–15).

Studies in the past have generated an enormous wealth ofknowledge on the events in signal-receiving cells. Before initiatingtheir effects on the signal-receiving cells, Wnt proteins undergoproper modification, sorting and secretion processes in the signal-producing cells (16–19). Recent identification of Wntless (Wls/Evi/Srt) (20–22), a regulator for Wnt production in Drosophila, hasdirected attention to the maturation, sorting and secretion pro-cesses in signal-producing cells. Given the extensive Wnt family inhigher organisms, it is not clear how many Wls genes are present andwhether Wls is essential for Wnt production. This study describes

Gpr177, the mouse orthologue of Drosophila Wls, required forembryogenesis. Disruption of Gpr177 disturbs axial patterning, aphenotype resembling the loss of Wnt3. This disruption not onlyaffects Wnt production, but also interferes with Wnt signaling. Asa Wnt transcriptional target, Gpr177 is elevated to promote Wntproduction in a positive feedback loop. Our results indicate that areciprocal regulation of Wnt and Gpr177 is essential for theestablishment of the mammalian A-P axis.

ResultsGpr177 Is Essential for Mouse Embryogenesis. We investigated howmany Wls exist, and whether Wls regulates the Wnt pathwayessential for mammalian development. By protein sequenceanalysis (NCBI HomoloGene) we found that Gpr177, whichshows high percentages of identity with Wls of human (96.1%),Drosophila (44.9%), and C. elegans (40.9%), is likely the mouseorthologue. No additional gene product shared significant ho-mology with fly and human counterparts. We then performedwhole mount RNA in situ hybridization to examine Gpr177expression in early embryogenesis. Gpr177 was expressed in theproximal epiblast at the junction between the embryonic andextraembryonic tissue, and later was more restricted to theprimitive streak and mesoderm extending to the distal tip of theembryo (Fig. 1 A–D). Its strong presence was found in bothposterior visceral endoderm and epiblast at the prestreak, butswitched to the mesoderm at late-streak (Fig. 1 E–H). Theexpression pattern of Gpr177 is reminiscent of Wnt3, which isrequired for axial patterning (5, 23).

To determine whether Gpr177 regulates Wnt production andthis regulation is essential for mouse development, we created amutant strain Gpr177lacZ, carrying an insertion of �-geo into theninth intron of Gpr177 (Fig. S1 A and B). The transgene insertiondisrupted the seven transmembrane domain which results ingeneration of a fusion transcript (Fig. S1C). PCR analysesfurther confirmed that the Gpr177 locus was altered by trans-gene-mediated mutagenesis (Fig. S1D). The mutant lacking thecarboxyl-terminal region of Gpr177 disabled its function as thetruncation interrupts its subcellular distribution and proteininteraction (see below). Mice heterozygous for Gpr177lacZ ap-peared normal and were fertile. We were not able to recoverGpr177 homozygous (Gpr177-/-) newborns or embryos afterE10.5, suggesting that they died during early embryogenesis.From E6.5 to E8.5, Gpr177�/� and �/- embryos consisted ofthree germ layers and underwent gastrulation to form organizedstructures, including extraembryonic ectoderm, chorion, allan-tois, head folds, and primitive streak (Fig. 1 I, K, M, N, and T).

Author contributions: J.F., M.J., A.J.M., H.-M.I.Y., and W.H. designed research; J.F., M.J.,A.J.M., H.-M.I.Y., and W.H. performed research; J.F., M.J., A.J.M., H.-M.I.Y., and W.H.analyzed data; and A.J.M. and W.H. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

1J.F., M.J., and A.J.M. contributed equally to this work.

2To whom correspondence should be addressed. E-mail: wei�hsu@urmc.rochester.edu.

This article contains supporting information online at www.pnas.org/cgi/content/full/0904894106/DCSupplemental.

18598–18603 � PNAS � November 3, 2009 � vol. 106 � no. 44 www.pnas.org�cgi�doi�10.1073�pnas.0904894106

However, Gpr177-/- embryos did not exhibit distinct structuresbut remained to grow as egg cylinders (Fig. 1 J and O). Themutants consisted of two layers of tissue, ectoderm and visceralendoderm. In addition, the mesoderm and primitive streak weremissing and ectoderm was composed of a thick layer of cells.Later, the ectoderm and visceral endoderm continue to growand the excess ectoderm becomes irregular and folded (Fig. 1 Land U). To examine the Gpr177 protein, we generated antibodiesrecognizing its carboxyl terminus, which was deleted in themutants. Immunostaining detected a strong presence of Gpr177in the control mesoderm at E7.5 (Fig. 1 P–R). However, theGpr177-positive mesoderm was absent in the mutants (Fig. 1S).

Disruption of Gpr177 Impairs Development of the Body Axis. We nextanalyzed Gpr177 mutants for the expression of specific markersduring gastrulation (Fig. 2). BMP4 is expressed in the extraem-bryonic ectoderm where its signals induce the proximal epiblast

to acquire posterior cell fates and restrict the formation of distalvisceral endoderm (DVE), which are precursors of anteriorvisceral endoderm (AVE) to the distal end (7, 24). The expres-sion of BMP4 apparently is not affected by the mutation,implying proper positioning of extraembryonic ectoderm duringthe proximal-distal (P-D) axis formation (Fig. 2 A and B; n � 5).To examine the formation of AVE, we examined the expressionof Cer1 (25), an AVE marker in early to mid-streak stageembryos and later in the definitive endoderm emanating fromthe node (Fig. 2C). In Gpr177-/- embryos, we detected Cer1 in

A B C D

E F G H

I JK L

M N O T

P Q S U

Fig. 1. Disruption of Gpr177 impairs embryogenesis in mice. (A–H) In situhybridization in whole mounts (A–D) and sections (E–H) reveals the expressionof Gpr177 in E6.25 (A, E, and F) and E7 (B), and E7.25 (C, D, G, and H) embryos.The approximate positions of E–H are shown by the dotted line in A and C.Gross morphological analysis of Gpr177�/� (I and K) and Gpr177-/- (J and L)embryos at E7.5 (I and J) and E8.5 (K and L). Sections of the Gpr177�/� (M, N,P–R, and T) and Gpr177-/- (O, S, and U), E7.5 (M–S), and E8.5 (T and U) embryoswere analyzed by histology (M–O, T, and U) and immunostaining of Gpr177(P–S). Arrowheads and arrows indicate the anterior and posterior mesoderm,respectively. AVE, anterior visceral endoderm; Ch, chorion; Ect, ectoderm; Epc,ectoplacental cone; NE, neural ectoderm; PS, primitive streak; VE, visceralendoderm; XEct, extraembryonic ectoderm. [Scale bars, 300 �m (A–D, I, and J);100 �m (E–H and P–S); 500 �m (K and L); and 200 �m (M–O, T, and U).]

A B D

E F G H

I J K L

M N O

R S TP Q

C

Fig. 2. Gpr177 is requiredforWntproductionandsignaling inpatterningofA-Paxis. (A–N and P–T) Molecular marker analysis of control (�/� and �/-) andGpr177 mutant (-/-) littermates at E6.5 (A and B) and E7.0–7.5 (C–N and P–T)determines the role of Gpr177 in early embryogenesis using in situ hybridizationof BMP4 (A and B), Cer1 (C and D), Otx2 (E and F), Hesx1 (G and H), Gsc (I and J),Brachyury (T) (K and L), Wnt3 (M and N), and Axin2 (P and Q), and GFP analysis ofAxin2 (R–T). The control embryos are shown with the anterior facing to the left.(O) Gpr177 is essential for Wnt production and signaling. Immunoblot analysis ofE6.5 and E7.5 embryos shows the level of Gpr177, Wnt3/3a, and �-catenin pro-teinsaffectedbytheGpr177mutation.Actin level isusedasa loadingcontrol.Thenumber represents the relative protein level of Wnt3/3a and �-catenin betweenGpr177�/�, Gpr177�/-, and Gpr177-/-. (R–T) Axin2GFP mouse strain, expressingGFP in the Axin2-expressing cells, was used to examine the activation of Wnt/�-catenin signaling in the Gpr177�/� (R and S) and Gpr177-/- (T) embryos duringgastrulation. AVE, anterior visceral endoderm; DE, definitive endoderm; PS,primitive streak. [Scale bars, 200 �m (A and B) and 300 �m (C–N and P–T).]

Fu et al. PNAS � November 3, 2009 � vol. 106 � no. 44 � 18599

DEV

ELO

PMEN

TAL

BIO

LOG

Y

the visceral endoderm although its expansion into the definitiveendoderm was absent, suggesting that AVE was establishedduring initial regional patterning (Fig. 2D; n � 3). However, thesubsequent development of the anterior ectoderm did not occur.Otx2 is expressed ubiquitously in all germ layers at the earlystreak-stage, but maintained only in the anterior region at themid-streak stage (Fig. 2E). We detected Otx2 throughout theentire ectoderm of Gpr177 mutants (Fig. 2F; n � 4). In addition,the expression of Hesx1 was not affected in the AVE (Fig. 2 Gand H; n � 3), but was altered in the forebrain of Gpr177-/-embryos. The uniform presence of Otx2 reflected an undiffer-entiated ectoderm rather than expansion of anterior neural cellfates. These data indicate that the posterior development of theembryo might also be defective. Gsc was expressed in the nodeadjacent to the anterior region of primitive streak and the newlyformed axial mesoderm (Fig. 2I). At the late-streak stage, theGsc-expressing progenitors of the notochord, identified in the nodelocated at the anterior end of the primitive streak, possess thecapability necessary (26) and sufficient to induce the neural axis (27,28). However, Gsc expression was abolished in the mutant, sug-gesting the lack of gastrulation organizer activity in the Gpr177mutant (Fig. 2 J; n � 3). Furthermore, mesoderm specificationrequires T (Brachyury), which is expressed in the primitive streakand axial mesoderm (Fig. 2K). Its expression in the most anteriorstreak region is dependent on Wnt signaling (29). At E7.5, we didnot detect its presence in the mutant, confirming the lack ofprimitive streak and mesoderm formation (Fig. 2L; n � 3).

Gpr177 Deficiency Alters Wnt Production and Signaling. The abovephenotypes are highly reminiscent of the ablation of Wnt3, whichis required for establishment of the primitive streak (5). Wetherefore examined the expression of Wnt3 whose transcripts werefound in the primitive streak, proximal epiblast, and visceralendoderm, at the junction between the embryonic and extraem-bryonic ectoderm with higher levels localized to the posteriorregion of both Gpr177�/� (Fig. 2M) and -/- embryos (Fig. 2N; n �5). Although the expression pattern of Wnt3 transcript does notseem to be altered, its protein production and signaling effect mightbe affected by Gpr177 deficiency. We therefore determinedwhether Gpr177 is essential for Wnt3 production and signaling inearly embryogenesis. Previous reports indicated that �-cateninsignaling is critical for the establishment of the A-P axis (13, 30, 31).Immunoblot analysis of E6.5 and E7.5 embryos showed that cellularlevels of �-catenin were drastically reduced by the mutation (Fig.2O). Using an antibody recognizing Wnt3/3a, the protein level wasunaffected at E6.5 (Wnt3 only) but accumulated at E7.5 (Wnt3 andWnt3a). Note that the mutants were fairly normal at the prestreakwhere Wnt3a is not present. Because of technical limitation, wecould not examine the expression pattern of Wnt3/3a proteins in thedeveloping embryo. Nonetheless, the results suggest that Wntexpression was not affected in the embryos. However, �-cateninwas not activated likely due to a secretion defect. Furthermore,Axin2 is regulated by �-catenin and LEF/TCF-dependent transcrip-tion, and has been widely used as a downstream target of Wnt (32,33). Its expression in the posterior region of the E7.5 embryo (Fig.2P) appeared to be missing in the mutant (Fig. 2Q; n � 2). Usingthe Axin2GFP mouse strain to label the Axin2-expressing cells, wedetected GFP signals in the Gpr177�/� but not Gpr177-/- embryosduring gastrulation, indicating that the canonical Wnt pathway isaffected by the mutation (Fig. 2 R–T; n � 2). Our findings suggest thatGpr177 acts downstream of Wnt3 and regulates its signaling in earlypatterning of the A-P axis. In addition to Wnt production, Gpr177is required for Wnt signaling during embryonic axis formation.

Subcellular Distribution of Gpr177. To determine the role of Gpr177in production of Wnt, we examined its localization within the cell.First, immunostaining showed that endogenous Gpr177 is mainlylocalized to cellular vesicles in the majority of cell types (n � 10)

A B

C D

E F G

H I

J K L

M N O

P Q

R S T

U V W

Fig. 3. Wnt regulates cellular distribution of Gpr177. (A–D) Three-dimensional images of immunostained cells reveal differential localizations ofendogenous Gpr177 in NEP (A), Neurosphere cells (B), MSC (C), and C57MG (D).Cut view (E–I) and 3-D imaging (J–O) analyses show that Gpr177 co-localizeswith a Golgi marker GM130 using superimposed imaging (H and L) or pseudo-coloring of the co-localization signal in white (I and M–O). NEP cells wereimmunostained with Gpr177 (E and J) and GM130 (F and K), and counter-stained with DAPI (G–O). Three serial sections along the front-view (Co Level1, 2, and 3) reveal the co-localization signal on the surface of Golgi (M–O).(P–T) High levels of Gpr177 lead to its accumulations in Golgi. Three-dimensional imaging of the endogenous Gpr177 in the vesicles of C3H10T1/2cells (P). Expression of a myc-tag full length Gpr177 (MT-Gpr177) showsalteration in subcellular distribution of Gpr177 (Q), co-localizing with GM130(R–T). Co-immunostaining of a MT-Gpr177 mutant lacking the carboxyl ter-minal region (U) and Calnexin (V) reveals their co-localization (W).

18600 � www.pnas.org�cgi�doi�10.1073�pnas.0904894106 Fu et al.

examined with an exception of neural stem cells (Fig. 3 A–D). Inneural epithelial progenitors (NEP) and neurosphere cells, isolatedfrom E12.5 neural tube and forebrain, respectively, Gpr177 washighly concentrated in the perinuclear area resembling Golgi inaddition to the vesicles (Fig. 3 A and B). Consistent with a previousstudy of Wls in Drosophila (34), the endogenous Gpr177 wasco-localized with a Golgi marker GM130 (Fig. 3 E–I) in cut viewanalysis. Using a combined 3-D imaging and co-localization anal-ysis, Gpr177 was also co-localized with GM130 on Golgi (Fig. 3J–L). In serial sectioned views of the 3-D images (Fig. 3 M–O), theco-localization signal (white color) appeared to be on the Golgiapparatus between the Gpr177-positive vesicles (green color) andGolgi (red color). The budding of the Gpr177 containing vesiclesseems to occur on the surface of Golgi and continue their intra-cellular trafficking routes.

To determine whether the differential localization of Gpr177 isattributed to neural stem cell specificity, we expressed a myc-taggedGpr177 (MT-Gpr177) protein in a variety of cell types (n � 6). Inall six cell types examined, including C3H10T1/2, MC3T3, C57MG,293T, mouse embryonic fibroblasts, and mouse mesenchymal stemcells, MT-Gpr177 displayed localization to Golgi in addition tocellular vesicles (Fig. 3 P–T). Furthermore, a MT-Gpr177 mutant,lacking the carboxyl-terminal region similar to the Gpr177-lacZfusion, exhibited a dislocated distribution pattern. Co-localizationof an ER marker Calnexin with this mutant suggests that it fails toenter the secretory pathway (Fig. 3 U–W). The results suggest thatthe Golgi accumulation of Gpr177 is not due to cell-type specificity.

The expression level of Gpr177 might dictate its cellular distribu-tion. It is conceivable that neural stem cells, expressing high levelsof Gpr177, are the only Wnt-secreting cell type we have examined.Cells with Gpr177 elevation are likely to be the signal-secreting cellsbecause of its role in assisting the production of Wnt proteins.

Gpr177 Is a Transcriptional Target of Wnt/�-catenin Signaling. To testwhether expression of Wnt might alter the cellular level anddistribution of Gpr177, we studied the transcriptional regulationof Gpr177. We isolated and characterized the Gpr177 regulatoryregion which contains 7 potential LEF/TCF binding sites todetermine whether the Gpr177 transcription is regulated by Wnt(Fig. 4A). The results indicated that Gpr177 transcription isstimulated by the dominant-activated mutant, �N�-cat or cat-CLEF1 (Fig. 4B). Deletion analysis of the Gpr177 regulatoryregion further showed that four potential LEF/TCF-binding sites(nos. 4–7) might be responsible for the transcriptional stimula-tion (Fig. 4C). Chromatin immunoprecipitation (ChIP) analysisdemonstrated that the expressed catCLEF1 bound to these fourLEF/TCF consensus sites in cells (Fig. 4 D and E).

Next, we determined which of these four sites is most criticalfor the activation of Gpr177 by the �-catenin and LEF/TCF-dependent transcription. Because the sites 4 and 5, as well as 6and 7, are closely linked to each other, a point mutation strategy(Fig. 4A) was used to disrupt each of these four sites (M4, M5,M6, and M7). Disruption of number 4 or 7 site, but not 5 or 6,significantly diminished the transcriptional stimulation (Fig. 4F).

A B

C

D E

FG

H I

J K

L M

N O

Fig. 4. Gpr177 is regulated by the canonical Wnt pathway. (A) Graphs illustrate the luciferase reporter constructs for Gpr177 promoter with the wild-type ormutation (cross) of LEF/TCF binding consensus sequences. (B) Expression of dominant activated catCLEF1 or ��-cat protein stimulates the transcription activityof a 3-kb Gpr177 promoter (P1) in 293T cells. Relative luciferase activity (RLA) determined the transcriptional activation of the Gpr177 promoter-luciferaseconstruct. The analysis of pGL3, a parent vector, shows background activity. (C) Fold of induction shows the effect of catCLEF1 on transcriptional activation ofthe deletion mutants. (D) 293T cells were transfected by increasing amounts of DNA plasmid (1, 3, and 9 �g) to express catCLEF1, analyzed by immunoblot (IB).(E) ChIP analysis reveals high affinity LEF/TCF binding sites (nos. 4–5 and 6–7) in the Gpr177 promoter. The order number of these sites is color coded with thoseshown in A. NC is a negative control, which analyzes the regulatory region of Gpr177 without LEF/TCF binding sequence. The controls are direct PCR analysesof LEF/TCF binding sites without ChIP. (F) Analysis of the promoter constructs containing point mutations further reveals the consensus sites required for �-cateninand LEF/TCF-dependent transcription. (G) IB analysis indicates the Gpr177 level elevated in the primary MEC cells by the MMTV-Wnt1 transgene. The expressionlevel of Actin was analyzed as a loading control. (H–K) �-gal staining of the virgin 2-month mammary glands in whole mounts (H and I) and sections (J and K)reveals the Wnt-dependent activation of Gpr177 in the mammary glands. The reporter expression from the Gpr177-lacZ knock-in allele was detected in theMMTV-Wnt1 transgenics (I and K) but not the controls (H and J). Three-dimensional imaging of the immunostained control (L) and MMTV-Wnt1 transgenic (M)MEC reveals distinct localization patterns of endogenous Gpr177. The endogenous Gpr177 distribution in C3H10T1/2 cells (N) is also altered by high levels ofHA-Wnt3A (O). The insets show co-immunostaining of Gpr177 with Golgi markers, GM130 (M) and GS28 (O). Immunostained cells were counterstained by DAPI(blue). G, Golgi; V, vesicle. [Scale bars, 500 �m (H and I) and 50 �m (J and K).]

Fu et al. PNAS � November 3, 2009 � vol. 106 � no. 44 � 18601

DEV

ELO

PMEN

TAL

BIO

LOG

Y

A combinatorial mutation of number 4 and 5 (M45) or number6 and 7 (M67) caused a reduction similar to the effect of M4 orM7 (Fig. 4F). However, the loss of both number 4 and 7 sites(M47), but not number 5 and 6 sites (M56) drastically abolishedthe �-catenin and LEF/TCF-dependent transcription (Fig. 4F).The transcriptional activity of M47 is about the same as M4567where all four sites are disrupted (Fig. 4F). The results stronglysuggest that Gpr177 is a direct Wnt downstream target whoseexpression is regulated by the �-catenin and LEF/TCF-dependent transcription.

Wnt Expression Alters the Cellular Level and Distribution of Gpr177.We examined whether the Gpr177 expression is stimulated byWnt that might provide a feedback mechanism to regulate itsproduction and signaling. Indeed, immunoblot analysis showedthat the steady state level of the endogenous Gpr177 increasedin the primary mammary epithelial cells (MEC) of MMTV-Wnt1compared to the controls (Fig. 4G). To test the stimulation ofGpr177 by Wnt in animals, we crossed the MMTV-Wnt1 trans-gene into mice carrying the Gpr177lacZ allele. This allele, whichcontains the �-geo reporter controlled by the Gpr177 locus,permits an examination of its endogenous expression activity. Invirgin mammary glands heterozygous for the Gpr177lacZ allele,no �-gal staining could be detected (Fig. 4 H and J). In contrast,the MMTV-Wnt1 transgenic littermates showed strong �-galstained signals, suggesting that the Gpr177 expression is stimu-lated (Fig. 4 I and K). Next, the cellular localization of Gpr177affected by Wnt was analyzed to further our investigation ontheir interactions. We examined primary MEC cells isolatedfrom MMTV-Wnt1 transgenic females and their control litter-mates. Compared to the controls, Gpr177 was located to Golgiin addition to cellular vesicles in the MMTV-Wnt1 cells (Fig. 4L and M). Furthermore, transient expression of HA-Wnt3A inC3H10T1/2 cells also led to an accumulation of endogenousGpr177 in the Golgi (Fig. 4 N and O). These data support thehypothesis that Wnt proteins modulate the cellular level and theexpression of Gpr177. As a direct target of Wnt, Gpr177 mightfacilitate their maturation, sorting, and secretion processes in afeedback regulatory loop.

Interactions of Gpr177 and Wnt Proteins. To investigate whether theWnt-dependent elevation of Gpr177 assists the Wnt productionthrough a feedback regulatory mechanism, we examined theirsubcellular localizations. Immunostaining showed that the ele-vated Gpr177, caused by expression of Wnt1 (Fig. 5 A–D),became co-localized together in the Golgi and vesicles. Co-immunoprecipitation further identified complexes containingGpr177 and Wnt proteins in cells (Fig. 5 E and F). GST pull downassay indicated that Gpr177 associates with Wnt1, Wnt3, andWnt5a proteins, which are highly expressed in neural stem cells(Fig. 5H). This association was not detectable in other cell typeswithout high levels of Wnt expression. Analysis of the GST-Gpr177 deletion mutants (Fig. 5G) further revealed that a shortN-terminal domain (amino acids 101–232) is required for Wntproteins to associate with Gpr177 (Fig. 5H). Thus, the Wnt-mediated elevation of Gpr177 interacts with Wnt proteins in afeedback loop. Although it remains possible that Gpr177 doesnot bind directly to Wnt, they appear to regulate each other ina reciprocal manner.

DiscussionThis study demonstrated an essential role of Gpr177, the mouseorthologue of Drosophila Wls, in establishment of the A-P axis.Genetic analysis in mice has revealed the requirement of manydifferent Wnt proteins in a wide range of developmental processes.The disruption of Wnt3 seems to exhibit the earliest developmentalabnormality. The Gpr177 mutation causes defects in the primitivestreak and mesoderm formation, highly resembling the loss of Wnt3

(5). This suggests an indispensible role of Gpr177 in Wnt sortingand secretion. The Gpr177-mediated Wnt production is essentialfor mammalian development.

Golgi accumulations of endogenous Gpr177 occur in neuralstem cells. This is in agreement with previous analyses ofexogenous Wls, over-expressed in transfected cells (20, 35–37).The Golgi accumulation is not due to the cell-type specificity, butrather the expression level within the cells. Indeed, high levels ofWnt proteins are found in the neural stem cells where theirassociation with Gpr177 can be detected. The results imply thatthe cellular localization of Gpr177 might serve as an indicator forWnt-producing cells. Indeed, cells expressing Wnt exhibit ele-vated levels of Gpr177, leading to Golgi accumulation. Wntexpression might modulate the distribution of Gpr177 thatdictates the trafficking routes in signal-producing cells. Thisfeedback regulatory mechanism ensures proper sorting pathwaysto be activated for the secretion of Wnt proteins (34–38). Incontrast, the main role of Gpr177 expressed at low levels in nonWnt-producing cells could help in generating a morphogengradient for long range effects through endocytosis and exocy-tosis. Further analysis on the trafficking routes of Gpr177-containing vesicles will elucidate the mechanism underlying theprocess of Wnt maturation, sorting and secretion.

We hypothesize that a reciprocal interaction between Wnt andGpr177 plays a key role in the regulation of their expression,subcellular distribution, binding, and organelle-specific association.The mechanisms underlying the reciprocal regulation in the Wnt-producing cells are necessary for the patterning of the A-P axis. Therequirement of Wnt signaling in development of various organssuggests that Gpr177 might be required for these developmentalprocesses (1, 2). Indeed, we have found that Gpr177 is expressed indifferent tissues, including kidney, lung, skin, intestine, brain, spinalcord, skeleton, eyes, excretion glands, ear, tooth, and palatal shelve.

A

B

C

D

D’

E F

G

H

Fig. 5. Wnt proteins bind to and co-localized with Gpr177. Three-dimensional imaging analysis of immunostained C57MG cells expressing MT-Wnt1 reveals co-localization (C) of endogenous Gpr177 (A) with MT-Wnt1(B) in both Golgi and vesicles. D and D’ display different angle views of C wherethe sectioned level is shown by purple rectangle. (E and F) IP-IB analysisidentifies protein complexes containing Gpr177 and Wnt in cells transfectedby HA-Wnt1 (E) or MT-Gpr177 (F). (G) A scheme of GST-Gpr177 fragments.Numbers indicate positions of amino acids. Endoplasmic reticulum signalsequence and transmembrane domains are highlighted in green and blue,respectively. (H) GST pull-down assay analyzes the association of Gpr177 withthe Wnt1, Wnt3, or Wnt5a protein complex.

18602 � www.pnas.org�cgi�doi�10.1073�pnas.0904894106 Fu et al.

As Wnt signaling is intimately involved in a variety of cancers andcongenital diseases in humans (3, 4), Gpr177 might also be essentialfor these processes. Creation of mouse strains permitting geneticinactivation of Gpr177 in a spatiotemporal-specific fashion prom-ises important insights into the reciprocal regulation of Wnt andGpr177 in development and disease. As there are 19 differentmembers of Wnt in mammals, it is not known whether Gpr177 isrequired for all of their productions. Targeted disruption of Gpr177in cells expressing a specific Wnt is likely to gain knowledge on thegenerality of its function.

MethodsDetails for experimental materials and analyses are described in the SI Meth-ods. In brief, the Gpr177 mutant strain was generated using an ES clone (BayGenomics). For embryo genotyping, yolk sacs or embryonic materials recov-

ered from paraffin sections were used in PCR analysis. Axin2GFP strain permitsinducible expression of GFP in the Axin2-expressing cells (39, 40). Care and useof experimental animals described in this work comply with guidelines andpolicies of the University Committee on Animal Resources at the University ofRochester. Isolation and culture of primary neurospheres, NEP, MEC, calvarialMSC, and other cell lines are described in the SI Methods (41–44). RNA probes(5, 25) were generated to analyze the gene expression pattern by in situhybridization (45). Histology, �-gal staining, immunostaining, immunoblot,protein precipitation, chromatin immunoprecipitation, and various DNA vec-tors used are described in the SI Methods (39, 40, 42, 45–47).

ACKNOWLEDGMENTS. We thank Richard Behringer, Edward De Robertis, andBrigid Hogan for reagents, C.-S. Victor Lin and Chris Proschel for technicalassistance, and the reviewers for comments and suggestions. This work wassupported by National Institutes of Health Grants DE15654 and CA106308(to W.H.).

1. Logan CY, Nusse R (2004) The Wnt signaling pathway in development and disease. AnnRev Cell Dev Biol 20:781–810.

2. Clevers H (2006) Wnt/beta-catenin signaling in development and disease. Cell 127:469–480.

3. van Amerongen R, Berns A (2006) Knockout mouse models to study Wnt signaltransduction. Trends Genet 22:678–689.

4. Grigoryan T, Wend P, Klaus A, Birchmeier W (2008) Deciphering the function ofcanonical Wnt signals in development and disease: Conditional loss- and gain-of-function mutations of beta-catenin in mice. Genes Dev 22:2308–2341.

5. Liu P, Wakamiya M, Shea MJ, Albrecht U, Behringer RR, Bradley A (1999) Requirementfor Wnt3 in vertebrate axis formation. Nat Gen 22:361–365.

6. Beddington RS, Robertson EJ (1999) Axis development and early asymmetry in mam-mals. Cell 96:195–209.

7. Tam PP, Loebel DA, Tanaka SS (2006) Building the mouse gastrula: Signals, asymmetryand lineages. Curr Opin Genet Dev 16:419–425.

8. Takaoka K, Yamamoto M, Hamada H (2007) Origin of body axes in the mouse embryo.Curr Opin Genet Dev 17:344–350.

9. Hamatani T, Carter MG, Sharov AA, Ko MS (2004) Dynamics of global gene expressionchanges during mouse preimplantation development. Dev Cell 6:117–131.

10. Wang QT, et al. (2004) A genome-wide study of gene activity reveals developmentalsignaling pathways in the preimplantation mouse embryo. Dev Cell 6:133–144.

11. Rossant J, Tam PP (2004) Emerging asymmetry and embryonic patterning in earlymouse development. Dev Cell 7:155–164.

12. Zeng L, et al. (1997) The mouse Fused locus encodes Axin, an inhibitor of the Wntsignaling pathway that regulates embryonic axis formation. Cell 90:181–192.

13. Chazaud C, Rossant J (2006) Disruption of early proximodistal patterning and AVEformation in Apc mutants. Development 133:3379–3387.

14. Morkel M, et al. (2003) Beta-catenin regulates Cripto- and Wnt3-dependent geneexpression programs in mouse axis and mesoderm formation. Development 130:6283–6294.

15. Kemler R, et al. (2004) Stabilization of beta-catenin in the mouse zygote leads topremature epithelial-mesenchymal transition in the epiblast. Development 131:5817–5824.

16. Coudreuse D, Korswagen HC (2007) The making of Wnt: New insights into Wntmaturation, sorting and secretion. Development 134:3–12.

17. Hausmann G, Banziger C, Basler K (2007) Helping Wingless take flight: How WNTproteins are secreted. Nat Rev 8:331–336.

18. Takada R, et al. (2006) Monounsaturated fatty acid modification of Wnt protein: Itsrole in Wnt secretion. Dev Cell 11:791–801.

19. Willert K, et al. (2003) Wnt proteins are lipid-modified and can act as stem cell growthfactors. Nature 423:448–452.

20. Banziger C, Soldini D, Schutt C, Zipperlen P, Hausmann G, Basler K (2006) Wntless, aconserved membrane protein dedicated to the secretion of Wnt proteins from signal-ing cells. Cell 125:509–522.

21. Bartscherer K, Pelte N, Ingelfinger D, Boutros M (2006) Secretion of Wnt ligandsrequires Evi, a conserved transmembrane protein. Cell 125:523–533.

22. Goodman RM, et al. (2006) Sprinter: A novel transmembrane protein required for Wgsecretion and signaling. Development 133:4901–4911.

23. Rivera-Perez JA, Magnuson T (2005) Primitive streak formation in mice is preceded bylocalized activation of Brachyury and Wnt3. Dev Biol 288:363–371.

24. Lawson KA, et al. (1999) Bmp4 is required for the generation of primordial germ cellsin the mouse embryo. Genes Dev 13:424–436.

25. Belo JA, et al. (1997) Cerberus-like is a secreted factor with neutralizing activityexpressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev68:45–57.

26. Belo JA, Leyns L, Yamada G, De Robertis EM (1998) The prechordal midline of thechondrocranium is defective in Goosecoid-1 mouse mutants. Mech Dev 72:15–25.

27. Sulik K, et al. (1994) Morphogenesis of the murine node and notochordal plate. DevDyn 201:260–278.

28. Beddington RS (1994) Induction of a second neural axis by the mouse node. Develop-ment 120:613–620.

29. Yamaguchi TP, Takada S, Yoshikawa Y, Wu N, McMahon AP (1999) T (Brachyury) is adirect target of Wnt3a during paraxial mesoderm specification. Genes Dev 13:3185–3190.

30. Kimura-Yoshida C, et al. (2005) Canonical Wnt signaling and its antagonist regulateanterior-posterior axis polarization by guiding cell migration in mouse visceralendoderm. Dev Cell 9:639–650.

31. Huelsken J, Vogel R, Brinkmann V, Erdmann B, Birchmeier C, Birchmeier W (2000)Requirement for beta-catenin in anterior-posterior axis formation in mice. J Cell Biol148:567–578.

32. Jho EH, Zhang T, Domon C, Joo CK, Freund JN, Costantini F (2002) Wnt/beta-catenin/Tcfsignaling induces the transcription of Axin2, a negative regulator of the signalingpathway. Mol Cell Biol 22:1172–1183.

33. Lustig B, et al. (2002) Negative feedback loop of Wnt signaling through upregulationof conductin/axin2 in colorectal and liver tumors. Mol Cell Biol 22:1184–1193.

34. Port F, et al. (2008) Wingless secretion promotes and requires retromer-dependentcycling of Wntless. Nat Cell Biol 10:178–185.

35. Franch-Marro X, et al. (2008) Wingless secretion requires endosome-to-Golgi retrievalof Wntless/Evi/Sprinter by the retromer complex. Nat Cell Biol 10:170–177.

36. Belenkaya TY, et al. (2008) The retromer complex influences Wnt secretion by recyclingwntless from endosomes to the trans-Golgi network. Developmental cell 14:120–131.

37. Yang PT, Lorenowicz MJ, Silhankova M, Coudreuse DY, Betist MC, Korswagen HC(2008) Wnt signaling requires retromer-dependent recycling of MIG-14/Wntless inWnt-producing cells. Dev Cell 14:140–147.

38. Pan CL, Baum PD, Gu M, Jorgensen EM, Clark SG, Garriga G (2008) C. elegans AP-2 andretromer control Wnt signaling by regulating mig-14/Wntless. Dev Cell 14:132–139.

39. Yu HM, Liu B, Chiu SY, Costantini F, Hsu W (2005) Development of a unique system forspatiotemporal and lineage-specific gene expression in mice. Proc Natl Acad Sci USA102:8615–8620.

40. Yu HM, Liu B, Costantini F, Hsu W (2007) Impaired neural development caused byinducible expression of Axin in transgenic mice. Mech Dev 124:146–156.

41. Yu HM, et al. (2005) The role of Axin2 in calvarial morphogenesis and craniosynostosis.Development 132:1995–2005.

42. Liu B, Yu HM, Huang J, Hsu W (2008) Co-opted JNK/SAPK signaling in Wnt/beta-catenin-induced tumorigenesis. Neoplasia 10:1004–1013.

43. Grandbarbe L, et al. (2003) Delta-Notch signaling controls the generation of neurons/glia from neural stem cells in a stepwise process. Development 130:1391–1402.

44. Welm BE, Tepera SB, Venezia T, Graubert TA, Rosen JM, Goodell MA (2002) Sca-1(pos)cells in the mouse mammary gland represent an enriched progenitor cell population.Dev Biol 245:42–56.

45. Chiu SY, Asai N, Costantini F, Hsu W (2008) SUMO-specific protease 2 is essential formodulating p53-Mdm2 in development of trophoblast stem cell niches and lineages.PLoS Biol 6:e310.

46. Liu B, Yu HM, Hsu W (2007) Craniosynostosis caused by Axin2 deficiency is mediatedthrough distinct functions of beta-catenin in proliferation and differentiation. DevBiol 301:298–308.

47. Hsu W, Mirando AJ, Yu HM (August 3, 2009) Manipulating gene activity in Wnt1-expressing precursors of neural epithelial and neural crest cells. Dev Dyn, 10.1002/dvdy.22044.

Fu et al. PNAS � November 3, 2009 � vol. 106 � no. 44 � 18603

DEV

ELO

PMEN

TAL

BIO

LOG

Y