A road to kidney tubules via the Wnt pathway

Transcription

1 Pediatr Nephrol (2000) 15: IPNA 2000 DEVELOPMENTAL BIOLOGY REVIEW Seppo J. Vainio Marika S. Uusitalo A road to kidney tubules via the Wnt pathway Received: 21 June 1999 / Revised: 31 March 2000 / Accepted: 31 March 2000 Abstract Classical in vitro studies indicate that tubule induction in the kidney mesenchyme is mediated by cellcell contacts between the inducer tissue and the metanephric mesenchyme. Induction is completed within the first 24 h, after which tubules will form because of stimulated cell proliferation, migration, and cell adhesion. Recent evidence has revealed an essential role for the secreted signals from the Wnt gene family. Of these, Wnt-4 is expressed in developing tubules and knocking out its function perturbed kidney development. More detailed studies demonstrated normal condensation, but tubules were missing. Subsequent experiments indicated that Wnt-4 is also a sufficient signal to trigger tubulogenesis. Cells that were engineered to express Wnt-4 not only induced tubulogenesis in the kidney mesenchyme of Wnt- 4 mutant embryos, but also induced tubules in the wild type mesenchyme. With the transfilter induction assay, Wnt-4-mediated induction was completed within the first 24 h, depending on the presence of proteoglycans and cell-cell contacts between the interactants. In addition, Wnt-4 autoinduced expression of its own gene and a panel of other components of the Wnt signalling pathway, such as frizzleds and a candidate Wnt antagonist from the secreted frizzled-related protein family. Taken together, the data provide evidence of an essential role for Wnt signal transmission and transduction pathways in the induction of kidney tubules, and the findings have paved the way for detailed molecular studies. The kidney as a model of embryonic induction: a brief history The kidney is a useful model to assay inductive signalling between the epithelial and the mesenchymal tissues, which is also the major driving force of organogenesis in most other organs. These interactions also regulate the initial formation of a predetermined field for organs during earlier developmental stages and play a role in the terminal differentiation of the various cell types [1]. By separating the interacting tissue layers, namely the epithelial ureter and the metanephrogenic mesenchyme, Grobstein [2] provided evidence in the 1950s that Key words Kidney organogenesis Tubule induction Wnt Frizzled Secreted frizzled-related protein S. Vainio ( ) University of Oulu, Department of Biochemistry, P.O.Box 3000, Linnanmaa, Finland Seppo.Vainio@oulu.fi Tel.: , Fax: S.J. Vainio M.S. Uusitalo Biocenter Oulu and Department of Biochemistry, Faculties of Science and Medicine, University of Oulu, Oulu, Finland Fig. 1A F Model systems used to analyze kidney tubulogenesis experimentally. A Ureter bud (u) grows into the kidney mesenchyme (m) and when separated (B) and recombined for 24 h, ureter branching and formation of tubules (t) from the nephrogenic zone (hatched) are induced and formed during subculture (C). Tubules can be induced experimentally with a piece of dorsal spinal cord (spc) (D), as well as with cells that express Wnts including Wnt-4 (E) that is also expressed in the spc. Both will lead to the development of tubules (t, hatched) in vitro (F) (f filter, wg cells that express Wnt protein)

2 152 Fig. 2A G Expression of some Wnts and Wnt pathway genes in the embryonic kidney. A Wnt-11 is expressed at the tips (arrows) of epithelial ureter bud that is thought to be the endogenous site of an in vivo inductive signal. Yellow star marks the induced and condensed kidney mesenchyme, days post coitum (DPC) 13.5 DPC. B A sample stained as a whole mount to demonstrate Wnt-11 gene expression in the ureter tips in the whole kidney, 14.5 DPC. C Wnt-4 is expressed in the nephrogenic mesenchyme and the pretubular aggregates that will generate the nephrons, 13.5 DPC. D A sample stained as a whole mount to demonstrate Wnt-4 gene expression in pretubular aggregates, E14.5. E A secreted frizzled-related protein, sfrp-2, is present as Wnt-4 in the pretubular aggregates DPC, whereas Frizzled-7 (F) is expressed throughout the nephrogenic zone that governs Wnt-4- and sfrp-2-positive areas, 14.5 DPC. G Wnt-7b is expressed in the presumptive collecting duct epithelium, 13.5 DPC nephrogenesis involves inductive signalling from the ureter bud to the kidney mesenchyme in order to turn on differentiation of tubules. Furthermore, studies with explant cultures, where interacting tissues are cultured either separately or in recombination in vitro, provided conclusive evidence that the kidney mesenchyme generates the nephron and many of its associated cell types (Fig. 1). One of the problems in analyzing kidney tubule induction is that after microsurgical separation, the ureter is a weak tubule inducer in explant cultures. Saxén [1], Grobstein [2] and later Ekblom et al. [3] developed experimental tools to analyze the kinetics of tubule induction more efficiently. They developed a transfilter assay system where the kidney mesenchyme is placed on the top and the heterologous inducer tissue under the filter (Fig. 1). Several embryonic tissues, including embryonic spinal cord, were able to replace the ureter bud for tubule induction. When the kinetics of the induction was studied more closely, mainly using a piece of a spinal cord, it was found that a full response in this model system required around 24 h of recombination. After this inductive period the inducer tissue could be removed, and segmented tubules were shown to develop into the induced mesenchyme during subculture [1, 3]. Based on these studies, the following model of tubule induction was formulated [1, 4]. After signal transmission (apparently by direct cell-cell contacts), the induced cells start to proliferate, migrate, and adhere to the pretubular cell aggregates, which polarize, epithelialize, and differentiate into nephrons. The ureter bud was expected to express an upstream inductive factor that would trigger tubulogenesis in the nephrogenic mesenchymal cells [1]. The assumption was that the spinal cord and other heterologous inducer tissues in vitro would present the same or a similar primary signal as the ureter bud in vivo to induce tubule formation in the mesenchyme. Differential expression of the Wnt genes during kidney development Wnts form a large family of secreted signals whose biochemical properties have been difficult to study due to problems in obtaining biologically active and soluble Wnt proteins [5, 6]. Much of our knowledge of Wnt functions in mammals is based on gene targeting experiments, which have proved to be an effective way to assay their developmental roles. Knock-out experiments in the mouse showed that Wnt-1 is important for midbrain development, Wnt-3a for gastrulation, Wnt-5a for outgrowing of the craniofacial area and the limbs, and Wnt- 7a for dorsal-ventral polarity of the limb [7]. Of the Wnt family members, at least Wnt-11, Wnt-7b, and Wnt-4 are expressed in the developing kidney [8]. Wnt-11 expression is upregulated in the Wolffian duct in the region of the presumptive ureter bud at 10.0 days post coitum before ureteric budding, and is localized to the newly formed ureter tips during subsequent kidney development (Fig. 2) [8]. Wnt-4 expression appears at 11.0 days post coitum (DPC) in the nephrogenic zone adjacent to the ureter bud, and is upregulated in the pretubular aggregates [9]. Wnt-7b expression appears a bit later at 13.5 DPC and is present in the presumptive col-

3 lecting duct throughout kidney morphogenesis [8]. Based on in situ hybridization studies, it was concluded that transcripts of these Wnt genes appeared sequentially and that they were differentially expressed during kidney development (Fig. 2). The segmental and sequential expression suggested specific roles for these Wnt genes in the developing kidney. Evidence that Wnt signalling is essential for kidney development and especially for differentiation of the tubules Of the Wnt genes, Wnt-4 was of a particular interest, as its expression occurred in the metanephric mesenchyme around 11.0 DPC, and thereafter became confined to the developing tubules (Fig. 2). Based on the localized expression pattern, Wnt-4 may have either a local function to regulate differentiation of the tubules or it may function as a signal to cells that surround the tubule. In the latter alternative, Wnt-4 could contribute to the differentiation of the stromal component. Targeted disruption of Wnt-4 in the mouse was used to investigate these alternatives [9]. Analysis of Wnt-4 null mice revealed perturbed kidney development. More-detailed studies indicated that glomeruli did not develop and that the kidney mesenchyme remained morphologically undifferentiated [9]. These genetic studies indicated a direct role for Wnt signalling in kidney organogenesis, as reported for Drosophila melanogaster, where wingless signalling controlled the formation of the malpighian tubules [10]. Analysis of the kidneys from the mutant embryos at 11.5 DPC indicated normal condensation of the pretubular mesenchymal cells and unchanged expression of the panel of developmentally regulated genes important for kidney development. These included Wilms tumor gene (WT-I) [11], the paired-box gene, Pax-2 [12], oncogene N-myc [13], and the c-ret, which is a receptor for another important signalling protein implicated in kidney development, and the glial cell line-derived neurotrophic factor (GDNF) [14 17]. The defects in the Wnt-4 mutant kidneys became clear at stages when induced mesenchymal cells should have started to assemble into pretubular cell aggregates. This process failed completely without Wnt-4. In addition to the failure to form tubules, expression of Pax 8 (another gene expressed exclusively in the pretubular cells) was lost [9, 18]. Based on these results, the conclusion was drawn that the early steps of kidney development occurred normally, but the epithelial transformation failed when Wnt-4 activity was removed. It is possible that Wnt-4 may regulate nephrogenesis via the Pax genes, but this remains to be analyzed. Hence, it appeared that Wnt-4 was specifically required in differentiation of the pretubular cell aggregates that generate the nephrons. An additional feature in the Wnt-4 knock-out mice was that ureter branching appeared initially unchanged. This suggested that branching of the ureter was independent of mesenchymal Wnt-4 signalling and the pretubular aggregation process, a proposal that is also supported by the in vitro studies [19]. One function of Wnt-4 may be to regulate cell adhesion, which is apparently necessary for subsequent nephrogenesis. Candidate targets for Wnt-4 to mediate adhesion are cadherins and integrins [20] (Fig. 3). The function of cadherin is regulated by β-catenin, and both these genes are components of the Wnt signal transduction pathway [5]. Furthermore, β-catenin binds to the cytoplasmic domain of cadherin when the Wnt signal is not present, but binds to transcription factors including Lef- 1/TCFs when the pathway is activated by a Wnt signal [21]. We speculate that when Wnt-4 is not present in the kidney, cadherin expression is not activated and no adhesion or subsequent tubulogenesis occurs. Consistent with this proposal, Lef-1, β-catenin (our unpublished data), E- cadherin [22], and a recently identified cadherin, cadherin-6 [23], are all expressed in the pretubular cell aggregates and could be targets of Wnt-4 signalling. However, Wnt-4 signalling could be regulated by proteoglycans (PGs) such as syndecan-1, which correlates with Wnt-4 expression in the tubules [24]. As a support for this suggestion, PGs are required for Wnt-mediated activities [25, 26] and may function by localizing the Wnt signalling in the kidney. Recently, an integrin-linked kinase was identified and shown to be regulated by the Wnt pathway. This suggests that Wnt signalling may also play a role in controlling integrin-mediated cell adhesion [27]. Several Wnts are expressed in the spinal cord that induces the kidney tubules 153 Several gene family members, including Wnt-1, -3a, -3, -4, -7a, and -7b [28] are expressed in the spinal cord, which is a robust inducer of tubules. Recent data from a double knock-out of Wnt-1 and Wnt-3a showed neural crest defects thus indicating that Wnts are active signalling components in the developing central nervous system in the mouse [29]. Wnt-1 is expressed at the dorsal most part of the spinal cord, in the roof plate, whereas Wnt-4 is expressed in the dorsal half most corresponding to the piece used in the classic transfilter assay as a heterologous inducer tissue. This correlation raised the possibility that Wnt signals could be responsible for the tubule inductive property of the spinal cord. The hypothesis was directly tested by engineering cells that ectopically expressed various Wnts, such as Wnt-1, -3a, -4, -5a, -7a, -7b, and -11. Cells that expressed Wnt-1 [30], -3a, -4, -7a and -7b all induced tubules when cultured with the separated kidney mesenchyme (Fig. 1), whereas non-transfected cells or cells that expressed Wnt-5a or Wnt-11 did not [31]. These inductive cells were also sufficient to activate expression of a panel of markers for the differentiated tubules. Interestingly, it was also found that cells transfected with the cdna encoding Wnt-4 protein induced Wnt-4 gene expression in the kidney mesenchyme. These

4 154 Fig. 3A,B A model of the Wnt pathway and possible association of some of its components with kidney tubule induction. A The ureter tip may express a Wnt, such as Wnt-11, that transiently triggers the Wnt pathway in adjacent mesenchymal cells (m) (arrow). As in the canonical Wnt pathway, a Wnt signals through its frizzled (Fz) receptor and dishevelled (Dsh) to inactivate zeste white-3 (zw3, GSK-3β). As a consequence β-catenin (b-cat) concentration increases in the cytoplasm (c) and β-catenin binds to TCF, a T-cell factor. In the nucleus (n) this complex may lead to activation of target genes including those for the nephrogenesis [8]. Activation may involve autoregulated expression of Wnt-4 (circled arrow) and Pax-8 genes [9]. B The tubule induction step is followed by induced mitosis (arrow) and an aggregation phase between the induced mesenchymal cells that may be mediated in part by cadherin. Wnt signalling, in this case Wnt-4, may be inactivated by a secreted frizzled-related protein sfrp. As a result β-catenin levels are kept low through interactions with zw3, adenomatous polyposis coli (APC) and axin. Aggregation is essential for the subsequent nephrogenesis and fails without the Wnt-4 signal, for example when it is genetically removed from the system. (u ureter, m kidney mesenchyme, um uninduced mesenchyme) results supported the model that Wnt-4 may induce tubule formation by autoregulating its own expression in the mesenchyme. Kidney tubule induction mediated by Wnt signals is redundant The fact that several Wnts are expressed in the spinal cord and are capable of inducing tubules while only Wnt-4 is present in the mesenchyme indicated clearly that Wnt signalling in the kidney is redundant. The redundancy in Wnt signalling also became evident from studies in which Wnt-4-deficient spinal cord rescued Wnt-4-deficient kidneys in organ cultures [9]. The interpretation of these results was that the other Wnts that are still expressed in the Wnt-4 mutant spinal cord would serve as signals to replace Wnt-4. This is apparently sufficient to rescue the tubulogenesis of the mutant kidney mesenchyme. At present, we do not have evidence that another Wnt family member takes over Wnt- 4 function at the post-inductive stages, but we are unable to completely rule this out. Although the current genetic and in vitro data strongly suggest a role for the Wnts in mediating tubule induction, final proof of this would be tubule induction with a recombinant Wnt protein. The results raised the possibility that the kidney mesenchyme would either express multiple frizzleds, which are likely receptors for the Wnts [32, 33], or that several Wnts could mediate their signalling via one or more frizzleds and lead to tubulogenesis. Indeed, our analysis indicated that receptors studied (6 frizzleds) and some of their secreted frizzled-related proteins (sfrps) were present in the embryonic kidney. However, of the Frizzleds, only Frizzleds-2, Fz3 and -7 (Fz) may be considered as candidate receptors for Wnt-4 and sfrp-2 as an antagonist. This suggestion was based on the findings that their expression correlates with Wnt-4 gene expres-

5 sion (Fig. 2), and was dependent on Wnt-4 signalling (our unpublished data). What have we learned from the data? The likely answer is that although Wnt-4 is expressed in the tubules from their early stages of differentiation throughout their morphogenesis, Wnt-4 functions transiently to induce tubules. Hence, if the contact and migration hypothesis for the tubule inductive mechanisms holds [1], we hypothesize that Wnt-4-deficient nephrogenic cells would obtain their Wnt-4 signal from cells that express Wnt-4 on the other side of the filter, or from the other Wnts present in the recombinant spinal cord in the case of the mutant tissues. Such signalling appears sufficient to turn on the morphogenetic program that leads to tubule formation. Hence, in such a situation each cell would need to get in touch with the Wnt-4 signal and then adhere to form tubules. How can we explain the fact that several genes that are considered as markers for tubule induction were still expressed in the Wnt-4 knock-out kidneys in a situation where other experimental data point to an inductive role for Wnt-4? As introduced, the model for tubule induction was that the ureter bud would express the inductive signal to trigger tubule differentiation in the kidney mesenchyme, and it was assumed that the spinal cord would express the same or a similar signal. The genetic evidence clearly shows that, in addition to Wnt-4, there is at least one other essential signal for nephrogenesis. Hence, the function of this currently unknown ureter-derived signal in vivo is apparently to trigger the expression of Wnt-4, which may then induce tubules via its autoregulatory mechanism. Recent knock-out data from another gene, Emx-2, also indirectly supports a critical role for Wnt-4 in tubule induction and suggests a means of identifying the inducer. Emx-2 is a homeobox-containing transcription factor and is exclusively expressed in the ureter bud. In the Emx-2 knock-out kidney, development is perturbed early, Wnt-4 gene expression is not induced, and no tubules are formed [34, 35]. The nature of the inductive signals of the heterologous inducer tissue in the transfilter assay From the 1950s, the transfilter assay (Fig. 1) has been used to study the dynamics of tubule induction [1]. Spinal cord has often been used as a heterologous inducer tissue, as it is a more-potent inducer then the natural inducer, ureter bud. The data concerning Wnt-4 signal transmission have raised the possibility that spinal cord is an efficient inducer tissue, as it expresses several Wnt genes and especially the Wnt-4 signal. Hence, the interpretation of the results is that the spinal cord acts as a robust inducer tissue, as it expresses the mesenchymal, "downstream signal" in the form of Wnt-4 and a panel of other Wnts that are also sufficient to induce tubules [31]. In this model, the spinal cord would present the Wnt-4 to the mesenchymal cells and trigger tubules by its autoregulative property [9, 31]. As Wnt-4 is expressed normally in the mesenchyme, the spinal cord signals would directly use the Wnt-4 signalling pathway that is functional in the kidney mesenchyme to trigger tubulogenesis (Figs. 1, 3). Could a Wnt also operate in vivo as a primary ureter-derived inductive signal to trigger differentiation of the tubules? Wnt-11 is expressed precisely at the newly formed ureter tips and, for this reason, is a candidate primary inductive signal for tubules (i.e., a signal upstream of Wnt-4). As discussed, Wnt-11 expression is upregulated in the Wolffian duct prior to the formation of the ureter bud, and Wnt-11 expression precedes induction of Wnt-4 and Wnt- 7b gene expression. However, current evidence does not support a role for Wnt-11. In the same assay, where Wnt- 4 and the other Wnts induced tubules, Wnt-11 failed to do so [31]. Furthermore, knocking out Wnt-11 function by gene targeting did not show a phenotype in kidney development (Vainio et al., unpublished data) and excludes roles of Wnt-11 at present. It is expected that the ureter-derived inductive signal also operates indirectly or directly to turn on the Wnt pathway in mesenchymal cells to activate nephrogenesis. The Wnt signal transduction pathway involves several components, including Lef/TCFs, catenin, and cadherin. The current model of the Wnt signalling pathway [6] fits well into the molecular dynamics of tubule induction. Upon induction via a Wnt, a frizzled receptor may become activated and, via different steps, β-catenin would eventually bind to Lef-1/TCF and translocate into the nucleus to regulate expression of genes that contribute to nephrogenesis. This induction step is apparently transient in the kidney tubules. Inactivation of the Wnt pathway, possibly by members of the sfrps present in the kidney and loss of receptor for Wnt-4 ([36, 37, 38]; Vainio et al. unpublished data), would be expected to lead to accumulation of β-catenin. β-catenin would bind to the cytoplasmic domain of cadherins also present in assembling tubules to induce cell adhesion (Fig. 3) [22, 23]. Relevance of these findings to humans 155 In humans, a well-known hereditary kidney disease, autosomal dominant polycystic disease (ADPKD), causes slowly progressive epithelial cyst formation. Mutations in the PKD1 and PKD2 genes (encoding polycystin 1 and 2, respectively) are detected in most of these patients. Moreover, mice lacking a functional PKD1 gene show similar enlarged cystic kidneys to those observed in individuals having a mutation. Studies with mice homozygous for polycystin 1 gene disruption reveal that polycystin 1 is required for normal elongation and maturation of the tubular structures [39], a process triggered and regulated by Wnt-4 signalling. A more-direct suggestion that polycystic kidneys could have defects in Wnt signalling comes from recent studies where polycystin 1 was shown to modulate the Wnt signalling cascade by stabilizing en-

6 156 dogenous β-catenin and stimulating TCF-dependent gene transcription in vitro [40]. Hence, these studies raise the interesting possibility that in addition to being involved in neoplastic growth [5], the Wnt pathway may also be associated with kidney diseases. In summary, the identification of an important role for Wnt signalling in normal and defective nephrogenesis has paved the way for more-detailed study. This may offer new avenues for the future development of novel treatments for kidney diseases. Acknowledgements We thank Dr. Brian Norledge for comments. This work was financially supported by the Sigrid Jusélius Foundation, Academy of Finland, Ministry of Education, University of Oulu, and Biocenter Oulu. References 1. Saxén L (1987) Organogenesis of the kidney. Cambridge University Press, Cambridge, UK 2. Grobstein C (1955) Inductive interactions in the development of the mouse metanephros. J Exp Zool 130: Ekblom P, Miettinen A, Virtanen I, Wahlstrom T, Dawnay A, Saxen L (1981) In vitro segregation of the metanephric nephron. Dev Biol 84: Lehtonen E (1976) Transmission of signals in embryonic induction. Med Biol 54: Cadigan KM, Nusse R (1997) Wnt signaling: a common theme in animal development. Genes Dev 11: The Wnt Gene Homepage. wntwindow.html 7. Uusitalo M, Heikkilä M, Vainio S (1999) Molecular genetic studies of Wnt signalling in the mouse. Exp Cell Res 253: Kispert A, Vainio S, Shen L, Rowitch DH, McMahon AP (1996) Proteoglycans are required for maintenance of Wnt-11 expression in the ureter tips. Development 122: Stark K, Vainio S, Vassileva G, McMahon AP (1994) Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4. Nature 372: Skaer H (1996) Cell proliferation and development of the Malpighian tubules in Drosophila melanogaster. Exp Nephrol 4: Kreidberg JA, Sariola H, Loring JM, Maeda M, Pelletier J, Housman D, Jaenisch R (1993) WT-1 is required for early kidney development. Cell 74: Torres M, Gomez-Pardo E, Dressler GR, Gruss P (1995) Pax-2 controls multiple steps of urogenital development. Development 121: Mugrauer G, Alt FW, Ekblom P (1988) N-myc proto-oncogene expression during organogenesis in the developing mouse as revealed by in situ hybridization. J Cell Biol 107: Schuchardt A, D'Agati V, Larsson-Blomberg L, Costantini F, Pachnis V (1994) Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature 367: Robertson K, Mason I (1997) The GDNF-RET signalling partnership. Trends Genet 13: Sariola H, Sainio K (1997) The tip-top branching ureter. Curr Opin Cell Biol 9: Vainio S, Muller U (1997) Inductive tissue interactions, cell signaling, and the control of kidney organogenesis. Cell 90: Plachov D, Chowdhury K, Walther C, Simon D, Guenet J-L, Gruss P (1990) Pax8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110: Davies J, Lyon M, Gallagher J, Garrod D (1995) Sulphated proteoglycan is required for collecting duct growth and branching but not nephron formation during kidney development. Development 121: Falk M, Salmivirta L, Durbeej M, Larsson E, Ekblom M, Vestweber D, Ekblom P (1996) Integrin alpha 6B beta 1 is involved in kidney tubulogenesis in vitro. J Cell Sci 109: Clevers H, Wetering M van de (1997) TCF/LEF factor earn their wings. Trends Genet 13: Vestweber D, Kemler R, Ekblom P (1985) Cell-adhesion molecule uvomorulin during kidney development. Dev Biol 112: Cho EA, Patterson LT, Brookhiser WT, Mah S, Kintner C, Dressler GR (1998) Differential expression and function of cadherin-6 during renal epithelium development. Development 125: Vainio S, Lehtonen E, Jalkanen M, Bernfield M, Saxen L (1989) Epithelial-mesenchymal interactions regulate the stagespecific expression of a cell surface proteoglycan, syndecan, in the developing kidney. Dev Biol 134: Reichsman F, Smith L, Cumberledge S (1996) Glycosaminoglycans can modulate extracellular localization of the wingless protein and promote signal transduction. J Cell Biol 135: Hacker U, Lin X, Perrimon N (1997) The Drosophila sugarless gene modulates Wingless signaling and encodes an enzyme involved in polysaccharide biosynthesis. Development 124: Novak A, Hsu SC, Leung-Hagesteijn C, Radeva G, Papkoff J, Montesano R, Roskelley C, Grosschedl R, Dedhar S (1998) Cell adhesion and the integrin-linked kinase regulate the LEF- 1 and beta-catenin signaling pathways. Proc Natl Acad Sci U S A 95: Parr BA, Shea MJ, Vassileva G, McMahon AP (1993) Mouse Wnt genes exhibit discrete domains of expression in the early embryonic CNS and limb buds. Development 119: Ikeya M, Lee SM, Johnson JE, McMahon AP, Takada S (1997) Wnt signalling required for expansion of neural crest and CNS progenitors. Nature 389: Herzlinger D, Qiao J, Cohen D, Ramakrishna N, Brown AM (1994) Induction of kidney epithelial morphogenesis by cells expressing Wnt-1. Dev Biol 166: Kispert A, Vainio S, McMahon AP (1998) Wnt-4 is a mesenchymal signal for epithelial transformation of metanephric mesenchyme in the developing kidney. Development 125: Wang Y, Macke JP, Abella BS, Andreasson K, Worley P, Gilbert DJ, Copeland NG, Jenkins NA, Nathans J (1996) A large family of putative transmembrane receptors homologous to the product of the Drosophila tissue polarity gene frizzled. J Biol Chem 271: Bhanot P, Brink M, Harryman Samos C, Hsieh J-C, Wang Y, Macke JP, Andrew D, Nathans J, Nusse R (1996) A new member of the frizzled family from Drosophila functions as a Wingless receptor. Nature 382: Pellegrini M, Mansouri A, Simeone A, Boncinelli E, Gruss P (1996) Dentate gyrus formation requires Emx2. Development 122: Miyamoto N, Yoshida M, Kuratani S, Matsuo I, Aizawa S (1997) Defects of urogenital development in mice lacking Emx2. Development 124: Leyns L, Bouwmeester T, Kim SH, Piccolo S, DeRobertis EM (1997) Frzb-1 is a secreted antagonist of Wnt signaling expressed in the Spemann organizer. Cell 88: Leimeister C, Bach A, Gessler M (1998) Developmental expression patterns of mouse sfrp genes encoding members of the secreted frizzled related protein family. Mech Dev 75: Lescher B, Haenig B, Kispert A (1998) sfrp-2 is a target of the Wnt-4 signaling pathway in the developing metanephric kidney. Dev Dyn 213: Lu W, Peissel B, Babakhanlou H, Pavlova A, Geng L, Fan X, Larson C, Brent G, Zhou J (1997) Perinatal lethality with kidney and pancreas defects in mice with a targeted Pkd1 mutation. Nat Genet 17: Kim E, Arnould T, Sellin LK, Benzing T, Fan MJ, Grüning W, Sokol SY, Drummond I, Walz G (1999) The polycystic kidney disease 1 gene product modulates Wnt signaling. J Biol Chem 274: