Development of the metanephric kidney requires the concerted interaction of two tissues, the epithelium of the ureteric duct and the metanephric mesenchyme. Signals from the ureter induce the metanephric mesenchyme to condense and proliferate around the ureter tip, reciprocal signals from the mesenchyme induce the ureter tip to grow and to branch. Wnt genes encode secreted glycoproteins, which are candidate mediators of these signaling events. We have identified three Wnt genes with specific, non-over-lapping expression patterns in the metanephric kidney, Wnt-4, Wnt-7b and Wnt-11. Wnt-4 is expressed in the condensing mesenchyme and the comma- and S-shaped bodies. Wnt-7b is expressed in the collecting duct epithelium from 13.5 days post coitum onward. Wnt-11 is first expressed in the nephric duct adjacent to the metanephric blastema prior to the outgrowth of the ureteric bud. Wnt-11 expression in Danforth’s short-tail mice suggests that signaling from the mesenchyme may regulate Wnt-11 activation. During metanephric development, Wnt-11 expression is confined to the tips of the branching ureter. Maintenance of this expression is independent of Wnt-4 signaling and mature mesenchymal elements in the kidney. Moreover, Wnt-11 expression is maintained in recombinants between ureter and lung mesenchyme suggesting that branching morphogenesis and maintenance of Wnt-11 expression are independent of metanephric mesenchyme-specific factors. Interference with proteoglycan synthesis leads to loss of Wnt-11 expression in the ureter tip. We suggest that Wnt-11 acts as an autocrine factor within the ureter epithelium and that its expression is regulated at least in part by proteoglycans.

The development of vertebrate organs requires intricate cell-cell communications to assure the concerted program of cell growth, differentiation and morphogenesis. Renal development is characterized by an interaction between an epithelial and a mesenchymal compartment: the nephric (Wolffian) duct and the nephrogenic mesenchyme, both derived from the intermediate mesoderm. In mammals, this interaction results in the formation of three embryonic kidneys, the pronephros, the mesonephros and the metanephros, the latter of which is responsible for removal of nitrogenous waste and physiological control of salt balance.

Metanephric development in the mouse begins when the ureteric bud, an outgrowth of the nephric duct, contacts the undifferentiated metanephric blastema around 10.5 days post coitum (days p.c.). Signals from the ureter induce the metanephric mesenchyme to condense and proliferate around the ureter tip, whilst reciprocal signals from the mesenchyme induce the ureter tip to grow and branch (Grobstein, 1953, 1955). The induced mesenchyme then aggregates and undergoes an epithelial transformation. Cytodifferentiation and interaction with the vasculature results in the formation of the glomerular and tubular epithelia of the nephron. In turn, the ureter gives rise to the collecting duct system of the kidney (for reviews see Saxen, 1987; Bard et al., 1994; Patterson and Dressler, 1994).

The development of transfilter culture techniques using isolated metanephric mesenchyme has allowed the partial characterization of the requirements for ureter morphogenesis and metanephric induction. Organ culture experiments have demonstrated that the extracellular matrix is required for branching morphogenesis in the kidney (Ekblom, 1981; Roskelly et al., 1995). Chemicals known to interfere with proteoglycan metabolism partially or completely abolish ureter morphogenesis (Davies et al., 1995; Klein et al., 1989; Lelongt et al., 1988; Platt et al., 1987). A similar result was reported when polyanionic compounds were added to kidney cultures (Ekblom et al., 1978). Further, neonatal mouse kidney epithelial cells undergo morphogenesis in matrigel supplemented with FGF (Traub et al., 1990), and Madin-Darby canine kidney epithelial cells can form tubules in collagen gel in the presence of hepatocyte growth factor (Montesano et al., 1991).

The induction by the ureter and other heterologous inducers, most notably spinal cord, depends on cell contacts. The induction process itself is relatively slow and cannot be transferred by induced mesenchymal cells to uninduced cells (Grobstein, 1953, 1955, 1956; Saxen and Lehtonen, 1978; Saxen et al., 1976; Unsworth and Grobstein, 1970; Wartiovara et al., 1974). Although many growth factors have been identified in the kidney (Hammerman, 1995), none of these has been shown to individually induce tubulogenesis. A combination of FGF2 and a pituitary extract can induce tubulogenesis, suggesting that tubule induction is a multistep process mediated by soluble and insoluble factors (Perantoni, 1991; Perantoni et al., 1995).

Recently, it was shown that members of the Wnt family are both sufficient and necessary for tubule development. Wnt-4 is expressed in the condensing mesenchyme and the comma- and S-shaped bodies in the metanephric kidney. Gene targeting in ES cells has demonstrated that Wnt-4 acts as an autoinducer of mesenchymal aggregation (Stark et al., 1994). Cells expressing Wnt-1 can induce tubule formation in isolated metanephric mesenchyme (Herzlinger et al., 1994). Wnt-1 is not expressed in the kidney, but is expressed in the dorsal spinal cord, the strongest known inducer of tubulogenesis (Wilkinson et al., 1987). It is conceivable that Wnt-1 mimics the activity of a Wnt family member specifically expressed in the kidney, either Wnt-4 in the mesenchyme or an unknown Wnt in the ureter.

In order to gain insight in the complexity of Wnt gene function in the developing kidney, we have systematically analyzed expression of all available members of the mouse Wnt gene family (14 in total) (Lee et al., 1995) in the metanephric kidney. Here we report that, in addition to Wnt-4 (Stark et al., 1994), Wnt-7b and Wnt-11 show specific sites of non-overlapping expression. Wnt-7b is expressed in the maturing collecting ducts, while Wnt-11 expression is restricted to the ureter tips. We provide evidence that activation of Wnt-11 expression in the nephric duct depends on signals from the metanephric blastema. Maintenance of expression in the ureter tips, however, is independent of specific metanephric mesenchymal signals as Wnt-11 expression persists as the ureter undergoes morphogenesis in heterotypic mesenchyme. Proteoglycans appear to play a major role in the cell or matrix interactions that regulate Wnt-11 expression.

Mouse stocks

Mice heterozygous for the Danforth’s short-tail (Sd) mutation were purchased from the Jackson Laboratory. The mutation was maintained on the background RSV/Le Re/Re Sd/+ Va/+, (re-rex, Va-varitint waddler). Heterozygotes were identified by an absent or rudimentary tail. Wnt-4 heterozygotes were derived and genotyped as described previously (Stark et al., 1994). Embryos for the analysis of the normal pattern of Wnt-11 expression and for kidney dissections were derived from matings of Swiss Webster wild-type animals (purchased from the Jackson Laboratory). For timed pregnancies, plugs were checked in the morning after mating, noon was taken as 0.5 days of gestation.

Isolation and sequencing of Wnt-11 cDNAs

A Wnt-11 cDNA fragment corresponding to the highly conserved exon 4 was cloned using a PCR amplification strategy on 9.5 days p.c. embryonic cDNA (Gavin et al., 1990). The Wnt-11 PCR-fragment was labeled with [32P]CTP by random priming (GibcoBRL Random Prime Labeling Kit) and used to screen a newborn mouse kidney cDNA library (Wada et al., 1993). Hybridization was performed under stringent conditions in Church buffer at 68°C (Church and Gilbert, 1984). Approximately 1×106 plaques were screened and 19 positive plaques purified. 11 independent cDNAs were subcloned into pBlue-script SKII using the ExAssist System (Stratagene). cDNAs were sequenced (Sanger et al., 1977) with a Sequenase Kit (USB).

In situ hybridization

Whole-mount in situ hybridization was performed as described by Parr et al. (1993) and modified according to Knecht et al. (1995). Digoxigenin probes were synthesized using the Digoxigenin RNA Labeling Kit (Boehringer Mannheim). A probe to Wnt-11 was generated from the full-length cDNA pnk3 by linearization with XhoI and transcription with T3 RNA polymerase. A probe corresponding to approximately 1.2 kb of the 3′-part of the coding region of c-ret was generated from the plasmid pmc-ret (a kind gift of F. Constantini) by linearization with BamHI and transcription with T3 RNA polymerase. A probe corresponding to nucleotides 3612-4951 of the c-ros cDNA was obtained by linearizing the plasmid pmc4 (a kind gift of C. Birchmaier) with XbaI and transcribing with T7 RNA polymerase. Riboprobes were used at approximately 1 μg/ml.

Stained embryos, kidneys and cultures were transferred into 80% glycerol and photographed on Ektachrome Tungsten 64 color slide film (Kodak) using a Leica Wild M10 photomicroscope. Composites were generated using Adobe Photoshop v3.0 and Canvas v3.5 on a Power Macintosh.

In situ hybridization analysis on sections was performed according to published procedures (Wilkinson et al., 1987). Generation of PCR fragments used as in situ probes was described by Gavin et al. (1990). The in situ probes for Wnt-4 and Wnt-7b were described before (Parr et al., 1993), the Wnt-11 riboprobe was prepared from the PCR-fragment.

Organ culture techniques

Metanephric kidneys and lungs from 11.5 days p.c. wild-type Swiss Webster embryos were dissected in PBS. In some cases, parts of the mesonephric duct derivatives were left attached to increase the size of the specimen, which facilitated later in situ hybridization analysis. For recombination experiments, ureters were manually separated from metanephric mesenchyme, and lung mesenchyme from lung epithelium, following a 2 minute incubation in 3% pancreatin/trypsin (GibcoBRL) in Tyrode’s solution. Metanephric kidneys were placed on 4–6 mm diameter 0.1 μm Nucleopore filters (Costar) supported by stainless steel grids on the surface of the culture medium (Dulbecco’s modified Eagle’s medium (Sigma #D5796) supplemented with 10% fetal calf serum (Hyclone), 2 mM glutamine (GibcoBRL #1273), 1× penicillin/streptomycine (GibcoBRL #0511). Medium was modified with one or more of the following chemicals: 30 mM NaClO3 (Sigma #S3171), 10-100 μg/ml pentosan polysulphate (Sigma #P8275), 50 μM suramin (CB Chemicals, Woodbury, CT), 100 μg/ml chondroitin-6-sulphate from shark cartilage (ICN), 100 μg/ml low molecular weight heparin (CalBiochem), chondroitinase ABC (Sigma #C2905), heparitinase III (Sigma #H8891). In recombination experiments, lung mesenchyme was placed adjacent to one or two ureters or ureter tips on one filter. Samples were incubated in 5% CO2 in air at 37°C. For all kidney culture experiments, a minimum of 6, in most cases 12, specimens were processed.

For in situ hybridization analysis, filters were submerged in cold methanol for 10 seconds and then fixed in 4% paraformaldehyde in PBS overnight prior to stepwise transfer into methanol and storage at –20°C.

Whole-mount β-galactosidase staining

Metanephric kidney cultures from transgenic experiments were processed for β-galactosidase activity according to published procedures (Whiting et al., 1991). The cultures were stained for periods ranging from 30 minutes to 12 hours according to the strength of expression. The reaction was stopped by washing in PBS and post-fixation in 4% paraformaldehyde/PBS overnight. Specimen were then processed through a graded series into 80% glycerol and photographed as described.

Identification of Wnt genes with specific expression patterns in the metanephric kidney

In order to identify Wnt genes specifically expressed in the metanephric kidney, PCR fragments from all available mouse Wnt genes (14 so far) were used to generate antisense ribo-probes for in situ hybridizations on sections of 14.5 days p.c. metanephric kidneys. Three Wnt genes with specific and non-overlapping expression domains were identified: Wnt-4, Wnt-7b and Wnt-11 (Christiansen et al., 1995).

In the 14.5 days p.c. metanephric kidney, Wnt-4 is expressed in the mesenchymal condensates, their simple epithelial derivatives, the comma- and S-shaped bodies, and in the centrally located stroma. Wnt-7b transcripts are restricted to the collecting ducts and Wnt-11 expression to the tips of the ureter (Fig. 1). The expression and requirement for Wnt-4 in kidney development has been addressed earlier (Stark et al., 1994), the expression of Wnt-7b in the collecting ducts makes it unlikely that it is involved in inductive events between the ureter and the mesenchyme. In contrast, Wnt-11 expression correlates with sites of epithelial-mesenchymal interactions, and growth and branching morphogenesis of the ureter. Wnt-11 is the first secreted protein with such an expression pattern making it a candidate mediator of these phenomena. We therefore focused our work on the characterization of Wnt-11.

Fig. 1.

Expression of Wnt genes in the metanephric kidney. In situ hybridization with [35S] ribopobes was used to detect specific expression of Wnt-4 (A), Wnt-7b (B) and Wnt-11 (C) in sections of a 14.5 days p.c. metanephric kidney. Wnt-4 is expressed in the mesenchymal condensates and the comma- and S-shaped bodies, and in the centrally located stroma, Wnt-7b in the collecting ducts and Wnt-11 in the ureter tips.

Fig. 1.

Expression of Wnt genes in the metanephric kidney. In situ hybridization with [35S] ribopobes was used to detect specific expression of Wnt-4 (A), Wnt-7b (B) and Wnt-11 (C) in sections of a 14.5 days p.c. metanephric kidney. Wnt-4 is expressed in the mesenchymal condensates and the comma- and S-shaped bodies, and in the centrally located stroma, Wnt-7b in the collecting ducts and Wnt-11 in the ureter tips.

In situ hybridization analysis of Wnt-11 expression during murine development

Prior to urogenital development, we detect Wnt-11 expression first in a punctate pattern in the posterior half of the embryo at 6.75 days p.c., where the primitive streak has formed (Fig. 2A). At 7 days p.c., two expression domains can clearly be distinguished. The node and, at the posterior end of the embryo, the extraembryonic mesoderm and base of the allantois (Fig. 2B,C). Expression in the node disappears at around 8 days p.c. whereas the expression in the posterior trunk persists in the tailbud until at least 11.5 days p.c. (Fig. 2D-H). At 7.75 days p.c. (before somite formation), we detect Wnt-11 transcripts in the anlage of the heart, which is located rostral-lateral to the invaginating foregut anlage (Fig. 2D). Expression in the forming heart tube continues until at least 9.5 days p.c. and is confined to the myocardial layer (not shown). At around 8.25 days p.c., Wnt-11 expression is detected in the most rostral somites at the dorsalmost aspect, probably marking the future dermatome (Fig. 2E). At 9.5 days p.c., expression is weak in the newly forming and stronger in more mature somites (Fig. 2F,G). During tail formation (11.5 days p.c.), expression is very strong in the newly forming somites at the tail tip. Rostrally, expression becomes progressively restricted to the caudal half of the somite (Fig. 2H). At 10.5 days p.c., we detect Wnt-11 in the branchial arches (Fig. 2H and not shown). Wnt-11 shows a particular dynamic expression pattern in the limb bud. Wnt-11 expression accompanies the emerging forelimb bud at 9.25 days p.c. (Fig. 2G). An initially broad distal domain of staining is later confined to a distal ectodermal strip of cells identical to the apical ectodermal ridge. Finally, from 10.5 to 11 days p.c., expression is observed in the condensing skeletal rudiments (Fig. 2H) and later in the perichondrium (data not shown).

Fig. 2.

Spatial expression of Wnt-11 between 6.75 days and 11.5 days p.c. of mouse development as detected by whole-mount in situ hybridization. (A) 6.75 days p.c., weak Wnt-11 expression in scattered cells in the posterior half of the embryo. (B) 7.0 days p.c., onset of expression in the node and in the extraembryonic mesoderm including the allantois. (C) 7.5 days p.c., enhanced expression in these two regions. (D) Wnt-11 expression comes on in the primitive heart anlage, and is maintained in the posterior region. (E) 8.5 days p.c., Wnt-11 is strongly expressed in the heart tube and the posterior trunk, weak expression in the newly forming somites. (F) 9 days p.c., Wnt-11 is expressed along the length of the mesonephric duct. (G) Wnt-11 expression accompanies the outgrowth of the forelimb buds, (H) 10.5 days p.c., expression is maintained in the somites, in the apical ectodermal ridge and the proximal region of the limb and is initiated in the first and second branchial arch. no, node; h, primitive heart anlage; nd, mesonephric duct.

Fig. 2.

Spatial expression of Wnt-11 between 6.75 days and 11.5 days p.c. of mouse development as detected by whole-mount in situ hybridization. (A) 6.75 days p.c., weak Wnt-11 expression in scattered cells in the posterior half of the embryo. (B) 7.0 days p.c., onset of expression in the node and in the extraembryonic mesoderm including the allantois. (C) 7.5 days p.c., enhanced expression in these two regions. (D) Wnt-11 expression comes on in the primitive heart anlage, and is maintained in the posterior region. (E) 8.5 days p.c., Wnt-11 is strongly expressed in the heart tube and the posterior trunk, weak expression in the newly forming somites. (F) 9 days p.c., Wnt-11 is expressed along the length of the mesonephric duct. (G) Wnt-11 expression accompanies the outgrowth of the forelimb buds, (H) 10.5 days p.c., expression is maintained in the somites, in the apical ectodermal ridge and the proximal region of the limb and is initiated in the first and second branchial arch. no, node; h, primitive heart anlage; nd, mesonephric duct.

Expression in the urogenital system is first detected in the mesonephric duct at around 9 days p.c. at a very low level (Fig. 2F). Expression follows the posterior elongation of the duct (Fig. 2G). At 10.5 days p.c., the mesonephric duct has reached the position of the metanephric mesenchyme which is located at approximately the level of the hindlimb buds. At this time, we see a strong Wnt-11 signal in the epithelium of the region of the nephric duct that is facing the metanephric mesenchyme (Fig. 3A). Thus, expression is asymmetric in the nephric duct at this time. This expression domain demarcates the region of the nephric duct that will form the ureter bud a few hours later. Expression of Wnt-11 is confined to the rounded tip of the invading ureter whereas the stalk region is devoid of signal (Fig. 3B). At 11 days p.c., the ureter bud branches for the first time and the Wnt-11 expression domain is split in two (Fig. 3C). Consequently, when the first branching event is completed at 11.5 days p.c., Wnt-11 expression is again confined to the tip of both branches of the ureter (Fig. 3D). At more advanced stages of metanephric development, the inductive interactions take place in a narrow zone in the periphery of the kidney. Whole-mount in situ hybridization of 12.5 to 18.5 days p.c. kidneys shows a spotted pattern of Wnt-11 expression in a surface view of the kidney (Fig. 3E). Sections reveal that Wnt-11 expression is precisely confined to the growing tips of the ureter in this peripheral region where nephrogenesis is continuing (Fig. 3F).

Fig. 3.

Wnt-11 expression in the metanephric kidney. Posterior trunk regions of embryos and isolated metanephric kidneys were stained as whole mounts for Wnt-11 expression. (A) Wnt-11 expression is first seen at 10.5 days p.c. in the epithelium of the ureter adjacent the metanephric mesenchyme. (B,C) With the ingrowth of the ureteric bud (10.75 days p.c., 11 days p.c.), the signal is only detected at the rounded tip. (D) The expression domain splits in two after the bifurcation of the ureter has been completed at 11.5 days p.c. (E) Wnt-11 expression as seen in 16.5 days p.c. metanephric kidney stained as a whole mount. (F) Sectioning reveals that expression is confined to the tips of the growing ureter in the nephrogenic zone

Fig. 3.

Wnt-11 expression in the metanephric kidney. Posterior trunk regions of embryos and isolated metanephric kidneys were stained as whole mounts for Wnt-11 expression. (A) Wnt-11 expression is first seen at 10.5 days p.c. in the epithelium of the ureter adjacent the metanephric mesenchyme. (B,C) With the ingrowth of the ureteric bud (10.75 days p.c., 11 days p.c.), the signal is only detected at the rounded tip. (D) The expression domain splits in two after the bifurcation of the ureter has been completed at 11.5 days p.c. (E) Wnt-11 expression as seen in 16.5 days p.c. metanephric kidney stained as a whole mount. (F) Sectioning reveals that expression is confined to the tips of the growing ureter in the nephrogenic zone

Regulation of Wnt-11 expression in the metanephric kidney

Activation of Wnt-11 expression

The upregulation of Wnt-11 expression in the epithelium contacting the metanephric blastema suggests that signals from the metanephric mesenchyme are involved in Wnt-11 activation. In order to address this issue, we studied Wnt-11 expression in embryos homozygous for the Danforth’s short-tail (Sd) mutation. Heterozygotes are characterized by a short or absent tail and the occasional absence of one kidney. Homozygous mutant embryos show a severe truncation of the axis and gross urogenital defects. Kidneys are hypoplastic or agenic, a phenotype that was attributed to a failure of metanephric mesenchyme induction due to a delay or absence of ureter outgrowth (Gluecksohn-Schoenheimer, 1943, 1945). Rather than influencing the ureter directly, the Sd mutation most likely interferes with the necessary proximity of the nephric duct and the metanephric blastema, which is required for outgrowth of the ureteric bud (Gluecksohn-Waelsch and Rota, 1963).

We harvested embryos at 11-11.5 days p.c. when homozygous Sd embryos are readily distinguishable due to the occurrence of edemas in the tail tip and spina bifida in the posterior trunk. Heterozygotes show only a slightly thinned tail at this time (Gluecksohn-Schoenheimer, 1945). The frequency of these phenotypes was as expected for a Mendelian trait. From 46 embryos isolated in total, 16 were classified as wild-type, 26 as Sd/+ and 13 as Sd/Sd. All wild-type embryos showed Wnt-11 expression in the tips of the bifurcated ureter. From the 26 embryos scored as Sd/+, 12 had a wild-type staining pattern, 5 had bilateral but unequal staining, 5 showed unilateral staining only and 4 had very weak or no detectable staining. In the homozygous mutants we observed three different phenotypes. In 8 cases we did not detect any staining but the ureter was absent; in three cases, we noticed a unilateral staining at the tip of a long unbranched ureter stalk and, in 2 cases, staining was weak but bilateral (Fig. 4). These results reveal that Wnt-11 expression is not strictly dependent on Sd. They are compatible with the notion that Sd does not affect kidney development directly but rather disturbs (secondarily) the proximity between the ductal elements and the metanephric mesenchyme. The missing contact between metanephric blastema and the nephric duct prevents ureter outgrowth and Wnt-11 activation.

Fig. 4.

Wnt-11 expression in Danforth’s short-tail mice. (A,B) Wnt-11 expression in the posterior trunk and kidneys of Sd/+ embryos. Wnt-11 expression is lost in the most caudal somites, but still present in the ureter tips (see arrow). (C-H) Wnt-11 expression in the posterior trunks and kidneys of Sd/Sd embryos. All embryos are characterized by large edema in the tail. Wnt-11 expression in the somite is partially or completely lost, but persists in the tail bud. Wnt-11 expression is found in the tip of the ureter on both sides (D), on one side (F) or is absent (H) (see arrow).

Fig. 4.

Wnt-11 expression in Danforth’s short-tail mice. (A,B) Wnt-11 expression in the posterior trunk and kidneys of Sd/+ embryos. Wnt-11 expression is lost in the most caudal somites, but still present in the ureter tips (see arrow). (C-H) Wnt-11 expression in the posterior trunks and kidneys of Sd/Sd embryos. All embryos are characterized by large edema in the tail. Wnt-11 expression in the somite is partially or completely lost, but persists in the tail bud. Wnt-11 expression is found in the tip of the ureter on both sides (D), on one side (F) or is absent (H) (see arrow).

Maintenance of Wnt-11 expression in the ureter tips

After the initial activation in the nephric duct, Wnt-11 expression is strictly confined to the tips of the ureter as long as branching morphogenesis and mesenchymal induction occur. As reciprocal signals from the induced mesenchyme undergoing tubulogenesis are thought to regulate ureter development (for review see Saxen, 1987), it is conceivable that mesenchymal factors are required to maintain and restrict Wnt-11 expression to this region of the ureter.

Mice homozygous for the Wnt-4 mutation fail to form pretubular cell aggregates and, consequently, do not develop more mature mesenchymal structures (Stark et al., 1994). We analyzed Wnt-11 expression in the Wnt-4 mutant background in whole kidneys and in sections by in situ hybridization at 12.5 and 14.5 days p.c. (Fig. 5). Wnt-11 expression persists at the tips of the ureter until at least 14.5 days p.c. though, at 14.5 days p.c., the tips are very much reduced in number and they do not show the half moon form indicative of tip branching observed in wild-type kidneys (Fig. 5D). Thus, Wnt-11 expression is independent of Wnt-4 signaling and also of signals provided by pretubular cell aggregates or epithelial bodies. However, the clear reduction of ureter tips as noted previously (Stark et al., 1994) suggests that the fine tuning of ureteric growth and branching critically depends on the intact architecture of the mesenchymal and epithelial elements.

Fig. 5.

Wnt-11 expression in kidneys mutant for Wnt-4. Wnt-11 is expressed in the ureter tips in the mutants, although the number of branches is clearly decreased in comparison to the wild-type. (A) 12.5 days p.c. +/+, (B) 12.5 days p.c. −/−, (C) 14.5 days p.c. +/+, (D) 14.5 days p.c. −/−. Embryos were derived from matings of Wnt-4 heterozygotes and genotyped individually, kidneys were stained as whole-mounts for Wnt-11 expression.

Fig. 5.

Wnt-11 expression in kidneys mutant for Wnt-4. Wnt-11 is expressed in the ureter tips in the mutants, although the number of branches is clearly decreased in comparison to the wild-type. (A) 12.5 days p.c. +/+, (B) 12.5 days p.c. −/−, (C) 14.5 days p.c. +/+, (D) 14.5 days p.c. −/−. Embryos were derived from matings of Wnt-4 heterozygotes and genotyped individually, kidneys were stained as whole-mounts for Wnt-11 expression.

Earlier reports indicated that metanephric mesenchyme is unique in its ability to support growth and branching of ureteric epithelium (Bishop-Calame, 1965a,b; Saxen, 1987). We recombined ureters isolated at 11.5 days p.c. after the first branching (‘T-stage’) with lung mesenchyme from the same embryo. Unexpectedly, we found that lung mesenchyme is able to support branching morphogenesis of the ureter though with a pattern more reminiscent of lung epithelium (Fig. 6A). We believe this effect was not due to contaminating metanephric mesenchyme since isolated ureters degenerated rapidly when cultured on nucleopore filters and tubule induction never occurred in our cultures. We obtained identical results when we recombined ureter tips with lung mesenchyme (data not shown) suggesting that the growth and branching potential of the ureter resides in this morphologically distinct structure.

Fig. 6.

Growth and branching of the ureter occurs in a recombinant with lung mesenchyme. (A) Isolated ureters grow and branch excessively in lung mesenchyme as seen in a bright-field image. (C) Wnt-11 expression and (D) c-ret expression are confined to the rounded tips of the ureter; (B) the ureter still expresses Pax-2. 11.5 days p.c. ureters (bifurcated once) and lung mesenchyme from the same embryo were recombined on a nucleopore filter. Specimen were grown for 2 days (C,D) and 4 days (A,B), respectively, before subjecting them to β-gal staining (A) or whole-mount in situ hybridization (C,D).

Fig. 6.

Growth and branching of the ureter occurs in a recombinant with lung mesenchyme. (A) Isolated ureters grow and branch excessively in lung mesenchyme as seen in a bright-field image. (C) Wnt-11 expression and (D) c-ret expression are confined to the rounded tips of the ureter; (B) the ureter still expresses Pax-2. 11.5 days p.c. ureters (bifurcated once) and lung mesenchyme from the same embryo were recombined on a nucleopore filter. Specimen were grown for 2 days (C,D) and 4 days (A,B), respectively, before subjecting them to β-gal staining (A) or whole-mount in situ hybridization (C,D).

To determine whether the ureteric epithelium preserved its molecular identity under these conditions, we examined Pax-2 expression in the explants. Pax-2 encodes a transcription factor that is both expressed in the ureter and required for urogenital development (Dressler et al., 1990; Torres et al., 1995). We used a transgenic line that expresses β-galactosidase exclusively in the nephric duct, its derivatives and the ureter under the control of Pax-2 genomic sequences (A. Kispert, D. H. Rowitch and A. P. McMahon, unpublished observations). We detected β-galactosidase activity in the epithelium until at least day 6 of culture suggesting that the epithelium retains is ureteric identity in the recombinants (Fig. 6B). Moreover, Wnt-11 expression is also maintained and localized to the tips for at least 4 days (Fig. 6C). Intriguingly, c-ret, a receptor tyrosine kinase which is normally coexpressed with Wnt-11 at the ureteric tips remains coexpressed with Wnt-11 in the recombinants suggesting a common mechanism of regulation (Fig. 6D). Thus, maintenance of Wnt-11 and c-ret expression in the ureter tips is not kidney mesenchyme specific. Only a general permissive environment may be required to maintain ureter viability and Wnt-11 transcription.

Requirement of proteoglycans for Wnt-11 expression

We were interested to investigate whether the extracellular matrix (ECM), and in particular proteoglycans (macromolecules consisting of a protein moiety and a covalently attached often highly sulphated glycosaminoglycan (GAG) (Kjellen and Lindahl, 1991)), could be an essential factor in the maintenance of Wnt-11 expression. Proteoglycans have been shown to be expressed both in the ureter and the induced mesenchyme (Vainio et al., 1989) and are required for ureter branching and tubule induction, but not for nephron maturation (Davies et al., 1995; Platt et al., 1987). We addressed this problem by administering compounds known to interfere with the integrity and function of proteoglycans, to kidney rudiment cultures, taken at 11.5 days p.c. when only one branch of the ureter had formed (T-stage). After 24 hours in control medium, the ureter had branched several times and tubule induction was evident by the appearance of mesenchymal condensates (Fig. 7A,B). Expression of Pax-2 (Fig. 7B), Wnt-11 (Fig. 7C) and the receptor tyrosine kinases c-ret and c-ros (not shown) confirmed that growth and branching of the ureter in vitro approximated the in vivo condition. For all experimental treatments, at least 6, in most cases 12, kidney rudiments were cultured and processed for any given concentration or time point.

Fig. 7.

Sulphated proteoglycans are required for Wnt-11 expression. All kidney rudiments were isolated at 11.5 days p.c. and grown in culture for various times with supplementation of the culture medium as indicated. Cultures were fixed and stained as whole mounts for Wnt-11 expression using in situ hybridization or probed for Pax-2 expression using a β-galactosidase assay. (A) Bright-field image of a 48 hour metanephric kidney culture. Ureter branching and induced tubules are clearly visible. (B) Pax-2 staining outlines the regular pattern of ureter branching in a kidney rudiment cultured for 48 hours. The ampulla-like ureter tips are evident. (C) Wnt-11 expression marks the numerous ureter tips after 48 h of growth in culture. (D,E) Kidney rudiments grown in medium supplemented with 30 mM NaClO3 for 24 hours do not show ureter growth. Wnt-11 expression is lost in the ureter tips (*, but not in the mesonephric duct to the left, D), whereas Pax-2 expression can be detected even after 96 h in culture (E). (F,G) Kidney rudiments were grown for 24 hours in medium containing 30 mM NaClO3 and were then transferred into standard medium for 48 hours. Numerous induced tubules and clear condensates were scored. Wnt-11 expression is very spotty and irregular (F), the ureter branching pattern is very disturbed as evidenced by Pax-2 expression (G). (H,I) Chondroitin-6-sulphate at 100 μg/ml had no effect on Wnt-11 expression (H) and ureter growth and branching as seen with the Pax-2 marker (I) after 24 hours culture. (J,K) Supplementation of the medium with 100 μg/ml heparin leads to a drastic downregulation of Wnt-11 expression after 24 hours, and almost a halt in ureter growth (J). After 72 hours, Pax-2 expression uncovers a severely growth-retarded ureter with very limited branching (K). Kidney rudiments grown in medium containing 50 μM suramin (L,M) and 100 μg/ml pentosan polysulphate (N,O), respectively, have completely lost Wnt-11 expression at the ureter tips after 24 hours (L,N). Pax-2 expression persists after 72 hours in the completely growth-arrested ureter.

Fig. 7.

Sulphated proteoglycans are required for Wnt-11 expression. All kidney rudiments were isolated at 11.5 days p.c. and grown in culture for various times with supplementation of the culture medium as indicated. Cultures were fixed and stained as whole mounts for Wnt-11 expression using in situ hybridization or probed for Pax-2 expression using a β-galactosidase assay. (A) Bright-field image of a 48 hour metanephric kidney culture. Ureter branching and induced tubules are clearly visible. (B) Pax-2 staining outlines the regular pattern of ureter branching in a kidney rudiment cultured for 48 hours. The ampulla-like ureter tips are evident. (C) Wnt-11 expression marks the numerous ureter tips after 48 h of growth in culture. (D,E) Kidney rudiments grown in medium supplemented with 30 mM NaClO3 for 24 hours do not show ureter growth. Wnt-11 expression is lost in the ureter tips (*, but not in the mesonephric duct to the left, D), whereas Pax-2 expression can be detected even after 96 h in culture (E). (F,G) Kidney rudiments were grown for 24 hours in medium containing 30 mM NaClO3 and were then transferred into standard medium for 48 hours. Numerous induced tubules and clear condensates were scored. Wnt-11 expression is very spotty and irregular (F), the ureter branching pattern is very disturbed as evidenced by Pax-2 expression (G). (H,I) Chondroitin-6-sulphate at 100 μg/ml had no effect on Wnt-11 expression (H) and ureter growth and branching as seen with the Pax-2 marker (I) after 24 hours culture. (J,K) Supplementation of the medium with 100 μg/ml heparin leads to a drastic downregulation of Wnt-11 expression after 24 hours, and almost a halt in ureter growth (J). After 72 hours, Pax-2 expression uncovers a severely growth-retarded ureter with very limited branching (K). Kidney rudiments grown in medium containing 50 μM suramin (L,M) and 100 μg/ml pentosan polysulphate (N,O), respectively, have completely lost Wnt-11 expression at the ureter tips after 24 hours (L,N). Pax-2 expression persists after 72 hours in the completely growth-arrested ureter.

Chlorate competes with sulphate in the synthesis of phospho-adenosine-5′-phosphosulphate, the sulphate donor used by sulphotransferases in the sulphation of polysaccharides (Farley et al., 1978). The addition of chlorate to the culture medium therefore selectively probes for the requirement of the glycosaminoglycan moiety of proteoglycans. Recently, Davies et al. (1995) reported that ureteric bud growth and branching in kidney rudiments is reversibly inhibited by chlorate in the culture medium. Surprisingly, nephron development is apparently unaffected under these conditions. Inhibition is concentration dependent and correlates quantitatively with the deprivation of sulphated GAGs. Chlorate concentrations between 15 mM and 30 mM were found to block growth and branching of the ureter completely. Ureters exhibited a T-shape even after 96 hours in culture. Nephron maturation in these cultures continued seemingly unaffected. In 30 mM chlorate, Wnt-11 expression was completely abolished after 24 hours, whereas the ureteric marker Pax-2 was expressed until at least 96 hours of incubation (Fig. 7D,E). Davies et al. (1995) noted that the effect on the ureter is reversible upon removal of the chlorate after 30 hours or competition with a low concentration of sulphate. However, in our hands, we observed major defects in ureter growth and branching after only 18 hours incubation in chlorate, but we did notice a massive wave of new tubule induction after transfer into fresh medium. Branching was highly ‘compressed’ and ureter shape disturbed as demonstrated by Pax-2 expression (Fig. 7G). Wnt-11 was expressed in few distinct spots and some smeary background (Fig. 7F). Kidney rudiments cultured for 48 hours in medium containing 30 mM chlorate retained their T-shape after another 48 hours incubation in standard medium but never showed Wnt-11 expression (not shown).

Chondroitin sulphate and heparin/heparan sulphate are major constituents of the proteoglycan complex. Both sulphated proteoglycans can be selectively degraded with the specific glycanases chondroitinase ABC and heparitinase III, respectively (Davies et al., 1995). 0.5 U/ml chondroitinase ABC and heparitinase III, respectively, had no effect on ureter shape and Wnt-11 expression, higher concentrations of heparitinase III (2.5 U/ml) and a combination of both enzymes (0.5 U/ml chondroitinase, 0.5-2.5 U/ml heparitinase III) resulted in branching inhibition and a partial loss of Wnt-11 expression (not shown).

Thus, sulphated proteoglycans are important for the maintenance of Wnt-11 expression and chondroitin sulphate, heparan sulphate and heparin might constitute a major fraction of the glycosaminoglycans required for Wnt-11 regulation.

The effect of chlorate on ureter growth and branching in kidney cultures might result from the loss of bound factors from the matrix. Indeed, Wnts, fibroblast growth factors and a variety of other secreted factors are thought to bind to matrix components, including proteoglycans. We investigated the effect of three sulphated glycosaminoglycans and of suramin, all of which release Wnt proteins into the medium in cell culture systems (Burrus and McMahon, 1995). 100 μg/ml chondroitin-6-sulphate had no effect on ureter morphology and Wnt-11 expression (Fig. 7H,I). However, kidney rudiments cultured for 3 days in 100 μg/ml heparin, exhibited very limited ureter growth and branching, although tubule induction was normal. Wnt-11 expression was found to be severely reduced after 24 hours of culture (Fig. 7J,K). Kidneys grown in medium containing 10-100 μg/ml pentosan polysulphate for 24 hours showed no Wnt-11 expression in the ureter tips, tubule formation was severely reduced, and growth and branching of the ureter were completely inhibited (Fig. 7L,M). Finally, we tested the effect of suramin on kidney cultures. In 50 μM suramin, the ureter epithelium survived and expressed the ureter marker Pax-2 even after 4 days of culture. However, ureter growth was completely arrested, tubule induction extremely rare and Wnt-11 expression lost after 24 hours (Fig. 7N,O).

To address whether the loss of Wnt-11 expression in these cultures was solely the result of the loss of the characteristic ampulla-like structure at the ureter tips, which is not evident in chlorate, pentosan polysulphate and suramin cultures after 24 hours, we compared a time course of Wnt-11 and c-ret and c-ros (not shown) expression in 30 mM chlorate (Fig. 8). Already after 5 hours, we detected a dramatic reduction of Wnt-11 expression (Fig. 8C) and after 8 hours expression was almost completely abolished (Fig. 8E). In contrast, expression of c-ret and c-ros, which exhibit a similar pattern, were only slightly reduced after 8 hours (Fig. 8D,F). After 24 hours, both markers were completely lost from the tips but a weak expression was detected throughout the ureter (Fig. 8J). Thus, the loss of Wnt-11 precedes gross morphological changes in the ureter (Fig. 8D,F).

Fig. 8.

Kinetics of Wnt-11 downregulation in cultures of metanephric kidneys treated with NaClO3 and pentosan polysulphate, respectively. Kidney rudiments were isolated at 11.5 days p.c. (bifurcated once) and cultured in standard medium (A,B) and in medium supplemented with 30 mM NaClO3 (C-J) and 100 μg/ml pentosan polysulphate (K-N), respectively. Cultures were fixed after 2 hours, in the case of the standard medium, after 5 hours (C,D), 8 hours (E,F), 16 hours (G,H), and 24 hours (I,J) when grown in medium supplemented with NaClO3 and after 3 hours (K,L) and 6 hours (M,N), respectively, in the pentosan polysulphate experiments. Cultures were subsequently subjected to whole-mount in situ hybridization to detect Wnt-11 expression (A,C,E,G,I,K,M) and c-ret expression (B,D,F,H,J,L,N), respectively. Loss of Wnt-11 expression is a rapid consequence of chlorate treatment and is even accelerated in the pentosanpolysulphate-treated cultures. Wnt-11 downregulation precedes loss of c-ret transcription and morphological changes at the ureter tip.

Fig. 8.

Kinetics of Wnt-11 downregulation in cultures of metanephric kidneys treated with NaClO3 and pentosan polysulphate, respectively. Kidney rudiments were isolated at 11.5 days p.c. (bifurcated once) and cultured in standard medium (A,B) and in medium supplemented with 30 mM NaClO3 (C-J) and 100 μg/ml pentosan polysulphate (K-N), respectively. Cultures were fixed after 2 hours, in the case of the standard medium, after 5 hours (C,D), 8 hours (E,F), 16 hours (G,H), and 24 hours (I,J) when grown in medium supplemented with NaClO3 and after 3 hours (K,L) and 6 hours (M,N), respectively, in the pentosan polysulphate experiments. Cultures were subsequently subjected to whole-mount in situ hybridization to detect Wnt-11 expression (A,C,E,G,I,K,M) and c-ret expression (B,D,F,H,J,L,N), respectively. Loss of Wnt-11 expression is a rapid consequence of chlorate treatment and is even accelerated in the pentosanpolysulphate-treated cultures. Wnt-11 downregulation precedes loss of c-ret transcription and morphological changes at the ureter tip.

Since chlorate exerts its effects intracellularly and acts only on newly synthesized proteoglycans, its effects should be retarded compared to those substances that directly affect the matrix. Not surprisingly, we found that the kinetics of Wnt-11 down-regulation accelerated in cultures that had been supplemented with 100 μg/ml pentosan polysulphate. Even after only 2-3 hours of incubation, Wnt-11 expression was almost completely abolished, well before c-ret transcription was lost in the ureter tips and preceding visible morphological changes (Fig. 8K-N).

The development of the mammalian kidney can schematically be divided into three processes. First, the metanephric blastema forms at the caudal end of the intermediate mesoderm. Second, the ureter bud emerges from the nephric duct, enters the metanephric blastema, grows and branches. Third, mesenchymal cells condense, epithelialize and differentiate to form the mature nephron.

We demonstrate here that Wnt-11 activation in the nephric duct is dependent on the metanephric blastema and its subsequent expression correlates with branching morphogenesis of the ureter. These results are consistent with the mesenchymal blastema providing a signal or signals that activate Wnt-11 and lead to ingrowth of the ureteric bud. Support for the existence of a mesenchymal signal promoting ureter outgrowth comes from genetic studies. For example, in the Wilms tumor-1 (WT-1) mutant, the metanephric blastema undergoes apoptosis and subsequently, in the absence of the mesenchyme, the nephric duct fails to bud (Kreidberg et al., 1993). Similar studies on Sd kidneys in vivo and in culture indicate that abnormalities in formation, growth and differentiation of the ureteric bud that result in the partial or complete failure of the metanephric mesenchyme differentiation have their origins in altering the interactions between the nephric duct and the blastema as intact kidney rudiments of Sd homozygous embryos are able to undergo differentiation in culture (Gluecksohn-Schoenheimer, 1943;, 1945; Gluecksohn-Waelsch and Rota, 1963).

Interestingly, we noticed a frequent unilateral loss and unequal bilateral expression of Wnt-11 in Sd heterozygotes, which most likely reflects variation in the proximity of the metanephric mesenchyme and the nephric duct tissues rather than direct genetic interactions. A second model, that positional cues intrinsic to the nephric duct autonomously regulate the activation of Wnt-11, is clearly not supported by our data.

Our experiments specifically addressed the regulation of Wnt-11 activation, which correlates with outgrowth of the ureteric bud. Additional experiments will be necessary to address whether Wnt-11 is required for bud outgrowth. The observation that Wnt-11 activation precedes budding supports such a model.

Recombination experiments in culture have suggested that following invasion of the kidney mesenchyme, survival, growth and branching of the ureter are also controlled by factors specific to the metanephric mesenchyme (Bishop-Calame, 1965a,b; Saxen, 1987). Our findings contradict this view. In our experiments, the ureter epithelium grows and branches in lung mesenchyme though the branching pattern resembled that of lung rather than ureteric bud epithelium. Further, Wnt-11 expression is retained and remains restricted to the tips of the ureter. Thus, it is unlikey that metanephric-specific factors are required for branching morphogenesis of the ureter. While we cannot exclude that our recombinants are contaminated with metanephric mesenchyme, it is unlikely because isolated ureters without lung mesenchyme degenerated and we never observed tubule induction in the recombinants. Unfortunately, there is no suitable marker to assay for small numbers of contaminating cells. Interestingly, it was thought for some time, that mouse submandibular epithelium critically depends on specific submandibular mesenchymal factors for branching. However, it now appears that lung mesenchyme is also able to support budding and cytodifferentiation of salivary epithelium (Lawson, 1972, 1974).

One interpretation of our results is that the ureteric epithelium contains an intrinsic morphogenetic program, and that the metanephric mesenchyme provides nonspecific survival/proliferation factors. Interestingly, Nogawa and Ito (1995) have reported that branching morphogenesis of mouse lung epithelium occurs in mesenchyme-free conditions. All that is required is a basement membrane matrix and the soluble growth factor aFGF. Further, neonatal mouse kidney epithelial cells (Traub et al., 1990) and Madin-Darby canine kidney epithelial cells can form tubules on manipulations of the matrix and addition of some soluble growth factors (Montesano et al., 1991). Our data suggest that the metanephric mesenchyme plays two roles in ureter branching. First, it provides general proliferative factors, a role that can be mimicked by heterologous mesenchymes. Second, it controls the ‘kidney’-specific rate and mode of ureter morphogenesis probably by the concerted interaction of more mature mesenchymal elements with the ureter. Our data also show that the localization of Wnt-11 expression to the tips of the ureter is independent of any specific signals provided by the kidney mesenchyme though the precise cellular dynamics of expression may be modulated specifically.

Numerous studies have pointed to the importance of the extracellular matrix (ECM) in epithelial-mesenchymal interactions in general and in kidney development in particular (Roskelly et al., 1995). The ureter epithelium is surrounded by a basal lamina consisting of type IV collagen, glycoproteins and proteoglycans (Ekblom, 1981). Induction of mesenchymal cells results in the remodeling of the interstitial type of ECM into an epithelial type (Aufderheide et al., 1987; Ekblom et al., 1980). Components known to interfere with proteoglycan metabolism and structure have profound effects on kidney development. In kidney cultures grown in medium supplemented with β-D-xyloside, a competitor of xylosylated core proteins at the level of galactosyltransferase, synthesis of macromolecular proteoglycans is decreased. Kidney development is perturbed and, in particular, branching morphogenesis is abnormal (Klein et al., 1989; Lelongt et al., 1988; Platt et al., 1987). Recently Davies et al. (1995) showed that chlorate, a competitive inhibitor of sulphate in the sulphation of proteoglycans, has an effect similar to that of β-D-xyloside in inhibiting branching and growth of ureters and tubule induction in cultures of kidney rudiments. In both cases, nephron maturation occurred normally suggesting that sulphated proteoglycans are not required for nephron development.

We find that Wnt-11 expression is rapidly lost on chlorate treatment. This is followed by a downregulation in expression of the receptor tyrosine kinases c-ret and c-ros and loss of the ampulla-like morphology at the termini of the ureteric buds. These morphological and molecular changes most likely explain the absence of tubule induction under these conditions. Our findings suggest that proteoglycans are an important component in a signaling process that maintains Wnt-11 expression and the integrity of the tips of the ureter bud. As our recombination experiments demonstrate that only the tips are necessary for branching morphogenesis, loss of tips most likely accounts for the cessation of ductal development. At present, it is not clear whether proteoglycans themselves and/or some associated factors modulate the development of the ureteric bud. For example, recent studies have pointed out that proteoglycans act as low affinity receptors for many growth factors including fibroblast growth factors and transforming growth factors (for review see Schlessinger et al., 1995; Rapraeger, 1995). Moreover, Wnts are thought to interact with ECM components (Bradley and Brown, 1990; Burrus and McMahon, 1995; Papkoff and Schryver, 1990). All of these factors may be displaced from matrix association by polysulphated compounds, which, in our experiments, block ureteric bud development and tubule induction. The effect of the polysulphated compounds on ureter branching correlates directly with their effect on Wnt-11 expression. Wnt-11 expression in the ureter tip was unaffected by chondroitin sulphate, reduced by heparin and completely abolished by pentosan polysulphate and suramin in the culture medium. Loss of Wnt-11 expression is a rapid response to addition of pentosan polysulphate arguing for a direct interference with a signaling pathway regulating Wnt-11 expression, possibly Wnt-11 itself.

Our data argue that Wnt-11 is likely to play an important role in regulating development of the metanephric kidney possibly in the branching morphogenesis of the ureter epithelium. In Drosophila melanogaster, wingless is expressed in the Malpighian tubules and is required for proliferation in the morphogenesis of this organ (Skaer and Martinez Arias, 1992) pointing to a possible evolutionary conservation of Wnt gene function in the ductal epithelium during development of excretory structures. Wnt-11 is coexpressed with the receptor tyrosine kinase c-ret in the nephric duct and in the ureter tips during normal kidney development, in Wnt-4 mutant kidneys and in heterologous recombination experiments. Moreover, both genes are downregulated under conditions that interfere with the structure and function of proteoglycans. Taking the requirement of c-ret for ureter growth and branching into account (Schuchardt et al., 1994), it is tempting to speculate that Wnt-11 and c-ret both act in the ureter tip as part of a cascade of genes controlling growth and branching.

Finally, we cannot exclude additional roles for Wnt-11 in kidney development. Intriguingly, Herzlinger et al. (1994) have shown that cells expressing Wnt-1 are able to induce tubu-logenesis in isolated metanephric mesenchyme. Wnt-1 is expressed in dorsal spinal cord (Wilkinson et al., 1987), the strongest heterologous inducer of tubulogenesis (Grobstein, 1956). However, Wnt-1 is not expressed in the kidney, suggesting that Wnt-1 mimics a real Wnt-like inducer. The expression pattern of Wnt-11 in the metanephric kidney is fully compatible with the notion that Wnt-11 is this endogenous inducer. We cannot rigorously exclude this possibility at present, but thus far we have been unable to induce tubule formation using conditions in which Wnt-1 acts as a strong inducer (A. K. and A. P. M., unpublished observation). Consequently, we favor a model in which Wnt-11 is required within the ureter as a survival/proliferation factor.

We would like to thank Scott Lee and Andy Dudley for critically reading the manuscript and helpful discussions. We are grateful to Frank Constanini for the c-ret probe, to Carmen Birchmaier for the c-ros probe and Yashpal Kanwar for the newborn kidney cDNA library. A.K. is a Postdoctoral Fellow of the Human Science Frontier Program Organization. S.V. was supported by fellowships from the Alfred Kordelin Foundation, the Academy of Finland, the Finnish Cultural Foundation and the European Molecular Biology Organization. D. H. R. is a Howard Hughes Medical Institute Physician Postdoctoral Fellow. Work in A. P. M.’s laboratory is supported by grants from the NIH.

Aufderheide
,
E.
,
Chiquet-Ehrismann
,
R.
and
Ekblom
,
P.
(
1987
).
Epithelial- mesenchymal interactions in the developing kidney lead to expression of tenascin in the mesenchyme
.
J. Cell Biol
.
105
,
599
608
.
Bard
,
J. B. L.
,
McConell
,
J. E.
and
Davies
,
J. A.
(
1994
).
Towards a genetic basis for kidney development
.
Mech. Dev
.
48
,
3
11
.
Bishop-Calame
,
S.
(
1965a
).
Sur le comportement en greffes chorio- allantoidiennes, de l’urétère de l’embryon de poulet associé á mesenchymes héterologues
.
C. R. Acad. Sci. Paris
260
,
281
283
.
Bishop-Calame
,
S.
(
1965b
).
Études d’associations héterologues de l’urétère ed de differents mesenchymes de l’embryon de poulet, par la technique de greffes chorio-allantoidiennes
.
J. Embryol. Exp. Morph
.
14
,
247
253
.
Bradley
,
R. S.
and
Brown
,
A. M. C.
(
1990
).
The proto-oncogene int-1 encodes a secreted protein associated with the extracellular matrix
.
EMBO J
.
9
,
1569
1575
.
Burrus
,
L. W.
and
A. P.
McMahon
(
1995
).
Biochemical analysis of murine Wnt proteins reveals both shared and distinct properties
.
Exp. Cell Res
.
220
,
363
373
.
Christiansen
,
J. H.
,
Dennis
,
C. L.
,
Wicking
,
C. A.
,
Monkley
,
S. J.
,
Wilkinson
,
D. G.
and
Wainwright
,
B. J.
(
1995
).
Murine Wnt-11 and Wnt- 11 have temporally and spatially restricted expression patterns during embryonic development
.
Mech. Dev
.
51
,
341
350
.
Church
,
G. M.
and
Gilbert
,
W.
(
1984
).
Genomic sequencing
.
Proc. Natl. Acad. Sci. USA
81
,
1991
1995
.
Davies
,
J.
,
Lyon
,
M.
,
Gallagher
,
J.
and
Garrod
,
D.
(
1995
).
Sulphated proteoglycan is required for collecting duct growth and branching but not nephron formation during kidney development
.
Development
121
,
1507
1517
.
Dressler
,
G. Y.
,
Deutsch
,
U.
,
Chowdhury
,
K.
,
Nornes
,
H. O.
and
Gruss
,
P.
(
1990
).
Pax2, a new murine paired-box-containing gene and its expression in the developing excretory system
.
Development
109
,
787
795
.
Ekblom
,
P.
(
1981
).
Formation of basement membranes in the embryonic kidney: an immunohistological study
.
J. Cell Biol
.
91
,
1
10
.
Ekblom
,
P.
,
Alitalo
,
K.
,
Vaheri
,
A.
,
Timpl
,
R.
and
Saxen
,
L.
(
1980
).
Induction of basement membrane glycoprotein in embryonic kidney: possible role of laminin in morphogenesis
.
Proc. Natl. Acad. Sci. USA
77
,
485
489
.
Ekblom
,
P.
,
Nordling
,
S.
and
Saxen
,
L.
(
1978
).
Inhibition of kidney tubule induction by charged polymers
.
Cell Diff
.
7
,
345
353
.
Farley
,
J. R.
,
Nakayama
,
G.
,
Cryns
,
D.
and
Sagel
,
I. H.
(
1978
).
Adenosine triphosphate sulphurylase from Penicillium chrysogencium: equilibrium binding, substrate hydrolysis and isotope exchange studies
.
Arch. Biochem. Biophys
.
185
,
376
390
.
Gavin
,
B. J.
,
McMahon
,
J. A.
and
McMahon
,
A. P.
(
1990
).
Expression of multiple novel Wnt-1/int-1-related genes during fetal and adult mouse development
.
Genes Dev
.
4
,
2319
2332
.
Gluecksohn-Schoenheimer
,
S.
(
1943
).
The morphological manifestation of a dominant mutation in mice affecting tail and urogenital system
.
Genetics
28
,
341
348
.
Gluecksohn-Schoenheimer
,
S.
(
1945
).
The embryonic development of mutants of the Sd-strain in mice
.
Genetics
30
,
29
38
.
Gluecksohn-Waelsch
,
S.
and
Rota
,
T. R.
(
1963
).
Development in organ tissue culture of kidney rudiments from mutant mouse embryos
.
Dev. Biol
.
7
,
432
444
.
Grobstein
,
C.
(
1953
).
Inductive epithelio-mesenchymal interactions in cultured organ rudiments of the mouse
.
Science
118
,
52
55
.
Grobstein
,
C.
(
1955
).
Inductive interactions in the development of the mouse metanephros
.
J. Exp. Zool
.
130
,
319
340
.
Grobstein
,
C.
(
1956
).
Transfilter induction of tubules in mouse metanephrogenic mesenchyme
.
Exp. Cell Res
.
10
,
424
440
.
Hammerman
,
M. R.
(
1995
).
Growth factors in renal development
.
Sem. Nephrol
.
15
,
291
299
.
Herzlinger
,
D.
,
Qiao
,
J.
,
Cohen
,
D.
,
Ramakrishna
,
N.
and
Brown
,
A. M. C.
(
1994
).
Induction of kidney epithelial morphogenesis by cells expressing Wnt-1
.
Dev. Biol
.
166
,
815
818
.
Kjellen
,
L.
and
Lindahl
,
U.
(
1991
).
Proteoglycans: structure and functions
.
Ann. Rev. Biochem
.
60
,
443
475
.
Klein
,
D. J.
,
Brown
,
D. M.
,
Moran
,
A.
,
Oegema
,
T. R. J.
and
Platt
,
J. L.
(
1989
).
Chondroitin sulfate proteoglycan synthesis and reutilization of β-D- xyloside-initiated chondroitin/dermatan sulfate glycosaminoglycans in fetal kidney branching morphogenesis
.
Dev. Biol
.
133
,
515
528
.
Knecht
,
A. K.
,
Good
,
P. J.
,
Dawid
,
I. B.
and
Harland
,
R. M.
(
1995
).
Dorsal- ventral patterning and differentiation of noggin-induced neural tissue in the absence of mesoderm
.
Development
121
,
1927
1936
.
Kreidberg
,
J. A.
,
Sariola
,
H.
,
lLoring
,
J. M.
,
Maeda
,
M.
,
Pelletier
,
J.
,
Housman
,
D.
and
Jaenisch
,
R.
(
1993
).
WT-1 is required for early kidney development
.
Cell
74
,
679
691
.
Lawson
,
K. A.
(
1972
).
The role of mesenchyme in the morphogenesis and functional differentiation of rat salivary epithelium
.
J. Embryol. Exp. Morph
.
27
,
497
513
.
Lawson
,
K. A.
(
1974
).
Mesenchyme specificity in rodent salivary gland development: the response of salivary epithelium to lung mesenchyme in vivo
.
J. Embryol. Exp. Morph
.
32
,
469
493
.
Lee
,
S. M. K.
,
Dickinson
,
M. E.
,
Parr
,
B. A.
,
Vainio
,
S.
and
McMahon
,
A. P.
(
1995
).
Molecular genetic analysis of Wnt signals in mouse development
.
Sem. Dev. Biol
.
6
,
267
274
.
Lelongt
,
B.
,
Makino
,
H.
,
Dalecki
,
T. M.
and
Kanwar
,
Y. S.
(
1988
).
Role of proteoglycans in renal development
.
Dev. Biol
.
128
,
256
176
.
Montesano
,
R.
,
Matsumoto
,
K.
,
Nakamura
,
T.
and
Orci
,
L.
(
1991
).
Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor
.
Cell
67
,
901
908
.
Nogawa
,
H.
and
Ito
,
T.
(
1995
).
Branching morphogenesis of embryonic mouse lung epithelium in mesenchyme-free culture
.
Development
121
,
1015
1022
.
Papkoff
,
J.
and
Schryver
,
B.
(
1990
).
Secreted int-1 protein is associated with the cell surface
.
Mol. Cell. Biol
.
10
,
2723
2730
.
Parr
,
B. A.
,
Shea
,
M. J.
,
Vassileva
,
G.
and
McMahon
,
A. P.
(
1993
).
Mouse Wnt genes exhibit discrete domains of expression in the early embryonic CNS and limb buds
.
Development
119
,
247
261
.
Patterson
,
L. T.
, and
Dressler
,
G. R.
(
1994
).
The regulation of kidney development: new insights from an old model
.
Curr. Opin. Genet. Dev
.
4
,
696
702
.
Perantoni
,
A. O.
(
1991
).
Induction of tubules in rat metanephrogenic mesenchyme in the absence of an inductive tissue
.
Differentiation
48
,
25
31
.
Perantoni
,
A. O.
,
Dove
,
L. F.
and
Karavanova
,
I.
(
1995
).
Basic fibroblast growth factor can mediate the early inductive events in renal development
.
Proc. Natl. Acad. Sci. USA
92
,
4696
4700
.
Platt
,
J. L.
,
Brown
,
D. M.
,
Granlund
,
K.
,
Oegema
,
T. R. j.
and
Klein
,
D. J.
(
1987
).
Proteoglycan metabolism associated with mouse metanephric development: morphologic and biochemical effects of β-D-xyloside
.
Dev. Biol
.
123
,
293
306
.
Rapraeger
,
A. C.
(
1995
).
In the clutches of proteoglycans: how does heparan sulfate regulate FGF binding?
Chem. & Biol
.
2
,
645
649
.
Roskelly
,
C. D.
,
Srebrow
,
A.
and
Bissell
,
M. J.
(
1995
).
A hierarchy of ECM- mediated signalling regulates tissue-specific gene expression
.
Curr. Op. Cell Biol
.
7
,
736
747
.
Sanger
,
F.
,
Nicklen
,
S.
and
Coulson
,
A. R.
(
1977
).
DNA sequencing with chain terminating inhibitors
.
Proc. Natl. Acad. Sci. USA
74
,
5463
.
Saxen
,
L.
(
1970
).
Failure to demonstrate tubule induction in a heterologous mesenchyme
.
Dev. Biol
.
23
,
511
523
.
Saxen
,
L.
(
1987
).
Organogenesis of the Kidney. Cambridge, UK: Cambridge University Press
.
Saxen
,
L.
and
Lehtonen
,
E.
(
1978
).
Transfilter induction of kidney tubules as a function of the extent and duration of intercellular contacts
.
J. Embryol. Exp. Morph
.
47
,
97
109
.
Saxen
,
L.
,
Lehtonen
,
E.
,
Karkinen-Jaaskelainen
,
M.
,
Nordling
,
S.
and
Wartiovara
,
J.
(
1976
).
Are morphogenetic tissue interactions mediated by transmissible signal substances or through cell contacts?
Nature
259
,
662
663
.
Schlessinger
,
J.
,
Lax
,
I.
and
Lemmon
,
M.
(
1995
).
Regulation of growth factor activation by proteoglycans: what is the role of the low affinity receptors?
Cell
83
,
357
360
.
Schuchardt
,
A.
,
D’Agati
,
V.
,
Larsson-Blomberg
,
L.
,
Constantini
,
F.
and
Pachnis
,
V.
(
1994
).
Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret
.
Nature
367
,
380
383
.
Skaer
,
H.
and
Martinez-Arias
,
A.
(
1992
).
The wingless product is required for cell proliferation in the Malpighian tubule anlage of Drosophila melanogaster
.
Development
116
,
745
754
.
Stark
,
K.
,
Vainio
,
S.
,
Vassileva
,
G.
and
McMahon
,
A. P.
(
1994
).
Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4
.
Nature
372
,
679
683
.
Torres
,
M.
,
Gomez-Pardo
,
E.
,
Dressler
,
G. R.
and
Gruss
,
P.
(
1995
).
Pax-2 controls multiple steps of urogenital development
.
Development
121
,
4057
4065
.
Traub
,
M.
,
Wang
,
Y.
,
Szczesny
,
T. M.
and
Kleinman
,
H. K.
(
1990
).
Epidermal growth factor or transforming growth factor a is required for kidney tubulogenesis in matrigel cultures in serum-free medium
.
Proc. Natl. Acad. Sci. USA
87
,
4002
4006
.
Unsworth
,
B.
and
Grobstein
,
C.
(
1970
).
Induction of kidney tubules in mouse metanephric mesenchyme by various embryonic mesenchymal tissues
.
Dev. Biol
.
21
,
547
556
.
Vainio
,
S.
,
Lehtonen
,
E.
,
Jalkanen
,
M.
,
Bernfield
,
M.
and
Saxen
,
L.
(
1989
).
Epithelial-mesenchymal interactions regulate the stage-specific expression of a cell-surface proteoglycan, syndecan, in the developing kidney
.
Dev. Biol
.
134
,
382
391
.
Wada
,
J.
,
Liu
,
Z. Z.
,
Alvares
,
K.
,
Kumar
,
A.
,
Wallner
,
E.
,
Makino
,
H.
and
Kanwar
,
Y. S.
(
1993
).
Cloning of a cDNA for the α subunit of mouse insulin-like growth factor I recptor and the role of the receptor in metanephric development
.
Proc. Natl. Acad. Sci. USA
90
,
10360
10364
.
Wartiovara
,
J.
,
Nordling
,
S.
,
Lehtonen
,
E.
and
Saxen
,
L.
(
1974
).
Transfilter induction of kidney tubules: correlation with cytoplasmic processes into Nucleopore filters
.
J. Embryol. Exp. Morph
.
31
,
667
682
.
Whiting
,
J.
,
Marshall
,
H.
,
Cook
,
M.
,
Krumlauf
,
R.
,
Rigby
,
P. W. J.
,
Stott
,
D.
and
Alleman
,
R. K.
(
1991
).
Multiple spatially specific enhancers are required to reconstruct the pattern of Hox-2.6 gene expression
.
Genes Dev
.
4
,
180
189
.
Wilkinson
,
D. G.
,
Bailes
,
J. A.
,
Champion
,
J. F.
and
McMahon
,
A. P.
(
1987
).
Expression of the proto-oncogene int-1 is restricted to specific neural cells in the developing mouse embryo
.
Cell
50
,
79
88
.