The bladder, the largest smooth-muscle organ in the human body, is responsible for urine storage and micturition. P63, a homolog of the p53 tumor-suppressor gene, is essential for the development of all stratified epithelia, including the bladder urothelium. The N-terminal truncated isoform of p63, ΔNp63, is known to have anti-apoptotic characteristics. We have established that ΔNp63is not only the predominant isoform expressed throughout the bladder, but is also preferentially expressed in the ventral bladder urothelium during early development. We observed a host of ventral defects in p63-/- embryos, including the absence of the abdominal and ventral bladder walls. This number of ventral defects is identical to bladder exstrophy, a congenital anomaly exhibited in human neonates. In the absence of p63, the ventral urothelium was neither committed nor differentiated,whereas the dorsal urothelium was both committed and differentiated. Furthermore, in p63-/- bladders, apoptosis in the ventral urothelium was significantly increased. This was accompanied by the upregulation of mitochondrial apoptotic mediators Bax and Apaf1, and concurrent upregulation of p53. Overexpression ofΔ Np63γ and ΔNp63β in p63-/- bladder primary cell cultures resulted in a rescue,evidenced by significantly reduced expressions of Bax and Apaf1. We conclude that ΔNp63 plays a crucial anti-apoptotic role in normal bladder development.

Bladder exstrophy (BE) is a serious congenital anomaly affecting one in 36,000 live births (Martinez-Frias et al.,2001). In cases of BE, the ventral abdominal and bladder walls are either absent, leaving the bladder cavity exposed, or covered only by an amniotic sac; the pubic bones, external genitalia and rectus abdominis muscles are separated along the midline, whereas the anus is displaced ventrally,often with an associated narrowing or atresia. Treatment of BE is complicated and involves major reconstructive surgical procedures. Although advanced parental age (Boyadjiev et al.,2004), familial links (Shapiro et al., 1984) and racial predilection(Roberts et al., 1995) imply a genetic cause, the molecular mechanisms underlying the formation of BE remain unknown. As such, our understanding of the pathogenesis of BE remains limited to that provided by a previous descriptive study(Muecke, 1964).

During embryogenesis, the cloacal cavity at the posterior end of the embryo is partitioned by the uro-rectal septum into the ventral urogenital sinus(UGS) and the dorsal hindgut. The UGS subsequently develops into the urethra,bladder and urachus. The UGS epithelium differentiates into a stratified transitional epithelium, known as the urothelium, whereas the mesenchyme of the UGS differentiates into the lamina propria and the smooth muscle of the bladder, known as the detrusor muscle. Interaction between the UGS epithelium and its mesenchyme is crucial for proper development of the detrusor muscle,as previous studies have shown that the UGS epithelium provides key signaling input that promotes differentiation of the UGS mesenchyme into smooth muscle(Baskin et al., 1996b).

Homologs p53, p63 and p73 comprise the p53 gene family(Levine, 1997; Murray-Zmijewski et al.,2006). p63 is highly expressed in all stratified epithelia and its expression can be detected in the urothelium as early as E11.5 and persisting thereafter (Kurita and Cunha, 2001; Kurita et al., 2004a; Kurita et al.,2004b). We therefore hypothesize that p63 plays a role in bladder urothelium development that, in turn, affects bladder development. p63-/- mice exhibit severe developmental anomalies,including failure of skin morphogenesis, truncation of limbs and craniofacial abnormalities (Mills et al.,1999; Yang et al.,1999). The specific mechanism underlying the regulation of epithelial stratification and development by p63 is not fully delineated and remains controversial. Some investigators suggest that failure of epithelial stratification in the absence of p63 is related to a lack of commitment (Koster et al.,2004; Mills et al.,1999), whereas others suggest that it results from a defect in epithelial cell proliferation (McKeon,2004; Yang et al.,1999).

p63 expresses multiple N-terminal isoforms, known as TAp63 and ΔNp63, because of the presence of an alternative promoter located in intron 3. The full-length isoform, TAp63, contains a transactivation (TA) domain similar to the TA domain of p53. TAp63 is capable of activating numerous p53target genes, promoting cell-cycle arrest(Yang et al., 1999) and inducing apoptosis (Jacobs et al.,2005). Conversely, the truncated isoform, ΔNp63,acts in a dominant-negative manner towards the TA isoforms of p63 and p53(Yang et al., 1998).Δ Np63 has been shown to inhibit apoptosis(Jacobs et al., 2005) and to promote stem-cell proliferation in vitro(Moll and Slade, 2004). In addition to these N-terminal isoforms, alternative splicing at the C-terminus of p63 generates three isoforms: α, β and γ. In combination with the N-terminal isoforms, six p63 isoforms can be generated (Yang et al.,1998).

In the current study, we find that the ΔNp63 isoform is the predominant isoform in the ventral bladder throughout development. In the absence of p63, the abdominal and ventral bladder walls are absent;these defects epitomize the BE complex in humans. In addition, the ventral epithelium of the p63-/- bladder is neither committed to stratification nor differentiated, and exhibits significantly increased apoptotic activity. Pro-apoptotic mediators Bax and Apaf1are upregulated in the p63-/- bladder. Restoration ofΔ Np63β or ΔNp63γ protein levels in p63-/- bladders partially rescues expression of Bax and Apaf1. Furthermore, absence of p63 in the bladder epithelium leads to failure of induction of the adjacent UGS mesenchyme, resulting in a significant reduction of mesenchymal proliferation. Taken together, these observations lead us to conclude thatΔ Np63 plays a crucial anti-apoptotic role in the development of the ventral bladder epithelium.

p63-/- mutant mice genotyping

p63-/- mutant mice were bred on a C57B16 background(Mills et al., 1999). Homozygous embryos were identified by phenotype. Heterozygous embryos were genotyped by PCR (primers 5′-GTGTTGGCAAGGATTCTGAGACC-3′ and 5′-GGAAGACAATAGCAGGCATGCTG-3′).

Histochemistry and immunohistochemistry

Specimen sections (7′m) were stained with 50% hematoxylin and 0.5%Eosin in 70% ethanol. Carbohydrates were stained with periodic acid and Schiff reaction (PAS, Surgipath). Alkaline phosphatase (AP) reaction was studied by treating slides with 0.1 M Tris-buffered solution (pH 9.5) followed by the addition of BM purple AP substrate (Roche).

Immunochemistry was performed as follows: after quenching the endogenous peroxidases with 3% H2O2 in 10% methanol, the antigens were retrieved by boiling the slides in an antigen-unmasking solution (H-3300,Vector Laboratories). The sections were blocked with blocking reagent (Roche). Primary antibodies at the following dilutions were applied: cytokeratin 18(CK18) (1:100, Santa Cruz Biotechnology), p63 (4A4, 1:100, Santa Cruz Biotechnology), TAp63 (1:20, Santa Cruz Biotechnology),Δ Np63 (1:100, gift from Dr K Nylander)(Nylander et al., 2002), p53 (1:250, Abcam, ab26), p73 (1:200, Abcam, ab17230),villin (1:100, Santa Cruz Biotechnology), uroplakin 3 (undiluted, Santa Cruz Biotechnology), smooth-muscle α-actin (undiluted, Sigma Chemicals),cleaved caspase-3 (1:100, Sigma Chemicals), Msx1 (1:500, Covance Research Products) and smooth-muscle heavy-chain myosin (1: 2000, Santa Cruz Biotechnology). Appropriate secondary antibodies were applied at 1:200 dilutions. Avidin-biotin-peroxidase complex (ABC)-buffer washing was followed by substrate diaminobenzidine (DAB) staining. Cell proliferation was assayed by 5-bromo-2′deoxyuridine (BrdU) incorporation (animals were injected with 100 μm BrdU per gram of bodyweight). Apoptosis was studied using the terminal deoxynucleotidyl transferase biotin-dUTP nick end-labeling (TUNEL)assay and FragEL DNA Fragmentation Detection Kit (Calbiochem).

RNA extraction, qPCR and RT-PCR

Total RNA was extracted using the RNeasy Mini Kit (Qiagen). cDNA was synthesized using the SuperScript II First-Strand Synthesis Kit (Invitrogen),then purified using a QIAquick PCR Purification Kit (Qiagen). Quantitative polymerase chain reactions (qPCRs) were performed using commercially available p53 primers (SuperArray Bioscience, PPM02931A-24) and selfdesigned primers: p53 (5′-CACCTCACTGCATGGACGATC-3′,5′-GTCTGCCTGTCTTCCAGATACTCG-3′, T: 59.1°C); p73(5′-CAAGAAGGCAGAGCATGTGA-3′,5′-TCATACGGCACAACCACACT-3′, T: 50.1°C);′-actin(internal control, 5′-CCTTTTCCAGCCTTCCTTC-3′,5′-TACTCCTGCTTGCTGATCC-3′, T: 55.0°C); Bax(5′-CGAGCTGATCAGAACCATCA-3′,5′-CTCAGCCCATCTTCTTCCAG-3′, T: 50.1°C); and Apaf1(5′-GAGAAAACCCTGAGGCACAA-3′,5′-TAATTAAAGCGGCTGCTCGT-3′, T: 50.4°C). The relative expressions were analyzed according to Pfaffl's methods(Pfaffl, 2001).

Reverse transcriptase-polymerase chain reactions (RT-PCRs) were performed using the following primers: ΔNp63(5′-CAATGCCCAGACTCAATTTAGTGA-3′,5′-GGCCCGGGTAATCTGTGTTGG-3′, 221bp, T: 51.4°C); TAp63(5′-AACCCCAGCTCATTTCTCTG-3′,5′-GGCCCGGGTAATCTGTGTTGG-3′, 449 bp, T: 57.0°C); p63′ (5′-ACGGGGTGGAAAAGAGATGGTC-3′,5′-AAGAGACCGGAAGGCAGATGAAG-3′, 919 bp, T: 59.5°C); p63α (5′-GACTTGCCAAATCCTGACA-3′,5′-AAGAGACCGGAAGGCAGATGAAG-3′, 619 bp, 55.1°C); and p63β (5′-CTCCCCGGGGCTCCACAAG-3′,5′-AAGAGACCGGAAGGCAGATGAAG-3′, 338 bp, T: 56.2°C).

Immunoblot

Immunoblot was performed as previously described(Qiu et al., 2004). Briefly,the cultured cells were washed twice with 1×PBS and lysed using a solubilizing buffer (1×PBS containing 1% Nonidet P-40, 1% deoxycholate,5 mM EDTA, 1 mM EGTA, 2 mM PMSF, 0.1 mM leupeptin, 100 KIU/mL Trasylol and 0.5μm ALLN), and an equal amount of cell lysates were resolved on 8% SDS-PAGE mini gels. Following SDS-PAGE, the protein was transferred electrophoretically for 18 hours at 4°C onto PVDF. The membranes were blocked with a 4%solution of fat-free dry milk powder, incubated with the primary antibodies(anti-Bax antibody, Upstate Cell Signaling, cat. no. 06-499, 1:500;anti-Apaf1, Chemicon, cat. no. MAB3504, 1:500; and anti-β-actin antibody, Sigma Chemicals, cat. no. A 5441, 1:1000), washed, and incubated with a secondary antibody conjugated to horseradish peroxidase. Membranes were then incubated in an enhanced chemiluminescence-detection reagent (Amersham Pharmacia Biotech) and exposed to Kodak Hyperfilm (Eastman Kodak). Films were developed and quantitative analysis was performed using an imaging densitometer.

Organ culture, primary cell culture and transfection

The dissected bladders were cultured in 50% BGJb medium (Invitrogen), plus 50% A10 (Wisent) culture medium with supplements of transferrin (20 ng/ml),insulin (10 ng/ml) and epithelial growth factor (10 ng/ml). Primary cells were cultured with Eagle's minimum essential medium (EMEM; Wisent) and 20% fresh bovine serum (FBS). Green fluorescent protein (GFP)-tagged recombinant bicistronic adenoviruses with ΔNp63γ,Δ Np63β, and TAp63γ constructs were generated, purified and titered as previously reported(Jacobs et al., 2005). The adenoviruses were added to the culture media at the ratio of 5-10 pfu per cell.

In situ hybridization

In situ hybridization of paraffin sections with a DIG-labeled RNA probe was performed as previously described (Hui and Joyner, 1993). Briefly, the dewaxed slides were pre-fixed with 4%paraformaldehyde, permeabilized with proteinase K (0.02 mg/ml), and treated with 0.2 M HCl solution and 0.1 M triethanoloamine solution (TEA), plus 0.025 ml acetic anhydride/liter of TEA. The slides were then hybridized with 4.0μg/ml of DIG-labeled RNA probes (DIG labeling mix; Roche) in a formamide/sodium citratesodium chloride (SSC) buffer in a 55°C oven overnight. The slides were then washed with a 5×SSC/formamide solution. and treated with RNAse-A, 2×SSC and 0.2×SSC before being blocked with blocking reagent (Roche). Anti-DIG alkaline phosphatase antibody (Roche)was then applied, followed by BM purple AP substrate (Roche).

p63 deficiency leads to bladder exstrophy

In humans, BE complex is evidenced by a cluster of ventral midline defects,including: (1) ventral abdominal- and ventral bladder-wall defects; (2) bifid external genitalia; and (3) separation of the pubic bones(Fig. 1A,B). All p63-/-embryos examined (n=12) developed bladder abnormalities. Four embryos developed BE with ventral bladder- and abdominal-wall defects (with and without membrane cover), bifid external genitalia (Fig. 1C,D) and umbilical hernia. The remaining eight embryos developed dilated bladders with both thin lamina propria and thin muscle layers. The sagittal sections of E18.5 p63-/--mutant embryos demonstrated the full complement of BE (Fig. 1E,F),as evidenced by: (1) ventral abdominal-wall defect; (2) ventral bladder-wall defect covered with a thin membrane; (3) absence of pubic symphysis at the midline (i.e. separation of the pubic bones); (4) absence of external genitalia at the midline (i.e. bifid genitalia); (5) umbilical hernia; and (6)ventral translocation of the anus. Sections of even younger embryos demonstrated that the separation of external genitalia was evident at E11.5(Fig. 5A,B). Transverse sections through the p63-/--embryo pelvis confirmed BE(ruptured membrane and separation of pubic bones; Fig. 1G,H). In summary, the p63-/- mutant mouse phenotype recapitulates the full spectrum of human BE complex.

p63 is expressed in bladder epithelium throughout its development and ΔNp63 is the predominant isoform

To define the role of p63 in bladder development, the ontogeny of p63 expression in the bladder was examined by immunohistochemistry from gestational days E11.5 to E17.5 using a 4A4 pan-p63 antibody(Fig. 2A-D). p63 was initially expressed in the distal-ventral UGS epithelium, hindgut and skin overlying the genital tubercle of E11.5-E12.5 embryos(Fig. 2A,B). p63expression then extended to the epithelium over the body and dome of the bladder at E15.5 and E17.5, respectively(Fig. 2C,D). Next, the expression pattern of the different N-terminal isoforms was studied using antibodies specific to either TAp63 or ΔNp63(Nylander et al., 2002). Expression of the ΔNp63 isoform (detected by anti-ΔNp63 antibody) was found to be similar to that of the pan-p63 antibody (detected by 4A4), suggesting thatΔ Np63 represents the predominant isoform during bladder development. ΔNp63 expression began in the ventral UGS epithelium at E11.5 and extended to the rest of the epithelium later in development (Fig. 2E-H).Δ Np63 was also expressed in the epithelium overlying the urogenital tubercle (Fig. 2E,F). By contrast, TAp63 (detected by anti-TAp63) was expressed only transiently from E11.5 to E12.5 in the epithelium of the distal hindgut and in its communication with the UGS(Fig. 2I,J). TAp63expression decreased markedly in the distal hindgut after E14.5(Fig. 2K,L and data not shown)and was not expressed in the skin overlying the urogenital tubercle.

To verify the dominant p63 N-terminal isoform expressed in wild-type bladders, RT-PCR was performed on RNA extracted from E15.5 wild-type bladders using primers specific to ΔNp63 and TAp63,respectively. This analysis confirmed that the predominant N-terminal isoform of p63 during early bladder development was ΔNp63(Fig. 2M). Unlike in the skin,where p63α is the predominant C-terminal isoform(Westfall et al., 2003; Yang et al., 1998), RT-PCR detected only p63γ and p63β in the bladder epithelium (Fig. 2M). Thus, we concluded that ΔNp63γ and ΔNp63β are the predominant isoforms of p63 expressed in bladder epithelium during development.

p63 expression is ventrally restricted during early bladder development

To understand better why p63-/- embryos develop ventral midline defects, the p63 expression pattern during early bladder development was studied using immunohistochemistry. Although there was widespread p63 expression throughout the skin and urothelium in E18.5 embryos (data not shown), sagittal sections of E11.5 embryos showed that p63 expression was restricted to the ventral epithelia of the urogenital tubercle and UGS, the tail bud, the oral epithelium(Fig. 3A, arrows), and the apical ectodermal ridge (data not shown). Horizontal pelvic sections of E11.5 embryos confirmed that p63 expression was present in the skin overlying the urogenital tubercle (Fig. 3B) and ventral UGS epithelium(Fig. 3C). In later gestational-stage embryos, p63 expression was stronger and epithelial stratification was more advanced in the ventral skin compared with that of the dorsal skin (Fig. 3D-F). In summary, p63 expression in early bladder- and skinepithelia development is ventrally restricted.

p63-deficient bladder epithelium is abnormal along the dorso-ventral axis

To determine whether the stratification of the endoderm-derived urothelium is affected similarly to that of the ectoderm-derived epithelium in the absence of p63, E18.5 p63-/- bladders were stained with hematoxylin and Eosin. This analysis revealed that, whereas wild-type bladder epithelium differentiates into stratified transitional urothelium (Fig. 4A,C), the bladder epithelium of p63-/- mutants fails to stratify and remains as a single layer (Fig. 4B,D). Differences in epithelial morphology were also noted along the dorso-ventral axis. The dorsal epithelium of p63-/-bladder consisted mainly of simple cuboidal cells(Fig. 4D), whereas the ventral epithelial cells were primarily simple squamous cells(Fig. 4B, arrow).

As p63 has previously been shown to be essential in ectodermal epithelial commitment and/or differentiation(Mills et al., 1999), the role of p63 in bladder development was examined using wellestablished markers for epithelial differentiation in E18.5 embryos. K18, a marker of the endoderm or uncommitted epithelium, is not expressed in stratified or differentiated epithelia (Koster et al.,2004). We noted that, in mature wild-type bladder urothelium, K18 expression was absent (Fig. 4E,G). In the p63-/- bladder, whereas the dorsal epithelium did not express K18 (Fig. 4H), the ventral epithelium retained K18 expression, indicating that it was uncommitted to stratification(Fig. 4F). To further determine the status of epithelial differentiation in bladder tissue, the expression of uroplakin 3, a marker for terminally differentiated urothelium, was studied(Wu et al., 1999). Uroplakin 3 was strongly expressed in mature wild-type bladder urothelium(Fig. 4I,K). In the p63-/- bladder, uroplakin expression was reduced in the dorsal epithelium (Fig. 4L) and undetectable in the ventral epithelium(Fig. 4J). This suggests that,whereas the p63-/- ventral bladder epithelium is undifferentiated, the dorsal epithelium is capable of differentiation, even in the absence of p63 (Fig. 4K,L). As null mutation of p63 has been reported to be associated with intestinal metaplasia(Signoretti et al., 2005; Yang et al., 1999), intestinal markers were also examined in the p63-/- bladder epithelium. This analysis revealed that intestinal transformation does not occur in the p63-/- bladder epithelium (Fig. S1 in the supplementary material). In summary, null mutation of p63 was noticed to affect the development of bladder epithelium differentially along the dorso-ventral axis, ultimately resulting in uncommitted and undifferentiated ventral bladder epithelium.

Apoptosis is increased in p63-deficient bladder epithelium

ΔNp63 is known to act as a naturally occurring dominant negative. It has been shown to counteract the pro-apoptotic actions of TAp63 and p53 in vitro(Yang et al., 1998). We have shown that ΔNp63 is the major isoform of p63 expressed in the developing ventral bladder, and that both the mesenchyme and epithelium of the ventral UGS develop abnormally in the absence of ΔNp63(Fig. 5A,B). As such, we hypothesized that p63 acts as a pro-survival protein in the developing bladder, thus preventing the apoptosis of ventral UGS epithelium during development. To directly test this hypothesis, we examined the amount of apoptosis in p63-/- bladders by TUNEL assay and cleaved caspase-3 expression. The number of TUNEL-positive cells in the ventral UGS epithelium of E11.5 p63-/- mutants was significantly higher than that of wild-type controls(Fig. 5C,D) (44% versus 9%,Student's t-test, P<0.05). This increase in apoptosis was further corroborated by an observed increase in cleaved caspase-3 expression in p63-/- ventral UGS epithelium (35% versus 5%, Student's t-test, P<0.05) (Fig. 5E,F). In comparison, we noted minimal apoptotic activity in the dorsal epithelia of both wild-type and p63-/- bladders, as determined by TUNEL assay and cleaved caspase-3 expression(Fig. 5C-F). We also compared the apoptotic activities (percentage of cleaved caspase-3-positive cells) of skin overlying the p63-/- and wild-type urogenital tubercles (12.8±2.7% and 2.7±1.7%) and found the difference between them was statistically significant (Student's t-test, P<0.01). This phenomenon of increased apoptosis in the absence of p63 does not appear to be restricted to the ventral UGS. Oral-cavity epithelium, which also expresses high levels of p63 during early development (Fig. 3A), was noted to have a statistically significant increase in apoptosis in the p63-/- embryo (Fig. 5G,H, 49.0±1.0% versus 9.0±1.0%, Student's t-test, P<0.05). In summary, our data show that apoptosis is increased in the epithelia of the ventral UGS, as well as in other epithelial structures where p63 expression is normally high during early development.

Developmental apoptosis, important in normal organogenesis, is understood to involve the mitochondrial apoptotic pathway(Vaux and Korsmeyer, 1999). As such, we analyzed the expression of the mitochondrial apoptotic mediators Bax and Apaf1 in E12.5 and E13.5 wild-type and p63-/- bladders by qPCR. In the p63-deficient bladders, the relative expressions of Bax and Apaf1 were increased at both E12.5 and E13.5 (Fig. 5I,J). Taken together, our data showed that, in the absence ofΔ Np63, there was increased mitochondrial apoptotic activity in the developing bladder.

βNp63 is anti-apoptotic during bladder development

To confirm the anti-apoptotic role of ΔNp63 during bladder development, E13.5 p63-/- bladder primary cell cultures were infected with bicistronic adenoviruses expressing TAp63γ,Δ Np63β, ΔNp63γ and green fluorescent protein (GFP), or GFP alone, for 24 hours. The cells were then harvested and the expressions of Bax and Apaf1 were quantified with immunoblots. Their gel tensiometry readings were compared. Compared with the GFP-transfected control, transfection with ΔNp63β orΔ Np63γ adenoviruses significantly reduced the expressions of both Bax (P<0.05) and Apaf1(P<0.01), whereas transfection with TAp63γadenovirus led to an increase in the expressions of both Bax and Apaf1 (P<0.05) (Fig. 6A,B). To confirm the anti-apoptotic role of ΔNp63,organ cultures of E13.5 p63-/- bladders were infected with bicistronic adenoviruses expressing ΔNp63γ and GFP or GFP alone for 24 hours. Bax expression was examined by qPCR. We observed a more than 50% reduction of Bax relative expression in theΔ Np63γ-infected p63-/- bladder. These data suggest that ΔNp63, the predominant isoform of p63 in the bladder, plays an anti-apoptotic role during bladder development.

Apoptosis of bladder cells of E12.5 p63-/-animals is associated with upregulation of p53 and p73 expression

We then examined whether the expressions of p53 and p73were affected in p63-/- bladders. Immunohistochemical staining showed an upregulation of p53 expression in the ventral aspect of E11.5 p63-/- UGS (Fig. 6C,D). Additionally, there also appeared to be an upregulation of p73 expression in p63-/- UGS(Fig. 6E,F). We proceeded to quantify the p53 and p73 mRNA expressions in E12.5 and E13.5 wild-type and p63-/- bladders using real-time PCR analysis. There was a compensatory upregulation of the relative expressions of both p53 and p73 in E12.5 p63-/-bladders, compared with those of the wild-type controls (Student's t-test, P<0.05). Interestingly, in E13.5 p63-/- bladders, the expressions of both p53 and p73 were downregulated (Fig. 6G,H, Student's t-tests, P<0.01). The transient upregulation of p53 and p73 expression in p63-/- bladders co-incides temporally with increased apoptosis (Fig. 5C-F).

Failure of ventral UGS mesenchymal induction and proliferation in the absence of epithelial ΔNp63

Appropriate epithelial-mesenchymal interaction is essential for normal bladder development; in the absence of bladder epithelium, bladder smooth muscle does not develop normally (Baskin et al., 1996a). To examine how the p63-null epithelium affects the adjacent mesenchyme, we studied the expression of Msx1, a homeobox gene that is induced in the mesenchyme by an epithelium-derived signal. In the wild-type control embryos, Msx1 was expressed in the ventral subepithelial mesenchyme of E13.5 bladders, suggesting expression of Msx1 in the mesenchyme by the ventral UGS epithelium(Fig. 7A, arrow). By contrast,the expression of Msx1 is greatly reduced or absent in the sub-epithelial mesenchyme of E13.5 p63-/- bladders(Fig. 7B). To assess whether the p63-positive epithelium release mediator(s) that are crucial for homeostasis of the mesenchyme, we studied, by mRNA in situ hybridization, the expression of Fgf8, another marker normally induced in the apical ectodermal ridge by p63 (Mills et al., 1999; Yang et al.,1999). Fgf8 is normally expressed in the mesenchyme of E11.5 wild-type UGS (Maruoka et al.,1998) (Fig. 7C). We found that the expression of Fgf8 was downregulated in the p63-/- ventral UGS(Fig. 7D). These results suggest that p63 deficiency in UGS epithelium is associated with a failure of induction in the adjacent UGS mesenchyme, especially ventrally. Epithelial-mesenchymal transition (EMT) is a cellular mechanism during which certain cells switch from an epithelial to a mesenchymal status. During development, EMT is involved in neural-crest migration, heart morphogenesis and formation of palate mesenchymal cells from the oral epithelium on E13.5 in mice (Larue and Bellacosa,2005). We questioned whether bladder epithelial cells undergo EMT,as this might explain why the failure of epithelial development is associated with failure of mesenchymal development. We studied the expression of Snai1(m-snail), which induces the epithelial-mesenchymal transition(Barrallo-Gimeno and Nieto,2005; Cano et al.,2000). We could not detect any Snai1 expression in either the wild-type or p63-/- E14.5 bladders by in situ hybridization (Fig. 7E,F). These results failed to demonstrate the role of EMT at this stage of bladder development.

To explain the paucity of smooth muscle in the ventral bladder wall,mesenchymal cell proliferation was studied with the incorporation of BrdU. Msx1 is commonly expressed in regions of rapid proliferation(Bendall and Abate-Shen, 2000)and Fgf8 regulates survival and proliferation in the anterior heart field(Park et al., 2006). In the absence of epithelial ΔNp63 expression, mesenchymal expression of both Msx1 and Fgf8 were decreased. This was accompanied by a reduction in cell proliferation in both the epithelium and mesenchyme,especially ventrally (Fig. 7G,H). The difference in cell proliferation between p63-/- and the wild-type control was statistically significant (ANOVA, dorsal epithelium and mesenchyme: P<0.01;ventral epithelium and ventral mesenchyme: P<0.0001)(Fig. 7I). Taken together, in the absence of ΔNp63, the ventral bladder epithelium fails to induce expression of Msx1 and Fgf8 in the adjacent mesenchyme. This is associated with a decreased mesenchymal cell proliferation.

Smooth-muscle differentiation is disturbed in p63-deficient bladders

Msx1 is known to repress terminal differentiation(Bendall and Abate-Shen, 2000). To determine the effect of Msx1 downregulation on smooth-muscle differentiation, the expression of smooth-muscle heavy-chain myosin, which is present only in mature smooth-muscle cells(Owens, 1995), was studied immunohistochemically. In E14.5 wild-type bladders, smooth-muscle heavy-chain myosin expression was absent or very weak, whereas, in the p63-/- bladder, its expression was strong in the thin ventral bladder wall (Fig. 8A,B). This suggests that the absence of Msx1 in the ventral mesenchyme allowed premature smooth-muscle differentiation in the adjacent mesenchyme. Despite premature smooth-muscle differentiation, the E18.5 p63-/- bladder contained little or no smooth muscle ventrally, but did retain a thin layer of smooth muscle dorsally(smooth-muscle α-actin). Moreover, the lamina propria was either greatly reduced or absent in the p63-/- bladder(Fig. 8C,D). In addition,unlike the wild-type bladder detrusor muscle, which displayed well-organized smooth-muscle stratification, the dorsal smooth muscle in the p63-/- bladder was disorganized and non-stratified(Fig. 8E,F).

Our experimental findings support a number of conclusions. First,Δ Np63 is preferentially expressed in the ventral UGS during early bladder development. Second, in the absence of ΔNp63,ventral bladder epithelial development is abnormal because of an increase in ventral bladder epithelial apoptosis. Third, ΔNp63 prevents this ventral bladder epithelial apoptosis by, at least partially,downregulating the mitochondrial apoptotic pathway. Finally, in the absence ofΔ Np63, there is decreased cell proliferation in the UGS mesenchyme. The increased epithelial apoptosis and decreased cell proliferation in both epithelium and mesenchyme ultimately results in bladder exstrophy in p63-/- embryos.

Early ventral p63 expression and ventral midline defects in p63-/- mutants

In the current study, we demonstrated that, during early development, p63 is preferentially expressed in the epithelia of ventral structures, including the genital tubercle(Fig. 3B), oral cavity(Yang et al., 1999) and ventral UGS (Fig. 3C)(Kurita et al., 2004a). In the absence of p63, development of these ventral structures is defective,and is manifested as a truncated maxilla, cleft palate, ventral pelvic-wall defect (Ince et al., 2002),bifid genitalia and bladder exstrophy. We have established thatΔ Np63 is the predominant p63 isoform throughout bladder development. In a zebrafish model, ΔNp63 has been noted to be required for ventral specification, with loss of ΔNp63resulting in a reduction of ventral (non-neural) ectoderm, whereas overexpression of ΔNp63 expands the ventral ectoderm(Bakkers et al., 2002).Δ Np63 is also a direct target of Bmp4(Bakkers et al., 2002), a morphogen vital for correct ventral patterning(Lemaire and Yasuo, 1998). In light of these findings, our results suggest that ΔNp63 may be a ventralizing protein in mammalian development and absence ofΔ Np63 may account for the ventral midline defects observed in p63-/- embryos.

Msx1 is a mesenchymal marker, the expression of which is induced by adjacent epithelial tissue (Jowett et al.,1993). In p63-/- embryos, in which limb buds are absent or vestigial, Msx1 expression in the progress zone beneath the apical ectodermal ridge of the limb bud is greatly reduced or absent(Mills et al., 1999; Yang et al., 1999). Our findings further suggest that epithelial-mesenchymal interaction also plays an important role in ventral bladder development. Msx1 expression in the ventral mesenchyme is deficient in the p63-/- bladder,suggesting that p63-deficient epithelium fails to induce appropriately the adjacent mesenchyme (Fig. 6A,B). This decreased Msx1 expression is associated with a reduction in UGS mesenchymal proliferation and premature terminal differentiation of the smooth muscle. Interestingly, Msx1 is also a ventralizing signal responsible for mesoderm patterning under the regulation of Bmp4 in Xenopus (Takeda et al., 2000). In summary, a failure of mesenchymal induction may be responsible for the changes observed in p63-/- ventral UGS mesenchyme.

Notably, the specific epithelial signal to the UGS mesenchyme remains undefined. A possible candidate protein for this role is the secreted diffusible morphogen sonic hedgehog (Shh). Shh is known to participate in numerous developmental processes involving the epithelial-mesenchymal interaction (Ingham and McMahon, 2001). It also promotes proliferation and inhibits differentiation in renal mesenchymal cell development(Yu et al., 2002). Furthermore, Shh is expressed in the UGS epithelium during early bladder organogenesis (Bitgood and McMahon,1995; Mo et al.,2001). We found that the expression of Shh was reduced in the ventral p63-/- UGS, where the epithelium is squamous and uncommitted (Fig. S2 in the supplementary material). The reduction of Shh signaling may contribute to the reduction in cell proliferation and premature terminal differentiation in ventral bladder mesenchymal development. This remains to be determined.

Temporospatial restriction of p63 expression determines epithelial commitment to stratification and differentiation

The exact role of p63 during epithelial development is controversial (McKeon, 2004). p63 either commits the epithelium to stratification(Mills et al., 1999) or maintains epithelial proliferation (Yang et al., 1999). Koster et al. suggested that TAp63initiates epithelial commitment and that ΔNp63 is responsible for epithelial differentiation (Koster et al., 2004). Our study shows that, during the developmental period examined (E11.5-E17.5), ΔNp63 expression began in the ventral UGS and progressively extended to the remaining bladder urothelium. In the p63-/- bladder, a clear phenotypic difference is noted in the epithelium along the dorso-ventral axis of the bladder. The ventral epithelium is squamous, with almost no adjacent smooth muscle, whereas the dorsal epithelium is cuboidal with a thin layer of disorganized muscle. The distal ventral UGS epithelium, which is destined to become the urethra,transforms into an intestine-like epithelium(Signoretti et al., 2005). The ventral epithelium remains both uncommitted (positive for K18) and undifferentiated (negative for uroplakin 3). The dorsal epithelium, however,is both committed (negative for K18) and differentiated (positive for uroplakin 3). These results suggest that the timing of p63 expression in normal bladder development determines the extent of developmental delay. Thus, the absence of p63 during early ventral bladder development affects both epithelial commitment and differentiation, whereas, in dorsal epithelium, where p63 is normally expressed later, p63deletion does not appear to affect either commitment or differentiation.

ΔNp63 is prosurvival in ventral bladder development

This study provides in vivo evidence that ΔNp63 is anti-apoptotic during bladder development. In vitro study showed thatΔ Np63 can compete for the apoptotic target-gene site or form a transactivationincompetent heterocomplex with p53 or TAp73,thus inhibiting apoptosis (Yang et al.,2002). In our study, apoptotic activity was increased in the ventral UGS epithelium in the absence of ΔNp63(Fig. 5D,F). Expression of the mitochondrial apoptotic mediators Bax and Apaf1 was also elevated in p63-/- bladders; elevated Bax and Apaf1 expressions in p63-/- bladders were rescued by the overexpression of either ΔNp63β orΔ Np63γ. This rescue is corroborated by a previous study,in which ectopic ΔNp63α expression in the epidermis reduces epidermal susceptibility to ultraviolet light-induced apoptosis(Liefer et al., 2000). In developing sympathetic neurons, where ΔNp73 is the predominant isoform, a p73 knockout leads to increased apoptosis in a fashion similar to the p63 knockout in the ventral bladder. Overexpression ofΔ Np73 rescues the sympathetic neurons from apoptosis induced by withdrawal of the nerve growth factor(Pozniak et al., 2000). Our results not only confirm the functional consistency of ΔN isoform proteins of the p53 family, but also demonstrate the anti-apoptotic role of the ΔNp63 isoform during normal mammalian development.

ΔNp63 is also detected in oral carcinoma and the intensity of its expression increases with the severity of dysplasia(Nylander et al., 2002),suggesting an oncogenic role or stem-cell pluripotency factor for theΔ Np63 isoform. The possible mechanism may involve a p53 target gene, p21,as ΔNp63α binds the p21 promoter, represses its transcription and permits cell-cycle progression (Westfall et al.,2003). The scenario is similar to ΔNp53, which is tumorigenic (Mowat et al.,1985). Overexpression of a C-terminal dominant-negative fragment of p53Np53)(Shaulian et al., 1992) in human urothelial cells has been reported to increase the cell-proliferation rate (Shaw et al., 2005). In this study, p63-/- mutation has been shown to be associated with a significant reduction of cell proliferation in the ventral bladder epithelium. This suggests that ΔNp63 also promotes epithelial proliferation in mammalian bladder development. Autocrine regulation of urothelial cell proliferation via the EGFR signaling loop observed in urothelial regenerative response could also play a role in urothelial development (Varley et al.,2005).

Our data showed that deletion of p63 is associated with compensatory upregulation of p53 expression in the bladders of younger embryos (E11.5-E12.5). This p53 upregulation may further contribute to the apoptosis observed in the ventral p63-/-bladder, in addition to the protein-protein and protein-target gene interactions. Although the expression of p73 in p63-/- bladders is also upregulated, its role in inducing apoptosis is uncertain, as the predominant isoform of p73 expressed in bladder has not been studied.

In conclusion, we have established a p63-/- murine model for BE. We have found that ΔNp63 is expressed initially in the ventral bladder urothelium and possesses a ventralizing property. Although the complete bladder urothelium fails to stratify in the absence of p63, ventral urothelial development is more delayed than that in the dorsal epithelium, being both uncommitted and undifferentiated. We found thatΔ Np63 is the predominant isoform in the bladder. WithoutΔ Np63, urothelial apoptosis is increased and cell proliferation is reduced. We also noted a concurrent upregulation of p53expression. Overexpression of ΔNp63γ andΔ Np63β rescue the expression levels of the mitochondrial apoptotic mediators Bax and Apaf1 in p63-/- bladders. We conclude that ΔNp63plays a crucial anti-apoptotic role during ventral bladder development.

This work was supported by grants from the Canadian Institute of Health Research (#57889) and March of Dimes Birth Defects Foundation, USA(#FY02-154).

Bakkers, J., Hild, M., Kramer, C., Furutani-Seiki, M. and Hammerschmidt, M. (
2002
). Zebrafish DeltaNp63 is a direct target of Bmp signaling and encodes a transcriptional repressor blocking neural specification in the ventral ectoderm.
Dev. Cell
2
,
617
-627.
Barrallo-Gimeno, A. and Nieto, M. A. (
2005
). The Snail genes as inducers of cell movement and survival: implications in development and cancer.
Development
132
,
3151
-3161.
Baskin, L. S., Hayward, S. W., Sutherland, R. A., DiSandro, M. J., Thomson, A. A., Goodman, J. and Cunha, G. R. (
1996a
). Mesenchymal-epithelial interactions in the bladder.
World J. Urol.
14
,
301
-309.
Baskin, L. S., Hayward, S. W., Young, P. and Cunha, G. R.(
1996b
). Role of mesenchymal-epithelial interactions in normal bladder development.
J. Urol.
156
,
1820
-1827.
Bendall, A. J. and Abate-Shen, C. (
2000
). Roles for Msx and Dlx homeoproteins in vertebrate development.
Gene
247
,
17
-31.
Bitgood, M. J. and McMahon, A. P. (
1995
). Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo.
Dev. Biol.
172
,
126
-138.
Boyadjiev, S. A., Dodson, J. L., Radford, C. L., Ashrafi, G. H.,Beaty, T. H., Mathews, R. I., Broman, K. W. and Gearhart, J. P.(
2004
). Clinical and molecular characterization of the bladder exstrophy-epispadias complex: analysis of 232 families.
BJU Int.
94
,
1337
-1343.
Cano, A., Perez-Moreno, M., Rodrigo, I., Locascio, A., Blanco,M., de Barrio, M., Portillo, F. and Nieto, A. (
2000
). The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression.
Nat. Cell Biol.
2
,
76
-83.
Hui, C. C. and Joyner, A. L. (
1993
). A mouse model of greig cephalopolysyndactyly syndrome: the extra-toesJ mutation contains an intragenic deletion of the Gli3 gene.
Nat. Genet.
3
,
241
-246.
Ince, T. A., Cviko, A. P., Quade, B. J., Yang, A., McKeon, F. D., Mutter, G. L. and Crum, C. P. (
2002
). p63 Coordinates anogenital modeling and epithelial cell differentiation in the developing female urogenital tract.
Am. J. Pathol.
161
,
1111
-1117.
Ingham, P. W. and McMahon, A. P. (
2001
). Hedgehog signaling in animal development: paradigms and principles.
Genes Dev.
15
,
3059
-3087.
Jacobs, W. B., Govoni, G., Ho, D., Atwal, J. K., Barnabe-Heider,F., Keyes, W. M., Mills, A. A., Miller, F. D. and Kaplan, D. R.(
2005
). P63 is an essential proapoptotic protein during neural development.
Neuron
48
,
743
-756.
Jowett, A. K., Vainio, S., Ferguson, M. W., Sharpe, P. T. and Thesleff, I. (
1993
). Epithelial-mesenchymal interactions are required for msx 1 and msx 2 gene expression in the developing murine molar tooth.
Development
117
,
461
-470.
Koster, M. I., Kim, S., Mills, A. A., DeMayo, F. J. and Roop, D. R. (
2004
). p63 is the molecular switch for initiation of an epithelial stratification program.
Genes Dev.
18
,
126
-131.
Kurita, T. and Cunha, G. R. (
2001
). Roles of p63 in differentiation of Mullerian duct epithelial cells.
Ann. N. Y. Acad. Sci.
948
,
9
-12.
Kurita, T., Medina, R. T., Mills, A. A. and Cunha, G. R.(
2004a
). Role of p63 and basal cells in the prostate.
Development
131
,
4955
-4964.
Kurita, T., Mills, A. A. and Cunha, G. R.(
2004b
). Roles of p63 in the diethylstilbestrol-induced cervicovaginal adenosis.
Development
131
,
1639
-1649.
Larue, L. and Bellacosa, A. (
2005
). Epithelial-mesenchymal transition in development and cancer: role of phosphatidylinositol 3′ kinase/AKT pathways.
Oncogene
24
,
7443
-7454.
Lemaire, P. and Yasuo, H. (
1998
). Developmental signalling: a careful balancing act.
Curr. Biol.
8
,
R228
-R231.
Levine, A. J. (
1997
). p53, the cellular gatekeeper for growth and division.
Cell
88
,
323
-331.
Liefer, K. M., Koster, M. I., Wang, X. J., Yang, A., McKeon, F. and Roop, D. R. (
2000
). Down-regulation of p63 is required for epidermal UV-B-induced apoptosis.
Cancer Res.
60
,
4016
-4020.
Martinez-Frias, M. L., Bermejo, E., Rodriguez-Pinilla, E. and Frias, J. L. (
2001
). Exstrophy of the cloaca and exstrophy of the bladder: two different expressions of a primary developmental field defect.
Am. J. Med. Genet.
99
,
261
-269.
Maruoka, Y., Ohbayashi, N., Hoshikawa, M., Itoh, N., Hogan, B. L. and Furuta, Y. (
1998
). Comparison of the expression of three highly related genes, Fgf8, Fgf17 and Fgf18, in the mouse embryo.
Mech. Dev.
74
,
175
-177.
McKeon, F. (
2004
). p63 and the epithelial stem cell: more than status quo?
Genes Dev.
18
,
465
-469.
Mills, A. A., Zheng, B., Wang, X. J., Vogel, H., Roop, D. R. and Bradley, A. (
1999
). p63 is a p53 homologue required for limb and epidermal morphogenesis.
Nature
398
,
708
-713.
Mo, R., Kim, J. H., Zhang, J., Chiang, C., Hui, C. C. and Kim,P. C. (
2001
). Anorectal malformations caused by defects in sonic hedgehog signaling.
Am. J. Pathol.
159
,
765
-774.
Moll, U. M. and Slade, N. (
2004
). p63 and p73:roles in development and tumor formation.
Mol. Cancer Res.
2
,
371
-386.
Mowat, M., Cheng, A., Kimura, N., Bernstein, A. and Benchimol,S. (
1985
). Rearrangements of the cellular p53 gene in erythroleukaemic cells transformed by Friend virus.
Nature
314
,
633
-636.
Muecke, E. C. (
1964
). The role of the cloacal membrane in exstrophy: the first successful experimental study.
J. Urol.
92
,
659
-667.
Murray-Zmijewski, F., Lane, D. P. and Bourdon, J. C.(
2006
). p53/p63/p73 isoforms: an orchestra of isoforms to harmonise cell differentiation and response to stress.
Cell Death Differ
.
13
,
962
-972.
Nylander, K., Vojtesek, B., Nenutil, R., Lindgren, B., Roos, G.,Zhanxiang, W., Sjostrom, B., Dahlqvist, A. and Coates, P. J.(
2002
). Differential expression of p63 isoforms in normal tissues and neoplastic cells.
J. Pathol.
198
,
417
-427.
Owens, G. K. (
1995
). Regulation of differentiation of vascular smooth muscle cells.
Physiol. Rev.
75
,
487
-517.
Park, E. J., Ogden, L. A., Talbot, A., Evans, S., Cai, C. L.,Black, B. L., Frank, D. U. and Moon, A. M. (
2006
). Required,tissue-specific roles for Fgf8 in outflow tract formation and remodeling.
Development
133
,
2419
-2433.
Pfaffl, M. W. (
2001
). A new mathematical model for relative quantification in real-time RT-PCR.
Nucleic Acids Res.
29
,
e45
.
Pozniak, C. D., Radinovic, S., Yang, A., McKeon, F., Kaplan, D. R. and Miller, F. D. (
2000
). An anti-apoptotic role for the p53 family member, p73, during developmental neuron death.
Science
289
,
304
-306.
Qiu, W., Kohen-Avramoglu, R., Rashid-Kolvear, F., Au, C. S.,Chong, T. M., Lewis, G. F., Trinh, D. K., Austin, R. C., Urade, R. and Adeli,K. (
2004
). Overexpression of the endoplasmic reticulum 60 protein ER-60 downregulates apoB100 secretion by inducing its intracellular degradation via a nonproteasomal pathway: evidence for an ER-60-mediated and pCMB-sensitive intracellular degradative pathway.
Biochemistry
43
,
4819
-4831.
Roberts, D. J., Johnson, R. L., Burke, A. C., Nelson, C. E.,Morgan, B. A. and Tabin, C. (
1995
). Sonic hedgehog is an endodermal signal inducing Bmp-4 and Hox genes during induction and regionalization of the chick hindgut.
Development
121
,
3163
-3174.
Shapiro, E., Lepor, H. and Jeffs, R. D. (
1984
). The inheritance of the exstrophy-epispadias complex.
J. Urol.
132
,
308
-310.
Shaulian, E., Zauberman, A., Ginsberg, D. and Oren, M.(
1992
). Identification of a minimal transforming domain of p53:negative dominance through abrogation of sequence-specific DNA binding.
Mol. Cell. Biol.
12
,
5581
-5592.
Shaw, N. J., Georgopoulos, N. T., Southgate, J. and Trejdosiewicz, L. K. (
2005
). Effects of loss of p53 and p16 function on life span and survival of human urothelial cells.
Int. J. Cancer
116
,
634
-639.
Signoretti, S., Pires, M. M., Lindauer, M., Horner, J. W.,Grisanzio, C., Dhar, S., Majumder, P., McKeon, F., Kantoff, P. W., Sellers, W. R. et al. (
2005
). p63 regulates commitment to the prostate cell lineage.
Proc. Natl. Acad. Sci. USA
102
,
11355
-11360.
Takeda, M., Saito, Y., Sekine, R., Onitsuka, I., Maeda, R. and Maeno, M. (
2000
). Xenopus msx-1 regulates dorso-ventral axis formation by suppressing the expression of organizer genes.
Comp. Biochem. Physiol.
126B
,
157
-168.
Varley, C., Hill, G., Pellegrin, S., Shaw, N. J., Selby, P. J.,Trejdosiewicz, L. K. and Southgate, J. (
2005
). Autocrine regulation of human urothelial cell proliferation and migration during regenerative responses in vitro.
Exp. Cell Res.
306
,
216
-229.
Vaux, D. L. and Korsmeyer, S. J. (
1999
). Cell death in development.
Cell
96
,
245
-254.
Westfall, M. D., Mays, D. J., Sniezek, J. C. and Pietenpol, J. A. (
2003
). The Delta Np63 alpha phosphoprotein binds the p21 and 14-3-3 sigma promoters in vivo and has transcriptional repressor activity that is reduced by Hay-Wells syndrome-derived mutations.
Mol. Cell. Biol.
23
,
2264
-2276.
Wu, H. Y., Baskin, L. S., Liu, W., Li, Y. W., Hayward, S. and Cunha, G. R. (
1999
). Understanding bladder regeneration:smooth muscle ontogeny.
J. Urol.
162
,
1101
-1105.
Yang, A., Kaghad, M., Wang, Y., Gillett, E., Fleming, M. D.,Dotsch, V., Andrews, N. C., Caput, D. and McKeon, F. (
1998
). p63, a p53 homolog at 3q27-29, encodes multiple products with transactivating,death-inducing, and dominant-negative activities.
Mol. Cell
2
,
305
-316.
Yang, A., Schweitzer, R., Sun, D., Kaghad, M., Walker, N.,Bronson, R. T., Tabin, C., Sharpe, A., Caput, D., Crum, C. et al.(
1999
). p63 is essential for regenerative proliferation in limb,craniofacial and epithelial development.
Nature
398
,
714
-718.
Yang, A., Kaghad, M., Caput, D. and McKeon, F.(
2002
). On the shoulders of giants: p63, p73 and the rise of p53.
Trends Genet.
18
,
90
-95.
Yu, J., Carroll, T. J. and McMahon, A. P.(
2002
). Sonic hedgehog regulates proliferation and differentiation of mesenchymal cells in the mouse metanephric kidney.
Development
129
,
5301
-5312.