In mice, metanephric kidney development initiates between embryonic day (E) 10.5 and 11.0 when the caudal-most portion of the Wolffian duct (WD), known as the ureteric bud (UB), branches and invades the dorsally located metanephric mesenchyme (MM). The UB undergoes repeated rounds of branching morphogenesis throughout the embryonic period, halting shortly after birth (Yu et al., 2004; Schedl, 2007; Dressler, 2009)
Several genes have been identified as critical to UB development in embryonic kidney (Schedl, 2007; Affolter et al., 2009; Michos, 2009). Gdnf is a growth factor that promotes UB outgrowth (Moore et al., 1996; Pichel et al., 1996; Sánchez et al., 1996) and is expressed in the MM and signals to the WD expressing the Ret receptor tyrosine kinase (Schuchardt et al., 1994, 1996; Durbec et al., 1996; Vega et al., 1996) and coreceptor Gfrα1 (Cacalano et al., 1998; Enomoto et al., 1998). In turn, Ret activation results in the initiation of intracellular signaling cascades that mediate UB outgrowth. In addition to outgrowth, Gdnf-Ret signaling also plays a central role in the subsequent UB branching (Shakya et al., 2005; Costantini and Shakya, 2006). Following UB invasion of the MM, Gdnf becomes expressed in the cap mesenchyme surrounding each UB tip, and the expression of Ret becomes localized to the UB tips. Ret is a downstream target of Gdnf signaling and forms part of a positive feedback loop with another component, Wnt11 (Majumdar et al., 2003). Wnt11 is dependent upon Gdnf signaling for its UB tip-specific expression pattern and subsequently contributes to the maintenance of Gdnf expression in the adjacent cap mesenchyme. In mice lacking Ret, Gdnf, Gfrα1, or Wnt11, the UB fails to form, leading to renal agenesis or poor branching, leading to severe renal hypodysplasia.
Additional genetic studies in mice have further revealed a critical role for the β-catenin/Gata3 signaling pathway, which maintains the epithelial cells in an undifferentiated state in UB and prevents inappropriate responses to signals promoting UB differentiation (Grote et al., 2008; Marose et al., 2008; Schmidt-Ott and Barasch, 2008). Hoxb7-Cre mediated removal of β-catenin from the mouse UB/Wolffian duct epithelium leads to the premature expression of gene products normally associated with the differentiated kidney collecting duct system including the water channel protein, Aquaporin-3 (Aqp3), and the tight junction protein isoform, ZO-1α+. Reciprocally, expression of a stabilized form of β-catenin appears to block differentiation of the ducts (Marose et al., 2008; Schmidt-Ott and Barasch, 2008). Grote et al. (2008) have reported that Gata3 inactivation with Hoxb7-Cre causes premature nephric duct cell differentiation and loss of Ret expression, and they have further identified Gata3 as a mediator of β-catenin function and demonstrated that the β-catenin/Gata3 pathway prevents premature cell differentiation. Interestingly, no difference in DBA and Zo1+ expression levels can be observed between wild-type and Ret-deficient embryos leading Grote et al. to conclude that Gata3 acts independently of Ret to maintain an undifferentiated state in the duct epithelium (Grote et al., 2008). Although some genetic studies on the function for maintaining the UB/Wolffian duct in an undifferentiated state have been reported as described above, the mechanisms that prevent the inappropriate differentiation of the duct remains largely unknown.
Lgr4 (Leucine-rich repeat containing G-protein-coupled receptor 4) is a gene that was identified from an EST database as a novel GPCR (G-protein-coupled receptor) that, along with Lgr5–Lgr8, has high homology to glycoprotein hormone receptors such as FSHR, LH/CGR, and TSHR (Hsu et al., 1998; Loh et al., 2000). At present, the ligands for Lgr4, Lgr5, and Lgr6 have not been identified. Our Lgr4 null mice, completely lacking exon18, which encodes the whole transmembrane domain of Lgr4, showed embryonic/neonatal lethality with the additional phenotype of hypoplastic kidneys and an increased concentration of plasma creatinine (Kato et al., 2006). More recently, Lgr4 insufficiency has been shown to cause impaired morphogenesis of other organs such as epididymis and efferent ducts (Mendive et al., 2006; Hoshii et al., 2007; Li et al., 2010), eyelid (Kato et al., 2007), gall bladder (Yamashita et al., 2009), cystic duct (Yamashita et al., 2009), erythroid precursor cells (Song et al., 2008), iris (Weng et al., 2008), bone (Luo et al., 2009), and hair follicle (Mohri et al., 2008). We have also shown that Lgr4 has a critical function in the reproductive organs of female mice (Mohri et al., 2010). As yet, however, the ligand(s) for Lgr4 remains unidentified, as do the mechanisms by which Lgr4 regulates kidney morphogenesis.
In our previous report, the 129Ola × C57BL/6J hybrid background occasionally gave us neonatal null mutants (Kato et al., 2006), but here we demonstrate that no neonatal Lgr4−/− mice could be obtained on a strict C57BL/6J background (n=54) (the homozygous Lgr4 null mice with the C57BL/6J background are referred to as Lgr4−/−). In the Lgr4−/− embryos, the sizes of the kidneys were significantly smaller than those of wild-type littermates, and all of the embryos displayed dilated tubules at the late stage of kidney development (E16.5). Interestingly, we observed reduced expression of the genes related to branching morphogenesis such as Ret and Wnt11 in the kidney of Lgr4−/− embryos. We discuss the essential function for Lgr4 in maintaining cells of the UB in an undifferentiated state. These data suggest an essential role of Lgr4 in the normal development of the mammalian kidney.
Increased Incidence of Embryonic Lethality in Lgr4−/− Mice in the C57BL/6J Background
In our previous study, we reported that Lgr4 null mice generated on a 129Ola × C57BL/6J hybrid background were born at a significantly lower rate than expected 25% after crossing Lgr4 heterozygous mice (Kato et al., 2006). In almost all cases, the 129Ola × C57BL/6J Lgr4 null mice that survived in utero died shortly after birth, and we observed striking renal hypoplasia accompanied by elevated concentrations of plasma creatinine. Histological analysis of the P0 null mouse kidney showed a notable decrease in the total number and density of the glomeruli (Kato et al., 2006). We then backcrossed our 129Ola × C57BL/6J hybrid Lgr4 mice to C57BL/6J mice for at least 10 generations in order to obtain Lgr4 knockout mice with the C57BL/6J background (the homozygous Lgr4 null mice with the C57BL/6J background are referred to as Lgr4−/−). We observed no neonatal Lgr4−/− mice (Table 1). Furthermore, we sacrificed pregnant females at E18.5 and checked the genotype of the embryos (n=43). Only four embryos were Lgr4−/− and two of them were already dead (Table 1). These data suggest that most Lgr4−/− embryos die before birth.
Table 1. Number of Mice of Each Genotype at Indicated Days
E15.5 (n = 63)
E18.5 (n = 43)
E19.5 (n = 54)
Lgr4 Is Expressed During Kidney Development
RT-PCR analysis with previously established primer sets (Kato et al., 2007) showed that Lgr4 was normally expressed in the kidney of Lgr4+/+ embryos and that Lgr4−/− embryos lacked expression (Fig. 1A). We also examined the expression profile of Lgr4 during kidney development in wild-type embryos by RT-PCR analysis and found that Lgr4 was expressed during the late stages of embryogenesis (Fig. 1B). To analyze the localization of Lgr4 expression in embryonic kidney tissues, in situ hybridization was performed. Lgr4 mRNA was localized in the renal vesicles and tubule progenitor epithelium (including comma-shaped body and S-shaped body) (Fig. 1C). Signals were not detected in the negative control experiments (data not shown).
Defects of Lgr4 Result in Dilated Tubule Formation
As illustrated in Figure 2, the kidneys of Lgr4−/− embryos were smaller than those of Lgr4+/+ embryos indicating renal hypoplasia (Fig. 2A, B). Histological analysis revealed that Lgr4−/− kidneys formed tubules that were not obviously different from those of Lgr4+/+ littermates until E15.5. By E16.5, all of the Lgr4−/− kidneys clearly showed dilated tubules (cysts) (Fig. 2C–F; n>5). Although two studies have revealed the requirement of Lgr4 for kidney morphogenesis, cystic phenotypes during developmental stages have not been mentioned, and no detailed analysis of Lgr4 in kidney development has yet been described (Mazerbourg et al., 2004; Kato et al., 2006).
Impaired Branching Morphogenesis Is Observed in the Lgr4−/− Kidneys
Gdnf-Ret signaling is required for the maintenance of Wnt11 expression in the UB tip cells, and Wnt11 promotes Gdnf expression in the surrounding MM, suggesting that Gdnf, Ret, and Wnt11 participate in an autoregulatory feedback loop that regulates ureteric branching morphogenesis.
As illustrated in Figure 2B, Lgr4−/− embryos exhibited smaller sized kidneys at E16.5. Next, we collected metanephric kidneys from embryos (Lgr4+/+ or Lgr4−/−) at E15.5, subjected them to RNA extraction, and examined the effect of Lgr4 deletion on the expression of ureteric tip cell-specific genes using quantitative RT-PCR. In contrast to Lgr4+/+ kidneys, the expression of Ret (Lgr4+/+ vs. Lgr4−/−: 0.95±0.10 vs. 0.31±0.04, P=0.0084) and Wnt11 (Lgr4+/+ vs. Lgr4−/−: 1.00±0.06 vs. 0.53±0.01, P=0.0002) was markedly reduced in Lgr4−/− kidneys at E15.5. On the other hand, the expression of Gdnf (Lgr4+/+ vs. Lgr4−/−: 0.97±0.21 vs. 0.82±0.14, P=0.65) in the surrounding MM was not affected in the Lgr4−/− kidneys at E15.5 (Fig. 3A). To further investigate the abnormal gene expression in UB tip cells in Lgr4−/−, we analyzed Ret and Wnt11 expression by in situ hybridization at E15.5, which demonstrated that it was similarly reduced, as shown in Figure 3A (n=3) (Fig. 3B–E).
Lgr4−/− Kidneys Show Premature Differentiation With Reduced Expression of Gata3 and Lef1
During the development of metanephric kidneys, DBA is downregulated in ureteric tip cells (Michael et al., 2007; Grote et al., 2008; Marose et al., 2008). Here, DBA staining performed on metanephric kidney at E15.5 showed that the Lgr4−/− tips of UB were clearly positive in comparison with the Lgr4+/+ tips (n>4) (Fig. 4A–D). Furthermore, we examined the expression of another marker of the differentiated collecting duct, the water channel protein, Aqp3 (Marose et al., 2008). Similar to DBA, Aqp3 protein was not detectable in the Lgr4+/+ UB but was expressed in the Lgr4−/− UB (n=4) (Fig. 4E, F). These results strongly suggest that the UB of Lgr4−/− underwent premature differentiation.
A similar phenotype was reported in mice lacking β-catenin and Gata3 (Grote et al., 2008; Marose et al., 2008). Grote et al. (2008) have proposed a genetic cascade in which Gata3 acts downstream of β-catenin to prevent premature cell differentiation and they also have reported that Gata3 acts independently of Ret to maintain the cells in an undifferentiated state (Grote et al., 2008). We collected metanephric kidneys from Lgr4+/+ or Lgr4−/− embryos at E15.5 and subjected them to RNA extraction. Quantitative RT-PCR detected a reduction of Gata3 expression in Lgr4−/− kidneys, although this decrease was not statistically significant (Lgr4+/+ vs. Lgr4−/−: 1.07±0.15 vs. 0.63±0.11, P=0.097) (Fig. 5A). In situ hybridization was used to confirm this reduction in Gata3 expression (n=3) (Fig. 5B, C). Recently, it has been reported that an appropriate amount of β-catenin is necessary to maintain UB cells in an undifferentiated state (Marose et al., 2008). Additionally, the importance of β-catenin/TCF/Lef1 signaling in the ureteric bud epithelial differentiation has been reviewed (Schmidt-Ott and Barasch, 2008). Through the use of quantitative RT-PCR and immunohistolochemical analysis, we found that the level of expression of β-catenin in the kidneys of Lgr4−/− appeared similar to that in Lgr4+/+ kidneys (Fig. 5D), and no obvious difference in the level of activation of β-catenin in the Lgr4−/− UB was detected (n=3) (Fig. 5E, F). Immunostaining was then used to examine the expression of Wnt effectors Lef1 in the UB at E15.5 in Lgr4−/−. In Lgr4−/− embryos, although the expression of Lef1 was unaltered in the renal vesicles and tubule progenitor epithelium, a significant reduction was observed in the UB of Lgr4−/− (n=4) (Fig. 5G, H; UB). These results suggested a possible signal cascade by which Lgr4 maintains UB cells in an undifferentiated state via the β-catenin/TCF/Lef1 signaling pathway in normal kidney morphogenesis (see Discussion section). Wnt/β-catenin/TCF/Lef1 signaling is also required to induce the nephron formation (Park et al., 2007; Schmidt-Ott et al., 2007; Schmidt-Ott and Barasch, 2008). Lef1 signals in RV and TU in the Lgr4−/− embryos showed that nephron formation was normally induced in the Lgr4−/− embryos.
Two studies have revealed the requirement of Lgr4 for kidney morphogenesis, but no detailed analysis of Lgr4 in kidney development has yet been described (Mazerbourg et al., 2004; Kato et al., 2006). Mazerbourg et al. (2004) reported that the size of Lgr4 mutant kidneys was reduced but, other than size, no description suggesting any mechanism of Lgr4 in the development of kidney was given. In addition, they reported no significant differences in the structure of the kidney, serum glucose, AST, BUN or urine BUN and creatinine levels between wild-type and homozygous newborn mice (Mazerbourg et al., 2004). On the other hand, Kato et al. (2006) previously demonstrated that Lgr4 null mice generated on a 129Ola × C57BL/6J hybrid background were born at a significantly lower rate than the expected 25% after crossing Lgr4 heterozygous mice. The 129Ola × C57BL/6J Lgr4 null mice that survived in utero died shortly after birth in almost all cases, and histological analysis of the P0 null mice kidney only showed a notable decrease in the total number and density of the glomeruli. In addition to morphological changes, serological tests showed elevated concentrations of plasma creatinine (Kato et al., 2006). In these two reports, detailed analyses to clarify the functions of Lgr4 in the kidney morphogenesis were not carried out.
In this study, Lgr4 knockout mice on a 129Ola × C57BL/6J hybrid background were backcrossed for at least 10 generations to C57BL/6J mice. On the C57BL/6J background, loss of Lgr4 resulted in severe hypoplasia with cystic kidneys by E16.5. Quantitative RT-PCR and in situ hybridization analysis showed reduced expression of branching morphogenesis-related genes (Ret and Wnt11) in Lgr4−/− kidneys. The Lgr4−/− kidneys further showed signs of premature differentiation of the UB with reduced expression of Lef1 and Gata3 in the UB, and the cysts of Lgr4−/− kidneys were composed of DBA-positive cells (see Supp. Fig. S1, which is available online). The hypoplasia phenotype may be due to loss of Ret-Wnt11 leading to premature termination of branching morphogenesis or a failure in maintaining undifferentiated cells of the UB tips. The importance of Lef1 function in Wnt signaling has been long established in numerous models of development and morphogenesis. In such developmental signal transduction cascades, Wnt can stimulate the transcriptional activation of numerous developmental genes through the activation of β-catenin-Lef1 complexes in the nucleus; Lef1 is also a downstream target of the canonical Wnt pathway (Hovanes et al., 2001; Filali et al., 2002; Handrigan and Richman, 2010). Therefore, expression of Lef1 indicates when and where the canonical Wnt pathway is active during kidney development. We observed a significant reduction of Lef1 in the Lgr4−/− embryos, and this result suggested that Wnt signaling might be blocked. Furthermore, the expression of another canonical Wnt-signaling target gene Axin2 (Jho et al., 2002; Grote et al., 2008; Handrigan and Richman, 2010) could not be detected in Lgr4−/− embryos, supporting our speculation (Fig. 5I, J; UB). Our results provide the basis for a new model in which the Lgr4 pathway makes a critical contribution in controlling the cell state in the UB via β-catenin/TCF/Lef1 signaling (Fig. 5K). In wild-type kidneys, Lgr4, which encodes a transmembrane receptor, is expressed and localized to renal vesicles and more differentiated epithelial cells. Subsequently, it may promote the secretion of its downstream molecule(s) (illustrated as X). The X may directly or indirectly regulate canonical Wnt pathway, followed by appropriate differentiation via Gata3 expression in the UB. Grote et al. (2008) have identified Gata3 as a mediator of β-catenin function and demonstrated that the β-catenin/Gata3 pathway maintains the cell in an undifferentiated state. Despite the fact that Lef1 and Axin2 expression mostly disappeared, in situ hybridization demonstrated significantly decreased but not absent Gata3 expression in the UB of Lgr4−/− embryos. This suggested the presence of a cross-talking pathway, which might induce the expression of Gata3, not via Lgr4 or the Wnt pathway, and another signaling pathway maintaining the UB in an undifferentiated state, not via Gata3. Further study with β-catenin mutants and Lgr4−/− mice might clarify the mechanism that maintains the UB in an undifferentiated state. Furthermore, crossing Lgr4 mutants with a new Hoxb7-Gata3 TG line might reveal more about Lgr4 pathway and its role in kidney development.
In the present work, we also observed reduced expression of branching morphogenesis-related genes (Ret and Wnt11) in the Lgr4−/− kidneys (Fig. 3). It has been previously reported that Ret expression is regulated by several genes, including β-catenin, Gata3, Emx2, retinoic acid, as well as others (Mendelsohn et al., 1999; Schedl, 2007; Bridgewater et al., 2008; Grote et al., 2008). Our results suggest that Lgr4 pathway makes a critical contribution to controlling the β-catenin/TCF/Lef1 signaling and Gata3 expression in UB via molecule(s) (X). Hence, reduced expression of Ret observed in Lgr4−/− kidneys may be caused by the impaired regulation of β-catenin/TCF/Lef1 signaling or Gata3. On the other hand, in our preliminary data, the expression of Ret in Lgr4−/− kidneys was similar to Lgr4+/+ at E14.5 (data not shown). Further analysis of the relationship between Lgr4 and Ret expression will provide new insights into the biology of kidney development.
In summary, we have shown a novel function of Lgr4 in regulating the state of epithelial cells, providing an important clue to the role of Lgr4 in not only kidney formation but the morphogenesis of other duct structures such as efferent ducts (Mendive et al., 2006; Hoshii et al., 2007; Li et al., 2010) and cystic ducts (Yamashita et al., 2009). In this study, we could not identify the candidate(s) of factor X. Identification of the candidate(s) of factor X using microarray analysis followed by ISH with the identified gene and analysis of the UB of its knockout mice would further advance our understanding the role of Lgr4 in kidney morphogenesis.
Lgr4 knockout mice (129Ola × C57BL/6J hybrid background) (Kato et al., 2006) were backcrossed 10 times with C57BL/6J male and female mice. To obtain appropriate embryos, Lgr4+/− female mice were housed with adult Lgr4+/− male mice overnight. The homozygous Lgr4 mice with C57BL/6J background are referred to as Lgr4−/−. The morning of the day of plugging was considered E0.5. The care and use of mice in this study were approved by the Institution Animal Care and Use Committee of Tohoku University.
Histology and Immunostaining
For the kidney histology, the embryos were fixed in 4% formaldehyde. After dehydration, the samples were embedded in paraffin, and the paraffin blocks were sectioned at 5-μm thickness and stained with H&E (hematoxylin-eosin). In the immunological staining, which was detected with avidin–biotinylated enzyme complex or with Alexa 594-labeled goat anti-mouse antibody (Invitrogen; 1:200), paraffin-embedded sections of kidney were analyzed with the following antibodies and conjugate: rabbit monoclonal antibodies against Lef1 (Cell Signaling Technology, Danvers, MA; 1:200), mouse monoclonal antibodies against active-β-catenin (ABC) (Millipore, Billerica, MA; 1:1,000), rabbit polyclonal antibodies against Aqp3 (Millipore; 1:2,000), and FITC conjugated dolichos biflorus agglutinin (DBA) (Vector Labs, Burlingame, CA; 1:200). These antibodies were diluted with 5% goat serum or 1% BSA in TBS (Tris-buffered saline) and applied overnight at 4°C. DBA-lectin staining was visualized with FITC.
In Situ Hybridization
In situ hybridization analysis was performed on paraffin or frozen sections of embryonic kidney with probes for Lgr4, Ret, Wnt11, Gata3, and Axin2. The probes were generated by PCR amplification of the coding sequence of these genes. Non-radioactive in situ hybridization was performed as described previously (Kawamata et al., 2007). The samples were prehybridized in hybridization buffer (50% formamide, 5×SSC, 1 mg/ml yeast tRNA, 2% Blocking Powder (Roche, Nutley, NJ), 0.1% Tritonx-100, 0.5% CHAPS, 5 mM EDTA, 50 μg/ml heparin) without an RNA probe for 1 hr and then hybridized with an RNA probe for 16 hr. Hybridization was detected using an anti-DIG Fab (Roche) coupled to alkaline phosphatase using NTB/BCIP solution (Roche).
Metanephric kidneys from 4–6 embryos (Lgr4+/+ and Lgr4−/−) at E15.5 were collected and subjected to RNA extraction with TRIZOL reagent. cDNA was then synthesized from 2 μg of total RNA in 20 μl of the reaction mixture according to standard procedures. For quantitative PCR, cDNA was added to 20 μl of the reaction mixture containing 10 μl THUNDERBIRD SYBR qPCR Mix (TOYOBO) and 1 μl of 12.5 μM primers (forward and reverse). For each sample, a parallel reaction was set up with acidic ribosomal phosphoprotein PO (arbp) primers for the endogenous control. The primer sequences used were described previously or as shown in Table 2 (Kato et al., 2007; Mohri et al., 2008). The reactions were run in a DNA Engine Opticon System (MJ Research, Tokyo, Japan).
Table 2. Primer Sets Used in This Study
Product size (bp)
We thank Dr. G. Yamada and Dr. T. Maruyama for their valuable support. We are very grateful to Dr. M.Yanagita for reading the manuscript. Y.M. was supported by Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists.