If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported in part by National Institutes of Health Grants DK-42921, DK-57328, and DK-67565. Further support was provided by the Van Andel Research Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Alterations in Wnt/β-catenin signaling have been linked to abnormal kidney development and tumorigenesis. To gain more insights into the effects of these alterations, we created mice carrying a conditional deletion of the Apc tumor suppressor gene specifically in the renal epithelium. As expected, the loss of Apc leads to increased levels of β-catenin protein in renal epithelium. Most of these mice die shortly after birth, and multiple kidney cysts were found upon histological examination. Only rarely did these animals survive to adulthood. Analysis of these adults revealed severely cystic kidneys associated with the presence of renal adenomas. Our results confirm an important role for proper regulation of Wnt/β-catenin signaling in renal development and provide evidence that dysregulation of the pathway can initiate tumorigenesis in the kidney.
Kidney disease is a major worldwide health problem. In the United States alone, ∼4.5% of adults over 20 years of age (7.4 million persons) have physiological evidence of chronic kidney disease as determined by a moderately or severely reduced glomerular filtration rate (
), and almost 12,000 deaths each year are caused by this cancer. Whereas great progress has been made recently in identifying genetic changes underlying kidney diseases, more needs to be done to effectively address these important clinical problems. This includes work to characterize the signaling pathways related to these diseases and to create model systems on which the factors that initiate these diseases can be assessed. One attractive candidate for such a signaling pathway is the Wnt/β-catenin pathway.
Wnts are a family of secreted proteins that bind to a receptor complex including a frizzled molecule and either Lrp5 or Lrp6. This binding initiates an intracellular signaling cascade that eventually results in the stabilization of the cytoplasmic β-catenin protein (
). Normally, the cytoplasmic (non-cadherin-associated) levels of β-catenin are low because of the constitutive activity of the serine/threonine protein kinase glycogen synthase kinase 3 (GSK3). In the absence of an upstream Wnt signal, GSK3 phosphorylates β-catenin on serine and threonine residues in its N terminus, targeting it for ubiquitin-dependent proteolysis. When the appropriate Wnts are present, GSK3 kinase activity is inhibited, leading to stabilization of β-catenin in the cytoplasm and the nucleus. Nuclear β-catenin associates with members of the Lef/TCF family of DNA-binding proteins, leading to up-regulation of specific target genes (
). One way that this occurs is via loss of the APC tumor suppressor gene. APC is necessary for the proper regulation of β-catenin; in its absence, β-catenin is not degraded via GSK3-dependent phosphorylation (
). These patients carry germline-inactivating mutations in the APC gene, and loss of the wild-type copy leads to polyp initiation. In addition, ∼85% of all sporadically occurring colon tumors have inactivation of both APC alleles (
). In all cases, the loss of APC is associated with stabilization of the β-catenin protein in the cytoplasm. Mutations in β-catenin itself, which eliminate the sites phosphorylated by GSK3, are also found in colon tumors as well as many other tumor types (
). While the major function of APC is thought to be its regulation of β-catenin, APC also interacts with a multitude of other cellular proteins including axin, plakoglobin, asef, kinesin superfamily-associated protein 3 (KAP3), EB1, microtubules, and the human homolog of Drosophila Discs large (hDLG) (
Given its important role in regulating β-catenin and other cellular processes as well as its down-regulation in some RCCs, we decided to directly test whether loss of Apc leads to the development of kidney disease in a mouse model. To do this, we conducted the present study by using the cre-lox system (
). We generated mice homozygous for the Apc-flox allele that also contained the Ksp-Cre transgene (Ksp-Cre;Apc580S/580S). We report here that these mice develop early-onset polycystic kidney disease that results in neonatal death in the large majority of these mice. The rare surviving animals develop severe cystic disease, with some of the cysts having an appearance consistent with a diagnosis of cystic nephroblastoma or renal adenoma.
Transgenic Mice and Genotyping—Ksp-Cre transgenic mice containing 1.3 kb of the Ksp-cadherin promoter linked to the coding region of Cre recombinase were established previously (
) were generated from ES cells obtained from Tetsuo Noda. Genotyping was performed by PCR analysis of tail DNA. The primers used for Cre recombinase genotyping were as follows: 5′-CGATGCAACGAGTGATGAGGTTC-3′ (CreF) and 5′-GCACGTTCACCGGCATCAAC-3′ (CreR). Apc580S genotyping was carried out using the following primers: 5′-GTTCTGTATCATGGAAAGATAGGTGG-3′(APC-flox-P3) and 5′-CACTCAAAACGCTTTTGAGGGTTGAT-3′ (Apc-flox-P4).
Histology, Immunohistochemistry, and Immunofluorescent Staining—The mice were mainly euthanized by continuous inhalation of CO2. In some cases, 2.5% avertin (2,2,2-tribromoethanol), 0.5–0.6 ml, was injected intraperitoneally for immobilization of the mice, followed by the intraventricular perfusion of phosphate-buffered saline (PBS) containing 4% paraformaldehyde for tissue fixation. The kidneys were isolated and fixed in 10% neutral buffered formalin overnight, embedded in paraffin, and sectioned at 3- to 4-μm thickness. A part of the sections were stained in hematoxylin/eosin (H&E). For immunohistochemical staining, the sections were deparaffinized and epitope retrieval was performed by heating the sections at 95 °C in 10 mm citrate buffer pH 6.0 for 25 min. After treating in 0.3% hydrogen peroxide for 30 min in room temperature, the sections were blocked by normal sera according to the instructions for the VECTASTAIN Elite ABC kit (POD; Vector Laboratories), incubated with primary antibodies at 4 °C overnight, visualized by VECTASTAIN Elite ABC kit, and counterstained in hematoxylin. The primary antibodies used were rabbit anti-mouse APC polyclonal Ab (1:100, SC-896, Santa Cruz Biotechnology), anti-β-catenin mouse monoclonal Ab (1:100, cat. no. 610154, BD Biosciences Pharmingen), and rat anti-mouse Ki-67 monoclonal Ab (1:25, TEC-3, DAKO, Glostrup, Denmark).
For the immunofluorescent and lectin staining, the paraformaldehyde-fixed, paraffin-embedded sections were dewaxed, rehydrated in graded ethanol, washed in PBS, and then pretreated with 10 mm sodium citrate (pH 6.0) in a microwave oven for 20 min or PBS containing 20 μg/ml proteinase K at room temperature for 15 min. Sections were blocked in PBS containing 10% goat serum, 0.1% bovine serum albumin, and/or 1:5 mouse IgG, then incubated with the following antibodies: rabbit anti-Tamm Horsfall protein at 1:100 dilution (gift from J. Hoyer, University of Pennsylvania), rabbit anti-Na-K-Cl cotransporter (NKCC2) at 1:400 dilution (gift from M. Knepper, National Institutes of Health), rabbit anti-Na-Cl cotransporter (NCC) at 1:100 dilution (gift from M. Knepper) (
), mouse anti-E-cadherin at 1:200 dilution (BD Biosciences), rabbit anti-aquaporin-2 at 1:400 dilution (gift from M. Knepper), rabbit anti-aquaporin-3 at 1:400 dilution (Chemicon), rabbit anti-polycystin-2 (YCC2) at 1:150 dilution, rabbit anti-polycystin-1 at 1:200 dilution (Zymed Laboratories), rabbit anti-HNF-1β at 1:200 dilution (gift from M. Pontoglio, Pasteur Institute) (
), and/or mouse anti-acetylated tubulin at 1:1000 dilution (Sigma-Aldrich). In some experiments (Fig. 10, F and H), the antibody was preincubated in PBS containing 25 μg/ml blocking peptide (Zymed Laboratories Inc.) for 30 min at room temperature. After incubation with the primary antibody for 2 h at room temperature or overnight at 4 °C, sections were washed in PBS and incubated with 1:400 dilution of anti-rabbit IgG or anti-mouse IgG conjugated to cyanine 3 (Jackson ImmunoResearch Laboratories), Alexa Fluor 488 (Molecular Probes), or Alexa Fluor 594 (Molecular Probes). Some sections were co-stained with TRITC-coupled Dolichos biflorus agglutinin or FITC-coupled Lotus tetragonolobus agglutinin (Vector Laboratories), as described previously (
). Slides were mounted in Vectashield (Vector) containing 1:1000 DAPI and photographed under epifluorescence illumination using a Zeiss Axioplan 2 microscope. Images were analyzed using OpenLab software (Improvision). Sections from mutant and wild-type animals were treated identically and photographed under identical exposure conditions.
Sections for double immunofluorescent staining were deparaffinized, rehydrated, and then boiled in 10 mm sodium citrate buffer (pH 6.0) for 10 min to increase antigen exposure. Nonspecific antibody-binding sites in the tissue sections were blocked by incubation in 5% donkey serum (30 min at room temperature). The sections were then incubated overnight at 4 °C with both the rabbit anti-Apc (1:50) and mouse anti-β-catenin (1:50) antibodies described above. The sections were then washed and incubated in rhodamine-conjugated secondary donkey anti-rabbit (1:150) and FITC-conjugated secondary donkey anti-mouse antibody (1:150) (Jackson ImmunoResearch Laboratories, West Grove, PA) for 30 min at room temperature. Confocal fluorescent microscopy was performed to evaluate the slides.
Laser Capture Microdissection (LCM) and PCR Amplification—To confirm that Apc exon 14 had been deleted in the neoplastic renal epithelium, LCM was performed on both neoplastic tissue and histologically normal glomeruli. Sections (8 μm) were cut from the paraffin blocks, stained with H&E, dehydrated through a graded ethanol series to xylene, and then allowed to air dry in a desiccator. LCM was then performed using a PixCell IIe LCM system (Arcturus Engineering, Inc., Mountain View, CA) following manufacturer's protocols. Captured cells attached to the polymer film surface on the CapShur LCM caps (Arcturus Engineering, Inc.) were put directly into 0.5-ml microcentrifuge tubes and incubated with 150 μl of digestion buffer from PicoPure DNA extraction kit (Arcturus Engineering, Inc.) at 65 °C for 24 h. After boiling for 10 min to inactivate the proteinase K in the buffer, the DNA was ready for PCR amplification.
Primers were designed to amplify the region flanked by loxP sites in the 580S allele. A third primer specific for the PGK-neo cassette in the Apc580S allele was also used. Sequences were as follows: 5′-CCTGTTCTGCAGTATGTTATCATTC-3′ (P1); 5′-AAGACACCAAGTCCAAAGCACAC-3′ (P2); and 5′-CTATCAGGACATAGCGTTGG-3′ (P3). PCR was performed using Platinum PCR SuperMix High Fidelity (Invitrogen) and 300 nm each primer. Three DNA samples were analyzed: 1) laser-captured renal neoplastic epithelium from a Ksp-Cre(+)/Apc580S/580S mouse; 2) tail DNA from the same mouse, and 3) tail DNA from a mouse homozygous for the wild-type Apc allele. The PCR conditions were the following: 94 °C/2:00 (94 °C/15 s; 55 °C/30 s; 68 °C/2:30) (35×). PCR products were analyzed on a 1% agarose ethidium bromide gel.
Blood Chemistry—The blood samples from twelve 1- to 2-day-old neonates in two litters were collected. Among them, there were four Apc580S/580S neonates, five Ksp-Cre/Apc580S/+ neonates, and three Ksp-Cre/Apc580S/580S neonates. The blood samples from four adult litters at the ages of 5–12 weeks old with each of the three different genotypes were also collected. The plasma of the mice was separated by centrifuging the heparinized blood at 3,500 × g for 15 min. For quantitative determination of BUN, the VetScan Chemistry Analyzer (Abaxis, Inc., Union City, CA) was used. The plasma of neonates was diluted 1:5 with saline prior to the measurement. The plasma of adult mice was diluted 1:3.
Statistics—Mann-Whitney nonparametric test was applied for the comparisons of the BUN levels. A one-tailed p value <0.05 was considered statistically significant.
Ksp-cre/Apc580S/580S mice develop severe cystic kidney disease associated with kidney neoplasia. The expression of Cre recombinase in Apc580S mice creates an allele that encodes a protein that prematurely truncates the Apc protein at codon 580, creating a functionally null allele (
). These mice were crossed to mice carrying the floxed allele of Apc (Apc580S/580S)to generate Ksp-cre/Apc580S/+ mice, which were subsequently crossed to Apc580S/580S mice. These matings would be predicted to generate litters in which one-fourth of the mice have the Ksp-Cre;Apc580S/580S genotype. Upon routine monitoring of these litters, we found that several neonates became moribund, necessitating sacrifice. PCR analysis identified these animals as the Ksp-Cre;Apc580S/580S genotype. Analysis of blood urea nitrogen (BUN) levels in these mice showed significant elevation in the neonatal period (Fig. 1). Histological analysis of kidneys from these neonates revealed the presence of multiple cysts (Fig. 1), consistent with the idea that kidney dysfunction leads to the neonatal death.
Most Ksp-Cre/Apc580S/580S mice die within 2–3 days of birth. We genotyped 540 weanling mice (3 weeks of age) resulting from crosses between Ksp-Cre/Apc580S/+ and Apc580S/580S mice. If there were no neonatal lethality associated with loss of Apc in the kidney, one would predict that one-fourth of the resulting offspring present at weaning would be Ksp-Cre/Apc580S/580S. Consistent with the observed neonatal lethality, only 6 of the 540 mice surviving until weaning were of the Ksp-Cre/Apc580S/580S genotype. These animals were noticeably smaller than their littermates (3–4 grams less in weight). Plasma BUN levels were elevated in these animals and they became moribund at 8–12 weeks of age (Fig. 1). Necropsy of these animals revealed the presence of severely enlarged kidneys containing multiple cysts (Fig. 2A). Histological examination of the kidneys showed they contained simple cysts lined by a single layer of epithelial cells; glomerular cysts; multilayered cysts lined by a hyperplastic epithelium; and adenomas (Fig. 2, B and C). Examination of other organs revealed no other abnormalities, and no evidence of metastatic disease was found in the adult Ksp-Cre/Apc580S/580S mice.
Confirmation of Apc Loss in the Kidneys of Ksp-Cre/Apc580S/580S Mice—To verify that cre-mediated recombination results in loss of Apc in the kidney, we performed immunohistochemistry (IHC) and PCR-based assays. IHC for Apc protein revealed expression in tubules of control animals (Fig. 3, B and C), and loss of Apc expression in the hyperplastic epithelial lining of cysts (Fig. 3D). Consistent with the incomplete penetrance of cre expression in kidney tubules, mosaic expression of cre is observed in some kidney tubules (Fig. 3D). To verify on the genomic level that Apc was lost in kidney cells, we isolated cells in the hyperplastic cyst walls by laser capture microdissection and performed PCR analysis (Fig. 3F). Deletion of exon 14 in the Apc gene (and thus inactivation of the protein) was confirmed in the hyperplastic epithelial cells (Fig. 3G).
Cytoplasmic Levels of β-Catenin Are Elevated in Kidney Cells of Ksp-Cre/Apc580S/580S Mice—We predicted that loss of Apc would lead to elevated levels of cytoplasmic β-catenin. We performed IHC for β-catenin and found that its level was dramatically elevated in Ksp-Cre/Apc580S/580S kidney tissue (Fig. 4). This included elevated levels in the hyperplastic lining of cysts. Again, we observed a mosaic pattern of overexpression in some kidney tubules, consistent with the mosaicism observed for loss of Apc expression (Figs. 3D and 4, E–G). We also anticipated that cell proliferation would be increased in these cystic kidneys. To test this, we stained kidney tissue for the proliferation marker Ki67 and found a dramatic increase in proliferating cells within cysts of Ksp-Cre/Apc580S/580S mice, consistent with deregulated growth (Fig. 4J).
Tubular Origin of the Simple Cysts—To assess the tubular origins of the simple cysts, sections from neonate (2-day-old) and adult (8-week-old) Ksp-Cre/Apc580S/580S kidneys were stained by antibodies or lectins that label specific nephron segments. The markers included Lotus tetragonolobus agglutinin (LTA), which labels proximal tubules (
). This analysis showed that cysts were derived from Bowman's capsule, proximal tubules, thick ascending limbs, distal convoluted tubules, and collecting ducts (Fig. 5). These findings are consistent with the known expression of Cre in Ksp-Cre mice in all tubular epithelial cells extending from Bowman's capsule to collecting ducts (
). Importantly, NKCC2 and NCC, which are normally located in the apical membrane of renal tubular cells (Fig. 5D), were also located in the apical membrane of cyst epithelial cells (Fig. 5, C and E). These results suggested that a major defect in epithelial polarity was not present in kidneys containing Apc-deficient cells.
To assess the origins of the cysts, adjacent sections from neonate and adult Ksp-Cre/Apc580S/580S kidneys were co-stained with LTA, THP, NKCC2, and DBA. The percentages of simple cysts that were labeled with each marker were then measured. Only those cysts that could be unequivocally identified in adjacent sections were scored. At 2 days, 43% of cysts were labeled with LTA, indicating proximal tubule origin; 19% expressed THP and/or NKCC2, indicating thick ascending limb of loop of Henle origin; and 34% were labeled with DBA, indicating collecting duct origin. 5% of the cysts were not labeled with any of the markers tested. At 8 weeks of age, 7% of cysts were labeled with LTA, 19% expressed THP and/or NKCC2, 9% were labeled with DBA, and 65% were not labeled with any of the markers. The larger number of cysts that did not express any markers at 8 weeks of age compared with 2 days of age suggests that the cyst epithelial cells were progressively de-differentiated. Simple cysts, multilayered cysts, and adenomas all expressed E-cadherin in the plasma membrane at sites of cell-cell contact, indicating that they were epithelial in origin (Fig. 6). Antibody staining showed that simple cysts derived from collecting ducts (evidenced by labeling with DBA) expressed aquaporin-2 (Fig. 6C) and aquaporin-3 (Fig. 6D). Moreover, similar to what is found in normal collecting ducts, aquaporin-2 was localized exclusively in the apical membrane, whereas aquaporin-3 was localized exclusively in the basolateral membrane, further indicating the absence of a defect in epithelial cell polarity. In contrast, the large multilayered cysts and adenomas observed in the older kidneys were labeled with DBA, indicating that they originated from collecting ducts. However, the cells comprising these structures did not express aquaporin-2 or aquaporin-3, indicating that they were relatively undifferentiated.
Expression of Polycystin-2, Polycystin-1, Hepatocyte Nuclear Factor-1β, and the Presence of Cilia—Given the potential relationship between Wnt signaling and polycystins (
), we evaluated the expression of these proteins in Ksp-Cre/Apc580S/580S mice. Polycystin-2 was highly expressed in many of the cysts (Fig. 7, A and B), including both simple cysts and multilayered cysts. Higher magnification images showed that polycystin-2 was localized in the basal cytoplasm, similar to what has been observed in wild-type renal tubules (Fig. 7, C and D). Polycystin-1 was expressed in renal tubules but was absent from glomeruli and blood vessels in control mice, consistent with previous reports (
). The staining was blocked by preincubation of the antibody with the peptide immunogen, indicating that the staining was specific (Fig. 7F). Polycystin-1 was expressed in the multilayered cysts of Ksp-Cre/Apc580S/580S mice, and the level of expression was comparable to an adjacent non-cystic tubule. Fig. 7H shows peptide blocking photographed under identical exposure conditions.
Next, we examined the expression of hepatocyte nuclear factor (HNF)-1β in the kidneys of Ksp-Cre/Apc580S/580S mice. Mutations of HNF-1β have been shown to produce glomerular cysts and multilayered hyperplastic cysts (
) similar to those seen in the Ksp-Cre/Apc580S/580S mice. Fig. 7I demonstrates HNF-1β expression in a Ksp-Cre/Apc580S/580S kidney. Expression of HNF-1β was seen in the nuclei of cells lining the simple cysts (Fig. 7J), and is also seen in the nuclei of tubules from a wild-type kidney (Fig. 7K). HNF-1β was also expressed in the nuclei of the multilayered cysts (Fig. 7L) at a comparable level to that seen in adjacent non-cystic tubules.
Defects in cilia regulation or function are associated with cystic kidney disease (
). To gain more insight into the nature of the cysts, cilia were stained with an antibody to acetylated tubulin (Fig. 7, M–P). Cilia were present on the apical surface of tubules from control mice and in cysts from 2-day-old Ksp-Cre/Apc580S/580S kidneys. In the 8-week-old Ksp-Cre/Apc580S/580S kidneys, fewer cilia were present, but they could be clearly seen on the surface of some cyst epithelial cells. However, the large multilayered cysts completely lacked cilia. Taken together, these results suggest that the development of renal cysts and neoplasms in Ksp-Cre/Apc580S/580S mice was not caused by the loss of polycystin-1, polycystin-2, or HNF-1β.
In this study, we shown that kidney-specific inactivation of the Apc gene (Ksp-Cre/Apc580S/580S mice) leads to the development of early-onset cystic disease associated with perinatal lethality. Rare surviving animals develop large, multilayered cysts with high levels of proliferative cells. These hyperplastic regions also contained areas of pleiotropy and focal necrosis, consistent with a diagnosis of renal adenoma or renal carcinoma in situ. However, we did not see evidence for invasion of these hyperplastic cells into regions of more normal kidney, and no evidence of metastatic progression to other organs was seen. These results are consistent with previous reports (
), where expression of a mutant, proteolysis-resistant form of β-catenin led to cystic renal disease. This work extends this observation to show that inactivation of Apc (and associated up-regulation of endogenous β-catenin) leads to early-onset renal failure in most animals. In addition, these studies reveal the presence of multilayered cysts lined with hyperplastic cells with characteristics of neoplasia. Thus, dysregulation of the canonical Wnt signaling pathway leads to developmental abnormalities in the kidney that can predispose to the subsequent development of renal cancer.
The cysts in Ksp-Cre/Apc580S/580S kidneys were derived from multiple cell types in the kidney including Bowman's capsule, proximal tubules, thick ascending limbs, distal convoluted tubules, and collecting ducts. These abnormalities are consistent with the expression pattern of Ksp-Cre transgene (
). Mutations in the PKD1 gene account for ∼85% of ADPKD. A majority of the remaining cases (10%) are associated with mutations in the PKD2 gene. Polycystin-1 (PC-1), the protein expressed from the PKD1 locus, encodes a large (462 kDa) 11-pass transmembrane protein (
). We assessed whether the regulation of polycystin-1 or other proteins associated with cystic renal disease were altered in Ksp-Cre/Apc580S/580S mice. Surprisingly, the absence of Apc in our animal model did not result in the loss of polycystin-1, polycystin-2, HNF-1β, or cilia in renal tubular epithelia. This implies that the formation of the cysts in the present model was due to a novel but unknown mechanism.
Germline mutations in the APC gene are found in the majority of familial adenomatous polypopsis (FAP) patients (
). In addition, there are a number of hyperplastic and neoplastic diseases associated with the mutation of the APC gene in FAP patients, including the congenital hypertrophy of the retinal pigment epithelium, periampullary carcinoma, Gardner's syndrome with epidermoid skin cysts and benign osteoid tumors of the mandible and long bones, benign fibromatosis, hepatoblastoma, papillary carcinoma of the thyroid, adrenocortical tumors, and Turcot syndrome with multiple colorectal polyps and cerebellar medulloblastoma (
). The evidence presented here confirms that loss of Apc and associated stabilization of the β-catenin protein can lead to renal tumor development.
We thank Jason Martin, Bryn Eagleson, and the rest of the VARI vivarium staff for assistance in animal husbandry; Pam Swiatek and the VARI Mouse Germline Modification Core Facility for assistance; and Bree Berguis, Eric Hudson, Jim Resau, and the rest of the VARI Histology Core Facility for support. We also thank Sheri Holmen for constructive discussion on the project and Sunshine Kucholtz and Cassandra Zylstra for technical support. We thank Rod Bronson and Robert Sigler for advice on histology. We thank Mark Knepper, Steven Somlo, John Hoyer, and Marco Pontoglio for antibodies. We thank David Nadziejka for critically reading this manuscript. We also thank Tetsvo Noda for Apc580S ES cells.