| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Carcinogenesis,* The University of Texas M. D. Anderson Cancer Center, Science Park–Research Division, Smithville, Texas; and CIIT Centers for Health Research, Research Triangle Park, North Carolina
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Homozygous deficiency of the Tsc2 gene leads to embryonic lethality. Tsc2Ek/Ek embryos die in midgestation, during the rat equivalent of mouse embryonic days 9.5 to 13.5, when Tsc2 is strongly expressed in embryonic neuroepithelium.17 During this period, embryos lacking functional Tsc2 display dysraphia and papillary overgrowth of the neuroepithelium, indicating that loss of Tsc2 gene function disrupts the normal development of this tissue. This is also a key period for the development of the metanephric kidney, which emerges at approximately embryonic day 12 to 12.5. However, overlap between the window of lethality and this developmental window has precluded an examination of the effect of loss of the Tsc2 gene on the development of the kidney.
Eker rat RCCs arise as a result of loss of function of the wild-type Tsc2 allele, leading to the development of chromophilic lesions, primarily in the renal cortex. In spontaneous tumors, Tsc2 inactivation generally occurs as a result of loss of heterozygosity (LOH) at the Tsc2 locus,7,11,18
although point mutations in the carcinogen-induced RCC have also been reported.10,13,18
Loss of the normal Tsc2 allele occurs early in the process of tumorigenesis19
and loss of the wild-type Tsc2 allele in preneoplastic lesions coincides with overexpression of transforming growth factor (TGF)-
,20
which is also overexpressed in human RCC.21,22
Interestingly, renal cysts develop frequently in Eker rats, and so are a significant feature of tumor development in this animal model. Preneoplastic dysplasias, both spontaneous and carcinogen-induced, often develop as cystic lesions in both the proximal and distal nephron.23
Carcinogens such as the nephrotoxicant and nephrocarcinogen hydroquinine and its active metabolite Tris-hydroquinine (Tris-HQ) also induce renal adenomas and RCCs in Eker rats.19
Like spontaneous tumors, the tumors that develop in animals treated with these compounds have also lost the normal Tsc2 allele.19
Laser capture microdissection has been used to show that LOH occurs in preneoplastic lesions induced with Tris-HQ.19
Preneoplastic lesions induced by Tris-HQ, dysplasias described as "toxic tubules," are primarily cystic. These lesions have proliferative polyploid projections of dysplastic cells from cyst walls of tubules that have undergone regeneration after injury.19
Both the adenomas and carcinomas that develop after loss of tuberin function also often have a prominent cystic component.23
Thus, in the Eker rat model, there appears to be a link between cystagenesis and tumorigenesis, implicating the Tsc2 gene in both processes.
Recently, we identified a functional link between tuberin and polycystin-1, the product of the polycystic kidney disease 1 (Pkd1) gene.24 Autosomal-dominant polycystic kidney disease is the most common potentially lethal genetic disease in humans.25-27 Most autosomal-dominant polycystic kidney disease occurs as a result of defects in the PKD1 gene that result in loss of function of polycystin-1 and the development of epithelial cysts in the kidney, liver, and less frequently, other organs, including the pancreas, spleen, and arachnoid and seminal vesicles.28,29 The TSC2 and PKD1 genes are adjacent on human chromosome 16, and cases of extremely severe infantile autosomal-dominant polycystic kidney disease have been described in which the predisposing genetic alteration is a co-deletion of one allele each of TSC2 and PKD1.30-32 In addition to this genetic evidence that these two genes participate in this disease, a functional linkage has been established between tuberin and polycystin-1. Tuberin is required for appropriate cellular localization of polycystin-1 to the basal-lateral aspect of the plasma membrane and in tuberin-null cells, polycystin-1 fails to localize to adherens junctions at the cell surface and becomes sequestered in the Golgi.24 Loss of the Tsc2 gene was shown to result in functional inactivation of polycystin-1 in the absence of genetic alterations at the Pkd1 locus,24 suggesting that in the Eker rat model, the cystic characteristic of preneoplastic lesions and tumors may be a result of functional inactivation of polycystin-1 in Tsc2-null cells.
In this report, we describe the consequence of loss of the Tsc2 suppressor gene during development and the genetic basis for a severe form of polycystic kidney disease that occurs in very young Eker rats. In these animals, loss of tuberin as a consequence of LOH of Tsc2 during embryogenesis resulted in a severe form of bilateral polycystic kidney disease and the development of extrarenal neoplasms in very young animals. We used Tsc2 and Pkd1 polymorphisms to assess the genetic status of both of these genes in kidney lesions from affected animals and EKT2, a cell line derived from polycystic kidney epithelial cells. These data established that while the Pkd1 gene was intact in affected animals, LOH at the Tsc2 locus resulting from chromosome duplication and nondisjunction during embryonic development resulted in mosaicism for loss of tuberin function, subsequent development of polycystic kidney disease, and neoplastic transformation of renal tubular epithelial cells.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The EKT2 cell line was established from a polycystic/neoplastic kidney of Eker rat CIIT-2. EKT2 cells were grown in DF8/3T3-conditioned medium33 containing 10% fetal bovine serum and were maintained at 37°C in a humidified incubator with 5% CO2. To determine the doubling time for EKT2 and the ERC18 cell line,12 cells were plated in triplicates in 24-well dishes at 5 x 104 per well in growth medium. Cells were harvested by trypsin at 24-hour intervals and counted with a Coulter counter (Coulter Corp., Hialeah, FL). Doubling times (t1/2) were calculated from the logarithmic portion of the growth curves.
Karyotyping and Fluorescence in Situ Hybridization
Karyotyping and fluorescence in situ hybridization analysis were performed to confirm the status of Pkd1 alleles in EKT2 cells. Cells were disassociated by trypsin treatment, centrifuged at 300 x g for 5 minutes, resuspended in 7.5 mmol/L of KCl at room temperature for 15 minutes, and fixed in methanol:glacial acid (3:1, v:v) mixture. Cells were washed three times with fresh fixative, placed on microscope slides and air-dried. Metaphase chromosomes were analyzed by quinacrine/Hoechest 33258 staining. The PKD1 probe for fluorescence in situ hybridization analysis was labeled with biotin-16-dUTP (Boehringer Mannheim, Indianapolis, IN) by nick-translation, purified by ethanol precipitation, and dissolved in 20 µl of formamide. The probe was denatured at 95°C and mixed with hybridization buffer consisting of one part of bovine serum albumin (Boehringer Mannheim), two parts of 10x standard saline citrate (SSC), and two parts of 50% dextran sulfate (Sigma, St. Louis, MO). Denatured chromosomes were incubated with hybridization buffer containing labeled probe at 37°C for 15 hours in a humidified chamber. After hybridization, the slides were washed sequentially at 37°C in 50% formamide/2x SSC, 2x SSC, and 1x SSC for 15 minutes, with a final wash in 4x SSC for 5 minutes. The slides were immersed in 70 µl of 4x SSC containing 3 µg/ml of fluorescein isothiocyanate-avidin (Vector Laboratories, Burlingame, CA), and 1% bovine serum albumin for 45 minutes at 37°C. The slides were washed for 5 minutes in 4x SSC, 4x SSC containing 0.05% Triton X-100, and 4x SSC, mounted in anti-fade solution [10% phosphate-buffered saline (PBS), 1% diazabicyclooctane, 1 µg/ml 4',6'-diamidino-2-phenylindole, and 1 mg/ml p-phenylenediamine, all from Sigma]. Labeled chromosomes were analyzed with an Olympus fluorescence microscope.
Tumorigenicity
Tumorigenicity of EKT2 cells was assessed using the nude mice assay. Mice were injected subcutaneously at two sites per animal with EKT2 cells suspended in 0.5 ml of sterile PBS. Six mice were injected with 106 cells and another six received 0.5 x 106 cells per site. After 3 months, all of the mice were terminated using CO2 euthanasia and examined for lesions. In addition to visual examination, injection sites were dissected and evaluated histologically.
Western Blot Analysis
For tuberin detection, tissues and cells were lysed in lysis buffer [PBS containing 0.1% sodium dodecyl sulfate, 1% Nonidet P-40, 0.5% deoxycholic acid, and 1 mmol/L phenylmethanesulfonyl fluoride (Sigma), 1 µg/ml aprotinin (Sigma), 1 µg/ml leupeptin (Sigma)] on ice for 60 minutes and centrifuged at 10,000 x g at 4°C for 20 minutes to remove the nuclei. Protein concentration was determined with BCA Protein Assay Reagent (Pierce, Rockford, IL). Thirty µg of protein from each lysate was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electrotransferred to polyvinylidene difluoride membranes. Nonspecific antigens were blocked by incubation at room temperature for 1 hour with 5% nonfat milk in TBST buffer (20 mmol/L Tris-HCl, pH 7.4, 136 mmol/L NaCl, and 0.05% Tween-20). The membranes were incubated with tuberin antibody C-20 (diluted 1:250; Santa Cruz Biotechnology, Santa Cruz, CA) for 2 hours in 1x TBST containing 0.5% bovine serum albumin, washed three times in TBST at room temperature for 10 minutes and once in TBS, and incubated with secondary horseradish peroxidase-conjugated anti-rabbit antibody (diluted 1:2000; Santa Cruz Biotechnology) for 1 hour in 1x TBST containing 1% nonfat milk. The membranes were washed three times in 1x TBST, and tuberin was detected by using the LumiGLO Chemilumine Substrate Kit (Kirkegaard and Perry Laboratories, Gaithersburg, MD).
RNA and DNA Isolation
Affected tissues were dissected free of normal tissue, frozen in liquid nitrogen, and ground with a mortar and pestle under liquid nitrogen. Cultured cells were washed twice with phosphate-buffered saline (PBS), then removed directly from their plates by scraping into guanidine isothiocyanate according to standard methods. Total RNA and genomic DNA were extracted from the cell lines and frozen tissues with the RNeasy and DNeasy kits (Qiagen Inc., Valencia, CA) and quantitated spectrophotometrically. Genomic DNA was also isolated with the DNeasy Kit (Qiagen Inc.) from formalin-fixed and paraffin-embedded tissues that were deparaffinized with xylene.
Northern Analysis
Aliquots of poly(A)+ RNA (5 µg) from EKT2 cells and normal rat kidney were denatured, separated in 1.0% formaldehyde-agarose gels, transferred to nitrocellulose membranes by capillary blotting, and baked at 80°C for 2 hours, according to standard methods. cDNA probes for rat TGF-34
and GAPDH were labeled with [-32P]dCTP to a specific activity of >108 cpm/µg by random oligonucleotide priming according to the manufacturer’s recommendations (Prime-It II; Stratagene, La Jolla, CA). Hybridization with nitrocellulose membranes was performed in 40% formamide/3x SSC at 42°C. Washes were at 35°C in 1x SSC/0.1% sodium dodecyl sulfate. Membrane filters were exposed to X-ray film with intensifying screens for 48 to 72 hours.
Southern Blot and DNA Genotyping
High-molecular weight genomic DNA was isolated from tissues and the EKT2 and ERC15 cell lines by proteinase K (Promega, Madison, WI) digestion at 37°C for 16 to 18 hours, extraction with phenol/chloroform/isoamyl alcohol (24:24:1, v:v:v), and treatment with DNase-free RNase (Boehringer Mannheim). The DNA was ethanol precipitated, dissolved in TE buffer (pH 8.0), and quantitated spectrophotometrically. Fifteen µg of each DNA sample was digested with XbaI and BglII, electrophoresed in 1.0% agarose gels, and transferred to nitrocellulose membranes (Schleicher & Schuell, Keene, NH). Hybridization was performed in 5x SSC containing 1x Denhardt’s, 50 mmol/L KPO4, 0.25% (W:V) sodium dodecyl sulfate, 250 µg/ml salmon sperm DNA (Ambion, Austin, TX), 50% formamide, and 106 cpm/ml 32P-labeled Pkd1 probe at 42°C overnight. The membranes were washed twice for 5 minutes each time in 1x SSC at room temperature and four times for 30 minutes each time in 0.1x SSC containing 0.05% sodium dodecyl sulfate at 50°C. The Pkd1 probes for Southern analysis were obtained by reverse transcriptase (RT)-polymerase chain reaction (PCR) using normal rat kidney RNA. RT-PCR products were gel-purified and labeled with [-32P]dCTP using the Prime-it II Random Primer Labeling kit (Stratagene). The primers used to generate the rat Pkd1 3' fragment were 5'-TGG ACA CCA CTC AGT ATT ACC-3' (forward) and 5'-CAC AGT AGT CCT GCC CTT GCT-3' (reverse). The primers used to generate the Pkd1 5' fragment were 5'-ACT TCC TCC CTG CCC ATT-3' (forward) and 5'-CCT GGT AAC CTT GGA GAC TA-3' (reverse). We examined EKT2 cells and tissues from Eker rats for Tsc2 LOH by using our previously developed PCR-high-performance liquid chromatography (HPLC) approach.18
Briefly, we performed PCR using a common forward primer for both the wild-type and mutant Tsc2 alleles (located in intron 30) and reverse primers specific for the mutant (located in the Eker mutation) and wild-type alleles (located in exon 31) for 35 cycles with denaturation at 95°C for 30 seconds, annealing at 56°C for 30 seconds, and elongation at 72°C for 30 seconds. Under these conditions, a 180-bp product specific for mutant Tsc2 allele and a 240-bp product specific for the wild-type Tsc2 allele were amplified. After amplification, 5 to 10 µl of PCR mixture were analyzed on the WAVE HPLC system with a DNA separation column (Transgenomic, Santa Clara, CA) under nondenaturing conditions. Under these conditions, PCR fragments were separated by HPLC based on their size, with the 180-bp product being eluted at 4.3 minutes and the 240-bp product eluted at 5.3 minutes.
Tsc2 Polymorphism Detection by Denaturing HPLC (DHPLC)
To identify DNA polymorphisms in the wild-type and mutant Tsc2 alleles, we amplified 17 overlapping RT-PCR fragments covering the entire coding region of the Tsc2 gene by using the primer pairs shown in Table 1
. The RT-PCR was performed according to standard protocols and the annealing temperatures for PCR were calculated for each pair of primers based on their sequence. To detect the heteroduplexes, the RT-PCR products were denatured at 95°C for 5 minutes and allowed to reanneal by slow cooling to room temperature. Five to 10 µl of each RT-PCR mixture was analyzed by denaturing HPLC (DHPLC). The gradient of acetonitrile and melting temperature (Table 1)
for each RT-PCR Tsc2 fragment were determined with WAVE 4 software (Transgenomic) depending on the fragment sequence and size. Three alternative melting temperatures (predicted, 1°C lower, and 1°C higher than predicted) were used to confirm the presence or absence of heteroduplexes indicative of a nucleotide difference. To identify the specific nucleotides that differed between the wild-type and mutant Tsc2 alleles, we sequenced the fragments that formed heteroduplexes with an ABI 377 automatic sequencer by using standard methods.
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We observed three instances of animals from colonies of Eker rats (CIIT-1, CIIT-2, UTMDA-3), two males and one female, that became moribund or died at an extremely young age (< 3 months) with severe bilateral renomegaly. Necropsy of these animals revealed extensive bilateral polycystic kidney disease. The characteristic macroscopic finding was severe bilateral renomegaly with cyst formation. In each animal, both kidneys were massively enlarged and exhibited diffuse bilateral involvement, with the exception of UTMDA-3, which had bilateral disease but segmental involvement of the right kidney. The affected portions of the right and left kidneys of UTMDA-3 were identical in macroscopic appearance to the diffuse bilateral renomegaly in CIIT-1 and CIIT-2. The kidneys were enlarged with multiple cysts that were generally spherical and unilocular and ranged from 1 mm to 1 cm in diameter. A few cysts were multilocular and cylindrical or fusiform. The kidneys maintained a reniform shape and had a smooth cortical surface. On cut surface, the cysts were distributed homogeneously throughout the cortical and medullary regions and obliterated the normal cortical and medullary zones. Intercystic tissue was abnormal in color and consistency and appeared as gray firm infiltrate. Gray nodular foci (1 mm to 1 cm in size) were scattered throughout the cortical and medullary regions in intercystic regions. The renal pelvis and papilla were not discernable structures with the exception of the unaffected region of right kidney of UTMDA-3.
Microscopic examination of the massively enlarged kidneys revealed a polycystic renal parenchyma with relatively little normal kidney tissue (Figure 1)
. The most prominent feature of the admixture of cysts was that the epithelial lining of the cyst walls consisted of proliferative neoplastic cells. Small cysts were lined by so-called "chromatophilic renal tubular epithelial cells" that piled up in multiple layers to form "atypical hyperplasias." These dysplasias have been described previously as "preneoplastic" lesions in the rat.23
In some instances, these lesions consisted of a simple layer of cuboidal to low-columnar epithelium, whereas in other areas the lesions formed intraluminal papillary excrescences that obliterated the cystic spaces. The larger affected regions were similar to cystpapillary renal cell adenomas and, in some areas, carcinomas because of the central areas of necrosis and cellular atypia. Throughout the kidneys the normal renal parenchyma was replaced by cysts whose walls contained an admixture of histological tumor patterns including solid, tubular, cystpapillary, trabecular, and comedo variants. There were no discrete areas of neoplasia within affected kidneys that exhibited instead a diffuse involvement. Normal renal structures including glomeruli were limited to intercystic cortical and medullary regions, except in the unaffected region of the right kidney of UTMDA-3, which had relatively normal renal parenchyma at the ventral pole.
|
In UTMDA-3, additional tumors were observed in both the uterus and spleen, which are very frequent sites of tumor development in older Tsc2 Ek/+ rats but have not been previously reported to develop tumors in such young animals. The histological appearance of these proliferative lesions in young Eker rats was similar to that previously reported in older Tsc2 mutant rats. Interestingly, however, the angiomatous proliferation in the spleen of UTMDA-3 presented as a diffuse organ involvement rather than as a focal circumscribed lesion previously noted in the adult-onset cases. Microscopically the splenic lesion was characterized by diffuse proliferation in red-pulp regions of vascular channels lined by endothelial cells devoid of cellular atypia (Figure 2)
. Histologically, these splenic vascular proliferations have been described as benign vascular neoplasms (hemangiomas) although they may represent a hamartomatous change.2
The extensive bilateral polycystic kidney disease in each of the affected animals, the diffuse involvement of the spleen with vascular proliferation, and involvement of multiple organs in such young animals, suggested that a genetic alteration during embryogenesis, rather than multiple independent somatic events, was responsible for the observed pathophysiology.
|
We established a primary explant from a portion of the polycystic kidney from animal CIIT-2 to facilitate the phenotypic and genotypic analysis of affected cells.24 To characterize this cell line, cells from this explant were cultured and serially passaged in vitro >100 population doublings to derive the cell line designated EKT2. We determined the karyotype, doubling time in vitro and tumorigenicity in vivo to characterize the phenotype of EKT2 cells, and determined how they compared to RCC-derived cells in which Tsc2 was inactivated somatically.
As shown in Figure 3
, EKT2 cells were nearly diploid, except for a single translocation involving chromosomes 4 and 6 [+der(6)t(4:6)(q11:q24)] resulting in trisomy of the q arm of chromosome 4. EKT2 cells also contained two cytogenetically normal copies of chromosome 10, where both the Tsc2 and Pkd1 genes are located. Using a cosmid containing the rat Pkd1 gene, we were able to demonstrate by fluorescence in situ hybridization analysis that EKT2 cells retained two copies of the Pkd1 gene (data not shown). Consistent with their near diploid karyotype, EKT2 cells were not tumorigenic in nude mice (0 to 24 sites injected), similar to RCC-derived cell lines from Eker rats.12
These cells had a doubling time of 26 hours, and similar to other cell lines exhibiting LOH at the Tsc2 locus, EKT2 cells also overexpressed TGF- (Figure 4)
, confirming that expression of this growth factor is associated with loss of Tsc2 gene function.20,35
|
|
To directly assess the tuberin status of the EKT2 cell line and affected and unaffected tissues, Western analysis was performed on frozen tissues and cell lysates. As previously reported for EKT2 cells, the affected regions from both the right and left kidneys of UTMDA-3 and the uterine leiomyoma from this animal did not express tuberin, whereas several unaffected tissues, including the brain and liver did (Figure 5)
. Because EKT2 cells retained two copies of chromosome 10, lack of tuberin expression was not because of monosomy 10. Therefore, we performed a genetic analysis to determine the mechanism responsible for loss of tuberin in the affected lesions and EKT2 cells.
|
Originally the Eker mutation was identified in the Wistar strain of rats and was subsequentially crossed and maintained on a Long-Evans background. We hypothesized that polymorphisms in the Pkd1 gene that we had previously identified24
were because of the retention of sequences characteristic of its Wistar origin. Therefore, while the wild-type Tsc2 allele was from the Long-Evans background, we analyzed the full coding region of the Tsc2 gene to identify strain-specific DNA polymorphisms that could discriminate between the Wistar and Long-Evans Tsc2 alleles. Using 17 PCR primer pairs designed to span the entire coding region of the Tsc2 gene (Table 1)
and DHPLC, we identified a nucleotide change in fragment 4 (exons 10 to 12) (Figure 6)
. Sequencing of this fragment revealed that exon 10 of the wild-type Tsc2 gene (the Long-Evans allele) contained a C at position 1092 (codon 364), whereas the inactive allele containing the Eker mutation (the Wistar allele) had a T at this position. The T-to-C substitution at position 1092 did not alter the amino acid sequence, indicating that this substitution is a DNA polymorphism between the Wistar and the Long-Evans rat strains.
|
. An XbaI polymorphism that discriminated between a 9.0-kb fragment in Long-Evans rats and a 5.6-kb fragment in Wistar rats at the 3' end of Pkd1 and a BglII polymorphism that discriminated between a 7.0-kb fragment in Long-Evans rats and a 7.4-kb fragment in Wistar rats at the 5' end of Pkd1 were identified. That these polymorphisms could discriminate between Pkd1 alleles adjacent to the mutant (Wistar) and the wild-type (Long-Evans) Tsc2 alleles was demonstrated by using Tsc2Ek/+ (ERC15 cells), which retain heterozygosity at the Tsc2 locus and contain two Pkd1 alleles (Figure 7)
.
|
To determine whether inactivation of the wild-type allele of the Tsc2 gene in the EKT2 cell line and affected kidneys occurred at the DNA level or the RNA level, we combined LOH analyses using a previously developed PCR-HPLC approach18
and the Tsc2 exon 10 polymorphism described above. At the DNA level, the allele ratio for the 240-bp wild-type and 180-bp mutant PCR products from heterozygous cells was empirically determined by HPLC to be 1.23 ± 0.24 (n = 24) and LOH was defined as an allele ratio >1.71 (ie, >2 SD away from the mean).18
Calculation of allele ratio from the HPLC tracings shown in Table 2
demonstrated that both affected kidneys from UTMDA-3 and EKT2 cells had lost the wild-type Tsc2 allele (Figure 8A)
. This finding was confirmed at the RNA level by using the Tsc2 exon 10 polymorphism (Figure 8B)
. Both affected kidneys from UTMDA-3 and the EKT2 cell line expressed only RNA transcripts from the mutant Tsc2 allele. Consistent with loss of the wild-type Tsc2 allele, both EKT2 cells and affected kidneys exhibited loss of the Pkd1 allele adjacent to the wild-type Tsc2 gene (Figure 8C)
. However, Southern analysis demonstrated that EKT2 cells contained two copies of the Wistar Pkd1 allele adjacent to the mutant Tsc2 gene, consistent with the presence of two copies of chromosome 10 in these cells. Loss of the wild-type Tsc2 allele in the polycystic kidney of CIIT2 from which EKT2 cells were derived was further confirmed by PCR analysis of paraffin sections from the affected kidney (Figure 9)
. The presence of two copies of chromosome 10 in EKT2 cells, both containing mutant Tsc2 alleles, and the fact that the polycystic kidney disease in CIIT-2 was bilateral suggested that chromosome nondisjunction during embryogenesis caused the loss of the wild-type Tsc2 allele. However, germline homozygosity for the Eker mutation resulting in loss of Tsc2 function is embryonic lethal,17
suggesting that the affected animals were in fact mosaics.
|
|
|
. Similar to EKT2 cells and the affected kidney of CIIT-2, the affected kidney, spleen, and uterine leiomyoma from UTMDA-3 had lost the normal Tsc2 allele and had only mutant Tsc2 alleles (Table 2)
. Consistent with the absence of the wild-type allele, RNA isolated from affected kidney, spleen, and uterine leiomyoma was transcribed exclusively from the mutant allele, whereas RNA from unaffected tissues (stomach, brain, and liver) and the unaffected portion of the right kidney contained transcripts from both Tsc2 alleles (Figures 8 and 10)
. These data indicate that affected animals were indeed mosaics, composed of affected organs containing cells that had lost the wild-type Tsc2 allele (kidney, spleen, and possibly uterus) and normal organs (such as stomach, brain, and liver) that retained an apparently normal wild-type Tsc2 allele. This mosaicism appears to be responsible for the development of severe, bilateral polycystic kidney disease and early onset lesions in the spleen and uterus in animals that were otherwise phenotypically normal.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Although the histological and cytological patterns of the neoplastic epithelial proliferations in the cystic kidneys in the cases described in this report were quite diverse, there were no other tumor types within the renal parenchyma. This is important because nephroblastomas, renal mesenchymal tumors, and transitional cell tumors have all been reported in rats and are histologically quite different from the epithelial tumors of the tubular epithelia. Nephroblastomas in rats have been well characterized and usually have a characteristic tripartite pattern, consisting of blastemal, epithelial, and stromal elements.37 These are composed of nests, cords, and islands of poorly differentiated cells, morphologically resembling the embryonal renal blastema. In the cases reported here, despite early involvement of the kidney, there was no evidence of involvement of renal cell types other than differentiated renal tubular epithelial cells. None of the cystic kidneys had any evidence of nephroblastoma development, nor any evidence of formation of other renal cell tumors such as the angiomyolipoma commonly noted in patients with tuberous sclerosis. The involvement of the spleen in UTMDA-3 was also unusual in that the lesion developed very early and had a diffuse red pulp involvement. The numerous well-formed vascular channels lined by relatively normal appearing endothelial cells made this lesion more reminiscent of a hamartomatous vascular proliferation rather than a true neoplastic hemangioma. The role of Tsc2 in vascular pathobiology remains to be determined, but vascular neoplasms or hamartomas are common in the spleen and occur more rarely in other visceral, subcutaneous, and dermal sites in Eker rats.2,38
The bilateral nature of the polycystic kidney disease and extent of splenic involvement in the affected animals suggests that loss of the wild-type Tsc2 gene occurred early in the development of these organs. Cytogenetic analysis and Tsc2 and Pkd1 LOH studies indicated that the affected cells contained two copies of the Tsc2 gene carrying the Eker mutation and two copies of the adjacent Pkd1 gene. The result was genetic inactivation of the Tsc2 gene, loss of its gene product, tuberin, and apparent functional inactivation of polycystin-1, the product of the Pkd1 gene that requires tuberin for normal function. The cytogenetic and LOH data suggested that chromosome nondisjunction may have been responsible for loss of the wild-type Tsc2 gene in at least one of these cases, as the chromosome carrying the wild-type Tsc2 gene was replaced with the one carrying the mutant allele. That a very young animal developed uterine leiomyoma, which normally develops in only mature females (>10 months old) is also consistent with an early loss-of-function event. In the kidney, the consequence of this apparent developmental loss of Tsc2 was early development of severe renal cysts leading to polycystic kidney disease and neoplastic transformation of renal epithelial cells.
The cystic lesions in the polycystic kidney disease may be ascribed to loss of normal polycystin-1 function. As demonstrated previously, when polycystin-1 is expressed in the absence of tuberin, it becomes sequestered in the Golgi and fails to transit appropriately to the plasma membrane, resulting in functional inactivation of the protein.24 Tubular dysmorphogenesis observed during the development of polycystic kidney disease has been hypothesized to result from altered extracellular matrix interactions of the tubular epithelial cells.39 Polycystin is co-localized at the basolateral aspect of the cell membrane with E-cadherin and ß-catenin in adherens junctions.40-42 Loss of polycystin localization to the plasma membrane in the absence of tuberin could potentially alter cell-cell and cell-matrix interactions, contributing to both the altered cell proliferation and altered production of extracellular matrix, which are hallmarks of polycystic kidney disease.
The Pkd1 gene is 63 bp away from the Tsc2 gene, and these two genes are transcribed in a tail-to-tail manner, with overlapping 3'-untranslated regions.43 The Tsc2 allele in which the Eker mutation occurs was originally derived from a Wistar rat.6,44 The spontaneous mutation that arose in these animals was subsequently crossed onto a Long-Evans background. We established that for the Tsc2 exon 10 polymorphism, the mutant Eker allele is of Wistar origin and the wild-type allele is Long-Evans. Our data indicate that despite the large size of the Tsc2 and Pkd1 genes and fact that the Eker mutation has been retained on the Long-Evans background for >20 years, most likely these genes appear to be in linkage disequilibrium, as a result of their close proximity and the continued selection for the Eker mutation in this animal cancer model.
Most importantly, renal cells that had lost the Tsc2 gene in the affected kidney were neoplastic cells of the type associated with the transformation of mature tubule epithelial cells (adenocarcinoma) rather than embryonic lesions (nephroblastoma) associated with transformation of developing metanephric kidney. Nephroblastoma is associated with loss of function of the Wilms’ tumor suppressor gene (Wt-1).45,46 A previous study of Eker rats treated transplacentally with nitrosamines demonstrated that the induction of mutations in the Tsc2 gene during kidney development very frequently results in early-onset RCCs that lack Wt-1 mutations, rather than in nephroblastoma.47 The proliferative renal tubular cyst linings with cytological and histological characteristics of transformed mature tubule epithelial cells in the polycystic kidneys reported here further suggests that the function of the Tsc2 tumor suppressor gene is conditional on the stage of kidney development, affecting renal epithelial cells but not the nephroblasts from which they arise. These data suggest that loss of tuberin function may have little consequence for nephroblasts until these cells have passed beyond the mesenchymal-epithelial transition and developed into epithelial cells. At this later period of kidney development, the absence of tuberin has a dramatic impact, causing these epithelial cells to become transformed.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by National Institutes of Health grants CA63613 (to C. L. W.), CA16672, and ES07784.
Accepted for publication October 17, 2002.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Related articles in Am J Pathol:
This article has been cited by other articles:
 |
|
 |
|
  Y.-X. Zhang, H. Meng, D.-R. Zhong, Y.-X. Jiang, Q. Dai, and H. Zhang Cardiac Rhabdomyoma and Renal Cyst in a Fetus: Early Onset of Tuberous Sclerosis With Renal Cystic Disease J. Ultrasound Med., June 1, 2008; 27(6): 979 - 982. [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Marx and W. F. Simonds Hereditary Hormone Excess: Genes, Molecular Pathways, and Syndromes Endocr. Rev., August 1, 2005; 26(5): 615 - 661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Ward and D. E. Devor-Henneman Mouse Models of Human Familial Cancer Syndromes Toxicol Pathol, January 1, 2004; 32(1_suppl): 90 - 98. [Abstract] [PDF]
|
|
|
|
 |
|
 D. M. Hockenbery Nailing Down a Link between Tuberin and Renal Cysts Am. J. Pathol., February 1, 2003; 162(2): 369 - 371. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
