Wilms Tumor 1

A number sign (#) is used with this entry because Wilms tumor-1 (WT1) is caused by heterozygous mutation in the WT1 gene (607102) on chromosome 11p13.

Wilms tumor is the most common renal tumor of childhood, occurring with an incidence of 1 in 10,000 and with a median age of diagnosis between 3 and 4 years of age. Wilms tumours are thought to develop from abnormally persistent embryonal cells within nephrogenic rests. Histologically, Wilms tumor mirrors the development of the normal kidney and classically consists of 3 cell types: blastema, epithelia, and stroma (summary by Slade et al., 2010).

Genetic Heterogeneity of Wilms Tumor

Susceptibility to Wilms tumor is genetically heterogeneous. WT2 (194071) is caused by mutation in the H19/IGF2-imprinting control region (ICR1; 616186) on chromosome 11p15. WT3 (194090) represents a locus mapped to chromosome 16q. WT4 (601363) represents a locus mapped to chromosome 17q12-q21. WT5 (601583) is caused by mutation in the POU6F2 gene (609062) on chromosome 7p14. WT6 (616806) is caused by mutation in the REST gene (600571) on chromosome 4q12.

Mutations in the BRCA2 gene (600185) have also been reported in Wilms tumor. Rare somatic and constitutional disruption of the HACE1 gene (610876) has also been reported in Wilms tumor.

Somatic mutations in the glypican-3 gene (GPC3; 300037) have been described in Wilms tumor. Somatic mutations in the WTX gene (300647) on the single X allele in tumors from males and on the active X allele in tumors from females have also been described.

Wilms tumor (WT) is one of the most common solid tumors of childhood, occurring in 1 in 10,000 children and accounting for 8% of childhood cancers. It is believed to result from malignant transformation of abnormally persistent renal stem cells which retain embryonic differentiation potential (Breslow and Beckwith, 1982; Rahman et al., 1996).

The risk of Wilms tumor is increased in association with several recognizable congenital malformation syndromes, although these cases account for less than 5% of all clinical patients with Wilms tumor (Tsuchida et al., 1995).

The 'WAGR' syndrome (194072) is characterized by susceptibility to Wilms tumor, aniridia, genitourinary abnormalities, and mental retardation; WAGR is a 'contiguous gene syndrome' in which a constitutional deletion on chromosome 11p13 affects several contiguous genes, resulting in a constellation of defects (Schmickel, 1986; Park et al., 1993).

Meadows et al. (1974) described a family in which the mother had congenital hemihypertrophy (235000) and 3 of her children had Wilms tumor. A fourth child had a urinary tract anomaly. In 1 of the children the Wilms tumor was bilateral and in a second it was multicentric.

Bond (1975) found associated congenital anomalies in 5 of 11 cases of bilateral Wilms tumor and in only 3 of 76 cases of unilateral Wilms tumor.

Beckwith (1998) provided useful data on the age at diagnosis of the first Wilms tumor in cases of syndrome-associated WT. Among 121 cases of Beckwith-Wiedemann syndrome (BWS; 130650), 96% were diagnosed by age 8 years; the oldest BWS patient had WT detected at 10 years, 2 months. Among 203 patients with hemihyperplasia, 94% were detected by age 8; the oldest HH patient had WT detected at 12 years, 4 months. Among 61 WAGR patients, Wilms tumor was detected in 98% by age 6 years; the oldest WAGR patient had WT detected at 7 years, 3 months. Among 52 patients with Denys-Drash syndrome (DDS; 194080), WT was detected in 96% by age 5 years; the oldest DDS patient had WT detected at 6 years of age.

Germline mutations cause less than 5% of Wilms tumors; most WTs are sporadic (Bove, 1999). However, numerous instances of multiple sibs with Wilms tumor have been described (Fitzgerald and Hardin, 1955). Strom (1957) described a family with 5 cases in 3 generations. A healthy male had 2 affected children (out of 5) by 1 wife and 1 affected child by another wife. A sister and an aunt of his had died in infancy or early childhood of abdominal tumor. Brown et al. (1972) reported the occurrence of Wilms tumor in 4 members of 3 successive generations of a family: the proband, a girl, her mother, aunt, and grandfather. The presence of Wilms tumor was histopathologically confirmed in 3 of the 4 cases. The right kidney was affected first in all. The aunt eventually developed Wilms tumor of the left kidney leading to her death at age 7. Jolles (1973) described Wilms tumor in a 30-month-old girl and hypernephroma in her 67-year-old paternal grandmother.

Matsunaga (1981) concluded that inheritance in familial cases, 'which constitute less than 1% of all' cases, is autosomal dominant with incomplete penetrance and variable expressivity. About 20% of familial cases are bilateral; about 3% of sporadic cases are bilateral. Bilateral cases may also be familial. Matsunaga (1981) further concluded that his 'host resistance model,' in which possession of a suppressor gene results in failure of tumor induction in carriers of the tumor gene, fit the data.

Knudson and Strong (1972) reviewed and summarized data on 58 familial cases of Wilms tumor. They concluded that bilateral tumors are more likely to be familial, that familial tumors result from 2 mutations, 1 germinal and 1 somatic, and that sporadic tumors result from 2 somatic mutations. Work of Fearon et al. (1984), Koufos et al. (1984), Orkin et al. (1984), and Reeve et al. (1984) demonstrated that homozygosity of 11p change is present in Wilms tumor, thus providing support for the Knudson hypothesis.

Comings (1973) proposed that dominantly inherited tumors may arise through inactivation or loss of a pair of regulatory genes that normally suppress the expression of a structural transforming gene.

From gene dosage studies, Narahara et al. (1984) concluded that both the CAT and the WAGR loci are in the chromosome segment 11p1306-p1305, with CAT distal to WAGR. Nakagome et al. (1984) concluded that deletion of the middle part of the 11p13 band is crucial to the WAGR syndrome; others had suggested that the distal half is of critical importance.

Turleau et al. (1984) reviewed a total of 42 cases. On the basis of a boy with deletion of most of 11p13, low catalase, nephroblastoma, chordee and cryptorchidism but normal irides and no mental retardation, Turleau et al. (1984) suggested that the determinant of aniridia may be separate from that for nephroblastoma. The authors pointed out that in all published cases with aniridia the distal half of 11p13 is deleted whereas in their presently reported case there was 'a tiny residual distal segment.' The observation might suggest the order cen--CAT--WILMS--aniridia--tel; however, Narahara et al. (1984) placed the catalase locus distal to the WAGR locus.

Mapping studies by de Martinville and Francke (1983, 1984) appeared to rule out a close physical association between HRAS1 (190020) and the region responsible for predisposition to Wilms tumor. Deletion, however, may bring them together. They placed HRAS1, HBB and insulin in the 11p15-p14.2 segment. By somatic cell hybridization, Junien et al. (1984) found that HRAS1 maps to 11p15.5-p15.1. In 4 cases of deletion of 11p13 with WAGR, they found that the restriction enzyme digestion patterns typical of HRAS1 were present. Thus, HRAS1 is not deleted in WAGR, a finding consistent with the difference in mapping. Reeve et al. (1984) demonstrated loss of HRAS in a sporadic case of Wilms tumor. Pointing out the conflicting evidence on the location of HRAS, they concluded that until the chromosomal location of HRAS has been determined with certainty, one cannot exclude a possible functional involvement of this oncogene in Wilms tumor development. By molecular genetic studies of cells from a patient with aniridia-Wilms tumor, Michalopoulos et al. (1985) concluded that a deletion visible cytologically in 11p13 deleted the catalase loci but not the LDHA locus, which is proximal, nor insulin, gamma-globin loci, HRAS1 and calcitonin, which are located distally.

Van Heyningen et al. (1985) studied 5 persons with constitutional deletions of 11p. All had aniridia; 2 had had a Wilms tumor removed. Using a cDNA probe for catalase, they showed that the CAT locus was deleted in 4 of 5 and that it must be proximal to the Wilms and aniridia loci. HBB and CALC were deleted in none; therefore, these loci are likely to be outside the region 11p15.4-p12. A region of 11p associated with Wilms tumor has also been tied to rhabdomyosarcoma and hepatoblastoma (Koufos et al., 1985); see WT2 (194071). All 3 of these rare embryonal cancers occur in the Beckwith-Wiedemann syndrome (130650).

By use of RFLPs that map to 11p, Raizis et al. (1985) detected mitotic recombination as the mechanism of homozygosity in a Wilms tumor. Their findings showed that insulin and beta-globin had come to homozygosity in the tumor but PTH remained heterozygous. Thus, PTH must be proximal to 11p13, the cytologically determined site of the Wilms tumor 'gene.' Scoggin et al. (1985) showed that E7-associated cell-surface antigen encoded by chromosome 11 and defined by a monoclonal antibody is deleted in cases of WAGR. This antigen is probably the same as that previously called 'a1' (151250). The studies in cases of WAGR with small deletions of 11p permitted regionalization of the assignment of antigen a1 to 11p13.

In 4 cases of Wilms tumor, Reeve et al. (1985) found that transcripts of insulin-like growth factor II (IGF2; 147470) were highly elevated as compared with adjacent normal kidney. By in situ hybridization, Reeve et al. (1985) mapped the IGF2 gene to 11p14.1, close to the WAGR locus. Francke (1990) pointed out that the Wilms tumor site is close to that of IGF2, which is a candidate gene for Wilms tumor at that site.

Lewis and Yeger (1987) mapped the Wilms tumor region with 4 clones that were derived from the area of deletion in 11p, together with somatic cell hybrids containing chromosome 11 from leukemic T cells with translocation t(11;14), from fibroblasts from a familial aniridia patient with translocation t(4;11), and from lymphoblasts from a patient with Wilms tumor and deletion of 11p but no aniridia. The following order was deduced: centromere--CAT--T-cell break--aniridia break--FSHB--telomere.

Porteous et al. (1987) used chromosome-mediated gene transfer to provide an enriched source of DNA markers for 11p. They defined 10 distinct regions of 11p, 5 of which subdivided band 11p13. They also mapped 2 independent 11p13 translocation breakpoints to within the smallest region of overlap (SRO) defined by WAGR deletions. The first came from a patient with familial aniridia, and the second was found in a neonate with the clinical features of Potter facies and the pathologic features of genitourinary dysplasia, with urethral and ureteral atresia and bilateral undescended testes. Porteous et al. (1987) raised the question of whether Wilms tumor and genitourinary dysplasia are alternative manifestations of mutation at the same locus.

Using the fluorescence-activated cell sorter to select a series of somatic cell hybrids with deleted translocated chromosome 11 segregated from its normal homolog, Seawright et al. (1988) analyzed these cell hybrids with gene-specific probes and for cell-surface marker expression to order the markers and find an SRO for WAGR. They found that FSHB maps distal to WAGR and CAT maps proximal to it. Two translocation breakpoints in 11p13 (1 associated with familial aniridia and 1 with a sporadic case of congenital renal dysfunction resulting from urethral and ureteral atresia) mapped within this SRO.

Puissant et al. (1988) reported a patient with WAGR and a de novo reciprocal translocation 46,XY,t(5;11)(q11;p13). On Southern blot analysis, the gene encoding catalase had been deleted, but the gene encoding the beta subunit of follicle-stimulating hormone (FSHB) was intact.

Using a range of probes for chromosome 11, Mannens et al. (1988) demonstrated that loss of heterozygosity in Wilms tumors may not necessarily involve the proposed Wilms tumor locus at 11p13 and may be limited to 11p15.5. A separate gene coding for genitourinary dysplasia (symbolized GUD) was suggested by Bonetta et al. (1989), who found that the deletion breakpoint of a translocation t(11;2)(p13;p11) in a patient with Potter facies and genitourinary dysplasia mapped to the same 225-kb pulsed field gel electrophoresis fragment as did the fragment deleted in Wilms tumor. Van Heyningen et al. (1990) suggested that the Wilms tumor gene itself may be responsible for abnormalities of genitourinary development in WAGR as a pleiotropic effect. The suggestion was based on the observations that the tumor predisposition and the genitourinary malformations map to precisely the same area and that the WT candidate gene shows expression in both the developing kidney and gonads.

Grundy et al. (1988) and Huff et al. (1988) presented evidence that another form of autosomal dominant Wilms tumor does not map to 11p. In a multipoint linkage analysis of a family segregating for Wilms tumor, Grundy et al. (1988) excluded the mutation from both 11p13 and 11p15. In a second family, the 11p15 alleles lost in the tumor were derived from the affected parent, thus excluding this region as the location of the inherited mutation. Huff et al. (1988) likewise studied a Wilms tumor family with 7 DNA markers that spanned the 11p13 region and found no linkage.

Schwartz et al. (1991) used a highly informative CA repeat polymorphism within the WT1 gene to study a family with 6 persons with Wilms tumor in 3 generations. They demonstrated that a predisposition to WT did not segregate with the polymorphism; furthermore, linkage analysis excluded WT predisposition from 11p15 also.

In the cells of a Wilms tumor unassociated with the WAGR syndrome and with normal constitutional chromosomes, Kaneko et al. (1981) found an interstitial deletion involving the region 11p14-p13.

Reeve et al. (1985) proposed that IGF2 is the (or a) transforming gene in Wilms tumor. Scott et al. (1985) pointed out that Wilms tumor is histologically indistinguishable from the early stages of kidney development. In 12 sporadic cases of Wilms tumor, Scott et al. (1985), like Reeve et al. (1985), found that expression of the IGF2 gene was markedly increased relative to adult tissues, but was comparable to the level of expression in several fetal tissues including kidney, liver, adrenal, and striated muscle. Although this may merely reflect the stage of tumor differentiation, the possibility that IGF2 is involved in the transformation process was raised.

Weissman et al. (1987) explored the role of the 11p13 deletion in Wilms tumor by introducing a normal human chromosome 11 into a Wilms tumor cell line by means of the microcell transfer technique. The ability of the cells containing the normal chromosome 11 to form tumors in nude mice was completely suppressed.

Dao et al. (1987) examined the karyotype and chromosome 11 genotype of normal and tumor tissues from 13 childhood renal tumor patients. Tumors of 8 of the 12 Wilms tumor patients showed molecular evidence of loss of 11p DNA sequences by somatic recombination (4 cases), chromosome loss (2 cases), and recombination (2 cases) or chromosome loss and duplication. One malignant rhabdoid tumor in a patient heterozygous for multiple 11p markers did not show any tumor-specific 11p alteration. One of the patients had Perlman syndrome (renal hamartomas, nephroblastomatosis, and fetal gigantism; 267000).

Kumar et al. (1987) demonstrated deletion of 11p14-p12 in a Wilms tumor removed from a 9-month-old boy with aniridia. Kozman et al. (1989) found loss of alleles from 11p in a Wilms tumor in a 37-year-old male. The finding indicated a common pathogenesis of childhood and adult types and suggested that molecular genetic studies may be useful in the differentiation of Wilms tumor from renal cell carcinoma or sarcoma when the histologic findings are unclear.

Whereas morphologic transformation of normal human cells by BK virus (BKV) and by BKV DNA and its subgenomic fragments occurs in very low frequency, de Ronde et al. (1988) found that 4 individuals with various deletions in the short arm of one chromosome 11 were unusually susceptible to morphologic transformation. They suggested that the susceptibility might be explained by a 'transformation suppressor' locus situated within the deleted region. The deleted region included that of WAGR; the 'transformation suppressor' locus may be identical to the Wilms tumor locus.

Francke (1990) commented on the significance of the putative gene for Wilms tumor. She compared the finding with that in retinoblastoma where a single gene locus has been found to be responsible. She reviewed the evidence for at least 3 genes capable of producing Wilms tumor: one in 11p13, one in 11p15.5 (194071) and at least one other not situated in either of these regions (194090). In the case of Wilms tumor, it is possible that changes at several sites are collaborative, or perhaps more likely that changes at several alternative sites result in the same tumor. A third model is that of hierarchical gene interaction. If the function of the gene at 11p13 is to turn off the gene at 11p15.5, then loss of 11p13 expression would have the same effect as mutation or allele loss at 11p15.5.

In 2 cases in which a Wilms tumor contained a somatic WT1 mutation, Park et al. (1993) found that nephrogenic rests in the same kidneys had the identical mutation. Thus, nephrogenic rests and Wilms tumor are topographically distinct lesions that are clonally derived from an early renal stem cell. Inactivation of WT1 appears to be an early genetic event that can lead to the formation of nephrogenic rests, enhancing the probability that additional genetic hits will lead to Wilms tumor.

In the case of both P53 (191170) and RB1 (614041), characterization of the tumor suppressor genes was provided by the dramatic growth-suppressing properties when the genes were reintroduced into cells containing inactivated endogenous genes. Similar studies in Wilms tumors had been complicated by the existence of multiple genetic loci implicated in different subsets of tumors and by the unavailability of appropriate target cell lines. To obtain an appropriate cell line for studying WT1 function, Haber et al. (1993) inoculated minced human Wilms tumors subcutaneously into nude mice and then adapted the tumor explants to growth in vitro. They found that 1 cell line could be propagated indefinitely in tissue culture without loss of tumorigenic potential. Transfection of each of 4 wildtype WT1 isoforms suppressed the growth of these cells. The endogenous WT1 transcript in these cells was devoid of exon 2 sequences, a splicing alteration that was also detected in varying amounts in all Wilms tumors tested but not in normal kidney. Production of this abnormal transcript, which encodes a functionally altered protein, may represent a distinct mechanism for inactivating WT1 in Wilms tumors.

Miyagawa et al. (1998) focused on the ectopic formation of skeletal muscle in Wilms tumors. They presented evidence supporting a negative regulatory role for WT1 in myogenesis. Their findings suggested that the metanephric-mesenchymal stem cells of the kidney may have the capacity to differentiate into skeletal muscle cells as well as epithelial cells. Normally, the expression of WT1 appears to prevent this ectopic differentiation program from being activated. In vitro studies suggested that WT1 may play a direct role in suppressing the formation of skeletal muscle.

Rakheja et al. (2014) reported the whole-exome sequencing of 44 Wilms tumors, which identified missense mutations in the microRNA (miRNA)-processing enzymes DROSHA (608828) and DICER1 (606241), and novel mutations in MYCN (164840), SMARCA4 (603254), and ARID1A (603024). Examination of tumor miRNA expression, in vitro processing assays, and genomic editing in human cells demonstrated that DICER1 and DROSHA mutations influence miRNA processing through distinct mechanisms. DICER1 RNase IIIB mutations preferentially impair processing of miRNAs deriving from the 5-prime arm of pre-miRNA hairpins, while DROSHA RNase IIIB mutations globally inhibit miRNA biogenesis through a dominant-negative mechanism. Both DROSHA and DICER1 mutations impair expression of tumor-suppressing miRNAs, including the LET7 family (see 605386), which are important regulators of MYCN, LIN28 (see 611043), and other Wilms tumor oncogenes. Rakheja et al. (2014) concluded that these results provided insights into the mechanisms through which mutations in miRNA biogenesis components reprogram miRNA expression in human cancer and suggested that these defects define a distinct subclass of Wilms tumors.

Imprinting

Schroeder et al. (1987) found that in 5 patients with Wilms tumor and 2 others previously reported, there was a loss of chromosome 11 alleles and that these alleles in all 7 cases were of maternal origin. All of these tumors were sporadic. The authors concluded that the initial mutation, either germinal or somatic, must have occurred on the paternal chromosome. There was no occupational history pointing to an increased risk of mutation in the fathers and, on the average, paternal age was not increased. They stated that the probability of all 7 patients losing the maternal allele in their Wilms tumor tissues, if the loss is indeed random, is less than 1%. By means of RFLPs, Huff et al. (1990) demonstrated that 7 of 8 de novo deletions of band 11p13 were of paternal origin. The 1 case of maternal origin was unremarkable in terms of the size or extent of the deletion, and the child developed Wilms tumor. Transmission of 11p13 deletions by both maternal and paternal carriers of balanced translocations has been reported, although maternal inheritance predominates. These data, in addition to the general predominance of paternally derived, de novo mutations at other loci, suggested that increased frequency of paternal deletions is due to an increased germinal mutation rate in males.

Jeanpierre et al. (1990) found loss of maternal alleles from the 11p15 area of the maternal chromosome in Wilms tumor tissue and a constitutional deletion of 11p13 of the maternal chromosome. There have been other instances in which the 11p region involved in loss of heterozygosity (11p15) is different from the region involved in hereditary predisposition (11p13). See 194071.

Haber et al. (1990) described a sporadic, unilateral Wilms tumor in which 1 allele of the WT1 candidate gene contained a 25-bp deletion spanning an exon-intron junction and leading to aberrant mRNA splicing and loss of 1 of the 4 zinc finger consensus domains in the protein. The mutation was absent in the affected person's germline, consistent with the somatic inactivation of a tumor suppressor gene. In addition to the intragenic deletion affecting 1 allele, loss of heterozygosity at loci along the entire chromosome 11 indicated an earlier chromosomal nondisjunction and reduplication. Haber et al. (1992) presented evidence that this mutation of the WT1 gene behaves as a dominant negative, suppressing the function of the wildtype protein by a trans-dominant mechanism. They suspected this because the mutated allele was found to be coexpressed with the wildtype allele in a sporadic Wilms tumor. They therefore tested the ability of this mutant WT1 allele, containing an in-frame deletion within the DNA-binding domain, to transform primary baby rat kidney cells. The mutant WT1 gene was found to cooperate with the adenoviral E1A gene in transforming baby rat kidney cells. The wildtype WT1 gene in all of its alternatively spliced forms neither suppressed E1A-induced focus formation nor cooperated with E1A.

Ton et al. (1991) demonstrated that the smallest region of overlap between deletions causing Wilms tumor was a 16-kb segment of DNA encompassing one or more of the 5-prime exons of the zinc finger gene located on 11p13, together with an associated CpG island. This supported the authenticity of the zinc finger gene as the disease locus.

Kakati et al. (1991) described a family in which a son had bilateral WT and an extra ring chromosome in the lymphocytes and in kidney tissue. The size of the ring varied considerably from cell to cell. A daughter had unilateral WT and an abnormal clone containing a small ring chromosome in PHA-stimulated and EBV-transformed lymphocytes. The mother, who was unaffected, had a karyotype similar to that of the daughter with WT. Kakati et al. (1991) hypothesized that the son's large ring chromosome was an amplified form of the small ring inherited from the mother. Chromosome 11 was cytogenetically normal in all cells examined in the affected children and the unaffected mother.

Varanasi et al. (1994) analyzed the structural integrity of the entire WT1 gene in 98 sporadic Wilms tumors. By PCR-SSCP, they found that mutations in the WT1 gene are rare, occurring in only 6 tumors analyzed. In 1 sample, 2 independent intragenic mutations inactivated both WT1 alleles, providing a singular example of 2 different somatic alterations restricted to the WT1 gene. The data, together with the previously ascertained occurrence of large deletions/insertions in WT1, defined the frequency at which the WT1 gene is altered in sporadic tumors.

Nordenskjold et al. (1995) screened 27 cases of 46,XY females with gonadal dysgenesis who had previously been screened for and found not to carry SRY gene mutations (480000) to determine whether isolated gonadal dysgenesis might be due to WT mutations. Using denaturing gradient gel electrophoresis, they found a heterozygous point mutation in exon 8 in 1 of these patients: arg366-to-his, which had previously been described in a case of Denys-Drash syndrome (607102.0004). Reevaluation of the clinical data confirmed the diagnosis of Drash syndrome. Based on these results, Nordenskjold et al. (1995) concluded that isolated gonadal dysgenesis is not caused by mutations in the WT1 gene.

White et al. (2002) identified 2 nonconservative single base changes in the GPC gene (300037.0006-300037.0007) in Wilms tumor tissue only, implying a possible role of GPC3 in Wilms tumor development.

Lu et al. (2002) studied 18 cases of Wilms tumor with a novel gene-expression profiling method that targets individual chromosomes: comparative expressed sequence hybridization (CESH). Relative overexpression of genes on the long arm of chromosome 1 was shown in all tumors that subsequently relapsed, but in none of those from patients in remission.

Anglesio et al. (2004) analyzed the HACE1 gene (610876), which is located 50 kb downstream of the chromosome 6q21 breakpoint of a nonconstitutional t(6;15)(q21;q21) rearrangement in sporadic Wilms tumor. Although the HACE1 locus was not directly interrupted by the translocation in the index Wilms case, HACE1 expression was markedly lower in tumor tissue compared with adjacent normal kidney. HACE1 expression was virtually undetectable in the SK-NEP-1 Wilms tumor cell line and in 4 of 5 additional primary Wilms tumor cases compared with patient-matched normal kidney. There was no evidence of HACE1 mutation or deletion, but hypermethylation of 2 upstream CpG islands correlated with low HACE1 expression in tumor samples.

Slade et al. (2010) identified a constitutional de novo balanced translocation t(5;6)(q21;q21) in a boy who developed bilateral Wilms tumor at age 6 months. Breakpoint analysis showed that the translocation transected intron 6 of the HACE1 gene. Further analysis of the HACE1 gene in 450 individuals with Wilms tumor identified 1 patient with a constitutional truncating mutation (W364X) who inherited the mutation from her unaffected mother, suggesting either reduced penetrance or that the mutation was an unrelated finding. Slade et al. (2010) concluded that abrogation of HACE1 activity may predispose to the development of Wilms tumor, although HACE1 mutation is rare and makes only a small contribution to disease incidence.

In 2 brothers who both developed Wilms tumor and brain tumors, Reid et al. (2005) identified 2 truncating BRCA2 mutations (600185.0027; 600185.0031).

In 51 Wilms tumors tested for both gene copy alterations and intragenic mutations, Rivera et al. (2007) found inactivation of the WTX gene (300647) in approximately one-third. Tumors with WTX mutations lack WT1 mutations. In contrast to autosomal tumor suppressor genes, WTX is inactivated by a monoallelic 'single-hit' event targeting the single X chromosome from males and the active X chromosome from females.

Regev et al. (2008) reported maternal transmission of a nonsense mutation in the WT1 gene (607102.0027). The mother had Wilms tumor in infancy and decreased fertility in adulthood, and her son displayed genitourinary abnormalities, including glanular hypospadias with chordee and bilateral undescended testes, gonadal dysgenesis with gonadoblastoma foci, and intraabdominal Mullerian derivatives. The boy also had ventricular septal defect by echocardiography; no Wilms tumor was detected up to 6 years of age. Regev et al. (2008) stated that the nonsense mutation demonstrates the lack of correlation between genotype/phenotype and mutation position in the WT1 gene, the presence of intrafamilial variability, and the effect of gender on severity of genitourinary anomalies.

Royer-Pokora et al. (2010) described the establishment and characterization of long-term cell cultures derived from 5 individual Wilms tumors with WT1 mutations. Three of these tumor cell lines also had CTNNB1 (116806) mutations and an activated canonic Wnt (164820) signaling pathway as measured by beta-catenin/T cell-specific transcription factor transcriptional activity. Four lines showed loss of heterozygosity of chromosome 11p due to mitotic recombination in 11p11. Gene expression profiling revealed that the WT cell lines were highly similar to human mesenchymal stem cells (MSCs), and FACS analysis demonstrated the expression of MSC-specific surface proteins CD105 (ENG; 131195), CD90 (THY1; 188230), and CD73 (NT5E; 129190). The stem cell-like nature of the WT cells was further supported by their adipogenic, chondrogenic, osteogenic, and myogenic differentiation potentials. By generating multipotent mesenchymal precursors from paraxial mesoderm in tissue culture using embryonal stem cells, gene expression profiles of paraxial mesoderm and MSCs were described. Using these published gene sets, the authors found coexpression of a large number of genes in WT cell lines, paraxial mesoderm, and MSCs. Lineage plasticity was indicated by the simultaneous expression of genes from the mesendodermal and neuroectodermal lineages. The authors concluded that WTs with WT1 mutations have specific traits of paraxial mesoderm, which is the source of kidney stromal cells.

Schumacher et al. (1997) identified 19 hemizygous WT1 gene mutations/deletions in tissue samples from 64 patients. The histology of the tumors with mutations was stromal-predominant in 15, triphasic in 3, blastemal-predominant in 1, and unknown in 2 cases. Among 21 patients with stromal-predominant tumors, 15 had WT1 mutations and 10 of these were present in the germline. Of the patients with germline alterations, 6 had associated genitourinary (GU) tract malformations and a unilateral tumor, 2 had a bilateral tumor and normal GU tracts, and 2 had a unilateral tumor and normal GU tracts. Three mutations were tumor-specific and were found in patients with unilateral tumors without genital tract abnormalities. These data demonstrated the correlation of WT1 mutations with stromal-predominant histology, suggesting that a germline mutation in WT1 predisposes to the development of tumors with this histology. Twelve mutations were nonsense mutations resulting in truncation at different positions in the WT1 protein, and only 2 were missense mutations. Of the stromal-predominant tumors, 67% showed loss of heterozygosity, and in 1 tumor a different somatic mutation in addition to the germline mutation was identified. Thus, in a large proportion of a histopathologically distinct subset of Wilms tumors, the classic 2-hit inactivation model, with loss of a functional WT1 protein, is the underlying cause of tumor development.

By linkage analysis, McDonald et al. (1998) found evidence of a predisposition gene for Wilms tumor at 19q13.4 in 5 families. Furthermore, affected members of 2 other families demonstrated loss of heterozygosity (LOH) in this region. In addition, 10% (15 of 148) of sporadic Wilms tumors analyzed in their laboratory showed LOH at multiple markers on 19q (Ruteshouser et al., 2001).

Using polyclonal antibodies, Dressler and Douglass (1992) detected high levels of PAX2 (167409) expression in the epithelial cells of human Wilms tumors. In the mouse they showed by immunocytochemistry that expression of the Pax2 gene was localized to the nuclei of condensing mesenchyme cells and their epithelial derivatives in the developing kidney. The data suggested that Pax2 is a transcription factor that is active during the mesenchyme-to-epithelium transition in early kidney development and in Wilms tumor. Pax2 is a member of the family of genes identified in the mouse on the basis of a common protein coding domain, the paired box, first described in the Drosophila segmentation genes 'paired' and 'gooseberry.' The Pax genes are expressed during embryogenesis in a tissue-restricted manner. The gene mutant in Waardenburg syndrome type I (193500) is the human homolog of the Pax3 gene (606597) and the gene mutant in aniridia (106210) is the human homolog of the Pax6 gene.

Patek et al. (1999) reported that heterozygosity for a targeted murine Wt1 allele, which truncates zinc finger-3 at codon 396, induced mesangial sclerosis characteristic of Denys-Drash syndrome in adult heterozygous and chimeric mice. Male genital defects were also evident, and there was a single case of Wilms tumor in which the transcript of the nontargeted allele showed an exon 9 skipping event, implying a causal link between Wt1 dysfunction and Wilms tumorigenesis in mice. However, the mutant protein with the truncation at codon 396 accounted for only 5% of Wt1 protein in both heterozygous embryonic stem cells and the Wilms tumor. This has implications regarding the mechanism by which the mutant allele exerts its effect.