Physiol. Genomics 16: 287-300, 2004;
doi:10.1152/physiolgenomics.00164.2003
1094-8341/04 $5.00
Received 26 September 2003;
accepted in final form 10 December 2003.
Physiological Genomics 16:287-300 (2004)
1094-8341/04 $5.00 © 2004 American Physiological Society
Invited Review
Insights into the physiological role of WT1 from studies of genetically modified mice
Maria Teresa Discenza1 and
Jerry Pelletier1,2
1 Department of Biochemistry
2 McGill Cancer Center, McGill University, Montreal, Quebec H3G 1Y6, Canada
 |
ABSTRACT
|
|---|
Discenza, Maria Teresa, and Jerry Pelletier. Insights into the physiological role of WT1 from studies of genetically modified mice. Physiol Genomics 16: 287-300, 2004; 10.1152/physiolgenomics.00164.2003.—The identification of WT1 gene mutations in children with WAGR and Denys-Drash syndromes pointed toward a role for WT1 in genitourinary system development. Biochemical analysis of the different WT1 protein isoforms showed that WT1 is a transcription factor and also has the ability to bind RNA. Analysis of WT1 complexes identified several target genes and protein partners capable of interacting with WT1. Some of these studies placed WT1, its downstream targets, and protein partners in a transcriptional regulatory network that controls urogenital system development. We review herein studies on WT1 knockout and transgenic models that have been instrumental in defining a physiological role for WT1 in normal and abnormal urogenital development.
kidney development; mouse models
|
INTRODUCTION
|
|---|
WILMS TUMOR (WT), a pediatric kidney cancer, has been the subject of intense clinical and basic research for several years. The implication of the Wilms tumor 1 (WT1) tumor suppressor gene in the etiology of WT illustrated the impact that genetic alterations can have on both development and tumorigenesis. Biochemical studies of the WT1 gene and gene product, as well as the study of patients and genetically modified mice with mutations in WT1 have all contributed in defining a role for WT1 during development.
|
The WT1 Gene, mRNA Transcripts, and Protein Isoforms
|
|---|
The WT1 gene is located at human chromosome position 11p13 (Fig. 1) and on mouse chromosome 2. Both the mouse and human genes span 
50 kb of genomic DNA comprising 10 exons that encode mRNAs of 3 kb (16, 37, 38). A map of the WT1 gene structure is shown in Fig. 2.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 2. Schematic diagram of the genomic organization of WT1 (A) and of the WT1 protein (B). WT1 is encoded by 10 exons. Exons 7, 8, 9, and 10 encode WT1 zinc fingers I, II, III, and IV, respectively. The WT1 zinc finger domain is involved in both DNA and RNA binding and harbors multiple nuclear localization signals (NLS). Alternatively spliced sequences are represented by open boxes. Alternative splicing of exon 5 and alternative usage of two different splice donor sites at the 3' end of exon 9 produce different WT1 isoforms. The three translation initiation sites that have been identified are indicated in exon 1. The TGA translational stop codon is indicated. The three different amino-terminal domains produced by alternative translation start site usage are represented by different shades of gray. An RNA editing event, which alters the amino acid at position 280 from Leu to Pro, is indicated on the WT1 protein as well as the boundaries for the transcription regulation domains and the self-association domain.
|
|
WT1 exon 5 and exon 9 are alternatively spliced (37, 47). Inclusion of exon 5 inserts 17 amino acids between the NH2- and COOH-terminal domains of WT1 (Fig. 2). The WT1 gene encodes a protein with typical characteristics of a transcription factor (Fig. 2). The NH2-terminal domain of WT1 is proline- and glutamine-rich. The COOH-terminal domain of WT1 contains four zinc fingers, each one encoded by an exon, encompassing exons 7 to 10. The zinc fingers are of the Cys2-His2 variety, similar to the DNA-binding domain of the Drosophila Krüppel protein. Two nuclear localization signals have been identified in the zinc finger domain of WT1, one being in zinc finger 1 and the other in zinc fingers 2 and 3 (13). The use of an alternative splice donor site at the end of exon 9 results in the incorporation of three amino acids, lysine, threonine, and serine (KTS), between zinc fingers three and four (Fig. 2). The WT1 proteins have molecular masses of 52–54 kDa depending on the inclusion or exclusion of the two splice inserts (93). The 52-kDa protein corresponds to the WT1 isoform that lacks both alternatively spliced sequences, named WT1(-/-), and the 54-kDa protein corresponds to the WT1 isoform that contains both alternatively spliced sequences, named WT1(+/+). The WT1 isoform containing the exon 5 splice insert, but lacking the KTS insert, is named WT1(+/-), and conversely, the WT1 isoform lacking the exon 5 insert but containing the KTS insert is named WT1(-/+). Much work has been done in studying the effect of the KTS splice insert on DNA-binding and functional properties of WT1 where the presence or absence of the 17 amino acid exon 5 splice insert is of little consequence to its function. In these cases, WT1 protein isoforms are simply referred to as WT1(+KTS) and WT1(-KTS) to reflect the presence or absence of the KTS splice insert, respectively. Similarly, WT1 isoforms are often referred to as WT1(+exon 5) and WT1(-exon 5) to reflect the presence or absence of the of exon 5 splice insert.
Three translation initiation sites have been identified in the WT1 mRNA (Fig. 2). The four WT1 protein isoforms described above of molecular masses between 52 and 54 kDa are generated through the use of the major initiator AUG. Larger WT1 protein isoforms have been identified that are generated by translation initiation at an in-frame CUG codon upstream of the initiator AUG and results in WT1 protein isoforms of 60–62 kDa (14). Internal translation initiation at an in-frame AUG codon downstream of the initiator AUG generates smaller WT1 protein isoforms with molecular masses of 36–38 kDa (117).
The diversity of WT1 protein isoforms is further increased by an RNA editing event, detected in adult rat kidney and human testis, at nucleotide 839 of the WT1 mRNA that replaces leucine-280 in the WT1 protein by a proline (Fig. 2) (121). Therefore, taking into account all possible modifications, at least 24 different WT1 protein isoforms would be expected to be generated.
|
Wilms Tumor
|
|---|
WT is a nephroblastoma that occurs with a frequency of 1 in 10,000 births (81). The tumor is embryonic in origin, it is detected in children usually during the first 5 years of life and can occur unilaterally or bilaterally. A WT arises when condensed metanephric mesenchymal cells of the developing kidney fail to properly differentiate (76). When these cells persist in their undifferentiated state, they form a nephrogenic rest (8, 18). The majority of nephrogenic rests become dormant or regressing rests, and only a minority proliferate to become hyperplastic rests. A rest that undergoes genetic or epigenetic changes to become a neoplastic rest will consequently proliferate further to produce a benign lesion (adenomatous rest) or a malignant WT (reviewed in Refs. 10, 26). WTs often have triphasic histology in which areas of blastema, epithelia, and stroma can be observed. This pattern is suggestive that mesenchymal cells had already started to respond to the inductive signals that initiate nephrogenesis to produce a partial pattern of differentiation (9). Variants of WTs also occur in which one cell type predominates, such as blastema or stroma. The clinical treatment of WT involves surgical resection of the tumor or the affected kidney, which may be followed by chemotherapy and sometimes radiation therapy. Children are cured in 85% of cases (42). Anaplastic WTs (containing abnormal nuclear morphological features), however, are a subtype of tumors (<5%) displaying poor prognosis with many remaining refractory to treatment. This variant has been linked to a higher frequency of p53 mutations (5, 6).
|
WT Genetics
|
|---|
Most WTs are unilateral and are thought to arise from sporadic, somatic mutations. Ten to fifteen percent of sporadic WTs are caused by inactivating mutations of the WT1 tumor suppressor gene (20, 21, 46, 75, 127). WT1 mutations represent an early event in the etiology of WT as they have been found in nephrogenic rests (100). In the remaining cases of sporadic WT, the WT1 gene is not mutated and is expressed at high levels. It is believed that these tumors have a predisposing mutation in a gene functioning either upstream or downstream of WT1 or in another cellular pathway leading to hyperproliferation of nephrogenic rests. Genetic predisposition to Wilms tumorigenesis usually results in bilateral and/or multifocal disease and occurs in 5–10% of WT cases (42). Predisposition to WT is often found in association with one of four syndromes. These syndromes are WAGR, Denys-Drash, Beckwith-Wiedemann, and Frasier. Their characteristics are summarized in Table 1. The WT1 gene has been documented to be mutated in WAGR, Denys-Drash, and Frasier syndromes (4, 12, 65, 103, 104).
|
WT Associated Syndromes Involving Mutations in WT1
|
|---|
WAGR syndrome.
Children suffering from WAGR syndrome present with WT, aniridia (hypoplasia of the iris), genitourinary malformations (hypospadia, cryptochidism, horseshoe kidneys, renal hypoplasia) and mental retardation (106). The genes responsible for WT and aniridia in children with WAGR syndrome were identified by fine mapping analysis of cytogenetic deletions in chromosome 11p13 (Fig. 1A; Table 1) (16, 38, 125). Hemizygous deletion of the Pax 6 gene is responsible for aniridia associated with WAGR, whereas hemizygous deletion of the WT1 gene is associated with the genitourinary system malformations of this syndrome and with predisposition to the development of WT (2, 11, 16, 38, 53, 56, 104, 125).
Denys-Drash syndrome.
Denys-Drash syndrome (DDS) is another disease of the genitourinary system with predisposition to WT (Table 1) (24, 28). The genitourinary system malformations presented by children with DDS are much more severe than those presented by children with WAGR, and the incidence of WT is higher. The severity of genital abnormalities observed in DDS varies and can include severe sexual ambiguity. Internal Müllerian duct structures are sometimes found in males (41, 80), and in several cases, individuals with grossly normal external and internal female genitalia have gonads containing only testicular tissue. Thus a wide range of reproductive tract abnormalities are observed.
The glomerular nephropathy, caused by diffuse mesangial sclerosis (DMS), is the most consistent feature of DDS and leads to end-stage renal disease (48, 58). The glomerular capillary tuft is surrounded by a layer of glomerular extracellular matrix (ECM), which in turn is surrounded by a visceral epithelial layer consisting of podocyte cells. The glomerular ECM consists of two regions, the glomerular basement membrane and the glomerular mesangial matrix. The mesangial cells secrete the mesangial matrix and are in direct contact with the arteries of the glomerular capillary tuft. In DMS, all mesangial cells in all the glomeruli of the kidney produce and secrete excess fibrotic material leading to a massive accumulation of ECM surrounding the capillary tuft. This causes the arteries in the capillary tuft to collapse, subsequently resulting in hypertension. Furthermore, the podocyte cells are often underdeveloped in children with DDS.
Children with DDS contain a germ line mutant allele of the WT1 gene, and analysis of WTs from three DDS patients indicated somatic loss of heterozygosity at the second wild-type WT1 locus (103). In 1991, Pelletier et al. (103) investigated 12 children with DDS and observed that each of the 12 children had inherited missense mutations in the zinc finger region of the WT1 gene. Furthermore, it was observed that 8 of the 12 patients had exactly the same mutation. This mutation consists of an arginine to tryptophan (R to W) change at residue 394 WT1 zinc finger III and is sufficient to abrogate DNA binding by the WT1 protein to a high-affinity DNA binding site (Fig. 1B) (103). Other point mutations occurring in the zinc-finger region of the WT1 gene which interfere with DNA binding by WT1 have also been identified in children with DDS (Fig. 1B) (3, 12, 116). Several laboratories showed that WT1 can self-associate through its NH2-terminal domain and subsequently analyzed the effect of WT1 association with a DDS mutant WT1 (33, 90, 112). The data indicated that association of wild-type WT1 with DDS mutant WT1 resulted in an inhibition of WT1 transcriptional activity and in an alteration of the subnuclear localization of wild-type WT1. This suggests that mutant DDS proteins may act in a dominant-negative fashion and disrupt the function of the wild-type WT1 protein in DDS. The more severe phenotype observed in individuals with DDS compared with WAGR individuals may, therefore, be explained by a higher functional loss in WT1 activity in DDS. It remains possible, however, that the product of the DDS WT1 allele may disrupt normal genitourinary developmental pathways by dominantly interacting with other proteins.
Frasier syndrome.
Frasier syndrome is a rare disorder characterized by male pseudohermaphroditism (involving an XY karyotype, female external genitalia and streak gonads), glomerulopathy (involving focal segmental glomerulosclerosis where only some glomeruli are affected, and within those glomeruli, only certain segments of the capillary tuft are affected), and the development of genitourinary tumors (gonadoblastoma, and to a lesser extent, WT) (Table 1) (35, 50, 64). Frasier syndrome is associated with a point mutation in the intron 9 donor splice site of one WT1 allele, resulting in the loss of expression of the WT1(+KTS) isoform from that allele (Fig. 1C) (4, 65). Therefore, a change in ratio of the +KTS to -KTS isoforms is responsible for the developmental defects observed in Frasier syndrome. The different WT1 isoforms may have overlapping but not necessarily identical functions, and therefore the loss or reduction of one isoform could not be fully compensated for by the activity of one of the other isoforms.
|
Expression Pattern of the WT1 Gene
|
|---|
The association of genitourinary system malformations in humans with germ line mutations in the WT1 gene indicates that WT1 is a key regulatory factor controlling development of the genitourinary system. To be able to assess the relevance of mouse models harboring WT1 lesions, it is important to analyze the role that WT1 plays in normal physiology and development.
WT1 expression pattern during kidney development.
Studies using in situ hybridization and immunohistochemistry have shown that WT1 is expressed during all three stages of mammalian kidney development (1, 15, 94, 105, 108, 109). The earliest expression of WT1 in the mouse kidney has been reported at embryonic day 9 (E9) (1, 109). At this stage, WT1 is expressed over the urogenital ridge, in the degenerating pronephroi and in the mesonephroi. In particular, WT1 is expressed in the early vesicle stage of the mesonephroi and in the glomerular podocyte cells of the mesonephroi. In human embryos, the earliest time point at which WT1 expression has been analyzed is at around week 7 of gestation (94, 108). Similar to the mouse embryo, WT1 expression is also detected in the glomerular podocyte cells of the mesonephroi.
Metanephric kidney development commences at around E11.5 in the mouse (E36–E37 in humans). Just prior to this (mouse E11, human E34–E35), there are low levels of WT1 expression in the loose, uninduced metanephric mesenchyme. As the ureteric bud emerges from the nephric duct (mouse E11.5, human E36–E37), the metanephric mesenchyme condenses around the tip of the bud and begins to strongly express WT1. During nephrogenesis, WT1 expression remains high in the renal vesicles and comma-shaped bodies. At the S-shaped body stage, WT1 expression is turned off in all cells except those cells destined to become the podocyte cells of the glomeruli. In the adult kidney, WT1 expression remains high in the podocyte cells of the glomeruli.
WT1 expression pattern during gonad development.
In mammals, both the kidneys and the gonads are derived from the urogenital ridge (39). As mentioned previously, WT1 is expressed in the urogenital ridge early on in urogenital system development (55, 94, 105, 108, 109). In particular, WT1 is expressed in the sex cords of the genital ridge portion of the urogenital ridge, which gives rise to the gonads. WT1 is expressed in the mesenchymal compartments of the developing male and female gonads (1, 55, 94, 105, 108, 109). In the adult, WT1 expression is restricted to the Sertoli cells of the testes and the epithelial and granulosa cells of the ovary. WT1 is also expressed in the uterus (embryonic and adult), the oviduct, and endometrium of the female (105).
Other sites of WT1 expression.
Apart from the developing kidney and gonad, WT1 is also expressed during development of the spleen and the mesothelium (a lining of cells surrounding organs in the thoracic and abdominal cavities) (1, 94, 105, 108, 109). Common features of these tissues are that they are all mesodermal in origin and that they experience a switch from mesenchymal to epithelial cells at the time of WT1 expression. These same studies have also shown that WT1 is expressed in the liver, the thymus, certain areas of the brain and spinal cord, and in the abdominal wall musculature. Other studies have shown that WT1 is expressed in stem cells of the bone marrow (36), during development of the retina (128), and in the proliferating coelomic epithelium, the developing diaphragm and limb, the septum transversum, the early proepicardium, the epicardium, and the subepicardial mesenchyme (91, 92).
|
Expression Ratios and Patterns of WT1 +/-KTS and WT1 +/-exon 5 Splice Isoforms
|
|---|
Haber et al. (47) were the first to describe WT1 alternative splicing. Using the RNase protection assay (RPA), the authors quantitated the four WT1 splice isoforms and reported that the ratio of all four isoforms does not vary temporally and spatially during normal kidney development and is 8.3:2.5:3.8:1 for WT1(+/+):WT1(+/-):WT1(-/+):WT1(-/-) (47). This corresponds to ratios of 2.25:1 for the +/-exon 5 isoforms and 3.5:1 for the +/-KTS splice isoforms. The data presented in this publication, however, is difficult to interpret given the presence of multiple protected bands obtained due to the use of a long RPA probe. Subsequent studies have further examined the expression of the four WT1 splice isoforms during kidney development and in different species. Expression data from these publications, obtained from reverse transcription-PCR (RT-PCR) and RPA assays, are consistent with each other but do not corroborate the expression ratios obtained by Haber et al. (47).
The WT1 KTS splice isoform is found in species from all gnathostome classes (61). Furthermore, data shows that the +/-KTS splice isoform ratio does not vary during development of the kidney and that interspecies differences are not evident (49, 113). The +/-KTS ratio is in the range of 1.1–1.5:1 (49, 113).
Unlike the WT1 KTS splice isoform, the WT1 exon 5 splice isoform is only expressed in mammals (61). Analysis of WT1 isoform expression in human, mouse, and rat tissue has shown that there are differences in exon 5 splicing during kidney development, as well as species-specific differences in the expression of WT1 +/-exon 5 splice isoforms. RT-PCR analysis of human and mouse normal tissue, as well as human tumor tissue, showed differences in exon 5 splicing during human kidney development (113). All four WT1 splice isoforms were expressed equally in human fetal kidney, whereas human adult kidney expressed approximately twofold excess of exon 5 containing isoforms. In the same study, RT-PCR analysis of WT1 splice isoforms in WTs generally reflected the ratios seen in fetal kidneys. Work done by Iben et al. (57) confirmed the above results through the analysis of native WT1 protein in human kidney and WT samples. Mouse tissue, however, showed a relative lack of exon 5-containing transcripts. Other studies have also shown a lack of WT1 exon 5 transcripts in mouse tissue, including mouse ovarian and oviduct tissue (66), mouse uterine tissue (113), and mouse testis (61). Analysis of rat tissue, however, has shown that the uterine stroma expresses predominantly the WT1 exon 5 isoform and that, similar to humans, rat testes express an excess of exon 5-containing transcripts to a ratio of 4:1 (61, 113). Given the differences in expression patterns and ratios of the WT1 exon 5 splice isoforms between humans and rodents, the interpretation of WT1 function from phenotypic abnormalities of mouse models showing alterations in exon 5 isoform expression may not be extendable to the situation during human development.
|
DNA Binding and Transcriptional Regulation by WT1
|
|---|
To identify WT1 target genes, many studies first identified WT1 interacting sequences by using binding site selection experiments with oligonucleotides, whole genomic PCR, and DNase I footprint analysis (reviewed in Ref. 85). In general, WT1(-KTS) binds to GC-rich sequences, such as the Egr-1-like consensus sequence, 5' GCGGGGGCG 3' (110), and the WTE site, 5' GCGTGGGAGT 3' (96), or to sites containing (5'TCC3')n motifs (132).
The WT1 protein contains domains that, when fused to the GAL4 DNA-binding domain, can function independently to repress or activate transcription. A repression domain has been identified in amino acids 85–124, and an activation domain has been identified in amino acids 181–250 (133) (Fig. 2). Similarly, fusion of the 17 amino acids of the first alternatively spliced exon (exon 5) to GAL4 shows that this region is capable of repressing transcription (134), suggesting that this alternative splicing event functions to modify the transcriptional regulatory properties of WT1. The mechanism of WT1-mediated transcriptional repression has been analyzed via cotransfection assays and transcriptional run-on experiments to test the ability of WT1 to repress different classes of activation domains previously shown to stimulate the initiation and elongation steps of RNA polymerase II transcription (72). In these assays, it was revealed that WT1 represses all three classes of activation domains: those that stimulate initiation, elongation, and both initiation and elongation.
Transfection assays using reporter gene constructs and WT1 expression vectors also show that WT1 is able to activate and repress different target genes. WT1 has been shown to regulate the expression of several classes of genes, including growth factors, growth factor receptors, transcription factors, extracellular or secreted proteins, as well as others (reviewed in Ref. 119). Many of these WT1 downstream target genes have been validated only in transient cotransfection assays, where the effects of chromatin and other interacting proteins are not as they would be in a more native experimental setting. Also, interpretation of the transcriptional regulatory properties of WT1 from the results of transient cotransfection assays can be biased by technical aspects of how the experiments were performed. In these studies, WT1 may activate or repress transcription depending on which expression vector is used to synthesize WT1, the architecture of the promoter regulating the expression of the reporter gene, and the cells lines used in the experiments (reviewed in Ref. 85). Several studies have examined endogenous, differential gene expression in cell lines inducible for WT1 expression or in cell lines constitutively expressing ectopic WT1 compared with cell lines lacking WT1. The use of microarray analysis, Northern blotting, Western blotting, and suppression subtractive hybridization PCR (44) has identified several genes whose endogenous expression is altered upon expression of WT1, including the epidermal growth factor receptor gene (EGFR) (32), the epidermal growth factor family member amphiregulin (70), the orphan nuclear receptor gene Dax-1 (63), podocalyxin, a major structural membrane protein of glomerular podocytes (99), the gene encoding the retinoblastoma (Rb) associated protein RbAp46 (44), and the receptor tyrosine kinase regulator Sprouty1 (43). The expression of the above-mentioned genes was also shown to correlate with WT1 expression in the kidney and/or gonad.
|
Posttranscriptional Regulation by WT1
|
|---|
It is speculated that WT1 is also involved in posttranscriptional regulatory pathways due to its ability to bind RNA and to its association with members of the splicing machinery. WT1 has been shown to bind to the murine IGF-II mRNA transcript via its zinc finger region, where WT1(+KTS) binds with greater affinity than WT1(-KTS) (17). A biological role for WT1 in the posttranscriptional regulation of IGF-II, however, has not been demonstrated. Using an RNA selection method to identify WT1 binding ligands from a random RNA pool, three groups of RNA ligands specifically recognized by WT1 were identified (7). This study showed that WT1 zinc fingers II to IV are necessary and sufficient for RNA binding.
Several pieces of data suggest a role for WT1 in splicing. WT1 has a distinct subnuclear expression pattern, where WT1(+KTS) isoforms preferentially colocalize with molecules implicated in mRNA splicing to characteristic nuclear "speckles," whereas WT1(-KTS) appears to be expressed diffusely throughout the nucleoplasm (69). Larsson et al. (69) also showed that WT1 coimmunoprecipitates with proteins of the splicing apparatus. Furthermore, WT1 has been shown to copurify with nuclear poly(A)+ ribonucleoprotein (68), to directly associate with the constitutive splicing protein U2AF65, and to become incorporated into spliceosomes in an in vitro splicing assay (22). These data suggest that WT1 is involved in posttranscriptional processing of RNA, but a functional involvement of WT1 in splicing has never been demonstrated. Indeed, in vitro experiments aimed at demonstrating a direct role for WT1 in splicing using HeLa cell nuclear extracts have failed to do so (J. Pelletier, unpublished data).
|
WT1 Protein Partners
|
|---|
In addition to the documented association with splicing factors, WT1 has been shown to interact with a number of additional proteins, many of which are also transcription factors and modulate WT1 activity. Table 2 provides an overview of the proteins identified by various protein-protein interaction assays as being interacting partners of WT1. The characterization of WT1 protein partners may lead to the identification of important pathways involved in cellular proliferation and differentiation. The following is a brief description of the implications of WT1 interactions with Par-4 and BMZF2. Discussion of WT1’s interaction with p53 and Pax2 is presented in a later section of this review.
Par-4 (for "prostate apoptosis response-4"), a protein involved in the pro-apoptotic response of prostate cancer cells, binds to the zinc finger region of WT1 (59). Cotransfection of WT1 and Par-4 resulted in suppression of the growth-inhibitory activity of WT1, correlated with inhibition of WT1 transactivation and stimulated WT1-mediated repression. Other studies have shown that Par-4 interacts with the 17-amino acid alternatively spliced exon of WT1 (114). When fused to the GAL4 DNA-binding domain, this 17-amino acid region acts as an autonomous transcription activation domain whose activity is dependent on association with Par-4.
Bone marrow zinc finger 2 (BMZF2) is a potential transcription factor with 18 zinc fingers, where zinc fingers VI—X are involved in protein-protein interactions with the zinc finger region of WT1 (71). Cotransfection of WT1 and BMZF2 results in inhibition of transcriptional activation by WT1. BMZF2 protein is nuclear and has been shown to be expressed predominantly in fetal tissues (71). Furthermore, fusion of BMZF2 to the GAL4-DNA binding domain revealed the presence of an active repressor domain within BMZF2. BMZF2 may potentially interfere with the transcriptional regulatory functions of WT1.
|
Mouse Models Involving the WT1 Gene
|
|---|
Biochemical analyses of the WT1 gene product have been useful in studying DNA binding, protein-protein interactions, and the transcriptional activity of the different WT1 isoforms. From these data, much insight was gained into how the different WT1 isoforms function in the cell. Organogenesis, however, is a complex process involving the interplay of many different factors and cross talk between several tissues and cell lineages. Given that these processes can only be studied in the context of developing animals, gene disruption in the mouse is an invaluable tool with which to study the physiological properties of WT1. The generation of mouse models harboring mutant WT1 alleles has supported the idea that WT1 is a key regulator of urogenital system development. These mouse models have allowed for a more thorough investigation of WT1 expression patterns, elucidation of details concerning the regulation of WT1 gene expression, and insights into the role played by WT1 during urogenital system development. A summary describing these mouse models is presented in Supplementary Table 3 (available at the Physiological Genomics web site).1
|
The SeyDey Mouse
|
|---|
The Dickie small-eye (SeyDey) allele in mice is an interstitial deletion on chromosome 2 encompassing the WT1 and Pax6 genes (40). This mutation is comparable to the gross cytogenetic deletions identified at human chromosome 11p13 in children with WAGR syndrome. Individuals with WAGR and SeyDey/+ mice have similarly malformed eyes (40). Renal neoplasms, however, have not been detected in SeyDey/+ mice, and their kidneys appear histologically normal. Furthermore, the developmental anomalies observed in the gonads, genitalia, and mesonephric structures of WAGR patients are not present in SeyDey/+ mice. This suggests that the fundamental events that regulate urogenital system development in mice may be less sensitive to WT1 gene dosage. Although not fully understood, the failure of SeyDey/+ mice to develop WTs may relate to the fact that mouse kidneys represent a smaller or more short-lived target cell population than human kidneys. Mouse kidneys are 100-fold smaller than human kidneys and development occurs over a period of 31 wk in humans compared with 21 days in mice. Therefore, this decreases the opportunity for the occurrence of postzygotic mutations during differentiation of the metanephric blastema. Additionally, there are (at least) two tumor suppressor genes configured on the same chromosome (11p13 and 11p15) in humans, whereas in mice, these two loci are on separate chromosomes (chromosome 2 and 7). Thus, in humans, reduction to hemizygosity for these two loci can be achieved by a single genetic lesion, whereas two independent events would be required in mice.
|
The WT1 Null Mouse
|
|---|
To improve our understanding of the role for WT1 in urogenital development, and to elucidate putative functions in other developmental processes, mice were generated carrying a targeted mutation of WT1 (67). Initially, C57BL/6 heterozygous mutant mice analyzed up to 10 mo of age were reported to appear normal and did not develop tumors (67). In a more recent study of the same WT1 heterozygotes, but bred on a mixed genetic background, analysis of the mutant mice up to 400 days after birth revealed that 40% of the mice died within this period, where the majority of them developed severe diffuse glomerulosclerosis (Supplementary Table 3) (86). Furthermore, glomerulosclerosis was accompanied by severe albuminuria and the loss of nephrin, CD2AP, 
-actinin-4, and WT1 expression (86).
Homozygous mutant mice did not complete gestation, and induction of the metanephros did not occur (Supplementary Table 3). At E10 and E11, there were two- to threefold fewer mesonephric tubules present (but structurally normal in appearance). In mouse embryos, two distinct sets of mesonephric tubules are observed throughout mesonephric development (115). Four to six pairs of cranial mesonephric tubules develop as outgrowths of the Wolffian (or nephric) duct, whereas the remaining majority of tubules are caudal tubules which differentiate similarly to metanephric nephrons. The murine mesonephric tubules degenerate by apoptosis, except in males where the cranial tubules become the epididymal ducts. It was shown that in WT1 null embryos, the reduction in number of mesonephric tubules is due to the fact that these embryos only develop the cranial mesonephric tubules but not the caudal ones (115). Therefore, WT1 appears to regulate only development of the caudal mesonephric tubules.
Explant cultures of the mutant metanephric mesenchyme with embryonic spinal cord, a very strong inducer of mesenchymal to epithelial transformation, did not result in induction of the mesenchyme, suggesting that apoptosis of the metanephric mesenchyme was not a consequence of the absence of an inducing ureteric bud. This indicates that WT1 is required for the survival and early differentiation of the metanephric mesenchyme and that mesenchymal cell death is the result of an autonomous cell defect. Work by Donovan et al. (27) in 1999, however, has shown that the prospective metanephric mesenchyme that appears in WT1-/- embryos has indeed begun to differentiate toward the nephrogenic lineage, regardless of the absence of WT1. They examined the mutant mesenchyme for the expression of genetic markers of early mesenchymal differentiation and showed that Pax2, Six2, and GDNF ("glial cell line derived neurotrophic factor") mRNAs were present in the metanephric mesenchyme of WT1-/- embryos to levels comparable to wild-type metanephric mesenchyme of the same embryonic day. They also compared the histological appearance of WT1-/- mesenchyme with the surrounding mesenchyme of the urogenital ridge and demonstrated that the WT1-/- mesenchyme had a distinct histological appearance. From this study, it would appear that the initial differentiation of the metanephric mesenchyme does not require the presence of the ureteric bud or the intrinsic expression of WT1. Although Pax2 mRNA is present in the WT1-/- mesenchyme, no Pax2 protein can be detected in the mutant mesenchyme (67), indicating that WT1 may be required posttranscriptionally for the expression of Pax2.
WT1+/- mice have been analyzed on a 129/Sv inbred genetic background (66). Similar to the C57BL/6 WT1+/- mice (67), the 129/Sv WT1+/- mice had no apparent kidney phenotype. Although WT1+/-129/Sv males were fertile, WT1+/-129/Sv females produced no implanted embryos (Supplementary Table 3). Recall that WT1 is expressed in the gonad during development, as well as in the adult ovaries (epithelial and granulosa cells), oviduct, and uterus (105). Approximately 50% of WT1+/- females generated from a backcross of WT1+/-129/Sv:C57BL/6 F1 hybrids to 129/Sv were fertile, indicating the presence of a WT1 modifier gene that affects survival of the preimplantation embryo. There was no detectable change in the levels of WT1 protein or mRNA splice isoforms in the oviducts, ovary, and uterus of WT1+/- females, suggesting that this modifier gene may act downstream of WT1.
Another group also demonstrated that the penetrance of the WT1-/- phenotype depends on the existence of one or more modifier gene(s) (51). The WT1 null mutation was bred into the strains MF1, BALB/c, and C3H. Genotypic analysis of the offspring from intercrosses of heterozygous F1 animals revealed that WT1-/- animals were born in each case (Supplementary Table 3). Focusing on the MF1 background, the authors also observed a dramatic failure of spleen development, in addition to the well-characterized phenotypic abnormalities of the C57BL/6 WT1-/- embryo (67). Enhanced apoptosis was detected in the primordial spleen cells of the MF1 WT1-/- embryos. This mouse model will therefore be useful in studying involvement of WT1 in spleen development.
|
A DDS Mouse Model
|
|---|
A DDS mutation (named WT1tmT396) has been introduced into embryonic stem (ES) cells (102). Targeted ES cells produced wild-type and mutant transcripts of the WT1 gene in equal amounts, but analysis of the protein products revealed that the mutant protein accounted for only 5% of the total WT1 protein produced. The low levels of WT1tmT396 mutant protein produced were, however, sufficient to induce urogenital abnormalities in mice when these ES cells were used to generate transgenics (Supplementary Table 3). Low levels of DDS WT1 protein would be able to exert a strong dominant-negative effect if protein complexes containing wild-type WT1 and mutant WT1 were more stable than, or formed preferentially over, complexes containing only wild-type WT1. A single case of WT was observed where the transcript of the nontargeted allele harbored a mutation caused by an exon 9 skipping event, thus providing the sole example in mice associating WT development with WT1 dysfunction. This mouse model also demonstrates that some of the phenotypes associated with DDS can be caused by a mutant WT1 allele. In situ marker analysis of chimeric kidneys showed that mutant ES cells colonized the kidneys as efficiently as did wild-type cells (101). Furthermore, analysis of sclerotic glomeruli showed that all cells of the glomerular lineage, including mesangial cells, podocytes, and parietal epithelial cells of Bowman’s capsule, were colonized by mutant ES cells in chimeric kidneys (101). Therefore, reduced expression of the DDS WT1 allele did not affect specification of the glomerular lineage early in nephrogenesis. Immunohistochemistry was used to analyze the levels of WT1 and Pax2 in podocytes of sclerotic and nonsclerotic chimeric glomeruli (101). The authors conclude that glomerular sclerosis is not due to loss of WT1 expression by the podocytes and that podocyte Pax2 expression is a consequence of glomerular sclerosis and reflects reexpression rather than persistent expression due to the loss of repression by WT1. Although these conclusions may be correct, interpretation of the immunohistochemistry data may be ambiguous for several reasons. First, the degree of glomerular sclerosis may affect the presence of proteins, such that loss of protein expression in highly sclerotic glomeruli may be due to tissue damage and therefore may not directly correlate with protein levels in a viable cell. Also, comparing nonsclerotic with sclerotic glomeruli within the s
TOP
ABSTRACT
INTRODUCTION
Y; 2) 366 R