Loss of chromosome 11p alleles in cultured cells derived from Wilms' tumours

Hallmark discoveries in the biology of Wilms tumour

The modern study of Wilms tumour was prompted nearly 50 years ago, when Alfred Knudson proposed the 'two-hit' model of tumour development. Since then, the efforts of researchers worldwide have substantially expanded our knowledge of Wilms tumour biology, including major advances in genetics - from cloning the first Wilms tumour gene to high-throughput studies that have revealed the genetic landscape of this tumour. These discoveries improve understanding of the embryonal origin of Wilms tumour, familial occurrences and associated syndromic conditions. Many efforts have been made to find and clinically apply prognostic biomarkers to Wilms tumour, for which outcomes are generally favourable, but treatment of some affected individuals remains challenging. Challenges are also posed by the intratumoural heterogeneity of biomarkers. Furthermore, preclinical models of Wilms tumour, from cell lines to organoid cultures, have evolved. Despite these many achievements, much still remains to be discovered: further molecular understanding of relapse in Wilms tumour and of the multiple origins of bilateral Wilms tumour are two examples of areas under active investigation. International collaboration, especially when large tumour series are required to obtain robust data, will help to answer some of the remaining unresolved questions.

Orthotopic Wilms tumor xenografts derived from cell lines reflect limited aspects of tumor morphology and clinical characteristics

BACKGROUND

Wilms tumor (WT) is a pediatric tumor of the kidney, the treatment of which includes heavy chemotherapy. Affected children would likely benefit from more targeted therapies with limited side effects. Establishment of relevant orthotopic WT xenografts is important to better understand mechanisms of WT growth and for preclinical drug testing.

PROCEDURE

Here we established and characterized orthotopic xenografts from WT cell lines WiT49, CCG-99-11, and WT-CLS1 to ascertain in what aspects each of them recapitulated WT histology, immunophenotype, invasion, and metastatic spread.

RESULTS

WiT49 xenografts recapitulated near triphasic WTs with clear WT1 staining and anaplastic features, but with tumor restricted to the kidney. On the contrary both CCG-99-11 and WT-CLS1 xenografts conveyed metastatic disease. CCG-99-11 showed a blastemal phenotype whereas WT-CLS1 xenografts did not properly reflect any specific WT subtype.

CONCLUSIONS

From the three tested cell lines, orthotopic WiT49 xenografts best reflect the triphasic pattern of classical WT.

Characterization of 17.94, a novel anaplastic Wilms' tumor cell line

Despite considerable advances in understanding the molecular pathogenesis of Wilms' tumor (WT), its cell biology is less well understood, partly due to the paucity of established WT cell lines. We report here the establishment of a new anaplastic WT cell line, 17.94, which expressed NCAM, SALL1, and CITED1-phenotypic features expected of metanephric blastema-derived cells. Treatment of 17.94 cells with 12-O-Tetradecanoylphorbol 13-acetate caused morphological changes, which led to reduced NCAM and SALL1 expression, but expression of vimentin was maintained, indicating a potential for stromal differentiation. The 17.94 cell line contained a TP53 mutation, consistent with the anaplastic histology of the original tumor, but lacked mutations in WT1, WTX, or CTNNB1, which are the other genes involved in WT pathogenesis. The 17.94 cells showed no loss of heterozygosity at 7p, 11p, or 16q; however, DNA hypermethylation was detected at several loci, including the H19 differentially methylated region (indicative of loss of imprinting of IGF2 at 11p15) and at the PCDH@ gene clusters at 5q31. The derivation of the 17.94 cell line should help to further dissect the genetic-epigenetic interactions involved in the pathogenesis of WT.

Characterization of primary Wilms tumor cultures as an in vitro model

Functional analysis of gene candidates and testing of novel therapeutics in Wilms tumors (WT) has been hampered by the lack of in vitro model systems. WT are characterized by a spectrum of histological appearances, but published cell lines are mostly derived from rare anaplastic variants or even non-WT. There has been some success in establishing primary cultures, but these are often poorly characterized or only derived from less frequent WT1 mutant tumors. We report the generation of a set of primary WT-cell cultures using a simple cultivation protocol. Our cultures could be established after preoperative chemotherapy and irrespective of histological subtypes or genetic alterations. The presence of tumor-specific genetic alterations validates these cultures as being tumor-derived. Genetic characterization is of utmost importance as some cultures with similar morphological appearance lacked such alterations and either represent clonal variants or normal cells. By immunohistochemistry, the cells are either epithelial or more mesenchymal, and the latter exhibiting a longer life span with 30 or more passages before undergoing senescence. This may be related to WT being embryonal tumors with a strong differentiation potential that may prevail in vitro. Telomeres progressively shorten with cultivation, but their length does not predict lifespan. hTERT transfection may partly allow establishment of immortalized lines, because 2/7 cultures avoid senescence even in later passages. Importantly, these cells can be efficiently manipulated by transfection, making them a useful model system for in vitro testing.

Frequency and timing of loss of imprinting at 11p13 and 11p15 in Wilms' tumor development

Epigenetic changes occur frequently in Wilms' tumor (WT), especially loss of imprinting (LOI) of IGF2/H19 at 11p15. Our previous results have identified imprinted transcripts (WT1-AS and AWT1) from the WT1 locus at 11p13 and showed LOI of these in some WTs. In this article, we set out to test the relationship between LOI at 11p13 and 11p15 and their timing in WT progression relative to other genetic changes. We found a higher level (83%) of 11p13 LOI in WT than of 11p15 LOI (71%). There was no correlation between methylation levels at the 11p13 and 11p15 differentially methylated regions or between allelic expression of WT1-AS/AWT1 and IGF2. Interestingly, retention of normal imprinting at 11p13 was associated with a small group of relatively late-onset, high-stage WTs. An examination of genetic and epigenetic alterations in nephrogenic rests, which are premalignant WT precursors, showed that LOI at both 11p13 and 11p15 occurred before either 16q loss of heterozygosity (LOH) or 7p LOH. This suggests that these LOH events are very unlikely to be a cause of LOI but that LOH may act by potentiating the effects of overexpression of IGF2 and/or WT1-AS/AWT1 that result from LOI.

The parathyroid hormone-responsive B1 gene is interrupted by a t(1;7)(q42;p15) breakpoint associated with Wilms' tumour

Wilms' tumour (WT) has a diverse and complex molecular aetiology, with several different loci identified by cytogenetic and molecular analyses. One such locus is on chromosome 7p, where cytogenetic abnormalities and loss of heterozygosity (LOH) indicate the presence of a Wilms' tumour suppressor gene. In order to isolate a candidate gene for this locus, we have characterized the breakpoint regions at a novel constitutional chromosome translocation (t(1;7)(q42;p15)), found in a child with WT and skeletal abnormalities. We identified two genes that were interrupted by the translocation: the parathyroid hormone-responsive B1 gene (PTH-B1) at 7p and obscurin at 1q. With no evidence for LOH at 1q42, we focused on the characterization of PTH-B1. We detected novel alternately spliced isoforms of PTH-B1, which were expressed in a wide range of adult and foetal tissues. Importantly, expression of two isoforms were disrupted in the WT of the t(1;7) patient. We also identified an additional splice isoform expressed only in 7p LOH tumours. The disruption of PTH-B1 by the t(1;7), together with aberrant splicing in sporadic WTs, suggests that PTH-B1 is a candidate for the 7p Wilms' tumour suppressor gene.

Low frequency of genetic lesions in Wilms tumors by representational difference analysis

Genomic representational difference analysis (RDA) was carried out on a total of nine Wilms tumors and one cystic partially differentiated nephroblastoma (CPDN; a sub-type of Wilms) to look for novel genetic deletions involving tumor suppressor genes. Genomic DNA from either short-term cultured Wilms tumor cells or a WT xenograft was used to create driver representations, and genomic DNA from matched normal kidney or normal kidney cells grown in short-term culture was used to create the tester. Genuine difference products were obtained from only one of the tumors. However, none of these fragments were found to be deleted in the original tumor biopsy, microdissected tumor or in the lung metastasis from this patient. It is, therefore, likely that the deletions were due to random losses associated with the genetic instability of the cultured cells from this particular tumor. We did not isolate difference products from any of the other tumors, showing that they did not have chromosomal losses, homozygous deletions or regions of LOH that were detectable by RDA.

Bone marrow micrometastases in a patient with localized Wilms' tumor

The case of a 7-year-old boy presenting at diagnosis with a localized (stage III) Wilms' tumor of favorable histology is presented. Immunocytologic analysis of bone marrow aspirates revealed cells positive for neural cell adhesion molecule (NCAM) and negative for class I major histocompatibility complex (MHC) antigens. These cells were interpreted as deriving from the tumor blastemal component. Postoperatively the child underwent radiotherapy and chemotherapy, and he remains free of disease 12 months after completion of therapy. In patients with nonmetastatic Wilms' tumor at onset, the evaluation of the actual frequency of occult marrow involvement and the assessment of its clinical significance may necessitate further investigation.

Germline and somatic abnormalities of chromosome 7 in Wilms' tumor

Although a gene (WT1) located at chromosome 11p13 is implicated in the development of Wilms' tumor (WT), there is evidence that genes on other chromosomes are also involved. A WT patient presented with a constitutional balanced translocation between chromosomes 1 and 7, t(1;7)(q42;p15), the breakpoints of which could represent a WT predisposition gene in this patient. Cytogenetic analysis of the tumor from this patient revealed an acquired abnormality of the other chromosome 7, resulting in an isochromosome of the long arm and a 46,XY,t(1;7)(q42;p15)c,i(7)(q10) karyotype. The regions of the translocation breakpoints were investigated in a series of 24 WTs using Southern blot analysis. This confirmed the monosomy of 7p and trisomy of 7q in the tumor of the translocation patient, and in addition a loss of chromosome 7p alleles was identified in a WT of a bilaterally affected patient. In addition, two WTs were shown to have an extra copy of chromosome 7 alleles. Multiple copies of chromosome 1q alleles, probably resulting from secondary changes, were observed in two WTs, one of which was also associated with a trisomy of chromosome 7. These results indicate that 7p may contain a tumor suppressor gene involved in WT development, and that duplications of 7q also may play a role in WT development.

Low frequency of mutations in the WT1 coding region in Wilms' tumor

A series of twenty unselected Wilms' tumors were analysed for alterations in the WT1 tumor suppressor gene. The entire coding region of WT1 was amplified by RNA-PCR, and then screened for mutations by single-strand conformational polymorphism analysis (SSCP). This method was shown to be capable of detecting point mutations in the WT1 gene, by using an experimentally produced mutation. A single mutation, a 226 bp intragenic deletion, was detected in a tumor from a patient with the WAGR syndrome. These results suggest that alterations in the WT1 gene may be involved in only a subset of Wilms' tumors, and that other loci need to be investigated as potential suppressor genes in sporadic Wilms' tumors.

Molecular and cellular heterogeneity of Wilms' tumor

We developed a Wilms' tumor-cell culture system to investigate the molecular basis of nephrogenesis and oncogenesis. Several distinct fractions of cells were isolated and characterized from the same tumor specimen. The cells exhibited striking differences in morphology, immunocytochemical staining profiles and cytogenetics. One fraction contained cells with features of epithelium; other cell fractions resembled partially differentiated mesenchyme (blastema or stroma). While the Wilms' tumor-suppressor gene WT1 was not altered, loss of heterozygosity (LOH) and an insertion in intron I of the p53 tumor-suppressor gene occurred in the tumor and the cultured cell types. LOH for RB was detected only in the cultured cells. These findings are consistent with a model of tumor initiation in a pluripotent cell that is able to undergo subsequent differentiation along multiple different lines and which mimics normal nephrogenesis.

Paternal origin of 11p15 duplications in the Beckwith-Wiedemann syndrome. A new case and review of the literature

A boy suffering from the Beckwith-Wiedemann syndrome (BWS) was found to have partial trisomy of the short arm of chromosome 11 [46,XY,der(5)t(5;11)(p15.2;p14)]. Both his parents were phenotypically normal, but his father carried a balanced translocation between chromosomes 5 and 11 [46,XY,t(5;11)(p15.2;p14)]. DNA analysis of polymorphic markers on 11p15 confirmed the paternal origin of the duplicated material in the child. This case is the sixth report of paternal duplication of 11p15 in BWS. These results are discussed in relation to the possible role of genomic imprinting in BWS and in Wilms' tumor.

Suppression of tumorigenicity in Wilms tumor by the p15.5-p14 region of chromosome 11

Wilms tumor has been associated with genomic alterations at both the 11p13 and 11p15 regions. To differentiate between the involvement of these two loci, a chromosome 11 was constructed that had one or the other region deleted, and this chromosome was introduced into the tumorigenic Wilms tumor cell line G401. When assayed for tumor-forming activity in nude mice, the 11p13-deleted, but not the 11p15.5-p14.1-deleted chromosome, retained its ability to suppress tumor formation. These results provide in vivo functional evidence for the existence of a second genetic locus (WT2) involved in suppressing the tumorigenic phenotype of Wilms tumor.