Genetic mosaicism in normal tissues of Wilms' tumour patients.

We describe the partial loss of heterozygosity (LOH) at chromosome 11p loci in normal tissues (normal kidney and/or blood) from four of 67 Wilms' tumour patients. Autologous tumour DNA showed complete loss of the same, maternally derived, alleles. These observations indicate that the normal tissues were mosaic for cells heterozygous and homozygous for 11p markers and that tumours subsequently developed from the homozygous cells that had undergone an 11p somatic recombination event. We suggest that LOH for 11p alleles is compatible with normal growth and differentiation and is significant pathologically only when accompanied by other genetic alterations.

RNA expression of the WT1 gene in Wilms' tumors in relation to histology.

BACKGROUND: On the basis of accumulating data, the recently isolated WT1 gene is a Wilms' tumor gene and a putative tumor suppressor gene. These findings include expression in developing fetal kidney, intragenic deletions in tumors, and germline mutations in predisposed individuals. Wilms' tumors, which exhibit a broad range of differentiation, are composed of three cell types: blastema, epithelium, and stroma. PURPOSE: The purpose of this study was to investigate the relationship between WT1 gene expression and histologic composition in Wilms' tumors in an effort to elucidate how the WT1 gene functions in proliferation of these histologic components. METHODS: We used Northern blot hybridization to study WT1 gene expression by messenger RNA (mRNA) accumulation in 20 tumors of varying histology and in adjacent uninvolved kidney tissue. In two patients, tumors were also compared before and after therapy. RESULTS: Tumors that were predominantly blastemal expressed high amounts of WT1 mRNA, whereas predominantly stromal tumors expressed either low or undetectable amounts. Blastemal tumors that were predominantly poorly differentiated expressed WT1 mRNA at higher levels than those that were more well differentiated. Although we expected that a putative tumor suppressor gene like WT1 would generally be expressed at lower levels in tumor than in normal kidney, this was true only in predominantly stromal cells. One of the two patients studied before and after therapy had a dramatic response to therapy accompanied by a decline in WT1 gene expression and disappearance of blastemal and epithelial elements. CONCLUSIONS: A correlation was observed between WT1 gene expression and histology of the tumors. Level of expression was inversely related to the degree of differentiation in blastemal tumors and in the patient with a dramatic response to therapy. These results, in conjunction with the observation that WT1 mRNA is abundant in normal fetal kidney, suggest that WT1 gene expression is related to kidney development, especially in differentiation of blastemal components. IMPLICATIONS: Further studies to search for alterations of the WT1 gene in tumors and to identify regulatory factors in gene expression will increase understanding of the role of this gene in normal development and tumorigenesis.

Frequent association of beta-catenin and WT1 mutations in Wilms tumors.

The etiology of Wilms tumor, an embryonic kidney tumor, is genetically heterogeneous. One Wilms tumor gene, WT1, which encodes a zinc finger transcription factor, is mutated in 10-20% of Wilms tumors, but it is still not clear what critical cellular pathway(s) is affected by these mutations. Recently beta-catenin mutations have been reported in 6 of 40 (15%) of Wilms tumors. Beta-catenin is the central effector in the Wnt signal transduction pathway, and deregulation of beta-catenin signaling is critical in the development of a number of malignancies. The observation of beta-catenin mutations in Wilms tumors suggests that abrogation of the Wnt signaling pathway also plays a role in some Wilms tumors. To assess the relationship of WT1 mutations vis-à-vis beta-catenin mutations in Wilms tumor, we analyzed 153 primary tumors, and 21 of 153 (14%) carried beta-catenin mutations. Surprisingly, we observed a highly significant (P = 3.6 x 10(-13)) association between WT1 and beta-catenin mutations; 19 of 20 beta-catenin-mutant tumors had also sustained WT1 mutations. By analogy to the patterns of concordant and discordant gene mutations observed in other tumors, our data suggest that mutation of WT1 and beta-catenin affects two different cellular pathways, both of which are critically altered in at least a subset of Wilms tumors.

Sequential loss of heterozygosity at microsatellite motifs in preinvasive and invasive head and neck squamous carcinoma.

Studies of sequential molecular alterations in noninvasive and invasive head and neck squamous carcinoma are few in number. Consequently, the genetic changes associated with the neoplastic transformation of these carcinomas have not been defined. To identify chromosomal alterations in preinvasive and invasive head and neck squamous carcinoma, we analyzed DNA from microdissected normal squamous epithelium, severe dysplasia, and invasive carcinoma samples from 20 patients for loss of heterozygosity (LOH) at microsatellite loci by multiplex PCR. Twenty-five microsatellite repeats on chromosomes 3p, 5q, 8p, 9p and 9q, 11q, 17p, 17q, and 18p and 18q regions were used. In informative cases, LOH in noninvasive lesions was observed in 9p (28%), 9q and 18q (10%), 11q and 17p (7%), and 3p and 18p (5%). A high incidence of LOH in invasive carcinoma was observed at 9p (72%), 8p (53%), 3p (47%), 9q (35%), and 11q (33%). LOH was also associated with DNA aneuploidy, high tumor stage, and poor histological differentiation. Our results indicate that: (a) the high incidence of LOH at loci on chromosomes 9p, 8p, 3p, 9q, and 11q implicate these regions in head and neck squamous carcinoma tumorigenesis; (b) 9p loci alterations are manifested in the early development of these tumors; (c) LOH is correlated with poor prognostic clinicopathological factors; and (d) LOH at 8p loci appears to be associated with the tumor's aggressive features.

Loss of heterozygosity for chromosomes 16q and 1p in Wilms' tumors predicts an adverse outcome.

We have prospectively analyzed Wilms' tumors from 232 patients registered on the National Wilms' Tumor Study for loss of heterozygosity (LOH) on chromosomes 11p, 16q, and 1p. These chromosomal aberrations were found in 70 (33%), 35 (17%), and 21 (12%) of the informative cases, respectively. LOH for two of these regions occurred in only 25 cases, and only one tumor harbored LOH at all three sites. There was no statistically significant association between LOH at any of the three regions and either the stage or histological classification of the tumor. Patients with tumor-specific LOH for chromosome 16q had relapse rates 3.3 times higher (P = 0.01) and mortality rates 12 times higher (P < 0.01) than patients without LOH for chromosome 16q. These differences remained when adjusted for histology or for stage. Patients with LOH for chromosome 1p had relapse and mortality rates three times higher than those for patients without LOH for chromosome 1p, but these results were not statistically significant. In contrast, LOH for chromosome 11p had no effect on measures of outcome. These molecular markers may serve to further stratify Wilms' tumor patients into biologically favorable and unfavorable subgroups, allowing continued use of the clinical trial mechanism in the study of Wilms' tumor.

Nonlinkage of 16q markers to familial predisposition to Wilms' tumor.

Wilms' tumor (WT), a childhood cancer of the kidney, occurs in both familial and sporadic forms. Chromosome 11 genes have been implicated in the etiology of WT, and mutations in a gene at chromosomal band 11p13, WT1, have been identified in a few WT cases. However, 11p13 has been excluded as the site of the predisposition mutation segregating in several large WT families, which implies the existence of a non-11p familial predisposition gene. Recently, loss of heterozygosity for 16q markers located between chromosomal bands 16q13 and 16q22 has been reported in approximately 20% of sporadic Wilms' tumors. To determine if this region of 16q harbors the non-11p familial WT gene, a genetic linkage study of five WT families was undertaken. Using multipoint analyses, we ruled out genetic linkage of familial WT predisposition to 16q.

Evidence for genetic heterogeneity in familial Wilms' tumor.

Wilms' tumor (WT), a childhood kidney cancer, occurs both sporadically and, less frequently, in a familial context. Genetic linkage studies of several large WT families have excluded the one cloned WT gene, WT1, as the locus responsible for familial predisposition. These data demonstrate the existence of a familial predisposition gene distinct from WT1 and, more broadly, imply that the genetic etiology of WT is heterogenous. However, it has been unknown whether the predisposition observed in large WT families is also heterogenous or perhaps is due to mutations at a single locus. Recently, examination of a large French-Canadian WT family has demonstrated genetic linkage to 17q12-q21. We report here the results from a genetic linkage study of six WT pedigrees. Analyses of genotype data from eight loci within the 17q12-q21 region in these families resulted in cumulative lod scores of <-4.0 through the region, thereby excluding linkage. The ability to rule out the 17q region as the site of a predisposition gene in several of these pedigrees individually demonstrates the existence of more than one gene that predisposes to WT in large pedigrees and again emphasizes that the etiology of WT is genetically heterogenous.

Linkage of familial Wilms' tumor predisposition to chromosome 19 and a two-locus model for the etiology of familial tumors.

Familial predisposition to Wilms' tumor (WT), a childhood kidney tumor, is inherited as an autosomal dominant trait. For most WT families studied, the 11p13 gene WT1 and genomic regions implicated in tumorigenesis in a subset of tumors can be ruled out as the site of the familial predisposition gene. Following a genome-wide genetic linkage scan, we have obtained strong evidence (log of the odds ratio = 4.0) in five families for an inherited WT predisposition gene (FWT2) at 19q13.3-q13.4. In addition, we observed loss of heterozygosity at 19q in tumors from individuals from two families in which 19q can be ruled out as the site of the inherited predisposing mutation. From these data, we hypothesize that alterations at two distinct loci are critical rate-limiting steps in the etiology of familial WTs.

Absence of PPP2R1A mutations in Wilms tumor.

Evidence from genetic linkage analysis indicates that a gene located at 19q13.4, FWT2, is responsible for predisposition to Wilms tumor in many Wilms tumor families. This region has also been implicated in the etiology of sporadic Wilms tumor through loss of heterozygosity analyses. The PPP2R1A gene, encoding the alpha isoform of the heterotrimeric serine/threonine protein phosphatase 2A (PP2A), is located within the FWT2 candidate region and is altered in breast and lung carcinomas. PPP2R1B, encoding the beta isoform, is mutated in lung, colon, and breast cancers. These findings suggested that both PPP2R1A and PPP2R1B may be tumor suppressor genes. Additionally, PP2A is important in fetal kidney growth and differentiation and has an expression pattern similar to that of the Wilms tumor suppressor gene WT1. Since PPP2R1A was therefore a compelling candidate for the FWT2 gene, we analysed the coding region of PPP2R1A in DNA and RNA samples from affected members of four Wilms tumor families and 30 sporadic tumors and identified no mutations in PPP2R1A in any of these 34 samples. We conclude that PPP2R1A is not the 19q familial Wilms tumor gene and that mutation of PPP2R1A is not a common event in the etiology of sporadic Wilms tumor.

Analysis of possible WT1 RNA processing in primary Wilms tumors.

WT1 RNA processing abnormalities have been suggested to play a role in the development of Wilms tumor by reports of editing at codon 280 in the rat WT1 transcript (codon 281 in humans) and aberrant splicing of exon 2 in WT1 transcripts from Wilms tumor xenograft cell lines. Both events result in a functionally changed WT1 protein and are potential mechanisms of altering normal protein function in the absence of WT1 DNA mutations. To determine whether either of these RNA processing events occurs in primary Wilms tumors, we analysed WT1 mRNA from 15 primary tumors. There was no evidence of WT1 RNA editing at codon 281, and only one primary tumor displayed aberrant splicing of exon 2. Sequence and Southern analysis of DNA from this tumor did not reveal any alteration in or around exon 2. These results suggest that neither RNA editing at codon 281 nor aberrant exon 2 splicing is a frequent mechanism of WT1 alteration during tumorigenesis.

Structural analysis of the human nov proto-oncogene and expression in Wilms tumor.

We have cloned and sequenced the nov gene (novH) which is the homolog of the chicken nov proto-oncogene overexpressed in avian nephroblastomas. The novH gene is highly conserved and encodes a putative IGF-binding protein similar to that of chicken. We report that relative to autologous normal kidney expression of novH is elevated in Wilms tumors containing predominantly stromal elements and is inversely correlated in these tumors to the expression of WT1. Our results suggest that the regulation of IGFII expression by WT1 and increase of novH in Wilms tumors might be interrelated and represent a key element in tumor development in human.

Frequent loss of imprinting at the IGF2 and H19 genes in head and neck squamous carcinoma.

Genomic imprinting is an inherited epigenetic phenomenon that results in parental-origin-specific gene expression in somatic cells. Relaxation or loss of this feature in certain genes has been demonstrated in several pediatric and adult neoplasms, suggesting an association with tumorigenesis. We analysed 64 primary untreated head and neck squamous carcinoma for the loss of imprinting in the IGF2 and H19 genes to determine the implications of this alteration in the development and progression of these tumors. Forty-nine (77%) of the 64 tumors were informative for imprinting analyses of these genes. IGF2 and H19 were imprinted in all normal squamous epithelium examined. Twelve (37.5%) of 32 tumors informative for H19 and 11 (40.7%) of 27 tumors informative for IGF2 manifested loss of imprinting. Ten tumors were informative for both genes, of which four maintained the constitutional imprinting and six showed loss of imprinting at either H19 or IGF2. These data suggest that loss of imprinting at the IGF2 and H19 loci play a role in the oncogenesis of head and neck carcinoma.

Wilms tumor genes.

Multiple 'WT' genes exist. The WT1 gene at chromosomal band 11p13 has been cloned and is known to be important in the etiology of at least some tumors by virtue of the identification of both germline and somatic mutations in WT patients. Genes at 11p15 and 16q are also involved, either as initiating or tumor progression events. An unlocalized familial predisposition gene is also known to be important etiologically. The identification of several genes that are involved in the etiology or progression of WT, the preferential loss of maternally derived alleles in tumor tissue, and the observed reduction to 11p homozygosity in normal tissue DNA from some patients, all strikingly indicate that a simple, one-locus-'two-hit' genetic model for WT is inadequate. The question is not if this model needs to be modified, but how it should be modified, or if it is even valid enough to be a starting point for understanding the genetics of Wilms tumor. To begin to address this, several questions can be asked. Do all Wilms tumors carry mutations at the WT1 locus? Do both alleles at the WT1 locus need to be inactivated or lost for tumorigenesis? Or, instead, do some WT1 mutations act dominantly? Do patients with bilateral disease carry germline mutations as originally hypothesized, or, as more recently suggested, is bilateral disease the result of early somatic mutations, genomic imprinting, or multifactorial inheritance? Must mutations at an 11p15 locus and/or 11p15 LOH accompany WT1 mutations, or do 11p13 and 11p15 mutations act independently of each other? Have tumors from familial WT cases (who do not carry germline WT1 mutations) sustained somatic mutations at the WT1 locus, the 11p15 locus or the 16q locus? Conversely, do tumors from sporadic WT patients carry somatic mutations at the non-11p familial predisposition gene? Will most tumors be found to carry mutations at the same one or two loci, but differ only with regard to whether the mutations are somatic or germline? Are effects of genomic imprinting layered over, so to speak, a framework of classically mendelian mutations, or in some cases is imprinting the mechanism by which genes are inactivated or their normal function modulated? Although not definitive, there are data that bear on some of these questions. Germline mutations have been observed in patients with bilateral tumors, but may not prove to be a universal feature of bilateral disease.(ABSTRACT TRUNCATED AT 400 WORDS)

Treatment with nephrectomy only for small, stage I/favorable histology Wilms' tumor: a report from the National Wilms' Tumor Study Group.

PURPOSE: Children younger than 24 months with small (< 550 g), favorable histology (FH) Wilms tumors (WTs) were shown in a pilot study to have an excellent prognosis when treated with nephrectomy only. PATIENTS AND METHODS: A study of nephrectomy only for the treatment of selected children with FH WT was undertaken. Stringent stopping rules were designed to insure closure of the study if the true 2-year relapse-free survival rate was 90% or lower. RESULTS: Seventy-five previously untreated children younger than 24 months with stage I/FH WTs for which the surgical specimen weighed less than 550 g were treated with nephrectomy only. Three patients developed metachronous, contralateral WT 1.1, 1.4, and 2.3 years after nephrectomy, and eight patients relapsed 0.3 to 1.05 years after diagnosis (median, 0.4 years; mean, 0.51 years). The sites of relapse were lung (n = 5) and operative bed (n = 3). The 2-year disease-free (relapse and metachronous contralateral WT) survival rate was 86.5%. The 2-year survival rate is 100% with a median follow-up of 2.84 years. The 2-year disease-free survival rate (excluding metachronous contralateral WT) was 89.2%, and the 2-year cumulative risk of metachronous contralateral WT was 3.1%. CONCLUSION: Children younger than 24 months treated with nephrectomy only for a stage I/FH WT that weighed less than 550 g had a risk of relapse, including the development of metachronous contralateral WT, of 13.5% 2 years after diagnosis. All patients who experienced relapse on this trial are alive at this time. This approach will be re-evaluated in a clinical trial using a less conservative stopping rule.

Allelic loss and replication errors at microsatellite loci on chromosome 11p in head and neck squamous carcinoma: association with aggressive biological features.

The frequent loss of heterozygosity (LOH) demonstrated at chromosome 11p regions in several sporadic malignancies has suggested the presence of tumor suppressor genes at these locations. To obtain detailed mapped incidence of the microsatellite alterations at these regions and to investigate the possible correlation between the genotype and the pathobiological characteristics of head and neck squamous carcinoma, we analyzed paired DNA samples from normal mucosa and primary tumor specimens from 56 patients with these tumors. Our results show that 50.9% of the tumors had microsatellite alterations at one or more of these loci. LOH was manifested in 45. 5% and instability in 5.5% of the tumors. 11p15 loci showed more frequent LOH (39.6%) than those of 11p13 (29.3%) and 11p11-12 (18. 8%); the D11S988 (11p15) marker showed the highest single locus incidence of LOH (29.7%). Eight tumors (22.2%) demonstrated simultaneous LOH at both the 11p15 and 11p13 regions. LOH was significantly associated with poor histological differentiation, DNA aneuploidy, and high proliferative activity in these neoplasms. Our study extends the involvement of the 11p13 and 11p15 regions to head and neck squamous tumorigenesis and indicates that the terminal loci of 11p may harbor a tumor suppressor gene(s) associated with the progression of these tumors.

Wilms tumor genetics.

Wilms tumor (WT), a sporadic and familial childhood kidney tumor, is genetically heterogeneous. One WT gene, WT1 at 11p13, has been cloned, but only a minority of WTs carry detectable mutations at that locus. WT1 can also be excluded as the predisposition gene in most WT families, implying the existence of other WT genes. Studies of WT families have demonstrated that familial predisposition is also heterogeneous and involves at least two other loci besides WT1. In addition to WT1 and the familial predisposition genes, a role for other genes in the development of WTs is implied by the somatic occurrence of genetic and epigenetic alterations such as loss of heterozygosity and loss of imprinting in tumors and, rarely, the observation of nonchromosome-11 constitutional aberrations in WT patients. Determining the pattern of presence or absence of these various genetic alterations in tumors and elucidating the function of the genes involved will provide a better understanding of the cellular processes that are critical for normal cell growth and differentiation, but are abrogated in the course of tumorigenesis.

SCE frequencies in rabbit lymphocytes as a function of time after an acute dose of cyclophosphamide.

Following acute and chronic exposures to various chemicals in vivo, the average SCE frequency in human and rabbit lymphocytes has generally been shown to decrease with time posttreatment. The rate of this decline varies, however, and little data have been published pertaining to the decrease in SCEs soon after exposure. To gain more information about the immediate decline in SCEs with time, we injected rabbits with a single dose of 35 mg/kg cyclophosphamide (CP) and determined SCE levels in circulating lymphocytes at various times 5 h to 2 weeks after treatment. We observed a rapid decline in SCE frequencies within 5 days, and by 10 days post-exposure the SCE levels were back to control values. The distribution of SCEs among cells and the number of circulating lymphocytes were also analyzed at each time. Within 2-3 days posttreatment we observed a rapid loss of cells with high SCE levels concomitantly with a rapid decline in circulating lymphocytes and a decrease in the average SCE frequency. When the number of lymphocytes began to increase, the number of cells with normal SCE values also increased. By 10-11 days after CP, the lymphocyte count had recovered, the SCE frequency had returned to control levels, and the distribution of SCEs among cells was almost identical to the control distribution. These data, in addition to published information on rabbit lymphocyte lifespan, suggest that the decline in SCE levels with time posttreatment is a function of lymphocyte turnover.

Evaluation of chemically induced cytogenetic lesions in rabbit oocytes. III. A postimplantation analysis of streptonigrin effects.

The frequency of consistent structural chromosome aberrations was determined for 108 fetuses (day 14 of gestation) from 20 female rabbits treated before conception with 90 micrograms/kg of streptonigrin (NSC-45383). The findings by karyotype analysis of 1.85% of the fetuses with consistent aberrations was compared with 6.32% frequency previously observed in 6-day (preimplantation) blastocysts from identically treated females and found to be significantly reduced. The interpretation of the findings is that the developmental events of implantation and placentation are effective in eliminating from further gestation the majority of the rabbit conceptuses with consistent structural chromosome abnormalities. The data are also discussed relative to the predictive nature of somatic cell chromosome damage in women of childbearing age being an estimator of risk of producing offspring with structural chromosome aberrations. The data show that preconceptional treatment of female rabbits with streptonigrin caused a shift in the cytogenetically determined sex ratio of the 14-day fetuses, but not of the 6-day blastocysts. The significant decrease in relative numbers of males observed is consistent with the induction of lethal mutations on the X chromosome. This finding is discussed in conjunction with the timing and parental specificity of X chromosome inactivation during mammalian embryonic development.