Ruthild G. Weber

Alterations of the Tumor Suppressor Genes CDKN2A (p16INK4a), p14ARF, CDKN2B (p15INK4b), and CDKN2C (p18INK4c) in Atypical and Anaplastic Meningiomas

We investigated 67 meningothelial tumors (20 benign meningiomas, 34 atypical meningiomas, and 13 anaplastic meningiomas) for losses of genetic information from chromosome arms 1p and 9p, as well as for deletion, mutation, and expression of the tumor suppressor genes CDKN2A (p16(INKa)/MTS1), p14(ARF), CDKN2B (p15(INK4b)/MTS2) (all located at 9p21) and CDKN2C (1p32). Comparative genomic hybridization and microsatellite analysis showed losses on 1p in 11 anaplastic meningiomas (85%), 23 atypical meningiomas (68%), and 5 benign meningiomas (25%). One atypical meningioma with loss of heterozygosity on 1p carried a somatic CDKN2C mutation (c.202C>T: R68X). Losses on 9p were found in five anaplastic meningiomas (38%), six atypical meningiomas (18%), and one benign meningioma (5%). Six anaplastic meningiomas (46%) and one atypical meningioma (3%) showed homozygous deletions of the CDKN2A, p14(ARF), and CDKN2B genes. Two anaplastic meningiomas carried somatic point mutations in CDKN2A (c.262G>T: E88X and c.262G>A: E88K) and p14(ARF) (c.305G>T: G102V and c.305G>A: G102E). One anaplastic meningioma, three atypical meningiomas, and one benign meningioma without a demonstrated homozygous deletion or mutation of CDKN2A, p14(ARF), or CDKN2B lacked detectable transcripts from at least one of these genes. Hypermethylation of CDKN2A, p14(ARF), and CDKN2B could be demonstrated in one of these cases. Taken together, our results indicate that CDKN2C is rarely altered in meningiomas. However, the majority of anaplastic meningiomas either show homozygous deletions of CDKN2A, p14(ARF), and CDKN2B, mutations in CDKN2A and p14(ARF), or lack of expression of one or more of these genes. Thus, inactivation of the G(1)/S-phase cell-cycle checkpoint is an important aberration in anaplastic meningiomas.

Juvenile polyposis: massive gastric polyposis is more common in MADH4 mutation carriers than in BMPR1A mutation carriers

Abstract. Juvenile polyposis syndrome (JPS) is an autosomal dominant predisposition to multiple juvenile polyps in the gastrointestinal tract. Germline mutations in the MADH4 or BMPR1A genes have been found to be causative of the disease in a subset of JPS patients. So far, no genotype-phenotype correlation has been reported. We examined 29 patients with the clinical diagnosis of JPS for germline mutations in the MADH4 or BMPR1A genes and identified MADH4 mutations in seven (24%) and BMPR1A mutations in five patients (17%). A remarkable prevalence of massive gastric polyposis was observed in patients with MADH4 mutations when compared with patients with BMPR1A mutations or without identified mutations. This is the first genotype-phenotype correlation observed in JPS.

Characterization of Genomic Alterations in Hepatoblastomas : A Role for Gains on Chromosomes 8q and 20 as Predictors of Poor Outcome

As data on the genomic alterations in hepatoblastoma (HB) are limited, 34 HB tumors and three HB cell lines were screened for DNA copy number changes by comparative genomic hybridization. The average number of chromosomal imbalances per tumor was 2.3 +/- 0.5 (mean +/- SEM) with gains sevenfold more frequent than losses. The most frequent gains of chromosomal material in HB tumors were on 2q (44%), 1q (41%), 2p (29%), 20 (24%), 22q (18%), 8q (15%), 8p and 12q (9% each), as well as 7q, 12p, and 17 (6% each) and the only recurrent loss was on 4q in 12% of cases. Highly amplified sequences were identified in four tumors and mapped to 2q24 in two cases, to 8q in two cases (once to 8q11.2-q13 and once to 8q11.2-q21.3) as well as to 10q24-q26 in one case. In one cell line, highly amplified DNA sequences were mapped to 7p and 8q. Comparison to previously published data on this series of HB revealed that the number of chromosomal imbalances was significantly higher in HB tumors with loss of heterozygosity on 11p (P = 0.03), whereas in five of 10 HB biopsies without chromosomal imbalances, beta-catenin gene mutations were found. HB patients were divided into a good (no evidence of disease) and a poor (died of disease) outcome group according to their clinical course after standard therapy. Two alterations were found to be significantly associated with poor outcome: gain on 8q (P = 0.007) and gain on 20 (P = 0.009). In summary, our analysis allowed the identification of gains on chromosomes 1q and 2 as hallmark DNA copy number changes in HB with 2q24 as a critical chromosomal band. Furthermore, this study provided evidence that gains on 8q and 20 play a role as markers of prognostic significance in HB.

Comprehensive analysis of genomic alterations in gliosarcoma and its two tissue components

Gliosarcoma is a variant of glioblastoma multiforme characterized by two components displaying gliomatous or sarcomatous differentiation. We investigated 38 gliosarcomas for aberrations of tumor‐suppressor genes and proto‐oncogenes that are commonly altered in glioblastomas. Amplification of CDK4, MDM2, EGFR, and PDGFRA were found in 11% (4/35), 8% (3/38), 8% (3/38), and 3% (1/35) of the tumors, respectively. Nine of 38 gliosarcomas (24%) carried TP53 mutations. PTEN mutations were identified in 45% (9/20) of the investigated tumors. Twenty gliosarcomas were analyzed by comparative genomic hybridization (CGH). Chromosomal imbalances commonly detected were gains on chromosomes 7 (15/20; 75%), X (4/20; 20%), 9q, and 20q (3/20, 15% each); and losses on chromosomes 10 and 9p (7/20, 35% each), and 13q (3/20, 15%). Five different high‐level amplifications were mapped to 4q12–q21 (1 case), 6p21 (1 case), 7p12 (2 cases), proximal 12q (4 cases), and 14q32 (1 case) by CGH. Southern blot and/or differential PCR analyses identified amplification of PDGFRA (4q12), CCND3 (6p21), EGFR (7p12), CDK4 (12q14) and/or MDM2 (12q14.3–q15), and AKT1 (14q32.3) in the respective tumors. Separate analysis of the gliomatous and sarcomatous components of eight gliosarcomas by CGH after microdissection and universal DNA amplification revealed that both components shared 57% of the chromosomal imbalances detected. Taken together, our data indicate that the genomic changes in gliosarcomas closely resemble those found in glioblastomas. However, the number of chromosomes involved in imbalances in gliosarcomas was significantly lower than that in glioblastomas, indicating a higher genomic stability in gliosarcomas. In addition, we provide further support for the hypothesis that the gliomatous and sarcomatous components are derived from a single precursor cell clone, which progressed into subclones with distinct morphological features during tumor evolution. According to our data, gain/amplification of genes on proximal 12q may facilitate the development of a sarcomatous phenotype.

Chordoid Glioma of the Third Ventricle: Immunohistochemical and Molecular Genetic Characterization of a Novel Tumor Entity

Chordoid glioma of the third ventricle was recently reported as a novel tumor entity of the central nervous system with characteristic clinical and histopathological features (Brat et al., J Neuropathol Exp Neurol 57: 283–290, 1998). Here, we report on a histopathological, immunohistochemical and molecular genetic analysis of five cases of this rare neoplasm. All tumors were immunohistochemically investigated for the expression of various differentiation antigens, the proliferation marker Ki‐67, and a panel of selected proto‐oncogene and tumor suppressor gene products. These studies revealed a strong expression of GFAP, vimentin, and CD34. In addition, most tumors contained small fractions of neoplastic cells immunoreactive for epithelial membrane antigen, S‐100 protein, or cytokeratins. The percentage of Ki‐67 positive cells was generally low (< 5%). All tumors showed immunoreactivity for the epidermal growth factor receptor and schwan‐nomin/merlin. There was no nuclear accumulation of the p53, p21 (Waf‐1) and Mdm2 proteins. To examine genomic alterations associated with the development of chordoid gliomas, we screened 4 tumors by comparative genomic hybridization (CGH) analysis. No chromosomal imbalances were detected. More focussed molecular genetic analyses revealed neither aberrations of the TP53 and CDKN2A tumor suppressor genes nor amplification of the EGFR, CDK4, and M DM2 proto‐oncogenes. Our data strongly support the hypothesis that chordoid glioma of the third ventricle constitutes a novel tumor entity characterized by distinct morphological and immunohistochemical features, as well as a lack of chromosomal and genetic alterations commonly found in other types of gliomas or in meningiomas.

Allelic gain and amplification on the long arm of chromosome 17 in anaplastic meningiomas.

Using comparative genomic hybridization (CGH) we have previously identified amplification at 17q21‐qter as a common aberration in anaplastic meningiomas but not in atypical or benign meningiomas (19). To define the amplified genomic region, we analyzed 44 meningeal tumors, including 7 benign meningiomas of World Health Organization (WHO) grade I, 19 atypical meningiomas (WHO grade II) and 18 anaplastic meningiomas (WHO grade III) at 46 chromosome 17 loci (including 42 17q loci). In line with the CGH data we found evidence of increased numbers of alleles on 17q. The incidence rose with malignancy grade, culminating at 61% (11 of 18 cases) in the anaplastic meningioma group. The majority of cases showing increased allele numbers had, on average, low‐level allelic gains (relative increase in allele dosage of 2‐ to 5‐fold). Amplification of alleles (defined here as an average relative increase in allele dosage of more than 5 times) was detected in 2 anaplastic meningiomas. The amplification patterns in these tumors defined a number of common regions of amplification/increased allele copy number, the best defined include one between D17S790 and D17S1607 and one between D17S1160 and PS6K. Real‐time PCR analysis of the PS6K candidate gene revealed no high‐level amplification despite this affecting adjacent loci. Our findings are fundamental for the identification of the gene(s) in 17q22‐q23 that is (are) the target(s) for increased copy number in anaplastic meningiomas and possibly other tumor types.

Autoradiographic Visualization of A1 Adenosine Receptors in Rat Brain with [3H]8‐Cyclopentyl‐1,3‐Dipropylxanthine

A1 adenosine receptors were labeled in rat brain sections with the antagonist [3H]8‐cyclopentyl‐1,3‐dipropylxanthine ([3H]DPCPX) and visualized at the light microscopic level using autoradiography. The specific binding of [3H]DPCPX to the sections showed the pharmacological characteristics of A1 adenosine receptors and was accompanied by very low levels of nonspecific binding. Whereas GTP had no significant effect on [3H]DPCPX binding to rat brain membranes, the addition of 100 μM GTP increased the apparent affinity of [3H]DPCPX to tissue sections fivefold (from 1.83 to 0.35 nM), enhancing it to the affinity measured in membranes. However, GTP altered neither the binding capacity nor the distribution of binding sites in tissue sections. It is suggested that a competitive antagonism with endogenous adenosine explains the lower affinity of [3H]DPCPX in the absence of GTP. The autoradiographic pattern of [3H]DPCPX binding was characteristic for A1 adenosine receptors. Distinct labeling of the different layers of the cerebellar cortex was shown by photomicrographs generated with the coverslip technique. In addition, several fiber tracts were found to be labeled. The high selectivity for A1 adenosine receptors and low nonspecific binding of [3H]DPCPX, the ability to produce high‐resolution autoradiograms, together with the fact that the effects of endogenous adenosine can be eliminated by the addition of GTP make [3H]DPCPX a very useful tool in the autoradiographic study of A1 adenosine receptors.

Analysis of human meningiomas for aberrations of the MADH2, MADH4, APM‐1 and DCC tumor suppressor genes on the long arm of chromosome 18

We have previously reported that losses of genomic material from the long arm of chromosome 18 are frequent in atypical and anaplastic meningiomas but rare in benign meningiomas. In the present study, we have investigated a series of 37 meningiomas for mutation and expression of 4 tumor suppressor genes (MADH2, MADH4, APM‐1 and DCC) located at 18q21. Comparative genomic hybridization or loss of heterozygosity analysis showed losses on chromosome 18 that included sequences from 18q21 in 15 of 37 tumors. Mutation analysis of APM‐1revealed a missense mutation (c. 1819G>A: G607S) in 1 atypical meningioma. None of the tumors showed mutations of MADH2 and MADH4 or loss of detectable transcripts from MADH2, MADH4, APM‐1and DCC. In contrast to human brain tissue, normal leptomeninges and meningiomas showed preferential expression of a DCC splice variant lacking 60 base pairs from exon 17. Taken together, our data do not support a significant role for MADH2, MADH4, APM‐1 and DCC alterations in the pathogenesis of meningiomas. The targeted gene that is inactivated in most meningiomas with 18q losses remains to be identified.

Autoradiographic visualization of A1-adenosine receptors in brain and peripheral tissues of rat and guinea pig using 125I-HPIA

A1-adenosine receptors were identified in sections of rat brain and guinea pig kidney with the radioiodinated agonist 125I-N6-p-hydroxyphenylisopropyladenosine (125I-HPIA) using in vitro autoradiography. The affinities of adenosine receptor ligands in competing with 125I-HPIA binding to tissue sections were in good agreement with those found in membranes, and indicate that the binding site represents an A1-adenosine receptor. The distribution of 125I-HPIA binding sites in rat brain sections was similar to the pattern of [3H]N6-cyclohexyladenosine ([3H]CHA) binding sites determined previously, with highest densities in the hippocampus and dentate gyrus, the cerebellar cortex, some thalamic nuclei and certain layers of the cerebral cortex. In the guinea pig kidney 125I-HPIA labelled longitudinal structures in the medulla. This study demonstrates that 125I-HPIA allows the autoradiographic detection of A1 adenosine receptors in the brain and peripheral organs and has the advantage of short exposure times.

High-resolution comparative hybridization to combed DNA fibers

Comparative genomic hybridization (CGH) has proven to be a comprehensive new tool to detect genetic imbalances in genomic DNA. However, the resolution of this method carried out on normal human metaphase spreads is limited to low copy number gains and losses of ≥ 10 Mb. An improved resolution allowing the detection of copy number representations of single genes would strongly enhance the applicability of CGH as a diagnostic and research tool. This goal may be achieved when metaphase chromosomes are replaced by an array of target DNAs representing the genes of interest. To explore the feasibility of such a development in a model system we used cosmid MA2B3, which encompasses about 35 kb in the vicinity of exon 48 of the human dystrophin gene. Linearized cosmid fibers were attached to a glass surface and aligned in parallel by “molecular combing”. Two-color fluorescence in situ suppression hybridization was performed on these cosmid fibers with probe mixtures containing different ratios (ranging from 1:2 to 4:1) of biotin- and digoxigenin-labeled MA2B3 cosmid DNAs. For each mixture fluorescence ratios were determined for 40–50 individual combed DNA molecules. In two series comprising a total of 651 molecules the median fluorescence ratio measurements revealed a linear relationship with the chosen probe ratios. Our study demonstrates that fluorescence ratio measurements on single DNA molecules can be performed successfully.