Abstract
We have previously reported on the amplification and overexpression of the MDM2 proto-oncogene in a subset of malignant gliomas without TP53 mutation (G. Reifenberger et al., Cancer Res., 53: 2736–2739, 1993). Here, we show that the MDM4 (MDMX) gene located on 1q32 is a further target for amplification in malignant gliomas. MDM4 codes for a Mdm2-related protein that can bind to p53 and inhibits p53-mediated transcriptional transactivation. We investigated a series of 208 gliomas (106 glioblastomas, 46 anaplastic gliomas, and 56 low-grade gliomas) and identified 5 tumors (4 glioblastomas and 1 anaplastic oligodendroglioma) with MDM4 amplification and overexpression. Several other genes from 1q32 were found to be coamplified with MDM4, such as GAC1 in five tumors, REN in four tumors, and RBBP5 in three tumors. Additional analyses revealed that the malignant gliomas with MDM4 amplification and overexpression carried neither mutations in conserved regions of the TP53 gene nor amplification of the MDM2 gene. Taken together, our data indicate that amplification and overexpression of MDM4 is a novel molecular mechanism by which a small fraction of human malignant gliomas escapes p53-dependent growth control.
Introduction
Gliomas are the most common primary tumors of the central nervous system. The molecular genetic alterations associated with the development and progression of these tumors have been intensively studied over the past years (for recent reviews, see Refs. 1, 2). Common aberrations identified in well-differentiated gliomas of WHO grade II include mutation of the TP53 gene on 17p13 in about 30–50% of diffusely infiltrative astrocytomas (1, 2, 3), as well as the loss of genetic information on chromosome arms 1p and 19q in up to 80% of oligodendrogliomas (4, 5, 6). Malignant progression to anaplastic gliomas (WHO grade III) and glioblastoma (WHO grade IV) is associated with the acquisition of multiple additional genetic changes. These include mutation and/or homozygous deletion of the tumor suppressor genes CDKN2A, RB1, and PTEN, as well as amplification of various proto-oncogenes, such as EGFR, CDK4, MDM2, and PDGFRA (1, 2). It has also become apparent that many of the genetic alterations detected in gliomas affect genes that encode members of the pRb14- and/or p53-dependent growth regulatory pathways. The pRb1-dependent G1-S-phase cell cycle checkpoint is impaired in a significant fraction of anaplastic gliomas and the vast majority of glioblastomas, due to RB1 mutation, or homozygous CDKN2A deletion, or amplification and overexpression of CDK4, CDK6, CCND1, or CCND3 (2, 7, 8, 9, 10). In addition, p53-dependent pathways are frequently altered in diffusely infiltrative astrocytomas including glioblastomas, due either to TP53 mutation or MDM2 amplification and overexpression (2, 3, 11). Homozygous CDKN2A deletion associated with loss of p14ARF expression represents a further aberration that likely contributes to the impairment of p53 function in a considerable fraction of malignant gliomas (12).
The Mdm4 (Mdmx) protein has been identified as a novel p53-binding protein that shares structural and functional properties with Mdm2 (13, 14). Like Mdm2, Mdm4 can bind to the NH2-terminal part of p53 and, thereby, inactivates the function of p53 as a transcriptional activator (13, 14). In addition, Mdm4 has been shown to bind to Mdm2, an interaction that is mediated by the COOH-terminal RING finger domains of both proteins and results in the inhibition of Mdm2 degradation (15). The human MDM4 gene has been mapped to chromosome band 1q32 (14), a region repeatedly shown to contain amplified sequences by CGH analyses of human malignant gliomas (5, 16, 17). In addition, we have previously reported one anaplastic oligodendroglioma with amplification of the renin gene (REN) on 1q32 (4), and other authors have demonstrated amplification and overexpression of a novel gene from 1q32 termed GAC1 (“glioma amplified on chromosome 1”) in three glioblastomas and one anaplastic astrocytoma (18).
In the present study, we report on data that suggest that the MDM4 gene is the major target for 1q32 amplifications in malignant gliomas. Other genes located in this chromosomal region, including GAC1, REN, and RBBP5, may be coamplified in gliomas with MDM4 amplification. The absence of TP53 mutations and MDM2 amplification in all of the tumors with MDM4 amplification indicates that amplification and overexpression of MDM4 represents an alternative molecular mechanism for inactivating the p53-dependent growth control in malignant gliomas.
Materials and Methods
Tumor Samples.
The tumors were selected from the tumor tissue collection of the Department of Neuropathology, Heinrich-Heine-University, Düsseldorf, Germany. All of the tumors were classified according to the WHO classification of tumors of the central nervous system (19). The tumor series consisted of 106 glioblastomas of WHO grade IV (84 classic glioblastomas, 11 giant cell glioblastomas, and 11 gliosarcomas), 46 anaplastic gliomas of WHO grade III (6 anaplastic astrocytomas, 27 anaplastic oligodendrogliomas, 6 anaplastic oligoastrocytomas, and 7 anaplastic ependymomas), 49 gliomas of WHO grade II (13 astrocytomas, 16 oligodendrogliomas, 6 oligoastrocytomas, and 14 ependymomas), and 7 gliomas of WHO grade I (3 pilocytic astrocytomas, 3 myxopapillary ependymomas, and 1 subependymoma). Parts of each tumor were snap-frozen immediately after operation and stored at −80°C. To ensure that the tumor pieces taken for molecular genetic analysis contained a sufficient proportion of tumor cells, histological evaluation of a representative part of each of these pieces was performed. Only tumor pieces with an estimated tumor cell content of 80% or more were used for molecular genetic analysis. As reference tissue for the expression studies at the mRNA level, we used nonneoplastic cerebral tissue (cortex and white matter) from the temporal lobe of an adult patient operated on for chronic epilepsy.
DNA and RNA Extraction.
Extraction of high-molecular-weight DNA and RNA from frozen tumor tissue was carried out by ultracentrifugation as described previously (7). Extraction of high-molecular-weight DNA from peripheral blood leukocytes was performed according to a standard protocol.
Duplex PCR Analyses.
All of the 208 tumors were screened for amplification of MDM4 and GAC1 using duplex PCR analyses. The tumors showing amplification of these genes were additionally analyzed for coamplification of the REN, ELF3, ELK4, and PTPN7 genes from 1q32, and amplification of the MDM2 gene from 12q14–q15. The individual primer sequences used for amplification of fragments from each of these genes, as well as the reference genes GAPDH (12p13), IFNG (12q15), and DES (2q35) are listed in Table 1. The primer sequences for duplex PCR analysis of MDM2 have been reported (20). Each PCR reaction was performed in a total volume of 10 μl using 10 ng of template DNA. PCR conditions, including cycle number (26–28 cycles), MgCl2 concentration and annealing temperature, were optimized for each duplex PCR assay. The PCR products were separated on 3% agarose gels and ethidium bromide-stained bands were recorded by the Gel-Doc 1000 system (Bio-Rad, Hercules, CA). Quantitative analysis of the signal intensities obtained for each target gene and the corresponding reference gene was performed with the Molecular Analyst software (version 2.1, BioRad). Only increases in the target:reference gene ratio of more than 5-times the ratio obtained for constitutional DNA were considered as gene amplification.
Southern Blot Analysis.
The gliomas showing amplification of genes from 1q32 by duplex PCR were investigated by Southern blot analysis using PCR-generated probes for MDM4 (a 221-bp fragment covering nucleotides 907–1127 of the MDM4 coding sequence, GenBank accession no. AF007111), GAC1 (a 234-bp fragment covering nucleotide 1846–2079 of the GAC1 coding sequence, GenBank accession no. AF030435), REN (a 215-bp fragment spanning exon 4, nucleotides 1291–1505 of GenBank accession no. M26900), and RBBP5 (a 285-bp fragment corresponding to nucleotides 449–733 of the RBBP5 coding sequence, GenBank accession no. X85134). The primers used for PCR amplification of these probes are listed in Table 1. The individual PCR products were purified after separation by agarose gel electrophoresis using the Jetsorb DNA isolation kit (Genomed, Bad Oeynhausen, Germany). For Southern analysis, 2.5 μg of DNA was digested with the restriction enzyme TaqI, separated on 0.8% agarose gels, and alkali-blotted to Hybond-N+ membranes (Amersham Life Sciences, Uppsala, Sweden). The membranes were hybridized with DNA probes labeled with [α-32P]dCTP by random priming. Hybridized membranes were exposed to imaging plates (Fuji, Kanagawa, Japan) and analyzed using the FLA-2000 Phosphor and Fluorescent Image Analyzer (Fuji). As reference for the assessment of gene copy number, the blots were hybridized with probes for the variable number of tandem repeat locus D2S44 on 2q21.3–q22 (pYNH24, obtained from American Type Culture Collection) and a PCR-generated probe (20) for the CCNA gene on 4q25–q31. Densitometric analyses of gene copy number was performed with the Mac BAS Version 2.5 software (Image Reader, Version 1.4E, and Image Gauge, Version 3.0) supplied with the FLA-2000 analyzer. Only a relative increase in signal intensity of more than 5-times that of normalized constitutional DNA was considered as gene amplification.
Mutational Analyses.
SSCP/heteroduplex analysis for mutations in exons 4–10 of the TP53 gene was carried out as described (21). In brief, PCR products from each exon were separated by electrophoresis on 10% nondenaturing polyacrylamide gels at 120 V for 16 h. Each fragment was analyzed at room temperature and at 4°C. After electrophoresis, the SSCP/heteroduplex band patterns were visualized by silver staining of the gels.
Duplex Reverse Transcription-PCR Analyses.
A total of 142 tumors (75 glioblastomas, 6 anaplastic astrocytomas, 15 anaplastic oligodendrogliomas, 5 anaplastic oligoastrocytomas, 6 anaplastic ependymomas, 11 astrocytomas, 8 oligodendrogliomas, 6 oligoastrocytomas, 7 ependymomas, and 3 pilocytic astrocytomas), including the 5 tumors with MDM4 and GAC1 amplification, were investigated for expression of transcripts from these genes by duplex reverse transcription-PCR using β-2-microglobulin (B2MG) transcript levels as reference. The tumors with amplification of genes from 1q32 were additionally analyzed by duplex reverse transcription-PCR for expression of transcripts from RBBP5 and REN. Three μg of total RNA from each tumor were reverse-transcribed into cDNA in a total volume of 50 μl using random hexanucleotide primers and Superscript reverse transcriptase (Life Technologies, Inc., Eggenstein, Germany). PCR conditions, including cycle number (24–30 cycles), MgCl2 concentration and annealing temperature were optimized for each PCR reaction. The respective primer sequences are listed in Table 1. Agarose gel electrophoresis and densitometric analysis of the PCR products was carried out as described above. The ratio between target mRNA and B2MG mRNA signal intensities was calculated for each tumor and normalized to the target mRNA:B2MG mRNA ratio determined for nonneoplastic brain tissue.
Expression of the recently described MDM4-short (MDM4-S) splice variant (22), which carries a 68-bp deletion covering nucleotides 343–410 of the MDM4 cDNA sequence (GenBank accession no. AF007111), was assessed by reverse transcription-PCR using primers flanking the deleted sequence (Table 1). The ratio between the MDM4-S splice variant and the normal MDM4 transcript was densitometrically determined and normalized to the MDM4-S:MDM4 ratio determined for nonneoplastic brain tissue.
Northern Blot Analyses.
Northern blot analysis was carried out as described (10, 11). A synthetic oligonucleotide probe complementary to bases 101–150 of the GAPDH mRNA (GenBank accession no. XO1677) was used to assess variations in RNA loading of the Northern blots. Hybridized membranes were analyzed with the FLA-2000 imager as described above.
Immunohistochemical Analyses.
Paraffin sections of the five gliomas with MDM4 amplification were immunohistochemically stained for p53 (clone DO7, Dako, Glostrup, Denmark) and Mdm2 (clone SMP14, Dako) using the indirect avidin-biotin-peroxidase method for detection of antibody binding. To enhance immunoreactivity, sections were pretreated by microwave heating in 10 mm citrate buffer (pH 6.0) three times for 10 min each. The primary antibodies were applied at a concentration of 1 μg/ml for an incubation period of 16 h at room temperature. This was subsequently followed by incubations with (a) biotinylated rabbit-antimouse-immunoglobulin antiserum (Dako) for 30 min and (b) avidin-biotin-peroxidase-complex (Dako) for 30 min. Antibody binding was visualized with 3′3-diaminobenzidine (Sigma, Deisenhofen, Germany) as the chromogenic substrate. Between each pair of steps, sections were washed in PBS two times for 10 min each. Negative controls were performed by omission of the primary antibody and its substitution with an irrelevant mouse monoclonal antibody. All of the sections were counterstained with hemalum.
Results
The entire series of 208 gliomas was screened by duplex PCR for MDM4 and GAC1 amplification. Four (4%) of 106 glioblastomas and 1 (4%) of 27 anaplastic oligodendrogliomas demonstrated increased target:reference gene ratios indicating amplification of these genes (Fig. 1). Four of these five tumors showed evidence for REN coamplification by duplex PCR analysis. Amplification of MDM4, GAC1, and REN was confirmed by Southern blot analysis (Fig. 2). In addition, Southern blot analysis revealed amplification of the RBBP5 gene in three of the five tumors with MDM4 amplification (Fig. 2). Densitometric determination of the signal intensities obtained by Southern blot hybridization revealed amplification levels between 5- and 6-fold for the glioblastomas GB31D and GB35D, about 10-fold for the glioblastoma GB216D, approximately 15-fold for the anaplastic oligodendroglioma AO11D, and about 25-fold for the glioblastoma GB112D. In each tumor, copy numbers of the coamplified GAC1, REN, and RBBP5 genes were elevated to similar levels. None of the tumors with MDM4 amplification demonstrated amplification of ELK4, ELF3, and PTPN7. Two (GB96D and AA7D) of the remaining 203 gliomas showed evidence for low-level gains of MDM4 and GAC1 copy number by duplex PCR analysis (Fig. 1). On Southern blotting, these tumors demonstrated an approximately 2-fold increase in MDM4 and GAC1 gene dosage (Fig. 2), which was not considered gene amplification.
All five of the tumors with MDM4 amplification demonstrated enhanced expression of MDM4 transcripts by duplex reverse transcription-PCR (Fig. 3 a). Densitometrical analysis revealed increases in MDM4:B2MG ratios ranging from 5-fold (GB31D and GB35D) to more than 10-fold (GB112D, GB216D, and AO11D) relative to the nonneoplastic brain reference. Investigation of 137 gliomas without MDM4 amplification detected MDM4 transcripts in all of the tumors. Only two gliomas without MDM4 amplification (1 ependymoma and 1 anaplastic ependymoma) demonstrated a more than 4-fold increase in the MDM4:B2MG ratio. All of the other gliomas without MDM4 amplification showed MDM4:B2MG ratios between 0.2-fold and 3-fold relative to nonneoplastic brain tissue.
A selected series of 30 gliomas, including 2 of the 5 gliomas with MDM4 amplification (GB216D and AO11D), was additionally investigated for MDM4 mRNA expression by Northern blotting. A major MDM4 transcript of about 10 kb and an additional transcript of about 15 kb were expressed in nonneoplastic brain tissue and all of the gliomas investigated. Both of the transcripts were overexpressed in GB216D and AO11D as compared with nonneoplastic brain tissue and gliomas without MDM4 amplification (Fig. 3 b).
The recently reported MDM4-S splice variant (22) was detected by reverse transcription-PCR in nonneoplastic brain tissue and all of the gliomas investigated (Fig. 3 c). The presence of this splice variant was confirmed by DNA sequencing (data not shown). Densitometrical analysis revealed that the MDM4-S:MDM4 ratio varied considerably from tumor to tumor. The mean MDM4-S:MDM4 ratio was significantly higher in glioblastomas (mean ratio, 4.2; range, 0.6–13.7; n = 75) than in anaplastic gliomas (mean ratio, 1.6; range, 0.6–4.0; n = 33) and WHO grade II gliomas (mean ratio, 1.9; range, 0.4–8.9; n = 32) (t test, P < 0.001). Glioblastomas with MDM4 amplification showed no significant difference in the mean MDM4-S:MDM4 ratio as compared to glioblastomas without MDM4 amplification (t test, P = 0.75).
Expression of GAC1 transcripts was found to be elevated in four of five malignant gliomas with GAC1 amplification, with GAC1:B2MG ratios ranging from 2-fold to more than 5-fold relative to nonneoplastic brain tissue (Fig. 3,a). One glioblastoma (GB31D) demonstrated GAC1 amplification but no increase in GAC1 mRNA expression, i.e., the normalized GAC1:B2MG ratio in this tumor was only 0.8. Investigation of 137 gliomas without GAC1 amplification demonstrated GAC1 transcripts by reverse transcription-PCR in 116 tumors. Two ependymomas and 1 anaplastic ependymoma (including the two tumors with elevated MDM4 mRNA levels) demonstrated GAC1:B2MG ratios of more than 4-fold relative to nonneoplastic brain tissue. In the remaining tumors, GAC1:B2MG ratios were generally below 2.5 times the ratio determined for nonneoplastic brain tissue, with the majority of tumors (114 of 134) showing ratios equal to, or lower than, that obtained for nonneoplastic brain tissue. The four malignant gliomas with REN amplification showed elevated REN:B2MG transcript ratios relative to nonneoplastic brain tissue (Fig. 3,a). Similarly, the three malignant gliomas with RBBP5 amplification overexpressed the respective mRNA (Fig. 3 a).
SSCP/heteroduplex analysis of exon 4–10 of the TP53 gene did not reveal aberrantly migrating bands in any of the five malignant gliomas with MDM4 amplification. None of these tumors showed amplification of the MDM2 gene (Fig. 1). Histologically, four of the tumors with MDM4 amplification demonstrated the classical morphological features of glioblastoma multiforme. Tumor AO11D was an anaplastic oligodendroglioma with prominent formation of polar spongioblastoma-like architectures. All of the tumors with MDM4 amplification were primary malignant gliomas, i.e., none of the patients had a history of previous surgery for a lower-grade glioma. Immunohistochemical staining of sections from these tumors for p53 and Mdm2 proteins remained completely negative in GB31D, GB216D, and AO11D. Small fractions of tumor cells (<5%) showed nuclear immunoreactivity for p53 and Mdm2 in tumors GB35D and GB112D.
Discussion
The TP53 gene is a frequent target for mutations in human gliomas, in particular in tumors of astrocytic differentiation (1, 2, 3). The TP53 gene product (p53) is critically involved in the regulation of cell cycle progression, apoptosis, and DNA repair (for review, see Ref. 23). The activity of p53 is regulated by the Mdm2 protein, which can bind to p53 and inhibits p53-mediated transcriptional activation of several genes including MDM2 itself (for review, see Ref. 24). Furthermore, the binding of Mdm2 to p53 results in nuclear export and proteolytic degradation of p53 by the ubiquitin-proteasome pathway (24). Amplification and overexpression of the MDM2 gene has been detected in various human tumor types (for review, see Ref. 25), including about 8–10% of malignant gliomas (11). In malignant gliomas, MDM2 amplification and overexpression are exclusively found in tumors without TP53 mutation, suggesting that both alterations are alternative mechanisms for the inactivation of p53 function (11).
The present study reports on the first demonstration of amplification and overexpression of the MDM4 gene on 1q32 in human tumors. The MDM4 gene product has been identified as a p53-binding protein with structural homology to Mdm2 (13, 14). Like Mdm2, Mdm4 contains a p53-binding domain at the NH2 terminus, a centrally located zinc finger domain, and a RING finger domain at the COOH terminus (13, 14). Binding of Mdm4 to p53 has been shown to inhibit p53-mediated transcriptional transactivation of other genes (13, 14). Mdm4 can also bind to Mdm2 and, thereby, inhibits Mdm2 degradation (15). In contrast to Mdm2, however, Mdm4 expression is neither induced by DNA damage nor transcriptionally regulated by p53 (13).
We identified MDM4 amplification and overexpression in five primary malignant gliomas. None of these tumors showed detectable TP53 mutations or MDM2 amplification, which suggests that MDM4 amplification and consecutive overexpression is a further (alternative) mechanism by which the p53-dependent growth control is inactivated in malignant gliomas. Immunohistochemistry for p53 and Mdm2 revealed either no staining of tumor cells or nuclear immunoreactivity restricted to less than 5% of tumor cells. Thus, the overexpression of Mdm4 in gliomas with MDM4 amplification does not result in an accumulation of the p53 and/or Mdm2 proteins.
We also investigated our tumor series for expression of the recently identified MDM4-S splice variant (22). This splice variant was constitutively expressed in nonneoplastic brain tissue as well as in gliomas of all of the major types. Among the tumors without MDM4 amplification, the general MDM4 transcript level was not elevated in glioblastomas as compared with anaplastic gliomas and low-grade gliomas. However, the mean expression level of MDM4-S relative to MDM4 was significantly higher in glioblastomas. The truncated protein encoded by MDM4-S has been reported to more efficiently suppress p53-mediated transcriptional transactivation and induction of apoptosis than full-length Mdm4 (22). Thus, it is possible that the increased expression of MDM4-S in glioblastomas further contributes to the functional impairment of p53-dependent regulatory pathways in these tumors.
The GAC1 gene has been recently identified as a gene located on 1q32.1 that is amplified in malignant gliomas (18). The GAC1 gene product is a transmembrane protein belonging to the leucine-rich repeat superfamily (18). The biological function of Gac1 is presently unknown. However, its structural homology to other leucine-rich repeat proteins known to be involved in cell adhesion or signal transduction suggests a possible function as cell-adhesion molecule or receptor protein (18). We found GAC1 coamplified in all five of the gliomas with MDM4 amplification from our series. In contrast to MDM4 amplification, however, GAC1 amplification was not invariably associated with overexpression at the transcript level. This finding argues for MDM4 being the more important amplification target than GAC1. Nevertheless, it is possible that GAC1 amplification and overexpression may provide an additional growth advantage to glioma cells.
We have previously reported on the amplification of REN in the anaplastic oligodendroglioma AO11D (4). Northern blot analysis of AO11D, however, showed no overexpression of REN transcripts (4). Similarly, Almeida et al. (18) observed coamplification of REN with GAC1 in two glioblastomas but no REN overexpression on Northern blotting. In the present study, we identified four malignant gliomas with coamplification of REN and MDM4 (including AO11D). Employing reverse transcription-PCR, we found that REN mRNA is expressed at low levels in nonneoplastic brain tissue and at higher levels in all of the gliomas with REN amplification, including AO11D. These data are in line with an immunohistochemical study reporting on the detection of renin expression in 8 of 10 glioblastomas investigated (26). The negative findings obtained previously by Northern blotting (4, 18) are, thus, likely due to the limited sensitivity of this method. The biological significance of renin expression in glioma cells is unclear at present. It is possible that increased expression of renin can provide a growth advantage, for example, by promoting microvascular proliferation (26). Furthermore, tumor-associated overexpression of renin has been reported as a rare cause of paraneoplastic arterial hypertension (27, 28). However, review of the clinical files of the patients whose gliomas showed REN amplification revealed no record of pre- or postoperative arterial hypertension.
Investigation of other candidate genes from 1q32 revealed coamplification of RBBP5 in three of five tumors with MDM4 amplification. The RBBP5 gene product is a Mr 66,000 nuclear protein that can bind to the E1A-binding pocket B of pRB1 and that preferentially associates with underphosphorylated pRB1 (29). The biological significance of this interaction is unknown, and it remains to be shown whether RBBP5 amplification and overexpression provides a selective growth advantage to glioma cells. Interestingly, overexpression of another pRb1-binding protein termed Bog (B5T-overexpressed gene) has recently been shown to confer resistance to transforming growth factor beta-1-mediated growth suppression and results in neoplastic transformation of normal rat liver cells (30). No amplification of the ELF3 and ELK4 genes from 1q32, both coding for members of the ets family of transcription factors (31, 32), was found in the gliomas with MDM4 amplification. In addition, we did not detect amplification of the PTPN7 (HePTP) gene, which encodes a hematopoietic tyrosine phosphatase and has been reported to be amplified and overexpressed in preleukemic myeloproliferative disorders (33).
In summary, our data indicate that amplification and overexpression of MDM4 represents a novel molecular mechanism by which a small subset of malignant gliomas escapes p53-dependent growth control. Coamplification and overexpression of other genes from 1q32, including GAC1, REN, and RBBP5, may provide an additional growth advantage in some gliomas with MDM4 amplification. Because comparative genomic hybridization studies have detected amplification of sequences from 1q32 not only in malignant gliomas but also in other human tumors, for example, breast and ovarian carcinomas, small cell lung carcinomas, osteosarcomas, and certain types of hematopoietic neoplasms (for review, see Ref. 34), it is likely that MDM4 represents an important amplification target in various human tumor types.
Demonstration of MDM4, GAC1, and MDM2 gene amplification in human malignant gliomas by duplex PCR. The numbers on top of the panels: 1, GB96D; 2, GB31D; 3, GB35D; 4, GB112D; 5, constitutional DNA; 6, GB216D; 7, AO11D; 8, GB110D; 9, AA7D; 10 and 11, constitutional DNA; 12, TP365MG (positive control for MDM2 amplification). Signal intensities are significantly increased (>5-fold relative to normalized constitutional DNA) for MDM4 and GAC1 in tumors GB31D, GB35D, GB112D, GB216D, and AO11D. Tumors GB96D and AA7D demonstrated evidence for low copy number gains. Note that none of the tumors with MDM4 amplification showed MDM2 amplification.
Demonstration of MDM4, GAC1, and MDM2 gene amplification in human malignant gliomas by duplex PCR. The numbers on top of the panels: 1, GB96D; 2, GB31D; 3, GB35D; 4, GB112D; 5, constitutional DNA; 6, GB216D; 7, AO11D; 8, GB110D; 9, AA7D; 10 and 11, constitutional DNA; 12, TP365MG (positive control for MDM2 amplification). Signal intensities are significantly increased (>5-fold relative to normalized constitutional DNA) for MDM4 and GAC1 in tumors GB31D, GB35D, GB112D, GB216D, and AO11D. Tumors GB96D and AA7D demonstrated evidence for low copy number gains. Note that none of the tumors with MDM4 amplification showed MDM2 amplification.
Demonstration of MDM4, GAC1, REN, and RBBP5 gene amplification in human malignant gliomas by Southern blot analysis. Case numbers are given on the top of the blots (GB, glioblastoma; AO, anaplastic oligodendroglioma). Southern blotted tumor DNA (T) and corresponding leukocyte DNA (B) from four patients were hybridized with radiolabeled probes for MDM4, GAC1, RBBP5, and the reference locus D2S44. There is coamplification of MDM4, GAC1, REN, and RBBP5 in tumor AO11D and coamplification of MDM4, GAC1, and REN in tumor GB112D. The glioblastoma GB96D showed about a 2-fold increased gene dose of MDM4 and GAC1, whereas glioblastoma GB110D demonstrated normal copy numbers of all of the three genes. Right ordinate, approximate sizes of hybridized restriction fragments in kb.
Demonstration of MDM4, GAC1, REN, and RBBP5 gene amplification in human malignant gliomas by Southern blot analysis. Case numbers are given on the top of the blots (GB, glioblastoma; AO, anaplastic oligodendroglioma). Southern blotted tumor DNA (T) and corresponding leukocyte DNA (B) from four patients were hybridized with radiolabeled probes for MDM4, GAC1, RBBP5, and the reference locus D2S44. There is coamplification of MDM4, GAC1, REN, and RBBP5 in tumor AO11D and coamplification of MDM4, GAC1, and REN in tumor GB112D. The glioblastoma GB96D showed about a 2-fold increased gene dose of MDM4 and GAC1, whereas glioblastoma GB110D demonstrated normal copy numbers of all of the three genes. Right ordinate, approximate sizes of hybridized restriction fragments in kb.
a, analysis of MDM4, GAC1, RBBP5, and REN mRNA expression in malignant gliomas by duplex reverse transcription-PCR. The numbers on top of the panels: 1, GB96D; 2, GB31D; 3, GB35D; 4, GB112D; 5, nonneoplastic brain tissue; 6, GB216D; 7, AO11D; 8, GB110D; 9, AA7D; 10, nonneoplastic brain tissue; 11, normal kidney tissue. Right ordinate, sizes of amplified fragments in bp. A fragment from each transcript was coamplified with a 114-bp fragment from the B2MG mRNA. All of the tumors with MDM4 amplification demonstrate increased MDM4 mRNA levels relative to nonneoplastic brain tissue. GAC1 mRNA levels are elevated in 4 of 5 tumors with GAC1 amplification whereas RBBP5 transcripts are overexpressed in 3 of 3 tumors with RBBP5 amplification. REN transcripts are abundant in normal kidney (Lane 11) and weakly expressed in nonneoplastic brain tissue. The four gliomas with REN amplification showed elevated levels of REN mRNA relative to nonneoplastic brain tissue. +, tumors with amplification of the respective genes. b, Northern blot analysis of MDM4 mRNA expression. The same blot was hybridized with probes for MDM4 and GAPDH. Lanes 1–6: gliomas without MDM4 amplification; Lane 7: glioma with MDM4 amplification (GB216D); Lane 8: nonneoplastic brain tissue. There is an increased expression of MDM4 transcripts in GB216D. Right ordinate, approximate transcript sizes in kbp. c, expression of the MDM4-S splice variant in gliomas. Shown are results obtained for eight glioblastomas (top panel), eight anaplastic gliomas (middle panel), and eight WHO grade II gliomas (bottom panel). The MDM4-S transcript is coexpressed with the normal MDM4 transcript in nonneoplastic brain tissue and all tumors. The MDM4-S:MDM4 ratio is higher in glioblastomas than in anaplastic or WHO grade II gliomas. Glioblastomas with MDM4 amplification (Lanes 1–2, upper panel) show similar MDM4-S:MDM4 ratios as glioblastomas without MDM4 amplification (Lanes 3–8, upper panel). Right ordinate, sizes of amplified fragments in bp. NB, nonneoplastic brain tissue.
a, analysis of MDM4, GAC1, RBBP5, and REN mRNA expression in malignant gliomas by duplex reverse transcription-PCR. The numbers on top of the panels: 1, GB96D; 2, GB31D; 3, GB35D; 4, GB112D; 5, nonneoplastic brain tissue; 6, GB216D; 7, AO11D; 8, GB110D; 9, AA7D; 10, nonneoplastic brain tissue; 11, normal kidney tissue. Right ordinate, sizes of amplified fragments in bp. A fragment from each transcript was coamplified with a 114-bp fragment from the B2MG mRNA. All of the tumors with MDM4 amplification demonstrate increased MDM4 mRNA levels relative to nonneoplastic brain tissue. GAC1 mRNA levels are elevated in 4 of 5 tumors with GAC1 amplification whereas RBBP5 transcripts are overexpressed in 3 of 3 tumors with RBBP5 amplification. REN transcripts are abundant in normal kidney (Lane 11) and weakly expressed in nonneoplastic brain tissue. The four gliomas with REN amplification showed elevated levels of REN mRNA relative to nonneoplastic brain tissue. +, tumors with amplification of the respective genes. b, Northern blot analysis of MDM4 mRNA expression. The same blot was hybridized with probes for MDM4 and GAPDH. Lanes 1–6: gliomas without MDM4 amplification; Lane 7: glioma with MDM4 amplification (GB216D); Lane 8: nonneoplastic brain tissue. There is an increased expression of MDM4 transcripts in GB216D. Right ordinate, approximate transcript sizes in kbp. c, expression of the MDM4-S splice variant in gliomas. Shown are results obtained for eight glioblastomas (top panel), eight anaplastic gliomas (middle panel), and eight WHO grade II gliomas (bottom panel). The MDM4-S transcript is coexpressed with the normal MDM4 transcript in nonneoplastic brain tissue and all tumors. The MDM4-S:MDM4 ratio is higher in glioblastomas than in anaplastic or WHO grade II gliomas. Glioblastomas with MDM4 amplification (Lanes 1–2, upper panel) show similar MDM4-S:MDM4 ratios as glioblastomas without MDM4 amplification (Lanes 3–8, upper panel). Right ordinate, sizes of amplified fragments in bp. NB, nonneoplastic brain tissue.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Supported by grants from the Deutsche Forschungsgemeinschaft (SFB400 C5), the Deutsche Krebshilfe/Dr. Mildred Scheel Foundation (10-1361-Re2), and the Schäfersnolte Foundation.
The abbreviations used are: pRb1, retinoblastoma protein; SSCP, single-strand conformation polymorphism; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
Primers used for the amplification and expression analyses
Genes . | Primer sequences . | GenBank accession no. . | Fragment sizes (bp) . | Applicationsa . |
---|---|---|---|---|
MDM4 | 5′-GAGTGGCAGTGTACTGAATGC-3′ | AF007111 | 221 | 1, 4 |
5′-GCATCTTTAGGTCTAACGACAG-3′ | ||||
MDM4 | 5′-CTCAGTGTCAACATCTGACAG-3′ | AF007111 | 197 | 2 |
5′-CATATGCTGCTCCTGCTGATC-3′ | ||||
MDM4 | 5′-CACTTTAGCCACTGCTACTAC-3′ | AF007111 | 167/100 | 3 |
5′-GCTCTGAGGTAGGCAGTGTG-3′ | ||||
GAC1 | 5′-AGACGGTTAGTGTGGTTGTGG-3′ | AF030435 | 234 | 1, 4 |
5′-GGGTAATGTTGTAGCTGTGGG-3′ | ||||
GAC1 | 5′-AGATGACGGACCACTGTTTGC-3′ | AF030435 | 300 | 2 |
5′-CCACAACCACACTAACCGTCT-3′ | ||||
RBBP5 | 5′-ACTCCGATTTGAACGTTGTGGC-3′ | X85134 | 285 | 2, 4 |
5′-ATTCCTGCATAGGTTCAGGCTC-3′ | ||||
REN | 5′-GACAAACCTGAGCCTCTGTC-3′ | M26900 | 215 | 1, 4 |
5′-GTGGGTCAGGAGAGGCCTG-3′ | ||||
REN | 5′-GTCGTCTTTGACACTGGTTC-3′ | M26900 | 191 | 2 |
5′-GTGATGATGTCCTGGCTGAG-3′ | ||||
ELK4 | 5′-GAATCCAGCCGAGAAACTGGC-3′ | NM_001973 | 335 | 1 |
5′-CTGTGTCGATGTCTGGGTGAG-3′ | ||||
ELF3 | 5′-GACTGGTGCTTCTCGGAGCTC-3′ | AF016295 | 178 | 1 |
5′-GTCCCAGTACTCTTTGCTCAGC-3′ | ||||
PTPN7 | 5′-CACAGTTGCACCTTCTGTAG-3′ | X53364 | 183 | 1 |
5′-GAGTCTGAAGAATTGACAGTG-3′ | ||||
DES | 5′-TTCGATGTACATTTCCATCA-3′ | M63391 | 170 | 1 |
5′-ACAACTAACTGAAAAGAGAG-3′ | ||||
GAPDH | 5′-AACGTGTCAGTGGTGGACCTG-3′ | J04038 | 160 | 1 |
5′-AGTGGGTGTCGCTGTTGAAGT-3′ | ||||
IFNG | 5′-GCAGAGCCAAATTGTCTCC-3′ | J00219 | 94 | 1 |
5′-GTCTTCCTTGATGGTCTCC-3′ | ||||
B2MG | 5′-ACCCCCACTGAAAAAGATGA-3′ | AB021288 | 114 | 2 |
5′-ATCTTCAAACCTCCATGATG-3′ |
Genes . | Primer sequences . | GenBank accession no. . | Fragment sizes (bp) . | Applicationsa . |
---|---|---|---|---|
MDM4 | 5′-GAGTGGCAGTGTACTGAATGC-3′ | AF007111 | 221 | 1, 4 |
5′-GCATCTTTAGGTCTAACGACAG-3′ | ||||
MDM4 | 5′-CTCAGTGTCAACATCTGACAG-3′ | AF007111 | 197 | 2 |
5′-CATATGCTGCTCCTGCTGATC-3′ | ||||
MDM4 | 5′-CACTTTAGCCACTGCTACTAC-3′ | AF007111 | 167/100 | 3 |
5′-GCTCTGAGGTAGGCAGTGTG-3′ | ||||
GAC1 | 5′-AGACGGTTAGTGTGGTTGTGG-3′ | AF030435 | 234 | 1, 4 |
5′-GGGTAATGTTGTAGCTGTGGG-3′ | ||||
GAC1 | 5′-AGATGACGGACCACTGTTTGC-3′ | AF030435 | 300 | 2 |
5′-CCACAACCACACTAACCGTCT-3′ | ||||
RBBP5 | 5′-ACTCCGATTTGAACGTTGTGGC-3′ | X85134 | 285 | 2, 4 |
5′-ATTCCTGCATAGGTTCAGGCTC-3′ | ||||
REN | 5′-GACAAACCTGAGCCTCTGTC-3′ | M26900 | 215 | 1, 4 |
5′-GTGGGTCAGGAGAGGCCTG-3′ | ||||
REN | 5′-GTCGTCTTTGACACTGGTTC-3′ | M26900 | 191 | 2 |
5′-GTGATGATGTCCTGGCTGAG-3′ | ||||
ELK4 | 5′-GAATCCAGCCGAGAAACTGGC-3′ | NM_001973 | 335 | 1 |
5′-CTGTGTCGATGTCTGGGTGAG-3′ | ||||
ELF3 | 5′-GACTGGTGCTTCTCGGAGCTC-3′ | AF016295 | 178 | 1 |
5′-GTCCCAGTACTCTTTGCTCAGC-3′ | ||||
PTPN7 | 5′-CACAGTTGCACCTTCTGTAG-3′ | X53364 | 183 | 1 |
5′-GAGTCTGAAGAATTGACAGTG-3′ | ||||
DES | 5′-TTCGATGTACATTTCCATCA-3′ | M63391 | 170 | 1 |
5′-ACAACTAACTGAAAAGAGAG-3′ | ||||
GAPDH | 5′-AACGTGTCAGTGGTGGACCTG-3′ | J04038 | 160 | 1 |
5′-AGTGGGTGTCGCTGTTGAAGT-3′ | ||||
IFNG | 5′-GCAGAGCCAAATTGTCTCC-3′ | J00219 | 94 | 1 |
5′-GTCTTCCTTGATGGTCTCC-3′ | ||||
B2MG | 5′-ACCCCCACTGAAAAAGATGA-3′ | AB021288 | 114 | 2 |
5′-ATCTTCAAACCTCCATGATG-3′ |
1, amplification analysis by duplex PCR; 2, expression analysis by duplex reverse transcription-PCR; 3, demonstration of the MDM4-S splice variant by reverse transcription-PCR; 4, probe generation for hybridization of Southern and Northern blots.
Acknowledgments
We thank Dr. M. C. Sabel for kindly reviewing the clinical files of the patients whose gliomas showed amplification of genes from 1q32.