Cancer Research Grants Frontiers in Basic Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hartmann, C.
Right arrow Articles by Louis, D. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hartmann, C.
Right arrow Articles by Louis, D. N.
[Cancer Research 62, 4100-4108, July 15, 2002]
© 2002 American Association for Cancer Research

Molecular Biology and Genetics

Transcript Map of the 3.7-Mb D19S112–D19S246 Candidate Tumor Suppressor Region on the Long Arm of Chromosome 191

Christian Hartmann, Loki Johnk, Gaspar Kitange, Yanhong Wu, Linda K. Ashworth, Robert B. Jenkins and David N. Louis2

Molecular Neuro-Oncology Laboratory, Department of Pathology and Neurosurgical Service, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02129 [C. H., L. J., D. N. L.]; Division of Laboratory Genetics, Mayo Clinic and Foundation, Rochester, Minnesota [G. K., Y. W., R. B. J.]; and Lawrence Livermore National Laboratory, Livermore, California and Department of Energy Joint Genome Institute, Walnut Creek, California [L. K. A.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Allelic losses of the q13.3 region of chromosome 19 have been documented in malignant gliomas, neuroblastomas, and ovarian carcinomas, strongly suggesting the presence of a 19q13.3 tumor suppressor gene. Deletion mapping in tumors over the past decade has narrowed the candidate region considerably but has produced partially conflicting results, with some small candidate regions defined only by isolated tumors with deletions. Mutation and expression screening of genes from the most likely candidate regions has failed to identify the gene of interest, perhaps because of the conflicting deletion mapping data. The recently increased public availability of human genomic sequence, combined with improved bioinformatics capabilities, has now made it possible to map much larger candidate regions in considerable detail. We have manually generated a transcript map that spans most of the 19q13.3 tumor suppressor candidate region, from D19S219 to D19S246, with a resolution and quality superior to that of computer-generated maps. These results are presented in the hope that an improved map of the candidate region will facilitate further widespread screening and eventual identification of the gene or genes deleted in human gliomas, neuroblastomas, and ovarian cancers.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Allelic losses of the long arm of chromosome 19 in a variety of human malignancies strongly suggest that 19q harbors a tumor suppressor gene. Loss of 19q is found commonly in all three major types of diffuse human malignant gliomas (1, 2, 3, 4, 5) as well as in epithelial ovarian cancers and neuroblastomas (6, 7, 8) . Indeed, this region is already of clinical importance, because 19q loss in combination with 1p loss is associated with a greater likelihood of chemosensitivity, more durable chemotherapeutic responses, and longer overall survival in patients with anaplastic oligodendrogliomas (9, 10, 11) . There has, therefore, been considerable interest in identifying the putative tumor suppressor gene from this candidate region.

Deletion mapping studies of malignant gliomas have gradually narrowed the candidate region over the past decade (Fig. 1)Citation . A common region of loss in astrocytomas was initially found between 19q13.11 and 19q13.4 (12 , 13) and was then reduced to a 9.5-Mb area between D19S178 on 19q.13.2 and D19S180 on 19q13.4 (14) and to a 4-Mb region on 19q13.3 between APOC1 and HRC (15) . Subsequently identified 1.4-Mb deletion areas between D19S412 and STD (16) and between D19S241E and D19S596 (17) defined a minimal common deletion region of 150 kb (17) . Sequencing of this 150-kb candidate region revealed GLTSCR1, EDH2, GLTSCR2, and SW, but no mutations or expression abnormalities were found in these genes (18 , 19) . Recent deletion mapping in neuroblastomas has demonstrated a commonly deleted region between D19S412 and D19S606, overlapping the larger glioma candidate regions but excluding the 150-kb glioma candidate region (7) . On the other hand, one glioma has been reported with an interstitial loss of a 425-kb region between D19S219 and D19S112 at 19q13.3 (20) , centromeric to the probable glioma/neuroblastoma locus. Conversely, epithelial ovarian cancer deletions apparently overlap more telomerically on 19q13.3, between D19S246 and the KLK gene family (8) . There is thus controversy in assigning the most likely location for the 19q tumor suppressor, with some candidate regions based only on a small number of deletions in genetically unstable tumors, and it remains possible that the region contains more than one tumor suppressor locus. These difficulties have thwarted identification of the gene, or genes, of interest.



View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Prior mapping studies demonstrating partially overlapping common deletion areas (between D19S241E and STD) on 19q13.3, spanning 5.5 Mb; 1, Rubio et al. (15) ; 2, Yong et al. (20) ; 3, Rosenberg et al. (16) ; 4, Smith et al. (17) ; 5, Mora et al. (7) in neuroblastoma; 6, Bicher et al. (8) in ovarian cancer.

 
Although traditional gene-hunting strategies required mapping a region of interest to a few hundred kilobases, the public availability of extensive draft sequences from the human genome and of computer bioinformatics programs has made it possible to map much larger candidate regions. This event has in turn raised the possibility that a larger transcript map could be probed more effectively for the 19q gene than smaller maps following more traditional approaches. Here we present a manually generated transcript map that spans most of the 19q13.3 tumor suppressor candidate regions, from D19S219 to D19S246, with a resolution superior to that of currently available computer-based maps. The map covers a candidate region defined by many tumor breakpoints, reducing the likelihood that genomic instability in individual tumors could direct gene identification strategies to an incorrect locus. This large transcript map thus lays the groundwork for the prioritization of 19q tumor suppressor gene candidates and the determination of their involvement in these important human malignancies by further high-throughput screening studies.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Physical Map.
An approximately 3.7-Mb physical map spanning the region from D19S112 to D19S246 was generated using 28 BACs3 and 19 cosmids/fosmids sequenced by LLNL (Table 1)Citation . Publicly available scaffolds GA_x2KMHMR58TM, GA_x2KMHMR58UK, GA_x2KMHMR58JG, GA_x2KMHMR58W5, GA_x2KMHMR58UK, GA_x2KMHMR592L, GA_x2KMHMR58HM, GA_x8WDQ41, GA_x8WDQ6N, GA_x8WDQ6T, and GA_x8WDQ7U from Celera (Celera Genomics, Rockville, MD) were then applied to control for sequence ordering and quality and also to fill gaps. Results from an earlier shotgun sequencing project of the minimal deletion region based on BAC 284K17 were also included (18) . BAC and cosmid/fosmid order was also accomplished using the chromosome 19q arm metric physical map from LLNL.4 Discrepancies were resolved within the human genome BAC clone map, based on shared restriction fragments from the Genome Sequence Center of Washington University School of Medicine, St. Louis.5 Sequence assembly was accomplished with SeqMan II 5.0 (DNAstar, Madison, WI) and Staden-Package 2000.0.6


View this table:
[in this window]
[in a new window]

  		Table 1 List of all BACs, fosmids, and cosmids between D19S219 and D19S246 sequenced by LLNL in centromeric to telomeric order

 
Identification of Genes, CpG Islands, and Promoter Regions.
Genes were identified by two different strategies. Nucleotide similarity searches with BLAST 2.0 were performed using the nr (restricted to Homo sapiens), hEST, and mEST (parameters: expect = 0.0001, word size = 11, filters = low complexity and for nr also human repeats) databases. Database queries were executed locally and on National Center for Biotechnology Information Blast servers. Results were assessed manually to test the logical stringency of gene candidates, using criteria such as full-length completeness, multiple appearance of sequence motifs, and correct splicing sites. Genomic sequences were also analyzed by the gene prediction programs GrailEXP 2.07 and Genscan.8 If only hESTs and predicted exons were available, results from both strategies were manually evaluated to test the importance of possible gene candidates. CpG islands were identified by GrailEXP, CpG Islands9 and WebGene10 and rejected if a distance of >7 kb separated the island from exon 1 of a gene candidate. The value of predicted CpG islands was further evaluated by overlapping the region with predicted promoter sites generated by the promoter prediction programs Neural Network Promoter Prediction,11 Promoter 2.0 Prediction Server,12 WWW Signal Scan,13 and Transfac.14

Further Characterization of Genes.
Expression patterns were tested using SAGE,15 with the mRNA sequence of each gene analyzed according to SAGE tags, to determine whether there was higher expression in normal brain than in glioma or higher expression in low-grade than in high-grade glioma. Possible functions of the gene products were determined by literature review and mRNA sequence analysis. To this end, the mRNA sequence was translated by Gene Runner 3.05 (Hastings Software, Hudson, NY) in all three possible reading frames, blasted against protein-nr, and analyzed for conserved protein motifs by RPS-blast using the Conserved Domain Database based on Pfam and Smart. Because genes that regulate key cellular functions, including tumor suppressor genes, are often highly conserved during evolution, the 107 genes were compared among different species using protein BLAST and by BLAST comparison of the genomic sequence against the mouse EST database.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Physical Map.
Three sequence contigs were assembled that cover the entire 3.7-Mb region (Fig. 2)Citation : a centromeric contig of 1852 kb (Fig. 2a)Citation , a small middle contig of 108 kb (Fig. 2b)Citation , and a telomeric contig of 1791 kb (Fig. 2c)Citation . The size of the two remaining gaps is unknown, but each is likely to be <10 kb, given the few and small remaining sequences from the corresponding draft version BAC sequence files. Seven gaps of known size (average size, 8 kb) remain in the centromeric contig and 6 gaps in the telomeric contig, but it is unlikely that these gaps contain new genes; the surrounding regions contain no common gene characteristics (incomplete cDNA alignments, incomplete exons, or CpG islands), and the small size of the gaps would only allow single-exon genes. Comparison of publicly available sequences with the previously mapped 150-kb minimal deletion area (17) and with sequence material from an earlier shotgun sequencing project (18) revealed no significant differences. Most of the markers defining the margins of the common regions of overlap were identified on the three contigs; the microsatellite marker D19S412 was found at position 1,067,000 in the centromeric contig, and the STD gene was identified near position 450,000 of the telomeric contig (Fig. 1Citation , number 3; Ref. 16 ). The microsatellite marker D19S241E was seen at position 1,833,000 in the centromeric contig, D19S596 at position 1,275,000 of the telomeric contig, and the 150-kb minimal deletion area aligned between positions 248,000 and 359,000 in the telomeric contig (Fig. 1Citation , number 4; Ref. 17 ). The recently mapped common deletion region in neuroblastomas was found between position 1,067,000 (D19S412) of the centromeric contig and 42,000 (D19S606) of the telomeric contig (Fig. 1Citation , number 5; Ref. 7 ). All three common regions of overlap span the middle contig and the two gaps of unknown size. The microsatellite marker D19S219 was identified at position 51,000 of the centromeric contig, the marker D19S112 at 437,000 on the same contig (20) . Interestingly, when compared with the published data, the two markers D19S246 (position 1,668,000 on the telomeric contig) and HRC (position 1,680,000 on the telomeric contig) were found in reverse order and much closer to each other (Fig. 1Citation , numbers 1 and 6) than reported previously (8 , 15) .



View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Gene map between D19S219 and D19S246, spanning 3.7 Mb. a, centromeric contig (CC, 1852 kb). b, middle contig (MC, 107 kb). c, telomeric contig (TC, 1791 kb). Arrows behind gene names, position of gene on the forward or reverse DNA strand; italic gene names, genes tested for mutation in earlier studies; lines pointing to a genomic spot, the central position of the corresponding gene. The minimal deletion region is based on Smith et al. (17) .

 
Transcript Map.
BLAST and gene prediction programs identified 107 possible genes in the three contigs, some with multiple copies and 85 (79%) with CpG islands close to exon 1. Fig. 2Citation presents the basic features of the map, and some gene details are given in Table 2Citation . The following section highlights features of some of the genes on each of the three contigs.


View this table:
[in this window]
[in a new window]

  		Table 2 List of all genes between D19S219 and D19S246

 
Forty-three genes mapped to the centromeric contig, some of particular interest as candidate genes. Between D19S219 and D19S112 is the myotonic dystrophy region containing the DMPK and DMWD genes. Approximately 80 kb telomeric to D19S112 lies the NOVA2 gene encoding an RNA-binding protein (21) . The next telomeric gene, PGLYRP, has similarities with to the tumor necrosis family superfamily, whereas the newly identified gene TGIF3 contains a HOX homeobox domain. The mRNA sequence DKFZp434J0226 was seen at two locations on the same genomic DNA strain, enclosing the gene RPL12. The second copy of DKFZp434J0226 exhibited no first exon. In both cases, the sequence is not translatable, but matching hESTs can be found; therefore, it remains unclear whether this is a processed pseudogene or whether noncoding RNA is the ultimate product. More than 900 kb centromeric to D19S412, the previously cloned and analyzed candidate gene PPP5C (22) appeared on the centromeric contig. The 7-exon putative tumor suppressor gene Rho-GAP p190-A (23 , 24) was located 300 kb centromeric to the common deletion area. The 7-exon gene KIAA1722 also maps to the same genomic sequence, suggesting that it is a larger splicing variant of p190-A. Although there are no significant differences in the translated amino acid sequence, the noncoding exons 6 and 7 differ greatly in size between the two transcripts. The p53 up-regulated modulator of apoptosis (PUMA)/Bcl2-binding component 3 (bbc3) gene and SUMO-1 activating factor (SAE1) genes were located 40 and 100 kb centromeric to D19S241E, respectively, and were covered by three common deletion regions.

Four genes were identified in the small middle contig. The centromeric/telomeric orientation of this contig was based on assembling multiple BACs, with BAC position determined by the LLNL BAC map. Although it was possible to assign C5aR, C5L2, and KIAA0134 completely to the middle contig, positions 1 to 373 of DKFZp547H236 were not identified in the sequence. Presumably, the first exon or exons of the gene are located within the gap between the middle and the telomeric contig, because the subsequent exon of DKFZp547H236 aligns very close to the telomeric border of the contig.

The telomeric contig contained 61 genes. Although most chromosomes have a gene density of 7–15 genes/Mb (chromosome 19, 20 genes/Mb; Ref. 25 ), the telomeric contig has nearly 41 genes/Mb. Within the minimal deletion area, no new genes were identified, aside from the already cloned and analyzed GLTSCR1, EDH2, GLTSCR2, and SW (18) . Interestingly, two mRNA sequences, FLJ00085 and DKFZp434C031, are splicing variants of GLTSCR2. Although GLTSCR2 has an mRNA length of 1516 bp, FLJ00085 has a length of 4186 bp, covering all exons and introns 6, 7, and 10 of GLTSCR2. Alternatively, the 1762-bp DKFZp434C031 starts with exon 6 of GLTSCR2. Exons 5 and 8 of DKFZp434C031 are longer than the corresponding exons 10 and 13 of GLTSCR2, and the last two exons of DKFZp434C031 are not part of the GLTSCR2 sequence. Only partial hESTs were found that align to these alternate sequences. Another 1468-bp sequence, which is 96% identical to GLTSCR2, was described as P60, coding for a protein that interacts with the herpes simplex virus 1 regulatory proteins ICP22 and ICP0 (26) . No coding sequences were identified within 120 kb centromeric to the 150-kb minimal common deletion area. Only isolated hESTs and single predicted exons were seen, suggesting an absence of genes in this region. However, because of a 10-kb gap within this area, there is still the possibility of an unidentified gene. On the other hand, 9 kb telomeric to the minimal deletion area, a 496-bp putative single-exon gene was found with an open reading frame of 115 amino acids, showing homologies to the ribosomal protein L23a. Near the UniGene cluster Hs.104894, recently described as epididymal sperm binding protein (ELSPBP1; Ref. 27 ), the gene CaBP5 was localized. However, there was no evidence for exons 4 to 6 of CaBP3, described as being on the opposite strand and sharing the reverse sequence of exons 1 to 3 of CaBP5 (28) .

The 10-exon gene CARD8/KIAA0955 was found 400 kb telomeric to the 150-kb minimal common deletion area. The genes NDPP1 and DACAR also align to the CARD8 sequence; however, they cover only exon 3 to the beginning of exon 10. By assembling the corresponding UniGene cluster Hs.10031 and aligning the resulting consensus sequence, an additional 160-bp exon "0" was identified 6.2 kb upstream to exon 1. The UniGene cluster Hs.310472 (internal name JAZ-1) aligns to the genomic sequence near CARD8 and contains a KRAB box in the translated protein, thereby identifying it as another member of the zinc finger family. Another promising candidate within the same region is the glutamate-rich WD repeat protein (GRWD), with 7 exons and a length of 2240 bp. The putative apoptosis-associated tyrosine kinase 2 (AATK2) gene was found close to GRWD. This gene was assembled based on the cDNA KIAA1883, UniGene cluster Hs.281328, other hESTs and GrailEXP-generated exon candidates and appears to code for a tyrosine kinase that is similar to AATK (or AATYK). The phosphoinositol 3-phosphate binding protein-1 (PEPP1) gene was found telomeric to D19S596. This gene, which contains a pleckstrin homology domain and codes for a signaling protein of unknown function, was initially described as being expressed only in melanocytes and melanoma cells (29) . However, the SAGE tags were also found in different gliomas, indicating a higher expression in low-grade than high-grade tumors. The two apoptosis-regulating genes, GADD34 and BAX, which are within close proximity to each other, were identified 130 and 220 kb telomeric to D19S596, respectively. Because early studies mapped the candidate gene BAX to 19q13.3, it was one of the first genes screened for mutations (30) . An interesting cluster of genes was observed 100 kb centromeric to D19S246; 8 copies of HCG-ß and 3 copies of NT6 were present on both strands, falling within RUVBL2 and NT4/5. The 4716-bp cDNA KIAA0654 aligned with D19S246 in the telomeric contig. FLJ22319 (3886 bp) and FLJ13788 (1616 bp) are two other cDNAs that are identical to KIAA0654. The 1440-bp mRNA of Liprin-alpha 3 aligns to position 2586-4110 of the KIAA0654 sequence, indicating splicing variants of this gene. Whereas earlier mapping studies placed the HRC gene centromeric to D19S246 (8 , 15) , the genomic DNA sequence revealed HRC to be 20 kb telomeric to the microsatellite marker.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
On the basis of the draft version of the Human Genome Project, publicly available genomic sequences that cover the common deletion regions between D19S219 and D19S246 were assembled into three contigs and were annotated for genes. Despite success in assembling these large contigs, some gaps remain. Presumably, the two gaps of unknown size are highly repetitive and cannot be sequenced. The 13 gaps (averaging 8 kb in size) in the centromeric contig and telomeric contig sequences are in regions not yet finished by the Joint Genome Institute but are covered by publicly available Celera material. These data were of high enough quality to estimate the size of the remaining gaps, using algorithms provided by the company.

The draft results of the Human Genome Project estimated the number of genes at 30,000–45,000 (25 , 31) ; however, >95,000 UniGene clusters are listed in the database. One explanation for this discrepancy is the presence of splicing variants, resulting in multiple UniGene clusters for one gene. In support of this hypothesis, many possible variants were identified within the annotated 19q13.3 region, and these variants often represented already characterized genes. Evaluation of splicing variants is relatively simple in cases such as KIAA1722/Rho GAP p190-A, DKFZp547H236/human Meis3, or KIAA0654/Lin-7b, because extant hEST material covers all relevant genomic areas, making it easier to verify that these are splicing variants. On the other hand, for genes such as GLTSCR2, for which FLJ00085 and DKFZp434C031 illustrate two alternate splicing variants (that include intronic areas of GLTSCR2), no matching hESTs can be identified. Such splicing variants may have low levels of expression, perhaps limited to specific tissues, or there could be general problems related to single-clone-for-single-gene cDNA sequences. Because of the nature of these cDNAs, the mRNA source may be premature and not completely spliced (32 , 33) . Aligning hESTs from the corresponding UniGene cluster with genomic DNA provides another method to test for possible splicing variants. For instance, using this approach, an additional 160-bp CARD8 exon and a longer (43-bp) variant of PLA2G4C exon 15 were found.

New computer-based strategies thus make it possible to extend the candidate region dramatically, with a higher resolution, and with the identification of multiple putative variant forms. Functional classification, based on in silico comparison of the translated amino acid sequence with well-characterized proteins and known protein domains, predicted new genes involved in apoptosis (CARD8, GRWD, and AATK2), cell adhesion (KIAA0654), developmental regulation (DKFZp547H236), or transcriptional control (DKFZp434I1930, DKFZp434M154, and JAZ-1). Although such an approach can be helpful in reducing the number of candidate genes, it is important to recall that some known tumor suppressors did not, when initially identified, fit into an expected category of tumor-related genes. One such gene was the NF2 gene, which encodes a moezin-, ezrin-, radixin-like protein (34) . Additionally, many genes based on single-clone-for-single-gene methods did not contain any known protein domain or show similarities with characterized genes (e.g., DKFZp564M1019, FLJ10922, and KIAA1183). The putative function of candidate genes is therefore helpful when relationships can be identified with known tumor-related genes, but negative results clearly do not exclude such genes as candidates.

We also examined expression patterns using the UniGene and SAGE databases. Although the evaluation of clustered hESTs from UniGene gives clear information concerning nervous system expression, different potential expression profiles of mutated tumor suppressor genes complicate interpretation of quantitative SAGE-based expression data from brain tumors. For instance, transcriptional levels can be reduced to zero in the presence of deletions, decreased mRNA stability, or mutations affecting the regulation of transcription. Expression can be reduced to 50% with a loss of one allele, whereas the transcriptional level of the remaining mutated allele is not affected or even elevated. For instance, an incorrectly folded protein may be proteolyzed faster, raising transcriptional activity via a positive feedback mechanism. Different expression patterns can result from distinct alterations in a single tumor suppressor gene, depending on type and location of the mutation (e.g., TP53). Moreover, SAGE-based expression data are limited by the number of available libraries, the lack of tumor genotyping, and tissue heterogeneity of normal brain libraries. Custom cDNA microarrays could eliminate these SAGE-based problems and would be a logical next step in using expression data to evaluate tumor suppressor gene candidates in large chromosomal areas.

In the large candidate region, a number of genes involved in apoptosis, cell adhesion, developmental regulation, and cell cycle control are promising candidates. TGIF3 (TALE homeobox TG-interacting factor 3), between D19S219 and D19S412, is a new and highly conserved member of the TGIF protein family. Although the 1562-bp TGIF (18p11.3) and the 1647-bp TGIF2 (20q11.2-12) have 3 exons, TGIF3 has a cDNA length of 608 bp and presumably only 2 exons. TGIF homeodomain transcription factors contribute to Hox-mediated developmental programs (35) , and up-regulation of TGIF2 occurs in ovarian cancer cell lines (36) . Because of the major cDNA sequence similarities to TGIF, SAGE could not determine an expression pattern for TGIF3; both genes contain the same sequence for the SAGE tags. Falling between D19S412 and D19S241E, the conserved protein kinase D2 (PKD2) gene is covered by three common deletion regions and is widely expressed in many tissues, particularly in highly proliferative organs. Although PKD/PKCµ can be cleaved by caspase 3, thereby sensitizing cells to the cytotoxic effects of various genotoxic agents (37) , the highly similar PKD2 gene does not contain the cleavage site. Therefore, PKD2 may serve a different function in apoptosis (38) .

The highly conserved, putative tumor suppressor gene Rho-GAP p190-A maps more centromeric than initially reported (23) . Functionally, p190-A can suppress ras-induced malignancy, and p190-A antisense RNA facilitates malignant transformation (39) . However, SAGE expression data showed a highly variable distribution in the available glioma libraries. The nearby PUMA, or bbc3, gene is closer to D19S241E. This gene, which has a highly similar mouse homologue, has two p53-binding sites in its promoter and when expressed, results in extremely rapid apoptosis by binding to Bcl-2 and Bcl-XL (40 , 41) . SAGE tags were unfortunately only available for four glioblastoma libraries, but in all cases these glioblastomas had low expression levels when compared with other tumors. Another candidate is DKFZp547H236, which is highly homologous to the mouse TALE homeoprotein Meis3. Interestingly, the highly similar family member Meis1 was isolated as a common site of viral integration in a mouse model of acute myeloid leukemia (42) . Overexpression of this gene, along with the presence of Hoxa9, induced acute myeloid leukemia, suggesting an oncogenic function (43) . SAGE data show two groups of expression for this gene and illustrate the utility of the available SAGE information, normal brain and low-grade glioma libraries with high levels of expression versus glioblastomas having 20–50% comparative expression levels.

The cDNA DKFZp434I1930 mapped centromeric to the 150-kb small deletion region, containing a Myb-like DNA-binding domain that was found with RPS-BLAST. Neither brain-derived ESTs nor useful SAGE tags were available, but its location and putative function make it a an attractive candidate, along with CARD8, which maps between STD and D19S596. Two recent binding studies showed partially contradictory results for the protein that harbors a CARD domain. In the first study, the protein was named tumor-up-regulated CARD-containing antagonist of caspase nine (TUCAN). It was found that TUCAN interferes with binding of Apaf1 to procaspase-9 and suppresses caspase activation induced by the Apaf1 activator, cytochrome c, thereby suggesting more an oncogenic function (44) . The authors of the second study named the protein CARDINAL and found no interaction with procaspase-9 or other caspases. Their results indicate a negative regulation of the nuclear factor-{kappa}B pathway in a proinflammatory context (45) . SAGE data indicated a lower expression level in high-grade, compared with low-grade, gliomas. Interestingly, a comparison of mRNA and protein sequence to other species yielded no similarities.

The GRWD gene is another candidate gene found in this region. pBLASTing the protein sequence of this gene indicated a high level of conservation: Mus musculus, Drosophila melanogaster, Caenorhabditis elegans, Saccharomyces cerevisiae, S. pombe, Arabidopsis thaliana, and Gallus gallus all express highly analogous proteins. The same BLAST applied to human sequence yielded similarities to the human retinoblastoma-binding protein 4/p48 (G1-S transition control protein-binding protein IEF-8502), a protein that negatively regulates ras in yeast (46) . Furthermore, the conserved protein domain WD40 (repeated three times in the sequence) is found in many genes involved in apoptosis, transcriptional regulation, and cell cycle control, such as APAF-1, TAFII-100, and CDC4. The corresponding SAGE tag for GRWD can be found in multiple glioma libraries, but the differences in expression do not reveal any significant pattern. The putative apoptosis-associated tyrosine kinase 2 (AATK2) gene (KIAA1883) was found close to GRWD and contains a serine/threonine protein kinase domain. The highly similar AATK (or AATYK) gene on chromosome 17q25.3 is expressed during growth arrest and apoptosis of myeloid cells (47) and induces and promotes neuronal differentiation (48) . Interestingly, AATK2 specific SAGE tags were found in five normal brain libraries and only in one pilocytic astrocytoma. No high-grade glioma library contained a tag, indicating no expression of this gene in malignant gliomas. Ten hESTs derived from brain, fetal brain, kidney, and colon libraries indicated neuronal expression, which was similar to proteins from different species.

The 4716-bp cDNA KIAA0654 overlaps directly with D19S246. The resulting protein contains both an ezrin/radixin/moesin motif, implying similarities with the tumor suppressor gene NF2, and a tyrosine kinase motif. FLJ22319 (3886 bp), FLJ13788 (1616 bp), and Liprin-alpha 3 (mRNA length of 1440 bp) all align to the KIAA0654 sequence, indicating splicing variants of this gene. The putative splicing variant Liprin-alpha 3 lacks the ezrin/radixin/moesin domain. The Liprin family has been described as transmembrane protein-tyrosine phosphatases, involved in regulation of cell-matrix interactions and having similarities to the APC tumor suppressor gene (49) . Interestingly, SAGE data revealed two groups of glioblastomas, differing in expression by a factor of 3 to 4.

This list of genes spans most of the common deletion regions mapped to 19q13.3 and thereby sets the stage for mutation screening as well as evaluations for other mechanisms of gene inactivation. One such possible mechanism relates to the remarkable number of CpG islands on chromosome 19. Most chromosomes have 5–15 islands/Mb (average, 10.5 islands/Mb); however, chromosomes 16, 17, and 22 have 19–22 islands/Mb. Strikingly, chromosome 19 is an extreme outlier, with 43 islands/Mb (25) . This can be only partially explained by the high density of genes on chromosome 19, because the CpG islands/gene ratio still separates chromosome 19 from all others (average, 1.05 CpGs/gene; chromosome 19, 2.25 CpGs/gene). The remarkable density of CpG islands increases the attractiveness of the hypothesis that CpG hypermethylation might be a primary mechanism of silencing a 19q tumor suppressor gene. Curiously, in the oligodendroglioma subtype of malignant glioma, allelic loss typically affects large portions of the long arm of 19q, rather than the smaller, interstitial deletions described above for astrocytic malignant gliomas (16) . At least 27 genes with tumor-related functions map between D19S219 and D19S246, raising the possibility that allelic losses of a large portion of 19q might be related to either combined silencing of multiple genes by hypermethylation or haploinsufficiency for a combination of these genes (50) . Methods to evaluate these possibilities, however, lag behind the well-defined approaches to mutation screening. Therefore, although the results of the Human Genome Project have been exceedingly helpful in mapping tumor-related regions and in identifying candidate genes, similar advances are necessary to facilitate the next step of evaluating such genes on a large scale. In this light, the present detailed map is reported in the hope that it will lay the foundations for additional physical mapping and the eventual identification of the gene that is lost so commonly in human gliomas, neuroblastomas, and ovarian cancers.


View this table:
[in this window]
[in a new window]

  		Table 2A Continued

 

   FOOTNOTES
 
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.

1 Supported by NIH Grants CA69285 (to D. N. L.) and CA85799 (to R. B. J.). Back

2 To whom requests for reprints should be addressed, at Molecular Neuro-Oncology Laboratory, CNY6, Massachusetts General Hospital, 149 Thirteenth Street, Charlestown, MA 02129. Phone: (617) 726-5510; Fax: (617) 726-5079; E-mail: louis{at}helix.mgh.harvard.edu Back

3 The abbreviations used are: BAC, bacteria artificial chromosome; LLNL, Lawrence Livermore National Laboratory; hEST, human expressed sequence tag. Back

4 Internet address: http://greengenes.llnl.gov/genome-bin/loadmap?region = mq. Back

5 Internet address: http://genome.wustl.edu/gsc/human/Mapping/index.shtml. Back

6 Internet address: http://www.mrc-lmb.cam.ac.uk/pubseq/staden_home.html. Back

7 Internet address: http://grail.lsd.ornl.gov/grailexp. Back

8 Internet address: http://genes.mit.edu/GENSCAN.html. Back

9 Internet address: http://www.ebi.ac.uk/cpg. Back

10 Internet address: http://www.itba.mi.cnr.it/cgi-bin/webgene/webgene_launcher. Back

11 Internet address: http://www.fruitfly.org/seq_tools/promoter.html. Back

12 Internet address: http://www.cbs.dtu.dk/services/promoter. Back

13 Internet address: http://bimas.dcrt.nih.gov/molbio/signal. Back

14 Internet address: http://transfac.gbf.de/TRANSFAC/index.html. Back

15 Internet address: http://www.ncbi.nlm.nih.gov/SAGE. Back

Received 3/ 1/02. Accepted 5/ 9/02.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. von Deimling A., Bender B., Jahnke R., Waha A., Kraus J., Albrecht S., Wellenreuther R., Fassbender F., Nagel J., Menon A. G., et al Loci associated with malignant progression in astrocytomas: a candidate on chromosome 19q. Cancer Res., 54: 1397-1401, 1994.[Abstract/Free Full Text]
  2. Reifenberger J., Reifenberger G., Liu L., James C. D., Wechsler W., Collins V. P. Molecular genetic analysis of oligodendroglial tumors shows preferential allelic deletions on 19q and 1p. Am. J. Pathol., 145: 1175-1190, 1994.[Abstract]
  3. Kraus J. A., Koopmann J., Kaskel P., Maintz D., Brandner S., Schramm J., Louis D. N., Wiestler O. D., von Deimling A. Shared allelic losses on chromosomes 1p and 19q suggest a common origin of oligodendroglioma and oligoastrocytoma. J. Neuropathol. Exp. Neurol., 54: 91-95, 1995.[Medline]
  4. Bello M. J., Leone P. E., Vaquero J., de Campos J. M., Kusak M. E., Sarasa J. L., Pestana A., Rey J. A. Allelic loss at 1p and 19q frequently occurs in association and may represent early oncogenic events in oligodendroglial tumors. Int. J. Cancer, 64: 207-210, 1995.[Medline]
  5. Nakamura M., Yang F., Fujisawa H., Yonekawa Y., Kleihues P., Ohgaki H. Loss of heterozygosity on chromosome 19 in secondary glioblastomas. J. Neuropathol. Exp. Neurol., 59: 539-543, 2000.[Medline]
  6. Amfo K., Neyns B., Teugels E., Lissens W., Bourgain C., De Sutter P., Vandamme B., Vamos E., De Greve J. Frequent deletion of chromosome 19 and a rare rearrangement of 19p13.3 involving the insulin receptor gene in human ovarian cancer. Oncogene, 11: 351-358, 1995.[Medline]
  7. Mora J., Cheung N. K., Chen L., Qin J., Gerald W. Loss of heterozygosity at 19q13.3 is associated with locally aggressive neuroblastoma. Clin. Cancer Res., 7: 1358-1361, 2001.[Abstract/Free Full Text]
  8. Bicher A., Ault K., Kimmelman A., Gershenson D., Reed E., Liang B. Loss of heterozygosity in human ovarian cancer on chromosome 19q. Gynecol. Oncol., 66: 36-40, 1997.[Medline]
  9. Cairncross J. G., Ueki K., Zlatescu M. C., Lisle D. K., Finkelstein D. M., Hammond R. R., Silver J. S., Stark P. C., MacDonald D. R., Ino Y., Ramsay D. A., Louis D. N. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J. Natl. Cancer Inst., 90: 1473-1479, 1998.[Abstract/Free Full Text]
  10. Smith J. S., Perry A., Borell T. J., Lee H. K., O’Fallon J., Hosek S. M., Kimmel D., Yates A., Burger P. C., Scheithauer B. W., Jenkins R. B. Alterations of chromosome arms 1p and 19q as predictors of survival in oligodendrogliomas, astrocytomas, and mixed oligoastrocytomas. J. Clin. Oncol., 18: 636-645, 2000.[Abstract/Free Full Text]
  11. Ino Y., Betensky R. A., Zlatescu M. C., Sasaki H., Macdonald D. R., Stemmer-Rachamimov A. O., Ramsay D. A., Cairncross J. G., Louis D. N. Molecular subtypes of anaplastic oligodendroglioma: implications for patient management at diagnosis. Clin. Cancer Res., 7: 839-845, 2001.[Abstract/Free Full Text]
  12. von Deimling A., Louis D. N., von Ammon K., Petersen I., Wiestler O. D., Seizinger B. R. Evidence for a tumor suppressor gene on chromosome 19q associated with human astrocytomas, oligodendrogliomas, and mixed gliomas. Cancer Res., 52: 4277-4279, 1992.[Abstract/Free Full Text]
  13. Ransom D. T., Ritland S. R., Kimmel D. W., Moertel C. A., Dahl R. J., Scheithauer B. W., Kelly P. J., Jenkins R. B. Cytogenetic and loss of heterozygosity studies in ependymomas, pilocytic astrocytomas, and oligodendrogliomas. Genes Chromosomes Cancer, 5: 348-356, 1992.[Medline]
  14. von Deimling A., Nagel J., Bender B., Lenartz D., Schramm J., Louis D. N., Wiestler O. D. Deletion mapping of chromosome 19 in human gliomas. Int. J. Cancer, 57: 676-680, 1994.[Medline]
  15. Rubio M. P., Correa K. M., Ueki K., Mohrenweiser H. W., Gusella J. F., von Deimling A., Louis D. N. The putative glioma tumor suppressor gene on chromosome 19q maps between APOC2 and HRC. Cancer Res., 54: 4760-4763, 1994.[Abstract/Free Full Text]
  16. Rosenberg J. E., Lisle D. K., Burwick J. A., Ueki K., von Deimling A., Mohrenweiser H. W., Louis D. N. Refined deletion mapping of the chromosome 19q glioma tumor suppressor gene to the D19S412–STD interval. Oncogene, 13: 2483-2485, 1996.[Medline]
  17. Smith J. S., Tachibana I., Lee H. K., Qian J., Pohl U., Mohrenweiser H. W., Borell T. J., Hosek S. M., Soderberg C. L., von Deimling A., Perry A., Scheithauer B. W., Louis D. N., Jenkins R. B. Mapping of the chromosome 19q-arm glioma tumor suppressor gene using fluorescence in situ hybridization and novel microsatellite markers. Genes Chromosomes Cancer, 29: 16-25, 2000.[Medline]
  18. Smith J. S., Tachibana I., Pohl U., Lee H. K., Thanarajasingam U., Portier B. P., Ueki K., Ramaswamy S., Billings S. J., Mohrenweiser H. W., Louis D. N., Jenkins R. B. A transcript map of the chromosome 19q-arm glioma tumor suppressor region. Genomics, 64: 44-50, 2000.[Medline]
  19. Pohl U., Smith J. S., Tachibana I., Ueki K., Lee H. K., Ramaswamy S., Wu Q., Mohrenweiser H. W., Jenkins R. B., Louis D. N. EHD2, EHD3, and EHD4 encode novel members of a highly conserved family of EH domain-containing proteins. Genomics, 63: 255-262, 2000.[Medline]
  20. Yong W. H., Chou D., Ueki K., Harsh G. R. T., von Deimling A., Gusella J. F., Mohrenweiser H. W., Louis D. N. Chromosome 19q deletions in human gliomas overlap telomeric to D19S219 and may target a 425 kb region centromeric to D19S112. J. Neuropathol. Exp. Neurol., 54: 622-626, 1995.[Medline]
  21. Ueki K., Ramaswamy S., Billings S. J., Mohrenweiser H. W., Louis D. N. ANOVA, a putative astrocytic RNA-binding protein gene that maps to chromosome 19q13.3. Neurogenetics, 1: 31-36, 1997.[Medline]
  22. Yong W. H., Ueki K., Chou D., Reeves S. A., von Deimling A., Gusella J. F., Mohrenweiser H. W., Buckler A. J., Louis D. N. Cloning of a highly conserved human protein serine-threonine phosphatase gene from the glioma candidate region on chromosome 19q13.3. Genomics, 29: 533-536, 1995.[Medline]
  23. Tikoo A., Czekay S., Viars C., White S., Heath J. K., Arden K., Maruta H. p190-A, a human tumor suppressor gene, maps to the chromosomal region 19q13.3 that is reportedly deleted in some gliomas. Gene (Amst.), 257: 23-31, 2000.[Medline]
  24. Settleman J., Narasimhan V., Foster L. C., Weinberg R. A. Molecular cloning of cDNAs encoding the GAP-associated protein p190: implications for a signaling pathway from ras to the nucleus. Cell, 69: 539-549, 1992.[Medline]
  25. Lander E. S., Linton L. M., Birren B., Nusbaum C., Zody M. C., Baldwin J., Devon K., Dewar K., Doyle M., FitzHugh W., et al Initial sequencing and analysis of the human genome. Nature (Lond.), 409: 860-921, 2001.[Medline]
  26. Bruni R., Fineschi B., Ogle W. O., Roizman B. A novel cellular protein, p60, interacting with both herpes simplex virus 1 regulatory proteins ICP22 and ICP0 is modified in a cell-type-specific manner and is recruited to the nucleus after infection. J. Virol., 73: 3810-3817, 1999.[Abstract/Free Full Text]
  27. Saalmann A., Munz S., Ellerbrock K., Ivell R., Kirchhoff C. Novel sperm-binding proteins of epididymal origin contain four fibronectin type II-modules. Mol. Reprod. Dev., 58: 88-100, 2001.[Medline]
  28. Haeseleer F., Sokal I., Verlinde C. L., Erdjument-Bromage H., Tempst P., Pronin A. N., Benovic J. L., Fariss R. N., Palczewski K. Five members of a novel Ca(2+)-binding protein (CABP) subfamily with similarity to calmodulin. J. Biol. Chem., 275: 1247-1260, 2000.[Abstract/Free Full Text]
  29. Dowler S., Currie R. A., Campbell D. G., Deak M., Kular G., Downes C. P., Alessi D. R. Identification of pleckstrin-homology-domain-containing proteins with novel phosphoinositide-binding specificities. Biochem. J., 351: 19-31, 2000.[Medline]
  30. Chou D., Miyashita T., Mohrenweiser H. W., Ueki K., Kastury K., Druck T., von Deimling A., Huebner K., Reed J. C., Louis D. N. The BAX gene maps to the glioma candidate region at 19q13.3, but is not altered in human gliomas. Cancer Genet. Cytogenet., 88: 136-140, 1996.[Medline]
  31. Venter J. C., Adams M. D., Myers E. W., Li P. W., Mural R. J., Sutton G. G., Smith H. O., Yandell M., Evans C. A., Holt R. A., et al The sequence of the human genome. Science (Wash. DC), 291: 1304-1351, 2001.[Abstract/Free Full Text]
  32. Wolfsberg T. G., Landsman D. A comparison of expressed sequence tags (ESTs) to human genomic sequences. Nucleic Acids Res., 25: 1626-1632, 1997.[Abstract/Free Full Text]
  33. Hillier L. D., Lennon G., Becker M., Bonaldo M. F., Chiapelli B., Chissoe S., Dietrich N., DuBuque T., Favello A., Gish W., Hawkins M., Hultman M., Kucaba T., Lacy M., Le M., Le N., Mardis E., Moore B., Morris M., Parsons J., Prange C., Rifkin L., Rohlfing T., Schellenberg K., Marra M., et al Generation and analysis of 280,000 human expressed sequence tags. Genome Res., 6: 807-828, 1996.[Abstract/Free Full Text]
  34. Trofatter J. A., MacCollin M. M., Rutter J. L., Murrell J. R., Duyao M. P., Parry D. M., Eldridge R., Kley N., Menon A. G., Pulaski K., et al A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell, 72: 791-800, 1993.[Medline]
  35. Yang Y., Hwang C. K., D’Souza U. M., Lee S. H., Junn E., Mouradian M. M. Three-amino acid extension loop homeodomain proteins Meis2 and TGIF differentially regulate transcription. J. Biol. Chem., 275: 20734-20741, 2000.[Abstract/Free Full Text]
  36. Imoto I., Pimkhaokham A., Watanabe T., Saito-Ohara F., Soeda E., Inazawa J. Amplification and overexpression of TGIF2, a novel homeobox gene of the TALE superclass, in ovarian cancer cell lines. Biochem. Biophys. Res. Commun., 276: 264-270, 2000.[Medline]
  37. Endo K., Oki E., Biedermann V., Kojima H., Yoshida K., Johannes F. J., Kufe D., Datta R. Proteolytic cleavage and activation of protein kinase Cµ by caspase-3 in the apoptotic response of cells to 1-ß-D-arabinofuranosylcytosine and other genotoxic agents. J. Biol. Chem., 275: 18476-18481, 2000.[Abstract/Free Full Text]
  38. Sturany S., Van Lint J., Muller F., Wilda M., Hameister H., Hocker M., Brey A., Gern U., Vandenheede J., Gress T., Adler G., Seufferlein T. Molecular cloning and characterization of the human protein kinase D2. A novel member of the protein kinase D family of serine threonine kinases. J. Biol. Chem., 276: 3310-3318, 2001.[Abstract/Free Full Text]
  39. Wang D. Z., Nur E. K. M. S., Tikoo A., Montague W., Maruta H. The GTPase and Rho GAP domains of p190, a tumor suppressor protein that binds the Mr 120,000 Ras GAP, independently function as anti-Ras tumor suppressors. Cancer Res., 57: 2478-2484, 1997.[Abstract/Free Full Text]
  40. Nakano K., Wousden K. H. PUMA, a novel proapoptotic gene, is induced by p53. Mol. Cell, 7: 683-694, 2001.[Medline]
  41. Yu J., Zhang L., Hwang P. M., Kinzler K. W., Vogelstein B. PUMA induces the rapid apoptosis of colorectal cancer cells. Mol. Cell, 7: 673-682, 2001.[Medline]
  42. Moskow J. J., Bullrich F., Huebner K., Daar I. O., Buchberg A. M. Meis1, a PBX1-related homeobox gene involved in myeloid leukemia in BXH-2 mice. Mol. Cell. Biol., 15: 5434-5443, 1995.[Abstract]
  43. Kroon E., Krosl J., Thorsteinsdottir U., Baban S., Buchberg A. M., Sauvageau G. Hoxa9 transforms primary bone marrow cells through specific collaboration with Meis1a but not Pbx1b. EMBO J., 17: 3714-3725, 1998.[Medline]
  44. Pathan N., Marusawa H., Krajewska M., Matsuzawa S., Kim H., Okada K., Torii S., Kitada S., Krajewski S., Welsh K., Pio F., Godzik A., Reed J. C. TUCAN, an antiapoptotic caspase-associated recruitment domain family protein overexpressed in cancer. J. Biol. Chem., 276: 32220-32229, 2001.[Abstract/Free Full Text]
  45. Bouchier-Hayes L., Conroy H., Egan H., Adrain C., Creagh E. M., MacFarlane M., Martin S. J. CARDINAL, a novel caspase recruitment domain protein, is an inhibitor of multiple NF-{kappa}B activation pathways. J. Biol. Chem., 76: 44069-44077, 2001.
  46. Qian Y. W., Wang Y. C., Hollingsworth R. E., Jr., Jones D., Ling N., Lee E. Y. A retinoblastoma-binding protein related to a negative regulator of Ras in yeast. Nature (Lond.), 364: 648-652, 1993.[Medline]
  47. Gaozza E., Baker S. J., Vora R. K., Reddy E. P. AATYK: a novel tyrosine kinase induced during growth arrest and apoptosis of myeloid cells. Oncogene, 15: 3127-3135, 1997.[Medline]
  48. Raghunath M., Patti R., Bannerman P., Lee C. M., Baker S., Sutton L. N., Phillips P. C., Damodar Reddy C. A novel kinase, AATYK induces and promotes neuronal differentiation in a human neuroblastoma (SH-SY5Y) cell line. Brain Res. Mol. Brain Res., 77: 151-162, 2000.[Medline]
  49. Serra-Pages C., Medley Q. G., Tang M., Hart A., Streuli M. Liprins, a family of LAR transmembrane protein-tyrosine phosphatase-interacting proteins. J. Biol. Chem., 273: 15611-15620, 1998.[Abstract/Free Full Text]
  50. Devilee P., Cleton-Jansen A., Cornelisse C. J. Ever since Knudson. Trends Genet., 17: 569-573, 2001.[Medline]



This article has been cited by other articles:


Home page
 Mol. Biol. CellHome page
F. Okahara, K. Itoh, A. Nakagawara, M. Murakami, Y. Kanaho, and T. Maehama
Critical Role of PICT-1, a Tumor Suppressor Candidate, in Phosphatidylinositol 3,4,5-Trisphosphate Signals and Tumorigenic Transformation
Mol. Biol. Cell, November 1, 2006; 17(11): 4888 - 4895.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
M. Alaminos, V. Davalos, S. Ropero, F. Setien, M. F. Paz, M. Herranz, M. F. Fraga, J. Mora, N.-K. V. Cheung, W. L. Gerald, et al.
EMP3, a Myelin-Related Gene Located in the Critical 19q13.3 Region, Is Epigenetically Silenced and Exhibits Features of a Candidate Tumor Suppressor in Glioma and Neuroblastoma
Cancer Res., April 1, 2005; 65(7): 2565 - 2571.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hartmann, C.
Right arrow Articles by Louis, D. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hartmann, C.
Right arrow Articles by Louis, D. N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online