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 Supplementary Data
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 Song, S. W.
Right arrow Articles by Zhang, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Song, S. W.
Right arrow Articles by Zhang, W.
[Cancer Research 65, 3562-3567, May 1, 2005]
© 2005 American Association for Cancer Research

Molecular Biology, Pathobiology, and Genetics

Inactivation of the Invasion Inhibitory Gene IIp45 by Alternative Splicing in Gliomas

Sonya W. Song, Gregory N. Fuller, Hong Zheng and Wei Zhang

Department of Pathology and Brain Tumor Center, University of Texas M.D. Anderson Cancer Center, Houston, Texas

Requests for reprints: Wei Zhang, Cancer Genomics Core Lab, Department of Pathology, University of Texas M.D. Anderson Cancer Center, Box 85, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: 713-745-1103; Fax: 713-792-5549; E-mail: wzhang{at}mdanderson.org.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The invasion inhibitory protein 45 (IIp45) we recently identified was underexpressed in glioblastoma multiforme, the most malignant form of glioma. The IIp45 gene is located at chromosome 1p36 where frequent deletions have been reported in various types of tumors, including gliomas, raising the possibility that IIp45 may be a classic tumor suppressor gene that can be inactivated by frequent point mutations. To test this hypothesis, we sequenced the IIp45 gene in 59 diffuse glioma samples of different grades and histologic subtypes and identified a possible point mutation or a rare polymorphism in only one sample (1.7%), suggesting that IIp45 is not a classic tumor suppressor gene such as p53. Instead, reverse transcription-PCR and subsequent sequencing results revealed a tumor-specific IIp45 spliced isoform (IIp45S) in 20 of 59 (34%) gliomas examined, particularly in glioblastoma multiformes, including native tissue samples (15 of 25; 60%) and cell lines (5 of 5; 100%). The alternative splicing event is independent of 1p36 deletion, which is not common in glioblastoma multiforme. The IIp45S transcript was not detected in any of 18 normal organs, including fetal and adult brain. We determined that the IIp45S isoform results from exclusion of IIp45 exon 7 and encodes a variant protein that carries a COOH terminus different from that of IIp45 due to a frame-shift mutation. IIp45S protein was undetectable in glioma tissues, although IIp45S mRNA was prevalent. We found that IIp45S, once translated, is rapidly degraded by an ubiquitin-proteasome mechanism. Thus, the IIp45 gene is inactivated by a tumor-specific alternative splicing that generates an aberrant and unstable IIp45 isoform in infiltrative gliomas.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Gliomas constitute the most common type of primary brain tumor. Although the highest grade of glioma, glioblastoma multiforme, rarely disseminates beyond the central nervous system, it is extremely invasive, killing almost exclusively by local invasion, which renders conventional treatment modalities, such as surgery and irradiation, ineffective in most cases. Thus, there is a pressing need for a better understanding of glioma invasion to develop new therapeutic strategies that block the invasion process. We recently reported the identification of an invasion inhibitory protein gene, IIp45, which inhibits glioma cell migration and invasion both in vitro and in vivo (1), functions that are opposite from its binding partner, insulin-like growth factor binding protein 2 (IGFBP2; ref. 2). In contrast to IGFBP2, which is reactivated and overexpressed in about 80% of highly invasive glioblastoma multiformes (3, 4), the expression of IIp45 is reduced in glioblastoma multiformes (1).

The IIp45 gene is located at chromosome 1p36 (refined in 1p36.22), where frequent deletions have been reported in a wide spectrum of cancers, including oligodendrogliomas (81%; refs. 5, 6), astrocytomas (25%; refs. 5, 6), neuroblastoma (30%; ref. 7), breast cancer (61%; ref. 8), colon and rectum cancer (55%; ref. 8), and prostate cancer (40%; ref. 9). The frequent deletions observed at 1p36 in multiple cancer types suggests that a tumor suppressor gene or genes may be located in the region, which has prompted mutation screening of a few genes located in this region. One of these genes is p73 (1p36.33), a p53 homologue (10, 11). However, extensive mutation studies showed that p73 is rarely mutated in human tumor cells (12–14). Instead, differential expression of splice variants of p73 seems associated with tumor malignancy. The p73 {alpha} variant has been found to predominate in malignant tumors but not in benign tumors or normal tissues (10–12, 15).

Accumulating evidence has shown that tumor-specific alternative splicing is an important mechanism that regulates gene expression, and thereby its protein functions, during cancer development (16). The genes encoding calcitonin and fibronectin were the first genes identified of which aberrant RNA splicing correlates with tumorigenesis (17, 18). In gliomas, the fibroblast growth factor receptor 1 (FGFR-1) gene is aberrantly spliced during progression (19, 20). Exclusion of the {alpha} exon generates the FGFR-1ß isoform, which has a 10-fold higher affinity for its ligands compared with the FGFR-1{alpha} isoform that bears the {alpha} exon. The level of FGFR-1ß was found to significantly correlate with the degree of astrocytic tumor malignancy, suggesting that FGFR-1ß provides tumor cells with enhanced FGFR-1-initiated signaling, which confers a growth advantage.

To determine whether IIp45 is frequently mutated in gliomas and falls under the rubric of a classic tumor suppressor gene, we conducted a mutation screening of 59 glioma samples. We found that IIp45 mutation is rare at best in gliomas. Instead, a tumor-specific and glioma progression-associated alternative splicing of IIp45 was identified that generates an unstable IIp45 mutant protein and thus inactivates IIp45 in gliomas.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Clinical samples, tumor cell lines, and DNA and RNA isolation. Glioma tissues were provided by the M.D. Anderson Cancer Center Brain Tumor Bank. Glioma cell lines (U251, U87, and T98G) and the HEK cell line were obtained from American Type Culture Collection (Rockville, MD). LN229 and D54 were obtained from Dr. Alfred Yung at M.D. Anderson Cancer Center. Genomic DNA was extracted using the Qiagen Genomic DNA Extraction Kit. Total RNA was extracted using Trizol extraction methods (Molecular Research Center, Inc., Cincinnati, OH). Normal total RNA was purchased from Clontech Laboratories, Inc. (Palo Alto, CA).

PCR, sequencing, and single nucleotide polymorphism analysis. The nine exons of IIp45 were amplified from genomic DNA using primers located in the introns ~50 bp away from each exon to ensure that splicing consensus site sequences were included in the amplified fragments (Supplementary Table S1). The promoter region was amplified using primers located separately in exon 1 and in the 5' flanking region (1.014 kb upstream of exon 1). All PCR reactions were done under the conditions of 94°C for 30 seconds, 58°C for 30 seconds, and 72°C for 1 minute for 35 cycles followed by 72°C for 7 minutes. PCR products were purified using the Qiagen MinElute PCR purification Kit (Valencia, CA) and sequenced on an ABI 3700 DNA Analyzer. Single nucleotide polymorphisms (SNP) were identified using Sequencher software version 4.0.5 (Gene Codes Co., Ann Arbor, MI).

Reverse transcription-PCR and splicing analysis. Reverse transcription-PCR (RT-PCR) was done using total RNA from normal tissues, glioma tumor samples, and cell lines as described previously (1). Primers for PCR amplification of cDNA (Supplementary Table S1) were located in the 5' and 3' untranslated regions of IIp45 exons 1 and 9, respectively, producing a product of 1,175 bp including the entire coding region. Alternative splicing events were determined by analyzing the genomic DNA sequence of IIp45 obtained from the human genome database and the cDNA sequence of IIp45 (Genbank accession no. AK024020) against the consensus splicing site (GT to AG).

Cell culture, cloning, and transfection. Cells were maintained in DMEM/F12 with 10% fetal bovine serum in a 5% carbon dioxide humidified incubator as described previously (1). IIp45S was cloned into pcDNA 3.1 vector and transfected into glioma cells by GenePorter Transfection Reagents (Gene Therapy Systems, San Diego, CA).

Antibodies and Western blotting. A rabbit polyclonal anti-IIp45S antibody was generated recognizing an epitope (N-CPRSLGPTSHGRSPED-C) derived from the COOH-terminal region of IIp45S (Bethyl Laboratories, Inc., Montgomery, TX). An anti-IIp45 antibody was made as described previously (1). A monoclonal anti-HA antibody was generously provided by Dr. Xiangwei Wu (Baylor College of Medicine, Houston, TX). Western blotting was done as described previously (1).

RNA and protein stability measurements. To measure RNA stability, the parental LN229 cells or the stable clones expressing IIp45 or IIp45S (4 x 105 cells) were incubated in the presence of RNA synthesis inhibitor actinomycin D (5 µg/mL) for 4 to 16 hours. Total RNA was extracted and analyzed by semiquantitative RT-PCR similarly as described above except that 20 and 30 cycles were applied separately for the stable clones and the parental cells. PCR primers (forward primer 5' GCTGGCGCCTCAGGCCATACC 3' and reverse primer 5' GGAGTCAGTCCTCAGGGC 3') were used to amplify two splicing isoforms of IIp45. For proteasome inhibitor treatment, cells were treated with 20 µmol/L of MG132 (Sigma, St. Louis, MO) for indicated periods of time, and cell extracts were immunoblotted with the anti-IIp45 or anti-IIp45S antibody. To observe the protein degradation, the cells were incubated with 10 µg/mL of CHX for various time points. Cell extracts were then analyzed by immunoblotting with anti-IIp45 or anti-IIp45S antibody.

Immunoprecipitation and ubiquitination assays. Immunoprecipitation was done as described previously (1). For ubiquitination assays, LN-229 cells were transfected with different combinations of plasmids expressing IIp45 (2 µg), IIp45S (2 µg), or HA-ubiquitin (1 µg). Twelve hours after transfection, the cells were treated with 20 µmol/L of MG132 for 12 or 24 hours and lysed in lysis buffer as described previously (1). The IIp45S-ubiqutin complex was immunoprecipitated by the polyclonal anti-IIp45 antibody, analyzed on SDS-PAGE gel electrophoresis, and blotted with a monoclonal anti-HA antibody.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Mutation screening of genomic DNA of IIp45 in glioma samples. We screened genomic DNAs from 59 glioma tissues and 6 cell lines for mutations by directly sequencing exons 1 to 9 plus adjacent intron regions and a part of the promoter region of the IIp45 gene. One possible mutation was found in a glioblastoma multiforme tissue sample at codon 206 in exon 4, resulting in a change from Arg to Trp. However, because the normal tissue from this sample was not available, we cannot exclude the possibility that this may be a rare polymorphism. Even if this instance is considered a true mutation, the overall low frequency of IIp45 mutation (1.7%) suggests that point mutation is not a common mechanism for inactivation of IIp45 in gliomas. We did not detect any mutations in the splicing consensus site sequences. The sequencing project identified six SNPs located in the promoter region {c(–147)t}, exon 2 (N99K), exon 2 (P142S), intron 2 (g1101a), exon 3 (E167K), and intron 6 (a8231g).

Identification of glioma-specific alternative splicing of IIp45. In our initial sequence analysis, we also amplified by PCR and sequenced IIp45 cDNAs from some glioma and normal samples. PCR results showed that many glioma tissues and cell lines generated either two products or a predominantly smaller product, whereas all of the normal tissues produced only the larger form (Fig. 1A). Sequence analysis indicated that the larger band found in both normal and glioma samples corresponds to the full-length cDNA of IIp45 and the smaller band is a spliced form of IIp45 that misses exon 7 (termed IIp45S; Fig. 1B). The excision of IIp45 exon 7 in the tumor samples results in a frame-shift in the open reading frame at the COOH terminus starting at codon 282 and generates an IIp45 isoform (IIp45S) that is different from IIp45 in the COOH-terminal region (Fig. 1C). A SMART program (http://smart.embl-heidelberg.de) domain analysis showed that IIp45, but not IIp45S, has a four-helical up-and-down bundle structure (D1e85a) that is shared among members of the cytochrome superfamily. IIp45S, instead, obtains two zinc finger domains in its COOH termini (Fig. 1C).



View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Alternative splicing of IIp45. A, IIp45 cDNAs in normal tissues, glioma samples, and cell lines were amplified by RT-PCR using equal amounts of total RNA. Two isoforms, IIp45 and IIp45S, were detected in different ratios in different tissues. Only IIp45 was detected in normal tissues. B, cDNA samples representing IIp45, IIp45S, or a mixture of both were sequenced, which revealed that IIp45S was an alternative splicing product. Top, full-length IIp45 cDNA from normal brain tissue; middle, spliced IIp45 isoform; bottom, both isoforms of IIp45 from individual glioma tissues. Arrow, splicing site. C, schematic illustration of the alternative splicing event and the amino acid sequences of the two isoforms at the COOH-terminal region. Exon 7 is spliced out in IIp45S, resulting in a frame-shift and resultant change of the COOH-terminal sequences. D1e85a is a structure domain present in IIp45 but absent in IIp45S. Instead, IIp45S carries two zinc finger domains (zf-HIT and ZnF-C3H1).

 
We thus investigated the presence of the IIp45S transcript in 59 glioma tissues, 6 tumor cell lines, and 18-pooled normal human tissues, including normal fetal and adult brain tissues by RT-PCR using equal amounts of total RNA. The frequency of the IIp45S isoform correlates with glioma grade (Table 1). Sixty percent of glioblastoma multiformes (15 of 25) expressed the IIp45S transcript, among which 20% (5 of 25) expressed only IIp45S but not IIp45. In comparison, only ≤20% (11-20%) of the lower grades of glioma (O, AO, and AA) produced both isoforms, and none of them produced only IIp45S. Thus, we observed an increase in IIp45 alternative splicing during glioma progression. Although the 1p36 location of IIp45 gene prompted us to carry out detailed mutational and splicing analyses, our findings showed that the alternative splicing of IIp45 transcript is not linked to allelic loss of 1p36 because an inverse correlation of IIp45 alternative splicing and 1p36 loss of heterozygosity is seen in A, AO, and glioblastoma multiforme. This is not particularly surprising because many genes showed tumor specific alternative splicing patterns in the absence of allelic loss (17–20).


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

  		Table 1. IIp45S isoform identified in glioma tissues by RT-PCR

 
Absence of IIp45S protein in gliomas. To examine the expression of the IIp45S protein in gliomas, we generated a rabbit polyclonal antibody that recognizes a specific epitope in the COOH-terminal region of IIp45S. Western blotting analysis showed that the antibody specifically recognized IIp45S but not IIp45 when protein lysates from LN229 glioma cells transfected with a pcDNA-IIp45S, pcDNA-IIp45, or an empty vector pcDNA, were examined (Fig. 2A). Furthermore, we noticed that the protein level of IIp45S was lower than that of IIp45 expressed in these cells when using an anti-IIp45 antibody that recognizes an epitope in the NH2-terminal region shared by IIp45 and IIp45S (Fig. 2B). We attempted to establish IIp45S stable clones of the LN229 cells and, consistently, we found that the stable clones expressed a very low level of IIp45S (data not shown). Furthermore, we examined the effect of IIp45S on cell invasion using a matrigel invasion chamber assay described in our previous study (1). IIp45S had less inhibitory effect on cell invasion than IIp45. One possible explanation for this observation could be the lower protein level of IIp45S compared with IIp45 in the stable clones (data not shown).



View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Expression of IIp45S and IIp45 in gliomas. A, an anti-IIp45S antibody was produced that specifically detects the IIp45S protein but not the IIp45 protein expressed in LN229 cells. Total proteins isolated from cells transfected with empty vector or plasmids expressing IIp45 or IIp45S were analyzed in a Western blotting assay. B, reduced protein levels of IIp45S were observed in transfected glioma cells compared with that of IIp45. Three different glioma cell lines (LN229, SNB19, and U251) were transfected with equal amounts of expression vectors for IIp45 and IIp45S. Total proteins from transfected cells were analyzed for IIp45 and IIp45S expression using an antibody that recognizes both isoforms. C, expression of IIp45S and IIp45 in glioblastoma multiformes (GBM). IIp45S and IIp45 mRNAs were detected by RT-PCR in six glioblastoma multiformes (top). The expression of IIp45S and IIp45 proteins in these tumors was examined by Western blotting using an anti-IIp45S or an anti-IIp45 antibody. IIp45S was undetectable in all of the glioblastoma multiformes.

 
With the IIp45S-specific antibody, we determined the expression of IIp45S in primary glioma tissues. Based on the RT-PCR data (Fig. 2C, top), proteins from three IIp45S-dominant and three IIp45-dominant glioblastoma multiforme tissue samples were evaluated by Western blotting. IIp45S was not detected in any of the glioblastoma multiforme samples, whereas IIp45 was detected in the IIp45-dominant glioblastoma multiformes and very weakly in the IIp45S-dominant glioblastoma multiformes (Fig. 2C, bottom). The fact that the IIp45S protein was undetectable, although its mRNA was prevalent suggests that IIp45S is an unstable protein in glioma cells.

Excision of IIp45 exon 7 produces highly unstable spliced product IIp45S. To test the possibility that IIp45S is unstable and degraded by an ubiquitin-proteasome mechanism, the LN229 stable clones expressing IIp45S or IIp45 (as a control) were treated with a protein synthesis inhibitor, CHX, or a ubiquitin-proteasome inhibitor, MG132, which allows observing ubiquitination of a target protein without its degradation. After CHX treatment, IIp45S was quickly degraded within 30 minutes and disappeared after 4 hours. In contrast, MG132 treatment resulted in an increase of IIp45S that reached its maximum level 6 hours post-treatment (Fig. 3A-B). In comparison, IIp45 is more stable and MG132 has significantly less effect on IIp45 accumulation. (Fig. 3A-B).



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Stability of IIp45S and IIp45. A, LN229 stable clones expressing IIp45S or IIp45 were treated with the protein synthesis inhibitor CHX or the proteasome inhibitor MG132, and total proteins isolated from the cells at different time points after treatment were analyzed for IIp45 and IIp45S expression by Western blotting assays. IIp45S rapidly decreased over time after CHX treatment but accumulated after MG132 treatment, whereas IIp45 was relatively stable after these treatments. B, Western blotting results were quantified by densitometry and normalized to the amount of actin. Protein levels in the untreated cells are indicated as 100%. C, mRNA stability of IIp45S and IIp45 in transfected LN229 cells. The same stable clones as in (A) were treated with the RNA synthesis inhibitor actinomycin D (Act. D) and their mRNA was subjected to semiquantitative RT-PCR analysis. mRNA level of IIp45S or IIp45 was relatively stable compared with that of Myc mRNA after treatment. D, parental LN229 cell line was treated as in (C) and the endogenous mRNA for IIp45S or IIp45 was examined as in (C). IIp45S or IIp45 mRNA was not decreased within 4 hours of treatment, whereas Myc mRNA declined rapidly.

 
To further verify that the failure to detect IIp45S in gliomas is not due to faster mRNA decay of IIp45S, we did a semiquantitative RT-PCR using total RNA isolated from IIp45S- or IIp45-expressing cells or their parental cells incubated in the presence of actinomycin D, an RNA synthesis inhibitor. The levels of both IIp45S and IIp45 mRNA, either exogenously or endogenously transcribed, remained stable in the treated cells over time without obvious decay (Fig. 3C-D). Myc mRNA, a positive control, disappeared quickly 4 hours after the same treatment (Fig. 3C-D) as reported previously (21). These experiments provide evidence that the lack of IIp45S protein in glioma cells is a result of protein instability and degradation of IIp45S by the ubiquitin-proteasome pathway.

IIp45S is degraded by the ubiquitin-proteasome pathway. We further verified this notion by directly examining IIp45S ubiquitination. To create a "trackable" ubiquitin on the target proteins, LN229 cells were cotransfected with a HA-tagged ubiquitin expression vector and pcDNA-IIp45S or pcDNA-IIp45, followed by MG132 treatment. The expression of IIp45S, IIp45, or HA-ubiquitin in the transfected cells was confirmed (Fig. 4A). IIp45S or IIp45 was then pulled down with anti-IIp45 antibody and their respective ubiquitination was detected by Western blotting using an anti-HA antibody. The marked ubiquitination of IIp45S, but not IIp45, was detected after MG132 treatment (Fig. 4B) indicating that the ubiquitination degradation pathway does in fact target IIp45S.



View larger version (54K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Ubiquitination of IIp45S. LN229 cells were cotransfected with a HA-Ub expression vector and IIp45S or IIp45 expression plasmid, followed by MG132 treatment. A, expression of IIp45S, IIp45, and HA-Ub in the transfected cells was examined by Western blotting. Decreased IIp45S was observed in cells cotransfected with HA-Ub. Multiple bands of HA-Ub were detected in cells transfected with HA-Ub, indicating not only the expression of HA-Ub but also the presence of cellular ubiquitination of multiple proteins in the cells. Some nonspecific bands were observed in all cells. B, protein lysate was subjected to immunoprecipitation with an anti-IIp45 antibody. Immunoprecipitates were analyzed by immunoblotting with an anti-HA antibody. Marked ubiquitination of IIp45S but not IIp45 was detected.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Genetic mutation is a hallmark feature of classic tumor suppressor genes, which generally have one allele deleted and the second allele mutated, such as in the cases of p53 (22) and PTEN (23), or inactivated by promoter methylation, such as in the case of p16 (24). Increasing evidence, however, suggests that alternative splicing is also an important alternative mechanism for the inactivation of tumor suppressor genes (25–29). Alternative splicing is well documented to increase protein complexity through the production of different splicing variants that encode distinct protein products (16). Emerging findings have shown that alternative splicing can down-regulate gene expression by generating unstable mRNA isoforms or protein products (25–29). Our present study has shown that the IIp45 gene, a negative regulator of tumor invasion, is regulated in such a manner in gliomas.

Recent reports have shown that alternative splicing reduces or inactivates the expression of some tumor suppressor genes, such as p73 (15), NF-1 (25, 26), CP-4 (27), or FHIT (28). Similar to what we have found for IIp45 in the present study, the COOH-terminal region of p73 has also been reported to be altered by alternative splicing, leading to the production of six p73 isoforms ({alpha}, ß, {gamma}, {delta}, {varepsilon}, {zeta}). These isoforms differ in transcriptional activity on p53-responsive promoters (10, 30). p73 {alpha} mRNA is the most abundant isoform of p73 in malignant tumors but its protein product (p73 {alpha}) has a much lower transactivation ability compared with p73 ß (10, 12, 30, 31). The COOH-terminal region of p73 {alpha} contains a regulatory domain that regulates p73 protein stability by interacting with an ubiquitin-proteasome degradation pathway (15), a finding that is very similar to what we observed in the case of IIp45S in this study. Thus, accumulating evidence suggests that alternative splicing-induced protein instability may represent a new paradigm for gene inactivation in cancers.

The molecular mechanisms that regulate tumor-specific alternative splicing have therefore become an important area for research. Two possible scenarios are commonly considered: one is mutation of cis-regulatory sequences in target genes such as splicing sites or splicing silencer or enhancer elements, leading to aberrant splicing; the other is alteration in the trans-regulators of splicing (19). Our mutation screening did not identify any mutations within RNA splicing sites of IIp45, ruling out the possibility that splicing site mutations alter IIp45 RNA splicing decisions in gliomas. Thus, the possible causes of alternative splicing of IIp45 in gliomas might be changes in splicing silencer elements, splicing enhancer elements, or trans-regulators of splicing of IIp45. Our bioinformatics sequence analysis of IIp45 genomic sequences spanning from introns 6 to 7 identified putative binding sites for polypyrimidine tract-binding protein (PTB), splicing silencer, within intron 7. It has been reported that overexpression of PTB is correlated with glioma progression (32). Therefore, one possibility is that the increased PTB in glioma cells suppresses the normal splicing of IIp45 intron 7 that results in the exon 7 skipping. This possibility will be tested in the future.

In summary, our present study illustrates that inactivation of the invasion inhibitory gene IIp45 is not via frequent genetic mutation. Instead, IIp45 is inactivated in gliomas through a tumor-specific alternative splicing mechanism, which is unrelated to the frequent deletion of 1p36 in tumors. The tumor-specific alternative splicing of IIp45 generates a mutant isoform encoding a highly unstable protein product that is recognized and degraded by the ubiquitin-proteasome machinery. This study supports the notion that alternative splicing is the predominant mechanism for inactivation of a group of tumor suppressor genes that include p73 and IIp45.


   Acknowledgments
 
Grant support: M.D. Anderson Cancer Center and National Cancer Institute/NIH Cancer Center Support Grant (M.D. Anderson Cancer Center).

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.

We thank Shouming Kong and Ellen Taylor for their technical assistance and Dr. Xiangwei Wu for providing the pcDNA HA-Ub plasmid.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 9/20/04. Revised 1/23/05. Accepted 2/ 2/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Song SW, Fuller GN, Khan A, et al. IIp45, an insulin-like growth factor binding protein 2 (IGFBP-2) binding protein, antagonizes IGFBP-2 stimulation of glioma cell invasion. Proc Natl Acad Sci U S A 2003;100:13970–5.[Abstract/Free Full Text]
  2. Wang H, Wang H, Shen WP, et al. Insulin-like growth factor binding protein 2 enhances glioblastoma invasion by activating invasion-enhancing genes. Cancer Res 2003;63:4315–21.[Abstract/Free Full Text]
  3. Fuller GN, Chang HR, Hess KR, et al. Reactivation of Insulin-like growth factor binding protein 2 expression in glioblastoma multiforme: a revelation by parallel gene expression profiling. Cancer Res 1999;59:4228–32.[Abstract/Free Full Text]
  4. Sallinen SL, Sallinen PK, Haapasalo HK, et al. Identification of differentially expressed genes in human gliomas by DNA microarray and tissue chip techniques. Cancer Res 2000;60:6617–22.[Abstract/Free Full Text]
  5. Smith JS, Perry A, Borell TJ, et al. Alterations of chromosome arms 1p and 19q as predictors of survival in oligodendrogliomas, astrocytomas, and mixed oligoastrocytomas. J Clin Oncol 2000;18:636–5.[Abstract/Free Full Text]
  6. Smith JS, Alderete B, Minn Y, et al. Localization of common deletion regions on 1p and 19q in human gliomas and their association with histological subtype. Oncogene 1999;18:4144–52.[CrossRef][Medline]
  7. Janoueix-Lerosey I, Novikov E, Monteiro M, et al. Gene expression profiling of 1p35-36 genes in neuroblastoma. Oncogene 2004;23:5912–22.[CrossRef][Medline]
  8. Ragnarsson G, Eiriksdottir G, Johannsdottir JT, Jonasson JG, Egilsson V, Ingvarsson S. Loss of heterozygosity at chromosome 1p in different solid human tumours: association with survival. Br J Cancer 1999;79:1468–74.[CrossRef][Medline]
  9. Gibbs M, Stanford JL, McIndoe RA, et al. Evidence for a rare prostate cancer-susceptibility locus at chromosome 1p36. Am J Hum Genet 1999;64:776–87.[CrossRef][Medline]
  10. Zwahlen D, Tschan MP, Grob TJ, et al. Differential expression of p73 splice variants and protein in benign and malignant ovarian tumours. Int J Cancer 2000;88:66–70.[CrossRef][Medline]
  11. Loiseau H, Arsaut J, Demotes-Mainard J. p73 gene transcripts in human brain tumors: overexpression and altered splicing in ependymomas. Neurosci Lett 1999;263:173–6.[CrossRef][Medline]
  12. Takahashi H, Ichimiya S, Nimura Y, et al. Mutation, allelotyping, and transcription analyses of the p73 gene in prostatic carcinoma. Cancer Res 1998;58:2076–7.[Abstract/Free Full Text]
  13. Aoki T, Miyamoto A, Marubashi S, et al. Clinical value of alterations in p73 gene, related to p53 at 1p36, in human hepatocellular carcinoma. Int J Oncol 2004;24:441–6.[Medline]
  14. Shan L, Yang Q, Nakamura Y, et al. Frequent loss of heterozygosity at 1p36.3 and p73 abnormality in parathyroid adenomas. Mod Pathol 2001;14:273–8.[CrossRef][Medline]
  15. Lee CW, La Thangue NB. Promoter specificity and stability control of the p53-related protein p73. Oncogene 1999;18:4171–81.[CrossRef][Medline]
  16. Song SW, Cote G, Wu CL, Zhang W. Alternative splicing: genetic complexity in cancer. In Zhang W and Shmulevich I, editors. Computational and statistical approaches to genomics. Boston: Kluwer Academic Publishers; 2002. p. 277–97.
  17. Steenbergh PH, Hoppener JW, Zandberg J, et al. Calcitonin gene related peptide coding sequence is conserved in the human genome and is expressed in medullary thyroid carcinoma. J Clin Endocrinol Metab 1984;59:358–60.[Abstract/Free Full Text]
  18. Castellani P, Siri A, Rosellini C, et al. Transformed human cells release different fibronectin variants than do normal cells. J Cell Biol 1986;103:1671–7.[Abstract/Free Full Text]
  19. Cote G, Bruno IG, Jin W. Abnormal regulation of RNA splicing in gliomas. In Zhang W and Fuller GN, editors. Genomic and molecular neuro-oncology. Massachusetts: Jones and Bartlett Publishers. 2004; p. 165–83.
  20. Yamaguchi F, Saya H, Bruner JM, Morrison RS. Differential expression of two fibroblast growth factor-receptor gene is associated with malignant progression in human astrocytomas. Proc Natl Acad Sci U S A 1994;91:484–8.[Abstract/Free Full Text]
  21. Dani C, Blanchard JM, Piechaczyk M, El Sabouty S, Marty L, Jeanteur P. Extreme instability of myc mRNA in normal and transformed human cells. Proc Natl Acad Sci U S A 1984;81:7046–50.[Abstract/Free Full Text]
  22. Kyritsis AP, Bondy ML, Xiao M, et al. Germline p53 gene mutations in subsets of glioma patients. J Natl Cancer Inst 1994;86:344–9.[Abstract/Free Full Text]
  23. Li J, Yen C, Liaw D, et al. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 1997;275:1943–7.[Abstract/Free Full Text]
  24. Costello JF, Berger MS, Huang HS, Cavenee WK. Silencing of p16/CDKN2 expression in human gliomas by methylation and chromatin condensation. Cancer Res 1996;56:2405–10.[Abstract/Free Full Text]
  25. Kaufmann D, Leistner W, Kruse P, et al. Aberrant splicing in several human tumors in the tumor suppressor genes neurofibromatosis type 1, neurofibromatosis type 2, and tuberous sclerosis 2. Cancer Res 2002;62:1503–9.[Abstract/Free Full Text]
  26. Thomson SA, Wallace MR. RT-PCR splicing analysis of the NF1 open reading frame. Hum Genet 2002;110:495–502.[CrossRef][Medline]
  27. Pio R, Zudaire I, Pino I, et al. {alpha} CP-4, encoded by a putative tumor suppressor gene at 3p21, but not its alternative splice variant {alpha} CP-4a, is underexpressed in lung cancer. Cancer Res 2004;64:4171–9.[Abstract/Free Full Text]
  28. Huiping C, Kristjansdottir S, Bergthorsson JT, et al. High frequency of LOH, MSI and abnormal expression of FHIT in gastric cancer. European J Cancer 2002;38:728–35.
  29. Martin A, Tsui HW, Tsui FW. SHP-1 variant proteins are absent in motheaten mice despite presence of splice variant transcripts with open reading frames. Mol Immunol 1999;36:1029–41.[CrossRef][Medline]
  30. Ueda Y, Hijikata M, Takagi S, et al. Transcriptional activities of p73 splicing variants are regulated by inter-variant association. Biochem J 2001;356:859–66.[CrossRef][Medline]
  31. Ozaki T, Naka M, Takada N, et al. Deletion of the COOH-terminal region of p73{alpha} enhances both its transactivation function and DNA-binding activity but inhibits induction of apoptosis in mammalian cells. Cancer Res 1999;59:5902–7.[Abstract/Free Full Text]
  32. Jin W, McCutcheon IE, Fuller GN, et al. Fibroblast growth factor receptor-1 {alpha}-exon exclusion and polypyrimidine tract-binding protein in glioblastoma multiforme tumors. Cancer Res 2000;60:1221–4.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
Y. Yu, X. Jiang, B. S. Schoch, R. S. Carroll, P. M. Black, and M. D. Johnson
Aberrant Splicing of Cyclin-Dependent Kinase-Associated Protein Phosphatase KAP Increases Proliferation and Migration in Glioblastoma
Cancer Res., January 1, 2007; 67(1): 130 - 138.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
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 Song, S. W.
Right arrow Articles by Zhang, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Song, S. W.
Right arrow Articles by Zhang, W.

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