Cancer Research Translational Cancer Medicine 2008: Cancer Clinical Trials and Personalized Medicine Susan G. Komen for the Cure-AACR Outstanding Investigator Award for Breast 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 Cell Growth & Differentiation
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 Kumar, R.
Right arrow Articles by Callen, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kumar, R.
Right arrow Articles by Callen, D. F.
[Cancer Research 65, 11304-11313, December 15, 2005]
© 2005 American Association for Cancer Research

Molecular Biology, Pathobiology and Genetics

FBXO31 Is the Chromosome 16q24.3 Senescence Gene, a Candidate Breast Tumor Suppressor, and a Component of an SCF Complex

Raman Kumar1, Paul M. Neilsen1, Joanne Crawford2, Ross McKirdy1, Jaclyn Lee1, Jason A. Powell2, Zarqa Saif1, Julie M. Martin1, Marc Lombaerts3, Cees J. Cornelisse3, Anne-Marie Cleton-Jansen3 and David F. Callen1

1 Breast Cancer Genetics Group, Dame Roma Mitchell Cancer Research Laboratories, Department of Medicine, University of Adelaide and Hanson Institute, Institute of Medical and Veterinary Science Adelaide; 2 Centre for Medical Genetics, Department of Cytogenetics and Molecular Genetics, Women's and Children's Hospital, North Adelaide, South Australia, Australia; and 3 Department of Pathology, Leiden University Medical Centre, Leiden, The Netherlands

Requests for reprints: David F. Callen, Breast Cancer Genetics Group, Dame Roma Mitchell Cancer Research Laboratories, Institute of Medical and Veterinary Science, Hanson Institute Building, Frome Road, Adelaide, South Australia 5000, Australia. Phone: 618-8-222-3450; Fax: 618-8-222-3217; E-mail: david.callen{at}imvs.sa.gov.au.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
A BAC located in the 16q24.3 breast cancer loss of heterozygosity region was previously shown to restore cellular senescence when transferred into breast tumor cell lines. We have shown that FBXO31, although located just distal to this BAC, can induce cellular senescence in the breast cancer cell line MCF-7 and is the likely candidate senescence gene. FBXO31 has properties consistent with a tumor suppressor, because ectopic expression of FBXO31 in two breast cancer cell lines inhibited colony growth on plastic and inhibited cell proliferation in the MCF-7 cell line. In addition, compared with the relative expression in normal breast, levels of FBXO31 were down-regulated in breast tumor cell lines and primary tumors. FBXO31 was cell cycle regulated in the breast cell lines MCF-10A and SKBR3 with maximal expression from late G2 to early G1 phase. Ectopic expression of FBXO31 in the breast cancer cell line MDA-MB-468 resulted in the accumulation of cells at the G1 phase of the cell cycle. FBXO31 contains an F-box domain and is associated with the proteins Skp1, Roc-1, and Cullin-1, suggesting that FBXO31 is a component of a SCF ubiquitination complex. We propose that FBXO31 functions as a tumor suppressor by generating SCFFBXO31 complexes that target particular substrates, critical for the normal execution of the cell cycle, for ubiquitination and subsequent degradation. (Cancer Res 2005; 65(24): 11304-313)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Numerous cytogenetic and molecular studies of breast cancer have identified frequent loss of heterozygosity (LOH) of the long arm of chromosome 16 (1). We undertook detailed studies to define more precisely the region of chromosome 16 most frequently affected by LOH and therefore the likely location of the breast cancer tumor suppressors. A total of 712 sporadic breast tumors were screened, and it was established that a 2.4-Mb segment of the chromosome 16q24.3 terminal band was one of the minimum regions of LOH (2). These findings have subsequently been substantiated by a pooled analysis of various published LOH studies that confirmed band 16q24.3 as a region of preferential loss (1). All 104 genes in this LOH region were identified, and candidate genes were selected based on their known or predicted function as well as their expression in breast cancer cell lines as determined by real-time reverse transcription-PCR (RT-PCR; ref. 3). Large variations in expression of FBXO31 and CBFA2T3 (also known as MTG16) were observed in breast cancer cell lines similar to that found for established tumor suppressor genes. Functional studies were consistent with CBFA2T3 as a breast cancer tumor suppressor (4). Studies suggest that more than one tumor suppressor gene may reside in regions of cancer LOH; for example, the 3p LOH region in small cell lung carcinoma contains several tumor suppressor genes (5). Therefore, we have continued to characterize additional candidate tumor suppressor genes in the 16q24.3 region.

Previous studies showed that the microcell-mediated transfer of human chromosome 16q fragments caused senescence in human- and mouse-immortalized cell lines (6). This senescence-associated region was subsequently localized to a 360-kb yeast artificial chromosome (YAC) d792t2 mapping between the markers D16S498 and D16S476 (7) and more recently to an 85-kb BAC clone 346J21 (8). Transfer of this BAC into several cell lines, including the human breast cancer line MCF-7, was shown to restore cellular senescence. These findings are consistent with the location of a senescence gene within the BAC 346J21, and this gene may also function as a tumor suppressor gene. Mapping of the two previously identified potential tumor suppressor genes locates FBXO31 within the YAC d792t2, whereas CBFA2T3 is 1.7 Mb distal to this YAC. We report mapping of the BAC 346J21 with respect to the FBXO31 gene and investigate if the properties of FBXO31 are consistent with a senescence and tumor suppressor gene. In addition, because FBXO31 encodes an F-box protein and other proteins with F-box domains have been shown to be variant components of SCF complexes (9), we have investigated whether FBXO31 mediates its function as part of an SCFFBXO31 E3 ubiquitin ligase complex.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell lines and antibodies. 184V and 48RS are finite life span human mammary epithelial cells (HMEC) derived from two different patients, whereas 184A1 is a nonmalignant immortally transformed cell line derived from normal breast epithelium (10). These cell lines were grown in MCDB-170 medium (Invitrogen, Carlsberg, CA). GP-293 cells were purchased from BD Biosciences (San Jose, CA). All other cell lines were purchased from the American Type Culture Collection (Manassas, VA) and were grown in the recommended media. Stable cell lines expressing FLAG-HA-FBXO31 were generated by G418 selection of 293T cells transduced with a pQCXIN-based retroviral construct. Antibodies used were rat anti-HA (Roche Diagnostics; Indianapolis, IN); mouse anti-myc and mouse anti-Cul-1 (Santa Cruz Biotechnology, Santa Cruz, CA); mouse anti-Skp1 (BD Biosciences); and rabbit anti-Skp1, rabbit anti-Roc-1, and rabbit anti–mitogen-activated protein/extracellular signal-related kinase 2 (NeoMarkers, Fremont, CA). A rabbit polyclonal antibody was raised against a 15-amino-acid synthetic peptide representing the COOH terminus (amino acids 525-539; QAFDEMLKNIQSLTS) of the FBXO31 protein conjugated to keyhole limpet hemocyanin. The polyclonal antibody was then affinity purified on Sepharose beads coupled with the same peptide.

Plasmids. Appropriate protein-coding open reading frames (ORF) were PCR amplified from adult brain marathon cDNA (BD Biosciences) and cloned in-frame with the epitope tags into either pCMV-myc or pCMV-HA (BD Biosciences) mammalian expression vectors to generate clones expressing myc-FBXO31 and HA-Skp1 proteins. The sequence coding for amino acids 51 to 109, containing the F-box domain, were deleted from the pCMV-myc-FBXO31 construct using overlap PCR to generate pCMV-myc-FBXO31{Delta}F. For retroviral-mediated ectopic expression studies, fragments coding myc-FBXO31, myc-FBXO31{Delta}F, and myc-p53 were also cloned into the pLNCX2 vector (BD Biosciences). For the cell cycle analysis, FBXO31, FBXO31{Delta}F, and p53 coding fragments were cloned in-frame with the enhanced green fluorescent protein (EGFP) ORF into pEGFP-C1 (BD Biosciences) to generate constructs capable of expressing EGFP fusions. To generate stable FBXO31 expressing cell lines, the FBXO31 ORF with FLAG and hemagglutinin (HA) epitope tags was cloned into the pQCXIN retroviral vector. The sequences of all constructs were confirmed by DNA sequencing.

Human specimens. Primary breast tumors with known histology, stage, and differentiation grade had 16q LOH status determined as reported previously (2). Patient material was obtained on approval of local medical ethics committees.

Northern blot. Multiple Tissue Northern Blot (BD Biosciences) carrying polyadenylated RNA from various human tissues was probed with a 32P-labeled FBXO31 DNA fragment. The 224-bp DNA probe was generated by PCR amplification from the full-length clone using the following primers: forward, 5'-CTTCACCGATATAGACAC and reverse, 5'-GGCCGTACATGCACTCCACTG and radiolabeled with {alpha}-32P-dCTP using the Megaprime DNA labeling system (Amersham Biosciences, Piscataway, NJ).

Real-time reverse transcription-PCR. FBXO31 expression was determined by real-time RT-PCR with the forward primer: 5'-CCGGCGGGAGGCAGGAGGAGT and reverse primer: 5'-GCGGCGGTAGGTCAGGCAGTTGTCG. Using TRIzol (Invitrogen, Carlsbad, CA), total RNA was isolated from pellets of cell lines grown to ~80% confluence, or paraffin blocks of breast tumors using 20 x 20 µm tissue sections with at least 50% tumor. Two micrograms of total RNA were used as template for cDNA in a final volume of 20 µL. Each real-time RT-PCR reaction (10-minute activation of the polymerase at 95°C, 45 cycles of 15 seconds at 95°C, 1 minute at 60°C, and signal detection at 60°C) contained 0.2 µL of this cDNA and used SYBR Green (qPCRTM Core kit, Eurogentec, Seraing, Belgium or Bio-Rad iQ Supermix) on a Bio-Rad iCycler (Bio-Rad, Hercules, CA). The geometric mean expression of three housekeeping genes (HNRPM, CPSF6, and TBP) was used to normalize the expression of FBXO31 in breast tumor cDNA according to published methods (11). For cell line expression, the housekeeping gene cyclophilin A was used to normalize the expression of FBXO31, because previous studies determined this was a transcript with minimal variation in such cell lines (3).

Cell-based assays. To generate amphotropic recombinant retroviruses, GP2-293 cells were transfected with pVSV-G (BD Biosciences) and various gene constructs cloned into the pLNCX2 vector using LipofectAMINE 2000 transfection agent (Invitrogen). Forty-eight hours after transfection, the culture medium containing the retrovirus was collected, filtered, and used to transduce the appropriate cell line at ~50% confluence. Transduction was enhanced by adding 8 µg/mL polybrene and centrifuging plates at 1,250 rpm for 30 minutes. To determine the frequency of colony formation, after 24 hours, the cells from each treatment were plated in six-well plates at concentrations of 4 to 10 x 103 cells/mL in selective media containing G418, and colonies were counted in triplicate wells after growth for a further 2 to 3 weeks. The empty vector was used as a negative control. The transduction efficiency was monitored by immunofluorescence detection of the myc-tag. Cellular senescence was assayed using a kit from Cell Signaling Technologies (Beverly, MA) based on senescence-specific acidic ß-galactosidase activity (12). Senescent cells stained intense blue after overnight incubation at 37°C. Colonies of >30 cells containing >10 blue staining cells were scored as senescent. Proliferation assays were determined on transduced cells selected for 10 to 14 days in the presence of G418. These cells were plated at 10% to 20% confluence in 96-well plates and at various times samples (n = 6-8) assayed by incubating cells for 2 hours with the CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Madison, WI) and measuring absorbance at 490 nm.

Cell synchronization. Cells were synchronized at G1-S phase using a double thymidine block. Cells were grown in the presence of 2 mmol/L thymidine (Sigma-Aldrich, St. Louis, MO) for 24 hours and then washed and grown in fresh medium without thymidine for 10 hours. Cells were cultured in the presence of 2 mmol/L thymidine for a further 16 hours and then released from the G1-S block by washing twice with fresh medium. Cells collected at various time points following release from the second thymidine block were lysed in 50 mmol/L Tris-HCl (pH 7.5), 250 mmol/L NaCl, 1% Triton X-100, 1 mmol/L EDTA, 50 mmol/L NaF, 0.1 mmol/L Na3VO4, 1 mmol/L DTT, 1x protease inhibitors (Roche, Indianapolis, IN) on ice for 15 minutes. Lysed samples were clarified by centrifugation and then assayed for protein concentration using bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL).

Cell cycle analysis. MCF-7 or MDA-MB-468 cells were transfected with pEGFP-FBXO31, pEGFP-FBXO31{Delta}F, pEGFP-p53, or pEGFP-C1 using LipofectAMINE 2000 (Invitrogen). Cells were collected 24 and 48 hours after transfection and treated as described (13). Briefly, cells were pelleted (300 x g for 5 minutes at 4°C), washed twice with cold PBS, resuspended in 500 µL of cold PBS, and then fixed for 1 hour at 4°C by adding of 500 µL fixation solution (2% w/v paraformaldehyde in PBS, pH 7.2). The fixed cells were pelleted, washed with cold PBS, resuspended in 1 mL of 70% ethanol added dropwise to the pellet while vortexing, and then incubated overnight at 4°C. The next day, the cells were pelleted and resuspended in 1 mL propidium iodide solution (40 µg/mL with 100 µg/mL RNase A) for 30 minutes at 37°C in the dark and analyzed on a FACScan flow cytometer (BD Biosciences, San Jose, CA), and the multivariate data were collected using CellQuest software (BDIS, Pittsburgh, PA). Cell cycle analysis of DNA histograms was done using ModFit LT V2.0 (Verity, Topsham, ME) software.

Western blot analysis. Cells were lysed in 150 mmol/L NaCl, 1% Triton-X, 50 mmol/L Tris-HCl (pH 8) with complete protease inhibitor cocktail (Roche); sonicated; and centrifuged. Clarified cell lysates or immunoprecipitated protein samples were resolved on SDS-PAGE and transferred on to Hybond-C Extra (Amersham Biosciences). Membranes were probed with various primary antibodies and detected with appropriate horseradish peroxidase–conjugated secondary antibodies using enhanced chemiluminescence detection system (Amersham Biosciences) using standard protocols (14).

Coimmunoprecipitation and affinity purification. Plasmids (2 µg each) expressing either myc-FBXO31 or HA-Skp1 or both were transfected into 5 x 105 HEK293T cells in 35-mm wells. Twenty-four hours after transfection, cells were harvested and lysed as for Western blot analysis. Clarified lysates were incubated with either anti-myc or anti-HA-conjugated agarose beads (Sigma-Aldrich). Beads were washed twice with 150 mmol/L NaCl, 1% Igepal CA-630, 0.5% sodium deoxycholate, 0.1% SDS, 50 mmol/L Tris-HCl (pH 8), then 20 mmol/L Tris-HCl (pH 7.5), and the protein complexes were eluted with 1x protein-loading buffer (0.0625 mol/L Tris-HCl [pH 6.8], 2% SDS, 10% glycerol, 5% 2-mercaptoethanol; ref. 15). Input and immunoprecipitated proteins were resolved by Western blot analysis, and membranes were immunoblotted with rat anti-HA, mouse anti-myc, mouse anti-Cullin-1, and rabbit anti-Roc-1 antibodies. HEK293T and HEK293T cell lines stably expressing FLAG-HA-FBXO31 protein were used to affinity purify the tagged FBXO31. Cells were lysed in 20 mmol/L Tris-HCl (pH 8), 200 mmol/L KCl, 5 mmol/L MgCl2, 1 mmol/L EDTA, 0.1% Tween 20, 0.2 mmol/L phenylmethylsulfonyl fluoride, 1x protease inhibitors (Roche) on ice for 15 minutes. Clarified lysates were incubated for 2 hours with anti-FLAG M2 agarose (Sigma-Aldich). The beads were then washed with the lysis buffer, and the protein complexes were eluted with the same buffer containing 250 µg/mL FLAG peptide (Roche). The inputs and eluates were analyzed by Western blotting with rat anti-HA, rabbit anti-Roc-1, mouse anti-Cullin-1, and mouse anti-Skp1 antibodies.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Analysis of 16q24.3 senescence region. The relative positions of the FBXO31 gene together with the YAC d792t2, the BAC 346J21, and the various sequence-tagged sites (STS) used previously to define the location of the BAC (8) are given in Fig. 1A. The sequenced extremities of the BAC 346J21 are defined from the available end sequences with accession nos. B15982 and B15983 available at the National Center for Biotechnology Information (NCBI). In BAC 346J21, the marker D16S3063 was reported to be distal to D16S3048 as determined from the mapping of STSs and overlapping BAC clones (8). However, the order of markers as defined by the NCBI genomic sequence places D16S3063 proximal to D16S3048 and some 14 kb proximal to the STS 344A17.T7. The STS 346J21.T7, together with WI15838, WI12410, and the 3' end of FBXO31 were all located within 200 bp. The YAC d792t2 was purchased and shown by PCR with STS primers derived from the cDNA sequence BX119888 to possess an interstitial deletion extending at least 1.2 kb (Fig. 1A). These results were confirmed by multiplex PCR with additional STSs present on the YAC (data not shown). The YAC d792t2 is a chimera because it also possesses a segment of chromosome 11q13 (7). The BAC 346J21 only contains 290 bp of the 3' end of FBXO31 with the majority of the gene positioned telomeric to this BAC. The genomic region encompassed by the BAC does not contain any predicted CpG islands and contains 44% low copy repeats as determined by RepeatMasker. The NCBI build 35.1 of the human genome sequence predicts three possible transcripts completely contained within the genomic region of the BAC 346J21 (as defined by the location of the end sequences B15982 and B15983 from chromosome 16 base 85,835,220 to 85,920,731): LOC388305, LOC400552, and LOC440391. Based on the absence of introns for LOC400552 and LOC440391, the lack of significant representation of any of the three transcripts in expressed sequence tag (EST) libraries, and the absence of any motifs or species conservation of the predicted proteins, these are unlikely to represent functionally transcribed genes. Because well-defined genes within the confines of this BAC were absent, it was decided to further characterize the FBXO31 gene, located within the YAC d792t2 and immediately adjacent to the BAC 346J21.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 1. Location and tissue expression of the FBXO31 gene and presence of an F-box domain in the encoded protein. A, location of the FBXO31 gene. Positions of various markers are determined by analysis of the genomic sequence of this region of chromosome 16. Positions of the YAC d792t2 and the BAC 346J21 are shown (7, 8). B, Northern blot of polyadenylated RNAs from various human tissues showing a 3.6-kb FBXO31 transcript. Arrows, location of molecular weight markers. C, alignment of F-box domains from FBXO31 and other selected F-box proteins. Boxed, individual amino acids identical among all five F-box sequences; underlined, when present in three to four sequences. Nucleotide sequences are derived from Genbank, NCBI.

 
Northern blot and in silico analyses. The presence of a 3.6-kb RNA band on a Northern blot shows that FBXO31 expression is negligible in bone marrow, highly expressed in brain, and expressed at similar levels in the other 10 human tissues, including breast (Fig. 1B). FBXO31 mRNA contains a 1,620-bp ORF that is transcribed from the genomic DNA with nine exons spanning ~52 kb. The FBXO31 transcript has an exceptionally short 23-base 5'-untranslated region. In silico analysis of the EST database and 5' rapid amplification of cDNA ends (data not shown) failed to identify any additional 5' transcript sequence. In addition, the presence of a 2.48-kb CpG island (66% G + C and with 9% CpG) encompassing the first 350-bp exon of FBXO31 also supports the 5' start of the transcript. FBXO31 encodes a 539-amino-acid protein with a predicted molecular mass of 61 kDa. The predicted amino acid sequence of FBXO31 does not have significant homology to known proteins, except for a consensus F-box domain at the amino terminus (Fig. 1C). The 40-amino-acid F-box domain is involved in protein-protein interactions and is present in a large family of proteins (16). FBXO31 also contains six minimal (RxxL) destruction box (D-box) motifs that are hallmark of proteins degraded via the anaphase-promoting complex/cyclosome (APC/C; refs. 17, 18).

Ectopic expression of FBXO31 induces cellular senescence in MCF-7. Colonies of MCF-7 growing in the presence of G418 after retroviral transduction were stained for the presence of the senescence-specific acidic ß-galactosidase activity (12). Senescent cells had intense blue staining with a large flattened morphology. The MCF-7 cell line has a spontaneous background of senescent cells usually evident as occasional blue staining cells (Fig. 2A). Following several weeks of growth, 35% to 40% of colonies ectopically expressing FBXO31 showed senescence compared with 5% to 7% of senescent colonies with vector alone (Fig. 2B-C; Table 1). Colonies from cells transduced with FBXO31{Delta}F were similar to the vector alone, suggesting that the F-box motif of FBXO31 is critical for this senescent function. In subsequent experiments, MCF-7 cells ectopically expressing an EGFP-tagged FBXO31 showed that the presence of cell senescence was correlated to the expression of FBXO31 (Fig 2D-I). In MCF-7 colonies, senescent cells were frequent in regions with EGFP fluorescence, whereas there were few senescent cells in regions without EGFP.



View larger version (84K):
[in this window]
[in a new window]
  		Figure 2. FBXO31 ectopic expression induces cellular senescence in MCF-7 cells. Staining of senescent cells in colonies of MCF-7 ectopically expressing empty vector (A) and FBXO31 (B and C). Colonies of MCF-7 stained for senescence (D, F, and H) and fluorescent images showing the ectopically expressing EGFP-FBXO31 (E, G, and I).

 

View this table:
[in this window]
[in a new window]
  		Table 1. MCF-7 senescent colonies ectopically expressing FBXO31

 
Alteration of FBXO31 expression effects growth of breast cell lines. Because it was determined that ectopic expression of FBXO31 could induce senescence in the breast cell line MCF-7, we investigated if this gene could function more generally as a tumor suppressor. Ectopic expression of a tumor suppressor in a cancer cell line would be expected to inhibit the ability of cells to initiate colonies on plastic and inhibit cell proliferation. FBXO31 was ectopically expressed in the breast cancer cell lines MCF-7 and MDA-MB-468 by retroviral transduction and colony formation on plastic assessed. Both cell lines showed a reduction in colony formation, compared with the vector control, when transduced with retroviruses expressing FBXO31 (Fig. 3A-B). This reduction in colony formation averaged 50% for MCF-7 from five independent experiments and 77% for MDA-MB-468 from six independent experiments (Fig. 3C). In all experiments, an expected severe inhibition in colony formation (generally 95% compared with the vector control) was observed in cells transduced with retrovirus expressing the p53 protein. Ectopic expression of FBXO31{Delta}F, a construct with a deletion of the F-box domain, showed an intermediate reduction in colony formation (MCF-7) or no significant reduction (MDA-MB-468) compared with the vector control (Fig. 3A-B).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 3. Ectopic expression of FBXO31 reduced the growth characteristics of breast cancer cell lines MCF-7 and MDA-MB-468. The MCF-7 (A) and MDA-MB-468 (B) breast cancer cell lines were transduced with recombinant retroviruses expressing FBXO31, FBXO31{Delta}F, p53 (positive control), and empty vector (negative control). Number of colonies growing on plastic dishes in the presence of G418, 2 weeks after transduction. C, average effect of ectopic expression of FBXO31 on colony growth, compared with vector alone, from five independent experiments with MCF-7 and six experiments with MDA-MB-468. D, FBXO31 ectopic expression inhibited MCF-7 cell growth. Growth assays were initiated from cultures transduced with retroviruses expressing FBXO31, FBXO31{Delta}F, p53, and empty vector and selected for 14 days in media containing G418.

 
The effect of FBXO31 ectopic expression on the proliferation of MCF-7 cells was also determined by generating cultures of G418-resistant MCF-7 cells following retroviral transduction. Growth curves of these cell lines showed that cells expressing FBXO31 or p53 had a similar reduction in their proliferation (after 72 hours, 34% and 38%, respectively) compared with the vector control (Fig. 3D). Cells expressing FBXO31{Delta}F had an intermediate 20% reduction of proliferation compared with the control.

FBXO31 expression and mutation screening. To further investigate the possible role of FBXO31 as a tumor suppressor, the relative expression of FBXO31 was determined by real-time RT-PCR in a panel of breast cell lines (Fig. 4A). Results showed levels of expression in finite life span HMEC and nonmalignant immortalized cells that averaged 7.9 times greater than the average expression breast cancer cell lines. The chromosome 16q LOH of the breast cancer cell lines has been determined previously (19). The average relative expression of FBXO31 in the cell lines MCF-7 and BT20, without chromosome 16q LOH, was significantly higher than the cell lines T47-D, MDA-MB-231, MDA-MB-468, and SKBR3 with chromosome 16 LOH (1.27 versus 0.58, P < 0.01). Based on the findings of reduced expression in breast cancer cell lines, expression was then assessed by real-time RT-PCR in a panel of primary breast tumors selected to have at least 50% tumor cells as estimated from H&E-stained sections. Presented in Fig. 4B is the relative FBXO31 expression in tumors classified by analysis of polymorphic markers (2) into those without chromosome 16q LOH and those with LOH for the entire long arm of chromosome 16. There was a trend for lower FBXO31 expression in tumors with 16q LOH (10 of 15 tumors had expression of <50% of the average expression in normal breast) compared with tumors without LOH of 16q (6 of 11 tumors).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 4. FBXO31 expression as measured by real-time RT-PCR is reduced in breast cell lines and tumors. A, FBXO31 expression in various breast cell lines. B, FBXO31 expression in breast tumors with and without 16q LOH. Horizontal dashed line, at 50% of the average expression in the normal breast samples.

 
Based on the observation of reduced FBXO31 expression in some tumors, it was considered possible that the FBXO31 gene may function as a tumor suppressor. Tumor-restricted mutations can provide definitive evidence for a tumor suppressor gene (20). Accordingly, screening for the presence of mutations in FBXO31 was undertaken on DNA isolated from a panel of 24 breast cancer cell lines by single strand conformational analysis (SSCA). In addition, DNA from 44 breast tumors with known LOH of the long arm of chromosome 16, and the DNA from the normal peripheral blood lymphocytes of the same patients were also screened. Primers for SSCA were designed to bind within the intronic sequences such that coding and flanking intronic regions could be analyzed (data available upon request). No tumor specific mutations were detected for FBXO31 both in breast cancer cell lines and tumor DNA.

FBXO31 is cell cycle regulated. Following the findings that ectopic expression of FBXO31 suppresses growth, a fluorescence-activated cell sorting–based cell cycle analysis was used to determine if the mechanism of this growth suppression is due to a specific effect on the cell cycle. Cell cycle analysis was done on the breast cancer cell line MDA-MB-468 transiently expressing the EGFP-tagged proteins for 24 and 48 hours (Fig. 5). Analysis of the asynchronous cell populations showed a 13% increase in the proportion of EGFP-FBXO31-expressing G1 cells from 24 to 48 hours (Fig. 5A-B) compared with the 3% increase observed in only EGFP-expressing G1 cells in the same time period (Fig. 5G-H). Ectopic expression of EGFP-FBXO31{Delta}F resulted in profiles similar to the control with a 3% increase in G1 cells from 24 to 48 hours (Fig. 5C-D). However, when EGFP-p53 was ectopically expressed, there was a 17% increase in G1 cells from 24 to 48 hours (Fig. 5E-F). Similar results were observed in independent experiments and in the MCF-7 cell line (data not shown). These observations suggest that a block in the cell cycle at G1 is the probable cause of the observed negative effect of FBXO31 ectopic expression on growth of breast cancer cell lines.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 5. FBXO31 ectopic expression blocks cells at G0-G1 of the cell cycle. Cell cycle curves of EGFP-positive asynchronous MDA-MB-468 cultures 24 and 48 hours after transfection with EGFP fusion constructs of FBXO31 (A-B), FBXO31{Delta}F (C-D), p53 (E-F), and empty vector (G-H).

 
The cell cycle regulation of FBXO31 levels was then investigated. The endogenous FBXO31 levels were determined using an affinity-purified rabbit polyclonal anti-FBXO31 antibody. Specificity of the anti-FBXO31 antibody was confirmed by Western blot analysis on protein lysates of MCF-10A and HEK293T cells transiently expressing myc-FBXO31. The protein samples showed a band at 78 kDa, suggesting that this antibody could specifically detect endogenous and ectopically expressed FBXO31 (data not shown). FBXO31 levels at different stages of the cell cycle were investigated in MCF-10A and the breast cancer cell line SKBR3 by collecting samples at various time points following release from a double thymidine block. For both cell lines, the variation in levels of FBXO31 at different times after release of the block shows that this protein is cell cycle regulated (Fig. 6A). The cell cycle synchronization was confirmed by change in the levels of cyclin B1. These levels were at a maximum at 6 to 10 hours for MCF10A and 8 to 12 hours for SKBR3. Cell progression from mitosis is associated with decreased levels of cyclin B1. For both cell lines, the levels of FBXO31 were a maximum at similar times, from 8 to 16 hours after release (Fig. 6A). Comparison with the cyclin B1 cell cycle profile indicate that levels of FBXO31 protein were at a maximum at late G2 to early G1 phase.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 6. FBXO31 is cell cycle regulated and is a component of the SCF complex. A, MCF-10A and SKBR3 cells were synchronized with a double thymidine block, harvested at the indicated time points after release from the block, and analyzed by Western blot analysis. Blots were probed with anti-FBXO31, anti-cyclin B1, and anti-MEK2 antibodies. B, HEK293T cells were transiently transfected with constructs expressing individually or combinations of myc-FBX031, myc-FBX031{Delta}F, or HA-Skp1 proteins as shown. Total cell lysates (lanes 1-5) were immunoprecipitated (lanes 6-10) with either anti-myc (top three blots) or anti-HA (bottom three blots) antibodies. Input (lanes 1-5) and immunoprecipitated (lanes 6-10) proteins were resolved by SDS-PAGE and probed with appropriate antibodies to detect FBXO31, FBXO31{Delta}F, Skp1, Cullin-1, and Roc-1 proteins. C, lysates from HEK293T and a stable HEK293T clone expressing FLAG-HA-FBXO31 were affinity purified using anti-FLAG M2-coated agarose beads. Input and affinity purified proteins were resolved by SDS-PAGE and probed with appropriate antibodies to detect FBXO31, Skp1, Cullin-1, and Roc-1 proteins.

 
FBXO31 exists as a component of SCFFBXO31 complex. Immunoprecipitation was used to determine if FBXO31 exists as a component of the SCF ubiquitination complex. Reciprocal experiments on HEK293T lysates prepared from cells transfected with constructs either expressing myc-FBXO31, myc-FBXO31{Delta}F, HA-Skp1, both myc-FBXO31 and HA-Skp1, or both myc-FBXO31{Delta}F and HA-Skp1 proteins showed that FBXO31 specifically interacts with the Skp1 protein (Fig. 6B). The FBXO31-Skp1 interaction was effectively eliminated when the F-box was deleted in cells transfected with the myc-FBXO31{Delta}F construct. Endogenous Cullin-1 and Roc-1 were also present in the immunoprecipitations from cells transfected with constructs expressing myc-FBXO31/HA-Skp1. The trace amounts of Cullin-1 and Roc-1 detected in anti-myc immunoprecipitations from cells transfected with myc-FBXO31{Delta}F and/or HA-Skp1 were considered to originate from nonspecific interactions, because Cullin-1 and Roc-1 were not detected in the reciprocal anti-HA immunoprecipitations done on the same lysates. To provide additional evidence for FBXO31 as a component of an SCFFBXO31 complex, stable HEK-293T-based cell lines were generated by retroviral transfer of FLAG-HA-FBXO31. The expression of the tagged FBXO31 protein was at near endogenous levels (data not shown). Cell lysates from these stable cell lines were affinity purified using anti-FLAG antibody, and the presence of endogenous Skp1, Cullin-1, and Roc-1 proteins was confirmed by Western blot analysis (Fig. 6C). Because Skp1, Cullin-1, and Roc-1 constitute invariant components of the SCF complexes (21, 22), it is concluded that FBXO31, either ectopically expressed or expressed in stable clones at near physiologic levels, exists as part of an SCFFBXO31 complex.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Previous studies determined that the YAC d792t2, and subsequently the transfection of a retrofitted BAC 346J21, caused cellular senescence when transferred into cell lines, and this was inferred to be caused by a gene termed SEN16 (7, 8). Our analysis of the YAC d792t2 showed an interstitial deletion was present in the region encompassed by the BAC, whereas analysis of the genomic sequence contained within the BAC 346J21 did not identify any potential candidate senescence genes. We chose to characterize the gene FBXO31, as it was located within the YAC d792t2. FBXO31 possessed properties consistent with the SEN16 senescence gene, because ectopic expression of FBXO31 in the breast cell line MCF-7 (one of the cell lines used in the SEN16 studies; ref. 8) resulted in an increase in the number of senescent colonies. In addition, colonies of transduced MCF-7 showed the expression of FBXO31 was correlated with the presence of regions of cellular senescence. In vitro systems using ectopic expression driven by highly active promoters can lead to nonphysiologic consequences. However, it is unlikely that the observed functional consequences of FBXO31 ectopic expression reflect nonphysiologic expression, because the senescence phenotype observed in MCF-7 cells (Fig. 2B-C) was similar to that previously reported (8). In addition, we have observed that G418-selected cells with retrovirally induced expression of FBXO31 have near physiologic levels of protein.

Although the data presented here are consistent with FBXO31 being the previously identified cellular senescence gene SEN16 (8), the location of FBXO31 is immediately distal to the BAC 346J21. There are several possible explanations for the apparent absence of the ORF of the FBXO31 gene in the BAC 346J21. First, the BAC may be rearranged and actually contain the FBXO31 gene. The possibility of a rearranged clone is suggested by the reversal of the order of the markers D16S3063 and D16S3048 in the reported physical mapping of BAC 346J21 (8). The second possibility is that uncharacterized DNA rearrangements in the clones containing the BAC result in changes in endogenous FBXO31 expression. MCF-7 transfected with the retrofitted BAC 346J21 were identified as rare G418-resistant colonies, but these could not be further characterized due to the absence of polymorphic markers unique to the introduced BAC. Therefore, there is no information regarding whether the introduced DNA is integrated, the sites of integration and the copy number. Studies suggest that it is extremely difficult to obtain stable cell lines with intact introduced DNA when transformed with large fragments of DNA, and alternative vector systems have been specifically designed to alleviate this problem (23).

Previously, based on breast cancer LOH studies, the chromosome band 16q24.3 was delineated as the smallest region of overlap and therefore the likely location of one or more breast cancer tumor suppressor genes (2). Therefore, the possibility was investigated that FBXO31 has a broader role as a breast tumor suppressor. The data presented suggest that this may be the case. The expression of the FBXO31 gene was reduced in both breast tumor cell lines and sporadic primary breast tumors compared with nonmalignant breast epithelium (Fig. 4A and B). In addition, there was a significantly reduced expression in breast cancer cell lines with 16q LOH compared with those without 16q LOH. Similarly, in primary tumor samples, there was a trend for a higher proportion of tumors showing reduced expression when LOH of 16q was present. However, complete elimination of FBXO31 expression was not observed in breast tumors, because the maximum reduction of the relative FBXO31 expression determined by RT-PCR was about 20% of the expression in normal breast. This suggests that down-regulation of FBXO31 is a likely scenario for function as a tumor suppressor but that residual levels of expression are retained.

Additional studies were then undertaken to further investigate the potential function of FBXO31 as a tumor suppressor. Ectopic expression of FBXO31 in the breast cancer cell lines MCF-7 and MDA-MB-468 resulted in a moderate inhibition of colony growth on plastic and reduced the proliferation of MCF-7 cells. Cell cycle analysis of the breast cancer cell lines ectopically expressing FBXO31 was consistent with a block in the cell cycle at G1. Investigation of endogenous FBXO31 expression in synchronized MCF-10A and SKBR3 cells show FBXO31 protein levels are at a maximum from late G2 to early G1 phase (Fig. 6A). The timing of FBXO31 destruction is consistent with APC-mediated degradation (17). Cell cycle analysis showed that inhibition of growth of breast cancer cell lines by ectopic FBXO31 expression is likely to be caused by cells not progressing normally past G1 phase of the cell cycle (Fig. 5). These results are consistent with the analysis of replicative senescence in primary cells where entry into senescence is associated with the accumulation of cells at G1 phase (24). In summary, the expression and functional studies and the role of FBXO31 in the cell cycle are consistent with a tumor suppressor role of FBXO31. Tumor haploinsufficiency for FBXO31 as a consequence of being in a region of frequent LOH in breast cancer is likely because tumors possess a low level of expression, and it is predicted that the protein is still subject to cell cycle variation. It is suggested that FBXO31 haploinsufficiency contributes to the acquisition by tumor cells of escape from senescence and the normal controls of cell proliferation. These conclusions are further supported by a breast cancer gene expression microarray study, which predicted the clinical outcome in breast cancer based on the combined relative expression profiles of 70 genes (25). Represented in these 70 genes is FBX031 (denoted as contig 51464).

The predicted amino acid sequence of the FBXO31 protein does not have significant homology to known proteins except for the presence of a 40-amino-acid F-box domain at the amino acid terminus. F-box proteins can exist as part of SCF ubiquitin ligase complexes that are involved in diverse cellular functions, including signal transduction, control of G1-S progression, and orderly execution of the cell cycle (26, 27). Skp1, Cullin-1, and Roc-1 are invariant proteins of the SCF complex, whereas the F-box proteins that bind to Skp1 are the components that impart functional specificity. For example, Skp2 specifically binds phosphorylated p27 resulting in a critical role in the degradation of p27 and therefore a control of S phase entry of the cell cycle (28).

Our results support the existence of a SCFFBXO31 complex. First, FBXO31 coimmunoprecipitated with Skp1, Cullin-1, and Roc1 (Fig. 6B) and copurified with FBXO31 in stable cell lines expressing tagged FBXO31 at near physiologic levels (Fig. 6C). Second, analysis of the nonmalignant breast cell line MCF-10A and the breast tumor cell line SKBR3 showed that FBXO31 is cell cycle regulated (Fig. 6A). Third, in silico analysis showed that FBXO31 protein has six minimal D-box (RxxL) motifs that are the hallmark of proteins degraded via the APC/C (17). FBXO31 shares all these features with the Skp2 protein present in the SCFSkp2-Cks1 complex (18), suggesting a functional similarity.

It is therefore proposed that FBXO31 is likely to form a functional SCFFBXO31 complex that would recruit and ubiquitinate specific proteins for subsequent degradation. Most F-box proteins bind to their substrates through regions located COOH-terminal to the F-box motif (29). F-box proteins notated as "FBXO" do not have recognizable substrate binding domains; however, as analysis of FBXO proteins progresses, new substrate binding domains are being recognized. For example, Fbx7 has been shown to recruit a substrate; the region responsible was shown to be a proline-rich region and was found in two other FBXO proteins (30). The COOH-terminal part of FBXO31 contains a 175-amino-acid region that is unusually rich (25%) in glycine and arginine residues that may represent the substrate binding region. Regions rich in glycine and arginine have been implicated as protein binding domains, although such reported domains have a more defined glycine/arginine repeat structure than present in FBXO31 (30, 31). It is proposed that the substrates of the SCFFBXO31 complex that are ubiquitinated and subsequently degraded are critical cell cycle proteins, and we are presently working towards their identification.


   Acknowledgments
 
Grant support: National Health and Medical Research Council of Australia grant 207703 (D. Callen) and Faculty of Health Sciences and Department of Medicine, University of Adelaide and the Royal Adelaide Hospital.

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 Dr. David Millband (Hanson Institute, Adelaide, Australia) for useful discussions and Dr. Martha Stampfer (Life Sciences Division, Lawrence Berkeley National Labs, Berkeley, CA) for generously providing the cell lines 184V, 48RS, and 184A1.


   Footnotes
 
Note: Z. Saif was on leave from the National Institute for Biotechnology and Genetic Engineering, Faisalabad, Pakistan.

Received 3/21/05. Revised 8/26/05. Accepted 9/29/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Miller BJ, Wang D, Krahe R, Wright FA. Pooled analysis of loss of heterozygosity in breast cancer: a genome scan provides comparative evidence for multiple tumor suppressors and identifies novel candidate regions. Am J Hum Genet 2003;73:748–67.[CrossRef][Medline]
  2. Cleton-Jansen A-M, Callen DF, Seshadri R, et al. Loss of heterozygosity mapping at chromosome arm 16q in 712 breast tumors reveals factors that influence delineation of candidate regions. Cancer Res 2001;61:1171–7.[Abstract/Free Full Text]
  3. Powell JA, Gardner AE, Bais AJ, et al. Sequencing, transcript identification, and quantitative gene expression profiling in the breast cancer loss of heterozygosity region 16q24.3 reveal three potential tumor-suppressor genes. Genomics 2002;80:303–10.[CrossRef][Medline]
  4. Kochetkova M, McKenzie OL, Bais AJ, et al. CBFA2T3 (MTG16) is a putative breast tumor suppressor gene from the breast cancer loss of heterozygosity region at 16q24.3. Cancer Res 2002;62:4599–604.[Abstract/Free Full Text]
  5. Zabarovsky ER, Lerman MI, Minna JD. Tumor suppressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers. Oncogene 2002;21:6915–35.[CrossRef][Medline]
  6. Reddy DE, Sandhu AK, DeRiel JK, Athwal RS, Kaur GP. Identification of a gene at 16q24.3 that restores cellular senescence in immortal mammary tumor cells. Oncogene 1999;18:5100–17.[CrossRef][Medline]
  7. Reddy DE, Keck CL, Popescu N, Athwal RS, Kaur GP. Identification of a YAC from 16q24 carrying a senescence gene for breast cancer cells. Oncogene 2000;19:217–22.[CrossRef][Medline]
  8. Kaur GP, Reddy DE, Zimonjic DB, de Riel JK, Athwal RS. Functional identification of a BAC clone from 16q24 carrying a senescence gene SEN16 for breast cancer cells. Oncogene 2004;24:47–54.[CrossRef]
  9. Kipreos ET, Pagano M. The F-box protein family. Genome Biol 2000;1:REVIEWS3002.[Medline]
  10. Stampfer MR, Yaswen P. Human epithelial cell immortalization as a step in carcinogenesis. Cancer Lett 2003;194:199–208.[CrossRef][Medline]
  11. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002;3:RESEARCH0034.[Medline]
  12. Dimri GP, Lee X, Basile G, et al. A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc Natl Acad Sci U S A 1995;92:9363–7.[Abstract/Free Full Text]
  13. Schmid I, Sakamoto KM. Analysis of DNA content and green fluorescent protein expression. In: Robinson JP, Darzynkiewitz Z, Dobrucki J, et al., editors. Current protocols in cytometry. New York: John Wiley & Sons; 2003. unit 7.16.
  14. Bonifacino JS, Estaban C, Dell A, Springer TA. Immunoprecipitation. In: Ausubel FM, Brent R, Kingston RE, et al., editors. Current protocols in molecular biology: New York: John Wiley & Sons; 2005. section 10.16.
  15. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5.[CrossRef][Medline]
  16. Jin J, Cardozo T, Lovering RC, Elledge SJ, Pagano M, Harper JW. Systematic analysis and nomenclature of mammalian F-box proteins. Genes Dev 2004;18:2573–80.[Free Full Text]
  17. Zur A, Brandeis M. Timing of APC/C substrate degradation is determined by fzy/fzr specificity of destruction boxes. EMBO J 2002;21:4500–10.[CrossRef][Medline]
  18. Bashir T, Dorrello NV, Amador V, Guardavaccaro D, Pagano M. Control of the SCF(Skp2-Cks1) ubiquitin ligase by the APC/C(Cdh1) ubiquitin ligase. Nature 2004;428:190–3.[CrossRef][Medline]
  19. Callen DF, Crawford J, Derwas C, Cleton-Jansen A-M, Cornelisse CJ, Baker E. Defining regions of loss of heterozygosity of 16q in breast cancer cell lines. Cancer Genet Cytogenet 2002;133:76–82.[CrossRef][Medline]
  20. Paige AJ. Redefining tumour suppressor genes: exceptions to the two-hit hypothesis. Cell Mol Life Sci 2003;60:2147–63.[CrossRef][Medline]
  21. Winston JT, Koepp DM, Zhu C, Elledge SJ, Harper JW. A family of mammalian F-box proteins. Curr Biol 1999;9:1180–2.[CrossRef][Medline]
  22. Cenciarelli C, Chiaur DS, Guardavaccaro D, Parks W, Vidal M, Pagano M. Identification of a family of human F-box proteins. Curr Biol 1999;9:1177–9.[CrossRef][Medline]
  23. Wade-Martins R, Frampton J, James MR. Long-term stability of large insert genomic DNA episomal shuttle vectors in human cells. Nucleic Acids Res 1999;27:1674–82.[Abstract/Free Full Text]
  24. Shelton DN, Chang E, Whittier PS, Choi D, Funk WD. Microarray analysis of replicative senescence. Curr Biol 1999;9:939–45.[CrossRef][Medline]
  25. van 't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–6.[CrossRef][Medline]
  26. DeSalle LM, Pagano M. Regulation of the G1 to S transition by the ubiquitin pathway. FEBS Lett 2001;490:179–89.[CrossRef][Medline]
  27. Reed SI. Ratchets and clocks: the cell cycle, ubiquitylation and protein turnover. Nat Rev Mol Cell Biol 2003;4:855–64.[CrossRef][Medline]
  28. Carrano AC, Eytan E, Hershko A, Pagano M. SKP2 is required for ubiquitin-mediated degradation of the CDK inhibitor p27. Nat Cell Biol 1999;1:193–9.[CrossRef][Medline]
  29. Cardozo T, Pagano M. The SCF ubiquitin ligase: insights into a molecular machine. Nat Rev Mol Cell Biol 2004;5:739–51.[CrossRef][Medline]
  30. Hsu JM, Lee YC, Yu CT, Huang CY. Fbx7 functions in the SCF complex regulating Cdk1-cyclin B-phosphorylated hepatoma up-regulated protein (HURP) proteolysis by a proline-rich region. J Biol Chem 2004;279:32592–602.[Abstract/Free Full Text]
  31. Friesen WJ, Dreyfuss G. Specific sequences of the Sm and Sm-like (Lsm) proteins mediate their interaction with the spinal muscular atrophy disease gene product (SMN). J Biol Chem 2000;275:26370–5.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Mol Cancer ResHome page
R. Kumar, J. Manning, H. E. Spendlove, G. Kremmidiotis, R. McKirdy, J. Lee, D. N. Millband, K. M. Cheney, M. R. Stampfer, P. P. Dwivedi, et al.
ZNF652, A Novel Zinc Finger Protein, Interacts with the Putative Breast Tumor Suppressor CBFA2T3 to Repress Transcription
Mol. Cancer Res., September 1, 2006; 4(9): 655 - 665.
[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 Kumar, R.
Right arrow Articles by Callen, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kumar, R.
Right arrow Articles by Callen, D. F.

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 Cell Growth & Differentiation