| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
1 Division of Genetics, Department of Pediatrics, University of California, Irvine, California; 2 Laboratory Medicine and Radiation Oncology, Cancer Genetics and Breast Oncology, UCSF Comprehensive Cancer Center, University of San Francisco, San Francisco, California; and 3 Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
UMS is an autosomal dominant condition with variable clinical features. Mammary gland abnormality ranges from hypoplasia to absence of breasts in males and females (11) . The disease gene was identified as TBX3 by linkage analysis (9 , 10) . TBX3 consists of an N-terminal half, a DNA binding domain, and a COOH-terminal half. The function of TBX3 is to repress gene expression (12, 13, 14) . The repression domain is located in the COOH-terminal half from amino acid 567 to 623 (15) . The nuclear localization signal was found at amino acids 292–297 (16) . The clinical features of UMS suggest that TBX3 is required for mammary gland growth and development. The organogenesis of breast development and oncogenesis share several common features (17) . Both processes require cell proliferation, modulation of cell death (apoptosis), invasion of surrounding tissues, and neovascularization. In organogenesis, the cells generally are relatively undifferentiated. Similarly, in cancer progression, cells that contribute to neoplasias tend to appear relatively undifferentiated or dedifferentiated (18) . In breast cancer, the uncontrolled growth of abnormal cells in breast tissue suggests that TBX3 may play a role in the disease pathogenesis.
Studies of TBX2 provide another line of evidence that calls for a detailed investigation of the role of TBX3 and its isoforms in breast cancer. It recently was found that TBX2, a homologue of TBX3, is amplified in a subset of breast cancers and breast cancer cell lines. TBX2 is active in inhibiting senescence of mouse embryo fibroblast (MEF) cells (14) . TBX3 and TBX2 share a similar expression pattern. In mouse mammary glands, tbx3 is expressed in the mammary buds, which are epithelial structures (19) ; tbx2 is expressed in mesenchyme. A significant homology between TBX2 and TBX3 was found not only in the T-box domain but also in the other domains, suggesting that TBX2 and TBX3 may share similar functions (10) . These similarities warrant an investigation of the roles of TBX3 in breast cancer.
Recent studies also have made TBX3 an attractive candidate for the control of mammary development and neoplastic progression. TBX3 (GenBank accession no. NM_005996) was found to be able to immortalize MEF cells (15 , 20) . Expression of p19ARF (Alternative Reading-Frame protein), an inhibitor of murine double minute 2, is repressed by TBX3 (16 , 20) . In mice, p19, a mouse homologue of human p14, interacts with and inhibits murine double minute 2 activity from shuttling p53 to the cytoplasm. Therefore, inhibition of p19 destabilizes p53. In humans, TBX3 also was found to inhibit p14 by directly binding to its promoter region (13) . This pathway plays an important role in proliferation and senescence. Wild-type TBX3, but not TBX3 with missense mutations that cause UMS, has the ability to repress the human p14 promoter (20) . TBX3 inhibits the p53 pathway from suppressing apoptosis, facilitates cell transformation, and blocks myogenic differentiation (16) . By cotransfection with myc or Ras, TBX3 was found to be able to transform MEF cells, which is correlated with its ability to repress p19 expression. The data suggest that overexpression of TBX3 may be associated with tumor formation by allowing breast epithelial cells to undergo additional rounds of cell division in an otherwise growth/proliferation-suppressive environment.
Multiple isoforms of TBX3 have been identified (10) . TBX3 + 2a (GenBank accession no. NM_016569) is particularly interesting. TBX3 + 2a was first identified in a human senescent fibroblast cell and infant human brain cDNA library and is produced by alternative splicing. TBX3 encodes a 743-amino acid peptide, with amino acids 105–285 in the T-box domain. Compared with TBX3, TBX3 + 2a carries an extra 20-amino acid sequence in the middle of the DNA binding domain (LAFPSDHATWQGNYSFGTQT, amino acids 220–240 of TBX3 + 2a). This structural difference suggests its distinctive function in gene regulation. It is possible that TBX3 and TBX3 + 2a act on different targets.
This study examined TBX3 and TBX3 + 2a expression patterns in multiple human and mouse tissues. We found that TBX3 and TBX3 + 2a are widely expressed. However, alternative splicing of TBX3 and TBX3 + 2a are tissue specific and species specific. Compared with TBX3, TBX3 + 2a has a distinct function because it accelerates senescence instead of inhibiting senescence of MEF cells. By using an in vitro oligonucleotide binding assay with nuclear extract that overexpressed TBX3, we found TBX3 and TBX3 + 2a have distinct binding targets. To examine the roles of TBX3 and TBX3 + 2a, we performed TaqMan quantitative PCR to quantify their expression levels on 28 different breast cancer cell lines. Our results suggest that TBX3 and TBX3 + 2a may play a role in human breast cancer.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmid Construction and Preparation of the Replication-Defective Retroviruses.
PFB-Neo and pFB-Neo-TBX3 were gifts from Dr. Peter Hurlin (University of Oregon). To construct pFB-Neo-TBX3 + 2a, a 5'-end fragment of TBX3 + 2a was PCR-amplified. The PCRs were carried out using pT77TBX3-long as a template and the following primers: the forward primer, (pT77TBX3R1): CGA CGA ATT CTA GAA ATA ATT TTG TTT; and the reverse primer, (TBX3 + 362 R): GCG AAG CTT GGG GAA GAA CGG CGG CTG GTG AC.
The resultant PCR product was digested with BamHI and EcoRI and replaced the 5'-end fragment of pFB-Neo-TBX3. The plasmid was sequenced and confirmed.
To produce the replication-defective retroviruses, the aforementioned constructs were transfected into packaging cell PT67 (gift of Xing Dai Lab at UCI). The transfection was performed with the Invitrogen Transfection Kit. The cells were grown in DMEM with 10% fetal bovine serum and selected with 500 µg/ml G418 for 2 weeks; selection started 2 days after transfection. Forty-eight hours before harvesting the viruses, the medium was changed back to G418-free medium. The supernatants were filtered, and the titers were determined in NIH mouse fibroblast cells by following the manufacturer’s instructions (Strategene, La Jolla, CA).
MEF Cell Preparation.
MEF cells were prepared and propagated in our laboratory. Thirteen-day-old mouse embryos were harvested. After removing the heads and internal organs, the embryos were passed through 20-gauge syringe needles and digested with trypsin for 5 min and then grown in 100-mm Petri dishes with DMEM supplanted with 10% FBS at 3 dishes/embryo. The cells were harvested and frozen to –150°C. These cells were considered as passage 1. To determine the effects of TBX3 and TBX3 + 2a on the senescence of MEF cells, replication-defective retroviruses were used to infect MEF cells (passage 1). The cells were grown in DMEM supplanted with 10% fetal bovine serum and selected with 500 µg/ml of G418 for 1 week, starting 48 h after infection. MEF cells infected with the aforementioned replication-defective retroviruses were seeded into six-well plates. The experiment was performed in triplicate, and the cells were maintained in DMEM with 500 µg/ml and 10% fetal bovine serum and split 1:3 every 3.5 days. The time was determined based on the fact that the cells would not be completely confluent (
80%). The cells were counted with a hemocytometer under a microscope, and the cell numbers were recorded. At passage 6, one set of the six-well dishes was stained with methylene blue.
Nuclear Extract Preparation and Gel Shift Assay.
Nuclear extract was prepared from MEF cells infected with replication-defective retroviruses at passage 4 or from NIH mouse fibroblast cells infected with replication-defective retroviruses. To perform a nuclear extract gel shift, 4 µg of nuclear extract protein were incubated with P32 end-labeled T-site oligonucleotide. The resultant mixture was subjected to electrophoretic separation in 6% polyacrylamide gel. The gels were dried and exposed to Kodak X-ray film (Rochester, NY).
Western Blot Analysis.
Western blot analysis was performed with nuclear extract of passage 4 MEF cells infected with pFB-Neo, pFB-Neo-TBX3, pFB-Neo-TBX3 + 2a. Forty µg nuclear extract protein were separated on 3% SDS-PAGE gel and transferred to a nitrocellulose membrane. The membrane was probed with monoclonal anti-HA antibody (Santa Cruz Biotechnology, Santa Cruz, CA) as the primary antibody and horse antimouse IgG conjugated with horseradish phosphotase as the second antibody. The membranes were developed with SuperSignal Western Femto Maximum Sensitivity Substrate Kit (Pierce, Rockford, IL).
Cell Lines and TaqMan Real-Time PCR.
The human breast epithelial cell line MCF-12A and the breast cancer cell lines MDA-MB-435, AU-565, BT474, BT 483, BT 549, CAMA-1, MCF 7, T47D, UACC-812, ZR-75–1, and ZR-75–30 were originally obtained from the American Type Culture Collection (Manassas, VA). HCC-70, HCC-1500, HCC-1599, HCC-1937, HCC-2185, and HCC3153 cell lines were obtained from Dr. Gazdar, University of Texas Southwest Medical Center (Ref. 21
; for information on the remaining cell lines, please refer to Refs. 22, 23, 24
). The cell lines were maintained with Dr. Gray at UCSF. For TaqMan real-time PCR, total cellular RNA was purified from malignant and nonmalignant breast cancer cell lines with TRIzol reagent (Invitogen). Ten ng of total RNA were used to perform TaqMan real-time PCR with the following primers and probes: the upper primer, 5'GGATGTCCAAAGTCGTCA3'; the lower primer, 5'GCTGGTATTTGTGCATGGAGTTCA3'; probe 1 (for TBX3), TTTGTGCATGGAGTTCAATATAGTAAATC; and probe 2 (TBX3 + 2a), TACCAAAACTATAATTCCCCTGCCACGT.
Probe 1 and probe 2 were labeled with Fam and Tet, respectively. 18s RNA TaqMan Kit was used to quantify the 18s RNA as an internal control, and the probe in 18s RNA kit was labeled as Vic (Applied Biosystems). TaqMan PCR was prepared with an ABI 3700 real-time PCR machine, and the data were analyzed with ABI software version 2.1 (Applied Biosystems).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, respectively, allowing us to amplify two isoforms in a single reverse transcription-PCR reaction and calculate their ratio. Reverse transcription-PCR was performed with total RNA from a variety of human tissues. The TBX3 T-box region is amplified with the SuperScript One-Step RT-PCR System (Invitrogen). As shown in Fig. 1A
, TBX3 and TBX3 + 2a were expressed in all of the tissues examined; the ratio between these two isoforms (TBX3 versus TBX3 + 2a) varied from tissue to tissue. This result suggested that alternative splicing could be tissue specific. TBX3 generally is the dominant isoform. The PCR products were quantified with the RNA labchip in an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). In the colon, the ratio of TBX3 to TBX3 + 2a is >99% (Lane 18, Fig. 1A
). However, in lung tissue, the ratio is close to 1:1 (Lane 7, Fig. 1A
). In normal breast tissue, the ratio is 2:1 (Lane 20). This result suggests that the ratio of TBX3 to TBX3 + 2a is tissue dependent. The gel-purified fragments were subcloned into a pCR TA cloning vector (Invitrogen) and sequenced. As predicted, 487-bp and 547-bp fragments originated from the TBX3 gene produced by alternative splicing (Fig. 1B)
. The wide expression pattern of TBX3 is consistent with the clinical features of multiple organ involvement in UMS and lethality of the homozygote knockout of tbx3, suggesting that TBX3 plays an important role in fundamental cellular function.
|
, although tbx3 also is dominant in all tissues, the ratio between tbx3 and tbx3 + 2a is consistent in a narrow range, from 32–38%. This result suggests that the alternative splicing pattern of TBX3 is species specific.
TBX3 and TBX3 + 2a Are Functionally Distinctive.
The structural difference in the DNA binding domains of TBX3 and TBX3 + 2a suggests their distinctive functions. TBX3 was found to be able to inhibit senescence in MEF cells, and we performed tests to determine whether TBX3 + 2a has the same function (15)
. In this experiment, MEF cells at passage 1 were infected with replication-defective retrovirus (pFB-Neo-TBX3 and pFB-Neo-TBX3 + 2a) to overexpress TBX3 and TBX3 + 2a, respectively. pFB-Neo was used as a control. As shown in Fig. 2
, TBX3 is able to immortalize MEF cells, suggesting inhibition of senescence. In contrast, in cells harboring TBX3 + 2a, acceleration of senescence occurs, and growth is slower than in the control (pFB-Neo). Cells infected with pFB-Neo-TBX3 have passed 15 passages without senescence. Cells infected with pFB-Neo and pFB-Neo-TBX3 + 2a stopped growing at eight and six passages, respectively (data not shown). The result is consistent with the growth curve (data not shown), suggesting that TBX3 and TBX3 + 2a have distinctive functions in inhibiting senescence of MEF cells.
|
-ATP end-labeled T-site (T-site, TGACACCTAGGTGTGAAATT; Refs. 15
, 25
). Carlson et al. (15)
previously showed TBX3 binded to this sequence. Our experiment confirmed the binding. As shown in Fig. 3
, Lane 2, TBX3 from NIH mouse fibroblast cells gave a background binding due to endogenous expression of tbx3, and overexpression of TBX3 increased such a binding (Lane 3). To test the specificity of TBX3 binding, an unlabeled oligonucleotide of the T-site was used for competitive binding. As shown in Lanes 4–7, the unlabeled T-site oligonucleotide could compete away the signals in a dosage-dependent manner. However, the binding with the nuclear extract of NIH mouse fibroblast cells infected with replication-defective pFB-Neo-TBX3 + 2a (Lane 8) showed no difference in binding activity compared with the endogenous binding (Lane 2). The binding activities of nuclear extract from TBX3 + 2a were compared with nuclear extract from NIH mouse fibroblast cells infected with control vector (pFB-Neo; data not shown). The result also confirmed the signal of Lane 8 (Fig. 3A)
is caused by endogenous expression of tbx3. To confirm that ectopically expressed TBX3 + 2a protein was present in the nuclear extract, the nuclear extract used for in vitro oligonucleotide binding was subjected to Western blot analysis. Because pFB-Neo-TBX3 and pFB-Neo-TBX3 + 2a carried the HA tag, the fusion protein was detected with monoclonal anti-HA antibody. As shown in Fig. 3B
, TBX3 + 2a is present in the nuclear extract used for in vitro oligonucleotide binding.
|
, TBX3 is able to immortalize MEF cells. TBX3 also is significantly homologous to TBX2, which was found amplified in a subset of primary human breast cancers. To test whether TBX3 and/or TBX3 + 2a expression is aberrant in breast cancer cell lines, we performed real-time PCR to quantify TBX3 and TBX3 + 2a in 28 breast cancer cell lines and 1 nonmalignant breast epithelial cell line. As shown in Fig. 4A
, the PCR primers were designed so that a single set of primers would amplify TBX3 and TBX3 + 2a. Probe 2 is complimentary to the exon 2a sequence and specifically detects TBX3 + 2a. In contrast, probe 1, which was used specifically to detect TBX3, sits at the junction of exon 2 and exon 3. Probe 1 and probe 2 are labeled as Fam and Tet, respectively. TaqMan quantitative PCR was performed in triplicate and normalized with 18 s RNA (labeled as Vic), which also was quantified with TaqMan quantitative PCR. The first line, MCF-12A, is a nonmalignant mammary gland epithelial cell line. The relative levels of TBX3 and TBX3 + 2a were compared in this cell line. As shown in Fig. 5B
, one-third of the breast cancer cell lines overexpressed TBX3, and the ratios of TBX3 + 2a to TBX3 also were aberrant (Fig. 5C)
. The majority of cell lines expressing a high level of TBX3 showed a decreased ratio of TBX3 + 2a/TBX3 except for CAMA-1 (Fig. 5B)
. These results suggest that TBX3 and TBX3 + 2a may play a role in breast cancer. Seven cell lines overexpressing TBX3 in quantitative PCR, one control cell line (MCF-12A), and two cell lines expressing comparable levels of TBX3 were analyzed by Southern blot analysis. The results are consistent with the findings in the real-time PCR (data not shown). These results suggest that the TaqMan PCR we used here is a reliable and much more sensitive method to quantify the TBX3 and TBX3 + 2a levels. We conclude that TBX3 is overexpressed and the ratios of TBX3 + 2a and TBX3 are altered in a subset of breast cancer cell lines.
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The differences between the DNA binding domains of TBX3 and TBX3 + 2a suggest binding to different targets. As shown in Fig. 6
, the 20 amino acid difference likely interferes with target binding, suggesting that TBX3 and TBX3 + 2a have distinct targets and functions, as shown in Fig. 2
and Fig. 3
. In an in vitro oligo-binding assay, TBX3 was found to bind to TGACACCTAGGTGTGAAATT (15)
, a target of the T-box peptide that was originally identified in random oligonucleotide selection experiments by using the bacterially expressed Xenopus laevis T-box domain as a protein source (25)
. The crystal structure of the Brachyury T-box bound to DNA confirms that the protein binds as a dimer (26)
. Our results showed that the extra 20 amino acids in the DNA binding domain of TBX3 + 2a abolished the binding to this target. One possible explanation for the failure of the TBX3 + 2a protein to bind to the TBX3 target would be the difference in its DNA binding domain. Identification of the targets for TBX3 + 2a is in progress.
|
The molecular mechanism of TBX3 overexpression still is unknown. Among 27 known estrogen receptor (ER) status breast cancer cell lines, 12 are ER positive and 15 are ER negative. Nine of 12 ER-positive cell lines overexpressed TBX3. BT-474, CAMA-1, MDA-MB-175VII, ZR-75–1, and ZR-75–30 are among the group expressing the highest TBX3 level and are all ER positive. Only 4 of 12 ER-negative call lines overexpress TBX3, and the levels are all relatively low. This result suggests that TBX3 expression is associated with ER status. The estrogen receptor is a transcription factor. A typical estrogen responsive element is palindromic GGTCANNNTGACC (28) . We have searched the TBX3 promoter region –5000 to +1000. There is no perfect palindromic sequence in this region. However, estrogen receptors may regulate gene expression through a variant sequence or non-nuclear events. Gene expression often is regulated through an estrogen responsive element-like sequence, imperfect palindromic element, or even half element. Multiple half-estrogen elements were found in the human TBX3 promoter region. The significance of these elements is unknown. Because TBX3 also is a transcription factor that could regulate other gene expression, we searched the T-box binding site in the promoter region for T-box 3 binding sites. The following T-box binding sites were found: –4686, –4489, –3740, and +516. The significance of these transcription binding sites again will need to be studied further in cultured cells and primary tumors.
TBX3 and TBX3 + 2a may become a biomarker for diagnosis and serve as a therapeutic target for additional treatment. There is a possibility that study of TBX3 and TBX3 + 2a in breast cancer will open a new avenue to understanding the pathogenesis of one of the leading causes of death in women, breast cancer.
|
ACKNOWLEDGMENTS |
|---|
|
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.
Requests for reprints: Taosheng Huang, Division of Genetics, Department of Pediatrics, Robert R. Sprague Hall 314, University California, Irvine, Irvine, CA 92697. Phone: 949-824-9346; Fax: 949-824-9776; E-mail: huangts{at}uci.edu
Received 2/19/04. Revised 5/ 6/04. Accepted 5/21/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Patil, S. A. Lee, E. Macias, E. T. Lam, C. Xu, K. D. Jones, C. Ho, M. Rodriguez-Puebla, and X. Chen Role of Cyclin D1 as a Mediator of c-Met- and {beta}-Catenin-Induced Hepatocarcinogenesis Cancer Res., January 1, 2009; 69(1): 253 - 261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rodriguez, E. Aladowicz, L. Lanfrancone, and C. R. Goding Tbx3 Represses E-Cadherin Expression and Enhances Melanoma Invasiveness Cancer Res., October 1, 2008; 68(19): 7872 - 7881. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Georges, G. Nemer, M. Morin, C. Lefebvre, and M. Nemer Distinct Expression and Function of Alternatively Spliced Tbx5 Isoforms in Cell Growth and Differentiation Mol. Cell. Biol., June 15, 2008; 28(12): 4052 - 4067. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R Copley The animal in the genome: comparative genomics and evolution Phil Trans R Soc B, April 27, 2008; 363(1496): 1453 - 1461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Yarosh, T. Barrientos, T. Esmailpour, L. Lin, P. M. Carpenter, K. Osann, H. Anton-Culver, and T. Huang TBX3 Is Overexpressed in Breast Cancer and Represses p14ARF by Interacting with Histone Deacetylases Cancer Res., February 1, 2008; 68(3): 693 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Abrahams, S. Mowla, M. I. Parker, C. R. Goding, and S. Prince UV-mediated Regulation of the Anti-senescence Factor Tbx2 J. Biol. Chem., January 25, 2008; 283(4): 2223 - 2230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Palena, D. E. Polev, K. Y. Tsang, R. I. Fernando, M. Litzinger, L. L. Krukovskaya, A. V. Baranova, A. P. Kozlov, and J. Schlom The Human T-Box Mesodermal Transcription Factor Brachyury Is a Candidate Target for T-Cell-Mediated Cancer Immunotherapy Clin. Cancer Res., April 15, 2007; 13(8): 2471 - 2478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-A. Renard, C. Labalette, C. Armengol, D. Cougot, Y. Wei, S. Cairo, P. Pineau, C. Neuveut, A. de Reynies, A. Dejean, et al. Tbx3 Is a Downstream Target of the Wnt/{beta}-Catenin Pathway and a Critical Mediator of {beta}-Catenin Survival Functions in Liver Cancer Cancer Res., February 1, 2007; 67(3): 901 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Guray and A. A. Sahin Benign breast diseases: classification, diagnosis, and management. Oncologist, May 1, 2006; 11(5): 435 - 449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-L. Cai, W. Zhou, L. Yang, L. Bu, Y. Qyang, X. Zhang, X. Li, M. G. Rosenfeld, J. Chen, and S. Evans T-box genes coordinate regional rates of proliferation and regional specification during cardiogenesis Development, May 15, 2005; 132(10): 2475 - 2487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhao, T. Tong, and Z. Zhang Expression of the Leo1-like domain of replicative senescence down-regulated Leo1-like (RDL) protein promotes senescence of 2BS fibroblasts FASEB J, April 1, 2005; 19(6): 521 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. W. Vance, S. Carreira, G. Brosch, and C. R. Goding Tbx2 Is Overexpressed and Plays an Important Role in Maintaining Proliferation and Suppression of Senescence in Melanomas Cancer Res., March 15, 2005; 65(6): 2260 - 2268. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
