| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Molecular Virology Section, Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland 20892-0460
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Studies over the last 15 years have revealed the p53 protein to be a critical "guardian of the human genome" (16) . Findings of rapid and robust p53 transcription and synthesis after exposure of mammalian cells to genotoxic agents (17, 18, 19) have contributed important insights to p53 function. On the basis of this type of evidence, one can infer a feedback linkage between DNA damage and induction of p53, which serves to enforce the repair of genetic lesions (20 , 21) . Analogously, one could similarly reason that a definition of the signals that provoke expression of MSC proteins would aid in elucidating their functions.
We identified previously and cloned the cDNA for the 718 amino acid human MAD1 protein and showed hsMAD1 to be a binding partner for the 205 amino acid MAD2 protein (10) . Although MAD2 has been studied extensively and found to function as a negative regulator of the anaphase promoting complex (22, 23, 24) , MAD1 has been investigated poorly, and its activities are largely unclear. Toward clarifying its role in cell growth and division, we sought to understand the regulated expression of MAD1 in human cells. Here we describe the cloning and the characterization of the hsMad1 promoter. Unexpectedly, expression of hsMad1 was found to not be M but to be G1 phase specific. Additionally, the hsMad1 promoter was not responsive to MTI but was activated strongly by phorbol ester. We also found enhanced activation of hsMAD1 expression by a gain-of-function p53 mutant. On the basis of our data, we propose hsMAD1 as a p53 downstream cellular factor that potentially links p53 to aneuploidy in human cells.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reporters and Assays.
A fragment containing the 5' most coding exon of hsMAD1 and 1.5-kb upstream nucleotides were amplified by PCR of human genomic DNA using the following primer pairs: 5'-TGGCGTCTTCCATGGGCCGCTCGCAGCCAGCTTGCCGCCGC-3' 5'-GGGATTAGATCTGGAACAATTAGGGAAGACGTATGGTC-3'. To construct hsMad1-promoter reporters, the amplified PCR product was digested with BglII and NcoI and cloned into pGL3-Basic vector (Invitrogen). The resulting construction was confirmed by DNA sequencing. Deletions and mutations of the hsMad1 promoter were introduced by PCR-based mutagenesis and also confirmed by sequencing. For p53 activation of hsMAD1, HeLa, HCT116, and HCT116p53-/- cells were transfected with hsMad1 reporter plasmids with either WT or mutant p53 expression vectors using Lipofectamine (Life Technologies, Inc.). Expression vectors for mutant p53 281G and 143A were gifts from Dr. Tslty (25)
. ß-galactosidase reporter pCMVß (Promega) was used to normalize transfection efficiency. Cells were harvested 24 h after transfection, and luciferase or ß-galactosidase activities were measured by using the substrates from Promega or Tropix, respectively. For comparisons of mRNA levels, matched tumor/normal expression array (BD Clontech) was examined to compare hsMAD1 mRNA expression between tumor tissue and the corresponding normal tissue from the same individual, according to manufacturer’s protocol. Briefly, the array blot was hybridized with hsMAD1 full-length cDNA probe labeled with 
-32P and exposed to X-ray film. Human ubiquitin cDNA probe provided by the manufacturer was used to ensure valid comparisons.
Cell Cultures.
The human colorectal cancer cell line HCT116 and the human cervix cancer cell line HeLa were obtained from American Type Culture Collection. HCT116 p53-/- cell line was a gift from Dr. Vogelstein (26)
. HCT116, HCT116 p53-/-, and HeLa cells were maintained in DMEM, supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 µg/ml streptomycin. Stable transformants of HeLa or HCT116 cells with the MAD1 1.5-kb promoter luciferase reporter plasmids were selected by 200 µg/ml Zeocin (Invitrogen), and Zeocin-resistant colonies were screened by luciferase assays. Several different clones were isolated and examined as a pool of five different clones for further analyses to exclude the possibility of clone-specific activities. Cell pools were treated with 100 ng/ml TPA (Sigma Chemical Co.), 20 ng/ml TNF- (Life Technologies, Inc.), 0.5 µM DAR (Sigma Chemical Co.), 1.01 µM DOR (Sigma Chemical Co.), 200 ng/ml NOC (Sigma Chemical Co.), 5 nM Actinomycin-D (Sigma Chemical Co.), or UV-irradiated (100 J/m2
), and then they were harvested every 3 h for luciferase activities. Primary PBMCs were from anonymous normal donors from the NIH blood bank. Normal foreskin fibroblasts Hs27 and Hs68 were from American Type Culture Collection.
[3H]Thymidine Incorporation.
HeLa MAD1-LUC and HCT116 MAD1-LUC cells were cultured without serum for 48 h. The serum-starved, G0-synchronized cells were then released with 2% serum. 1 µCi of [3H]Thymidine pulsed for 2 h; the cells were harvested every 3 h to measure incorporated [3H].
Computer Analysis for Human Genomic DNA.
Mad1 genomic sequence upstream of the 5' most coding exon was homology searched by BLAST. A putative transcription start site was identified within an 1.5-kb fragment of upstream sequence by an algorithm for promoter prediction of eukaryotic sequences.3
Cis-elements were predicted by TFSEARCH.4
Western Blotting.
Cell lysates were resolved by electrophoresis in 10% polyacrylamide gels and then transferred to Immobilon-P membranes (Millipore). The blots were incubated with anti-ß actin antibody (Sigma Chemical Co.), anti-p53 antibody Pab1801 (Santa Cruz Biotechnology), or anti-MAD1 polyclonal serum raised to amino acids 324–498 (10)
and visualized by chemiluminescence (Tropix).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The computer analysis also predicted a major transcription start site (designated by arrow, Fig. 1A
), which was confirmed experimentally by 5' RACE analysis of hsMad1 mRNAs (data not shown).
|
) was tested in transient transfections into three different human cells, HeLa, HCT116, and HCT116 p53-/-. Compared with background activity from the promoterless pGL3 luciferase vector, the -1502 to +35-pGL3 plasmid produced 50–100-fold higher luciferase units (Fig. 1B)
. This finding is consistent with the prediction that this fragment contains a housekeeping promoter.
To further define the minimal sequences required for transcription, eight promoter mutations generated by progressively deleting sequences from the 5' end were constructed. Activities from these hsMad1 mutant promoters were compared with control pGL3 plasmid, as well as a pGL3 plasmid containing the -1502 to +35 sequence inserted in a reversed orientation (Fig. 1B
, top and bottom lines). Luciferase activities were normalized to cotransfected CMV-ß-galactosidase reporter. Multiple assays performed separately in HeLa, HCT116, and HCT116 p53-/- cells yielded highly reproducible and consistent values (Fig. 1B)
. Among the deleted promoters, the -670 to +35, the -487 to +35, the -373 to +35, and the -313 to +35 constructions had high luciferase activities. 5' deletions commencing from either -133 or -73 significantly decreased activity. Further deleting the promoter to a -31 to +35 fragment reduced expression to that indistinguishable from background (Fig. 1B)
. Taken together, the results confirmed a promoter activity for the human 1.5-kb fragment and delineated an essential core promoter that spans positions -73 to +35.
Cis-acting Transcriptional Elements in the hsMad1 Promoter.
Next, we systematically queried the relative contribution to activity by various putative transcription factor motifs located between -133 and +35 (Fig. 1C)
. Using EcoRI linkers, we individually disrupted by linker substitution the Sp1 (-99/-95), GATA (-75/-68), Nkx-2 (-41/-36), GATA (-32/-24), and GC (-23/-12; +7/+17) elements (Fig. 1C)
. As shown in Fig. 1C
, disruption of Nkx-2- (GCAAGTG; -41/-36) and the promoter distal Sp1 (TGCGC; -99/-95) motifs reduced basal promoter activity. By contrast, mutation of neither GATA (-32/-24) nor the two transcriptional start site proximal GC boxes (-23/-12; +7/+17) affected promoter function. No individual EcoRI linker substitution had the drastic effect on transcription as that produced by truncating the promoter to a -31 to +35 fragment (Fig. 1C)
. These results suggest that multiple factors may interact redundantly within the -133 to +35 region to promote hsMad1 transcription.
Cell Cycle Phase-specific Expression of hsMAD1.
The definition of the hsMad1 promoter provided an opportunity to investigate its regulated expression during the cell cycle. Because hsMAD1 is an MSC component, a priori, one might expect its expression to be coincident with cellular entry into mitosis. To examine this possibility, we derived independent cell clones based on parental HeLa and HCT116 cells (HeLa MAD1-LUC and HCT116 MAD1-LUC), which were transfected stably with the 1.5-kb hsMad1 promoter luciferase reporter. For cell cycle expression assay, pools of individual HeLa MAD1-LUC or HCT116 MAD1-LUC clones were examined.
Cells were first serum starved for 48 h and then released into complete medium containing 2% fetal bovine serum and sampled every 3 h thereafter for luciferase activity. Duplicate sets of cells were identically processed in parallel except that 1 µCi of [3H]thymidine was added to one set for 2 h before cell sampling. Incorporation of [3H] into acid-insoluble form was monitored for cellular DNA synthesis.
HeLa MAD1-LUC cells, after 48 h of serum starvation, were largely quiescent as reflected by both low luciferase activity and [3H] counts (Fig. 2
, left). When cells were released from this G0-like state by exposure to fetal bovine serum, luciferase activity and [3H]incorporation rose progressively. In HeLa MAD1-LUC cells, luciferase values plateaued between 6 and 15 h after release, whereas [3H]incorporation reached a maximum at 15 h and dropped precipitously thereafter (Fig. 2
, left). Similar profiles were observed in HCT116 MAD1-LUC cells (Fig. 2
, right). In both HeLa MAD1-LUC and HCT116 MAD1-LUC cells, the peak rise in luciferase activity chronologically preceded the corresponding [3H]curves. The simplest interpretation of these profiles is that hsMad1 expressed preferentially before the S phase of the cell cycle (i.e., in G1). Thereafter, the concordant cessation of luciferase activity and [3H]incorporation indicated that the hsMad1 promoter was minimally active in post-S phases (i.e., in G2 or M). These results address transcription from the hsMad1 promoter. Elsewhere, we have found that synthesis of the hsMAD1 protein is also under translational and post-translational controls (10)
.
|
), DNA-damaging agent (UV light), RNA-synthesis inhibitor (actinomycin D), topoisomerase inhibitors (DAR and DOR), and microtubule inhibitory agent (NOC).
MTIs damage the mitotic spindle and activate MSC function (4
, 5) . Ordinarily, one might expect NOC treatment to induce hsMad1 luciferase activity. Surprisingly, neither HeLa MAD1-LUC nor HCT116 MAD1-LUC cells changed in baseline luciferase activity when exposed to NOC (Fig. 3)
. Indeed, in HeLa MAD1-LUC cells, to the exclusion of all other agents, TPA was the sole treatment that robustly activated the hsMad1 promoter (Fig. 3
, left). A similar induction by TPA was also verified in HCT116 MAD1-LUC cells (Fig. 3
, right). However, here, we found that topoisomerase inhibitors, DAR and DOR, also enhanced hsMad1 luciferase expression. We do not understand fully what cell-specific differences between HCT116 and HeLa might explain the differential DAR/DOR response; however, independent of this result, both cells demonstrated that the hsMad1 promoter is strongly stimulated by mitogenic agent, TPA, but not by microtubule disrupting agent, NOC.
|
).
|
) was assessed by Western blotting. We internally normalized hsMAD1 from PBMCs to cellular ß-actin protein and then compared these values to corresponding values from transformed Jurkat (a human lymphoblastic leukemia cell line) and K562 (a human chronic myelogenous leukemic cell line) cells. In side-by-side comparisons, all three PBMCs conveyed similarly low values for hsMAD1. By contrast, Jurkat and K562 cells had normalized hsMAD1 values that exceeded their untransformed counterparts by 5–20-fold.
Many reasons could explain differences between cell lines and primary cells. To more directly correlate amounts of hsMAD1 with cellular proliferation, we induced PBMCs to proliferate by treatment with IL-2 and PHA. Exposure to IL-2 + PHA activates otherwise quiescent mononuclear cells to divide. When compared with unstimulated PBMCs (Fig. 4B
, Lane 2), proliferating PBMCs treated with IL-2 + PHA for 4 days (Fig. 4B
, Lane 4) increased hsMAD1 expression by >6-fold and showed a normalized value that actually exceeded that found for Jurkat cells (Fig. 4B
, Lane 1). That enhanced hsMAD1 expression correlated with cellular proliferation is fully consistent with above findings (Fig. 3)
that the hsMad1 promoter responds efficiently to mitogenic stimulation.
To extend the generality of the findings beyond PBMCs, we also checked a second type of primary human cells, foreskin fibroblasts. Two independent foreskin fibroblast cultures (Hs27 and Hs68; Fig. 4C
) were examined. They (Fig. 4C
, Lanes 2 and 3
) were compared with immortalized African green monkey kidney cells (CV-1; Fig. 4C
, Lane 4), SV40-transformed CV-1 cells (Cos; Fig. 4C
, Lane 1), and human colon (SW480; Fig. 4C
, Lane 5) and cervical (HeLa; Fig. 4C
, Lane 6) cancer cells. In agreement with PBMC findings, both foreskin fibroblasts had much reduced levels of hsMAD1 when compared with the adherent immortalized/transformed cells. Separately, we confirmed that Jurkat and K562 do propagate more rapidly than unstimulated PBMCs and that Cos, CV-1, SW480, and HeLa cells divide faster in culture than either Hs27 or Hs68 (data not shown).
To further evaluate hsMAD1 expression and cellular proliferation, we next examined eight pairs of normal tumor stomach tissue cDNA libraries. The normal tumor cDNA pairs were assessed for MAD1 mRNA levels and then compared after normalizing for control "ubiquitin" mRNA. In four of eight pairs, normalized MAD1 expression was clearly higher in the tumor tissue (Fig. 4D)
, whereas no normal sample had obviously higher MAD1 levels than its tumor counterpart. Taken together with the results in Fig. 4, A–C
, the data provide a consistent correlation between heightened hsMAD1 expression and increased cellular proliferation.
A Gain-of-Function p53 Mutant Up-Regulated Expression of hsMad1.
In considering the hsMAD1 expression profile in cells, we recalled previously that hsMad1 transcript was 1 of only 14 cellular mRNAs that were induced robustly by a p53-expressing adenovirus vector in colorectal cancer line DLD-1 (28)
. Although not recognized at the time of publication, the p53-induced gene 9 (PIG9) in Polyak et al. (28)
was, in fact, hsMad1. Because other studies have linked aberrant cellular ploidy with p53 mutations (25
, 29) and because we observed heightened hsMAD1 expression in cancer cells (Fig. 4)
, we sought to understand better the possibility that p53 might be an upstream regulator of hsMad1. Accordingly, we compared WT p53 and two p53 mutants (143A and 281G; Ref. 25
) for respective abilities to activate hsMad1. When p53WT, p53 143A, and p53 281G were introduced separately into HeLa cells, ambient expression of hsMAD1 was perturbed minimally by either p53 WT or p53 143A. By contrast, hsMAD1 was increased significantly (
9.5-fold) by p53 281G in a dose-dependent manner (Fig. 5A)
.
|
, although both p53 WT and p53 143A perturbed expression 2-fold, p53 281G induced expression in excess of 6-fold.
To understand how p53 281G activated the hsMad1 promoter, we separately assayed in HeLa cells the full-length hsMad1 promoter luciferase plasmid and eight hsMad1 promoter mutants (Fig. 5C)
. Reproducibly, deletion mutants -133 to +35 and -73 to + 35, and linker insertion mutants -100/-95, -71/66, -41/-36, and -30/-25, maintained responsiveness to p53 281G. By contrast, individual disruptions in two GC boxes (-20/-15 and +9/+14) dramatically abolished p53 281G responsiveness. Interestingly, these two transcription site proximal GC boxes bear resemblance to the canonical 5' XXXC(A,T)(T,A)GYY 3' (where X = purines and Y = pyrimidines) p53-responsive motif (30)
, as well to the p53-responsive GC sequence first characterized for the SV40 GGGCGG sequences (18)
.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Here we sought to characterize cis- and trans-factors that influence the expression of the hsMad1 promoter in an attempt to better understand how MAD1 functions. We began with the identification of a 1.5-kb human genomic fragment as the provisional hsMad1 promoter. Analysis of this sequence revealed it to be TATA-less and highly GC rich with significant promoter activity (Fig. 1)
. Within this promoter, one can visualize many degenerative forms of the canonical Sp1-GGGCGG-motif or its closely related hexanucleotide GC box sequences (Refs. 27
and 33
; Fig. 1
). On the basis of the absence of a TATA element and the presence of multiple GC motifs, we interpret the hsMad1 promoter to be characteristic of promoters for housekeeping genes.
In studying the regulated expression of hsMad1, we came across several unexpected findings. Contrary to our a priori assumption that an MSC protein should be expressed chronologically proximal to the onset of mitosis, we found the hsMad1 promoter to be basally active in G2 and M but preferentially active in the G1 phase of the cell cycle (Fig. 2)
. Additional observations that spindle disrupting MTI, NOC, failed to induce, whereas mitogenic agent, TPA, did robustly induce the hsMad1 promoter were also unanticipated (Fig. 3)
. Add to these results a correlation between levels of hsMAD1 in cells and the proliferation state of the cells (Fig. 4)
, one is then struck by the possibility that hsMAD1 likely has cellular functions in addition to its expected role in the MSC. Indeed, we do not currently understand the gamut of functions described by MSC proteins. In this vein, it is instructive to note that mice lost for MSC function through "knockout" of the murine Mad2 gene are embryonically lethal (6)
. The embryonically lethal phenotype is formally inconsistent with proteins that serve purely an inducible checkpoint function, as exemplified by the normal development to maturity of p53 knockout mice (34)
. Hence, our current results coupled with emerging observations5
that overexpressed hsMAD1 promotes cyclin D expression and cell cycle progression in serum-arrested cells suggest that this protein has function(s) essential for cellular proliferation beyond that needed for MSC activity.
It is intriguing that hsMad1 is activated by WT p53 (28)
and activated even better (in our experiments) by a gain-of-function p53 mutant (Fig. 5)
. Mutation of p53 is the most common genetic alteration found in human cancers (35)
, and tumors that express gain-of-function p53 mutants have a worse prognosis than p53-null cancers (36)
. The tumor-promoting activity of mutant p53 emanates in part from its distinct transactivating ability when compared with WT p53. Thus, several cellular proliferation-related genes, such as c-myc, epidermal growth factor receptor, proliferating cell nuclear antigen, and multidrug-resistance-1, have been reported to be activated selectively by mutant p53 but not WT p53 (37)
. Recognizing that activation of genes by p53 is somewhat cell background dependent (38)
, within the context of the cells used in this study, our findings suggest that hsMad1 is another growth-related cellular gene activated preferentially by some p53 mutants. The finding that certain stomach cancers have higher levels of hsMad1 transcripts (Fig. 4D)
may indicate the presence of such gain-of-function p53 mutations in those tumors.
Mechanisms as to how mutant p53 transactivates cellular genes appear to be variable and not understood completely. The IL-6 promoter, e.g., is transactivated by p53 mutant protein through CAAT/enhancer binding proteinß motifs (39)
. In the case of the hsMad1 promoter, we find it to be activated by p53 281G through two start site proximal GC boxes (Fig. 5)
, which resemble closely the WT p53-responsive GC sequence described for the SV40 promoter (40)
. Indeed, our results, consistent with the findings of Polyak et al. (28)
assayed in a different cellular background, indicate that WT p53 can also slightly activate hsMad1, albeit (in our context) to a level of <2-fold (Fig. 5)
.
In considering the biological relevance of activation by p53, we note the reported linkage between cellular aneuploidy and aberrant p53 function (25 , 29 , 41) . Indeed, interestingly, p53 281G falls into a category of gain-of-function p53 mutants that specify for a RSC phenotype (25) . Accordingly, one could view hsMAD1 as a p53-downstream MSC factor whose activity is modulated differentially by p53 mutants. The unexpected finding here that hsMAD1, in addition to its MSC function, has a proliferation-associated role raises the possibility that WT-p53 and RSC-p53 activate divergent hsMAD1 properties. Conceivably, mild hsMad1 induction represents an attempt by WT-p53, in response to chromosomal mis-segregation, to invoke the MSC function of hsMAD1. On the other hand, robust induction of hsMad1 by RSC-p53 may activate a proliferative hsMAD1 activity that overrides its MSC function. Pending a better understanding of hsMAD1’s non-MSC function(s), additional studies are needed to clarify how the interplay between hsMAD1 and WT, or mutant, p53 might either mitigate or enhance the propensity for development of aneuploidy in mammalian cells.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be addressed, at Building 4, Room 306, 9000 Rockville Pike, Bethesda, MD 20892-0460. Phone: (301) 496-6680; Fax: (301) 480-3686; E-mail: kj7e{at}nih.gov 
2 The abbreviations used are: MSC, mitotic spindle assembly checkpoint; MAD, mitotic arrest deficiency; MTI, microtubule inhibitor; RACE, rapid amplification of cDNA end; TNF, tumor necrosis factor; TPA, 12-O-tetradecanoylphorbol-13-acetate; CMV, cytomegalovirus; DAR, daunorubicin; DOR, doxorubicin; NOC, nocodazole; PBMC, peripheral blood mononuclear cell; IL, interleukin; PHA, phytohemagglutinin; WT, wild type; RSC, relaxed spindle checkpoint.
3 Internet address: http://www.fruitfly.org/seq_tools/promoter.html.
4 Internet address: http://www.cbrc.jp/research/db/TFSEARCHJ.html.
5 K-T. Jeang, unpublished observations.
Received 11/21/01. Accepted 2/27/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Mizuarai, K. Yamanaka, and H. Kotani Mutant p53 Induces the GEF-H1 Oncogene, a Guanine Nucleotide Exchange Factor-H1 for RhoA, Resulting in Accelerated Cell Proliferation in Tumor Cells. Cancer Res., June 15, 2006; 66(12): 6319 - 6326. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Tabakin-Fix, I. Azran, Y. Schavinky-Khrapunsky, O. Levy, and M. Aboud Functional inactivation of p53 by human T-cell leukemia virus type 1 Tax protein: mechanisms and clinical implications Carcinogenesis, April 1, 2006; 27(4): 673 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Zhang, M. Li, W. Ma, Y. Hou, Y.-H. Li, S.-W. Li, Q.-Y. Sun, and W.-H. Wang Localization of Mitotic Arrest Deficient 1 (MAD1) in Mouse Oocytes During the First Meiosis and Its Functions as a Spindle Checkpoint Protein Biol Reprod, January 1, 2005; 72(1): 58 - 68. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Jeong, H.-J. Shin, S.-J. Kim, G.-H. Ha, B.-I. Cho, K.-H. Baek, C.-M. Kim, and C.-W. Lee Transcriptional Abnormality of the hsMAD2 Mitotic Checkpoint Gene Is a Potential Link to Hepatocellular Carcinogenesis Cancer Res., December 1, 2004; 64(23): 8666 - 8673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. S. Chun and D.-Y. Jin Transcriptional Regulation of Mitotic Checkpoint Gene MAD1 by p53 J. Biol. Chem., September 26, 2003; 278(39): 37439 - 37450. [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 |
