The Oncologist, Vol. 5, No. 6, 501-507,
December 2000
© 2000 AlphaMed Press
NCI All Ireland Cancer Conference Proceedings |
Uncovering Functionally Relevant Signaling Pathways Using Microarray-Based Expression Profiling
D. Paul Harkin
The Queen's University of Belfast Cancer Research Centre and Belfast City Hospital Trust, Belfast, Ireland
Correspondence:
D. Paul Harkin, M.D., The Queen's University of Belfast Cancer Research Centre, Belfast City Hospital, Lisburn Road, Belfast 9, Ireland. Telephone 44-1232-263911; Fax: 44-28-90-263744.
 |
ABSTRACT
|
|---|
The introduction of microarray technology to the scientific and medical communities has fundamentally altered the way in which we now address basic biomedical questions. Microarrays technology facilitates a more complete and inclusive experimental approach where alterations in the transcript level of entire genomes can be simultaneously assayed in response to a variety of stimuli. Conceptually different approaches to the development of microarray technology have resulted in the generation of two different array formats: oligonucleotide arrays and cDNA arrays.
The application of microarray and related technologies to identify specific targets of defined genes that have clearly been implicated in cancer progression requires a specific experimental approach. The objective of this approach is to define changes in transcriptional profile that occur in response to modulating the expression level of the gene to be studied. The resulting altered expression profile can then be viewed as a blueprint by which that gene effects its cellular function.
We have used oligonucleotide array-based expression profiling in collaboration with Affymetrix to identify downstream transcriptional targets of the BRCA1 tumor-suppressor gene as a means of defining its function. BRCA1 has been implicated in at least three functional pathways, namely, mediating the cellular response to DNA damage, as a cell cycle checkpoint protein and in the regulation of transcription. The physiological significance of these properties and their implications for the function of BRCA1 as a tumor-suppressor gene remain to be defined.
Key Words. Oligonucleotide arrays • Expression profiling • BRCA1 • Target genes • GADD45
|
INTRODUCTION
|
|---|
Microarrays are systematic arrays of cDNAs or oligonucleotides of known sequence that are printed or synthesized at discrete loci on a glass or silicon surface (Fig. 1
). The introduction of microarray technology to the scientific and medical communities has fundamentally altered the way in which we now address basic biomedical questions. Microarrays technology facilitates a more complete and inclusive experimental approach where alterations in the transcript level of entire genomes can be simultaneously assayed in response to a variety of stimuli. This genome-wide approach to transcriptional analysis or "transcriptional profiling" provides comparative data on the relative expression level of individual transcripts within an organism, and relates this to alterations that occur as a consequence of a defined cellular stimulus. This new holistic approach has generated additional problems relating to data management and a requirement for sophisticated methods of analysis to extract biologically relevant data from the mass of primary information.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. Two main microarray formats are currently available. A) Oligonucleotide arrays which were initially pioneered by Affymetrix are generated using a combination of oligonucleotide synthesis and photolithography. A photolithographic mask is used to generate localized areas of photodeprotection on a glass slide that has been coated with linker molecules containing a photochemically removable protecting group. Specific dNTPs are then chemically coupled at the deprotected site facilitating the synthesis of specific oligonucleotide sequences. A series of different photolithographic masks are used with an intervening dNTP coupling reaction to generate the desired array. Other methods of generating oligonucleotide arrays rely on depositing a presynthesised oligonucleotide onto the array. B) cDNA arrays are generated by robotically printing double-stranded cDNAs of known sequence onto a glass slide at a predetermined spatial orientation. The printing is achieved by a computer controlled robotic arm that moves in three dimensions and which contains up to 12 pen tips which deposit a precise volume of purified cDNA onto the glass which has been coated with amino silanes or amino-reactive silanes. Other methods of printing are also available such as piezo or ink-jet delivery. The spotted cDNA is then cross-linked to the slide by UV irradiation.
|
|
Conceptually different approaches to the development of microarray technology have resulted in the generation of two different array formats.
|
OLIGONUCLEOTIDE ARRAYS
|
|---|
The first oligonucleotide arrays were generated using a combination of oligonucleotide synthesis and photolithography to synthesize specific oligonucleotides in a predetermined spatial orientation on a solid surface such as glass or silicon [1, 2]. Affymetrix (Santa Clara, CA; http://www.affymetrix.com/), has pioneered this technology and currently has generated a number of different commercially available array products including human, mouse and various model organisms. The arrays are generated by attaching synthetic linker molecules that have been modified with a photochemically removable protecting group to a solid support such as glass or silicon. A photolithographic mask is then applied through which ultraviolet (UV) light is passed generating localized areas of photodeprotection to which protected dNTP are then attached in a chemical coupling reaction. Each photolithographic mask applied generates different areas of photodeprotection on the solid substrate and, using a combination of these masks with an intervening chemical coupling step, the desired probes are synthesized at the sites specified in the original design [3]. An additional feature of oligonucleotide arrays is that each gene included on the array is represented by up to 20 different oligonucleotides spanning the entire length of the coding region of that gene. Moreover each of these oligonucleotides is paired with a second mismatch oligonucleotide in which the central base in the sequence has been changed. The combination of probe redundancy and inclusion of a mismatched control sequence greatly reduces the rate of false positives obtained from this type of approach.
For expression profiling-based comparisons, fluorescently labeled probes are generated from test and reference samples. For Affymetrix-based oligonucleotide arrays, fluorescent probes are generated by reverse transcribing total RNA using an oligo-dT primer containing a T7 polymerase site. Amplification and labeling of the cDNA probe is achieved by carrying out an in vitro transcription reaction in the presence of a biotinylated dNTP, resulting in the linear amplification of the cDNA population (approximately 30-100-fold). The biotin-labeled cRNA probe generated from test and reference samples is then hybridized to separate oligonucleotide arrays, followed by binding to a streptavidin-conjugated fluorescent marker. Detection of bound probe is achieved following laser excitation of the fluorescent marker and scanning of the resultant emission spectra using a scanning confocal laser microscope. The differential fluorescent signal is then represented as alterations in transcriptional profile between the two samples compared.
|
CDNA ARRAYS
|
|---|
cDNA arrays have been pioneered by the Brown Laboratory at Stanford University [4] (http://cmgm.stanford.edu/pbrown/mguide/index.html) and are generated by robotically printing double-stranded cDNA onto a solid support such as glass silicon or nylon. The success of this technology has come in part from the development of robots capable of precisely depositing a defined quantity of arrayed cDNAs at predetermined locations on the solid support. The other essential ingredient is access to sequence verified and array formatted cDNA clones. This is necessary to facilitate the rapid manufacture of arrays and ensure that the location and identity of each cDNA on the array is known. Prior to printing, cDNA clones which are stored as glycerol stocks, are polymerase chain reaction (PCR) amplified using well-defined vector-based primers flanking the cDNAs. The amplified products are then gel-purified and re-arrayed into 96- or 384-well plates followed by printing onto the desired solid support. Currently sequence-verified and array-formatted cDNA clone sets are available from Incyte Genomics (Palo Alto, CA; http://www.synteni.com/) and Research Genetics (Huntsville, AL; http://www.resgen.com/) for a cost of £6 per clone, and are supplied as glycerol stocks in 96- or 384-well plates.
For cDNA-based expression profiling experiments, total RNA is first extracted from the experimental samples to be compared and fluorescently labeled with either cye3- or cye5-dUTP in a single round of reverse transcription. Cye3 and cye5 are preferentially used because they are readily incorporated by reverse transcription; they exhibit good photostability and most importantly, are widely separated in terms of their excitation and emission spectra. The fluorescently labeled cDNA probes are hybridized to a single array in a competitive hybridization reaction. Detection of hybridized probes is achieved by laser excitation of the individual fluorescent markers, followed by scanning using a confocal scanning laser microscope. The raw data are represented as a normalized ratio of cye3:cye5 and digitally color coded such that red represents genes transcriptionally upregulated in the test versus the reference, green represents genes downregulated and yellow represents those genes that exhibit no difference between test and reference samples.
|
APPLICATIONS OF MICROARRAY TECHNOLOGY
|
|---|
Microarray technology is currently being used in a high-throughput approach to study gene expression and sequence variation on a genomic scale. This review is primarily focused on the use of microarray technology in RNA expression studies as a means of identifying signaling pathways regulated by key genes implicated in tumorigenesis. The term expression profiling, however, encompasses a wide variety of different experimental strategies that use alterations in transcriptional profile as a means to explain the molecular basis of how specified experimental models respond to particular stimuli or changes in homeostasis, whether induced or occurring naturally. The questions being asked determine to a large extent the design of the array experiment. In tumor-profiling experiments the objective is to identify transcriptional changes that may be causative or occur as a consequence of the transition from the normal to the tumor phenotype. This approach also allows the identification of molecular fingerprints that facilitate the classification of different tumor types, thereby providing a molecular means of diagnosis in patients. In these approaches it is essential to carry out multiple independent experiments in a large cohort of samples to isolate biologically relevant changes from spurious results that may arise as a result of genetic heterogeneity between individual samples. A number of studies have recently been published which clearly illustrate this concept. Alizadeh et al. [5] have successfully used this approach to identify molecularly distinct subclasses of diffuse large B-cell lymphoma that were not distinguishable by conventional means. A similar study has identified a molecular fingerprint comprising approximately 50 genes, isolated from a total of over 6,000 that can reliably differentiate between acute myeloid leukemia and acute lymphoblastic leukemia [6]. Both of these studies relied on multiple individual experiments using a variety of different tissue samples to reproducibly identify tumor type-specific molecular determinants.
|
IDENTIFICATION OF SPECIFIC TRANSCRIPTIONAL TARGETS
|
|---|
The application of microarray and related technologies to identify specific targets of defined genes that have clearly been implicated in cancer progression requires a different experimental approach. The objective of this approach is to define changes in transcriptional profile that occur in response to modulating the expression level of the gene to be studied. The resulting altered expression profile can then be viewed as a blueprint by which that gene effects its cellular function. In this approach the investigator has greater control over the question being asked as the experimental variables are artificially controlled. Therefore multiple repetitions of the experiment are not required and in general the experiment carried out in duplicate is sufficient to generate reliable data. Comparing transcriptional profiles as a means of defining downstream signaling pathways has previously been validated by various researchers using alternative techniques such as differential display [7] and serial analysis of gene expression [8]. The advent of microarray technology has, however, dramatically simplified the experimental design required in this approach allowing the simultaneous identification of all potential targets. Currently the major drawback to using arrays in this manner is that it is entirely dependent on the state of knowledge of the genome under investigation. However, the rapid completion of the human genome sequencing project and the speed at which various other genomes have been or are in the process of being sequenced will in due course make this issue redundant. The major criticism leveled at this type of approach is that the expression levels achieved in these artificial systems are not physiologically relevant. While this is undeniably true, the onus is on the investigator to devise additional experiments to confirm that genes identified in this manner represent physiologically relevant targets.
|
THE BRCA1 TUMOR-SUPPRESSOR GENE
|
|---|
We have used oligonucleotide array-based expression profiling in collaboration with Affymetrix to identify downstream transcriptional targets of the BRCA1 tumor-suppressor gene as a means of defining its function [9]. BRCA1 is mutated in the germline of women with a genetic predisposition to breast and ovarian cancer [10]. Germline mutations of BRCA1 are found in half of breast-ovarian cancer pedigrees and approximately 10% of women with early onset of breast cancer, irrespective of family history [11]. Somatic inactivation of BRCA1 is rare in sporadic breast cancers [12], however, mutations have been observed in approximately 10% of sporadic ovarian cancers, suggesting potentially distinct genetic mechanisms for these two diseases [13]. It has also been reported that reduced BRCA1 protein expression is observed in the majority of sporadic breast cancers, indicating that epigenetic mechanisms may also play a role in regulating BRCA1 expression [14]. Most mutations identified to date result in premature truncation of the BRCA1 protein and, consistent with its role as a tumor-suppressor gene, exhibits loss of heterozygosity of the wild-type allele in tumor specimens.
BRCA1 encodes a unique 1863 amino acid phosphoprotein that is predominantly localized to the nucleus. An N-terminal ring finger motif mediates a physical interaction with BARD1, another ring finger protein of unknown function [15]. Sequence analyses have identified a previously undescribed C-terminal BRCT motif, postulated to play a role in cell cycle checkpoint control in response to DNA damage [16]. Consistent with these observations BRCA1 becomes hyperphosphorylated in response to various DNA damaging agents including 
-irradiation, an effect that is mediated in part by the ataxia telangiectasia (ATM) [17] and chk2 kinases [18]. BRCA1 has also been shown to colocalize with RAD51, the mammalian homologue of bacterial recA, involved in homologous recombination and repair of double-strand breaks in DNA following ionizing radiation [19]. Moreover, a defect in transcription-coupled repair of oxidative-induced DNA damage in mouse embryo fibroblasts with attenuated BRCA1 function [20], suggests that BRCA1 may play a more general role in mediating the cellular response to DNA damage.
BRCA1 has also been implicated in cell cycle checkpoint control, becoming hyperphosphorylated during late G1 and S and transiently dephosphorylated early after M phase [21]. Overexpression of BRCA1 has been shown to induce a G1/S arrest in human colon cancer cells [22] and genetic instability has been observed in BRCA1 exon 11 isoform-deficient cells resulting from a defective G2/M checkpoint and centrosome amplification [23].
A specific role has been postulated for BRCA1 in transcriptional regulation. The C-terminal domain has been shown to contain a potent transactivation domain when fused to a heterologous DNA binding motif [24]. BRCA1 has also been shown to be a component of the RNA polymerase II holoenzyme complex, potentially through its ability to associate with RNA helicase A [25]. In addition BRCA1 has been shown to physically associate with the transcriptional regulators p53 [26], CtIP [27], c-myc [28], and the histone deacetylases HDAC1 and HDAC2 [29].
BRCA1 therefore has been implicated in at least three functional pathways, namely, mediating the cellular response to DNA damage, as a cell cycle checkpoint protein and in the regulation of transcription. The physiological significance of these properties and their implications for the function of BRCA1 as a tumor-suppressor gene remain to be defined.
|
GENE EXPRESSION SYSTEMS
|
|---|
A problem with generating model expression systems to study gene function by expression profiling is that the gene under investigation may, under conditions of forced expression, have a negative effect on cell growth. This problem is illustrated by the difficulty in generating cell lines with constitutive overexpression of genes such as p53 or BRCA1, where forced expression can lead to growth suppression or cell death. In order to circumvent this problem we used the tet-off inducible expression system [30] to generate cell lines with highly regulated inducible expression of BRCA1 [9]. The tet-off inducible expression system is based on the introduction to the cells of a chimeric transactivator comprised of the tet repressor fused to the VP16 transactivation domain. This transactivator, which is inactive in the presence of tetracycline, can bind in the absence of tetracycline to promoters containing the tet operator sequence, such as that used to drive expression of BRCA1 in the above study. The use of this type of expression system therefore facilitates the alteration of a single parameter within the experimental design, namely the induction of BRCA1. The advantage of this approach is that the genetic background of the two populations to be compared by oligonucleotide arrays is essentially identical, the only exception being the induction of BRCA1 in one population (Fig. 2). The question being asked therefore is clearly defined thereby dramatically reducing the expected output in terms of targets from the array experiment. Biotinylated cRNA probes derived from cells in which exogenous BRCA1 was switched off (+ tet) and switched on (- tet) were hybridized to high-density oligonucleotide arrays representing 6,800 known transcripts and expressed sequence tags (ESTs). A number of BRCA1 transcriptional targets were identified in this screen, with the gene exhibiting the greatest degree of differential signal intensity being the stress and DNA damage-inducible gene GADD45 [9] (Fig 3).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 2. Basic design of the experimental set-up used to identify transcriptional targets of the BRCA1 tumor-suppressor gene. Total RNA was extracted from cells in which exogenous BRCA1 was switched off (+ tet) or switched on (– tet). Biotinylated cRNA probes were generated from the two RNA populations and hybridized to individual Affymetrix prototype oligonucleotide arays. The resultant fluorescent signal was scanned and analyzed identifying a number of BRCA1 transcriptional targets [9].
|
|
View larger version (64K):
[in this window]
[in a new window]
|
Figure 3. Fluorescent image generated using the Affymetrix human cancer G110 array containing approximately 1,700 genes that have been implicated in cancer. Each gene on the array is represented by 16 probe pairs, one being wild-type and one containing a mismatch at the central nucleotide. Individual arrays were hybridized with biotinylated cRNA probes generated from cells in which exogenous BRCA1 was induced or repressed. The image shows two of the genes, GADD45 and ATF3 which were identified (and confirmed by Northern blot analysis) as being transcriptional targets of the BRCA1 tumor-suppressor gene. (Printed with permission from Affymetrix.)
|
|
|
FUNCTIONAL RELEVANCE OF IDENTIFIED TARGET GENES
|
|---|
Specific target genes identified by this approach must subsequently be verified by an alternative method such as Northern blot analysis or quantitative reverse transcriptase-PCR to confirm induction in response to the defined stimulus (in this case induction of BRCA1). Using this approach a total of 23 genes and ESTs identified as having increased expression following BRCA1 induction by array hybridization were confirmed as demonstrating at least a twofold induction by Northern blot analysis [9]. Genes confirmed as being transcriptionally activated or repressed in response to the defined stimulus may then be studied further to evaluate their role in signaling pathways regulated by the inducing agent. For the BRCA1 study described above, the induction of GADD45 by BRCA1 was correlated with BRCA1-mediated activation of the c-jun N-terminal kinase/stress-activated protein kinase JNK/SAPK pathway. Furthermore activation of JNK/SAPK was shown to be required for BRCA1-mediated apoptotic cell death in this cell line model system thereby providing a model for BRCA1-mediated apoptosis [9]. This approach has also been clearly illustrated in a strategy similar to that described for BRCA1 in which transcriptional targets of the Wilms' tumor-suppressor gene (WT1) were identified by oligonucleotide array-based expression profiling. The major target gene identified was amphiregulin, a member of the epidermal growth factor family. WT1 was shown to bind directly to the amphiregulin promoter resulting in potent transcriptional activation. Further in vivo studies indicated that the amphiregulin expression pattern mirrored the highly restricted expression pattern of WT1 during fetal kidney development and that recombinant amphiregulin stimulated epithelial branching in organ cultures of embryonic mouse kidneys [31]. Both these studies therefore clearly illustrate the benefits of this experimental approach in terms of its use in defining physiologically relevant target genes. Similar approaches using different technologies further illustrate the need for functional characterization of identified target genes to confirm their physiological relevance in mediating cellular signaling pathways regulated by the genes under investigation. A recent report published in Proceedings of the National Academy of Science from the Vogelstein Laboratory (Baltimore, MD) serves as an example of the need for additional confirmation of identified target genes. In this report Yu et al. used a modified version of the tet-off inducible expression system to define downstream transcriptional targets of the p53 tumor-suppressor gene [8]. They identified a total of 34 genes that exhibited at least a 10-fold upregulation in response to inducible expression of p53. What was surprising about this study was the observed heterogeneity of the response when evaluated in different cell lines derived from the same tissue of origin. For example, of 33 genes studied only nine were shown to be induced in a panel of five unrelated colorectal cell lines, 17 were induced in a subset and eight were not induced in any of the five cell lines examined. This would suggest a high degree of cell type specificity, possibly reflecting the requirement for cell type specific p53 transcriptional coactivators. Further variability was observed when induction of the identified targets were assayed in response to the clinically relevant chemotherapeutic agents adriamycin and 5-fluorouracil (5-FU) previously shown to activate the p53 response. Only six of the genes identified were induced by both agents, suggesting clear target specificity depending on the nature of the inducing signal. Even more surprising, however, was the observation that for the majority of the genes identified, p53 was not absolutely required for induction in response to adriamycin and 5-FU. This suggests that these agents do not act exclusively through p53, emphasizing the redundancy that is inherent in the majority of signaling pathways.
The concern, therefore, in this type of experimental approach is not necessarily the use of overexpression systems, but rather the requirement for a further physiological screen to confirm the relevance of particular target genes identified. Such issues represent a very important concern in the interpretation of studies of this nature and considerable effort must therefore be expended to ensure the physiological relevance of such observations.
|
ACKNOWLEDGMENT
|
|---|
I would like to acknowledge Dr. Daniel Haber in whose lab the BRCA1 work was carried out, Jamie Bean who contributed significantly to this work, and Jonathon Oliner from Affymetrix who carried out the oligonucleotide array experiments.
|
REFERENCES
|
|---|
-
Lockhart DJ, Dong H, Byrne MC et al. Expression monitoring by hybridization to high density oligonucleotide arrays. Nat Biotechnol 1996;14:1675-1680.[CrossRef][Medline]
-
Wodicka L, Dong H, Mittmann M et al. Genome-wide expression monitoring in Saccharomyces cerevisiae. Nat Biotechnol 1997;15:1359-1367.[CrossRef][Medline]
-
McGall G, Labadie J, Brock P et al. Light-directed synthesis of high-density oligonucleotide arrays using semiconductor photoresists. Proc Natl Acad Sci USA 1996;93:13555-13560.[Abstract/Free Full Text]
-
Schena M, Shalon D, Davis RW et al. Quantitative monitoring of gene expression patterns with complimentary DNA microarray. Science 1995;270:467-470.[Abstract/Free Full Text]
-
Alizadeh AA, Eisen MB, Davis RE et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403:503-511.[CrossRef][Medline]
-
Golub TR, Slonim DK, Tamayo P et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531-537.[Abstract/Free Full Text]
-
Tanaka H, Arakawa H, Yamaguchi T et al. A ribonucleotide reductase gene involved in a p53-dependent cell-cycle checkpoint for DNA damage. Nature 2000;404:42-49.[CrossRef][Medline]
-
Yu J, Zhang L, Hwang PM et al. Identification and classification of p53-regulated genes. Proc Natl Acad Sci USA 1999; 96:14517-14522.[Abstract/Free Full Text]
-
Harkin DP, Bean JM, Miklos D et al. Induction of GADD45 and JNK/SAPK-dependent apoptosis following inducible expression of BRCA1. Cell 1999;97:575-586.[CrossRef][Medline]
-
Miki Y, Swensen J, Shattuck-Eidens D et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994;266:66-71.[Abstract/Free Full Text]
-
FitzGerald M, MacDonald D, Krainer M et al. Germ-line BRCA1 mutations in Jewish and non-Jewish women with early-onset breast cancer. N Engl J Med 1996;334:143-149.[Abstract/Free Full Text]
-
Futreal P, Liu Q, Shattuck-Eidens D et al. BRCA1 mutations in primary breast and ovarian carcinomas. Science 1994;266:120-122.[Abstract/Free Full Text]
-
Berchuck A, Heron KA, Carney ME et al. Frequency of germline and somatic BRCA1 mutations in ovarian cancer. Clin Cancer Res 1998;4:2433-2437.[Abstract/Free Full Text]
-
Wilson CA, Ramos L, Villasenor MR et al. Localization of human BRCA1 and its loss in high-grade non-inherited breast carcinoma. Nat Genet 1999;21:236-240.[CrossRef][Medline]
-
Wu LC, Wang ZW, Tsan JT et al. Identification of a RING protein that can interact in vivo with the BRCA1 gene product. Nat Genet 1996;14:430-440.[CrossRef][Medline]
-
Koonin EV, Altschul SF, Bork P. BRCA1 protein products: functional motifs. Nat Genet 1996;13:266-268.[CrossRef][Medline]
-
Cortez D, Wang Y, Qin J et al. Requirement of ATM-dependent phosphorylation of brca1 in the DNA damage response to double strand breaks. Science 1999;286:1162-1166.[Abstract/Free Full Text]
-
Lee JS, Collins KM, Brown AL et al. hCds1-mediated phosphorylation of BRCA1 regulates the DNA damage response. Nature 2000;404:201-204.[CrossRef][Medline]
-
Scully R, Chen J, Plug A et al. Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell 1997;88:265-275.[CrossRef][Medline]
-
Gowen L, Avrutskaya AV, Latour AM et al. BRCA1 required for transcription-coupled repair of oxidative DNA damage. Science 1998;281:1009-1012.[Abstract/Free Full Text]
-
Ruffner H, Verma IM. BRCA1 is a cell cycle-regulated nuclear phosphoprotein. Proc Natl Acad Sci USA 1997;94:7138-7143.[Abstract/Free Full Text]
-
Somasundaram K, Zhang H, Zeng YX et al. Arrest of the cell cycle by the tumour-suppressor BRCA1 requires the CDK-inhibitor p21 WAF1/CiP1. Nature 1997;389:187-190.[CrossRef][Medline]
-
Xu X, Weaver Z, Linke SP et al. Centrosome amplification and a defective G2-M cell cycle checkpoint induce geneticinstability in BRCA1 exon 11 isoform-deficient cells. Mol Cell 1999;3:389-395.[CrossRef][Medline]
-
Monteiro ANA, August A, Hanafusa H. Evidence for a transcriptional activation function of BRCA1 C-terminal region. Proc Natl Acad Sci USA 1996;93:13595-13599.[Abstract/Free Full Text]
-
Anderson S, Schlegel B, Nakajima T et al. BRCA1 protein is linked to the RNA polymerase II holoenzyme complex via RNA helicase A. Nat Genet 1998;19:254-256.[CrossRef][Medline]
-
Ouichi T, Monteiro ANA, August A et al. BRCA1 regulates p53-dependent gene expression. Proc Natl Acad Sci USA 1998;95:2302-2306.[Abstract/Free Full Text]
-
Yu X, Wu LC, Bowcock AM et al. The C-terminal (BRCT) domains of BRCA1 interact in vivo with CtIP, a protein implicated in the CtBP pathway of transcriptional repression. J Biol Chem 1998;273:25388-25392.[Abstract/Free Full Text]
-
Wang Q, Zhang H, Kajino K et al. BRCA1 binds c-Myc and inhibits its transcriptional and transforming activity in cells. Oncogene 1998;17:1939-1948.[CrossRef][Medline]
-
Yarden RI, Brody LC. BRCA1 interacts with components of the histone deacetylase complex. Proc Natl Acad Sci USA 1999;96:4983-4988.[Abstract/Free Full Text]
-
Gossen M, Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 1992;89:5547-5551.[Abstract/Free Full Text]
-
Lee SB, Huang K, Palmer R et al. The Wilms' tumour suppressor WT1 encodes a transcriptional activator of amphiregulin. Cell 1999;98:663-673.[CrossRef][Medline]
Received August 15, 2000;
accepted for publication August 15, 2000.
This article has been cited by other articles:
|
|
|
|
 
M. A. Lanaspa, N. E. Almeida, A. Andres-Hernando, C. J. Rivard, J. M. Capasso, and T. Berl
The tight junction protein, MUPP1, is up-regulated by hypertonicity and is important in the osmotic stress response in kidney cells
PNAS,
August 21, 2007;
104(34):
13672 - 13677.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. P. Harkin
Genomics and the Impact of New Technologies on the Management of Colorectal Cancer
Oncologist,
October 1, 2006;
11(9):
988 - 991.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. T. Lynch, C. L. Snyder, J. F. Lynch, B. D. Riley, and W. S. Rubinstein
Hereditary Breast-Ovarian Cancer at the Bedside: Role of the Medical Oncologist
J. Clin. Oncol.,
February 15, 2003;
21(4):
740 - 753.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Romero, H. Kuivaniemi, and G. Tromp
Functional Genomics and Proteomics in Term and Preterm Parturition
J. Clin. Endocrinol. Metab.,
June 1, 2002;
87(6):
2431 - 2434.
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y. Tagaito, V. Y. Polotsky, M. J. Campen, J. A. Wilson, A. Balbir, P. L. Smith, A. R. Schwartz, and C. P. O'Donnell
A model of sleep-disordered breathing in the C57BL/6J mouse
J Appl Physiol,
December 1, 2001;
91(6):
2758 - 2766.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
