|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology
Gene Discovery Using the Serial Analysis of Gene Expression Technique: Implications for Cancer ResearchFrom the Department of Adult Oncology, Dana-Farber Cancer Institute and Harvard Medical School, Boston, MA; and Department of Pathology, Duke University Medical Center, Durham, NC. Address reprint requests to Kornelia Polyak, MD, PhD, Department of Adult Oncology, Dana-Farber Cancer Institute, 44 Binney St, D740C, Boston, MA 02115; email: kornelia_polyak{at}dfci.harvard.edu ABSTRACT ABSTRACT: Cancer is a genetic disease. As such, our understanding of the pathobiology of tumors derives from analyses of the genes whose mutations are responsible for those tumors. The cancer phenotype, however, likely reflects the changes in the expression patterns of hundreds or even thousands of genes that occur as a consequence of the primary mutation of an oncogene or a tumor suppressor gene. Recently developed functional genomic approaches, such as DNA microarrays and serial analysis of gene expression (SAGE), have enabled researchers to determine the expression level of every gene in a given cell population, which represents that cell population’s entire transcriptome. The most attractive feature of SAGE is its ability to evaluate the expression pattern of thousands of genes in a quantitative manner without prior sequence information. This feature has been exploited in three extremely powerful applications of the technology: the definition of transcriptomes, the analysis of differences between the gene expression patterns of cancer cells and their normal counterparts, and the identification of downstream targets of oncogenes and tumor suppressor genes. Comprehensive analyses of gene expression not only will further understanding of growth regulatory pathways and the processes of tumorigenesis but also may identify new diagnostic and prognostic markers as well as potential targets for therapeutic intervention. UNTIL RECENTLY, the approach to understanding the molecular basis of complex biologic processes such as human development and cancer was to study the behavior of genes one at a time. Two recently developed technologies, oligonucleotide or cDNA microarrays and serial analysis of gene expression (SAGE), allow researchers to determine the expression pattern of thousands of genes simultaneously.1,2 One of the main differences between the two technologies is that microarray preparation requires prior knowledge of the sequence of the gene transcripts to be analyzed. This is a serious limitation, even for organisms with completely sequenced genomes such as humans, because genome annotation and gene prediction remain technical challenges.3 SAGE, however, can be used to analyze gene expression in organisms whose genomes are largely uncharacterized. Another advantage of SAGE over microarrays is the portability of the data generated. Although SAGE data are directly comparable, the differences in microarray formats and normalization methodologies make direct comparison of data sets between microarray platforms very difficult. SAGE not only can accurately determine the absolute abundance of mRNAs but also can detect even slight differences in expression levels between samples. Indeed, a recent comparison of the gene expression profiles obtained using oligonucleotide microarrays (6,000 genes; Affymetrix GeneChip, Santa Clara, CA) and SAGE from RNA isolated from monocytes and granulocyte-macrophage colony-stimulating factor–induced macrophages highlighted this strength of the SAGE technique. Although the expression level differences detected by the two techniques were in fairly good agreement for abundant and highly differentially expressed (> 10-fold difference) genes,4 many of the differentially expressed genes identified by SAGE were either not present on the GeneChip or the fold difference detected was not significant. Clearly, the first part of this problem will be solved once microarrays of oligonucleotides or cDNAs representing all expressed genes become available. Unfortunately, Ishii et al did not analyze RNA by a third quantitative technique, such as real-time or digital polymerase chain reaction (PCR) or cDNA library screening. Consequently, the origin of the disparity between SAGE and GeneChip that they reported remains to be resolved. However, mRNA quantitation seems more accurate with SAGE than with microarrays. Microarrays have advantages, too, of course, because they are relatively easy to use and more suitable for high-throughput applications. As this brief comparison indicates, the choice of technique is determined by the specific experiment to be performed. Expression profiling of hundreds of tumor samples is certainly more efficient with the use of microarrays. SAGE, however, seems to be a better choice for the identification of genes and alternatively processed transcripts that are unique to a specific cell type and for the analysis of previously uncharacterized organisms. SAGE is based on the following two principles:
The generation of a SAGE library involves sequential enzymatic steps. The process is depicted in Fig 1 and described briefly below. Double-stranded cDNA is generated from mRNA isolated from the cells or tissues of interest and immobilized on streptavidin-coated magnetic beads by virtue of biotin residues incorporated through the use of biotinylated oligonucleotide deoxythymidine primer during reverse transcription. The cDNA is then cleaved with a frequent cutting restriction enzyme to ensure that every cDNA is cleaved at least once. NlaIII is the most frequently used enzyme and cuts DNA at an average of every 256 bp. The most 3'-end of the cDNA (up to the most 3' NlaIII site) is then collected on the beads and ligated to a linker. This linker has a recognition site for a type IIS restriction enzyme (eg, BsmfI) and a PCR primer site. Type IIS restriction enzymes cut DNA a certain number of bases away from the recognition sequence; therefore, a short fragment of the cDNA (SAGE tag) remains attached to the linker but is cleaved from the beads. These tags are blunt-ended, ligated to each other to form ditags, and used as templates for PCR amplification. Cleavage of this PCR product with NlaIII releases the ditags, which are isolated, concatenated, subcloned into an appropriate vector, and sequenced. Theoretically, inefficient enzymatic reactions that occur during the generation of a SAGE library can lead to inaccurate data, but this problem can be avoided by rigorous testing of the reagents and the use of appropriate controls. The analysis of the SAGE data is performed using the SAGE software (Johns Hopkins University, Baltimore, MD), which extracts the tags from the sequence and determines their abundance and identity. Moreover, because a typical SAGE experiment involves the comparison of at least two different libraries, the statistical significance of gene expression differences between libraries is calculated for each transcript detected.
Several technical modifications have been made to the original SAGE protocol to improve its efficiency.6-8 The most significant advance was the development of so-called micro-SAGE or SAGE-lite, which has enabled comprehensive gene expression profiles to be established from as few as 5,000 cells, thus facilitating analysis of samples obtained by microscopic dissection of frozen tissue.9-11 The introduction of an additional PCR amplification step after the cDNA synthesis reaction has allowed a further decrease in the amount of material required for successful use of SAGE, and comprehensive gene expression profiles have been derived from as few as eight oocytes.12 Although including PCR amplification steps may alter the representation of the mRNAs, this approach can be useful for the identification of cell type-specific transcripts. SAGE has been used to analyze various normal and diseased cell types in the belief that deciphering the molecular differences between them will further understanding of their principal biologic processes. Studies have been performed with the aim of extending understanding of atherosclerosis, human immunodeficiency virus infection, and the function of normal liver, muscle, thyroid, and various hematopoietic cells.13-20 Particular emphasis has been placed on the analysis of various cancer types. Some of the advances made in this area of research are reviewed next.
SAGE: A LOOKING GLASS FOR CANCER
In Search of Tumor Markers In the past few years, SAGE studies have been performed in patients with colon, pancreatic, lung, bladder, ovarian, and breast cancers; glioblastomas; and medulloblastomas.23-31 Several known genes and a number of uncharacterized genes have been found to be up- or downregulated in these various cancers. Examples of these differentially expressed genes are listed in Table 2. In most instances, these SAGE analyses were performed with a limited number of tissue samples; usually one or two pairs of primary tumor and corresponding normal tissue. For this reason, many of the genes highlighted in the SAGE experiments have been examined and subsequently have been validated in multiple tumor and normal tissue pairs using a variety of approaches, including Northern blot analysis, real-time PCR, mRNA in situ hybridization, and immunohistochemistry. Because one of the goals of these studies was to identify global tumor markers, the expression of these genes was analyzed in multiple normal and cancerous tissue types. An ideal tumor marker has high specificity and high sensitivity and should be easy to assay in clinical practice. One of the genes found to be highly expressed in pancreatic cancers—tissue inhibitor of metalloproteinase type I (TIMP-1)—seemed to have all of these characteristics.23,32 Using a simple enzyme-linked immunosorbent assay, the serum level of TIMP-1 was demonstrated to be elevated significantly in patients with pancreatic cancer, but TIMP-1 alone was inadequate as a tumor marker because it gave positive results in only approximately 30% of patients with pancreatic cancer. However, in combination with two other markers—CA19-9 and carcinoembryonic antigen—60% of the 85 patients with pancreatic cancer in the study were diagnosed successfully with high specificity and without numerous false positives. Similarly, protein gene product 9.5 and several cell surface molecules seem to be promising candidates as tumor markers for lung cancer and glioblastoma multiforme, respectively.27,33 Thus comprehensive analysis of gene expression has the potential to identify candidates that—either alone or in combination—can improve the accuracy of cancer diagnosis.
Identifying Pathways in Pediatric Tumors The gene expression profiles established from the pediatric tumor, medulloblastoma, turned out to be particularly informative.28 In addition to clarifying the cellular origin of these tumors, analysis of the profiles led to the identification of a key pathway for the regulation of differentiation by virtue of its dysfunction in these cells. Two of the most highly expressed genes in medulloblastoma are zinc finger protein 1 (ZIC1) and orthodentical homolog 2 (OTX-2). Both encode proteins with known developmental functions. ZIC1 is the putative human homolog of the Drosophila Opa pair-rule gene that is required for the expression of wingless and engrailed. Interestingly, mutations in other components of the wnt (ie, the mammalian homolog of wingless) pathway, such as beta-catenin (ß-catenin), have also been implicated in human medulloblastomas. OTX2 is the human homolog of the Drosophila homeobox gene orthodenticle. This fly gene is known to be involved in the development of rostral brain regions, and, similarly, elimination of OTX2 function in the mouse causes defects in rostral neuroectoderm development leading to a headless phenotype. Interestingly, in normal cerebellum, both ZIC1 and OTX-2 are very selectively expressed in the cells of the external and internal granular layers and in those cells migrating between them. The high expression of these genes in medulloblastomas supports the hypothesis that these tumors originate from precursor cells located in these layers. In addition, the restricted expression pattern of OTX2 makes this gene a particularly promising candidate as a diagnostic marker for medulloblastoma.
From Phenotype to Genotype
A Portrait of Neovasculature
GROWTH REGULATORY PATHWAYS The estrogen-signaling pathway in breast and gynecologic cancer and the androgen-signaling pathway in prostate cancer are among the most important to our understanding of these malignancies.36,37 Despite the well-established roles that these hormones play in cancer initiation and progression, little is known about the mechanisms that account for their tumorigenic effects. The actions of estrogens and androgens are mediated by their receptors, the ER, and the androgen receptor. These proteins are members of the nuclear hormone receptor family of ligand-dependent transcription factors.38-40 With this fact in mind, it seems reasonable to suppose that comprehensive analysis of gene expression profiles after estrogen or androgen treatment might identify critical downstream targets of these pathways. A SAGE analysis of LNCaP prostate cancer cells treated with a synthetic androgen (R1881) led to the identification of a novel androgen-responsive, prostate-specific gene, PMEPA1.41 In situ hybridization and Northern blot analysis of multiple tissue types revealed that PMEPA1 is highly expressed only in the prostate gland and is virtually absent from all other cell types. Moreover, analysis of androgen-independent prostate cancers demonstrated elevated levels of PMEPA1 mRNA. This finding indicates that PMEPA1 may play an important role in androgen-mediated tumorigenesis. Interestingly, the PMEPA1 gene is localized to chromosome 20q13.2–20q13.33, a region frequently amplified in prostate and other cancer types. However, analysis of PMEPA1 mRNA levels in androgen-responsive primary prostate tumors did not reveal increased expression levels, suggesting that overexpression of PMEPA1 may be important in androgen-independent but not androgen-dependent tumors. Two similar studies of the effects of estrogen treatment also successfully identified hormone-responsive genes. Comprehensive gene expression profiles were established from the MCF-7 human ER+ breast cancer cell line both before and after estrogen treatment.42 In one study, SAGE libraries were prepared 24 hours after treatment, a relatively late time point.43 In the other, the analyses were made much earlier, at 3 and 10 hours after the addition of the hormone.43 Comparison of the gene expression profiles not surprisingly revealed them to be rather dissimilar. However, the expression of several known estrogen targets—including cathepsin D and pS2—were found to be altered in both cases. The analysis of estrogen-mediated gene expression changes shortly after treatment turned out to be the more informative, emphasizing the importance of timing of sample collection. One interesting novel estrogen target gene identified was Wnt-1 inducible signaling pathway protein 2 (WISP-2). WISP-2 previously had been identified as a gene upregulated in a Wnt-1-transformed mouse mammary epithelial cell line. Several members of the chaperone family were also induced by estrogen. Included in this group were HSP90,HSP60,HSC71,FKBP4, and E2-induced gene 1 (E2IG1), a novel member of the HSP20 family. In addition, many proliferation-related genes were found to be upregulated by estrogen as well. These were genes such as cyclin D1;CCT2 (a regulator of cyclin E folding); Ran/TC4 (a component of the nuclear pore complex); calmodulin subunits; and several paracrine-autocrine factors, including stanniocalcin 2 (involved in Ca2+ metabolism) and Inhibin Bß. The tumor-related proteins caveolin 1, nm23H1, and D52L1 also were identified as estrogen targets. A number of novel genes were highlighted in the analysis and were designated E2IG2 to E2IG5. The function of these genes is currently unknown, although their sequence characteristics and homologies to other proteins suggest that they may participate in estrogen-mediated tumorigenesis. The identification of novel estrogen-responsive genes not only will further our understanding of the cancer promoting effects of estrogen but also may provide candidate diagnostic markers and potential therapeutic targets. For example, E2IG1 and stanniocalcin 2 were found to be expressed exclusively in ER+ breast cancers. This feature highlights these genes as potentially powerful breast cancer biomarkers.
TARGETS OF TUMOR SUPPRESSOR GENES AND ONCOGENES Another study aimed at identifying p53 transcriptional targets was performed in the DLD-1 human colon cancer cell line. This line undergoes apoptosis after infection with a replication-defective adenovirus-encoding p53 (Ad-p53).50 Cells infected with a lacZ-expressing adenovirus (Ad-lacZ) served as controls. Comparison of the two SAGE libraries generated led to the identification of 14 transcripts that were specifically induced at least 10-fold by p53. These were named PIGs, for p53-induced genes. Only two of these genes—p21 and PIG8/Ei24—were known p53 targets, with the majority representing novel genes. Interestingly, several of the PIGs seemed to encode proteins known or predicted to participate in the regulation of cellular redox status. The well-documented role of reactive oxygen species (ROS) in apoptosis, coupled with the identification of a number of PIGs with redox function, led to the hypothesis that p53 induces apoptosis by increasing the generation of ROS, resulting in oxidative stress, mitochondrial damage, and subsequent cell death. This hypothesis was later confirmed in biochemical experiments. Although ectopic expression of any single PIG was insufficient to induce apoptosis, blocking induction of PIGs did markedly inhibit cell death. This study not only identified several novel direct targets of p53 but also was one of the first to demonstrate that analysis of global gene expression profiles can identify a group of genes that function in a common pathway to illuminate the underlying mechanism of a complex biologic process.
Analysis of gene expression profiles established from the HCT116 human colorectal cancer cell line before and after gamma-irradiation ( Finally, an experiment performed with a derivative of the DLD-1 cell line expressing a tetracycline-inducible p53 (tet-p53) identified yet another set of p53-induced genes. These genes—named PETs, for p53 early transcripts—were found by analyzing gene expression profiles at time points soon after p53 induction.53 The almost complete lack of overlap between the set of genes affected by Ad-p53 and tet-p53 is intriguing. The expression profiles were examined at different times after p53 transduction (16 hours for Ad-p53 and 9 hours for tet-p53); thus this comparison demonstrates that the genes induced by p53 can vary dramatically, even in the same cell type. In summary, the analysis of transcriptomes after p53 expression has determined that p53 exerts its diverse cellular functions by influencing the expression of a large group of genes (Fig 2). Moreover, profound variability seems to exist with regard to the extent, timing, and p53 dependence of the expression of these genes. This variability is most likely due to differences in the cellular context (eg, cell type, extra- and intracellular signals) in which p53 exerts its tumor-suppressive function.
Adenomatosis Polyposis Coli/ß-Catenin/T-Cell Factor Pathway The high frequency of mutation in the earliest forms of colon cancer demonstrates that the elimination of wild-type adenomatosis polyposis coli (APC) gene function is required for the initiation of colon carcinogenesis.54,55 One of the tumor-suppressive functions of APC is the recruitment of ß-catenin into multimolecular protein complexes that include GSK3ß kinase and a scaffold protein called axin.56,57 Once recruited to this complex, ß-catenin is phosphorylated by GSK3ß and subsequently degraded by the ubiquitin proteasome pathway. ß-catenin, in addition to other proteins such as the cadherins and desmoplakin, binds to the T-cell factor (TCF) family of transcription factors and activates gene transcription.58,59 Mutation of ß-catenin or inactivation of APC renders ß-catenin resistant to degradation and leads to an increase in ß-catenin/TCF-mediated transcription. However, the critical targets of this pathway are unknown. In an attempt to determine the effect of wild-type APC expression on the transcription profiles of human colon cancer cells, SAGE libraries were generated from a derivative of the HT-29 cell line (whose endogenous APC is mutated) expressing exogenous wild-type APC under the control of a metallothionein promoter.60 Induction of APC expression by the addition of zinc to the culture medium results in growth inhibition and apoptosis. These cellular effects were accompanied by the rapid downregulation of 16 transcripts. The most prominent of these was the c-myc oncogene, which was repressed within 6 hours of APC induction. This finding suggested that c-myc might be a direct target of the ß-catenin/TCF-4 pathway, a hypothesis that subsequently was demonstrated to be correct when two functional TCF-4 binding sites were identified in the c-myc promoter region. Moreover, the ability of a dominant-negative TCF-4 to block c-myc expression in cells containing a mutant APC or ß-catenin protein demonstrated that ß-catenin/TCF complexes are required for c-myc expression in human colon cancer cells. The identification of c-myc as a downstream target of the APC/ß-catenin/TCF pathway explains two intriguing characteristics of human colorectal tumors. First, despite the well-documented overexpression of c-myc mRNA and protein in most colon cancers, regardless of cancer stage, genetic rearrangements (eg, amplification, translocations) involving the c-myc genomic region are rare.61,62 Second, and in contrast to many other cancer types, mutations in the p16/p15,cyclin D1, cyclin-dependent kinase-4 (CDK4), Rb pathway are very rare in human colorectal tumors. Because c-myc can bypass p16- and pRb-mediated growth inhibition, the activation of c-myc through the ß-catenin/TCF pathway may alleviate the need for elimination of these genes.
Another intriguing target of the APC/ß-catenin/TCF pathway identified was the peroxisome proliferator-activated receptor
In summary, SAGE analysis of human colon cancer cells after ectopic expression of wild-type APC has identified two important downstream targets of the APC/ß-catenin/TCF pathway (Fig 3). One of these targets (c-myc) provides a link to enhanced cellular proliferation, and the other (PPAR
Thec-myc Oncogene The c-myc oncogene encodes a transcription factor of the helix-loop-helix leucine zipper family and has been implicated in a variety of cancer types.64 It is well established that c-myc induces proliferation by promoting cell cycle entry, but the downstream target genes responsible for this function are still largely unknown. In an attempt to elucidate the mechanism by which c-myc enhances proliferation, SAGE libraries were generated from serum-starved human umbilical vein endothelial cells (HUVEC) infected with a replication-defective adenovirus-expressing c-myc or a control adenovirus-expressing green fluorescence protein.65 The expression of c-myc is necessary but not sufficient to induce cell cycle entry in HUVEC in the absence of serum. Analysis of gene expression profiles after ectopic expression of c-myc in serum-starved cells therefore allowed the identification of direct c-myc target genes and not merely genes affected as a secondary consequence of c-myc-induced proliferation. Because of its role in cell cycle regulation, CDK4 was one of the most intriguing c-myc target genes identified. CDK4 associates with D-type cyclins to regulate G1 phase progression and is demonstrably essential for cell proliferation.66 Previous studies had demonstrated that many of the effects of c-myc overexpression, including inhibition of cell cycle arrest induced by serum starvation, p53 activation, and transforming growth factor-beta treatment, can be mimicked by CDK4 overexpression.67-69 In addition, both c-myc and CDK4 can immortalize primary cells, and both are frequently overexpressed in human tumors.70-73 The direct activation of the CDK4 gene by c-myc was verified by the identification of four consensus c-myc binding sites in the promoter regions of both the human and mouse CDK4 genes. Further analysis of these putative c-myc binding sites demonstrated that they are both necessary and sufficient for the induction of the CDK4 mRNA by c-myc. Given that the c-myc gene was recently identified as a downstream target of the APC/ß-catenin/TCF pathway (Fig 3) and that this pathway is frequently disrupted in colorectal cancers, the concomitant overexpression of CDK4 in these tumors indicates that CDK4 is likely to be a relevant target of the c-myc oncogenic pathway.74 In the past few years, the use of SAGE as a means of analyzing global gene expression profiles has increased. These analyses have provided a wealth of new information regarding the differences between normal and cancerous tissue at the molecular level. In addition, several potential diagnostic and prognostic markers as well as a number of candidate therapeutic targets have been identified. Cancer researchers now can see into the function of cells at a far higher resolution than was considered possible even a few years ago. In much the same way as microscopes and telescopes revolutionized cell biology and astronomy, comprehensive gene expression analysis is likely to accelerate progress in molecular biology dramatically. This technological advance promises a revolution in the way that cancer patients are diagnosed and treated. ACKNOWLEDGMENTS Supported in part by National Cancer Institute Cancer Genome Anatomy Project contract no. S98-146A. REFERENCES
1.
Velculescu VE, Zhang L, Vogelstein B, et al: Serial analysis of gene expression. Science 270: 484-487, 1995
2.
Chee M, Yang R, Hubbell E, et al: Accessing genetic information with high-density DNA arrays. Science 274: 610-614, 1996
3.
Venter JC, Myers EW, Li PW, et al: The sequence of the human genome. Science 291: 1304-1351, 2001 4. Ishii M, Hashimoto S, Tsutsumi S, et al: Direct comparison of GeneChip and SAGE on the quantitative accuracy in transcript profiling analysis. Genomics 68: 136-143, 2000[Medline] 5. Velculescu VE, Madden SL, Zhang L, et al: Analysis of human transcriptomes. Nat Genet 23: 387-388, 1999[Medline]
6.
Powell J: Enhanced concatemer cloning: A modification to the SAGE (Serial Analysis of Gene Expression) technique. Nucleic Acids Res 26: 3445-3446, 1998
7.
Kenzelmann M, Muhlemann K: Substantially enhanced cloning efficiency of SAGE (Serial Analysis of Gene Expression) by adding a heating step to the original protocol. Nucleic Acids Res 27: 917-918, 1999 8. Angelastro JM, Klimaschewski LP, Vitolo OV: Improved NlaIII digestion of PAGE-purified 102 bp ditags by addition of a single purification step in both the SAGE and microSAGE protocols. Nucleic Acids Res 28: E62, 2000
9.
Datson NA, van der Perk-de Jong J, van den Berg MP, et al: MicroSAGE: A modified procedure for serial analysis of gene expression in limited amounts of tissue. Nucleic Acids Res 27: 1300-1307, 1999
10.
Virlon B, Cheval L, Buhler JM, et al: Serial microanalysis of renal transcriptomes. Proc Natl Acad Sci U S A 96: 15286-15291, 1999
11.
Peters DG, Kassam AB, Yonas H, et al: Comprehensive transcript analysis in small quantities of mRNA by SAGE-lite. Nucleic Acids Res 27: e39, 1999 12. Neilson L, Andalibi A, Kang D, et al: Molecular phenotype of the human oocyte by PCR-SAGE. Genomics 63: 13-24, 2000[Medline] 13. de Waard V, van den Berg BM, Veken J, et al: Serial analysis of gene expression to assess the endothelial cell response to an atherogenic stimulus. Gene 226: 1-8, 1999[Medline]
14.
Chen H, Centola M, Altschul SF, et al: Characterization of gene expression in resting and activated mast cells. J Exp Med 188: 1657-1668, 1998
15.
Hashimoto S, Suzuki T, Dong HY, et al: Serial analysis of gene expression in human monocyte-derived dendritic cells. Blood 94: 845-852, 1999
16.
Hashimoto S, Suzuki T, Dong HY, et al: Serial analysis of gene expression in human monocytes and macrophages. Blood 94: 837-844, 1999
17.
Hashimoto SI, Suzuki T, Nagai S, et al: Identification of genes specifically expressed in human activated and mature dendritic cells through serial analysis of gene expression. Blood 96: 2206-2214, 2000
18.
Suzuki T, Hashimoto S, Toyoda N, et al: Comprehensive gene expression profile of LPS-stimulated human monocytes by SAGE. Blood 96: 2584-2591, 2000
19.
Angelastro JM, Klimaschewski L, Tang S, et al: Identification of diverse nerve growth factor-regulated genes by serial analysis of gene expression (SAGE) profiling. Proc Natl Acad Sci U S A 97: 10424-10429, 2000 20. Yamashita T, Hashimoto S, Kaneko S, et al: Comprehensive gene expression profile of a normal human liver. Biochem Biophys Res Commun 269: 110-116, 2000[Medline]
21.
Lal A, Lash AE, Altschul SF, et al: A public database for gene expression in human cancers. Cancer Res 59: 5403-5407, 1999
22.
Lash AE, Tolstoshev CM, Wagner L, et al: SAGEmap: A public gene expression resource. Genome Res 10: 1051-1060, 2000
23.
Zhang L, Zhou W, Velculescu VE, et al: Gene expression profiles in normal and cancer cells. Science 276: 1268-1272, 1997
24.
Hibi K, Liu Q, Beaudry GA, et al: Serial analysis of gene expression in non-small cell lung cancer. Cancer Res 58: 5690-5694, 1998
25.
Nacht M, Ferguson AT, Zhang W, et al: Combining serial analysis of gene expression and array technologies to identify genes differentially expressed in breast cancer. Cancer Res 59: 5464-5470, 1999 26. Bahrenberg G, Brauers A, Joost HG, et al: Reduced expression of PSCA, a member of the LY-6 family of cell surface antigens, in bladder, esophagus, and stomach tumors. Biochem Biophys Res Commun 275: 783-788, 2000[Medline]
27.
Loging WT, Lal A, Siu IM, et al: Identifying potential tumor markers and antigens by database mining and rapid expression screening. Genome Res 10: 1393-1402, 2000
28.
Michiels EM, Oussoren E, Van Groenigen M, et al: Genes differentially expressed in medulloblastoma and fetal brain. Physiol Genomics 1: 83-91, 1999
29.
Hough CD, Sherman-Baust CA, Pizer ES, et al: Large-scale serial analysis of gene expression reveals genes differentially expressed in ovarian cancer. Cancer Res 60: 6281-6287, 2000 30. Parle-McDermott A, McWilliam P, Tighe O, et al: Serial analysis of gene expression identifies putative metastasis-associated transcripts in colon tumour cell lines. Br J Cancer 83: 725-728, 2000[Medline] 31. Gunnersen JM, Spirkoska V, Smith PE, et al: Growth and migration markers of rat C6 glioma cells identified by serial analysis of gene expression. Glia 32: 146-154, 2000[Medline] 32. Zhou W, Sokoll LJ, Bruzek DJ, et al: Identifying markers for pancreatic cancer by gene expression analysis. Cancer Epidemiol Biomarkers Prev 7: 109-112, 1998[Abstract]
33.
Hibi K, Westra WH, Borges M, et al: PGP9.5 as a candidate tumor marker for non-small-cell lung cancer. Am J Pathol 155: 711-715, 1999
34.
Caron H, van Schaik B, van der Mee M, et al: The human transcriptome map: Clustering of highly expressed genes in chromosomal domains. Science 291: 1289-1292, 2001
35.
St Croix B, Rago C, Velculescu V, et al: Genes expressed in human tumor endothelium. Science 289: 1197-1202, 2000
36.
Nandi S, Guzman RC, Yang J: Hormones and mammary carcinogenesis in mice, rats, and humans: A unifying hypothesis. Proc Natl Acad Sci U S A 92: 3650-3657, 1995 37. Jenster G: The role of the androgen receptor in the development and progression of prostate cancer. Semin Oncol 26: 407-421, 1999[Medline] 38. Freedman LP: Multimeric coactivator complexes for steroid/nuclear receptors. Trends Endocrinol Metab 10: 403-407, 1999[Medline] 39. Freedman LP: Strategies for transcriptional activation by steroid/nuclear receptors. J Cell Biochem Suppl 32-33: 103-109, 1999 40. Freedman LP: Increasing the complexity of coactivation in nuclear receptor signaling. Cell 97: 5-8, 1999[Medline] 41. Xu LL, Shanmugam N, Segawa T, et al: A novel androgen-regulated gene, PMEPA1, located on chromosome 20q13 exhibits high level expression in prostate. Genomics 66: 257-263, 2000[Medline] 42. Inadera H, Hashimoto S, Dong HY, et al: WISP-2 as a novel estrogen-responsive gene in human breast cancer cells. Biochem Biophys Res Commun 275: 108-114, 2000[Medline]
43.
Charpentier AH, Bednarek AK, Daniel RL, et al: Effects of estrogen on global gene expression: Identification of novel targets of estrogen action. Cancer Res 60: 5977-5983, 2000
44.
Hollstein M, Sidransky D, Vogelstein B, et al: p53 mutations in human cancers. Science 253: 49-53, 1991 45. Prives C, Hall PA: The p53 pathway. J Pathol 187: 112-126, 1999[Medline]
46.
Ko LJ, Prives C: p53: Puzzle and paradigm. Genes Dev 10: 1054-1072, 1996 47. Burns TF, El-Deiry WS: The p53 pathway and apoptosis. J Cell Physiol 181: 231-239, 1999[Medline] 48. el-Deiry WS: Regulation of p53 downstream genes. Semin Cancer Biol 8: 345-357, 1998[Medline] 49. Madden SL, Galella EA, Zhu J, et al: SAGE transcript profiles for p53-dependent growth regulation. Oncogene 15: 1079-1085, 1997[Medline] 50. Polyak K, Xia Y, Zweier JL, et al: A model for p53-induced apoptosis. Nature 389: 300-305, 1997[Medline] 51. Hermeking H, Lengauer C, Polyak K, et al: 14-3-3 sigma is a p53-regulated inhibitor of G2/M progression. Mol Cell 1: 3-11, 1997[Medline] 52. Chan TA, Hermeking H, Lengauer C, et al: 14-3-3 sigma is required to prevent mitotic catastrophe after DNA damage. Nature 401: 616-620, 1999[Medline]
53.
Yu J, Zhang L, Hwang PM, et al: Identification and classification of p53-regulated genes. Proc Natl Acad Sci U S A 96: 14517-14522, 1999 54. Kinzler KW, Vogelstein B: Lessons from hereditary colorectal cancer. Cell 87: 159-170, 1996[Medline] 55. Polakis P: The adenomatous polyposis coli (APC) tumor suppressor. Biochim Biophys Acta 1332 (3): F127-F147, 1997[Medline] 56. Gumbiner BM: Carcinogenesis: A balance between beta-catenin and APC. Curr Biol 7: R443-R446, 1997[Medline] 57. Bienz M: APC: The plot thickens. Curr Opin Genet Dev 9: 595-603, 1999[Medline]
58.
Morin PJ, Sparks AB, Korinek V, et al: Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science 275: 1787-1790, 1997
59.
Korinek V, Barker N, Morin PJ, et al: Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma. Science 275: 1784-1787, 1997
60.
He TC, Sparks AB, Rago C, et al: Identification of c-MYC as a target of the APC pathway. Science 281: 1509-1512, 1998
61.
Calabretta B, Kaczmarek L, Ming PM, et al: Expression of c-myc and other cell cycle-dependent genes in human colon neoplasia. Cancer Res 45 (12 Pt 1): 6000-6004, 1985 62. Finley GG, Schulz NT, Hill SA, et al: Expression of the myc gene family in different stages of human colorectal cancer. Oncogene 4: 963-971, 1989[Medline] 63. He TC, Chan TA, Vogelstein B, et al: PPARdelta is an APC-regulated target of nonsteroidal anti-inflammatory drugs. Cell 99: 335-345, 1999[Medline] 64. Grandori C, Cowley SM, James LP, et al: The Myc/Max/Mad Network and the transcriptional control of cell behavior. Annu Rev Cell Dev Biol 16: 653-699, 2000[Medline]
65.
Hermeking H, Rago C, Schuhmacher M, et al: Identification of CDK4 as a target of c-MYC. Proc Natl Acad Sci U S A 97: 2229-2234, 2000
66.
Sherr CJ: The Pezcoller lecture: Cancer cell cycles revisited. Cancer Res 60: 3689-3695, 2000 67. Ewen ME: p53-dependent repression of cdk4 synthesis in transforming growth factor-beta-induced G1 cell cycle arrest. J Lab Clin Med 128: 355-360, 1996[Medline] 68. Latham KM, Eastman SW, Wong A, et al: Inhibition of p53-mediated growth arrest by overexpression of cyclin-dependent kinases. Mol Cell Biol 16: 4445-4455, 1996[Abstract] 69. Reed MF, Liu VF, Ladha MH, et al: Enforced CDK4 expression in a hematopoietic cell line confers resistance to the G1 arrest induced by ionizing radiation. Oncogene 17: 2961-2971, 1998[Medline]
70.
Khatib ZA, Matsushime H, Valentine M, et al: Coamplification of the CDK4 gene with MDM2 and GLI in human sarcomas. Cancer Res 53: 5535-5541, 1993 71. Masciullo V, Scambia G, Marone M, et al: Altered expression of cyclin D1 and CDK4 genes in ovarian carcinomas. Int J Cancer 74: 390-395, 1997[Medline] 72. Kim H, Ham EK, Kim YI, et al: Overexpression of cyclin D1 and cdk4 in tumorigenesis of sporadic hepatoblastomas. Cancer Lett 131: 177-183, 1998[Medline] 73. Brison O: Gene amplification and tumor progression. Biochim Biophys Acta 1155: 25-41, 1993[Medline] 74. Ikeda K, Monden T, Tsujie M, et al: Cyclin D, CDK4 and p16 expression in colorectal cancer [in Japanese]. Nippon Rinsho 54: 1054-1059, 1996[Medline] Submitted November 30, 2000; accepted March 6, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
-irradiation) identified yet another important p53 target gene.
gene. Homologs of the 14-3-3
(PPAR
DNA binding complexes. 
