New opportunities for uncovering the molecular basis of cancer

Identification of candidate cancer genes involved in human retinoblastoma by data mining

OBJECTIVE

The objective of this study was to discover potential cancer-related genes involved in retinoblastoma (RB) tumorigenesis.

MATERIALS AND METHODS

Using a data-mining tool called cDNA Digital Gene Expression Displayer (DGED) and serial analysis of gene expression DGED from the Cancer Genome Anatomy Project (CGAP) database, eight cDNA libraries and five serial analysis of gene expression libraries from retinoblastoma (RB) solid tumors and normal retina tissues were analyzed. The deregulated genes were classified into major families using information from Gene Ontology. Several candidate cancer-related genes were analyzed by real-time reverse transcription polymerase chain reaction (RT-PCR) and immunohistochemistry (IHC) on tissue microarrays (TMA) of RB and human normal retina samples.

RESULTS

A total of 260 genes with deregulated expression emerged when examined by DGED from the CGAP database. Functional classification of these genes not only provided an interesting insight into RB tumorigenesis but also facilitated target identification for RB therapeutics. Several candidate genes were confirmed by real-time RT-PCR and IHC analysis on TMA and were found to be associated with RB genesis through text-mining in Information Hyperlinked over Proteins. The results also implicated MCM7 and WIF1 as promising therapeutic targets for RB, but further validation is needed.

Overview of metastasis assays

During tumor progression, cells acquire genetic and proteomic changes as they transform from normal to hyperplastic, through dysplasia, to carcinoma in situ, and finally to invasive and metastatic. The time course of progression may extend as far back as 10 years prior to diagnosis. Discerning the mechanism whereby tumor cells execute metastatic dissemination may provide the foundation necessary for successful treatment of the disease. For example, direct genetic evidence has linked in situ breast cancer to invasive carcinoma of the breast supporting the generally accepted assumption that carcinoma in situ of the breast is a clonal expansion of hyperproliferating cells. This in turn may provide a more comprehensive and/or functionally directed target strategy for intervention and prevention of breast cancer. This overview provides a picture of the processes related to metastasis and the experimental approaches used to study these processes.

Pan-genome isolation of low abundance transcripts using SAGE tag

The SAGE (serial analysis of gene expression) method is sensitive at detecting the lower abundance transcripts. More than a third of human SAGE tags identified are novel representing the low abundance unknown transcripts. Using the GLGI method (generation of longer 3' EST from SAGE tag for gene identification), we converted 1009 low-copy, human X chromosome-specific SAGE tags into 10210 3' ESTs. We identified 3418 unique 3' ESTs, 46% of which are novel and originated from the lower abundance transcripts. However, nearly all 3' ESTs were mapped to various regions across the genome but not X chromosome. Detailed analysis indicates that those 3' ESTs were isolated by SAGE tag mis-priming to the non-parent transcripts. Replacing SAGE tags with non-transcribed genomic DNA tags resulted in poor amplification, indicating that the sequence similarity between different transcripts contributed to the amplification. Our study shows the prevalence of novel low abundance transcripts that can be isolated efficiently through SAGE tags mis-priming.

Linking the human cytogenetic map with nucleotide sequence: the CCAP clone set

We present the completed dataset and clone repository of the Cancer Chromosome Aberration Project (CCAP), an initiative developed and funded through the intramural program of the U.S. National Cancer Institute, to provide seamless linkage of human cytogenetic markers with the primary nucleotide sequence of the human genome. Spaced at 1-2 Mb intervals across the human genome, 1,339 bacterial artificial chromosome (BAC) clones have been localized to chromosomal bands through high-resolution fluorescence in situ hybridization (FISH) mapping. Of these clones, 99.8% can be positioned on the primary human genome sequence and 95% are placed at or close to their precise nucleotide starts and stops. This dataset can be studied and manipulated within generally available public Web sites. The clones are available from a commercial repository. The CCAP BAC clone set provides anchors for the interrogation of gene and sequence involvement in oncogenic and developmental disorders when the starting point is the recognition of a structural, numerical, or interstitial chromosomal aberration. This dataset also provides a current view of the quality and coherence of the available genome sequence and insight into the nucleotide and three-dimensional structures that manifest as Giemsa light and dark chromosomal banding patterns.

Bioinformatics in Otolaryngology

This article reviews the recent literature on microarray analysis, bioinformatics techniques, and genomics in relation to the study of carcinogenesis of head and neck cancers.

Large-scale Transcriptome Analyses Reveal New Genetic Marker Candidates of Head, Neck, and Thyroid Cancer

A detailed genome mapping analysis of 213,636 expressed sequence tags (EST) derived from nontumor and tumor tissues of the oral cavity, larynx, pharynx, and thyroid was done. Transcripts matching known human genes were identified; potential new splice variants were flagged and subjected to manual curation, pointing to 788 putatively new alternative splicing isoforms, the majority (75%) being insertion events. A subset of 34 new splicing isoforms (5% of 788 events) was selected and 23 (68%) were confirmed by reverse transcription-PCR and DNA sequencing. Putative new genes were revealed, including six transcripts mapped to well-studied chromosomes such as 22, as well as transcripts that mapped to 253 intergenic regions. In addition, 2,251 noncoding intronic RNAs, eventually involved in transcriptional regulation, were found. A set of 250 candidate markers for loss of heterozygosis or gene amplification was selected by identifying transcripts that mapped to genomic regions previously known to be frequently amplified or deleted in head, neck, and thyroid tumors. Three of these markers were evaluated by quantitative reverse transcription-PCR in an independent set of individual samples. Along with detailed clinical data about tumor origin, the information reported here is now publicly available on a dedicated Web site as a resource for further biological investigation. This first in silico reconstruction of the head, neck, and thyroid transcriptomes points to a wealth of new candidate markers that can be used for future studies on the molecular basis of these tumors. Similar analysis is warranted for a number of other tumors for which large EST data sets are available.

Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation

Human embryonic stem (hES) cells hold promise for generating an unlimited supply of cells for replacement therapies. To characterize hES cells at the molecular level, we obtained 148,453 expressed sequence tags (ESTs) from undifferentiated hES cells and three differentiated derivative subpopulations. Over 32,000 different transcripts expressed in hES cells were identified, of which more than 16,000 do not match closely any gene in the UniGene public database. Queries to this EST database revealed 532 significantly upregulated and 140 significantly downregulated genes in undifferentiated hES cells. These data highlight changes in the transcriptional network that occur when hES cells differentiate. Among the differentially regulated genes are several components of signaling pathways and transcriptional regulators that likely play key roles in hES cell growth and differentiation. The genomic data presented here may facilitate the derivation of clinically useful cell types from hES cells.

Down's syndrome-associated Single Minded 2 gene as a pancreatic cancer drug therapy target

We report here a pancreatic cancer drug therapy utility of a gene involved in Down's syndrome. Single Minded 2 gene (SIM2) from Down's Syndrome Critical Region was expressed in pancreatic cancer-derived cell lines as well as in tumor tissues, but not in the normal pancreas. A related member of the SIM family, SIM1, did not show similar specificity. Inhibition by antisense technology of one of the isoforms of SIM2, the short-form (SIM2-s) expression in the CAPAN-1 pancreatic cancer cell line, caused a pronounced growth inhibition and induced cell death through apoptosis. The specificity of antisense was inferred from inhibition of SIM2-s mRNA but not the related members of SIM family. In view of the high mortality rate of pancreatic cancer patients, these findings have important implications for the future of pancreatic cancer treatment.

The generation and utilization of a cancer-oriented representation of the human transcriptome by using expressed sequence tags

Whereas genome sequencing defines the genetic potential of an organism, transcript sequencing defines the utilization of this potential and links the genome with most areas of biology. To exploit the information within the human genome in the fight against cancer, we have deposited some two million expressed sequence tags (ESTs) from human tumors and their corresponding normal tissues in the public databases. The data currently define approximately 23,500 genes, of which only approximately 1,250 are still represented only by ESTs. Examination of the EST coverage of known cancer-related (CR) genes reveals that <1% do not have corresponding ESTs, indicating that the representation of genes associated with commonly studied tumors is high. The careful recording of the origin of all ESTs we have produced has enabled detailed definition of where the genes they represent are expressed in the human body. More than 100,000 ESTs are available for seven tissues, indicating a surprising variability of gene usage that has led to the discovery of a significant number of genes with restricted expression, and that may thus be therapeutically useful. The ESTs also reveal novel nonsynonymous germline variants (although the one-pass nature of the data necessitates careful validation) and many alternatively spliced transcripts. Although widely exploited by the scientific community, vindicating our totally open source policy, the EST data generated still provide extensive information that remains to be systematically explored, and that may further facilitate progress toward both the understanding and treatment of human cancers.

What can digital transcript profiling reveal about human cancers?

Important biological and clinical features of malignancy are reflected in its transcript pattern. Recent advances in gene expression technology and informatics have provided a powerful new means to obtain and interpret these expression patterns. A comprehensive approach to expression profiling is serial analysis of gene expression (SAGE), which provides digital information on transcript levels. SAGE works by counting transcripts and storing these digital values electronically, providing absolute gene expression levels that make historical comparisons possible. SAGE produces a comprehensive profile of gene expression and can be used to search for candidate tumor markers or antigens in a limited number of samples. The Cancer Genome Anatomy Project has created a SAGE database of human gene expression levels for many different tumors and normal reference tissues and provides online tools for viewing, comparing, and downloading expression profiles. Digital expression profiling using SAGE and informatics have been useful for identifying genes that have a role in tumor invasion and other aspects of tumor progression.

Identification of Down's syndrome critical locus gene SIM2-s as a drug therapy target for solid tumors

We report here a cancer drug therapy use of a gene involved in Down's syndrome. Using bioinformatics approaches, we recently predicted Single Minded 2 gene (SIM2) from Down's syndrome critical region to be specific to certain solid tumors. Involvement of SIM2 in solid tumors has not previously been reported. Intrigued by a possible association between a Down's syndrome gene and solid tumors, we monitored SIM2 expression in solid tumors. Isoform-specific expression of SIM2 short-form (SIM2-s) was seen selectively in colon, prostate, and pancreatic carcinomas but not in breast, lung, or ovarian carcinomas nor in most normal tissues. In colon tumors, SIM2-s expression was seen in early stages. Antisense inhibition of SIM2-s expression in a colon cancer cell line caused inhibition of gene expression, growth inhibition, and apoptosis. The administration of the antisense, but not the control, oligonucleotides caused a pronounced inhibition of tumor growth in nude mice with no major toxicity. Our findings provide a strong rationale for the genes-to-drugs paradigm, establish SIM2-s as a molecular target for cancer therapeutics, and may further understanding of the cancer risk of Down's syndrome patients.

Gene discovery in oral squamous cell carcinoma through the Head and Neck Cancer Genome Anatomy Project: confirmation by microarray analysis

The near completion of the human genome project and the recent development of novel, highly sensitive high-throughput techniques have now afforded the unique opportunity to perform a comprehensive molecular characterization of normal, precancerous, and malignant cells, including those derived from squamous carcinomas of the head and neck (HNSCC). As part of these efforts, representative cDNA libraries from patient sets, comprising of normal and malignant squamous epithelium, were generated and contributed to the Head and Neck Cancer Genome Anatomy Project (HN-CGAP). Initial analysis of the sequence information indicated the existence of many novel genes in these libraries [Oral Oncol 36 (2000) 474]. In this study, we surveyed the available sequence information using bioinformatic tools and identified a number of known genes that were differentially expressed in normal and malignant epithelium. Furthermore, this effort resulted in the identification of 168 novel genes. Comparison of these clones to the human genome identified clusters in loci that were not previously recognized as being altered in HNSCC. To begin addressing which of these novel genes are frequently expressed in HNSCC, their DNA was used to construct an oral-cancer-specific microarray, which was used to hybridize alpha-(33)P dCTP labeled cDNA derived from five HNSCC patient sets. Initial assessment demonstrated 10 clones to be highly expressed (>2-fold) in the normal squamous epithelium, while 14 were highly represented in the malignant counterpart, in three of the five patient sets, thus suggesting that a subset of these newly discovered transcripts might be highly expressed in this tumor type. These efforts, together with other multi-institutional genomic and proteomic initiatives are expected to contribute to the complete understanding of the molecular pathogenesis of HNSCCs, thus helping to identify new markers for the early detection of preneoplastic lesions and novel targets for pharmacological intervention in this disease.

An international database and integrated analysis tools for the study of cancer gene expression

Researchers working collaboratively in Brazil and the United States have assembled an International Database of Cancer Gene Expression. Several strategies have been employed to generate gene expression data including expressed sequence tags (ESTs), serial analysis of gene expression (SAGE), and open reading-frame expressed sequence tags (ORESTES). The database contains six million gene tags that reflect the gene expression profiles in a wide variety of cancerous tissues and their normal counterparts. All sequences are deposited in the public databases, GenBank and SAGEmap. A suite of informatics tools was designed to facilitate in silico analysis of the gene expression datasets and are available through the NCI Cancer Genome Anatomy Project web site (http://cgap.nci.nih.gov).

Construction of cDNA libraries from microdissected benign and malignant thyroid tissue

cDNA libraries were constructed from thyroid epithelial cells gained by laser capture microdissection for gene expression analysis of the progression of thyroid cancer. Six histologically diverse thyroid tissue specimens were used. A mean of 93 ng of total RNA was gained per tissue sample from a mean estimated number of 25,000 microdissected cells per sample. Analysis of randomly selected clones from six libraries showed an average insert size of 600 (range, 300-1500) bp. Preliminary sequencing of clones selected from the six libraries indicates a range of 46% to 62% known genes per library, 4% to 25% anonymous expressed sequence tags per library, and 15% to 43% novel expressed sequence tags per library. Thyroglobulin was found in normal thyroid epithelium and follicular thyroid adenoma, whereas calcitonin precursor transcripts were found in medullary thyroid carcinoma. We demonstrate production of high-quality cDNA libraries of microdissected tissue of the thyroid, which should prove useful for gene expression analysis of human thyroid tumors.

Expression of hpttg proto-oncogene in lymphoid neoplasias

Pituitary tumor-transforming gene (pttg) is a distinct proto-oncogene which is expressed in certain normal tissues with high proliferation rate and in a variety of tumors. PTTG is the vertebrate analog of yeast securins Pds1 and Cut2 with a key role in the regulation of sister chromatid separation during mitosis. Impairment of PTTG regulated functions is expected to lead to chromosomal instability and aneuploidy. Human pttg (hpttg) is abundantly expressed in Jurkat T lymphoblastic lymphoma cells but not in normal peripheral blood leukocytes. To obtain additional data on the potential role of hpttg in lymphomagenesis we selected 150 cases of lymphoid tumors for the assessment of hpttg expression in tumor tissues. Immunohistochemical studies on formalin-fixed, paraffin-embedded tissues revealed hPTTG in 38.8% of B-cell lymphomas, 70.2% of T-cell lymphomas, and 73.1% of Hodgkin's lymphomas. Among B-cell lymphomas, the most frequently immunostained tumors were plasma cell tumors, diffuse large cell lymphomas, and follicle center cell lymphomas. In Hodgkin's disease, immunoreactivity was mainly noted in Reed-Sternberg cells. In conclusion, the frequent overexpression of hpttg in many histological subtypes of lymphoma suggests the involvement of this proto-oncogene in lymphomagenesis.

Identifying novel transcripts and novel genes in the human genome by using novel SAGE tags

The number of genes in the human genome is still a controversial issue. Whereas most of the genes in the human genome are said to have been physically or computationally identified, many short cDNA sequences identified as tags by use of serial analysis of gene expression (SAGE) do not match these genes. By performing experimental verification of more than 1,000 SAGE tags and analyzing 4,285,923 SAGE tags of human origin in the current SAGE database, we examined the nature of the unmatched SAGE tags. Our study shows that most of the unmatched SAGE tags are truly novel SAGE tags that originated from novel transcripts not yet identified in the human genome, including alternatively spliced transcripts from known genes and potential novel genes. Our study indicates that by using novel SAGE tags as probes, we should be able to identify efficiently many novel transcripts/novel genes in the human genome that are difficult to identify by conventional methods.

Comprehensive resource: Skeletal gene database

The Skeletal Gene Database (SGD) is an integrated resource that provides comprehensive information about bone-related genes, mRNA, and proteins expressed in human and mouse, with rich links to numerous other electronic tools. SGD contains expressed sequence tag (EST) data from all the skeletal-related cDNA libraries that are available to the public. It supplies the query/data access analytic tools for users to search and compare each gene expressed in skeletal tissue(s). The results derived from EST tissue expression profiling will allow users to get the data on the mRNA copy numbers of each gene expressed in each tissue and its normalized value. From the SGD, researchers can obtain information regarding the name, symbol, size, exon/intron number, chromosomal location, LocusLink, and related disease (if any is known) of each gene. This electronic compendium also furnishes information on the protein of the corresponding gene including the protein size (amino acid number and molecular weight). It provides swift and ready access to other useful databases including OMIM, UniGene and PUBMED. The data will be updated regularly in step with current and future research, thereby providing what we hope will serve as a highly useful source of information and a powerful analytic tool to the scientific community.

A research-based tumor tissue bank of gynecologic oncology: characteristics of nucleic acids extracted from normal and tumor tissues from different sites

This article describes a gynecology and pathology-oriented tumor tissue bank that is approaching the research requirements of modern molecular oncology and compared characteristics of nucleic acids extracted from preserved tissues. Through August 2000, 8869 specimens, including fresh neoplastic tissues and normal counterparts, body fluids (ascites, tumor content, and blood), and cervical scrapings, were procured from 1853 patients. DNA and RNA were extracted from a random sampling of normal (n = 50) and tumor (n = 53) tissues from the uterine cervix (n = 47), endometrium (n = 24), and ovary (n = 32). As expected, tumor tissues conferred a higher yield of DNA (1.56 +/- 1.24 versus 0.94 +/- 0.72 microg/mg tissue, P = 0.001) and RNA (5.04 +/- 6.21 versus 2.12 +/- 1.76 microg/ml, P < 0.001) than normal tissues. However, the RNA message abundance, as measured by RNA yield/DNA yield, was not different between tumor and normal tissues. With a similar content of DNA in the endometrium, uterine cervix, and ovary, RNA yield was higher in the endometrium than the others (P = 0.013). In tumors from these three sites, similar yields of DNA and RNA were noted. Overall the yield of DNA remained unchanged from specimens preserved for as long as 7 years, although at this length of storage, RNA yield became lower and variable. This study provides the basic characteristics of nucleic acids derived from normal and tumor tissues and ensures future research utility of these frozen specimens.

Bioinformatics tools for whole genomes

The advent of whole-genome data resources--not only sequence but also other genome-scale data collections such as gene expression, protein interaction, and genetic variation--is having two marked, complementary effects on the relatively new discipline of bioinformatics. First, the veritable flood of data is creating a need and demand for new tools for dealing adequately with the deluge, and, second, the unprecedented extent, diversity, and impending completeness of the data sets are creating opportunities for new approaches to discovery based on computational methods.

Gene expression profile of human bone marrow stromal cells: high-throughput expressed sequence tag sequencing analysis

Human bone marrow stromal cells (HBMSC) are pluripotent cells with the potential to differentiate into osteoblasts, chondrocytes, myelosupportive stroma, and marrow adipocytes. We used high-throughput DNA sequencing analysis to generate 4258 single-pass sequencing reactions (known as expressed sequence tags, or ESTs) obtained from the 5' (97) and 3' (4161) ends of human cDNA clones from a HBMSC cDNA library. Our goal was to obtain tag sequences from the maximum number of possible genes and to deposit them in the publicly accessible database for ESTs (dbEST of the National Center for Biotechnology Information). Comparisons of our EST sequencing data with nonredundant human mRNA and protein databases showed that the ESTs represent 1860 gene clusters. The EST sequencing data analysis showed 60 novel genes found only in this cDNA library after BLAST analysis against 3.0 million ESTs in NCBI's dbEST database. The BLAST search also showed the identified ESTs that have close homology to known genes, which suggests that these may be newly recognized members of known gene families. The gene expression profile of this cell type is revealed by analyzing both the frequency with which a message is encountered and the functional categorization of expressed sequences. Comparing an EST sequence with the human genomic sequence database enables assignment of an EST to a specific chromosomal region (a process called digital gene localization) and often enables immediate partial determination of intron/exon boundaries within the genomic structure. It is expected that high-throughput EST sequencing and data mining analysis will greatly promote our understanding of gene expression in these cells and of growth and development of the skeleton.

Identification of alternate polyadenylation sites and analysis of their tissue distribution using EST data

Alternate polyadenylation affects a large fraction of higher eucaryote mRNAs, producing mature transcripts with 3' ends of variable length. This variation is poorly represented in the current transcript catalogs derived from whole genome sequences, mostly because such posttranscriptional events are not detectable directly at the DNA level. Alternate polyadenylation of an mRNA is better understood by comparison to EST databases. Comparing ESTs to mRNAs, however, is a difficult task subjected to the pitfalls of internal priming, presence of intron sequences, repeated elements, chimerical ESTs or matches with EST from paralogous genes. We present here a computer program that addresses these problems and displays ESTs matches to a query mRNA sequence to predict alternate polyadenylation and to suggest library-specific forms. The output highlights effective polyadenylation signals, possible sources of artifacts such as A-rich stretches in the mRNA sequences, and allows for a direct visualization of EST libraries using color codes. Statistical biases in the distribution of alternative mRNA forms among EST libraries were systematically sought. About 1450 human and 200 mouse mRNAs displayed such biases, suggesting in each case a tissue- or disease-specific regulation of polyadenylation.

The Cancer Genome Anatomy Project: new resources for reading the molecular signatures of cancer

The Cancer Genome Anatomy Project (CGAP) has built informational, technological, and physical resources to interface genomics with basic and clinical cancer research. The CGAP web site (http://cgap.nci.nih.gov) provides informatics tools for in silico analysis of the CGAP datasets as well as information for accessing each of the CGAP resources. Published in 2001 by John Wiley & Sons, Ltd.

Sox8 gene expression identifies immature glial cells in developing cerebellum and cerebellar tumours

Sox8 is a member of the E subgroup of Sox genes, the other members of which are Sox9 and Sox10, both of which are implicated in specific human disorders. Recently, Sox8 homologues have been cloned in chick, mouse and human and have been shown to be strongly expressed in the embryonic and adult brain. Nevertheless, the cell types that express Sox8 have not been determined. We show here that Sox8 is expressed in immature glia in the developing cerebellum. Sox8 is also expressed in scattered cells in the cerebellar tumour, medulloblastoma. This gene therefore provides an early glial marker that may provide more detailed insight into the cellular makeup and consequent behaviour of medulloblastomas.

Genome analysis with gene-indexing databases

The recent release of the draft sequence and the eventual completion of the human genome present the scientific community with a rich source of data to mine. Yet, these data are content poor in the absence of additional correlative information. Expressed sequence tag (EST) datasets and their associated gene indices have existed for many years, and represent the first attempt at understanding the complexity of the genome. These datasets remain extremely important as information sources and, in particular, as tools for analyzing the completed genomes. Here, we discuss the nature of ESTs and their associated tools and gene-indexing databases. In particular, we will compare three EST gene indices (UNIGENE, Merck Gene Index Version 2.0 and Doubletwist CAT), discuss how these gene indices are applied for both genome analysis and drug discovery, and demonstrate their importance as a complementary dataset to the annotated human genome.

The use of common genetic polymorphisms to enhance the epidemiologic study of environmental carcinogens

Overwhelming evidence indicates that environmental exposures, broadly defined, are responsible for most cancer. There is reason to believe, however, that relatively common polymorphisms in a wide spectrum of genes may modify the effect of these exposures. We discuss the rationale for using common polymorphisms to enhance our understanding of how environmental exposures cause cancer and comment on epidemiologic strategies to assess these effects, including study design, genetic and statistical analysis, and sample size requirements. Special attention is given to sources of potential bias in population studies of gene--environment interactions, including exposure and genotype misclassification and population stratification (i.e., confounding by ethnicity). Nevertheless, by merging epidemiologic and molecular approaches in the twenty-first century, there will be enormous opportunities for unraveling the environmental determinants of cancer. In particular, studies of genetically susceptible subgroups may enable the detection of low levels of risk due to certain common exposures that have eluded traditional epidemiologic methods. Further, by identifying susceptibility genes and their pathways of action, it may be possible to identify previously unsuspected carcinogens. Finally, by gaining a more comprehensive understanding of environmental and genetic risk factors, there should emerge new clinical and public health strategies aimed at preventing and controlling cancer.

Chemoprevention of superficial bladder cancer

The development, evaluation and approval of promising agents for bladder cancer prevention (chemoprevention)depends upon the rational integration of four key components: a) Agents (pharmaceuticals, biologics and nutrients); b) Biomakers (intermediate endpoints that predict for clinical response and risk reduction; c) Cohorts (well defined high risk target populations d) Designs (efficient trial designs linked to the clinical phase of development). The promise of this overall strategy is the ability to conduct faster, smaller and more cost effective trials which incorporate validated surrogate endpoints rather than conventional clinical endpoints (cancer incidence, recurrence and survival). Current National Cancer Institute (NCI) phase III bladder cancer chemopreventive trials in progress are described. Since most patients with superficial (transitional cell) bladder cancer present with early disease (Ta, T1, Tis lesions) that frequently recurs and is easily accessible by serial cystoscopy and urine cytology, bladder cancer serves as a powerful clinical for conducting prevention trials of new agents for a tobacco related malignancy.

Sequence databases and microarrays as tools for identifying prostate cancer biomarkers

Identification, acquisition, and assessment of molecular markers that could be adopted as surrogate endpoints for evaluating a response to prostate cancer intervention strategies is highly desirable. Recent advances in the fields of genomics and biotechnology have dramatically increased the quantity and accessibility of molecular information that is relevant to the study of prostate carcinogenesis. One major advance involves the construction of comprehensive databases that archive gene sequences and gene expression data. This information is in a format suitable for virtual queries designed to distinguish the molecular differences between normal and cancer cells. A second major advance uses robotic tools to construct microarrays comprising thousands of distinct genes expressed in prostate tissues. Such arrays offer a powerful approach for monitoring the expression of thousands of genes simultaneously and provide access for techniques designed to assess patterns or "fingerprints" of gene expression that may ultimately be used as signatures of response to therapeutic intervention.

Prostate cancer chemoprevention: Strategies for designing efficient clinical trials

A chemoprevention (CP) strategy has evolved for conducting efficient clinical trials for prostate cancer (PCa) prevention. It integrates five key components, including agents, biomarkers, cohorts, designs, and endpoints. The rationale for the CP strategy relates to the natural history of prostate cancer. There is a wide array of natural and synthetic agents that hold promise for inhibiting, reversing, or modulating the transition from normal to precancer and from precancer to cancer. These agent classes include antiandrogens, antiestrogens, phytoestrogens, antioxidants, anti-inflammatory (proapoptotic) agents, antiproliferation/antidifferentiation agents, signal transduction modulators of receptor tyrosine kinase and ras farnesylation, antiangiogenesis agents, insulinlike growth factor (IGF)-1, peroxisome proliferator-activator receptor modulators (-gamma and -delta), and gene-based interventions. Biomarkers and endpoints are guided by the level of evidence required (eg, phase 1, 2, 3). Two candidate surrogate endpoints (SE) based on histology are high-grade prostatic intraepithelial neoplasia (HGPIN) and computer-assisted image analysis of dysplastic lesions. Phase 1 trials use standard endpoints of safety, pharmacokinetics and limited pharmacodynamics. Phase 2 trials use endpoints of modulation of biomarkers and correlation with histology. Phase 3 trials use endpoints of clinical benefit, such as cancer incidence reduction and quality of life. Validation of a biomarker as a SE involves correlation of the biomarker with clinical benefit. Cohorts (target populations) for phase 2/3 trials include the general population of men over age 50 with a normal prostate-specific antigen (PSA), subjects with a strong family history of PCa, subjects with elevated PSA/negative biopsy, and subjects with HGPIN/negative biopsy. These at-risk populations reflect key individual risk factors (age, race, serum PSA [free/total]; serum IGF-1/IGF binding protein (IGFBP)-3; 1, 25(OH)(2) D3; family history of PCa; carriers of PCa susceptibility genes [ELAC2, CYP3A4, SRD5A2, etc.]; and histology such as atypia and HGPIN) that could be combined into a multivariate risk model for PCa. The probability of cancer risk (recurrence) is a key factor that impacts on the clinical trial design (power, sample size, and primary endpoint). Multivariate predictive mathematical models for biochemical recurrence after radical prostatectomy by decreasing sample size and time to clinical outcomes maximize trial efficiency and identify the patients most likely to benefit from secondary prevention. The two large primary prevention trials, Prostate Cancer Prevention Trial/Seleninium and Vitamin E Chemoprevention Trial (PCPT/ SELECT), in low- and average-risk subjects have sample sizes of 18,000 to 32,000, with a treatment duration of 7 years to detect a 25% reduction in biopsy-proven PCa. Subjects with HGPIN have the highest known cancer risk (approximately 50% at 3 years), and thus require a small sample size (n = 450) to detect a 33% reduction in cancer incidence. A schema involving three sequential trials for agent registration is described. In summary, a CP strategy that incorporates well-defined agents, clinical and validated SE, and high-risk cohorts defined by genetic and acquired risk factors in a series of well-designed randomized controlled trials provides an efficient pathway for evaluating and approving new agents for PCa prevention.

A bacterial artificial chromosome library for sequencing the complete human genome

A 30-fold redundant human bacterial artificial chromosome (BAC) library with a large average insert size (178 kb) has been constructed to provide the intermediate substrate for the international genome sequencing effort. The DNA was obtained from a single anonymous volunteer, whose identity was protected through a double-blind donor selection protocol. DNA fragments were generated by partial digestion with EcoRI (library segments 1--4: 24-fold) and MboI (segment 5: sixfold) and cloned into the pBACe3.6 and pTARBAC1 vectors, respectively. The quality of the library was assessed by extensive analysis of 169 clones for rearrangements and artifacts. Eighteen BACs (11%) revealed minor insert rearrangements, and none was chimeric. This BAC library, designated as "RPCI-11," has been used widely as the central resource for insert-end sequencing, clone fingerprinting, high-throughput sequence analysis and as a source of mapped clones for diagnostic and functional studies.

Identification and characterization of SEB, a novel protein that binds to the acute undifferentiated leukemia-associated protein SET

SET, the translocation breakpoint-encoded protein in acute undifferentiated leukemia (AUL), is a 39-kDa nuclear phosphoprotein and has an inhibitory activity for protein phosphatase 2A (PP2A). SET is fused to a putative oncoprotein, CAN/NUP214, in AUL and is thought to play a key role in leukemogenesis by its nuclear localization, protein-protein interactions and PP2A inhibitory activity. Here, we describe the isolation and characterization of a novel cDNA encoding a protein with 1542 amino-acid residues that specifically interacts in a yeast two-hybrid system as well as in human cells with SET. This new protein, which we name SEB (SET-binding protein), is identified as a 170-kDa protein by immunoprecipitation with a specific antibody and is localized predominantly in the nucleus. SEB1238--1434 is determined as a SET-binding region that specifically binds to SET182--223. SEB also has an oncoprotein Ski homologous region (amino acids 654--858), six PEST sequences and three sequential PPLPPPPP repeats at the C-terminus. SEB mRNA is expressed ubiquitously in all human adult tissues and cells examined. The SEB gene locus is assigned to the chromosome 18q21.1 that contains candidate tumor suppressor genes associated with deletions in cancer and leukemia. Although the function of SEB is not known, we propose that SEB plays a key role in the mechanism of SET-related leukemogenesis and tumorigenesis, perhaps by suppressing SET function or by regulating the transforming activity of Ski in the nucleus.

Prostate cancer genomics

The molecular processes contributing to cancer of the human prostate gland are under intensive investigation. Methods used for discovering genetic alterations involved in prostate neoplasia include family studies designed to map hereditary disease loci, chromosomal studies to identify aberrations that may locate oncogenes or tumor suppressor genes, and comprehensive gene expression studies. These studies determine how various molecular signaling pathways influence or reflect the process of carcinogenesis. However, a comprehensive overview of the cell is necessary to understand all of the dynamic interactions between genes, their protein products, and the network of cellular processes resulting in tumorigenesis. Unraveling the complexity of these systems in a timely manner involves the integration of computers, miniaturization, and automation into molecular biology. New biotechnologies such as the development of automated DNA sequencing and complementary DNA microarrays allow for a systematic, "discovery-driven" approach. These and other technologies afford a comprehensive view of biology and pathology that have the potential to fully characterize the processes involved in neoplasia and therefore provide potential targets for the therapy of prostate and other cancers.

Expression sequence tag-specific full-length cDNA cloning: actin cDNAs

Current strategies for cDNA cloning are based on construction of cDNA libraries and colony screening. The process of obtaining a full-length cDNA clone can be highly time and labor intensive. Using the human actin gene as a model target cDNA, we have developed an RNA-capture method for rapid cloning of full-length cDNAs. The approach involves the capture of mRNA with expressed sequence tag (EST)-derived, biotin labeled antisense "capture" primers and streptavidin-coated magnetic beads. Full-length cDNA is then synthesized from purified EST-specific mRNA and cloned directly into plasmid vectors. The results of using beta-actin-based capture primers on cytoplasmic RNA were the isolation of both beta- and gamma-actin cDNA clones. Of the 16 actin-specific cDNA clones analyzed, 15 (93%) were full-length. This approach for cloning full-length cDNAs from available ESTs or partial cDNA sequences will facilitate a more rapid and efficient characterization of gene structure and function.

A computational reconstruction of the adult human heart transcriptional profile

The reconstruction of the transcriptional profile of the adult human heart was attempted, by applying a bioinformatic and computational approach to UniGene data. A catalogue of 2077 expressed genes was produced. Over 1000 entries of the catalogue corresponded to putative novel genes. Highly expressed genes accounted for about 20% of the total. Almost all genes expressed in adult heart resulted to be active in at least one additional tissue and about 90% were found in over five additional tissues. A genomic map of 1364 genes expressed in heart, which also indicated chromosomal location, was produced, which could be conveniently used for the discovery of the determinants of gene-orphan heart diseases and for the detection of clusters of highly expressed genes. The catalogue and the genomic map of genes expressed in adult human heart are available on Internet at the sites: http://telethon.bio.unipd.it/GETProfiles/heart and http://telethon.bio.unipd.it/GETMaps/heart.

In silico cloning of novel endothelial-specific genes

The endothelium plays a pivotal role in many physiological and pathological processes and is known to be an exceptionally active transcriptional site. To advance our understanding of endothelial cell biology and to elucidate potential pharmaceutical targets, we developed a new database screening approach to permit identification of novel endothelial-specific genes. The UniGene gene index was screened using high stringency BLAST against a pool of endothelial expressed sequence tags (ESTs) and a pool of nonendothelial ESTs constructed from cell-type-specific dbEST libraries. UniGene clusters with matches in the endothelial pool and no matches in the nonendothelial pool were selected. The UniGene/EST approach was then combined with serial analysis of gene expression (SAGE) library subtraction and reverse transcription polymerase chain reaction to further examine interesting clusters. Four novel genes were identified and labeled: endothelial cell-specific molecules (ECSM) 1-3 and magic roundabout (similar to the axon guidance protein roundabout). In summary, we present a powerful novel approach for comparative expression analysis combining two datamining strategies followed by experimental verification.

Mining the human genome using microarrays of open reading frames

To test the hypothesis that the human genome project will uncover many genes not previously discovered by sequencing of expressed sequence tags (ESTs), we designed and produced a set of microarrays using probes based on open reading frames (ORFs) in 350 Mb of finished and draft human sequence. Our approach aims to identify all genes directly from genomic sequence by querying gene expression. We analysed genomic sequence with a suite of ORF prediction programs, selected approximately one ORF per gene, amplified the ORFs from genomic DNA and arrayed the amplicons onto treated glass slides. Of the first 10,000 arrayed ORFs, 31% are completely novel and 29% are similar, but not identical, to sequences in public databases. Approximately one-half of these are expressed in the tissues we queried by microarray. Subsequent verification by other techniques confirmed expression of several of the novel genes. Expressed sequence tags (ESTs) have yielded vast amounts of data, but our results indicate that many genes in the human genome will only be found by genomic sequencing.

Gene expression profiles in squamous cell carcinomas of the oral cavity: use of laser capture microdissection for the construction and analysis of stage-specific cDNA libraries

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer among men in the developed world affecting the oral cavity, salivary glands, larynx and pharynx. Utilizing tissue from patients with HNSCC, we sought to systematically identify and catalog genes expressed in HNSCC progression. Here, we demonstrate the successful use of laser capture microdissection for procuring pure populations of cells from patient tissue sets comprised of oral squamous cell carcinomas (OSCCs) and matching normal tissue. From the estimated 5000 cells procured for each sample, we were able to extract total RNA (14.7-18.6 ng) of sufficient quality to transcribe GAPDH by reverse transcriptase-polymerase chain reaction (RT-PCR). The RNA was used for the synthesis of blunt-ended, double-strand complementary DNAs (cDNAs) by oligo (dT)-mediated reverse transcription, followed by addition of linkers. Primers specific for these linkers with uracil deglycosylase-compatible ends were used to amplify these cDNAs by PCR and the product was subcloned into the pAMP10 cloning vector. Ninety-six clones from each of six libraries were randomly sequenced and results indicated that 76-96% of the inserts represent either anonymous expressed sequence tags (ESTs) (25-48%), known genes (9-29%) or novel sequences (27-51%), respectively, with very little redundancy. These results demonstrate that high quality, representative cDNA libraries can be generated from microdissected OSCC tissue. Furthermore, these finding suggest the existence of at least 132 novel genes expressed in our cDNA libraries, which may have a role in the pathogenesis of HNSCC, and may represent novel markers for early detection as well as targets for pharmacological intervention in this disease.

Laser capture microdissection in pathology

The molecular examination of pathologically altered cells and tissues at the DNA, RNA, and protein level has revolutionised research and diagnostics in pathology. However, the inherent heterogeneity of primary tissues with an admixture of various reactive cell populations can affect the outcome and interpretation of molecular studies. Recently, microdissection of tissue sections and cytological preparations has been used increasingly for the isolation of homogeneous, morphologically identified cell populations, thus overcoming the obstacle of tissue complexity. In conjunction with sensitive analytical techniques, such as the polymerase chain reaction, microdissection allows precise in vivo examination of cell populations, such as carcinoma in situ or the malignant cells of Hodgkin's disease, which are otherwise inaccessible for conventional molecular studies. However, most microdissection techniques are very time consuming and require a high degree of manual dexterity, which limits their practical use. Laser capture microdissection (LCM), a novel technique developed at the National Cancer Institute, is an important advance in terms of speed, ease of use, and versatility of microdissection. LCM is based on the adherence of visually selected cells to a thermoplastic membrane, which overlies the dehydrated tissue section and is focally melted by triggering of a low energy infrared laser pulse. The melted membrane forms a composite with the selected tissue area, which can be removed by simple lifting of the membrane. LCM can be applied to a wide range of cell and tissue preparations including paraffin wax embedded material. The use of immunohistochemical stains allows the selection of cells according to phenotypic and functional characteristics. Depending on the starting material, DNA, good quality mRNA, and proteins can be extracted successfully from captured tissue fragments, down to the single cell level. In combination with techniques like expression library construction, cDNA array hybridisation and differential display, LCM will allow the establishment of "genetic fingerprints" of specific pathological lesions, especially malignant neoplasms. In addition to the identification of new diagnostic and prognostic markers, this approach could help in establishing individualised treatments tailored to the molecular profile of a tumour. This review provides an overview of the technique of LCM, summarises current applications and new methodical approaches, and tries to give a perspective on future developments. In addition, LCM is compared with other recently developed laser microdissection techniques.

Identifying potential tumor markers and antigens by database mining and rapid expression screening

Genes expressed specifically in malignant tissue may have potential as therapeutic targets but have been difficult to locate for most cancers. The information hidden within certain public databases can reveal RNA transcripts specifically expressed in transformed tissue. To be useful, database information must be verified and a more complete pattern of tissue expression must be demonstrated. We tested database mining plus rapid screening by fluorescent-PCR expression comparison (F-PEC) as an approach to locate candidate brain tumor antigens. Cancer Genome Anatomy Project (CGAP) data was mined for genes highly expressed in glioblastoma multiforme. From 13 mined genes, seven showed potential as possible tumor markers or antigens as determined by further expression profiling. Now that large-scale expression information is readily available for many of the commonly occurring cancers, other candidate tumor markers or antigens could be located and evaluated with this approach.

Genomic views of the immune system*

Genomic-scale experimentation aims to view biological processes as a whole, yet with molecular precision. Using massively parallel DNA microarray technology, the mRNA expression of tens of thousands of genes can be measured simultaneously. Mathematical distillation of this flood of gene expression data reveals a deep molecular and biological logic underlying gene expression programs during cellular differentiation and activation. Genes that encode components of the same multi-subunit protein complex are often coordinately regulated. Coordinate regulation is also observed among genes whose products function in a common differentiation program or in the same physiological response pathway. Recent application of gene expression profiling to the immune system has shown that lymphocyte differentiation and activation are accompanied by changes of hundreds of genes in parallel. The databases of gene expression emerging from these studies of normal immune responses will be used to interpret the pathological changes in gene expression that accompany autoimmunity, immune deficiencies, and cancers of immune cells.

Comprehensive genetic analysis of cancer cells

Human cancer is viewed as a disorder of genes originating from the progeny of a single cell that has accumulated multiple genetic alterations. The genetic alterations include point mutation, chromosomal rearrangements and imbalances. Amplifications primarily involve oncogenes whose overexpression leads to growth deregulation, while deletions commonly target tumor suppressor genes that control cell cycle checkpoints and DNA repair mechanisms. With the advent of molecular cytogenetics procedures for global detection of genomic imbalances and for multicolor visualization of structural chromosome changes, as well as the completion of human genome mapping and the development of microarray technology for serial gene expression analysis of the entire genomes, a significant progress has been made in uncovering the molecular basis of cancer. The major challenge in cancer biology is to decipher the molecular anatomy of various cancers and to identify cancer-related genes that now comprise only a fraction of human genes. The complete genetic anatomy of specific cancers would allow a better understanding of the role of genetic alterations in carcinogenesis, provide diagnostic and prognostic markers and discriminate between cells at different stages of progression toward malignancy. This review highlights current technologies that are available to explore cancer cells and outlines their application to investigations in human hepatocellular carcinoma.

SAGEmap: a public gene expression resource

We have constructed a public gene expression data repository and online data access and analysis, WWW and FTP sites for serial analysis of gene expression (SAGE) data. The WWW and FTP components of this resource, SAGEmap, are located at http://www.ncbi.nlm.nih. gov/sage and ftp://ncbi.nlm.nih.gov/pub/sage, respectively. We herein describe SAGE data submission procedures, the construction and characteristics of SAGE tags to gene assignments, the derivation and use of a novel statistical test designed specifically for differential-type analyses of SAGE data, and the organization and use of this resource.

Competitive hybridization: theory and application in isolation and quantification of differentially regulated genes

Competitive hybridization is a simple yet powerful method that was developed to screen cDNA libraries for differentially regulated genes. The method is based on competition between unlabeled cDNA from the mRNA of one sample and labeled cDNA from another sample. By manipulating the amount of competing unlabeled cDNA, background signals from the nonregulated genes can be increased or reduced, enabling the signals from differentially regulated genes to be contrasted and to be identified in a quantitative manner. To demonstrate the feasibility of the method, we screened a citrus cDNA library for ethylene-induced genes and identified three genes with different levels of ethylene induction. The mathematical basis of the method and its possible application in gene chip technology are discussed.

Comprehensive analyses of prostate gene expression: convergence of expressed sequence tag databases, transcript profiling and proteomics

Several methods have been developed for the comprehensive analysis of gene expression in complex biological systems. Generally these procedures assess either a portion of the cellular transcriptome or a portion of the cellular proteome. Each approach has distinct conceptual and methodological advantages and disadvantages. We have investigated the application of both methods to characterize the gene expression pathway mediated by androgens and the androgen receptor in prostate cancer cells. This pathway is of critical importance for the development and progression of prostate cancer. Of clinical importance, modulation of androgens remains the mainstay of treatment for patients with advanced disease. To facilitate global gene expression studies we have first sought to define the prostate transcriptome by assembling and annotating prostate-derived expressed sequence tags (ESTs). A total of 55000 prostate ESTs were assembled into a set of 15953 clusters putatively representing 15953 distinct transcripts. These clusters were used to construct cDNA microarrays suitable for examining the androgen-response pathway at the level of transcription. The expression of 20 genes was found to be induced by androgens. This cohort included known androgen-regulated genes such as prostate-specific antigen (PSA) and several novel complementary DNAs (cDNAs). Protein expression profiles of androgen-stimulated prostate cancer cells were generated by two-dimensional electrophoresis (2-DE). Mass spectrometric analysis of androgen-regulated proteins in these cells identified the metastasis-suppressor gene NDKA/nm23, a finding that may explain a marked reduction in metastatic potential when these cells express a functional androgen receptor pathway.

Screening poly(dA/dT)- cDNAs for gene identification

Many genes expressed in the human genome have not been identified despite intensive efforts. We observed that the presence of long poly(dA/dT) sequences in the 3' end of cDNA templates contributes significantly to this problem, because the hybrids formed randomly between poly(dA) and poly(dT) sequences of unrelated cDNA templates lead to loss of many templates in the normalization/subtraction reactions. The low abundant copies, which account for the majority of the expressed genes, are affected in particular by this phenomenon. We have developed a strategy called screening poly(dA/dT)(-) cDNAs for gene identification to overcome this obstacle. Applying this strategy can significantly enhance the efficiency of genome-wide gene identification and should have an impact on many functional genomic studies in the postgenome era.