MAGE-F1, a novel ubiquitously expressed member of the MAGE superfamily.

Most known members of the MAGE superfamily are expressed in tumors, testis and fetal tissues, which has been described as a cancer/testis or "CT" expression pattern. We have identified a novel member of this superfamily, MAGE-F1, which is expressed in all adult and fetal tissues tested. In addition to normal tissues, MAGE-F1 is expressed in many tumor types including ovarian, breast, cervical, melanoma and leukemia. MAGE-F1 is encoded on chromosome 3, identifying a sixth chromosomal location for a MAGE superfamily gene. The coding region of MAGE-F1 is contained within a single exon and includes a microsatellite repeat. Sequence analysis and expression profiles define a new class of ubiquitously expressed MAGE superfamily genes that includes MAGE-F1, MAGE-D1, MAGE-D2/JCL-1 and NDN. The finding that several MAGE genes are ubiquitously expressed suggests a role for MAGE encoded proteins in normal cell physiology. Furthermore, potential cross-reactivity to these ubiquitously expressed MAGE gene products should be considered in the design of MAGE-targeted immunotherapies for cancer.

Tissue classification with gene expression profiles.

Constantly improving gene expression profiling technologies are expected to provide understanding and insight into cancer-related cellular processes. Gene expression data is also expected to significantly aid in the development of efficient cancer diagnosis and classification platforms. In this work we examine three sets of gene expression data measured across sets of tumor(s) and normal clinical samples: The first set consists of 2,000 genes, measured in 62 epithelial colon samples (Alon et al., 1999). The second consists of approximately equal to 100,000 clones, measured in 32 ovarian samples (unpublished extension of data set described in Schummer et al. (1999)). The third set consists of approximately equal to 7,100 genes, measured in 72 bone marrow and peripheral blood samples (Golub et al, 1999). We examine the use of scoring methods, measuring separation of tissue type (e.g., tumors from normals) using individual gene expression levels. These are then coupled with high-dimensional classification methods to assess the classification power of complete expression profiles. We present results of performing leave-one-out cross validation (LOOCV) experiments on the three data sets, employing nearest neighbor classifier, SVM (Cortes and Vapnik, 1995), AdaBoost (Freund and Schapire, 1997) and a novel clustering-based classification technique. As tumor samples can differ from normal samples in their cell-type composition, we also perform LOOCV experiments using appropriately modified sets of genes, attempting to eliminate the resulting bias. We demonstrate success rate of at least 90% in tumor versus normal classification, using sets of selected genes, with, as well as without, cellular-contamination-related members. These results are insensitive to the exact selection mechanism, over a certain range.

Support vector machine classification and validation of cancer tissue samples using microarray expression data.

MOTIVATION: DNA microarray experiments generating thousands of gene expression measurements, are being used to gather information from tissue and cell samples regarding gene expression differences that will be useful in diagnosing disease. We have developed a new method to analyse this kind of data using support vector machines (SVMs). This analysis consists of both classification of the tissue samples, and an exploration of the data for mis-labeled or questionable tissue results. RESULTS: We demonstrate the method in detail on samples consisting of ovarian cancer tissues, normal ovarian tissues, and other normal tissues. The dataset consists of expression experiment results for 97,802 cDNAs for each tissue. As a result of computational analysis, a tissue sample is discovered and confirmed to be wrongly labeled. Upon correction of this mistake and the removal of an outlier, perfect classification of tissues is achieved, but not with high confidence. We identify and analyse a subset of genes from the ovarian dataset whose expression is highly differentiated between the types of tissues. To show robustness of the SVM method, two previously published datasets from other types of tissues or cells are analysed. The results are comparable to those previously obtained. We show that other machine learning methods also perform comparably to the SVM on many of those datasets. AVAILABILITY: The SVM software is available at http://www.cs. columbia.edu/ approximately bgrundy/svm.

Evolution of Antp-class genes and differential expression of Hydra Hox/paraHox genes in anterior patterning.

The conservation of developmental functions exerted by Antp-class homeoproteins in protostomes and deuterostomes suggested that homologs with related functions are present in diploblastic animals. Our phylogenetic analyses showed that Antp-class homeodomains belong either to non-Hox or to Hox/paraHox families. Among the 13 non-Hox families, 9 have diploblastic homologs, Msx, Emx, Barx, Evx, Tlx, NK-2, and Prh/Hex, Not, and Dlx, reported here. Among the Hox/paraHox, poriferan sequences were not found, and the cnidarian sequences formed at least five distinct cnox families. Two are significantly related to the paraHox Gsx (cnox-2) and the mox (cnox-5) sequences, whereas three display some relatedness to the Hox paralog groups 1 (cnox-1), 9/10 (cnox-3) and the paraHox cdx (cnox-4). Intermediate Hox/paraHox genes (PG 3 to 8 and lox) did not have clear cnidarian counterparts. In Hydra, cnox-1, cnox-2, and cnox-3 were not found chromosomally linked within a 150-kb range and displayed specific expression patterns in the adult head. During regeneration, cnox-1 was expressed as an early gene whatever the polarity, whereas cnox-2 was up-regulated later during head but not foot regeneration. Finally, cnox-3 expression was reestablished in the adult head once it was fully formed. These results suggest that the Hydra genes related to anterior Hox/paraHox genes are involved at different stages of apical differentiation. However, the positional information defining the oral/aboral axis in Hydra cannot be correlated strictly to that characterizing the anterior-posterior axis in vertebrates or arthropods.

Negative selection: a method for obtaining low-abundance cDNAs using high-density cDNA clone arrays.

The identification of the entire complement of genes expressed in a cell, tissue, or organism provides a framework for understanding biological properties and establishes a tool set for subsequent functional studies. The large-scale sequencing of randomly selected clones from cDNA libraries has been successfully employed as a method for identifying a large fraction of these expressed genes. However, this approach is limited by the inherent redundancy of cellular transcripts reflecting widely variant levels of gene transcription. As a result, a high percentage of transcript duplications are encountered as the number of sequenced clones accrues. To address this problem, we have developed a negative hybridization selection method that employs the hybridization of complex cDNA probes to high-density arrays of cDNA clones and the subsequent selection of clones with a null or low hybridization signal. This approach was applied to a cDNA library constructed from normal human prostate tissue and resulted in the reduction of highly expressed prostate cDNAs from 6.8 to 0.57% with an overall decline in clone redundancy from 33 to 11%. The selected clones also reflected a more diverse cDNA population, with 89% of the clones representing distinctly different cDNAs compared with 67% of the randomly selected clones. This method compares favorably with cDNA library re-association normalization approaches and offers several distinct advantages, including the flexibility to use previously prepared libraries, and the ability to employ an iterative screening approach for continued accrual of cDNAs representing rare transcripts.

Comparative hybridization of an array of 21,500 ovarian cDNAs for the discovery of genes overexpressed in ovarian carcinomas.

Comparative hybridization of cDNA arrays is a powerful tool for the measurement of differences in gene expression between two or more tissues. We optimized this technique and employed it to discover genes with potential for the diagnosis of ovarian cancer. This cancer is rarely identified in time for a good prognosis after diagnosis. An array of 21,500 unknown ovarian cDNAs was hybridized with labeled first-strand cDNA from 10 ovarian tumors and six normal tissues. One hundred and thirty-four clones are overexpressed in at least five of the 10 tumors. These cDNAs were sequenced and compared to public sequence databases. One of these, the gene HE4, was found to be expressed primarily in some ovarian cancers, and is thus a potential marker of ovarian carcinoma.

Messenger RNA translation state: the second dimension of high-throughput expression screening.

Technological advances over the past 10 years have generated powerful tools for parallel analysis of complex biological problems. Among these new technologies, DNA arrays have provided an important experimental approach for identifying changes in the levels of individual mRNA molecules during important cellular transitions. However, cellular behavior is dictated not by mRNA levels, but by the proteins translated from the individual mRNA species. We report a high-throughput method for simultaneously monitoring the translation state and level of individual mRNA species. Messenger RNAs from resting and mitogenically activated fibroblasts were separated, according to degree of ribosome loading, into well-translated and under-translated pools. cDNA probes generated from these fractions were used to interrogate cDNA arrays. Among approximately 1,200 genes analyzed, less than 1% were found to be translationally regulated in response to mitogenic activation, demonstrating the strong selectivity of this regulatory mechanism. This high-throughput approach is shown to be an effective tool for superimposing translation profile on mRNA level for large numbers of genes, as well as for identifying translationally regulated genes for further study.

Monitoring gene expression profile changes in ovarian carcinomas using cDNA microarray.

The development of cancer is the result of a series of molecular changes occurring in the cell. These events lead to changes in the expression level of numerous genes that result in different phenotypic characteristics of tumors. In this report we describe the assembly and utilization of a 5766 member cDNA microarray to study the differences in gene expression between normal and neoplastic human ovarian tissues. Several genes that may have biological relevance in the process of ovarian carcinogenesis have been identified through this approach. Analyzing the results of microarray hybridizations may provides new leads for tumor diagnosis and intervention.

An expressed-sequence-tag database of the human prostate: sequence analysis of 1168 cDNA clones.

The human prostate is a complex glandular organ with functional development under hormonal regulation. Diseases of the prostate result in significant morbidity and mortality in the form of benign prostatic hypertrophy and prostate adenocarcinoma. The characterization of the molecular framework of the human prostate at the level of expressed genes will facilitate the understanding of normal and pathological prostate biology. The purposes of this study were to acquire an initial assessment of the qualitative and quantitative diversity of gene expression in the normal human prostate and to determine the extent that genes with prostate-restricted expression can be assessed using an expressed sequence tag approach. We have constructed a directional cDNA library from normal adult human prostate tissue and partially sequenced the 5' end of 1168 randomly selected cDNA clones, resulting in more than 400 kb of DNA sequence. Homology searches of the sequenced cDNAs against the GenBank and dbEST databases revealed that 43% of the sequences are identical to human genes whose functions are known, 5% are similar but not identical to known genes in humans or lower organisms, 5% match the mitochondrial genome, 9% are composed of interspersed DNA repeats, 30% are homologous to sequences in the dbEST database without a described function, and 6% are novel sequences. A total of 780 distinct species were identified. In addition to the 74 novel transcripts, 4 genes, prostate-specific antigen (PSA), prostate secretory protein (PSP), prostate acid phosphatase (PAP), and human glandular kallekrein 2 (HK2), have no homologous sequences in the databases that originate from sources other than prostate and thus may represent genes with prostate-restricted expression. Sequences matching PSA, PSP, and PAP each accounted for > 1% of the total ESTs and represent highly abundant transcripts, correlating with the abundance of these proteins in the prostate gland. No novel transcripts were represented by more than one EST and thus are expressed at levels much lower than the known prostate-specific genes.

Inexpensive handheld device for the construction of high-density nucleic acid arrays.

We report an easy-to-use, 384-pin handheld arraying and replicating device (ARD) for constructing high-density replicas of nucleic acids and E. coli transformants. We have modified an existing 384-pin tool to include a novel guide system to ensure vertical pin movement and enhance reproducibility. An asymmetric rectangular multiplexing frame is designed to increase the array density to 1536 dots on a standard microplate-size nylon membrane and to reduce the time and effort involved in producing array replicas. Our initial studies used the ARD to construct 1536-dot arrays of ovarian cDNA clones. We have hybridized these arrays with 32P-labeled probes, which resulted in distinctive signals for either visual interpretation or semiautomatic spot detection and signal integration.

High-throughput plasmid mini preparations facilitated by micro-mixing.

We have developed a reliable high-throughput plasmid isolation system using a 96-well plate format. This system combines a novel glass bead micro-mixing method with modified alkaline lysis and Sephacryl S-500 DNA purification procedures. Mechanical forces generated by vortexing glass beads inside each well of the 96-well plates ensure that the bacterial pellets are homogeneously resuspended, the cells are completely lyzed, and the resulting bacterial lysates are thoroughly mixed with the potassium acetate solution. The vortexing speed and duration for glass bead mixing have been standardized to facilitate plasmid DNA yields without significant adjustments.

HOM/HOX homeobox genes are present in hydra (Chlorohydra viridissima) and are differentially expressed during regeneration.

Hydra, a diblastic animal consisting of two cell layers, ectoderm and endoderm, is one of the most ancient animals displaying an anteroposterior axis with a head and a foot developing from an uncommitted gastric region. As such, hydra is an interesting model for studying the presence and function of homeobox genes in a phylogenetically old organism. By screening a Chlorohydra viridissima cDNA library with a 'guessmer' oligonucleotide, we have cloned several such cnidarian homeobox-containing genes (cnox genes). Two of these, cnox1 and cnox2, display labial and Deformed type homeodomains respectively and could represent two ancestral genes of the HOM/HOX complexes; cnox3 exhibits some similarity to the BarH1 and the distal-less type homeodomains and a fourth gene is highly related to the msh/Hox7 type of homeodomain. We used quantitative PCR to study levels of expression of these genes along the body axis and during head regeneration. In all cases, the expression in heads was stronger than that in the gastric region. cnox1 transcripts dramatically peaked within the first hours of head regeneration, whereas cnox2 and cnox3 reached their maximal levels 1 and 2 days after cutting respectively. This differential expression of homeobox genes at various stages of regeneration suggests that they play specific roles in regenerative processes.