Detection of diluted gene expression alterations using cDNA microarrays.

The use of DNA microarrays has spanned numerous disciplines of life science research. Despite the volume of studies utilizing this technology, no consensus exists on basic issues such as the determination of significantly altered genes in a given experiment, often leading to either false-negative or false-positive data. In this report, we study the effect of dilution of biological alterations on the detection level of gene expression differences using cDNA microarrays. We propose that subtle alterations in transcript levels of genes below the 2-fold level should be considered when replicate hybridizations are performed, because these subtle gene expression changes may be due to a robust response in few cells. We measured the effect of dilution of gene expression and found that differences in gene expression between the two cell lines assayed (HaCaT and MCF-7) were detected even after a 20-fold dilution factor. These results better our understanding of biological alterations that comprise a relatively small percentage of an assayed organ and help in the interpretation of gene expression data.

Progress in the application of DNA microarrays.

Microarray technology has been applied to a variety of different fields to address fundamental research questions. The use of microarrays, or DNA chips, to study the gene expression profiles of biologic samples began in 1995. Since that time, the fundamental concepts behind the chip, the technology required for making and using these chips, and the multitude of statistical tools for analyzing the data have been extensively reviewed. For this reason, the focus of this review will be not on the technology itself but on the application of microarrays as a research tool and the future challenges of the field.

Discovery in toxicology: mediation by gene expression array technology.

Toxicogenomics is a term that represents the merging of toxicology with novel genomics techniques. Data generated in the new-age era of toxicology is relatively complex, requires new bioinformatics tools for adequate interpretation, and allows for the rapid generation of testable hypotheses. Hazard identification and risk assessment processes will advance from the use of genomics techniques, which will lead to greater understanding of mechanism(s) of action of toxicants, development of novel biomarkers of exposure and effect, and better identification of sensitive subpopulations.

Arsenic disrupts cellular levels of p53 and mdm2: a potential mechanism of carcinogenesis.

The antitumor protein p53 plays a critical role in DNA repair. Inorganic arsenic exposure is associated with a wide variety of human tumors, particularly of the skin. To investigate how inorganic arsenic might interfere with DNA repair and lead to greater incidence of hyperkeratosis and skin tumors, we exposed human keratinocytes (HaCaT) to environmentally relevant concentrations of arsenite for 14 days. Arsenite reduced p53 levels while concomitantly increasing the p53 regulatory protein mdm2 levels in a dose- and time-dependent manner. We propose the disruption of the p53-mdm2 loop regulating cell cycle arrest as a model for arsenic-related skin carcinogenesis and it may be important in tumors with elevated mdm2 levels.

Arsenic binding proteins from human lymphoblastoid cells.

Arsenic is a ubiquitous contaminant of drinking water and food. The mechanisms of the toxic action of inorganic arsenic are unknown. We report the isolation of proteins having a high affinity for arsenic in the +3 oxidation state that are induced by arsenite (AsIII) in human lymphoblastoid cells. The arsenic-binding proteins were isolated using a p-aminophenylarsine oxide affinity column. At least four proteins of 50, 42, 38.5 and 19.5 kDa were isolated by elution with 10 or 100 mM 2-mercaptoethanol. Two proteins were tentatively identified as tubulin and actin on the basis of their molecular weights and previously reported affinity for the arsenic column. The identities of the remaining proteins are unknown. Heme oxygenase 1 was induced by AsIII but did not bind to the arsenic affinity column. We conclude that AsIII induces multiple proteins that have variable affinities for arsenic in the +3 state as judged by the concentration of 2-mercaptoethanol required for their elution. The arsenic binding motif of these proteins may involve three thiol groups arranged 3-6 A apart by the tertiary structure of the protein as suggested by others. These proteins may serve as high affinity binding sites for AsIII and may be involved in the biological action of AsIII.