Bernard W Futscher

PMID: 17674622;Abstract:

High dimensionality has been a major problem for gene array-based cancer classification. It is critical to identify marker genes for cancer diagnoses. We developed a framework of gene selection methods based on previous studies. This paper focuses on optimal search-based subset selection methods because they evaluate the group performance of genes and help to pinpoint global optimal set of marker genes. Notably, this paper is the first to introduce tabu search (TS) to gene selection from high-dimensional gene array data. Our comparative study of gene selection methods demonstrated the effectiveness of optimal search-based gene subset selection to identify cancer marker genes. TS was shown to be a promising tool for gene subset selection. © 2007 IEEE.

PMID: 21684831;PMCID: PMC3230447;Abstract:

Background: Chronic arsenic exposure is a worldwide health problem. How arsenic exposure promotes a variety of diseases is poorly understood, and specific relationships between experimental and human exposures are not established. We propose phenotypic anchoring as a means to unify experimental observations and disease outcomes. Objectives: We examined the use of phenotypic anchors to translate experimental data to human pathology and investigated research needs for which phenotypic anchors need to be developed. Methods: During a workshop, we discussed experimental systems investigating arsenic dose/exposure and phenotypic expression relationships and human disease responses to chronic arsenic exposure and identified knowledge gaps. In a literature review, we identified areas where data exist to support phenotypic anchoring of experimental results to pathologies from specific human exposures. Discussion: Disease outcome is likely dependent on cell-type-specific responses and interaction with individual genetics, other toxicants, and infectious agents. Potential phenotypic anchors include target tissue dosimetry, gene expression and epigenetic profiles, and tissue biomarkers. Conclusions: Translation to human populations requires more extensive profiling of human samples along with high-quality dosimetry. Anchoring results by gene expression and epigenetic profiling has great promise for data unification. Genetic predisposition of individuals affects disease outcome. Interactions with infectious agents, particularly viruses, may explain some species-specific differences between human pathologies and experimental animal pathologies. Invertebrate systems amenable to genetic manipulation offer potential for elaborating impacts of specific biochemical pathways. Anchoring experimental results to specific human exposures will accelerate understanding of mechanisms of arsenic-induced human disease.

PMID: 2104539;Abstract:

This study was initiated to determine if DNA-damaging chemotherapeutic agents can suppress the expression of oncogenes. The effects of three structurally related bifunctional alkylating agents on the steady state mRNA levels of c-myc, c-fos, N-ras, and β-actin in the human colon carcinoma cell line Colo320HSR were examined. Colo320HSR has an amplified c-myc oncogene, which is highly overexpressed, and is assumed to be one of the transforming genes of this cell line. Two concentrations of mechlorethamine, L-phenylalanine mustard, and 4-hydroperoxycyclophosphamide, which produced 1 or 3 log cell kills were used to examine the effects of drug exposure on the expression of specific genes. Steady state mRNA levels were measured by Northern blot analysis. Following a 1-h drug exposure, RNA was isolated from cells at 0, 6, 12, and 24 h following drug removal. The agents used produced changes in the expression of specific genes, and all three did so in a similar fashion. Immediately following drug removal, the steady state expression of c-myc in treated cells was increased 2- to 3-fold compared to control. At 6 and 12 h following drug removal, c-myc levels were depressed 2.5- to 5-fold. By 24 h, c-myc expression approached, but remained below, control levels. Immediately following drug removal, c-fos levels were increased 3- to 4-fold, and from 6 to 24 h following drug removal, c-fos levels gradually returned to, or fell below low basal levels. During the 24-h time course, drug treatment had little or no effect on the steady state levels of N-ras or β-actin. These data support the hypothesis that alkylating agents may suppress the expression of specific transforming genes.

PMID: 16229802;PMCID: PMC1501934;Abstract:

Epigenetic control participates in processes crucial in mammalian development, such as X-chromosome inactivation, gene imprinting, and cell type-specific gene expression. We provide evidence that the p53-inducible gene 14-3-3σ is a new example of a gene important to human cancer, where epigenetic mechanisms participate in the control of normal cell type-specific expression, as well as aberrant gene silencing in cancer cells. Like a previously identified cell type-specific gene maspin, 14-3-3σ is a p53-inducible gene; however, it participates in G2/M arrest in response to DNA-damaging agents. 14-3-3σ expression is restricted to certain epithelial cell types, including breast and prostate, whereas expression is absent in nonepithelial tissues such as fibroblasts and lymphocytes. In this report, we show that in normal cells expressing 14-3-3σ, the , 14-3-3σ CpG island is unmethylated; associated with acetylated histones, unmethylated histone H3 lysine 9; and an accessible chromatin structure. By contrast, normal cells that do not express 14-3-3σ have a methylated 14-3-3σ CpG island with hypoacetylated histones, methylated histone H3 lysine 9, and an inaccessible chromatin structure. These findings extend the spectrum of cell type-specific genes controlled partly by normal epigenetic mechanisms, and suggest that this subset of genes may represent important targets of epigenetic dysregulation in human cancer. Copyright © 2005 Neoplasia Press, Inc. All rights reserved.