Science Inventory

DOSE RESPONSE FROM HIGH THROUGHPUT GENE EXPRESSION STUDIES AND THE INFLUENCE OF TIME AND CELL LINE ON INFERRED MODE OF ACTION BY ONTOLOGIC ENRICHMENT (SOT)

Citation:

Black, M., A. Karmaus, M. Martin, J. Naciff, G. Daston, Russell S. Thomas, M. Andersen, AND B. Wetmore. DOSE RESPONSE FROM HIGH THROUGHPUT GENE EXPRESSION STUDIES AND THE INFLUENCE OF TIME AND CELL LINE ON INFERRED MODE OF ACTION BY ONTOLOGIC ENRICHMENT (SOT). Presented at Society of Toxicology Annual Meeting, San Diego, CA, March 22 - 26, 2015. https://doi.org/10.23645/epacomptox.5178799

Impact/Purpose:

The L1000 is a HT genomics platform that measures 1000 landmark genes to then computationally predict expression across a whole human genome equivalent array. We used the L1000 system with visualization tools to assess gene expression changes and ontologic enrichment for 9 agrochemicals at 9 concentrations across 5 cell lines at 6 and 24 hr after treatment. These findings suggest selection of appropriate time points and the use of multiple cell models should be considered in HT genomics strategies designed to inform chemical MOA.

Description:

Gene expression with ontologic enrichment and connectivity mapping tools is widely used to infer modes of action (MOA) for therapeutic drugs. Despite progress in high-throughput (HT) genomic systems, strategies suitable to identify industrial chemical MOA are needed. The L1000 is a HT genomics platform that measures 1000 landmark genes to then computationally predict expression across a whole human genome equivalent array. We used the L1000 system with visualization tools to assess gene expression changes and ontologic enrichment for 9 agrochemicals at 9 concentrations across 5 cell lines at 6 and 24 hr after treatment. The cell lines were the metabolically competent HepaRG line and 4 cancer cell lines: A549 (lung), A673 (bone), HT29 (colorectal) and MCF7 (breast). Genes significant for a monotonic dose response (log likelihood ratio test with permutations) were used for analyses. Differential gene expression levels varied significantly across cell lines and times, with no one cell line consistently responsive for any chemical at both times. A broad assessment of the 9 chemicals showed that A549 and HT29 cells (at 6 hr) and HepaRG and A673 cells (at 24hr) were typically the most responsive. At 6hr, A549 and HT29 cells tended to show significant gene enrichment of cell cycle mitotic processes. At 24hr, A673 cells tended to show gene enrichment of immune response, disease and cellular metabolic pathways. Comparison of HepaRG and A673 cell responses (at 24hr) for fenbuconazole, associated with liver effects, showed enrichment of common parent pathways (DNA repair, cell signaling) in both, but several child categories of these were identified only in HepaRGs. These findings suggest selection of appropriate time points and the use of multiple cell models should be considered in HT genomics strategies designed to inform chemical MOA.This abstract does not necessarily reflect U.S. EPA policy.

Record Details:

Record Type:DOCUMENT( PRESENTATION/ POSTER)
Product Published Date:03/26/2015
Record Last Revised:08/04/2016
OMB Category:Other
Record ID: 322930