Citation:

Martin, M. T., A Karmaus, M. Black, J. Naciff, G. Daston, E. Carney, M. Andersen, R. S. Thomas, AND B. Wetmore. Comparison of L1000 and Affymetrix Microarray for In Vitro Concentration-Response Gene Expression Profiling (SOT). Presented at SOT annual meeting, San Diego, CA, March 22 - 26, 2015. https://doi.org/10.23645/epacomptox.5178763

Impact/Purpose:

poster presented as SOT annual meeting in San Diego, CA on March 25, 2015

Description:

Advances in high-throughput screening technologies and in vitro systems have opened doors for cost-efficient evaluation of chemical effects on a diversity of biological endpoints. However, toxicogenomics platforms remain too costly to evaluate large libraries of chemicals in concentration-response. The current study evaluates gene expression profiling using the L1000 platform (Genometry), whereby 1000 landmark genes are quantified and a computational approach is applied to infer the expression of the remaining ~21000 genes in the genome. Human cell lines (HepG2, HepaRG, Ishikawa, MCF7) were treated with 34 unique chemicals at three concentrations for 6 hours. Purified RNA extracts were halved for gene expression evaluation by Affymetrix microarray and L1000 technologies to be conducted on the same samples. The raw normalized expression values generated by each platform were compared per gene, revealing a linear correlation of R2 = 0.63. Comparison of the gene expression profiles revealed differences in the dynamic range of fold-expression achieved per gene as well as differences in the number of differentially expressed genes (as calculated via ANOVA to account for multiple concentrations), though Connectivity Mapping (cmap; Broad Institute) identified similar profiles. For example, valproic acid (10-1000 uM) treatment in MCF7cells resulted in the differential expression of 373 and 453 genes in the Affymetrix microarray and L1000 platforms, respectively, with only 22 genes in common. However, the cmap for both datasets returned a valproic acid signature of histone deacetylase inhibition. In summary, despite differences in raw expression values, fold changes and genes identified, both platforms succeed in identifying biological signatures of chemical-mediated disruption. This abstract does not necessarily reflect US EPA policy.

URLs/Downloads:

Record Details: