You are here:
Development of transcriptomics-based biomarkers for selected endocrine disrupting chemicals in zebrafish (Danio rerio)
Citation:
WANG, R., D. C. BENCIC, A. D. BIALES, R. W. FLICK, J. M. LAZORCHAK, DAN VILLENEUVE, AND G. T. ANKLEY. Development of transcriptomics-based biomarkers for selected endocrine disrupting chemicals in zebrafish (Danio rerio). Presented at SETAC 2011, Boston, MA, November 13 - 17, 2011.
Impact/Purpose:
The objective of this task is to develop molecular indicators to evaluate the integrity and sustainability of aquatic fish, invertebrate, and plant communities (GPRA goal 4.5.2). Specifically, this subtask aims to evaluate methods for the measurement of: fish and invertebrate community composition, especially for morphologically indistinct (cryptic) species population genetic structure of aquatic indicator species and its relationship to landscape determinants of population structure (to aid in defining natural assessment units and to allow correlation of population substructure with regional stressor coverages) genetic diversity within populations of aquatic indicator species, as an indicator of vulnerability to further exposure and as an indicator of cumulative exposure patterns of temporal change in genetic diversity of aquatic indicator species, as a monitoring tool for establishing long-term population trends.
Description:
Genome-wide transcriptional profiling by microarrays provides a powerful platform for gene expression-based biomarker discovery. After their wide acceptance in human disease diagnosis, prognosis, and drug discovery, these gene signatures are increasingly being adopted for environmental toxicology applications as well. Built upon our previous work in zebrafish of reverse-engineered transcription factor (TF) networks and their statistical linkage to various treatment conditions of endocrine disrupting chemicals (EDCs), a systematic, computationally- intensive search using genetic algorithm coupled with support vector machine was conducted for candidate gene signatures capable of distinguishing individual chemical treatment conditions from their controls. Microarray data from zebrafish brain (male, female, or combined), ovary, and testis were analyzed. The search within each tissue type was either confined to hundreds of individual TF networks si;gnificantly associated with EDCs of differing mechanisms/modes of action, or across the entire genome. In agreement with our previous findings, tissue type was found to be critical to biomarker search. Brain yielded most candidate gene signatures, followed by ovary and testis. By chemical, prochloraz had the greatest number of candidates, then flutamide, fipronil, and enthinyl estradiol. A subset of these candidates will be evaluated using an independently derived microarray dataset generated from additional EDC-exposed zebrafish samples.