Water pollution remains a major environmental concern, with increased toxic by-products being released into water bodies. Many of these chemical contaminants persist in the environment and bio-accumulate in aquatic organisms. At present, toxicological tests are mostly based on laboratory tests, and effective methods for monitoring wild aquatic environments remain lacking. In the present study, we used a well-characterized toxic chemical, 3,3′,4,4′,5-polychlorinated biphenyl (PCB126), as an example to try to identify common biomarker genes to be used for predictive toxicity of this toxic substance. First, we used two laboratory fish models, the zebrafish (Danio rerio) and medaka (Oryzias latipes), to expose PCB126 to obtain liver transcriptomic data by RNA-seq. Comparative transcriptomic analyses indicated generally conserved and concerted changes from the two species, thus validating the transcriptomic data for biomarker gene selection. Based on the common up- and downregulated genes in the two species, we selected nine biomarker genes to further test in other fish species. The first validation experiment was carried out using the third fish species, Mozambique tilapia (Oreochromis mossambicus), and essentially, all these biomarker genes were validated for consistent responses with the two laboratory fish models. Finally, to develop universal PCR primers suitable for potentially all teleost fish species, we designed degenerate primers and tested them in the three fish species as well as in another fish species without a genomic sequence available: guppy (Poecilia reticulata). We found all the biomarker genes showed consistent response to PCB126 exposure in at least 50% of the species. Thus, our study provides a promising strategy to identify common biomarker genes to be used for teleost fish analyses. By using degenerate PCR primers and analyzing multiple biomarker genes, it is possible to develop diagnostic PCR arrays to predict water contamination from any wild fish species sampled in different water bodies.
