Brain transcriptomics in response to beta-naphthoflavone treatment in rainbow trout: The role of AhR signaling.

Polychlorinated biphenyls (PCBs) exposure disrupts steroid production in teleostean fishes. While this suppression of plasma steroid levels is thought to involve aryl hydrocarbon receptor (AhR) signaling, the target tissues impacted and the molecular mechanisms involved have rarely been addressed. We tested the hypothesis that AhR activation downregulates genes involved in neuroendocrine function, including the control of brain-pituitary-interrenal (BPI) and -gonadal (BPG) axes in rainbow trout. To elucidate receptor-specific signaling, we utilized a pharmacological approach using beta-naphthoflavone (BNF) and resveratrol (RVT) as AhR agonist and antagonist, respectively. The gene expression pattern in the brain was analysed using a low-density targeted trout cDNA array enriched with genes encoding proteins involved in endocrine signaling, stress response and metabolic adjustments. Upregulation of AhR and CYP1A1 gene expression with BNF and the inhibition of this response by RVT confirmed AhR-dependent signaling. RVT by itself impacted only a few genes, while BNF treatment significantly modulated the transcript level of 49 genes, many of which are involved in the neuroendocrine control of stress and reproduction. Of these, only 27% of the BNF-mediated transcriptional response was blocked by RVT, suggesting molecular regulation of neuroendocrine pathways that are also AhR-independent. Gene expression pattern for select genes seen with the microarray analysis was also confirmed using quantitative real-time PCR. Overall, our results reveal for the first time that BNF disrupts several key genes involved in the neuroendocrine control of stress and sex steroid biosynthesis, while the mode of action involves both AhR-dependent and -independent pathways in trout. Keywords: Aryl hydrocarbon receptor, rainbow trout, brain transcriptomics, resveratrol, beta-naphthoflavone The details of the experimental protocol was published recently (Aluru and Vijayan, 2006) and we only used the unstressed groups from this study for the brain transcriptomics analysis. Briefly, groups of 8 fish each were randomly distributed in 12 tanks (100 liter) and acclimated for two weeks prior to the experiment. Treatments were assigned to the tanks (3 tanks per treatment x 4 treatments) randomly and the fish were fed with feed laced with either vehicle (ethanol laced feed) or vehicle containing resveratrol (RVT; 20 mg/kg body mass/day), β -naphthoflavone (BNF; 10 mg/kg body mass/day) or a combination of BNF and RVT (RBNF; 20 mg RVT and 10 mg BNF/kg body mass/day) for 5 days. The feed was laced by evenly submerging in 95% ethanol alone (control) or ethanol containing RVT, BNF or RBNF to ensure adequate coating of the pellets. The ethanol was allowed to evaporate by air-drying and the feed was stored in a cool and dry place. The method and duration of feeding was based on our previous study (Aluru et al., 2005). Microarray experiments were designed to comply with minimum information about microarray experiment (MIAME) guidelines. The development and application of the targeted trout cDNA microarray has been described in detail elsewhere (Wiseman et al., in review). For hybridization, each sample was co-hybridized with a reference RNA pool (RP), made by pooling equal amounts of RNA from all the samples (4 fish per treatment X 4 treatments). Hence, there were a total of 16 array slides for this experiment, each consisting of a sample and a reference RNA pool. Dye swapping was carried out for each treatment such that two slides had samples labeled with Cy3 and reference RNA with Cy5, while the other two was the opposite.