You are here:
Hypothesis testing with computational modeling: linking aromatase inhibition with plasma vitellogenin dynamics in fathead minnows
Citation:
Cheng, W., Z. Qiang, A. Schroeder, Dan Villeneuve, G. Ankley, AND R. Conolly. Hypothesis testing with computational modeling: linking aromatase inhibition with plasma vitellogenin dynamics in fathead minnows. Society of Environmental Toxicology and Chemistry, Salt Lake City, UT, November 01 - 05, 2015.
Impact/Purpose:
Endocrine disruption in fish can affect reproduction and population trajectories. This abstract describes quantitative modeling as part of a larger EPA research project focusing on understanding effects of an aromatase inhibitor on vitellogenin levels in fish. Aromatase inhibition and vitellogenin production are key events in an adverse outcome pathway (AOP) linking exposure of fish to certain endocrine disrupters with adverse effects on reproduction.
Description:
Fadrozole inhibits aromatase (CYP19A), a key enzyme that converts testosterone to estradiol (E2). In fish, E2 concentrations control hepatic synthesis ofthe glycolipoprotein vitellogenin (VTG), an egg yolk precursor protein essential to oocyte development and larval survival. When fathead minnows wereexposed to fadrozole continuously for 8 days and then held in control water (without fadrozole) for an additional 20 days, we observed dose-dependentreductions in plasma E2 and VTG concentrations. While VTG concentrations dropped as soon as E2 level declined at the onset of fadrozole exposure,when fathead minnows were depurated in control water the recovery of VTG was delayed relative to that of E2. In an effort to understand the differentialresponses at the onset and offset of fadrozole exposure, a computational model of the hypothalamic-pituitary-gonadal (HPG) axis was developed and usedto evaluate two alternative hypotheses regarding the regulation of VTG. The first hypothesis involved a signaling network motif called type I coherentfeedforward loop operating in the liver, where hepatic synthesis of VTG requires both direct E2 signaling and an intermediate transcription factor induced byE2. The second hypothesis involves another type of network motif, a negative feedback loop operating in the ovary, where the synthesis of the ovariantransporter used to uptake VTG from the blood is negatively regulated by accumulated VTG in the ovary. For both loops, ultrasensitivity, which allowssignal amplification and threshold, was implemented by using Hill equations. Both models were able to recapitulate the observed differential VTG dynamics,suggesting these network motifs may operate in fathead minnows. While computational modeling cannot, by itself, reveal the actual biological identity of theregulatory circuits, it may, as we showed here, identify data gaps and propose candidate structures that could be investigated in the laboratory. Follow-upexperimental work is needed to test the two alternative hypotheses and differentiate between them. This is an abstract does not necessarily reflect EPApolicy. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.