Occurrence and biological effect of exogenous steroids in the Elkhorn River, Nebraska, USA
Published in Science of the Total Environment 388:1-3 (December 15, 2007), pp. 104–115; doi: 10.1016/j.scitotenv.2007.08.001 Copyright © 2007 Elsevier B.V. Used by permission. http://www.elsevier.com/locate/scitotenv
Recent studies of surface waters in North America, Japan, and Europe have reported the presence of steroidogenic agents as contaminants. The current study has three objectives: 1) to determine if steroidogenic compounds are present in the Elkhorn River, 2) to determine if sediments collected from the Elkhorn River can act as a source of steroidogenic compounds to aquatic organisms, and 3) to determine if site-specific biological effects are apparent in the hepatic gene expression of fathead minnows. Evidence was obtained using three approaches: 1) deployment of polar organic chemical integrative samplers (POCIS), 2) deployment of caged fathead minnows, and 3) a laboratory experiment in which POCIS and fish were exposed to sediments from the deployment sites. Deployment sites included: the Elkhorn River immediately downstream from a Nebraska wastewater treatment plant, two waterways (Fisher Creek and Sand Creek) likely to be impacted by runoff from cattle feeding operations, and a reference site unlikely to be impacted by waste water inputs. The POCIS extracts were analyzed for a number of natural steroids and metabolites, as well as four different synthetic steroids: ethinylestradiol, zearalonol, 17β-trenbolone and melengestrol acetate. Estrogenic and androgenic metabolites, as well as progesterone and trace levels of melengestrol acetate were detected in POCIS deployed at each site. POCIS deployed in tanks containing field sediments from the four sites did not accumulate the synthetic steroids except for ethinylestradiol, which was detected in the aquarium containing sediments collected near the wastewater treatment plant. Fish deployed in Sand Creek and at the wastewater treatment plant experienced significantly elevated levels of gene expression for two genes (StAR and P450scc) relative to those deployed in Fisher Creek. Fish exposed to the sediments collected from Sand Creek had significantly higher levels of hepatic StAR and P450scc gene expression than did fish exposed to sediments from the two other field sites, as well as the no-sediment control tank. In conclusion: 1) detectable levels of steroidogenic compounds were detected in passive samplers deployed in the Elkhorn River, 2) sediments do not appear to be a significant source for steroidogenic compounds, and 3) site-specific differences were found in mRNA expression among the different treatment groups of fish; however, a functional explanation for these differences is not readily forthcoming.