Produced water extracts from North Sea oil production platforms result in cellular oxidative stress in a rainbow trout in vitro bioassay E. Farmen a,b, * , C. Harman a , K. Hylland a,b , K.-E. Tollefsen a a Norwegian Institute for Water Research, Gaustadallèen 21, N-0349 Oslo, Norway b University of Oslo, Department of Biology, P.O. Box 1066, Blindern, N-0316 Oslo, Norway article info Keywords: Rainbow trout Oncorhynchus mykiss Toxicity Oxidative stress Hepatocyte abstract Produced water (PW) discharged from offshore oil industry contains chemicals known to contribute to different mechanisms of toxicity. The present study aimed to investigate oxidative stress and cytotoxicity in rainbow trout primary hepatocytes exposed to the water soluble and particulate organic fraction of PW from 10 different North Sea oil production platforms. The PW fractions caused a concentration-depen- dent increase in reactive oxygen species (ROS) after 1 h exposure, as well as changes in levels of total glu- tathione (tGSH) and cytotoxicity after 96 h. Interestingly, the water soluble organic compounds of PW were major contributors to oxidative stress and cytotoxicity, and effects was not correlated to the content of total oil in PW. Bioassay effects were only observed at high PW concentrations (3-fold concentrated), indicating that bioaccumulation needs to occur to cause similar short term toxic effects in wild fish. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Produced water (PW), the water separated from oil and gas dur- ing oil production, is by far the largest waste effluent from the off- shore oil and gas production industry (Utvik, 1999). Such discharges generally increase with the age of the production well and the total volume from the Norwegian sector is predicted to reach more than 250 million m 3 in 2010 (OD, 2007). Since little is known about the fate and sub-lethal effects of produced water, bio- logical effect assessment and toxicity characterisation is of great importance. Several factors are essential in assessing the environ- mental risk of oil-related compounds to marine organisms, such as composition, biodegradation, bioaccumulation and biological effects. It is known that PW effluents contain potentially toxic compounds such as polycyclic aromatic hydrocarbons (PAH), alkyl- phenols (AP), heavy metals, various production chemicals and bio- cides (Brendehaug et al., 1992). Dilution of PW will however lead to low concentrations of PW components in the receiving waters, and acute toxic concentrations for fish will only be present in the immediate vicinity of the discharge point (Sommerville et al., 1987). The only routine testing currently performed on PW is chem- ical analyses, such as the determination of concentrations of total oil and some selected target compound groups. Suggestions of imple- mentation of biological effects assessment has thus led to the appli- cation of various biological test methods such as assessment of acute toxicity, endocrine disruption, developmental toxicity, geno- toxicity, and arylhydrocarbon receptor (AhR)-mediated toxicity (Brendehaug et al., 1992; Stephens et al., 2000; Meier et al., 2007a,b; Hylland et al., 2008). Such in vivo studies, however, often require long term exposures to detect toxic effects and are restric- tive in terms of a limited number of exposure combinations that can be tested simultaneously. In contrast, small scale bioassays such as the use of in vitro cultured primary cells has several advantages in toxicity screening such as cost- and time efficiency, small sample requirement and possibilities for high-throughput screening (Cast- ano et al., 1996). In combination with different sample extraction methods such as solid-phase extraction (SPE) and organic extraction in general, environmental contaminants can be concentrated to lev- els that facilitate chemical and biological testing. Consequently, this approach has been used for screening single chemicals, mixtures and complex environmental samples, by utilizing classical biomark- ers and effect endpoints (Gagne and Blaise 1997; Schirmer et al., 1997; Tollefsen et al., 2007; Tollefsen et al., 2006). Oxidative stress is caused by excessive amounts of reactive oxy- gen species (ROS) such as hydroxyl radicals, hydrogen peroxides and superoxides. These radicals are produced in cells following exposure to a range of different xenobiotics (Livingstone 2001). The highly reactive properties of ROS molecules make them a potential threat to normal cellular function since dysfunctional enzymes, lipid peroxides and DNA damage, can in different ways result in malfunctioning, necrotic or apoptotic cells (Halliwell and Gutteridge, 1999). ROS have furthermore been linked to several dis- eases, including cancer and neurodegenerative disorders (Frenkel, 1992; Halliwell, 1994; Markesbery 1997; Weisburger, 2001). To 0025-326X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2010.01.015 * Corresponding author. Address: Norwegian Institute for Water Research, Gaustadallèen 21, N-0349 Oslo, Norway. E-mail address: eff@niva.no (E. Farmen). Marine Pollution Bulletin 60 (2010) 1092–1098 Contents lists available at ScienceDirect Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul