Biomarker candidate discovery in Atlantic cod (Gadus morhua) continuously exposed to North Sea produced water from egg to fry

Aquatic Toxicology 96 (2010) 280–289

Contents lists available at ScienceDirect

Aquatic Toxicology journal homepage: www.elsevier.com/locate/aquatox

Biomarker candidate discovery in Atlantic cod (Gadus morhua) continuously exposed to North Sea produced water from egg to fry Anneli Bohne-Kjersem a,∗ , Nicolai Bache b , Sonnich Meier c , Gunnar Nyhammer d , Peter Roepstorff b , Øystein Sæle d,e , Anders Goksøyr a,f , Bjørn Einar Grøsvik a,c a

Department of Molecular Biology, University of Bergen, PB 7800, N-5020 Bergen, Norway Department of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark Institute of Marine Research, PB 1870, N-5817 Bergen, Norway d Department of Biology, University of Bergen, PB 7800, N-5020 Bergen, Norway e National Institute of Nutrition and Seafood Research, PB 2029, N-5817 Bergen, Norway f Biosense Laboratories AS, Thormøhlensgt. 55, N-5020 Bergen, Norway b c

a r t i c l e

i n f o

Article history: Received 30 July 2009 Received in revised form 3 November 2009 Accepted 8 November 2009 Keywords: Biomarker candidates Egg Fry Larvae Proteomics Oil Produced water

a b s t r a c t In this study Atlantic cod (Gadus morhua) were exposed to different levels of North Sea produced water (PW) and 17␤-oestradiol (E2 ), a natural oestrogen, from egg to fry stage (90 days). By comparing changes in protein expression following E2 exposure to changes induced by PW treatment, we were able to compare the induced changes by PW to the mode of action of oestrogens. Changes in the proteome in response to exposure in whole cod fry (approximately 80 days post-hatching, dph) were detected by two-dimensional gel electrophoresis and image analysis and identified by MALDI-ToF-ToF mass spectrometry, using a newly developed cod EST database and the NCBI database. Many of the protein changes occurred at low levels (0.01% and 0.1% PW) of exposure, indicating putative biological responses at lower levels than previously detected. Using discriminant analysis, we identified a set of protein changes that may be useful as biomarker candidates of produced water (PW) and oestradiol exposure in Atlantic cod fry. The biomarker candidates discovered in this study may, following validation, prove effective as diagnostic tools in monitoring exposure and effects of discharges from the petroleum industry offshore, aiding future environmental risk analysis and risk management. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Produced water is considered to be by far the most dominant discharge from offshore oil production, containing a variety of compounds, including oil products such as mono-, di- and polyaromatic hydrocarbons (PAHs), alkyl phenols, and several heavy metals, which have previously been shown to cause adverse effects in fish (Neff, 2002). Some of these compounds are reported to be endocrine disrupting chemicals (Evanson and Van Der Kraak, 2001; Santodonato, 1997), that affect steroidogenesis (Evanson and Van Der Kraak, 2001; MacLatchy and Van Der Kraak, 1995), reproductive organs and reproduction (Meier et al., 2007a,b), in fish eggs, larvae, and fry (Hahn, 2001; Rolland, 2000). PAHs and metals present in produced water are also reported to cause oxidative stress (Hasselberg et al., 2004; Livingstone, 2001), possibly resulting in mutagenesis (Machala et al., 2001) and carcinogenesis (Santodonato, 1997).

∗ Corresponding author. Present address: Det Norske Veritas, Johan Berentsensvei 109-111, N-5020 Bergen, Norway. Tel.: +47 90 09 44 57; fax: +47 55 58 96 83. E-mail address: [email protected] (A. Bohne-Kjersem). 0166-445X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.aquatox.2009.11.005

Biological effects of exposure occurring at the early stages of development are of special interest, as these stages are found to be more sensitive to xenobiotic exposure (Bern, 1992; Colborn et al., 1993). In many teleost species, this period of enhanced vulnerability is from fertilization through the yolk sac embryo stage, which is prior to the larvae stage. However, studies have shown that food and water-borne exposure contributes significantly also to fry mortality (Rolland, 2000). There has been an unprecedented decline in commercial marine fish species world-wide, resulting in a growing concern for the future viability of fishery resources and prompting a search for causes explaining this decline (Jackson et al., 2002). In many coastal ecosystems cod fish and other marine vertebrates are functionally or entirely extinct. Overfishing is believed to be the main reason for this (Worm et al., 2006), although pollution is considered an important contributing factor to reported declines (Lotze and Milewski, 2004). Atlantic cod was selected for this study as it is one of the key fish species, together with herring, for risk assessment studies in connection with oil and gas activities in Norwegian waters. Development of improved and more sensitive biomarkers may contribute with new knowledge that is important to risk management, as drilling and exploration of wells in new areas is progressing.

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

At present, it is difficult to detect and monitor effects of exposure before they become a threat to health and reproduction and causing effects at the population level. Many field studies report late occurring adverse effects in various teleost species, often not visible in the exposed fish, but rather in its progeny (see reviews by Rolland, 2000; Schwaiger et al., 2002). Also several reports suggest that adverse effects are induced at lower levels of exposure than known biomarkers can detect (Denslow et al., 2001; Reid and MacFarlane, 2003). Hence, there is a need for more sensitive biomarkers that can act as early warning signals to detect and monitor the effects of oil and produced water on marine fish. In this study, we have exposed Atlantic cod (Gadus morhua) at early life stages to different concentrations of produced water (Oseberg C, the North Sea) and 17␤-oestradiol (E2 ), a natural oestrogen, continuously for 90 days from egg to fry stage. Our aim was to investigate produced water-induced changes in cod fry, and, by comparing them to the responses induced by E2 , to see if any of these responses represent a similar mode of action (MOA). 2. Materials and methods Acetic acid, Coomassie brilliant blue CBB G250, orthophosphoric acid, urea, thiourea, tris (hydroxymethyl)-ammonium methane, acetonitril, methanol and trifluoroacetic acid were all purchased from Merck (Damstadt, Germany). Ethanol was obtained from Arcus Kjemi (Oslo, Norway), and bovine serum albumin, BSA, 3-(3-cholamidopropyl)-dimetylammonio-1-propansulfonate, CHAPS, Triton X-100, dl-dithiothreitol, DTT, iodoacetamide, and alpha-cyano-4-hydroxycinnamic acid, CHCA, were purchased from Sigma–Aldrich (St. Louis, MO, US). AmpholineTM 3.5–10, DryStrip Cover fluid, 18 cm IPG strips pH 4.5–5.5 and IEF Electrode strips were purchased from GE Healthcare (Uppsala, Sweden). Ammonium dodecyl sulphate, SDS, agarose, 30% acrylamide/bis solution, 37.5:1, ammonium persulphate, TEMED, and Precision Plus ProteinTM standard were purchased from Bio-Rad (Hercules, CA, US). Trypsin was obtained from Promega (Madison, WI, US), while Poros 20R2, Reverse Phase packing, was purchased from Applied Biosystems (Foster City, CA, US). Siliconised tubes were provided from Sorenson, BioScience, Inc. (Salt Lake City, UT, US). Peptide Calibration Standard and MTP384 Target plate polished steel TF were purchased from Bruker Daltonics (Leipzig, Germany). 2.1. Experimental design The experimental set up was carried out in accordance with “The Code of Ethics of the World Medical Association (Declaration of Helsinki) for animal experiments”, and approved by The National Animal Research Authority in Norway. The experimental design was made to answer whether environmentally realistic doses of produced water (PW) gave effects on the developing cod egg, larvae and fry at chronic exposures. For this reason PW at 0.01%, 0.1% and 1% were selected as exposure concentrations. Modelling studies have calculated that PW may be diluted approximately 1:30 at 10 m, 1:100 at 100 m, and 1:1000, 1 km from the outlet pipe (Neff, 2002). Modelling calculations and in-field measurements in the North Sea have shown concentrations of dispersed oil around the largest oilfields to be approximately 1–3 ppb, which roughly corresponds to a dilution factor of 1:10,000 (Rye et al., 1998). Parallel treatments with 10 ␮g/l 17␤-oestradiol were included in order to compare responses after PW with oestrogenic responses. All treatments were performed in triplicates, as well as the control (clean sea water). The exposure was carried out for 90 days, starting March 25, 2004, on fertilized cod eggs and ended at early fry stages June 22, 2004. Light and temperature regime were held as natural conditions in the Bergen area, Norway at this time of year. Cod larvae

281

were fed natural zooplankton after end of the yolk sac stage until the end of the exposure period. The eggs used in this study were obtained from wild cod caught in Tysfjorden in Lofoten, Norway. For spawning, one male and one female fish were placed in a spawning tank and the resultant eggs collected from a filter placed over the runoff outlet. To ensure a realistic level of biological variation in this study, we mixed eggs collected from five separate pairs of spawning cod. The average egg diameter (D) from each spawning pair was measured, and the number of eggs calculated according to the following formula: N (number of eggs per ml) = 1222 × D−2.71 (Kjesbu, 1989). Using this formula, 60,000 eggs (12,000 from each of the five pairs) were added to 100 l of water in each exposure tank. Three identical 100 l tanks, each containing 60,000 fertilized cod eggs (1–2 days old), were subjected to one of the five different treatment regimes described above for 23 days (in March/April 2004), until day 3 posthatch (dph). The larvae were counted and 6500 from each tank were transferred to a fresh 100 l tank and the treatment continued for 67 days (473-day-degrees), through the whole start-feeding phase. At the end of this 90-day period (June 22, 2004) the fish were approximately 2 cm in length and 10 mg in weight (dry weight). The dilution factor in the tanks was controlled by gas chromatography/mass spectrometry analysis of alkyl phenols as described in Boitsov et al. (2004). The average result of two measurements of C2 alkyl phenols taken throughout the experiments showed the dilution factors to range from 30% to 110% of the nominal doses. 2.2. Sample preparation Whole individual cod fry (approximately 80 days post-hatching, dph) were homogenised in 6× (v/w) re-hydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 20 mM DTT, 0.5% (v/v) Triton X-100, 0.5% (v/v) Ampholine 3–10, bromphenol blue) using a Potter Elvehjem homogeniser with a Teflon® pestle. The pestle was run up and down 6 times during homogenisation. The homogenised sample was subsequently centrifuged at 13,000 × g at room temperature for 20 min. The protein concentration was determined using a plate reader-modified Bradford’s assay (Bradford, 1976). 2.3. Two-dimensional gel electrophoresis (2DE) 500 ␮g of sample was diluted in re-hydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 20 mM DTT, 0.5% (v/v) Triton X-100, 0.5% (v/v) Ampholine 3-10, bromphenol blue) and added to IPG strips pH 4.5–5.5 (covering the majority of proteins in the sample, results not shown), 18 cm (Görg et al., 2000; O’Farrell, 1975). Four individuals from each treatment-group were analyzed by 2DE. Due to a limited amount of sample, less protein (135–250 ␮g) from samples exposed to 0.1% produced water was diluted in buffer and added to the IPG strips. Only three individuals from this group were analyzed by 2DE. The strips were re-hydrated for a minimum of 12 h and focused on a Multhiphor II unit (GE Healthcare) according to the manufacturer’s guidelines. The strips were equilibrated 15 min at room temperature in 0.25% DTT-containing SDS equilibration buffer (50 mM Tris pH 8.8, 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, bromphenol blue) and 15 min in a 4.5% iodoacetamide-containing SDS equilibration buffer (Görg et al., 2000; O’Farrell, 1975) prior to separation in the second dimension on 9% SDS-PAGE gels (Laemmli, 1970). Gels were run in an Ettan Dalt Twelve unit (GE Healthcare) at 1 W/gel, 20 ◦ C, for approximately 17 h. The gels were then stained with colloidal Coomassie (Neuhoff et al., 1988). Coomassie-stained gels were scanned on GS-800 Calibrated Densitometer flatbed scanner (Bio-Rad) using PDQuest 7.2.0 software (Bio-Rad). The gels were scanned with high resolution: 127.0 ␮m × 127.0 ␮m.

282

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

Global protein expression was analyzed using Delta 2D image analysis software (version 3.4) from Decodon (Greifswald, Germany): TIFF images of the 2DE gel images were up-loaded in Delta 2D and grouped according to treatment. The images were warped using a group-warping strategy and ‘exact’ warping. A union fusiongel was produced from all the images within the control, high level exposure (1.0% PW) group and the oestradiol-treated group. No filters were used. Spots were detected on the fusion-gel and validated spots were transferred to all the other gel images. Normalisation was performed, based on total intensity including all spots on the gel. The spot data from the all images quantitation table was exported to a cvs-file in Excel, which was successively uploaded in the Ludesi Interpreter (www.ludesi.com, Lund, Sweden), together with the gel TIFF images. Gel comparison, and ANOVA (p ≤ 0.05 and fold change ≥2) were performed using Ludesi Interpreter. Any changes detected in difficult gel areas or any changes resulting from bad spot detection, were excluded to limit the number of false positives.

2.5. Database searches The cod EST database used for protein identification included more than 50,000 EST sequences from cod from a project at the Institute of Marine Research, Norway, NIFES (National Institute of Nutrition and Seafood Research) and groups affiliated to the National Research Council, Canada. The annotation of the sequences was performed by the Computational Biology Unit at the University of Bergen, producing approximately 6000 contigs and more than 30,000 singlets. The EST sequences are derived from cDNA libraries of cod exposed to an array of environmental pollutants, such as PCBs and heavy metals, and purified mRNA sequences from several different tissues. RNA was also purified from cod captured in the harbours of Bergen (Norway) and Trondheim (Norway), and therefore assumed to be exposed to a battery of pollutants, as well as cod exposed to hypoxia and hyperoxia, http://www.codstress.olsvik.info/Results.html (Olsvik et al., 2005). 2.6. Histology

2.4. Mass spectrometry In-gel digestion of protein spots was performed as described by Shevchenko et al. (1996). Briefly, excised gel plugs were washed in digestion buffer (50 mM NH4 HCO3 , pH 7.8/acetonitrile: 60/40) and dried by vacuum centrifugation. Modified trypsin (10 ng/␮l) dissolved in 50 mM NH4 HCO3 , pH 7.8, was added to the dry gel pieces and incubated on ice for 1 h. After removing the supernatant, additional digestion buffer was added and the digestion was continued at 37 ◦ C over night. The supernatant of the digestion was analyzed directly by off-line LC-MALDI without peptide extraction. Peptide separation was performed using an inert nanoflowHPLC system from LC-Packings (Ultimate; Switchos2; Famos; LC Packings, Amsterdam, The Netherlands) equipped with a Probot MALDI spotting device (LC Packings, Amsterdam, The Netherlands). 10 ␮l of the in-gel digested protein(s) was acidified with 0.5 ␮l 10% TFA and loaded onto a home-made 1-cm fused-silica pre-column (100 ␮m i.d., ReproSil-Pur C18 AQ, 3-mm; Dr. Masch GmbH), and then eluted onto a home-made 10-cm fused-silica analytical column (75 ␮m i.d., ReproSil-Pur C18 AQ, 3-mm; Dr. Masch GmbH), using a flow of 200 nl/min, and a gradient from 100% solvent A (H2 O/TFA, 100:0.1, v/v) to 50% solvent B (H2 O/acetonitrile/TFA, 20:80:0.1, v/v/v) over 40 min, then 50–100% solvent B in 5 min. Column effluent was mixed in a 1:9 ratio with MALDI matrix (2 g/l acyano-4-hydroxycinnamic acid in H2 O/acetonitrile/TFA, 30:70:0.1, v/v/v) through a 25 nl mixing tee (fused-silica capillary TSP050375; Composite Metal Services Ltd., UK) and spotted onto a MALDI plate in 30 s fractions. The LC-MALDI preparation was analyzed on an ABI 4700 (Applied Biosystems, Framingham, MA, USA) proteomics analyzer using an automated workflow. A maximum of 10 peptides per spot with a s/n > 80 was automatically selected for MS/MS. CID was performed at collision energy of 1 kV with an indicated collision gas pressure of ∼1 × 10−6 Torr. The spectra were annotated and analyzed using Data ExploreTM v. 4.5 (Applied Biosystems) The MS2 data from the entire off-line LC-MALDI experiment were combined into a single mass list using an in-house developed script (Jakob Bunkenborg, SDU, DK). The sequence data from the MS–MS runs (mascot-generic format, mgf files) were used in database searches in the cod expressed sequence tag (EST) database and/or the NCBInr database for protein identity using the MASCOT search engine (Perkins et al., 1999; Steen and Mann, 2004). The following parameters were used for the database search: Missed cleavage 1: Significance threshold p < 0.05; MS1 mass accuracy 100 ppm, MS2 mass accuracy; 0.25 Da; partial modifications: carbamidomethyl (C) and oxidation (M).

Whole fry samples were fixed in 4% buffered paraformaldehyde and embedded in paraffin. Four micrometers longitudinal sections were mounted on untreated object lamellas. Mallory phosphotungstic acid-hematoxylin (PTAH) staining for cross-striations in muscle was done according to Carson (1990). Since tissue was fixed in formalin, sections were post-fixed in Zenker fixative containing 5% acetic acid, post-deparaffinization. To speed up the ripening process of the PTAH stain potassium permanganate was added. This reduces the natural ripening time of 4–6 months to 2 weeks. After staining, slides were dehydrated in graded ethanol and cleared in xylene before mounted in DPX (Fluka). Distance from the start point of one A-band to the next A-band was measured with a computer aided stereological tool (CAST 2, Olympus Denmark) that is calibrated with the microscope (Zeiss). Measurements were done with 1000× magnification. 2.7. Statistical analysis Hierarchical clustering (Ward), 3D scatterplots, discriminant analysis, and principal component analysis were performed in JMP v 7.0.1 for Mac (SAS Institute Inc.) on spot volume data normalized to spot mean, subsequently log 2-transformed and normalized to larvae weight. 3. Results Exposure of cod to 1.0% produced water (PW) during egg and larvae stage resulted in a significant increase in mortality compared to control. At sampling, the three tanks exposed to 1% PW contained 0, 12 and 15 individuals from the original 6500 larvae per tank giving survival rates of 0.14 ± 0.12%, after 67 days of exposure. Survival of larvae in the three control chambers were 8.7 ± 1.9%. Survival in the treatments with 10 ␮g/l E2 , 0.01% PW, 0.1% PW were from 5.5% to 7%, but not significantly different from the control groups. The survival rate of 8.7% in the control tanks may be compared with the high natural mortality rate for early stages of cod as reported by Sundby et al. (1989) and Kristiansen et al. (1997). Mortality rate from hatching until early juvenile stage (2–3 months age) in the Lofoten area was estimated to be between 7% and 8% per day in 1983–1985 based on abundance of eggs in the spawning area and juvenile abundance in the nursery areas further north (Sundby et al., 1989). Morphological changes were observed among several of the 1.0% PW-treated fishes, although not statistically significant compared to the control group (Meier et al., submitted). Analysis of histological sections showed that there were significant differences

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

283

Table 1 Distance between A-bands in muscle fibres from cod larvae exposed to produced water and 17␤-oestradiol. Group

Day 66

Control 0.1% PW 1.0% PW E2

1.71 1.77 1.67 1.83

± ± ± ±

Day 80 0.01ab 0.01ac 0.04b 0.06c

1.76 ± 0.03a 1.77 ± 0.01a –* 1.78 ± 0.01a

Different letters in superscripts indicate significant differences between groups (ANOVA, p < 0.05). * No fry available for analysis.

Fig. 1. Mallory PATH stained muscle sections from 17␤-oestradiol-treated fish (A) and control (B) sampled 80 days post-hatching (dph). The scale bar is 10 ␮m. The dark stained A-bands and light stained I-bands are indicated. The Z line is found in the I-band (hard to see in figure) and the A-band is bisected by a lightly staining H zone.

in muscle fibres between E2 -exposed and control fish sampled at day 66 post-hatching (Fig. 1), whereas there were no structural differences found in the fry sampled at 80 dph (Table 1). There was also a significant difference in the cod fry weight between the 1.0% PWtreated group compared to the 0.1% and E2 -treated groups in the cod fry sampled for proteomics analysis. Individuals in these two groups were of significantly lower weight compared to the 1.0% PW group (not shown). The proteomic analysis included 4 individuals within each treatment-group and 3 individuals from the group treated with 0.1% (medium level) PW. Exposure of cod to produced water and

E2 from egg to fry stage (90 days of exposure) resulted in a number of protein changes observable on 2D gels (Fig. 2). Treatment with 10 ␮g/l E2 induced more changes in the 2DE pattern compared to the changes induced by exposure to produced water (Fig. 2). The 2DE and image analysis of the global protein expression of whole cod fry exposed to produced water and E2 showed a significant change in protein expression (ANOVA: p < 0.05; fold change ≥2) of 84 proteins (Table 2). Of the 84 protein changes, 27 were significantly changed after exposure to 0.01% PW, another 24 have changed levels after exposure to 0.1% PW, and 8 additional proteins have changed levels after exposure to 1.0% PW, whereas 61 proteins were shown to have changed levels after exposure to 10 ␮g/l E2 . Differentially expressed proteins were isolated from a control gel and from a gel of an E2 -treated sample (Fig. 3), and trypsindigested prior to MS/MS analysis. The MS/MS data were used for database searches for protein identification, where 13 of totally 24 protein spots were successfully identified (Table 3). Several of the proteins in the trains of spots in the upper part of the gel were identified as skeletal muscle myosin heavy chains (MHC), as were several of the differentially expressed protein spots. However, the identified MHCs show some conflicting results regarding protein expression: Those in the upper trains of spots appeared to be up-regulated after exposure to produced water and downregulated following E2 exposure, whereas those proteins identified as myosin with a lower molecular weight appeared to be downregulated by PW treatment and up-regulated after exposure 10 ␮g/l

Fig. 2. Comparison of protein changes induced by produced water treatment and 17␤-oestradiol exposure of cod fry. Left: protein changes induced by 1.0% PW (orange) compared to control (blue) detected by 2DE and image analysis in whole fry individuals. Right: protein changes induced by 10 ␮g/l E2 (orange) compared to control (blue) detected by 2DE and image analysis in whole fry individuals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

284

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

Table 2 Overview of differentially expressed proteins in cod fry exposed to produced water (PW) and 17␤-oestradiol (E2) found by Delta 2D image analysis. Spot

ASV

9 11 17 61 75 90 88 92 94 107 109 113 118 123 139 145 234 238 239 241 243 244 247 254 256 257 261 262 265 276 282 283 306 308 309 325 326 327 329 331 331 334 335 337 339 344 348 349 352 356 355 357 357 361 366 371 372 374 376 384 386 401 402 404 408 409 410 415 417 418 418 419 422

3988 8945 773 556 1940 574 1620 683 3921 1598 1105 2216 3527 6259 30138 1199 1112 626 854 922 691 674 501 1123 662 1055 551 673 600 694 546 652 935 554 1334 1246 818 948 665 1168 1029 565 504 866 1249 811 2502 1329 1479 1806 1493 816 1370 1069 785 1322 1468 2464 509 2148 4243 886 785 1025 8095 13843 28071 1741 3050 2071 992 676 7715

CV

0.01% PW

0.975 0.673 1.186 0.954 1.053 0.712 0.806 0.844 0.523 0.774 0.513 0.704 0.973 0.658 0.465 0.036 0.659 0.852 1.018 0.914 0.813 0.87 0.952 0.777 0.775 0.651 0.74 0.67 0.832 0.811 0.723 0.0242 0.843 0.907 0.988 0.917 0.758 0.858 0.769 0.963 0.99 0.764 0.886 1.051 0.759 0.69 0.469 0.675 0.935 0.693 0.671 0.696 0.685 0.589 1.146 0.804 0.553 0.592 0.617 0.584 0.43 0.62 0.758 0.647 0.885 0.793 0.491 0.879 0.734 0.677 1.026 1.153 0.755

2.1 1.2 0.02 0.2 0.4 0.6 0.6 0.2 0.8 1.4 1.1 0.5 0.2 2.3 0.5 0.4 0.7 0.1 0.05 0.4 0.5 0.9 0.2 0.4 0.9 0.8 0.7 0.5 0.9 0.3 0.7 3.0 1.5 1 0.9 1.6 1 2.3 0.9 0.4 1.1 0.6 0.8 1.2 1.3 2.3 0.9 1.1 0.5 6.3 1 0.7 1.6 1.4 2.0 0.4 1.2 1 1.6 4.3 0.6 0.9 2.2 2.0 3.0 1.5 1 0.9 1.6 1 2.3 0.9 0.4

0.1% PW 1 0.5 0.2 0.4 0.8 0.2 0.4 0.3 0.5 1.8 0.2 0.2 0.2 1 0.5 0.4 2.1 0.8 0.5 0.2 0.4 0.8 0.4 0.3 0.9 0.9 0.4 0.7 0.7 0.8 2.4 1.4 0.7 1 1.3 1.9 1 0.7 1.1 0.3 0.4 1.2 2.3 1.9 2.3 0.7 0.5 1.1 0.4 2.2 0.8 1.3 2.2 1.2 1.7 0.7 1 1.3 6.1 16.2 0.3 1.3 1.4 0.9 1.4 0.7 1 1.3 1.9 1 0.7 1.1 0.3

1.0% PW

E2 (10 ␮g/l)

ID

2.7 1.5 3.5 1.6 3.2 0.6 1.3 0.4 0.7 0.9 0.7 0.6 1.1 2.1 0.8 0.4 0.6 1.3 1.0 0.3 0.4 0.4 0.6 0.4 0.5 0.5 0.4 0.4 1 2.2 0.6 0.5 0.6 0.4 1.2 0.7 0.5 0.8 0.9 0.8 0.5 0.7 0.4 0.9 0.7 0.5 1 0.9 0.6 0.4 0.4 2.6 0.4 0.9 0.3 1 0.8 0.7 0.9 1.2 0.8 0.7 0.8 0.4 0.5 0.6 0.4 1.2 0.7 0.5 0.8 0.9 0.8

0.08 0.2 1 0.4 0.8 0.2 0.4 0.2 0.4 3.6 0.5 0.4 0.07 0.8 0.4 0.08 2.0 2.4 2.9 1.7 2.0 3.3 1.6 1.8 3.0 2.2 1.8 1.8 3.0 0.8 1.6 2.1 2.3 1.8 6.6 4.6 2 1.3 2.4 0.6 0.9 2.3 2.4 2.7 4.7 2.9 0.3 4.3 1.9 4.4 1.9 6.5 2.9 2.7 3.7 3,0 4.9 3.6 3.7 6.4 0.9 2.0 5.4 1 2.1 2.3 1.8 6.6 4.6 2 1.3 2.4 0.6

MHC

ATP synthase

Krt4 protein

MHC MHC Keratin k8b MHC

Alpha-actinin

MHC

Hsc71 Skeletal alpha-actin

Muscle alpha-actin Muscle alpha-actin

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

285

Table 2 (Continued ) Spot

ASV

423 423 440 441 442 452 455 456 459 460 462

12857 13570 1566 2966 1140 3283 1063 13266 1667 3522 3383

CV

0.01% PW

0.1% PW

1.0% PW

0.614 0.892 0.84 1.055 0.866 0.982 1.164 0.382 0.496 1.017 0.633

1.1 0.6 0.8 1.2 1.3 2.3 0.9 1.1 0.5 6.3 1

0.4 1.2 2.3 1.9 2.3 0.7 0.5 1.1 0.4 2.2 0.8

0.5 0.7 0.4 0.9 0.7 0.5 1 0.9 0.6 0.4 0.4

E2 (10 ␮g/l) 0.9 2.3 2.4 2.7 4.7 2.9 0.3 4.3 1.9 4.4 1.9

ID

Muscle alpha-actin

Average spot volumes (ASV) are listed to show spot intensities. Fold up or down-regulation after the different treatments are listed together with coefficient of variation (CV). Significantly up- or down-regulated spots are shown in bold (ANOVA: p < 0.05, fold change ≥2). Number of individuals was 4 per group, except for group 0.1% PW where n was 3. Protein identifications (ID) are shown when obtained.

E2 . Hence, these protein changes appear to reflect different protein modifications of myosin depending on exposure. Several of the other differentially expressed proteins were identified as fast skeletal muscle alpha-actin and appeared to be up-regulated by PW exposure and down-regulated by E2 treatment. Hsc71, an Hsp70, appeared to be down-regulated by 0.1% PW. Alpha-actinin was also found to be down-regulated by produced water treatment (both 0.1% and 1.0%). Exposure to 0.01% and 0.1% PW caused a downregulation of keratin. E2 treatment appeared to down-regulate ATP synthase, which is important in cellular energy production. Two-way hierarchical clustering analysis, performed on weightnormalized data to exclude the influence of weight differences on the parameters, illustrates that the E2 group clusters separate from all produced water groups, and that the high dose (1.0%) produced group is difficult to separate from the controls (Fig. 4A). A similar pattern appears from the principal component analysis (Fig. 4B), which also clearly separates the low (0.01%) and medium (0.1%) PW exposures from the control. By applying stepwise discriminant analysis as an approach to select protein variables that can predict the group classification of the individual fry samples, we were able

to select a set of three protein spots that separated the groups sufficiently to avoid any misclassifications (Fig. 5A). These were spots 9, 17 and 109. Of these, only spot 9 had been identified by MS/MS, as myosin heavy chain protein (Table 2). However, using the whole set of 14 identified protein spots, a correct classification of all samples were also obtained (Fig. 5B), indicating that all the identified proteins can be considered useful biomarker candidates when used in a combined multi-biomarker approach. 4. Discussion In this study, we have detected changes in protein expression of cod fry exposed to produced water and 17␤-oestradiol by 2DE and image analysis. Changes in protein expression patterns were also confirmed by hierarchical clustering and principal component analysis. Of the proteins shown to be differentially expressed (ANOVA: p < 0.05 and fold change ≥2), most of the changes occurred following exposure to low level, 0.01%, and medium level, 0.1%, of produced water (27 and 24 spots out of 84, respectively), corresponding to a distance between 100 and 1000 m from the point of

Table 3 Differentially expressed proteins from cod fry exposed to produced water and 17␤-oestradiol identified by MS/MS. Spot no. 9 94

Accession number

Protein Identity

Queries matcheda

Database

Scoreb

Seq. cov. %

Theoretical pI/Mw (kDa)

XP 708916.1

Myosin, heavy polypeptide 1, skeletal muscle (MHC) [Danio rerio] ATP synthase subunit beta [Cyprinus carpio] Krt4 protein [Danio rerio] Myosin, heavy polypeptide 1, skeletal muscle (MHC) [Danio rerio] Myosin, heavy polypeptide 1, skeletal muscle (MHC)[Danio rerio] Simple type II keratin k8b (S2) [Oncorhynchus mykiss] Myosin, heavy polypeptide 1, skeletal muscle (MHC) [Danio rerio] Alpha-actinin [Danio rerio]c Myosin heavy chain (MHC) [Cyprinus carpio] Heat shock cognate 71 kDa protein (Hsc71) [Ictalarus punctatus] skeletal alpha-actin [Carassius auratus] Fast skeletal muscle alpha-actin [Gadus morhua] Fast skeletal muscle alpha-actin [Gadus morhua]

29 (177)

NCBI

1586/31

13

5.66/287609.6

11 (67)

Cod EST

421/27

38

5.05/55247.3

4 (150) 2 (79)

Cod EST NCBI

58/27 94/31

10 2

5.34/54027.2 5.66/287609.6

2 (45)

NCBI

63/32

1

5.66/287609.6

3 (373)

Cod EST

90/27

4

5.14/58976.5

3 (45)

NCBI

51/31

2

5.66/287609.6

8 (53) 1 (44)

NCBI NCBI

179/30 39/31

11 1

5.17/103860.2 5.14/110155.2

6 (49)

NCBI

364/31

12

5.19/71340.4

3 (46) 4 (47)

NCBI Cod EST

92/31 74/28

11 16

5.23/42141.1 5.23/41974.9

4 (47)

Cod EST

135/28

17

5.23/41974.9

Q9PTY0

118 243

AAH66728 XP 708916.1

244d

XP 708916.1

247

CAA63300.1

254d d

XP 708916.1

344 366d

AAN77132 BAA09069.1

386d

P47773

402d 418d

P49055 AAM21702

441d

AAM21702

The following search criteria were used: modification: carbamidomethyl (C), oxidation (M); peptide tolerance: 100 ppm; MS/MS tolerance: 0.25 Da; enzyme: trypsin; monoisotopic. In NCBI-searches taxonomy was specified: Actinopterygii (ray-finned fish). p < 0.05. a The number of query fragments that match/the total number of query fragments. b The score of query’s matched in the database/the treshold score of homology (identity). c The protein was identified as an unnamed protein and was further characterised by homology in blast. d No preanalytical LC-run.

286

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

Fig. 3. Localisation of differentially expressed proteins on the 2DE gels. The significant protein changes found by image analysis of PW- and oestradiol-treated whole cod fry (ANOVA: p < 0.05, fold change ≥2) are indicated on representative 2DE gels by their spot numbers. Abbreviations used: PW; produced water-treated sample, E2 ; oestradiol-treated sample.

discharge from an oil platform (calculations based on Neff, 2002; Rye et al., 2000). These results indicate an effect on the cod fry proteome at low levels of produced water, and appreciable distances from the point of discharge. A significant increase in mortality was observed among fry exposed to high levels of produced water (1% PW), however, only minor responses were observed in the pro-

teome of the surviving individuals in this group. The high mortality rate after exposure to 1% PW could mean that only the individuals that had low susceptibility to produced water were left to be analyzed from this group. This could influence the results and lead to responses in this group more similar to the control group. Although several of the proteins affected by low levels of produced water exposure (0.01% PW and 0.1% PW) also were affected by E2 treatment, the responses were in many cases opposite. Multivariate analyses (hierarchical clustering and principal component analysis) clearly separated E2 -treated individuals from the rest of the samples, indicating other effects from low dose of PW than those obtained after E2 . In a continuation of this study, early juveniles exposed to 1% PW had a significant induction of vitellogenin, while no significant changes were observed after 0.1% and 0.01% PW treatment (Meier et al., submitted). Myosin and alpha-actin were among the identified proteins that were differentially expressed following exposure to produced water and oestradiol. These are proteins known to be important for myogenesis (muscle development) (Steinbacher et al., 2006; Ono et al., 2006; Thiebaud et al., 2001; Venkatesh et al., 1996). Produced water treatment induced a change in the expression of myosin, which was different to that induced by the oestradiol treatment. In contrast, alpha-actin showed a similar response in protein expression after both produced water (0.01% and 0.1%), and oestradiol treatment, however high levels of PW resulted in a down-regulation of alpha-actin. At the larvae stage increased activity increases the requirements for respiration, and locomotor activity is closely linked to both muscular and gill development (Falk-Petersen, 2005). Alpha-actin and myosin are also linked to ovary contraction and retinal rod structure (Van Nassauw et al., 1991; Chaitin and Burnside, 1989; Poplinskaia, 1995; Williams et al., 1992). During larvae stage the rods seem to be involved in movement perception and may be particularly important in predator avoidance (Blaxter, 1986; O’Connell, 1981). Produced water also resulted in a down-regulation of alphaactinin, another protein linked to retinal structure and perception (Williams et al., 1990). Hence, the changes in the levels of myosin, alpha-actin, and alpha-actinin suggest an impact of produced water on the fast skeletal muscle development, essential for somatic growth and perception, potentially impairing general growth and development of cod fry. These results may also explain the differences in larvae size observed between cod fry exposed to 1.0% PW and 10 ␮g/l oestradiol. However, the larger size of 1.0% PW-treated larvae may be explained by the ample access to food for the few individuals left in the tank. The food supply was kept constant while the high mortality in this group significantly lowered the number of animals consuming the food. Hsc71, a heat shock protein belonging to the Hsp70 family, was found to be down-regulated by produced water. The Hsp70 proteins are known to be able to bind hydrophobic protein domains and function as chaperones that assemble and stabilize multi-protein complexes, translocation of polypeptides across cell membranes, and aiding protein folding (Feder and Hofmann, 1999; Young and Hartl, 2002). Hsp70 is reported to associate with Hsp90, which is of importance to the conformation of the ligand binding domain of the nuclear steroid receptors (and the Ah receptor), as reviewed by (Pratt, 1997). Even though Hsp70 proteins are mostly associated to up-regulation as responses to oxidative stress, a down-regulation of Hsc71, as observed in this study may possibly lead to adverse responses. Numerous studies indicate the usefulness of the Hsp70 response in environmental monitoring, as reviewed by Ryan and Hightower (1996) and Mukhopadhyay et al. (2003). The filament proteins keratin K8b (S2) and cytokeratin 4 (Krt4), were both found to be down-regulated by produced water in this study. The keratins are known to be involved in maintenance of

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

287

Fig. 4. Multivariate analysis of proteome changes in Atlantic cod fry exposed to produced water (PW) and 17␤-oestradiol (E2 ). (A) Two-way hierarchical clustering analysis. Weight- and spot volume-normalized data were analyzed in JMP v 7.0.1 (SAS Institute, Inc.). Individual samples (left axis) are labeled according to group (1–4, control ; 5–8, 0.01% PW ; 9–11, 0.1% PW ; 12–15, 1.0% PW 䊉; and 16–19, 10 ␮g/l E2 , x). Lower axis indicates spot number. (B) Principal component analysis. Weight- and spot volume-normalized data were analyzed in JMP v 7.0.1 (SAS Institute, Inc.). Individual samples are labeled according to group (control ; 0.01% PW ; 0.1% PW ; 1.0% PW 䊉; and 10 ␮g/l E2 , x).

cellular architecture and to provide mechanical resistance against stress (Kirfel et al., 2002). The down-regulation of keratin suggests an effect on tissue integrity. Some studies also report keratins to be involved in apoptosis and signal transduction (Schaffeld and Schultess, 2006). Taken together, the identified responses from produced water exposure seen in this study involve structural, cytoskeletal, and signalling proteins, which are important for muscle development, rod/retina function, cellular signalling and morphology (tissue integrity). These functions are also important for swim performance and predator escape (Falk-Petersen, 2005). Treatment with 10 ␮g/l E2 lowered the protein expression of ATP synthase, which may lower the energy supply of the cell. This may in turn lead to decreased activity of energy-consuming pro-

cesses, such as anabolic processes, neuronal function, mobility, and predator escape. Several proteins were differentially expressed in the present study, which we were not able to identify. Reasons for this may be an incomplete cod EST sequence database, and limited amounts of protein available from isolated spots. Using a stepwise, linear discriminant analysis to select a set of protein variables sufficient to predict the correct classification of each individual fry sample, we were able to find three proteins that performed this task well. Of these three, myosin heavy chain spot 9 was the only identified protein. However, when using a larger set comprising all 14 identified proteins, we were also able to classify all samples to the correct group. As in the hierarchical cluster and principal component analyses, the high dose (1.0%) produced water

288

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289

Fig. 5. Linear, stepwise discriminant analysis of proteome changes in Atlantic cod fry exposed to produced water and 17␤-oestradiol. (A) Canonical plot with minimal set of protein variables necessary to obtain correct classification of all samples using all protein spots. (B) Canonical plot with only identified proteins as predictors (identified proteins, see Table 1). Weight- and spot volume-normalized data were analyzed in JMP v 7.0.1 (SAS Institute, Inc.).

individuals were closer to the controls than to any other group. Although the 14 identified protein spots may work as biomarkers, some of the unidentified protein spots selected by the discriminant analysis appear to have even stronger predictive power, and further attempts will be made to identify these proteins. Discriminant analysis has been successfully applied in proteomics studies, exposing variables that effectively predict group classification of individual samples (e.g. Imre et al., 2008; Cecconi et al., 2008). Before the changes in protein expression identified in this study can be applied as biomarkers in future bio-monitoring, they will have to pass a verification and validation process, where the differential protein expressions will need to be confirmed by more targeted assays and techniques, e.g. using specific antibodies, and the analysis must be extended to a larger number of samples including a broader range of controls and cases to confirm sensitivity and assess specificity (Rifai et al., 2006).

5. Conclusion In this study we were able to detect alterations in the proteome of whole cod fry induced by exposure to low levels of produced water. Different patterns of response were observed between 17␤-oestradiol and produced water exposure, indicating different modes of action. Myosin heavy chain, fast skeletal muscle alpha-actin, Hsc71, alpha-actinin, ATP synthase, and keratin were identified as potential biomarker candidates of the effects of produced water and 17␤-oestradiol on cod fry. Such responses reflect a potential impairment of the general growth and develop-

ment of cod fry. Impairment of growth and development may have important economic consequences for fisheries as the biomass of the fish stocks can be reduced. The protein changes observed may function as early warning signals preceding more adverse biological effects among individuals exposed to produced water. These biomarker candidates will have to be validated before use in environmental monitoring of exposure and effects of discharges from the petroleum industry offshore, aiding future environmental risk analysis and risk management. Acknowledgements This study was supported by the Norwegian Research Council grants 164423/S40 and 141213/720 and by Total E&P Norway. The instruments for off-line LC–MS and MS/MS were funded by the Danish Research Agency through the Danish Biotechnology Instrument Centre. We want to thank Professor Audrey Geffen (University of Bergen) for valuable discussions on statistical approaches. References Bern, H.A., 1992. The development of the role of hormones in development—a double remembrance. Endocrinology 131 (5), 2037–2038. Blaxter, J.H.S., 1986. Development of sense organs and behaviour in teleost larvae with special reference to feeding and predator avoidance. Trans. Am. Fish. Soc. 115, 98–114. Boitsov, S., Meier, S., Klungsøyr, J., Svardal, A., 2004. Gas chromatography–mass spectrometry analysis of alkylphenols in produced water from offshore oil installations as pentafluorobenzoate derivatives. J. Chromatogr. A 1059 (1/2), 131–141.

A. Bohne-Kjersem et al. / Aquatic Toxicology 96 (2010) 280–289 Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Carson, F., 1990. A self-instructional text. In: Histotechnology. ASCP, IL, pp. 156–157. Cecconi, D., Tessari, M., Willé, D.R., Zoli, M., Domenici, E., Righetti, P.G., Carboni, L., 2008. Serum proteomic analysis during nicotine self-administration, extinction and relapse in rats. Electrophoresis 29 (7), 1525–1533. Chaitin, M.H., Burnside, B., 1989. Actin filament polarity at the site of rod outer segment disk morphogenesis. Invest. Ophthalmol. Vis. Sci. 30 (12), 2461–2469. Colborn, T., vom Saal, F.S., Soto, A.M., 1993. Developmental effects of endocrinedisrupting chemicals in wildlife and humans. Environ. Health Perspect. 101 (5), 378–384. Denslow, N.D., Lee, H.S., Bowman, C.J., Hemmer, M.J., Folmar, L.C., 2001. Multiple responses in gene expression in fish treated with estrogen. Comp. Biochem. Physiol. Part B: Biochem. Mol. Biol. 129 (2/3), 277–282. Evanson, M., Van Der Kraak, G.J., 2001. Stimulatory effects of selected PAHs on testosterone production in goldfish and rainbow trout and possible mechanisms of action. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 130 (2), 249–258. Falk-Petersen, I.B., 2005. Comparative organ differentiation during early life stages of marine fish. Fish Shellfish Immunol. 19 (5), 397–412. Feder, M.E., Hofmann, G.E., 1999. Heat-shock proteins, molecular chaperones, and the stress response: evolutionary and ecological physiology. Annu. Rev. Physiol. 61, 243–282. Görg, A., Obermaier, C., Boguth, G., Harder, A., Scheibe, B., Wildgruber, R., Wiess, W., 2000. The current state of two-dimensional electrophoresis with immobilized pH-gradients. Electrophoresis 21, 1037–1053. Hahn, M.E., 2001. Dioxin toxicology and the aryl hydrocarbon receptor: insights from fish and other non-traditional models. Mar. Biotechnol. 3 (Suppl. 1), S224–S238. Hasselberg, L., Meier, S., Svardal, A., 2004. Effects of alkylphenols on redox status in first spawning Atlantic cod (Gadus morhua). Aquat. Toxicol. 69 (1), 95–105. Imre, T., Kremmer, T., Héberger, K., Molnár-Szöllosi, E., Ludányi, K., Pócsfalvi, G., Malorni, A., Drahos, L., Vékey, K., 2008. Mass spectrometric and linear discriminant analysis of N-glycans of human serum alpha-1-acid glycoprotein in cancer patients and healthy individuals. J. Proteomics 71 (2), 186–197. Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsfjord, L.W., Bourque, B.J., et al., 2002. Historical overfishing and the recent collapse of coastal ecosystems. Science 293, 629–637. Kirfel, J., Peters, B., Grund, C., Reifenberg, K., Magin, T.M., 2002. Ectopic expression of desmin in the epidermis of transgenic mice permits development of a normal epidermis. Differentiation 70 (1), 56–68. Kjesbu, O.S., 1989. The spawning activity of cod, Gadus morhua L. J. Fish Biol. 34, 195–2006. Kristiansen, T.S., Jorstad, K.E., Ottera, H., Paulsen, O.I., Svasand, T., 1997. Estimates of larval survival of cod by releases of genetically marked yolk-sac larvae. J. Fish Biol. 51 (Suppl. A), 264–283. Laemmli, U.K., 1970. Cleavage of structural proteins during assembly of head bacteriophage-T4. Nature 227 (5259), 680–685. Livingstone, D.R., 2001. Contaminant-stimulated reactive oxygen species production and oxidative damage in aquatic organisms. Mar. Pollut. Bull. 42 (8), 656–666. Lotze, H.K., Milewski, I., 2004. Two centuries of multiple human impacts and successive changes in a North Atlantic food web. Ecol. Appl. 14 (5), 1428–1447. Machala, M., Vondracek, J., Blaha, L., Ciganek, M., Neca, J.V., 2001. Aryl hydrocarbon receptor-mediated activity of mutagenic polycyclic aromatic hydrocarbons determined using in vitro reporter gene assay. Mutat. Res. 497 (1/2), 49–62. MacLatchy, D.L., Van Der Kraak, G.J., 1995. The phytoestrogen beta-sitosterol alters the reproductive endocrine status of goldfish. Toxicol. Appl. Pharmacol. 134 (2), 305–312. Meier, S., Andersen, T.C., Lind-Larsen, K., Svardal, A., Holmsen, H., 2007a. Effects of alkylphenols on glycerophospholipids and cholesterol in liver and brain from female Atlantic cod (Gadus morhua). Comp. Biochem. Physiol. C Toxicol. Pharmacol. 145 (3), 420–430. Meier, S., Grøsvik, B.E., Nyhammer, G., Geffen, A., Knag, A.C., Makhotin, V., et al., 2007b. Effects on development, sex differentiation and reproduction on Atlantic cod exposed to produced water during early life stages. In: SETAC Europe 17th Annual Meeting, Porto, Portugal. Meier S, Morton, H.C., Nyhammer, G., Grøsvik, B.E., Makhotin, V., Audrey Geffen, A., Boitsov, S., Kvestad, K.A., Bohne-Kjersem, A.B., Goksøyr, A., Folkvord, A., Klungsøyr, J., Svardal. A., submitted. Development of Atlantic cod (Gadus morhua) exposed to produced water during early life stage. Effects on embryos, larvae, and juvenile fish. Mar. Environ. Res. Mukhopadhyay, I., Nazir, A., Saxena, D.K., Chowdhuri, D.K., 2003. Heat shock response: hsp 70 in environmental monitoring. J. Biochem. Mol. Toxicol. 17 (5), 249–254. Neff, J.M., 2002. Bioaccumulation in marine organisms. In: Effect of Contaminants From oil well Produced Water. Elsevier B.V., Amsterdam, p. 452. Neuhoff, V., Arold, N., Taube, D., Ehrhardt, W., 1988. Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background

289

at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9 (6), 255–262. O’Connell, C.P., 1981. Development of organ systems in the northern anchovy Engraulis mordax and other teleosts. Am. Zool. 21, 429–446. O’Farrell, P.H., 1975. High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250 (10), 4007–4021. Olsvik, P.A., Lie, K.K., Jordal, A.E., Nilsen, T.O., Hordvik, I., 2005. Evaluation of potential reference genes in real-time RT-PCR studies of Atlantic salmon. BMC Mol. Biol. 6, 21. Ono, Y., Liang, C., Ikeda, D., Watabe, S., 2006. cDNA cloning of myosin heavy chain genes from medaka Oryzias latipes embryos and larvae and their expression patterns during development. Dev. Dyn. 235 (11), 3092–3101. Perkins, D.N., Pappin, D.J., Creasy, D.M., Cottrell, J.S., 1999. Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20 (18), 3551–3567. Poplinskaia, V.A., 1995. The cytostructure and morphogenesis of the rod outer segments. Ontogenez 26 (1), 5–21. Pratt, W.B., 1997. The role of the hsp90-based chaperone system in signal transduction by nuclear receptors and receptors signaling via MAP kinase. Annu. Rev. Pharmacol. Toxicol. 37, 297–326. Reid, D.J., MacFarlane, G.R., 2003. Potential biomarkers of crude oil exposure in the gastropod mollusc. Austrocochlea porcata: laboratory and manipulative field studies. Environ. Pollut. 126 (2), 147–155. Rifai, N., Gillette, M.A., Carr, S.A., 2006. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat. Biotechnol. 24 (8), 971–983. Rolland, R.M., 2000. A review of chemically-induced alterations in thyroid and vitamin A status from field studies of wildlife and fish. J. Wildl. Dis. 36 (4), 615–635. Ryan, J.A., Hightower, L.E., 1996. Stress proteins as molecular biomarkers for environmental toxicology. EXS 77, 411–424. Rye, H., Reed, M., Ekrol, N., 1998. Sensitivity analysis and simulation of dispersed oil concentrations in the North Sea with the PROVANN model. Environ. Model. Software 13, 421–429. Rye, H., Johansen, Ø., Reed, M., Ekrol, N., Deqi, X., 2000. Exposure of fish larvae to hydrocarbon concentration fields generated by subsurface blowouts. Spill Sci. Technol. Bull. 6 (2), 113–123. Santodonato, J., 1997. Review of the estrogenic and antiestrogenic activity of polycyclic aromatic hydrocarbons: relationship to carcinogenity. Chemosphere 34 (4), 835–848. Schaffeld, M., Schultess, J., 2006. Genes coding for intermediate filament proteins closely related to the hagfish “thread keratins (TK)” alpha and gamma also exist in lamprey, teleosts and amphibians. Exp. Cell Res. 312 (9), 1447–1462. Schwaiger, J., Mallow, U., Ferling, H., Knoerr, S., Braumbeck, T., Kalbfus, W., et al., 2002. How estrogenic is nonylphenol? A transgenerational study using rainbow trout (Onchrhynchus mykiss) as a test organism. Aquat. Toxicol. 59 (3–4), 177–189. Shevchenko, A., Jensen, O.N., Podtelejnikov, A.V., Sagliocco, F., Wilm, M., Vorm, O., et al., 1996. Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. Proc. Natl. Acad. Sci. U.S.A. 93 (25), 14440–14445. Steen, H., Mann, M., 2004. The ABC’s (and XYZ’s) of peptide sequencing. Nat. Rev. Mol. Cell Biol. 5 (9), 699–711. Steinbacher, P., Haslett, J.R., Six, M., Gollmann, H.P., Sanger, A.M., Stoiber, W., 2006. Phases of myogenic cell activation and possible role of dermomyotome cells in teleost muscle formation. Dev. Dyn. 235 (11), 3132–3143. Sundby, S., Bjørke, H., Soldal, A., Olsen, S., 1989. Mortality rates during the early life stages and year class-strength of northeast Arctic cod (Gadus morhua L.). In: Rapports et procés-verbaux des Réunions Conseil international pour l’Exploration de la Mer, 191, pp. 351–358. Thiebaud, P., Rescan, P.Y., Barillot, W., Ralliere, C., Theze, N., 2001. Developmental program expression of myosin alkali light chain and skeletal actin genes in the rainbow trout Oncorhynchus mykiss. Biochim. Biophys. Acta 1519 (1/2), 139–142. Van Nassauw, L., Harrisson, F., Callebaut, M., 1991. Localization of smooth-muscle markers in the ovaries of some ectothermic vertebrates. Anat. Rec. 229 (4), 439–446. Venkatesh, B., Tay, B.H., Elgar, G., Brenner, S., 1996. Isolation, characterization and evolution of nine pufferfish (Fugu rubripes) actin genes. J. Mol. Biol. 259 (4), 655–665. Williams, D.S., Arikawa, K., Paallysaho, T., 1990. Cytoskeletal components of the adherens junctions between the photoreceptors and the supportive Muller cells. J. Comp. Neurol. 295 (1), 155–164. Williams, D.S., Hallett, M.A., Arikawa, K., 1992. Association of myosin with the connecting cilium of rod photoreceptors. J. Cell Sci. 103 (Pt 1), 183–190. Worm, B., Barbier, E.B., Beaumont, N., Duffy, J.E., Folke, C., Halpern, B.S., et al., 2006. Impacts of biodiversity loss on ocean ecosystem services. Science 314 (5800), 787–790. Young, J.C., Hartl, F.U., 2002. Chaperones and transcriptional regulation by nuclear receptors. Nat. Struct. Biol. 9 (9), 640–642.