Induction of Vitellogenesis in 17A-Ethinylestradiol-Exposed Rainbow Trout (Oncorhynchus Mykiss): a Method Comparison

Comparative Biochemistry and Physiology Part C 132 (2002) 483–492

Induction of vitellogenesis in 17a-ethinylestradiol-exposed rainbow trout (Oncorhynchus mykiss): a method comparison

Tim Verslycke*, Gert F. Vandenbergh, Bram Versonnen, Katrien Arijs, Colin R. Janssen

Laboratory of Environmental Toxicology and Aquatic Ecology, Ghent University, J. Plateaustraat 22, B-9000 Ghent, Belgium

Received 22 February 2002; received in revised form 27 June 2002; accepted 2 July 2002

Abstract

( ) Juvenile rainbow trout, Oncorhynchus mykiss, were exposed to the synthetic estrogen 17a-ethinylestradiol EE2 ( ) ( ) through injection 1, 10, 25 and 50 mgEE2yg fishyweek and via water exposure 1, 10 and 100 ng EE2yl . After seven (injection and water exposure) and 14 days (only for water exposure), blood and plasma vitellogenin concentrations were quantified using indirect endpoints, i.e. plasma alkaline-labile phosphorus (ALP), plasma protein and plasma calcium. In addition, the relative gonad (GSI) and liver weight (HSI) were recorded. Actual plasma vitellogenin concentrations were measured with an enzyme immunoassay. Only fish injected with 50 mgEE2yg fish had a significantly higher gonad weight. No concentration-dependent changes in the HSI were detected in fish exposed via the water, but a significant dose-dependent increase of the HSI was observed in fish injected with EE2 . Exposure of rainbow trout to EE2 had a significant effect on all tested plasma parameters. Plasma protein, phosphoprotein and calcium concentrations were significantly higher after two weeks exposure to 100 ng EE22yl. Fish injected with 10, 25 and 50 mgEEyg fish exhibited increased plasma protein concentrations after 1 week. Compared to the controls, plasma ALP and calcium levels were significantly higher in all injected fish. A significant and positive correlation was observed between all three plasma parameters and between these indirect parameters and the actual plasma vitellogenin concentrations. These findings indicate that both the plasma ALP and the plasma calcium assay have a similar sensitivity as that of available ( ) antibody-based assays EIA , at least in EE2 exposure studies, and thus these assays can provide a rapid, simple and cost-effective alternative to available immunoassays. ᮊ 2002 Elsevier Science Inc. All rights reserved.

Keywords: Rainbow trout; Oncorhynchus mykiss; Vitellogenin; Ethinylestradiol; ALP; Calcium; EIA; Method comparison; Endocrine disruption

1. Introduction Vtg is present in very low concentrations in the plasma of immature or male organisms. However, ( ) Vitellogenin Vtg is an estrogen induced yolk- estrogenic compounds can also act on hepatic precursor lipophosphoprotein which is present in estrogen receptors to induce synthesis of vitello- the blood of oviparous vertebrates and inverte- genin (Pelissero et al., 1993). Vtg production by ( brates during vitellogenesis Bergink and Wallace, males and juveniles can therefore be considered as ) 1974 . During vitellogenesis the liver of females a biomarker of exposure to environmental estro- is stimulated to produce Vtg, which in turn, is gens (Heppell et al., 1995; Kime et al., 1999). incorporated into the yolk of developing oocytes. A number of different techniques has been used *Corresponding author. Tel.: q32-9-264-37-07; fax: q32- to indirectly detect blood Vtg levels. Frequently 9-264-37-66. used methods involve the biochemical determina- E-mail address: [email protected] (T. Verslycke). tion of phosphoprotein (as alkali-labile protein

1532-0456/02/$ - see front matter ᮊ 2002 Elsevier Science Inc. All rights reserved. PII: S1532-0456Ž 02. 00111-4 484 T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 phosphorus, ALP), total protein or calcium (Yaron compared and suggestions are made for future use et al., 1977; Whitehead et al., 1978; Korsgaard of these methods as fast and cost-effective alter- and Petersen, 1979; Scott et al., 1980; Nath and natives to the available immunoassays. Sundararaj, 1981; Craik and Harvey, 1984; Tinsley, 1985; Copeland et al., 1986; Parker and McKeown, 2. Methods 1987; Kramer et al., 1998; Christensen et al., ) 1999 . Other methods are based on immuno- 2.1. Fish agglutination (Le Bail and Breton, 1981), densi- ( tometry following electrophoresis Van Bohemen Eighty-four Juvenile rainbow trout, with an et al., 1981; Allner et al., 1999), radial immunod- ( ) average mass of 15 g, were obtained from an iffusion Hara, 1978 and radioimmunoassay aquaculturist (De Keijzerberg, The Netherlands). (Idler et al., 1979; Campbell and Idler, 1980; So ) A 10 day acclimation period preceded the exposure et al., 1985; Sumpter, 1985 . Recently, different period. During acclimation, the fish were kept in immunoassays (EIA) have been developed for ( 200-l glass tanks with carbon-filtered tap water measuring Vtg in various teleost species Specker and a biofilter. One-third of the culture water was and Sullivan, 1994; Bon et al., 1997; Brion et al., renewed manually every 2 days. 2000). Although these assays are sensitive, their use is mostly limited to one species due to the a complexity of the Vtg molecule. Moreover, these 2.2. Exposure to 17 -ethinylestradiol assays are expensive and limited in their use due to the lack of species-specific antibodies. In the first experiment, 48 fish were exposed to ( In the present study, blood Vtg levels were 1, 10 and 100 ng EE2yl and a solvent control 0.1 ) measured in rainbow trout Oncorhynchus mykiss ml EtOHyl water for 1 weeks. After 1 week (Teleostei: Salmonidae) exposed to 17a-ethinyles- exposure, 24 fish were sampled and the remaining ( ) fish were collected and analyzed at the end of the tradiol EE2 ; water-borne and injected by different exposure period. The exposure medium was direct and indirect methods. EE2 is the main active component of contraceptive pills and is also more renewed every 2 days. EE2 concentrations were resistant to breakdown than natural estrogens (Sole´ checked with liquid chromatography coupled with ( ) et al., 2000). This synthetic compound has at least multiple mass spectrometry LC-MSn and were been partially held responsible for the estrogenic within 10% of the nominal concentrations. activity of many effluents (Desbrow et al., 1998; For the second experiment, fish were injected Routledge et al., 1998; Larsson et al., 1999). with 1, 10, 25 and 50 mgEE2yg fishyweek in half ( The rainbow trout is one of the most widely doses twice a week e.g. 0.5 mgEE2yg fish twice ) used fish species in ecotoxicology and its biology a week for the 1 mg EE2 dose . Solvent controls ( ) is well suited for laboratory studies and for in situ received the vehicle 100 ml of peanut oil only, monitoring for the presence of environmental whereas controls were not injected. In both exper- estrogens (Harries et al., 1996; Jobling et al., iments, fish were exposed in a temperature-con- ( ) 1996; Thorpe et al., 2000). Furthermore, radioim- trolled room 20 8C , with an average water munoassays (Sumpter, 1985) and enzyme immuno temperature of 18 8C. assays (Bon et al., 1997) for Vtg detection in this species are available. 2.3. Collection of plasma, liver and gonad This study evaluates the potential use of indirect indicators of plasma Vtg levels. The total protein, On day 7 and day 14 (for the water exposure alkali-labile phosphoprotein and calcium concen- only), blood samples were collected from the trations in the plasma of rainbow trout exposed to caudal artery using heparinized syringes and trans-

EE2 were determined. To validate these indirect ferred into heparinized ice-cooled vials containing parameters, actual plasma Vtg concentrations were aprotinin (Sigma–Aldrich, Belgium) at a final also determined in a selected number of samples concentration of 10 mlyml. Plasma collected after with a commercially available rainbow trout vitel- centrifugation (1700=g, 10 min, 4 8C), was logenin EIA (Biosense, Norway). In addition, the shock-frozen in liquid nitrogen and kept at y80 gonadosomatic (GSI) and hepatosomatic (HSI) 8C. Liver and gonads were dissected, weighed and index were measured. The different techniques are immediately frozen in liquid nitrogen and kept at T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 485 y80 8C for further analysis. The GSI and HSI or time-dependent effects were observed on the were calculated as follows: GSI. Exposure of rainbow trout to EE had a signif- GSI (%) 2 icant effect on all four tested plasma parameters s(total gonad weightytotal body weight) (protein, ALP, Ca and Vtg). The plasma protein =100 concentration was significantly higher after 2 ( )s( ) ( ) HSI % total liver weightytotal body weight weeks exposure to 100 ng EE2yl P-0.01 , but =100 was not significantly different from the solvent

control at 1 and 10 ng EE2yl. A significant time- 2.4. Vitellogenin analysis dependent induction of the plasma protein concentration was noted at the highest exposure The plasma total protein concentration was concentration (P-0.05; Fig. 1a). Plasma phospho- determined according to Bradford (1976) with protein concentrations were significantly induced ( ) bovine serum albumin Sigma–Aldrich, Belgium by EE22 at 100 ng EE yl after 1 week exposure as protein standard. Alkaline-labile phosphorus, as (P-0.001). Plasma ALP concentrations were sig- ortho-phosphate released from vitellogenin, was nificantly higher after two weeks in the 1 and 10 extracted from 30 ml of plasma according to ng EE2yl treatments compared to those observed Wallace and Jared (1968). ortho-Phosphates were after 1 week exposure to the same treatment, but determined through a colorimetric measurement of were not significantly higher than plasma ALP acidified phosphomolybdate complexes using a concentrations in the solvent control (Fig. 1b).A commercially available kit (Sigma–Aldrich, Bel- similar induction was observed for the plasma gium). Total plasma calcium levels were measured calcium and plasma vitellogenin concentrations of ( ) by atomic absorption spectrophotometry EE2 -exposed fish Fig. 1c,d . (SpectraAA-100, Varian) in 1:10 dilutions (20 ml ) plasma in 2 ml 1% HNO3 . Plasma vitellogenin 3.2. Exposure via injection levels were determined in two plasma samples of each treatment (no plasma was left to determine The GSI of rainbow trout injected with 50 mg ) Vtg in the 1 ng EE2yl treatment at day 7 with a EE2yg fishyweek was significantly increased over commercially available EIA kit for rainbow trout naive fish after 1 week of exposure (P-0.05), but (Biosense, Norway). All samples were analyzed this increase was not significant when compared in duplicate. to the solvent control. The weight of the liver increased in a dose-dependent way and was sig- 2.5. Statistical analysis nificantly higher than the solvent control after ( exposure to 25 and 50 mgEE2yg fishyweek P- All data were checked for normality and homo- 0.05; Fig. 2b)(Table 1). Plasma protein concen- geneity of variance using Kolmogorov–Smirnov trations were significantly higher in fish injected and Bartlett’s test, respectively, with a?.05. The with 10, 25 and 50 mgEE2yg fishyweek as ( effects of the EE2 treatment and exposure time compared to the solvent control P-0.05; Fig. were tested for significance with one-way analysis 2a). Plasma ALP and calcium concentrations were of variance (ANOVA) using Scheffes´ test (Statis- induced in all exposure doses after one week (P- tica௣, Statsoft Inc., USA). 0.01; Fig. 2b,c). A similar induction was observed

in the plasma vitellogenin concentrations of EE2 - 3. Results injected fish as determined with EIA (Fig. 2d).

3.1. Exposure via the water 3.3. Correlation between protein, calcium, ALP and vitellogenin concentrations The HSI, was significantly higher after 2 weeks exposure to 10 and 100 ng EE2yl compared to 1 To examine the relationship between the differ- week exposure to the same concentrations (Table ent plasma parameters, pair-wise linear regression 1). However, compared to the control fish, no analysis was performed on both the water-borne significant effects on the HSI of exposed fish were and injection data (Fig. 3a,b). All parameters were observed on day 7 and 14. No clear concentration- significantly and positively correlated (P- 486 T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 b ) ) ) ) 0.16 0.07 0.14 0.06 or after " " " " ) ( ( ( ( 2 EE ( 100 1.41 0.17 1.13 0.20 c,d c ) ) 0.30 0.05 " " ( ( -ethinylestradiol a 50 1.75 0.30 c,d ) ) 0.06 0.04 " " ( ( 3 for all treatments in the injection experiment. s 25 1.75 0.26 n ; ) 5 a c ) ) ) ) ) ) s n ( 0.03 0.22 0.26 0.21 0.02 0.04 " " " " " " ( ( ( ( ( ( 10 0.19 1.00 1.43 1.48 0.20 0.17 GSI day 14 y a ) ) ) ) ) ) 0.04 0.22 0.39 0.14 0.03 0.03 " " " " " " ( ( ( ( ( ( and control . ) ) 5 1 0.22 0.93 1.33 1.30 0.25 0.16 s n . 0.05 ( ) - a P ) ) ) ) ) ) ( 0.01 - P 0.17 0.11 0.01 0.17 0.01 0.04 ( " " " " " " HSI day 7 ( ( ( ( ( ( y of juvenile rainbow trout after 7 days and 14 days of water-borne exposure to 17 . ) ) Solvent control 1.01 1.17 0.16 1.09 0.24 0.21 0.05 GSI ( - ) ) . ) P ( 0.08 0.06 0.05 " " ( ( - P ( Control 1.01 0.13 ) 2 GSI at day 7 in the same concentration y week and gonadosomatic index y ) ) l ) HSI g fish y l ( y y g ng ( m ng ( ( 2 2 2 ) ) ) ) ) ) S.D. S.D. S.D. S.D. S.D. S.D. " " " " " " ( ( ( ( ( ( 6 for all treatments in the water-borne exposure, except control HSI HSI GSI HSI GSI GSI Significantly different from HSI at day 7 inDenotes the significant differences same from concentration solvent control Significantly different from HSI Denotes significant differences from control s 7 day exposure EE 7 day exposure EE a b c d n 14 day exposure EE Injection Table 1 Mean hepatosomatic index 7 days exposure via injection with EE Water-borne exposure T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 487

Fig. 1. Mean plasma protein (a), ALP (b), calcium (c) and vitellogenin (d) concentrations in juvenile rainbow trout after 7 days and ( ) ( ) 14 days water-borne exposure to ethinylestradiol EE2 via the water. * Denotes significant differences from solvent control P-0.05 , ** (P-0.01), *** (P-0.001); 8 denotes significant differences from the mean protein concentration at day 7 in the same treatment (P-0.05), 88 (P-0.01), 888 (P-0.001); ns6 for protein, ALP and calcium, ns2 for Vtg; n.d.snot determined.

0.0001). Plasma ALP and calcium concentrations techniques (e.g. hepatosomatic index, gonadoso- exhibited the highest correlation and the overall matic index, plasma protein, plasma phosphopro- correlation in the injection experiment was better tein, plasma calcium) to measure induced plasma than in the exposure via the water. Actual plasma Vtg levels as alternatives to the more expensive vitellogenin concentrations were also measured in and species-specific antibody-based assays. two plasma samples for each treatment (water- Vtg induction in general results in enhanced borne and injection) and compared with plasma liver metabolism leading to an enlargement of the ALP, Ca and protein concentrations in the same liver and consequently an increased hepatosomatic samples. A positive and significant correlation was index (Medda et al., 1980; Korsgaard et al., 1983; found between actual plasma vitellogenin concen- Haux and Nordberg, 1985; Christensen et al., trations and concentrations of the plasma parame- 1999). Although a significant time-dependent ters which indirectly quantify vitellogenin (Fig. induction of the liver weight was observed in the 3c). water-borne exposures, no significant concentration-dependent effects were found (i.e. no differ- ) 4. Discussion ences between EE2 -exposed fish and control fish . However, the liver weight of the injected fish was Exposure to environmental estrogens induces significantly higher after 1 week in the 10 mg the production of the female-specific protein Vtg EE2yg fish treatment. This corroborates the find- in teleost fish. A number of different techniques ings of Christiansen et al. (1998) on increased have been used to determine plasma Vtg levels as HSI of juvenile rainbow trout (46–147 g) injected ( ) a biomarker of exposure to these estrogen mimics. with 5 mgEE2yg fish 2.85"0.07 after 9 days The present study evaluates the value of indirect as compared to the carrier control (1.02"0.14). 488 T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492

what we found after 7 days (1.75"0.3). Although the HSI has the advantage of being a relatively simple parameter to measure, it should be considered as a rather crude indicator of estrogen exposure. In addition, incautious dissection of the liver can lead to large variability and inaccuracies in the HSI leading to less discriminative power (Sil- versand, 1996; Allen et al., 1999a,b; Tremblay and Van Der Kraak, 1999). Furthermore, enlargement of the liver can also be a simple toxic effect. Due to the small size of the gonads of the immature rainbow trout, large variability in the GSI results was observed. Only fish injected with

50 mg EE2 had significantly larger gonads than the control fish after 1 week of exposure. Jobling et al. (1996) reported a reduction in the relative gonad weight of adult rainbow trout exposed to 2

ng EE2yl for 3 weeks. Similar results were found by Scholz and Gutzeit (2000) with juvenile medaka (Oryzias latipes) exposed to 10 and 100 ng

EE2yl for 2 months. The reduction of ovarian growth might be attributed to, for example, ovary- specific toxicity andyor interference with the release of gonadotropins, that have been shown to stimulate ovarian development (Chan, 1976; Jala- bert, 1976). In our study, the exposure period was probably too short and the fish too small to observe the described effects. Even in adult female fish the relative gonad weight might be of limited use since the GSI in unexposed fish changes drastically during the reproductive season. For example, in 2 year old female rainbow trout relative gonad weight rises from 0.5% during reproductive season up to 18% just before spawning (Bon et al., 1997). Furthermore, ovarian growth is also affected by environmental parameters such as oxygen content, temperature and pH (Parker and McKeown, 1987), making the GSI an extremely variable and hard to interpret biomarker. Plasma protein concentrations in rainbow trout were significantly higher in controls after injection ( ) ( ) ( ) Fig. 2. Mean plasma protein a , ALP b , calcium c and with 10, 50 and 100 mgEE2yg fish or exposure vitellogenin (d) concentrations in juvenile rainbow trout inject- to 100 ng water-borne EE yl for 2 weeks. Emmer- ( ) 2 ed with ethinylestradiol EE2 after 1 week. * Denotes signif- ( ) ( ) ( - ) ( - ) sen and Emmersen 1976 , Tinsley 1985 , Bon icant differences from control P 0.05 ,** P 0.01 ; 8 ( ) denotes significant differences from solvent control (P-0.05), et al. 1997 found a similar induction in plasma 88 (P-0.01); ns3 for protein, ALP and calcium, ns2 for Vtg. protein concentrations in fish injected with, respectively, 5, 2 and 1.5 mg17b-oestradiolyg fishy Exposure time has an important effect on the HSI. week. No literature data were found on plasma

In a previous similar injection experiment with protein concentrations in fish injected with EE2 , juvenile rainbow trout treated with 50 mgEE2ygy but our findings suggest a significant induction in fishyweek, we found a HSI of 3.42"0.81 after 14 the same concentration range as seen with 17b- days (unpublished data), which is almost twice of oestradiol. Effects on plasma protein concentra- T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 489

Fig. 3. Correlation between the different plasma parameters in juvenile rainbow trout exposed to 17a-ethinylestradiol via the water (a) and via injection (b). Correlation between the vitellogenin concentrations measured with EIA and the other plasma parameters in 25 plasma samples of juvenile rainbow trout exposed to 17a-ethinylestradiol via wateryinjection (c).

tions in rainbow trout exposed to EE2 via the tions. Induced plasma Vtg levels can be detected water were not as pronounced as after injection in rainbow trout at concentrations as low as 2 ng ( with EE2 . In general, the plasma protein concen- EE2yl by specific immunoassays. Jobling et al., tration was a less sensitive endpoint than the other 1996; Larsson et al., 1999). In this study, however, measured parameters. no significant induction was found in the plasma

Rainbow trout exposed to 100 ng EE2yl for 7 vitellogenin concentrations in the 1 and 10 ng days had significantly higher ALP and plasma EE2yl treatments. Our results indicate that both calcium levels. The significantly higher plasma the plasma ALP and the plasma calcium assay can

ALP and calcium concentration observed in the 1 be used with equal sensitivity in EE2 water expo- and 10 ng EE2yl treatment after 14 days when sure studies when fish are exposed for a prolonged ( ) compared to 7 days in the same treatment, may period e.g. at least 2 weeks . EE2 concentrations indicate that longer exposure would have resulted in surface waters of up to 7 ng EE2yl have been in an increase in even the lowest test concentra- reported (Desbrow et al., 1998; Larsson et al., 490 T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492

1999), which means that both the ALP and calci- tration range. However, the ALP protocol is quite um assay can potentially be used under field elaborate and a lot of manipulations are required. conditions to indirectly detect the induction of In contrast with the ALP determination, the calci- vitellogenesis. Current research with rainbow trout um assay is simple and rapid, requiring only a exposed to effluents of sewage treatment plants dilution of the sample and the AAS measurement. for 3 weeks confirms the successful use of ALP Assay variability for the calcium assay in this and calcium assays in the field (Versonnen et al., study was consequently approximately half of ALP in preparation). assay variability. Since calcium and ALP were ( 2 ) Injection of rainbow trout with 1 mgEE2yg fish well correlated in both experiments R )0.9 ,we induced significantly higher plasma calcium and suggest the calcium method to be the preferred plasma ALP levels after 7 days (P-0.01). Induc- indirect assay for the detection of vitellogenin in tion in plasma ALP and calcium concentration was estrogen exposed fish. also found in fish injected with 2–5 mg17b- oestradiolyg fishyweek (Emmersen and Emmer- 5. Conclusions sen, 1976; Tinsley, 1985). Purdom et al. (1994), Christiansen et al. (1998), Sole´ et al. (2000) Rainbow trout, exposed to the synthetic estrogen reported an increase in plasma vitellogenin in fish 17a-ethinylestradiol had significantly higher plas- injected with 5 mgEE2yg fish after 9 days. Our ma vitellogenin levels as measured by indirect results and those of these reports suggest that the methods, i.e. plasma alkaline-phosphorus, plasma plasma calcium and ALP assays have an equal protein and plasma calcium. These results were sensitivity in the tested concentration range. In significantly correlated with actual plasma Vtg addition, the results of the ALP and Ca assays concentrations. The plasma calcium and ALP correlated well with the results of the immuno assays can be used as rapid, easy and inexpensive assay. The strong induction in all measured plasma alternatives to the available immunoassays with parameters, even in the lowest EE2 exposure doses, similar sensitivity. In future studies the use of these could indicate a pharmacological rather than a assays under field conditions will be evaluated. physiological response. This strong induction is also apparent in the bimodal spreading of the data. Acknowledgments Future injection experiments with EE2 in which (- lower injection doses are used 1 mgEE2yg This research was supported by a research grant fishyweek) might result in a better linear response. of the Flemish Institute for the Promotion of Plasma ALP, calcium and protein levels were Scientific and Technological Research in Industry significantly and positively correlated in rainbow (IWT). We wish to especially recognize the work trout exposed to EE2 . All three parameters signif- of Kristof Titeux and the valuable technical assis- icantly correlated with actual plasma Vtg concen- tance of Barbara Deryckere and Jill Van trations. Vtg is the only phosphorus containing Reybrouck. protein in the blood of oviparous vertebrates ( Emmersen and Petersen, 1976; Whitehead et al., References 1983; Tyler et al., 1988) and contains 0.6–0.8% protein-bound phosphorus (Silversand, 1996). The Allen, Y., Matthiessen, P., Scott, A.P., Haworth, S., Feist, S., presence of alkali-labile protein phosphorus in fish Thain, J.E., 1999a. The extent of oestrogenic contamination plasma is specifically associated with the lipo- in the UK estuarine and marine environments-further sur- phosphoprotein Vtg. This is clearly shown in the veys of flounder. Sci. Total Environ. 233, 5–20. linear regression between ALP and Vtg data (Fig. Allen, Y., Scott, A.P., Matthiessen, P., Haworth, S., Thain, J.E., ) ( ) Feist, S., 1999b. Survey of estrogenic activity in United 3c . Wallace 1970 showed that one atom of Kingdom estuarine and coastal waters and its effects on calcium is associated with every protein phosphate gonadal development of the flounder Platichthys flesus. group in the Vtg complex. Our study demonstrates Environ. Toxicol. Chem. 18, 1791–1800. that ALP and calcium levels were both strongly Allner, B., Wegener, G., Knacker, T., Stahlschmidt-Allner, P., elevated by EE . The induction of plasma ALP 1999. Electrophoretic determination of estrogen-induced 2 protein in fish exposed to synthetic and naturally occurring was approximately three times higher than for chemicals. Sci. Total Environ. 233, 21–30. plasma calcium and protein, which may result in Bergink, E.W., Wallace, R.A., 1974. Estrogen-induced synthe- a higher discriminative power in the low concen- sis of yolk proteins in roosters. Am. Zool. 14, 1177–1193. T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492 491

Bon, E., Barbe, U., Rodriguez, J.N., et al., 1997. Plasma environmental estrogens. Environ. Health Perspect. 103, vitellogenin levels during the annual reproductive cycle of 9–15. the female rainbow trout (Oncorhynchus mykiss): establish- Idler, D.R., Hwang, S.J., Crim, L.W., 1979. Quantification of ment and validation of an ELISA. Comp. Biochem. Physiol. vitellogenin in Atlantic salmon, Salmo salar, plasma by 117B, 75–84. radioimmunoassay. J. Fish. Res. Board Can. 36, 574–578. Bradford, M.M., 1976. A rapid and sensitive method for Jalabert, B., 1976. In vitro oocyte maturation and ovulation in quantitation of microgram quantities of protein utilizing the rainbow trout (Salmo gairdneri), northern pike (Essox principle of protein-dye binding. Anal. Biochem. 72, lucius), and goldfish (Carassius auratus). J. Fish. Res. 248–254. Board Can. 33, 974–988. Brion, F., Rogerieux, F., Noury, P., et al., 2000. Two-step Jobling, S., Sheahan, D., Osborn, J.A., Matthiessen, P., Sump- purification method of vitellogenin from teleost fish species: ter, J.P., 1996. Inhibition of testicular growth in rainbow rainbow trout (Oncorhynchus mykiss), gudgeon (Gobio trout (Oncorhynchus mykiss) exposed to estrogenic alkyl- gobio) and chub (Leuciscus cephalus). J. Chromatogr. B737, phenolic chemicals. Environ. Toxicol. Chem. 15, 194–202. 3–12. Kime, D.E., Nash, J.P., Scott, A.P., 1999. Vitellogenesis as a Campbell, C.M., Idler, D.R., 1980. Characterization of an biomarker of reproductive disruption by xenobiotics. Aqua- oestradiol-induced protein from rainbow trout serum as culture 177, 345–352. vitellogenin by the composition and radio-immunological Korsgaard, B., Petersen, I.M., 1979. Vitellogenin, lipid and cross-reactivity of ovarian yolk fractions. Biol. Reprod. 22, carbohydrate metabolism during vitellogenesis and pregnan- 605–617. cy and after hormonal induction in the blenny, Zoarces Chan, K.K., 1976. Effect of synthetic luteinizing hormone- viviparus (L.). Comp. Biochem. Physiol. 63B, 245–251. releasing hormone (LH-RH) on ovarian development in Korsgaard, B., Emmersen, J., Petersen, I.M., 1983. Estradiol- Japanese medaka, Oryzias latipes. Can. J. Zool. 55, induced hepatic protein synthesis and transaminase activity 155–160. in the male flounder Platichthys flesus. Gen. Comp. Endo- Christensen, L.J., Korsgaard, B., Bjerregaard, P., 1999. The crinol. 50, 11–17. effect of 4-nonylphenol on the synthesis of vitellogenin in Kramer, V.J., Miles-Richardson, S., Pierens, S.L., Giesy, J.P., the flounder Platichthys flesus. Aquat. Toxicol. 46, 211–219. 1998. Reproductive impairment and induction of alkaline- Christiansen, T., Pedersen, K.L., Korsgaard, B., Bjerregaard, labile phosphate, a biomarker of estrogen exposure, in P., 1998. Estrogenity of xenobiotics in rainbow trout (Oncor- fathead minnows (Pimephales promelas) exposed to water- hynchus mykiss) using in vivo synthesis of vitellogenin as borne 17b-oestradiol. Aquat. Toxicol. 40, 335–360. a biomarker. Mar. Environ. Res. 46, 137–140. Larsson, D.G.J., Adolfsson-Erici, M., Parkkonen, J., et al., Copeland, P.A., Sumpter, J.P., Walker, T.K., Croft, M. Vitello- 1999. Ethinyloestradiol—an undesired fish contraceptive? genin levels in male and female rainbow trout (Salmo Aquat. Toxicol. 45, 91–97. gairdneri Richardson) at various stages of the reproductive Le Bail, P.Y., Breton, B., 1981. Rapid determination of the cycle. Comp. Biochem. Physiol. 83B, 487–493. sex of puberal salmonid fish by a technique of immunoag- Craik, J.C.A., Harvey, S.M., 1984. A biochemical method for glutination. Aquaculture 22, 367–375. distinguishing between sexes of fishes by the presence of Medda, A.K., Dasmahapatra, A.K., Ray, A.K., 1980. Effects yolk protein in the blood. J. Fish Biol. 25, 293–303. of estrogen and testosterone on the protein and nucleic acid Desbrow, C., Routledge, E.J., Brighty, G.C., Sumpter, J.P., contents of liver, muscle and gonad and plasma protein Waldock, M., 1998. Identification of estrogenic chemicals content of male and female (vitellogenic and non-vitello- in STW effluent. 1. Chemical fractionation and in vitro genic) singi fish (Heteropneustes fossilis)(Bloch). Gen. biological screening. Environ. Sci. Technol. 32, 1549–1558. Comp. Endocrinol. 34, 427–436. Emmersen, B.K., Emmersen, J., 1976. Protein, RNA and DNA Nath, P., Sundararaj, B.I., 1981. Isolation and identification of metabolism in relation to ovarian vitellogenic growth in the female-specific serum lipophosphoprotein (vitellogenin) in flounder Platichthys flesus (L.). Comp. Biochem. Physiol. the catfish, Heteropneustes fossilis (Bloch). Gen. Comp. 55B, 315–321. Endocrinol. 32, 184–190. Emmersen, B.D., Petersen, I.M., 1976. Natural occurrence, Parker, D.B., McKeown, B.A., 1987. Effects of pH andyor and experimental induction by estradiol-17-b, of a lipo- calcium-enriched freshwater on plasma levels of vitellogenin phosphoprotein (vitellogenin) in flounder (Platichthys fle- and Ca2q and on bone content during exogenous vitelloge- sus,L.). Comp. Biochem. Physiol. 54B, 443–446. nesis in rainbow trout (Salmo gairdneri). Comp. Biochem. Hara, A., 1978. Sexual differences in serum proteins of chum Physiol. 87A, 267–273. salmon and the purification of female-specific protein. Bull. Pelissero, C., Flouriot, G., Foucher, J.L., Bennetau, B., Le Gac, Jpn. Soc. Sci. Fish 44, 689–693. F., Sumpter, J.P., 1993. Vitellogenin synthesis in cultured Harries, J.E., Sheahan, D.A., Jobling, S., et al., 1996. A survey hepatocytes; an in vitro test for the estrogenic potency of of estrogenic activity in United Kingdom inland waters. chemicals. J. Steroid Biochem. 44, 263–272. Environ. Toxicol. Chem. 15, 1993–2002. Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, N.C., Tyler, Haux, C., Nordberg, B., 1985. The influence of estradiol-17b C.R., Sumpter, J.P., 1994. Estrogenic effects of effluents on the liver content of protein, lipids, glycogen and nucleic from sewage treatment works. Chem. Ecol. 8, 275–285. acids in juvenile rainbow trout Salmo gairdneri. Comp. Routledge, E.J., Sheahan, D.A., Desbrow, C., Brighty, E.J., Biochem. Physiol. 81B, 275–279. Waldock, M., Sumpter, J.P., 1998. Identification of estrogen- Heppell, S.A., Denslow, N.D., Folmar, L.C., Sullivan, C.V., ic chemicals in STW effluent. 2. In vivo responses in trout 1995. Universal assay of vitellogenin as a biomarker for and roach. Environ. Sci. Technol. 32, 1559–1565. 492 T. Verslycke et al. / Comparative Biochemistry and Physiology Part C 132 (2002) 483–492

Scholz, S., Gutzeit, H.O., 2000. 17-a-ethinylestradiol affects (Oncorhynchus mykiss). Environ. Toxicol. Chem. 19, reproduction, sexual differentiation and aromatase gene 2812–2820. expression of the medaka (Oryzias latipes). Aquat. Toxicol. Tinsley, D., 1985. A comparison of plasma levels of phospho- 50, 363–373. protein, total protein and total calcium as indirect indices of Scott, A.P., Bye, V.J., Baynes, S.M., 1980. Seasonal variations exogenous vitellogenesis in the crucian carp Carassius in sex steroids of female rainbow trout (Salmo gairdneri carassius (L.). Comp. Biochem. Physiol. 80B, 913–916. Richardson). J. Fish Biol. 17, 587–592. Tremblay, L., Van Der Kraak, G., 1999. Comparison between Silversand, C., 1996. Vitellogenesis in Teleost Fish, a Study the effects of the phytosterol beta-sitosterol and pulp and of Vitellogenin and Egg Lipids. Vasastadens Bokbinderi, paper mill effluents on sexually immature rainbow trout. Goteborg¨ 84 pp. Environ. Toxicol. Chem. 18, 329–336. So, Y.P., Idler, D.R., Hwang, S.J., 1985. Plasma vitellogenin Tyler, C.R., Sumpter, J.P., Bromage, N.R., 1988. In vivo uptake in landlocked Atlantic salmon (Salmo salar Ouananiche): and processing of vitellogenin in the rainbow trout Salmo isolation, homologous radioimmunoassay and immunologi- gairdneri. J. Exp. Zool. 246, 171–179. cal cross-reactivity with vitellogenin from other teleosts. Van Bohemen, Ch.G., Lambert, J.G.D., Peute, J., 1981. Annual Comp. Biochem. Physiol. 81B, 63–71. changes in plasma and liver in relation to vitellogenesis in the female rainbow trout, Salmo gairdneri. Gen. Comp. Sole,´´ M., Porte, C., Barcelo, D., 2000. Vitellogenin induction Endocrinol. 44, 94–107. and other biochemical responses in carp, Cyprinus carpio, Wallace, R.A., 1970. Studies on amphibian yolk. IX. Xenophus after experimental injection with 17a-ethynylestradiol. Arch. vitellogenin. Biochim. Biophys. Acta 215, 176–183. Environ. Contam. Toxicol. 38, 494–500. Wallace, R.A., Jared, D.W., 1968. Studies on amphibian yolk. Specker, J.L., Sullivan, C.V., 1994. Vitellogenesis in fishes: VII. Serum phosphoprotein synthesis by vitellogenic females status and perspectives. In: Davey, K.G., Peter, R.E., Tobe, and oestrogen-treated males. Can. J. Biochem. 46, 953–959. ( ) S.S. Eds. , Perspectives in Comparative Endocrinology. Whitehead, C., Bromage, N.R., Forster, J.R.M., 1978. Seasonal National Research Council, Ottowa, Ontario, Canada, pp. changes in reproductive function of the rainbow trout (Salmo 304–315. gairdneri). J. Fish Biol. 12, 601–608. Sumpter, J.P., 1985. The purification, radioimmunoassay and Whitehead, C., Bromage, N.R., Breton, B., 1983. Changes in plasma levels of vitellogenin from the rainbow trout, Salmo serum levels of gonadotropin, oestradiol-17b and vitellogen- gairdneri. In: Lofts, B., Holmes, W.M. (Eds.), Current in during the first and subsequent reproductive cycles of Trends in Comparative Endocrinology. Hong Kong Univer- female rainbow trout. Aquaculture 34, 317–326. sity Press, Hong Kong, China, pp. 355–357. Yaron, Z., Terkatin-Shimony, A., Shaham, Y., Salzer, H., 1977. Thorpe, K.L., Hutchinson, T.H., Hetheridge, M.J., Sumpter, Occurrence and biological activity of oestradiol-17b in the J.P., Tyler, C.R., 2000. Development of an in vivo screening intact and ovariectomised Tilapia aurea (Cichlidae: Teleos- assay for estrogenic chemicals using juvenile rainbow trout tei). Gen. Comp. Endocrinol. 33, 45–52.