Uptake of 17β-estradiol and biomarker responses in brown trout (Salmo trutta) exposed to pulses

Environmental Pollution 159 (2011) 3374e3380

Contents lists available at SciVerse ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Uptake of 17b-estradiol and biomarker responses in brown trout (Salmo trutta) exposed to pulses Jacob J.G. Knudsen, Henrik Holbech, Steffen S. Madsen, Poul Bjerregaard* Institute of Biology, University of Southern Denmark, Campusvej 55, DK-5230 Odense, Denmark

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 April 2011 Received in revised form 29 July 2011 Accepted 19 August 2011

In streams, chemicals such as 17b-estradiol (E2) are likely to occur in pulses. We investigated uptake and biomarker responses in juvenile brown trout (Salmo trutta) of 3- or 6-h pulses of concentrations up to 370 ng E2 L
1. Uptake by the fish was estimated from disappearance of E2 from tank water. A single 6-h pulse of 370 ng E2 L
1 increased the plasma vitellogenin concentration, liver Era- and vitellogeninmRNA. Exposure to 150e160 ng E2 L
1 for 6 h increased vitellogenin in one experiment but not in another. Two 6-h pulses had a larger effect one pulse. Brown trout in the size range 24e74 g took up E2 linearly with time and exposure concentration with a concentration ratio rate of 20.2 h
1. In conclusion, the threshold for induction of estrogenic effects in juvenile brown trout at short term pulse exposure appears to be in the range 150e200 ng E2 L
1. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Salmo trutta Estrogen Pulse exposure Vitellogenin E2-uptake

1. Introduction Estrogens discharged from wastewater treatment plants (WWTP) may cause feminisation of male fish in contaminated rivers and estuaries (Sumpter, 2005; Desforges et al., 2010). Estrone (E1), 17b-estradiol (E2) and ethinylestradiol (EE2) are the three estrogenic species known to contribute most of the estrogenicity in wastewater from WWTPs (Desbrow et al., 1998). Recent investigations have shown that WWTPs are not the only sources for addition of estrogens to the aquatic environment: Discharges of estrogens from scattered houses in the open land (Stuer-Lauridsen et al., 2006), leaching from farmland treated with liquid manure (Dyer et al., 2001; Stuer-Lauridsen et al., 2005; Matthiessen et al., 2006; Hildebrand et al., 2006; Kjaer et al., 2007) and discharges from dairy farms (Matthiessen et al., 2006) may contribute to the estrogenic load in streams. Discharges from simple septic tanks in the open land may have estrogenic activities as high as 400 ng 17b-estradiol equivalents (EEQ) L
1 (StuerLauridsen et al., 2005, 2006) and Kjaer et al. (2007) observed that precipitation incidents caused E1 and E2 to leach into drainage water for as long as three months after application of manure to farmland. This type of leaching and discharges from houses in the open land are likely to result in pulsed rather than constant exposure of the fish inhabiting the streams e especially in headwater

* Corresponding author. E-mail address: [email protected] (P. Bjerregaard). 0269-7491/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2011.08.039

streams with limited dilution capacity. Yet most of our knowledge about concentrationeresponse relationships and thresholds for estrogenic responses in various fish species has been obtained in laboratory studies with constant exposure regimes. Therefore, the question is: What does an EC50-value (15 ng E2 L
1) for induction of vitellogenin in juvenile brown trout over 8 days constant exposure (Bjerregaard et al., 2008) tell us about the potential impacts of exposure to shorter pulses of e.g. 400 ng EEQ L
1? Induction of the synthesis of the yolk-precursor protein vitellogenin in juvenile and male fish is the most well established biomarker of estrogenic contamination of the aquatic environment (Sumpter and Jobling, 1995; Harries et al., 1997), but other sensitive and more rapidly responding molecular biomarkers of estrogenic activity include hepatic transcript levels of vitellogenin and the hepatic estrogen receptor a, as these are known to represent estrogen-responsive genes (Pakdel et al., 1991; Burki et al., 2006). Several E2-injection experiments have demonstrated the effect of acute estrogen exposure. Most of these have used relatively high E2 doses (0.1e5 mg kg
1; rainbow trout (Oncorhynchus mykiss) (Leguellec et al., 1988; Pakdel et al., 1991; Donohoe and Curtis, 1996; Arukwe et al., 2001), fathead minnow (Pimephales promelas) (Korte et al., 2000), sheepshead minnow (Cyprinodon variegatus) and brown trout (Sherry et al., 1999)). The data from these experiments are, however, difficult to extrapolate to a natural scenario, where pulse exposure to lower and waterborne concentrations often occurs. Only a few studies have investigated the lower threshold (dosage and duration) for waterborne estrogenic exposure. Yamaguchi et al. (2005) water-exposed Japanese medaka

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

(Oryzias latipes) to E2 for 8 h and reported LOEC of 0.1 mg E2 L
1 for vitellogenin-mRNA induction and <0.1 mg E2 L
1 for ERa-mRNA induction. Panter et al. (2000) exposed fathead minnow to intermittent 1e3 d exposures of E2 over a period of 120 days and plasma vitellogenin concentrations were elevated to approximately the same degree as those in chronic exposure experiments using the same concentration. Thus, the response was higher than in continuous exposure to the equivalent time-weighted average concentration, suggesting a “memory-effect” of previous exposures. The brown trout is widely distributed in headwater streams in Europe and has been used as a monitoring organism for endocrine disruption (Korner et al., 2005; Vermeirssen et al., 2005; Burki et al., 2006). There are reports of elevated plasma concentrations of vitellogenin in male brown trout from a number of streams (Bjerregaard et al., 2006, 2008; Kelly et al., 2010). The purpose of the present investigation was to determine E2-uptake and use different molecular biomarkers (ERa-mRNA, vitellogenin-mRNA, vitellogenin protein) to investigate the threshold for the estrogenic effect in brown trout upon pulsed exposure to 17b-estradiol.

3375

were exposed to pulses of E2 (SigmaeAldrich, Brøndby, Denmark) by being transferred by netting to steel exposure tanks with the desired concentration of E2 for a period of 3 or 6 h (Table 1). In experiment 1, a static system was used (3 h exposure) and flow-through systems were used in experiment 2e6 (6 h exposure). In the flowthrough system, administration of water and E2 was controlled by two peristaltic pumps (Ole Dich Instrument Makers, Copenhagen, Denmark). E2 was dissolved in 96% ethanol before administration. During the experiment, tap water and stock solution were mixed to the desired concentration before entering the steel tanks. Submerged circulation pumps positioned in the tanks were used to ensure complete and uniform dispersal of the test compounds. Only the ethanol was added to control tanks. To characterise actual exposure concentrations and to follow the removal of E2 from the water phase during the exposure, water samples for quantification of E2 were taken during the pulse exposures as indicated in Table 1. 2.3. Experiments The brown trout were subjected to various types (duration, exposure concentrations, single/repeated pulses) of pulsed exposure to E2 as presented in Table 1. After the pulse exposure, the fish were transferred by netting to flow-through tanks with uncontaminated water and blood samples were taken from anaesthetized (50e100 mg MS-222 L
1 [ethyl 3-aminobenzoate methanesulfonate salt, SigmaeAldrich, Brøndby, Denmark]) fish during the days after the exposure as indicated in Table 1. In experiment 3, the liver was removed for analysis of ERamRNA and vitellogenin-mRNA. 2.4. Analytical procedures

2. Materials and methods 2.1. Experimental animals Sexually immature brown trout (Salmo trutta) were obtained during 2006 from Funen Salmon Fishing, Elsesminde, Denmark, where they had been reared from hatch. The juvenile fish were the offspring of parent brown trout caught during the previous year in Funen in streams on their way back to breed. At Elsesminde the juvenile fish had been kept in 16 m3 tanks with approximately one water exchange per day at 7e9  C and ambient light:dark regime. The water supply consisted of 99.9% re-circulated, filtered groundwater and 1& groundwater. The fish were brought up on commercial trout food (Aller Mølle, Brande, Denmark). In the laboratory, the fish were acclimatized (12.5e14  C) in flow-through systems of 55 or 112.5 L stainless steel tanks for 4e30 days. During the acclimatization and experimental periods the fish were held at a 12:12 light:dark cycle and they were fed commercial (Aller Aqua, Brande, Denmark) trout feed (0.5e1% body weight/day). 2.2. Experimental set-up During experiments, the fish were held in the stainless steel tanks (55 or 112.5 L) with flows of tap water (groundwater) between 130 and 173 L per 24 h. The fish

2.4.1. Vitellogenin analysis Vitellogenin was measured in either plasma samples (Exp. 1e4 and 6) or whole body homogenates (Exp. 5). Blood was collected from the caudal vessels from MS-222 anaesthetized fish by means of a heparinised (5000 IU ml
1) syringe, transferred to heparinised Eppendorf tubes on ice, centrifuged (3 min, 20,000 rpm, 4  C) and the supernatant was stored at
80  C in aliquots until measurement. Whole body homogenates were prepared by first anaesthetising and killing the fish with an overdose of buffered MS-222. After being weighed, fish were placed separately in microcentrifuge tubes on ice, crushed with a small plastic pistil and immediately mixed with two times the body weight of ice cold homogenisation buffer (50 mM TriseHCl, pH 7.4; 1% protease inhibitor cocktail (SigmaeAldrich, MO, USA)). The homogenates were centrifuged (20 min, 50,000 g, 4  C) and the supernatants were stored at
80  C in aliquots until measurement. A homologous, direct non-competitive sandwich Enzyme Linked Immuno Sorbent Assay (ELISA) as described by Bjerregaard et al. (2006) was used to quantify the vitellogenin concentration in plasma or whole body homogenates. 2.4.2. RT Q-PCR Livers removed from the decapitated fish were immediately frozen in liquid nitrogen and stored at
80  C until RNA isolation. Total RNA was isolated by the Trizol method according to the manufacturer’s manual (Invitrogen, Carlsbad, CA,

Table 1 Salmo trutta. Experimental conditions. Actual concentrations and fish weights are given as mean  Standard error of mean. Exp

[E2] (ng L
1)

Num-ber of tanks

Fish pr. tank

Nomi-nal

Actual

1

0 25 50 100 200

<1 24  2 48  4 87  9 182  18

1 1 1 1 1

2

0 200

<1 156  11

3

0 400

4

Exposure

Weight (g)

Month

Day in experiment

Duration (h)

12 12 12 12 12

0 0 0 0 0

3 3 3 3 3

53.0 63.4 61.9 61.0 60.9

3.0 5.0 4.5 4.5 4.8

Feb

2 5

13 13e15

0 0

6 6

44.6  3.4 60.0  2.7

<1 370  5

2 4

16 10e11

0 0

6 6

0 200

<1 159  6

2 2

10 10

0 0

5

0 200

<1 206  2

2 2

25/15 30/20

6

0 100

<1 64  6

1 1

50/35/20 50/35/20

a

Repeated sampling from the same fish.

Age

Temp.  C

Sampling Water (hours in pulse)

Vitellogenin (days)

1

0, 1, 2, 3

3a , 6 a

12.5

Mar

1

0, 3, 6

2, 4, 6

12.5

49.3  3.2 53.6  2.9

May

1þ

0, 3, 6

1, 2, 4, 6

12.5

6 6

74.2  6.8 70.3  5.4

Jul

1þ

0, 3, 6

1, 4

15.5

0, 2 0, 2

6 6

1.80  0.07 1.8  0.1

Aug

0þ

0, 3, 6

4, 6

14

0, 2, 4 0, 2, 4

6 6

28.6  1.1 23.6  0.8

Dec

0þ

0, 3, 6

4, 6, 8

12.5

    

3.1. E2 removal from exposure tanks E2 concentrations decreased gradually after the transfer of the fish to the exposure tanks and the decrease in concentrations was fairly consistent among replicate tanks or pulses (Fig. 1A). In the groups of fish with individual average weights between 24 and 74 g, the removal of E2 from each tank correlated (r2 ¼ 0.74, p < 0.001) with the amount of biomass added (Fig. 1B), indicating that E2 lost from the water phase was taken up into the fish. For these fish, the rate for E2-uptake increased linearly with the E2 concentration in the water (E2uptake (ng/g/h) ¼ 0.0202 * [E2]water (ng/L) þ 0.376; r2 ¼ 0.96, p < 0.001) giving a concentration ratio (between water and organism) rate of 20.2 h
1 in the concentration range 24e370 ng E2 L
1 (Fig. 1C). Smaller fish (1.8  0.1 g) showed a considerably higher concentration ratio rate (123  25 h
1).

400 4

300 2

200 2

100

5 3

0

0

3

6

Time (hours) 14 13 12 11 r = 0.86 p < 0.001

10 9 8 7

500

1000

1500 -1

Biomass (g tank )

C -1

-1

E2 uptake (ng g h )

3. Results

500

B

2.4.3. Quantification of E2 in water samples Measurement of actual water concentration of E2 was performed as described by Rose et al. (2002) using solid phase extraction followed by LC-MS analysis. EE2 was used as internal standard for E2. The detection limit for E2 was approximately 1 ng L
1. 2.4.4. Data treatment and statistical analysis Statistical analyses were made by the statistical program SAS 9.1 (Cary, NC, USA). Parametric tests were used except in one case where the criteria for normal distribution and variance homogeneity were not met. A non-parametric test was then used. Logarithmic transformation was used in all parametric analyses. The ELISA data for experiment 3 were analysed by two-way analysis of variance (ANOVA) followed by a Bonferroni adjusted Fishers’s Least Significant Difference test because there was significant interaction. The other ELISA data were analysed by two-way ANOVA and QPCR data were analysed by a Students’s t-test, except in the case of vitellogenin-mRNA where a non-parametric KruskaleWallis test was used.

-1

A

-1

USA). The purity was checked by the A260/A280 ratio, which was always 1.6e1.8. Following isolation, RNA was quantified using a RiboGreen RNA Quantitation Kit (Molecular Probes, Eugene, OR, USA) according to the manufacturer’s manual. One microgram RNA was then treated with 1 unit RQ1 DNase (Promega, Madison, WI, USA) for 30 min at 37  C in a total volume of 20 mL followed by 5 min at 75  C to inactivate RQ1 DNase. First strand cDNA was synthesised from 1 mg total RNA following the instructions of the SuperScriptII RNase H-Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA, USA). Expression of ERa- and vitellogenin-mRNA was analysed by SYBR-based quantative PCR analysis using primers specific for Salmo salar (Table 2). Gene-specific expression was normalised to expression of elongation factor-1a. Polymerase chain reactions were done with Brilliant SYBR Green QPCR Master Mix (Stratagene, La Jolla, CA, USA) on a Mx3000P thermocycler (Stratagene, La Jolla, CA, USA). PCR reactions contained 1 mL cDNA (50 ng RNA), 150 nM of each primer and 12.5 mL 2 Brilliant SYBR Green Master Mix in at total volume of 25 mL. All QPCR reactions were performed as follows: 10 min of polymerase activation at 95  C, 40 cycles of 95  C for 30 s and 60  C for 1 min. Melting curve analysis was performed following each reaction to confirm the presence of only a single product of the reaction. Negative control reactions were carried out for representative samples using RNA that had not been reverse transcribed to control for the possible presence of genomic DNA contamination. Genomic DNA was present but never constituted more than 1:32,768. No-template control reactions were also performed to verify that there was no cDNA contamination or primeredimer amplification in the reactions. The results were normalised to elongation factor-1a by the 2
DDCtmethod (Livak and Schmittgen, 2001) and presented as arbitrary numbers.

E2 concentration (ng L )

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

E2 removal (% h )

3376

35 25 15 10 8 6 4 2 0

0

100

200

300

400 -1

Exposure concentration (ng E2 L ) Fig. 1. Salmo trutta. (A): Concentrations of E2 in the exposure tanks during exposures. C: Exp. 1. -: Exp. 2. ;: Exp. 3. :: Exp. 4. ,: Exp. 5. A: Exp 6. Numbers refer to number of replicate tanks. Mean  SEM are given (although most SEMs are smaller than the symbols). (B): E2 removed in each tank as a function of the fish biomass in the tank. (C): E2-uptake into the fish calculated from removal of E2 from the water phase. Filled symbols: Average weight of the fish between 24 and 74 g. Open symbols: Fish weight: 1.8 g.

E2 concentrations were below the detection limit in water samples from the storage and control tanks. Table 2 Primer sequences, accession number and properties for isoform specific primers for S. Salar. Name Forward primer 50 e30

Reverse primer 50 e30

VTG

AGA AGG AAG CAC CCA AY049952 89 GGA AT TGA GCA AGA TGA TGG X89959 119 CTT TG GCA CCC AGG CAT ACT AF321836 71 TGA AAG

ATG GCG AAG TCA GAC AGG AG ERa GGA TGT GTG GAG GGT ATG G EF-1a GAG AAC CAT TGA CAA CTT CGA GAA G

Accession Amplicon number for (bp) Salmo salar

3.2. Estrogenic effects Exposure to E2 concentrations between 24 and 182 ng L
1 for 3 h (experiment 1) did not result in elevated vitellogenin levels in plasma samples taken 3 and 6 d after the pulse exposure (results not shown). Likewise, 6 h exposure to average concentrations of 156  11 ng E2 L
1 (experiment 2) did not affect plasma vitellogenin levels 2, 4, 6 and 10 d after the exposure (results not shown). Exposure to 370 ng E2 L
1 for 6 h (experiment 3) increased the plasma vitellogenin levels 2, 4 and 6 days after the exposure

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

10

5

10

4

10

3

10

2

Males only

A

B *

*

-1

Plasma vitellogenin (ng ml )

10

3

10

2

*** 10

4

1

4

A

4

1

**

***

4

Days after exposure 10

3

Fig. 3. Salmo trutta. Vitellogenin concentrations in plasma of brown trout exposed to 0 (,) or 159 (-) ng E2 L
1 for 6 h. Mean  SEM for 7e10 fish (\ þ _) in (A) and 5e7 fish (_) in the insert (B). * indicates difference from control at 0.05 level.

100 3

10

B

***

100 10

***

4

-1

10 1 0.1

n.a.

-2

n.a.

10 100

Liver α - receptor- mRNA

10

Whole body vitellogenin (ng g )

Liver vitellogenin-mRNA

Plasma vitellogenin -1 (ng ml )

(Fig. 2A), whereas only a trend was seen at day 1. Although still elevated relative to the control group, plasma vitellogenin levels in the exposed group decreased significantly (p < 0.0016) from day 4 to day 6.Exposure to the 6-h pulse of 370 ng E2 L
1 induced a 7-fold increase in ERa- (p < 0.0001) one day post exposure (Fig. 2B). ERa-mRNA levels still tended to be elevated on day 6 (p ¼ 0.0501). Vitellogenin-mRNA was elevated 6000-fold on day 1 (p ¼ 0.0004), and still significantly higher than control values on day 6 (p ¼ 0.0003; Fig. 2C). In experiment 4, 6 h exposure to 159 ng E2 L
1 only caused a trend (p ¼ 0.09) of augmented plasma vitellogenin levels (Fig. 3A) 1 d after the exposure, when all the fish were considered. If only the male fish in this experiment were considered (Fig. 3B), the plasma vitellogenin levels were significantly elevated on both day 1 (p ¼ 0.024) and 4 (p ¼ 0.034). Exposure to two 6 h pulses of 206 ng E2 L
1 with a 48 h interval (experiment 5) caused a significant (p < 0.001) increase in whole body vitellogenin level four days after the second exposure (Fig. 4). Only a trend (p ¼ 0.09) of increase was observed 4 days after the first pulse.

3377

C ***

10

1

n.a.

n.a.

2

4

*** 10

3

100

10

1

6

Days after exposure Fig. 2. Salmo trutta. Vitellogenin concentrations in plasma (A) and relative levels of vitellogenin- (B) and ERa-receptor (C) mRNA in liver of brown trout exposed to 0 (,) or 370 (-) ng E2 L
1 for 6 h ** or *** indicate difference from control at 0.01 or 0.001 level, respectively. n.a.: not analysed. Mean  SEM for 8e11 fish.

1

2

Pulse number Fig. 4. Salmo trutta. Vitellogenin concentrations in whole body brown trout 4 days after exposure to one or two 6-h pulses of 0 (,) or 206 (-) ng E2 L
1. Mean  SEM for 10 fish. The two pulses were given with a 48 h interval. *** indicates difference from control at 0.001 level.

3378

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

Exposure to three 6 h pulses of 64 ng E2 L
1 with 48 h intervals (experiment 6) did not affect plasma vitellogenin levels 4, 6 and 8 days after the first pulse (results not shown). 4. Discussion 4.1. E2-uptake In all of the exposure experiments, concentrations of E2 in the water were gradually reduced after addition of fish to the tanks. This reduction is most likely caused by uptake of E2 into the fish as observed in experiments with rainbow trout exposed to di-2-ethylhexyl phthalate (DEHP) (Tarr et al., 1990) and the herbicide trifluralin (Schultz and Hayton, 1994). In experiments with rainbow trout and propylparaben and butylparaben in the same exposure system as used in the present experiments, concentrations also decreased when the fish were added to the tanks (Bjerregaard et al., 2003; Alslev et al., 2005). The fact that concentrations increased to the original values in the flow-through systems after the fish had been removed indicates that removal of the chemicals by the fish rather than microbial degradation caused the decrease in the concentrations. Likewise, at 8 d exposure to E2 in the same flow-through system concentrations were 65  14 ng E2 L
1 with fish in the tanks and 134  2 ng E2 L
1 before fish were added and after they had been removed (Bjerregaard et al., 2008). The likelihood that the E2 removed from the tank water is actually taken up into the fish via the gills is also supported by the investigation by Maunder et al. (2007) who measured a very rapid uptake of E2 into the blood of sticklebacks Gasterosteus aculeatus during 6 h of exposure (from ambient water concentrations between 500 and 700 ng E2 L
1). Uptake rates for chemicals in fish from the water phase depend on the n-octanol:water partitioning coefficient (KOW) (Neely, 1979) for the chemicals such that uptake rates increase up to approximately Log KOW ¼ 6 and decrease for higher values (Mckim et al., 1985). In investigations in the same flow-through system as used in the present investigation (only utilising slightly larger rainbow trout and 8e11 days’ exposure), uptake rates for propylparaben (Bjerregaard et al., 2003) and butylparaben (Alslev et al., 2005) in rainbow trout were 6.8 h
1 and 16.2 h
1, respectively, and uptake rates for E2 brown trout was 21.5 h
1 (calculated from Bjerregaard et al., 2008). Propylparaben, butylparaben and E2 have log KOW values of 3.04, 3.57 and 3.9, respectively, and in this KOW range the uptake rates are linearly correlated (Fig. 5) with the KOW values, corroborating previous results (Neely, 1979; Mckim et al., 1985; McKim and Erickson, 1991).

-1

U p take rate (h )

25 20 15 10 5 0

3

3.2

3.4

3.6

3.8

4

Log Kow Fig. 5. Relation between uptake rates in trout and octanol-water partition coefficient for propylparaben (:) (Bjerregaard et al., 2003), butylparaben (-) (Alslev et al., 2005) and E2 (C) (Bjerregaard et al., 2008) over 8e11 days’ exposure. B: Present investigation.

Uptake rates for E2 in the present investigation were six-fold higher in smaller (1.8 g) than in larger (24e74 g) juvenile brown trout. Physiological processes generally scale with body size in double logarithmic plots with exponents often in the range 0.6e0.8; thus, the exponent for gill ventilation volumes in rainbow trout is 0.73 (Hughes, 1984). Uptake rates for synthetic chemicals from the water into fish are closely related to the gill ventilation volumes. Uptake of DDT in the mosquito fish Gambusia affinis scaled with body sizes with exponents of 0.75e0.77 (Murphy and Murphy, 1971) and in rainbow trout uptake of trifluralin and DEHP scaled with exponents of 0.66 (Schultz and Hayton, 1994) and 0.44 (Tarr et al., 1990), respectively. Although only two different size classes of brown trout were used in the present investigation, the scaling exponent for uptake of E2 (0.6) does not seem to deviate in any systematic way from the exponents for the synthetic chemicals. 4.2. Biomarker responses The threshold for induction of vitellogenin synthesis in brown trout after single 6-h exposures appears to be in the range between 150 and 200 ng E2 L
1 which is approximately one order of magnitude higher than the EC50-value after continuous exposure over 8 d (Bjerregaard et al., 2008). There were, however, quantitative differences in the magnitude and duration of the response both with respect to dosage and type of biomarker. A 370 ng E2 L
1 pulse induced 7-fold increase in liver ERamRNA whereas a 6000-fold increase in vitellogenin-mRNA was observed one day after exposure. This was followed by a roughly 100-fold increase in the plasma vitellogenin level peaking on day 4 after exposure. By comparison, a 6-h pulse of 200 ng E2 L
1 only induced a 5-fold increase in plasma vitellogenin on day 1, which was cleared on day 4 after exposure. A dose dependent vitellogenin response has also been reported in previous injections experiments with E2 and EE2 in rainbow trout (Purdom et al., 1994; Sherry et al., 1999). A higher fold induction of the vitellogenin-mRNA level relative to the ER-mRNA level was also observed in rainbow trout (Oncorhynchus mykiss) injected with E2 (Pakdel et al., 1991; Arukwe et al., 2001). Yamaguchi et al. (2005) on the other hand found a more similar induction of the two mRNAs in male medaka (Oryzias latipes) exposed to waterborne E2. Even though stimulation of the vitellogenin-mRNA occurs relatively rapidly it requires transcription of the ER gene (Arukwe et al., 2001; Yamaguchi et al., 2005). This conforms with the presence of an estrogenresponsive element (ERE) in the promotor of the rainbow trout (O. mykiss) ERa gene (Ledrean et al., 1995). A fast increase in vitellogenin-mRNA (as in experiment 3 of the present investigation) was observed in E2-injection experiments in rainbow trout (Leguellec et al., 1988; Pakdel et al., 1991; Arukwe et al., 2001), fathead minnow (Pimephales promelas) (Korte et al., 2000) and sheepshead minnow (Cyprinodon variegatus) (Bowman et al., 2000). Furthermore, Yamaguchi et al. (2005) showed that 100e10,000 ng E2 L
1 pulses for 8 h induced vitellogenin-mRNA in the liver at the end of the experiment. The induction of plasma vitellogenin, however, takes more time to develop because it takes time to build protein. In the present experiment 3, a significant elevation was seen on day two, whereas previous E2 pulse experiments have shown induction of plasma vitellogenin in less than 24 h (Bowman et al., 2000; Korte et al., 2000), on the second day (Arukwe et al., 2001) and on the third day (Leguellec et al., 1988) after exposure. Differences in fish species, gender, size, metabolic status and water temperature may contribute to these differences in vitellogenin time course. In experiment 5, whole body vitellogenin was analysed due to the fish size being too small for blood sampling. This experiment showed that two 6-h pulses were more effective in inducing

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

a vitellogenin response on day 4 than a single pulse. This so-called “memory effect” has previously been observed by Bowman et al. (2000), Leguellec et al. (1988) and Pakdel et al. (1991) in sheepshead minnow and rainbow trout given repeated E2-injections. A faster induction and/or a higher vitellogenin-mRNA or circulating vitellogenin level was obtained in response to the second injection (Leguellec et al., 1988; Pakdel et al., 1991; Bowman et al., 2000). Bowman et al. (2000) suggested that stabilisation and increased half-life of vitellogenin-mRNA was induced by E2. In addition, Panter et al. (2000) showed that intermittent and continuous waterborne E2 exposure of male fathead minnow were equipotent with regard to elevation of plasma vitellogenin. Intermittent exposure actually resulted in a higher effect than continuous exposure to the equivalent time-weighted average concentration (Panter et al., 2000). Due to this memory effect, repeated or interrupted environmental pulse-exposures may lead to a lower threshold concentration and a stronger estrogenic response. The different parameters assayed may all be used as molecular biomarkers of estrogenic activity. However, they show different sensitivity and time course development in response to an estrogenic pulse. This is as expected, partly due to the different roles in the vitellogenic pathway and due to the time it takes to build protein compared to mRNA. Vitellogenin-mRNA seems by far to be a more sensitive and longer lasting biomarker than ER-mRNA. On the other hand vitellogenin plasma levels may be more robustly changed with a longer half life than mRNA levels after estrogenic exposure. This was also the case in the rainbow trout, fathead minnow and sheepshead minnow (Leguellec et al., 1988; Pakdel et al., 1991; Bowman et al., 2000). In one of the experiments (Fig. 3) statistically significant responses were seen in male, but not in female juvenile brown trout. In a field survey using one to three year old brown trout, females had higher median plasma vitellogenin levels than males (Bjerregaard et al., 2008). Since differences in vitellogenin levels may exist between even sexually immature male and female brown trout it is very important e in field surveys as well as in laboratory experiments e to keep track of the sex of the fish used in the investigations. Although very small (w1.8 g) brown trout take up E2 from the water approximately 6 times faster than larger (w50 g) trout there is no evidence from the data that the higher uptake in the smaller fish is also reflected in a higher sensitivity to the E2 exposure (Figs. 3 and 4). The present investigation was not, however, specifically designed to address this question, which may have some interest since some of the phenomena related to endocrine disruption (Sumpter, 2005) in the environment are expected to be caused by exposure of the early (hence small) life stages of various fish species. Acknowledgements This investigation was supported by grants from the Danish Natural Science Research Council. The authors thank Bente Frost Jacobsen for skilful assistance with the LC-MS analyses. References Alslev, B., Korsgaard, B., Bjerregaard, P., 2005. Estrogenicity of butylparaben in rainbow trout Oncorhynchus mykiss exposed via food and water. Aquatic Toxicology 72, 295e304. Arukwe, A., Kullman, S.W., Hinton, D.E., 2001. Differential biomarker gene and protein expressions in nonylphenol and estradiol-17 beta treated juvenile rainbow trout (Oncorhynchus mykiss). Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 129, 1e10. Bjerregaard, L.B., Korsgaard, B., Bjerregaard, P., 2006. Intersex in wild roach (Rutilis rutilis) from Danish sewage effluent-receiving streams. Ecotoxicology and Environmental Safety 64, 321e328.

3379

Bjerregaard, P., Andersen, D.N., Pedersen, K.L., Pedersen, S.N., Korsgaard, B., 2003. Estrogenic effect of propylparaben (propylhydroxybenzoate) in rainbow trout Oncorhynchus mykiss after exposure via food and water. Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 136, 309e317. Bjerregaard, P., Hansen, P., Larsen, K.J., Erratico, C., Korsgaard, B., Holbech, H., 2008. Vitellogenin as a biomarker for oestrogenic effects in brown trout, Salmo trutta: laboratory and field investigations. Environmental Toxicology and Chemistry 27, 2387e2396. Bowman, C.J., Kroll, K.J., Hemmer, M.J., Folmar, L.C., Denslow, N.D., 2000. Estrogeninduced vitellogenin mRNA and protein in sheepshead minnow (Cyprinodon variegatus). General and Comparative Endocrinology 120, 300e313. Burki, R., Vermeirssen, E.L.M., Korner, O., Joris, C., Burkhardt-Holm, P., Segner, H., 2006. Assessment of estrogenic exposure in brown trout (Salmo trutta) in a Swiss midland river: integrated analysis of passive samplers, wild and caged fish, and vitellogenin mRNA and protein. Environmental Toxicology and Chemistry 25, 2077e2086. Desbrow, C., Routledge, E.J., Brighty, G.C., Sumpter, J.P., Waldock, M., 1998. Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environmental Science & Technology 32, 1549e1558. Desforges, J.P.W., Peachey, B.D.L., Sanderson, P.M., White, P.A., Blais, J.M., 2010. Plasma vitellogenin in male teleost fish from 43 rivers worldwide is correlated with upstream human population size. Environmental Pollution 158, 3279e3284. Donohoe, R.M., Curtis, L.R., 1996. Estrogenic activity of chlordecone, o, p0 -DDT and o, p0 -DDE in juvenile rainbow trout: induction of vitellogenesis and interaction with hepatic estrogen binding sites. Aquatic Toxicology 36, 31e52. Dyer, R.A., Raman, D.R., Mullen, M.D., Burns, R.T., Moody, L.B., Layton, A.C., Sayler, G.S., 2001. Determination of 17b-estradiol concentrations in runoff from plots receiving dairy manure. The Society for Engineering in Agricultural, Food, and Biological Systems 1, 2107. Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, M., Sumpter, J.P., Taylor, T., Zaman, N., 1997. Estrogenic activity in five United Kingdom rivers detected by measurement of vitellogenesis in caged male trout. Environmental Toxicology and Chemistry 16, 534e542. Hildebrand, C., Londry, K.L., Farenhorst, A., 2006. Sorption and desorption of three endocrine disrupters in soils. Journal of Environmental Science and Health Part B-Pesticides Food Contaminants and Agricultural Wastes 41, 907e921. Hughes, G.M., 1984. General anatomy of the gills. In: Hoar, W.S., Randall, D.J. (Eds.), Fish Physiology. Gills. Part A, vol. X. Anatomy, Gas Transfer, and Acid-Base Regulation Academic Press, Orlando, pp. 1e72. Kelly, M.A., Reid, A.M., Quinn-Hosey, K.M., Fogarty, A.M., Roche, J.J., Brougham, C.A., 2010. Investigation of the estrogenic risk to feral male brown trout (Salmo trutta) in the Shannon International River Basin District of Ireland. Ecotoxicology and Environmental Safety 73, 1658e1665. Kjaer, J., Olsen, P., Bach, K., Barlebo, H.C., Ingerslev, F., Hansen, M., Sorensen, B.H., 2007. Leaching of estrogenic hormones from manure-treated structured soils. Environmental Science & Technology 41, 3911e3917. Korner, O., Vermeirssen, E.L.M., Burkhardt-Holm, P., 2005. Intersex in feral brown trout from Swiss midland rivers. Journal of Fish Biology 67, 1734e1740. Korte, J.J., Kahl, M.D., Jensen, K.M., Pasha, M.S., Parks, L.G., LeBlanc, G.A., Ankley, G.T., 2000. Fathead minnow vitellogenin: complementary DNA sequence and messenger RNA and protein expression after 17 beta-estradiol treatment. Environmental Toxicology and Chemistry 19, 972e981. Ledrean, Y., Kern, L., Pakdel, F., Valotaire, Y., 1995. Rainbow trout estrogenreceptor presents an equal specificity but a differential sensitivity for estrogens than human estrogen-receptor. Molecular and Cellular Endocrinology 109, 27e35. Leguellec, K., Lawless, K., Valotaire, Y., Kress, M., Tenniswood, M., 1988. Vitellogenin gene-expression in male rainbow trout (Salmo gairdneri). General and Comparative Endocrinology 71, 359e371. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)(-Delta Delta C) method. Methods 25, 402e408. Matthiessen, P., Arnold, D., Johnson, A.C., Pepper, T.J., Pottinger, T.G., Pulman, K.G.T., 2006. Contamination of headwater streams in the United Kingdom by oestrogenic hormones from livestock farms. Science of the Total Environment 367, 616e630. Maunder, R.J., Matthiessen, P., Sumpter, J.P., Pottinger, T.G., 2007. Rapid bioconcentration of steroids in the plasma of the three-spined stickleback Gasterosteus aculeatus exposed to waterborne testosterone and 17b-oestradiol. Journal of Fish Biology 70, 678e690. Mckim, J., Schmieder, P., Veith, G., 1985. Absorption dynamics of organic chemical transport across trout gills as related to octanol water partition-coefficient. Toxicology and Applied Pharmacology 77, 1e10. McKim, J.M., Erickson, R.J., 1991. Environmental impacts on the physiological mechanisms controlling xenobiotic transfer across fish gills. Physiological Zoology 64, 39e67. Murphy, P.G., Murphy, J.D., 1971. Correlations between respiration and direct uptake of DDT in the mosquito fish Gambusia affinis. Bulletin of Environmental Contamination and Toxicology 6, 581e588. Neely, W.B., 1979. Estimating rate constants for the uptake and clearance of chemicals by fish. Environmental Science & Technology 13, 1506e1510. Pakdel, F., Feon, S., Legac, F., LeMenn, F., Valotaire, Y., 1991. In vivo estrogen induction of hepatic estrogen-receptor messenger-RNA and correlation with

3380

J.J.G. Knudsen et al. / Environmental Pollution 159 (2011) 3374e3380

vitellogenin messenger-RNA in rainbow trout. Molecular and Cellular Endocrinology 75, 205e212. Panter, G.H., Thompson, R.S., Sumpter, J.P., 2000. Intermittent exposure of fish to estradiol. Environmental Science & Technology 34, 2756e2760. Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, N.C., Tyler, C.R., Sumpter, J.P., 1994. Estrogenic effects of effluents from sewage treatment works. Chemistry and Ecology 8, 275e285. Rose, J., Holbech, H., Lindholst, C., Norum, U., Povlsen, A., Korsgaard, B., Bjerregaard, P., 2002. Vitellogenin induction by 17 beta-estradiol and 17 alphaethinylestradiol in male zebrafish (Danio rerio). Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 131, 531e539. Schultz, I.R., Hayton, W.L., 1994. Body size and the toxicokinetics of Trifluralin in rainbow trout. Toxicology and Applied Pharmacology 129, 138e145. Sherry, J., Gamble, A., Fielden, M., Hodson, P., Burnison, B., Solomon, K., 1999. An ELISA for brown trout (Salmo trutta) vitellogenin and its use in bioassays for environmental estrogens. Science of the Total Environment 225, 13e31. Stuer-Lauridsen, F., Kjølholt, J., Høibye, L., Hinge-Christensen, S., Ingerslev, F., Hansen, M., Krogh, K.A., Andersen, H.R., Halling-Sørensen, B., Hansen, N., Køppen, B., Bjerregaard, P., Frost, B., 2005. In: Environmental Project, no. 977 (Ed.), Survey of Estrogenic Activity in the Danish Aquatic Environment. Danish Environmental Protection Agency, Copenhagen, Denmark, pp. 1e170.

Stuer-Lauridsen, F., Kjølholt, J., Høibye, L., Hinge-Christensen, S., Ingerslev, F., Hansen, M., Krogh, K.A., Andersen, H.R., Halling-Sørensen, B., Hansen, N., Køppen, B., Bjerregaard, P., Frost, B., 2006. In: Environmental Project, no. 1077 (Ed.), Survey of Estrogenic Activity in the Danish Aquatic Environment. Part B. Danish Environmental Protection Agency, Copenhagen, Denmark, pp. 1e49. Sumpter, J.P., 2005. Endocrine disrupters in the aquatic environment: an overview. Acta Hydrochimica et Hydrobiologica 33, 9e16. Sumpter, J.P., Jobling, S., 1995. Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environmental Health Perspectives 103, 173e178. Tarr, B.D., Barron, M.G., Hayton, W.L., 1990. Effect of body size on the uptake and bioconcentration of di-2-ethylhexyl phthalate in rainbow trout. Environmental Toxicology and Chemistry 9, 989e995. Vermeirssen, E.L.M., Burki, R., Joris, C., Peter, A., Segner, H., Suter, M.J.F., BurkhardtHolm, P., 2005. Characterization of the estrogenicity of swiss midland rivers using a recombinant yeast bioassay and plasma vitellogenin concentrations in feral male brown trout. Environmental Toxicology and Chemistry 24, 2226e2233. Yamaguchi, A., Ishibashi, H., Kohra, S., Arizono, K., Tominaga, N., 2005. Short-term effects of endocrine-disrupting chemicals on the expression of estrogenresponsive genes in male medaka (Oryzias latipes). Aquatic Toxicology 72, 239e249.