Tissue-specific suppression of estrogen, androgen and glucocorticoid receptor gene expression in feral vitellogenic male Mozambique tilapia

Chemosphere 69 (2007) 32–40 www.elsevier.com/locate/chemosphere

Tissue-specific suppression of estrogen, androgen and glucocorticoid receptor gene expression in feral vitellogenic male Mozambique tilapia Chang-Beom Park a, Akihiro Takemura b, Neelakanteswar Aluru c, Yong-Ju Park b, Byung-Ho Kim d, Chi-Hoon Lee a, Young-Don Lee a, Thomas W. Moon e, Mathilakath M. Vijayan c,* b

a Marine and Environmental Research Institute, Cheju National University, Jeju 695-814, South Korea Sesoko Station, Tropical Biosphere Research Center, University of the Ryukyus, 3422 Sesoko, Motobu, Okinawa 905-0227, Japan c Department of Biology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 d Jeju Hi-Tech Industry Development Institute, Jeju 690-121, South Korea e Department of Biology, University of Ottawa, Ottawa, Ontario, Canada K1N 6N5

Received 27 November 2006; received in revised form 20 April 2007; accepted 25 April 2007 Available online 12 June 2007

Abstract While vitellogenesis in male fish is commonly used as a biomarker of xenoestrogen exposure, very little is known about the impacts associated with this unusual protein synthesis in feral populations. To this end, a recent study showed elevated circulating vitellogenin (VTG) levels in male Mozambique tilapia (Oreochromis mossambicus) collected from the Aja but not Tengan Rivers in Okinawa, Japan. Here we investigated whether this unusual protein synthesis in male fish from the Aja River affect transcript abundance of estrogen (ER), androgen (AR) and glucocorticoid (GR) receptors in the liver, brain and testis. The detection of plasma VTG levels (100 lg ml1) in male tilapia confirmed xenoestrogenic exposure in the Aja, but not the Tengan River. This protein induction was not associated with any changes in the reproductive capacity as assessed by sperm mobility and testis histology in the Aja fish. Plasma levels of estradiol-17b, 11ketotestosterone and cortisol were not significantly different between fish from the two rivers. Quantitative real-time PCR revealed a significant reduction in transcript levels of ERa and ERb, GR and ARa but not ARb, in the livers of tilapia from the Aja compared with the Tengan River. There were no significant changes in any of the steroid receptor transcript levels in either the brain or testis between the two rivers. Overall, our results imply that xenoestrogen exposure and VTG synthesis may lead to disruption of liver responsiveness to sex steroids and glucocorticoid stimulation in feral male fish.  2007 Elsevier Ltd. All rights reserved. Keywords: GR; ER; AR; VTG; Oreochromis mossambicus; Xenoestrogen

1. Introduction Vitellogenin (VTG) is the precursor of egg yolk protein in oviparous teleosts (Bun-Ng and Idler, 1983; Mommsen and Walsh, 1988; Lazier and MacKay, 1993; MacKay et al., 1996). It is synthesized in the liver of female fish in response to estradiol-17b (E2)-induced estrogen receptor *

Corresponding author. Tel.: +1 519 888 4567x32035. E-mail address: [email protected] (M.M. Vijayan).

0045-6535/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.04.072

(ER) activation (Lazier and MacKay, 1993; Jalabert, 2005). This well characterized mode of action of E2 in inducing VTG synthesis forms the basis for the use of this egg yolk precursor as an exposure biomarker for estrogen and/or estrogen-mimics in the aquatic environment (Denslow et al., 1999). Specifically, the presence of detectable levels of circulating VTG in male or immature fish is a good indication of exposure to chemicals that mimic ER activation (Sumpter, 1998; Kime, 1999; Jobling and Tyler, 2003). While most studies on endocrine disruption have

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

focused on measuring VTG levels as a biomarker of xenoestrogen exposure (Sumpter, 1998; Rotchell and Ostrander, 2003; Navas and Segner, 2006), very few studies have actually examined the impact of this unusual protein synthesis on other functions that may affect target tissue responsiveness in male fish. Steroid hormone receptors, including ER and the androgen (AR) and glucocorticoid (GR) receptors are key mediators of target tissue responsiveness to steroid hormone action. Indeed, changes in receptor content or function could adversely affect neuro/endocrine regulation of reproductive and metabolic homeostasis (Vijayan et al., 2005; Goksoyr, 2006; Tabb and Blumberg, 2006). For instance, reduced brain GR levels in response to polychlorinated biphenyl (PCB) exposure was shown to impair the functioning of the hypothalamus–pituitary–interrenal axis to stress in Arctic charr (Salvelinus alpinus; Aluru et al., 2004). Also, reduced GR levels in response to arsenate exposure depressed cortisol-induced glucose production in trout hepatocytes (Boone et al., 2002). Similarly, effects of xenobiotics on the reproductive axis are thought to be mediated via estrogen and androgen receptors (Goksoyr, 2006; Tabb and Blumberg, 2006). Indeed, endocrine disruption in feral fish populations is an issue of growing concern as the number of reports on reproductive abnormalities in fish increase (Jobling et al., 1998; Sumpter, 1998; Kime, 1999; Jobling and Tyler, 2003; Goksoyr, 2006). However, while the majority of studies rely on circulating VTG levels as a biomarker of exposure to xenoestrogens, there is a paucity of information on other endocrine markers, including receptor responses in feral fish. Against this backdrop, we tested the hypothesis that unusual VTG synthesis in feral male fish will suppress energy demanding pathways, including steroid responsiveness in Mozambique tilapia (Oreochromis mossambicus). To this end, ER, AR and GR gene expression was used as marker for steroid responsiveness as previous studies linked these receptor transcript abundance to steroid signaling (Vijayan et al., 2003; Sabo-Attwood et al., 2004). A previous study confirmed elevated circulating VTG levels in male tilapia only in the Aja but not Tengan Rivers in Okinawa, Japan (Ogasawara et al., 2000). This provided a natural setting to sample fish from two rivers with differing VTG levels. The Tengan River tilapia represents male fish from most rivers in Okinawa showing undetectable levels of VTG. Consequently, male tilapia were collected from these two rivers to assess plasma VTG and steroid hormone levels, reproductive capacity and mRNA abundance of ER, AR and GR in the liver, brain and testis of these fish.

33

the Tengan and Aja Rivers in July. The rivers are 15 km apart with similar characteristics, including strong tidal influence and only domestic sewage outfall entering the river. The difference in salinity between the sampling sites (Tengan: fresh water; Aja: brackish water) here simply reflect the distance from the Ocean and, indeed, the fish in both rivers move freely between fresh water and seawater. As a previous study did show that male tilapia in the Aja but not the Tengan Rivers were vitellogenic (Ogasawara et al., 2000), we used male fish from these two rivers to investigate the impact of VTG induction on plasma steroid levels and target tissue steroid receptor transcript abundance. Fish were transported to the Sesoko Station, Tropical Biosphere Research Center, University of the Ryukyus, and held separately in two fresh water tanks (500 l). The fish were maintained under natural light conditions with aeration for two days to allow recovery from the stress of capture and transport prior to sampling. 2.2. Sampling Groups of eight male fish were quickly sampled by netting the fish from the tank and anesthetizing them with 2phenoxyethanol (1:1000; Kanto Chemicals, Tokyo, Japan). The length and weight of each fish were recorded to determine the condition factor (CF = (body weight in g times (total length3 in cm)1 · 100). Blood was sampled from the caudal vein using a heparinized syringe; the plasma was collected by centrifugation (9600g for 5 min) and frozen at 70 C for later analyses. Fish were dissected and testis weight was recorded to determine the gonadosomatic index (GSI = (testis weight times body weight1) · 100). Pieces of liver, testis and brain were frozen immediately for later determination of ER, AR and GR mRNA abundance. A small piece of testis was fixed in Bouin’s solution for histological examination. 2.3. Measurement of sperm motility

2. Materials and methods

Sperm motility was determined as described previously (Morita et al., 2003). Briefly, sperm was collected from the dissected testes by inserting a fine disposable transfer pipette (Iuchiseieido, Japan) into the sperm duct. Semen (approximately 106–108 sperm ml1) was immediately diluted in water (45 ll) on a glass slide using a fine glass capillary tube, and covered with a cover slip. Sperm movements were recorded using a video recorder (SLV-LF1; Sony, Japan) and a CCD camera (cs 226; Olympus, Japan) mounted on a phase contrast microscope (Optiphoto, Nikon) exactly as explained previously (Morita et al., 2003).

2.1. Experimental fish

2.4. Histological observation

Male tilapia were collected using a cast net from an area near to their spawning habitat (spawning nests visible) in

Fixed testis pieces were dehydrated in an ethanol series and embedded in histoparafin (m.p. 56–58 C, Histologie;

34

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

Merck, Darmstadt, Germany). The embedded tissue was sectioned at 5 lm thickness, and stained with hematoxylin-eosin for observation under a light microscope (Olympus BS50, Tokyo, Japan) and images captured with a digital camera (DP-50, Olympus, Tokyo, Japan).

Table 1 List of primers used for partial cloning of steroid receptors in Mozambique tilapia, Oreochromis mossambicus

2.5. Measurement of plasma vitellogenin

ERa

Name ERa

ARa

A sandwich ELISA (enzyme-linked immunosorbent assay) method was used for the determination of plasma VTG exactly as previously reported for this species (Takemura and Kim, 2001). Conjugation of horseradish peroxidase to the VTG antibody was carried out in accordance with the method of Nakane and Kawaoi (1974). The detection limit of the VTG assay is 3.2 ng ml1. 2.6. Measurement of plasma steroid hormones Plasma levels of estradiol-17b (E2), 11-ketotestosterone (11-KT) and cortisol were measured by ELISA according to established protocols for this species (Vijayan et al., 2001; Rahman et al., 2000). While cortisol was measured directly from diluted (20·) plasma, E2 and 11-KT were measured after solvent extraction of plasma. Briefly, plasma samples (100 ll) were extracted three times with 2 ml diethyl ether (Kanto Chemical) and vortexed for 30 s. An aliquot of the diethyl ether-steroid solution was transferred to an assay tube, the solvent evaporated and the steroids reconstituted in 200 ll 50 mmol/l borate buffer (pH 7.8) containing 1% bovine serum albumin (assay buffer). 2.7. RNA extraction and cDNA synthesis Total RNA was extracted using TRI-reagent (Molecular Research Center Inc., Cincinnati, OH, USA) following the manufacturer’s instructions. RNA was DNase treated and the concentration of RNA was determined spectrophotometrically at 260 nm wavelength using a NanoDrop ND-1000 UV–VIS Spectrophotometer (Wilmington, DE, USA). First-strand cDNA was synthesized with 1 lg total RNA using ImProm-IITM Reverse Transcription System (Promega, Madison, WI, USA) according to the manufacturer’s manual. Briefly, total RNA was heat denatured (70 C) and cooled on ice. The sample was used in a 20 ll reverse transcriptase reaction using 0.5 lg oligo d(T) primers and 1 mmol/l each dNTP, 20 U ribonuclease inhibitors, and 40 U M-MuLV reverse transcriptase. The reaction was incubated at 42 C for 1 h and stopped by heating at 72 C for 15 min. After cDNA synthesis, the reaction mixture was diluted to a final volume of 100 ll by adding 80 ll nuclease-free water.

ARa GR

Primer sequences Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

5 0 -ATCAGCCATTTGGGATGCTC-3 0 5 0 -TGCCTTGAGGTCCTGAACTG-3 0 5 0 -GCCTCAGTTCCTCAGACTGC-3 0 5 0 -TCCAACATCTCCAGCAACAG-3 0 5 0 -GTCCCTGCTCAGCATCCTAC-3 0 5 0 -TCATACATGCTGGACGCTTG-3 0 5 0 -AACTCCCACCTGGTCTTCCT-3 0 5 0 -CAAGGTCTGGGGCAAAGTAA-3 0 5 0 -GCAGGTTTCCCACGATGAGTAC-3 0 5 0 -CCACGCTCAGGGATTTATTCAC-3 0

(AB045212), GR (D66874) and b-actin (AB037865) were amplified by polymerase chain reaction (PCR) with primer sets designed using tilapia sequences (Table 1). The PCR products were electrophoresed on a 1% agarose gel and the bands with appropriate sizes were excised from the gel and purified following the manufacturer’s protocol (Promega). The PCR products were cloned into the pGEM-T easy vector system (Promega) and then sequenced using the Thermo Sequence Fluorescent Labeled Primer Cycle Sequence kit with 7-deaza-dGTP in an ALF Express IITM DNA Sequencer System Version 2.1 (Amersham Biosciences, Piscataway, NJ). The amplified cDNA fragments were confirmed using the BLAST program (National Center for Biotechnology Information, NIH, USA). 2.9. Quantitative real-time polymerase chain reaction (qPCR) 2.9.1. Standard curve The primer sequences used for qPCR were designed using Primer Express V 2.0 software (ABI Prism, CA, USA) and are shown in Table 2. Primer sets were selected so that the amplified cDNA would span two exon/intron boundaries of each gene thus eliminating amplification of genomic DNA. The length of the amplicon was 100– 300 bp in length and the primer melting temperatures were Table 2 List of primers used for quantitative real-time polymerase chain reaction (qPCR) in Mozambique tilapia, Oreochromis mossambicus Name b-actin ERa ERb ARa

2.8. Cloning and sequencing of target genes and b-actin

ARa

Partial sequences of ERa (accession number AM 284390), ERb (AM 284391), ARa (AB045211), ARb

GR

Primer sequences Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

5 0 -TACCACCATGTACCCTGGCATC-3 0 5 0 -TACGCTCAGGTGGAGCAATGA-3 0 5 0 -GGCTCAGCAGCAGTCAAGAA-3 0 5 0 -TGCCTTGAGGTCCTGAACTG-3 0 5 0 -ACCTTCCGGCAGCAGTACAC-3 0 5 0 -TCCAACATCTCCAGCAACAG-3 0 5 0 -GTCCCTGCTCAGCATCCTAC-3 0 5 0 -TCACTCCCATCCATGACAGC-3 0 5 0 -CAGCCTCAATGAATTGGGAGA-3 0 5 0 -ATCCCAAGGCAAACACCATC-3 0 5 0 -GCAGGTTTCCCACGATGAGTAC-3 0 5 0 -CCACGCTCAGGGATTTATTCAC-3 0

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

2.9.2. Quantification The PCR reaction components and conditions were exactly as noted in the previous section and for every test sample qPCR for both target and housekeeping genes was performed. The following PCR cycle was used for gene amplification: initial denaturation at 95 C for 1 min; 40 cycles: denaturing at 95 C for 5 s, annealing and extension at 60 C for 1 min. To ensure the specificity of the PCR amplicons, a dissociation curve analysis was carried out by raising the temperature of the sample slowly from 60 to 95 C until the last step of the PCR reaction (Park

a Vitellogenin (μg . ml-1)

200

150

**

100

50

et al., 2007). These curves confirmed a single amplified product and the absence of primer–dimer formation (data not shown). The transcript levels were obtained from their respective standard curves and were normalized to b-actin (housekeeping gene). The threshold cycles (Ct) for b-actin did not show significant differences between the two rivers and hence used for normalization. 2.10. Statistical analysis Data are shown as mean ± standard error of the mean (SEM). Logarithmic transformation of the data set was carried out wherever necessary to meet the assumptions of homogeneity of variance prior to statistical analyses. Data comparisons between the two rivers were carried out using unpaired Student’s t-test (SPSS version 13.0; SPSS Inc., Chicago, USA). Statistical significance was set at p < 0.05. 3. Results 3.1. Plasma VTG and hormone levels Plasma VTG levels were below the assay detection limit (63.2 ng ml1) in the Tengan River fish whereas the Aja River fish had levels of circulating VTG of 100 lg ml1 (Fig. 1a). There were no significant differences in plasma E2, 11-KT or cortisol concentrations in fish from the two rivers (Figs. 1b–d).

b

300

17β-estradiol (pg . ml-1)

set to 60 C. Standard curves for the genes were constructed using serial dilutions of plasmid vector stock with inserted target sequences. A series of 10-fold dilutions of the plasmid DNA encoding for the target genes and b-actin were prepared and included in each amplification reaction to generate a standard curve as described previously (Park et al., 2007). Copy numbers for each plasmid DNA were calculated as (DNA concentration, ll lg1) · 106 pg lg1 · 1 pmol(660 pg · plasmid size)1 · Avogadro number · 1012. qPCR was performed using an ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Every PCR reaction was performed in triplicate in a final volume of 20 ll with 2 · SYBR premix Ex Taq (Takara), 0.2 lmol/l each of forward and reverse primers, and 4 ll fivefold diluted plasmid DNA (standard) as template.

250

150 100 50

Aja

3000

d

2500

Cortisol (ng. ml-1 )

11-ketotestosterone (pg . ml-1)

Tengan

2000 1500 1000 500 0

200

0

0

c

35

Tengan

Aja

Tengan

Aja

Tengan

Aja

25 20 15 10 5 0

Fig. 1. Plasma vitellogenin (a), estradiol-17b (b), 11-ketotestosterone (c) and cortisol (d) levels in male mozambique tilapia (Oreochromis mossambicus) from the Aja and Tengan Rivers. All values represent mean + SEM (n = 68). Plasma vitellogenin levels in Tengan River fish were below the assay detection limit, but the minimum detection limit of 3.2 ng ml1 was used for statistical analyses. * denotes significant differences between the two rivers (Student’s t-test, p < 0.05).

36

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

3.2. Morphological and reproductive parameters

3.3. Estrogen receptor mRNA abundance

Total length and body mass of tilapia from the Aja River was significantly greater than the Tengan River fish. However, no significant difference in the condition factor was observed between the two rivers (Table 3). Gonadosomatic index and sperm motility were higher in fish from the Aja compared with the Tengan Rivers (Table 3). Histological observation of the testes showed that the Aja River fish contained few spermatozoa and appeared to be in the partially spent to spent stage of testes development (Figs. 2C and D). The Tengan River fish were in the early maturing stage of gonadal development as seen by the large numbers of spermatozoa and spermatids (Figs. 2A and B).

The transcript abundance of both ERa and ERb isoforms was several-fold higher in tilapia liver relative to the brain and testis (Figs. 3a and b). In the liver, both ER isoforms were significantly lower in the Aja fish compared with the Tengan fish (Figs. 3a and b). There were no significant differences in ER isoform transcript abundance in either the brain or testis between the two rivers (Figs. 3a and b). 3.4. Androgen receptor mRNA abundance As seen with ERs, the AR isoform transcript abundance was also several-fold greater in the liver compared with the

Table 3 Total length, body mass, gonadosomatic index (GSI), sperm motility and condition factor (CF) of Mozambique tilapia, Oreochromis mossambicus, from the two rivers River

Total length (cm)

Body mass (g)

GSI

Sperm motility (%)

CF

Tengan Aja

22.9 ± 1.72 35.6 ± 0.34*

213.8 ± 27.3 735.0 ± 31.2*

0.09 ± 0.03 0.34 ± 0.05*

32.6 ± 11.1 88.7 ± 3.6*

1.82 ± 0.2 1.91 ± 0.1

Values represent means ± SEM (n = 6–8 fish). * Significantly different from the Tengan River (p < 0.05).

Fig. 2. Histological stage classification of testes in Mozambique tilapia, Oreochromis mossambicus, from the two rivers. (A and B) Two types from testes of Tengan River tilapia; (C and D) two types from testes of Aja River tilapia. (A) early mature stage (scale bar = 100 lm); (B) mature stage (scale bar = 100 lm); (C) partial spent stage (scale bar = 400 lm); (D) spent stage (scale bar = 400 lm). St, spermatid; Sz, spermatozoa.

C.-B. Park et al. / Chemosphere 69 (2007) 32–40 30

ERα/β-actin

20

a

Tengan Aja

100 75

Tengan Aja

50

*

10

ARα/β-actin

a

37

2

25

*

2

1 1 0 Testis

b

Liver

0

Brain

300

b

Testis

Liver

Testis

Liver

Brain

200 175

200

ARβ/β-actin

ERβ/β-actin

150 100 50

125 100

*

75

25

50 25 Testis

Liver

0

Brain

Fig. 3. Estrogen receptor isoform (ERa and ERb) transcript levels in various tissues of tilapia captured from the Aja and Tengan Rivers. All values represent mean + SEM (n = 6–8). * denotes significant difference in the liver ER isoform transcript levels between the two rivers (Student’s t-test, p < 0.05).

testis and brain. There were significantly lower ARa mRNA levels but not ARb mRNA levels in the liver of tilapia from the Aja River compared with the Tengan River (Figs. 4a and b). Neither testis nor brain showed any significant differences in AR gene expression between the two rivers (Fig. 4). 3.5. Glucocorticoid receptor mRNA abundance Overall, glucocorticoid receptor (GR) mRNA levels were twofold greater in tilapia liver compared with the brain and testis (Fig. 5). Similar to the ERs and ARs, the transcript abundance of GR was significantly lower in the liver of the Aja fish compared with the Tengan fish. There was no significant difference in GR receptor mRNA abundance in the brain and testis of the Aja fish compared with the Tengan fish (Fig. 5). 4. Discussion A novel finding from this study is the tissue-specific suppression of ER, AR and GR gene expression in the liver of male tilapia collected from the Aja compared with the Ten-

Brain

Fig. 4. Androgen receptor isoform (ARa and ARb) transcript levels in various tissues of tilapia from the Aja and Tengan Rivers. All values represent mean + SEM (n = 6–8). * represent significant difference in liver ARa transcript levels between the two rivers (Student’s t-test, p < 0.05).

100 Tengan Aja 80

GR/β-actin

0

60

40

20

* 0

Testis

Liver

Brain

Fig. 5. Glucocorticoid receptor (GR) transcript levels in various tissues of tilapia captured from the Aja and Tengan Rivers. All values represent mean + SEM (n = 6–8). * represent significant difference in liver GR transcript levels between the two rivers (Student’s t-test, p < 0.05).

gan Rivers. The detection of circulating VTG levels in male fish confirms the presence of xenoestrogens in the Aja River and agrees with earlier reports of the presence of chemicals inducing ER signaling in Okinawan rivers (Takemura and Sin, 1999; Ogasawara et al., 2000), includ-

38

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

ing the Aja River (Ogasawara et al., 2000). Indeed, the detection of VTG in male fish is a well established biomarker for xenoestrogen exposure in the aquatic environment (Sumpter, 1998; Tyler et al., 1998; Denslow et al., 1999; Kime, 1999; Jobling and Tyler, 2003; Garcia-Reyero et al., 2006). To our knowledge this is the first study to examine the expression of ER, AR and GR in feral male fish showing detectable VTG levels. While studies have examined the impact of xenoestrogens on ER gene expression (Yadetie et al., 1999; Arukwe, 2001; Luo et al., 2005; Greytak and Callard, 2007; Meucci and Arukwe, 2006; Vetillard and Bailhache, 2006), the majority of these studies focused on shorter-term laboratory exposures. The general indication from these previous studies is that ER gene expression is induced in an isoform- and tissue-specific manner by xenoestrogens (Meucci and Arukwe, 2006; Vetillard and Bailhache, 2006; Greytak and Callard, 2007) and is consistent with the autoregulation of ER by E2 observed in teleosts (Urushitani et al., 2003; Menuet et al., 2004; SaboAttwood et al., 2004). However, this does not seem to be the case in our study where liver ER isoform transcript abundance was lower in the Aja River fish despite higher VTG levels compared with the Tengan fish. This response was not unique to ER isoforms because liver ARa and GR transcript levels were also downregulated in fish from the Aja River. Similar to our results, a recent study with feral killifish collected from a xenoestrogenic site also showed a lower abundance of ERa in different tissues of reproductively active female fish (Greytak and Callard, 2007). However, unlike our study, the killifish had depressed plasma E2 levels which may have played a role in the lower ER abundance (Greytak and Callard, 2007). The lack of significant differences in plasma E2, 11-KT and cortisol levels in fish between the rivers (Figs. 1b–d) argues against a role for circulating steroids in modulating their receptor transcript abundance in the present study. Although the reason for tissue-specific suppression of steroid receptors, seen only in the liver but not testis or brain, is unclear, the results do point to a targeted disruption of liver steroid signaling in vitellogenic male fish. This is especially the case given the positive correlation between steroid receptor transcript levels and target tissue responsiveness to steroids in fish (Vijayan et al., 2003; Sabo-Attwood et al., 2004). Although the mechanism(s) involved in the suppression of liver steroid receptor gene expression is not known, we hypothesize that enhanced liver energy demand associated with VTG synthesis may be a factor for this tissue-specific response seen in male tilapia. Indeed liver is the primary target organ for xenoestrogen-induced VTG synthesis and this de novo protein synthesis is energy demanding (Mommsen and Walsh, 1988). It is proposed that nearly 80% of oxygen demand in trout hepatocytes is associated with protein synthesis (Pannevis and Houlihan, 1992; Mommsen, 1997). Consequently, the energy re-partitioning in the liver associated with xenoes-

trogen-induced protein synthesis may compromise other energy demanding pathways that are critical for homeostasis (Vijayan et al., 1997; Vijayan et al., 2001; Boone et al., 2002). For instance, we showed recently that unusual liver protein (VTG) synthesis associated with E2 stimulation in male tilapia reduced gill sodium pump activity (an energy demanding process) and compromised seawater acclimation in this euryhaline species (Vijayan et al., 2001). Also, higher heat shock protein (hsp) 70 and cytochrome P450 1 A expression in response to b-naphthoflavone exposure was associated with a depressed gluconeogenic capacity in rainbow trout liver (Vijayan et al., 1997). Similarly, arsenate-treated hepatocytes with elevated hsp70 content showed a reduction in cortisol-mediated glucose production (Boone et al., 2002). Taken together, the enhanced liver energy demand for synthesis of proteins encoded by ER-responsive genes in male tilapia in the Aja River may compromise other critical but energy demanding pathways, including steroid receptor synthesis. While this metabolic hypothesis appears to support the depressed liver steroid receptor transcript levels, we cannot discount the possibility that other factors, including other chemicals, may directly impact steroid receptor expression in the liver. However, that seems unlikely given that both the Aja and Tengan Rivers have very similar characteristics and only receives domestic sewage outfall. The Aja River, in addition receives effluents from a tofu plant and the occurrence of VTG in male tilapia only downstream from this plant, but not upstream suggests phytoestrogen exposure (Ogasawara et al., 2000). The lack of any fundamental changes in testicular morphology or sperm motility in this study supports the absence of reproductive impairment in response to this xenoestrogen exposure in the Aja River tilapia (Ogasawara et al., 2000; Seki et al., 2002). The higher GSI and sperm motility in the fish from the Aja compared with the Tengan River may simply be related to the reproductive status and behaviour of the fish, especially since this fish is a repeat spawner and the reproductive stages are not synchronized (Takemura et al., 1998). As is common to field sampling, we were unable to have a true size control for this study. However, as the condition factor remained the same between the two groups, it is probable that the larger size fish belonged to an older age class. Another factor that was different between the two sampling sites was the salinity of the medium. The Aja fish were collected from a brackish site, while the Tengan fish came from a fresh water site. However, both fish were transported back to the laboratory and allowed to recover in fresh water for two days prior to sampling. The recovery from transportation and handling stressors was evident from the lack of a cortisol response in the two groups coupled with the low estimated levels of this steroid, values that are well within unstressed levels reported for teleost fishes (Barton et al., 2002). However, it is well established that the VTG response in male fish is indeed due to xenoestrogenic stimulation and occurs independent of size and/or salinity of the medium. Consequently, it is unlikely that the

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

VTG levels seen only in the male Aja fish, but not the Tengan fish, was due to the size and/or medium salinity differences observed between the two groups. Altogether, our results highlight the potential for disruption of target tissue steroid signaling in feral vitellogenic male fish. In conclusion, our results for the first time demonstrate a tissue-specific suppression of ER, AR and GR transcript abundance in the liver of feral vitellogenic male tilapia. The observation that all three receptor changes were evident only in the liver, but not testis or brain of male tilapia, leads us to propose a metabolic hypothesis for the disruption of steroid signaling. Specifically, the unusual protein synthesis (VTG) in male fish liver, and the associated metabolic cost, disrupts other energy demanding pathways including steroid receptor expression in tilapia. As VTG levels in male fish are a biomarker of xenoestogenic contamination, our results supports the presence of phytoestrogens in the Aja River (Ogasawara et al., 2000). However, vitellogenesis did not affect the reproductive output of male tilapia in the Aja River. Overall, our results suggest that xenoestrogen exposure and VTG synthesis disrupts liver steroid receptor transcript abundance, which may lead to reduced liver responsiveness to sex steroids and glucocorticoid stimulation in male fish. Acknowledgements This study was supported by the 21st Century COE program ‘‘The Comprehensive Analyses on Biodiversity in Coral Reef and Island Ecosystems in Asian and Pacific Regions’’ from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and the Natural Sciences and Engineering Research Council (NSERC, Canada) Discovery Grant program. The technical assistance rendered by Dr. Morita and Ms. McGuire is gratefully acknowledged. This is a contribution from the International Summer Program 2005 of the 21st Century COE program. References Aluru, N., Jorgensen, E.H., Maule, A.G., Vijayan, M.M., 2004. PCB disruption of the hypothalamus–pituitary–interrenal axis involves brain glucocorticoid receptor downregulation in anadramous Arctic charr. Am. J. Physiol. Regul. Integr. Comp. Physiol. 287, R787–R793. Arukwe, A., 2001. Cellular and molecular responses to endocrinemodulators and the impact on fish reproduction. Mar. Pollut. Bull. 42, 643–655. Barton, B.A., Morgan, J.D., Vijayan, M.M., 2002. Physiological and condition-related indicators of environmental stress in fish. In: Adams, S.M. (Ed.), Biological Indicators of Stress in Fish, second ed. American Fisheries Society, Bethesda, MD, pp. 111–148. Boone, A.N., Ducouret, B., Vijayan, M.M., 2002. Glucocorticoid-induced glucose release is abolished in HSP70-accumulated trout hepatocytes. J. Endocrinol. 172, R1–R5. Bun-Ng, T., Idler, D.R., 1983. Yolk formation and differentiation in teleost fishes. In: Hoar, W.S.M., Randall, D.J. (Eds.), Fish Physiology, vol. IXa. Academic Press, Orlando, FL, pp. 373–397. Denslow, N.D., Chow, M.C., Kroll, K.J., Green, L., 1999. Vitellogenin as biomarker of exposure for estrogen or estrogen mimics. Ecotoxicology 8, 385–398.

39

Garcia-Reyero, N., Barber, D.S., Gross, T.S., Johnson, K.G., Sepulveda, M.S., Szabo, N.J., Denslow, N.D., 2006. Dietary exposure of largemouth bass to OCPs changes expression of genes important for reproduction. Aquat. Toxicol. 78, 358–369. Goksoyr, A., 2006. Endocrine disruptors in the marine environment: mechanisms of toxicity and their influence on reproductive processes in fish. J. Toxicol. Environ. Health A 69, 175–184. Greytak, S.R., Callard, G.V., 2007. Cloning of three estrogen receptors (ER) from killifish (Fundulus heteroclitus): differences in populations from polluted and reference environments. Gen. Comp. Endocrinol. 150, 174–188. Jalabert, B., 2005. Particularities of reproduction and oogenesis in teleost fish compared to mammals. Reprod. Nutr. Dev. 45, 261–279. Jobling, S., Tyler, C.R., 2003. Endocrine disruption, parasites and pollutants in wild freshwater fish. Parasitology 126, S103–S108. Jobling, S., Nolan, M., Tyler, C.R., Brighty, G., Sumpter, J.P., 1998. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32, 2498–2506. Kime, D.E., 1999. A strategy for assessing the effects of xenobiotics on fish reproduction. Sci. Total Environ. 225, 3–11. Lazier, C.B., MacKay, M.E., 1993. Vitellogenin gene expression in teleost fish. Mol. Biol. Frontiers 2, 391–405. Luo, Q., Ban, M., Ando, H., Kitahashi, T., Kumar Bhandari, R., McCormick, S.D., Urano, A., 2005. Distinct effects of 4-nonylphenol and estrogen-17 beta on expression of estrogen receptor alpha gene in smolting sockeye salmon. Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 140, 123–130. MacKay, M.E., Raelson, J., Lazier, C.B., 1996. Up-regulation of estrogen receptor mRNA and estrogen receptor activity by estradiol in liver of rainbow trout and other teleostean fish. Comp. Biochem. Physiol. C: Pharmacol. Toxicol. Endocrinol. 115, 201–209. Menuet, A., Le Page, Y., Torres, O., Kern, L., Kah, O., Pakdel, F., 2004. Analysis of the estrogen regulation of the zebrafish estrogen receptor (ER) reveals distinct effects of ERalpha, ERbeta1 and ERbeta2. J. Mol. Endocrinol. 32, 975–986. Meucci, V., Arukwe, A., 2006. The environmental estrogen, 4-nonylphenol modulates brain estrogen receptor and aromatase (CYP19) isoforms gene expression patterns in Atlantic salmon (Salmo salar). Mar. Environ. Res. 62, S195–S199. Mommsen, T.P., 1997. Growth and metabolism. In: Evans, D.H. (Ed.), The Physiology of Fishes, second ed. CRC Press, Boca Raton, FL, pp. 65–97. Mommsen, T.P., Walsh, P.J., 1988. Vitellogenesis oocyte assembly. In: Hoar, W.S., Randall, D.J. (Eds.), Fish Physiology, vol. IXa. Academic Press, San Diego, pp. 347–406. Morita, M., Takemura, A., Okuno, M., 2003. Requirement of Ca2+ on activation of sperm motility in euryhaline tilapia Oreochromis mossambicus. J. Exp. Biol. 206, 913–921. Nakane, P.K., Kawaoi, A., 1974. Peroxidase labelled antibody. A method of conjugation. J. Histochem. Cytochem. 22, 1084–1091. Navas, J.M., Segner, H., 2006. Vitellogenin synthesis in primary cultures of fish liver cells as endpoint for in vitro screening of the (anti)estrogenic activity of chemical substances. Aquat. Toxicol. 80, 1–22. Ogasawara, K., Takemura, A., Takano, K., 2000. Detection of a femalespecific protein, vitellogenin, in the plasma of male tilapias inhabiting Okinawan rivers. Biol. Mag. Okinawa 38, 1–9. Pannevis, M.C., Houlihan, D.F., 1992. The energetic cost of proteinsynthesis in isolated hepatocytes of rainbow-trout (Oncorhynchus mykiss). J. Comp. Physiol. B: Biochem. Systemic Environ. Physiol. 162, 393–400. Park, Y.J., Park, J.G., Hiyakawa, N., Lee, Y.D., Kim, S.J., Takemura, A., 2007. Diurnal and circadian regulation of a melatonin receptor, MT1, in the golden rabbitfish, Siganus guttatus. Gen. Comp. Endocrinol. 150, 253–262. Rahman, M.S., Takemura, A., Takano, K., 2000. Correlation between plasma steroid hormones and vitellogenin profiles and lunar periodicity in the female golden rabbitfish, Siganus guttatus (Bloch). Comp. Biochem. Pysiol. B: Biochem. Mol. Biol. 127, 113–122.

40

C.-B. Park et al. / Chemosphere 69 (2007) 32–40

Rotchell, J.M., Ostrander, G.K., 2003. Molecular markers of endocrine disruption in aquatic organisms. J. Toxicol. Environ. Health B: Crit. Rev. 6, 453–496. Sabo-Attwood, T., Kroll, K.J., Denslow, N.D., 2004. Differential expression of largemouth bass (Micropterus salmoides) estrogen receptor isotypes alpha, beta, and gamma by estradiol. Mol. Cell. Endocrinol. 218, 107–118. Seki, M., Yokota, H., Matsubara, H., Tsuruda, Y., Maeda, M., Tadokoro, H., Kobayashi, K., 2002. Effect of ethinylestradiol on the reproduction and induction of vitellogenin and testis–ova in medaka (Oryzias latipes). Environ. Toxicol. Chem. 21, 1692–1698. Sumpter, J.P., 1998. Feminized responses in fish to environmental estrogens. Toxicol. Lett. 82–83, 737–742. Tabb, M.M., Blumberg, B., 2006. New modes of action for endocrinedisrupting chemicals. Mol. Endocrinol. 20, 475–482. Takemura, A., Kim, B.H., 2001. Effects of estradiol-17b treatment on in vitro and in vivo synthesis of two distinct vitellogenins in tilapia. Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 129, 641–651. Takemura, A., Sin, D.-Ho., 1999. Detection of vitellogenin in the sera of male tilapia inhabiting the Okukubi River, using an enzyme linked immunosorbent assay (ELISA). Biol. Mag. Okinawa. 37, 1–7. Takemura, A., Sin, D-Ho., Takano, K., 1998. Annual reproductive cycle of female tilapia living in Okukubi river on the Okinawan island. Biol. Mag. Okinawa 36, 11–18. Tyler, C.R., Jobling, S., Sumpter, J.P., 1998. Endocrine disruption in wildlife: a critical review of the evidence. Crit. Rev. Toxicol. 28, 319– 361.

Urushitani, H., Nakai, M., Inanaga, H., Shimohigashi, Y., Shimizu, A., Katsu, Y., Iguchi, T., 2003. Cloning and characterization of estrogen receptor alpha in mummichog, Fundulus heteroclitus. Mol. Cell. Endocrinol. 203, 41–50. Vetillard, A., Bailhache, T., 2006. Effects of 4-n-nonylphenol and tamoxifen on salmon gonadotropin-releasing hormone, estrogen receptor, and vitellogenin gene expression in juvenile rainbow trout. Toxicol. Sci. 92, 537–544. Vijayan, M.M., Pereira, C., Grau, E.G., Iwama, G.K., 1997. Metabolic responses associated with confinement stress in tilapia: the role of cortisol. Comp. Biochem. Physiol. C: Pharmacol. Toxicol. Endocrinol. 116, 89–95. Vijayan, M.M., Takemura, A., Mommsen, T.P., 2001. Estradiol impairs hyposmoregulatory capacity in the euryhaline tilapia, Oreochromis mossambicus. Am. J. Physiol. Regul. Integr. Comp. Physiol. 281, R1161–R1168. Vijayan, M.M., Raptis, S., Sathiyaa, R., 2003. Cortisol treatment affects glucocorticoid receptor and glucocorticoid-responsive genes in the liver of rainbow trout. Gen. Comp. Endocrinol. 132, 256–263. Vijayan, M.M., Prunet, P., Boone, A.N., 2005. Xenobiotic impact on corticosteroid signaling. In: Mommsen, T.P., Moon, T.W. (Eds.), Biochemistry and Molecular Biology of Fishes, vol. 5. Elsevier, New York, pp. 365–394. Yadetie, F., Arukwe, A., Goksoyr, A., Male, R., 1999. Induction of hepatic estrogen receptor in juvenile Atlantic salmon in vivo by the environmental estrogen, 4-nonylphenol. Sci. Total Environ. 233, 201– 210.