- Email: [email protected]
Effects of copper on cortisol receptor and metallothionein expression in gills of Oncorhynchus mykiss
Aquatic Toxicology 51 (2000) 45 – 54 www.elsevier.com/locate/aquatox
Effects of copper on cortisol receptor and metallothionein expression in gills of Oncorhynchus mykiss Zhi Chao Dang a, Gert Flik a, Bernadette Ducouret b, Christer Hogstrand c, Sjoerd E. Wendelaar Bonga a, Robert A.C. Lock a,* b
a Department of Animal Physiology, Uni6ersity of Nijmegen, Toernooi6eld, 6525 ED Nijmegen, The Netherlands Endocrinologie Mole´culaire de la Reproduction, Uni6ersite´ de Rennes,, Campus de Beaulieu, 35042 Rennes Cedex, France c King’s College London, School of Health and Life Sciences, Di6ision of Life Sciences, Franklin – Wilkins Building, 150 Stamford Street, London SE1 8WA, UK
Received 3 December 1999; received in revised form 28 February 2000; accepted 1 March 2000
Abstract Effects of waterborne Cu (2.4 mM) on the expression of glucocorticoid receptor (GR) and metallothionein (MT) in the branchial epithelium of freshwater rainbow trout (Oncorhynchus mykiss) was studied by immunocytochemistry. After 5 days of Cu exposure, the number of GR-immunoreactive (GR-ir) cells in the gill epithelium had decreased, whereas the number of MT-ir cells had increased. Localization of GR in chloride cells was achieved by double staining for Na+/K+-ATPase; other cell types were identified on the basis of their topology. GRs were present in the chloride cells in both the filaments and lamellae, in respiratory cells in the lamellae, in pavement cells, basal layer cells and undifferentiated cells in the filaments. Co-localization of Na+/K+-ATPase and MT revealed that MT was expressed in chloride cells, both in filaments and lamellae. Occasionally, MT immunoreactivity was found in pavement cells and in undifferentiated cells. By double staining for Na+/K+-ATPase and GR, for Na+/K+-ATPase and MT and for GR and MT, we can conclude that after 5 days of Cu stress there are chloride cells that express GR and MT, GR or MT alone or neither of the two proteins. This apparent functional heterogeneity of branchial chloride cells may reflect a limited window when chloride cell subpopulations show an adaptive response to Cu. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Glucocorticoid receptor; Metallothionein; Fish gills; Chloride cell; Cu; Immunocytochemistry
1. Introduction
* Corresponding author. Tel.: +31-24-3652476; fax: + 3124-3652714. E-mail address: [email protected] (R.A.C. Lock).
The glucocorticoid receptor (GR) is an important mediator of cortisol signals. By using steroid binding assay GR expression has been detected in fish gills, liver, intestine, brain and muscle, with gill tissue having the highest cortisol-binding ac-
0166-445X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 0 ) 0 0 1 0 2 - 8
46
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
tivity (Chakraborti et al., 1987; Lee et al., 1992; Tujague et al., 1998). In rainbow trout Oncorhynchus mykiss, gill tissue sensitivity to cortisol is correlated positively with GR concentration (Shrimpton and McCormick, 1999). It has been shown that cortisol and physical stressors, such as confinement, downregulate GR expression in fish gills (Maule and Schreck, 1990; Shrimpton and Randall, 1994). However, no information is available on the effects of chemical stressors, such as heavy metals, on gill GR expression. Moreover, branchial GR expression has been studied mostly by steroid binding assay, and the cellular localization of GR in the gills of rainbow trout remains to be determined. Cortisol plays an important role in osmoregulatory adaptation, as it stimulates the proliferation and differentiation of the ion-transporting cells (chloride cells, CCs) of the gills as well as the Na+/K+-ATPase expression in these cells (Flik and Perry, 1989; Wendelaar Bonga and Lock, 1992; Seidelin et al., 1999). Consensus exists that exposure of fish to waterborne Cu leads to decreased branchial Na+/K+-ATPase activity and increased plasma cortisol levels (Brown, 1993; McDonald and Wood, 1993; Wendelaar Bonga, 1997). In a recent study (Bury et al., 1998), it was shown that cortisol protects against Cu induced necrosis and promotes apoptosis in tilapia gill chloride cells in vitro. Thus, Cu and cortisol have chloride cells as target and increased metallothionein (MT) synthesis may represent a protective mechanism. Concomitantly increased plasma cortisol and branchial MT were observed in cadmium exposed tilapia (Fu et al., 1990). High doses of dexamethasone (40 mg/kg) and cortisol (2.8 mg/ml medium) have been shown to stimulate MT synthesis in vivo in crucian carp liver and kidney (Muto et al., 1999), and in vitro in primary cultures of rainbow trout liver cells (Hyllner et al., 1989). However, evidence for MT expression in gills directly controlled by cortisol is lacking so far. The inducibility of MT by metals in gills of rainbow trout is also under debate. Laure´n and McDonald (1987) found no MT induction in rainbow trout after exposure of fish to 0.85 mM Cu for up to 28 days. Indeed, Cu may not be a very
strong inducer of MT expression in rainbow trout (Olsson et al., 1998). However, Laure´n and McDonald (1987) used a method to quantify MT that may not have recognized Cu–MT. In contrast, induction of MT expression has been clearly established, both at the gene and protein level in rainbow trout gills and other tissues and cell types after exposure to Cd (Norey et al., 1990), Zn (Hogstrand et al., 1995; Mayer et al., 1996; Olsson, 1996), or Ag (Hogstrand et al., 1996; Mayer et al., 1996). We have demonstrated recently by means of immunocytochemistry that waterborne Cu induces MT synthesis in tilapia gills and MT-immunopostitivity was found in all branchial cell types (except mucous cells) within 5 days (Dang et al., 1999). As a start to understand the response of trout to sublethal levels of Cu, we here report an inventory of the GR and MT expression in the gills of fish during 5 days of exposure to 2.4 mM Cu in the water. We applied immunocytochemistry to study the effects of Cu exposure on GR and MT in the branchial epithelial cells of rainbow trout. Using confocal laser scanning microscopy (CLSM), colocalization of GR and Na+/K+-ATPase (for chloride cells) and of MT and Na+/K+-ATPase were carried out to identify the cell types in the branchial epithelium that harbor these proteins; co-localization of GR and MT was also done to address the relationship between MT induction and GR distribution in the gills.
2. Materials and methods
2.1. Fish Young brood of rainbow trout, O. mykiss, was obtained from a local trout farm. Fish were kept in round 500-l tanks containing continuously aerated, filtered, and recirculating (20% fresh in part) Nijmegen tapwater (14°C; pH 7.6). The concentrations of the main ions in the water (in mM) are: Na+ (0.4), Ca2 + (0.7), Mg2 + (0.2), Cl− (0.8). Fish were fed Trouvit pellets, at 2% of their body weight per day. Lights were on for 12 h/day.
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
2.2. Experimental design Two weeks before the start of the experiment, sixteen fish (mean weight 4293 g) from the stock tank were divided into two groups of eight fish each in 80-l round tanks. Water conditions were as above. The experiment started by pumping a 10 mM stock Cu solution (added as s.a. Cu(NO3)2) into one tank. The flow rate was 500 ml/min during the first 5 h and 100 ml/h afterwards. In this way, the concentration in the tank gradually rose from a control level of 90 nmol/l Cu to approximately 2.4 mmol/l (equivalent to around 150 mg/l) after 5 h (Dang et al., 1999). The exposure period was 5 days, during which period the fish were not fed. At the end of the experiment, fish from both tanks were anaesthetized in a neutralized MS 222 solution (0.1 g/l), killed by spinal dissection and the second gill arch was removed from the left side for immunohistochemistry.
2.3. Cortisol Plasma cortisol was measured by radioimmunoassay according to Balm et al. (1994).
2.4. Immunostaining for GR or MT Bouin-fixed (24 h) gills were embedded in paraffin, cut at 7 mm and mounted on poly-Llysine coated slides (Sigma, St. Louis, MO, USA). An avidin–biotin– peroxidase complex (ABC) based method (Hsu et al., 1981) for immunocytochemical detection of GR and MT was chosen. Endogenous peroxidase activity was quenched by 20 min incubation in methanol containing 2% H2O2 at room temperature. Slides were then rinsed in two changes of 0.05 M Tris-buffered saline (150 mM NaCl containing 0.03% Triton X-100, pH 7.6; TBS TX) for 10 min each and incubated with 20% normal goat serum for 30 min. For detection of MT, a polyclonal antiserum raised in rabbit against perch (Perca flu6iatilis) MT (dilution 1: 4000) was used (Hogstrand and Haux, 1990). This antibody has been successfully used for tilapia (Dang et al., 1999), rainbow trout and brown trout (Burkhardt-Holm et al., 1999).
47
For GR detection, a polyclonal antiserum raised in rabbit against rainbow trout GR (dilution 1: 2000) was applied (Tujague et al., 1998). Gill sections with application of GR or MT antiserum were incubated overnight at room temperature. Biotinylated goat anti-rabbit IgG was used as second antibody for MT or GR for 1 h at room temperature; next, peroxidase-conjugated streptavidin (ABC kit, VECTOR Laboratories, prepared at least 30 min before use) was applied for a further 1 h. Between each step the sections were washed twice for 10 min in TBS TX solution. Thereafter, 3-3%-diaminobenzidine (DAB) in TB buffer (0.05 M Tris-buffered saline pH 7.6) with H2O2 (0.03%) was applied at room temperature. Finally, the sections were dehydrated and mounted in Entellan®. In controls the first antiserum was omitted.
2.5. Double immunostaining 2.5.1. Na+/K+-ATPase and MT After removing the paraffin in xylene and graded alcohol solutions, endogenous peroxidase was blocked and slides were rinsed in two changes of TBS TX for 10 min each. A monoclonal antibody to avian Na+/K+-ATPase (code, IgGa5, Developmental Studies Hybridoma Bank, Johns Hopkins, University School for Medicine, Baltimore and the Department of Biological Sciences, University of Iowa, Iowa City, USA, dilution 1:100) and the above mentioned MT antibody were simultaneously applied and incubated overnight at room temperature. Biotinylated goat anti-rabbit IgG probed with streptavidin-FITC was used to visualize MT. Texas-Red-conjugated goat-anti-mouse was used to probe Na+/K+-ATPase. Incubations at room temperature lasted 1 h in all cases. Between each step the sections were washed twice in TBS TX. 2.5.2. Na+/K+-ATPase and GR The above mentioned monoclonal Na+/K+ATPase and GR antisera were simultaneously applied and incubated overnight at room temperature. Texas-Red-conjugated goat-anti-rabbit (for GR) and FITC-conjugated goat-anti-mouse (for chloride cells) were used simultaneously to probe for GR and Na+/K+-ATPase, respectively.
48
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
2.5.3. MT and GR Sections were incubated with the above mentioned MT antiserum overnight at room temperature and biotinylated anti-rabbit IgG was used as a second antiserum for 1 h. Next, GR antiserum was applied overnight. Finally, Texas-Red-conjugated goat-anti-rabbit (for GR) and streptavidinFITC were simultaneously used to probe for GR and MT, respectively. 2.6. Confocal laser scanning microscopy (CLSM) A confocal laser scanning microscope (CLSM, MRC-600, BioRad) was used to study the colocalization of MT and Na+/K+-ATPase, GR and Na+/K+-ATPase, as well as GR and MT. All images were recorded and processed with an IBM compatible computer.
2.7. Quantification Gill MT-ir cells were quantified by a video image data analyzing system (VIDAS; Kontron, Germany). Sagittal sections of gill arches were cut at 7 mm and every 8th section was mounted; per fish at least five sections were collected on the slides and quantified. On each section, filaments that contained lamellae were randomly taken; five different areas (two filaments in each area, each filament with 0.3 mm length) were quantified. For each fish, the numerical density of immunoreactive cells was expressed as number of cells per mm length of filament. Seven fish were analyzed per group (n=7). GR-ir areas were quantified by VIDAS. The same section collection strategy as for MT was used for GR. The threshold light intensity was fixed according to indication of computer for control and Cu exposed tissues. Five different areas from each section were randomly chosen and the surface of the stained GR-ir areas was measured. The results are expressed as GR-ir mm2/mm2.
2.8. Statistics Data are presented as means 9S.D. Differences among groups were assessed by two-tailed
t-test for unpaired observations using Instat® software.
3. Results
3.1. GR and MT in branchial epithelium In control trout, GR-ir cells with varying staining intensities were located both in the filamental and the lamellar epithelium; numerous GR-positive cells both in the filaments and lamellae were strongly stained (Fig. 1A). The staining intensity of GR-ir cells in the filaments and lamellae had clearly decreased after 5 days of Cu exposure, although some cells were still strongly stained (Fig. 1B). The GR-ir cell number had decreased by 73% from 35 01799780 mm2/mm2 (range: 17 600–43 600 mm2/mm2) in controls to 9446 9 7572 mm2/mm2 (range: 3400–24 600 mm2/mm2) in 5 days of Cu-exposed fish (PB 0.001; Fig. 2). In control fish, few MT-ir cells were found in the interlamellar area of filaments and occasionally in the lamellae (Fig. 3A). However, after 5 days of Cu exposure, numerous MT-ir cells with a strong staining intensity were seen; most of these cells were located in the interlamellar area (Fig. 3B), some were in the lamellae and in that case stained either weakly or strongly. After 5 days of Cu exposure, MT-ir cell number increased from 5.59 0.9 in controls to 45.59 11.3 per mm filament length in Cu exposed fish (PB 0.001; Fig. 2).
3.2. Cortisol Plasma cortisol levels were 117.6 933.5 ng/ml in trout exposed to Cu for 5 days, an increase of 253% over controls (PB 0.001; Fig. 4).
3.3. Localization Confocal laser scanning microscopy (CLSM) examination of gills from Cu-exposed rainbow trout (Fig. 5) immuno-labeled for GR (red color, A), for Na+/K+-ATPase (green color, B) or double labeled for both (yellow color, C) showing that GR expression was present both in filamental
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
49
Fig. 1. GR immunostaining in the gills of control and Cu (2.4 mmol/l) exposed rainbow trout. In control fish, numerous GR-ir cells both in the filaments and lamellae were stained strongly (A). In Cu exposed fish, GR-ir cell number in the filaments and lamellae had decreased; the staining intensity of GR-ir cells varied with some of cells staining strongly (B). Magnification: 320 ×.
CCs (i.e. the Na+/K+-ATPase positive cells) and lamellar CCs. Not all CCs were GR-positive. In the filaments, GR-ir cells not stained for Na+/K+ -ATPase were found, which were identified as pavement cells (PCs), basal layer cells (BLCs) and undifferentiated cells (UCs) according to their topology. In the lamellae, the GR-ir cells that did not stain for Na+/K+-ATPase were respiratory cells (RCs). Gills immuno-labeled for Na+/K+ATPase (red color, D), for MT (green color, E), or double-labeled for these epitopes (yellow color, F), showed that CCs in the filamental and lamellar epithelium of Cu exposed fish contained MT. Occasionally, MT was present in PCs and UCs. Double staining for GR (red color, G) and for MT (green color, H) revealed that some MT-ir cells in the filaments and lamellae were also GR-ir cells, with MT distributed more in the apical part than GR (yellow color, I). We also found some MT-positive but GR-negative CCs.
Wood, 1993; Wendelaar Bonga, 1997). Waterborne Cu enters all types of branchial cells (Dang et al., 1999) and thus may result in a direct interaction between Cu and the GR. Metal ions such as Cu and Cd have been demonstrated to interfere directly with steroid binding to GR as well as with the receptor binding to the GREs of
4. Discussion Our results demonstrate that Cu exposure of rainbow trout decreases the number of branchial GR-ir cells. This decrease may well result from heavy metal-GR interactions. Fish gills are prime targets for acute Cu toxicity (McDonald and
Fig. 2. Quantification of GR and MT in the gills of control rainbow trout and of fish exposed to 3.2 mmol/l Cu for 5 days. Data are means 9 S.D. (n = 7). Significant differences are indicated with asterisks, PB0.001.
50
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
Fig. 3. MT immunostaining of gills in control and Cu (3.2 mmol/l) exposed rainbow trout. In control fish, a few MT staining cells were present in the filament (A). After Cu exposure, more MT-ir cells appeared in the filament (B). Magnification: 320 ×.
the genome (Freedman et al., 1988; Dunderski et al., 1992; Makino et al., 1996). Furthermore, these metals may disturb the cross-talk of the GR with other transcription factors that determine genome activation or repression (Makino et al., 1996; Reichardt and Schutz, 1998). However, elevated levels of cortisol may also contribute to the downregulation of GR. Prolonged elevation of plasma cortisol during waterborne Cu exposure, observed in our study on rainbow trout has been reported earlier for this species as well as for other species (Munoz et al., 1991; Brown, 1993). Ligand-binding techniques have shown that cortisol or dexamethasone treatments decrease GR numbers in the liver and brain of rainbow trout (Pottinger, 1990; Lee et al., 1992) as well as in the gills of coho salmon (Maule and Schreck, 1990; Shrimpton and Randall, 1994). In chum salmon cortisol treatment decreased the numbers of GR and the overall tissue immunoreactivity (Uchida et al., 1998). The elevation of cortisol levels after Cu exposure in our experiment is further evidence that Cu evokes a stress response (Brown, 1993; Wendelaar Bonga, 1997). It is well-known that stressors such as netting, crowding or air exposure elevate plasma cortisol levels, and downregulate GR in fish gills (Maule and Schreck, 1990; Shrimpton and Randall, 1994). The classic mechanism of
downregulation of GR by cortisol include decreased receptor affinity (Maule and Schreck, 1990; Shrimpton and Randall, 1994), shortening of receptor half-life (McIntyre and Samuels, 1985), and reduction of gene transcription (Rosewicz et al., 1988). As the 5% flanking region of the rainbow trout MT-A gene contains a putative glucocorticoid responsive element (GRE) with significant similarity to mammalian sequence
Fig. 4. Plasma cortisol levels in control and 5 days of Cu-exposed fish. Data are means 9S.D. (n =8); asterisks, PB 0.001.
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
51
Fig. 5. Confocal laser scanning microscopy of gill sections from Cu exposed rainbow trout double-labeled for GR (red color, A) and for Na+/K+-ATPase (green color, B) showing that GR expression was present in the Na+/K+-ATPase positive chloride cells (CCs; yellow color, C) of filaments and lamellae; other cell types were defined according to their topology, and were Na+/K+-ATPase negative as shown by merging pictures A and B. Magnification: 400 × . Immunostaining for Na+/K+-ATPase (red color, D), for MT (green color, E) and merged pictures of double staining for both (yellow color, F) showed that MT was present in the CCs. Magnification: 600 × . Pictures of GR immunostaining (red color, G) and MT immunostaining (green color, H) were merged to show that MT and GR were often but not consistently present in the same cells. Magnification: 400 × .
52
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
(Olsson et al., 1995), it could be argued that cortisol may mediate expression of MT during Cu exposure. There is evidence that stress or administration of glucocorticoids can induce MT expression in the liver of mullet, red drum (Baer and Thomas, 1990), largemouth bass (Weber et al., 1992) and crucian carp (Muto et al., 1999). However, attempts to induce MT by glucocorticoids through administration to plaice (Overnell et al., 1987) and rainbow trout (Hyllner et al., 1989) have been unsuccessful. Thus, so far there is no evidence that endogenous cortisol alone can induce expression of MT in tissues of rainbow trout (Hyllner et al., 1989). According to northern blot analyses for MT-mRNA as well as to reporter gene assays in a rainbow trout-derived cell line (RTG-2), glucocorticoids do not activate the MTA promoter (Hogstrand, unpublished observation). However, it is possible that cortisol enhances the metal-mediated stimulation of MT synthesis either by interacting with GR and GRE or by increasing the concentration of Cu or Zn in gill cells. Such a heavy metal and cortisol synergism was not considered in studies on the mullet, red drum, largemouth bass, and crucian carp, where it was concluded that glucocorticoid induction of MT is of physiological significance (Baer and Thomas, 1990; Weber et al., 1992; Muto et al., 1999). Our finding that GR and MT are not always colocated in the same chloride cells may mean either that MT synthesis does not require cortisol or that the cortisol signal produced downregulation of GR. The opposite changes of MT and GR in response to Cu in our study further indicate that cortisol alone does not stimulate MT synthesis in gill epithelium in vivo. In line with this conclusion we found that confinement for 5 days and cortisol feeding, treatments that both increase plasma cortisol to the same levels as Cu treatment does, did not induce MT synthesis in tilapia gill epithelium in vivo (Dang et al., unpublished). In rainbow trout, chloride cells, pavement cells, respiratory cells, undifferentiated cells, and basal layer cells express GR. Similar observations were made on chum salmon; the highest amount of GR expression was seen in chloride cells (Uchida et al., 1998). The difference in GR density and distri-
bution between rainbow trout (GR not dominantly present in chloride cells) and chum salmon (GR dominantly present in chloride cells) may relate to species differences. We observed the same pattern for GR expression in Atlantic salmon as was published for chum salmon. Co-localization of MT and Na+/K+-ATPase (for chloride cells) revealed that stimulation of MT expression after 5 days of Cu occurred only in chloride cells. This observation is in agreement with a study conducted on rainbow trout exposed to a sewage plant effluent, in which increased MT expression was observed in chloride cells only (Burkhardt-Holm et al., 1999). Similarly, in gills of Cu-exposed tilapia, MT induction occurred in the chloride cells. However a clear difference with regard to stimulation of MT expression in branchial epithelium exists between these two species, as tilapia shows the strongest induction of MT in pavement cells, respiratory cells and basal layer cells (Dang et al., 1999). For GR and MT staining, a marked cellular heterogeneity in branchial chloride cells was observed in our study. We found four types of chloride cells, i.e. GR-ir chloride cells, MT-ir chloride cells, both GR- and MT-ir chloride cells, and both GR- and MT-ir negative chloride cells. That not all chloride cells are MT-positive was also observed in another study on rainbow trout (Burkhardt-Holm et al., 1999). This heterogeneity may relate to the developmental stage of the chloride cells, because induction of MT expression by Cu may well be restricted to the early stages of cell development (Dang et al., 1999). It may well be that the window for expression of the GR is different in trout and this may explain the occurrence of cells with and without GR and with or without MT. MT expressed in chloride cells coexpressing GR might restore the function of the GR in branchial epithelia after Cu exposure, because metal interaction with GR can be reversed by the metal binding chemical dithiothreitol (DTT) as well as by the sulfhydryl modifying reagents 2-mercaptoethanol and N-acetyl-L-cysteine (Dunderski et al., 1992; Makino, et al., 1996) and thus likely also by MT because MT has high heavy metal binding capacity (Hussain et al., 1995). As the responsiveness of branchial Na+/
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
K+-ATPase to cortisol is correlated with gill GR concentrations (Shrimpton and McCormick, 1999), a restoration of functional GR by MT may allow the fish to cope more easily with stressors such as Cu. MT in chloride cells may well protect the Na+/K+-ATPase from Cu blockage and restore its function (Hussain et al., 1995). It has been well established that Cu induces apoptosis and necrosis of chloride cells (Wendelaar Bonga, 1997; Bury et al., 1998). This phenomenon indicates that different chloride cells react in a different way to Cu. Indeed, we have found in tilapia gills that different chloride cells accumulate different amounts of Cu after Cu exposure (Dang et al., 1999). It is possible that the amount of Cu accumulated by the cells may be related to the type of cell death; necrosis is generally associated with lethal cell/membrance damage, whereas the controlled apoptotic cell death leaves the cell intact. Moreover, cell death by necrosis involves a higher risk of inflammation and associated tissue damage than cell death by apoptosis (Wendelaar Bonga and Van der Meij, 1989). It could well be that the type of cell death relates to the amounts of GR and/or MT present in the cells. We here show that some CCs of rainbow trout do not contain GR and MT. As cortisol protects against Cu induced necrosis and induces apoptosis in CCs (Bury et al., 1998), CCs without GR may be more vulnerable to necrosis than apoptosis.
References Baer, K.N., Thomas, P., 1990. Influence of capture stress, salinity and reproductive status on zinc associated with metallothionein-like proteins in the livers of three marine teleost species. Mar. Environ. Res. 29, 277–287. Balm, P.H.M., Pepels, P., Helfrich, S., Hovens, M.L.M., Wendelaar Bonga, S.E., 1994. Adrenocorticotropic hormone in relation to interrenal function during stress in tilapia (Oreochromis mossambicus). Gen. Comp. Endocrinol. 96, 347– 360. Brown, J.A., 1993. Endocrine responses to environmental pollutants. In: Rankin, J.C., Jensen, F.B. (Eds.), Fish Ecophysiology. Chapman and Hall, London, pp. 276–295. Burkhardt-Holm, P., Bernet, P., Hogstrand, C., 1999. Increase of metallothionein-immunopositive chloride cells in the
53
gills of brown trout and rainbow trout after exposure to sewage treatment plant effluents. Histochem. J. 131, 339 – 346. Bury, N.R., Li, J., Flik, G., Lock, R.A.C., Wendelaar Bonga, S.E., 1998. Cortisol protects against copper induced necrosis and promotes apoptosis in fish gill chloride cells in vitro. Aquat. Toxicol 40, 193 – 202. Chakraborti, P.K., Weisbart, M., Charkraborti, A., 1987. The presence of corticosteroid receptor activity in the gills of the brook trout, Sal6elinus fontinalis. Gen. Comp. Endocrinol. 66, 323 – 332. Dang, Z.C., Lock, R.A.C., Flik, G., Wendelaar Bonga, S.E., 1999. The metallothionein response in gills of Oreochromis mossambicus exposed to copper in fresh water. Am. J. Physiol. 277, R320 – 331. Dunderski, J., Stanosevic, J., Ristic, B., Trajkovic, D., Matic, G., 1992. In vivo effects of cadmium on rat liver glucocorticoid receptor functional properties. Int. J. Biochem. 24, 1065 – 1072. Flik, G., Perry, S.F., 1989. Cortisol stimulates whole-body calcium uptake and the branchial calcium pump in freshwater rainbow trout. J. Endocrinol. 120, 75 – 82. Freedman, L.P., Luisi, B.F., Korszun, Z.R., Basavappa, R., Sigler, P.B., Yamamoto, K.R., 1988. The function and structure of the metal coordination sites within the glucocorticoid receptor DNA binding domain. Nature (London) 334, 543 – 546. Fu, H., Steinebach, O.M., Van den Hamer, C.J.A., Balm, P.H.M., Lock, R.A.C., 1990. Involvement of cortisol and metallothionein-like proteins in the physiological responses of tilapia (Oreochromis mossambicus) to sublethal cadmium stress. Aquat. Toxicol. 16, 257 – 270. Hogstrand, C., Haux, C., 1990. A radioimmunoassay for perch (Perca flu6iatilis) metallothionein. Toxicol. Appl. Pharmacol. 103, 56 – 65. Hogstrand, C., Reid, S.D., Wood, C.M., 1995. Calcium versus zinc transport through the gills of freshwater teleost fish. J. Exp. Biol. 298, 337 – 348. Hogstrand, C., Galvez, F., Wood, C.M., 1996. Toxicity, silver accumulation and metallothionein induction in freshwater rainbow trout during exposure to different silver salts. Environ. Toxicol. Chem. 15, 1102 – 1108. Hsu, S.M., Raine, L., Fanger, H., 1981. Use of avidin-biotinperoxidase complexes (ABC) and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem. 29, 577 – 580. Hussain, S., Anner, R.M., Anner, B.M., 1995. Metallothionein protects purified Na+/K+-ATPase from toxicity in vitro. In Vitro Toxicol. 8, 25 – 30. Hyllner, S.J., Andersson, T., Haux, C., Olsson, P.E., 1989. Cortisol induction of metallothionein in primary culture of rainbow trout hepatocytes. J. Cell. Physiol. 139, 24 – 28. Laure´n, D.J., McDonald, D.G., 1987. Acclimation to copper by rainbow trout, Salmo gairdneri: physiology. Can. J. Fish. Aquat. Sci. 44, 99 – 104. Lee, P.C., Goodrich, M., Struve, M., Yoon, H.I., Weber, D., 1992. Liver and brain glucocorticoid receptor in rainbow trout, Oncorhynchus mykiss: down-regulation by dexamethasone. Gen. Comp. Endocrinol. 87, 222 – 231.
54
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
Makino, Y., Tanaka, H., Dahlman-Wright, K., Makino, I., 1996. Modulation of glucocorticoid-inducible gene expression by metal ions. Mol. Pharmacol. 49, 612–620. Maule, A.G., Schreck, C.B., 1990. Glucocorticoid receptors in leukocytes and gill of juvenile coho salmon (Oncorhynchus kisutch). Gen. Comp. Endocrinol. 77, 448–455. Mayer, G.D., Leach, D.A., Olsson, P.-E., Hogstrand, C., 1996. Transcriptional activation of the rainbow trout metallothionein-A gene by Zn and Ag. In: Andre´n, A.W., Bober, T.W. (Eds.), Proceedings of the 4th International Conference on the Transport, Fate and Effects of Silver in the Environment. University of Wisconsin, Madison, WI, pp. 179 – 184. McDonald, D.G., Wood, C.M., 1993. Branchial mechanisms of acclimation to metals in freshwater fish. In: Rankin, J.C., Jensen, F.B. (Eds.), Fish Ecophysiology. Chapman and Hall, London, pp. 297–332. McIntyre, W.R., Samuels, H.H., 1985. Triamcinolone acetonide regulates glucocorticoid-receptor levels by decreasing half-life of the activated nuclear-receptor form. J. Biol. Chem. 260, 418 – 427. Munoz, M.J., Carballo, M., Tarazona, J.V., 1991. The effect of sublethal levels of copper and cyanide on some biochemical parameters of rainbow trout along subacute exposition. Comp. Biochem. Physiol. 100C, 577–582. Muto, N., Ren, H.W., Hwang, G.S., Tominaga, S., Itoh, N., Tanaka, K., 1999. Induction of two major isoforms of metallothionein in crucian carp (Carassius cu6ieri ) by airpumping stress, dexamethasone, and metals. Comp. Biochem. Physiol. 122C, 75–82. Norey, C.G., Cryer, A., Kay, J., 1990. A comparison of cadmium-induced metallothionein gene expression and metal ion distribution in the tissues of cadmium-sensitive (rainbow trout, Salmo gairdneri ) and tolerant (stone loach, Noemacheilus barbatulus) species of freshwater fish. Comp. Biochem. Physiol. C. 97, 221–226. Olsson, P.E., Kling, P., Erkell, L.J., Kille, P., 1995. Structural and functional analysis of the rainbow trout (Oncorhyncus mykiss) metallothionein-A gene. Eur. J. Biochem. 230, 344 – 349. Olsson, P.-E., Kling, P., Hogstrand, C., 1998. Mechanisms of heavy metal accumulation and toxicity in fish. In: Langston, W.J., Bebianno, M.J. (Eds.), Metal Metabolism in Aquatic Environments. Chapman and Hall, London, UK, pp. 321 – 350 Chapter 8. Olsson, P.E., 1996. Metallothioneins in fish: induction and use in environmental monitoring. In: Taylor, E.W. (Ed.), Toxicology of Aquatic Pollution: Physiological, Cellular and Molecular Approaches. Cambridge University Press, Cambridge, UK.
.
Overnell, J., McIntosh, R., Fletcher, T.C., 1987. The enhanced induction of metallothionein by zinc, its half-life in the marine fish Pleuronectes platessa, and the influence of stress factors on metallothionein levels. Experientia 43, 178 – 181. Pottinger, T.G., 1990. The effect of stress and exogenous cortisol on receptor-like binding of cortisol in the liver of rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 78, 194 – 203. Reichardt, H.M., Schutz, G., 1998. Glucocorticoid signalling — multiple variations of a common theme. Mol. Cell. Endocrinol. 146, 1 – 6. Rosewicz, S., McDonald, A.R., Maddux, B.A., Goldfine, I.D., Miesfeld, R.L., Logsdon, C.D., 1988. Mechanism of glucocorticoid receptor down-regulation by glucocorticoids. J. Biol. Chem. 263, 2581 – 2584. Seidelin, M., Madsen, S.S., Byrialsen, A., Kristiansen, K., 1999. Effects of insulin-like growth factor-I and cortisol on Na+/K+-ATPase expression in osmoregulatory tissues of brown trout (Salmo trutta). Gen. Comp. Endocrinol. 113, 331 – 342. Shrimpton, J.M., McCormick, S.D., 1999. Responsiveness of gill Na+/K+-ATPase to cortisol is related to gill corticosteroid receptor concentration in juvenile rainbow trout. J. Exp. Biol. 202, 987 – 995. Shrimpton, J.M., Randall, D.J., 1994. Downregulation of corticosteroid receptors in gills of coho salmon due to stress and cortisol treatment. Am. J. Physiol. 267, R432 – 438. Tujague, M., Saligaut, D., Teitsma, C., Kah, O., Valotaire, Y., Ducouret, B., 1998. Rainbow trout glucocorticoid receptor overexpression in Escherichia coli: production of antibodies for western blotting and immunohistochemistry. Gen Comp. Endocrinol. 110, 201 – 211. Uchida, K., Kaneko, T., Tagawa, M., Hirano, T., 1998. Localization of cortisol receptor in branchial chloride cells in chum salmon fry. Gen. Comp. Endocrinol. 109, 175 – 181. Weber, D.N., Eisch, S., Spieler, R.E., Petering, D.H., 1992. Metal redistribution in largemouth bass (Micropterus salmoides) in response to restrainment stress and dietary cadmium: role of metallothionein and other metal-binding proteins. Comp. Biochem. Physiol. 101C, 255 – 262. Wendelaar Bonga, S.E., Lock, R.A.C., 1992. Toxicants and osmoregulation in fish. Neth. J. Zool. 42, 478 – 493. Wendelaar Bonga, S.E., Van der Meij, J.C.A., 1989. Degeneration and death, by apoptosis and necrosis, of the pavement and chloride cells in the gills of the teleost Oreochromis mossambicus. Cell Tissue Res. 419, 454 – 459. Wendelaar Bonga, S.E., 1997. The stress response in fish. Physiol. Rev. 77, 591 – 625.
Effects of copper on cortisol receptor and metallothionein expression in gills of Oncorhynchus mykiss Zhi Chao Dang a, Gert Flik a, Bernadette Ducouret b, Christer Hogstrand c, Sjoerd E. Wendelaar Bonga a, Robert A.C. Lock a,* b
a Department of Animal Physiology, Uni6ersity of Nijmegen, Toernooi6eld, 6525 ED Nijmegen, The Netherlands Endocrinologie Mole´culaire de la Reproduction, Uni6ersite´ de Rennes,, Campus de Beaulieu, 35042 Rennes Cedex, France c King’s College London, School of Health and Life Sciences, Di6ision of Life Sciences, Franklin – Wilkins Building, 150 Stamford Street, London SE1 8WA, UK
Received 3 December 1999; received in revised form 28 February 2000; accepted 1 March 2000
Abstract Effects of waterborne Cu (2.4 mM) on the expression of glucocorticoid receptor (GR) and metallothionein (MT) in the branchial epithelium of freshwater rainbow trout (Oncorhynchus mykiss) was studied by immunocytochemistry. After 5 days of Cu exposure, the number of GR-immunoreactive (GR-ir) cells in the gill epithelium had decreased, whereas the number of MT-ir cells had increased. Localization of GR in chloride cells was achieved by double staining for Na+/K+-ATPase; other cell types were identified on the basis of their topology. GRs were present in the chloride cells in both the filaments and lamellae, in respiratory cells in the lamellae, in pavement cells, basal layer cells and undifferentiated cells in the filaments. Co-localization of Na+/K+-ATPase and MT revealed that MT was expressed in chloride cells, both in filaments and lamellae. Occasionally, MT immunoreactivity was found in pavement cells and in undifferentiated cells. By double staining for Na+/K+-ATPase and GR, for Na+/K+-ATPase and MT and for GR and MT, we can conclude that after 5 days of Cu stress there are chloride cells that express GR and MT, GR or MT alone or neither of the two proteins. This apparent functional heterogeneity of branchial chloride cells may reflect a limited window when chloride cell subpopulations show an adaptive response to Cu. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Glucocorticoid receptor; Metallothionein; Fish gills; Chloride cell; Cu; Immunocytochemistry
1. Introduction
* Corresponding author. Tel.: +31-24-3652476; fax: + 3124-3652714. E-mail address: [email protected] (R.A.C. Lock).
The glucocorticoid receptor (GR) is an important mediator of cortisol signals. By using steroid binding assay GR expression has been detected in fish gills, liver, intestine, brain and muscle, with gill tissue having the highest cortisol-binding ac-
0166-445X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 0 ) 0 0 1 0 2 - 8
46
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
tivity (Chakraborti et al., 1987; Lee et al., 1992; Tujague et al., 1998). In rainbow trout Oncorhynchus mykiss, gill tissue sensitivity to cortisol is correlated positively with GR concentration (Shrimpton and McCormick, 1999). It has been shown that cortisol and physical stressors, such as confinement, downregulate GR expression in fish gills (Maule and Schreck, 1990; Shrimpton and Randall, 1994). However, no information is available on the effects of chemical stressors, such as heavy metals, on gill GR expression. Moreover, branchial GR expression has been studied mostly by steroid binding assay, and the cellular localization of GR in the gills of rainbow trout remains to be determined. Cortisol plays an important role in osmoregulatory adaptation, as it stimulates the proliferation and differentiation of the ion-transporting cells (chloride cells, CCs) of the gills as well as the Na+/K+-ATPase expression in these cells (Flik and Perry, 1989; Wendelaar Bonga and Lock, 1992; Seidelin et al., 1999). Consensus exists that exposure of fish to waterborne Cu leads to decreased branchial Na+/K+-ATPase activity and increased plasma cortisol levels (Brown, 1993; McDonald and Wood, 1993; Wendelaar Bonga, 1997). In a recent study (Bury et al., 1998), it was shown that cortisol protects against Cu induced necrosis and promotes apoptosis in tilapia gill chloride cells in vitro. Thus, Cu and cortisol have chloride cells as target and increased metallothionein (MT) synthesis may represent a protective mechanism. Concomitantly increased plasma cortisol and branchial MT were observed in cadmium exposed tilapia (Fu et al., 1990). High doses of dexamethasone (40 mg/kg) and cortisol (2.8 mg/ml medium) have been shown to stimulate MT synthesis in vivo in crucian carp liver and kidney (Muto et al., 1999), and in vitro in primary cultures of rainbow trout liver cells (Hyllner et al., 1989). However, evidence for MT expression in gills directly controlled by cortisol is lacking so far. The inducibility of MT by metals in gills of rainbow trout is also under debate. Laure´n and McDonald (1987) found no MT induction in rainbow trout after exposure of fish to 0.85 mM Cu for up to 28 days. Indeed, Cu may not be a very
strong inducer of MT expression in rainbow trout (Olsson et al., 1998). However, Laure´n and McDonald (1987) used a method to quantify MT that may not have recognized Cu–MT. In contrast, induction of MT expression has been clearly established, both at the gene and protein level in rainbow trout gills and other tissues and cell types after exposure to Cd (Norey et al., 1990), Zn (Hogstrand et al., 1995; Mayer et al., 1996; Olsson, 1996), or Ag (Hogstrand et al., 1996; Mayer et al., 1996). We have demonstrated recently by means of immunocytochemistry that waterborne Cu induces MT synthesis in tilapia gills and MT-immunopostitivity was found in all branchial cell types (except mucous cells) within 5 days (Dang et al., 1999). As a start to understand the response of trout to sublethal levels of Cu, we here report an inventory of the GR and MT expression in the gills of fish during 5 days of exposure to 2.4 mM Cu in the water. We applied immunocytochemistry to study the effects of Cu exposure on GR and MT in the branchial epithelial cells of rainbow trout. Using confocal laser scanning microscopy (CLSM), colocalization of GR and Na+/K+-ATPase (for chloride cells) and of MT and Na+/K+-ATPase were carried out to identify the cell types in the branchial epithelium that harbor these proteins; co-localization of GR and MT was also done to address the relationship between MT induction and GR distribution in the gills.
2. Materials and methods
2.1. Fish Young brood of rainbow trout, O. mykiss, was obtained from a local trout farm. Fish were kept in round 500-l tanks containing continuously aerated, filtered, and recirculating (20% fresh in part) Nijmegen tapwater (14°C; pH 7.6). The concentrations of the main ions in the water (in mM) are: Na+ (0.4), Ca2 + (0.7), Mg2 + (0.2), Cl− (0.8). Fish were fed Trouvit pellets, at 2% of their body weight per day. Lights were on for 12 h/day.
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
2.2. Experimental design Two weeks before the start of the experiment, sixteen fish (mean weight 4293 g) from the stock tank were divided into two groups of eight fish each in 80-l round tanks. Water conditions were as above. The experiment started by pumping a 10 mM stock Cu solution (added as s.a. Cu(NO3)2) into one tank. The flow rate was 500 ml/min during the first 5 h and 100 ml/h afterwards. In this way, the concentration in the tank gradually rose from a control level of 90 nmol/l Cu to approximately 2.4 mmol/l (equivalent to around 150 mg/l) after 5 h (Dang et al., 1999). The exposure period was 5 days, during which period the fish were not fed. At the end of the experiment, fish from both tanks were anaesthetized in a neutralized MS 222 solution (0.1 g/l), killed by spinal dissection and the second gill arch was removed from the left side for immunohistochemistry.
2.3. Cortisol Plasma cortisol was measured by radioimmunoassay according to Balm et al. (1994).
2.4. Immunostaining for GR or MT Bouin-fixed (24 h) gills were embedded in paraffin, cut at 7 mm and mounted on poly-Llysine coated slides (Sigma, St. Louis, MO, USA). An avidin–biotin– peroxidase complex (ABC) based method (Hsu et al., 1981) for immunocytochemical detection of GR and MT was chosen. Endogenous peroxidase activity was quenched by 20 min incubation in methanol containing 2% H2O2 at room temperature. Slides were then rinsed in two changes of 0.05 M Tris-buffered saline (150 mM NaCl containing 0.03% Triton X-100, pH 7.6; TBS TX) for 10 min each and incubated with 20% normal goat serum for 30 min. For detection of MT, a polyclonal antiserum raised in rabbit against perch (Perca flu6iatilis) MT (dilution 1: 4000) was used (Hogstrand and Haux, 1990). This antibody has been successfully used for tilapia (Dang et al., 1999), rainbow trout and brown trout (Burkhardt-Holm et al., 1999).
47
For GR detection, a polyclonal antiserum raised in rabbit against rainbow trout GR (dilution 1: 2000) was applied (Tujague et al., 1998). Gill sections with application of GR or MT antiserum were incubated overnight at room temperature. Biotinylated goat anti-rabbit IgG was used as second antibody for MT or GR for 1 h at room temperature; next, peroxidase-conjugated streptavidin (ABC kit, VECTOR Laboratories, prepared at least 30 min before use) was applied for a further 1 h. Between each step the sections were washed twice for 10 min in TBS TX solution. Thereafter, 3-3%-diaminobenzidine (DAB) in TB buffer (0.05 M Tris-buffered saline pH 7.6) with H2O2 (0.03%) was applied at room temperature. Finally, the sections were dehydrated and mounted in Entellan®. In controls the first antiserum was omitted.
2.5. Double immunostaining 2.5.1. Na+/K+-ATPase and MT After removing the paraffin in xylene and graded alcohol solutions, endogenous peroxidase was blocked and slides were rinsed in two changes of TBS TX for 10 min each. A monoclonal antibody to avian Na+/K+-ATPase (code, IgGa5, Developmental Studies Hybridoma Bank, Johns Hopkins, University School for Medicine, Baltimore and the Department of Biological Sciences, University of Iowa, Iowa City, USA, dilution 1:100) and the above mentioned MT antibody were simultaneously applied and incubated overnight at room temperature. Biotinylated goat anti-rabbit IgG probed with streptavidin-FITC was used to visualize MT. Texas-Red-conjugated goat-anti-mouse was used to probe Na+/K+-ATPase. Incubations at room temperature lasted 1 h in all cases. Between each step the sections were washed twice in TBS TX. 2.5.2. Na+/K+-ATPase and GR The above mentioned monoclonal Na+/K+ATPase and GR antisera were simultaneously applied and incubated overnight at room temperature. Texas-Red-conjugated goat-anti-rabbit (for GR) and FITC-conjugated goat-anti-mouse (for chloride cells) were used simultaneously to probe for GR and Na+/K+-ATPase, respectively.
48
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
2.5.3. MT and GR Sections were incubated with the above mentioned MT antiserum overnight at room temperature and biotinylated anti-rabbit IgG was used as a second antiserum for 1 h. Next, GR antiserum was applied overnight. Finally, Texas-Red-conjugated goat-anti-rabbit (for GR) and streptavidinFITC were simultaneously used to probe for GR and MT, respectively. 2.6. Confocal laser scanning microscopy (CLSM) A confocal laser scanning microscope (CLSM, MRC-600, BioRad) was used to study the colocalization of MT and Na+/K+-ATPase, GR and Na+/K+-ATPase, as well as GR and MT. All images were recorded and processed with an IBM compatible computer.
2.7. Quantification Gill MT-ir cells were quantified by a video image data analyzing system (VIDAS; Kontron, Germany). Sagittal sections of gill arches were cut at 7 mm and every 8th section was mounted; per fish at least five sections were collected on the slides and quantified. On each section, filaments that contained lamellae were randomly taken; five different areas (two filaments in each area, each filament with 0.3 mm length) were quantified. For each fish, the numerical density of immunoreactive cells was expressed as number of cells per mm length of filament. Seven fish were analyzed per group (n=7). GR-ir areas were quantified by VIDAS. The same section collection strategy as for MT was used for GR. The threshold light intensity was fixed according to indication of computer for control and Cu exposed tissues. Five different areas from each section were randomly chosen and the surface of the stained GR-ir areas was measured. The results are expressed as GR-ir mm2/mm2.
2.8. Statistics Data are presented as means 9S.D. Differences among groups were assessed by two-tailed
t-test for unpaired observations using Instat® software.
3. Results
3.1. GR and MT in branchial epithelium In control trout, GR-ir cells with varying staining intensities were located both in the filamental and the lamellar epithelium; numerous GR-positive cells both in the filaments and lamellae were strongly stained (Fig. 1A). The staining intensity of GR-ir cells in the filaments and lamellae had clearly decreased after 5 days of Cu exposure, although some cells were still strongly stained (Fig. 1B). The GR-ir cell number had decreased by 73% from 35 01799780 mm2/mm2 (range: 17 600–43 600 mm2/mm2) in controls to 9446 9 7572 mm2/mm2 (range: 3400–24 600 mm2/mm2) in 5 days of Cu-exposed fish (PB 0.001; Fig. 2). In control fish, few MT-ir cells were found in the interlamellar area of filaments and occasionally in the lamellae (Fig. 3A). However, after 5 days of Cu exposure, numerous MT-ir cells with a strong staining intensity were seen; most of these cells were located in the interlamellar area (Fig. 3B), some were in the lamellae and in that case stained either weakly or strongly. After 5 days of Cu exposure, MT-ir cell number increased from 5.59 0.9 in controls to 45.59 11.3 per mm filament length in Cu exposed fish (PB 0.001; Fig. 2).
3.2. Cortisol Plasma cortisol levels were 117.6 933.5 ng/ml in trout exposed to Cu for 5 days, an increase of 253% over controls (PB 0.001; Fig. 4).
3.3. Localization Confocal laser scanning microscopy (CLSM) examination of gills from Cu-exposed rainbow trout (Fig. 5) immuno-labeled for GR (red color, A), for Na+/K+-ATPase (green color, B) or double labeled for both (yellow color, C) showing that GR expression was present both in filamental
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
49
Fig. 1. GR immunostaining in the gills of control and Cu (2.4 mmol/l) exposed rainbow trout. In control fish, numerous GR-ir cells both in the filaments and lamellae were stained strongly (A). In Cu exposed fish, GR-ir cell number in the filaments and lamellae had decreased; the staining intensity of GR-ir cells varied with some of cells staining strongly (B). Magnification: 320 ×.
CCs (i.e. the Na+/K+-ATPase positive cells) and lamellar CCs. Not all CCs were GR-positive. In the filaments, GR-ir cells not stained for Na+/K+ -ATPase were found, which were identified as pavement cells (PCs), basal layer cells (BLCs) and undifferentiated cells (UCs) according to their topology. In the lamellae, the GR-ir cells that did not stain for Na+/K+-ATPase were respiratory cells (RCs). Gills immuno-labeled for Na+/K+ATPase (red color, D), for MT (green color, E), or double-labeled for these epitopes (yellow color, F), showed that CCs in the filamental and lamellar epithelium of Cu exposed fish contained MT. Occasionally, MT was present in PCs and UCs. Double staining for GR (red color, G) and for MT (green color, H) revealed that some MT-ir cells in the filaments and lamellae were also GR-ir cells, with MT distributed more in the apical part than GR (yellow color, I). We also found some MT-positive but GR-negative CCs.
Wood, 1993; Wendelaar Bonga, 1997). Waterborne Cu enters all types of branchial cells (Dang et al., 1999) and thus may result in a direct interaction between Cu and the GR. Metal ions such as Cu and Cd have been demonstrated to interfere directly with steroid binding to GR as well as with the receptor binding to the GREs of
4. Discussion Our results demonstrate that Cu exposure of rainbow trout decreases the number of branchial GR-ir cells. This decrease may well result from heavy metal-GR interactions. Fish gills are prime targets for acute Cu toxicity (McDonald and
Fig. 2. Quantification of GR and MT in the gills of control rainbow trout and of fish exposed to 3.2 mmol/l Cu for 5 days. Data are means 9 S.D. (n = 7). Significant differences are indicated with asterisks, PB0.001.
50
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
Fig. 3. MT immunostaining of gills in control and Cu (3.2 mmol/l) exposed rainbow trout. In control fish, a few MT staining cells were present in the filament (A). After Cu exposure, more MT-ir cells appeared in the filament (B). Magnification: 320 ×.
the genome (Freedman et al., 1988; Dunderski et al., 1992; Makino et al., 1996). Furthermore, these metals may disturb the cross-talk of the GR with other transcription factors that determine genome activation or repression (Makino et al., 1996; Reichardt and Schutz, 1998). However, elevated levels of cortisol may also contribute to the downregulation of GR. Prolonged elevation of plasma cortisol during waterborne Cu exposure, observed in our study on rainbow trout has been reported earlier for this species as well as for other species (Munoz et al., 1991; Brown, 1993). Ligand-binding techniques have shown that cortisol or dexamethasone treatments decrease GR numbers in the liver and brain of rainbow trout (Pottinger, 1990; Lee et al., 1992) as well as in the gills of coho salmon (Maule and Schreck, 1990; Shrimpton and Randall, 1994). In chum salmon cortisol treatment decreased the numbers of GR and the overall tissue immunoreactivity (Uchida et al., 1998). The elevation of cortisol levels after Cu exposure in our experiment is further evidence that Cu evokes a stress response (Brown, 1993; Wendelaar Bonga, 1997). It is well-known that stressors such as netting, crowding or air exposure elevate plasma cortisol levels, and downregulate GR in fish gills (Maule and Schreck, 1990; Shrimpton and Randall, 1994). The classic mechanism of
downregulation of GR by cortisol include decreased receptor affinity (Maule and Schreck, 1990; Shrimpton and Randall, 1994), shortening of receptor half-life (McIntyre and Samuels, 1985), and reduction of gene transcription (Rosewicz et al., 1988). As the 5% flanking region of the rainbow trout MT-A gene contains a putative glucocorticoid responsive element (GRE) with significant similarity to mammalian sequence
Fig. 4. Plasma cortisol levels in control and 5 days of Cu-exposed fish. Data are means 9S.D. (n =8); asterisks, PB 0.001.
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
51
Fig. 5. Confocal laser scanning microscopy of gill sections from Cu exposed rainbow trout double-labeled for GR (red color, A) and for Na+/K+-ATPase (green color, B) showing that GR expression was present in the Na+/K+-ATPase positive chloride cells (CCs; yellow color, C) of filaments and lamellae; other cell types were defined according to their topology, and were Na+/K+-ATPase negative as shown by merging pictures A and B. Magnification: 400 × . Immunostaining for Na+/K+-ATPase (red color, D), for MT (green color, E) and merged pictures of double staining for both (yellow color, F) showed that MT was present in the CCs. Magnification: 600 × . Pictures of GR immunostaining (red color, G) and MT immunostaining (green color, H) were merged to show that MT and GR were often but not consistently present in the same cells. Magnification: 400 × .
52
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
(Olsson et al., 1995), it could be argued that cortisol may mediate expression of MT during Cu exposure. There is evidence that stress or administration of glucocorticoids can induce MT expression in the liver of mullet, red drum (Baer and Thomas, 1990), largemouth bass (Weber et al., 1992) and crucian carp (Muto et al., 1999). However, attempts to induce MT by glucocorticoids through administration to plaice (Overnell et al., 1987) and rainbow trout (Hyllner et al., 1989) have been unsuccessful. Thus, so far there is no evidence that endogenous cortisol alone can induce expression of MT in tissues of rainbow trout (Hyllner et al., 1989). According to northern blot analyses for MT-mRNA as well as to reporter gene assays in a rainbow trout-derived cell line (RTG-2), glucocorticoids do not activate the MTA promoter (Hogstrand, unpublished observation). However, it is possible that cortisol enhances the metal-mediated stimulation of MT synthesis either by interacting with GR and GRE or by increasing the concentration of Cu or Zn in gill cells. Such a heavy metal and cortisol synergism was not considered in studies on the mullet, red drum, largemouth bass, and crucian carp, where it was concluded that glucocorticoid induction of MT is of physiological significance (Baer and Thomas, 1990; Weber et al., 1992; Muto et al., 1999). Our finding that GR and MT are not always colocated in the same chloride cells may mean either that MT synthesis does not require cortisol or that the cortisol signal produced downregulation of GR. The opposite changes of MT and GR in response to Cu in our study further indicate that cortisol alone does not stimulate MT synthesis in gill epithelium in vivo. In line with this conclusion we found that confinement for 5 days and cortisol feeding, treatments that both increase plasma cortisol to the same levels as Cu treatment does, did not induce MT synthesis in tilapia gill epithelium in vivo (Dang et al., unpublished). In rainbow trout, chloride cells, pavement cells, respiratory cells, undifferentiated cells, and basal layer cells express GR. Similar observations were made on chum salmon; the highest amount of GR expression was seen in chloride cells (Uchida et al., 1998). The difference in GR density and distri-
bution between rainbow trout (GR not dominantly present in chloride cells) and chum salmon (GR dominantly present in chloride cells) may relate to species differences. We observed the same pattern for GR expression in Atlantic salmon as was published for chum salmon. Co-localization of MT and Na+/K+-ATPase (for chloride cells) revealed that stimulation of MT expression after 5 days of Cu occurred only in chloride cells. This observation is in agreement with a study conducted on rainbow trout exposed to a sewage plant effluent, in which increased MT expression was observed in chloride cells only (Burkhardt-Holm et al., 1999). Similarly, in gills of Cu-exposed tilapia, MT induction occurred in the chloride cells. However a clear difference with regard to stimulation of MT expression in branchial epithelium exists between these two species, as tilapia shows the strongest induction of MT in pavement cells, respiratory cells and basal layer cells (Dang et al., 1999). For GR and MT staining, a marked cellular heterogeneity in branchial chloride cells was observed in our study. We found four types of chloride cells, i.e. GR-ir chloride cells, MT-ir chloride cells, both GR- and MT-ir chloride cells, and both GR- and MT-ir negative chloride cells. That not all chloride cells are MT-positive was also observed in another study on rainbow trout (Burkhardt-Holm et al., 1999). This heterogeneity may relate to the developmental stage of the chloride cells, because induction of MT expression by Cu may well be restricted to the early stages of cell development (Dang et al., 1999). It may well be that the window for expression of the GR is different in trout and this may explain the occurrence of cells with and without GR and with or without MT. MT expressed in chloride cells coexpressing GR might restore the function of the GR in branchial epithelia after Cu exposure, because metal interaction with GR can be reversed by the metal binding chemical dithiothreitol (DTT) as well as by the sulfhydryl modifying reagents 2-mercaptoethanol and N-acetyl-L-cysteine (Dunderski et al., 1992; Makino, et al., 1996) and thus likely also by MT because MT has high heavy metal binding capacity (Hussain et al., 1995). As the responsiveness of branchial Na+/
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
K+-ATPase to cortisol is correlated with gill GR concentrations (Shrimpton and McCormick, 1999), a restoration of functional GR by MT may allow the fish to cope more easily with stressors such as Cu. MT in chloride cells may well protect the Na+/K+-ATPase from Cu blockage and restore its function (Hussain et al., 1995). It has been well established that Cu induces apoptosis and necrosis of chloride cells (Wendelaar Bonga, 1997; Bury et al., 1998). This phenomenon indicates that different chloride cells react in a different way to Cu. Indeed, we have found in tilapia gills that different chloride cells accumulate different amounts of Cu after Cu exposure (Dang et al., 1999). It is possible that the amount of Cu accumulated by the cells may be related to the type of cell death; necrosis is generally associated with lethal cell/membrance damage, whereas the controlled apoptotic cell death leaves the cell intact. Moreover, cell death by necrosis involves a higher risk of inflammation and associated tissue damage than cell death by apoptosis (Wendelaar Bonga and Van der Meij, 1989). It could well be that the type of cell death relates to the amounts of GR and/or MT present in the cells. We here show that some CCs of rainbow trout do not contain GR and MT. As cortisol protects against Cu induced necrosis and induces apoptosis in CCs (Bury et al., 1998), CCs without GR may be more vulnerable to necrosis than apoptosis.
References Baer, K.N., Thomas, P., 1990. Influence of capture stress, salinity and reproductive status on zinc associated with metallothionein-like proteins in the livers of three marine teleost species. Mar. Environ. Res. 29, 277–287. Balm, P.H.M., Pepels, P., Helfrich, S., Hovens, M.L.M., Wendelaar Bonga, S.E., 1994. Adrenocorticotropic hormone in relation to interrenal function during stress in tilapia (Oreochromis mossambicus). Gen. Comp. Endocrinol. 96, 347– 360. Brown, J.A., 1993. Endocrine responses to environmental pollutants. In: Rankin, J.C., Jensen, F.B. (Eds.), Fish Ecophysiology. Chapman and Hall, London, pp. 276–295. Burkhardt-Holm, P., Bernet, P., Hogstrand, C., 1999. Increase of metallothionein-immunopositive chloride cells in the
53
gills of brown trout and rainbow trout after exposure to sewage treatment plant effluents. Histochem. J. 131, 339 – 346. Bury, N.R., Li, J., Flik, G., Lock, R.A.C., Wendelaar Bonga, S.E., 1998. Cortisol protects against copper induced necrosis and promotes apoptosis in fish gill chloride cells in vitro. Aquat. Toxicol 40, 193 – 202. Chakraborti, P.K., Weisbart, M., Charkraborti, A., 1987. The presence of corticosteroid receptor activity in the gills of the brook trout, Sal6elinus fontinalis. Gen. Comp. Endocrinol. 66, 323 – 332. Dang, Z.C., Lock, R.A.C., Flik, G., Wendelaar Bonga, S.E., 1999. The metallothionein response in gills of Oreochromis mossambicus exposed to copper in fresh water. Am. J. Physiol. 277, R320 – 331. Dunderski, J., Stanosevic, J., Ristic, B., Trajkovic, D., Matic, G., 1992. In vivo effects of cadmium on rat liver glucocorticoid receptor functional properties. Int. J. Biochem. 24, 1065 – 1072. Flik, G., Perry, S.F., 1989. Cortisol stimulates whole-body calcium uptake and the branchial calcium pump in freshwater rainbow trout. J. Endocrinol. 120, 75 – 82. Freedman, L.P., Luisi, B.F., Korszun, Z.R., Basavappa, R., Sigler, P.B., Yamamoto, K.R., 1988. The function and structure of the metal coordination sites within the glucocorticoid receptor DNA binding domain. Nature (London) 334, 543 – 546. Fu, H., Steinebach, O.M., Van den Hamer, C.J.A., Balm, P.H.M., Lock, R.A.C., 1990. Involvement of cortisol and metallothionein-like proteins in the physiological responses of tilapia (Oreochromis mossambicus) to sublethal cadmium stress. Aquat. Toxicol. 16, 257 – 270. Hogstrand, C., Haux, C., 1990. A radioimmunoassay for perch (Perca flu6iatilis) metallothionein. Toxicol. Appl. Pharmacol. 103, 56 – 65. Hogstrand, C., Reid, S.D., Wood, C.M., 1995. Calcium versus zinc transport through the gills of freshwater teleost fish. J. Exp. Biol. 298, 337 – 348. Hogstrand, C., Galvez, F., Wood, C.M., 1996. Toxicity, silver accumulation and metallothionein induction in freshwater rainbow trout during exposure to different silver salts. Environ. Toxicol. Chem. 15, 1102 – 1108. Hsu, S.M., Raine, L., Fanger, H., 1981. Use of avidin-biotinperoxidase complexes (ABC) and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem. 29, 577 – 580. Hussain, S., Anner, R.M., Anner, B.M., 1995. Metallothionein protects purified Na+/K+-ATPase from toxicity in vitro. In Vitro Toxicol. 8, 25 – 30. Hyllner, S.J., Andersson, T., Haux, C., Olsson, P.E., 1989. Cortisol induction of metallothionein in primary culture of rainbow trout hepatocytes. J. Cell. Physiol. 139, 24 – 28. Laure´n, D.J., McDonald, D.G., 1987. Acclimation to copper by rainbow trout, Salmo gairdneri: physiology. Can. J. Fish. Aquat. Sci. 44, 99 – 104. Lee, P.C., Goodrich, M., Struve, M., Yoon, H.I., Weber, D., 1992. Liver and brain glucocorticoid receptor in rainbow trout, Oncorhynchus mykiss: down-regulation by dexamethasone. Gen. Comp. Endocrinol. 87, 222 – 231.
54
Z.C. Dang et al. / Aquatic Toxicology 51 (2000) 45–54
Makino, Y., Tanaka, H., Dahlman-Wright, K., Makino, I., 1996. Modulation of glucocorticoid-inducible gene expression by metal ions. Mol. Pharmacol. 49, 612–620. Maule, A.G., Schreck, C.B., 1990. Glucocorticoid receptors in leukocytes and gill of juvenile coho salmon (Oncorhynchus kisutch). Gen. Comp. Endocrinol. 77, 448–455. Mayer, G.D., Leach, D.A., Olsson, P.-E., Hogstrand, C., 1996. Transcriptional activation of the rainbow trout metallothionein-A gene by Zn and Ag. In: Andre´n, A.W., Bober, T.W. (Eds.), Proceedings of the 4th International Conference on the Transport, Fate and Effects of Silver in the Environment. University of Wisconsin, Madison, WI, pp. 179 – 184. McDonald, D.G., Wood, C.M., 1993. Branchial mechanisms of acclimation to metals in freshwater fish. In: Rankin, J.C., Jensen, F.B. (Eds.), Fish Ecophysiology. Chapman and Hall, London, pp. 297–332. McIntyre, W.R., Samuels, H.H., 1985. Triamcinolone acetonide regulates glucocorticoid-receptor levels by decreasing half-life of the activated nuclear-receptor form. J. Biol. Chem. 260, 418 – 427. Munoz, M.J., Carballo, M., Tarazona, J.V., 1991. The effect of sublethal levels of copper and cyanide on some biochemical parameters of rainbow trout along subacute exposition. Comp. Biochem. Physiol. 100C, 577–582. Muto, N., Ren, H.W., Hwang, G.S., Tominaga, S., Itoh, N., Tanaka, K., 1999. Induction of two major isoforms of metallothionein in crucian carp (Carassius cu6ieri ) by airpumping stress, dexamethasone, and metals. Comp. Biochem. Physiol. 122C, 75–82. Norey, C.G., Cryer, A., Kay, J., 1990. A comparison of cadmium-induced metallothionein gene expression and metal ion distribution in the tissues of cadmium-sensitive (rainbow trout, Salmo gairdneri ) and tolerant (stone loach, Noemacheilus barbatulus) species of freshwater fish. Comp. Biochem. Physiol. C. 97, 221–226. Olsson, P.E., Kling, P., Erkell, L.J., Kille, P., 1995. Structural and functional analysis of the rainbow trout (Oncorhyncus mykiss) metallothionein-A gene. Eur. J. Biochem. 230, 344 – 349. Olsson, P.-E., Kling, P., Hogstrand, C., 1998. Mechanisms of heavy metal accumulation and toxicity in fish. In: Langston, W.J., Bebianno, M.J. (Eds.), Metal Metabolism in Aquatic Environments. Chapman and Hall, London, UK, pp. 321 – 350 Chapter 8. Olsson, P.E., 1996. Metallothioneins in fish: induction and use in environmental monitoring. In: Taylor, E.W. (Ed.), Toxicology of Aquatic Pollution: Physiological, Cellular and Molecular Approaches. Cambridge University Press, Cambridge, UK.
.
Overnell, J., McIntosh, R., Fletcher, T.C., 1987. The enhanced induction of metallothionein by zinc, its half-life in the marine fish Pleuronectes platessa, and the influence of stress factors on metallothionein levels. Experientia 43, 178 – 181. Pottinger, T.G., 1990. The effect of stress and exogenous cortisol on receptor-like binding of cortisol in the liver of rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 78, 194 – 203. Reichardt, H.M., Schutz, G., 1998. Glucocorticoid signalling — multiple variations of a common theme. Mol. Cell. Endocrinol. 146, 1 – 6. Rosewicz, S., McDonald, A.R., Maddux, B.A., Goldfine, I.D., Miesfeld, R.L., Logsdon, C.D., 1988. Mechanism of glucocorticoid receptor down-regulation by glucocorticoids. J. Biol. Chem. 263, 2581 – 2584. Seidelin, M., Madsen, S.S., Byrialsen, A., Kristiansen, K., 1999. Effects of insulin-like growth factor-I and cortisol on Na+/K+-ATPase expression in osmoregulatory tissues of brown trout (Salmo trutta). Gen. Comp. Endocrinol. 113, 331 – 342. Shrimpton, J.M., McCormick, S.D., 1999. Responsiveness of gill Na+/K+-ATPase to cortisol is related to gill corticosteroid receptor concentration in juvenile rainbow trout. J. Exp. Biol. 202, 987 – 995. Shrimpton, J.M., Randall, D.J., 1994. Downregulation of corticosteroid receptors in gills of coho salmon due to stress and cortisol treatment. Am. J. Physiol. 267, R432 – 438. Tujague, M., Saligaut, D., Teitsma, C., Kah, O., Valotaire, Y., Ducouret, B., 1998. Rainbow trout glucocorticoid receptor overexpression in Escherichia coli: production of antibodies for western blotting and immunohistochemistry. Gen Comp. Endocrinol. 110, 201 – 211. Uchida, K., Kaneko, T., Tagawa, M., Hirano, T., 1998. Localization of cortisol receptor in branchial chloride cells in chum salmon fry. Gen. Comp. Endocrinol. 109, 175 – 181. Weber, D.N., Eisch, S., Spieler, R.E., Petering, D.H., 1992. Metal redistribution in largemouth bass (Micropterus salmoides) in response to restrainment stress and dietary cadmium: role of metallothionein and other metal-binding proteins. Comp. Biochem. Physiol. 101C, 255 – 262. Wendelaar Bonga, S.E., Lock, R.A.C., 1992. Toxicants and osmoregulation in fish. Neth. J. Zool. 42, 478 – 493. Wendelaar Bonga, S.E., Van der Meij, J.C.A., 1989. Degeneration and death, by apoptosis and necrosis, of the pavement and chloride cells in the gills of the teleost Oreochromis mossambicus. Cell Tissue Res. 419, 454 – 459. Wendelaar Bonga, S.E., 1997. The stress response in fish. Physiol. Rev. 77, 591 – 625.