Killifish are euryhaline teleosts that adapt to rapid changes in the salinity of the seawater. It is generally accepted that acclimation to seawater is mediated by cortisol activation of the glucocorticoid receptor (GR), which stimulates CFTR mRNA expression and CFTR-mediated Cl− secretion by the gill. Because there is no direct evidence in killifish that the GR stimulates CFTR gene expression, quantitative PCR studies were conducted to test the hypothesis that cortisol activation of GR upregulates CFTR mRNA expression and that this response is required for acclimation to seawater. Inhibition of the GR by RU-486 prevented killifish from acclimating to increased salinity and blocked the increase in CFTR mRNA. In contrast, inhibition of the mineralocorticoid receptor by spironolactone had no effect on acclimation to seawater. Thus acclimation to increased salinity in killifish requires signaling via the GR and includes an increase in CFTR gene expression. Because arsenic, a toxic metalloid that naturally occurs in the aquatic environment, has been shown to disrupt GR transcriptional regulation in avian and mammalian systems, studies were also conducted to determine whether arsenic disrupts cortisol-mediated activation of CFTR gene expression in this in vivo fish model and thereby blocks the ability of killifish to acclimate to increased salinity. Arsenic prevented acclimation to seawater and decreased CFTR protein abundance. However, arsenic did not disrupt the GR-induced increase in CFTR mRNA. Thus arsenic blocks acclimation to seawater in killifish by a mechanism that does not disrupt GR-mediated induction of CFTR gene expression.
- cystic fibrosis transmembrane conductance regulator, chloride channel
- ion transport
- chloride secretion
- environmental toxicant
the euryhaline teleost Fundulus heteroclitus (i.e., killifish) is a model organism that is utilized extensively to study the homeostatic mechanisms regulating salt balance and to study the effects of environmental toxicants on cellular function. Killifish can acclimate to abrupt and dramatic changes in the salinity of seawater (14, 25–27, 29, 35). The ability to acclimate to increased salinity requires the gills to switch from NaCl absorption (freshwater) to NaCl secretion (seawater) to maintain NaCl homeostasis (25, 44). Cl− secretion by mitochondrion-rich cells in the gill is a two-step process: Cl− uptake across the basolateral membrane is mediated by the Na+-K+-2Cl− (NKCC1) cotransporter, and the Cl− that enters the cell via the NKCC1 cotransporter is secreted into seawater via CFTR Cl− channels located in the apical plasma membrane (25). Cl− secretion generates an apical-negative, transepithelial voltage that provides the driving force for the paracellular secretion of Na+. Upregulation of NaCl secretion subsequent to increased salinity is mediated by an increase in the number of mitochondrion-rich cells and by increases in CFTR, Na+-K+-ATPase, and NKCC1 (25, 26, 28, 35). Plasma levels of several hormones, including growth hormone, insulin-like growth factor, and cortisol, rise when the salinity is increased, and these hormones play an important role in acclimation (11, 23, 27). In killifish, transfer from freshwater to seawater is accompanied by a rapid (1 h) and dramatic increase in plasma cortisol levels that precedes increased CFTR gene expression (25, 27, 29, 35). Two glucocorticoid receptors (GR) have been identified in trout, and three have been identified in cichlid fishes (9, 12). Although there is no evidence that fish synthesize aldosterone, mineralocorticoid receptors (MR) have been identified in fish and cortisol activates MR in fish (5, 12). It has long been assumed that cortisol activation of the GR is required for increased CFTR gene expression in killifish and thereby the ability to acclimate to increased salinity. Although the importance of cortisol to acclimation in killifish is widely accepted, there is little direct proof to support this concept. A recent study examined the role of the GR in the ability of killifish to acclimate to a change in salinity from 10 to 150% seawater (26). RU-486, a GR antagonist, reduced cAMP-stimulated CFTR-mediated Cl− secretion in killifish opercular membranes. Although this observation is consistent with a role of the GR in acclimation to seawater, the effect of RU-486 on CFTR mRNA was not evaluated. Thus one of two goals of this study was to test the hypothesis that cortisol activation of the GR is required for increased CFTR mRNA expression and thereby the ability of killifish to acclimate to increased salinity. The second goal, described below, was to determine whether environmental arsenic disrupts cortisol activation of the GR, thereby blocking the increase in CFTR mRNA expression and the ability of killifish to acclimate to increased salinity.
Arsenic disrupts hormone-GR signaling and is considered the number one environmental toxicant of concern both worldwide and in the United States, according to the World Health Organization and the Agency for Toxic Substances and Disease Registry. Arsenic is a toxic metalloid that naturally occurs in the environment. Although arsenical compounds are used to treat several diseases, exposure to inorganic arsenic in occupational settings, at toxic waste sites, and from natural sources in the environment, including well water, is a major health concern (1, 18, 20, 21). Chronic exposure to arsenic-contaminated water or air has been linked to cancers of the skin, lung, bladder, liver, and kidney and increased risk of type 2 diabetes, vascular disease, cardiovascular disease, and reproductive and developmental disorders (1, 18). The Environmental Protection Agency (EPA) has set a human health consumption criterion for total dissolved arsenic in seawater of 0.0175 parts per billion (ppb), a level that is estimated to pose increased risk to human health if exceeded (31). However, these concentrations are often exceeded, even in clean coastal waters that typically range from 1 to 3 ppb (31), and in polluted seawater, arsenic concentrations >1,000 ppb (13 μM) have been reported (7). The current EPA-regulated U.S. drinking water standard is 10 ppb.
The mechanism(s) by which arsenic increases disease risk is not clear. Arsenic at low concentrations (i.e., ∼1–100 ppb) is a potent endocrine disrupter, altering hormone-activated gene transcription mediated by the GR (6, 16). Chronic exposure to arsenic inhibits CFTR transcription in human T84 colon epithelial cells in vitro (22). In addition, acute exposure to arsenic reduces CFTR-mediated Cl− secretion (IC50 = 305 ppb) across the opercular membrane of the killifish, in part by inhibiting mitochondrial respiration (39). Because so little is known about the cellular mechanism(s) by which arsenic causes human disease and affects the function of ion channels and transporters and because arsenic disrupts transcriptional activation by hormone-bound GR, studies were conducted to determine whether arsenic affects the ability of killifish to acclimate to an increase in salinity by reducing cortisol-GR-stimulated CFTR gene expression.
We report that the GR antagonist RU-486 inhibited the salinity-induced increase in CFTR gene expression, as determined by quantitative PCR, and the ability of killifish to acclimate to increased salinity. By contrast, the MR antagonist spironolactone had no effect on acclimation. In addition, arsenic blocked the ability of killifish to acclimate to increased salinity. However, contrary to our hypothesis, arsenic did not alter cortisol-stimulated GR activation of CFTR gene expression, although it substantially decreased CFTR protein abundance.
MATERIALS AND METHODS
Studies were performed in compliance with Institutional Animal Care and Use guidelines approved by Mount Desert Island Biological Laboratory (MDIBL; A3562-01) and Dartmouth Medical School (A3259-01). Killifish were collected from Northeast Creek (Bar Harbor, ME) and held in aquaria containing running seawater at the MDIBL (Salisbury Cove, ME) or artificial seawater (Instant Ocean) at Dartmouth College for at least 2 wk to ensure acclimation to seawater. The osmolality and Cl− concentration were similar in seawater and Instant Ocean. Fish were acclimated to freshwater by a reduction in salinity of the water to 10% seawater for 2 wk and then replacement of the 10% seawater with “soft” freshwater for 2 wk (4) for an additional 2 wk (27). Acclimation to seawater was achieved by transferring freshwater-acclimated fish to 100% seawater. Some fish were exposed to arsenic in the seawater (sodium arsenite). Identical results were obtained at the MDIBL and Dartmouth.
Freshwater-acclimated fish were either untreated or injected intraperitoneally with vehicle (ethanol control), RU-486 (40 and 100 μg/g body wt), a GR antagonist, or spironolactone (100 μg/g body wt), a MR antagonist. The concentrations of these drugs have been used by others to induce steroid receptor blockade in fish (26, 34, 43). Subsequent to IP injection, fish were either returned to freshwater or placed in 100% seawater. Results in the untreated and vehicle-injected fish were similar (see below). Thus IP injections, which have the potential to induce stress and increase cortisol levels, had no overt effect on the parameters measured in this study.
Blood and tissue collection.
Fish were anesthetized in tricaine (MS-222; 0.1%), and blood was obtained by cardiac puncture in heparinized hematocrit tubes. Plasma was obtained by centrifugation (10,000 g, 3 min) and was stored at −80° C until Cl− concentration (P[Cl−]) was measured by amperometric titration using a Labconco chloridometer. Tissue was collected for isolation of mRNA for quantitative RT-PCR or for analysis of arsenic levels by inductively coupled plasma-mass spectrometer (ICP-MS; see below).
Acute (96 h) toxicity tests were performed according to recommendations given by the U.S. EPA (Environmental Monitoring Systems Laboratory, Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine Organisms) to examine the effects of arsenic on freshwater- and seawater-acclimated fish, as well as fish exposed to arsenic in freshwater and transferred to seawater (15). For the latter, freshwater-acclimated fish were exposed to arsenic in the seawater for 96 h. To determine the effects of this arsenic exposure on the ability of the fish to acclimate to changing salinity, these fish were transferred immediately to seawater containing no arsenic for 48 h, and mortality was observed during this period. A group of fish that were similarly exposed to arsenic in freshwater and returned to freshwater containing no arsenic for 48 h served as a control for the transfer experiment. No mortality was observed in this group. Arsenic test solutions were prepared with natural seawater or very soft reconstituted freshwater (4) from stock made with NaAsO2 (purum grade; Fluka, Buchs, Switzerland) dissolved in deionized water. Toxicity tests were conducted in 1-liter polyethylene aquaria (10 fish/aquarium) at 13°C with three replicate aquaria per treatment. Test water was aerated and replaced daily, and water samples were taken at the start and conclusion of each test for measurement of total ammonia, pH, salinity, and temperature. There were no appreciable differences in water quality parameters over the test periods.
A series of tests was also conducted to determine the time course of mortality following transfer from freshwater to seawater in fish receiving drug treatments (detailed above) and exposed to arsenic in freshwater. Test design was similar to that described above, with the exception that six replicate aquaria, each containing 10 fish, were used per treatment group, and mortality was assessed at 0, 24, 48, 72, and 96 h following transfer. Also, these arsenic tests differed from those described above in that fish were exposed to arsenic in freshwater for 48 h and then either retained in freshwater or transferred to seawater containing the same concentration of arsenic.
Isolation of RNA and quantitative RT-PCR.
Quantitative RT-PCR experiments were conducted to examine the effects of increased salinity, arsenic, and RU-486 on CFTR gene expression by the gill. To isolate total RNA from gill, tissue was obtained as described above and immediately stored in RNAlater (Ambion, Austin, TX). Total RNA was isolated from 30 mg of tissue [from 3 animals to reduce animal-to-animal variation (19)] per observation using the RNAeasy mini kit (Qiagen, Valencia, CA). RNA was treated with DNase (DNA-Free; Ambion) to remove contaminating DNA. Total RNA was quantified using spectrophotometric optical density (OD260/280) measurements (NanoDrop; NanoDrop Technologies, Rockland, DE), and RNA quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Wilmington, DE). Two-step RT-PCR was performed with 1 μg of total RNA (Retroscript reverse transcriptase; Ambion) and random decamers. Primers and probe for real-time PCR were synthesized using the Assays-by-Design service (Applied Biosystems, Foster City, CA). The sequences for killifish CFTR, NKCC1, and the α1a-subunit of the Na+-K+-ATPase were submitted for primer/probe design, and probe target was set to a predicted exon-exon splice junction. The sequences of the primers and probe used (5′ to 3′) are as follows: CFTR exon 24 forward, TGCAAGCATGGAGGAAAGC; CFTR exon 24 reverse, CCGGAAGGTTCCAGTCAGAAT; CFTR exon 24 internal probe, CCGCAGAAAGTCTTT; Na+-K+-ATPase α1a forward, GCTTTCTTCTCCACCAACTGTATTG; Na+-K+-ATPase α1a reverse, CGGTCGCCGGTGTTAATGA; Na+-K+-ATPase α1a internal probe, CTGCCAGAGGAATTG; NKCC1 forward, GGCTGGTGCCAACATTTCTG; NKCC1 reverse, AAGAAGGGTTCCTTTGGGTATAGC; and NKCC1 internal probe, CCTTGCTGACCCACAGATG.
The probes were labeled with 6-FAM dye with a minor groove-binding modification and nonfluorescent quencher on the 3′ end. The Assays-by-Design primers and probe, premixed to a concentration of 18 μM for each primer and 5 μM for each probe (equivalent to a 20× mix), were combined with TaqMan Universal master mix (Applied Biosystems) and cDNA diluted in RNase-free H2O in a 20-μl reaction and placed in a 96-well-format spectrofluorometric thermal cycler (ABI Prism 7700 sequence detection system). Duplicate and/or triplicate reactions of each sample were incubated at 95° for 10 min, followed by 40 cycles of 15 s at 95° and 1 min at 60°. In preliminary studies, quantitative RT-PCR products were run on an LMP agarose gel to confirm product size, subcloned into pCR4-TOPO (Invitrogen, Carlsbad, CA), and submitted for sequence analysis to confirm identity of the products. Dilutions of CFTR, Na+-K+-ATPase, and NKCC1 plasmid DNA prepared from the killifish quantitative RT-PCR products were used to construct a standard curve. The standard curves showed a correlation coefficient close to 1 (R2 > 0.99) and were linear over a 6-log range. Equivalent amplification efficiencies of standard and target molecules were observed, and SYBR Green melting curve dissociation analysis and sequencing revealed a single PCR product for each target. Raw data were analyzed; baseline and threshold values were set and gene expression was interpolated using the external standard curves. The cDNA generated during reverse transcription and used as template was quantified (NanoDrop), and data were calculated as gene expression (fg/ng cDNA).
ICP-MS analysis of tissue arsenic.
Gill tissue arsenic levels were measured by the Dartmouth Superfund Trace Metal Core Facility using a cold vapor/hydride generation magnetic sector ICP-MS (ICP-MS Element, Finnigan MAT) according to the methods of Klaue and Blum (20). Briefly, killifish were anesthetized as described above and rinsed with ultrapure water, and gill tissue was dissected, weighed, and stored frozen at −80°C. Gill tissue was digested under pressure using a sequential combination of nitric acid (optima grade) and hydrogen peroxide (optima grade).
Western blot analysis.
Western blot analysis of CFTR (clone 24-1, 1:500 dilution; R&D Systems), Na+-K+-ATPase (Na+-K+-ATPase a5 supernatant, 1 μg/ml; Developmental Studies Hybridoma Bank, University of Iowa), NKCC1 (T4, 1 μg/ml; Developmental Studies Hybridoma Bank), actin (clone C4, 1:1,000 dilution; MP Biomedicals), and heat shock protein 70 (HSP70; clone C92F3A-5, 1:2,000 dilution; Stressgen) in gill tissue was performed as described in detail previously (42). In preliminary studies, arsenic had no effect on actin and HSP70 protein abundance as determined by Western blot analysis.
Analysis of data.
Data are presented as means ± SE. Statistical significance of experimental maneuvers was determined using the paired or unpaired Student's t-test or ANOVA and the Tukey-Kramer multiple comparison test using GraphPad Instat (version 3.0a for Macintosh; GraphPad Software, San Diego, CA). A P value <0.05 was considered significant.
GR, but not MR, is required for seawater acclimation.
To determine whether activation of the GR is essential for killifish to acclimate to increased salinity, we left freshwater-acclimated fish untreated (control) or injected them intraperitoneally with either vehicle (ethanol, sham) or RU-486 (40 and 100 μg/g), a GR antagonist, and they were immediately either returned to freshwater or placed in 100% seawater (Fig. 1). There was no significant mortality of control or sham fish returned to freshwater or placed in seawater. All RU-486-treated fish returned to freshwater survived. By contrast, there was significant mortality (∼90% at 96 h in the 100 μg/g group and ∼50% at 96 h in the 40 μg/g group) in RU-486-treated fish transferred to seawater. These results are consistent with the conclusion that activation of GR by endogenous cortisol is required for killifish to successfully acclimate to increased salinity.
Recent studies have suggested that cortisol may activate the MR in fish and thereby play a role in acclimation to seawater (5, 12). To determine whether activation of MR is essential for killifish to acclimate to increased salinity, we injected freshwater-acclimated fish intraperitoneally with spironolactone (100 μg/g), a potent and specific antagonist of MR, and they were immediately either returned to freshwater or placed in 100% seawater. All spironolactone-treated fish returned to freshwater or placed in seawater survived (Fig. 1). These results are consistent with the conclusion that activation of the GR, but not the MR, by endogenous cortisol is required for killifish to acclimate to increased salinity.
The GR antagonist RU-486 disrupts Cl− homeostasis.
To examine the effect of RU-486 on Cl− homeostasis, we collected blood from fish for measurements of P[Cl−]. As shown in Fig. 2, transfer of freshwater-acclimated killifish to seawater increased P[Cl−]. This increase in P[Cl−] was greater in fish treated with RU-486. These data indicate that the RU-486-induced mortality in fish transferred from freshwater to seawater is most likely caused by the inability to maintain P[Cl−] by excreting Cl− via CFTR in the gills.
The GR antagonist RU-486 blocks seawater induction of CFTR gene expression.
Transfer of freshwater-acclimated fish to seawater is thought to enhance CFTR mRNA expression via increased cortisol, which activates the GR. Thus studies were conducted to determine whether RU-486 inhibits acclimation to seawater, at least in part, by blocking the cortisol-GR-induced increase in CFTR mRNA expression. First, studies were conducted to examine the time course of increased CFTR mRNA levels following transfer from freshwater to seawater. Transfer of freshwater-acclimated killifish to seawater increased CFTR mRNA expression in gill after a delay of 24 h (Fig. 3). This observation confirms several previous studies (27, 35) and established the 24-h time point following transfer to seawater as optimal for investigating changes in CFTR mRNA. At this time point, RU-486 completely inhibited the 2.5-fold increase in CFTR mRNA that was observed in fish transferred from freshwater to seawater (Fig. 4). This observation provides the first direct evidence in killifish that activation of the GR by endogenous cortisol mediates an increase in CFTR mRNA expression when fish are transferred from freshwater to seawater.
Arsenic inhibits seawater acclimation.
To determine whether arsenic is a potential endocrine disruptor in killifish, we performed arsenic toxicity tests in fish acclimated to freshwater and seawater and in freshwater-acclimated fish transferred to seawater. The effects of arsenic on mortality, evaluated for 48 h following arsenic exposure, are presented in Table 1. The LC10 (lethal concentration to 10% of fish) and the LC50 (lethal concentration to 50% of fish) values were equivalent in freshwater- and seawater-acclimated fish. However, in freshwater fish transferred to seawater, the LC50 value decreased significantly from 230.4 to 163.2 μM. Thus arsenic was more toxic to fish stressed with a freshwater-to-seawater challenge.
To examine in more detail the time course of arsenic toxicity, we treated freshwater-acclimated fish with arsenic (106 or 160 μM) for 48 h and then either retained them in freshwater containing the same concentration of arsenic or transferred them to seawater containing the same concentration of arsenic (Fig. 5). There was no mortality of control fish (no arsenic) returned to freshwater or transferred to seawater. In addition, there was no mortality of arsenic-exposed fish at the lower dose when placed in seawater. By contrast, after 48–96 h in seawater containing 160 μM arsenic, there was significant mortality (∼45% at 96 h). Together, these data demonstrate that arsenic in the water inhibits the ability of killifish to acclimate to an increase in salinity at concentrations that are not otherwise toxic to fish not challenged with increased salinity.
Arsenic disrupts Cl− homeostasis.
To determine whether arsenic caused mortality by disrupting Cl− homeostasis, we measured P[Cl−] in freshwater-acclimated fish transferred to seawater. Arsenic at nonlethal concentrations (66 and 106 μM) significantly increased P[Cl−] in freshwater-acclimated fish exposed to seawater (Fig. 6). A higher concentration of arsenic (160 μM), which also caused some mortality, elicited an even greater increase in P[Cl−] when freshwater-acclimated fish were exposed to seawater. These data suggest that arsenic disrupts Cl− homeostasis.
Arsenic does not inhibit seawater induction of CFTR gene expression.
Our initial hypothesis was that arsenic prevents acclimation to seawater by disrupting GR activation of CFTR mRNA expression. To test this hypothesis, we exposed freshwater-acclimated fish to arsenic (66, 106, or 160 μM) for 48 h and then directly transferred them to seawater containing the same arsenic concentrations for 24 h. Subsequently, CFTR mRNA in gill was measured using quantitative RT-PCR (Fig. 7). As observed in earlier studies, transfer of freshwater-acclimated fish to seawater increased CFTR mRNA levels (25, 26, 28, 35). However, contrary to our hypothesis, arsenic concentrations of 66, 106, or 160 μM did not significantly reduce CFTR mRNA levels at 24 h (compared with control fish). Because arsenic caused significant mortality only after 48 h (see Fig. 5), we speculated that arsenic might decrease CFTR mRNA levels at 48, 72, or 96 h, but not at 24 h. Thus we conducted studies to examine the longer-term effects of arsenic on CFTR gene expression (Fig. 8). However, prolonged exposure to arsenic [96 h, a time point associated with significant (45%) mortality] also failed to decrease CFTR mRNA levels. These results do not support our initial hypothesis that arsenic prevents acclimation to seawater by disrupting GR activation of CFTR gene expression.
Arsenic decreases CFTR protein abundance.
Additional studies were conducted to determine whether arsenic abrogated the ability to acclimate to increased salinity by reducing CFTR protein levels. In these experiments, freshwater-acclimated fish were exposed to arsenic (160 μM) or vehicle for 48 h and then directly transferred to seawater containing the same arsenic concentrations or to vehicle for 1, 2, or 3 days. Subsequently, CFTR protein in gill was measured using Western blot analysis (Fig. 9). In the absence of arsenic, transfer from freshwater to seawater elicited a small increase in CFTR protein expression in gill (Fig. 9). However, in fish exposed to arsenic and transferred from freshwater to seawater, CFTR protein abundance was negligible and significantly less at every time point compared with fish not exposed to arsenic. Thus, although arsenic did not alter the increase in CFTR mRNA in freshwater fish transferred to seawater, arsenic dramatically reduced CFTR protein abundance in freshwater fish transferred to seawater. Thus it is likely that arsenic blocks acclimation to seawater by reducing CFTR protein abundance.
Arsenic has no effect on Na+-K+-ATPase or NKCC1 gene expression or protein abundance.
Cl− secretion by mitochondrion-rich cells in gill is a two-step process: Cl− uptake across the basolateral membrane is mediated by the NKCC1 cotransporter, and the Cl− that enters the cell across the basolateral membrane is secreted into seawater via CFTR Cl− channels located in the apical plasma membrane (25). The Na+-K+-ATPase, by maintaining a low intracellular Na+ concentration, facilitates Cl− uptake across the basolateral membrane via NKCC1. Thus arsenic also may have disrupted Cl− homeostasis by affecting Na+-K+-ATPase and/or NKCC1 gene expression or protein abundance. Arsenic (160 μM) did not affect Na+-K+-ATPase (Fig. 10) or NKCC1 (Fig. 11) mRNA expression or protein abundance when fish were transferred from freshwater to seawater.
Arsenic accumulates in killifish gill.
Given the effects of arsenic on P[Cl−] and the lack of overt toxicity at water concentrations up to 106 μM, we were interested in the extent to which arsenic actually accumulates in the tissues of the killifish over the course of these experiments. Arsenic levels were measured in the gills of control and exposed killifish by using high-resolution ICP-MS. Freshwater-acclimated fish were maintained in freshwater or transferred to seawater containing 0, 106, or 160 μM arsenic for 96 h (Fig. 12). The background arsenic concentration was 0.002 ± 0.001 μmol/g in the gill of freshwater fish and 0.004 ± 0.001 μmol/g in seawater fish in nominally arsenic-free seawater (Fig. 12; P = not significant). Gill arsenic levels increased significantly to 0.015 μmol/g in fish exposed to 106 μM arsenic in the water, a 2.4-fold increase, and increased to 0.057 μmol/g in fish exposed to 160 μM arsenic in the water, a 12-fold increase (Fig. 12). Thus, despite high concentrations of arsenic in the water, the internal concentrations of arsenic in the gills of exposed fish were relatively modest. The relevance of these tissue levels of arsenic to levels measured in fish exposed to arsenic in the environment and with regard to endocrine disruption are discussed below.
The major new observations in this report are that the GR antagonist RU-486 inhibits CFTR gene expression, increases P[Cl−], and blocks the ability of killifish to acclimate to an increase in salinity from ∼0 to 100% seawater. By contrast, spironolactone, an antagonist of MR, had no effect on acclimation to seawater. Thus the MR does not appear to be essential for killifish to adapt to increased salinity. Arsenic also blocked the ability of killifish to adapt to increased salinity by reducing CFTR protein abundance. However, arsenic had no effect on the GR activation of CFTR mRNA expression, suggesting that arsenic affects CFTR protein abundance at one or more posttranscriptional steps.
Our results with RU=486 are consistent with recent observations that RU-486 inhibits cAMP-CFTR-mediated Cl− secretion by killifish operculum (26). Together with our quantitative RT-PCR results, these data are consistent with the conclusion that endogenous cortisol activates the GR and thereby increases gill CFTR mRNA expression. This enhances CFTR-mediated Cl− secretion by gill and operculum, thereby facilitating the ability of fish to acclimate to increased salinity. It should be noted that in the study by Marshall et al. (26), RU-486 did not cause mortality in fish transferred from brackish water (10% seawater) to 150% seawater over a 48-h period. By contrast, there was significant mortality (∼90%) over 96 h in our study, in which RU-486-treated fish were transferred from freshwater to 100% seawater. This is likely due to differences in experimental protocols, primarily the degree to which the salinity was changed, which in the present study was more severe and likely to impose more stress on osmoregulatory mechanisms.
Our data demonstrate that the GR and not the MR plays an essential role in acclimation to seawater and upregulation of CFTR mRNA expression in killifish gill and operculum. RU-486, a potent GR antagonist, but not spironolactone, an MR antagonist, blocked acclimation to seawater. RU-486 also blocked the seawater-stimulated increase in CFTR mRNA. It is not known whether the MR is expressed in killifish, although the MR is expressed in other teleosts. Thus the negative result with spironolactone could be due to an absence of MR in killifish. It also must be considered that endogenous cortisol may activate CFTR mRNA expression and Cl− secretion by binding to other steroid receptors such as the progesterone or estrogen receptors. However, we think this is unlikely because of the known ligand specificity of these highly conserved receptors in other systems and because estrogen has been shown to inhibit osmoregulation in killifish (24). Activation of the estrogen and progesterone receptors also reduces CFTR Cl− secretion in pancreatic and T84 cells (36, 41). Moreover, progesterone receptors are not highly expressed in the gill (32). Together, these results strongly point to the GR being a major regulator of this adaptive response in killifish.
Arsenic prevented acclimation to increased salinity by a mechanism that did not involve disruption of GR activation of CFTR mRNA expression. Measurements of cellular arsenic accumulation in gill revealed that arsenic in the water significantly increased tissue levels of arsenic to 0.057 μmol/g, a value comparable to levels previously observed to disrupt hormone regulation of gene expression in killifish liver and mammalian cells. For example, in killifish exposed to low levels of arsenic (10.4 μM arsenic in 20% seawater) for 2 wk, liver arsenic concentration increased to 0.13 μmol/g, a concentration that inhibited cortisol activation of LDH-B mRNA expression in liver but not cortisol activation of phosphenolpyruvate carboxykinase mRNA expression (5a). These observations may explain, at least in part, why in the present study arsenic had no effect on the GR stimulation of CFTR mRNA expression. Nonetheless, the ability of arsenic to block acclimation to seawater, increase P[Cl−], and decrease CFTR protein abundance at concentrations that did not alter CFTR mRNA expression suggest that other mechanisms are principally responsible for this block in the adaptive response. Since the GR mediates many cellular processes, these effects could result from disruption of GR signaling by arsenic at these other steps (2, 40, 46), but given the many cellular targets and effects of arsenic, it also is possible that these effects are mediated by other targets and other mechanisms.
Because arsenic dramatically decreased CFTR protein abundance, increased P[Cl−], and also caused significant mortality, these data suggest that arsenic disrupted the ability of fish to regulate NaCl homeostasis. Several mechanisms that do not involve disruption of GR activation of CFTR mRNA expression should be considered. First, arsenic dramatically reduced CFTR protein abundance, an observation consistent with the view that arsenic has posttranslational effects on CFTR. Recent studies have demonstrated that low-level acute and chronic arsenic exposure leads to upregulation of genes associated with the proteosome-mediated protein degradation pathway (45) (Hamilton JW, Ihnat MA, Davey J, Hampton T, and Andrew A, unpublished studies) and accumulation of ubiquitinated proteins (8). Thus it is likely that arsenic reduced CFTR protein abundance, at least in part, by activating proteosomal degradation of CFTR. In a previous study, we reported that arsenic rapidly (minutes) and reversibly inhibited CFTR-mediated Cl− secretion across the operculum by inhibiting mitochondrion respiration (39). It is unlikely that this effect of arsenic extended to the 48- to 96-h treatment period in this study, because inhibition by arsenic of mitochondrion respiration is transient, and after 48–96 h, arsenic has no effect on mitochondrial respiration in either operculum or kidney (Miller D, Shaw JR, and Stanton BA, unpublished observations). Consistent with the view that the concentration of arsenic used in the present study did not cause cell stress, arsenic had no effect on actin or HSP70 expression as determined by Western blot analysis (see materials and methods). Moreover, arsenic did not affect Na+-K+-ATPase or NKCC1 mRNA expression or protein abundance. Thus the chronic effects of arsenic on killifish do not appear to involve a general cellular stress response or to be nonspecific. Arsenic may have other effects on cellular function that reduce CFTR-mediated Cl− secretion, including effects on signaling pathways, and other posttranslational mechanisms that affect CFTR function. Elucidation of other mechanism(s) by which arsenic disrupts osmoregulation in killifish is under investigation and is beyond the scope of the present report.
It is interesting to note that the adverse effect of arsenic in these experiments was exacerbated when the animals were stressed with an increase in salinity. The same arsenic concentrations that were lethal in this salinity challenge were not overtly toxic in either freshwater- or seawater-adapted animals. This is a common paradigm for arsenic effects, where it may appear to have little or no obvious effects by itself but in combination with another agent or stressor will greatly exacerbate the other condition. For example, inorganic arsenic alone has been consistently negative for carcinogenesis in animal bioassays, but arsenic in drinking water greatly enhances the tumorigenicity of ultraviolet radiation in a mouse model (33). Similar synergistic or potentiating effects of arsenic have been seen for endocrine disruption (6, 16), angiogenesis (17, 38), and DNA damage and repair (3, 13). Human epidemiology studies also have shown that arsenic is strongly synergistic with cigarette smoking for risk of lung cancer (10, 37). This has interesting experimental and practical implications in evaluating the pathophysiological effects of arsenic.
Although the concentration of arsenic in seawater that blocked acclimation in the current study is higher than that observed even in polluted seawater, we believe the data in this study are environmentally relevant, because the tissue arsenic levels we observed in our 24- to 96-h experiments are lower than the tissue values observed in wild killifish collected from polluted waters. For example, in a California study, killifish were collected from contaminated water and had tissue arsenic levels of ∼0.41 μmol/g (30), a value some seven times higher than that of the level of 0.057 μmol/g we observed in our experiments. Thus, although higher external levels may be required experimentally in short-term studies to achieve sufficient intracellular levels, it is important to compare tissue and intracellular levels rather than external levels when comparing effects.
In summary, we report that the GR appears to be essential for killifish to successfully acclimate to increased salinity. By contrast, the MR is not essential for killifish to acclimate to increased salinity. In addition, arsenic blocked the ability of killifish to acclimate to increased salinity. Interestingly, this mechanism does not appear to involve disruption of GR activation of CFTR gene expression. However, the ability of arsenic to block acclimation, while simultaneously increasing plasma chloride levels and decreasing CFTR protein abundance suggests that these effects are mediated by an effect of arsenic on CFTR abundance and/or function. We hypothesize that arsenic acts as an endocrine disruptor in killifish by altering one or more GR-mediated posttranscriptional steps that regulate CFTR protein abundance. Future studies need to focus on determining whether this involves other arsenic-sensitive GR-mediated events and/or other non-GR-dependent pathways.
This study was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01 DK-45881 (to B. A. Stanton), National Institute of Environmental Health Sciences (NIEHS) Superfund Basic Research Program Project Grant P42 ES-07373 (to B. A. Stanton, J. R. Shaw, J. W. Hamilton), NIEHS Center for Membrane Toxicity Studies at Mount Desert Island Biological Laboratory (MDIBL) Grant P30 ES-03828 (to B. A. Stanton and J. R. Shaw), a Research Development Program Grant from the Cystic Fibrosis Foundation (to B. A. Stanton), a Cystic Fibrosis Foundation award (to C. R. Stanton), a pilot project grant from Dartmouth's Center for Environmental Health Sciences (to B. A. Stanton and J. R. Shaw), and a MDIBL New Investigator Award (to J. R. Shaw). K. Gabor and L. Durant were supported by the National Center for Research Resources Maine IDeA Network of Biomedical Research Excellence Grant 2-P20-RR016463-04. E. Hand, A. Lankowski, and R. Thibodeau were supported by a National Science Foundation Research Experience for Undergraduates site at MDIBL (NSF DBI-0453391).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society